DL413iD

REV 1

p+

Radio, RF*
and Video
% _~
Applkhions

MOTOROLA

MOTOROLA
@

All products are sold on Motorolas Terms & Conditions of Suppl~$ae;ng

a product covered by this document I he Customer agrees to be bound
constitutes
or forms pafl of a contract (with the exception of the contents
by I hose Terms & Conditions and nothng contained in ttis docuhnt
of this Notice). A copy of Motorola,s Terms a Conditio,~s of ~pply is avail~le on CeqUeS1.

,.,
Motorola

reserves

the tight to make chang&+$~$~Yudher

notice to any products

herein.

Motorola

makes

no warranty, repr-entation

or

guarantee regarding Ihe suittility of its p~~,@

f? any pa ficular purpose, nor does Motorola assume any Habitily atising ot of ihe apphcation
or use of any product or circuit, and sp*@@g,~sclaims
any and all liabdhy, including without timitation con=quential or incidental damages,
TypicalParameters ~n and do varp~~~faril
appl!~t ions. All operating parameters, including Typical, must be validated for each customer
aPPl)cal ion by customefsi~hni calw~otoroladms
not mnvey any Iimnse under its patent rights nor the rights of others. Motorola products
are not designed, intended, O( ~.ho~ed
for use as mmpnents
in systems inlended for surgiwl implant into the body, or other applications
intended to SUpPOfl or sust~RTe~a$dr
ay other application in which the Iailure at the Motorola product muld create a situation where personal
iniu~ or death may occu,$~,,~,$
Buyer purch%e or use Motorola products for any such unintended or unauthorized apph~tion, Buyer shall
indemnify ad hold Mo~rol&,$Ad its olfimm, employees, subsidiaries, affiliates, and dist ributors harmless against all claims, COS!S,damages, and
expenses, and re:~~~@2~torney
fees arising out of, directly or indirectly, any claim at personal injury or death associated with such unintended
or unauthorized ~,
d~n II such claim alfeges that Motorola was negligent regarding the design or manufacture of the pad. Motorola and @ are
registered tr$$emk~kf
Motorola, Inc. Motorola, Inc. is an Equal ~potiunity/Alfirma! ive Action Employer.
,..
.1

.*$?,

*..

~1. ..+
Thwt$~%~er
should ensure that it has the most up to date version of the document by contacting its local Motorala office. This document
s$~(~es
any earlier d%mentat ion relining to the products relerred to herein. The informa!on contained in this document is current at the date
,.,

>,M:~~~LcatiOn.
,<,
~.,,
?!,!. $
,,>::~i.>.,,
..i>,,!
.+..,~~$,,
,,,,,,
.?>~~.

<:&*, -,,.
, \ >$,~.
.<~..,.
ii

II may subsequently be updated, revised or withdrawn.

Includes literature available at June 1994

All trademarks recognized.
@ MOTOROU
All Hghts

I NC.
Resewed

Rrst Edition DL4131D, 1991

DL413/D Rev. 1, 1994
Ptinted in Great Bribin by Tavistock Press (Bedford) Ltd. 5000 9/94

Preface

This compilation

of Application

etc. was originally

Notes, Engineering

to that are currently

and may only be supported

is referenced.
Also

included

documents
devices

in the series

that

may

by the European

Bulletins,

bterature

Design

of Motorola

available

for limitedU~~R%ution
in a specific region
~~
by the country,,@,,,,,,~
@@~?nof the document in which it
*,, ,1+/
*,
t.,(?!,
.\t;>\
for co~~T&~e}ess
and historical

no longer,,.($~~%ailable

are referenced

Centre

Concepts,

Ltd,,
. . ~.i
in Milton Keynes, England, and has subsequently
gained worldwide accepta+g,~$p
%...Lii~+.
Because of the worldwide popularity of the Application
Manuals Series itJ$~~,pd]f3.,
~.,
,,f ?!,,t,..
.
tant for the reader to take note of the following:
,,~,
.*
-~,
,., .,,.
,
The various Application
Notes, Engineering
Bulletins,
Design i$~~88pts,
etc.
*,$,
\.,t.*
which are included
were developed
at Design Centres s~[~~@gltally located
,;a:.$*}+*K$.
throughout
the global community
and many were originq~!y%~tfen
to support a
.,$*\>:
local need. Whilst the basic concepts of each of the ~o~~~afions
included may
?:{$
,N<:++.
have broad global applicability,
specific Motorola,Jq@mkonductor
parts may be
referred

published

or perha~,s}sirnply,

individually

the original

significance
because

document

are

obsolete

is out of print,

Such items are marked in ttis:~able of Contents, Cross Reference, Abstracts and
2*.
on the first page of th+e%$%ument with the letters HI to indicate that these
documents
All the
included

are inclu,,~kf~~
Historical
, .~c~
-a
~,),
.,$~

Applicatiq~~i$#tes,
to ~~~~~~e

products. gM~~$er,
Series@rtR~&ser
~.?
~ \:yi
con~rm~~~oduct
,..+~
.,,,,
.ri#ikyou
,<;:hh,~+
..,,,
?>
!$3,%>
>,:*.~,~:~;,i,
\
i,.,L ,,
$tl~ +1,
$$: :.:,*:.
,>$.>~;.!t
.l.+ti.,~
$...\::\
,J:,$\~
.,,,, .

Engineering

the users
before

should
availability

Information

attempting

contact

Bulletins,

knowledge

only.
Design

to design-in

the local Motorola

and if application

Concepts,

and understanding

support

a device

referenced

supplier

or sales

is available.

etc. are

of Motorolas
in this
office

to

Other books in this series include:

DL40a/D Rev. 1

8-bit MCU Applications

DL409/D Rev. 1

16/32-bit Applications

DL4 10/D Rev. 1

Power Applications

DL41 l/D Rev. 1

Communications

DL4 12/D Rev. 1

Industrial Control Applications

Manual
Manual

Manual

Applications
,,
~~~:;,.

-...-_.-

_ ...

Contents
page
Device Cross Reference ........ ....... ..... .. .. .. ....... .. .. ..... ..... . ..... ........ .. ..... .. ...... ....... ....... .. . ........ 9
Abstrscts of Applications
Applications

Documents

..

Documents

. 13
~:t.,,,

AN438
AN448
AN460
AN463
AN479
AN749
AN756
AN790
AN879
AN925
AN932
ANI019
AN102O
AN102I
AN1022
ANI025
AN1 027
AN1028
AN1029
AN103O
ANI032
AN I 033
ANI034
AN I 037
ANI039
AN104O
AN1041
AN1044

Contents (continued)

Device

Cross Reference

This quick-reference list indicates where specific

components are featured in applications documents
reproduced in this Manual.
2N6439 ...................................... EB77
CA2820 ..................................... ANI022
CR2424 ..................................... ANl O2l
AN1047
.....
.. .
AN1306
CR2425 . . . . . . .
. ....... ..... ANI021
AN1047
LTIOOI . . .. . . . . . ..
. ....... AN102O
LT1817 ...................................... ANl O2O
LT1829 ...................................... ANl O2O
LT5839 ...................................... ANl O2O
MCIOE1 II ................................ AN14O5
MCI OE211 ................................ AN1405
MC1OH6OX
AN1402
MCI OH641 ................................ ANI4O5
MC10H64x . ... . . . . . . . . . . AN1401
MC1OH66O . . . . . . . .. .. . . . . AN1092
MC68HC05B6
EB411
MC68HC05E0
AN460

10

11

12

Abstracts
intederance on a multi-sync colour monitor. It uses a
Iowmst MC44602P2current mode controller-designed
apecificallyfor driving high voltage bipolar transistors
with an MJH1 8010 switchmode power transistor.
,*\
AN749
Broadband
Transformers
and
.i:~.~~+,
Broadband Combining Techniques f@- ~~),

AN438
300 W, 8k108MHz
Amplifier Using the
TP1940 MOSFETS Push-Pull Transistor
Provides the design of an efficient 300W amplifier with
high power gain, compact physical layout and operation on a 50V power supply. It uses the TPI 940, a high
power, high gain, broadband push-pull Power MOSFET with low Reverse Transfer Capacitance. Includes
circuit, pafls list, PCB aflwork and component layout.
FLOP
Teletext
AN448
Microcontrollers

This application note provides a numQ,#$,Of~~ractical

examples of broadband transform~fdk.~~
applications. It includes detailed design ko~~~~~and
parformancedata, and discusses pw~~@M!ngtechniques
that are useful in design ing$,~@\~@@er RF amplifiers.

using M6805

The -7 members of Motorolas M68HC05 MCU family

provide a cost-effective method of adding On Screen
Display (OSD) to TVs and VCRs. This note describes
an example of Full Level One Feature (FLOF) Teletext
control software written for the MC68HC05T7to control
type 5243 Teletext chips. Around 3K bytes of ROM are
used, allowing the code to fit with tuning, OSD and
stereo functions into the 7.9 Kbytesofthe MC68HC05T7.
The example software includes the Spanish implementation of Packet 26; Packet 26 allows for the substitution of specific characters for a particular country.
AN460
An RDS Decoder
MC68HC05E0

AN756
Crystal SwitcMog&:hods
MC12061 Oscillato+~e,

for MC12060/

This repofldiscu~~~h~thods
of using diodes to select
se~esrasOna~~*~@se
lactrO nlcally. Circuit designs
suitable for ~~~~~ %Yystal frequencies from 10OkHz tO
20 MHz are ~~~~}dped, with emphasis on minimizing
freque,~~pulling,
Although developed for use with the
MCI~o~~~W
MC1 2061 integrated circuit crystal oscillat~~s?$~ techniques will generally be useful in any
~,&p,~?&$]on where it is necessary to select electronically
,: we of a group of crystals with minimum disturbance to
t$.~,.y tti~ series resonant frequency of the selected cystal.

Using the

The Radio Data System (RDS) adds digital datacaPa- ~,

bility to VHF FM transmissions
on band II (87.5 &:$
108 MHz). The system is in use in the UK and in sey~r~ ~
other European countries, and it is intended that ~~~a~
be adopted eventually by most of Western Europej it~
defined by EBU Technical Document 324,4: lnf8rmation is transmitted in groups of four 26-Qj~~:ks
on a
supressed 57kHz sub-carrier. This n~:~~sdtibes
an
MC66HC05E0-based clock/radioa~l~~~l~n:
it includes
*/: $
a complete software listing.
.:.!,y~
,*,Y
.$I,>,.!,$l,..
h.~,
68HC05K0 lnfra-~:$+~~~ote
COntrO/
AN463
In addition to the samei~P~<#nd
registers as other
members of the M68&@~$amily
the MC66HC05K0
has a 15-stage
multl;$~~pn
timer and 10 bidirectional
1/0 lines. A mas$jb~mrs
available for software programmable pg&~<$~@s on all the 1/0 pins; 4 of the Pins
are capable~ga$erating
interrupts. It is ideally suited
for remot$-co~f$et keyboard applications because the
pull-do@~#and
the interrupt drivers on the poti pins
allow$~~~rds
to be built without any external compon~k;$gcept
the keys themselves. This application
~~
use of many of the on-chip features to control a
:,~~~~ra-red
remote control.
.,<a+.ix
~$
-.f-

ttAh790
Thermal Rating of RF Power Transistors
,\?
Reliability is of primary concern to most transistor
users. The degree of reliability achieved in practice is
controlled by the device user because he determines
environmental conditions and the stress levels applied.
Knowledge of the basic physical properties of the
mate flals, and the methods used to calculate thermal
resistance, will assist the usar in transistor selection
and equipmant design. This nota clarifies and corrects
some long-standing industry-wide assumptions about
thermal resistance and high temperature derating,
AN879
Monomax: Application
of the MC13001
Monochrome
Television Integrated Circuit
This application note prasents a complete 12 black
and white line-operated television raceiver including
artwork for the printed circuit board. It is intended to
provide a good starting point for the first-time user.
Some of the most common pitfalls are overcome and
the significance of component selections and locations
are discussed.
AN925
UHF Preamplifier
Dual-Gate GaAs FET

Universal Input Voltage Range Power

.,,:$ 4$.@79
$?
Supply for High Resolution Monitors with
>,,+!.
,,~
Multi-Sync Capability
$..
,.{::,..i,,i+,+
,
..\.$:
,,
This note describes an easy-to-build, high performance,
,$
low cost 100W flyback power supply, able to work on
any mains supply from 85Vac to 265Vac, and from
40Hz to 100Hz. It is automatically synchronised to the
horizontal scanning frequency for minimum screen

Centers

on Budget

The signal-to-noise ratio of a communications system

can ba improved by increasing tha power of the transmitter, increasing the gain of the antenna, or improving
the sensitivity of the receiver. A Iow-noisa preampliiar
is an economical solution for receiver enhancement
and this note describes the design, construction and
performance of a 400-512MHz preamplifier using Motorolas dual-gate GaAs FET.

13

-.

Abstracts
AN932
Application
Encoder

of the MC1377

(continued)
ANI025
Reliability Considerations
in Design
and Use of RF Integrated Circuits

Co/our

RF integrated circuits located at strategic points in a

CATV system feature prominently in the overall
reliability assessment. Low noise and distoflion require
*,N
.*5,.. ,~.,,,
state-of -the-art transistor structures. Gold metallization.,;,.+$Y;YN+W
,Y..
thermal equilibrium and automated process con~b~~t$%:a
have resulted in transistor lifetimes of over 100 ~~%+$.~
An ovewiew of the physics of construction invq&~~~~$
the die and interconnects is discussed, tog,@$&~@h a
definition of major reliability terms and ati:~$~o~~ction
to hardware and software m icroci rcui&@~&i&y tools.
.&.<>,,
:~.l
ANI027
Reliability/Performanc6~#sp&ts
of
CA TV Amplifier Design
~k$>p

The MC1377 is and economical, high quality, RGB

encoderfor NTSC or PAL applications. It accepts RGB
and composite sync inputs, and delivers a 1V &p
composite NTSC or PAL video output into a 75Q load.
It can provide its own colour oscillator and burst gating,
or it can easily be driven from external sources. Performance viflually equal to high-cost studio equipment
is possible with common colour receiver components.
AN1OI9
Decoding Using the TDA3330, with
Emphasis on Cable In/Cable Out Operation
The TDA3330 is a Composite Video to RGB Colour
Decoder originally intended for PAL and NTSC colour
TV receivers and monitors so its data sheet concentrates on picture tube drive. This practical application
note supplements the data sheet by providing circuits
for video cable drive as used in video processing, frame
store and other specialized applications, and expands
on TDA3330 functional details. Includes PCB atiwork
and layout of an evaluation board.
AN102O
A High-Performance
Video Amplifier
Hjgh Resolution
CRT Applications

Discusses the reliability a~y$R~~&& offered by the RF

hybrid amplifier used in G~k!$,r~plications.
The active
part of the hybtid is +$@\$f~hsistor metallization,
ballasting and rug~~~ne~are
reliability-related factors
that must be co@~{@{~.d by the device engineer when
designing a higb~~~,ormance CATVtransistor. Veflical.

for

This note describes a state-of-the-art

video amplifier
making use of the superior performance characteristics
of Motorola CRT dtiver transistors.
In particular, it
shows the high speed obtainable with low DC power
consumption. The circuit is insensitive to load varia.}$~
tions and interconnect methods.
..lJ$.,
<$%

AN1021
A Hybrid Video Amplifier
Resolution CRT Applications

and horizont~l.hge~metry and device distoflion are performanc~~etawfactors

that must also be taken into
accou,pt. %$~ relationship between these factors is
exa~+?,ed, ahd Iifetest data is presented to illustrate the
~%~nti~es gained by careful device design.
,,1 x1:,.

,$

for High .~.$,a.>.~:

.>.;,.,..,,:.
.,$! ....

Many of the 1024x 1024

and 1280x 1024,~[}@f@~kHz
horizontal sweep rate CRTs used in ,~ti~~
and
high resolution graphics applications~W~&t
realized
their potential performance beca~o$tfie
speed of
their vidao amplifiers. The CR2@4 ~h&CR2425 video
amplifiers are hybrid circuitq,Q~x~.~@d for high resolution CRT applications. The~~a~re
less than 2.9ns rise
and fall time for a 40V q%~$~fing,
and provide a low
...

AN{~S@
35/50 Watt Broadband
(160-240MHz)
3v~:&sh-Pull
TV Amplifier Band Ill
*

The main design aim for this broadband ultra-linear

push-pull amplifier wastokeepthe
design as simple as
possible, in order to obtain the best performance from
the two TPV375transistors
and to minimisetha cost. A
further target was to obtain the maximum gain by
reducing input matching circuit losses. Includes circuit,
background description, Smith charts and PCB layout.
ANI029
TV Transposers
Po = 0.5 W[1.OW

Band IV and V

Describes the performance of a470-860MHz broadband

ultra linear amplifier designed for use in band IV and V
TV transposers.
The design is based on the TPV596,
and is intended to be as inexpensive and straightforward as possible: the load line is defined to provide the
correct match for peak power; VSWR at the collector is
less than 2:l; input matching isdesigned toprovldeflat
gain with decreasing frequency;
and the design is
optimized with a CAD program.

AN1022
Mechag,*a@d
Thermal Considerations
in Using RF,Lf~&k@~ybrid Amplifiers
,... .
Motorolaath&~fi)*
hybrid amplifiers are medium power
(0.2W t~<~~~>ower
output) broadband devices (1 to
10O~,w~Jh~t
are biased in a class A mode for linear
opek$tio$.
~o ensure a proper electrical and mechani~~
* ~,.,
,sa!,~nY&#~ce with adequate RF and thermal character~~~~~certain guidelines are presented here so that the
~$$~kign
engineer can obtain maximum electrical per{ formance and the longest operating life.

ANI030
1 W/2 W Broadband
Band IV and V

TV Amplifier

Describes the design and performancaof a470-860MHz

broadband linear amplifier for use in band IV and VTV
transposers, based on a TPV597 transistor. The design uses a reflection technique to achieve an insetiion
loss of 6dB per octave with OdB for the highest frequency. Two amplifiers are connected together with

14

.-

...

Abstracts

(continued)
a pair of TP9383 transistors in push-pull configuration;
TP9383 is a double-diffused silicon epitaxial transistor
using gold metallization and diffused ballast resistors
for long operating life and ruggedness.

3dB quadrature hybrids to create a balanced amplifier

avoiding the inconvenience of needing a good match of
reflected power.
AN1032
How Load VSWR Affects
Circuits

Non-Linaar
AN1039

If your amplifiers pass lab tests but fail QC testing, the

testing environment - not the product is most likely at
fault! Oftan the culprit is correlation of test systemsRF
Correlation occurs only when target error limits are
adhered to on acontinuous
basis among two or more
testing stations. Such correlation is essential for nonlinear RF and microwave power amplifiers, whose
circuits are extremely sensitive to the impedance of
their loads. It iseasyto
compensate for the inseflion
loss errors in an attenuator, but much more difficult to
compensate for load VSWR.
ANI033
Match lmpedancaa
Anrp/ifiera

in Microwave

The key to successful solid-state microwava poweramplifier design is impedance matching. In any highfrequency power-amplifier
design, improper impedance matching will degrade stability and reduce circuit
efficiency. At microwave frequencies, this consideration is even more ctitical, since thetransistors
bondwire inductance and base-to-collector capacitance be-~
come significant elements in inDut/outout
imoedan&l.>,.
,,
network design. includes table ofcharacterist;c
ie~+.
ante and velocity factor for various width/height ra?~
~*..
and various materials.
>+$,
ANI034
Three Balun Designs
Amplifiers

for Pw~_~
s~$llt
.~;::
.,, \$\,
Single RF power transistors s~~~~.,?atlsfy
todays
design ctiteria; several davice~.~wsfpbe
coupled to
obtain the required amplifiqPt&@Wpower.
The pushpull technique is often cho%~~~tause
it allows input
and output impedancq~?~.,b$connec ted
in series for
RFoparation.
Baiun#ra~?brmers
provide the kay to
push-pull design,,$~$~~develops
three balun-transformers, culmin~{ng ~ith a microstrip version, None of
the baluns ~~<$fi~
nor were the parasitic elements
compensateatwl,~ this way, their daviatton from their
theoret~~~~pe}brmance
could be evaluated more eas,,,.
ily. L. \\,;it
.:.:,$
~.,%;
AN~@# <Z.Solid Stste Power Amplifier,
.. ,*W
FM, 88-?08MHz

470-860 MHz Broadband Amp/if[&,~:~

. .\
This note describes an ultra linear bro~~~~~
(470860MHz) amplifier developed for TV t[a~,po,~r
applications. Tha amplifiar incorporates+wi~t~V593
transistors.
Each transistor is usa~ ~.b~~
a separate
broadband amplifier which a[&~*~ned
with 3dB
hybrids. Includes circuit, p%?~kt ad PCB layout.
,::.~ .*?$%J
..*
~JJa
AN104O
Mounting Co~id&ona
for Po wer
Semiconductors
S:<,:$
~~~,,
.>?!
The operating e&~r&%ent
is a vital factor in setting
current and pq~~~$at~ngs of a semiconductor device.
Reliability is.j~~~~%~~ considerably for relatively small
reductions i~.~~~tion temperature.
Faulty mounting
not on&&~~cre2*es the thermal gradient between the
device~~~~%$ heat sink, but can also cause mechanical
d#>~-XThis
comprehensive note shows correct and
~%~k{~t
methods of mounting all types of discrete
,.~~-,
..
#ckages,
and discusses methods of thermal system
\\4*
.~,, &luatiOn.
:.*.,.
,.~ti1041
Mounting Procedures
for Very High
i
Power RF Transistors
High power (200 -600W) RF semiconductors
such as
tha MRF153.., and MRF141G... series dissipate an
abnormally large amount of heat within a small physical
araa.
Heat sink material, surface finish, mounting
screws, washers and screw torque ara extremely im portantfactors in ensuring reliability, This note explains
why.
ANI044
The MC1378
Video Synchronizer

A Monolithic

Composite

Tha MCI 378 provides an interface between a remote

composita colourvideo source and local RGB. On-chip
circuitry can lock a local computer to the remote source,
switching between local and remote signals to generate composite video overlays. This detailed note describes local and remote operation, picture-in-picture
applications and the design of test fixturas to help
system development.
Printed circuit artwork for an
evaluation board is provided. The NTSC/PAL colour
encoder is similar to the MC1377, discussed in detail in
AN932,

: y% solid state power amplifier in a high efficiency FM

~,,
,.,.,$, ,,# transmitter can be made by operating a number of
,s :., ~ building block amplifiers in parallel. This note dascribes
\i::\\\:,r\z<.i<)\
such a building block amplifier with high output power,
*i..
i;s.l>\+.,.,
high gain, good collector efficiency and broadband (88I:e:,., .:,,
,:.,
108 MHz) frequency res~nsa.
The design is simple,
reproducible and reliable, and is suitable for several
architectures. The amplifier has been developed using

ANI047
Electrical Characteristics
of the CR2424
and CR2425 CRT Driver Hybrid Amplifiers
Desctibes the circuit and thermal characteristics of the
CR2424 and CR2425 CRTdriver hybrid amplifiers, and
discusses three different methods of protecting against
damage by a tube arc. Provides details of bandwidth
and rise and fall times.

15

AN1061

Reflecting

on Transmission

Abstracts

(continued)

Line Effects

ANI 122

ANI080
External-Sync
Power Supply with
Universal Input Voltage Range for Monitors
As the resolution of colour monitors increases, the
performance and features of their power supplies becomes more critical. EM1/RFl generated by switching
power supplies can adversely affect resolution if switching frequency is not synchronised to horizontal scanning frequency.
This 90W flyback switching supply
demonstrates the use of new high-performance
devices in a low-cost design, and includes a new universal
input voltage adapter.
AN1092
Driving
ECL System

Hjgh Capacitance

DRAMs

Running

the MC44802A

PLL Circuit

The MC44802Aprovides
the Phase Locked Loop (PLL)
portion of a tuning circuit intended for TV, FM radio and
set-top convetier applications up to 1.3GHz; a cOmplete tuning circuit is formed by adding a Voltage
Se
Controlled Oscillator (VCO) and mixer, The data shee~~:.$$$
recommends use of an MCU for sending the CO@% \$,~?.
bytes that setthetuningfreq
uency, This notede~r~a$
a serial (IIC) intedace with an MC68HC1 1E9,j#a~~er
design the information is sufficiently gen@r*b&~llow
almost any MCU to be used. Includes ~,$~c%~
pro...:,,$$+
:.:.}\.
:
gram listing.
.,,.
.\J\~-:$
AN1207
TheMC145170
in Basiq?bHP%tid VHF
.~:f
a.,.<?
Oscillators
,,i:\>>.,..:\
~,?.
Frequency synthesisers $st[~the
MC145170 use
digital dividers which ar~ w~$!ly
under MCU control,
Tuning in less than a ~iil~hnd
can be achieved, and
the device can gene}~>many
frequencies from a
single reference,$~$[ce; the overall frequency capability
ranges from ~~~~~~~to
160MHz, Typical applications
include the~:aT~eY oscillator
in transmitters,
local
oscillatosf?~tivers,
cellular phones, and multiple
synchro~~>d clocks in computers and other systems,
,\*+.~ ~,$$.

In recent years, microprocessors and digital logic have

seen substantial increases in line drive capability. The
fast rise and fall times of modern devices make an
understanding of transmission lines and their effects on
system reliability a necessity. Includes a procedure for
assessing possible transmission line problems in practical designs.

in an

In systems where speed and efficiency are of utmost

importance, designers often mixtechnologiesto
achieve
the right combination of speed, power, cost and processing capability. Motorolas Emitter Coupled Logic (ECL)
makes it possible to operate up to 1GHz clock rates.
However, ECL speeds are not necessay in memo~
that is not accessed evey clock cycle - a large CMOS
DRAM is cheaper and uses less power and boq@
space than ECL memo~. The MC1 OH/1 OOH660,~4~%,
ECL-TTL Load Reducing DRAM Driver was d@s@~d
;, ~+::>:<:$
as a translator for such applications.
.$:.s. :?+

ANl~.~~bThermal

Distortion

in Video Amplifiers

$~h$fmal distotiion is a problem in many high resolution

t}&~%&o amplifiers, It occurs when there are instantane~~$$Y:VkUS
power changes in the transistor stages, and if the
%+ problem remains uncompensated it leads to the visual
effect known as smearing. This note discusses what
smearing is, what causes thermal distortion, how to
measure it, and how to compensate for it,
AN1401
Using SPICE to Analyze the Effects
of Board Layout on System Skew when
Designing with the MC IO/100H640
Family
of Clock Drivers

AN1 106
Considerations
in Usjng the ~%~$1
and MH W851 Series RF Power MW$&
, *i..
The MHW801 and MHW851 seri~,o?~~er
modules
are designed for use in cellw
Mhble
radios. A
considerable amount of app,li:~Qo@ information is included in the data sheet; $~3t~s provides additional
information concernin~~~b~r.~l
electrical considerations, noise characte~~ic$. gain control, circuit considerations and mog~~ting:~}.:~
. .$.S*
AN1 107
Unde~~*g
RF Data Sheet
Paramet*q3.~~$
, .:.\:\, ~.
The data~+t
is often the only source of information
about$~~c~aracte
rist ics and capabil ity of a product,
Th~#is ~~cially
true of RF devices, which have many
,,.@4!@&#pecifications.
It is therefore important that the
x~afacturer
anddesignerspeak
a common language.
~+qm~~s paper reviews the significance of the quoted
$>~Ualues and highlights critical characteristics,
Descrip tions cover the procedures used to obtain impedance
and thermal data, the importance of test circuits, low
noise considerations and linearity requirements.

Illustrates the complex influences of board layout on

the total skew of a system when designing with the
MCI OH/100H64x family of clock drivers. Discusses
transmission line theo~ and the vafious termination
techniques, and presents guidelines to assist designers in analyzing board layouts and loading schemes
using SPICE simulations to predict and minimise the
total skew of a system,
AN1402
MC1OI1OOHOO Translator
SPICE Modelling
Kit

Family

1/0

The difficulties of designing high-speed, controlledimpedance PC boards-and

the expense of reworking
them makes it essential fordesignersto
model circuit
petiormance prior to committing to a layout. This note
provides sufficient information for basic SPICE analysis on the interconnect traces driving or being driven by
the H600, H601, H602, H603, H604, H605, H606
and H607 translator chips. It includes schematics of
the input, output and ESD structures, and package

16

Abstracts

(continued)
ICS, temperature and DC supply voltage to help the
designer to predict the amount of frequency pull in a
particular design.

models which may affect the waveforms.

A SPICE
parameter set for the referenced devices is provided.
AN1404
ECLinPS Circuit Performance
Standard VIH Levels

at NonEB77
A 60 Watt 225-400MHz
2N6439

,%
$>,1
.{:: ,,,..J.:\*.,
This bulletin describes a 60 watt, 28 vo~bt~~band
amplifier coveting the 225-400 MHz miQ\&~[~~munications band, The amplifier may be y@~*~N~~y as a 80
watt output stage in a 225-400 lws~ti~sm
itter; by
using two of these amplifiers co@!~e4#tth
quadrature
coupiers a 100 watt outpu]l~~~j$~r
stage may be
constructed. The circuit is ~,~gmbd to be driven from a
50 ohm source and wor$~ntd~%ominal
50 ohm load,
$?..:*,
~\t,
A 1 Watf, ,@WAmp[ifj@r
EB89

When ECLinPS devices are intedaced to other technologies there may be times when the input voltages do
not meet the specification detailed in the ECLinPS data
book.
This application note discusses the consequences of driving ECLinPS devices with an Input
Voltage HIGH level which is outside the specification.
AN1405

ECL Clock Djstributjon

Techniques

Clock skew - the time difference between supposedly

simultaneous clock transitions within a system is one
of the main factors limiting system performance at high
frequencies.
If clock skew can be reduced, designers
can increase performance without using faster logic or
more complex and more expensive architectures.
Emitter Coupled Logic (ECL) technologies
offer a
number of advantages over the CMOS and TTL alternatives; this note describes the advantages, the three
skew problem areas, and methods of clock distribution
to minimise skew.
EB27A
Get 300 Watts PEP Linear Across
30MHz from thjs Push-Pull Amplifier

This S-band ,,,,~:~~?~~~erfeatures

simplicity
and
repeatability h,d~i~~~fig 8dB minimum gain at 1 watt
output on a 2%~#Spply. It uses an MRF2001 transistor
in acom@~n b&3e, class Cconfiguration, and istunable
frorn,,,2~~%@.35 GHz. Applications include micowave
co~ti~lo~tions
and other systems requiring medium
Pw@6arr0wband
amplification. The Bulletin stresses
#@ importance Of physical construction as well as
,.. e%cttical design.
,.>~:.:
,

2 to
,f

Includes circuit, PCB artwork and layout for a 300~$$

push-pull linear amplifier based on two MRF422s~@$$:
signed to operate over the 2 to 30 MHz band.
}%
MC1723 voltage regulator is used as a bia~su~y.
,+?:,
\
The Common
EB29
Advantages

Emltfer

TO-39+&&ks
~~t>l
,.,,,s~p,~4\<.,
,.,:
\.

The common Emitter TO-39 pac~~,~j,krs

from conventional TO-39s or TO-5S in tW~.&~~@mitter not the
collector is connected to $~~~~al
case. With NPN
transistors this configuraQol&\~~Wws direct connection
of the canto RF and n$Saf~@.~C ground in many class
B and C circuits. The& ar~two impoflant advantages:
by connecting thSs*e~@
RF ground, emitter inductance is reduced bd ~ain increased by 3 to 5 dB over
that of com@~&$~ionventionally
wired transistors,
And the cas&~~wy be directly pressed, clipped, or
solder@<f~the Heat sink with no effect on RF petiorm~~~
,{>,.+
~ ,..
ante.~~ ~~1~.
.+.*.>,
..;\, ,~
!*
EB@;+~Predict

Frequency

Accuracy

Amplifier

$2b90
Low-Cost VHF Amplifier Has Broadband
Performance
,,.,
This bulletin presents two VHF amplifier designs intended for FMor CWservice in the 136-174 MHz band,
Both feature the Motorola MRF260 and MRF262 plastic
encased VHF transistors which are rated at 5.o W and
15 W power output respectively, The devices are packaged in a standard To-220 silicone epoxy case with the
emitter wired to the metal tab and centre lead of the
device, This common emitter configuration results in
good RF performance, improved thermal conductivity,
and ease of mounting in an RF amplifier by connecting
the transistor mounting flange to RF and DC ground.
60 Waft VHF Amp/ifier Uses
EB93
Splifting/Combining
Techniques
Proven combining techniques can be used to obtain
higher output power and added reliability at VHF.
Simple matching networks and power transistors with
moderate gain can produce performance comparable
to that of a single-stage amplifier with a larger, more
expensive device. Though not the ultimate answer, the
splitter/combiner method has distinct advantages over
designs that force transistors into a parallel configuration, This 60 Wamplifier operates from 150 to 175 MHz
and features two low-cost MRF264 transistors. The
design uses a modified Wilkinson combiner technique
to produce 60W output with a drive level of 15W.

for

:,):@G~2060 and MC12061 Crystal Oscillator

+
~jrcuifs
($
,,;.J:~?$.{,.,
.SCrystal oscillators are used to generate a precise and
,1:.
*$\::,<:+:\
:$,$ highly stable signal. Such circuits typically provide this
signal at a frequency close to the resonant frequency of
,Sa.w.,:ti,
i\:Fy,
a>.
their crystal. However, circuit components and other
~.
factors external to the crystal influence its natural
resonance to some degree, an effect often referred to
as pulling or warping. This bulletin discusses the
variation in crystal frequency as a function of different

17

Abstracts
EB107
Mounting Considerations
RF Power Modules

for Motorola

The packaging used for Motorola RF Power Modules

consists of a copper flange on which the ceramic
substrates are soldered, and a non-conductive cover
which is either a snap-on design or attached by epoxy,
The substrates are either 96/~Alumina, 95.57~Alumina,
or 99A Beryllium Oxide, and are attached to the copper
flange using lead-tin or iridium based soft solders. This
bulletin discusses the mechanical factors that should
be considered when mounting these modules in equipment.
EB41 1

A Digital

Video Prototyping

System

This bulletin describes a Digital Video Prototyping

System (DVPS) developed using Motorolas latest multimedia devices, togetherwith a PC-based Field Programmable Gate Array (FPGA) development system. It is
designed to provide a fast and effective means of
prototyping and demonstrating digital video processing
functions. A Reference Section lists datasheets and
user manuals containing detailed descriptions
and
information on the devices. The DVPS has been successfully used to implement two TV sub-systems,
namely a Picture-In-Picture Processor and a4:3to 16:9
Picture Processor, which are also described,

(continued)

Applicati

19

20

300W, 88-I08MHz Amplifier using the

TPI 940 MOSFETS Push-pull Transistor
By Georges Chambaudu
Motorola Semiconducteurs

Bordeaux SA

Option 2 (with C9s and without

C9p)

GA (dB)
18.3

65.4

19.1

68

19.6

66.6

.:3: ~J$opper
heat spreader must be mounted on, or laid on top of, a heat sink with thermal grease interface.
...

.,.;><,,3 ., *
,,

,t~

b. Increasing the value of feedback resistors R8 and R9. This will change the Gain-Frequency
Input VSWR, The value of Cl must be raised.
5. In addition to the normal cooling of the units, some airflow is recommended
boards.

21

slope and

over the top side of the amplifier

+V
INPUT

SCALE: 1.5

<

6.8 -8.2

KQ l/4W

Thermistor,
2KQ
R8, R9

10KQ

model

L2

Ferrite

L3, L4

Lead

FET

TP1 940

T1

9:1 impedance

AWG

ratio (input
D. semi
I = 11 mm

4:1 impedance

or other
Products

2.

300

W,

8-108

MHz amphfier
23

schematic

semi

I = 9 mm
must

of suitable
~pe

wi[b

(see

3)

Figure

transformer)

rigid co-ax,,
(see

Figure

be loaded

dimensions

ferrite

cores,

Corporation

and 9367021002

figure

transformer)

tigid co-ax.,

ratio (output
O.D.

transformer

toroids

0,2 I.D.

of R8 and R9, 0.6 total.

O 062.0

L = 19 mm,

Wire,

1.5 VH Total

lengths

O 09V

Technologies

10OQ

#l 6 enamelled

beads,

L = 28 mm,

(Tl

at 75C

or EMC

or equivalent

10 turns

25Q,

at 25 C/2.5KQ

PPR515-20-3

5310

L1

T2

on FET gf~)

1/2W

KDI ~rofilm

25Q,

(depending

and pafis

with

list

ferrite

and Vi of 35-40,

such

as Fair-Rite

E and I tvpes

respectively

with
3)

9467012002

Cenue

Hgh Imp@an@

AN448

FLOF Teletext using M6805 Microcontrollers

BV Peter Topping
M CU Applications
Motorola Ltd, East Kilbride

1. INTRODUCTION
,*$

...
\.3\

The T members of the MC68HC05 familv of MCUS provide a convenient and c@,~@#&tve method of adding
on-screen-displav (OSD) to TVs and VCRs. As well as the 6&character OSD ca~@$,~theV
include 8 Kb~es of
ROM (adequate forTeletext, frequencv-svnthesis, stereo and OSD), 320 bvt~~~:$~~,
a 1&bit timer and 8 pulse
width-modulated
D/A converters. The MC68HC05T7 also includes IIC haf~wd~ and, bv using a 56-pin package,
4 ports of 1/0 independent of the OSD, serial and D/A outputs. It is th~~.&itable for large full-feature chassis.
The MC68HC05T1 is in the middle of the price/performance
ra@*~@
includes most of the features of
the MC68HC05T7 but in a 4&pin package. This is achieved bv sh.@~%.~0 with the other Pin functions (Spl, OSD,
D/A). Even if all these features are used, there is sufficient l/O~most
applications.

.
.
,.~?w,

The MC68HC05T2 is a 16K upgrade of the MC68HCO:~k~n&$he

MC68HC05T3 a 24K version with increased
RAM (512 bvtes) and enhanced OSD (112 characters a~d,% rows of OSD buffer). The low cost MC68HC05T4
has 5 Kbytes of ROM and 96 bvtes of RAM mahw$$$?able
in simpler (eg mono, non-Teletext) applications.
The T4 and T7 also include a 14-bit D/A converter to~cilitate voltage svnthesis tuning. There are EPROM (and
OTP) versions of the T3 (including T1 and T~em&ation), T4 and T7.
,:$$t:,~,
This application note describes an exa~~~$a~#Teletext control software written for the MC68HC05T7 which
directlv controls Teletext chips of th~~p&$~43.
Spanish FLOF Teletext (level 1.5) is handled using packet W26,
If no CCT teletext chip is present o~~$~,~ bus (as indicated bv the lack of an acknowledge), all Teletext functions
are disabled in software. Abo~t~wes
of ROM ara used allowing the code to fit into the 7.9K b~es available
.,+~..!
,,
in an MC68HC05T7 along wjtti~fitng,
OSD and stereo functions.
.,\@,.~(t,?,i,,!
The software in the in.~~~$$hsting has been written for the MC68HC05T7 but could, with a little modification,
be implemented o,n,,~$h@&M6805microcontrollers. A microcontroller without IIC hardware can be used as long
as additional so~r~~~~included
to facilitate the IIC bus using l/O pins. An example of ilC master 1/0 driven
software can,~e fwd
in application note AN446.

2. FLOF TELETEXT

FEATURES

@tilllWel One Feature (FLOF) Teletext utilises ghost packets to provide features in addition to those available
,,$.~,~~~the original CCT Teletext. The prima~ enhancement is the provision of a menu with a choice of four linked
,@,,~t{~~ges selectable by the user with a single press of one of four coloured buttons on the remote control. The menu
$,:*::. Itself is sent in the ghost page using packet 24 while the linked page numbers are contained in packet 27. In
**
addition to linked pages, packets 26 and 30 are used. Packet 26 allows for the substitution of selected characters
in the displav bv special characters specific to a particular count~. This example application includes the Spanish
implementation of packet 26. The broadcast service data packet (8~0) is used to get the initial (index) page for
each channel and to display station identification information.

25

Ghost packets handled

The FLOF menu information contained in this page extension packet is transferred by the microcomputer to
row 24 of the display chapter. When links are disabled because there is no packet 27 (destination code O)or when
bit 4 of bvte 43 is O, row 24 is blank.
W26 :
Optional handling of modes 1XXXX,01111 and 00010 in accordance with the Spanish Teletext specificati~~~~~
the additional characters which are available in the 5243 CCT chip are handled. The feature can be di:,a~f,Q$~th
a hardware link on an l/O pin (see figure 1) so that the software can be used at level 1.0 in non-Spanl$h &o@ties
...~:!t}l)$..
.,,, ,$>
also using packet 26.
..,.,
.3,$$
.y. .,, $.!
W27 :

~,

$.1.$, p<,

This packet contains the linked page numbers for the red, green yellow, blue and ind~~~la~~) keys. Bit 4 on the
link control byte (bVte 43) is used to determine if these links are enabled (1) or disa~~~~}. When enabled, the
Spanish specification requiras that bits 1,2 and 3 be used to enable the green, yello~,m
blue links respectively.
This use of these bits is not defined in the World Teletext Specification. For t~~eas~n their use is selectable bv
a hardware link (see figure 1). If these bits are not used, all links (if enable~~~~~~> will be taken from packet 27
but will be automatically disabled if the broadcast links are default (FFSmF) ,#r invalid.
g,
$
8/30 :

.. .-,],
The broadcast service packet is used to supplv the index pa~~nu~ber on exit from standbv and (if teletext is not
stopped) after a channel change. BVtes 10-30 of this pac~~~~~~;~$plaved for 5 seconds on exit from standbv and
*. ,i.,
(if teletext is not stopped) after a channel change.
-\J
..>
.:.. ~,

~b,
,,*<,%p

The software listing is in two parts, The f~~~,~~~~ontains the idle loop and IIC routines from the main W control
part of the MC68HC05T7 applica~%.;~
Idle loop controls the timing of eve~thing
performed bv the
microprocessor, scans the local k~~qW~ checks whether or not an IR command has been received, etc. It also
monitors the relevant flags in t,~~xkxt
chip and performs the tasks (eg fetching linked pages) which have to
be performed independently \~,,.\::\.,\i
,Q*{~~uests
for the user.
t.$
.$,

,.-,,. .?-

The second and main w~~s

the Teletext module itself. It contains all the subroutines required to carry out
automatic and user$re~~~ted Teletext activity. Both modules use the same RAM allocation file (RAMT8.S05)
which is include~$$:~tie listing of the Teletext module. This listing also includes a svmbol Cross-reference table.
~.:,~r},l.?.$J
Figure 1 sh~~~%&plifiedc
ircuit diagram of the application. Most of the MC68HC05T7S 1/0 is used for purposes
*,.,,~>,..
other thaJT&~~&xt and is not shown in detail. Communication with the 5243 Teletext chip is via an IIC bus in which
*>,,.*
the T~+WalWys the master, The function of the three l/O pins used for Teletext is described under Ghost packets
hati~:i~nd
Inputs and Outputs,
..!,
! .,%
.,,t~f
.- .,,,,
+:Al~#Sion of this Teletext SOftware has been implemented on an MC68HC05C4 for use in a N where the other
~ontrol functions were handled by a separate microcontroller. The signal from the IR pre-amp was fed into the
C4 which used Teletext commands to control a 5243 via a software IIC bus. Non-Teletext commands were re
generated bv the C4 and sent to the other microcontroller. This arrangement allows Teletext to be added to a
chassis which was originally designed without considering Teletext.

26

.,,, ~;y.
In the exampl~+%on
the idle loop code is in the main W control software module rather than in the teletext
module, Listi~~.*!>:,..
l$hows the relevant parts of this module. The loop time is 12.8mS and it is at this rate that the
timing c.@&ters used by Teletext (CNT1 and CNT4) are incremented. The standby condition is checked first; if
the W &Y@$h standby then there is no IIC activi~ and hence no reading from, or writing to, the 520. If the W
h~si~~~~xited from standby, as indcated by the flag 3, STAT2, then Teletext is initialised using the sub-routine
~~~@T. This subroutine writes to the 5243s control and mode registers (R5, R6 and R7) and checks that the
~,,;:
,,,lt~acknowledge is present. If therewas no acknowledge, as indicated by flag 6, STAT7, then no further Teletext
,s-,~;,. ,&tivity is attempted.
,;.
~.i;t,
~!
ic..$
\:::c,,)*\\,
;J:
If an acknowledge is present, Teletext polling goes ahead, although it is suspended if there is a mute or time
~:,,. -.
display. A mute indicates that the channel has just been changed, or no channel is tuned. During time display,
$.
all other Teletext activity is suspended. R&initialisation using subroutine START2 is performed if flag 7, STAT5
is set by a change of the tuned frequency.

27

Counter CNT4 is used to delay the transfer of packets 24 (page extension- FLOF menu), 27 (links), 26 (enhanced
display characters) and the control bits from row 25 (display page) after the initial arrival of a page. When row 24
is read the 5243 FOUND flag is set to indicate that the arrival has been acted upon. If UPDATE is on then an update
indicator appears if the update control bit (C9) is set or if the subpage has changed or if it is the first arrival of the
page. The update display is performed by the subroutine ARRVD which clears the transient flags and enables
the required display, i.e. page no. in normal mode and the whole of row O in subpage mode. Any boxed
information (eg subtitles or newsflash) in the current page is also displayed, The last Teletext function performed
by the idle loop is the checking of the FOUND flag in the 52&. This is accessed via the IIC bus; it is on the last
~:t.,,,,
.*{:<{
:<+
(not displayed) row of the display page along with the current page and suppage numbers and the control bits. ,~..
~~~~
,:t~.
,.}i,,
~,?
,. ...
If there is a current Teletext transient (time, row O box or packet 8/30), the transient control branch from thee~~&,,~$
loop is executed. This routine checks to see if it is time to end the transient. If it is, the subroutine O,~\&~*
executed. It resets transients for both the OSD generated by the MC68HC05T7 and Teletext. The $f~+:a~>ne
RSTMD2 performs this function for Teletext. It is called from within the subroutine OSDLE (no~~~~~
,.,;!
.. $+);.
\. *t., .>
:cy~)
~~,:.,
,
5. REMOTE CONTROL FUNCTIONS
,,,.tt<,:
:?,,.
.~

:. ~;,.>).
...,+,,
.4,*,
.,
$/),,
,:.,
?:>

k!.....{?
.:,l:.+,.
la,>,
. ,,,,

W~T

0-9

?*.~~

*J$:$,:*:,t

Linked page access keys. The decoder stores four pages of t%t. These are the display page and the three pages
corresponding to the red, green and yellow links. Th~blue8nked page is not acquired in advance. In the absence
of FLOF data or if the links are disabled by the coqf?o$.~t in packet 27, the red key is Page+ 1 and the 9reen key
page-1. Under these circumstances the requex~~age
and the next three pages are acquired.
~?~J>3
,.- ,:. \,
Pc+/

,i><,.yi+t:,,

,1,,

~q,,

~1<,
. ..

These keys always select page+ l/p~$&k,,mgardless

and yellow keys, the page is dis#W~&mediately

of the availability of FLOF information. As with the red, green

if it is already in RAM.

INDEX
/i,..>.
,y,:,,.~:?t!..,.!s~.

This key operates as ~n ~,~onal

link with the difference that if the link is invalid the initial page from packet 8/

;?:i.,*{f>i~
,$

.,. *.

Text m@: $~ter subpage mode, (max. 3979). TV mode: Display time in topright-hand corner for 5 seconds.
Pres$~~&$Q$ key during a station identification display (packet 8#0 bytes 1030) can be used to extend this display
be~~~.~be five seconds it appears for, after a channel change.
,!>+
,,,:,,,,
~\~,,
?kc$
Halt acquisition, STOP is displayed instead of page number, Press again to restart, If acquisition has been
stopped by partially entering a new page number then this key can be used to return to the original page.

28

-___.

MIWNOMIX
Toggle between Teletext and mixed display. Use of this key causes the display of the top status row for 5 seconds
if it is not being displayed be~use the current page is a newsfiash or a subtitle, 5243 contrast reduction is enabled
in mixed mode.
FULuOP/BOT

UPDATE
Return to picture until a new version of the requested page arrives. When it ar~~{;~~$~ page no. is displayed in
the topright-hand corner, the key operates in both W and Teletext mode, si~+~~~Qt into N mode. Any boxed
information (alarm clock, newsflash or subtitle) will be displayed. In suw~
@ode the complete header is
.,,, >.$
displayed so that both page& subpage numbers can be seen. Cancel up-;Ytering
Teletext mode and then
.,!;.,4 .
,>....,
going back to W mode by pressing the Wflext key twice.
..... .... *,s.

,,

.$,,

!:!:e,,.:

The Teletext module (listing 2) comprises variou~t~~~~titines

which are used both by the idle loop and to perform
any Teletext actions initiated by commands from ~ lb remote control. They are described in the order in which
they appear in the listing.
..
*~:
.$..%:$.
,+~

~X
is executed when the WflEXT.Q$~o$~S pressed. Its function is to toggle between W mode and Teletext
mode. This flag routes the microprocessor to execute either TnOFF
mode. The flag O,STAT indicates tb,~c~~~~fit
or THON according to the curf&$R~.We. TXTON checks that Teletext hardware is present and does nothing if
there has been no IIC acknoW.~@~f,
however, a 5243 is present in the N, it clears all transients (OSDLE) and
sets up the Teletext mode. $&~R&;alisesthe control registers (R5 and R6) to display text and background both in
and out of boxes. For ,~~~f~hes
the set-up is text and background within boxes and picture outside, ~OFF
also resets transie~{$$$gt~?orces N mode and sync, Polling and updating continue as a background activi~,
.<<)-
$$t.p
~.\
,.,, .,
When the UP~~~&dY
is pressed the update flag 4, STAT2 is set and TXTOFF executed so the N is forced to
N mode. .,,,
If t~~$~s a current transient hold (eg time), the hold is cleared before TXTOFF is executed,
.+;

<>
J~.,
.t:$

The nu~~.ehtry
subroutine DIGITO branches to DIGITS in subpage mode but otherwise accepts any number
key~$~~ge
number input. Three digits are required, the pointer PDP holding the current position (O, 1 or 2 for
bu~;~s,
tens or units). During entry the flag 2,STATis set to stop Teletext activiw. The numbers have to be
;,!*%,y
:it
~-w~~ten to the topleft-hand corner of the display page as well as saved in RAM. Once all three digits have been
the page is requested and page acquisition
..:.S$,t~?~~$tered
..

restarted.

.%,,$:*>*
a,.

Ny..*l,\%.$ -

& ... ...

).},
,,:.

The codeat label GETIT makes this request after first checking whether or not the selected page has alreadv been
requested (it could be the current display page or an alreadv requested linked page), If it has, then a switch is made
to the chapter associated with the appropriate acquisition circuit and no new request is generated. If not, the new
request is made and the FOUND flag set.

29

6b. Subroutines:

Colours, INDEX, NPAGE and PPAGE

The four colour keys (Red, Green, Yellow and Blue) are primarily intended for selecting Teletext linked pages.
When pressed the chapter which corresponds to theappropriateacquisition
circuit is selected for display. If links
are dsabled (by the link control bit or because there is no packet 27), then the RED and GREEN keys select current
page +1 and-1 respectively. This choice is taken according to the state of flag 3, STAT3 which reflects the condition
of the link control bit in packet 27. The code executed by RED, if links are not in use, is the same as that executed
by the + function (NPAGE) which always selects the next page. Similarly the alternative GREEN function
(PPAGE) is the same as for the - key. The YELLOW and BLUE keys do nothing under these circumstances. ln,?t;~~
Spanish Teletext the GREEN, YELLOW and BLUE links can be individually inhibited, but the RED link is op~$~;.~
l*T>
,*.?,,
<..
inhibited if all links are off.
...+,.~~
,..:,>
{*,
ti,i,,\;-&.J-f,>
The chapter associated with the selected page is displayed immediately if it has already been request~~,~.hh$dill
normally be the case if a linked page (red, green or yellow) has been selected. The code at label LPK~_uted
if the page has already been requested. If not, a jump to CLRPD is performed. CLRPD is a label ~WQ~@DIGITO;
the code at CLRPD requests a new page just as if the page number had been entered man~{~$ If fhe required
acquisition circuit is the one already current, then the unstop code is executed. This ~w$l~e
9reen Page
bein~ooke~for
header to roll as though the page number had just been entered. This _sJh%t
something can
be seen to happen in the case where the linked page differs only from the current w~~~~~
subpage number.
Linked subpagesare not fully supported in this implementation astheyare rarely q~edh~5roadcasters and would
significantly increase the size of the software. When the chapter is changed the~ik~xt
PBLF (Page being looked
for) flag is checked. If it is low the FOUND flag is cleared. This forces the ~f~~~
of the links associated with
the new dsplay page. If the page is not already in, this will automatical~h~~.@
when it arrives so the FOUND
~}..,.)~
flag does not need to be cleared.
~*, ~,=,,,,

,,.><,

The BLUE (or cyan) key is different in that its page will not no~m~~~be immediately available (the four pages:
display, red, green and yellow occupy the four acquisition ci~.$q,~d
RAM chapters).
,.\.,..,.::.,
!.?y,.y
The INDEX (or black link) function is similar to BLUE except th~ if its link is not valid it defaults to the initial (index)
page number supplied by packet 8~0 (see subroubne GT~).
,.J~-~,.,.
!*.

6c. Subroutines:

LINK, GLPI, GL,F~$$R~H, CHCKI and NOTOKX

*::,~.:..~i,

The subroutine LINK allocates the three l~~~ages

(RED, YELLOW and GREEN) to the three free acquisition
circuits (not in use by the display pagq)~~$~this
it checks the page numbers in turn to see if they have already
been requested. If so they are lefij,~$h~~urrent
acquisition circuit. If they have not already been requested the
page number is put into a LIFO. @hly&~ are regarded as acceptable digits for page numbers; this is consistent
with the Spanish specificatiewugh
the additional HEX numbers (A-F) may be used experimentally or by
Teletext page generators:+~~i,~~~his first loop the subroutine GLP1 is used to get the linked page numbers from
packet 27, perform a d@&,o$@of the Hamming encoded data and calculate the new magazine number (page
hundreds) if differe~~~ro~%fiat of the display page. GLP1 uses subroutine SRCH to check if the page has already
been requested. l$!~r,~are no links, or if links are disabled, then displayed page +1, +2 and +3 are requested.

+*...

!,. .
**> -\:.i\..,.t~\\

The second ~~i~~p LINK allocates new page numbers to the remaining unused acquisition circuits. It uses GLP2
to clear t~~$$~a~ant chapters in the Teletext memory and make the new requests. Subroutine CHCK1 is used
to chec~-w@%er or not an acquisition circuit is in use before it is loaded with a new page number from the LIFO.
,k:,~i]?:i,:i~
~~~s@~~hod of organizing new page requests prevents unnecessary requests being made for pages already
r$@$ted.
This is particularly important when links are disabled and pages are being requested using the +
o*- functions, Under these circumstances when the page number is incremented (or decremented) only one
new page has to be requested (new display page+ 3), while page, page+ 1 and page+2 do not need to change and
can be left in their current acquisition circuits.

NOTOK3 and NOTOK2 handle the RED and GREEN functions when links are disabled. They are disabled if the
Iinkcontrol bit (packet 27 bit3, byte 43) is zeroorif there is nopacket27, These subroutines respectively increment
and decrement the current page number (units and tens). The current magazine number (page hundreds) is
not affected.

6d. Subroutines:

ROW24, W2B, R2B, GCYI, CUNK and DECODE

ROW24 is used to transfer ghost row 20 (packet 24) into the display chapter. This has to be done via the &~Cbus.
The loop reads two bytes via the IIC (subroutine R2B) bus from the ghost page and writes it to the di:~~:~ge
(subroutine W2B). The FOUND flag is then set to indicate that the arrival of the page has been regow$d
and
acted upon. This subroutine is only called by the idle loop and is used along with the other su,~m~s
which
., .,,+>:,
1:;,
,t*...
get information from the ghost page (CLINK, LINK and GET25).
~.
~t.i,,.2.~~
R2B and W2B use IIC routines RWD and SEND which are outwith the Teletext modul~:<~wsubroutines
will
differ according to the microprocessor in use. An MC68HC05C8 implementation would$~~
to use 1/0 lines (see
reference for suitable software) while the MC68HC05T7 can use its IIC hardw~~:~~.Th& routines used in this
example are included in the listing extract from the W control software mod~h,:$~~~f~g 1).
.,.. ,,,.
:*,.-: ..
The subroutine GCYI is used by LINK to store the data associated with th&~$M
an INDEX links. As explained
above, these pages will not be acquired in advance, the page number o~$$bng sent to an acquisition circuit if
..,.
.
,.~~
?
requested by an IR command.
,,l,h
,~&.~j\.r,

.,
~,..~:
$,,.+

CLINK fetches the link control byte from packet 27 if the destina~fon ~de is OK and, after decoding the Hamming
~:,?, -,,..,.,
,.*.
encoded data, transfers the bits to STAT3.
\\;~.~.
t:!:.

The Hamming decode subroutine DECODE corrects fa~~~gle bit errors. This is done with in-line code using the
table HAM (at the end of listing 2) as this uses lesq~w
than an algorithmic method.
!:\\,\
. ....!>
..\,<,d.$

6e. Subroutines:

MIX, TRANx, Tfi~8WOLD, and NOHOLD

The mixed display capability of the Telet,~~ ch/p(5243) is toggled using an IR key which calls the subroutine MIX,
When mixed mode is entered, interlao~d ~~adcast sync. (31 2/313) is selected because the non< nterlaced sync,
used for teletext is not suitable if a ~@e
is present on the screen, This is set up via the 5243 mode register R1.
The control registers R5 and R,~~$e@dated to provide the mixed display,
<2C*
, .t..d~

When returning to a non-m~$~~%$~lay, the code at NOMIX is used to rwonfigure

the control registers and to set
up a Teletext only312/39Q n~:tnterlaced sync. This sync. reduces adjacent line flicker in a pure Teletext display.
~.,~l);.s~!.,t$,;~
The subroutine T~~Q
#ts up a transient which retains a black background on the top row so that the page
~$c
~,,,, ,$~..
number, time et~,cafi~e seen clearly, This type of transient is also started if the page number or subpage number
~.,, !::,,,,...$,
is being ent~&J$Nmlxed
mode, Sub-routines TRAN 1, TRAN2 and TRAN3 are used to initialise the various
transientJ~spl~~These
displays are cancel led as discussed above by actions taken within the idle loop controlled
by the f@,@k4nning timer within the MC68HC05T7.
~..*?,.,**
J, . Si

Th&;.w$
~{e~~rs
, ~,. ;,

subroutines are used in conjunction

within the 5243.

with the IIC SEND routine to write to various subsets

of the

,,, ~~,~$ the Teletext STOP function is requested by an IR command the routine HOLD is executed. This is a toggled
.&~i~:+{: /unction when requested in this way, HOLD displays the word STOP in place of the page number and stops
~~,+,$.t
:+*
the display acquisition circuit by clearing the 5243 HOLD flag accessed via its page request register R3,
.+:,**
.::<$

~$.

NOHOLD is executed to restart the display acquisition circuit. It

corner. If a new page number has been partially entered, a press
return to the most recent page request. This takes only a single
number cause a HOLD. The completion of a page number entry

31

returns the page number to the topleft-hand

of STOP (executing an UNHOLD) will allow a
press as the start of the entry of a new page
(3 digits) causes a NOHOLD.

6f. Subroutines:

REVEAL, EXPTB and TIME

The REVEAL function muses anv hidden displav information to appear, It is controlled bya bit in the displav mode
register (R7), The software example leaves anv revealed information permanently displayed, If, however, it is
required that such information disappear when the page is updated (this mav be better for a quiz page), then the
two commented

out lines (80 and 81) in the idle loop should be enabled,

The display expand facility is controlled bv another two bits in R7, The EXPTB sub-routine cvcles through normal,
to~half double height and bottom-half double height.

st.,
\
.,.,..
. )\i.
, ::;~~.,?,i.
\\.$.
,> *, .{*.

The example application uses a single IB key (subroutine TIME) for both the displav of the Teletext clock and a$+j>
entry into subpage mode. IF the set is in TV mode then the time is displayed for 5 seconds. If the ~ is i@~%&i~
mode then subpage mode is selected. Subpage number entry is described in the following section. &&ke
Teletext clock is requested it appears (boxed) at the to~right-hand corner. It is removed by the idle,lo~~,~~nds
after the last press of the time button. When the time is being displaved all other Teletext act:l~~j~opped
using UCHOLD.
~ .!,..
v:.i~?,
..~.,

6g. Subroutines:

DIGITS, SUBPG, GET25 and GET26

.\&

*:*:+l;,;:>$,,.
DIGITS is the subpage version of DIGITO and uses similar code. More checks on,the ~.~ut data are required as
the four digits of the subpage number have different maximum values. These $~$wums
are 3 for thousands,
7 for the tens and 9 for the hundreds and units. These values reflect the sub@&~<~&mbers original useasa time
(24hr format). For tens and thousands a keved 8 becomes a O and a 9 b,@8~$$a 1; for thousands onlv 4, 5, 6
:jjek,
~
and 7 become O, 1, 2 and 3 respectively.
.>:., ~:,).

,.:,,,:/.

The code at the label SETIT is the subpage

and sets the FOUND flag.

equivalent of GETIT, -~kd

above. It requests the new su~page
ii: ,,
,S>
~<>
.
,,.\&l%,
:.]1,*
.:,~.
The subroutine SUBPG IS called when the TIME (or clockp%~j is pressed (TV in Teletext mode). It toggles
between normal mode and subpage mode. When s@p@e mode is entered the page number displav (P) is
replaced with *** to indicate the mode change and:?b:~rompt for the entry of a subpage number. Once all four
+,*Y
digits have been entered the new sub-page is re,%S~dby SETIT. The code at the label RSTR is used to exit from
this mode back to the normal (page number) ~:,~>$estoring the page number displav to the top-left-hand corner.
!j:i.>~ $*L...,

GET25 is used bv the idle loop to get $N~~~~&rmatlon stored in row 25 of the displav chapter. This row is not
d(splaved but contains various info~~~~h
used bv the control microprocessor. The current page number,
magazine number, sub-page nu@r~~#etext
control bits and the FOUND and PBLF flags are available. GET25
gets the required information,~~,$~~des
it in the RAM of the MC68HC05T7.

,,1,

,.,.

At the end of this subro@&~fi@J~/O line 7,portB is checked. If it is low, packet 26 is handled. If it is high, this
packet is disabled. Thi$w~@be
required if this application were to be used in a country other than Spain which
used packet 26. It ~.~~?ti
require to be switched off as the enhanced displav feature uses different characters
,* ~tf.,
In countries which do not use packet 26 (eg the UK) it does not matter whether or not
depending on th~,f?~%h~fi.
packet 26 is e~a~e~:
.

,?j$,,\ ~:\?

If packet:~& iS&abled, GET26 processes all packet 26 data present in the ghost page. The tables G2TAB, G3TAB
and ~3A~,:q$%tain the characters used to replace the character at the displav location defined bv each packet.
>,
>.:.,.,
\..,)
,,:,,t,.

32

6h. Subroutines:

GIP, W4T and SW4T

The subroutine GIP gets the initial (index) page from packet 8B0, It will be doing this as the set is brought
out of standby or just after a channel change. It mav thus initiallv get a poor signal (or there mav be no
Teletexti so it tries repeatedly until it finds a valid packet 8B0 format 1. If this is not found after 96 tries it
gives up and sets the flag 6,STAT2 to indicate that there is no packet 8B0 (or no Teletext). In this
circumstance it defaults to an index page number of 100.
st.,\

R24T transfers bytes 1&30 of the broadcasting service data packet (8/30) into the display chapter. It is c~~,$hce
a second for five seconds after power-on ora channel change. The data is transferred to row Oof t~,~w~
page
which can be displaved either at the bottom or, as in this example, the top of the screen. Thi:~:t$a$$~ti @sPlav
is setup using the subroutine SR24T if Teletext is present. If the flag 6, STAT2 has been set,,QW~~~!~*described
above then SR24T does nothing. The transient displav is terminated bv code executed aE;XM,qPProPriate time
b>%>.,t..:
.
..\,\;:,,
,,~ ~,h
from within the idle loop,
,,.. ,/
&\.4
,,
,.,. ~.$,~

7. INPUT AND OUTPUTS

%,~i($;.
~..*;:::,.
,\t,)
,.
.~..

.~\>,*,,:.:)~

Apart from the IIC bus, onlv three pins on the controlling microprocess~~&~~levant
to Teletext. Two inputs
select the usage of packets 26 and 27 and one output can be used to+~~nt?a.any hardware which requires to be
changed according to whether or not there is a W picture currently
%~~displaved.
In manv applications some
,%,,,
..*,
or all of these functions will not be required and could be elimin~ted~@m the software thus freeing up the pins
~\,,
>*,* *?..,..\
for other uses.
,..*.,,.;> ++

33

LISTING 1
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
S6
S7
58
59
60
61
62
63
64
65
66
61
68
69
70
71
72
73
74
7S
16
77
78
7s
80

........................................................
.
Idle
100P.
,
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

00000000
00000003
00000005
00000001
00000009
Oooooooc
Ooooooof
00000011
00000013
00000016
00000018
0000001.
0000001.
Oooooolf

00000022
00000025
00000028
0000002b
0000002.
00000031
00000034
00000037
0000003.
0000003.
0000003f
00000042
00000044
00000046
00000048
0000004b
0000004,
000000S1
00000054
00000057
0000005.
0000005.
0000005,
90000060
90000062
90000064
00000066
00000068
0000006b
0000006.

0d13Cd

>3.00
>3.00
>3.00
>cdOOOO
030104
>1600
Zosr
>010009
>1700
>,500
>Ifoo
>cd OOOO
>cd OOOO
>02004d
>02004.
>0.0041
>040044

BRCLR
lNC
1 NC
lNC

JSR

FON

BRCLR
BSET
BM
BRCU
BC LR
BC ~

JSR
ALRON

JSR
BRSET
BRSET
BRSET
BRSET

BRSET

>06003,
>0c003b
>Ofooos

BRSET
BRSET

BRCLR

>lfoo

>cd OOOO
>01001.
>b60D
.130
252.
>Cdoooo
>cd OOOO
>cd OOOO
>cd OOOO
>09000s
>0b0002
ad6e
>1100
>D600
>b?oo
.608
>b700
.6:9
>cd OOOO
>080104
>1000

DNTRS

BW
JSR
JSR

JSR

NOUP

82
83
84
8S
86
*,
88
89
90
91
$,

00000010
00000072
00000375
9000027S

>3f00
>04008b
>060088
>090085

C000C27b

>b600

94
95
96
97
98
99
:$0
>9:
:C2
:33
:54

0000007d
0000007f
OCOOC 081
3COO0 084
00000086
00009C88
ocoooc8a
30000084
$oo~s$ef
0000099>
00009C93

.150
2403
>..0000
>b600
.104
2603
>cd OOOO
>3 foo
>3,00
2?03
>..0000

BCLR
JSR
BRCLR
LDA

cm

N024

6,1s,, .
CNT1
CNT4
CNT5
KBD
1,PORTB, FOX
3,STAT2

OUT,,

JSR
8RCLR
BRCLR
BSR
BC LR
WA
STA
WA
S TA
LDA
JSR
BRSET
BSET

BCLR

E=.

JSR
CLR

3, STAT2, AUO.
3.STAT2
2;ST*T2
7,STATS
RESTRT
VCRPOLL
1,STAT2, F1
1,STAT4, F1
6,STAT1, F1
2,sTAT2, F1
5,STAT, F1
3, STAT4, F1
6,STAT4, F1
7,STAT5, DNTRS
7,STAT5
sTmT2
0.STAT2, N024
CNT4
*48
,1
CLINK
LXNK
R0W24
GET25
4,STAT2, NOUP
5,STAT2, NOUP
MRW
o,STAT2
Ac c
R8

R2B
4,1OBUF+l, F1
0,STAT2
5,R7
TxT2
CNT4
2,STAT2,1LP
3,STAT2,1LP
4,STAT. lLP

BRSET
BRSET
BRCLR
........
...........?
. ... *J.

F1

. . . . . . . .

.
.

CWME

TE~TEXT
TWSIENT
POW 24 DEMY
MUTE TMSIENT
KEY~=D
,1~,.

F1

BC ~

>0.0041

81

;:

I LP

......................

.
.

tra., ie,t, .

Clear mode bits (.h...el mode, 2-digit

Pr.q. .. . ,.t~y ... .J

.
.
.

End

Teletext

........................................................
RST~

BRCLR
JSR

0, STAT5, S0S2
RES

YES ,

Pr.
RESTORE

SOS2

BCm

2,STAT4

ME

SURE ITS PROGM

RsTm2
RST~3

B...
BCLR
BRCLR
BCLR
LDA
STA

4.STAT4
4; STAT
5,STAT, TXTR1
5,STAT
0S03
,S

RESET 0s. ,-s1,.,

RESET ~lN TMSIENT
TIK HOLD ?
YES, CLE~
IT

TXT2
2,STAT, TXTR1
NOTTH
0,R7
0, STAT, TXTR2
AC C

STOP T1~ EXIT FMSH

OTHER HOLD ?
NO, S0 CLE= HOLD
~X OFF RW 0
TELETEXT ?

S TA
JSR
BRSET

JSR

TXTR1

BCm

TXTR2

BRSET
,.,
STA
C LR
Jw

2-DIGIT

NO. ENTRY
DISP

R4

R7
TXT2

NO, ALL ~XES

YE,

34

OFF

?
MODE

FUG
F-G

139
140
141
142
,,3
144
145
146
147
148
149
150
151
152
153
154
1s5
156
1s7
158
159
160
161
162
163
164
165
166
161
168
169
170
171
172 Ooooooff
113 00000102
174 00000104
175
j;~ 00000105

......... ..................... ..........................

.
.
.
UWat.d
Page has arrived.
.
....... ............ ........... ..........................
>b600

ACC

K
R4
4,STAT
5,STAT
::
CLM
JSR mxoo.
6,STAT, SPm
BRSET

>b, oo
>1900
>IBOO
>:;0000
>0.000s
,606
>cd OOOO
.646
>b, oo
>b, oo
.603
>050002
.602
>b, oo
>..0000
.610

L.&

JSR

sPm

NNr

RESTRT

LDA
, TA
s T*
LDA
BRCU
LDA
s TA
Jm

MODE

SB-PAGE

No,

smLL w,

%XOOF
+$46
R5
R6
+s03
2,C3, NNF
+$02
R7
TXT2

.606
,b700
>bloo
>3f00
>cd OOOO

LDA
ST.
LDA
S TA
STA
C LR
JSR

013.03
>1,00
81

B. CL.
BSET
RTS

0, MR, ACKOK
6, STAT7

LDA
STA

,Roc
.1s,

BCLR

0,STAT5

>b7 00

TWSIENTS

KILL

>cc OOOO

178

119
180
181
182
183
104
165
186
187
188
189
190
19:
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
201
208
209
210
211
212
213
?,.

00000108
0000010.
0000010.
0000010,

>b600
>b700
>1100
81

Ooooolor
00000111

00000113
00000115
00000117

RES

-S

RTS

SEND

BSR

STX

>bfoo
>1100
>b600

BCLR

LDA

ad25

00000119 >b600
.180
OOOOO1lb
2606
Ooooolld
Ooooollf >b600
adlb
00000121
00000123 >3.00
00000125
00000127
00000128
Ooooolza
0000012.
0000012,

>beoo

00000130
00000132
00000133

1b3b

ti14

>3.00
>3,00
26.9

BSR
~BU

LDA

cm
BNE
LDA

DPNT
0,x

DATA

BUFFER

SHIFT

SS.3

DATA

~BU

DONE

5,MCR

S TO?

PO XNTER

DPNT

WI

9,
81

MSR
FDR

llC

SET-UP

90 KHz
EN=LE
1:C AS mSTER
T-SMI
TTER , sTmT

+SBO
MCR
.mR

7,SR, .
SEND
::AU

AIT

35

Ioms

(EEFRO

mITE)

231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
2S3
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
211
272
273
214
275
276
277
278
279

........................................................
.
.
i
llC
read.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

0000014b
0000014d
00000141
00000151
00000153
00000155
000001S7

adO.
>b600
>bl O1
>b600
.1,1
2602
>3cO0

Rm

00000159
0000015b
0000015d
0000015C
00000161
00000163
00000165

.dd9
,1100
,b60U
.ddd
>b600
add9
1b3b

REM

00000167
00000169
0000016b
00000166
OOO0016f
00000171
00000173

1a3b
>1000
>b600
adcf
193b
163b
b63d

BSR

LDA
:%
rNc
1

IICSU

O,WDR
m. R

.s,,
BSET
LDA
BSR
BC ~

BSET
LDA
BRC LR
BSET

K
mcLR

0f3.
fd
1b3b
b63d
>b700
9.
e:

BC L.

. . . . . . . .

00000180
00000182

BSR

BCLR
LDA
BSR
LDA
BSR
BCLR

.
.
.
,
00000175
300001?8
0000017,
0000017.
0000013,
Ooooollf

REml
10BUF
10BF+
-DR
*$*1
,-1
SUWR

s?

GET FIRST
1

~=

BYTE

UP

m?
YES,

=XT

SUB-~

S1,,

SUWR
SHIFT
5, MCR
5,MCR

O,WOR
mo,
SHIFT
4,MCR
3, MCR
m.
7,.s,,
.
3, M,.
m,
10BUF + 1
7, HSR,
5, UCR
NR
10BUF

E
CL1
RTS
.........

,-.. ... ,.

AST
GET
MD

3f3c
80

36

..__

IT

BYTE
BYTE

SO

SAE

IT

STOP

DRESS

LISTING2
............................................................................
.
.
.
.
.
.
.
U**6 With ~T8.
s05, OST. S05 TMT7. S05
mlT*lete.t/aDlstereo

program

,MC68HC05T7)

CCT

control

module

,sP,

.
.

in)

,,,
develowd
by Motorola
Ltd
for d-st
ratio P, Po,,,
. This software
: NO liability can ~ aceePte6 for it, ,. in any specific application.
- all
rights
rese,
ved
Original software coPyriqht Motorola

.
.

PORTA
PORTB
P~TC
P~TD
DDM
DDRB
DDRc
DDRO

.
27
27
:;

Teletext

ZQU
EQU

EQU
EOU
EQU
EQU
EQU
EQU

00000006
00000009
0000000,

Tim,, control reai, ter

$12
$13
$14
$15
$16
$17
S18
$19
$lC

18 OSD data
registers
register
color & status
color
3/4
register
Row address
character

size
Windo/Col xeqi, ter
Col-{
color register
Horizontal position delay
$39
$3A
$3B
$3C

$3D
Test 1, OS D/Tin,
Test 2, EPROM

S3L
S3F

37

r/PM

.. .

........................................................
allocation.

:;
21
27
21
27
27
27
27
27
21
21
27
27
27
27
27
27
27
27
27
27
21
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
21
21
;;

.
Teletext
W
.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

00000000
00000001
00000002
00000003
00000004
Ooouooos
00000006
00000007
00000008
00000009
0000OOOa
0000OOOb
0000000.
0000OOOd
OOOOOOOe
OOOOOOOf
00000010
00000011
00000012
00000013
00000014
0D000015
00000016
0000001?
00000020
00000021
0000002.
0000002d
00000030
00000033
00000036
00000039
0000003a
0000003,
0000003?
00000040
00000041
00000042
00000046

27
27
27
27
21

27
27
,.

mode
P.qe
~.9.

..

.q:

~,

..5:

.,,

..6:

.,,

dis~l. cha~ter
reqist. r
y

di,P1ay control register

disP1ay control register
dlSP1ay mode reqister
active
active
active
;;?

(normal)
(news/sub)

chaPter register
... register
.01register
d~ta =eq~,ter

3,a
4th
LINKED PAGE NO, LIFO BUFFER
PAGE No.
ACO PAGE
AC1 PACE

INPUT
NO.
No.

BUFFER

AC2 PAGE NO.

AC3 PAGE N..
CYW PAGE NO.
INDEX PAGE No.
~,.. . .. .. .A, u.m.

.
.

:;
;; 00000047
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
27
21
21
27
27
27
27
27
27
27
27
27
27
27
;;

SUB1
R]

,2
R3
c1
C2
C3
C4
C5
C6
SUB2
R4
.5
R6
R7
SUB3
R@
.9
R1O
,11
P.
PT
,
.,,0
FACE
FAGO
PAG1
PAG2
PAG3
?*GC
PAG1
PDP
ACC
WACC
~DR
DPNT
SU~R
1OBu r
STAT2

STAT3

....................................*~#.
..................
..,,
,?
PLL DIVIDE
PLL
DIVIDE
WORK I,.

00000048
00000049
0000004.
OOOO004b
OOOOO04C
0000004d
0000004.
O000004f
00000050
000000s1
00000052
000000S3
00000054
000000ss

00000056

WP
WCAL

MSB

WTIO
MTIO

LSB

COUNTER
XEYWMD

COUNTER

12,8.s {inc, fzee ring)

12,8ms (i., reset every 1S d, inq tra, ient)
3.25 S [de., automatic standby timeout)
12. 8.s (cleared for .0.24 delay when Paqe a=rives)
12.8.s (i., transient mute)
TMSIEMT
DISP~Y
SECONDS COUNTER
0: TV/TELETEXT
1: IIC R/W
2: HOLD
3: IR REPEAT INHIBIT
4: TWSIENT
DISPMY
ON
S: T1~
HOLD
6: SUB-PAGE MODE
7: lR TASK PENDING
0: KEY FUNCTION PERFO=D
1: WCAL RZPEATING
2: P/c PROG : 0, CM
: 1
3: MUTE (TmSIENT)
4: OSD STATUS TWSIENT
5: MUTE (BUTTON)
6: COINCIDENCE MUTE
7: SZ~CH
$55 AT RESET, $~ NO&WLLY
CURRENT PROCW
NU~ER
CURRENT CMNEL
NU~ER
CURRENT DISP~Y
NU=ER
lPRO=W)
CURRENT DISPUY
NU~ER
(CMNEL)
FINE TUNING REG1 STER
oLm
LZ=L
BR1 LLIMCE
LE~L
CODE OF PRESSED KEY (mCAL)
LED DISP~Y
M
1, INTERRUPT TE~
..

.
.
.
Pm
,,0.

cm
DISP
DISC
,,..

AVOL
BRIL
KEY
~ No
lRW1

.
,.

1RW2
1R~3
1RW4
DIFFH
DIFFL
lRH
IRL
IRcODE
1RCNT
lRCMCT
OLDIR

lR TIW
.
.

.
.

DIFF~RENCE

IR cODE BIT
COLUCTION

38

........................................................
.
.
.
m allocation
for Ste, eoto
.

27
27
::
27
,,
;1
27
27
27
27
;7
27
2?
27
27
27
27
27
27
27
27
27
27

....................... .......... .......... ....... ......

0000006.
0000006d
0000006.
0000006f
00000070
00000071
00000072
00000073
00000014
00000075
00000076
00000077

xl
LVL

00000018

sTAT 5
.
.
.
.
.
,
.

STAT 6
,
.
.
.
.
.
.

STAT7
.
.
,
.
.
,
.

21
27 00000079
2,
21
21
27
27
,7
51

21
;; 0000007.
27
>,

;+
;;
27
27

;?L
HW
TONE
-TRIx

1
1
1
1

K2

;;
;;
27
27
21
27
27
27
21
21
27
27
27
27
21
27
27
27
27
27
27
27
27
27
27
27
27
27
>7

X!},ROW ~,

0000007b
0000007.
OO00007d
Oooooole
oQo0007r
00000080
00000081
00000082
00000083
00000084
00000085
00000086
00000081
00000088
00000089
0000008,
OOOOOO@b
0000006.
Ooooooed
0000008e
OOOOO08C
00000090
00000091
00000092

R.

=010,
1/2 6 .tline enable
character site
colon,& end col-

ad~res,

Winao.
ROW 2, color
R.. atidre,s c
Window CO1O.
ROW 3, colouc
ROW address
Window C.1OUX
ROW 4, CO1OU,

1/2 6 otli. e enable

character size
end col112 otli.e enable
character ,1,,
end col1/2 outline enable

ROW ,d&,e,,

character

size

Wind. CO1O, , end .01-.

Row 5. ..10., 1[2 , otl,.,e enable
Row address
character ,1,,
Window
CO1OUZ end col ROW 6,
CO1OUZ 1/2 & Outline enable
R..
aaaze.,
6 character ,1,,
W1dow
CO1OU,
ed colROW 7, CO1OU, 112 .utl. ne enable
R.. add,,,, character sire
Window CO1O.. end colROW 8, C.1OU. 112 otlie e.abl,
Row address character ,1.,
Window CO1OU. & end col-

1
1
1
1

12

NSED

23 BYTES USED FOR STACK

(1 INTERRUPT MD
9 NESTED SUBS I

22
1
SECTION
w
SECTION

.M2,
128

C0~

R0M2

39

........................................................
.

:;

.
.
.
T.1.t.xt/Tv
switching.
.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

33
34
35
,.
.

31
38
39
40
41
42
43
44
45
46
47
48
49
50
5:
52
53
54
55
::

00000000
00000003
00000006
00000008
0000OOOb
0000OOOd
Ooooooof
00000011
00000013
00000015
00000018
0000001,
0000OO1d
00000020
00000022
00000024
00000026
00000028
0000002,
0000002.

58
59
60
61
62
63
64
65
66
67
68
69
70

ooo0002f
00000032
00000034
00000037
0000003.
0000003.
0000003f
00000041
00000043
00000045
00000047
00000049
0000004b

71
72
13
74
75

0000004d
0000004f
00000051
00000054
00000056

76
77
78
79
80
81
82
83
84
85
86
87
08
89
90
91
92
93
94
95

00000059
0000005b
Oooooosd
0000005f
00000061
00000063
00000065
00000067
00000069
0000006b
0000006d
0000006f
00000071
00000073
00000075
00000077
00000079
0000007.
0000007b

,,

,000037
>0.0074

WTX
TXTON

>1000
>edoooo
,616
>b700
>1900
,1900
,,roo
,obooo*
,Iboo
>040003
>cd OOOO
a 6..
,b700
.646
,b, oo
>b$oo
>b, oo

NOTT

>..0000
>0.0048

UPDATE

>1800

>090003
>cdOOOO
>1100
>cd OOOO
>1100
.610
>b700
>1900

TXTOFF
TXTOF

>lbOO
,603
>b, oc
>b, oo

RST

>3f00
>cd OOOO
.602
>Ccoooo
,.602
.139
221b
.130
2517
>.601
.139
2211
,130
250d
>.600
.137
2207
.130
2503
>b700
81
99
81

0, STAT, TXTOFf
6, STAT7,
PWIC
0, STAT
OsDx
#S16
RI
4, STAT
4, STAT2
7, STAT2
5, STAT, NOTT
5, STAT
2, STAT, NOTT
NOTTH
*SCC
R5
4$46
R6
ACC
R4
TW2
6, STAT7,
PWIC
4, STAT2
4, STAT, TXTOFF
NOTTH
0, STAT
OSDLE
O, STAT
*$10
RI
4, STAT
5, STAT
+s03
RS
R6
R7

TEST

40

- .-...___..-..

___

. ...

192
193
194
195
196
191
198
199
200
201
202
203
204
20s
206
207
20s
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
2<5
246
247
248
249
250
25>
252
253
,s.
255
256
257
258
259
,<.
261

........................................................
.
.
.
Red,

,
Green
& Yello
keys
.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
0000O1lb
DO OOO1ld

RED

00000120 >cd OOOO

00000123 >cd OOOO
252c
00000126
00000128 >..0000
0000012b >b601
2025
0000012d

NPAGE

GREEN

00000134 >cd OOOO

00000137 ,.d OOOO
0000013,
2518
0000013. >..0000
oo00013f
0.0103
00000142 >010061
00000145 >b602
00000147
200b

PPAGE

90000149
0000014.
0000014C
00000152
00000154
00000156
90000157
00000159
0000015.
0000015d
Ooooolsr
00000161
00000163
00000165
00000161
00000169
0DOO016C
0000016f
00000172
00000174
0000017?
0000017.
0000017c
0000017,
00000180
00000182
00000184
00000187
00 C00189

>07005.

YEL~

OCO1O3
>030054
>b603
>b70C

ICI
UT

GWK
1Go

3, STAT3, RED2
1 NDXP

JSR

NOTOK 3
LPT
C~D
ACC+ 1
LPT

Bm
Jw

RED2
LPT2

ooo0012f >3foo
00000131 >06000h

S0

FORCE M 1 NCRWNT
REQUESTED ?

A-Y
GETIT
NO,

PDF

LINKS

OX

DECRWNT
ALR-YREQUESTED ?
NO,

S0

Jm
BRSET
BRCLR
LDA
BM

C~D
6,PORTB, lGO
0,STAT3, -C
ACC+2
~T

NO,
GYC

GETIT
BITS

BRCLR
BRSBT
BRC~
LDA
STA

3,STAT3, -C
6,P0RTB, Icl
1,sTAT3, mc
ACC+3

LINKS
ON 2
GYC BITS
EN~~D
YELW
LINKS
ON

MRCE

EN-LED
LINK ON ?

GREEN

?
?

W3

Ls m

x2
X3 FOR PAGE POINTER

.3

TX

>::0000
2547
>b600
>bl OD
2604
>1400
2009
>Od0U03
>Cdoooo
>cd OOOO
>3f00
>050003
>cd OOOO
.60=
>b700
>b60 O
>b700
>b700
>Cdoooo
>1500
>..0000

JSR
BCS
UA

NTSAC
SKOSP
CMO
cOK
NOTH~

:E
SET
BM
~cLR
JSR
JSR
cm

=C LR
JSR

K
U&
ST*
STA

JSR
BC LR
J~

TEST
MC
W3
ACC
NTSAC
2, STAT
Cmo
6,STAT, SKOSP
OUTSP
RSTR
PDP
2, sTAT, NOTHLD
NOHOLD
4SOF
C6
W3
,4
ACC
CFND
2,STAT
TXT2

IS PAGE No. 0. ?
IF NOT ~RT
ACC NO
IF SW
ACC CCT
THEN FORCE UNSTOP

~..>>
CHECK ,@}~%& lF-&lGH DO NOTHING
IF m%. (PA~ FOwD)
C=~
FOUNDS
TO ~g~,:;>mCHING
OF LINKS.

................................+.
.+.4%$ ........:
,
.!.,*
.
.
~,,,
Index Cyan keys,
.
.
.... .....................*.... .... .......................
0000018.
.eOf
0000018. >cd OOOO
0000 D191 2414
00000193 >ccOOOO
90000196

>07000d

CYM

B. CL. .* S9.*TA+2.mc

LINKS

GYC
CYM

6,STAT
PDP
PAGO+ 2,X
Pu
C2
PAGO+l, X
PT
c1
WA
S TA
SUB
STA
JSR
LDA
STA
LDA
STA
LDA
ST%

LDA
s T*
JSR
LDA

%; OOOOO1df
298
299
300
301

ON

LIMS

NO,

:4
CLR
3,STAT3, G~K
BRSET
JSR mxP
JSR N0T0K2
BU LPT

mo

,Vboo

..

...

P DP

iEET
JSR

000001.1
000001,4
000001.7
Ooooo lea

>1500
>cdOOOO
>cdOOOO
>.dOOOO
>,.0000

CYOK

FAGO,
PB

RESET PAGE WDE

R1l

BC R
JSR

2,STAT
TXT38
TM 1
SfND
TXT1

J,R
JW

?
EN=UD
ON ?

~S18
.3
lNDX
PH
PAGO, X
PT
PACO+l, X
Pu
,AGO+2 ,x
Acc
.8
1,,
iiso

S TA
C LR
LDA
STA

JSR

ON
BITS
L1~

R9
:0
RESET HOLD F~G
DISPMY
TOP ROW
SET FOUXOB

42

....... ........................... ................. .....

.
.
G.C linked yqe no, allocate t. AC C,.
.
.
.,, ,..,...,,,., .............. .... ................. .....

303
304
305
306
307

i08

309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
3S0
351
352
353
354
355
...
357

...

00000 led
0000O1ef
oOoOO1rl
000001f3
0000O1fS
0000O1fl
000001f9
0000O1fb
0000O1fd
0000O1ff
00000201
00000204
00000206
00000208
0000020.
0000020.
0000020.
00000210
00000212
00000214
00000217
00000219
O000021b
0000021d
0000021f
00000221

>b600
ab04
>b700

E
STA
CLR

,601
>b100
a6ff
>b701
>b702
>b103
>cdOOOO
>3.00
>b600
>b700
ad43
2406
>beoo
>.700
2003
>cdOOOO
>b600
ab06
>b100
>b600
.103
25.1

R
K
STA

00000223
00000226
00000228
0000022.

>cdOOOO
>3f00
.604
>b700

0000022.
0000022,
00000230
00000232
00000234
00000236
00000239
0000023b
0000023.
00000240
00000242
00000245
00000248
0000024,
0000024.

>3.00
>be OO
>,600
alff
2612
>CdOOOO
>b600
>cdOOOO
>beOO
>.700
>cdOOOO
>cdOOOO
>b600
alO1
22de

0000024e

LIW

>3roo

STA
JS R
lNC

L WP

K
0s
R

BHS

E.
BM
JS
R

NOTFNC

NEXTC

STA
WA
cm
B~

,s

LmP

...

0000024f
00000251
00000253
00000255
00000258
0000025,
0000025d
0000025f
00000262
00000264
00000266
00000269
0000026b
0000026d
0000026f
00000272
00000274

cm
BH 1

co.

~[~,~~~::;
,,1
OK 1

<:i,,:~~gy~j~j;j; <!:::04
*~:~990000029,
? *OO 000002,0
.. .,
...+,--402 000002.3
403 000002,6
404 000002,8
405 000002ab
406 000002ad
401 000002. f
40e 000002b2
409 000002b4
410
411

412
413
414
415

000002b6
000002b9
000002bb
000002bd
000002bf

>;ggg
>..0000

,Ul
0K2

>cd OOOO
>b601
>cd OOOO
>b700
>b600
>cdOOOO
>b700
20c0

R2BJ2

>cdOOOO

NOTTH

.018
>b700
,604
>..0000

PULL
WACC
CHCK 1
CONT

Ace, x
UP
GLP2
COUNT
+$01
L ~,

LDA

COR

GU 1

00000276 >.
00000219 >0
0000027. >0
000 D027f %
00000281 &t+2
00000283 Sk
0000
85 72
Ooo@*Y ~ >058009

,39& OOB% 294 >0?0009

COUNT
COUNT
ACC, X
+ $FF
A mc

., s

81

>h600
a113
2203
>07006.
,610
>cdODOO
>b601
>ca OOOO
>b7D0
>b600
>Cd OOOO
>b700
>b600
>b700
>cd OOOO
>,600
>b700

GCYI
*4
.0..

STA
JS R
JSR
LDA

A WC

z
COUNT
43
Lwe
WAC.

mx

4 FOR GHOST ROWS

$FF
&cc +1
ACC+2
ACC +3
1NDXP
COUNT
W3
R1O
GL,l
NOTFNO
CONT
&cc ,x
NBXTC
PUSH
W3

c,,

mx

CWTER

~D

:;
4

K
DE
C
?$
WE
JSR
E

%5*

365
366
367
368
369
370
371
372
373
374
375
376
371
378
379
38o
381
382
383
384
305
386
387
3B8
389
390

ACC
+4
R8
COUNT

IF

INDEx

1..0.,

DECODE

UNITS

DECODE

TENS

R
Pu
:NDX
PAGO, X
P

CHECK FOR ZERO ?

FETCH CURRENT
mG
PAGE HUNDREDS

JSR

WLO

BRCLR
,,s,,
BSET
BM
BC~
JSR
BRcLR
BRsET
=ET
:x,
BRCLR
BRSET
BSET
BM
BCLR
Jm

3, 10BF, OKO
0,?, 1
0,PH
OKO
O,PH
-,.
2, IOBF, OE1
l,FH, PT1
1,,.
OK]
l,F
3, 10BUF, 0K2
2.PH, PU1
2,,
0X2
2,,.
Sncfi

JSR
LDA
JSR
STA
LDA
JSR
STA
BM

R2B
IOBF+l
DECODE
PU
10BUF
DECODE
PT
R2BJ1

JsR
SB

REL1
*$18

STA
LDA
JW

:;M

LINKS

0.

YES , RoW 16 FOR LINKED PAGE,

FETCH 2 ,1.. BYTES

~G BIT ZERO OK ?
NO, S0 ToGGLE

MG
NO,

BIT
ONE OK
S0 TOGGLE

MG
NO,

BIT
SO

TWO

OK

TOGGLE

FETCH 2 LINK BYTES

DECODE :Ts

DECODE

R,

43

CONTROL

.1,,

3,STAT3, NOTOX
416
R2B
IOBUF+l
DECODE
W2
10BUF
DECODE

TENs

NO

417
418
419
420
421
422
423
424
425
426
427
428
429
430
43!
432
433
434
435
436
4,7
438
439
44C
44:
442
443
444
<45
446
44?
4<8
4<9
<5C
45:
452
453
<s&
45s
456
<~458
459
462
46:
<62
<62
<6<
<65
<66
,6<68
<69
~.:
,..
~,;
<,3
<~<
4,5
476
4,7
<-,
!,9
<e:
48:
<82
483
484
485
<86
487
488
,89

........................................................
N. bits C., default

}.1 -1) link,.

........ ...... ........... .................... .......... .

>Idoo
>cd OOOO

.0,0.3
.0,0.

>,600
>b, oo
>b602

JSR
LDA
ST.
LDA
1 NCA

>:00
>b702
.139
2312
a630
>b, oo
>b702
>3,01
>b601
,:39
2304
,630
>b70:
>h60:
>b7Gc
29b2

BCLR

s,,
ST.
cm
B LS
LDA
ST&
STA
: NC
LDA

cm
NOV9A
Nov 9

NoTOK2

BLS
LDA
51A
LDA
S 1A
Bm

6,STAT
1WDx
PAGO ,X
,.
PAGE.2

CMCELL

SUE-PAGE

,
,..s.2
*$39
NOV9
ts30
Pu
?AGE+2
PAGE. 1
PAGE+ 1
+s39
N0V9
+s30
PAGE. 1
PAGE+ 1
,,

BC LR
LDA
ST,
LDA

6.STAT
PAGO. X
,
,AGE~2

STA
STA

Pu

cm

+$30
NOV9
+s39
Pu
PAGE.2
PAGE+ 1
PAGE.1
8$30
NOv 9
*$39
NOV9A

CWCLL

SUB-PAGE

DZCA
BH S
LDA
ST,
,1A

3EC
L>,
:2
.9A
B=

PAGE+2

.,
...... ......4,.,. ................................,$t.}tw,~t~{~>~

GLP2

STA
LSU
LSM
Lsu
ST*
LSM

R2

C2

&cc
CLE~
CWTER
INTO
lIC
ROW O

+2
R:O
$2
*$39
=RT
c!
+s39
mK

CO LUW

,.?3
+6
TPAU2
3,R8
TXT38
SFND
TX71L

cLEm
cwr
ER
WAIT
FOR IT
DON*T
CLEM
THIS
T1%
PUT PAGE NUWER
IN CHAPTER
SE?
FOUND
F~G
WD
REQUEST
IT

+8
,H, x
LIFO, X
PSHL

44

.- .

..

523
524
525
526
527
,28
529 0000036e
530 00000370
531
532
533
534
535
536
537
538
539
541
541
542
543
544
545
546

........... ........................... ..................

.
1s Page already 1 M
?
.
........ .......... ................. ................. ....
>3rO0
>b600

SRCH
~PS

>rboo
>:00
>bLOO
260.
>.601
>bloo
,..
>.602
>b100
210s
>3.00
>b600
.104
25.0

Cm
LDA
LS~
~D
TM
LDA
cm
,..
LDA

WACC
WACC
WACC
PAGO, X
PH
,,.1
PAGO+l

,*\

....

6iG0+ 2,x
Pu
FND2
WACC
WACC
44
WPS

1.lN1

,,,
554
555
556
557
558 00000395 >3.00

575
576
517
578
579
580
581
S82
583
S84
585
586 O00003b9
S87 000003bb
588 O00003bd
589
590
591
592
593
594
59s
596
59,
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
.%.

.,.=.

~*>?

..........................
CWTER

.. .

mD
R8
*2O
R2B

>b700

.614
.d5a

LDA
STA
LDA
STA

>,l=lwoooo

~ 00h9>cdOOOO

CF..

k8@03ec
250,
..{~ ;,00~D03ee >1900
20
>cdOOOO
>.601
>b701
>.602
>b102
81

000003fe
Sr
000003ff >,600
00000401 >e700
00000403
SC
00404
>,09
630 00000406
631 00000408

1NDXP

ABCF
PULL
PLLL

#$zo
Rll
P.
3,STAT3, BMK

5ZACE

IOBUF+,

YES, S0 USE DATA

R0W24 EN-LED

10BUF

PH
ACC

ST.
L.,
BSR
B Ls

R8
*24
,20
m,

BSZT

4,R11

LDA

s TA

*25
,9

SET FOUSD FUG

mrTE
1:
.0.

L.A
STA
LU,
Jw

:0

COLUW

::T32

JSR
Ec s
BCLR

JSR
LDA
STA
LDA
STA
RTs
C LRX
LDA
S TA
INCX
CPx
B~

RI 1

LDA

3U
5,,.::$i$Q00003f0
.,..,
~.,.
613 000003f2
000003f5
000003S7
000003f9
000003fb
000003fd

4 FOR CHOST .0.s

ROW 20

BACK TO
DISP~Y
CWTER

CP BLF
MC?
4,R11
SFND2

CLEM

FOUND FM.

INDX
PAGO+l, X
PAGE.1
PAGO*2, x
PACE,2

LIFO, X
PH, X

;ZLL

RTs

45

633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674

676
677
678
679
600
601
682
683
684
685
606
687
688
689
690
691
692

.,,

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.

.
Read
and rite
Sub,.
ti-..
.
.
Cyan
6 Index
links
. link
control
wt..
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
00000409
0000040b
000004Dd
00000410
00000412
00000414
00000416
0000D418

>b, oo
.606
>cd OOOO
>3.00
>3.00
,b600
.126
81

W2B

V5

R TS

00000419
0000041b
0000041d
0000041f
00000421
00000423
00000425
00000428
00000429
0000042,
0000042.
0000042,
00000430
00000432

>b700
,608
>b700
,604
>b700
>..00

R2B
R2BN9

sTA

>cdOOOO
42
42
a60b
>b700
,622
>b700
>..0000

00000435
00000437
00000439
0000043.
0000043.
00000441
00000443
00000445
00000448
0000044,

.613
>b700
>Cdoooo
.640
>cdOOOO
a61f
,b700
>Cdoooo
.650
>..0000

GCY I

0000D44d
OODD044f
00000451
000004S3
00000455
00000457
00000459
0000045.
0000045.
00000460
00000462
00000464
00000466
00000468
0000046,
0000046.
0000046,

>b600
ab04
>b700

CLINK

STA

cm

::T32
R1O
R1O
R1O
*38

LDA
LDA

R9

sTA

::B3

sTA
LDX
JSR

2
*SUB3
SEND22

UL

DEUY
TO
llC
TIMING

UUL
REm22

>3800
>3f00
.610
>cd OOOO
>b601
260.
.625
>b700
.610
adbl
>b601
ad03
>bf OO
81

WA
sTA
JW

K
JSR
%

CYW

*19
R1O

cm

4s40

DA
STA
JSR
LDA
JP

GW2
*31
R1O
Gw 1
*S50
GLP2

LDA
~D

Kc

sTA
cm
cm

STA73
R1O

LDA
JSR

+16
R2B
10BUF+ 1
NPK27
*37
RIO
+16
R2B
10 BUE+1
DECODE
STAT3

INDEX

z
B%

DA
SR
sTX

RT S
. . . . . . . . . . . . . . . . . ...

, . . . . . . ..l.-$*.

H.mig

d.c.d~.::;.

m,

ON 1

WA
sTA
BRSET
BSET
BM
BC LR

69*

0000046f
00000471
00000472
00000475
00 D00477

. . . . . . . . . . . . . . . . . . . . . . .
.
.
.
. . . . . . . . . . . . . . . . . . . . .

,,

.,/

703
104

SATISFY

694
695
696
. . .

699
700

24

RW

R9

E
lNC
INC
DA

>b700
5f
>d60000
>b100
2732

;;: ~oooo4,3
>b,
oo
.......
..

r,

..

.4..

. . . ...?.

708
709
710
711

0000047.
00000480
00000482
00000484

712
713
714
715
716
717

00000489
0000048.
0000048.
00000491

718
719
720
721
722
723
724
125
726
727
728
729
730

03000487

00000493
00000495
0000049
0000049

.-

-...

>1000
2002
>1100
>cdOOOO
2722
>d60000
>b700
>020004
>1200
2002
>130

tl~:i,
738
739

TRTW

LDA
sTA
BRSET
BSET

Twl

:~R
BS R
BEQ

m,
x
SUB2
2, SUB2,
2, SUB2
TW1+2
2, SUB2
SSUB
FND

z
BRSET
BSET
BM
BCLR
BS R
BEQ

m.x
SUB2
3,suB2,
3, SUB2
TH1+2
3, SUB2
SSUB
FND

FNDJ
TRTH

:::::::
Oooooab
000004bb
00 DO04bd

BSR
BEQ

0000049
0000049
000004,
000004.
000004&
00U&@>
0
0

7 *8$
Z*?
7~~; i D*@V 4 b
,,&c? 3 ***O 0004 b
<..Y*$

\\.

.,1
ad56
2750

SUB2
l,suB2,0N1
1, SUB2
0N1+2
1, SUB2
SSUB
FND

TW1

THI

46

-.

, .

.
....-.-...-. -

141
742
143
744
745
746
747
748
749
750
751
752
753
754
755
?56
751
758
7s9
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
771
778
7,9
780
781
782
783
784
785
786
787
788
789
,90
791
792
..+
794
795
796
797
798
...

#,,

..................................... ................. ..
.
.
M.,, Hamlq
decode.
.
..................................... ................. ..
000004bf

000004.2
000004.4
000004.7
000004.9
000004cb
000004cd
000004cf

>d60000
>b100
>080004
>1800

2002
>1900
ad44
273,
>d60000

>b700
>0,0004
>1.00
2002
>lbOO
ad32
272.
>d60000
>b700

>oc0004
>lCOO
2002
>Idoo
ad20
271a
>460000
>b, oo
>0.0004
>1,00

2002
>1 foo
adOe
2708
00000507
00000508
0000050.
0000050.
Oouoosof
00000512
00000513

00000515
00000517

::or
2203
>,.0000
>d60000
81
>b600
>b100
81

,,,0

FOl

TRF 1

F1l

TRS 1

LDA
S TA
BRSE1
BSET
Bm
BC LR
BsR

m,
x
SUB2
5,suB2, F11
5,SUB2
,11+2
5,SUB2
SSU8

BEQ

FND

LDA
S,,

w,
x
SB2
6,sB2, s11
6,sB2
,11+2
6,SUB2
3SUB
FND

BM
BCLR

BSR
BEQ
TRSE

LDA
S TA
8RsET

8SET
SE 1

Bm
BCLR

BSR
BEQ

INcx
.,.
FND

SSUB

w,
x
SUB2
7.SB2, SE,
7.SB2
SE1+2
1,*B2
SSUB
FND
*SOF

BH 1
Jm
LDA
RTS

FND
TM
NU.,

LDA

SUB
wl

cm
RTS

800 00000518 >0.0015

801
802
803
804
805
806
eol
808
809
810
$11
812 00000530
813 00000532
814 00000534
815 00000536
816 00000538
817 0000053,
818 0000053.
819 0000053.
, 820
821 00000540
a
822 00000542 >.
823 00000545 ,.~
824 Ooooc
825 Ooooc
826 0000~
821 00 *>
628 Oh Q680, ad15
829 @ b @55~
,606
830 O~O@*ti
>b700
U*ti 0800556
.607
* k@0558
~hloo
.P,i>
,*I,
\ $&~$O00055a
.605
0000055. >b, oo
.604
00000s5.
00000560 >b?oo
00000562 >.,00
00000564 >..0000

MIX

00000567
00000569
0000056b
00000S6d
0000056f
00000571
00000573
00000575
00000577

E.RO

.619
>b700
.606
>b700
>b600
>b700
>3f00
,605
>..0000

x
SUB2
4,SUB2, F01
4,sB2
F01+2
4,SUB2
SSUB
FND

BRSET
BSET
S1l

-,

LDA
S TA
BRSE1
BSET
BM
BC LR
BSR
EEQ

BRSET

R5
4$17
R6
TW 2
Bc L,
LDA
S TA
LDA
S TA
LDA
S TA
Bm

7,sTAT2
#$16
RI
4$CC
R5
*%46
R6
TW2

LDA
JSR
LDA

~XOOF

,s,

s46 FOR NOMIX F3SH{SUBT

M:XED,

S0

NOM:X

CCT, 312/312 SYNC

EN-LING
GHOST ROWS

*2
s?

SET-,

s,..

FOXCE

HE~ER

C LM
JSR

WXOON

BSET

4, ST.,
FRO

BSR
LDA
STA
LDA
STA

TXT2

:;R
4$01
R?

LDA
STA
LDA
STA
LDX
>W

:?

LDA
S TA
LOA
STA
LDA
S TA
C LR

025
,9

5s

u,

~XES

CONTROL

44
SUB2
#SUB2
SEND22
FoRCE

:;0
ACC
,8
RLl

LDA
JW

ALL

EN-LE
DI,,

DISPMY

TIER

::T32

47

,X.,X,

0.

HE~ER

8S2
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
88:
882
883
884
885
886
881
888
889
890
891
892
893
894
895
896
897
898
899
900
931
902
903
994
995
996
907
...

........................................................
.
.
old.
.
.... ................. ...................................
0000057.
0300057.
0000057f
00000581
00000583
0000058S
00000588
0000058.
00000s8.
0000058d
0D00058f
00000591
00000593
00000S96
00000599
0000059.
0000059.
0000059,

>3f00
>040062
>1400
>b600
7b7D0
>cdOOOO
>3f00
>ldOO
SF
ad2b
>b600
>b700
>cd OOOO

HOLD

Ucnom

>0.000.

000005.3
000005,5
000005,7
00 Doc5a9
000005ab
000005. d
000005Af
000005b1
000005b3
000005b5
000005b7

>3f00
>3f C0
,3f00
.60!
>b70@
.60.
>b700
.601
>b700
>aeOO
>ccOOOO

000005ba
000005bc
000005be
000005bf
000005.1
000005c3
0>0005.5
090005.6
000005,8
CB0005c,
Oooooscd
0CO005cf
30000562
30000Sd4
OCOOCSd?
0D0c05d9
090 C95dc
0000 OSde

>bf

OO

>3f00
4f
ad07
>b600
ab04
97
.604
>b?OO
>d60000
>b?oo
>d60001
>b700
,d6C002
>b700
>d60003
>b700
>cc 5000

R8

JsR
C LR
BCLR
CLRX
BSR
LDA
STA

UP
R9
6,STAT

0s,
8M
TXT1
TXTIL

sFm
5PM

D1SP8

NOROLD

kic

S,,

BC U
LDA

>1700
.604
adll
20ae

PDF
;, ~,

LDA

,s,
c ,x

>3f00

000005.0

C LR
BRSET
BSET

BRSE1
CLR
C LR
CLR
LDA
STA
~A
STA
LDA
S1,
LDX
JW

ROW
0
RESET

SUB-PAGE

WDE

DISP8

&cc
R8
UP

X9
3,R3
::M
TW3
6,STAT, SPU2
C3
C4
Cs
*$OF

C6
*1O
W1
*1
SUB1
$SUB1
SEND22
.3
R9
D15P4
W3
#4

51s?4

:0
LHOLD, X
R1l
wow+
1,x
PH
LHOLD.2 ,x
PI
LHOLD+ 3,X
Pc
TXT3

,...... ...................i

ACC

UP
lNDX
PAGO,

PH
+$18
.3
PAGO+l, X
,,
c1
PAG0.2, X
Pu
C2
TXT1

48

.-..

956
951
958
959
960
961
962
963
964
965
966
967
96E
969
970
971
972
973
914
915
976
977
97a
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
10D2
1003
1004
1005
1006
1001
1008
1009
1010
1011
1012
1013
1014
1015

......................... ...............................
.
.
.
Reveal, t.~lbot ton clock.
.
..... .......... ..................... ................. ...
>0,0004

0. S.,
BSET
Bm

>1.00

2016
>Iboo
2012
>07000b
>090004
,1100
2008
>1800
2004
>1600
>1900
>..0000

REV
EXPTB

>O. ood.

BRSET

>010003
>..0000
>0.0025
>b60D
>b700
>Caoooo
>1800
>1.00
4f
adlc
,61,
adle
>Cdoooo
8609
>b700
>Cdoooo
.646
>b, oo
>b700
>cd OOOO
.606
>b700

BRCLR

>Y700
.620
2008
>b700
a60b
2004
>b, oo
.60.
>b700
>b100
>b600
>b?oo
>3roo
.606
>C.000o

CmCK

;ZET
LDA
STA
JSR
BSET
BSST

CLBSR

NOBx

LDA

#30
mxOON
FRO
+$09
,7
TXT2
+$46
R5
R6
TXT2

BS k
JSR
LDA

STA

JSR

s%
STA

JSR
TAO

.....
5,STAT. TAO
ACC
R4
UCHOLD
4, STAT
S,STAT

LDA
STA
RTs

?W

STA

LDA
mxoo.

WXOOF

mx

BM
STA
LDA
Bu
STX
LDA
ST.
STA
mA
s TA
C LR
LDA
Jm

49

1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
:043
1044
1045
1046
1047
104E
1049
:050
1051
10S2
1053
1054
1055
1956
:057
:058
:059
:060
!06:
:062
:063
:064
:065
:066
1067
:068
:069
:070
.97>
:072
:073
:074
1075
:$76
:077
:078
:C79
:C8C
:58:
:382
:383
:084
:085
:086
:081
:988
:$89
:G9C
:C9:
:C92
:093
:094
:095
>096
. . . .

........................................................
.
.
sub-page n-r
et ry r.tie.
.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
000006.6
000006.9
000006ab
000006ad
000006a
C
O00006b1
000006b3
000006b5
000006b7
OOO006b9
000006bb
O00006bc
000006be
000006c0
000006.2
000006.4
000306.6
OOOO06C8
00 C006ca
000006..
000006ce
000006d0
000006d2
000006d4
OO0006d6
000006d8
000006da
0000 C6d.
000006de
000006,0
000006,2
000006.4
000006.6
000006,7
000006,9
O00006ec
000006ee
000006f
D
000006f2
000006f4
O00006f6
000006f8
000006fa
0000C6fC
000006fe
00000100
90000103
00000106
OOOC0708
oooco 70a
3000070.
0000070:
0000071!
00000713
00000715

>cdoooo
>b600
.010
>be OO
2704
.302
260f
.107
23o2
,008
Sd
2606
,103
2302
,004
ab30
>,703
,303
2?14
a62.
.301
2706
.30>
2?C4
>b704
>b705
>b706
,3c O0
2002
>3f00
>b600
,b700
4f
>b100
>cd OODO
.602
>b700
>b603
>b700
>b604
>b?oo
>b6C5
>b700
>b606
>b700
>cdOOOO
>cdOOOO
>b600
2661
.606
>Cd OOOO
>b603
>b, oo
>b604
>b700

T,STF

DIGITS

W2
416
?.?
,0

SDo

;~RTD
THOU
::CH
#a
SOCH
$ OR TD
!iRTD
SORTD

::30
PAGE+ 3,X
~LRP D
t $2A

HUN
SEN

smPD
SPCN

BEQ
CPx
BEQ
S TA
STA
51A
1 NC
BW
cm
LD.
STA
C LW
S,,
JSR

E
s%
LDA

s TA
LDA
S TA
WA
STA

J3R
JSR
B%
LDA
JSR
LDA
STA

s?

!:N
::N
FAGE+4
PAGE+ 5
FAGE+6
PDF
sPGN
PDP
Acc
R8
R9
WXOON
82
R:O
PAGE*3
R1l
PAGE+4
PH
PAGE+5
PT
FAGE.6

Pu
TXT3
TW1
PDP
Sw
;;BX
PAGE+3
.11
PAGE+4
PH

.
.
.
000307:7
;00097:9
0000 C7:k
GO09071d
O09C07:?
309 C3721
00000723
OC000725
00000727
39000?29
0000072b
0000072d

>b60:
>bl OC
>b602
>b700
>b603
>b70C
>b604
>b100
>b605
>b700
>b606
>b700

C4
PAGE+ S
,5
PAGE+ 6
C6
PAGE

PAGE

HUNDREDS

PAGE

REQUEST

*$18
;:C
?

TXT38

TXT3
TXT32

JSR
BC LR
JW

TXT1
2, STAT
SFND

LDA
cm
BNE
LDA
S,,

PH
,330
TXT3
*538
,H

HNDREDS

IT

REQUEST
NOHOm
~lTE
ONE

TO

FOND

LDA
STA
s%
LDX

1123
1124
1125
1126
:127
1128
1129

00 D0075d
0000075f
00000761
00000763
00000766
00000768
0000076b

>b600
>b700
>b60Q
>cd OOOO
.604
>..0000
81

SEN022

LDA
ST&
Jm

TPSTP

LDA
STA
LDA
3SR

PAGE
R3
ACC
,,.

Y$
R,,

s Pn

S~

HOLD DRING
SUB-PAGE
NUWER
ENTRY

ii

50

_.

-.-....... .

1131
1132
1133
1134
,135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
11S6
1151
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1112
1173
1174
1115
1176
1171
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
,,.

....................... ........... .......... ............

,
.
Sub (timed) Page,.
.
...... ........... .......... ........................... ..
0000076. >0.002.
0000076f >Icoo
00000771
ad..
00000773 >3r00
00000775 ,cdOOOO
00000778 >.600
0000077. >b700
0000077. >.601
0000077. >b, ol
00000780 >,602
00000182 >b,02
00000784
.62.
00000786 >h, oo
00000788 >b700
0000078. >b, oo
0000018. ,b, oo
Ooooo?ae >b600
00000190 >b700
00000192 >3foo
00000194
.602
00000196 >b700
00000198
adb2
0000019. >..0000
000

D079d

0o00019f
000007,1
000007,4
000007.6
000001,9

adOd

SBPG

,s,
CLR
JSR
LDA
, TA
LDA
S TA
LDA
S TA
LDA
S TA
S TA
S TA
S TA
,DA
S,,

>Idoo
>3foa

000007b0
000007b2
000001b4
000001b7
000007b9
000007bb
0000O1bd
000007bf
000007c1
000007.3
000007c5
0000OIC1
000001c9
000007cb
000007cd
000007cf
000007d1
000007d3
000007d5

.650
>b, oo
>cdOOOO
>,600
>b, oo
sole
>b700
>,601
>b700
>b, oo
>,602
>b700
>b700
>3foo
a602
>b, oo
>b600
>b, oo
>..0000

6,STAT, OUTSP
6,STAT
T,.,,
,.,
I.D.
PAGO,.
PA..
PAGO+ >,,
PA.. +1

C.,
.DA
S TA
BSR
J*
OUTSP

BSR
LDA
JSR
BCLR
JSR
JW

RSTR

BCLR
CLR
LDA
s,,
JSR
LDA
S,,
SUB
S TA
LDA
S TA
S,.
LDA
S TA
S TA
c LR
LDA
STA
LDA
S TA
JW

>b600
>cd OOOO
>1500
,cd OOOO
>..0000

000001,.
000001..

,RSET
BSET

........... ,.

....
1191

1192

,,..
....

................. ...... .................................

1194

ACC
R8

5,STAT2
42
RIO
*25
R2B
10BUF+1
;;6
5,STAT2
C6
IOBUF
SU82
LDA
S TA
LDA
,s3
LDA

cm
BEQ
BSET
, TA
LDA

cm
B=Q
,s,7
STA
MD
SCZR
..,,

.m
ST,
WA
STA
LDA
JSR
,CLR
,, CL,
BSET
L.,
,W
OEQ
,,.,
S,A

1243
1244 00000836
1245 00000839

Of O1O1
81

CGET26

XINUTE,

UNITS

MI NTES

COL W

TENS
4

, W...,

CBIT

:0

+2>
R2B
10EUF+ :
C4
SU4
5,STAT2
C4
10BUF
C3
S3
5,S,.,2
C3

ROW

tiOURS

UNITS

H.R, ,..S & CB:TS 5 6

4SOC
2, S?AT?
3,STAT,
ST*T7
STAT7

SAW CBITS 5 6 :. STAT?

CLEM
NENSF MS B:T
CLEM
SUBTITLE ,1,
COLm

::0
*25
R2B
3, SB2
1,10BUF+;.,TR5
3,SB2
SB2
.5
CGET26
5,S7.,2
C5

6 ,cONTROL 91 TS ]

Rvd
XFER

CBI?B

(UPDATE)

TO B:T 3 OF HINTES TZNS

(RE?MCIKG CB1T4 (E-SE) )

PACKET 26 ENm...

BRCLR

RTS

51

......... .......... .......... .......... .................

.
.
.
Proces, Packet 26 info,
.
.
.. .......... ............................... .......... ...

1247
1248
1249
1250
1251
0000083,
0000083.

a6ff
>b700

GET26

LDA
s TA

*SF,

0000083e
00000840
00000842
00000844
00000846
00000848
0000084.
0000084.
0000064,
00000050
00000852
00000855
00000857
0000085.
0000085.
0000085,
00000860

>3f01
>b600
.b04
>b700
>b601
>b7U0
>3.00
>b600
aloe
2303
>ccOOOO
>b600
>cdOOOO
>b601
>b100
26de
>3,01

mP26

c LR

ACC

.1.0+1

ST-T

NEW
RW

%
sTA
LDA
sTA
lNC

$
LIFO+l

GHOST

CWTER

YWA
B=
J=

+14
OKRW
END26
LIFO
R2B
IOBUF+ 1
LIFO
WP26
LIFo+l

00000862
00000864
00000866
00000868
0000086.
0000086c
0000086.
00000870

>b600
ab04
>b700
>3.01
>3c01
>b601
>b700
.126

Omow
E
YWA
BNE
DZC
mP62

00000812
230d
00000814 >3f O0
00000876
a6ff
00000878 >b?OO
0000087, >b600
0000087. >cdOOOO
00090S7f
20bd
0900088:
00000883
00000886
COOO0888
0000088.
0000088.
0900088,
S000:890
90CO0892
CC 00 0894

>b600
>cd OOOO
>b60:
>b700
>b6C0
>b701
>3.01
>b601
>b700
>cd OOOD

NxTCH

LIFO

R1O
LIFO
L1FO

LOA
mD
STA
INC
1NC
LDA
STA
cm

ACC

Bbs
CL.
LDA
STA
LDA
JSR
BM

NXTCH

w2B
WP26

WA
JSR
LDA
sTA
LD&
sTk
1NC
LDA
STA
JSR

LIFO
R2B
LOBUF+ 1
LxF0+8
1oBUF
L1F0+7
LT,O+l
LIFO+l
R1O
R2BN9

~i
LDA
-D
Lsm
LSM
STA
JSR
L,.

:
LIFO+
LIFO+
LIFO+
R1O

1
1

+38

R1O
~$FF
Rl 1
LIFO

10BUF
L1F0+6
LIFO+,
#$7c

)sAWMODE
RW

24 ?

SUBTMCT

40 FOR RW

RW24

L1F0+4
NC
R8
L1fO+5
R1O
B. C LR
LDA
cm
BEQ
C LRX
BC~
LDA

cm
BEQ

L?
:;
BLS

4, L1Fo+2. NOTD
LIF0+2
*S1O
NLD

DIACRITICAL
NULL
YES,

JUST

NULL

DIA

?
SEND

1,

(B

IT7-11

7,LIF0+6
CT-, X
L1FO+6
CHFND
+7

*91
TRNCH
CHNF

LDA
UD
S TA
BEQ

L1F0+2
4SOF
L1f0+3
CHNF

;E
cm

~TT
,,

BNE
SUB
Bm

52

--

.. -

1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1313
1374
1315
1376
1377
1378
1319
1380
1381
1382
13e3
1384
138s
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430

lob
2702
2049
.005
>b703

ooo0090e

0000090,
0000090.
0000090.
00000910
00000912
00000914
00000915
00000918
0000091.
0000091d
ooooo91r
00000921
00000922
00000923
00000925
00000928
0000092,
0000092c
0000092d
00000930
00000932
00000935
00000937
00000939
0000093,
0000D93b
0000093d
00000940

CED1
Uoc
GTT

>fib03

L1F0+3
c,-, x
WTCH

>b602

cm

211,
.102
263.

BEQ

cm
BNE

,.32

BCLR
CL3X

LDA

BNE
STM

;;fo
>d60001

201.
>1f06

G2B1T

,::0000
2603
>..0000
>b106
2704
5.

TN23

>460001

G23F

:;fo

STW

2004
NULD
CHNF
WTC H

,603
>cd OOOO
>ccOOOO

00000950

81

EMD26

0000095;
00000957
0000095d
00000963
00000969
0000096f
00000975
0000097b
00000981
00000981
0000098d

202021,02383
248426932740
289429 .72,,2
2cb.2d5e2ebe
2f7630cb37c?
388a39a73a.2
3c823d8c3e89
3re161f963e5
69 fd6be66cfe
11r879cc7ccr
1f7r00

G2Tm

00000990
00000996

51815 b8d5c&b
Sd8e5E
2000

MODE

L1FO+2
*$OF
G2B1T
+$02
END26

LDA

alor

.3B1T

lLEGAL

LIFO+3

>::DOOO
203,

00000942 >1,06
00000344 >b606
00000946 >b, oo
00000948
OOOO094a
0000094d

*11
CED 1
C HWf
+5

;:
BEQ
1 NCX
I . . .
B=
LDA
BM

.3,=, X
STM
-,62
LIFO+6
G32F

BCm
CLRX
LDA
BNE
Jm

TN32
G3T-+l, X
GOTCH
7,LIFO+6
G2Tm, x
ST~
mP 62
LIFO+6
G23F

;FQ
lNCX
lNCX
BM TN23
LDA G2Tm+1,
BM GOTCH
STA

7, L1FO+6
L1FO+6
*11

LDA
J..
Jm

+5
TXT32
mP62

BSET

LDA

,,s

...............

.................

FCB
FCB
,,,

$51, $81, $5B,s8D, $5c, $8B

$5D,$8E. S5F, $20, $O0
FCB
FCB
FCB
FCB

a
A
,

FCB
FCB
FCB
FcB
FCB
FC B
FCB
FCB
FCB
FCB

N032
.016

C LR
BR C LR
BSET
BR C LR
BSET
BR C LR

BSET
.08

BRCLR

,04

BRCLR
,S ET
,,,.,
BSET
R,,

.02
No 1

BSET

\
.:,0+5
1, ,1,0+7,.332

5, L1FO+5
0, LIF0+7, N016
4, LI,0.5
6, L1FO+8, N08
3, L:FO.5
S,L:FO+8, N04
2, LIFO+5
4, L:FO+8, X02
1,L:fo+s
2, L;Fs.8, N01
0, .:.0+5

53

........................................................
.
.
.

.
Fetch initial Page from 8/30 format 1,
.
.
........................................................
>Iboo
>ld OO
>ld OO
>3foo

.602
>cd OOOO
.604
>b700
,660
>b, oo
>3,00
2735
a601
>cd OOOO
>3f00
.617
>cd OOOO
>b601
26ec
a601
>b?oo
.630
>b700
a617
>cd OOOO
>b600
>b?9C
>b700
>b6$0
>h, o:
>b660
>b, c2
>b6C0
>D, oo
>cd OOOO
>Ccoooo
>b601
.:]~
2502
>:.00
.63:
>b, ::

GIP

5,ST&T
6, STAT
6,STAT2
PDP

CW
CLCL-

::H
*4

TXT
1

Re
TRYAG

T1~ HOLD
sUB-PAGE MODE
NO TXT FMG

lFNF
&l

AGAIN ?

ONLY

{GHOST)

96 TRYS

TPAV2
R1O
*23
R2S
10BUF+1
TRYAG

8/30 FOMT
COLUW

:;0
*S30
PH
*23
.2BJ2
PH

GETIND

BYTE

CWTER

496
W3
W3

1 FOR INITIAL PAGE

RESET PACE HUNDREDS

LINE 23 (PACKET 8/30)
INITIALISE

INDEX

[B~CK1

lPNF

P8300K

-ITE

m.

ON

MX

OFF AT 31 32

.
:;T32
+20
.3
R1O
+23
R2B
1OBuF +1
,S7F
R1l
10BUF
,$7F

ST-T2

JSR
J%
LDA
STA
,SR
LOA
JSR
JSR

lNITx?

WX

ST~T

RWING

RW

;!
*1O
R1O
*24
W28
W3
W3
W3
439
EA

SDLY
NIIcD

~lTE

OFF

@ COLUW

STmT

mlTING

~ITE

TO ROW 24

ALL

20

23 - PACKET 8/30

DONS

AT

COLWW

10

TXTOF

+20
TPAU2
+Soc
R8
:y3

CLEU

TPAU2
GIP

GET PAGE No. FROM 8/30

cWTER

54

..

........................................................

1572
1513
1514
1575

Set P ROW 24 trasiet.

........................................................

,.,.

>0.0022
>1.00

s.24,

>2:0000
>1700
.604
>cdOOOO
,604
>b100
.606
>b700
>b700
.684
>b700
>Ctioooo
,606
>b, oo
,1800

BRSET
BSET
CLM
JSR
BCLR
mA
JSR

NOTXTX

81

6,STAT2 ,NOTXTX
5,STAT

,1=

HOLD -

ACc 0
SWP
IT
3 BYTES

UP
3,R3
+4
SPU
*4
R4

YT;
LDA
STA ;;
STA R6
LDA 4s64
sTA R7
JSR TXT2
LDA
STA :&
BSET4, STAT
RTS

PUT 24 AT TOP

($44 FOR CURSOR]

.........................................................
,
.
for
conversion,
Tables
..0
Haminq

.
1604
1605
1606
1607
16o8
1609
1610
1611
1612

HEX-ASC1
decode

1+,

STQP,,

.~:}

............................... .........................

OOOOODID
00000b21
O0000b29
O0000b31

3031323334353637
0bOb53544f500a Oa
1502495 e6473382C
ti0c78C9Qa1b6fdea

NUH

LHOLD
w
we

:<

~i.:<

+?3,:,,
,$..

FCC
FCB

FcB
FCB
END

S*O1

cross-reference

.27
.27

DR

Awc
mm
MAL
AVOL
BCOL
B~K
Wr

.29
.27
216
614
.93
491
.27
343
1153
,27
345
.27
.27
.>,

220
.623
146
502
;:0
1162
662
.353

222
,506
106
561
1186
1120

229

261

263

179
584
1196

181
599
1251

20s
676
1274

288
929

mx
WXOOF

WXOON
BRIL
BROW
c1

C2
C3
C4

939
480

1086
500

1207

55

1179
942

1088

1182

309
945

316
981

317
1053

326
1125

Spol
CNT1
CNT3
CNT4
CNTS
COK

COR
CONT
CFBLF
CT=
cYm
cYOK
DD 1
DECODE
91FP.

D1f FL
DIGIT
DIGITO
DIGITS
DISC
DIS?
DISP4
,15P8
DPGN
DPNT
DW
EA
END26
Em
EXP
EXPTB
F1l
FIN I
FND

FND2
FNDJ
FO 1
FRO
FTNE
G23F
G2BIT
c2Tm
.32F
G3B:T
G3:m
GCY 1
GET25
GET* 6
G,TIND
GETIT
G1,
GmK
GLP:
GLP2
WTCH
GREEN
G??
3:
w.
.W 8
.,;G
HOLD
w.
HVL
fin
lAC
IGO
:G:
:G2
INDEX
lNDX
:NDXP
lNITXT
IOBUf
:PNF
IRCMCT
IRCNT
IRCODE
IRH
lRL
lR~:
1RM2
IRm3
IRU4
.1
K2
KEY
KOUNT
L~L
mo
mOLD
LIFO

.27
.27
.27
.27
.239
367
.27
613
1331
.;; ?
.574
372
.27
.27
103
,7
104
.27
.2,
893
867
128
.27
.27
.1541
1266
1310
967
::8
536
721
542
794
749
828
.27
1394
1371
1390
1381
.1375
1377
336
2:
>244
.1493
.158
;:9
323
352
1967
17
1352
385
702
.1610
952
19
1044
.,,

cros,-refere..

.369
312
.945
1366
261
375

320

325

332

339

341

407

572

690

.700

1S4

763

772

781

756

165

342

349

3s3

.1432

404

.105
.103
.1023
.898
.890
,130
1559
1373
.1447
.973
.967
760
539
730
~548
712
751
,842
.1398
.1388
1398
,1305

.1409

.761

.543
739
.730
.752
990

.1417

1385
.665
.1196
.1253
1513
259
.215
.365
.471
1386
.208
.1363
.388
714

.1468

,570

667
669
1399

672
614
.1403

723

132

.9s4
.858
.1048

977

903
626
1298
1364
1455

905
1254
1299
1369
1456

1290
1379

1392

213

218

1503

147

.1608
1256
1304
1375
14S7

931

1141

1172

395
1294

403
1296

406
1303

571
1485

573
1s04

595
1544

597
1547

603

689

952

1203

1260
1305
1380
1458

1262
1309
1368
1459

1263
1311
1393

1267
1320
1401

1270
1323
1402

1272
1327
1441

,2?7
1330
1448

1218
1331
1449

1219
1336
1450

1288
1338
14s1

1292
1347
1452

1407
.224

56

_ .....------

.2;*1
MTRIX
MIX

ME
NEXT.
N1l CD
NO1

N016
.02
N032
.04
N08
NOBX
NOCH
NOHOLD
NOMIX
NOTCF
NOTD
NOTFND
NOT HLD
NOT OK
N0T0K2
N0TOK3
NOTR
NOTROM
NOTT
NOTTH
NOTXTX
N0V9
NOV9A
NPAGE
.,.27
NTSAC
NULD
NUH
NXTCH
OK 0
ox 1
ox 2
OKROW
Oml R
ON 1
OSDL
OSDLE
OUT
OUTSP
P8300K
PAGO
PAG1
PAG2
PAG3
PAGC
PAGE

PAG1
?mIc
PDP
FH
PLLH1
P LLL
PLmW
PPAGE
PROG
PSHL

.;;4
327
,1565
1458
1450
1456
1448
1454
1452
148
114

240
800
1354
1330
324
239
368
212
202
.1560
1314
46
22
1578
432
.441
18
684
232
1333
787
1284
384
390
395
1265
*27
116
.27
.24
964
236
1506
.27
481
1180
.27
.27
.27
.27
.27
436
1062
1495
,27
.;:
1471
.27
448
1490
.27
.626
.27

.,o,

s-r,feren.e

.800
603
,329
.1460
.1452
.1458
.1450
.1456
.1454
805
116
859
,812
.1358
.1369
.328
.241
,424
.446
.423
1323
48
.:;95
439
463
.201
.692
.235
.1401
,1607
.1292
381
393
398
.1267
718

9a 7
.118
.915

.50
61

~lool

~411

.442

454

.389
.395
.400

4d4

461

.719

40
966
1137
.1508

63
970
.1161

972

::4

:;6

87
534

162
537

92
431
1064

119
441
1066

125
442
1074

126
*49
1076

1494

1491

1499

58
113

:;1

;;9

13$
48s
14s3

150
517
1509

.975

149
459
1091

84
145
3e5
627

;:8

.94
270

386
902

388
934

161
620
1093

283
940

285
1142

287
1144

380
:146

%25
1173

447
:17?

163
622
109S

166
1039
1098

169
1047
1123

421
1048
1143

430
1049
1145

435
:060
1147

858

1026

1050

:052

1070

1140

1:69

391
1010

392
1063

394
1077

396
:101

397
1111

399
1:50

426
::74

630
.211

.::
.517

520
316

408

843

682

538

904

938

1065

1151

1178

:496

15::

286

378

405

429

434

451

4S6

479

541

906

94:

:067

::52

803
322
920
1482
493
1535

814
365
948
1488
591
1546

499
1001
1524
596

568
1004
1533
605

569
1007
1541
615

583
1059
1552
848

609
1157

644
1185

64S
1200

646
12,2

666
1232

5,1
1261

Sao
>200

900

922

1009

106:

1075

1149

682

688

9s0

1202

1214

1234

1268

1293

1484

1543

488
982
995
996
965
511
1326
641

012
1011
1508
1589
967
S86
1415
650

936
1586

1100

1124

11?6

:582

968
600
1522
843

969
678
1566
B64

971
847

973
862

974
869

992
917

1591
946

871

891

918

1013

1056

PT
,,1
,U
Pul
PULL
PUSH
:?
R1O
R] 1

588

280
172
807
809
962
311
1259
497

ml

MB
W1O

.27
.27
383

389

.650

413
244
816
818
963
496
1276
607

.568

57

.-:,:

10s4
1:55

}:83

.661
17
199
411
924
962

.198
.205
923
.934
.965
.962

.929

.;!
.583
.:;
237
1313
.558
1071
.102(
,1564
776
1046
,24
656
.:085
:,6
.606
.2,
767
:317
235
:04:
:222
:217
129s
.2,
::31
:C29
.27
1051
;:7
:3
158
7::
.27
:9
.27
235
985
.27
.27
.2,
.27
.2,
.27
1391
137e
.2,
.27
731
779
.27
.27
979
980
:24
.,,

;34
:C27
:9
.27

1161
.1318
562
.1129

.116S

778
.1049
1121
e39
300
616
169
.1320
.237
,1052
.1226
,:~~~
.1208
.:033
:C34
.1053
109
.883
.1578
400
720
.:~~2
37
239
1104
44
199

-::93
.1380
886
706
748
790
652
660
,1137
,998
.126
220
736
.]~,*
.977

888

.1119

513

.605

926

1105

.7,0

:C36

.1038

415

824

874

.529
729

738

753

.884

1128

1473

762

771

780

;;6

46
297
:164
800
220

:;3
1168
801
222

48
446
1468
812
261

60
82?
1469
1198
263

58

977

1227

1228

1229

1230

887
707
749
837
655

708
150
838
1116

7:0
752
1209
1117

715
?s7
1235

716
758
1237

39
247
1137
45
209

257
.737

;:9
1138

103
915

110
978

17$
9eo

147o

1507

15?8

733
775

734
776

124
166

125
767

726
768

728
770

1126

1163

1502

1581

233
984

735
777

265

.>7

.;7

999

1594

.1390
137?
.27
.24
1023
1236
.702
144

1166

1165
.,78

;:7
643
298
1562
60
<.
.,, . ...
.....

GPD*TE

107
18

.834
1109
1015
925

975
.1113
.1114
,1107

290

351

1592

993
1158
1406
1188

991
1567
1531

1S38

863

810

930

963

u,

248
1068
850
512

173
.58

.185

1102

.-.----

5
W1
;: E
W3

,1,
,.1

5*1
.648
.27
.27
602
.27
1556
.27
.27
17
.183
107

.ross-refere..e
654
111
,641
224
1557
337

700
373
1289
226

703
3?7
1554
230

791
1024

835

884

1114

243

314

321

329

331

890

894

1477

1478

347

494

529

530

532

543

544

548

558

S60

1501

220
709

.710

59

1540

1550

1555

60

.-

AN460

An RDS

using the MC68HC05E0

Peter Topping,
MCU Applications Group,
Motorola Ltd., East Kilbnde

INTRODUHION

.,,
.$

~>?.::i,.

\i:>/\*.

The Radio Data System (RDS) adds a digital data capabili~ to the FM VHF transmlss~&s,~n band II (87.5
J.*$
r
to 108 MHz). This capability is in use in the UK and several other European counl[ik~, the Intention being
t.!t:+,}!x
that most of western Europe will eventually adopt it, The. specification
i>,~~fbd in EBU Technical
.,7
<.\.,,1,,
Document 3244 (see reference 1),
~
..~.,.~
,!$
J!
+i.++:,..
~:t,.t:<t~$!\
To transmit the data, a sub cartier is added at 57 KHz. This sub carrier isati~~~de
modulated with a shaped
hi-phase coded signal. The sub carrier itself is suppressed to avo~S&~@ modulated cross-talk in phaseIocked loop stereo decoders and to maintain compatibility with$~~q,%rman ARI svstem which uses the
same sub carrier frequency. Information is sent in groups of,>;
,bw+~~bit blocks. Each grouP of 104 bits is
one of several types containing different information. It i~~p to the broadcaster which features are
transmitted. The only constraints are that the specified ~&~rnat*ust be adhered to and that Pl, PW and TP
should always be included. Each group contains a @*W~reWtsubset of the RDS features; tablel lists all
currently defined RDS features

Information

Program identification
Program type
Program service name
Radiotext
Clock time and date
Alternative frequencies
Traffic announcement
Traffic program
Music/speech switch
Decoder Identification
Programme item number
Enhanced other networks
Transparent data channel
In-house data
,,,... .Y.$,,
,~,
).$ J*;?

\y$$

,..

The retrieval of data is carried out by a demodulator circuit which generates clock and data signals that can
be used by a microprocessor. Suitable demodulators which can perform this function include S~7579T,
TDA7330, U2231 and RDS hybrids. The block diagram of a typical application is shown in figure 1. The
microprocessor, in this case an MC68HC05E0, decodes the RDS data using the clock and data signals from
one of these demodulators and sends selected data to dot-matrix display modules.

61

.\!.,: ~

is,,
*.,l$;$..

,Jc

This application incorporates an alarm clock which, if permanently powered, ~~:w used to switch on the
radio supplying the RDS data, at the required alarm time, There is a second ~rrnbutput
intended to sound
output active. The control
an alarm. This output is cancelled when any key is pressed, leaving th$g~c~~ftil
output could be used to switch the power supply of the radio or the +di@stage. [f an audio mute is used,
,: :.i,,,..-$
RDS information can be updated even when the radio is off, Alt~#tlv&y
the decoder can be used simply
to display RDS data with its power being supplied from the ra,~o afid manually switched on and off.
.+:s~,,.

This application supports Pl, PTY, PS, RT, CT, TP, TA: ~S, Dl, PIN and EON (see table 1). These features
facilitate permanent display of the 8-digit sta~ion n~me (PS) and time (CT) and, on request, can display
program type (PTY), radiotext data (RT) am~?~ status of the other RDS features. EON data can be
displayed, but the retuning features asso$i%@~ith
AF and EON are not supported as there is no capability
to control the tuned frequency. In a car@@o~,&ONdata would be used to switch the radio to a station which
is broadcasting local traffic inform~~~~.~~ AF data to tune the radio to the strongest signal carving the
, ~.,,,.~y{+,)$,~
:+.
.tc !..
selected service.
.,,, *J~},
.;V4, ~$

PI is a two byte number wM~~hXlti8ntifies the country, coverage area and service, It can be used by the
,W..
control microprocessor QMR?&@tnormally intended for display. A change in PI code causes the initialisation
of all RDS data as it indj&~@
a.... ~\ h,.,.. that the radio has been retuned. This application also facilitates the display of
the current PI code$~, 1
i:,~~,
.,,,
PTY is a $bit d~~~r
which indicates the current program type, At present 16 of these types are defined.
Examples ~\,&$;no
programme type, Current affairs and Pop music, although the actual syntax
which is+~~~ayed is determined by the software of the controlling microprocessor, In this example PTY
can ,Q~~$~byed on request, Table 2 shows the display used for each PTY code.
.}.
,1.,

?:,,,\:].

t~& IS$he eight character name of the station and is permanently

>.:.,,l$t!)+..+i

displayed (except in the standby mode).

.i$,.

~~~~ is radiotext and constitutes a string of up to 64 characters which give additional information regarding
* the sewice or programme currently being transmitted. In this application, RT is displayed on request on the
1Migit dot-matrix displays using scrolling.

62

Table 2. ~

Display

Pm

o
1
2
3
4
5
6
7
8
9

10
11
12
13
14
15
l&31

Types

No program type
News
Current affairs
Information
Sport
Education
Drama
Culture
Science
Varied
Pop music
Rock music
Easy listening

for a 1&character
space.

scrolling display the software reduce~~l se~tiences of two or more spaces to a single
*J\,,
,,,,,,,,
.\;..<<~~
;f
,1,,\,+\ .$:
CT data is transmitted every minute on the minu~g<wp?ovides
a very accurate clock, traceable to national
standards. The (Modified Julian) date and local t~e variation are also transmitted. Time is permanently
displayed. In standby mode (see below) t~ das is displayed instead of the PS name. The MJD number,
which is the form in which the date is (e&ived, can also be displayed. The microprocessor converts this
+>..
number into day-of-week, day-of-mo~~~;;~,onth and year.
~\>$,il
:}. ~

AF would be used by a car rad@W~>@;ne to the strongest signal carrying the selected service. AF data,
:/k;,.$:$t
along with TDC and INH, is @lIQq@ In this application.

, :~$\\::?,:;,<

TA and TP are flags. T~~i;&~~#f the transmitter normally carries traffic information and TA is set If a traffic
announcement is in,~r~:~ss. The combination, TA=l and TP=O, is used to indicate that EON data is being
used to supply inf~rm~tlon on other networks including traffic announcements. The status of these flags
can be displayQ@&tie
combination, TA=TP=l, is brought out to a pin and can be used to control a LED
or external ~ar~de.
An example of this could be to demute the radio or switch from cassette when a
traffic a~@~@cernent is taking place.
.*,,,.li:.:i:,,i:~
MS ~~&4ngle
~j@~@ent
.,:% $:,)>

bit indicating either music or speech and is intended to be used to make a tone or volume

to a radios audio stage. The MS bit is displayed on request.

.,i~,?Q#coder information (Dl) constitutes four bits indicating the type of transmission (mono, stereo, binaural,
~,\,
,.),
..)j.},
~.,.!
)~!etc.). It is not currently in use in the UK but can be displayed as a number between O and 15.
,).
$?,,<,,+,,
,,i:,,r\
Programme item number or PIN is used to identify the programme currently being broadcast. The format
i...
is a 2-byte number which includes the scheduled time and date (day-of-month)
of the start of the
programme. It can be displayed as four hexadecimal digits or fully decoded to dayaf-month and time.

63

EON (Enhanced Other Networks) replaces the older ON format, If type 14 groups are used to provide EON
data then ~pe 3 groups (ON) will not be used (table 6 shows the currently defined group types). Type 14A
groups are used to send information about other networks, The PS name and principal frequencv of up to
11 other networks can be displaved. TVpe 14B groups are intended to be used to switch to traffic
announcements in a radio in which the microprocessor can control the tuned frequency.

64

Table 3. Syndromes
Block

Syndrome

A
B

2
3

Mnary
11 1101
11 1101
10 0101
11 1100
01 0101

c
c

Hex
1000
0100
1100
1100
1000

$03, $D8
$03, $D4
$02, S5C
$03,$CC
$01,$58

st.,
\
.,.,..
.)\i.
,::;~~..,?,i.
*.,>
,.y,,~,i:>,.ii.
,.
This syndrome test has to take place after each bit is received. The test inspects the last 26 bi~%.r$~~~d,
until a valid syndrome is found. In this application, only syndrome A is accepted durin~!t~%~$~bv-bit
svndrome check and the data is used only after four valid syndromes have been acquired. Ki@k&complex
,:.,T$.,:.
,. ,,;:.
algorithm could allow all syndromes to be accepted during initial synchronisation and re~ulr~$!ess than four
valid syndromes before thedata isused. This can reduce thetime taken toacquireitheWcode,
which is
,,..+,
.,:
also included in block 3 of type B groups, but increases the likelihood that raow~.data,
91vin9 a valid
svndrome, will be used in error. The bit rate is 1187.5 Hz so the control mi@b~}&~essor hasa lot to do
during this initial synchronisation. Once the first valid svndrome has been@~&#s ubsequents Yndrome
checks need be done only after the next 26 bits have been receiveda?~?bis
is when the next valid
syndrome would be expected. If it is not found, then the bit-bv-bit sV~.@$onis~tion check is re-started. Once
consecutive A, B, C(or C')and DsVndromes have been detected,.@ ~~~~?$te grou Phasbeen acquired and
~~
, .. ~w
the data can be used.
~{~~~<
.:\2q\,*
4

Four bits in block 2 determine the group type. Block 2 al:,~:con~@insTP and PW data. The use of the other
bits in blocks 2,3and4depends
upon thegroup Wp:x~~@blockl
alwavscontains
theplcode.
Table7
shows the structures of the group types which are ~n~$d in this application.
.::,,,
~
+.>.
...
.\,,.l,,+.~
JN:;:::*3,,$

,,

,;,..;,

,$lhCUIT
,,

Figure 3 shows the circuit diagram. As,di?@rent demodulator devices can be used, the circuitry for the
demodulator is not shown. The clo$~%~$he
demodulator interrupts the microprocessor on each POsitive
edge. At this time a data bit is av&&19 Bnd is read orr bit 2 of Port B. Both an LCD and a VFD module are
:t\ ~>.,
,...
If the LCD module is not connected, a pull-down resistor should
shown but normally only one ,~!~~~sed.
be connected to bit 7 of p~~~J%s the microprocessor uses this bit to check that the controller in the
module is ready to recei~~~~+c$mmand. If this bit is left open circuit, it may cause the software to hang up.
Alternatively the LCD ~iv~software
could be removed, allowing the use of port C for other purposes
.,:., ,:,.:~>.~~
Wth more l/O qwaI@~; additions to the software would allow access to the other control bits intended for
controlling exf&~,a~hardware. These include the MS bit, DI data (4 bits) and PIN (match with current time
and date)xf~pey tbuld be brought out to port pins in a manner similar to that used for the TA=TP=l signal.
The unu~$~~~~rt A and D pins could also be used for this purpose but in this application thev were used
duri~~%~g
by the EOBUG monitor (reference 2). The application could make use of the Port A and D Pins,
if. f~~~ing
was done on a development svstem which did not have this limitation.

.$ii~,. ,,>
,.+y ,:,,,.
,.r
, ~,-

65

LCD Module

,*
4
J-

21

23

0 PB1 P83
45

20

I
63

Al 2

65

Al 1

A1o M
A9 =
A,7

o
D7
m

27

Al 2

55

46

53

23

24

Al 1
A1o
A9

OIVF

A7

,0

AO

D7

,,

22
14

la

E
&

25A6

62

Al

C,

21

4
28

CSROM

Zxlm

PEO
PEI

Slwp at Alarm

42

PE2 43

Al&m Enable
3
Alarm Otipm

xoop

SO~ARE
The complete sofware is listed. The reset routine (START) sets up the 1/0 ports including the enabling of
some of the special functions available on port D. These signals (Al 5, Al 4, RW and the P02 clock) were
used during debug. The pins are not used in the final application, This also applies to all the port A pins which
are configured as outputs. External interrupts are enabled on positive edges so that the RDS clock can
interrupt the microprocessor when each data bit is available. Timer B runs as a real-time clock with interrup~,
every 125 ms. Correct operation of this clock in the absence of an RDS signal requires that a 4. 194,,@~?*>
c~stal be used (the trimmer on pin 6 should be a~usted for accurate timekeeping). Timer As pre-~taier~??
set up to divide by 64; this causes the idle loop to cycle at 64 Hz. The reset routine also initiali~@$?&&CD
,.~~Q.
module (the display shows Mon O inv 00:00 until a valid group 4A is rece[ved), clears the R~~+wn@ calls a
. ,,. s:.{
!t~<. ,.
subroutine (INITD) to initialise the RAM locations used for displaying data,
,t,,~*+,\.
-$.$..,$
Lines 114-118 and 193-208 are commented
EOBUG monitor (reference 2),

out as they are only relevant when jQe%~~&ing using the

.+
!):,-.>,.
.,)
,..>,,
,+~.:~..l.,.jte
,:,. <*:<$,
The idle loop (IDLE) regularly checks the local keyboard for a keypress, com~,x~~e
current time with the
alarm time and performs other tim~dependent functions related to the di~~y<~~dules
and the sleep timer.
*.j.?i
The keyboard software (KBD) scans the 4-key matrix for a keypress~~l~~ 16ms. If the same key is held
pressed for 3 successive scans, it acts on this key function by,@8?~& the relevant subroutine (ALARM,
ONOFF, SLEEP or RDS). This software also controls the rep~~~~,$f
the SLEEP and RDS keys. This rate
is set at 6Hz (after an initial 750ms delay) when the key:J,,@r&$&ed to change the alarm time and 1Hz for
their normal function. The other kevs do not reDeat if h@W..down. Table 4 shows the functions available in

Alarm

Sleep

RDS

mode alarm

mode sleep (On)

mode alarm ON
k

RT
Pm
PI
TMP
PIN(h)
PIN(d)
MJD
MS/D I
EON 1

I
EON 11

inc. hr/min

67

The On/Off key uses the subroutine ONOFF to toggle between ON and standby. A port pin (3, PORTE) can
be used to control the power to the VHF radio and/or other external hardware. In standby mode, with the
alarm disabled, the time and date are displayed. If the alarm is enabled, the alarm time is d;splayed. In the
ON mode the time is displayed along with the current RDS PS-name. Table 5 shows these display formats.
Table
Di~iay

#
:.,

,,$

5. Display formats

mode

Format

The Alarm key calls t~~~~b?outine ALARM which displays the current alarm status. A second press
changes the alarm @#&*S?atus. When the alarm is armed, the alarm time is displayed. In this mode the
On/Off key can ,~e~~~~ to select either hours or minutes findicated by flashing) and the Sleep and RDS
keys used to ~~:~rneht and decrement the settings. If the alarm has triggered then the first press of any
key cancel$ itl?walarm
display has one of the WO alarm formats shown in table 5 according to whether
or not tp:>~$k~ is armed. As all the keys have a special function in the alarm mode the only way to exit
this ~~,~;~$$to wait for a timeout. If no keys are pressed for 5 seconds, the mode returns to normal.
~i,% ..ik~,
,J~~4$,@6P key controls the sleeP timer. If the decoder is in the standby mode the first press of Sleep
$i*ches
it on and initialises the sleep time to 60 minutes. When the sleep timer is running, this is indicated
~P+,r~~ a flashing decimal point in the right-most character of the display modules. Subsequent presses of the
$ Sleep key decrement the time remaining by 5 minutes. When the sleep time has elapsed, the decoder
returns to standby, In the alarm set-up mode this key decrements the alarm time,

68

The RDS key uses subroutine RDS to step through the various RDS data which can be displayed. Holding
down this key steps through the displays at 1Hz. The displays are RT (scrolling), PTY, Pi, TmP, PIN (hex),
PIN (decoded), MJD, MS~l and EON (11 networks) as shown in table 5. In the alarm set-up mode this key
increments the alarm time.
The timer interrupt routine (TINTB) updates the RT scrolling pointers (DISP1 and DISP2). These pointers are
incremented regularly whether or not an RT display is active, In this wav, the software can be easil$~,l
converted to using a 2-line LCD module in which the top line is the normal displav of PS-name and tirn$,@Q:N:.
the lower line a permanent display of scrolling RT. The timer interrupt also decrements the sleep t:@r&&
updates the RAM locations used to store hours, minutes, seconds and eighth-seconds. Al~@&:@?ta
(except date and time) is cleared by this routine if no valid RDS data is detected for a period,$$,~&J#@onds.
::/,!.

.J:

.?**

Hardware interrupts are vectored to jump to SDATA where serial data i,sf~~~f$ed from the RDS
demodulator. The clock edge causes an interrupt and the first instruction re~+,$h~tiata
into the carry bit
of the condition code register. The bit is shifted into a 4-byte RAM regi:~~~$f~~he
matrix multiplication
performed. The state of flag 0, STAT2, determines if the multiplication is to~&place
after every bit or only
after all 26 bits have arrived. The multiplication is performed using t@x,QR
. ~+,..l,,. (instructionsfor every bit (two
are required as the 1@bit syndrome requires two bytes). As the t~~we
matrix (see figure 2) is the unity
matrix, the first 10 bits are transferred directly into the syndrom:~,~
locations (SYN). This, the omission
....
of any EOR #$00 instructions, the reordering of the bits an&ig> use of the index register for tempora~
storage help to reduce the length of inline code in this rtitne.
fie routine could be shortened by using a
loop but this would incur an unacceptable
penal~W~~ $Xecution time. Microprocessors
with two
accumulators would find this task a lot simpler a~$:~yjgker but an MC68HC05E0, at half its maximum
speed, can easily perform the calculation in the r~~~~
time.
\ ?~.
,:.
After the multiplication has been perfor~~d the resultant 1&bit number is compared with the allowed
syndromes (see table 3). The variable L<:~f@qprds the current block level. It is initially zero but incremented
each time a valid syndrome is found.i~w~t
is zero only syndrome A is accepted, if this is found then
, ~ l\?
syndrome B is expected 26 bits.+~~y.~% when LEV is one only syndrome B is accepted. If an invalid
syndrome is found LEV is clear,e~~~~<vndrome
confidence level CONF is decremented and the interrupt
ended.
When a valid syndrom,y~ ~&&~d, CONF in increased by 4 and the 16 data bits saved in the relevant bvtes
of TM PGRP. If the ys~g%wdrome is type D then a complete group has been received and all 8 bytes are
locations at GROUP. This double buffer means that the data in GROUP can be
transferred to th<~a,~g,$~
used while int~<u~s are overwriting TMPGRP with new data.
.s.+
.,,..
,,
The con<&
@level CONF is used to decide what should be done if the data becomes unreliable due to a
poor ,&b.I~~~~tO the receiver. When the first valid syndrome is found it is initialised to 42. Subsequent valid
sv~~~SS
Increment it by four and invalid ones decrement it by 1. If CONF falls below 41, then it is
.i.~%ed
that synchronisation has been lost and a bit-by-bit re-synchronisation is carried out. If it falls below
{ 10~ We signal is deemed unacceptable and the displays are re-initialised. The confidence level is not
,~~~~~#remented
bv the detection of a valid syndrome if it is higher than 56.
,,,>,,..
>,,.>

69

GROUPS

HANDLED

If a complete group has been received the data can be processed. The buffering used would allow this to
be done outside the interrupt but in this case there is sufficient time to do it within the interrupt. The PI
code is checked to see if it has changed. If it has changed the displavs are initialised. In an application using
the AF capability of RDS, more use would be made of the PI code.

,.<*.:+,&,
,$
Group types 1A and 1B ~~nr$~n the same data except for the repetition of the PI code in type 1B. The PIN
data is recovered a#$~&8
in RAM. This is intended for future use to control external hardware, for
example a tape f,ecowf.
This would facilitate the unattended recording of a pre-selected program. At
present this a~~a~ion
simplv allows the display of PIN data both in its raw hexadecimal form and fullv
decoded to,@~~*fionth
and time. Full use of PIN data would require continuousiv comparing the PIN daVof-montk@~,tlMe
with the current davaf-month and time enabling an 1/0 pin to be switched when there
is a rn,g*@.~.
,,:,.
.. :\,\\
-wc&$A
+)t,:
;$,~,>~
,
.~y ~$ data from blocks 3 and 4 is written to RAM according to the address included in block 2, There are
~;four address bits and four ASCII encoded bvtes giving the possibility of 64 characters. [f the Text WB fiag
changes state, the RT area in RAM is cleared, indicating that the message has changed. Group 2B is not
handled as it is rarely, if ever, used.

70

Group

4A

Two of the more complex tasks to be performed are required by the CT calculations for group 4A. These
are for the local time difference and the conversion of the MJD number into a recognizable date,
The broadcast time is Universal Coordinated Time (UTC), effectively the same as GMT. Time differences
from UTC, including summer (daylight saving) time, are sent as an offset of up to +/- 12 hours in half-ho~r,
$)&>;:$.,
increments.
,,,::! .,.~,\
., .?,:: t
The software includes 4-function, 9-digit integral BCD arithmetic which is used to decode the dat@$J@*he
~..
,,~.
. ;:til.
MJD number using the formulae:
~, .!,.,,,
,,..
$..,
,t
,.r:- ,,>),,,!
,?
,..)+
.
= int[(MJD-1 5078.2)/365.25]
Y
~.s.::,,F*.
*...7,1~:$,..
~
,.,;:
.*J,
$
M
= int[(MJD-1 4956.1 -int(Yx365.25) )/30.6001 ]
.,,., ,>\!
Ddy

MJD-14956-int(Yx365

If M=140r M=15,
then K=l;
else K=O

Group

Year

Y+K

Month

M-1-12K

.25) lnt(Mx30.6001)

~N~k *
i,,I\\.~1
...
,~,
,),,J....\.,.
,, , .$$.
.,.
.
.:
,,
~,..,&
.t.}ii:f>..$,$
.*.,.:~J\~
.
...~,
~.~
~,,.Itt.
~ .,
.,,.,;J*
$,X,
,$
,\,*,,!.,.
.:??:,
p :t.~+,
,$ .;<~::.~,?~Jf
,;,.
* ,.

,,.., ..:,.S
*}Q,\$,
,
~:~&i,

14A

This group contains EON data. A large amount of inf~~%~tl~h can be sent using this group, and it can take
up to two minutes for all the data to arrive after ,,~.i
th&~M$
has been retuned. Th]s application saves the PI
,*,)$,L
code, PS name and principal frequencv of up tO*~$;Vtetworks although more networks, each with many
frequencies, and other data (e g PTY(ON), Pl~$ON~, TA(ON) etc.) may be sent. Table 5 shows the format
of the EON dlsplav. All the Information sQ@~+nISreal data from the Black HIII transmitter in central Scotland,
....\ ,,*
\~,.3>T
.,<:
>.,$,:*\,,*
Displays
.t ,,*J]..
,,<\~
:T$?.,,*.
The software drives both a p~,~~~~~&~ module (based on an HD44780 driver with or without an HD441 00)
and a serial VFD module (baWti,Q*an MSC7128 driver) to give a choice of display tvpes. The displavs show
the same data (w[thin ~~Pt@~%tions of their character ROMS).
,?.
,,,.. *
The displav routin@~~~B~ is executed In the Idle loop if flag 3, STAT2 IS set. It is set every 125ms by timer
B interrupts, lf,,4J@~&.$~AT2 IS set, the display IS initlallsed, indicating no valld RDS data. The LCD module
is then updat&#..wl~h new data. Each time anvthing IS written to the module, the subroutine WAIT is used
.::X>:,!.
before th<fh~ite YSexecuted: th]s checks that the controller In the module is not busv. This IS Indicated bv
a low om,h%j$$so bit 7 on port C should have a pull-down resstor to satisfy this condition if an LCD module
.<. :~
is nat %i~g used.
~$.,
~:i~,
s?!
, ~$,
. >$
,.\~.
>;i,,
~.x],
...
,,...,
\. .T,,,,
\, ~~
~\
*,::~>.
%~
. t
~~.~~,h>,,,<
,~,
J,$
?I:p,li$
~,:,
:\*.
~$+.,.,,
>*

71

Table 7.
Block

Group

O and

15B

PS name

PI

chck

hck P
code
(as block 2 for 1561

chck
PI
GrouD

PIN data

c
hck A

or

code

15-11

day-of-month

10-6.

PI

chck

code

2A

chck
D

hour

5-0.

Group

minute

RT

2 ASCII

chck

characters

LI

PI
Group

15-1.

hck B

4A

MJD

EON
code:
10:
Group

14A

9-5:
4:
34:

(144)

O : hour (4)

code

O-3

TP flag
PW

code

hck

5-9:

TP (On) flag

1 @l

usage

1 215

code

15-12:
11-6:

hour (34)
mtnute

(90)

offset

sense

4-0:

offset

(44)

chck
D

information

: PS
chck

4:AF

chck

AF (map)

1 : not used
: not imp,

chck
PI (On)

The listing is shown for use with a divide by 8 multiplexing

an HD44780 and an HD441OO.

LCD module. This module will normally contain

If a divide by 16 module (HD44780 only) IS to be used then line 1294 should be replaced by line 1293 and
line 1371 commented out to Include the execution of the code on lines 1379 to 1392.

The different display formats are selected by checking the various flags and the relevant routine executed.
The normal display permanently shows PS name and time. As the Iocatlons in RAM used for hour:$~~:.
minutes contain binary numbers they are converted to BCD before being written to the relevant @$$$m
DISP. Once all 16 bytes in DISP have been loaded, a loop is used to send the data to the LCD ,rn~~e$

.\.,

y .,,.{*
, fi.o.>
,$
The VFD routine sends the same data as is shown on the LCD module to the serial VFD mo~{~, T&8 display
driver used has a different character set from the standard ASCI I set used by the LC~X~l#.
The table
VTAB is used to convert ASCII data into the required character in the VFD module. The~#ll
table INITF is
used to send the required inltialisatlon bytes to the VFD module. This module doe&@&,require
a busy check
..~,,,s>
.>5
but does require a delay between successive bytes. This is satisfied by th~, wt~ tbop within the serial
.Rr.,. ii.
,::),,,,?,,.>
output loop VFDF.
~~..,ih .,j
Alarm

functions
.\:F, .o.~l~..

The alarm time can be entered as described above. If the alarm,s&,@fi@bled (alarm time disptayed on first
press of the ALARM key, and permanently d]splayed n standQy &~Q#) then, at the alarm time, the auxiliary
control line WIII go high. This can be used to control externafi$$dware,
for example to switch on the VHF
radio supplying the RDS data. If the auxllia~ Ilne IS alrea,~htgh fdecoder fully on or on via the sleep timer),
then it simply stays high. The operation of the slee~~t$mebts not affected if blt O of port E is high, If this
>,\\,
1/0 line is low at the alarm time, then the sleep tl~?$~~,$$tlvated for an hour. This takes place whether the
decoder was previously on, off, or running the si~~:,~mer, and has the effect of switching the auxiliary line
low an hour after the alarm time, regardless of ,+1>
itsEonditlon prior to the alarm.
,,,
At the alarm time the alarm output wilt.~$~$e activated (active low) as long as it is enabled by bit 1 of port
E being held low. This IS Intended t?~~&~.an alarm sounder. When this output is active, a press of any key
cancels it until the next alarm. T,h~~:@an&ellatlondoes not affect the auxlllary output.
,$,:~~,y,,,>..
,... s
! 1.
..::::.:,<.,,:.>
~+:~tl,,
**.+
REFERENCES

,i.

,+,

APPENDIX

(listing) follows

73

0001
0002
0003
::::
0006
07

;8
.: -..
.

....
0025
0026
0021

0009
0030

MI

,,..

FIUTIC,J

....
0081
0082
0083
0084
9085
0086
008?

00.9

Ooa

0088
00%9
0090
0091
0092
0093
0094
0095
0096
0097
0098
rfi~~

OOcb

0100

74

0112 400
0113
0114
0115
0116
0117
0118
0119
0120
0121
0122
::::

.SW9
~lR)
T m
T m
.Swm

(k.;

:[-(,
!1

JW

srm
,MTA
Srm
:,%:

:K
m,
w

E5Er mmP
~m- A lWWm
:=W;LV &mwm&Pr

($~~~Iw
[!0403mlm
(,W USU,, 50,06 Wffi
(:0409mm
[m Usw. :Odoc mlw

....................................................
m... r.,,,,. *tp r<!ms
...................................................,

0157
0158 =02. ,6
0159 d30 cd
0160 e033 cd
0161 e036 cd
0162 439 4
0>63 43. &
.,<. .> ..

30
&
&
&
&
&
>.

65
.6
.6
.6
,6

0168 .046 lf
0169 .047
C170 .048
0171 44. 26 fa
0172
0173 ,04 c
0176 44.
0175
0116 .0s1
0171 454
017B ~56
0119 .059
0180 45.
0181 45=
0182 ,061
0183 e06d
0184 466
0185 .069
0186
0187
0188
0189
0190
0191
0192
0193

2d

75

DE.W;
DE-W)
UE-%;
Dc.m)
DE-W)

................ ...... ............................,,.

Id 1,

Iw,

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ,,.

.
0.
0.
07
01 09
03
C5
C3 02
c]

....................................................

0313
0314

0315

Kwbti

0316
0317
0318

.-C

.rt,

. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

0119 .111 .6 20
0320 .113 .. 02
0321 .11s 48
0322 ,116 ad <0
0323 e118 ,. 08
0324 elk b7 01
0325 ellc M 01
0326 lle .5 30
0327 .120 26 07
0328 .122 5.
0329 .123 26 tO
0330 .12s 3f .9
0331 .121 20 0.
0332
0333 .129 b6 01
0334 e12b ,4 fo
0335 e12d b> .9
0336 e12f 27 06
0337 e131 bl .9
0338 133 3f =
0339 el]$ 1. =
0340 ,131 b6 =
0341 .139 09 m 04
0342 .13. al b
0343 elle 20 08
0344 eldO al 03
0345 142 25 29
0346 1d4 27 b
0367 .146 al 30
0348 .148 22 06
0349 .14. b6 .9
3350 ,14. 27 19
0351 .14* 98
0352 e14f 8,
0353
0356 e150 h6 .9
03s5 .152 al 50
0356 .154 27 0,
0357 ,156 .1 90
03S8 .153 26 Of
0359 .1% Ob cb 0,
0360 ,1% 18 a
0361 e15f ]f m
0362 ,161 b6 .9
0363 .163 27 02
0364 ,165 99
0365 .166 81
0366 ,167 lb m
0367 ,169 19 ~
0368 e16b 3f m
0369 elti 98
0370 .16. 81
0371
0>72
037,
0314

$Co
1$08
M~B
mmB

:$30
L1

W2

..........................+..<
f..,
.........................................*..........
.... s
Bwu,

0375
0376
,,77
ij78 e16f 24
0379 .111 b6
0380 .1?3 .1
03B1 ,175 2?
0382 ,177 .1
0383 .179 27
~??~ e:7b 0.

,,

.....................

:$20
t2

e key

Iu,

26
.9
50
07
90
03
m 19

!>

0388 .182 bl .9
0389 ,184 21 Ob
0390 e186 .1 .1
0391 ,189 27 0<
0392 e18b
0393 e18c ;:
0394
0395
3,6

=Iti
e18e
.18,

5.
5.
,

.4

K,yhti

;q

table

.. .. .. . . .. . . .. .. .. . . .. . . . . .. .. .. .. . . .. .. . . .. .. . . . . ..

mm

77

0411
0418
0419
042D
0421
0422
0423 ,1.8 05 Od 4.
0426 <I- 07 cb Oh
0425 .1* 09 & 04
0,26 lbl 19 &
0427 lm 20 09
0628 elti >8 &
0429 elb7 20 05
0430 .ti9 d .8 0.
0431 lk 16 &
0432 lk lb cb
0433 ,1,0 .6 19
043, .,<2 b7 77
0,35 ,1.4 10 cb
0636 =><6 91
0,37
0438
0639
0440
0461
0442
0463
0464 .,<7 05 0, 2d
0445 .IU 07 & 1.
0446 .l& 09 cb 19
0447 eldO 0. & Ob
0648 ela la &
0,49 .]& 1. cb
0450 elm ,6 50
0651 .Idg bl 77
0452 eL& 10 cb
0453 =1* 8:
065,
0455 el& 0< cb 0,
0456 ,1,1 lb <b
0457 ,1.3 20 f2
0458 .1= ld cb
0,59 ,1., 20 ,,
0460
0661
0462
0463
0464
066S
0466
0467 ,1.9 cd ,8 0.
0668 elw 13 cb
0469 .1- 06 04 03
0470 elfl 16 04
0471 ,1F3 El
047> ,1?4 >, 0,
0,?3 e1C6 81
0,74 .>f, ,4 04
0,75 elf9 8,
0476
0411
Odla
0(79
048c
0081
0482
0481 elf. 05 06 <a
0684 ,,Fd Ob CL 03
0485 dOO cc & 79
0486 S03 04 cb LO
0487 &06 02 <b 06
0488 &09 .6 3<
0489 dob b? 78
0490 4M
12 cb
0491 ~Of & .8 0.
0492 412 16 cb
0693 tild 20 08
0494 a,6 b6 18
0495 d18 .0 C5
b, 78
0696 41.
0497 &,. 2b *
0498 .21, .6 19
0499 ,220 b, 77
0500 *22 10 cb
0501 ,226 20 ,b
0502
0503
0504
0505
,,

.......................................................

0531
053t
0535
0536
,,,
0538
0539
0540
0541
0542
0543
0544
0545
0546
0547
0548
0549
0550
0551
0552
0553
0554
0555
:::;

1,,.,-,
-,,
slam
t...
.
. . .. . . . . . . .. .. . . .. . . .. .. .. .. . . .. . . .. . .. . . . .. .. .. .. . . .. .
455
0.
&58
b6
~Sa
al
25e 24
e25e 3,
&bO 20
.262
3f
da
20
.266
M
~68
.1
*6.
24
-6c
k
46,
.6
e270 b7
&72
10
tild
81
.215
3E
417
20

cb 0.
W
3b
04
9d
0.
9d
08
9.
17
09
9.
50
17
cb

PIW

~Sm
%
Ms
1X

6, ~AT,,
Mm
1s9
m
ml N

lHR

W-W
m.
M1m

9,
fs

...........+...........................................

0558
0559
0560
0561
0562
0563
0564
0565
0566
0567
0568
0569
0510
0571
0572
0513
0576
057S
0S76
0577
0578
0579
0580
0581
0582
0583
058,
0585
0586
058?

&79
dlc
47,
.280
,282
&84
.286
.288
,28.
&8c
a28e
&90
.292
a94
&96
497
,299
e29b

0.
Id
21
h
20
86
bl
20
%
27
3.
,6
b7
10
81
,6
b7
20

& 0.
M
0,
%
0.
&
9d
06
9.
09
9.
50
71
cb

0595
0596
0597
0598
0599
D600
0601
0602
9603
0604
0605
0606
0601
0608
0609
0610
0611
0612
0613
,1.

4.9
&&
,2*
&af
&bl
da
&M
~b7
~b9
&k
&ti
~bf
&cl
&c3
&c5
&c7
d<9
dtid
.>. $

3C
a,
25
15
lb
16
3.
3a
3.
b6
al
25
15
3f
3f
3r
If
35
3[
,}

.0
58
02
.9
0,
.9
99
,7
79
19
50
10
m
8.
8E
90
9:
92
w
..

slam

B.,-,

tim

.. .............. ...... ........ ...... .......... .... .....

=~
T=

6, =AT4
Mm

~lN

DE
Bm

.
%
Im

;2
mm
D=

;&
WIN
T5SD
Am
Am

% SD
~
=m

;;~T
O. =AT4

11
9.
E3

7.9

....................................................

0645
0646
0647
0648
O,*,

mS
-t

cl-k
a bit

,.,,rrqt

(IR))
calculate
wrd,-

-d

... . . . . . . . . . . . . .. . .. . . .. .. . . . .. . . . . .. . .. .. . . .. .. . . ..

04
39
39
39
39
01
3.
27
,7
80

01 00
7d
7.
7b
7.
.9 Ob
94
C3
0,

.6
bl
bb
.4
g,
b6
b,
01
b6
an
b?
9f
;;

la
94
7.
03

0)
b6
a8
b,

7b
97
?d 0.
97
b
97

=ATA

mY 1

m-

MT.1

S03

:: 03
97
;: ;: 0.

s13

*SB
1

1$03

S13

1.mT.3,*23
m. 1
I$8F
m. 1

S23

2 WT.l. S43
m. I
ISk?
m,. 1

,s03

;: 02
ld 0.
91
e
97

1
D,mT.3,
1

m.

.8 a7
b, 97

09
b6
,8
bl

m.

w.

03
7d 0.
97
8f
9,

:26
BIT
MT
,3

$$02
-3

;! 01
b
b6
.8
~

7d 0.
97
*
97
01

0707 &72
0708 &7d
0709 *76
0710 477

.8
9,
Od
bb
,S
bl
9f
.8

n,!

.,

. .. . . .
,

.,7.

0712 d7a Of
0713 d7d b6
0714 d7f .8
011s d81 b?
07?6 ,383 gf
07;7 &84 .6
,!. .... .?
0719 ~81 01

. .. . .. ,

7d 0.
97
01
97

,53

1s03

s63

6.MT.3, s73
m. 1
,Sbi
w. ,

S13

7 ST. ,.S02
m. 1
I$=
m. 1

02
,d 0.
97
&
97

1$02
..,,:.<..

.s

01
7C 0,

so2

S12

. . ... ...
w.

01S4 .3.8 Od
075$ e3cb Lb
0756 .3.+ .8
0757 e3.f b7
>758 e3d1 Yf
0159 e>~ .8
,. .,,. .,

7< 0.
97
6.
97
01

S62

1s01
7. DAT.2.
1
#SW
3m. 1

m.
0763 e3& ;8 &
0764 e3& b7 97
0765 4&
9f
766 .,., ., ,

S3J

80

0775
0776
0717
0778
0119
0700
0781
0782
0783
0784
0785
0786
0781
0788
0789
0790
07?1
07?2
0791
0794
0795
0196
0797
0198
0799
0800
0801
0802
0803
080,
0805
0806
0807
0808
0809
0810
0811
0812
0813
0814
0815
0816
0817
0818
0819
0820
0821
0822
0823
0824
0825

.....................................................
....................................................
dfl
d f3
d fs
e3f7
d f9
.3 tb
A fd
e3ff
&ol

1982 -N
1883 =49=

b6
al
27
al
27
.1
27
3E

93
03
5d
02
22
01
10
93

b6
al
26
b6
.1
26
2Q

97
d8
31
96
03
2b
53

b6
al
26
M
.1
26
20

97

d4
23
96
03
Id
45

06 80 0.
bb 91
;)
;
b6 96
al 02
20 0.
b6
al
26
b6
al
27

97
c.
06
96
03
h

....................................................

Sa
26

m
f9

TXL!

81

....................................................
WL=
PI .*.
1.1>.= 11
w+,
,
.hlg&

lrtlt

All

blink

1s $4.

bl<=k

3s ,.>t

. . .... ......... .... ..... .. .... ...... .. .. .... .. .... ..

m

m,

c-m

k,

GW,.
P1.1

%
BQ

PI wm

-Iws

ZP

D1.

&P.
L
P1.1
C=
cLm
4 SAT2

SA~

mslm
N
~~~L1=

PI

%
%
JSR

.s,
%~

cm

m.
D~SPUY

, . . . . .. . . .. . .. .. .. .. .. .. .. .. .. . . .. . . .. .. . . . . .. . .. . ..
wt.
All

Ati
e6d6
4d0
da
.,dc

b6
.6
27
al
27

4*
.,.0
&&

al f8
27 10
20 57

m
blek

a,d ,,
2. A,

,,<.

blwk

(QW

158,

88
f8
Oa
08
06

82

DATA

....................................................
.
.
R-=.
grmP 0 & 158 (D1 6 WS1

0971
0974
0975
0976
.,,

....................................................

ii78

0979 afd b6 89
0980 Aff ad 03
0981 &Ol ?7
0982 402 b6 89
0983 504 M 40
0984 606 5d
0985 ,507 26 07
0986 ,509 11 bO
0987 &Ob dd
0988 =S0. 27 02
0989 .50. 10 bO
0990 610 .3 01
0991 .512 26 07
0992 e51d 13 bO
0991 616 td
0996 .517 21 02
0995 .519 12 M
0996 &U
a] 02
0997 e51d 26 07
0998 51f 15 M
0999 621 4d
1000 e522 21 02
1001 e524 ld bO
1002 *26 .3 03
1003 e528 26 07
1004 e52a 11 bO
1005 62. 4d
1006 .52d 27 02
1007 e52f 16 bO
1008
1009 .531 11 1010 433 07 89 02
1011 .536 10 m
1012 .538 m .6 18
1013
1014
1015
1016
1011
1018
1019
1020 *3b .1 10
1021 e53d 27 04
1022 &]f al 18
;:;: .541 26 Ob
1025 643 b6
1026 &45 bl
1027 &47 b6
1028 .549 bl
1029
1030 =54b e.
1031
1012

mo

ml

m2

W3
WZ
.....

8.
91
M
92
e6 18

.-. .. .. ----...........................................

,.,

83

1065
1066
1067
1068
1069
1070
1071
1072
1013
1014
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
,089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1:0,
1102
1103
1106
1105
1106
1107
1108
1109
1,10
1,,1
1112
1113
1114
111s
11:6
11:7
1:,8
1219
1,20
112>
1122
1123
112,
1125
1126
1:2?
1128
i129
:1>0
1131
1112
113;
1134
:13s
);36
1:,7
1138
1139
:149
1141
1:02
1!43
::,4
1145
:146
:147

....................................................
.
wins.
4A
QXWP

(ff)

.. .. . . .. .. . .. . . . . . . . .. . . . . .. .. .. .. . . . . .. . . .. .. .. .. ..
*82
,584
e586

.1 40
27 03
cc A
lb

689
&8b
esnc
&6e
&90
&92
693

b6
46
.4
b7
b6
46
b7

695
691
&99
69.
&9b
6%
6%

b6 8.
36 8b
46
44
64
4,
b7 9.

&9f
6.1

b6 &
b7 76

&ti
&&
&a7
&a8
A=
&&
Ad
&af
&bl
Ab)

b6
38
49
38
49
ad
31
3f
3f
1.

mwP

4A

89
01
74
&
75

h
W
W
31
9b
9a
99
Q

wm?E

k.]

>H

d,f

f erer,ce

A,s,,m,!t

. . .. .. .. .. .. . . . .. . .. .. .. .. .. .. . . .. . . .. .. .. . . .. .. . . ..
b6 W
,8
27 Se
24 32
,,
44
44
46
97
24
b6
ao
2.
.b
3.
b,

0.
9b
1.
04
3.
9.
9b

-,.7

LT60
m

;;
9,
43
4.
2. i4
& 18
b. 9=
3d
26
3d
26
3.
3.
3.
20

76
08
75
02
74
,5
,6
2e

,<.

b7 9.
20 2.

84

1177
1178
1179
1180
1181
1182
1183 e61b al .0
1184 &ld 27 03
1185 e61f m =6 ~
1186
118? 622 3f 95
1188 424 k 95
1189 426 d6 01 45
1190 429 bl 8.
1191 e62b 26 69
1192 A2d d6 01 46
1193 430 bl w
1194 .632 26 62
1195
1196
,,97
1198
1199
1200
1201 634 b6 89
1202 .636 ad OE
1203 638 al 04
1206 Ah
24 10
1205 dk 48
1206 AM bb 95
1207 d3f 97
1208 640 b6 &
1209 A42 d7 01 47
1210 445 b6 8b
1211 447 d7 01 48
1212 ,64. 20 .<
1211
1214 44C .1 04
1215 &4e 26 34

....................................................
k=-.

gxap 14 (m)

.......................

...!

85

................................................,..,
O,wl.y

,W

..1=,,-

.. .. . . .. . . .. .. . . .. . . .. .. .. .. .. .. .. .. . . . . .. .. .. .. .. . .

1353
1354
135s
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1361
1368
:;:;

el=
06 cb 05
,760 d .1 . .
e763 20 03
e7d5 cd .9
.748
e769
.74.
.74.
e750
.752
.154
,756
.759
~,~..

5f
4
14
.6
al
26
.6
cd
5<
=.

la

eb 6.
03
bl
ff
02
2d
& 65
,,

1371 e
1372
1313
1,,4
1375
.. .

WIT
2, mmD
D1,P.8, X
I Sm
CM2
: $2[1
CXCK

MmE
MTA
m
A m

Sm.

IT W

86

....................................................

1394
1395
1396
1397
,,,,

m.
....................................................

1410

1811 .190 Sf
1412 .391 bf
ld13 ,193 ,6
1414 .795 .1
1,. -,., ,.
.6

.7
bl
ff
>
2d
?f

-. .-

ad

:4..
d 09
.3 10
26 e
16 01
11 01
El

..

08

24
12
,,
10
13
5a
26
..

~L

D1S3

44

02
01
01
01
01

D1S4
1, W~B
0, mm
O,mm
1, ~W8

D1S4

f2
40

.7.5 .0 Of bO 00 80 00
90

D1S3

...

MTA
cm,

HIGH

KM WTA
-m
m

ml?F

. . . . .. .

.......

..............
.6
.7., b7
e7i0 b7
e?a b7
e7a .6
e7d6 03
e7@
ela

20
bl
k
co
e

cb 05
05 99 02
b7 cc

D1SP.1O
nIsP.11
XIN
CED
DISP.:3
DISP.:d
:$20
2,m8,
m
1$3A
,s,.;2

87

1500
1501
1502
1503
1504
1505
1506 *19
1507 =8*
1508 e81d
1509 .81f
1510 =820
1511 .822
1512 .823
1513 425
1514 M27
1515 &29
1516 &2c
1511 .82.
15,8 ,830
1519 .832
1520 e81,
1521 e816
1522 =838
1523 e8k
1524
152S
1S26
1521
1528
1529
1530
1531 e83b
1S32 &M
1533 e83f
1516 e841
1535 ,843
1536 ,846
1537 .848
1538 ,84.
1539 .*4<
15,0 .84.
;~;
.850
1541
,544
1545
1546
1547
1548
15,9
1550
,55,
1552
,553
1554
1555
1556
1557
1558
1559
1560
1561
1562

.................. ........... . ... ..............

m

d,wl.y

.. ... . . . . . .. . . .. .. .. . . .. .. .. . .
&
.3
25
SE
.6
42
b7
3f
h
d6
h
e]
3.
3.
b6
al
25
81

XW

WD3

m
~
BW
m
w
WL
7$
W,
Wx
s.
mm
N
mw
M,
EL,
m,

.. . ..

.. . . . .. . . . .. ..

FY

:16
X,x,
8,6
%
w
M
-.x
.7
,,,,,

Ws

K>,>2

k,1,
U.r> .

....................................................
...................................... .............
k
,6
al
26
Ob
3.
3.
b6
al
22
20

,8S2
1.
.85,
20
.856
lb
,858
b?
eB5a 5f
e85b ,6
=8% .7
ee5f 5.
e860 .3
,862
26
.864
b6
,866 b7
,868
8;

mm
SK P]

FS?
m,,
m
XLP1

88

....................................................
.
,
w-,*
[ala

.mA)

d>.,la,

.. .. . . . .. . . . .. .. . . . . .. . . .. .. . . .. .. ... . .. . . ... . .. .. . .

am-

m
CED
DISP
DISP.1
WIN
~D
DISP. 2
.1,,.3
-.1.
nlsP.

X
d

,x

&P2
m
CED
,s30
m
,s20
D: SP.,1
DISP. 12
KIN
crD
DISP.14
.,s,.,5
,s20
;if18,
mF
DISP.13
1642
1643
1644
1645
,646

i647
1648
1649
16S0
1651
1652
1633
1654
1655
1656
165.
1658
,659
1660
1661
1662
1663
1664
1665
1666
1667
1668
,669
167,

e8fa
dti
e8fe
.900
.901
,903
.905
?907
.909
,90.
.90,
,910
.912
,915
,911
.919

5f
d6
.7
5.
.3
23
b6
27
cd
bf
b?
M
cd
bf
b7
81

c
bl

02

DIP1
DWP

... .
CPX
B=

Of
f6
8i
10
& 48
k
M
90
& 48
k
bf

,15
V&,

PIN

.,.., .

89

1722
1123
1725
172S
1?26
1127
1728
1729
1730
1?31
1732
;;;;
1735
1736
1737
1738
1739
1740
1741
1742
1163
17d4
,745
1746
1147
1748
17d9
1750
1151
,752
1751
1754
1155
1756
1757
1758
1159
1760
1761
1762
1761
1764
1765
1766
1767
1,68
1769
,,70
17?,
,,72
,,,3
,,,4
,775
1776
1777
1718
,779
1780
1781
1182
1783
1784
1785
1786
1787
1788
1789
1790
1191
1792
1193
1794
1195
1796
1197
1198
,799
1800
1801
1802
1803
,,,

....................................................
....................................................
.972 5f
.973 db =
e976 .7 bl
.97*
5.
.979 .1 Of
~::
;:
;;
.97f
.981
.984
.986
,988
=9%
a9U
,98f
=99,

27
A
bf
b7
b6
d
bf
b7
8>

.99>
.993
,996
,998
.999
e99b
.9W
e99f
,9,,
e9ti
e9ti
.9.4
,9.7
,9.9
9ti
.9A
e9af
=9b,
9b3
e9b5
.9b7
,9h9
e9W
e9M
e9bf
,9<,
,9,3
.9?5
e!.?
,9.9
e9cb
e9cd
e9cf
Ml
,9a
=B5
e9a
e9d9
.9*
,9e9dc
,9M
.9,1
e9Q
,9,3
,9&
,94
.Yw
,9,<
.9=
e9f1
e9f3
=9F5

5*
d6
.7
5.
a>
21
b6
2.
4,
44
44
d
d
26
a.
bf
bl
d
27
a?
26
,6
b7
.6
b.,
25
.6
b?
,6
b?
a;
26
,6
b,
.6
b7
b6
.6
k
58
49
58
49
d
bf
b7
b6
a$
&
bf
b,
8,

22

10
& 48
k
M
92
*
an
k
bf
PIW

N
bl

32

DP M2
mP2

of
t6
91
to

& 84
30
02
20
b3
M
31
2d
3,
08
?3
b5
74
h
,2
08
6,
w,
64
M
33
08
?2
b5
64
S
91
07
92

mo

eb 84
*
k
92
3f
eb 84
k
bf

W MT
:,1,9,

,:
X

,15
MU>,

90

...................................... .............

1832
1833
1836
1835
1,1,

d,w,.y

.. . . .. . . .. .. .. . . .. . .. . . .. .. .. .. . . ... . . ... . .. . . .. . . ..

,8
116

,,,
DISP.8
DISP.9
m.2,
x
DISP
W.3,
X
DIW. 1
m.4.
x
D,SP,2
m.s,
x
*,*,.
3
~.6.
x
.,,,.,
m.7,
x
DISP.5
m.8,
x
.1s,.6
m,Y,
x
DISP.1

....
1882
1883
1886
1885
1886
,887
1888
1889
1890
1891
1892
189>
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906

-80
-82
s84
s86
-88
-k
eBb
-W
-8f
-91
-93
-9S
-97
-99
S9b
-w
-9C
-.1
-3
-5
-,7
-.9
-b
-d
-af

bb
26
,6
&
b7
97
M
26
.3
26
.6
h
b7
b6
A
b7
.6
bl
b6
&
b7
bb
&
b7
81

34
02
fO
30
b
35
06
20
02
fO
3C
&
36
30
m
2,
&
37
30
bf
38
30
co

w.
w,
rcm2
:;
*sm
,s30
DISP.9
*6
*$30
DISP. Lo
*7
1$30
DISP,l,
c-8
*S3G
DISP.12
*k
DrsP.13
;l;P.14
,.
D, SP,15

91

1941
1942
1943
1944
1945
1946
1947 -e
1948 eti
1949 af2
1950 efd
1951 -E5
1952 af7
1951 ef9
1954 efb
1955 -1.
1956 &OO
1957 &02
1958
1959
1960
1961
1962
1963
1964
1965 &O>
1966 A04
1967 &07
1966 &09
1969
1970
1971
1972
1973
1914
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
,.s,
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
. ...

&k
tic
&k
*11
-13
&lS
All
61,
*1c
*1,

....................................................
slq

dimley

....................................................
Sf
d6
.7
5.
.3
23
b6
d
bf
b7
81

e
U

52

SLm,
DISP,

Of
f6
78
* M
b9
k

x
X

115
s WP

:E
DIsP.8
DISP.9

Ws h D1 d,wlay
.
.. .. .. . . .. .. .. . .. . . . .. .. .. .. . . .. . ... .. .. .. .. . .. . .. ..

5f
d6 e 62
.7 bl

mm,

D!SP, X

5.
.3 Of
23 f6
01 m 04
,6 U
b7 b7

115
rmP

0,=AT3.M1M
DISP.6

b6 M
YXn
D:SP.11
DISP.14

d eb 84
bC &
b7 bf
81

................................................,,.,,..
-e,

b,rmw

t.

mwckd

,,,

Q.

.......................................................

. ..
*1E
&l
*21
626
*2B
*2b
+2d

*2f
~]1
&33
&35
*I7
*19
A).
.

2001 do
2002 *43
2003 &5
2004 ml

. .

54

bf ad
ef

d
3.
d
.6
b7
34
36
24
ae
bf
d
.,
.

mz

Wx

<$-..$:>,

cm
k,

06

5.
et 84
0,
.6
d
al
07
30
a.
=
13
5$

JSR
N
JSR
m

CWS
R. 8
cw
14
W6

,.-,

cm

m2zm
WR
m
Wx

nx
2$.
mx
.

d w 33
3. .6
26 .0
81

....
~A

92

.>:lL

2036
2037
2038
2039
2040
2041
2042
2043
2044 &65
2045 &67
2046 &69
2047 +6b
2048
2049 &6c
2050 &be
2051 &la
2052 &72
2053 &74
2054 676
2055 678
2056 &7a
2057 &le
2058 &7f
2059 &81
2060 683
2061
2062
2063
2064
2065
2066
2067
2068 &E4
2069 &86
2070 *8
2071 &8a
2072 &8c
2073 &8e
2074 &f
2075 &91
2076 &93
2077
2078 &96
2079 &98
2080 &9a
2061
2082 &9b
2083 &M
2084 &9f
2085 ~]
2086
2087 &
2088 &]
2089 &d
2090 *5
2091 *6
2092 &,
2093 &9
2094
2095
2096
2,,1

bl 02
18 03
19 03
81
15
16
19
If
18
b6
19
b7
0.
33
17
81

03
03
03
07
03
02
03
a?
.7 t5
07
03

ad
ad
bl
ah
ad

,.
13
.7
,6
08

........................................................

5a

2. f7
b6 .7
cc & 48
28 03
ab 06
81
ab 06
29 02
.0 06
81
97
54
54
54
54
.4 01
81

93

2140
2141
2142
2143
21,4
2145
2146 +C2 20 20 41 6, ,1
M 20 2d 20 20 df
46 46 20 20
2167 s02 20 53 40 20 61
64 65 20 2d >0 20
20 20 20 20
2148 -12 20 5, 50 20 2d
30 20 20 54 4> 20
2d ?0 30 20
2!49 =22 20 50 4, de 20
6f 2, 20 2d 20 20
20 20 20 20
2150 *.32 2d ?0 20 20 74
20 61 74 20 2d 2d
2, 2d 2d 20
2151 -42 20 6d da 20 64
79 20 2d 20 20 20
20 20 20 20
2152 K52 20 53 6, 65 65
20 20 20 30 20 M
69 6, 2. 20
2153 -62 20 4d 2f 53 20
53 20 20 20 44 49
20 20 30 20
2154
21S5
21S6
215?
2158
2159
2160
2161 =72 ,d ,! ,. 54 ,5
57 65 64 54 68 75
,6 ,2 69 ~, 1 ,,
3, 75 6.
2162
2163 -87 69 .. 76

.................

..!......

D,*la,

...

72
.t
:0
!.
ha
61
70
20

WD

.*Y

..,.

.. . . . . .. .. . . .. . . .

65

::,,,,,,)

.,

,,,ss

.. .. . .. . .. . .. . .. .. .. . . .. .. . .

~..~

W,r.,

p,,,

.:..

>>6,

2166
,),7
2168
2169
2170
>,77

.,,,,..

*W,.,.-,

, -c

.. . . .

SW,

.............................. ...... ..............

nw.m.

?fi. m,

~:c.i-

94

....................................................
m .kracter
set

2191
2192
2191
2194
2195
;;;:
2198
2199
2200
2201
2202
2203
220d
2205
2206
2201
2208
2209
22:0
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
222s
2226
2227
222s
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
22,,
2245
2246
22t~
2246
q~~g

,,,.
&
&
&
&

1,
7,
7.
7.

7,
1.
7.
1,

7.
7.
,,
7.

7,
7,
7,
7,

S7E S7E.$7E,S7E
$7E $7 E.$7E, $1E
S7E. $7,, S7E. S7E
$?E $7 E, S7E,57E

&
4.2
dc6
dw

,.
7=
;e
7.

,.
7.
7,
7=

,=
7,
7.
7.

7,
7.
7e
7.

$7E $,s,
s,,.
s7,
S7E. $7E, S?E. $lE
S,E. $7E. S7S. $~E
$7E, $7E, S7E. $?E

dm
&
*6
~

7.
7.
7,
3f

?b
7,
7.
7d

7.
7,
7,
3.

7,
7.
7.
7d

S7E. S7B
$TE. $7E
$?E,S7E
$3 F.S70

*
44
Ad
&

00
04
08
7,

01
05
09
7.

02 03
06 07
7d7e
7. 7.

~e
&f2
dE6
df.

7,
Od
:1
:5

Oa
0,
12
16

Ob
of
:3
:1

tife
s02
-06
wOa

19 la lb 1.
Id 1, If 20
21 22 21 7,
7, 7= 7. Td

-O.
=12
=16
=1,
-1.
=22
-26
-2.

2250
2251
22s 2
2253
22\4
2255
2256
225?
2258
2259
2260
:26:
*2,
2262 -IQ
226>
2266
2265
2256
225?
2268
2269
2270

,.,

$7E
S7A
$7E
S7D

smh
(1.

$00.$01
S02,
S04. S85,306
s08. s09.57D
$7E. !7,
$7.,

S03
$07
$7E
*7C

0121
4567
89
.>,

S- E.SCA
SOD. SOE.
$11. $,1,
$,5. $!6,

:.
$10
$14
$!8

$An<
DEFG
:.
L3

$l?.51A.
SIB
51 D. SLE. $:F
$2:. s22. s23
$?z. $?z.37E

$,C
$?1
$?E
$70

PO.
T.
,:YZI
,,.

7. 24 25 26
27 28 29 2A
2b 2< 2d 2.
2: 30 31 32

$?A. S24. $2S

?2- S2B $29
s2B. $2c. s2n
52 F.$30 .$31

$Jb
52*
S2E
332

33 36
17 38
lb 3<
?. 7,

35
39
3d
7.

$33. $34. s35

537 $38.$39
33 B.$3C. $3D
s~c S7E. S7Z

s36
$3A
5-E
5~E

bf .,
cd et

86

0.
10
14
18

36
b
e
.,

$7A,
S7E.
S7E.
$3E

Sos
SOF
s13
!:7

.
,

,
N,,
,
.
.

abc
defg
h,]k
Lm.

,.r.
,.
:;:[

95

2304
2305
2306
2307
2308
2309
23,0
2311
23,2
2113
231,
2115
2316
2311
2318
2119
2320
232,
2322
232?
232,
2325
2326
2327
2328
2329
2330
2331
2312
2333
2334
2335
23)6
2337
2338
2339
23,3
2741
2?42
~;, ,
234,
23;5
234,
>,4,

....................................................

,!.
., .,,,

,,

, ,..-,. ,.

. . ...,.,

96

2406
2407
2408
2609
2410
2411
2412
2413
2414
2415
2416
2a17
2418
2419
;:::
2422
2423
2424
2425
2,26
2427
2428
2429
24)0
2411
2432
2433
2436
2415
2*36
2431
2d38
2439
2d40
2441
2442
2443
2444
2445
24,6
2447
2448
2469
2450
2451
2452
2453
2454
2455
2456
2457
2458
2,59
2460
2d61
2462
2*63
2466
2465
2466
2461

....................................................
D.,., -

of

ED

,>*,.

R .- P / Q. r-.,,&x
a, exit X = IR, m

,,,

W.

wd

.. . . ... . . . .. . . ... . . . . ... .. ... . . . . . .. .. ... . . . . .. . . . ..

.f07
ef09
efoc
efO.
ef10
~;:

a.
cd
. .
bf
. .
:

54
ef 86
42
A
4b
~ 2.

DTV

ef17 bf d
ef19 .,
39
,f lb cd e
2.
efle
ef2D
.f22
ef24
ef25
,E27
ef2a
eE2c
eE2e
f30

.6
b7
.,
f6
26
&
26
20
b6
bl

09
,,
39

=632
ef34
f36
ef39
=f3b
,SU
ef3f
ef41
ef43
efd5
ef68
ef4a
.:4.
et,<
,f50
ef52
ef53
ef55
eE57
ef59
ef5b
.ES
ef5f
e[61
ef63

.=
bf
&
b6
27
k
6.
20
a.
cd
. .
.6
b7
.6
,7
5.
3.
26
6f
3=
b6
al
26
.,
81

4b
d
=
6d
W
06
.1
53
ei
6b
= 33
39
08
. .
07
08

07
ef 64
f6
33
. .
al

-..

w.

DIDITS

m.

SNlns

a.
fl
08
al
al
0.
dl
54

ffis

97

.C95
ef97
ef99
ef9b
.(9.
,[.0
ae 39
de90
a. 19
d*90
.,
30
d
,f
.6 07
b7 38
d
et
b6 53
h, 73

86

07

..

5d
bf ad
.,
30
bE a.
d
fo
. . 62
tie6d
d
fo
cd ef
bf ad
. . 66
cd e

e E&
efdf
.,,,

..
b!
.,

Yw

m
m

mx

98

.s.

Q .-

a3
07

E
JSR
,s.

P .2 .R .-

1OK(%W-!5O?8
3652s00
Y
IIMID-l:>

{,50782000)
2)
078

2)/365

m
,Di

2.

,s,

,>-. .,

Sd
ad
,2

98

2r,1

2607
2608
2609
2610
2611
2612
2613
2614
2613

Y=Y.
K
H= M->-

K.,2

.. .. .. .. . ... . . . . .. .. . .. . .. . . . . .. ... . .. .. .. .. . . .. . . .. .. . . . .. . .. .. .
6f
%
70
11
04
13
%
= 90
f05a )f 6f
605, 3. 70
Eose 3. 70
f060 20 06
E062 .6 Oa
f064 b7 70
f066 If 6f
f068 3a 70
fO& 81

f04d b6

a.

f074 cd fO
f077 cd e

2646
26,7
2666
26,9
2650
2651
2652
2653
265d
2655
2656
2657
26S6
2659
.,, .
..

fo,.
fo7c
f07e
foao
f082

,f
35
3f
If
81

f083

9f

P.-Y
Q .R <-

10K.365 .25
10K.Y .36S

25

R .-

,oK(Im,,

,YR
ml

66

f06d bf ad
f06f ., 42
fo71

15

T*

fodf 27
E051 al
fos]
25
fo55 . .
f057 d

f06b

KSD
K .,=10

tiM

EOd9 b6
fOdb 21

!.

2.
,3
.6

59
%
5b
5.

.36<

05
b7 al
,6 0,
::
b3 al
26 fB
7F

6f 01
6f 02
6f 03
8,

266,
2662
2663
2664
2665
,<.

....
2667
2668
2669
2670
2671
2672
2673
26?4
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689

f098 a,
f09.
d6
fow
.7
f09f
5.
fOaO 26

09
PO .3
2f
f8

fod 81
foal
fOa5
f0a8
foa
fO&
fOd

a. 09
d6 FO cc
,7 2f
5.
26 t8
81

to.,
10M
f0b3

. .
d6
,7

09
fO &
2(

f0b5 5.
fOM
fOM

26
8,

E8

27,4
2715

99

25),

100

AN463

68HC05K0
Tony Breshn,
MCU Applications Group,
Motorola Ltd., East Klbride,

Remote Control
Scotland

The MC68HC05K0 is a low cost, low pin count

single chip microcomputer with 504 bytes of user
ROM and 32 bytes of RAM. The MC68HC05K0 is
a member of the 68 HC05K series of devices which
are available in 16-pin DI L or SOIC packages.
It uses the same CPU as the other devices in the
68HC05 family and has the same instructions and
registers. Additionally, the device has a 15-stage
multi-function
timer and 10 general purpose
hi-directional 1/0 lines. A mask dption is available
for software programmable
pull-downs on all of
the 1/0 pins and four of the pins are capable of
generating interrupts.

individual code for the key+Th~$#~y code is then

converted to a instructi~nw
that is inserted into
the transmission coQ~/~&~which,
using a defined
protocol, is transm?~~~,.~ the television receiver.
The command is<~t~$$hually transmitted as long as
the key is b~$$+h,eld down.

,#\\.~,;*A
As the,rmo~~control
is
to us$~:$tt~e power as
b~ente!ng
STOP mode
.,,<~~~ed and effectively

battery powered it needs

possible. This is achieved
when no keys are being
switches off the device.

The device is ideally suited for remote-contrQ~,fi~, ~$e micro controller comes out of STOP mode
keyboard applications because. the pulldowns a\@~k~tipon
receipt of an interrupt
request that is
generated when a key is pressed.
the interrupt
drivers on the port pins allow:
keyboards
to be built without
any ,$extd~nal
components except the keys themselyek~$here is
Remote
Control
Keyboard
no need for external pull-up or pull-d~w~<~esistors,
.v~ :*:,
or diodes for wire&OR interrupts,~g~:th~se features
The 68 HC05K0 has ten general purpose 1/0 pins.
are already designed into the ?!.
d~w~~
One of these is used for the transmission signal
. ::l.~i$;l
*,.$ ..,.::,:.,$$,
t:,..)
output leaving nine pins for the keyboard control.
This application makes W(&% Many of the device
Of these, four pins on PortA have internal interrupt
features to control a~~~~~~d
television remote
request hardware. Using these four pins as inputs
control. The app(j~~~$
could be very easily
allows key presses to be detected without any
modified
to c~,~;f~li,.$ny device with a similar
external interrupt hardware. This leaves the five
transmission &Ot~ol.
It will run on any of the K
remaining pins for outputs.
devices w~a~ut~oditication.
.<)~!,:~,,,,<
Using the internal pull-down facility and the rising
\ >*+1>!
edge interrupt request on the four inputs permits
.$$&*ote
Control Specifications
interrupts to be generated. If the five outputs are
$,l,/~.+Q>
set to logic l, so driving an input from logic O
~~h~basic purpose of a television remote control is
to logic 1 when a key is pressed, an interrupt
,,~1~~}.~ transmit a control instruction to the television.
Using
this
can
be
generated.
request
**$~: -The instruction is generated by a keystroke on the
,:~.~$
,,L..\.
~
arrangement a five by four keyboard matrix can be
remote
control
keyboard. The detection
and
+ty,.k
. .
.y~,
used. An extra four keys can be controlled if the
decoding of a key press and the transmission
Vdd line is used to drive one row of four keys to
encoding is carried out by the remote control micro
logic 1. Therefore the maximum amount of keys
controller.
controllable becomes twenty four.

101

Figure 1 Keyboard layout with

~~

$!$,
.),,

<.

~.),?,
* .i .,* .

A depressed key will set one of the input columns

to logic 1. By scanning the columns, and setting
each row output to logic O and then checking if the
inputs all become logic O, the associated row for

TradS$Yssion
,7*:ta .,,.
,.$.*,,l;~,,,:

Protocol

The tran,~m?~$on protocol in this application is that

used b~::+.the MC1 44105 IR Remote

Control
Tra#*$teY.
It uses a bina~ coded 9bit data word
the key can be determined. If rotating the logic O
through the five output pins fails to identi~ a key
?W~tWLSB
being transmitted first. Each bit of the
~$~%mitted signal is in the form of a hi-phase pulse
column, then the key must be connected to the
:+)
Vdd line. This process gives an individual code for .,.+code modulated (PCM) signal, whose bit coding is
shown in figure 2. For a transmitted O there is a
each key which is a combination of the code ~&~~
512W pause followed by a 32kHz pulse train for
the column inputs and the row outputs. Thi~.iFaR
512ps. For a transmitted 1 there is 32kHz pulse
then be decoded to an instruction that<l~<i~~$~~bd
,$<..
.>,
train followed by a 512VS pause. This gives a bit
into the output signal for transmissio~s
>, ,$j~
+::~tt,$.!;
... .,*,*
\./.
time of 1024w for all bits. This is shown as figure 2.
+$:,<*y<.
\
Figure 1 shows the layout of the~~~dard
on the
left and the scanned and tragW}%~@ codes on the
right, The keyboard layout~~:~r~rates
the various
bit-n
television controls plus q~nt~ls for TELETE~. On

~1

the left hand side~~@J*~#odes

returned from
~.>,,.}/
scanning the ke~boaFd:~bre shown in the upper
right-hand cor~,~,.@f ~ach key and the code sent fOr

I
o

transmissiomj~f
t~t key instruction are shown in
.~.>!
.+,..!
the botto~}%~r-~and corner. The 1/0 pins for each
row an~~tiwn
are also shown for each key.
~;.:~~.*.\\
*,, ,~j
,.*?~~t
*J*.;.,.
*ti:.,4!
~t-$.,

I
I

~1

1024PS

I
,,1

I
I
Figure 2 Bit coding of PCM signal

102

of

The command timing in figure 3 shows that after the

several messages. Each command begins with a

start message of nine 1s followed by the message

start transmission
the mes%ge
is continually
retransmitted at intewals of 131 ms (approximately

appropriate to the kev pressed. This message is

repeatedly transmitted
until the kev is released.
The transmission
is terminated after the kev is
released bv a end message of nine 1s.

8Hz) until the kev is released. This is shown as time

(c). The control timing shows the nine bit instruction
111001110 being transmitted starting with the LSB.
The pre+it pause is equal to two bit petiods and is
followed by a start-bit of logical 1. The PUISe tral~i~~h,
the
transition
b~t~~~s
continuous
during

A complete

transmission

command

consists

Eve~ message consists of a pre-bit, a pr~bit

pause, a start bit and nine data bits. The pre-bit and
the start-bit are alwavs logical1. The pre-bit allows
for the set up of the automatic gain control in the
receiving preamplifier.
Figure 3 gives the exact
timing relationships for the transmissions.

(a)

transmitting a logic O and a logic1. The ~%~q$ng

pulse train has a frequencv of approxj~$t@s2kHz
with a mark-t~pace ratio of one to .&[~e. *N
...~ ,.,
.,,.*,
>*+
\\$>...
The signal for transmission is ou?~t:$rough
one port
pin and is used to drive an lR~debmplifier
circuit.
i ,Q*+*:/,i
,?y.,
,,yi,
,{! , >1.},

(c)

(c)

,, ...
&9*,M

t(mlng

,., ,, .Sfi,
.:$
~;:
.:,
~!13
~t,\l:
~,,.l~:
i). ,:1:):,,
.:> Message timing
f.,+,
:T.,J.
.
J4Y. *~,
Modulation

(al = 32.8ms

stan transmission

(b) = 13.3ms

stan command

(cl = 131ms

control transmission

(d) =51 2PS

(half-blt tlmel

(e) = 1 ,024ms

(bit t!me)

(f) = 3.072ms

(pre-pulse time)

(g) = 8W
(h) = 32PS
(1)

=51

2ps

(llfcarfier)
(half-b!t t!me)

Figure 3 Grcuit timing

103

3
3 b[t t[me

161fcarrler

16/fcarrler

Remote Control Operation

Figure 4 is a flow diagram showing the operation of
the remote control on power-up or reset. After the
initial set-up of the ports as inputs or outputs the
remote control goes into STOP mode, It will remain
in STOP mode as long as the device is not reset or
a key is not pressed. When a key is pressed an
interrupt request is generated. A short time delay
makes sure that it is a true key press and not noise
and also allows time for any switching effects on
the inputs to pass prior to checking the inputs.

$
setup ports
for keybrd

STOP

Figure 4 Flow diagram

104

Hardware
The remote control circuit is shown in figure 5. The
hardware consists of the keyboard, the oscillator
and the infra-red amplifier. The oscillator can bea
crystal ora ceramic resonator with a frequency of
2MHz, The oscillator frequency is important since
the transmission timing
internal clock frequency.

is based arounda

lMHz

The infra-red amplifier uses two transistors and two

standard diodes to limit the current through the IR
diodes to approximately lA. There isa need fora
Iarge capacitor close to the [R diodes because of
the high switching current of the circuit.
$*,>>
::kyk...:$:,,
~\\~,
~, !:,,

,,:.,.
. Q
.>, ~,.*

PBo
IRQ
PAO
PA1
PM
PA3

105

Sotiare
The listing of the remote control assembler code is
contained at the end of this application note. The
first section of the listing sets up the ports prior to
going into STOP mode and waiting for a key to be
pressed, PortA bits&3 are setup as inputs with the

dependent upon the receiver software.

In this
example the eight bit instruction 14 changes the
channel to number four. In another receiver
may
interpret
the
application
the
receiver
instruction code 14 as increase volume.

~,
,,,
.,,:.{,,
,,,..
.>
The transmission
routine is entered with tea,!;
..,8:,
. ,.
instruction for transmission in keyst3. AfteC,@~- $
pr~bit
and the start-bit are transmittti~$~~
instruction byte is rotated (LSB first) int~%~e:,.drry
flag, A logic 1 is sent for transmissio?,J~.~~fla9
iS

pull-downs enabled. Bits 4-7areset

up as outputs
logic 1 as is PortB bit 0, PortB bit 1 isset-upas
output logic Oto switch off thel R amplifier before
going into STOP mode.
The next section of code named presd is the
routine pointed to by the interrupt vector and is
entered when a key IS pressed. This routine first
calls the keyboard scanning routine to determine
which key has been pressed. It then calls the
decoding routine to convert the code from the
keyboard to a code that will be accepted by the
television. The start message is then transmitted
and is followed by the instruction message, There
isthenacheck
to see ifthesame
key is still being
pressed. If it is then the instruction message is
re-transmitted until the key is released and the end
message is transmitted.

set after rotation and a logic 6~,1i~~~ent for

transmission
if the flag is cl~d~d.+~ach
bit is
: **,,:k
transmitted
as shown in fi~,$i:h.
The routines
sendO and sendl
se~~~pa~pause of 512w
followed bya 32kHz ~u~~:&&n for 512ps and a
32kHz pulse train for~$~~
followed bya 512w
pause respective~~~~
the situation when a 1
follows a 0,,4&~la
pulse train of 1024us is
required. T,~W;@breaks
in this pulse train the
sendO r~flne
checks the next bit to be
...J*
trans@~:d
to see if a double length Pulse train
mu$F~~nsmitted,

The sendl

routine then has

t~~~~& that a double length pulse train has not

As the transmission protocol requires nine data bits
:f~p
sent in the previous one and a half bit periods
andonly one byte instructions are being decodeda
bbfore sending a pulse train,
,.,,~~
flag has to be set for the ninth bit of ,Ihe
The routine burst produces the 32kHz pulse train for
transmission
routine.
For the start an@,~~@
a duration set by a count in the accumulator, As the
transmissions this flag is set to 1 to giv~~$kne
instruction time for setting the PortB bit 1 pin higher
1s message, For all instructions the nin~,t@t,,i? Oso
:$.,
.,.,,.. ..,?./t
low is five clock cycles then the minimum processor
the flag is cleared.
,<.,1.:
~!*>*5
C;,,
f,tl\.;f*
clock period is derived by dividing the minimum
The decoding routine compare~~&$~bde
from the
output state time, which is B~when
the output is
key board scan routine with ~ata:$rtay keydat. On
high, by the minimum number of clock cycles to
a match it takes the cor,%wing
element from
change this state, This gives an internal clock period
the array tvdat a~i, ~., instruction
code for
of 8~k equalling 1.6w. Addinga three cycle delay
*:; <@ :
~;:~
transmission.
will require an internal clock period of 8&/8 = 1~,
,.i
,,,~. .%*
\w
,.1,$.$
allowing a 2MHz oscillator to be used,
Thevalues oftS,~~~@Struction codes shown in the
right-hand,ww,~+o? figure 1 are specific for the
Thecode size isapproximately
300 bytes, leaving
receiver ~~;E$ation. Each receiver using the same
memory space for more features to be added to the
comm~~~ations
protocol will receive the same
controller.
t...
ni~@@tWstruction
but what the instruction does is
,,,,.>{.:}
.,);l+,,,.
t:;},.,pr:i,+=

106

Debug
On applving power to the circuit the RESET vector
will initialise the program counter at the beginning
of the software. When examining the output at
PortBbitl
with anoscilloscope
orlogicanalvserit
should be noted that when t~ing

signal bv pressinga kev the first signal out will be

the start message of nine 1s. To capture the
instruction the kev should be held down and as the
instruction will be continually re-transmitted then
the capture can be initiated at this point.

to capture the

107

tisting

0026
0027
0028
0029
0030
0031
0032
0033
0034
0035
0036
0037
0038 0000
0039 0001
0040 0004
0041 0005
0042 0008
0043 0010
0044
0045 OOeO
0046
0047 00,0
0048 00e1
0049 00e2
OOjO OOe3
0051
0052
0053
0054
0055
0056
0057
0058
0059 02o0
0069
0061 0200 9.
0C62 0201 ad
0063 0203 9.
0C64 32048e
OC65 0205 20
0056
00$7 0207 a6
0048 02Q9 b7
0069 025b b7
0070 ;23ti
b?
0071 fi2Sf.6
0072 0211 bl
0073 0213 a5
0<174 0215 b7

04

~:$,..
f.
wrta O-3 ipts
4-7 as output.
set outputs high
0-3 plldow.
prtb 0-1 OLPU:S

:0
04
05
10
03
05
01
01

set

108

portb 0 high

0078
0079
0080
0081
0082
0083
0084
0085
0086
0087
0088
0089
0090
0091
0092
0093
0094
0095
0096
0097
0098
0099
0100
0101
0102
0103
0104
0105
0106
0107
0108
0109
0110
0111
0112
0113
0114
0115
0116
0117
0118
0119
0120
0121
0122
0123
0124
0125
0126
0127
0128
0129
0130
0131
0132

.,. ..,* *...*... ...... ... . ..... ., ..,..,.**,,.* . .. ... ... .... .. . .
.
* ~E W
~D
1S D=ODED FOR WSM1SS1ON
* ~E TWSM1SS1ON
PROmCOL KQU1 R= A START =SGE
OF 9
.
.
* O=
DL~~D
BY mE KY PR=SED CODE. %1S CODE 1S
+ cO~l WALLY RE~NsM1~D
IF ~E ~Y 1S HELD ~W
W ENO ,
~ CODE OF 9 O~S ~WIW~S
~E TWSM1SS1ON
Am ~E DWICE
,
.
. REWRNS ~ S~P MODE.
.............................................................,
0218
021a
021c
021e
0220
0222
0224
0226
0228
022,
022.
022e
0230
0232
0234
0236
0238
023a
023c
023e
0240
0241
0243
0245
0247
0249
024b
024d

ad
b6
b7
ad
12
a6
b7
ad
b6
b7
13
ad
b6
a4
26
ad
b6
bl
26
..
5a
26
20
12
.6
b7
ad
80

34
el
eO
67
e3
ff
e2
71
el
e2
e3
69
00
Of
Of
16
eO
el
07
C8

presd

s.da~

Cloop
fd
e3
e3
ff
e2
4.

edtn

keyst2
keyst 1
l,dflag
#$ff
keyst 3
tr.m~t
keyst2
keyst3
1,dflag
trm, t
Prt.
#$Of
edtrn
keys.
keystl
keysc2
edtr,
#$c8
tloop
sndaqn

1,dfiag
l$ff
keyst 3
zrnmit

,,, ,,, ...,.,.,. .... ..... .. J*.f.s.q&\$&-.

. . .. . . . . . . . . . . .. . .. . .. . . . . .

. wHm A w
1s pRESSED %.@ICE
corns OUT OF sm? MODE
.
r ~E WBOARD
1S SCAN~D~)\SEE WICH W
1S PRESSED
,.,,,..................by .....................................
024e
0251
0253
0255
0257
0259
025b
025c
025e
0260
0262
0264

cd
b6
b7
a4
27
ae
9f
b4
b7
bf
b6
,4

02 f.
00
eO
Of
a7
ef

027a
027c
G27e
0280
0149 0282
0150
0284
0151 0286
0152

.4
27
b6
.4
b7
10
81

,0
el
00
00
Of

Of
06
el
3f
el
01

mic for debonce

check 1f key press
; store inpc*
mask outputs
stop if o key pressed
set one cw low
read oup. t lines
combl,e with inputs
; store .eycode
: to find rw which clears inputs
check for ipcs cleared
: mask outputs
: zero in kq-pxess r. clears l.p.ts
: check :f last rw
: set lsb to 1
: try portb ocpt if ot Frca
, Cw next p:ta Otpt row

sta KL
#$ef

stx
lda
ad
beq
lSIX
icx
b..
b,,
Lwb

b..
Ida
Sta
bsr
bset
lda
Sta
bsr
lda
sta
bclr
b.,
1da
and
be
b..
1da
cmp
bne
1dx
decx
be
bra
bsec
1da
sta
bs r
rtI

got it

1da
sta
lti
stx
bclr
1da
ad
beq
1da
and
sta
bset
rts

keyscl
keyst2
wrta
Frta
*$Of
got

,c

ryb
m trow
keystl
keyst2
#$fO
Prta
O,portb
Prta
#$Of
90tit
keyst2
#$3f
keyst2

set all ports otpts high

Set POrtb 0 Otpt low
check for ,npts cleared
mask otpts
zero I key-?re. s row clears >puts
: set indiv,d.al code .lnce last row
; store code
h?gh again
set portb CO1-

O, Por:b

109

0153
0154
0155
0156
0157
0158
0159
0160
0161
0162
0163
0164
0165
0166
0167
0168
0169
0170
0171
0172
0173
0174
0175
0176
0177
0178
0179
0180
0181
0182
0183
0184
0185
0186
0187
0188
0189
0190
0191
0192
0193
0194
019j
0196
0197
0198
0199
0200
0201
0202
0203
0204
0205
0206
0207
0208
0209

.... .. . ... ..... ... ... . ... ..... ... ...... .... ..... .... .. .... ... .
, ~E DKODE ROUT1~ USES ~
ARSAYS 1T COMPA~S ~E W

, VA~
W1~ mE ARMY ~YDAT ~
WSN h ~%H
1S m~D
WE

CORR~rnNDIK
ELEmw
IN ~
ASSAY ~DAT BEcWS
~

.
~ *SM1~D
COOE.
........................................................*..*. .
.

0287
0289
028c
028e
0290
0291
0293
0296
0298

ae
d6
bl
27
5a
26
d6
b7
81

18
03 02
el
03
f6
03 la
el

dxh
Xtel

lti
lda
Cmp
~q
decx
b..
1da
sta
Its

match

#$18
keyda t,x
keyst 2
match

; &ta array offset co zero

look at each elaet of array
-pre
with key read
; &c&e
if retch
else tq next elemmt
nom If no match fwd
; get key ctie
; store cde to tramit

mtel
cvdac,x
keyst2

... .... .. . ... ..... ... ... . .. ....,.,,, ..+.,,... . ... .. ... ... . .. . .
. ~E
.

0299
029b
029d
02a0
02a3
02a6
02a9
02ab
02ad
02af
02b1
02b3
02b5
02b7
02b8
02ba
02bd
02bf
02cl
02c3
02c5
02c7
02c9
02cb
02cc
02ce

10
ad
cd
cd
cd
cd
ad
ae
34
25
ad
20
ad
5.
26
03
ad
20
ad
ae
ad
ad
ad
5a
26
El

e3
32
02
02
02
02
24
08
e2
04
28
02
18

fc
fc
fc
fc

TmSMISS1~

PAUSE,

, ~D

A STAST

PRO=OL
BIT

N1~

START BIT ARE ~lC

WQUIRW

A PW-BIT,

DATA B17S, WERE

1,

O,dflag
sendl
&tw
&tm
&t=
data
sendl
t$08
keyst 3
&tal
=. dO
bitst
sedl

f3
e3 04
10
02
18
18
35
33
31
f7

02cf 01 e3 04

110

A PRE-BIT

~S

P~-BIT

.
.

111

112

AN479

Universal Input Voltage Range

st.,
\
Power Supply for High Resolution
Monitors with Multi-sync Capability ,,,,
.,?
*,(~$;~
.:&
/y:*

By J. P. Bruniquel,
Integrated Circuits Application
Motorola S.A,, Toulouse

,1,
, f.,,,,~$~+.
... .,7
,. s,,.,+,!..
~+,~,t .
~:f~>?.::i,
\i:>/\*.
$.~,,e.{*.
~,!.:,,C
.
,.,
$.

Lab.,

ABSTRACT
This Application Note describes an easy to build, high performance,
supply, able to work on any mains supply from 85 Vac to 265 Vat, from
synchronised on the horizontal scanning frequency for minimum scr=
monitor, thanks to the versatile, high performance, low cost ~-t
associated with the state of the art switchmode power transist~~ ~J~l
.a .,.l.*>,.:
;~
$1:..+

$::1,.,
<,k,
*..,>\.
~$
.
~,t<.$
.
.+s
a,*

loW~&
10OW FLYBACK power
4@:~k:to 100 Hz. It is automatimlly
int&erence
on a multi-sync colour
mode controller M~4602P2,
8010.

lNTRODU@ldk
..:\*J&
.. ~,~::i!,

~.::,,

The MC44602 has been specifically designed to d~~e @,h voltage bipolar transistors. Its 1A source and
1.5A sink capability, with all the protection feat~~~~$~g~clated with flyback power supplies, make it ideal
.,:,~...f,.
for this kind of application.
*,,>,>
,$.

New multksync high resolution colour ~~nitors have horizontal frequencies in the range of 31,5 kHz to
85 kHz. The switchmode power supp~a3~ciated
with these high resolution colour monitors must be
synchronized to the horizontal freq$,q~~jn
order to reduce any EM1/RFl effects visible on the screen, An
important feature for an off line R~~@rWpply is that it can be automatically adapted to any mains voltage
without any hardware adaptati~
?~.i.gff
..,,,,.1.,
.,:1
~~~s..~
..t:;x
.t,:~.
, .i.~i.
~~!i.,.
> -!i,,

,e..,,
.\.,,,?
:.,.$~~~

SPECIFICATION

Universal input volm,~~$ac

to 265 Vat, 40 Hz to 100 Hz
.,..
Output voltage~~,~$a,,,.$
,$>. J.,
135V
0.4A
,.\.. ,:
87V
0.2A
,+::. ~..,:>
,+\,:,~
25V
.* ..ii.. .J
0.8A
,,,.
it .,...tl.
16V
0.3A
,,,+..~ts
.;,f::l,
6.3V
0.8A
.~yi:J!f, .
.,~~~(iower:
100W
,,,,%i~RS,@rtcircuit protection

on all outputs

*&$$:,Overioad protection
$2,.)
,?l:,$:,>L
t,\N Minimum eficiency: 80% at full load
s
Line regulation: < + 17.
Load regulation: <t 1Yo
External synchronisation:

from 31.5 kHz to 85 kHz

Low overall cost.

113

TOPOLOGY AND MODE OF OPERAnON CHOICE

For multi output voltages at 10OW output power, the best choice is the SINGLE ENDED FLYBACK
TOPOLOGY. The best price/performance ratio is offered by a combination of a high performance current
mode controller MC44602 and a MJH1 8010 switching planar power transistor.
Depending on timebase frequency and mains voltage, the power supply works in either a discontinuous or
a continuous current mode. Continuous current mode is for low mains voltage, and discontinuous current .,~.~
mode is for high mains voltage and low power. The continuous current mode at low mains voltage lowers,$,,,.
$:fi~k
the peak current (1Peak) on the transistor and as a consequence lowers the Vce sat, the Ibl and the loss~. ,~~>
,. :%.
At high mains voltage the discontinuous current mode allows lower switch-n
losses and lower str~~s+~~~+
the high voltage output diode. When the output diode has to switch current, its losses are higher ~{f~~%
.:*~\,,, ....,,!l~~
The losses on the output diode depend on its current during conduction and current durin~,~$wg,
In
discontinuous current mode there is no current in the diode at switch on. in continuous curr~.~t~de
there
is always current in the diode at switch on and the Trr of the diode (switching Iosses/ Qepefids on this
current. To accommodate a wide range of applications, the frequency of operati~&~@
be between
.>..
.~,{>$
:.:,,
31.5 kHz and 85 kHz.
* $.,
.> Qv
.$,,,
.:),,~,$,\J$
~;
The MC44602 has a separate synchronisation input which resets the oscillator.~~~
B 5V positive pulse is
applied. Since the oscillator of the MC44602 is working at twice the outP~~~!re@encv, the Power suPPIY
will be svnchronised at half the horizontal scanning frequencv resulting i~~~$~,disturbance on the screen
IS for the power transistor
with the svnchronisation occurring onlv every two lines. Another ax,~~e
which results in fewer switching losses, as it works at half the sca~n~
#equencV. Switching losses are
directlv related to the switching frequencv, since thev are the ~~m~for
each cvcle. The higher the
frequencv, the greater the losses.
...$+ :??
,!-:~,;.
~
A zener limits the input voltage to 4,7V on the svnc. input. $~$~,f~ure 1,) The svnchronisation transformer
is a toroidal bifilar core which receives the pulses from @.#%@ base of the monitor. The svnc. Pulse wili
have 5V amplitude and about 2 @ width. The main n~d.wurce
is the high di/dt occurring at switch off,
The power SUPPIVworks at half the scanning frequenc~ ~%the impact of that disturbance is divided bv two.
~.>
NOISE

,)?

CURRENT

Figure 1 Switch off screen polution

114

~ANSFORMER

DESIGN

Since the transformer plays one of the most important parts in the performance of a flyback power supply,
due to coupling and leakage inductance, the transformer was designed around a SMT47 multislots former
and a B3 GETV 53.18.18. ferrite core from THOMSON OREGA.
The feedback from the output voltage is magnetically realised by the auxiliary winding which performs good
load, line and cross regulation, without the need for an optocoupler.
~.+,\,
...
...>:.i~+.:
,\,v$
+. .<+>
,
,..~
..:
This auxiliary windng has three main functions (see MC44602 data sheet):
- .,.c/\\\
~~,,
.,
<>,,,
-.s:!,.,?
.,,.).
>.~.~,,,
....>.
Self supplv of the MC44602
>:,~,1.
~~i).
!: ~.,,i5,b.,t!
,,.,--i,
,:t,it:.Image of output voltage for regulation
8{)*?k**~2
.>.
.?..,, ,$.:

~$,. [.\;F.

Image of output voltage for overload detection.

*:3/o%y$;\

Since the power supply will work from 85Vac to 265Vac, the minimum

:.. ~k,..:
rectit@~~tage
*R,!
~,,, . .. ~

To provide a safety margin in worst case conditions

voltage U of 90V.

(low mains-high po#\#~us

.,,.
:$$\*,$,-$~
.,

U is 8542=1 20V.

.,>!

choose a minimum

DC

..* %iw+k+
Assuming an 80% efficiency with an output power of 100 w~i~e
.::.
,\

Input Power Pin is 100/0.8=125W

.:.

The maximum primary current occurs at minimum v~k~g~~ and minimum

..>,~,
$!
is 31.5 kH~2 = 15.25 kHz.
t...i.!ie>..>t!

. ~,.,,,:*
The transformer must be calculated for 15 kHz ~;~~um frequency.

115

switching

frequency

Fs which

The technique used is the multi slot developed and widely used by OREGA THOMSON. figures 2 and 3
depict the way to couple the different windings in order to achieve a high coupling; this ensures an
acceptable magnetic feedback signal and a low leakage inductance.

Lp Pnmay winding spht

into 4 sections 0, @, 0,0
Laux Auxilia~ winting
L1 High vohage seconda~
winding split into 3 sections

o, e, o
L2secondary

winding V2

L3 3acondary winding V3 V4

Figure 3 Physical winding position

116

For multislot

construction

we chose 2.13 Turns =26 Turns

Ns 135V

=Np*(l-D)*Ws+Vf)Nin

min D

=26~0.6* 136/100+ 0.4 = 53 turns =2,5Vflurn

Np=26T

0.5mm diameter

NS135V=53T
Ns87V=35T
Ns25V=1OT
NS 16V= 7T
Ns6.3V= 3T
N Vax =7T
All wires are enamelled grade 2
Leakage inductance <270

THE CONTROLLER
f,t,p\x\i;:.j,
fixed frequency,~urrent mode controller is the heart of the flyback power
~,$
,,<:..~,
,, ..+,.,
This circuit, specially designed for Qff*@e:bnd high voltage DC-DC converter applications with bipolar
>}~::~,.,:.l~,il
transistors, offers:
~$,,,$~
,,\..,.
-!~t,>t:..-.,.!
,~<i$
,, ,,~;

Separate high current souf$~~~slnk

outputs
The MC44602 high performance,
supply.

Unique overload and sh&$~~BOit protection

Thermal protection;..

Oscillator with sy~~i~$t

Current mode,,~e:$tion

Output de~

Autorn,#C f~wforward
compensation
Latc~&~Q~M
for cycle by cycle current limiting

ty

..

%?,}}
to 500 kHz output switching

frequency

adjustment

lN~&~~&d reference undervoltage

lockouts with hysteresis

117

sWRT

ET

SWCHRO

R&

><?.*.
,-~:.j.
~.~t
~,~
!

With a peak current of 7A, the stateof-theart

good choice.

planar skhmode
,.~~. y

bipolar power transistor MJHI 8010 is a

!:,\,,,.ri
Off control: for reverse @a$Ycurrent lb2, a zener limits the reverse voltage to 4.7V, and the 2.2 pH L2
inductor limits the di~~~~werse
current to avoid Ic current crowding during Toff,

The snubbing capacitor of 330pF limits the dv/dt of the transistor at switch off; see ANE424 and AN1 080.

118

THE OUTPUT DIODES

Since the power supply can work in continuous
diodes thanks to their low TRR,
For 135V output, maximum

current mode, the output diodes need to be ULTRAFAST

reverse voltage is Vout +(max Vcc/n) n= Ns/Np

119

SCHEMATIC DIAGRAM
Vac to
5VW

RF I
IL-R
3

n,,

IHIW1O

Cli
~w

Figure

100W power supply schematic

120

I I

TeS

lResks31.5Wzto85

Condtions

KHz

Line Reg
135V
87V
25V
16V
6.3V
Load Reg
135V
Ripple
135V

~n=85Vac

lV (31 KH$~\,j:~~185KHz)

~n = 85Vac

0.3V (50 Rr

~n = 265Vac

lV

~n = 265Vac

Qwwwz)

(~~~&Z)

0.3V (85 KHz)

Efficiency

Stan&by

Mode

P input
P input
Output short circuit

Safe on a~$i~tputs

MC44602P2

MJH1801O

MR508
1N4934
1N4732
M UR460
MUR41OOE
MUR420
MUR415
1N4148
1N4731

4
2
2
2
;
A
1

121

OSCILL~RAMS

Vin 110V= F41.SkHz V-100V~iv

ld~iv

lkl~v

Wn mVac

F+kHz

Vw=200V/Dv lti~w

Wn 220Vac F45kHz VW=2WVIUV l@2~v

figure

6 Typical results obtained

122

lb-l ~v

lb=l~v

CONCLUSION

AN 1080/D Application note

M C44602 Data sheet
MJH1 8010 Data sheet
ANE424/D Application note

123

124

AN749

Prepared

by:

H. Granberg
R F Circuits

Engineering

125

Transformers E and F are intended for input applications, although A in a smaller physical form is also suitable.
In E, the windings are photo etched on double sided
copper-Kapton* (or copper-fiber~ass) laminate. The dielectric thickness is 3 reds, and tie tinding area is 0.25 in2.

Ferrite plates (#r = 2000 to 3000) are cemented on each

side to improve the low frequency response. This type
transformer in the size shown, can hande power levels to
10 W. Figure 2 shows curves for laminate thickess versus
winding area for various impdance ratios.
Impedance ratios of this transformer are not limited to
integers as 1:1, 4:1 N: L, and the dc isolated primary
and secondary have an advantage in certain circuit config
urations. This design @ find its applications in hgh
volume production or where the smd physical size is of
main concern. Table 1 shows the winding config~~ation
and measured data of the transformer shown in%~>e 3.
*,.i,
** .,>;*%

0.5

0.3

0.1

0,7

0.9

Transformer size (Area In.z)

FIGURE 2 Laminate Thickn-

vwsus Winding Area

Solder

Area

68

.,.,.,
!\\
.,

Solder
Area 70

Ferrite

30

Material

Patent AoDlied

32.0

52.9

-j 30

In the transformer shown in Figure 1F and Table 2, a

regular antenna bdun core is employed (Indiana General
F684-I or equivalent). Lines A and B each consist of two
twisted ptirs of AWG #30 enameled wire. The line
impedances are measured as 32 Ohms, which is sufficiently close to the optimum 25 Ohms calculated for 4:1
impedance ratio. (Z. =-L).
Windings a and b are wound one on top of the other,
around the center section of the bahur core. Lme c should
have an optimum Zo of 50 Ohms. It consists of one pair of
AWG #32 twisted enameled wire with the Z. measured as
62 Ohs. The bdun core has two magnetictiy isolated
toroids on wtich c is wound, divided equaUy between
each. The inductance of c should approach the combtied
tiductance of Lties a and b (Reference 4, 6).
The reactance in the 50 Ohm port (BB) should measure
a minimum of + j 200. To achieve this for a 4:1 trarr~
former, a and b should each have ttiee turns, and for a
9:1 transformer, four turns. When the windings are connected as a 9:1 configuration, the optimum Z. is 16.6
Ohms, and a larger amount of hi~ frequency compensation
til be necessary. bwer impedance lines can b; retized
with heavier wires or by twisting more than two pairs together. (e.g., four pairs of AWG #36 enameled wire

82

fo,.

FIGURE 3 Detaild Structure of Transformer Shown in Figure 1E

*Trademark of E. I. DuPont, De Nemours and Co., Inc.

127

wordd resrdt in the Z. of approximately 18 Ohms.) Detded information on the manufacture of twisted wire
transmission hes can be found in References 2,4, and 8.

Multi-Turn

Winding

Threaded
Through Tubinw

TABLE 2lm~tirn
at Teminals BW
Tran#or=
Termitid
x Shown

a, 4 Turns
o
A

C, 14.5 Turns

A,
o
b, 4 1., s

f (MHZ)

Rp (Ohms)

Xp (Ohms)

1.0

53.0

+j 1S5

2,0

52,6

+j 330

4.0

52.9

+j 430

FIGURE

4 Physi~l
(Actul

Cmtiru~tQ@~Qf

Numbw

of: K$&&t

a ?%:1 Tmnsfomer
Shown)

A similar piece of larni~,~ii<$oldered to the opposite

8.0
53.1
+j 600
ends of the tubes, an~,,.~$s,c%~perfofl is divided into two
+j 750
16.o
53.2
sections, thus isolating ~~{~efidswhere the primary connec32.0
+j 3060
53.5
tions are made .t$~~fsecondary
winding is formed by
threadin8 wir$~~good
RF inscdating properties throu~
the tubes f~ ~e~~$quired number of turns.
Figure 1A shows one of the most practical designs for
Mtho&@t~d measurements indicate ne~i@ble differhigher impedance ratios (16 and up). The low impedance
encey$irr p&formance for various wire sizes and types
winding always consists of one turn, which fimits the avafl(s<~&#~ or solid), the largest possible diameter should be
able rat;os tointegers 1,4,9 - N. Data taken of this type
, @os& for lower resistive losses. The initial permeability of
of a 16:1 transformer is shown in Table 3, whle Figure 4
.l.L~:&~:%rrite sleeves is determined by the minimum inducttiustrates the physical construction. Two tubes, 1.4 long
R~;~Rce required for the lowest frequency of operation accordand 1/4 in darneter copper or brass form the prbnary
,,*
ing to the previous formda. Typical U,is can va~ from
winding. The tubes are electricdy shorted on one end by.a
800 to 3000 depending upon the cross sectional area and
piece o; copper-clad laminate wi;h holes for the tube~$$d;$}
lowest operating frequency. Instead of the ferrite sleeves,
the tube ends are soldered to the copper fofl. ~~$~~~e
a number of toroids which may be more readily avtiable,
spacing shordd be larger than the outside dia~~fe~ b$.fie
can be stacked.
:,.>,
.~
.,$
ferrite sleeves.
t.~t\\,,..
,, .$:,.:3,~
The coupling coefficient between the primary and sec.:.,v~~~$.a
ondary is almost a lo~rithmic function of the tube diameter and length. This factor becomes more important
with very high impedance ratios such as 36:1 and up,
where higher coupling coefficients are required. The losses
in the ferrite are determined by the frequency, permeabfity
and flux density. The approximate power hsrrdlirrg capabfity cm be calculated as in Reference 4 and 6, but the
ferrite loss factor should be tken into consideration. The
Lr in M magnetic materials is inversely proportional to the
frequency, rdthough very few manufacturers give this data.
Two other variations of this transformer are shown in
Figure 5. The smtier version is suitable for input matching,
and cm handle power levels to 20 W. It employs a stackpole dual bdun ferrite core 57-1845-24B. The low impedance winding is made of 1/8 copper braid. The portions
+i 3090
54.0
...
of braid going through the ferrite are rounded, and open~$
4.0
+j 5S00
54.0
ings are made in the ends tith a pointed tool. The high
8.o
-j 300
53.9
impedance Wirrdlng is threaded through the rounded
16,0
j 760
53.1
portions of the brtid, which was uncovered in each end of
32.o
53.2
4600
the ferrite core. (%e Fi~res 4 and 5.)

128

H
T

ES

~plifiers, where a cmrstant load impedance must be main.

tained. Sometimes the inputs can be simply pardieled, and
a partiaf system faifure would not have catastropMc effects,
but wifl merely resuft in increawd input VSWR.
For very high frequencies and narrow bandwidths, the
hybrid couplers may consist of ody lengths of transmission line, such as co-afid cable. The physical lengths of the
lines should be negligible compared to the highest operating frequency to minimize the resistive losses, and to avoid
possible resonances, To increase the bandwidth and improve the isolation characteristics of the line, it is necessary
to increase the impedance for non-transmission line currents (parallel currents) without effecting its physical
length. This can be done by loading the line with magnetic
material. Ideally, this material should have a linear BH
curve, high permeability and low losses over a wide frequency range. Forhigh.frequency applications, some ferrites
offer satisfactory characteristics, making bandwidths of
four or more octaves possible.
Depending upon the balance and phase differences between the sources, the currents should be mostly cancelled
in the bafun lines. In a bafanced condition, very little power
is dissipated in the ferrite cores, and most occurring losses
will be resistive. Thus, a straight piece of transmission line
loaded tith a high permeability ferrite sleeve, wifl give
better results than a multiturn toroid arrangement with
its inherent higher distributed winding capacitance.
It is customary to design the individual amplifiers for SO
Ohm input and output impedances for testing purposes
and standardization. SO-and 2j.Ohm co-axial cable can
then be employed for the transmission lines. Twisted wire
lines should not be used at po~ver levels higher than 100
Watts average, due to their higher dielectric losses,
Variations of the basic hybrid are shown in Figure 1$
and B where both are suitable for power dividi~ h,,+
combining.
s!J,,;:.yt,
,+,,.~,.
.:,\
The bdancmg resistors are necessary to mgi{~~~~ow
VSWR in case one of the SO-Ohm point~<~~~~% a high
impedance as a result of a transistor fti$$$$~kh; an input
power splitter, neither SO-Ohm port,~e&rW~e
subjected
to a short due to the base compen$atlo~~etworks,
should
a base-emitter junction short o,~%~$~open junction wtil
result in hdf of the input ~o~~ ~&ing dissipated by the
balancing resistor, the oth$$~~}~ stfll being delivered to the
amplifier in operation<~Thi@ration
is reversed when the
hybrid is used as an ~$~ut combiner. A transistor ftilure
wdl practicrdly ~w%~s<wiuse an increase in the amplifier
output impe~~~~]mpared
to the S~hm
load impedance it can.,~$t~$rded
as an open circuit. When only one
amplifie~f~ Qmrative, hdf of its output power will be dissipatg~J~~,@, the other half being delivered to the load.
Th#&$&ning active source will still see the correct load
*
,,
i~~~~hce,
which is a basic requirement in combining
li~ar amplifiers. The resistors (R) should be of noninductive type, and rated for 25% of the totaf power, unfess
some type of automatic shutoff system is incorporated.
The degree of isolation obtainable depends upon the frequency, and the overall design of the hybrid. Typical

,> }figures for 2 to 30 MHz opemtion are 30-40 dB. Figures 8A and B show 4 port totem pole structures derived from Figures 7A and 7B. Both can be used with even
number of sources ordy, e.g. 4, 8, 16, etc. For type 8B, it
is more practical to employ toroidd multi-turn lines, rather
than the straight line alternatives, discussed earlier, The
power output with various numbers of inoperative sources
can be calculated as follows, if the phase differences are
negligible: (Reference 2)
Pout=
where:

()

~ N1
P = Total power of operative sources
N = Total number of sources
NI = Number of operative sources

Assuming the most common situation where one out of

four amplifiers will fail, 7S% of the total power of the remaining active sources will be delivered to the load.
Another type of multiport hybrid derived from Figure
7A is shown in Figure 9. It has the advantage of beirrg cap
able of interfacing with an odd number of sources or loads.
[n fact, this hybrid can be designed for any number of
ports. The optimum values of the balancing resistors will
vary according to this and dso with the number of ports
assumed to be disabled at one time. Two other power
combining arrangements are shown in Figures 10 and 11.

130

FIGURE

9 Three-Port Hvbrid Arrangement

131

FIGURE

11 Four-Fort Hy&id System

132

REFEMNCES
1.
2.

3.

4.
5.

6.

Ruthoffi
Some Broad Band Transformers, IRE,
Volume 47, August 1957,
Pizdis-Couse: Broadband Transformer Design for
RF Transistor Amplifiers, ECOM-2989, U.S. Army
Electronics Command, Fort Monmouth, New Jersey,
Jdy 1968.
Lewis: Notes on Low Impedance H.F. Broad Band
Transformer Techniques, Collins Radio Company,
November 1964.
Mbers: Design of H. F. Wideband Power Transformers, Philips Application Information
#530.
Philips Telecommunication Review, Volume 30, No.
4, pp. 137-146, November 1972.

7.

8.

9.

134

Granberg, H.: Broadband Linear Power Amplifiers

Using Push-Pull Transistors, AN-593, Motorola Semiconductor Products Inc.
Granberg, H.: Get 300 Watts PEP Linear Across 2 to
30 MHz From This Push-Pull Amplifier, EB-27,
Motorola Semiconductor Products inc.
hfferson:
Twisted Wire Transmission Line, IEEE
Transactions on Parts, Hybrids and Pactiging. Vol.
PHP-7, No. 4, December 1971.
,.
Krauss-Mien: Designing Toroidd Transformers,,~$~,,J
Wideband Performance,
Elect~@l~~:$
Optimize
*;.: *:>;,,
August 1973.
>*,:*
,,,)..~ $!,$:,+
\,:,,,F::/...
,,
,.
,..~
,:~;,, .}*.2,.
. >,*
.:,1>>
.\h.\l;*t\.
\.,:\.
..
.:

AN756
CRYSTAL SWITCHING M~HODS FOR
MC12060~C12061
OSCILWTORS
..<. ,,
>?>,
.!$.;!,,

.>

y,, \*ii
This report discussesmethods of using diodes to selectseries
,,+
.,,<p.
.,,,>,>,
! ,/+:\
resonant crystals electronically. ~rcuit designs suitable for
:$., %tt:.~-~
use with crystal frequencies from 100 kHz to 20 MHz are t ~~~~
w! $,$,::
developed with emphasis being placed on minimizing fr%: +$~~~
quency pulling. Although developed for use with :~e?<,..
MC12060 and MC12061 integrated circuit crystal oscill~~~,
the techniques will, in general, be useful in any a~pl~woh
where it is desired to electronicallyselectone ou\,,S~~~
$roup
of crystalswith a minimum of disturbance to t~~,~~~s resonant frequency of the selected crystal.
,)zj %~,
!.,

:$:?

CRYSTAL SWITCHING ~i*RODS

MC1 2060/MC1206#;QSklLLATORS
ti:v%,~
I~RODUCTION
Cmstal switchinz

1(1\ >\
~.:)
>~

can be achieved electronically fro.. ..$

the M-C12060 and MWC
12061 crystal oscillator inte~-ate$~~>:
circuits by utilizing diodes as RF switches. The sw~tch@g
is controlled by applying a fomard bias to th~,~kde associated tith the desired crystil and applyin&.,@ &erse
bias to the remaining diodes related t:~t~~:-lected
~:{,*.! , >?,.
Crystils.
In addition
to functioning
wif$<]~#~C12060/
MC12061 ICS, the switching ci$~~~~f~ms
described
here mn also be used in othe~,ja~ll~tions
where it is
desired to electronically swi&& WJ$s-resonant
crystals
tith a minimum of freque&
~lling.
Advantages
to this]~~~fhg
scheme include the
folloting
%.~,,,,.>
~
,\..J/,:\i*;
1. Eliminates t~~n~d
to run high frequency signals
though ~tisch~nlcal
switch;
2. Pemit$~~t@ing
crystals from a remote position
+~~,
~lfi~mum of disturbance to the OscillatOr,
3, ~wl-a
RF radiation
4. ~+*&
easily to electronic scanning methods;
,*: *?rates
from a single polarity, low voltage supply

~{

FOR

specified for operating frequency ranges of 100 kHz to

2.0 Mm, and 2.0 MW to 20 MHz respectively. Their
outputs consist of a si~le-endsd
TTL signal, PIW complementa~
sine wave and ECL signala. The sine wave
outputs are mpable of dtiving an ac load of 50 ohms at

135

when an external resistor is used

500 mVD. D
. (tmical)
..
to increase the cument in the emitter follower output.
The ECL and TTL outputs are capable of driving five
and ten gate loads respectively.
Series resonant crystals conneckd btween Pins 5 and
6 are required for use with these oscillators. The total
effective ac series resistance (crystal series resonance
reaistince plus my additional resistance contributed by
stitching components) between these pins must be less
than 4 k ohms for the MC 12060, and less than 155 ohms
for the MC12061.
For additional information on these ICS, see the device dati sheet and Engineering Bulletins EB59 and
EB60.
%hematic diagrams for the MC12060 and MC12061
c~stal switching circuits are given in Figures 1 and 2
respectively. The same basic technique is employed for
each IC except that an additional diode-resistor pair (D6,
R18 through D1O, R22) is incorporated for the MC12060
to offset its greater sensitivity to ac loading.
The MPN3401 PIN diode and the MSD7000 PN junction diode are used to switch the mystals. The MSD7000
was selected for use with the MC 12060 oscillator because
of its low capacitance (1.5 pF ma. for VR = O volts). It
is also an economical dual diode in the configuration
needed for this circuit.
The MPN3401 is med with the MC12061 circuit because it offers a la~ge off-to-on impedance ratio for low
dc bias currents at frequencies within the range of the
MC12061.

+~,@v&edata sheet for proper

~Lxc~.tiOnspe~ainin9
to
~~~bs
L1:

Delevan

#2500-44

N~ronim

#Wee

Miller

R1 R3, R5, R7, R9

2200

(2.5 mH nominal

inductance)

&~hf@10

R2, R4, R6, R8, R1O = 4.3 kQ

R12-

#6302

Rl,~~,R21

= 82 k~

R16 = 1 k~

R17 = 5.1 kQ

FIGURE 1 Schematic Diagram of C~stal

applications.
10 kO

= 0.1 SF

~+hta~,w D1o = 1/2 MSD7000

,$.,?:.$<>,,~~_ x5
= 0,1 MHz- 2,0 MHz

~t$bg
,,,:,>,..,

for the MC12060.

FIGURE 2 Schematic Diagram of Crystal Switching for the MC12W1

136

DC BMS REQUI~~NTS
Forward bias for the desired crystal selecting diode
(Dl, D2, D3, D4, or D5) is applied by setting the five
position switch. The bias current is primarily set by R17
and R2 (R4, R6, ~, and RIO have identical functions to
R2 when they are switched-in). The four remaining sets
of bias resistors, corresponding to the unselected crystals,
add a smaller amount of current to the fomard-biased
diode. The total fomard bias current, ID, mn be described
by the formulas
V - 2VD

ID =

R2 + {R17 II

R3+R4+R19
4
(

VD
x)
For c12060);
)}
.>~)y.

VVD

ID =
R2

+ {R17

m)

R3+R4+R13

II

,,<,.

$,
.F.! ~,tx
#F. Typical series resonant crystals$:~,$&% frequency
rmge exhibit equivalent CX capa@@ Vti,~$s of 0.012 pF
to 0.003 pF and the maximum .*$~sistince
specification for the MC12061 is 155J~%~>~gain,
the requirements of both items 1 and.< a~,& are met.

VD
For c12061)

)1

While one diode (or one diode pair in tbe case of Figure
1) is always fomard biased, the remaining diodes are
.$\?.\~,
,,.:::]:&
reverse biased to minimize their capacitance, This is ac.
DECOUPHNG
U~~$~CTED
CRYSTALS
complished with a single polarity supply by using pullup
Isolating unselea
cfistals is very imprtant
from
resistors (R12, R13, R14, R15, and R16) from the positive
the s~ndpoint of>~~h@;zing
frequency pull of the sepotintial to each switch terminal. Therefore, the cathlected crystal, $~~~~uring
that the oscillator will lock
des of the diodes comesponding to the unselected cryson a new ~atil
~equency when switched from a pretals are pulled up to approximately the supply voltage.
~ o~5one,...$,xL,<??....
.,.
Since one diode (or diode pair) is always selectid, current
Th~;$]j$$&%e for decoupling unselected crystals is to
is flowing through R17 continuously, causing a volbge
p13@,&,~~~h
impedance
in series with them. The
drop. Therefore, the anodes of the unselected diodes will
,M~%~OOOtypically has 0.72 pF of shunt capacitance CD
be negative with respect to their cathodes. When using
,{,&~refe;h Fi~re 3) at VR = 1.6 volts, and the MPN3401
a 5.0 volt su~DIY, this reverse bias will be 1.6 volts for
,,% &ically
0.75 pF at 1.2 volts of reverse bias. Since RD
the MC12060- and 1.2 volts for the MC 12061 crystal
,,
:+is efiremely large for the reverse bias condition, the reswitihing array.
+:~,~i\>i*1
suiting diode RS resistance will not be exceptionally
,,. .$:*>
~t~
large and CS will approximakly
equal CD, This series
::,.
ADDITIONAL CONSIDERATIONS
capacitive
is 30 to 300 times greater than typical values
A sticient
amount of foward current thro~h
@
of equivalent cryshl series remnant capacitance (CX).
diode selecting the desired crystil is required ,~lnsure
Therefore.
the total series equivalent
capacitance
a low value for diode resistance RD (see Fi@& ~7x>This
Cscx
i?~.)..k,,:.>i+
is important for two reasons:
(CT = ~)
decreases by only 3.270 to 0.337c re1. To minimize the effects of diode<~~@a/?%y on the
crysta~s natural series resonan~<~ency.
spectively. This, combined with a low value for Rs, main2. To minimize the total effect~~~a?nal
resistance
tains considerable coupling between the unsele~d
crysbetween Pins 5 and 6 of t~e~$~~~ated circuit.
@l (s) and the oscillator. Thus, the oscillator may remain
FrOm Figure 3 it is apd%~,$kat
as RD is made
at the previous crystil frequency, or operate at some ransmaller, XS is decreased ~
@ is increased. A large
,. ., ,.,$,.
value for CS relative ~*k* *staUs
equivalent series
capacitive
is requir,~~%, #tisfy item 1.
The impedance ~ t~fiSD7000
diode with 0.45 mA
of bias currenti@s
amical
value of 115 3 = 114.6j6 ohms at le~,j~~,~ and 115- 8 = 113.8-j16 ohms at
2 MHz; res.u~ri&&S values are respectively 0.265 #F
and O.$~~~~~~ince
typical series resonant crystils in
this fre~~&y
range exhibit equivalent series capacib~~s~l~~s,
CX, ranging from 0.024 pF to 0.012 pF,
it~
~&A satisfied, Also, since the equivalent series re-

dom frequency.
To reduce this problem, a shunt resistor (RI, R3, R5,
R7, Rg) is added to each switchi~diode
(DI, D2, D3,
D4, D5) in Figures 1 and 2. This shunt resistor estsb.
lishes a new md lower value for RD in Figure 3, which
results in a new RS value much greater than the
maximum allowable effective resistance specification for
the MC12060~C12061,
Worst-case coupling effects occur at 2 MHz for the
MC12060 and 20 MHz for the MC12061. Referring to
Figure 3: assume CD is equal to 1 pF; this gives XD =

.,.$sh~ce
of the diode is much less than the maximum
%f(~~ctive
resistance
specification
(4 k ohms) for the
y;*t*..MC12060,
item 2 is satisfied.
..,.:,?
. ~.
\.
For the MC 12061 circuit, the diode foward bias cur~,}
rent is 1.15 mA, This current is sufficient to keep the
series impedance of the MPN3401 PIN diode low. At
2 M& the impedance is nominally 22 28 = 19.4-j10
ohms and at 20 MHz 3.3 37 = 2.6-jl.98 ohms. The
resulting CS values in this case are 0.008 pF and 0.004

+
2dcD

= 79.5 k ohms at 2 MHz, and 7.95 k ohms at 20

MHz. To maximize the series equivalent resistor (Rs),

the parallel resistor RD is made equal to the reactance
XD at the highest operating frequency. For the MC 12060,
the values of RD = XD = 79.5 k ohms give RS = XS
= 39.7 k ohms. Since RS is now much greater than 4 k
ohms, the unselected crystils till be virtually isolated

137

from the oscillator. This isolation will become greater

tith a decrease in frequency.
Using the same formdas to determine the required
RD and to calculate RS and XS at 20 Mfi
for the
MC12061 results in RD = XD = 7.95 k ohms, giving a
new value of RS = XS = 3.97 k ohms. This value of RS
is much greater than 155 ohms, the m~imum
effective
resistance specification for the MC 12061. Therefore, the
oscillator will now have sficient
isolation from the un.
selected crystals to prevent erratic performance.
The values used for RI, R3, R5, R7 and R9 are 82 k
ohms, and 10 k ohms for Figures 1 and 2 respectively.

as the to~l number of crystals to be switied

is increased. However, by using the stitihing
techniques
shown in Figures 1 and 2, any frequen~ pulling in addition to that for a single crystil connected directly to
Pins 5 and 6 (i.e. pulling caused by the ICS done) is
negligible blow approximately 1 MHz for the MC12060
md 15 MW for the MC 12061. Measurements of this
additional pulling are summarized in Table 1. Typical
frequency pulling values attributable to the ICS themselves are given in Table H. In this case the devices are
operating tith a single crys@l connected directly to Pins
5 and 6 with no crystil switching circuits. The Tabl@~,
values have been taken as a reference in estib}:~~K#$
the pulling (noted in Table I) caused by the s&i~~~g
networks. men
using the crystal switchin~. ~~s~!ts,
completi pulling from the crysta~s series .w:~nt
frequency is obtiined by algebraically add~~~,,~spective
values in Tables I and 11. For exarn~!~~,~>~lute crYstal
pulling for the five crystal switc~g\~@em
when selecting the nominal 1.0 M& ~ys~$~s
approximately
0.0040
+ 0.0031
= O.O~~$ercent.
Similarly, absolute pulling for the 8.0 ~W$iC~YStal becomes 0.004
+ 0.0001
=
0.0039.$@@~fit.
Pulling effects of the
stitihing
circuits ~h~$,,%~t~ing
the 0.2 Mfi crYs@l
offset pulling caused ~,~~$ IC b give approximately zero
abwlute crysta~ ~~11.
men de:~{~~~:a% trim capacitor can be added in series with,:~$ C%:SMISand adjusted to pull the oscillator
up in ~+$~~ssy?
~,ve~
options are possible to reduce ac loading for
b~.Jhe
tiC 12060 and MC12061 crystal switching cir,:~~~s. using a higher voltage supply for the bias networks
$,wi~$,allow larger values of bias resisters to be used at
,~~~~
same diode current, resulting in reduced loading.
*%~~Also, RF decoupling chokes may be added between re~.. i sisters R2, R4, R6, RS, and RIO and capacitors C6
through C1O. ~ere
frequency pulling is not as critical,
L1 in Figure 1 maybe eliminated. These options are left
to the discretion of the user.

OSCILLATOR
AC LOADING
Oscillator ac loading must be minimized to reduce
frequency pulling and sine wave distortion. For the circui- shon in Figures 1 and 2 the ac loading is primarily
attributable to the biasing networks for the five diodes
(D1D5). All bias elements contribute to an effective ac
load, regardless of which crystal position is selected. This
occurs hause
the W signal is coupled through the parallel capacitance (Co) of the unselected crystils.
Due to a greater sensitivity to ac loading of the
MC12060, additional elements are used in the switching
networks for this detice. An RF choke, Ll, is incorporated
to minimize the loading effects of the common bias resistor, R17. In addition, a modified approach is used to
bias diodes D1 through D5. The networks (D6, R18)
through (D1O, R22) are added to minimize ac loading
and, at the same time, supply sufficient forward current
tith a 5-voIt supply. One diode (D1D5) in the MSD7000
dual diode package is used to stitch the crystil and the
second diode (D6D1O) is used for reducing ac loading.
R18 through R22 are essential to supply a small amount
of current for reverse bias of diodes D 1D5 corresponding
.!.$!:
to the unselected crystals.
Loading and therefore
frequency
pulling w~~l, ~~$
greater for higher frequency crystils and till, i~~~e

MC12060

Device
Nom;nal

c~atai

One c~stsl

frequenv

(connected

%
#&+ .,;
.,{>+.,.l*,
direct~~~i$
..
(MHz)

MC120S1

0.1

0.2

0.5

1.0

2.0

2.5

8.0

13,4

20.0

Ref.

Ref.

Ref.

Ref.

Ref.

Ref.

Ref.

Ref.

Ref.

Pins 5 and 6)
TWO crystal

switching

Five crystal
Less

than

switchi~eqyate~
~\~\*:$
one He~p~,~r&easurement
~i$\&
., ,,~~ -

(+, ~
..~;.:,\
,.~.J,
.,.

\ ~,...+..(*
-.tal .$

NO@i~(

~&stsl

limited

TABLE

T 0.0005

-0.0006

+ 0.0035

-0.004

+ 0.0008

+0.0013

+ 0.0004

-0.005

~ 0.0005

+ 0.0006

+ 0.0031

-0.018

+ 0.0008

+ 0,0001

-0.0006

-0.023

2.5

8.0

to resolution

II. Typical

of test equipment.

Frequency

~ ~vim
frequencY

one Hem

Pull In Percem

for ICS Only

MC12060

(M~z)

0.1

~~?&.~rcent
,&J>\\,.\ ~
~,~than

PUII, measurement

limited

0.2

0.5

0.0005

to resolution

-0.0012

MC120S1
1.0
-0.0040

of test eqipmet.

138

2.0
-0.03

-0.0002

-0.004

13.4

20.0

0.01

-0.05

AN-790

THERMAL

OF RF POWER TRANSISTORS

.
~1$,<
~)

Reliability

is of primary concern to many users of transistor%~%e

degree of reliability achieved is controlled by the device usf~~~@
he determines
the stress levels applied
by his circuit and en~@rn>fital
)>.,,t,
conditions. This application note will permit the device ~se~.+~~imate
transistor
power

reliability

dissipation

from

the

circuit

designers

and case temperature.

.,

point 06$~#,
, .,,::~,
4${,,,*. ,:,:,,
.?<,,,\
\..4~**!,$p
,,,

namely

*$ .

,$,,$&perat.re-Dependent
Thermal Prop:J~es
Of , ~,. ,kOfSilicon
~
and Be~Ilia
The temperaturedependent
thermal corrductivities of
silicon and beryllium oxide have been measured a<~~>~
documented by many laboratories during the last twent~~~
silicon and beryllium oxide are seen in Figures 1 through
years. Only in rare cases has this information b,$en @3 and Table 1. The temperature ranges are somewhat
seminated by semiconductor device manufacturg!%to the
wider than are necessary for typical transistor operation,
:~,,
users. The purpose of this note is to clarify.,,\.j./.>,t,,.
& cB>rect
but are shown to emphasize the wide variation in thermal
some long-standing industry-wide assump~infi$.@hYchhave
conductivit] es. Fulkerson et a13 tabulate the values for
thermal conductivity
and resistivity of silicon from
been commonly maintained about theL~<&<~@ance
and
,,,!:*
?,,,\ ,,
high temperature derating.
10OK to 1350K (see Table 1), and they find that the
.p*,\$\a,,
Most manufacturers
data smetk$include
a single
thermal resistivity of silicon as a funclion of temperature
can
be estimated by a linear approximation over the
thermal resistance number (R@J~
~$d
use
this
number
to
,,+k ix,,.x
temperature range shown.
calculate a linear derating c~pst~f out to some specified
,,,.. b.
maximum junction tem~ia~y~!;%he number cited on the
(400 - 6600K)
data sheet was prob~~~$~~>sured in the 25C to SOC
l/k =-0,1171 +2.9S4X 10-3 T(K)
(1)
range, and assume~;,,,c~tant
over the whole range of
temperatures u~to tK&.%aximum specified junction tem(600 - 10SOK)
perature, H@#+,$f$en ha.~e you calculated a Junction
(2)
I/k = -0.9609 t 4.229 X 10-3 T(K)
temperatu~,.
~ fl a data sheet, as TJ = TA + (OJC)PD?
,, .:~ky
A similar least-square fit to Fulkersons data over the
Unfort~~~ly$the
thermal resistance of silicon increases
range 200 to 700K, within lYo, is given by:
by ,,QWkfi
2SC to 200C. The thermal resistance of
Introduction
The temperature-dependent

thermal

properties

B@ c~inges by 3W., if the case temperature goes from

$J?$5%$to 100C. Knowledge of the basic physical prop.Q~;**s of the materials and the methods used to calculate
$,~.~,,~rrd measure thermal resistance will assist the device user
iy,)
.,, in transistor selection and equipment design.

(200 - 700K)
l/k= -0.2286

+3.1683

X 10-3 T(oK)

(3)

Similarly for beryllia, one can fit the data of Elstorr et

a12 over the range of 200 to 800K, with equation (4).
(200 - 8o00K)
I/k = 1,943 X 10-S T (K)17

(4)

where k is the thermal conductivity in units ofwatts/cmK.

NOTE: K= C+ 273
139

TABLE

1-

%oothd

FIGURE

Data for Thermal Conductivity and

Smootid
T
[Kl

k
(W cm-l

deg-l)

of Silicon (Ref. 1I

W = llk
(cm deg W-l)

100

7.52

150

3 .8a

0.258

200

2.44

0.410

0.133

250

1.7a

0.563

300

1,40

0.716

350

1.15

400
450

0.939

0.870
1,065

0.825

1.212

500

0.736

1.359

550

0.863

1.508

600

0.604

1.656

650
700

0.555

I .803

0.500

1,999

750

0.452

2.210

800

0.413

2.420

850

0.380

2.634

900

0.351

2.S45

950

0.327

3.055

1000

0.306

3,268

1050

0.287

3.479

1100

0.273

3.65

1150

0.261

3.82

1200

0.251

3.97

1250

0.245

4.08

1300

0.241

4.14

1350

0.239

4.18

FIGURE

Temwrature Dewndwt

lM k

ORNL ValuS
I
I

1-

Thermal Cotiuctivitv

R~istivitV of Silicon (Ref. 3)

hermal Conductivity

2 Thermal Conductivity of BeO (Rd. 21

t
1.7L

2 1.6
5
~

1.5

s
k

1,4

E
z
~
~
z
%
E

1.3
1.2
11
1,0
0.9
0,8
0.70,6
0.5
0,4

\
\

resistivity. The calculated thermal resistance of the

beryllia piece (from the previous section) is mathematically
divided into fifty layers, each with 1/50 of the total BeO
thermal resistance. The first layer at the bottom is
assumed to have its temperature at the heat-sink ambient
with its thermal resistance value corrected to the proper
temperature using the equations for the temperaturedependent resistivity, The power flux through the first
layer then leads to its temperature rise, and this new
temperature determines the thermal resistivity value for
the second layer. Its temperature rise is calculated, and so
on, until the result for the top surface of the fift~~!$~~er
gives the temperature rise above the amb~s~$k.~: the
~:):+,
... ..).
*t\\~
beryllia piece.
~~~$:.?:l,.
The same method is used for the sil~~$~$e, using the
beryllia top surface temperature :~ t~~$arting
point,
and correcting the thermal resist~~$,&~~ each of fifty
layers based upon the temper~~$~~$dk:thelayer directly
,$,\,\..
..

Geometric Factors and Themsal Resistance Calculation

The thermal resistance of most silicon RF transistors is
controlled by the bulk properties of silicon and be~llium
oxide, geometry of the heat generating (base) areas, and
the temperature of the heat sink (case). The interfaces
generally are well behaved and contribute little to the
overall total thermal resistance if the device, die and
package elements are assembled and handled properly.
Die temperature calculations are performed in two
steps. The first uses the method of Linsted and Surtey4
to calculate the temperature distribution of a die by using
a double Fourier series solution to Laplaces equation.
Figure 4 shows the device geometry and some of the
boundary conditions. Equation (5) will calculate the
temperature rise at any (x,y,z) point in the die, where
A, B, C,D, F are die and heat-generating area boundaries.
Q is the heat input in watts, and k is the thermal conductivity of the material in watts/cmK (Linsteds equation).

beneath it, until the top surface of the silicon die result
gives the calculated die temperature for that particular
case of ambient temperature and power dissipation. The
results of these calculations indicate that the thermal
resistance of a given device is not a constant number, but
is a function of the dissipated power and the ambient
(case) temperature. Another result is that the junction
temperature of a device dissipating power will rise more
than 1C for a 1C rise in ambient temperature. because
of the increase in thermal resistance. Figures 6 through 9
show the calculated thermal resistance and die temperature for several different devices as a function of ambient
temperature and power dissipation.

terms. Figure 5 shows the treatmk<tk<tiultiple

base cell
transistors. Lines of symmet~<W#tW
adjacent base cells
are considered to be adiab~~~w$boundaries
as assumed
by Lindsted. The powe~~,di~pated
is assumed to be
equally shared amon~:?~~,~~e?eral base cells. The result of
s,.
this calculation is Jhe iw~erature
rise of the silicon chip,
assuming a co$~g~~.thermal resistance for bulk silicon.
The same miodt~~i:~$ed to calculate the temperature rise
for the ~~~{%!piece,
using the silicon die area as tht
power,:,d~w~ling
area for the beryllia, again assuming
thejttk;i%~l resistance of the beryllia as a constant. The
~Q~~al? resistances of the stiicon die and the beryllia
$W~ate
are in series, so adding the above numbers gives
::k,i..>Y.value for the thermal resistance of the device at a
... .*.>\:+
?# particular temperature and a power level low enough to
avoid the effects of the temperature variations of the
respective thermal resistances,
The second step in the thermal resistance calculation
takes into account the temperature-dependent
thermal

141

FIGURE 4-

Mdel

J/

o eatossfrom

for Hmt Flow

142

FIGURE 6 Junction Temperature and Thermal

FIGURE 7 Junction Temperatum and Thermal

Raistance asa Function of Power Disipated,

Flange (Heat 8ink) Temperature

Resistanceas a Function of Power Disipated,

Flange (Heat Snkl Temperature
~

:=

\
10
o
o

I
25

FIGURE

50
WGE

\
\
I
125
75
lm
(HEAT
SINUTEMPERATURE,
C

\,

\
Iw

8 Junction Temperature and Thermal

Resistanceas a Function of Power Dissipated,

Flange [Heat Snkl Temperature
00 ~

,.1

FUNGE (HEAT
SINK)
nMPERATURE!~l

01
0

I
25

I
I
125
75
lW
50
FWNGE IHEAT
SINKI
TEM~RATURE~Cl

-,
1%

143

Experimental

Verification

Figures 10 through 12 are plots showing the correlation of measured to calculated temperature for several
geometries. under various conditions of flange temperature (30C to 150C). supply voltage, drive power, and

Die Temperature
Actual temperature measurements are made with an
infrared microscope, Barnes Eng. Co. Model W2A. This
instrument uses an iridium antimonide diode photo-detector at liquid nitrogen temperatures to measure the
infrared radiance emitted from a 1.j mil spol on the
surface being examined. The [R radiance versus temperature cur},e is calibrated by measuring the radiance at
various known temperatures monitored by a calibrated
thermocouple while the device is heated by external
means. An experimental calibration is necessary because
the rsdiarrce output of the device at a given temperature is
a function of the a~;erage ernissivity in the area seen by
the microscope. and this average ernissivity is a function
of the geometry and processing history of the device
in question. The effecti~,e emissivity depends upon the
re[ati~e amounts of melal and sdicon and the infrared
[ransparerrc} of the varying thicknesses of Si02 ~ass
in the field of view, The calibration data of radiance
versus temperature can be least-squares curve fit to an
equation of [he form T = (A)(R)b, where A and b are the
fitted constants. and R The measured radiance.
The de~ice is then powered up in its circuil. and the
radl~nce data collected point-by-point around the surface
of the silicon die. A computer program inputs the array
of rad]ance data. calculates the actual temperature from
the calibration equation. and prints a map of the temperature profile. as well as some statistical information
about the temperature distribution.
Of-Calculated

FIGURE

10 Actual

vs Calculated

FIGURE

11-

Actual vs Calculated Die Temperatur-

Geomarw
9NL
BeO~icknms M mil

AverageTemperatce
Temwmture Range

,/

Die Temperatures

.1.,

FIGURE

12 Actual vs Calculated Die Temperatu-

Geometv lKF
w ~ckness @ mil

AvHage Temperature
Temwrature Range

dc Otssipation
\

Wv++tl
M

Im

I
1%

I
I
I
2m
2W
m
CALCULATEDDIE TEMPERATURE~C]

,~

I
3W

144

lm

1*

2M
2W
w
WCUUTEO DIE TEMPERAWRE ITI

3W

MTBF as a function of power and ambient temperature.

The temperature lines are valid for any combination of
supply voltage, efficiency and drive power, by reading the
power axis as power dissipated. The MTBF lines, because
of the current dependence, have been constructed based
upon the assumptions of 12.5-volt supply and SOYoefficiency, so that the power axis should be interpreted as
output power. It is possible to use the MTBF set of lines
at other conditions. Enter the graphs by reading the
power output parameter as power dissipated, and find the
MTBF, then scale the MTBF by the ratio sq~{e of
the T = SWOcurrent to the actual current,
..:,$j..a,,,:i,.$,
~

output load magnitude and phase angles from 50 Q to

over 30:1 VSWR. The calculated temepratures seem to be
somewhat higher than measured at the higher power
levels. The calculated temperatures are based on the
calculated power dissipation, disregarding RF losses in the
actual loads and circuits.
Metal Migration and Mean Time to Faifure
The calculated/observed
temperature agreements are
seen to be close enough so that the calculated temperature
can be used as the basis for reliability calculations of Mean
Time Before Failure (MTBF) for metal migration based
upGn Blacks5 work,
~TBF = (cross section)3
[2 . f(To)

(6)

Equation (6) is the equation used for calculating metal

migration lifetime, where the cross section refers to the
conducting stripe dimensions in cm2, and 1 is the current
in the stripe in amps, f(TO) is an Arrhenius function of
the stripe material, having the form:
(7)

f(T) = B exp (- @/KT)

The material dependent parameters B and @are shown in

Table 2, K is Boltzmans constant, and T is in degrees
Kelvin. A series of graphs (Figures 13 throu@ 16) have
been constructed, one for each device, that present the
results of the calculations of device temperature and

FIGURE

13 Metal Migration

- MTBF
10,W

u
Q

145

15

75

Iw

CASE ~MPERATURE

125 WC
~C)

150

FIGURE
10,W

\\

15 Metal Migration

M\\\\

MTBF

FIGURE

17 Geometry
12.5

Geometry

Code

125C

MRF421

Al

Al

AU

MRF422

Metal

MRF428A

MRF243

5NN

v~ c [v)

50

28

Al

lKF

Code to Standard Part Cr~-Reference

MRF316

MRF453/A
MRF4551A

lWC

Geometv 9N1
Mersl!rsdon La~e Cv$tal Glassed Al
Finger ~mensios Wdth 1,0 mil
Helgh! 1.5g
BeO lhickne~m
mil
Opemting bntit(ons VCC = 12,5V
V=WA

\\
&
\

250T

\l

275-C

Wc

lW
URE PC)
FIGURE

16 Metal Migration

1W,wo

125

1%

175

MTBF
Geometv lKF
Msrahzaion LaQe Cvstal

\
I \

Glass& Al
finoer fimenslos,
Wtdth 1,0 ~;
Htight l\&&
\,50.c

~,\
..\.

MTBF28 = MTBF12,5 X 5,02

Similarly, for 50 V operation:
MTBFso=MTBF12.5

X 16

Conclusion
We have discussed the elements of thermal resistance
and metal migration lifetime with particular attention
paid to their variation with temperature as functions
of power dissipation and ambient temperature.
Graphical presentations of the results are included
which should be useful to the device user who is interested
in better reliability in his application.
References
1, G. A, Slack, Journal of Applied Physics, 35, 3462,
1964.
2. J. Elston, J. DeGoer, and Z. Mihailovic, J. Nucl.,
Mater., 11,333,334,
1964.
3. Fulkerson, Moore, Williams, Graves, and McElroy,
Phys, Rev., 167,768-780.
4, Linsted and Surtey, IEEE Transactions on Electron
Devices, ED-19, 42, 1972.
5. Black, Proc. IEEE, 57, 1587, 1969.
6, Hall, ECOM, DAAB07-70C
0164, October 1971.

146

MONOMAX APPLICATION OF
THE MC13001 MONOCHROME
TELEVISION INTEGRATED CIRCUIT ~%..
,.f:~
..x?
.
,h~.
,,.%-.>
:~.b~
~.
+
:s.?.
$:
,F...~~
+$
~k.y .,J,
.\lk
.+.:..<t,,$,.
.h
.\,
*,
.~,.+ ,,
,.
\>/*,,,
k,,
w,.
Technical Consultants:
.<?,,:.$
~~~.~
~,-{..:$
\
.*.
Cl. Tsui, Hong Kong
\ .,...+, ~
.$ ., ttw,
Peter Bissmire, Geneva
, ..(,\,,,:*,
~t!.,,..,$,)li

Lowell Kongable, Phoenix

. . ,)),l~~.,
*.\, ..,,,>
Mike McGinn, Tempe
~i:+
,.
k
t~.~;
, .,.,.*
.,?$
,
*.&s*?;k@
Prepared

by

Ben Scott

This application
white line-operated

note presents a co,~~et~ 12 black and

. .x\,,\
television receiveti~cluding
artwork for

the printed circuit board. It is inten~,d toprovide a good starting point for the first-time user.,$o~.e of the most common
,$P .?,,
pitfalls are overcome, and t~~ st$nificance of component
~} !.. >.selections and location sar~,:@&$*sed. The design has only4
factory adjustments:
H. #~~,Height,
AGC Delay, and V.
tinearity, and there are n? alignments.
Note that while t~$~discusses MC13001
tuner AGC)

(525 line, positive

therd::t~r%$alsoparts for 625 line and negative

>\+ $!
tuner AGC, i~,,e~.~b%binations.
1$.i.! {X
,*, i$>,**
$,;s(.~
$
,t:s,
S:<:t
. ,~~
,,* .\*. i,
:;* ~~.~,:,
s
,.:,~,~,~,+,,1
,1,
:i
INTRODUCTION
~:t,\,?
pation. Special attention was given to ESD (electro.
static discharge) immunity on all pins. An extremely
Monomax has been o@$~&>2rket
since mid-1981. It
was originally
devd~@w#h
a joint effort between
stable horizontal oscillator was devised.
Additional features which resulted from this desigrr
Zenith and Motorol#for>he
purpose of creating a high
effort included: a completely integrated IF and detector
performance
B:&W >biver.
It was intended for all
with no detector tuning or external filtering components,
types of mon?$~,~me receivers, including the demandan on-chip dc contrast control which permits remote loing potiab~
~&i&obile
applications,
which require
cation of the control without shielded cable, and fully
immuni~~?,~~ise,
airplane flutter and multipath
black level clamped video with blanking and beam cursigna~.f@~l&ons.
Features suggested by these requirerent limiting. The combination of system functions in the
me@&$f&q~;ded. noise filtering and canceling,
dualMonomax chip permitted some elegant 9olutiorrs which
lg~h~zontal
PLL, countdown vertical, and a flexible
would not have been practical or economically feasible
f~+&,~ly9tem.
in more conventional desi~s,
$J,}twas also required that the resulting receivers be
It is not the purpose ofthie AN to describe the overall
~,~>+ow in component
and manufacturing
co9t. To meet
Monomax chip in any greater detail than is required
f! this objective, effort was made to minimize external
for understanding receiver design decisions. The reader
components
(especially
precision components)
and
is urged to obtain a copy of the MC13001 data eheet
adjustments.
Above all, the receiver was to be reliable, so the chip
available
from Motorola Literature Distribution
or
Linear Applications.
It contains
come of the basic
was designed to operate at low voltage and low dissi-

147

rtuu n c

monomax

runcx80nem

148

olocu

uiagram

output. The audio output section is usually a Ciass B

type, operated directly from 12 Vdc. An IC combining
the sound IF, detector, and audio output is ideal in this
architecture. TDA1 190 is an example which fits well
with Monomax.
Figure 4 shows the basic power supply structure for
the ac line operated type of design. This is the most
economical
and the most common approach for B&W
television in most of the world, and it is the subject of
much of this AN. Special thought was given to this
type of set in the design of the MC13001 itself. Note
that the horizontal oscillator and driver are s~pplied
through high value resistors directly from tha,~~~{ied
power line dc (120 V). Only 4.0 mA are .B~2+
;nto
Pin 18 to power the horizontal oscillat~i$<$~$m.
The
balance of the horizontal circuit is ~Jso~/bperated
so it is fully operable from the lin$$@~ly.
The horizontal section then produces the~~~l~:~dc for the rest
of Monomax (50 mA), and fo~,$W~$%~#rs, the sound IF,
the vertical output, etc., :b%Rt $50 mA in all. This
method avoids the pr~bl~m by developing
12 Vdc
directly from the lin~~~~~~~the waste of power in a
linear approach, t,~:$f?~k
components
for a switchmode de-de convW*M @@the cost of a Iinetransformer.
As in the previ~~~~~mple,
the TDA1 190 can be used
for the enti~esou~d system, but many designers prefer
to use a @#&.A,
line operated, discrete output stage,
and o~$~:~~$ standard sound IF/detector ICs, such as
MC~3~,
@A3065 or TBA120. This removes the 12 V
sw~~y
Npple caused by loud low-frequency
audio pas+~ag>~$but costs a small audio output transformer. This
,..,:~:

application information
which will not be repeated in
this note. Also recommended
is a paper entitled
Monomax
An Approach to the One-Chip TV by
Gerald Lunn and Mike McGinn of Motorola. This can
be obtained from the proceedings of the IEEE Chicago
Spring Conference
on Consumer Electronics,
June,
1981, or from Linear Applications,
Motorola.
Monomax is not difficult to apply. A functional TV
set is virtually assured on the first try. But as anyone
closely associated with television design can attest,
there are, in every new design, a number of small but
objectional problems which stubbornly resist solution.
The receiver described here does not represent the last
word, but it is pretty close to production quality, and
it includes solutions to some of the most common
beginners problems. In the following text, an attempt
will be made to explain component value choices and
locations
in terms of problems solved or behavior
avoided, so that the future experimenter will be alerted.
THE

BASIC

DECISIONS/POWER

SUPPLY

One of the first considerations

in a new TV design is
whether the set is to be ac/dc (12 Vdc operable) or ac
line only. Monomax fits well into either, and has been
used in production designs of both types.
Figure 3 shows the architecture of an ac/dc type
with all systems operated from 12 Vdc. In this case,
the horizontal output stage is of the boost type, to
minimize horizontal deflection current and make the
yoke easier to manufacture.
The flyback transformer
contains
auxiliary
windings
which provide supply
voltages for the video output, picture tube grids, and
vertical deflection. Sometimes the boost voltage of 20

f
E

1
T
H

7 %~$,:heapproach presented in the complete receiver in

.:>
~+t$ this AN, but it could be easily changed to the single-

output

Driver

zoo

R Ex~=

(See Data Sheet)

+80V
120 Vac

Ltne

AC/DC

FIGURE

3 Basic AC/DC

149

Adautsr

Architecture

18

1001
&

MC13001

MONOMM
19

17

0.001

..-.

It is important to use good bypass techniques on all

power supplies, not only for low frequencies, but also
for RF. It is critical in prevention of faint but objectional vertical lines in the picture, caused by horizontal deflection system waveforms
getting into the
supplies. Good high-frequency
bypasses on Pins 18
and 19. with resuect to Pin 16. are essential.
THE

IF

,,.

The four stage IF in the MC13001 has 80pV sensi,#%$Y,

ity, sufficient for excellent overall performance .W~~n -
used with an ordinarv tuner and a conventi@a~k~
input bandpass netw~rk. It is recommen~,$.@~$~he
input always be used differentially
to r~%,~kk
possibility of feedback problems. The di$$*klkl
input
capacitance
decreases from its norti&l 5!Y PP, to about
2.o pF, in the top 10 dB of gain-@%<~#f
the IF. This
can be used to narrow the i,a~$i~,#C
filter, at very
weak signals, to reduce ov&alkdetected
noise, and
,:\lR?+~i*>**
$
improve picture lock.
.$
+<
If a SAW (surface acou~c:~ave)
filter is used, as in
this AN, the above b:~fi,~pa)~walking
technique cannot be used. Furt~-~qrS,
if a SAW filter is used, an
*..:1
additional
fixe.d:.~~,n-preamplifier
is needed to over,,.,.
come the 20 tq,,~~d%)oss thus imposed. Nevertheless,
this appro~+: ~~~ become increasingly
popular with
the intr~~u:~n
of low cost SAW filters, because it
elimi@@~,~& crucial and time consuming
production
ali~$~$~i.
~$~,~e is a steadily increasing supply of SAW filters
1~{+%~ marketplace,
so some criteria for choosing the
beSt one for the design are in order. Bear in mind that
all of the video selectivity is concentrated
in the tuner
and the IF input filter in this design. In a B&W receiver,
it is important to obtain a good compromise of picture
and sound quality with a single selectivity channel.

..
This me$@s*&$#ing
color and sound subcarriers low
enough %~~+old 920 kHz beat generation in the detector, fid ye%not attenuating the sound so deeply that
g~p..$ ~und
quieting is irretrievably
10st. A well~r$b%n characteristic
for achieving
this goal is as
**t~h*n
in Figure 5, taken from tuner-mixer input to
~~J**&etector. Of this, some selectivity
comes from the
circuits, but most of it is provided by the
,,.,...:, mixer-tuned
.$SAW filter.
Table I shows some available types, data normalized
to O dB picture carrier, The major difference is the
deDth of 41.25 MHz. In this regard, the Toshiba
Fl~32U, Kyocera, and the muRata parts are best for
B&W design. The mixer-tuned circuits will supply the

+10

-10 ~

I
-20 3

--

. . . .

s
s
-3D =

-w

4725
Ad, Sd

-5 0
Freq MHz
40

41

FIGURE

150

42

43

44

45

46

41

5 IF Bandpsss Characteristic

48

TABLE 1 Some Available

SAW Filters

Toshiba
Relative

Response

Kyocera

muRata

F1032B

F1032U

F1032V

F 1052

-40

-48

-45

-40

-37

-37

41.25 Sound

-12

-16

-6.5

-25

-18

-19

42.17 Color

+1.0

Peak

+4.0

+4.0

4.0

4.0

4.0

39.75 Adjacent

Picture

Insertion

LOSS

Sound

SAF45MC

-45

-48

-47

-40

-42

-38

-18

-19

-18

-21

-23

-20

45.75 Picture
47.25 Adjacent

+4.0

KAF45MR-MA

027

0
,, ~ ,$
.7

,&< ~$~

Remember that AGC loops have a largqk~~o~~tof

gain, and fast AGC 100PS, with good ai%~~~~~fiutter
performance,
are especially vulnera$J&*&**dfiection
currents. only a few millivolts on tw &~@Tines from
i>+i~
stray fields or ground 100PS can~%~~$a significant
bar in the picture. Keep the t~e#AGC
lead away
from yoke leads. The small ~&as&$apacitor
on Pin 11
further reduces this prob~~~~ad
should be placed as
close to the MC13001 #*~&Wble.
Mrmomax was des#~~d.#o that in the strong signal
region, above the*&~,
1the IF gain is held constant
while AGC ac~.up~,
the RF stage in the tuner. This
means that a,$~~~l.amount
of IF AGC range may not
be accessi,@l&.~#t~e normal implementation.
Optimum
setting,.~ff~:,$ elay pot keeps the RF section at maximm &$#mFor RF signal levels of from <10 pV to
l.QmVr~~, using 40 dB of the IF AGC range. The tuner
i**~M likely to be able to provide more than 40-46 dB of
$ ~~ii;onal
AGC, which will accommodate
signal levels
+$,$:%@to approximately
200 mVrms. This is adequate for
AGC
*:~Y,J;~,
the Monopole
antenna
applications,
but certainly
The AGC system was implemented here essentially
ts.
doesnt offer a lot to spare. Above this level, the AGC
as described
in the Data Sheet, including the AGC y
system loses control, the receiver overloads and even.
speed-up capacitor between Pins 9 and 10. This Wps
tually falls out of sync, One way to improve this, and
the AGC airplane flutter response time fast, even<~h%
pick up the remaining 6.0 dB or so of IF AGC capabil.
the signal is strong enough to move the AG&$~Wthe
ity, is to put a resistor from Pin 11 to Pin 10. The value
tuner control region. The RF AGC delay+,a~{~*%~*s one
of the resistor will be about 33 k for delay resistor values
of only 4 factory adjustments.
Ideally;&t #buld
be
shown, but will have to be tailored to the particular tuner
made with a calibrated signal leve$$:~~~cceptable
used. This can also be accomplished by a resistor from
results can be obtained with a str~n&~@ih&air
signal
Pin 9 to Pin 10. This, in fact, is the only solution in parts
of this
and a switch type attenuator.;! #Q@~ussion
providing negative tuner AGC.
adjustment is contained in A~~en#ix I.
*,>:*!$;>AS
.
.
additional slight amount of narrowing required. The
F1032V part is too wide, and F1052 is too narrow.
These are intended for color receiver architectures of
different
types. The SAW manufacturers
loading
recommendations
should be adhered to closely to
prevent ghosts (before and after the picture) caused
by capacitive
feed-through
and/or
triple transit
reflections.
At the input of the MC13001, it is important to use
good bypass capacitors on Pins 2,4 and 6 with respect
to Pin 1 of the MC13001. The best value was found to be
a straight lead, low-inductance
0.02 ~F disc ceramic
for reducing the infamous channel 6 beat. Pickup in
this area is also a possible source of vertical scan bars
in the picture caused by horizontal sweep cuments. It
is desirable to keep the SAW filter close to Pins 1, 2, 4
and 6. See the PC board layout Figure 14, Also, the IF
preamplifier must be kept compact and well grounded
to prevent feedback and oscillation with the tuner.

I
I
1

40 - _ __

IF

,
400

10
SIGNALS~ENGW

SIGNALS7RENG78 lmV]
FIGURE

6 Monomax

AGC

(mV]

FIGURE 7 Modified
AGC Cuwas
[Rasistor from Pin 11 to Pin 10)

Behavior

151

THE SYNC

SEPARATORS

H Sync

Composite
sync is stripped from noise-cancelled
video in apeakdetecting
sync separator, as shown in
Figure 2. The time constants for setting the slice level
of the detector are connected at Pin 7. As always, there
is the compromise
between optimum noise immunity
and tilting of the slice level during vertical interval.
For best horizontal separation, a short time constant
is required. There is also an AGC anti-lockup system
which responds to the voltage at Pin 7. It also requires
a short time constant. A second, longer time constant
can be diode connected to the same pin, to prevent too
much charge-up during the vertical interval,
Composite sync is subsequently integrated internally
and fed to another amplifier whose emitter is brought
out at Pin 23. Satisfactory
vetiical
sync can be
obtained (internally) by simply connecting Pin 23 to
a divider. Weak signal performance can be improved by
using an RC network on Pin 23 to make the separation
self compensating,
as in the horizontal separator. Also
.4GC from Pin 9 can be fed to Pin 23 to improve airplane
flutter vertical hold.
FLYBACK

-@g+~YHold
0.47

-0,01
&

82k

100k
100k

&-

INPUT

The only flyback pulse input to the MC13001 is at

Pin 15. It takes care of keying the AGC, blanking the
video output stage, and phase locking the horizontal
system. The Pin 15 input is a base-emitter junction,
with a reverse polarity diode for protection. The input
requirement is for a negative-going
pulse of 0.6 mA,
but it is best to choose a pulse voltage
andseries
~esis.
tor to give about -2,0 mA peak. This will make the
effective width be the pulse width near its base,
HORIZONTAL

OSCILLATOR/AFC

Monomax contains a really unique group of featurgs

in this area: dual-loop; variable-loop-gain
(bandy&~tbJ
on the first (eync) PLL; externally adjustable p.$~~~$~
in the second PLL; simple flyback pulse inpa~$,a.~~~~ing no ramp generation. These are descri~,$~~% ~dtail
in the data sheet, and will not be repeatg~m~g?d,
Shown in Figure 8(a) are the first P&$~@ponents
as presented in earlier publication,~i~n~ffi
8(b) a new
variation which has been imple~hte~ifi
this receiver.
This very simple change retai,g$,$~g,,~al time constant
on the phase detector. The iqro&rnent
is the 13 k/22 k
....
divider which sets a 5.O..W$~~~~$for the return of the
longer time constant fil~~, $nce 5.0 V is the reference
level in the oscillat~~~ i~~~so the operating voltage at
Pin 12, and at Pi~j&.,~~err in-lock, The benefit, then, is
that the 0.47 ,#,+#$,:Sht have b charge up, so theres
very little fr~,~~~
pulling during power-up or powerdom. This$:*3audible
chi~s and momenta~ stresses
due to le~~~b$les on the horizontal output device. Also
the ~c~e~iocks-in
quickly, which is highly desirable
,.:,,.:,
w i~,,~&~ warm-up picture tubes,
~~~~
that the proper setting of the horizontal hold
~~~$~ol occurs when no average current flows through
tti 390 k resistor, either to, or from, the oscillator, A
simple alignment procedure is to set the average Pin 12
to Pin 13 voltage to zero by adjusting the hold control,
when locked to a standard broadcast signal, using a
high impedance voltmeter.

The se$o~,~%rizontal
phase detector compares the
fly back~bput
phase with that of the oscillator, and
dev,.~ops ~:proportional dc voltage, which is filtered at
P.ti?~,,eThis dc voltage then sets the slice level on the
*.:~.\$\
~~s~~la~or ramp to produce the output timing desired.
o,~?$~&*Figure
9(a). Picture phasing
can be adjusted
W~~~slightly by a high value resistor on Pin 14 to +8.0 V or
. ground. A 220 k to +8.0 V will move the uicture about
2.0 ps to the left. A 220 k to ground will move it 2.0 ~s
to the right,
Another application of Pin 14 provides a method of
changing the duty cycle of the horizontal output waveform from Pin 17. Normally,
the desired waveform
would be 5070. This has been assured in the MC13001
by operating the slicer at 31.5 kHz, This permits output
phasing correction without
changing duty cycle, as
shown in Figure 9(a). In some receivers, when large
amounts of dc power are drawn from the fly back, the
on time of the horizontal output may have to be more
than 5070 of the cycle. This can be accommodated
by
feeding back some driver collector signal to the second
phase detector filter, as shown in Figure 10. This
imposes alternate slice levels and hence, the desired
change of duty cycle. Some tentative values for a set
configured like the one in this AN are given in Figure
10. This was not actually used in the final design,
because it wasnt needed. It is supplied here as a reference for future designs having more pOWer
drain
from the horizontal
output. Bear in mind that the
driver collector voltage would be much lower in the
12 Vdc receiver architecture mentioned earlier, requiring much different values to implement this idea. A
practical limit of control by this technique is about
a 60/40 duty cycle. The 0,001 capacitors on Pin 17 and
the driver base are to soften
waveform edges, to
reduce their radiation into signal circuits,

many customers like to have one, but also because it

permits using a smaller coupling capacitor for the
yoke. The smaller coupling capacitor saves money and
reduces picture bounce, but introduces some curvature
which must be compensated.
Feedback to Pin 21 provides overall output stage linearization and prevention
of deflection current change with temperature. It is
also a handy place to feedback a variable parabolic
waveshape for linearity control, as shown in Figure 11.

Osc

(A) NORMAL
SHOWN

APPLICATION,
AT TWO

PHASING

CONDITIONS

*-%-~-Pin14

..ip

~____

(BI ORIVER

FIGURE

COLLECTOR

9 Second

1~

FEEOBACK

Phase Oetector

22k

TO PIN 14

Slicer

160DF

E SOUND

~$:.,$4,
y,,*, .....
FIGURE

10

Horizontal

Driver

Feedback,

Output

SECTION

The buffered video detector output at Pin 28 is a

wideband signal used for sound take-off, A ceramic
sound take-off filter and detector tank were chosen
to eliminate alignment steps. The MC1358 is a popular, multi-sourced,
FM IF, detector and dc volume
control. It can be used with conventional
L-C circuits
or the ceramic
devices shown here. The L.C application
costs less in piece parts, but has a higher manufacturing cost in assembly and alignment.
Keep in mind that a limiting IF produces a wide
spectrum of4.5 MHz harmonics. The sound IF grounds
should be kept together and returned to Pin 1 by a
single path as shown in the copper layout of Figure 14.
Also it is a good idea to keep the input of the sound IF
IC close to Pin 28 to reduce radiation of video IF harmonics, generated in the video detector, from getting
back to the tuner or IF input,
In the receiver described here, an ac volume control
has been used, A potentiometer
is placed between the
MC1358 detector output, Pin 8, and the post amplifier input, Pin 14, The dc volume control, Pin 6, is
grounded for maximum volume. If the volume control
is to be mounted some distance away, and deflection
pickup is likely, then the dc volume control could be the
better choice. This can be done by ac coupling Pin 8 to
Pin 10, and placing a variable 50 k pot from Pin 6 to
ground. The disadvantage
is that the control contour
is less predictable in the dc control configuration.
It is,
nonetheless, a production proven method.

F~4~@nded

QQ~L~~.?
~!,.~i.I>l,

....
, ~,
~:\.~\
:J\
,$${} ~;~,;,t,.$.

.:: ...
Aside from all o~~~>~$histication
of the countdown vertical syst$~ ~thin the Monomax chip, what
.1..,.!
remains to be,.@c,compilshed outside of the device is
fairly conveq%{wpa}. At Pin 20, there is an external
capacitor,,&Qg&~
from a high voltage, to produce a
goodli%a~~$%kp, It is discharged within the chip,
usual~~wv~ertical
sync, but sometimes by the count.
da#* ~$~cult when sync is momentarily
absent. It is
~~~gr,bnt
for the capacitor to be a good stable low ESR
!:. ~~e and to be located close to Pin 20 and grounded as
~i~~~$osely as possible to Pin 1 to avoid pickup of horizontal
t,,,s Sweep which could hurt interlace.
~.:;>
The approximately
1.5 Vp.p waveform on Pin 20 is
inverted and buffered to Pin 22 to drive the external
output circuit. In the receiver design in this AN, a fairly
conventional
vertical output stage has been used. An
optional
linearity control has been added, because

153

THE VIDEO

OUTPUT

totiion of high-frequency
detail, due to excessive loading of the video driver. This can be reduced by adding
a resistor between Pin 24 and the trap, and by returning the bottom of the trap to the video output stage
emitter. The compromise
chosen is shown in the full
schematic. Again, it is good tokeepthese
parts close
to Pin 24 to reduce radiation of video detector products
back to the tuner and IF front end.

Pin 24 provides up to 1.4 V, black-to-white

video
drive, black level clamped, with a widened and amplified blanking pulse added. This is sufficient to drive a
single stage common-emitter
video output transistor.
A dc voltage of O to 5.0 V applied to Pin 26, varies the
black-to-white
amplitude at Pin 24 from 1.4 V to 0.1 V
without changing the absolute black level of the output
voltage. Beam current limiting can also be used to
control maximum brightness. This is accomplished
by
circuit shown in Figure 12. As beam current increases,
the H.V. winding current flowing inthe39k
resistor,
pulls the Pin 27 voltage down, When Pin 27 falls
below about 1.0 V, the contrast begins to be reduced.
This circuit was not used in the complete receiver in
this AN, for reasons which will be explained shortly.
The black level clamp capacitor on Pin 25 is usually
shown connected to ground. It can also be connected to
+8.0 V to cause the screen to be blanked for about 1
second after turn-on. This permits the scan systems to
stabilize before the picture becomes visible. Note If
the brightness control design window is set too high,
the raster may still be visible during start-up.
There are several approaches tosound
trapping in
the video output stage series tuned L-C from the
video output base to ground; parallel tuned L-C in the
video output emitte~ or a ceramic shunt element in the
video output base circuit. All of these can be detrimental to picture quality, if not carefully done. The
ceramic element is in keeping with the no alignment
philosophy successfully implemented thus far, so there
was a strong motivation
to use it. However, shunt
loading Pin 24, if too severe, causes considerable
dis-

The video output circuit can take many forms.

Monomax was designed to accommodate
full dc coupling, as described earlier, However, many TV desig~,
ers, and users, dont like full dc coupling, beca:s~~~::~
sometimes seems to go too black, creating the su~@$~~
that some information
is hidden. Also, ~J~,~,,~y
coupled video output to picture tube cathd::%$bhlly
requires a negative voltage for at least q~&@~~e grids
for proper set-up at high contrast s~$@g&~inally,
fully dc coupled designs are har,Q@t&o#$otect
from
\,, \
power-off flash or spot burn.
.,.<,;
,$ ~?.,.
For these reasons the recei~#iQesdribed
in this AN
was a partially dc couple~~~~~~%his
puts the brightness control in the cat~q$$.clftuit,
removes the need
for the brightness limftlti~$~nfiguration,
and makss
spot/flash
preventi~~%$~ier,
(The diode and electrolytic in GI are f~~t$is ratter purpose).
<.~>
,.t ?...8.
In the vix$~u~put
stage emitter, some dc set-up
from the :.J~kV :$pply has been used to adjust the output dc le~~{~t$~lnimizeoverall
dissipation. Also some
additianaljyertical
blanking has been fed through a
died&$rom the top of the vertical yoke. This blanking
~f}]?.b~t accomplished
in the IC internally
in later

.,,,

1
.-

,.

.>.

.:,

.:,

.$

H.V.-

FIGURE

12 Beam Current Limiting

154

22k

means a horizontal (saddle) winding of about 3.4 mH

I AGC DELAY
ADJUSTMENT
and a vertical (toroid) winding of approximately
3.0 fl,
Ideally, a known antenna signal level of 1.0 mV
10 mH. Numerous substitutions
are available,
but
(300 Q balanced) or 500 ~V (75 Q unbalanced)
is supthe above values must be adhered to for this set
plied to the tuner input. This signal level corresponds
architecture,
to the threshold of snow
in the picture, for most
Horizontal
Output Transistor
The board was
receivers. With this signal level, the AGC delay pot is
designed for a TO-3 type, such as a BU205, BU204, or
turned until the RF AGC voltage just begins to rise,
MJ12003. A plastic TO-22O type MJE12007 will do the
and then is backed off slightly. The picture should be
job with some mechanical
revision. The important
snow-free. If the RF AGC is permitted to rise, the
parameters are V(BR)CEX = 1300 V and IC = 2.0A. A
picture will start to show some snow, which therefore
small amount of heat sinking, such as a U q~,annel
represents less than optimum overall performance. If
with 2flagsofl
square inch each isrecom-~.
A
thesetting
is backed-off toomuch,
the delay may be
mica or Thermally
isolator is sugges@,@-$educe
too large and mixer overload may occur at stronger
shock hazard to the experimenter. If a~~dq.:tiesign
is
signals.
>..,, !)..,,
contemplated, as referred to back i~El~e
3, a lower
The correctness of this setting should be checked
voltage, higher current part like BQ*~~il~
be required
at weaker and stronger signals. At weaker signals, say
for the horizontal output, along ~$~$a,~ffierent yoke and
6.0 dB down, it should not be possible to improve the
.,. . ,.3~~
flyback.
picture noise by resetting the RF Delay. At stronger
,?Y*i+.),,,
\..,\,k?,
signal, say 40 dB stronger, there should be neither
Vertical
Output Tra~sistd~s
It is possible to
snow oroverload
evident in the picture, although the
get by with aTO-9~$*plementary
pair, such as
distance between these two conditions, as a function of
MPS6560
and M,~~@~2,
or the new, tall TO-92,
delay setting, may be very narrow. The AGC system
MPSWOI and M~,~~$,However,
the authors opinion
should automatically
avoid these troubles. It may be
is that these~$,~~~a~ too hot, with dissipation
apnecessary to make a slight compromise to avoid overproaching
1 w~j?each,
worst case. Recommended
load, which may produce a slight amount of snow in
alternati$%~~,clude D40EI and D41E1 in the TO-202, or
the 1.0 mV picture.
TIP2~.,@~@-~~~30 in TO-220. No heat sink is required.
The above compromises can be achieved successfully
The &~vi&s need only V(BR)CEO
= 30 V and good
without calibrated
signals, with just a switchable
h#&.aPV6 A.
attenuator
and a strong signal. Starting at strong
,,# @!&eo Output Transistor
FOrthe 10ad value
signal, note the available AGC Delay setting range
,,\I.,*,
between picture overload and snow. Using the switched
%hOwn in this design, a case 152 uniwatt, such as
vi?
fiPSUIO, is best. The300V
V(BR)CEO isrrot needed,
attenuator. reduce thesi~nal
strerreth and make sure
that neither problem ap~ears. If n~cessary tweak the .J.A
,y,,~,$ but the device must besmall
geometry; i.e., high f.
,...,..-.*,l
Delay, but dont move outside the original range~.iw
and low Ccb to preserve picture resolution. A tall T-0.9Z
Eventually the picture will get snowy, but the control-?
or even an MPSA43, TO-92, can be used if the collector
will only be able to make it snowier. Setting,$ to t~e
!oad is increased to 6.8 k, but some picture quality will
optimum (just barely) should still be withi~~%~$poted
be lost,
range.
Audio Output Section
The transformer should
,<\,, ,,~,
be approximately
30:1 turns ratio, capable of handling
APPENDIX
II COMPONEN~$$:T{$$
1 watt into 8.0 Q. The output transistor should be set
up at about IQ = 12-14 mA, and should be capable of
CONSTRUCTION
DETAIL~~R;~~\~
1.5 W continuous dissipation. ATO-220 type MJE2360T,
In order to make the enclos~
~~bard
pattern easy
mounted on at least 3 square inches of aluminum is
to use, the following com&h{~#
are recommended:
suggested.
Remember
that thes:~q.~$~jttinent
tothisdesign
H. Driver
Stage In the prototype receiver, the
architecture and this $$ec~c design. Many variations
available driver transformer had only about 12:1 turns
are possible with ~i$tt~b~%design work.
1$,
ratio, This necessitated
a large wattage dropping
Flyback
&,.Go?$,JStar Type 154.028A with selfresistor to provide the rather low-voltage, high-current
contained
H*{~~~*r.ectifier. Certainly.
substitution
is
primary waveforms.
It would be better to obtain a
possible,~~~~y$areful
attention to pin-outs and taps
transformer
of 30:1 or so, to permit a more efficient
is reqyi~$~~e
primary is, of course, a 120 Vdc type,
driver stage. The 4.3 k/2.O W resistor could then be
whi$hk~~k~sponds to about 800 Vpp pOsitive pulse at
reduced considerably.
In either case a TO-92 driver,
P$\Q~$$?k 3 is a negative going pulse of 35 VP-P and
type iMPSA42, is a good choice.
~n ~ja a negative-gOing
pulse Of abOut 120 VP-P The
SUMMARY:
@s#!%$terminal,
which is internal in the above model,
$++~~~ld be a positive going pulse of about 12 kVP. P.
..
Figures 13 and 14 provide the copper pattern for the
$<~:~~ery little flexibility can be permitted on these values.
PC board and the component
locations.
Note that
...
Be
careful
to
watch
pin-outs
and
horizontal
polarity.
....
signal input circuits are compact and grounded near
Pin 1, Subsequently
these and all other circuits are
Yoke
Gold
Star Type 153-020A for 90 12
connected to the central ground at Pin 16, without being
20 mm neck picture tube. It requires approximately
interconnected beforehand. The full receiver schematic
1.0 Ap.p in both horizontal and vertical windings tO
is given in Figure 15.
give proper overscan in the 90 tube at 10-11 kV. This
APPENDIX

155

FIGURE

14-

Printed

Circuit,

Copper

Wde (not full size)

,..-

.,%
I
,

.,a.

-.

T.

.,,
F

,m

,m,
k

.,*

,,.,..

FIGURE

15 Complete Receiver Schematic

AN925
UHF PREAMPLIFIER CENTE~ ON BUDGET
DUAL-GATEGaAs FET
Prepared
Gary

Barbarl,

by

Appllcatlons

Engineer,

RF Products

and

INTRODUCTION
This note describes the design, construction and performance of a 400512 MHz preamplifier utilizing Motorolas GaAa dual-ga~ field-effect-transistor.
In two-way communications, the ability to receive a
transmitted signal depends on the sysbms signal-tonoise ratio (SN). The S/N can be improved by increasing
the output power of the transmitted by increasing the
gain of the antenn% or by improving the sensitivity of
the receiver. The fimt two solutions could be quite expensive. A low noise preamplifier would be an economical
solution for improving the receiver system noise figure.
,::

DESIGN

.$,$,.

~N

i+,:t,::}{}lt
...;

The main criteria in the selection of a transistor fo~~:,:$k.

is low noise figure coupled with sufficienbi
preamplifier
gain to minimize the second stage contributi~
to:.$he
system noise figure. The Motorola MRF966 J4Ra Gtis
dual-gate
field-effect
transistor
design~~ f@rUHF
/q,:,<t,k:J.,.:<
applications.
!,{.....$1.~~.~
Designing
impedance
transformtitse~i
networks

159

TABLE
S-Parameter
VDS

= 5V,

and NF Data
IDS

= 10 mA

The MRF966 was matched by means of slug tuners to

obtain the minimum noise figure. The optimum source
(rms) and load (rml) impedances were then measured
on a network analyzer.
The slug-tuned circuit used in this urocedure is illus-

The network required to transform the optimum

impedances to the required 50+hm murw and load was
designed using a Stith Chati. The input matching network is shown in Figure 2, At the input of the preamplifier it is necessary ti transform the 50-ohm input
impedance to the optimum source reflection coefficient
(rms). Taking the values from Table 1 for 450 MHz an
input matihing ci~uit can be designed using a series

FIGURE

2 Input Matching Network

50 Q

FIGURE 4 9:1 Transformer

FIGURE 3 Output Matching NeWork

160

------ ..-..-

capacitance, shunt inductice,

and shunt mpacitance.
Stirting at the input of the device (rms), the shunt
inductance moves the impedance h the value of 50 +
j145 ohms (point A); (the shunt capacitor is used to fine
tune the inductir along the constint adrnittince circle),
Finally the series capacitance transforms point A to the
desired 50 + jO ohms center (point B). The required
reactance value for the three componenk can be obtiined
directly from the Smith Chart. From Figure 2 the shunt
inductance moves the vector along the path to position
A. This move requires an XL of 104.2 ohms. Therefore
the shunt inductor has a value of 37 nH at 450 MHz. The
shunt capacitor needs to be variable from 0.810 pF to
accommodak
variations between devices. The series
capacitance rotites the input impedsnce from 50 + j 145
ohms to the center of the chsrt (50 + jO) ohms at pint
B. Therefore, the required capacitive reactance is j2,9 or
j145 ohms which leads to a value of 2.4 pF at 450 M*.
A vsriable capacitor (O.*1O pF) will also be uwd here
to fine tune the preamplifier for a specific frequency over
the 400512 Mfi band,

The output matching circuit could be designed using

a shunt mpacitir with a *ries inductor, but a more convenient matching tichnique involves a 9:1 transformer.
This simple matching tichnique although not presenting
the optimum load reflection coefficient
(rL) to the
W966
provides improved stability at the expense of
slight gain reduction. The 50 ohm im~dance of the load
is transformed to a value of 450 + j150 (point C, Figure
3). The materials needed b construct the transformer
sre inexpensive and readily available. The lumped element form of the transformer and the winding procedure
are show in Fimre 4.
Source self-bias is wd utilizing a 1000 resi%~@s,The
resister will set the operating current at ap~~-~%ly
20 rnA. Decoupling the source and Gate 2 is @Sfished
using 1000 pF and 56 pF chip caps. Gat:~~&.#itively
biased using a simple voltage divider ~.~k$A
low POSitive voltige mr the gate will lower .~e$~~e
figure and
increase the power gain. The compf~p,wamplifier
schematic and the parts list are s~$w~k figure 5.
,:,*>,
-:$$
.{,
..
....

D2

R3 7,5-tilohm

0,125 Watt

L1 2 Turns No. 16 AWG 0.375 Diameter

T1 4 Turns No, 30 AWG, Indiana General
U 1 MC78L05
Dl, D2 1N4001
Q1 MRF966
J1, J2 SMA-Type
Female Connectors
B Ferroxcube Bead 56-590-65

FIGURE 5

Schematic ~agram

161

Core,

F2062-1-Q1

TOP ot PCB
E-ECXRE h - mnromastoi

I
!lKOl hiti sire)

20

1.8

I A
x

1.6

f, FREQUENCY (MHz)
FIGURE 11 Gain at Noise figure and Optimum
Noise Figure vemus Frequency (Tuned @ 430 MHz)

a
E
~

A
\

K
/,

-80

k
m

//

\
,2

/ /

Is 11

71

\\

I/x

>

-4.0

II

16
I \\

//

-20
I
-24

1
/

IS221

/
2a
420
FIGURE
vemus

12

Input

Frequency

and Output

(Preamp

Tun4

Return

LOSS

I
440
FIGURE

@ 430 MHz)

versus

163

\ )

460
480
f, FREQUENCY (dB)
13 Input

Frquency

and Output

(Preamp

Tund

500
Return

LOSS

@W

MHz)

520

APPLICATIONOF THE MC1377

COLORENCODER
The MC1377 is an economical, high quality, RGB encoder for NT~~,~
or PAL applications. It accaptsred, green, blue, and compositqt$~$<v
inputs and delivers lVpp composite NTSC or PAL video ou@,ut*
a 75 ohm load. It can provide its own color oscillatorand bu~~~ting,
or it can be easilydriven from extarnal sourcas.Perform#*~~lr~ually
equal to high cost studio equipment is possible wit$~~h~on
color
receiver components. The following note is inten~}~~xplain
the
oparation of the device and guida the prospa~~e$~er in selecting
,,F....s~
the optimum circuit for his needs.
.\.*:*:
?:
!:+
.*O
if,~:;,y!~
.t

,.$

PREFACE

..<,,

,*$$?,
:.)j~,f
*:..J,,*

Y = .59G + .30R + .llB

Since this device has applications in color cameras,
J ::$
,, ~:)..
~,,<*$
RY
= .70R .59G - .llB
video games, video text and computer generated graph- tt$sk..
WY = .89B .59G - .30R
its, it may attract potential users who are skilled in co~~~?i$
.+;>.
Texts on the NTSC system will show that studio modputer architecture, but not familiar with the encoding:~f $>
ulation is done on a different set of orthogonal ~es called
color television. Perhaps they have spent extens~xe hotirs
viewing graphics on a full R, G, B widebaR&**$nitor.
I and Q. Also they will point out that I is a somewhat
wider bandwidth than Q. The MC 1377 does not permit
This preface is intended to caution that .~&<Qr hTSC
the circuit designer this refinement, but it should be
encoding, no matter how rigorously exe@te~+,Wfll
cause
,. ,..$
noted that very few monitors or receivers contain any
some degree of picture degradation..*,
p,wcess of en:: ~,.
coding involves some bandwidth r~$~m,~~;:which means
circuitry to process the unequal bandwidths. (This is the
10SSof high frequency detail, and ~i~-s
the possibility
only compromise of standards in the MC 1377 which cannot be circumvented by applirdtion means.) Rotation of
of spurious picture patterns, d . tq$d~mg and decoding
y,
the coordinate system from UQ to (WY)/(WY)
does not
system limitations. The oti~~
Wandards were established about 25 years a&~~~,#ll
probably be in use for
constitute any further compromise whatsoever, and it
makes the encoding formulae for PAL and NTSC the
many years to come. It ~: n@ the objective here to detail
same. It also aligns (&Y) with the axis of the NTSC
these standards as ,~~$*6rences1-4
are available. AP
color burst, for internal circuit simplicity and system
pendix A showspid~[$$ily
why some loss of information
accuracy.
and detiil is ~~~ed.
The MC l,~h~;t$itapable of encoding NTSC and PAL

REFERENCES

to virtu~~~.~~~o
standards. It also can be used for very
low CO* ~fifi$ations where appropriate, with some compr:,~@#picture
quality. It can readily drive the 750
i@tit,,M,a composite video monitir, or be used to drive a

1. Donald G, Fink, Television Engineering Handbook,

McGraw-Hill 1957.
2. Hazeltine Stiff, %nciples
of Color Television,
Wiley 1956.
3. Gerald Eastman, Television
Systems Measurements, Tektronix 1969.
4. G. N. Patchett, Color Television, The PAL System,
Norman Price 1976.

,W<W.pr
VHF modulator so that color television receivers
&$x.* be used.

Fi@re 1 shows a block diagram of the color encoder.

The three color inputs at Pins 3, 4, and 5 are matrixed
to produce chrominance envelopes, (RY) and (BY), and
luminance ( Y) by the standard NTSCpAL
formulae:

165

358
4.43
MHZ

-~

Vcc ~
~ol

.12Vdc

2V
~
14

19T

color

bandpass
transformer

17

PAL

82V
Regulator

Switch
0,1 80~

1
I

f
L(ne

Burst

Dr(ve

Flag

R-Y

B-Y
Svnc

Latch$ng
Ramp

+
15

Color

Gen
J
2
COmpOslle
Sync

0,001
g

and

Matrix

t
1

Difference

Lum,nance

Input

~:~

4+2+::157
.--_-Inp.ts:

1.0 V p-p

The (WY) and (WY) si~als drive two double ba~~

anced (double sideband suppressed carrier) modula~~~~~:i,,
whose carriers are set at 0 md 90, respectively. w,%,
NTSC mode, the outputs of these chroma modul@tQ@;~&
added to produce composite chroma. Burst ,q~~~:@
or
burst flag is applied to the (&Y) mo$~ljw~?n
the
negative direction to produce a burst PUIX ~ ~}teference
angle of 180. Composite chroma is a~p~~
and buffered to Hn 13 (to Dermit external~~n~%?~th control as
desired) and is the; fed back i~@.~k&
at Pin 10 to be
combined with the luminance
mp~nent. The luminance
signal is also looped out_ % ~h 6 to Pin 8 to permit
ineertion of a delay line /&!~,#ch the delay incurred in
the chroma channelt~e
to bandwidth reduction. The
.+! ,...
passive component$:$~$$n
the chroma and luma chan~els are like th,@,&&
in the most common implementation Of cOIO~t~@gv?&Onreceivers.
In PAL @@}~&~rst flag is driven into both modulators
equally :~~~kuce a 225/1350 burst phase. The output
phas~k~~.~,~~arity, of the (RY) modulator output is alter+<~~switched
from 90 to 270 on successive hori-

.}

~$~~{hnes,
before being combined with (EY), which
r~alns
at 0. The switching of the modulator polarities
foFPAL mode is driven by the Iatihing ramp generator
through the PAI,/NTSC control. This control allows PAL
switching when Pin 20 is open, and stops when Pin 20
is wounded. The PAL phase can be detected at Pin 20
and controlled by means of external logic. The PAL phase

166

,,$,...
,,\ ,rf

mn be reversed by sensing when Pin 20 is high and Pin

1 is low, and momentarily pulling Pin 20 to ground with
an external switch.
The color subcarrier source for the modulators can be
implemented by free running the on-chip crystal oscillator, or by external drive into Pin 17, or by a combination
of both methods, The common collector Colpitts oscillator
is completed by connecting a standard tv receiver cclor
crystal and capacitor divider as shown. The oscillator is
followed by a 90 phase shifter to provide the quadrature
signal to the (WY) modulator. The direct oscillator output is taken as reference O and is fed directly to the
(WY) modulator.
The composite sync input at Pin 2 performs three impofint
functions: it provides the timing (but not the
amplitude) for the sync in the final outpuu it drives the
black level clamps in the modulators and output amplifieq and it triggers the ramp generator at Pin 1, which
produces burst envelope and PAL switching signal.
The ramp generator at Pin 1 is a simple R C type in
which the pin is held low until the arrival of the leading
edge of sync, The rising ramp function passes through
two level sensors the first one stark the burst pulse
and the second stops it, Since the early part of the
exponential function is used, the timing provided is relatively accurate from chip-to-chip
and assembly-toassembly. Fixed components are usually adequati, The
ramp continues to rise for more than Yz of the line in-

4.4V

Umits
for dc

(al

Couplsd

1.0 v (p-p)

inputs
2.2V a
?

(b)

terval, thereby inhibiting burst generation on %alf interva~ pulses on vertical front and back porfies. Burst
is also inhibited if swc is wider than the time required
for the ramp b reach the sense levels, as is the cass
during vertical eync. The ramp method will produre burst
on the vertical front and back porches at full line intervals. In most applications, this discrepancy from standards will not cause any problem. If it is objectionable,
and if a proper burst envelope signal is available, then
it can be inje~d
into Pin 1 directly. Another method,
suitable for either PAL or NTSC, will be desmibed later.

1oo~/.
Green
Input
(Pin 4)

*.,

STANDARD
(c)
1.0v

1oo~).
(p-p)

Blue
Input
(Pin 5)

5.0

,d)

Compos,te
Outpul
[Pin 91

4,0

3.0 *
[e)

8.2 Max

, ~--------

--=

0.9 Max

-0,5

o
Min

~.
II

1.7 Min

, I

It
II

II
J

i~-

Sync
Input
(Pin 21

INPUT LEVELS

The signals into Pins 3,4, and 5 shouid~;p~%k 1 Vpp

for standard, fully saturated, color out~~~~efs as shown
in Figure 2. The levels are import?l~6~~&e
the IC will
generate a predetermined 0.6 Vp~f#fi:,Wd
0.6 VPP burst
at the output, and it will ne@&t~,~~p
input signals ti
produce the corresponding f~l l~rninance and chrominance amplitudes. The :&&~ts age internally biased and
present a 10 k input i~~@~e.
The 15PF input coupling
capacitors are suffi@%9VW prevent tilt during the 50 or
60 W vertical pe~j~~jaut
signals can be dc coupled (to
save the cost o~:~~~~i~acitors), provided that the signal
levels are ~.tw~
2.2 V and 4.4 V at all times, It is
essential *~@&,~he portion of each input which occurs
durin~$~~~c
interval represent black for that input,
be:,~~
it:Will be clamped to reference black in the color
-x!;*
and the output stage. A refinement such as
... a d]~erence between black and blanking level must be
.~$:@corpora@d in the RGB input signals if required.
$?$~,

..

.S!~,.:#THE SYNC INPUT

t,?, :$<.,
.,,\-

As shown in Figure 2, the sync input can be varied

over a wide latitude, but will require bias pull-up from
most sync sources. The important requirements are that
during the period between sync pulses, the voltage must
be above 1.7 V and below the 8,2 V internal regulator,
During sync, the voltage (negative going) must extend
below + 0.9 V and should not exceed 0.5 V (to prevent
substrate leakage in the IC). For PAL operation, correctly serrated vertical sync is necessary to properly trigger the PAL divider. In NTSC mode, simplified block
vertical sync can be used but the loss of proper horizontal
timing may cause top hook or flag waving in some
monitors. An interesting note is that composib video can
be used directly as a sync signal, provided that it meets
the sync input criteria.

THE LATCHING
GENERATOR

FIGURE

..?%,
<..!.$$
>:h
.*...
,?.,.% ,.>$

2 SIGNAL

(Circuit

Values

The recommended application is to connect a close

tolerance (5%) 0.001 KF capacitir from Pin 1 to ground
and a resistor of 51 k or 56 k from Pin 1 to the 8.2 V
internally regulated supply (Pin 16). This will produce
a burst pulse of 2.5 to 3,5 P in duration, as shown in
Figure 3. As the ramp on Pin 1 rises toward the charging
voltige of 8.2 V, it passes first through a burst start
thresholfl at 1.0 V, then a stop threshol& at 1.3 V, and
finally a rmnp reset threehold at 5.0 V. If the resistor is
reduced to 43 k, the ramp will rise more quickly, producing a narrower and earlier burst pulse (starting about

VOLTAGES

of Figure

RAMP (BURST FLAG)

1)

167

0.4 @ after sync and only about 0.6 w wide), The burst
will be wider and later if the resistor is raised to 62 k,
but more importantly, the 5.0 V reset point may not be
reached in one full line interval, resulting in loss of alternate burst pulses.
As mentioned earlier, the ramp method does produce
burst at full line intervals on the vertical porches. This
is not rigorously comect for studio applications, If external burst flag is available, a positive pulse of between
1.0 V and 1.3 V (absolute value) can be applied to Pin 1
in the NTSC mode. This approach must be handled carefully, because a square pulse smaller than 1.0 V will not
trigger the burst generator, and a square pulse larger
than 1.3 V will shut off the burst generator almost befor~,
it starts, This direct injection technique does not provi~,
the ramp to operate the PAL flip-flop. Another metR$#~J*
suitable for either PAL nor NTSC, is shown in ${**
It requires a vertical drive pulse, starting a&{@ ~~%ding edge of vertical blanking and as wide ?S ~~.~$rval
where burst is not wanted (usually 9 lin~,}$,wkls),
The
extra transistor and diodes in the cirq~lt:~~~ an abrupt
step at the beginning of each line .~fi~.~hich
inhibits
*
.+.
burst generation.
>,.,$/,,
:,~,$k
,:

...

.$.

It is also possible to do.$h;,$~.,

let the oscillator free
run on its own crys@~~~.,,~ also be capable of being
ovemidden from a~,,ext~.ua
?
source. An extra coupling
capacitor of 50 @.,~,!.
@W
\!
.... the external source to Pin 17,
and a si~al #%%~->~pp was adequate with the limited
experiment,~&n @tempted.
??..,t+.
:i~+

VOLT,~@~,$CONTROLLED

90

~$~,:illator
drives the (BY) modulator and a volt.
a~~~tfblled
phase shifter which produces an oscillator
~~~$$.bf 90 t 7 at the (RY) modulator. If it is necessary
r~$@.SdJust the angle to better accuracy, the circuit shown
$? Figure 6 can be used,
,,,.
Pulling Pin 19 up till increase the (R-Y)to
(B-YJ
angle by about 0.250/@,
Pulling Pin 19 down reduces
the angle by the same sensitivity. The nominal Pin 19
voltage isabout6.3V,
so the 12 Vsupply is best for good
control, even though it is unre~lated.
In most situations,
the result of an error of7 is very subtle to all but the
most expert eye. For effective adjustment, the simplest
approach is to apply RGB color bar inputs and use a
vectorwope. A simple bar generator giving R, G and B
outputi is shown in Appendix D.

THE COLOR
REFERm&Q;}
OSCILLATOR/BUFF@

RESIDUAL
FEEDTHROUGH
COMPONENTS

As stated earlier@ ~eg~neral

description, there is
an on-board comm~k~,~ctor
Colpitts color reference
oscillator with t~~%~slstir
base at Pin 17 and the emit.
*>J
.$.
ter at Pin 18.$Wn
used with a common low-cost tv
c~stal and~#&ifive
divider, about 0.65 Vpp will be
develope~:xht,~n
17. The adjustment of oscillator frequen~s~,@done
with a series 30 pF trimmer capacitor
,.,, :,
oveP~:WBlr angeofabout
l. OkHz. Oscillator frequency

As shown on the MC1377 data sheet (and in Figure

2 (d), the composite output at Pin 9 for fully saturated
color bars is about 2.6V
output with full chroma on
thelargest bars(cyanan fpred) being l.7Vpp. Thetypical
device, due to imperfections in gain, matrixing, and modulator balance, will exhibit about 20 mVpp residual color
subcarrier in both white and black, Both residuals can
bereduced ti less than 10mVppfor
the more exacting

sw$~be,adjusted
for each unit, keeping in mind that
m@mon]ters
andreceivere can pull in 1200 Hz.
Ran external color reference is to be used exclusively,
it must be continuous. The components on Pins 17 and
18 cm be removed, and the external source capacitively
coupled into Pin 17. The amplitude at Pin 17 should be
between 0.5 Vpp and 1,0 VPP either sine or square wave

applications. The black imbalance is primarily in the

modulators and can be nulled by sourcing or sinking
small currents into clamp Pins 11 and 12 as shown in
Figure 7. The nominal voltage on these pins is about 4.0
Vdc, so 8.2 V is capable of supplying a pull up source.
(Pulling Pin 11 domisin
the 00 direction, up is 180,
Pulling Pin 12 down is in the 90 direction, up is 270.)

168

4icll_1

F1
I

E 4sci -

ST

vm..
NT

,,

!.,-,

,..--,

*,,..? .
T

T ,<,,

,..

,.,;ti. \L

,,,.

15 W..

,.

1
,,06

w
I

R!,,

4,,

,,0,

8>)0

,,0,

,,06

,0,

FIGURE 5

170

-..-

..

R31
5,k

R27
,,0

[
r~,oh
9

... .

Iwl
l-l

,43

.,-.
1 ......
22 k

INTERNAL

SCHEMATIC

171

T,,e

II

II

Any dircctimr of correction maybe required from part to

part. (Notetiat pulling Pin 11 up can produce a residual
c-ier
on the horimntal back porch which is the same
Dhase as burst. and can result in an almost normal color
display even with burst not present.)

+12V

wok

FIGURE

6 AMUSTING

MODULATOR

ANGLE

+ 8.2 v

Hok
I

+
+ 8.2

FIGURE

7 NULLING

RESIDUAL

COLOR

CARRIER

V
IN BLACK

L.
,.

,.

FIGURE

.) ,.

pass circuit between Pins 13 md 10. For proper color

level in the composite output, a mid-band insetiion 10SS
of 3.0 dB is desired. The bandpass circuit shown in Figure
1, using the TOKO tied tuned transformer (see Appendix B) gives this result. One of many tv color IF bandpass
circuits could also be used. When such a bandwidth reduction is inserted, the chroma is delayed by approximately 35o ns (as shown in Figure 8).
This 350 ns delay results in a visible displacement of
the color and black and white information on the final
display. The solution is to place a delay line in the lu- $s~
minance path from Pins 6 to 8 to realign the twO co~~>$
ponents. Again, a normal tv receiver delay line c~~~ ~.
used, These delay lines are usually of 1.0 k m~*~@~R
characteristic impedance, and the resistors at:~~~&and
8 should be selected accordingly. A very co~~at~~mped
constant delay line is available from TD,~~t~~tAppendix
C for specifications). Some types of dq~~y ~~
have very
low impedances (approximately lQO d~%) and should
not be used, due to drive and p~~~disslpation
require,,~ .~.~
:,:~~,.
,{,.\~:
,,,$,
ments.
k.i.$
In some applications, it ~@ Mdossible b delete both
the bandpass transform~k:~~the
delay line. For in~L.
stance, when the RGR?JAYmation
itself is ve~ low
resolution, i.e., ve@ipa&
band (less than 1.5 MHz),
no cross-talk wo,u~,~~nerated
in the encoder (see Figure 9). Keep iq$rn~$@however, that the standard monitor
or receive~pwfi~~$@~l see an incomect luminance sideband at X$*his
points up the value of at least some
chro~ banbidth
reduction in the encoder. A simpler,
l:~~f, $&t bandpass circuit is shown in Figure 10(a). It
@ovi@s the proper insertion loss, approximately f 1.0
k@bandwidth,
and about 100 ns delay,
::t.$:\...$,k,*
The circuit shown in Figure 10(b) is even less costly,
but has about 6.0 dB greater loss, provides very little
>+s
bandwidth reduction except to remove the baseband
feedthrough, and produces essentially no delay.

~m

., .,

$.*.
,:
~.<\,i. 7,:,
;:.,.
~i,

1.0

Whiti carrier imbalance at ,th$,,oq@ut can only be

corrected by juggling the rel%$fv~$~e~;ls of R, G and B
inputs for perfect balance,~t,&wd
devices are tested
to be wtihin 5% of balanc~~t X11 saturation, Black balance should be adjus~d firsti~$ecause it affects all levels
of gray scale equall~:;~~$~~ is also usually some residual
baseband video q~t$: Wtoma output (Pin 13), which is
most easily o~~,~@
by disabling the color oscillator,
Typical dev~~~f~~
0,4 Vpp of residual luminance for
satura~$:.@$~&bSr
inputs. This is not a major problem
since P1m 13 ~s always coupled to Pin 10 through either
,,,,.\i\!
a ba,e~=k
a high pass filter, but it serves as a warning
@ p~y~~$ber attention to the coupling network.

2.0

3.03.584.0

5.0

(a] ENCOOER OUTPUT WITH LOW RESOLUTION

AND NO BANDPASS
TRANSFORMER

b
1.0

2.0

[b) STANDARD

3.03.584,0
RECEIVER

FIGURE

INPUTS

5.0

RESPONSE

*< .><.%., . ~

Ti&ECHROMA

COUPLING

It will be left to the designer to decide which, if any,

compromises are acceptable. Color bars viewed on a good
mon;tir can be used to judge acceptability of step luminance/chrominance
alignment and step edge transients, but signals containing the finest detail to be encountered in the system must also be examined before
settling on a compromise.

CIRCUITS

Without going deeply into the subject, it is generally

true that monitors and receivers have color IF 6.0 dB
bandwidths of * 0.5 MHz. It is therefore recommended
that the encoder should also limit the chroma bandwidth
to approximately ~ 0.5 MHz through insertion of a band172

0.001

0.001
*10

l.Ok

4
Pin 13
22 pH
(a) Insertion

& 1.0 MHz

= T
100 nsec

56 pF

0.001

l.Ok

11

..it..$,

,,,.
.

,,,.
,:J.,),,

... ..$
. .

4.7 k
(b) Insetiion

2.0 MHz

OT

10 OPTIONAL

THE O~PUT

CHROMA

COUPLING

.! .,,:..

..$jl

printed circuit board will be even more e~~~vef$ cooled.

The MC1377 is designed to operate$~=;unregula~d 10.8 to 13.2 volt dc power supp&~&*ice
current
into Pin 14 with open output is t#@li~t30
to 32 ~.
To provide a stable reference fo~.~~$~~p
generator and
the video output, a high quali~,~.~V
internal regulator
is provided. The 8.2 V regu~to~~n
supply up to 10 A
for external uses, with~~$~qctive
source impedance of
less than 1.0 ohm. ~1~ ,~~lator
is convenient for a
tracking dc refere~~@,@c
coupling the output to an RF
mtiulator. TWi~~~~#~-on
drift for the regulator is ap-

2J pF

Loss: 9.0 dB

Bandwidth:

FIGURE

UGURE

*,0

4
Pin 13

Delsy:

Mwq

39 pF

Loss: 3.0 d6

Bandwidth:
Delsy:

,) I

CIRCUITS

STAGE

The output amplifier normally produces about 2.0 VPP

and is intinded to be loaded with 150 ohms as shown in
Figure 11. This provides about 1.0 VPP into 75 ohms, an
proximately + 35~V over 12 minutes in otherwise staindustry standard level (RS-343), In some cases the input
ble ambie~~~~~ditions.
to the monitor maybe through a large coupling capacitor.
,. ,,.,.... ..
If so. it is necessarv to connect a 150 ohm resistor from
Pin 9 to ~ound to ~rovide a low impedance path to discharge the capacitor. The nominal average voltage at Pin
~&Pieceding
Application Note was intended to detail
9 is over 4.0 volts. The 150 ohm dc load causes the cument
,: $:+heapplication and basis of circuit choices for this versupply to rise another 30 mA (to approximately 60 mA
%$ile tv signal encoder. A complete MC1377 application
~,c
totil into Pin 14), Under this (normal) condition the total
~i{:.with the MC 1374 VHF modulator is show in Figure 12.
device dissipation is about 600 mW. The calculated worst ?*? ,*$>
The internal schematic diagram of the MC1377 is pro,:...,,~..;>.
case die temperature rise is 60C, but the typical devicb::vided in Figure 5. If further assistance is needed, contact
in a test socket is only slightly warm to the touch at room ~
Motorola Linear and Military IC Division, Applications
temperature. The solid copper 20-Pin lead fra~e insa
Engineering.
~i,*.>,,.

suy$A$?

173

APPENDH
In full RGB systems, three information channels are
wired from the signal source to the display to permit
unimpaired image resolution. The detail reproduction of
the system is limited only by the signal bandwidth and
the capability of the color display detice, Higher than
normal eweep rates may be employed to add more lines
within a vertical pefiod. Three separate projection picture tubes can be used to eliminate the shadow mask
limitations of a conventional color CRT.
Figure (b) below shows the baseban& components of
a studio NTSC signal. As in the previous example, ener~
is concentrated at multiples of the horizontal sweep frequency. The system is further refined by precisely locating the color subcarrier midway between luminance spectral components. This places all color spectra between
luminance
spectra and can be accomplished
in the
MC 1377 only if full interlaced external color reference
Spectral Energy Is Always
At Horizontal

Sweep

and sync are applied. The individual components of luminance and color can then be separated by use of a comb
filter in the monitor or receiver. This technique has not
been widely used in consumer products, due to cost, but
it is rapidly becoming less expensive and more common.
The unequal bandwidths of I and Q cannot be implemented with the MC1377, first because I md Q axes are
not used, and second, because outputs of the two color
modulators are added before any bandwidth reduction is
~.
it,,,
imposed. Most monitors and receivers compromise+, $~~m.:~
standar& quite a bit, by using responses as sha~?<,~fi?~.
Figure (c). Some crosstalk of luminance informA@~ 1~~
chroma, and vice versa, is always present. W~~~~/ptability of the situation is enhanced by ,t,~? w@ession
of the color carrier and the generally ll~~~d.~~ility
of
the CRT to display information aboF&+&.~t:WHz. If the
si~al from the MC 1377 is to be usegtariaarily
to drive
,., -,.$.,,.
conventional non-comb filtered ~~itom
or receivers, it
would be best to reduce the ~~&dth
at the MC 1377
to that of Figure (c) to les~q~,,~wtalk.

Concentrated

Frequency

Multiples

Red

Ill

II

Green

Ill

Blue

1
1

!1!

FIGURE
I
I

FIGURE

13(c) TYPICAL

FIGURE

13(dl COLOR

1/0 SYSTEM

versus

sHowING
FIGURE

13(b) NTSC

STANDARD

SPECTRAL

MONITOW

,$ f

VECTOR
(kY)/(5Y]

STANDARD

REMTIONSHIP,
SYSTEM
COLORS

CONTENT

174

..
-...._

....-

..-_._-__.

-__., -----

APPENDIX
A PROTOTYPE
TOKO

3.520

SAMPLE

.5

BANDPASS

NUMSER

TRANSFORMER

16SNNF-1WS4AG

-0,7 mm Pin Diameter

15,0 mm Max

CHROMA

/
!r

._.

B]
3

@s

Toko America
5552 West Touhv
Skokie,

Avenue

Connation

IL 60077

Diagrem

SOttOm

Wew

(31 2) 677-3640

y
APPENDIX
A PROTOTYPE
TDK SAMPLE

Skokie,-lL

@~#

.,

Time

D~Yti@E

NUMBE&~122~lD-lS33

MARKING:

(31 2) 679-q~:>\W
+!.$,>,;\$,
?:,
\
.
,$;,. ..$$
~
.$ ,: a. ?
,>,T*,
~.,.,.
j: s,.. . . ...
.,
,: >,,.,,.
h
,

3
..:N,}
., ?<*
?* .+?...,.
4
~y},

~.

PART NUMBER,

MANUFACTURERS

DATE CODE AND

IDENTIFICATION,

LEAD NUMBER.

kern

Spwfications

Delay

400 na *

10A

Impedance

1200 Ohms

Resistance

Less Than

Transient

Anenuation

,,

C+~,,){S*$

Response

with

20 ns Rise-Time

Input

Pulse

Pre-Shoot:

10le

15 Ohms
10A Max

Over-Shoot:

10O/. Max

Rise-Time:

120 ns Max

3 dB Max

175

at 6.0 MHz

APPENDIX

AN RGB PULSE GENERATOR

176

. ...-_.. -.. _

__

..._.-. ..._.____

.__..

ANIO19
NTSC Decoding Using the TDA3330, with
Emphasis on Cable In/Cable Out Operation

,jk,,
,,,

Prepared by
Ben Scott and Khalid Shah
Bipolar Analog IC ~vision
PREFACE
The

TDA3330

Decoder

is

originally

receivers

end monitors.

picture

tube

application

drive,
note

by providing

processing

cialized

applications,

details

of the

PAL

than

is intended

in video

video

for

The data

rather

circuits

.,{
...,

composite

intended

to

and

sheet

level

to supplement

circuits,
and

cable
frame

store,

toward
This

the data

sheet

such
and

upon

Sandcastle
neers.

outputs.

drive,

to expand

Color

color

is oriented

cable

for video

RGB

NTSC

pulse
in
tle

is used,

such

solution

ing

as shown

as possible.
must

in Figure

The

be close

able

for

prototy

mended.

In

layout,

most

except

respects

ture

artifacts,

beats

with

a good

ground

rap

plane

ground

part

is not

,
::\.
rec-~tp;

sensitive

eas{e~~

the

control
i$,.

,+,,+..:.,!Y,
>
.,. ~-:.,
,~~.~~

400

mVpp,

scope
is 3.58
a few
ing

measured

probe.
MHz

Keep
and

hundred

is enough

at

with

in mind

is part

a#$~q:R~Nary
~kMt~e

It is recommend@Y~aPPin
ascertain
9 alone,
later.

that~it

Of co~+%~$.gn

becon~:~@to
capaa~~d%~{educed

for

output

scope

later

and

into

then

buffer

Pin 9, permanently,

pF/10

portions

For th~
really

2.

Setting

black

3,

Gating

to cable

at Pin

control

burst

level
phase

be ohase-locked

(ACC)

amplitude

ifi the
to the

so that

(APC)

burst,

IF gain

constant;

R, G, B outputs,

detector

and

so that the VCO

See

the

block

2,

only

of lock.

BuRSTGATE

to

8 and

covered

follower)

can
G

Pin 9tuning

accordingly.

COMPOSITE
VIDEO

Figure

177

15 is

be grounded.

has 3 functions:

IF gain

to keep

the color

16 should

cable

needs

used
of the

load-

Pins
be

and the

blanking

blanking

11 is not

MQ

initiallv

will

(emittar

with

or out

Ieeve

adjustment

Pin

about

probe

observed

the

both
is only

blanking

extra

required.

is adjusted

in Figure

at the

to supply

TDA3330

pulse

blank-

portion

waveform

application.

Pin

is

in addition

vertical

blanking

not

color

the

the

the

pulse,

the

the

so

blanking

frequency

loop

The
9 be

~~t%,~ning,

A @Fqsed&?e

4.0

oscillator

of$p~~~-locked

Hz pul~+,~.~~ge.
+~,
to pus$%[h$$oscil

be

driver

are

bur~t-gate

can

~$~k,~:{~ould

tube

all that

1, Gating

THE OS~~@~~

amplitude

engi-

is 4,0 V, sometimes

is used

applications,

gram
oscillator

Sometimes

a blanking

burst-gate

The

~k

unw@nted:pic-

are much

layout.

.i?h~other
s$hdcastle
9 :? +%
*
decoder,

8 and

composite
blanking,

rizontal

which

i~, the~~jcture

plane

not

however,

$o~~ro~lde

IS accep%,.>::,~,i.~

socket

is definitely

the

noise

MEASURING
The

that

burst-g~ting~~ulse.

is 2.5 V.$,~#.J,heho

outR$:,#,%WTl,

PC board,

at Pins

profile

oscillator,

and

sided

as much

A low

Wirew

for the

or double

components

pins,

ping.

TECHNIQUES

11, with

oscillator

to the

whic~~~ga~~

are at 4.Qc:M~J$JJhk TDA3330,

is a single

TV

a 4.0 V blanking

on top of it, as shown

super
sandcas-

i.e. vert@}@~&horizontal

to the

spe-

functional

used

best

European
with

and a 10 V burst-g~i&~lse

present,

TDA3330.

CIRCUIT CONSTRUCTION
The

ter$j,t@

a O V ~%~line

Figure
1,sometim@
~~&~~xpression

as used

other

the

is a familiar

It is basically

1. Sandcastle

dia-

CHROMA
TRAP

COMPOSITE
o
VIDEO
INPcT
II

SANOCASTLE OR

CHROMA
BANDPASS
FILTER

CHROMA
IFOUTPUT

BuRST
GATE INPuT

1
T

~~

-:.,?

It is important

that

the

at least 8,0 V and timed

video,

as shown

or too

wide

ended,
and

VCO

video,
and

in Figure

it will

leading

obtained

lock.

The
the

or it can

to separate
an 8-8.5

same

Use

pulse

for

The
The

out~uf~j~%!~iks
~~l&.~?&ync
blgk

1~~1

Pi~~.l&}i~~#

puts

so that,

for

by the

in the peak to peak

trol

can

change

the

control

to 5. OVdc.

black
F~gure6.

and
The

Pin

is
sepSYNC

positive-going
8ov MlN

removed.
The

as the
contrast

chrominance
bar

with

no change

about

from
control,

wave-

of all out-

brightness

1.4 Vto

contrast

con-

6.7Vas

about
Pin

2.OV
19, is

figure

178

..--.

is l.OVdc

the burst-

color

The

18 is raised

The

about

be

put the

DC level

control,

from

is reduced

is reduced

::IE5$., ~

changed.

amplitude.

level

voltage

output
and

to make

constant

output

brightness

signal

voltaaeon
See

are

luminance
shape,

control

PATH

remains

example,

the

5.

completely

hue

when

part of the Pin

upset the black

14 are

thesame

is moved

then

almost

both

is 3.5 Vdc,

one-shots

be called

13 and

output

the

control

the

method

12,

and

co#%$,&!Wehanges
f~msmaintain

s.$,~~:~bs

pulse
of the

con#~s#j:~uration
@~{%er,

can

@uis~~W}etcher

Figure

5.0 Vdc;

the

blank@.~@s

by a coti~~~j~$

func@>~;
see
,},+-,
~,,,
J ,L.,,
:.: ~\a\~,,.

at

when

level:.~~+oW,@vel

t~..wpiies

l&co&jd

~: a~ maximum

iS tq?t~ak%~

Figure4::@n@er

,:@~}~hlNANCE
video,

the

l-his will @$*w#~$%e

first
gating
syh~~,~$lch
would

clamping

Pin 15 be

to incom~fg

pulse

after

a one-shot

pulse

gate

equipment

see

into

respect

in black

burst-gate

into Pin 16. (Pin

gate inhibit).
15 pulse from
level

errors

be generated

sync.

svnc

be present

separator;

ps wide

arated

still

pulse

with

3. If the

to serious

from

a sync

burst-gate

correctly

3. Burst-Gating

6,0 dB
40 dB

The

maximum

7 times
For

output

greater

than

a composite

at Pin

17, due

is about
max

0.35

at the

input
to the

outputs.

to 2.0

Pins

13, and

input

12,

amplitude

control
because
input,

for
the
while

signal
delay

Vpp,
14.

The

to whit%,

line

thi~e

correct

then-g~~~%about

Howevqr,&$~r~

that

~~~~stient

chr@~af~ma

lumina~~~$wponent
colo~~.component
,,J,>,,+\.>l:,}
~!:~ ,$.

Pin

anv

of the
proportion,
directlv
is almost

Figure

to

17.

;%0.5

matchi.~~.~%~stors.

tot,~~wclrcuit

due

is ~bout

white lev~~~t

of 1.0 Vpp,

input

which

requires

the

black

b/ackto

white to b/ack and @&&,,~out

Vpp

doubled

voltage,

the

Vpp
This

2.5 VPP
can

be

5.O Vpp

at

change

in

follows

ACC

that

of the

the

levels

pensated

for input

and

brightness

max

output

of input,

of the
is about

of anv

the

to Pin
aspects

Method

179

of Gating

from

set

it is not

changes.

Also

note

(black

level

out

peaks.

note:
into

15, there

The

If a proper

Pin 17, and

should

of the color

unchanged

5. Alternate

but

be

it is important
up

to

work

automatically
that

with
corn.

at 5.0 Vpp

6,7 V) there

will

upper

for

limit

out
be
the

10 V.

nal is AC coupled
applied

IF. Therefore,
can

positive

Troubleshooting

is

color

TDA3330

different

clipping

saturation
This

to the
note

Video

(positive)
a proper

be video

processing

video

sig-

burst-gate

out,

portions

IS

regardless
of the

IC

THE
The

chroma

by a simple

3.58

about

MHz

*0.5

of a color

thing

MHz

from

(6 dB)

from

The

look

mVpp

our

with

portion

7. The circuit

application

of burst

will

input

circuit

chroma

like Figure

in

100 mVpp

10-200

bandpass

bandwidth.

should

about

the composite

single-tuned

recommended

yield

PATH

is derived

bar pattern

components
should

CHROMA

input

circuit

at pin 22, but anY-

work.

The

output

chroma
IF is at Pin 24, where
the burst shculd
150 mvpp,
There may or may not be chroma

of the

be about
present,

EENpuTTopN
depending

on

(Both

tings.

contrast

the

controls

24, changing

the

burst

pulses.

have

picture

and

saturation

exactly

the

chroma

same

amplitude

N.

control

effect

set~~?[~?s,
.~::+.
a~~~,~

betw@d:%Q@

CHROMA INPUT TO PIN 22

n

GREEN OUTPUT PIN 13 NORMAL

8RffiHTNESS CHANGES SL4CK LEVEL FROM
1 4vTO67
V WITHOUT CHANGING PP
CONmST
CHANGES PP AMPLITUOE
WITHOUT MOVING BUCK LEVEL

NHITE

In

ACKF ~

:m-pfi4-NoRMALo
in time

k.sfiioned

..,:?;

,:*

The

chroma

chroma
(Note:

100 PF performs
it reduces

chroma
and

subcarrier

better

24

than

which

tte

outputs

to the.

capacitor.

1.0 nF on the data

component
to
can

technique

shown

is usually

not

demanding

is coupled

5 by a small

balance

in the

trimmer

tuning

for the

Pin

luminance

demodulator

5 by the

is a fine
able

of

from

Pins 4 and

sheet;
Tweaking

IF output

demodulators,

feedthrough.
reduce

be done

at Pins

in Figure

needed,

but is avail-

application.

5
- 12Vdc

47 PF
b

56 k

1(

47PF

Figure

6. Some

Normal

and Other

Waveforms

Figure

180

8. Optional

Tweak

of Demodulator

8. This

24

lk

residual

Balance

INLOCK0 TO 103mVpD
SINGLETRACE

COLOR LOCKUP
If the

required

Pin 24),

and

lockup
As

a matter

earlier,

oscillator

is present

if the oscillator

is just

noted

chroma

for

scope

adjustment.

be observed

as the trimmer

Lock-in

is about

and

PC board

and

the trimmer

probe

cannot

Instead,

Pin 7. Waveforms

range

to be running,

of adjusting

the

this

the AFC filter,

at Pins 4, 5 (same

is known

put

After

Ori Pin 9.

be put

as shown

Initial Setup Sequence for T~3$~dEvaluation

as

then

the

necting

on the

scope

in Figure

on

ordinary

(Radio

the typical

Shack)

3.58

socket
MHz

2.

provide

anoutput

10has
1.

to be able

two

amplifier.

additional

ltprovides

2.

to drive

of a trap,

and

It reduces

the

another

andan

additional

the

Oseries
level

R into

into

minimum

(7.16

DClevel

of the2N4401,

black

toreduce
subcarrier

the

the

cable

,:t$~

MHz)

bv m~a~~
.,.
.?,.*

has

been

reported

due

2:1 &qtWtIOn

bR.%~!$,&

to
the

as 0.35V,

for

the

sheet)

i~~,$,~@nallV

data

acterized
then

killed

and

Pin2~$}$~T::~

0.7 V, which

trol

lOW.J+O~/fi~#IS

can be fixed

to VG&%,~i~supplies

additional

tn~

~~h$ernal

latch

that

the

mode,
holds

This

is char-

decodes

Pin 3 is above
the saturation

bv putting
current

22 k from
into

con-

*)$:,,*.
,k ..
,,,:r.{*.
$<$

with

relatively

. stood).
becoming

simple

It is hoped
familiar

of functional

application
that

and

this
satisfied

circuitry

paper
with

will

capability
(once

assist

users

in

pH

Signal

In order

to

wav

down).

POTS

Note:

when

See

the

trace

100m

Vpp.

possiblv

Figure9.

Inductor

TDA3330,

to

minimize

POT

and then

#1)

colors

1 and5.
color

Red

#2

balance,

Composite

Video

the Saturation
output

to

Next

minim

POT

on the

minimize

input.

next

all

oscil-

subcarrier

Iookat

the Blue

izesubcarrier

readjusted

Suggestion

adjust

the

from

input.

can
[POT

demodulator

turn

#3to

Saturation

with

happen

from

POT

qualitv.

satisfied

is

TDA3330,

on Pin 17 of the

at the

(i. e., B-Y)

picture

adjust

to Pin 9, until

as possible,

R-Y)

color

POTs

the

Signal

as small

(i.e.,

#l,4,5and6

till

Next

Pin70fthe

chroma

adjust

adjust

U signal

of

Signal

Burst-Gate

less than

signal

Look

the VSignal
and

15

Burst-Gate
ends,

sets

Signal,
Videoln

Pin

connected

Variable

tune

chroma

and

the

#8

Svnc

Feedthrough.

fine

the

on

In (or, alternatively,

adjust

181

of

the sub-

atthe
at

Signal

\vill

the

#6)

middle

POT

mVppto

Adjust

middle

it.

This

-650
signal

10

and

andlook

Capacitor,

probe

set

Video

Figure3.

tilt flat)

picture

under-

the

Signal

of the

(Make

the

2N4402

in making

Width

the Svnc

oscilloscope

loscope

7
range

helps

to the Video

probe

the

todc.

signal

has a wide

This

width

See

from

to make

from

of the
Composite

aoDroximatelv

Put the

the

to pull Pin 3 low (have faith),

and
state so it can be varied
bv the

the

is [n Lock.

drops

Signal

Pin 3, caus-

SUMMARY

The TDA3330

VCO

remove

Pln 3

ns after

so that

the Variable

Chroma

and
1,5 V

~,$?~~~s<pin2to an open
~,Sa$pF~tiOn control.
,.,.>;<~.\.>..\,\.
J!, .::{,:.
.<,.,, .:~:.l>.,

3 to

Puttheoscilloscope

down
5

12 of

collector

POT #8 so that

Note:

Adjust

internally

(I3 in Figure

intermittently

~ils.;$~~?he

-250

signal

insufficient.

bvad~o~whlch

color

of the

#9 sothatthe

50 kfl).

Adjust

#7

the

an

is 1,0 Vpp.

and

(i.e.,

conto

adjustments.

3.5-44s.

Trv

f~im~$hefield

mod~~~i~?~current

seqoent

begins

brightnessconkfd$~~~~~hg.
~J/;t,,,;.{.
\
i$::,.>*:,>.

NTSC

#2

values

POT

MISCELLAN~~@iEMLiNS
It

~X~

TDA3330.

emitter

75 Oc~M~.{$$refore,

supplied

Set

&@r

0.7 Vdc~:~-?~i

can

POT
point

outputs

decoder:

out

,*l,?S$POT
#7 sets the Burst-Gate
shown
in FigureJ
<%
the Burst-Gate
DelaV relative
>. %$,.$4;$
Use a dual input oscilloscope
.:..,..,.:.$+,~.l
theresiduala~~,:
signal
and the Burst-Gate

Thedesign

anopportunity
of the color

of @~$+@*

at t@q, sig-&l

Blue

Board
In and

the subsequent
steps,
components
to optimize

it is necessarvto

benefits:

harmonic

the75

a cable,

Look

Signal

and

m~t~o~$~llow
t&@~~tJM~riabie

Sig~~~~?kis

crvstal.

In order

RedA,~~&~

transist~~.,~~ust

TV

BUFFERING THE OUTPUTS

SVnc,

performance

9 will

1,

pF with

the

a Com~$~te~deo

appropriate
RGB
in order, to adjust

is adlusted.

18-22

connecting

for

to optimize

doing

and

Brightness

Contrast

(POT

this

(POT

is to
#5)

#4 and Hue

are

approximately

Repeat

theabove

qualitvof

picture.

to

POT

right.

Next

sequence

until

~_

w
~

10UF

39 k
.,.

2,,

2N44Q2

l.l.cl I I 1
24

I I I I
23

22

~~> 20 .s~3

{ I
8

1
16

1
,5

1
14

1
13

figure 10. TDA3330 RGB NTSC Decoder Grcuit

182

183

,s

figure

184

.-.-. -.- .-.

_.

__

AN I 020
A High-Petiormance Video Amplifier
For High Resolution CRT Applications

*.,
...i,
.$$:><
.,:
,~$

,..,
1. INTRODUCTION
This application note describes the superior
performance characteristics of Motorola CRT
diver transistors in a stat~of.theart
video
amplifier, In patiicular, the high speed obtain.
able with low DC power consumption is shown
A circuit which is insensitive to load variations
and interconnect methods is given.
Il. APPROACH
The performance requlremen[s for lhe ampllfler are
lhese
Voltage Gam
Rtse and fall (Imes
Outpul
Overshoo[
Load capaclrance
Power suppl[es

20
3 nS
40 V p p min.
51Cmax
8 pF m(n.
60 V,5V, -5V

The emitter followers prov!de a combmed ouIpuI

slgnal from a low Impedance, or s[if~ source This
sllff source makes lhe entire clrtul[ msens!tlve to
load variations and [o different methods of connacl.
mg Ihe vldao amplther lo [he CRT

Bolh Ihe emlrter current and [$c4*r

currenl of [h/s stage followt,\@;:*@a9e
almosl
inslamaneously Comput@,i@aflon has shown
!hal the oarmoon ,@V@am&
than 1 ns. The
~ransconduc7an@,~~~$~tncreased du~ng the
transloon IIti~bv @ding lhe peaking-networ~
R6, CZ, ~~ Add~>lhis network IS verv much
Ill. THE CIRCUIT
Itke a@~&,Jhe rtse time m the probes of fasl
@I~@es
In [ha cascade circulr under discusA. The Input Grcuit
?&\@
peaking network compenmtes rise
Refer IO [he ctrcutt diagram n Figure 1. A Iasi ~+. >,
~S$~ detetiorallon at the collector by speeding
pulse generator IS required for accurale Perform 1$$ ~~ [he eml[[er ~urren[ 0101 This procedure
ante dala The Tekoonlx Model PG502 Is,,@>@od ~~musl be applred wlrh moderation since It may
example of a Dulse qenerator for oDt/muN*.
affect the Iarge-s!gnalwing capabilnv The
mance, versatlbry and pr!ce conslder@on3.$$
m
res(stor, R6, should be equal to or larger than
pulsegenera[or has a rise ome ~@I~&:J@ of ,8
R4 The capacitor, CZ, determines the length of
ns and an outoui ]moedance o~~ohms. A
ome durng nhlch peaking occurs The prodoct

The voltage gain IS ob[alned m a Iransconduclance

amplifter [n the form of a common emlrler,
common-base cascode c!rcutr In Ihls c!rcu(r Ihe load
B.$~e ~ascode Circuit
caoacirance IS tsolated from [he cascode bv a se! of
. ~f~
,.
??. ..i
complememarv emlrter-followers Thus, Ihe
S:P.$~,The
Common.emiRer Stige uses ~
capaclooe Ioadmg on the cascode IS low, wh[ch ~,,$:.
allows operat:on al a ~odera[e dlss]pallon level +t~,,.. >~LTl 001 transistor m a TO-39 package The
,:,$J-,
,,
, ~ t.%i&:
emmer current of 70 mA IS supplled from a -5
Trre emlrler Iollowers are eased at a Clas:.~~~:$.,
V source vta reslsiors .R~, and R5 For ac, only
operating poln[ They conducl only du(!:g$@W
RL al 15 ohms IS operative. R4 and the built.in
transmons, whlla charg!ng or dischW&+~~e~ERT
emltterballast resls[or of 1 fi ohms, determine
,.i
capaclrance This operaoon IS slml~r to ~$~eway
Ihe lransconduc7anceof 01, wh+ch IS [hen 60
highly effclent CMOS Iog!c l@-~&ls@
mNV

of R6 and C2 IS tvplcallv a few nanoseconds.

The trimmer, C3, can be used for lvterumng,
but IS usoally no~ Important and mav be omit
[ed If (here IS lead mduc[ance associated wtth
the palh horn the emoter of 01 through C3 10
ground, use of C3 mav cause mrging at high
frequencies
2, The commonbase stage uses an LT181 7
transistor in a TO117 package. Since the tran
stslor must dlsslpate cont)nuouslv some IWO
Watts of DC power, good healsnklng IS man
datorv The TO 117 package prowdes a hghconduclance [hermal palh 10 a hea[slnk or
chass+s.AI the same Ilme, It adds onlv mmlmaf
capaclrance 10 [he clrcutr

185

Figure

2E. 10 nsec

Pixels

Hgure

10 V p-p

2F. 10 nsec

Pixels

40 V p-p

186

--.-.-

-.

The common base stage has near UmIV currem

gam and ac[s as an impedance Transformer, pro
v(dlng a currem source al IIS collec[or This cur
rem charges lhe combtned collector capdcltance:
of ~, and the emlttet followers, 0~ and 01,
which add up to aboul 5 pF al [he operallng
po(m To this [o[al one musl add about one pF
of strav capacitance A load 01 pull up resls[or
of 430 ohms IS used at Ihe collwtor 01 [he
common base [ranslstor, L) The rise ome al
[hts potm may be calculated 10 be
II=

35.2

.PI.430.6PF=57nS

Th/s value s lm~mved bv the addloon of a

peak!ng COIIof 22PH Theore[lcally, [he rise
Ilme could be reduced bv up IO 40,0 Iwlthout
overshooll by op!lmizlng [he mduc:ance Due 10
Ihe non Ibnearna[ure 01 [he capaclances 10 he
compen~ led for here, dlfferen[ elfecls resuli for
rise and tall !Imes This sltuaoon requoes a com
promise resuloog t~ a prscllcal improvemem of
less [ban [he [heoret)cal transmoo ome Never
theless, 3 ns Iransi!lon omes are ob[atned al [he
colleclor 01 L bv means of the emllter peaking
dmu~ed earner
The LT181 7 IS packaged In a common base con

dlsplav wlrh h!gh comrast and manv oansnlons,

[he currem }n 03 and 04 mav become ap
preclable, causing the oans!s{ors to heal up The
eleva!ed Iuoctlon iempera[ure shifts the b)as
poim from Class B In the doec[lon 01 AB
If Ihe emmers of [hese lranslsmrs were con
necled d[recllv, a OC componem of current
would flow horn the 60 V supply Ihrough [he
dewces 10 ground This pole curren~ would fur
lher heal up [he junc[lons and mlghl lead 10
thermal !unawav In [he c)rcull described, this
souaflon IS prevemed horn occumng through [he
use 01 [he emttter siablbztog reslslors RIO and
RI Using capacnor, C:, prevems delerlora[ion
of ihe dvnamlc operation ot me CIrCUII

One may be tempted to use slower instruments,

such as a 200 MHz tvpe, and correct mathematlcallv for Ihe addttlonal transttlon ome con(rlbuted bv the wope. We do not recommeod
this approach since slower scopes appear to produce wave shape distortions which lead 10
misleading rice.ome values

IV. AMPLIFIER

PERFORMANCE

Figure 2 contains phonographsshowing rise.and fall

omes at 10 V and 40 V peak lo-peak W~@}$!sO
shown are some response cuwes g:,wl~,bV )he
well-known c!rcuo analvsls progr~mSP@:. Careful
..\, ~..,\\ .{:.
modelllng of Ihe semlcooductu$ ~$~cording
10
the theory 01 Gummel and ~@;$e3ulled m good
agreement berween c~~$ef W8 Iaboralorv
generated pedormwd,~,~$fi
addltlon, computer
analysls offers !m@ls, @flch cannel be obta(ned by
A smpler, more pr!ml[lve wav lo avoid lhermal
prac[ical me,~:m>b
problems, IS 10 use only one bias diode, or none
,,
.:~i,, .-:~i
al all Do!og rhls, however, has serious effecls
. ., +
on The gray scale Itneao[v al mid range
Sho~*}Wre
3 are the supeomposed ploIs of rhe
}~~~>,@ita& ai lhe base 01 01 and the output
\.\,, , *
~+{~7@$5cross the CRT capacitance The second set
4. The ou~ut circuit The lT1 839 and
% b~[s, Froure 4. dlsolavs Ihe collec!orcurrem wave
LT5839 !ranslstors have excellent oeak curren[..
~~
I;rn 01 O]-and the combmed emmer clrcults of lhe
handllng capabllltles Thetr emlrler curranrs r~%~+,.+,~
complememarv sel of emlrter followers, The collec]or
,itriu all\,mslantaoeouslv !O rhe base Volw$
>.*
current of 01 shows clearlv the effect of peaking,
Even when supplytng several hundr:~r ~~,,,~
lmtmduced bv lhe emmer clrculr components, 86, C?
amperes of peak charqmq currenl?tia,$bask [o
and CJ Nole that under full swtng coodlllons (40 V
emoter gain holds up will [..* lh~be
PP OU!PUI),the waveforms are not quiie svm.
possible 10 dove more elab~}a~.~$ad conllgura
memcal The effec[ on the [ransolon t!mes O( the
eons fhar a bare caoa$fli}~, Thts ab(ll[v mav
OUIPUIvoltage, however, IS mlnlmal.
ease lnlerconnecl p@i~/~$..$fie clrcuo described

flguraoofl This means tha[ Ihe lrans{s[or base IS

connected 10 rwo svmmemcal low mduc[ance
base leads As IS well known, base-lead mduc
fance mav cause Imslab!lltlesIn common.base
In Figure 1 IS p*@:,~hugh
ro accommodate
configurations To prevem this from nappemng,
a piece of shieldedwle belween the CRT and
~se damptng ress[ors, R- and R8. have been
[he vlde~ ampif~r A !wlnlead Ilne or a stngle
added The value of these res(slors depends ~n
wtre .&ectlon
mav also be used mslead of [he
ihe device bias poIm and !he clrcult Iavou[ If
sh!dt$~i ~ke The clrcuo IS nof only able to
ow!llallons occur, !he! would be near a Glqa
.i. *y}*orafe
Inierconneci ne~works, bu[ also
ha m higher, and [heret!re may no{ be ~een
$: % .~%dle subslanltallv larger CRT capaclrances
on anylhlng bul a sampltng osolloscope Thev ~,~
*:.:.
S$iwlihoul
s[gnlflcant
penaloes tn r(se aod tall
w(II atiecl rise [Imes and ou[put swing capah}~l~:>,
.:\ .*.:*~%,,,* omes~ For Instance, {his cIrcuII IS capable of dr!v
IV Insrabllllles mav be eas(l~derecled w*, .$t<%t.,
. .
Ing b pt wlrh j H ns Iransloon omes

10 all cases, ihe Dresence of additional reactive

ClrCdl[ eleme~!s causes [ne Oulput CI(CUI110 have
reso~ances which WIII cause r)ngtng or over
shools, If the output clrcu!t IS nol properly
damped To rh(s end, a variable res(s~or,RI?, IS
]mcludedm Ihe clrcull When adlusled for crlocal
dampmg, [he waveform WIII look smooth across
!he load mpacoance
In ihe demons[raoon clrcuo, lF\g 11, a 65 pf
chtp capacitor simulates the CRT cathode
capacitance. II IS connecred acres a spec!al jack,
which has been desgned for rhe Tekrmnlx FET
probe, TVpe 6201 Probe, jack and chip have a
combined capacitance of 8PF. The FET probe
mav be used In conjunction wnh Tektronix
=mplmg xopes or real-ttme wopw wnh hndwldths of 300 MHz or more,

187

The example shown n bo[h Fgures 3 and 4 cor.

responds to a pixel.ome of 10 ns, wh!ch IS fhe practical m!nimum for a svstem wnh 3 ns transmons.
When opera[lng contlnuouslv at this ra[e, approx(matelv 25mA of average current flows m each one
of [he emloer followers This causes a slgntflcant rise
In case tempera[ore for ~nesedevices II IS therelore
recommended thal cllp.on heat radla[ors be used.
There IS no elecmcal penaltv for Ibis measure, $\nce
the collectors are on ground po[enllal
Heatsinkng becomes absolutely mandatorv II one explores rhe Ilmirs of the ampl(fler bv operating ar 100
MHz and bevond.

V. CONCLUSION
An ampltf!er was developed whtch meets all needs of
a h)ghresolulton CRT monttor Wh(le practical considerations plaved an mponant pan In the clrcutt
realoaoon, the prlmarv purpose was 10 demonstrate
translsror capabllltv It IS hoped that enough
background Information was given IO allow the
reader to ta(lor hls ctrcult to his sDeclflc oeeds.

~LEO

844724

STUIEO.8407

!622
24,16 2z.

On

BMIC

8P

LfGEMo

ml

v (3001

lCGENO

. I IVlol,

v 1100)
.

V[3
TIME

.1--------

v (1OOI
00000.01

. 1--------

I 5000.00

! 5000.01

*k
30W0.01

.7 500001

45000.01

0.000001

7500D01

TIME

v (3001

60000.01

I 500D.

2000001

1000001

0000001

1000001

2000001

0000001

3 ?50102

7 500D02

1 125DOI

! 500001

00

.0

t
t

.
I

I
I

I
\
1

I
voltage Acro
;\

Ipl

Val!age

1 Catho&
3,s,1

Capt$l,ce

I
I

I
1
I

I
E
=

,~

I
I

Cm,tt,! c,

15

PNP Follo,

20

Voltage

Rgure

Plots

188

4. Computer

Generated

Current

Waveforms

ANI021
A Hybrid Video Amplifier
For High Resolution CRT Applications
Motorola RF Devices haa uasd their unique high
frequency RF semiconductor capabilities and thin
film hybrid axpertise to produce a hybrid video
amphfier with less than 2.9 ns fise and fail time
for a 40 V output swing, This video amplifier provides a low power dissipation solution to a problem that has been limiting the performance of
ultra high resolution CRT monitors: video ampfi.
fier speed. Many of the 1024x 1024 and
1280 x 1024 pixel, 84 kHz horizontal sweep rate
CRTs that are used in CAD/CAM and high resolw
tion graphics appficstions have not reafized their
potential performance because of the speed of
their video amplifiers. Video amplifiers with
3,54 ns rise and fail times often found in these
high resolution CRTs do not provide optimum
picture quahty when the CRT has approximately
10 ns to energize each pixel, A slow video amp
will produce dimmer vertical fines than horizon.
tal tines or may force monitor designers to other
compromises such as a slower sweep rate which
may produce flicker, or lower cathode voltage
which will produce a dimmer picture. The hybtid
described here solves these problems.

SUMMARY

CONSTRUCTION
A. Mechanical
The amphfier is housed in a proven package,
which consists of a plastic housing, attached to
an aluminum heatsink. Dimensions and pin con.
figurations are shown on the attached specification sheets, The circuit uses spec!al sillcon
transistors mounted on best spreaders on an
alumina substrate w(th thmfdm resistors and
gold metallzatimr The substrate is soldered to .t,$

heatsmk vers!on, CR2425, Thes$$$$Y~~age

stvles are shown in Frgure 1., The ~~ctrical
characteristics of these tW@~&pllfiers are !den.
~...~
Iical, The heatsink stvl~~!cs$hould
be based
on ease of mecham~a(~lec~~al interface, In both
cases, the heats~~~~~~~~ound
potential and
should be attache& $JFP?tlV to the chassis m ex.
ternal heatslnk,~or tiechanical stab[itv and heat
conduc~~h to ambtent,
~ If,.
Th&~B~iY4
hvbrid driver can also be supplled
. $<\.\>:4.,*
,Vn %:kb~eticallv
sealed oackaae. The hermetic

The Video Amol(flers, CR2424 and

CR2425, are hvbrid integrated circuits designed ~~(+k~~;h
is designated CR2424~ and can be
for tigh resolution CRT Video Amplifier applic~,,~~~<$~eened
to Mil Std 883 me!hod 5008.
tions. TheV are capable of delivering 40 vo~yl?wk
B. Electrical
peaktmpeak output with overshoot tVpi@,y I%s*
The clrcult uses btpolar sitcon transistors in a
than 5% into an 8.5pf load. TVpical$$B9@<
two.stage feedback amplifier configuration, The
transidon times are 2.6 nsec w\$,$#~p,@idth
output IS supplied bv emitter. followers. Because
of better than 130MHz. Thev &ve~cellent
ef the complementary c!rcuitrv emploved, there is
grav.scale Iinear[tV, sre d: <@~~@,,#*h do not reno need for a load (or pull.up) resistor.
qulre an external load-resl~~~~$
..
.*

.\
)$
?;,:,,.<7
,,;,

-i \\,,*J

\ify

,j::$@ @

Therrn~~*sls
e~n~.$s~e

of an amplifter design IS a very

to ensure amplifier rehabibtv. Heat

,#~~@~#f the most critical factors that deter@~@ how long the amplifier operates

under operating condlt(ons is absoiutelv

necessarv. The infrared mlcroscanner was used
for evaluation of the CRT hvbrid amphfier from
the standpoint of thermal resistance and
operating temperature.
W!th the heatsmk temperature stabi[zed at
60C, the maximum transistor Iunction
temperature was measured at 108C. This is a
verv safe value, especially for dewces w!th all
gold metalization as used here. The maximum
temperature occurs when the output voltage is
either at its lower or upper extreme. Under ths
condition the maximum power dissipation on the
die will be approximately 1.6W. Thus, the
thermal resistance can be calculated to be
30 clw.
Under normal operating conditlmrs [normal
operating conditions means an average picture
content) the hottest transistor wdl dissipate ap.
proximately 1W. Again, with the heatsink
temperature stabilized at 60 C, the transistor
junctmn temperature wdl be 60C + 30 C/W x
1W = 90 C, This is a very safe value for this
kind of amplifier for a long life t[me

CR2424

1235

INPUT

{CASE 714G-01,
Hgure
1. Package

Types
189

Jil

OUTPUT

+Vcc

CR2425

figure

I
7a9

STYLE

2. Pin Configuration

11

P/N CR2424

APPLICATIONS
L Output Chara~risti=
The hvbrid is intended to be used as the final
stage of very fast video circuits. Properly driven,
it can produce continuously alternating 10 nsec
pixels with 40 volts swing and excellent bright.
ness. The nominal Ioad.capacitance is 8,5pf.
Other values may be accommodated, since the
output voltage is supphed bv a pair of emitter
followers, andisfairlv
insensitive to changesin
load capacitance.
Often a wire connection of some length between
the output of the module and the CRT cathode
cannot be avoided. In this case a resonant circuit
is formed, which may cause o~ectlonable ringing
m overshoot at its resonant frequencv. To avo!d
this condition a dampmg resistor must be used in
series wtth the lead inductance. For critical
damping the value of this res!stor becomes

to the output voltage. The ratio between these

voltages is approximately 230. From the above
values, one mav calculate a Iow-frequencv input
impedance of ~15 ohms at Pin 1.
Pin 1 isanintemal
dcfeedback rrode and thus,
as we can see, has a low impedance looking in
from the outside. Pnr 1 must be fed froma
series network made up of a resistor with a
shunt capacitor for high frequencv pre-emphasis.
An appropriate input network is shown in Figure
7 andisincluded aspartof
the standard test
fixturing.

biased at about 30mA. The collector lead must

be bypassed for RF asclose tothetrarrsistoras
possible. For all common.collector (or common
base} circuits, a base resistor of ~200hms
is
recommended. It helps suppress spurious oscilla
tions, which mav occur in the GHz range and are
dfficult to detect. Ffetistors R1, R2 and R3, and
capacitor Cl and coil 11 are a~ustable for
desired circuit gain andresponse. TVpical values
may be:

Wth the input terminal open, a dcleuelof

approximately 1.4 volt ex!sts at this point. Under
thts condition the module output voltage is
approximately onehalf of the supply voltage
applied.

GENERAL CONSIDERATIONS
C. Frequ~~@~ponsa
A. Test Circuit

Areslstor is often deslredat this posttlon also

for protection against arcing. In practice, the op.
rimum value of res!stancemav bedetermmed ex
perimentally durirrg the bread-boarding stage.
Tvpicalualues are 50to 100 ohms. The lead.
inductance mav be artificially increased by a few
tenths of a microhenrv toobtalna
desired peak
ing effect. Any change in inductance will requ(re
readjustment of the dampmg resistance, as
stated bv Equation [1).
A short piece of cable (75 or 93 ohm) or 300
ohmtwimlead, termmated bv a capacitance, will
act similar to an inductance !n the frequencv
range involved. In this case a damping resistor
must also be used.

In the~~r~~~~~d
in many equipment specif[ca
tions fre~~n~
response and tise.times are often
t@3ad as having a ftxed relationship. Theequa.
.<.$?~:~?~uently quoted IS
The input IS driven from a fast pulse gener,at$,$~ ;/
such as the Tektronix model PG502. It is IWOP}S-~.ti10904) = .35 f3dB [21

The test circuit used to evaluate the hybrid

module IS shown in Figure 7.

It can be shown that (2) indeed apphes for the

simple case of a single-pole R.C network, In reah.
ty, wdeo amphfiers have much more complicated
transfer functions, and the above equation holds
true onlv m a very general wav.
Since the module IS dc co~ledthe
input drive
voltage must be+~$ed
such that the driving
wave form is~~te~taround
1.4 uolts. If the
pulse gener%i~
should not allow the setting
of the @?~&?*@biasing current, injected at
mod&~&L#
. .!.,, - ,.~ 1, through a resistor of more

t@~/~<$%Ofim, may be applied in order to ad.

quiescent point of the output
The output terminal of the hybrid Is not short
. ~f~~tiesired
mge
circuit proof. ArrV resistance from this point to ,W.+~Y~$.
either oround or B+ should not be less than @O ...
tive FEToscilloscope probe fitted wltha 100:1
voltage divider. This probe adds 1.5pf to the
load capacitance, bflrrging the total load
capacitance to B,5 pf.
It is seen from Figure$~~:~t*;c,
current swing of ~6\Q~~causes

an input
the Output

voltage to cha:$~~~>+MO
volts. The next plot
(see Fqure 4~~l\atg*Me
input voltage, as
measured,@,aR$@t
port to the output voltage.
Theam~{&iQ;
phase. inverting. The ratio betwe@M@Wvoltagesis
approximately 13.5.
From~:@4~~ove values, one mav calculate a low
freque~v input impedance of -240 ohms at
the RF input port.
Figure 5 Isa plot that relates the input voltage,
as measured immediately at module terminal 1,

Themput circuit contains a series resistor and

capacitor m parallel, which IS tuned for good
response when diving with a 50 ohm pulse.
generator. These components perform a RC
peaking clrcutt.

B. Practical

Circuits

Themodule is best driven froma Iowimpedance

source, such as an emitter follower. The reader
is invited to experiment with a circuit as shown
m Figure B.
Thedrlver

transistor can bean LT2001

190

In addition to the proper gain response, another

amplifier characteristic is of great Importance,
Smcea symmetrical square wave consists of a
fundamental frequencv and odd harmonics
thereof, the preservation of the phase.relationship
between all frequencv components, wtile passing
through the amplifier, must be guaranteed, Ttis
requirement is tantamount to specifying a inear
phase response or, In other terms, a uniform
delay. Amp[fiers having constant group delav ex.
hiblt smooth, monotonically decreasing frequency.
response curves. One must be warv of responses
which show ripple or peaking at high frequencies,
Although sometimes impressive m terms of band.
width, such amplifiers often have poor transient
response. Shown in Figure 6 is the sine-wave
frequency responseofthe
CR2424inlts
test fixture with the input variables previously
adjusted for best rlae and fall times. The output
voltage is 20 Vpeak.to-peak. The sine.wave sig.
nal generator has a 50 ohm internal impedance.
The -3d8 point occursat about 200 MHz, For
40 Voutput swings the -3dB bandwidth is
tvpicallv 145MHz. Actual, photographs of
CR2424 output waveforms driving a 8.5 pf load
are shown in Figure 9,

14

1.7

CRT Hvbrid Ampltier CR2424

12

1.65

10
1.6

6 \
+4
;2
o -

-2

155

15

:
~
=

145
1.4
I 35

-4
1.3

-6 -

1,25

-8
-lo

12

20
VOUT

figure

3. Output

40

D.C.

Voltage

versus

60
Input

Current

I
I
CRT Hvbrid Amplifier CR2424
I

3
\
2

1
0

20
VOUT

O.C,

40

,..,:,?:$;
$.. .*$*

60

20

40

Figure

Figure

7. Test

60

191

100
120
Frequencv (MHzI

6. Frequency

Figure8.

Circuit

80

Experimental

140

Response

Circuit

160

180

of CR2424

200

a.

?:a j.j,

; ,a:

ANI022
Mechanical and Thermal Considerations
in Using RF
Prepared by
Don Feeney
Motorola

RF

Devices

,<1,.

. .. .

t~.

Motorolas thin film hybrid amplifiers are medium power

(0.2 W to 2.0 W power output) broadband devices (t to
1000 MHz) that are biased in a class A mode for linear operation. To insure a proper electrical/mechanical interface with
adequate RF/thermal characteriWca, certain guidelines are
presented for the design engineer to obtain maximum electrical performance and the longest operating life.

One additional note of caution. DO ~Q~/&~~pt

to laP or
file the heatslnk of the hybrid amplif,~J$,NdV%nly does this
void the warranty (considered ~i~,,~{#g
by the manufacturer), but you can Induce su~:a~
cracting during the
machining operation. If you ne@ a s%rter heatslnk, consider
the hermetic package opti~$&~p
low profile package available on some models.,+~owa
RF I!near hybrid amplifiers
are shipped with a ~-~~-h~ sutiace flatness of *.002.
To
improve heatslnki~k,~~~al
grease can be used,
..i,,, 2

THERMAL CONSIDERATIONS

PRINTED ~\~,Ul~

ABSTRACT

BOARD INTERFACE

All MQ@K~#}RF linear hybrid amplifiers are internally

A question that often arises from engineers using ourhybrld
match2@Jo ~!nominal characteristic impedance of 50 or 75
amplifiers is What is the thermal impedance? Thermal
ohmSx@~~:6! the input and the output. This not only reduces
impedance (expressed as OJC) IS a very real and important
t~ e~$nal components normally required to match to these
parameter for the RF design engineer using discrete solld
~hedances
m discrete deslans, but It also stmolifies the
state devices. However, th!s term loses Its meaning in a multi.s*\ e ~iremen~s for Intefiacing p~nted c!rcuit board connections
stage hybrid amplifier, Each stage may be biased at different :: .$,
quiescent conditions resulting in different junction tempera+, $+... ~ for short path lengths, strip Ilne width has I]ttle effect on
*. RF Detiormance.
tures under a g(ven set of environmental conditions. A%J!\/;,
Motorola RF linear hybr[d amplifiers feature .020 diameter
fionally, hybrid circuit design engineers may speak of 6J~$}.
gold plated pinsl spaced at .100 centers. Nomtnal pin length
referring to the thermal impedance of a single traT$isto@Mie
is .460 (,375 for hermetic package).z There is provision for
mounted on a hybrid circuit using their partlcul~~~ssembly
a total of mne pins, but unused pins WIII be missmg (refer to
processes. However, this term has no mea::,~~$o we cuspin configuraflon diagram for the particular hybrid amplifier),
tomer using their product who can only c$~~~~~~e
power
.> ,~.,,,
~>
y,
Viewing the hybrid from the top, pin 1 is Identified on the left.
consumption of the total amplifier.
.$::,,~
~ .:,.
This is the RF input, usually transformer coupled.3 The two
To avoid this confusion, Motorola RF\@$~@s simply rates
adjacent pins are ground connections. The middle three pms
the maximum operating case tem~w~~or
their RF linear
are reserved for power supply connections. Positive polarlty
hybrid amplifiers. These amplifiers a~~$lgned
so that under
un!ts have the power supply (n pin located m the middle.4
the worst case operating co$~tt~,$]!he
maximum Iunctlon
Units designed to operate from a negative supply have the
temperature of any of the tr~si~?? die will be below 150C.
power supply connection offset one pin to the left to guard
This junction temperate+~ates
with our two years of
against inadvertent installation in an Improper test fixture. The
accumulated reliab!llt~ ~~qat~ which predicts an MTBF m
extreme right hand pin is the RF output, and the two adjacent
excess of 142 yea~g>s~~
$:;\ .:
pins are ground connections. All ground connections are
internally connected to the flange, except as noted on the
HEATSINK ~q,~R wYBRID
functional schematic (refer to particular data sheets),
Like all R~t,W;
devices, hybrid amplifiers require heatsmking $~~r~er
operation. How much heatsinking is necEXTERNAL COMPONENTS
essar~,~@, mLch as is required to maintain the case operAlthough it IS not specified as a requirement on the data
at[n&~*rature
at the maximum value under worst case
sheets, it is usually good RF practice to add a low impedance
~bie~ temperature and maximum supply voltage. The presRF bypass capacitor (e.g., 0.1 pF chtp capacitor) located
,+s?t-br
absence of the RF signal IS insignificant due to the
near the power supply pin. Additional decoupling is normally
:$$,$+ss A bias conditions. Reducing the supply voltage will
not required, However, some Motorola RF linear hybrids
$:,...,~~,decrease the power consumption, but it WIII also decrease
require external chokes and capacitors for proper operation.5
the Iineatity. Attach the hybrid amplifier directly to the chassis,
*\$*,.!.
Chip capacitors are recommended. A broadband 30 pH RF
to a module card sidewall, to a small baseplate, or to a mountchoke may be constructed by winding 30 turns of #36AWG
ing bracket that IS connected to one of the above. But before
magnet wire on a Ferroxcube 891 T05014C4 core (alternate
you complete your design, verify that the maximum case
core is Indiana General P/N CF 12001) With an accompa(flange) temperature for the hybrid ampl!fier is within the mannying order of hybrid amplifiers, this choke may be procured
ufacturers specified limits under your worst case operating
through Motorola.
conditions.

193

For Motorola hybrid amplifier model CA2820, the external

chokes isolate the transistor from the power supply. Positioning of these chokes will have an effect on the high frequency end of the amplitude response.
TEST

provisions for adapting this same test fixture for the low profile
package, the bent pin option, and the hermetic package
option are presented in Figures 8, 9, and 10.
1 Pin diameter for hermetic package is .018.
These
Dins will mate with sockets manufactured
bv
Amphenol (P/N 502-20071-572) and Barnes (P/N 027-01802).
3 Except for CA2820, which has an internal DC blocking
capacitor at the input.
st.,\
4 Except for CA2820 and CA2870. Refer to individual data~..,..:~:t~
..>\,. ,,,,.
.,*,$, Jt>
sheets.
f:.~ ,,t!/)}*.
\
5 e.g. CA2820, CA2870
$&!~/.,,,
+$?
~i.t:. .~.:
q
:+
..~-{f,,.+
,.*,.

FIXTURES

Figuresl through 10detail theassembly of standard test

fixtures for Motorolas line of RF linear hybrid amplifiers. Much
of this mechanical information will prove useful to the engineer who is designing one of these units into his equipment,
The details of the test fixture assembly for the CA2820 presented in Figure 7 apply to most of the standard RF linear
hybrid amplifiers (just substitute PC boards, adjust pin spacing, and remove external components as required), Special

1 ooo791
- -+

033 OIA

500350---

- -.-

.200075- .000 ;
z:
.

I
-q
z

*mm

o-~m~

~,:~~+fx$

----

,g$~

194

.312

4 -@

1
.250

&--

II

-9-

313
t
6561

-q_

500
,

. l

k,063+1375-~
r

50 ------~
0RILLANDTAP4

40, 4 PLACES

NOTES

1 And, men,, on, ,n, riches !O1,,, .,

2 Ma1er8al (,38 a,mlnm

figure

2.

, 005

Heatsink Base Plate Construction for Hybrid

Amplifier Test Firrture

Hybrid Amplifier Test Fifi;res

ll

75

,,,

,,

300

.10

MILL SLOT
.065
(BELDW PIN)

.-

AM PHENOL P/N US-625/U (50Q)

TROPOMETER
PIN UBJ-20(75Q)
figure 5. Spacer for Bent Pin Package Option to
Standard Hybrid Amplifier Teat Rxturea

Figure 6. Modifications

195

to BNC Connetior

nun
onoun ~..,,
p,.,,,.,,,,
EIGHT ~N WKETS
AMPHENOL
WN W2.~7?

572

01

10

SPACER ,FIGu~
PIN am
wWEW
WRMS

4,

SPACED AS REO,REO
RN %2 2M7,
572
PIN W?4,&02

PRINTED C(KU(T
~ARD
,F,GRE
,)

CIKUIT ~RD
,F,GuRE ,)

WINEO

FOUR %REWS

SOLDER

440

lHREAD

518

TO K

BMRD

PIN SKETS

,=
!W,
W5
SLOTTED 8NC CO NbECTORS
rOUR 5PACERS 00=
250
LENGTH = 2%

,,

!:

R*%
n

.,

TWO

U=
~

P!.

TRw

>,,

,,,

,,29.

UPAC,,O,S

CHIP

PIN WO%FHI04AZ

(o!

w.,v,4,.,)

000
CHpco

S0 7,

,FIG RE 6,
ID = .,6

ALUMINUM
OASE PmTE
,,,G RE 2)
CHO.SS

Dn

LENGTH

P,.

w050BF!03AZ

,.,

m!vale,,

MO NTING HOLES FOR HYBRID

SECURE WITH 632 ,,< XREW

AM PLIFt ER

,#*?
@%

l@

\.$%\ ..:,,,:,,.
., :~,)},\,\ ,1 ,.

Figure
Rgure

7. CA2820

Test

Fixture

Assembly

YT

NO
,.

10

01

(Case

SCREWS
LENGTH

652

714F-01

8. Text

@tu@

in L#$*e

~ssembly

for

Package

(Case

Hybrid

Amplifiers

714G-01)

TwREAC

SPACSR ,F!G R6 5!
,$. SWKETS
SPACED
.L( ..:,\.
, ., :
BARNES

4S REOL$PED

&OAPIER

(FIGuRE

3,

.,.
%KETSAS
REO, RED
AMPHENOL P!N 502 2~7> 572
BARNES P, N02701802

P,. 027 0!8C2

,R, N,EO

, RC,T

BOARD

,FIGu RE II

L ~
nnnn
Uuuu

n
I

flnnnh

,,

FOUR XREWS

w.

SLOPED
(FIGRE 6)
FOUR SPACERS
B

440

THREAD

58LENGTH

SOLDER PIN WKETS

SO THAT
= ?W
,

00

H
= 2
AL MINuM ~SE

SECURE
XREWS,

10 =
W5

BOARD

BNC WNNEC7WS
-

PMTE

2W

10 =

[FIGuRE

116

21

AOAPTEB WITH WO HEX

632 THRMD,
%% LENGTH

MONTING
HOLES FM HYBRID
AMPLIFIER
SECURE WITH 632 THRUD
,.,
LENGTH

.,,*, ,.,
a

FM:$&.
*
\y,..
;.,,

Text
with

Fixture
Bent

Assembly

Pin Option

for
(Case

Hybrid

Amplifiers

Figure

10. Test

Rxture

in Hermetic

71~-01)

Assembly
Package

for
(Case

Hybrid

Amplifiers

826-01)

196

-._.

----

ANI025
Reliability Considerations
of RF Integrated Circuits

in Design and Use

,,..:<*
~,+$<
.~
,:..u
,.\\:>
~
~:.,
T,lk.
-<l+
,* .,,::i,,

Prepared by
James Humphrey and George Luetigenau

.s. ~t:. ,.$

,, ~.\\\i~.~s
,,.

DEFINITIONS

ABSTRACT
Reliability

is s major

factor

in the

profitability

Rellab[i[fy
of itS prOpOrtlOnally

clrcult

figures

low

prominently

cost,

m the

overall

ture. This complex

and important
strategic
points in the system
Fortunately,
nology,

modern

which

design

draws

function

plc -

is located

manufacturing
from

under

at

gener-

requirements.
chips are the
circuit
Low

for,w~@,~twas
}~;,~,.$~
,.>.,..4.$,.
\ .$*,, ,..

IS a de%&

des[gned

cessation

of abillty

previously

Jb)~~~f*t
mortalities
:~,@ LYefime failure rates

most vital elements

of the RF
noise
and
distortion
require

an Item

WIII

task
$~lsfab~rlly
for a speclfled
used?.~;$he
purpose
Intended,
and

specrfied
f~ctioti..wlthln
the area ~f~n,~qst
.x,.,,.
(a),::O*~&
arrival

ated by mil!tary and apace activities,

assures
a degree of
reliability
whtch
is compatible
w!th the most stringent

Transistor
integrated

conditions

Failure

tech-

resources

to the

PerfOrM
a defined
Iength of tima, when

RF Integrated
reliability

IS related

Failure

and

extensively

the

~sk
,,, ,,.
$.,.. i\:*.?+
.~) ...**..,
pr:~b~~ty
that

R = Reliability

of CATV

Systems.
In Spite

.1},
,:::;:.:*<
..
.+>,*
,, ,t,.>+~

to operate

to perform

esfabl[shed

Iimlts

In

(random)

~~~~~ [@jEndoffe{wearot)
,J.,t,,>
state-of-the-art
franslstor
structures
Gold metallizatlon,
s.:.
thermal
equilibrium
by means
of diffused
balanclng
~;??t$?, 4>
~TBF (Mean Time Between
Failures)
resistors,
as well as automated
process
control
have .l~.$e$, R
The total measured
operating
time of a population
of
resulted
in translator
Itfetimes of over 100 years
,?*.
~il:.::..
~,jJ,i\,,
equtpment,
divided by the total number
of fatlures within
~.*$*>
the population
during the measured
period of t[me,
One of the Inherent
reliability
advantages
of ICS IS the ~~
reduced

number

character!stlc
duction
paths

of interconnects

The full bene[tt

of t~ts

is achieved
through
the use ok.&j@
con.
(n conjunction
with gold ,wi~~.,b~dlng,

mechanical
and

safe

conditions
ment

construction
Operating

The

goals.

scann[n&$e?~6
.,.,:. $;*Y$

or abnormal

verify

all

practical

the

ach!eve.

rellabil!ty
In order
t&. av@ld
control
the elec~fy~~$ti~rmal,
ment

surroundin&$~~$

respect
ha~$[~eenmade

RF

for

it applles

a normal

distribution

to failures

resulting

-tlm

R=e
Counteract

fhe best

lives,

and

wearout.

problems,
the user must
and mechanical
environ-

IC Much
progress
by the equipment

e-At

of
Where.

In this
indust~

R =
f=

Reliability

.=

MTBF

MissIon

or probability

m hours

A=

lNTRQQdQWON
.... ..

of success

bme m hours

,,~~$,..,.i.
,,

.>,,~\,.,L~,

of
from

BASIC RELIABILITY EQUATION

;;!

s~#&s~~*ay

value

generally,

!;,

Abuse

mean

integrlfy

temperat~<$$ktimder

infrared

of design

guaraw$~~gbtural

Life

Average

Perhaps
the single most dangerous
en$~~~~.high
reliability is excessive
heat, Careful, co~#UJ@rl,Wed
circuit
design
coupled
with
thermally
1~~~,~$
StreSS.free

aiure

ate

failures

hours

R~~b\\t~
@~,*nl

~onslderatlons
m the operatton

are
becoming
of CATV Systems,

increasingly
requlrlng
an

,,.;.a~~r~t!on
of mllltary
and aerospace
reltabllity
tech\,
,.i 40QY
Into fhe CATV business
Market
surveys
show a
.S*~i a~~ge
number
of MSOS
and consultants
consider
re.
*:~ , .\.lt.,.
!,?, ,., Iiabllity as a malor Item in equipment
selectlon

,.~
,.,..
..,*,
>:,:
~..y
.
\~:$+

A def!n(tlon
of major rellablllty
terms
IS !mportanf
with an Introduction
to mlcroc!rcult
reliability
tools
hardware

SYSTEM

RELIABILITY

1. When
the

components

are in sertes,

components

will

result

failure

in failure

along
(both

and software)

An overview
revolved
wtth

d!scusslon
of Phys!cs
the die and Interconnects

of

ConstructIon
must be pre-

Then:

sented.

197

=-

m.,,,,v
i,,,,,
-

=
=

R, xR2 xR, x---R.

i, +Al +A, +---A,

of any one of
of the

system

When

the

dancy)

same

components

neglecting,

device,
the
requirements:

are

for simplicity,

switchover

in parallel

(redun-

RELIABILITY

the dects!on-making

function

and

the

fad

The

safe

CURVE

following

operational

curve

represents

the typical

conddion

of

relabtl!ty.

Infant Mortallty Plus

Random Failures

/1=

\
i=

Random Fa!lures Only

4
I

Failure
Rate

,:
RELIABILITY
The

mllita~

has

PREDICTION
put considerable

the study of reliabihty.

IS MilitaW
Handbook
E/ecfronic

Equ/pment.

develop
matlcal

failure
models

mtldaw

agenc!es.

ALGORITHM
money

and

t{:%~~~

tl@<$~#$

One ve~ useful mllltav ~f,@e%t

2176,
Re/fabi/ify
Pr@~~of
This

handbook

sw~,.~w

@?h~~$fl\eraction

FAILURE

=
+

l,+A,L
ZL..N,

Ap,n,,

RATE

MODEL

+Il~,N,,
(Attached
(Package

~,

(Substrate
components

contribution)
contributions)

contnbutions)

to
Base

rate predictions
by the %$~~~~!? mathe
based on yeara of .@q
@llect!on
by
A discussion

BASE
i,

failure

Failure

rate in failures/10

rate

due

to the

and film processing

Failure
rate contrlbuttons

of

network

complexity

area which

hr

substrate
due

and

to

substrate

includes:

(a) Number
(b) Number

ofleadterm!nations
of film resistors

(c)

ofd,screte

Number

chip

dev!ces
(d)
The

Typeoff,lm
sum

of

the

(th,n versus
failure

thick)

ratea

for

each resistor as a function

required resistance
tolerance

of the

The

device

sum

of

the

attached

fallUre rates for semiconductors

and capacitors
The hybrid
to include

package
material

failure a@usted
and style

198

_--..

...-

199

Not much

Ieeway

existsin

reducing

the current

density

(unless f. IS sacrificed).
Changing trom aluminum
extends the fife at least by an order of magnitude.

Comparing
hybrid versus
show the following.

to gold
At high

temperatures
the difference
is even more pronounced.
At 150C,
thetlme
tometal
failure for gold metallizafion
microwave
transistors
is m excess of 1~ hours = 114
years. While this number is quite comfortmg,
at Ilberty to treat the subject
of transistor

StafiOflS
Indexed

chip

and

the

IC substrate

mounting

are used to eutecbc

collet mount the chip to
Ieadframes
Tight control of pressure and scrub

sequence
result In defect free attachment
may employ other methods of heatslnking,

Although one
e.g. berylllum

oxide
substrates
for
mechanical
complexity

part of the circuit,

and the reduced

optimal

presently

circuit

advantages

layout

resulting

from

reduction

the
(n

Foreach

capacitor

connects

are eliminated

transistor

parts of hybrid circuits IS the

ablllfy to reduce the number,

more conventional
in the mechanical

tems have drastically

improved

An

schematic

analysis

Hybtid

of

Amphfier

the

WIII Illustrate

MIL

byscreemngcomplete
advantages
of hybrid

will

standard

oftwolnter.

transistors

=
=
=

hybrid jumpers
pins

be

the

same

for

~~:~ai?~
.. ,* , >,\\..
..> ,>*!..\,,,.
~> ~,:.l.. ,,}

building 33dBO$Q~i~
circuitln
hybrid f~,@$:<
~:.$ ..~,,....~
:$~i.
,,,t
=
24

Adddueto

transformers

of four
to the

and the connec-

12
100

,tJ;~,**+~:%9, :;
a<:t..J

a.

= ~>,,
~~~$,
y.
p~

,N+$,$p 12? Additional inter-+,*,A:i.i

,~.. \>,.,
connects per
~.
.X$$>,*,
33d B function
..~,s
-...
? Q
2174.$$%
dlacusses
the reduction
In

Handbook

of prm~@:$J@lt

boards

as a direct

the holes re~l~~,<,,,#ighty-one

additional
volvedm ma~~~ohed
lscreteampifier.

multlple

holes

are

of
ln-

..:.

retiabll!ty.
the

three

~~+,\

Adddueto
chipcapacitors
Add due to resistors

reliability

approaches.
Constant
and metallurgical
sys-

on

amln!mum

Transformer
interconnects
hybrid or d!screte.

INTERCONNECTS

circuits over
Improvement

used,

body

Less dueto

and test them

of the major

of

minimum

can

to the solder joints at

connection
to the reslator
hon to the PC board.

Less due to active

One of the most Important

interconnect
system. The

For each film resistor

used, a minimum
interconnects
areellminated
corresponding

Adddueto

minor

temperature.

control thequafity,
functions,
is one

used,

one

corresponding

The increase
In lnterconnectsin
In discrete form over the same

the added
freedom
of

outweight

transistor

techniques,

the PC board are etimlnated

2

one!s
not
chip heat-

Automatic

each

interconnects

s[nklng too lightly. A proven method for removmg

heat
while at the same time obtamlng
a solld mechamcal
mount, has been toemploy
aheatspreader
between
the
stl!con

For

discrete

33dB

the point.

Having @einte#Connects
made early
ing <~~ue~~e,
before the subsequent

in the manufacturseries of tests and

ln~>;%ns,
*U*JM
. ..5 . .>

on end

has beneficial

.,,,

influence

1(

1.

Q,

Q,

R
R,,

R,,

T,

::

200

II

R,.

R,,

R,.

R,,

equlPment

Thecomplete

functional

system

mclud!ng

Advantages

interconnects

Istested,
screened
and Q. C.sampled
many trmes before
[t even meets upwlth
the PC board in the manufacturers

Easier to control

Interconnects
D,e

Disadvantages
Heatspreade!

Jumper Wnd

II ,

More deformation

at bond foot

Hard to form loops

,,,.

Ptn

MOUNT

transistor

connections

heatspreaders,
are soldered

the substrate
sutiace.
tightly confrolied solder
Due

of Gold Bonding

More expens!ve

Capac,tOr

process

COMPONENT
The

D,e Bond

Bonding

with gold die and substrate

Strength stable with time/temperature

Malleable not subject to cracking

subsystem

%Ider

of Gold

Compatible

chip

to the fact that theunits

atmosphere

andthoroughly

In the production
greatly reduced

capaclfors

to the metalllzatlon

This process
reflow furnace

process,

and

preprocessed
cleaned

on
Number
of

m a

In an inert

and respected

workmanship

pln

pattern

IS completed

problems

early
are

BONDS

,i:,t

Wire bonding
Alumtnum

was a maJor rellablllty

I:,\,, e,$,
~.)$,.>
~
,$.<::.:,!>;<\;\i.

Issue for years

has been one of the most widely

used bonding:<x
Strength (Gram)

systems m the hybrid industry for many years. T,he @~n

reason forth! sls that ultrasonic alummum syst~s
bond
at room

temperature

and,

other hybrid assembly

hence,

processes.

do not ixrf~
,
.+~t:,~+.
,/.!.i,:
..

Gold thermal
compression
ball borf#j.@&s
rel!able standard process (n the se@~*,#ctor
for years

However,

temperatures
hybrids.
prompted

have

therequire@,~]@+
kept

The recent
cha~~v~
to all
thedevelopm~to~:
~compafible

Com~%~-+&th
L:~>:~t
**

been
a
industry

30@ CbDnding

th(,s+!~~!~ue

ature gold wire bond)W.~@~@*

which
;~
.*!.,
elumlnum.
,{
~<,Y.J,t\;.t}.A:,.
Advantage
of ~,u~um
Bonds
:,>..
Low te~$$atur&
process

with

Out

of

Strength

mOSt

gold hybrids
10wtemPer.

Versus Time on Gold Versus

Aluminum Wire

by far out-PeflOrms
T=

150C

Al dte metal
Strength

,Nh

wed
~Fy*tO
loop (stiff)
}LS (t.
,+PDi,~vantages
of Aluminum
eonds
~>\
...(~~,i.1..<j Degrades with time /temperature
,.:>1,$
\\:,*{,.
Krkendall
voiding
**..., *$\,
.
Intermetall!c formation w[th gold
~~ 9.1,
\i:.J
Brittle and subject to cracks
~).
D!ff!cult to screen
Difficult

Alum,nm
1
T!me

to control

201

..

RELIABILITY
Following

ADJUSTMENT

IS a discussion

of the

in MIL Handbook
2 17B.
influences
on hybrid circuit
TEMPERATURE

These
relate
reliability.

ADJUSTMENT

to

great rellabll]ty
improvements
ing the case temperature.

can

factors

the

expansion).

external
The

FACTOR

Operating
temperature
is one of
factors
inrel[abihty
Ascanbe
seen

(c) Where there has been an extended

mterruptlon
[n
production
or a change
tn Ime personnel
(radical

FACTORS

w adjustment

n,

the most
Important
bythe
curve shown,
be obtained

factor
and
for

over

was

be expected
have

much

factor

used

module

10 can

controls
as

Thlsmaturlty

by lower.

has

Failure Rate Multiplier

of

dlt[ons
extend
duction

as

Is extremely

400,000

shipped

volume,
product

The

The
TCsse

80

MILHdbk

60

placement

refined
process

that

feedback

does

thermal

Introduce

CATV

cast[ng
the
Still,

we

first

have

to f[ne

tun~%e-.~.
+?.
~$.
,y;rw,!:,.>i;}
*
,,,f:
\\,;,,l.
.,,:.:?,),
...... -y,

fit ~@?&>~M@?table

deslgns+~e;}xtremely

RF IC ~om
widei$p;~~ons

a swing

the

time

.~a~~~~o?

!Mre external
In SySteM

environmental

effects,
whlchwlllcausen,
and 5,0

40

The user
laboratory

1/
20~
246

10

20
This

curve

shows

case temperature
rate as the same

that

Th!sad]ustment

a hybrid

c!rcutt,

of 10@C,
has four
circuit run at 50C.

ENVIRONMENTAL

ADJUSTMENT
factor

conditions

is based

fo~~~~~+~tofallbetween
1.0
:1,3hy\:,y\,
$.?* ,,%8,,
, ..:+.,
must strive $@$eep %e components
as close to
zero as p~)$t~i~.
~$.::i~.p.,
~{:>%

QUALITYAD&$TM@NT
.,.:,, ..,,.

mental

IS where

and

Important
In protecting
environment
condlttons

This

question

Mechanical

2176

mdust~

since

120
100

can
pro-

s
)$ \
?:t~,,,.<y
,,;,
the IC Optimum
re~t,.
*$,
depending
on firn$$ ,, @fi
\.iy

Since

defect
analysls
and
and elimlnate
defects.

con-

per!od

continuous

modules

1970

In

until

This

of

[mportant.

CATV

constantly
Improved
and
Iiabillty
IS an evolutionary

Due to Temperature

to apply

slab! l+zed

6 months

operating
times

FACTOR
on the

the

service

that the part will be exposed

failure

environ.

FACTORmo
factor

based

on the

quallty

T~q:,J&vels
;&~DWffbatlon
*~qU~rements

,,,:~
t
,,>,.~i::,~.,
,,,,
!%. \&Ft

take Into account

different
screening
levels.
levels and quallty
conformance
InspectIon
for the specified
class.

to during

operation.

grade

of the @duc~*Th[s
factor mod!fles
the rel!abllity
levels
by the&r~:ent
qual!ty levels specified
[n MIL STD 883,
Te$)v&tH6ds
and
Procedures
for M;croelectron;cs.

at a

n,

is the a@trnent

MIL STD 883

Class

MIL STD 883

Class

nQ
0.5
f .0

Vendor Equivalent
Class
MIL STD 883 Class C
Commercial

with Screening

Commercial

(No Screening)

A study

of the MIL

STD

883

5.0

300
500
750

Quality

Requirements

allow

a ve~ important
discussion
of cost versus
rehab!lity
As
could be expected
the test, manpower,
equipment,
bme
and pape
(ork go up rapidly
as the MIL STD Grade is
mcreasec
A relatlve
plot of this relat!onshlp
IS shown
below

CostVersus

Reliability

Increasin
W~lUre
rate predfcted
by this mechanical
model can
b$expected
to Increase
bya factor of (n~ = 10) under
any one of the following
conditions,
(a) New

device

(b) Where

ma)or

costs

Average Grade

in md!al production
changes

m design

or processes

have
Lower Grade

occurred

.-.

Upper Grade

Many

of the

MIL

unimportant
cost

versus

supplier
is wilting

Standard

Military

in fluencmg
refinability

requirements

CATVreliabillty.

curve

can make choice$

to pay for

IS real
astothe

CONCLUSIONS

seem

However,

the

and

the

type

of rellabilityhe

Many

equipment

ments

such
as
rellabll!ty

EQUIPMENT
It takes

a massive

manufacturing
volume,
have

caused

us

Many

of

the return

the

SCANNING

and
to

in order

for the

CATV

performance
constantly

revested

dollars

to meet

are

for

for

very

Th!s

X-RAY

capability,

to make
poslt[on

equipments

of secondary
emission
acter!st[c
energies.
The
elements

may then

VARIABLE
This

purpose

x-rays
relative

FREQUENCY

reliability

leverage

the

which
quantity

of

due to

to control
volume

quality

by screemng

of complex

standard

functions

are

A monometallic

Magnithe SEM,

allows

chemical

system,

I.e , gold

,<,

*;),(...*

dl~fit~<mi[zaf[on

and

are opt[mum

fo:,$e~@Wli&*
.,, ,,}, .c~,..~
...<.
Reliability
can betmproved
b~:a*@duality
cost tofhe
module
process.
This Incr&e@;
Cost may easily be
due to the lowefiXa[lu&&te

us

com-

by detection
possess
char.
and Iocatlon
of

on the CRT.

VIBRATION
test

of determining

of vibratron

(b) Abilrty
(c)Large

returned

is a:compllshed

be displayed

IS a destructive

predictions

of Interconnects

gold wtre bonding

of surface

of the SEM,

to determine
This

of fermass[ve

for

for

h!gh magnification

IS a feature

a microprobe
of a sample.

217B

future

ANALYSIS

which

circutts

are available
today both m equipof reliability
and in analytical
tools

Handbook

is subjective.

condttlons
not available
w+th optical
methods
fications
up to 100,000
times are possible
with
DISPERSIVE

The

required
the

MIL

(a) Reducbon

MICROSCOPE

allows

Hybrid

the

industry

standards
reinvest

on investment

ELECTRON

instrument

investment

requirements

quallty

which

This

capital

reliability
tools
for evaluation

in the specified

which

the

effect

IS performed

for

on component

frequency

the
parts

range,

*IL

Handbook

tronic

217B,

Re/iab;//ty

Pred)cfion

of

Elec-

Procedures

for

Equ!pment,

MIL Standard

883,

Test Methods

and

M!croelectron!cs.
bonds

INFRARED

MICROSCOPY

The

to

abil!ty

atmg conditions
a new product

,+.. ,.~.:

examme

,,, ,.$.
...\
a circu!t

thermall;~nd$t

Resolution

of O.~.f~~~ch
.:,~<, \
,,?..,+....i,,,.~

Flah{e,

175,

Reliab!llty

of lt,M/crowaves,
oper-

R Y

isabsolutely
necess~W $
~j,*tieslgnlng
or testing
a new ~r$~k~
The infrared

mlcroscanner
IS used for evalu*t!~&new
from the standpoint
of therm:l
~~:~[%ce
temperature

MIL Handbook
Handbook.

.)

or eutecbc

products
andoperat[ng

can beachleved

Scapple

preach
no/ogy,

and MTF
July

and

Device

The Long

F Z,

to Hybrid Thermal
October
1973

and Shor{

Keister,

Simplified

Design,So/idState

J R Black, Electromlgratlon
Failure
num Metallizatlon
for Semiconductor

Modesrn
Dev!ces,
Number

of fhe

/EEE,

Volume

57,

Ap.
Tech.
Aluml.
Pro.

9, Sep.

1969

C M. Ryerson,
S.L
Reliability
Notebook
September
George

Data

1972.

:eed/ngs
tember

Webster,
Volume

FG
Albrlght,
RADC
11, RADC. TR.67-108,

1967.
G

Luettgenau,

slstors, International
holm, 1972

203

M/croe/ecfron/c

Microwave
Microwave

Power

Conference,

Tran.
Stock-

204

ANI027
Reliability/Performance
Aspects
of CAW Amplifier Design
Prepared
Michael

by
D, McCombs

,,,
$.J\..
.~,
*,,~
,,,.?.
.,, ..~e:,
,,)),.>,

ABSTRACT

The ulttmate test Is+5,$$&w:how long a part operates in

The reliability advantages to be offered by the RF hybrid
the field without f~~~ ,~e best way to simulate this IS
ampllfier as used In CATV applications are discussed.
The ac!lve part of the hybrid ampllfler is the transistor.
by means of a~g<~~~~~ Life test data IS Included as a
Metalt(zatton, ballasting and ruggedness
are rellablllty
means of de.mofi~~f.atlng the results of a careful design.
,r..h.?t
related factors that must be considered by the devtce
:...,,
s\:>,$,..,,,,,,.
engineer when deslgnlng a high performance CATV tranIl. Wfi~~$S
RELIABILITY
sistor Vertical and horizontal geometry and device
* ~t,
distortion mechanisms are performance related factors
O~,Q@k~ion
could be that reliability IS something that
that must also be taken Into account The Interrelation
cam:~ost you money f you dont have (t. The dictionary
between these factors IS exammed Life test data is then
.:.$~defm$s rellabiilty as the quallty describing that which IS
presented to Illustrate the advantages to be gained by ,+,,l$.),~pendable
or honest To bu(ld honest transistors and
careful device design
i!: +$, amplifiers IS a noble concept but one which may be
:,4,,.,.,
. ~ dlfflcult to measure So In the evervdav sense, rellablllty
.4:>>.,
,.
,,,,.>,,,
,,>p,>
*,. ,.,
.>.:.\.
IS a somewhat abstract Idea that IS difficult to describe
\\\t,
quantttatlvely.
In
engineering,
however,
rellab!l(ty has an
1. INTRODUCTION
~.,
exact meaning
The cable televrs!on system operalor buys$~julpment
Rellab!l(ty is the probability of a device perwhtch he knows haS demonstrated a cer@ln Wlmum
forming Its purpose adequately for the pertod of
level of performance, or n other words.mi~ent
that
time Intended under the operating conditions
meets hls speclflcahons.
If he quesk@&~&la
perform.
,.. ~,,,
encountered
ante he can run various electrlc~.$:,~s~g~
to
check It
, ,,$,4<$.
,.f::.,;.*,-,.
.. , ?-:.s.,
Another quest]on that we woul~+~(~,fb be able to answer
IS, how long WIII hls equf~fi~t,:~perate
before It falls
costing hlm downttme and;e~>~~ Thts IS the question of
rellablllty and to unde,k@t@@.t@sIt IS necessarY to under.
stand the factors t~hl
~:..,,~,>.,..Into des)gnlng tor rellablllty.
.,:.\
,,.,.,)\$ll/,,i.

When an amplifter IS des!gned for a certain level of gain,

It may happen n practice that the gatn is less fhan that
called out in the specification
In certain cases this may
be acceptable lf the ampllf!er turns out to be very rellable.
However, another ampllfier. which suppl(es the full gatn
w!th ease, may breakdown in operation because Its
components are being taxed to their Ilmits This IS where
rellablllty enters the p{cture It IS possible to achieve full
performance
and still have state. of.the-art rel(ab!llty
We said that rellabihty is the capability of equipment not
to break down in operatjon. The measure of an equip.
ments refiahlity, then, is the frequency at which fatlures
occur in time. A failure is a malfunction which causea the
component
to violate the requirement
for adequate
performmce. The frequency of such failures IS called the
failure rate. The reciprocal of the failure rate is called the
mean time between failures or MTBF.

The primary Qull%%lock

of a rellable CATV ampllfler IS
the RF (nt@@@edclrcult This concept possesses many
advantag@&~q@ the PC board discrete design mcludlng
a red~~~:~{u~ber
of Interconnects and the ab{llty of the
ma~~$~~urbr
to effectively
test the system before
d~~w.?b
the equipment manufacturer
.#Wofi@ one step further, the basic constituent
of the
~~+Jn~#gratedclrcult IS the transistor Itself It IS in the design
,h~:i~,s$....
,Yyt:
+$+<,QT
this transistor that the ideals of high performance with
:$* fellability can be effectively realized
,:;;!s:.,>,.
~,i,,\ J},
:.1..
..?

A = Failure Rate
1
.
= MTBF
A

205

Ill. HYBRID CIRCUIT RELIABILITY

ADVANTAGES

Refeming to Figure 1, it is seen that there are three basic

types of failures: early, chance and wearout failures?
Early failures occur early in the life of a component and
result usually from poor manufacturing. These can be
eliminated by a burn-in process.

The hybrid circuit is the heart of the CATV smphfier. This

assembly must perform its duty while experiencing
a
variety of electrical and environmental extremes. If the
hybrid circuit should fail, then the cost to the system
operator is high. For this reason the hybrid circuit should
bean extremely refiable piece of equipment.

Wearout failures are a symptom of component aging.

These types of failures can be eliminated by either
replacing at regular intervals or by designing for longer
life than the intended life of the equipment if the components are Inaccessible.
Chance failures occur at random intervals and are due to
sudden
stress
accumulations
beyond
the design
strength of the component
Since the other failure types
are relatively easy to ellminate, performance reiiablllty
should be determined by the chance failures
For chance failures only, reliability may be expressed
the exponential relationship

One subtle advantage relates to the wear OY&?~k$~~v

components.
Replacement of a hybrid cir~~~~dhs
replacing every ampllfier component whi~~:$~g
the
clock on the entire amplifier as far as w~~~~~is
concerned Replacing a component (n a~sc~~~
amplifier
does not. All of the other discrete ~,m~whts
continue
*,...,..
to approach their wear out fife.
.$,.,,?jk+
.e,,)5

by

R (t) = e.it
where k IS the failure rate and t IS a g!ven operating time;
t must never exceed the useful I[fe of the device The
derivation of th!s reliabll!ty expression is found in the
Appendix.
System failures are caused by component failures When
components can fail only because of chance, the system
will fail only because of chance The design engineer IS
responsible for the reliability which IS characteristic of his
equipment If he desires to reduce the number of chance
failures which occur during the useful life period of his
equipment, he must keep several key points [n mind>

The metallizatton
system ..~~ttf~
hybrid is another
advantage The gold meta~$~~~~which
is used for interconnects on the hybrl~+~wlt
allows the designer to
have the high conwt!vlt$.
of gold for use in tying to.
gether the vario&.,ibponents
of the circuit, while
having the ad,~?~~k reliability advantage of a mono
metallic golQ,s~$te@ in wire bonding from the transistor
to the hy@M.SE&~ though the hybrid circuit utlllzes heat
slnkm~ffo ~%ce heat buildup, any bi-mefalllc interface
WIII b&+~~uscept[ble to failure due to intermetalllc form@}~h, %ese
gold-aluminum mtermetallics are more
b~${le~$han the parent metals, and they also are sus,{,,:,,~%~t%le to void formation due to the faster diffusion of
~.~~~~hlnum Into gold compared with gold mto aluminum
,x, !K!rkendall Effect). If a hybrid circuit is manufactured
using dte with aluminum metalllzation, It IS certalrrly preferable to use alumlnum for bonding. This is because the
gold. aluminum Interface will then occur on the substrate,
away from the heat of the transistor This is Important
since the formation of intermetalllcs, AuAl~ or Au, AIz.
IS accelerated by temperature
However, these interfaces, even though they occur on the substrate, are
nonetheless sensltlve to weakening Which intermetallic
compound IS formed depends on the amount of gold
available in the bonding area. If the gold IS thin then
Au, AI, will be formed. If the gold IS thicker then Au, AI,
will be formed The end result IS the same; votding and a
weak bond which eventually Ilfts The entire process
can be accelerated by thermal cycling whereby cracks
are formed in the brittle intermetallics
Data presented
later !Ilustrates the comparison between failure rates due
to bond lifts In alumlnum and gold systems.
Another advantage which hybrids enjoy over discrete
designs is the reduction of the number of interconnects.

++:$Mn
com~nents
to accept overstress;
the
~~>~~rmal operating point should be well below rated
.~,.,
\
~+1~$~Z values, including temperature.

An interconnect IS a potent!al failure point. Reduction of

the number of these points will result in a more reliable
system. A calculation of the additional interconnects
required in a typical discrete amplifier over the hybrid
equivalent shows an increase of 127 interconnects
in
the discrete version? Figure 2 summarizes hybrid life
test data.

~$ 2. Provide good packaging

with adequate
heat
sinking.
3, Design w!th as few components and interconnects
as possible.

206

..

....

So it is apparent that the hybrid structure la inherently

more reliable than a discrete assembly. But the heart of
the ampfifier, be it hybrid or discrete, IS the transistor.

Reliability

which is not so heavily doped so that the reslstlvity of

this layer is higher than that of the substrate. It is the
configuration of this epitaxial layer that is very Important
to the performance of the device. It is this layer that will
form the collector of the transistor
There are two
parameters of the epi layer that can be specified by the
engineer. One IS the thickness wd the other IS the
resistivity.
The reslshvity IS chosen from operating
voltage cons(deratlons
The transistor IS Intended for a
specific purpose and presumably the voltage at which it
WIII be operating IS known If the device WIII be ~j$a~:d at
20 volts in an amplifler, then the collector~~~%n
voltage of the transistor, BVCEO,should befhlg~?
than
20 volts to provide a safety cush!on. Th%@Wmenon
that occurs in a well-designed translst@IQt,&Wakdown IS
called avalanche Th!s occurs wh@ $$~~iciently
htgh
reverse voltage is placed across,, a:@,@~tiction A field is
formed across this junction a@G~/7~~
are accelerated
across the field. When the a~~~~ti voltage equals the
avalanche voltage a multi~,l?~tioneff ect occurs In which
atomic bonds are brok~~~$w.ihe
Junction breaks down.
This IS the collecto~~[~a~own
voltage and (t is propotiional inverselyt&~,qdoplng
level of the collector or
epl layer By s~~~~~~
epl mater) al, then, the designer
sets hls voltage ~~?atlng I(mit.
.~$~:$
The oth~~~~$:~arameter
of interest IS the thickness of

Data at 95C Case Temperature

m
Figure 2. Hybrid Circuit

IV.

RF TRANSISTOR
CONSIDERATIONS

Life Test Data

DESIGN

The performance
which can be obtained from the
amplif]er is determined, In the end, by the transistor Not
only must the transistor provide performance, however,
it must prov!de this performance for a reasonable length
of time. If the transistor fails, then the hybrid falls and
cost to the system operator IS the result
When the transistor engineer begins to destgn a dev(ce
for use in CATV amplifiers, then, he is faced wtth two
main requirements The device must offer a certain level
of performance and it must do its job rellably We WIIInow
investigate the RF transistor and the considerations that
go Into Its design

the la&;. ~~as been found that epl thichess

is closely
tiad$ln:@.&oth device rellabilltv and Performance
One
pa$~re(er that IS commonly, used to describe high.:~freqoency
transistors IS fT. Th!s IS the galr-bandwidth
r t~oduct
of the device or the frequency at which the
,,,,,
:*..,:i>,
short clrcu(t current gam h,, , equals
*,. y common-emitter,
1. Starting Material
i~,ak,;+~ unity. A high fl means to the circuit des(gner better wide
Modern transistors are built using what IS called w;~l
band aam Performance The f7 frequency can be related
planar technology. This name arises from the fact ttik~t.
to the-physical device m terms of the various delay times
ail areas of the transistor are found on the planar sukbce
throughout the trmslstor
If the delay that a earner sees
of the silicon wafer Figure 3 Illustrates a c~~:.section
in traveling through a device is less than In another
>, .,,,,.
device, then the f! for the device w(th the least delay is
higher. The thichess
of the epltaxial region IS related
directly to one of these delay times; namely the rSCCrc
time constant in the collector. The r., IS the collector
series resistance and to reduce this value for a given
resistivity, we must reduce the epl thichess.
There IS
another advantage to be gained from reduc(ng the epi
thichess
which relates to distortion performance. figure
4 shows a comparison of mtermodulation
dlstoflion
performance between two CATV transistors. The tran.
siators are identical m all respects except that one

$~ o~<tYPlcal transistor structure as built using the planar

,,*
, ~><~l~chnology
The first job of the designer IS to decide
~,i., - what start!ng material he wishes to use for his transistor.
.?;*:<<
~i,-?>
...
The starting material consists of a wafer of Sil!cOn.
.,:J.
...
,..,,
approximately
10 roils thick and typ(cally 2 Inches (n
diameter. This silicon has been grown m crystal form
while introducing a large concentration of Impurities. Thfs
substrate silicon, then, IS very heavily doped so that the
resistlvity is very low. On the surface of this low reslstivity silicon wafer is then grown a layer of slllcon

10
Figure 4.

207

100
1.(mA]
IMD Distortion
Performance
of EPI Thickness

1000
as a Function

device
than

was

on thin
gain

current

with

density
ness

It is seen

ep! material

at higher

thin

epi

available

higher

Thm

is the

CATV

side

material

IS made

sensitive
effect

the

epitaxial

the

starting

be insured

that

fT In

Another

fact

to

becomes
Wth

capacitances

and

primarily

for

devices

are

favors

the

this

design.

figure

IS a representation
The

base

of typical

area

is

transistor

dictated

by

con-

the

power

epi
Emitter
Contact

per-

good

Base
Contact

,,,.

tran.

there

is a

as the

achieve

more

IS lost and
then.

is as

epi

good

and

collector

the

more

transistor

wants

thm

as

to choose

possible

for

to avo[d complete
ballasting.

IS decided

constant

upon,

IS ava!lable

vertical

geometry.

then

which
The

has

wh{ch

be reduced

can

due to carrier

The relationship

it must

will yield

been

Importance

transistor

fT IS the delay

tb

used

frequency

CATV

the base region

are

High

thin epi the ballasting

is thick enough
some

very

is that

thinner

designer,

a process

the

time

increase

and

material

performance

high

parasitic

Geometry

Vertical

Once

The

which

and prov!de

configurations

transistors.

device

distortion

to yield

resistor

The

yet which

to

figurations.

of

reslstlve

Unfortunately

variations.

material

performance
depletlon

story.

collector

mesh

mterdigltated

epi thick-

like a larger

better

power

sensitive

of deblasmg

by the

and

modern

current

as the

acts

appears

transistor

ruggedness.

maximum

overlay

levels.

thinner

to voltage

of the

region

applications.

to the

performance

loses

higher

the

In

built

performance

because

then,

thicker

was

for this performance

that

resulting

5070

which

increases

depletion

materlai

for

negative

high

fact

This occurs

at these

was

distortion

reason

in a device

currents,

eplt=lal

sisters

an

better

The

The thtn epi device,

formance

2.

offers

collector-base

region.

which

that the device

levels.

is decreased.

the
at

built on epi material

the other.

of

discussed.
In order

movement

for this delay

to

through
IS

Wb

2,43 Deotn (NB /NBc)

,.,{?,,
.

s. ~t,ii,~~.

k~,
This

relationship

tranSlt

across

dlffuslon
NBC. The

the base

co-efficient,
point here

as the square
to produce
The

well

control

describes
region

of the base

understood

formity
implant
t!ons.

width.

of baSe.&?~~h?Wb,

A deslF~k@al

then
base

pr~es$w%n

be

to a po[~&&~@,@ver.
devt~e ~lds
~@i,jM@~ent

-the-ar~,pa$

IS

1/

used

to

Base
Contact

as narrower

go down due to
In the dlffuslon

widths

Another

regions

a@~~i~$

with

of Implantation

stea~#@.s~adlents
resuf!&&/&

In those

good

%/

uni.

In

higher

the

f!elds

IS that It makes

emitter

and shofier

and

base

transit

t!mes

Hgure

~~~fr:
;f.
,$
.,.

3.,@&Wal
.onM+,..
,:~~re
$~$tor

handling

Geometry
item

must

is ready

be

to be

considered
built.

before

A mask

set

the

use:

basic

There
these

are
are

device
three

transistor

must

contact
overlay,

be

dev!ce

WIII determine

iphery

IS

This

geometries
and

mesh

base

in

emitter

The

208

to

reg[on

junction
edges.

what

under
on the

caus}ng
Since

transistor.

of current
the

IS because

effect

Configurations

dissipate

at

a large transverse

base

de-blasmg

dec(ded

of the

amount

de-

be

Transistor

ava!lable

The

frequencies

configurahon

interdlgitated,

area

generated

active
the

Typical

what the

must

be determined

CATV

5.

requirements

enough

~~gned, or, In other words, It must

dev!ce WIII look like, physically.

upon

Mesh

w[dth

are possible,$thoug~;:by
taking advantage
of Ion
technolog~~~1~~
formation
of the device June -

possible

first,

Emitter
Contact

~//J

fOr, @@f~%

has a,,~[r~

dlffuslon

sought,
which

state-of

required

In terms

which

this parameter

process,

time

Deb; and dop(ng gra@$$~:.~@and

IS that this delay t(rnQi~@8ireCtlY

a transistor

base widths are

non. untformitles

the

most

This

of the current
how

and
In the

WIII have

of the

em!tter-

to pass

much

IS
per-

levels

drop occurs

portion

be

by the

emttter

bias

em!tter

It IS known

wh[ch

minimum

voltage

the

must

heat

to be handled
htgher

central

There

the

at the

current

the

dev!ce

will be required

late the amount

handle

this current.

of emitter
Two

examples
em(tter

now

IS to pack

collector-base

of possible
peripheries

base

mterdigitated
are shown

to safely
this amount

area

junction

COLD
,.,,,,.
,,,.,,
a,.

possible,

capacitance
designs

m F}gure

having
It IS seen

ED =

24

BA =

88

Eo/BA

slightly

higher

EP/BA

are

ratio The problem

with the square
the long emttter
fingers
required

considerable
part

voltage

of the

WIII develop

drop along

device

Not

IS not

only

duced,

but t WIII soon

des[gn

with the higher

voltage

drop

as

heat

In the

problem,
shoti

enough,

long

The

more

current

problem

middle

arm

and

ballast

reslst~{s

pulllng

ar%k~us

volt~$$~~that

or

diffused

~~~~:~~ficrocracklng

is

clause

device

site,

thus

IS now

of

the

current

Iimltmg

the

emitter

area

This

series
and

and

film

they

resistors
also

are

are
faced

susceptible
with

more

barr!er

and

are

reasons
IS good
depends

re-

available

for current

input

desirable
varies

directly

necessary

to

209

since

as

sible

for

for noise

gain
width

have

on the

handling

purposes,
figure

on base

Figure

resistance

they

seen

that

amount

and
two

gain

base
per.

of periphery

emitter
base

reduction

stripe

region,

width

6 illustrates

to

WIII allow

a given

and

rb as much

Incidentally,

too.

for

spacing
within

contact

parastttc

spacing

transistor

of the

d!oxide

since

susceptible

contact

A narrow

emitter
the

not

the

resistance
the

rel[able

contact

we

directly

the

the
of

to be placed

to reduce

good

of emitter

high

widths
A narrow

formance

pre-

to reduce

periphery

since

slt on top of the slllcon

are

contact

more

begins

they

resrstors

desirable

emitter

Important

con-

since

oxide

it IS also
the

An Important
point IS the type of ballast
Two types of res!stor
are popular,
thin
Thin

Diffused

emitter

in the

current

barrier

barrier

7.) The

site

the

drop

quite

emitter

Junctton

thermal

avoid
the
cracking

to this

Figure
the

to draw

solut!on

(See

a proportionate

has

are now

WIII tend

with

to

coupled

between

m series

Figure 7. Ballast Resistor Configurations

of

able

f[ngers

The

fingers.

share

the

@sign also

th;$ device

If an emitter-base

~~~~~~f~llure
$+~si~r
used
,., !~fil~?

i@%~~~iclent

k~~i?~er

m$$,e?,$~dn

The

are $:\~~,&sely

~.@&\,$~We

podlon

metalliz~!~

s!stan<~~~~

of over~~w

~,e ~~er

be:re-

An~$&eF:~~vantage

IS to add,,~a~st.$esistors

feeder
tact

Although

which

hot sp~ts

!t IS inher~N~&$efier

the cells

the active

The resul~f~;

and

is des~$~~~$~mce

con figuratlon14~%/.$<~

however

con flgurat]on.
Will result .~~$

used

fail because

a higher

performan~$

IS eliminated

since

wtth

with

their length.

being

WIII device

IS that

square

to one

aspect-ratlc

problem

configuration

dlsslpate

compared

possible

aspect
that

IS square

ratios

which

this

27

EP/BA Comparison
for Square vs
Rectangular Base Configuration

6.

design

that

Figure

that

~o..

1111[

IS

to calcu-

necessary

mto the smallest

reducing

equal

it is @ssible

periphery

The task

periphery

thereby

to handle,

of emitter

is
rb,

it IS

as posof

rb IS

the Impact

metallics and the wire knd faflures that result. Figure 10

illustrates life test data that shows an increaaed failure
rate due to bond failures in the aluminum-gold system.

Life Test at 95C Case Temperature

Unit
Pam Descri~ion

Murs
Accumulated

Wire amd
Failure
Nos

Wire Bwd
Failure Rate
%

Figure 8. Effect of Emitter Stripe Width

on Base Resistance

The last step in the construction of the transistor is the

depos[t[on of metalilzation so that contact can be made
to the emitter and base regions. (See Figure 9 ) The type

Collector

.Yx:,a.,:$,;,
. ..

of metal to be used IS an impodant decision~~~~$$wo

metals that are low enough In conductlvit~t$j~c~h
be
used for translator metalllzation are 901Q~ti~]&rninUm
Aluminum metallization has been USe@,t~~;$earS
aS a
conductor for trmststors
Its advan~:${are
that It IS a
well. understood process, it of fer~@go@slllcon
contact
without any bwrier metall!za\J~~:~Q,~ it is inexpensive
However, considering the ~~crq contact geometry of
the RF transistor and the<~qJ~Q#~t will be mounted On a
gold hybrid circuit, th~<${h,~ dec(sion IS considerably
easier to make. Fo:<@CAT%:*ans! stor, gold provides the
follow!ng advanta~~.@Qr
alumlnum
~,*1,.
1. Monom$@,~~~k
bonding SySteM.

It was shown earlier that if~.

,Mt#esirable
to ach,eve a
high Eo/BA ratio so as t~$-in
maximum performance
from a device. This ,#,ps Nleved
by placing the tran.
slstor contacts as q$~~~gether
as possible. The use of
such tight conJ@k;;~,ometrY
fOrCeS
the use of verY
narrow metaj+~~er+l The resulting high current densdles
can lead tq~j~ti~blffiy problems as a result of electromigratlo,~ ~?~trom,gration
is a phenomenon
which
occur$f$~metal films as a function of time, temperature,
an~{aW+r~$ density For any given temperature, a ce~a!n
:q~llbr$m
concentration of vacancies ex]sts in all metal
~:f~,~~:~elf
diffusion of metal ions throughout the film
:~$~
due to the metal ions being thermally activated Into
a~jacent vacancies. In the absence of any external
$ forces, the metal (on dlffuslon WIII be isotropic and will
result In no net accumulation or depletlon of mass m any
given site. In the presence of an electric field, however,
the metal Ions experience a force due to thetr charge,
inducing an Ionic flux toward the cathode end of the film.
In addition, the conduction flow of electrons m the metal
due to the electric field will cause electron scattering off
the activated ions and impart momentum to them mreducingan ionic flux toward the anod!c end of the film. In
good conductors, the momentum exchange force dominates the electrostatic force and results In a net mass
transport toward the anodic end of the film. The result is
an open clrcult m the metallization strip This void for.
matlon IS accelerated by high temperatures and current
dens!ty
Aluminum has exhibited a high susceptibility to electromigration for current densit!es above 1U6 A/cm? Such a
current density IS easily real[zed [n state-of-the-ati
RF
devices. For a given device geometry there are only two
alternatives to allow reduction of the current density in a
device Either the operating level can be reduced or a
metal can be selected which has a higher mass and activation energy. The operating level cannot be reduced
without a sacrifice in performance. We can still keep high
performance and reduce the current density by using
gold metalllzation. At 200C, experiments conducted on
Identical trm sisters with gold vs. aluminum metal lization
showed an improvement in mean Ilfe time of two orders
of magnitude using gold.

2. Electt@,N~aT$on resistance
-,,.J.,.,*
3. Lo~gB~act
resistance with elim!natlon of shofls
$db$~sillcon.metal
alloying
,$4:f ~~r~slon resistance
+?~~Qx!de step coverage.
~$}l~kows
~.:
.,.

use of tighter contact geometries.

Monometallic

Wire Bonding System

As has been described, it is desirable to have an all-gold

metal system for reasons of reliability A monometallic
system eliminates the formation of gold-alumlnum inter-

210

-...-...

_--

Contact Resistance

Step Coverage

Gold cannot be used as a single layer metallizatlon

because of its relatively low silicon eutect[c temperature
end Its poor adhesion to silicon and silicon dioxide. A
barrier layer must be employed to prevent gold diffusion
into the silicon and this bwner metal must offer good
adhesion to silicon, sillcon dioxide, and gold Such a
barrier is offered by a system utilizing platinum sil!cide,
titanium and tungsten. The platinum s[licide forms a good
ohmic contact with the silicon; the TI/W provides the
necessary diffusion barrier and offers good adhesion to
SiQ and silicon.

Gold offers tremendous improvements over aluminum in

rts ability to cover oxide steps without decrease in metal
thickness or cracking. (See Figure 1 1,) Aluminum is
deposited by means of evaporation In a vacuum where
the mean free path of the aluminum particle is long. This
means that equal coverage of all surfaces IS Impossible
even if the target is rotated during evaporation, The
plate-up gold system reduces step coverage problems
to insignificance.
~,
,,,~,,:.{,,
Narrow Contact Geometries
~\~iFf:,.,:.
J
The RF transistor
must have very fln~>~~}~%ntal
geometry to achieve the performance *~W.@U in a
CATV system. With aluminum metaliiz~8&$~@se narrow
finger widths are achieved by etch& ~Z~~@aluminum to
remove !t Such a process, if don&~~fy:,caref ully, will at
best result In fingers of unev@t~&
which are susceptible to high current dens~~$$and
the associated
reliability problems. The $r~ksys?em is capable Of Providing microwave geo~~Re*With
Insignificant variations
In line w!dths In f~?,~~&
geometry on present gold
CATV devices is .&~~~@r than some Iow.noise microwave devices w,~t~:~&on
the market today
.:.!<.

Aluminum has historically offered good ohmic contact

without the need for barrier metals. In RF devices, however, at current densities well below electromtgration
densit]es, a problem of formation of silicon/aluminum
alloy IS ever present resultlng in emrtter-base shorts. Any
hot spot formation will result m an increased alloying rate
and early failure.

Corrosion

Resistance

Under biased conditions, in a hum!d atmosphere, gold

has demonstrated a Irfetime more than 3 times that of
aluminum. The failure mode In aluminum IS electro.
mechamcal corrosion and gold is insensitive to this
phenomenon.
$ Evaporated Alumlnum

REFERENCES
Der,valt.n

of ,el,ab<l,ty exDress,o
R(l)

10 chance

la<lue$

= e,,

of x, ,,ems 8s :Ont,wsly
c,
If an OC,g,nal Dopulat,o
cayog so [ha! the<, are X ,tems at t,me t ,he .hanQe of
Doplat,on 8 one ,nterval dl 8sox dt D,v,ded by the !.!.1
Pop.lal,on Xal t lh,s g,ves the negat,ve ,ate at tih,ch the
DoD.lat80. changes a! !,met
~ =

Qx!t
=
xx@

!hen

>d! = ax

lteg,.1,.g

= 1.X

C = tX

fnC

x=x.

The C =

And X X

= e..

If!herate.

fdecay

L #,con,,

= e.,

Al

x x

the ttme per,oa be,g COnsde,ed

over

}Idt
o
!Orl=o

Q;

.jAdt
o

Since X X. ,So,obab,l,ty
1.1,.. Ihen
R(1, =

ant ,hen

ol$.rv,val
X X, = e.:

211

foradecay,ngooo
AI

M,ke FlahIe
Rel,ab,l!ty and blTF
Shonolll
M,c,owaves
J. IY 1972

The Lung and

2 J.me5Hmphrey
ad George L.eltgena
Rel!ab,l,ty
Cons, dera!,o.s
,. Ges,gn ad Use of RF Integrated
C,,c,ts
IEEE NCTE Co fe,ece
February 1976
3

Ell,o(t Ph,lol sky

Alm>nm
&ds
ProdCtS D,.,,,.

Des,gn
Lm,,s When Us,g GOIO
Moto,ota
Inc
Sem,cona.ctor

4 R Flan,e .,,0 M \Ve,,s

A Stdyof
the Advantages
of Gold Me!all, zat,on , lhe Man.lacl,,reof
M,crowa.e
7,s,s10,s
TRW Sem,.odc!o,s
Tech,cal No,,
5 lg.,
B=osky
Re18a08/,1, Tneo,y
P,ent,ce Hall 1961

,0

P,ac;>ce

6 J R Black
Electron, g(a!, on Fa!lre Moaes!n Alum,
m
Me, all,zat,on
10, Sem, con Oc,o<
0,.>., s
P,oceed(gs
01 ,, IEEE Volume 57 Nmbe,
9
SeOtember 1969

212

AN I 028
35/50 Watt Broadband (16W40
Push-Pull W Amplifier Band Ill
This note describes
the performance
transposers
and transmitters.
Devices

used

Basic

: two

amplifier

specifications
at

PO=35W

lMD(l)=48d

at

PO=50W

= 28 volts,

Total

= 4.4 A

carrier

General
The

8 dB,

dasign

principal
emplov

sound

atms

simple
and

The

ma!n

requirement

The

second

7 dB,

sideband

design

consideration
the

solutlon

reduce

was

match
was

us to obtain

the

and

maximum

losses

to obtain

the

be the

maximum

factors

led

us to choose

matching

the

optimal

$~{{~}mances

for sewice

in ~~~@~$+TV

bestp~$$lbye
* :i, >.,t. .4,,+:.*

IMD over the band.

in all the

ga!n &@~educlng
),*.
..J,:
., $;/,2,

the

clrcults

usln~~~~r(e~.wavelength
,, ,,.,
\
\\\, ~,..,<.,<?
source impedances
to lQ~ @~pes with low losses
.+,J}}J>:
transistors
In a push pull conf]guraffbn.

TWO TPV 375.

from

S+.;].

..,.*. t:
.l;~.
>,....
,..>,.
.powe~+~~h~he
best

output

must

minimum.
These

signal

cost.

to obtain

output

consideration

permitting

the

designed

.141.
,..
,+
,::~*
.,..
,.

,.~
\~.~..
~:.1.,.
.. .
,.>,s *:$*,
,.t,i.++.ryi.~
,,,..-.,, >,,...
.,
$T*,..&~..,
Pgai
=lOd
B
,\,>.*\$,,\t~<\,,:l
input VSWR
: < 1.$*
~
\,\.,*.$
.,
Output VSWR
: ,y~+~\;,5
...
,\ .,,., .,~~y
16 dB
,:,:+ i.$t:$,
.}. ~
., , ,.:~
.>, ~!.,?.,,.,,1~
- ~, >*,\\.i\$\

were

the

this

push Pull amplifier

Consideration

a relattvelv

simpl[fv

linear

visronc;arrier

ultra

375

lMD(l)=51d

v
(1)

TPV

of a broadband

MHz)

To obtain

band.

Input

matching

transformers

circuit

at the

losses

Input

and

to a

output

which
permit
us to :
reduce the load and

couple

Because

two

the output

transformation

and input

ratio

The first approach

TPV375
data sheet
then

optimized

Amplifier

is one

with

basic

block

The

Input

and

and

a matching

the

aid

r,@~{ed

of ($@@PACT)J
tt:is.p+,
. \.*;->
..) .,!..
>!.*,,,.,,.

diagram
foF~~~#hpllf!er
(0.3, >?
outputr@$rcuMs
are each

are !n series,
for a single

made
impedance

is shown

of two

are transformed

,,:~..
!!~, .:+..,#he
\
The

devices

can

balun

ratio

theorit!cal

Output
circuit
The output
circuit

The

maximum

(CC,

(AA)

impedances
The element

given in the
values were

1 and

networks

the

circuit

schematic

IS shown

: a quarter-wavelength

in Figure

transformer-

2.

balun

input

is shown

the

input

BB)

been

in series

the

potnt

collapse

the

to 50

bu!lt

using

microstrlP

[n Smith

Chart

near

real axis

impedances

elements

match

to be easilv

are shown

into higher

at this

completes

chosen

impedances

parallel

VSWR

in Figure

IS 1.80:1

4 and

VSWR

is 1.16:1

and

the

Impedances,

the

output

the output

match!ng
but primarily
output
quarter-wave-length

theoretical

have

the

the

amplifier

impedance

1.

input

is doubled

technology.

to

bv Capacitors

impedances

17

~,

FF,

around

The

quarter-

~.

is 2.8:1.

dances are hlgherthan

the input
transformer
ratio is lower.
The inductors
aid the
for collector
bias. The

impedances

be considered

transformer

transformation
maximum

the required

netw~[k.~
,,!!. \\/\ *>5+$

3 and

~w.~he

configuration,

stage.

from the input and output

levels using a Smith Chart.

in Figure

composed

The

~fb yayelength

due to the push-pull

ended

program.

The quarter-wave~~ng~
transformer
. i~$~r*i&*
Input
circutg
~~..)
: > $.~?
The input pircul%.ie shown
!n Figure
lo~j,&#s,@tor
input impedances
.!.!\.,.?
Thy~~~~.D~)
series elements
and
~fi~;q:
p

$:*
~J~,.,.r+:t
.1,J.\~
.>-,
>{,l.:(,,
~,

lmpeda&es

of that

for the circuit calW/&oriwas

and matched
tW?~&k~&per

Design

The

transistor

half

and

maximum

impedances
matching

experimental

on
network

Smith

VSWR

Chart.

IS stmpler

provide
for good stabllltv
at the
transformer
ratio IS 1.6
1,
the

maximum

213

experimental

VSWR

is 1,60

Il. Since

the

1.

output

impe -

and the quarter-wavelength

low

frequencies,

is 1.44:1.

and

are

used

Amplifier

Performances

IMD versus output power : Figure 5

Input and output return loss and VSWR

Ga!n

versus

1 dB

6ias

gain

frequency
point

The

glass

: Vce

and

The epoxy-Glass

Figure

compression

conditions

Technology

: see

layout
1/16

-Teflon

:70

= 28

W
Total

V :

= 4.4

A.

considerations

Inch (s, = 4.1)

l/50

= Figure

[rich

(c,

is used

= 2.55)

as board

Isusedfor

the

material

transformers

Wehaveconsidered
foramlcrostri
pllnethatafter
W (Width)
gible and wecansize
theground
conductor
to be3
Wwithout
former
has the following
characteristics
:

We

The

Can have

Good

vertical

anyimpedance
dimensions

values
are small

within
and

the

realizable

except
(see

the

input

thedeta!ls

and

ouput

Figure

from theconductor
strip edge
perturbing
thepropagatlon.

min-max

mechanical

for

limits,

reallbliity

is good,

repeatability.

transformers.

8).

st.,\
.,.,.,. *
, ::;~~..,?,i.
*.,>
,.y ,,~,i:>,.ii.

the fields aren~~ykv$~

This kind of.w~K
,$?:,:8<>$
.,?,. \c~,\~,..Y$
y;+,,,t,
,p&:i.\..4,$\}
,,~:
1+::
. ..\~
*, ~~
*,.,.
~.*l,,,~.\

The bias circu[ts

are Included
with
RF circuits
In order to give a compact
amplifler
the layouts
and the Figure
12 the physical
layout
of the push -pull amplifier,

Combined

pairs

of

push-pull

Amplifiers

: FigW~~$?Wand
,..:.. ~.;..~:...

In general

They
The

can

push-

recombined

advantage

loss)

several

pull

amplifiers

bypairwith

of using

lncomparison

th!s

kind

quadrature
of coupler

With

Good

additional

effoitboth

repeatabilttyin

relatively

for the

combiners
is that

h!ghorlginal

the

final

stage

(see

block

input

and

of the

T:$~$#,~smitter

diag~~~~ure
~
$>
out@;,
,&,WR

theinput

production

become

good

(>

rtn,

VSWRofthe

andoutpK~~VS~~c
without

20dB

amplifiers.

9).

push$~~lampllfier.
. i.
,.y
*$%,
,,,,
,\*l;.,
~~
Ganeral
Conclusions
*
\ .:+
Pushpull
techniques
simpllfy
therequlred
circuitry
and a~&%&~ed
losses.
,\.*>,
..<~.,,
Theproblems
assoclatedv/lth3d
B hVbrids incascade
~~rA$ertion
loss and imbalance
(,,.
in parallel
are requr red are m[n!mlzed.

w!ththe

are used

show

$..b

..! ,

11

$:...,
,,.*F

f.. t:+,, ,:;::

~h ,.!
~::~$::$ ~
?!<,,

ouldbe!

va~sb~~:com~onentsb

mprovedtol
eing

,2:

when

four

devices

1.

reauired.

RFC

d
TPV 375

,.., ,,

Rgurel.

Push-Pull Circuit

214

_.

.__.

--. -. -

m
,%836 ..+(wi!!
.,.
!

,-.+T200Mzll
InF&

I.
-K

Rgure 3. Input tircuit

215

m,.,tii,.m,--.,

Figure

4A. Input tircuit

216

-.,-

IWEDANCE

COORDINATES- 50.OHM

CHARACTERISTIC IWDANCE

,.,.$))
figure 4B. Output Circuit

217

:BE:9
0

10

POWER OUT
+
PEAKSYNC

30

20

Figure 5. IMD versus Output Power

dB
15

1
////////

I ~i /////
b.) Equivalent Grcuit
Figure 8.

218

TPV 375 Push-pull

Input

Ouadraf,e
combiner

Quadrature
comb, e,

Otpl

50 Q
TPV 375 push.pull

Hgure 9. Combined Pair of Push-Pull Amplifiers

[
100

mm

output

.. -.,

St,, p

(Z. = 40)

#
Board

material

: glass leflon;

1 /50

Rgure 11. PC Board Layout for Input and Output Quater-Wavelength Transformer (Not to Scale)

219

inch;

S, = 2,55

>. ~+,
,..

.:

..,.

~%

..

figure

12, 160-240

,.

MH~),*er

by :

220

..

_.

_.
___

..

AN I 029
TV Transposers Band IV and V PO = 0.5 W/l.0

W
+<,

.,,,. t>
. . . ...$.71,:.:?>

This note describes the performance of a broadband

vice in band IV and V T V transposers.
Device used :

(470-860

TPV 596.
Basic specs :
I, M.D. 60 dB max. at Po = 0.5 watts
Vce

= 20 volts; Ic = 200 mA

Pgain

= 11.5 dB min.

The approach

used is intended

1 ) The load line be defined

2) The
3)

The

4)

Use

5)

Use

forward

ultra linear amplifier

desig~~$,$~dw ser.~. ,tp

,
.>+
.....A},:.
>\+,)@,*
..*~:,\
,.,<,
b.,?
,:,~...
>,.$:
,1:
~ ,,,*!:,
~
.,;:,,p?
,.,, ;i;?f%,,t*b
~:; $:, ,
.,.,,.
.,,.,,,\.
c-l,
,?
+:,
:..,,\/
~ ,s,
.,:%,,
,:,},}.:
,$,;.
.S>,
., $<>,$CJ
,,
k,., .....~t)j,
and inexpensi@@~$#ol lows.

the correct match for pea@fo~$r

(P. sync).
.:.,f...,><,,..
.*,,
VSWR at the collector be less than 2:1.
,,.~~i+,,,
.,,,.
input match be designed to provide flat gain with deqr~$j~@ ~requency.
,?
..,,. ~
W,.l
~$<.
b
computer aided design.
~tj,
~$:,:+
a three tone norm
,. -lfh,
.~l.~.j ~,:.>
Pvislon
= B dB
,$ ..,,

Psound
Psideband

6)

to be straight

MHz)

to provide

= 7 dB
= 16 dB

+$:~$,.2
.-r~\
...)~
,,?V.,
Circuit realization to be a distributed designkgllt
upon teflon glass copper clad circuit bosrds. However
the design will be analized using Er =ZX
(t1,0,~

The input and output impedances were ~~~~$~om the TPV596 data sheet and plotted on a smith chart. First consider
the input. To have flat gain with an o~[~~~m collector load, the basic physics of a class <<A>)biased device defines
a gain slope of 6 dB/octave whic~~~SF*e
compensated for. The bend of interest is 470-860 MHz which is .915
octaves which implies that 5.25 d@~@~g&n must be compensated for if the device is PerfectlY matched at 860 MHz.
This means that a transmission:@$&.#5
.25 dB or a VSWR for 11.0:1 must be employed at 470 MHz. The input Z is
converted to Y on Smith Cha@;{~?+Whepoint at 860 MHz will intersect the constant conductance line equal to 1.0
(2O m 0) if it is rotated 0.~4++Q8*flg a 20 m U (5O Q) transmission line. Afier this rotation a capacitive stub or chip
capacitor is used to res~,~%.~ t~ susceptance at 860 MHz; A capacitive stub or a chip capacitor equal to 16.7 pF can
be used, and the res~ ts$fi~wn on Smith chart (l). It is interesting to note that the VSWR vs frequencv can be
adjusted for gain fla~-+tiy
selecting an optimum Z. for the capacitive stub. It is also obvious that the IOCUSof
impedances at th$~?~it
input can vary between the locus of points defined by using a chip capacitor, and the
imaginary axi~.~~Q%j@g a stub with 20 = X. Graph (11)is a plot of these results. Because infinite isolation doesnt
exist between$.~e:~utput
and input of anv transistor, and because the required network is verv simple, the input
circuit will$,@eo~~ized
empirically. A computed aided circuit will be defined for the output onlv. It is also indicated
that a c~,~%~tion
chip capacitor and stub mav provide the best results.
.,- .,~:,f,?~
The @Q#btdrcuit considerations were first determined using a Smith Chart approach. It must be ciearlv understood
th~$titiuter
optimization is onlv as good as the circuit configuration and associated computer instructions.
*?.,
, ~:,${$%+
,p$@
*pproach
,;$
~*\$, *?t.:,.tL*.:.>
i::~}sr
<.::,<,, .+.
:,,.,.1,
i.tt.,.
.,
{*

follows

1 ) The device output impedances

Smith
are first converted

Chart

(11)

to admittances

and plotted as the conjugate

(Y load).

2)

In order to allow easy collector lead soldering a Zo = 50 fl, 3 mm long transmission line is used. Since
the Smith chart IS normalized to 20 mij (50 f)) we can rotate toward the load directly as the chart is configured.

3)

Since the balance of the circuit used Yo = 10 m~; (100 f)) we next normalize the chart to 10 mii. 100 f)
transmission line was chosen as a good compromise between phvsical length requirements and ease of
realization on Teflon Glass.

221

4) The next element, a shorted shunt transmission line less than l/4 in length reduces the imaginary part
by moving each point of admittance along a line of constant conductance. The length was chosen to locate
the lowest frequency point (400 MHz) near the real axis so that the locus of points would be more equally
distributed about a 2.0: 1 VSWR circle.
5) The resultant locus of points are then rotated with a 10 m?l (1OO f)) transmission line to a degree which
locates the admittance point of 860 MHz near the line of constant conductance equal to 2.0 on Smith
Chart (It). This conductance is exactly equal to 20 m{l since the chart is normalized to 10 m??
6) The final step is to use a parallel resonant circuit which will reduce the imaginary
and lower frequencies.
The following

approach

was used to calculate

the element

values for the antiresonant

By observation of the smith chart it was decided to place the 460 and 860
the 2.0: 1 VSWR circle.
It then follows

wLc =

st.,
\

circuit

MHz points on or just inside, ~~s

,.y ,,~,i:>,.ii.
., *.* :
i.{,.?S$,:)d.
$.
..,* +>$.
,~,, ~t~i.c,
,!

that
at f, = 460 MHz

pacts at both the upper

,t
,,*L,
$?

0.4

!:.<$$:i,,
.,,+>.
..\.**.>,,
.,.. .h.$v
:::/$,
~,,
,+
1
,..
~~.,,,.\\,
,\~$:,,
at fz = 860 MHz
w2c
= 1,7
:1,.J..
.,,,!,
,,.,
W2 L
~f,:ii,,yj:i
~i,.
? ,* ,\,\
,, >+~,r~..
,$y
The 2 equations with 2 urtknows are solved with the following result,
*:*, .$,&
L = 0,189 nHy
):.>.,,
~.,:.i..
.....,!.
C = 496.11 pFd
,>
,,:~i:i
,..>
.*...!..
stnce we are normalized to 10 mij
,~,,.,?1. -
Lactual = 0.189/,01
nH = 18.9 nHy $ >~s
Cactual = 496,11 x 0.1 pF = 4,96 @F~,::t$
t~.~~,,.
~,Jt,..
.>.,
..!
7) The result IS normalized to 20 mii with the final result shown!.*?*\$**
.* :v;>
\
\}>\\,:,
,,>,. .x
\
%.>.??
.$,
WI L

Gra~L$&t~}*~hOws

GfiM~tIv)

t$j}>1,,
,. =!

w
1

j. ,,,,.
the various VSWR

shows the collector

calculated

load VSWR

compared

to the theoretical

for the calculated,

optimized,

J
100

50.4 mm

61.6 mm

best curve and the actual VSWR

and actual result

~t. .l~ ~+.

~aph (V) is a plot of the single ended amplifier results taken w~th a network analvzer. No component losses were
considered for the theoretical and optimized analysis. The f!nal circuit was also optimized empirically from
470-860 MHz using a network analyzer.
The following results are asummary
hybrid adaptation.

of performance,

bias conditions

circuit configuration

and recommended

222

. ..-

.-..

....---.----

starting Imp.
rotated Adm.
final Adm. o/Chip Cap.
f!nal Adm. u/10 Q Stub
final Adm. m/50 0 Stub

~=~
~
~

figure 1. Smith Chart (1)

223

224

..

Vs WR

Transmission
LOSS (dB)
+ !0.0

,8

>,

50 f) Stub
Actual

9,0

------

~s
?,o

.\

\.

8,0

~>

6S

30

It

II
\

N, \

It=

I I
0,9

225

0,0


------

Preopthm,zalto
P0st0ptlmlzat40n
Measured

Frequency

--rlgure

- - ., .,
5, ~rapn
v

-m

-----

I rvaso

. . . . .

Ampiirter

(MHz)
.

226

venormance

versus

Frequency

R fC

.hr

II

Vc c

A
figure

7. Clasa

A Bias

227

Circuit

TPV 596 BROADBAND

AMPLIFIER

228

AN I 030
1 W/2 W Broadband ~
This note describes the performance
in band IV and V TV transposers.

Amplifier

of a broadband

(470-860

MHz)

Device used : TPV 597

Basic specifications
lMD(l)=60d8
Vce
=
20V;
P*,i
=11.5d
B.

at PO=l
W
IC = 440 mA

Band IV and V

ultra linear amplifier designed for$a$e

~;.4~~ ~,:,,
,;&
.* .
;1.
,i$..~k;,..~
..~,,.
,,t.,
,,i: $.$.
*, ~ik.:$.i~4\!
~~,
,., ,,; *>,,.!:
.,?+,
,?>
~\
i)}
,.,>**tx:,
<v
!
,~:,. ,

(1) Vision carrier 8 dB, sound carrier 7 dB, sideband s{gnal 16 dB.
%,. x~p$
.~.~,.,,.:).
General design
considerations
~~k*ji$ . ..i,~
In general to obtain a flat gain for broadband
amplifiers which use ,ransisto~~~it~;boul
6 dB power gain
~:t+,,
,,:,,l.,}
variation per octave we can use two techniques
:
,.
feedback technique
(eg emitter resistor and a negative feedback with a<$~~~~ive circuit between the collector
...,>\\
and the base),
~. ~
or reflect the input or the output power selectively to have a~a~~~$n
loss of 6 dB per octave with O d8
.*
for the highest frequency,
~+
(There is also another technique which uses a selective .a~{$~w~tor),
With the feedback technique we can have a good input %@ output match. With the second
technique
we need to reflect the input power and have a good o,*ut
m~tch In order to obtain a good IMD. It means
the inpul VSWR is very high for the low frequencies.,..,..,,~, $*,
The second solution is simpler than the first and if ~~>~s~lwo
amplifiers connected together with 3 dB quadrature hybrids to have a balanced ampllfier this in$~~$$n~$hce disappears, We have chosen for thts amplifier this
second solution.
For the larger broadband
amp}~$~~r(eg 170-860
MHz) th!s solutlon
must
be re)ected and
the onlv acceptable solut[on is to use the feed bqcktechrrtque.
.,
\,~
Amplifier
design
.$~:$:,
,
The first approach fcr the circuit calc@A{iO% was made by us[ng the Sm(th Chart from the input and output
impedances given in the TPV 597 d@t~~&&t to have, al the input, a reflected power so that the gain will be flat
and at the output to obtain the b~$~,i~>bh possible.
IN PUT VSWR VERSUS

FREQU~&~#O
,$~j:i:~>.

OBTAIN

A FLAT GAIN :

The power

gain can be ap$pr~~%~$~ed by :

,, ..$
F2
,.:,,.. ...~.,~\f~
Gz~
,:
:\:,. ~r
() F
.?~.,\
. .\\..
F~ax is the freque~y
~r which power gain drops to unitv.
,7:>.$$+,,+?

The transmiss$&n ~@s due to the input

.$.2},
~is

reflection

is :
Z=l-lplz

Ps the r*~~~j~ori coefficient.

TO

h,p~~$, x~onstant
~ Y*,: ~.?

F2
,nax
[llplz]=Gl,

() F
the gain at the highest frequency used (F}l)

,\:i*;:i$
\~s..
...,t~.
~*~s

.,:!.,, ~:Iyl,,~,f
,)s.. y

we must have :
G% z

lpl~

~d+~~$~or
p*\,.\ *,.*
J,?
r.?,.,
i:,
VSWR

1 + IPI
=
llpl

F2
=

:
()

F
[01
l+
T1-(i)212
H

2 ,,2

229

F 2
l
FH
[01

Figure

1 shows

the theoretical

VSWR

We have defined
the input circuit
to this curve, and have assumed

versus

frequency

After we have calculated

separately
the input
means
of the global
amplifier
and the TPV597

RF

Program

equivalent

circuit

: Figure

an insertion

loss of O dB (implies

: Figure

and the output

S-parameters,

circuits,
with the

p = O) for 860

an input VSWR
is negligible.

we optimized
some
COMPACT
Program.

as close

of the

Calculated

gain

Calculated

and

Calculated

andemp!rical

and

parameters

gain

Input

: Figure

VSWR

output

: Figure

VSWR

: Figure

*.,
..:},+<..j:~$
~
i.,i.? ,.~$
~,
$? $~:i{,
Z>k
, f~,,:\ ;,<~+
........ . -.,.*.
\ ~\
S.,,*,.
)..,~,,
~e~
.,*.:;:,... .b~~~

Performance

IMD

versus

output

IMD

versus

frequency:

Input

Output

return

loss

Gain

versus

Bias

conditions

return

Technology

power:

and

loss

and

frequencv
: V,,

and

Figure

Figure
VSWR

: Figure

VSWR
: Frgure
= 20

layout

7A

7B
5

: Figure

-.,:.r ~\
.,si..,., ,.e$,l<~

,.>\\, -.*.}. $,
.~$b, .!.,
,;,,\
:?;
*J:;>,,/.~.
t,. ** h ~
,* ~-,.{.?
.<*
>,;:,,::{,)?$,+:

V;

I,

= 44o

mA

considerations

The
glass Teflon
l/16 inch (Er = 2.55)is
used
have a good contact
and repeatable
results.

as board

Figure

circuit;

8 shows

Combined

the

by

empirical

empirical

MHz.

as possible

Amplifier

with

from the TPV597

input impedance
to have
that output
circuit losses versus frequency

circuit

- Transistor

diagram

and

the

bias

material,

Figure

This

9 shows

substrate

the

@:$~~~&#ed
~.,,:>,

P@~&~d

to the

Iavout.

,:J::.R:,A-*<.
+,,
~,:$
,, *!],
~

Stage

heatsinkto

t~.:..-.:l~,
of a single transistor,
e~t@a
the capability
Th@y ca~$be combined bv pair wrth quadrature
combiners (See figure 10). Since quadrature combiners have tN&,~+bilitV to channel the reflected power from
the amplifier into the fourth port of the combiner it mean~~%
!Nput and output VSWR become verV low
(VSWR
<
1.2).
The power
gain is reduced
due to the c~p~er$insertion
loss bv 0.6 dB. Coupler
Imbalance
In many

instance

which

forces

should

also

the

the

power

designer

be taken

into

output

to

use

account

requirements

combinations

as causing

of

transposers

of transistors,

some

IMD

Q@~~~~t~on.

1,2

I 01
l__

From global amplifier and S-parameters

Empirical VSWR

400

500

600

goo
Figure

1. Input

230

100
VSWR

FM 9 DO

DD

FF
B

GG

ii

HH

..
L are g,ven

for z, = 1.

Figure
for

2. RF ~@&~ent
CO%*

Circuit

Program

CIRCUIT
DEFINITION

FREQUENCY
166
167
170
168
165
166

POLAR S PARAMETERS
FOR TW@ EE (TPV 597)

:
100
1 12
100
100
2 12
END

OPTIMIZATION

\
figure

3. Compact

231

(MHz)

Program

DATA

VARIABLES

()

GRADIENTS

(1):4.51899

(l).
(2)
(3)
(4)
(5)
(6)

(2) :32.0136
(3) :29.2938
(4) :72,2399
(5) :5.16145
(6) :3.53445
ERR. F. = 7.809
HOW MANY
WANT

ITERATIONS

INTERMEDIATE

SEARCH

.894864
.704452E-01
:
2.69282
:
:
.287748
:
1.68585
: .267730

BEFORE

PRINTS

INTERRUPTED,

FINAL

POLAR
FREO.
(MAGN
470.00
500.00
600.00
700,00
800.00
860.00

NEXT sTo P? , 0 REsu LTS IN FINAL

(YES = 1- No
ANALYSIS

= o)~ TYPE TWO

FOLLOWS

S-PARAMETERS

IN

NIJMBERS

ANALYSIS
: (1, J) :0

50.0

OHM

SYSTEM

S11
<ANGL)

0.88 <
134
0.85 <
128
0,75 <
92
059<
55
0.43 <
0.20 < ii

Figure

4. Gain

vereus

Frequency

232

. ,..-.. .._

..

.._.__

_..__

Od8

-4-.
__

-.

..,_
-,

~A,
. .

Y.
.=.

5d

\. -- -.
-+ - --
=.

.\<:L.%.
-

TPv

597

SINGLE

VCE = 20 v

IOd

STAGE
Ic = 440mA

v.,

N
\.
,

Cal.

15d

1 44

.1,

IMD (dB)

55
TPv

597 Single
CARRIER

8 dB

SOUNO

CARRIER

7 d9

SIOEBAND

Stago

VISION

16dB

-60

VCE =
1.

65

70

-75

IMD (dB)
60

65

700

Figure 7b. IMD versus Frequency

234

..

20 v

= 440 mA

.,

330pF

300 A

IIHIII.--[
A&*
.ll T

2N 2904

235

~ BIAS

iNPUT

input
(low VSWR)

~
output
(low VSWR)

236

.-...

AN I 032
How Load VSWR Affects Non-Linear Circuits
Prepared by
Don Murray
RF Devices ~vision
Lawndale, California
ReprintedfromRF Design Magazine
If your amphfiere test out fine in the lab but
fail QC testing, the testing environment
not the product is likely at fault.
Consider the following scenario: Youre designing
and implementing into production a broadband
Class C power amphfier, During your design
phase, you follow all the rules of science and
also dig into your bag of electronic tricks to
meet the design specification. Your design is
fabricated and tested successfully in the lab.
Twentyfive more units are built in the lab and
they, too, test out fine.
Confident that both design and production pro
cedures are satisfactory, you begin series produc
tion. But when the first units reach RF test, not
one meets specification. Yet when vou ret~eve
the units, thev test OK in the lab.

pedance of their loads, either in test svstems or

equipment environments. It is easy to compem
Sate fOr the inSSrtiOn ioSS errOrS in an attenuator,
but it is much more difficult to compensate for
variations in the input impedance difference bet.
ween attenuator pads, that is, the load VSWR.

fion of collector current and transistotdie

..~:,
temperature.
l?~i}\.\3?~!*,
The theoretical approach will qX&~~}~~

changes in ampkfier output @@X~@) for a

givenchange in load re~~~~{$~l).
~
or simpficitv, let u~ssfu~,tie
following
hypothetical con~i$~~~ch
are tVPiCal of
Lets ex~mine RF correlation on both an eMpiriCal todavs RF p@@:~~#stors,
.k+,i ,,,$::
and thaoretlcal level.

EMPIRICAL

HVpoth~$:,con~tions:

APPROACH

VL6$42%

The empirical approach is shown in Table 1,

,,:.$~AT~ 1.5V
where several test circmf loads (consisting of
.,<~~~x
50W
series attenuators, directional couplers and RF $~.~t,~equencY = 1,0 GHZ
switches) were assembled. The insertion ~,~Wan~i~olvlng for load re~stance:
input impedance of each load string wa~ti,,.t;.,:.!
~1 ~ (Vcc -VCESATI _ 702.25
= 7.02Q
measured, Following this, the intivid~~~,
2P0
100
were connected to a given test @c~,c@aining
a common base microwave po~~fiaritistor.
power meter used was al~ a cohwnt.

The

Additionally, assume that a simple two.section

impedance matching network matches the 7Q to
Table I shows inserl~@{&~~~sertion
10ss correc 50Q. let this two-section match consist of two
tions, inticated R@#~8&~nd
actual power data A14 wave transformers.

,.t.,),.,,.

of each load s*&,rnaximum

error of 0.52 dB
Given the conditions we have hypothesized, the
was detected w}~.~ standard deviation of .19
dB. A1l.these.&ads had a VSWR less than 1.1:1 RL of 7.02Q represents the collector load that
wd yield the best simultaneous satisfaction of
at t~?~,equencv tested, A VSWR of 1.1:1 is
Whats wrong with these amps? Probablv
detice efficiency, device gain, gain transfer
the published specifications of comnothina, This scenano, in one form or another, is be~,th%
characteristics, and saturated power.
all too-common in the design and manufacture of,, ***Y
available attenuators, directional
nontinear RF circuitrv. The culprit is correlatiom+$,.,yYquFfers atld RF switches from most leading
For minimum a, with a 2 section match, the
of test systems. A difference of .5 dB is {nw<;]$~~nufacturers.
A VSWR of 1.5:1 is a tv~ical
to fail units that are perfectlv good, rew~k~.
VSWR specification limit at 1.4 GHz. It must be transformation ratio of each section is
noted that manv users will gladlv pav an addi
unnecessary and expensive retesting $r e~
reworking, Still worse, a half dB$,m~w#pass
tional nominal charge for components meeting a
units that dont meet specs ,,,,.:+4,,
and!??ve$should
be tighter VSWR spec.
\i>,:>Q
?
,:!,
shipped.
$!,, ~:j~
THEORETICAL
APPROACH
, t:~$.y<{.:..
Such correlation error$Wdl ~lsrupt an even more The vehicle for the theoretical rfiscuss[on IS the
A 4 100
well known expression:
important functi~,~ ,:hatN:maintaining
product
continuity. A dw
~@lt in 1982 should perform
2.67.
PO = [Vcc -VCESAT)2
the same .~,~, ~Mcal model number device
7
built in, lw~:&other
wav of saying this is that
2RL
a d~~~~?bd
in a 1982 test system should
Where: PO = Power output
Zo 1st section = d@~the
same results when tested in a 1976
Vcc = Collector supplv voltage
~at,~.
The kev, of course, is RF correlation.
11 .44Q
VCESAT=
CollectorEmitter
saturation
$.~,q ,>,voltage
Zo 2nd section = v (71(2.671(50)
~$<~~at
is RF correlation? SimplV put, RF correla
,?;...,,
RL = load resistance.
-.
tion occurs when target error [mits are estab
~l$rs*
30,5BQ
tt.
rhis expression is vahd for a narrow range of fll
hshad and adhered to on a continuous basis
(10% ranga maximum], Over a wider range of
among two or more testing stations. Such corA14 @ 1 GHz = 2.95 = .075m
RL, significant changes in VCESAToccur as a
relation is essential to cost.effect production of
Table II shows the transformed impedance at the
function of RL. Output power varies with the
non.tirrear RF and microwave power amphfiers,
input of the matching network as a function of
whose circuits are extremelv sensitive to the im. square of VCESAT.VCESATis a verv strong func

7-5

~
=
v

237

Table

1. Micmwava

The wbicle Wd
1400 MHz wth

Load

Substitution

Circuit
Rasurn
LOS

Powsr
Lsuel

Load

I.lw

1
1
2
3
4

35 d6
16d6
15.5 dB
15.5 dB
15.5 dB
16dB
15.2 dB
16.2 dB
16,6 dB

7.7W
7.6W
7.65W
8.OW
7.2W
8.3W
7.75W
7,78W

5
6
7
8

Study

for this lest was a production test fixture and cortiation

in out oower of 1.1 W.

sample #2 for the TRW MRA 1417.6

broadband, h!gh.gain transistor.

Oelb

COl~Or
Current

Measured
lnse~ion
Loss

30.03

dB

.51 A
.5 A
.51 A
.51 A
.505 A
.51 A
.505 A
.503 A

30.03
39.66
39.66
39.8
30.16
39.7B
39.73
39.7

dB
dB
dB
dB
dB
dB
dB
dB

Calibration
Error
+.03
+.03
-.44
-.32
-.20
+.16
+.22
-.27
-.30

dB
dB
dB
dB
dB
dB
dB
dB
dB

Actual
Power

from
Reference

thm
7.75W
6.67W
7.1OW
7.63W
7.47W
7.B9W
72Bw
7.26w

cafibratlon
reference

Lmd Input
Rsturn

+.06
-.27
-.26

99.1
99.1
-77.5
-171.5

-40,2

-40.2
-34.1
-34. i
-30.1
-317
-32.7
-35.4

dB
dB
dB

6B.1
-128.0
-144.6
11.9
-111.9

,..

Laroest Oelta alter cahbration correcoon IS 0.52 dB

Me;n value of the measured power = 7.41 W.
Standard Deviation = .34W
Note

-30

Table

= .19 dB.

dB RETURN 10SS - Q of 0.03 and VSWR of 1.06:1.

11, RL Effects

on Output

(Q)
45

,.\..

Power

Transformed
Reismnce

Load Resiatanw

Load
(Q)

Cumulative
AdB

Output Power

[w)

630

AdB

\l
$~~xw?s

5573
.095

..~f,..\.\

,~~;~~~ke

$,.*
I

&&

46

a bad clrcult

look good.

N
analvsls was done for a s!nale freauencv
Theprob~em
!s compounded
In a-broadband
environment

bv requirements

for a good broad

band load Impedance.

47

TEST EQUIPMENT
Test

4B

equipment

ACCURACY

manufacturers

some verv impassive

howaver,

49

considered

Considating
51

thenumber

equipment

53

dB addad to the

earfier

VieIds a total

possible

real.w~impedance

rotated
data

match

also is appropriate

raal match
the input

Our example

in that

center

to the matching
but wont

presented.

change

utilizes

for convemence,

CONCLUSION

The

The data presented

how

on realitv

The author

for an imaginarV-tOof the VSWR

network

will

in magnttude

circle

at

be
from

varlation

intoa

relative

to 50Q.

50W
the

10.3W

load with

a VSWRof

The result

+ 5.3 W(+.435dBl.
(.87

Al Make

dB). Thtsis
a good circuit

is a power
The total

enough

to:

look bad, or

238

the power

that

we betieve

for combining

output
Delta

fails this test,

are at fault.

1.1:1
is

of

VSWR.

error
The

is now

50W

a few

com

bad thlngscanbe,

the correlation

of RFpower

tion conditions,
represents

of

posible

are !n order.

beheves

for the test

dB, which

in table

error magnitude
1.1:1

stfict

devices

qualtv

If more
Instead,

reason
on how

standards

shown

to matntain
in Table [Il.

and the
produc-

than an occasional

first

that

analvza

svstem

for tha additional

be tO.2

tolerance

under seties

do not assume

and then the test

target

should

is the optimum

need for easv repeatabihtv

vae~~~.impedances.

dB. Th!s

dB indicated

range of our amplifler

Now wasee

IS no small
is t.13

*.435

Into a load with

power

bv the test

7,05W.

ments

RF testing

This adds uptoatotal

*14%

output

.51 [,.2551
.68 (1.34)
.B7 [:.4351

*3%

+.13

of
55

+3%.

faced

t3%

However,

*,565dB.

54

is

than

ofvatiablesin

manufacturers,

achievement.

avadable

better

of the task

Vears

constructed

equipment

to beno

and tha magn!tude

52

in recent

of a well

using the latest

generallv

have produced

equipment

the accuracv

svstem
50

anaivm

were taken at

Impsdance
Angle

10s

-30.5
+ 36 dB
-.07 dB
-.16 dB

Measurammts

tha test

to determine
error.

a *0.2

device

the devices

Some

circuit

the
suggestions

dB correlation

are

Tablalll.
Suggmtions
1. Sertahze and document allcomponents

Notes

to the Maintenance
(attenuators,

of Correlation

5. Beselectlve when uslngcablestn

test systems. For
example, the MILC.17 speclf!caoon for RG cable

dirwtional couplers, power meters, detectors, etc.] of

the test svstem Do not dtsturb the system once
cafibrat!on has been performed. Cafibrate the system
once a month.

tvpes says that RG58 can have a character[stlclm

pedancefrom48to
52 Q(maxlmu
VSflRof
1.041) whmterminatedtn
a perfect 50 Qload.
6. 8everV selectlve when choostfig RFswrtches. The
VSWRof a mechanlcaiswltch
wdlvarv w)th time.

2. Require that Ioadshave acahbratlon return loss

>-35
dB (VSWR of 1.05:1) in frequencv band of
Interest

7. If possible, terminate the system with a 50 Qload

rather than an attenuator, Load manufacturers ne,$$z

3. Dedicate test svstems lospeclftc ctrcuitsor

products. This lsnecessarv for both correlation and
p!oduct continurtv

only consider the VSWR of a load. However, fortszq

attenuator, tradeoffs must be made betwe~~W,$
and frequency response Measure power an&i~h$~
performance parametersv[a cabbrate$ ~ct!o;~~
.$..,,.,,,**
coup fers.
~ ...V+.*
~!q,,,~:.
,,>,!1 .,*
;,*, .~~ .
b;v{
<$

4. Theplacemenl
of transistors ln the test f!xtures
must be uniform. For Instance, flanged transistors
should be QlaCed [n the test fixtures with the device
pushed towards collector load csrcultrv.

239

..

240

-.

. . .

.
.-

AN I 033
Match Impedances
Heres

in Microwave

Amplifiers

and youre on the way to successful solid-state

how to analyze input/output factors and to create

,,.~:t.,,,,
..t~>,<
,i$yh{:<{:<+

\\,.}i,,
.~,.
\
,:t~.
~,?

designs.
a practical .!,..\..
d$~~fi.

Prepared by
Roger DeBloois

The key to successful

solid-sbte
microwave
power-amplifier
design is impedance
matching.
In any high-frequency
power-amplifier
de
sign, improper impedance matching will degrade
stability and reduce circuit efficiency, At micrw
wave frequencies, this consideration
is even more
critical,
since the transistorg
bond-wire
inductance and base-to-collector
capacitance
become
significant
elements in input output impedance
network design,
In selecting
a suitible
transistor.
therefore.
:J.>1>,),
>
keep in mind-that
the input and output im~~
,.+:r~hf$ ca~acitance
w is physically practical and to
,.:
anew are critical along with power output, gain,,.,,,,=<:,,,.
>. ]rovide
the balance with high-quality
chip ca.<,++
....?i>. pacitors.
and efficiency,
The first section of the impedance
matching
Unless the selected transistor
is used at fr~~l <;
network is extremely
important
because it mn
quencie9 that are much lower than the m~xi~qm
degrade the 9tability of the amplifier
if it is not
operating
frequency,
the input
imp+~,nce
is
well designed, Depending on the d~ign frequency
largely inductive
with a small real .:Ra&. The
of the amplifier
and the transistor
selected, the
large inductance
is due to bond ~f%.i%at
conresonated
real impedance
can range from less
nect the transistor
chip to the k@@\~kad of the
package
and to the common-et~~e~bond
wires.
than 50 Q to much higher. When it ig below 50 Q,
an additional
IOW-PS matching
section can be
The small real part of the in,~~;~]~pedance
is due
conveniently
added to achieve the requird
50~
to the large geometrim
r~dK~~ to generate high
impedance at the input.
power at high freque~$ieS%,,,Q~e b~e bulk resistThe higher-impedance
case presenti
a gpecial
ance may be the ,,l+&!*,.
pre~~~nant
part of the rwl
,r$*,,.
input impedance. .~~
tt :::)
problem
if microgtrip
techniques
are used to
,,s?.?*..
build the matching network. The problem occurg
,Cb$~.
v.,..
.,>\
because the resonated impedance may be as high
Use microstrip ~g,k at input network
.Wi>.
i
as 300 Q. Reducing this to 50 fi by use of a lowThe fi,&X~@ most im~rtant
gtep in designing
pags network
configuration
r~uires
a seriesthe im~~a,~~%hing
network for the gelwted detransmission
line that will behave m an inductor.
vita,h%,,%~provide a shunt capacitance
that will
The rule of thumb is that the characteristic
impedanw of the transmission
line must h at Iewt
r~~~~~> the inductive component
of the input
~trn,~dance.
This step forms the low-pass ma~h.
twice the higher impedance before guch behavior
raults.
Examination
of the accompanying
@ble
,$j,iit~>i~ section of the network and should provide
~%e smallest possible transformed
impedance, To
shows that characteristic
impedance lines of great!~.;t
,,.,::,
k;-.
!,>,
,s,, s~++ minimize the inductive component, the input and
er than 100 Q are very narrow. Narrow trans~t,...
./?:,:,
?),,,.
common+lement
lead lengths must be kept short.
mi99ion lines (less than O.01-inch wide) should
~,s,
The resonating
capacitance
is generally
best
be avoided wherever pogsible, bwause repeatability of width dimensions
is poor. Also, the loss in
provided by a microstrip
stub. In some cases the
a narrow line may kome
exc~sive.
A better
stub producing
the required
capacitance
is so
solution is to use a quarter-wave
transmis9i0nlarge that a practical
circuit
size cannot
&
Iine transformer
with a characteristic
impedance
realized. It is &st then to distribute
as much of
241

equal to the square root

product:
Zo = V 50 Z,.

Make output bandwidth

of the 504

impedance

wider than input

The output impedance

of a microwave
~wer
transistor
is usually defined as the conjugate
of
the load impedance
required to achieve the device performance.
A typical output equivalent cir.
cuit is shown in Fig. 1. The capacitance
C,,,,, is
nearly equal to the collwtor-b~e
capacitance
C,,,,
specified
for the selected transistor.
L, is the
inductance of the bnd wireg uged to bridge from
the collector
mehllization
area to the pxkage
the inductive
output lead, and L,,,,,, represenk
effech
of the common element bond wirw,
For correct
operation
of the transistor,
the
ultimate load impedance must be transformed
to
a real impedance
across the current generator.
This real impedance
is determined
by

[Vr

~L=

layPiate

CY, ensures that the input power will be .x%W*

before the collector
1SWR
rises to ~~~ft%at
endanger
the tran9i9tor,
In this ~a~~~h% transistor is protected from of f-f requ~~@~*ration,
The amount of additional bandw~,dfR,.?~q uired for
protection
of the transistor
,@~e~M# on the ruggedness
of the t ransi stoL ,~<,~$ The man ufai turers
9Fecification9
fod ~v~!tR tolerance
and
input Q can be a guid~.g~~+terrnining
the bandwidth requirements
~ ~ae input matching
ne>
,...~ .$3,,:
work.
One twhniqq~%~~,$btaining
the requird
bandwidth ig to %@a?e
a portion of the capacitive
reactance +{hp.ti$
transistor
output
impedance
with a ~Q,pRl)~nductor. The shunt inductor can
also & u:~ to feed the collector
supply voltage
toJ~~&li~,~&nsistor. Add itional transformation
may
frOm

t~,

vie (sat)]:

.-Zr,,,,

The load impedance

presented
to the package
terminals will contain the real impedance
at the
current generator,
transformed
to a lower value
by the low-paqs L swtion formed by C,,,,, and the
par~itic
inductances
L, and L ,,
Usually the
reactive part of the load impdance
is made inductive to tune out the residual capacitance
of
the device.
The output matihing
netw,ork should be de.
signed so it h~ gr~ter
bandwidth
than the i~$
put matching network. Providing
a good collw$b$~
match, both above and below the de9ign frg~t$n- ~

@,.&@ined

2. With this typical microwave amp~~~?breadboard

a metal
the entire board can be so~~.~<?~
to provide a path for thermal g8@+ngY
.,. ,/.,<> ,,s,
out,

a low-pass

matching

9Ktion.

~{:;?
In a&dWon to matching the device impedances,
dire~?~?~r~ent paths must be provided to the base
a~~~QJ@ctOr Of the transistor.
The cOll~tor path
-<t:&Wo~ided by the shOtied stub in the im~da~ke-matching
network.
The base path requires
$* the addition of a choke from the base to ground.
The choke can be a lumped element or a distributed shorted stub of sufficient
impedance
to be
negligible
in the circuit.
A quarter-wavelength
stub is ideal. The narrowest
practical line should
be selected, In addition
a dc blocking capacitor
is required in the collector circuit. Also needed is
a bypass
capacitor
to prot,ide
the proper
ac
shorting point for the inductive stub in the collector-matching
network.
Selection of a blocking
capacitor
i9 relatively
straightforward.
The capacitor
should he chosen
to pro~,ide low loss at the operating
frequency
while maintaining
the capacitance
at a value
that inhibits low-frequency
oscillation.
The latter
is caused by the series capacitors
tendency
to
display
riging
reactance
with decre~~ing
frequency.
Blocking
@pacitors
mu9t be lar~e enough to
preserve coupling characteristics
do~vn to a frequency where the shunt-feed
chokes can effw-

242

Microstrip

Z. and veloci~
(Prepared

factor vs width-to-height

by Don Schulz,

Teflon
K = 2,55

Nr

K = 1.0
W/H

z.

Applications

Vp

z;

Epoxy
K = 4.25
Vp

I
I

1.000

93.199

0.654

1.026

136.171

1.000

88.941

0.653

129.916

1.000

84.745

0.652

1.247

z.

87.986

0.522

60.977

84.414

0.521

58.44 L~*j;:bv361

80.870

0.521

55$@27 :~#}0.360

0.520

W:%.*I

123.753

I
[

I
I

1.000

80.616

0.651

73.888
I

70.463

0.353

39.337

0.352

0.512

37.164

0.351

0.511

35.053

0.350

().510

33.007

0.349

* 45.451

0.509

31.030

0.347

60.524

1.516

111.746

1,000

72.597

0,650

57,+*

1.672

105,926

1,000

68.721

0.649

.843

100.242

1.000

64944

0.648

2.240

89.327

2.470

84.115

2.723

79.076

3.002

1.000

0.647

$~:~@4

.$$,%%?$
~ +51:$23
$:}i3

~&291

]t

1.000

57.714

1.030

54.271

] ,.$9if~Y*

42,709

1.000

50.951

?$~,*4++@4

40066

I 0.508
I 0.507

47.75$$; ..,%&O.643

37.527

0.506

44.6~$

1.000

74.218

61.273

o.~6

0.357

~o.51 3

0.651

0.354

76.565

94.706

0.359
0.:

41.568

I
~~77fit

1.000

1 %~Q4$85

I 0.518
0.517+>.,,%
,..
, 48.566

Vp *.l,,
>;:;:y\:!,,
.$s!
!+
.,.,,
1,,,,,$t,~q$

04$+%$$ r: 46.187
I

,,, k#T~..
43.853

67.090

117.692

2,032

Alumina
K = 9.6
Vp

1.375

29.123
I

27.289

0.346
I

25,531

0.345

69.546

1.000

< 0.643

35.094

0.505

23.849

0.343

65.065

1.000

$$$1.759

0.642

32.768

0.504

22.244

0.342

,, :,:, *.959

4,023

60.779

1.000

0.641

30.550

0.503

20.716

0.341

4.435

56.689

1 oq.@ t. 6+36.292

0.640

28.440

0.502

19.266

0.340

4.890

52.796

~.Q:6q:;#

33.760

0.639

26.439

0.501

17.892

0.339

5.391

49.100

31.360

0,639

24.544

0.500

16.594

0.338

0.499

15.370

0.337

14.218

0.336

5,944
I

7.224

>im:tpws?

7.965

,,:.3Q6\#33

8.781

~+$%;;33 .484
r 30.9}04

9.6tiN
RO.$j%~Ji
qp%b

28.491
-

.J
~~.i:.14,304
,,$.:y,~\\,,,i~ 15.770
$Jy,,.>.j,:!?,:.,
17.387
!,.(..., >.$:,,
.
,$
\\@.>
19.169
$.$!
F
i..
21.133

26,240

\
I

26.952

0.638
I

0.637

22.755
I

21.069

I C1,498

1,000

24.93E

0.637

19.485

0.497

13.138

0.335

1.000

23.047

0.636

17.998

0.497

12.125

0.335

1.000

21.275

0.635

16.606

0.496

11.179

0.334

1.000

19.618

0.635

15.305

0.495

10.295

1.000

18.071

0.634

14.091

1.000

16.629

0.634

12.961

I
,

0.495

9.472

0.333
I

0.494

8.707

0.332

,.493

7.996

0.331

1.000

14.043

0.633

10.937

0.493

7.338

0.331

1.000

12,888

0,632

10.033

0.492

6.728

0.330

1.000
II

15.172
14.3 185
13.162

1.000

11.818

0.632

1I

10.830

It 0.632
0.631

9.074

1.000
I

1.000
1.000

9.917

8.299

I 0.631
0.630

243

I1

9.198

0.492

8.425
7.713

I 0.491
I
0.491

7.056

0,490

6.451

0.490

6,164

I
I

0.3332

20.381
18.702

29.091
]

22.192

17.1, 48

23.300
25.688

42.@3~ $}l.OOO

0.344

3.310

, *i$~;@,@
45.60 Q:, :<thp~oo

0.356

3.649

6.553

142.506

1.131

ratio.

Engineer)

77.360

0.931

(W/H)

0.330

I
I

5.644

~
I

5,164

0.329

4.722

0.328

4.315

0.328

Table continued
Atr
K = 1.0

z.

W/H

Teflo,,
K = 2.55

v.

z.

EPOV
K - 4.25

v.

z.
5,894

4.483

41.843
*46.132
50.860

56.073

61.821

68.157

75,144

82.846
91.337

100,700

Alumina
K = 9.6

v.
0.490

z.

0:489--

Vp

3.%2

0.327

3.598
3.284

0.327
0.327

2.995

0.327 $ ,$.:$

8.370

1.000

5.266

0,629

4.089

0.489

2.731

7.634

1.000

4.801

0.629

3.727

0.488

2.489

,ym+~~

3.397
3.094

2.267

<7;$@.3,$~

I 0.488

6.960

1.000

6.343

! 1:000

1.000

I
I

5.779

5.264
4.792

4.362

1.000
1.000
I
1.000

3.969

I 1.000

3.611

1.000

4.376

0.629

3.987

I 0,629

3.632
3.307
3.010

I 0.628
! 0.628
I 0.628

I
1-

2.739

0.628

2.492

I 0.628

2.267

0.628

2.818
2.! i66
2.335
2.125

1.933

0.488
I 0.488
0.487

.<

0.32B*?,.,P

2.0(~w~:&>26
l$$80*tc~ 0.325

.,:@$kl

0.325

0.487 ~:$~k*i557

0.325

0.4871%:f~r$f1.417

0.325

] 0.4~Yl

1.:289

0.325

1.172

0.324

I
W7he~ p~~~ic materials
are uged, its a good
prac~$~$.to mea~ure the material
thickness
and
are comdi,~~~~~$~consta]lt,
because variations
.~~q$,ij~ a recent tegt the dielectric
constant of a
t~&~ptof
epoxy fiberglass
material was measured
~t 4.55 at 1 MHz and 4.25 at 5(JO MHz. If the
,.J$
~ manufacturers
~alue of 5.5 had been used for
the design of matching
networks,
considerable
error would have resulted.
The physical dimensions
of the matching circuitry may be calculated
from the data in the
table. The line lengths are scaled by the velocity
fact~r, which is equal to Z Z
in air for a
constant width-to-height
ratio, }V H.
The final design of a typical breadboard
microis shown
in Fig. 2. The ground
wai,e amplifier
areas on the top of the board are connected to
the microstrip
ground plane by 2-roil-thick foil
wrapped around the edges of the board and the
areas directly
under the emitter
leads of the
transistor.
The foil is secured to the top and bottom surfaceg \vith solder, Plating may be used
for production
units. The entire board can be
goldered to a metal plate to allo\Y connector
mounting and to pro~-ide a thermal path for the
heat generated
by the transistor.

tively short the respective

port to Xround, Coupling capacitor
should not be excessii,ely large,
or they may produce as much as l-d~ loss in
gain with a corresponding
decrease in efficiency
in the case of collector coupling capacitors.
The
Q of the coupling capwitor
determines
the acceptable range of capacitance
values and is g~perally in~,ersely related to capacitance.
*klL.
Bypass capacitors
are gelected by ana~,~{~ d?
the same considerations
as those for,~b~%~hg
capacitors.
A large bypass capacitor$~,w~~~lum
or electrolytic),
placed from the dc~~~~~bint
to
,4., &.
ground, prevents tendencies tow~;~j~~-frequency oscillation
in the circuit. Al~$:~%:.@Y be neces-

gary to add smaller bj~ass $#pa&h&6>sto preser~,e

stability over a wide ran,&&$~f*$requencies.
>..{?.
..,-
. .*F.
Adjust for bandwidth,,:@~~}sical

dimensions

The circuit d~gri~~

readjusted
quickly for
reg~f~~eents
through
use of a combandwidth
puter optimiz$~~~~sprogram
such ~ Magic, offered b}. &~~x@~sity Computing
of Dallas, Tex.
!Vhen th~~,$~$p is finished, electrical dimensions
m(]s:~,$l$i.~~hkerted to physical dimensions.
in the desi~n sequence, the di,$.$3t~#point
$$WitP!a*:tiaterial must be chosen. Three common?~$>~f%>dmaterials
are Teflon fiberglass,
epoxy
t)~:~:~)$rglass
and alumina, Above 500 MHz, epoxy
;,>$6.
$.fiberglass
exhibits too many losses to be a good
choice. Teflon fiher~lass can be used up to se\~eral gigahertz:
it has reasonable
dielectric
losses
and i.q e.:t.~yto prncess. Alumina, a ceramic, offers
a hi~h dielectric constant, good dimensional
consist{~rrcy and small circuit geomet~.

The initial tune-up of the amplifier

matching
circuits can be expedited
by use of a netlvork
analyzer and a precision load on the input or
output connector. The circuit can be adjusted to
match the nominal impedances
supplied by the
transistor
manufacturer.
Distributed
stubg are
purposely
made longer than necegsary and are
the
adjusted to the correct length by trimmirrgof

244

.-

..

foil OIL the capacitive

gtubg. The inducti~,e stub
in the output net~vork is adjusted by positioning
of the bypzqs capacitor
along the stub and the
adjacent ground plane.
Thig procedure
resul~s in a load line that is
fairly clmse to optimum.
A transistor
can now
be inserted in the circuit and the collector matching net~vork readjusted
for ma:<imum collector
efficiency.
Stub toners are used to match the
amplifier input impedance, so that only one variable at a time need be considered.
Initially
it
may be necessary to operate the transistor
at reduced collector voltage and power output to avoid
excessive
stregs. When maximum
efficiency
is
obtained, the stub tuner is removed and the input netwrork adjusted for minimum input VSWR.
Now lets design an impedance-matching

circuit

Lets congider a practical

example of a procedure for the design of impedance-matching
circuitry.
The sample
circuit
uges a TRW
2N5596 at 700 MHz ag the active device.
Specifications
for the completed amplifier ale.
----

=
Z;r; [ =

P,,.,
G,

7
Specifications

=
=

50 n,
50 n,
20 Iv,
i dB,

~5r{

minimum.

for the TRW 2N5596

\vith appropriate
values indicated for the sample
design is shown in Fig. 3. The input match is
achieved when the input impedance is resonated
with a capacitive susceptance
of 0.18 mhos. This
susceptince
is realized by use of a pair of capacitive microstrip
stubs. Each stub must exhibit a
reactance
of 2 x 1 0.18 mhos, or 11.1 f). The
length of the gtub may be calculated by
,*

,$i
~~:\,
,,,~;,
,..

For cage of adjustment,

the l$~w~~ 5? the
stubs should be Iesg than 60 degrgeh~!$~ause
capacitive rectance
ig a tingen@~# <t~ction,
the
reactive
variations
per uni~;.%*H
become increasingly severe past 60, d~~,~,
It is better to
decrease Z. rather tha~~
uie longer gtubs to
achieve higher mpaci@ri$&~~heref ore 20 S 1.732
X, < ]g.24 Q. Be$@\>$~$ ig cagier to shorten a
mic;wtrip
gtub $~~$<~
lengthen
it, the Z. of
15 Q, for exam~~;$~ovides
sufficient adjustment
r4ngetoac~~rnwate
device variations.
The ne+,$*
is to transform
the resonated
imped<p~~? t~ 50 Q, This is accomplished
by a
seri~f$~~~migsion
line with a characteristic
impe&*e
of 50 0. From Fig. 3, we see that the
:+~gtfi.of
this line can be directly determined
to
,}:i:,B@,0.062 wa,,elengths,
or 22.3 degrees, long. A
$,* Capacitive
gusceptance
of 0.040 mhos completeg

are:

P,, : = 20 f~ at 1 GHz,
= 55; minimum at 1 G~@\
7
GP=o dB minimum at 1 ~,H~?,~.
z,
= 2.5 + J4.O at 700 #<~$@
Z,., = 6,0 -- J12.5 at ~&b,,m>~Z

converting
the degign to microgtrip
dimensions,
it is convenient
to choose a Z,, for the second
stub that is equal to that gelected for the first.
Therefore:

,,,-kc,?~*\.

tan O=$=~=O.3,

In practice,
the gain of, ~J~g&.~mon-emitter
amplifier
decreages at a EJ~l;&S4 to 5 dB per
octave. The 2X5596 at 7,Q,0i~~z produceg about
7 dB of gain. Theref#~ ~w~roximately
4 }V of
drive will be requi:g~$~,$roduce
20 W of output
poiver. The collec$or ~$fficiency can be expected
to increase at$*:,@@r
frequency, but it is difficult to estimat~.b@au9e
it is a complex phenom~
,:,~,*
non, Ma~&f@ctuPer9 curves of typical behavior
.x!.,,
are usefWl*kQ*tput
power will not increase sig,.+,
nific,~;@$.,:lmhth the decreased frequency.
~Q~t&?*lciency-f requency
relationship
depends
@~.f@ylce f, and ballasting.
Heavily
ballasted
~~<~%istorg
tend to give increased
efficiency
as
,s~ ;.f.#quency
is decre~ed.
However, they level out
.i$:
! at a lower efficiency
+,,+s:,.
than a non ballasted
part
~., becauge of IR logges in ballast
resigtors.
The
$:T
..\$x,3b.,
f:......
average increase in efficiency as a result of de*
creasing
frequency
ig about 20 j; per octave.
Values from 10 to 407. per octave have been
measured.
The initial phage of the degign ig begt accomplished
on an immittance
chart. The chart

or () = 16.7 degreeg.
In thig case the length chosen is 20 degrees to
allow for some adj ugtment.
The output match ig achieved by partial resonating of the deviceg output impedance with an
inductive gusceptance.
While the amount of susceptance chogen ig arbitrary
at thig point, the
output
network
bandwidth
is affected
by the
value. From Fig, 3, we can determine
that 0.05
mhos is re(~uired for the first matching element.
This susceptance
is achieved by use of a ghorted
microstrip
stub. The length of the gtub may be
calculated
from the equation
L

If Z,, of the stub ig arbitrarily

chosen to be 50 Q,

() = 21.8 degreeg
Again, the stub ig made
cause it can he adjusted
245

somewhat longer beby sliding the chip

capacitor
(ac short) up or down the line length.
The remaining
transformation
is achieved by a
50-[) geries-transmission
line of 0.15 wavelentihs
(54 dexrees Ionz) and a ca~acitive
gusce~tance
of 0.014 mhos. Selwting a pair of 50-ohm microstrip lines to provide the susceptance
requires a
stub length of

*&:fTke immittance chart, with values specified for

example, Indicates the necessary rnduct,ve

$~&:Q&!gn
.,,
.*5.<
t<,.

x. = 2 x ~,;i-4

= 143 n

tan ~ .:. = 15:3.

= 0.350 = 19.3 degreeg.
A stub length of 25 degrees
will
an adequate
allowance
for adjustment
circuit. =

the
and

capacitive
stubs
Impedance
transformations
ed by 50-Q series.transmission
Ilnes.

provide
of the

are

achiev.

246

..-...- ,._ .__. -.__.

__________

AN I 034
Three Balun

Push-Pull

Amplifiers
.:,

,,

~.,?,i.
,.,.,..
::;~,~>.

*.,>
,.y ,, ~,i:>,.ii.
,.

: !

..

.:$ .~.}i:+

,.$~
*

LE RF power transistors seldom

S
Ivjsfytodaysdesignc
riteria;seeral devices in separate packages; or in the same packa~e
(balanced, push-pull Or dual transistors},
must be coupled
to obtain the required amplifier output power. Since highpower transistors
ha~re very low impedance, designers are
challenged to match combined devices to a load. They often
choose the push-pull technique because it allows the input
and output impedances of transistors
to be connected in
series for RF operation.
Balun-transformers
provide the key to push-pull design,
but they have not been as conspicuous in microwave c~rcuits
as at lower frequencies. Ferrite baluns have been applied
up to 30 MHz; others incorporating
coaxial transmission
lines operate in the 30-to-400. MHz range.

,\ .,1:>~>

The success of these two balun typ~~ 9h~u$~ prompt the

nl icrowa~,e designer to ask if balun:Wn$Formers
can be
included in circuits for frequenci.~{~?~$~OO
MHz. Theory
and experimental
results lead tot~e @phatic answer: yes!
Not only will baluns function ~,m~~tave
frequencies, but
a special balun can be d,:@$~
in microatrip form that
avoids the inherent co,~h%~~rirproblems
of coax,
On the next six pa~r~op
will observe the development
of three balun-trarrsti~~:
~-culminating
with the microstrip tersion. No~>W$g%he baluns was tuned nor were the
parasitic elemgnts @@pensated. In this way, tbe deviation
of the expa?l~~~ntal baluns from their theoretical
performanc$,,,~&@ %e evaluated more easily. Tbe frequency
limitatiq~s ~~osed
by the parasitic elements also were
obseF~~b&.W*e clearly.
~i. .$
.%!>:
.+):
*~i>
<,.,.,

1. A balun transform
a balanced
ayatem
symmetrical
(with
reepact
to ground)
to
with one side grounded.
balanced
system

that is
en un-

Without
halun-transformers,
the minim~lm device impedance
(real) that can be matched to 50 ohms with acceptable
bandwidth and loss is approximately
0.5 ohms, The
key to increasing
the transistors
,.4
output power is reducing this impedance ratio. Although
3-dB hy hrirt co]nbiners can double the maxim[]m po~c,er output, they lower the
matching ratio to only 50:1. Balun
transformers
can reduce the origi.
nal 100:1 ratio to 625:1 or less. The
design offers other advantages:
the
ha]uns and associated matching circ,~its have greater bandwidth, lower
losses, and reduced even-harmonic
@*&...-.
..
levels.

2. Baluns are not free of disadvanCoupling a pair of push-pull

amplifiers with 3-dB hybrids avoids
(for four-transistor
circuits) one of
these the higher broadband VSWRS
of halun-transformers.
A second disadvantage,
the lack of isolation beI ,,,. w ,,,,,),. 1/
,,,, f,, /JMs K,,,ld
tween the two transistors
in each
push-pull configuration,
is outweighed by the advantages of the balun design in reducing the critical
impedance ratio,

tages.

247

,Zmui,.

3. In this simple
mission
line, the

balun

that

ugeg

a coaxiel

trans-

wounded outer conductor makeg an

unbalanced termination,
and the floating end makeg
a balanced termination.
Charge conservation requires
that the currentg on the center and the outer conductors maintain
equal magnitude
and a 180-dewee
phase relationship at any point
along the line. By properly
choosing the length and characteristic impedance, this balun
can be designed to match devices to their loads. In the case
shown, if 0, = 90 dewees, tbe
matching condition is:

~i=q
,.,

<...

,.,

...

0(,

!..

4. By adding a second coaxial line, the basic balun can

be made perfectly symmetrical.
[n this symmetrical coaxial
balun, the bandwidth
(in terms of the input VSWR) is
limited
by the transformation
ratio, 50/2R, and the
leakages, which are represented
by lines B and C. If Z, =
50 ohms and R = 25 ohms, the bar,dwidth is constrained
only by the leakages.

5. The equivalent
circuit
for the
symmetrical
balun
shows the effect of the leakages (lines B and C) on its
performance. A broadband balun can be obtained by using
a relatively high characteristic
impedance for these leakage
lines. In theory, the construction
of the baluns insures
perfect balance.

:~,~,{%se parameters. The performance of an experimental

&~un will be compared with these theoretical results

\_,_,
...

.
. . . . .. .
,,

\_/
2,.,,.,,
M,,..,,.!.

w,...,,

8. Two
formers

X116 Iine-gection
Chebyahev
impedence
transbalun to a 50-ohm measmetch the experimental
urement system. The balun was tested from O 6 to 1.5 GHz,

C. PC, M,.,A,
<...

w,,.

248

!.,

0,..

...

.,O,
s,..
. . ..s..
...,

!,

m
~: sespwtivey
9. The measured phaae difference and insertion loss
difference, which indicate the maximum unbalance for
the Design 1 experimental
balun, are 3 degrees and 0.2 dB,

0,,,.

:7,

! ...

.O

M,..,.

.....

.,

=,.... .F. . ..!, ..

\.\\

10. The maximum VSWR measured for the first

design Is 1.5:1. Note the comparison between the
calculated and measured response. The performance
shown can be considered valid for amplifier applf~+~:<$,.>,jj)$
tions up to an octave range.
*.,, \\,
/: ,..:
$.:,
,, ;g
.?:~

11

. . . ...&

~_~

\.,

:~ 11 .RThe second balun dealgn adds two Identical

~ :jk~oax lines tothe simple balun just described. The
SS, ~Duts of the identical lines are connected in series
,$
,*,.. to the output of the first balun. By putting their
outputs in parallel, the final output becomes symmetrical. The output impedance is halved.

12. The equivalent circuit for the Design 2 balun

indicates that its bandwidth, in terms of input VSWR,
is limited by tbe transformation
ratios of tbe first and
second sections and the leakages represented by lines
B, C, E, and G. If the balun is designed with Z. =
50 ohms, and Z = Z, = 25 ohms, and if the load,
2R, is set at 2x6.25 ohms, all of the transmission
lines
will be connected to their characteristic
impedances,
In this case, the bandwidth
will be limited by the
leakage alone, and a broadband balun can be obtained
by choosing lines B, C, E, and G with relatively high
impedance and A/4 length for the center frequency.
The balun achieves a transformation
from 50 ohms to
twice 6.25 ohms without causing a standing wave in
the coaxial cables.

,.,0,,,,...

!...,

!,..,,

,,

.0,.

!.,

. . . . .

. . . . .. .

. .

I
249

13. The performance of the Design 2 balun can

be calculated ualng its equivalent circuit. The
calculated VSWR shows a response very close to the
simple coaxial balun (Fig. 10) because the new second
section has four times the bandwidth
of tbe first
section. This design and its two companions
are
intended to have octave bandwidths
centered at 1.1
GHz, the central frequency used in digtance measuring equipment (DME, 1.025 to 1.150 GHz) and tactical
air navigation (TACAN, 0.960to 1.215 GHz). For line
A:Z~ = 500hms, LA = 1799 roils, <, = 2.10 lines B,
C, E,and G: Z= 300hms, L= 1799 miis, {,,f= 2.23:
lines Eand F:Z, = 250hms, L = 1799 roils, c, =2,10.

14. Two
,V4
transformers
match the experimental
twosectlon
coaxiel
baluns
6.26to the 50-ohm
ohm impedance

7,,,1.,,.,,<,. k(.
,,ft,n
IIK).1(,.4M)
.WHZmnoe

Ued

15. The measured

ured insefllon
loss
two-section
coaxial

, the

load. Although
these transformers drastically
reduce the
bandwidth
(in terms of the
VSWR), they dont affect the
balance.

phase
difference
and measdifference
ere ploned
for the
balun
(Design
2). The max-

imum unbalances for these (WOmeasuremen tsoverthe

octave bandwidth
are 1 degree and 0.2 dB
16. The calculated
and meaaured
values
for the
Input
VSWR
for the Dealgn
2 balun
show close

a~eement
between the experimental
and predicted
performances. This indicates that the parasitic indllctors at the connections are negligible to at least 1.4
GHz. Moreover, the balun has excellent balance tu 1.4
GHz and achieves the 4:1 transformation
without
causing a standing wave in the coaxial line. Despite
the many excellent qualities of the Design 1 and
Design 2 baluns, the necessary coaxial line connection
limits them to approximately
2 GHz.
I

..>

J.

L
. . .. .

,..,,,0.

0,.,

,0

17. The problems

associated
wl~~?~ **previous

coaxlat
baluns
can be redu<lce@m~~@inated
by

using a balun that allows a l~ic~strip

coplanar
arrangement
of the input ,&&i$u%yJut
lines, \vhich
!(;?
greatly simplifies the corm~;l?as to the amplifier.
This balunconsists
of @$~Q~&line, A, connected in
series to three elemen$%ij$:the
center of tbe halfwatelength cavit~:~~?:act~~e open-circuit stub, B, ad
the A/4 output ,~w~$and
D.

I
.s!.3

. ...!..

18, The equivalent

circuit
version
balun shows lines

:,,

,0,,,..

. . . . .

,,

of fhe

Design

3 coaxial

Canal Dconnected to place

their input signals in antiphase,
thereby producing
two antiphase signals at their outputs, Transmission
line impedances and lengths are optimized to achieve
the correct input/output
transformation
ratio and a
good match across the desired bandwidth. If only one
frequency or a narrow bandwidth is desired, and all
len@hs are A/4, the matching condition 2./50 =
2Z,z/R, will occur. In this case, Z, [ZE=Zp) and Z,
have no significance except for loss,

I
250

19. Thecoplanar arrangement oflnputand output

Ilnaa can ba accompllahed with microatrlp technolOgY. The uppermost conductor plane contains input
IIne A, output lines Canal D, and the open stub B.
tiuDlinK between these lines is avoided by separating

,.W,

r.

..,.

,4.,

,, .,.

.-

. c. ,,,

them by at least one line width. The middle conductor

carries the ground plane for the lines. To avoid
radiation loss, the center conductor must extend at
least one line width to either side of the upper plane
circuit line. The balun resonant cavitv is formed by
the re~ion bet~veen the
middle and the lower con19
ductor planes. A hole for
o
the ca~rity is cut in the
I
circuit fixture, filled,with
dielectric
and
with the middle q%~S&!
plane.
The $~~~$o~nd
length of th~~,~..~
nominally a h~~~~{,e)ength
at
midbaa$~:.$kd$ avoid disturb#~ af-.?he field dist:jp~j~q~<$he cavity width
qs~+be
at Ieagt three
tlti
the width of the
mid~le conductor
plane.
&he arms of the balun cavity are folded to produce
two parallel and proximate
output transmission
lines.
This configuration
is more
suited
to coupling
two
transistors
than the original layout in which the
,;,
*l:.,,
two outputs were on op , ~Ad:.>.
::k
/ posite sides (Fig. 17).
*

,.,.,.,,..,

,3

~J:
i

,.T

,,/

::
*

> /
..+

,
/

LIM~ .A

~,<,
,.;,
,!?>

/
L!.,

,/

. .

,,,8
,:+
\\.:.~

,,,

. .

,.-,0

L.

L,.,. .

,,

,,.,

cc

,,,

->5

,,,

. .

,,.

-,.

-2,

,.-,,
a

!8.0

m!,

!o.l..
.,-

!77~,M

,.->,&*t

+~

.w,r,

.*~z

,L:ry:v;.~+
L,. -,0
,.~~>.:
,.,,.~>,~
.. .. . ,o.,y~t%~%,.nsa.s
...5..*PJ~.
,, ::$
,: ~,:, ~~1.?
..,.

!.*}>
a

154.**.%

20. The input equivalent circuit for the mlcrostrlp

version of the Design 3 balun allows its theoretical
performance to be calculated. Tbe design parameters
shown provide a microstrip circuit that can be compared with the coaxial baluns of Design 1 and Design
2. Transmission
line A and lines C and D are loaded
~ by their characteristic
impedancesin
this case, 50
and 25 ohms. The cavity and the stub impose the
principal frequency limitation.
The impedances
of
these elements are dictated by the properties of the
i available dielectric substrates (glass-Teflon 0.020 and
0.0625 inches thick).

Y-c,, ~ ~
:;:
),$
.:.4. ,!~l,,
..:*.,+.
.:j~;.
~,>~~,.,,., ,
,.$:t, 4
~.. ,
~
,J,

3
*3

,.,.

,.,l,

,,,

,,,

;=
.,
.7

0,

,,O.

,.

{,,.,,

,..T
..,.

,,

,,

SW .,

,0,s,..
3,

,k

!,

*!...*,

!.

!,

,P

21. Tha input VSWR can be calculated based on

tha equivalent circuit for the mlcroatrlp balurr. For
a one-octave bandwidth,
the input VSWR is lower
than 1.75:1. This calculated performance is similar to
that of the two previous balun desi~s.
Tbe design
of the microstrip
has theoretically
perfect balance.

251

THREE

BALUNS

FOR PUSH-PULL

22. Theequlvalent
clrcult of themlcroetrlpbalun
showg it during
performance
measurement
with
A/16 matching Iineg. The experimental
model ugeg
18.mil glas9-Teflon ({, =
,*
2.55) for the tap circuitg
and 62,5 mil glagg-Teflon
for the cavity.
Balance
propertied were meagured
with a 50-ohm sygtem,
which was transformed
to
25 ohms by the A/16 linesection
Chebyghev
impedance
transformers,
m
which have a bandwidth
T/,< .rwnme?,
f.{ mw.>f.#
&f.
*h,, u,from 0,960 to 1.215 GHz.
,,,# ,h, ,~-,,sf condu,f.r pk.,

AMPS
.
g

~+

23. The unbelance between output ports fora oneoctave bandwidth is shown in the measured
1.5de~ee maximum phase difference and 0.15-dB maximum insertion loss difference.

24. The

central
expected,

frequency

Is

10

percent

higher

but responge ig cioge to the calculated

valueg if relative frequency is considered, If the output
transformers
and their effect on input VSWR are
disregarded,
an octave bandwidth with a maximum
input VSWR of around 2.0:1 can be obtained. The 100MHz shift between the two curveg may be caused by
tbe improper determination
of the folded cal,ity s
electrical length. Similar calculation inaccuracies may
arise from effects at the balun junction and from tbe
electrical length of the stub. As in the calculated
response, the experimental microstrip balun performs
comparably
to the tivo coaxial designs.

thsn

.$ ~ F

..1,.,,,.: ,,.
.,/,q .
25.

The similarity
in the performance
of ~}~,~
balun
designs
within
the considered
ik$~ncy$
indicates that tbe parasitic ele~~~~$~~o
banda

not
significantly
affect the theoretical ~i:p~t[es.
The
frequency limit is higher than I.&&Hz?$* all three.
In the 0.960-to-l.215-GHz
ba~d.<~Jt?(TACAN
and
DME applications), each perf@Ym~ With satisfactory
balance. The table comp:\~tf~fl,@ain
characteristics
of the balun designs.
,J, :3
The phase differences ~~~~o degrees) for all three
baluns are similag$fw those experienced
\vith the
miniature
3-dB @h&fd,#ouplers that are normally
used to combi~$~~n~~tors
for microwave balanced
amplifiers. Q@~%&iinsertion
loss differences of the
baluns arq b*@F~O.2 dB for a one-octave bandwidth
compa@~~t~O.5
dB.
The~~hy$Jcally gimple microstrip balun eliminates
th.q~~-tlon
problem inherent in coaxial designs:
P%K$* variances that breed gtanding waves and
$~~b~ance.
Microstripping
the transmission
lines al;@W3 a designer to choose any value of characteristic
hpedance
of the lines. Consequently,
the microstrip
balun is botb more manageable and more controllable.
Since the balun load impedance
will vary with
frequency, the begt results will be obtained by simultaneously optimizing the balun parameter
w,ith those
of tbe matching network. The transistors
internal
prematching
network must be considered. *

?=-

.,

_,.J1,,,

,.

.708,

,!

0,1!>

1,,.!.

$j,<kik;,$$!c,

Performance of tie Three Balun Designs

Typs of
bahm

Coaxial I
(Des,gn 1)
Coaxial II
(Design 2)
Microstrip
(Oesign 3)

8alun
loads, R
(ohm@)

Maximum exparlmental unbalance

for one-octave
bandwidth
AMAG
A@()
(dB)

25
6.25
25

Theoretlcsl
Input VSWR
foc
960-1215
MHz

0.2

1.15:1

One-octavs
bandwidth
1.6:1

0.2

1.15:1

1,6:1

1,5

0.2

1.20:1

1.8:1

252

.---- .

._

.
-

----.-.-,--- -,-.- --

AN I 037
Solid State Power Amplifier
300 W FM

,4:,,*.,

.?=,s

>1.., >\\,.i\

*:::.,l:

..

When designing an FM amplifier the total efficiency

m,Jst be the first goal,

%,:,

Overall efficiency is the combination of good collect efficiency and high gain, To get a good collector
efficiency the transistors must be operated in class C and the load impedance should match the transistors output impedance at the operation power level. Class C amplifiers are non-linear units. The harmonic
content of the output signal of this tvpe of amplifiers can be verv high and their power wasted with an
important reduction in the efficiency.

253

This fact made advantageous

the use of balanced amplifiers. In such circuit arrangement
all the even
harmonic are Iargelv suppressed and the waste of power minimized. Push-pull amplifiers have also the
additional
advantages of connecting
in series for RF operation the input and output impedance of the
2 transistors. That makes considerable
easier to match the Input and output impedances of the transistor pair. However, as the !mpedance transformation
IS lower, the RF power losses are smaller and
the gain and efficiency higher.
Another important consideration
in the des(gn of an FM amplifier IS the ruggedness of the amplifier.
~:t.,,,,
,,. ..t~>,<
FM transmitters are often operated 24 hours per dav and sometimes remotlv controlled and in difficult
,i$yh{:<{:<+
access sites. The operating
point of the transistors should be chosen In a conservative
wav and the
\ \ .~,.
\
,:t~.
,.}i,,~,?
heat properlv evacuated. A thermo switch should be Incorporated
to the svstem. The ampllfier must
\. .?
also be able towithstand
output VSWR. Although
all transmitters use to Incorporate VSWR protection
~ia:~~+ty$
in their Interlock
svstems, the amplif!er must be designed with the capability of supporting
VSWR ~kF::,,:S&(
3.1 as a minimum. This point can be verv determinant
when considering that on a high efficiency clr~d$k. $
,,,, ./., ~:,.r
the collector voltage swing can be close to 3 times the collector wpplv voltage.
.,, i~~ \:\*
.,s
ii\ ~~i
~t)$
~> .N~
CIRCUIT
DESCRIPTION
-.$, *<.\~~

,,,

..::<.):t.;~f6

Circuit schematic is given in the Figure 1. At the amplifier input there is a two sectio~,%@&n>The
first
section, L}, consists of a short Ienght (= A/20) of 50 Q coaxial semirigid cable. TM WtW*conductor
.
w,!
,**
of the coaxial cable is grounded at the input side and floats at the output.
,t.$. ~:$...
\.F3):.,,.:~.:>\
The second section of the balun consists of two identical coaxial cables, Lz and LY\,.of the same length
that LI but with 25 Q characteristic
impedance. The ends of these two co@~&4s are interconnected
in
series at the input side (thus offering 50 f) impedance to L1) and in parallel@y,.~~~~$butput
of the section.
3.
t.
The combined balanced impedance will be therefore 12.5 Q at the out~~Qf
$$e balun. The inPut impedance of the transistor pair QI and Qz is transformed to 12.5 Q (2 ~,$j?~pwith
the LC network repre,.
sented in the schematic.
.,{ ~1~
~.,?
.<.:,\,;,,
If this balun is well charged bv 2 x 6.25 Q it is well capable $~,:m~octave
operation. However in this
case the LC network that transform the impedances of the t:~h%JoF pair has been optimized onlv bet,,1$
ween 88 and 108 MHz.
.:i.,i,~;,,,.,
,., ..,.,.,,
.:,N,.<,,,$.;
A similar balun circuit is used at the output of the amph~qr. *he main difference with the input balun
is that the coaxial cables are also used in the collect bias~~gbircuit. Care has been taken with the decoupling of the collect bias in order to avoid low fr~$uencv oscillations.
The collect impedence is higher
than the base impedance and therefore the LC<a8~&t transforming
network is verv slmp Ie, onlv L8, L9
..;.,\
~~>
and C,.
,*V
.,,,.,...s.
..,, .,.<)>
.*.>$,
t,,...
.,~$..
.>
y;~:l.:i+
.,)*.,,.;.,,,:
.\i.:..,
..:.?$:.
,,

88-108

MHz;

300 W 28 V

Figure 1. FM Broadband

254

Power Amplifier

COMPONENTS

LIST

c,

= 120 + 80 pF Chip capacitor

ATC

c,

= 220

pF Chip

capacitor

ATC

100

B
B

= 470

pF Chip

capacitor

ATC

100

c,

= 100

pF Chip

capac!tor

ATC

100

C8

= 27 pF Chip

c,,

c,,

c.,

c,,

C,o,

C,2,

c,,

c,,,

C,h,

c.

c,,,

c,.

= 1 000

pF Disc

= 10n

c,.

= 0,1

ATC

VF
~F/63

V Electrolytic

= 50 Q coaxial

cable

LJ, Lj

= 25 L1 coaxial

cable

33,2

L.

L,

L,,

L,

= Line

on substrate

: 15

x 5 mm

L,,

L9

= L!ne

on substrate

: 10

x 5 mm

p;n

B
B

capacitor

L,

Hair

100

= 1 000

cl,

capacitor

100

copper

foil

3,2

18

mm

(Teflon)

L = 110

mm

mm

(Teflon)

L = 110

mm

3 mm 0,3

mm thickness

L 10, L,,

= 25 Q coaxial

cable

@ 5 mm

(Teflon)

L = 110

mm

L 12

= 50 t} coaxial

cable

(Teflon)

L = 110

mm

L 13

= 15 turns

R,,

RI

R,

Q,,

=22

8 mm

1,4

5 mm

mm wire

Q112W

=471i2W

Q,

= TP 9383

-.

M
z

&

Rg;re

2. Component

Leyout

Im

110

f, FREQUENCY(MHz)
Figure 4. Gain

255

and

Efficiency

veraus

Frquency

256

..

- .. .

AN I 039

470-860

MHz
Broadband Amplifier

5W

5 W UHF TV TRANSPOSER
AMP1
WITH TWO TPV 593 TRANSIST(

.-.;-~.
.,?:,<.,
This application note describes an u@lIfiwr
broadband (470-860
M Hz) amplifier, developed for TV
transposer applications. The amplif{~~iff~prporates
two TPV 593 transistors,
,$,.
\

.\~,

~,

Each transistor is used to buil$j<~i$~e~rate broadband amplifier, The two Identical amplifiers are later
combined with 3 dB hybrids.., ;~~<,*
.~>;~..l.,
~,,x
,.,
.!.. ...\/,...\.
The TPV 593 transistor ,h~s ~~~ developed for TV class A application. It incorporates gold metal lization
and diffused ballast re~$t~,~,x$or ruggedness and Iinear;ty. Its DC current consumption is very low and
!makes it a good can~da~
For solar cell powered systems. Its basic specifications are :
+,.,yt.$,,xf:
Ic = 450 mA
, <;~$;c
= 25 V
~,t.t.~~+}
.~, $; G
=9dBat860
MHz
~., **t:v
,,.
IMD = 60d B at860MHz
and2W
output
,.>d:~,,){.
. ..:,). ,,
J*, ~~.,..?

..>.:.?.
The S parameters of the TPV 593 are given in the table below,
*~,,.,,:t,:p.
.!. ..i,
POLAR

S-PARAMETERS

S11
(MAGN

,,.

470.00
650.00
860.00

0.93
0.93
0.92

IN 50.0

IJO
165
162

(MAGN
1,50
1.06
0.79

SYSTEM

S12

S21
ANGL)

OHM

ANGL)
63.0
50.0
38.0

(MAGN
0.040
0.050
0.056

S22
ANGL)
50.0
54.0
54.0

(MAGN

ANGL)

0.55
0.60
0,65

166
169
169

257

S21
dB

K
FACT

3.52
0.51
2.00

1.01
1.04
1.15

POLAR

F
MHz

COORDINATES

SOURCE

OF

REFL.

MAGN.

SIMULTANEOUS

CONJUGATE

LOAD

COEFF.

ANGLE

REFL.

MAGN.

Two identical
single transistor
class A amplifiers
of a singte amplifier
will be considered.

will

be

combined

MATCH

COEFF.

Gmax

ANGLE

with

3 dB

dB

couplers.

First

the+@ ~@ti%
!
,*.?
+~?
,i: ,3
.$:,$,.,,..
,C,>,:k:.t.
.,$* .
~,!:\.\\
From theanalysis
of the variation
of the TPV593S21
parameter
with the frequency
itm@y&&<@ent
hat
there isa difference
of 5.52d
B between
470 and 860 MHz
If a flat gain is required
tbiR;$Qin
slope has
,J ...,,:
to be compensated.
The compensation
can be implemented
in two ways :
\ $,+Jt,,>/.T,
,,~,, , ~~,,.m~

~,.<
,,y

a)

b)

,s, :J.*
By placing
a selective
attenuator
at the input of the transistor
amplifier,
with an,x!:~~~ioss
860 MHz and which
increases
to 5.52 dB at 470 MHz. The insertion
loss incre~&Should
:L.r
transistor
gain slope,
,++*\?
X* *,l.$*)>,
Byselective
mismatch
attheinput
of thetransistor.
Theinput
circuit wil~,@~)I$@ei mpedancem
MHz, in order to gets gain as close as possible to the GA max. Frequency
$~pe~dent
mismatch
the gain slope. At 470 MHz a VSWR
as high as 11:1 will be neces~ry,~,.~as
been proved
mismatch
at the base terminal
of a transistor
power
am Dlifier doesy~bt
modifv
the Iinearitv
device

minimum
compensate

at
the

atchingat86O
will compensate
that impedance
behavior
of the

Asit
wasdeclded
tocombine
two amplifiers
with 3 dB g~~~;~the
method
b) was selected.
500hms
3dB
hvbrid
couplers
when
used with two identical
load%, pr@ide
a good VSWR atthe common terminal even If the Ioadsd!ffer
from 500hms.
Thereflect4,
*,&&y
is dissipated
as the 50 ohms load connected to the fourth
terminal
of the coupler
The coupl;~~$~~ves
as a selective
attenuator.
Figure
1 shows
the amplifier
arrangement
The use of a 3 dB couplqrtbspllt
the Input signal makes almost
compulsory
theuseof
thesame
type of c(rcult attheoutp$.
:

E
BIAS

BIAS

L50m A
25V

OUT

h
Figure

1. Block

Diagram

of Amplifier

The amplrfle~
must be as linear as possible
over the complete
UHF
band. A transistor
usually
requ!res
impedance
matching
at the collector
side for optimum
intermodulation.
output
circuitry
has been designed
for impedance
matching
all over the bands
IV and

258

power
amplifier
Therefore
the
V,

COMPONENTS

L.

65 line
50 line
= 50 line
= 7 turns

L,

C,-c,
c,
C,-c,
C,-c,

=
=
=
=

L,

L,

L,

TO

m=

Variable
6.8
10

pF

11
1.5
17
ID

wire

Airtronic

pF ATC

lnF+l

% g at 860 MHz
% g at 860 MHz
Yo g at 860 MHz
2 mm - Closely
Wound

: 5 mm

ATC

PART LIST

AT

- wire

5 mm
st.,\
.,.,... )\i.
, ::;~~.,?,i.
*.,>
,.y ,,~,i:>,.ii.
,.
.:, : !
..
,,,,* .:&/y:*

1 mm
7275,

.8-4.5

pF

100A
100A

OnF+l~+l

f. ,,,,~$~+.

,1,, ... .,7

,. s,, .,+,!..
~4\..8

OpF

The input circuit consist of a three section low pass

type matching
network.
To minimize
power losses
all the impedance
transformations
are made at a
low Q level, Variable
capacitor
Cl is adjusted
for
optimum
VSWR at 860 MHz. The tuning is straight
forward
and only a small
retouch
is necessary
after the collector
tuning.
The very constant
S22 of the TPV 593 transistor
makes extremely
simple to match the collector to
a 50 ohms load. L8 tunes the output
capacitance
of the device and is determined
for good matching
at the low end of the band. Only one low pass
section
is necessary.
Capacitor
C5, variable,
allows a good shaping
of the output VSWR,
Collector
tuning
should
be done
atier tuning
the
input,
The bias control
in Figure 3.

&
Figure

3, Class

A Bias Grcuit

259

circuitry

is classical

and

is given

CONSTRUCTIONAL
The printed
paid to the
foils ensure

circuit
ground
proper

DETAILS

board lay-out of the complete

amplifier
is given
returns,
Plated through
holes have been used
grounding
of parallel
capacitors
and connectors,

Thecouplers

have

This

isasinexpensive

solution

been

made

with

parallel

wire

in Figure
to ensure

4. Considerate
attention
Iow emitter
inductance.

should be
Wrapped

cable.

as a straightforward,

9 metal

ized

holes

s..\~,.\

k4~$r@$.
!,$

and

Return

Loss versus

Frequency

Pref =~w

------

)uT~

i
_+z7-

g
.
$

.7

~-

--

.._,

--*+

$
g

. --

.
Pref = 4 w

-60

Gain

L;

.*---

..

z
%
~

-.
.

,0

l!.

.--.-.-
.-

500

100

600
f, FREQUENCY

figure

6. Intermodulation

800

860

(MHz)

Distortion

versus

Frequency

260

.___._-

Figure

7. Output

Power

vereue

Input

figure

Power
f = 470

8. Wsion

to Sound

Cro=

Modulation

MHz
~

18

g
~

14

!
L
1-

:.

,/

1
P,, INPUT POWER (W)

Pout, VISION (W1

.:~::.,-t-;i:$

,.I:WL,,, ...
!!

\.\,

NOTE:

A/.

of sound

carrier

261

I 7 dB) when

vision

carrier

is switch

ON/OFF

262

...-. .-.-..

ANI040
Mounting Considerations
Semiconductors
Prepared
Staff

by Mll

for Power

~.+,\,
..
...>:.i~+.:
.<+>
,,*{*
+$;$..
.$;. ,+

Roehr

Consultant,

Motorola

Semiconductor

\\t:k::,t
<>~.,..
, .>\>
!\\@
~~?,
?J
~il .
.,..~.t.,~
,i.f,<.,~~

Sector

...%.~<
...\\,.
.,:.
s..<~~j
~?.:,
,\. ,;.:

TABLE OF CONTENTS
Introduction

. . . . . . . . . . . . . . .

Mounting

Sutiace

Interface

Decisions,

Insulation

. .

. . . . . . .

.,,

.,,

Considerations..

Fastener

and

Connecting
Cleaning

Circuit

Thermal

System

4
7

Appendix
Appendix

ATherm~l~O$%tance
B M~a~~@&ntof

Socket

Appendix

CSou:<&~of

13

Package

,,

Hardware

Fasten ing Techniques.

Free Airand

1
2

. .

Preparation.
. .

Characteristics.
,,,

,,

. . .

Mounting.

.,

,..

,.

and power

parably

linked

to their

lead-mounted

parts

ger is required
exceeding
high
trvs

ratings

to prevent

its rated

limit,

the junction
therebv

Except

running

the

semiconductors

half for a decrease

135C, (l)
pliea
Proper

Guidelines

impose

decreases

in junction

a 110=C limit

mounting

upon

minimizes

th~J/~,R@?ature

between
the semiconductor
cq$~,~~~%f~e
Most early life field failur~,,~%:~~wer
can betraced

to faultvrn:~~ib$

bv oneto
.(2)

heat exchanger.
semiconductors
With

die.

The

toawarpeds$$~c~,
plastic-p@

With

from

improperly

the widespread

kag~@>semiconducto

rs,

nearly

everything

used
die,

the

because

of the area

of the package;

screw

than

face

and)orburrs

around
but

expected,

tends

contact

all

to

Ifa

rough

were

poor

pull

is small

hole

and the

higher

displaced

mounting

prospect

Pusllc 80DY

of

mechani~~~.damag
eisverv
significant.
Mechanical
dam.,! \?&J,, ,,.
ageqanf@@@!rt
hecase
moisture
resistance
or crack the
se@:~&~,uctor
,:.. 1.,.

die.

Cho.Therm IS a registered trademark of Chromer,cs. Inc.

Grafoil ISa registered lrademarkof Uion Carbide
El Dupo.t
Kapton Isa registered lrademarkof

Figure

R. bber.Duc
IS a trademark
of MVID
Eng, neertg
SII Pad isa trademark
of Berq.,sr
Sy.c-Nut
,Sa trademark
of Iwshak.
proof
Thermas,l
,s8 ,eg,stered
trademark
and The,maf,lm
IS alradema,kof
Thermally,
lCePAK, F.11 Pak, POWERTAP ad Thermopad
arelrademarks
of Motorola,
lnc

263

1. Extreme

A Semiconductor

IC

Caae

of Improperly

(Distortion

Mounting

Exaggerated}

in the

practices

use of various
the

junc-

heatsinksur-

be covered,

mounting

the

distortion

area

is a much

thehole

not

screw
sleeve.

enough

bv the large

the result

temperature

the

a sheet-metal

the

consumed

the

or crack

a fiber-insulating

the

In addition

are bent

can crack

wires,

possiblvcausing

tion

would

done.

with

containing

to tighten

to crack

most

the leads

orooerlv

is fastened

hole

packaged
devices,
fa@ty
~@onting
generally
causes
unnecesaarilv
high,~~u~lj~
temperature,
resulting
in
\ ,.,..
reduced
compone~t
h?stime, although
mechanical
damage has occurr~~~$.~~kasion

of doing

interna~bondirig

into the hole,

illustration,

metal

if not

the

package
a 1/4

force

bowing

gradient

procedures.

19

To use the socket,

which,

break

package

SUP-

jum@~:@mperature

package,

The

Q@MOSt

of ,~$~@.~&,pOWer

oDeration

through

in%-

temperatu~/rd~16WC

for designers

,.,a~

fr~@j$,.,Jk

approxi~~~lv

17
18

Accessories.

an example

s~wcket mounted.

riskK:~$i@J

failure rate. Furthermore,

the semiconductor
field historv
indicated
that the failure
r~te

silicon

for

aheatexchantt>+~k.
temperature

16

Concepts
Interface.

wrw.
A tab mount TO-220 package is shown being used
~,\>s ~t?~placement
for a TO-213 AA (TO-66)
part which was

are inse-

environment.

atlowcurrents,

14
16

. . . . . . . . . . . . . . . . . .

Ev@~~?~~

$~s[%:~.ihows

of semiconductors

thermai

used

Handling,,fi[m&~ls
*.* ,$,.?,,
Board~,..~ . . ......

l@e~,
.~~ . . . . . . . . . . . . . . . . . . . . ...20
. .. ,4.%,+,,.%..
,., k:
?} ~v; ~~

INTRODUCTION
Current

and

In many

situations

be electrically
isolation

material

aswell,

which

addition,
duced

voltages

also

impose

further

its mounting
extent,

junction

possibility

if high

agencies

from

is, to some

raises

the

which
thus

the case of the semiconductor

isolated

arc-over

are

present.

creepage

complicates

places

additional

a thermal

operating

of

must

sutiace,

temperatures.
problems

regulating

specifications

Electrical

demands

In

is intro-

Various

distance

design,

TIR= TOTAL
INDICATOR
READING

The

isolator

upon

isolation

the

mounting

procedure.
Proper

mounting

orderly
1.

attention

Preparing

2. Apply
3.

procedures

the

mounting

ingathermal

Installing

usually

necessitate

REFERENCE
PIECE

DEvICE
MOUNTING
AREA

to the following:

the

surface

grease

insulator

(if required)

(if electrical

isolation

is

desired)
4.

Fastening

the assembly

5. Connecting
In this note,
eral

terms

theterminals
mounting

for

packages

will

into

fit

are

the

generic

Unique

requirements
particular

are given

discussed

in this

following

As

when

are

Body

Sufiace

Mount

A contains

with

C indicates

a brief

Appendix

intetiace
the type

review

B discusses

thermal

of thermal

of accessories

resis-

measurement

resistance

dif-

tests. Appen@$x

supplied

,,,J~,~,P*kagea
~~?~onductor
~ c&%afullV
agea
the

package.

In lower

is satisfactory
and

is free

applications,
is required,

be considered.
Flatne:a

Surface

a~~catkhs,

if it ap~$+:t~.fl~$

from

a mOre
Mounting

power

with

in hew ? ~~~h~
% of the

test

as indicated

norma~~

:~ified

as a fraction

The mounting

if.~~$~~~an
4 roils
~Q,:~a,,is satisfactory
~,.>*<,

per inch,
in most

the

and

minimum

finish
below

specimen

to that

in Figure

2. Flatness

of the

Total

sudace

flatness,

normal
cases.

for

Indicator
i.e, Ah,~lR,

extruded

an unnatural
and

several

is the
the

interface
and does

average

mean

of the

value

resistance,

a finish

is satisfactory;
not significantly

deviations

of sudace

height,

in the

a finer
lower

range

finish

both
For

material

the capabilities

using

leave

to conform
perwith

be detected

effect

inherent
holes
for

for

results

results

in hotter

extruded

die sets equipped

and

fixtures,

with

applied

proper

in the

aluminum
of properiv

in both fine-edge
stamped

in

later,

acceptable

are capable
when

but

which

blankto

sheet

heatsinks.

The

Class A progresaivedies

on four-post

imme-

(if plastic),

much

are seldom
used

used

a significant
out of contact

package

until

probof the

or the device

is imposed,

manufacturers

sheared-through

holding

attempts

second

holes

in the

two

mounted

pressure

pads

of 50

is costly

contact

The

as commonlv

holes are pierced

crater

by distortion

may often

in the

of trouble
a punched

can cause

surface

is not manifested

thick

or

and

stress

punched

heatsinks,
ing

cracks

failure.

relatively
utilizing

alumi -

crater

package,

around

This

hole

pressure

The first effect

usually

in the

heats ink indentation,

the

bv visual

Although

is

process.

mounting

bridge

hole

be damaged

of the

packages,

been a source

the area

of its best-dissipating

operation

of a

can

diately

metal

to 60 microinches
to achieve

only

an early-life

var-

Finish

Sutiace

as the

centage

done,

pack-

such that the hole for

the

have

in the

device

the heatsink.

bv comparing

~%~rd

The

mav

of the surface
treatment
must

holes

an

encapsulated
For these

be used

not exceed

frOm the
may suc-

to accommodate

plastic

the mounting

it to the shape

:~ :$#
determined

does

around

holes

of this condition.

size must

if not properly

package

a straight

larger

but manv

is depressed

lems.

high-power

with

mounting

heatsink

,,. ,
,*..:.,:: ...

reference

above

In

deta?%$~.~mtnation
h@,~,: &od surface
.,.
,d>
+.?

flat ~&s~

Re~$h~J~),

against

bushing

hole

the heatsink

dee~ ,., s~?~hes.

screw

because

[n general,
the heatsink mounting
su~~~~%ould
have
a flatness and flnlah comparable
to t~a~$f
fie semicon-

used

are intolerant

Punched

SURFACE PREPAR~f~J,r

>.:.,?~..j,

$utiace

milled

if prepared

generallv
should only be large enough
of the fastener.
The larger thick flange

bushing,

a smaller

!\:j,.,,:.::*,:
>:,,,,~
.$,$,,

MOUNTING

iance

In general,

satisfactory

having mOunting
holes removed
die location,
such as the TO-3,

&

w insulating

by a nu$[~%~,

of manufacturers,

Sutiace

are

with

condition.

M@~~J~&holes
t~~~ow>learance

Mount

concepts.

surface

applications,

,$u~,a~~$

filled

Most commercially
availwill require
spotfacing

.?,,\,

Mount

Appendix

also

in hi~wl~~~,er

were

finish,

Mount@$Holes

Plastic

ficulties

a cOpper

a ty~?~~~,?.microinch

the voidsw$cratches

in goo.~~fbfig
~..,.

Mount

Pressfit

edge

used

tools

Mount

Flange

ductor

when

with

joint com~~nd.~]
or extruw~,,~~tsinks

or machined

Stud

tance

package

Ther@il,#~?us~ng

that heatsink
finishe~~~t~een
16 and 64 ~-in
less than * 2.5A di~.@@K
in interface
thermal

resistance

pertaining

bv

showed
caused

a thermal
able cast

classes

conducted

(TO-3)

note.

defined:

Tab

Tests

TO-204

they

that

on data sheets

The

in gen-

of packages.

it is probable

classes

package.

are discussed

classes

developed,

generic

to

the

procedures

several

newer

tance,

to the circuit

13) Catalog #S7.HS-9 [1ea71, page a, Thermally,

Dallas, Texas 75381.oa3a.

resis-

In..,

PO Box a10a39,

264

----.

_- ..._

,---_

.... . .. .
-

When
with

mounting

extruded

Chamfers

must

transfer

a general

cleanup

because

be broken

they

mounting

raduce

stress.

to remove

device

practice

very

ia important.

burrs

and heatsink

Many

tion

and may punc-

It need

area

when

also

available

under

heatsinks

the

Another

yet

that

For

economy,

good

tance,

which

area.

To

thermal

when

package

brush,

Some

the

correct

while

and

Some

cyclic

compound

spread

experimentaquantity;

too much

copper

very

insert

may

too

permit

should

be applied

used

from

for

the semiconductor
of paints

the

it

to insulate

heataink,

hard

an easy installation

amount

of grease

aasembly

interface.

sinks;

be ass~bfwhith

excess

C6Q,~~ytid

sho:~

or

the

assembly
is
Examination
even

wetting

Insufficient
torque
thermal
resistance

of airborne

should

causes
of the

particulate

be wiped

acetone

,ti~,s$ntact

rnay-~ter

for Iowvoltage

,~$t$

moist,e~,ed:~~fh

the semi-

side
a

shOulQ.;&RpQ@@ around

~oti~.edeveal

To preve~$~~~:,mulation

resis-

anodized

should

tha edges of matin~.~$~l~~es.

a significant
in~@,&i.~h
the

is attached

high thermal

.ty*i;:$.
severa~,:w~{c&h-

amount,

each mating
surfa,~i~~
production,
assemblers
be trained
to slqw!y @ply
the specified
torque
\.J\, t,
though
an exc~~~~,~e b~ount
of grease appears
at

even

of the

to all heatainks.

is sometimas

it is necessary

allow

because

storage

small

across
should

of the oils and

and

correct

amounts
of grease applied
even~~+~&t@#e
mating
surface.
When
the aW~~&:,}~COrrect

of a dismantled

of its thin

properties

be cleaned

the

and heatsinks

perimeter
of each mating
sutia:e:;~
slowly torqued
to the recommq@&~>~ue.

or chromate-

because

of the joint.

samples

different
of each

are

No treatment

ia iridite,

manufacture

because

autiaces

the

area,

to determine

determine

ductor

mounting

Heatsinka

plated

only

paint

However,

painted

the

a nickal

resistance

of the paint where

conductor

from

mounting

electrical

in the

required

contact

Iintless

material.

help

not fill all the voids,

resistance

to
Anod-

but negligible

finish

low

It need

collect

a practice

is usually

corrosion.

is required.

aluminum

has

sinks,

removal

have

offers

oxidation.

or

excess
will

the entire

ia necessary

little will

is necessary.

which

surface,

electrical

semiconductor
treatad

dip,

black-anodized

prevent

be removed

contact

which

are

and

only

electrical

of this sutiace

resists

ability

in significant

insulation.

ovar

a spatula

remova

package

some compound
to remain
between
well mated
metal
sutiaces
where
it will substantially
increase the th$rmal

aluminum

izing results

using
to

of the

evenly

cause

material.

radiation

layer

lightly

rotation

However,

which

thin

wiped

heat

Treetmant

improve

films

drilled,

surface

avoided

between

isolation

Sutiace

are

and increase

must

poor contact

acid

be

sutiace

the edges
ture

holes

aluminum,

or

away

alcohol.

These

plastic-encapsulated

the package

and cause

matter,

using

a cloth
solvents

devices,

a leakage

as they

path or carry

a~~ su&tances
which
might
attack tha aamiconductor
manufacturers
will provide anodized
meeting
specific insulation
voltage~y.t>$~+.,
~ The silicone oil used in most greases
haa been found
.,
requirements,
usually up to 400 volts,
$!$>$
~,..:.$~o evaporate
from hot sufiaces
with time and become
It is also necessary
that the surface
be free froms~,
deoosited
on other cooler surfaces.
Conseauendv.
manforeign material,
film, and oxide (freshly bared alumifi,~
ufacturera
must determine
whether
a microscopically
forms an oxide layer in a few seconds).
Immediatel$,@rlor
thin coating
of silicone
oil on the entire assembly
will
to assembly,
it is a good practice to polish t$~ mounting
pose any problems.
It may be necessary
to enclose comarea with No, 000 steel wool, followed
b~)~n%<~etone
or
ponents
using grease. The newer synthetic
baae greases
alcohol
rinse.
show far less tendency
to migrate
or creep than those
,>/.,:
,,,:. ,.\}
made with a silicone
oil base. However,
their currently
applications.
Some
or painted
sutiaces

a@.

,.

,.~
,
INTERFACE
DEc~J~.~

When

any

pated,

must

mating

surfac:~+~~

interface

high

amoqq&@~~&er

something

between
the

significant

and

thermal

quits

b$~ ~.p&

~,sia@6ke

several

yea%

called

grease,

~~$b:$~n

HQ&~~~~F@he

joint

6VCM/in
are highly

pafiially

Thermal
Joint

Compounds
compounds

conductive
base fluid
time

fill voids

particles
which

do not

spread

well,

when

have

therefore,

when

of grease

is generally

have

under

the

conformal

for

more

Data

pressure.

they

oil or other

dry

should

be evenly

the

The

rating

pack-

is shown

more

washers

valuable

resistance;
are

used,

joint

of the

for a greased

does

use

compound
insulator.

tive

with

pad

compounds

made

applied

(patent

in a

265

bare

joint

and a joint

from

Highly

data

Grafoil,

compressible,

when
it

conduc-

KON-DUX.

flake-like

of Figure

indicates

Another

is called

oriented,

using

grease.

for grease

as grease.

Aavid

grain

pads

with

in the data
the

and

the need

conductive

obtained

is required;
as well

a unique,

pending).

dry

to be a replacement

perform

grease

manufactur-

To minimize

offer

is shown

isolation

available

with

use grease,

compound,

using

equipment

petiormance

is claimad

indeed

some

vendors

approximate

graphite

of assembly

problem,

several

no electrical

synthetic

consistency

of theaa

breakdown

on several

contact

mandatory.

not, or cannot,

3. Grafoil

some

the

surface,

insulating

of the difficulty

grease,

which

and

of fine zinc or other

a grease-like

Since

are

dielectric

conditions

in lowering

mica

evaporation

ers will

developed

(Greaae)
in a silicone

less, they
and

Pada

Because

as

the

are

the

becomes

Conductive

of tha joint.

of problems,

to raplace

are a formulation

maintains

and temperature.

pads

the

improves

mounting

rougher

often

also

are

of compounds

different

1. The

range
conductivity

cost is higher.

the effect

under

grease

thermal

a significant

resistance

a number

pads

in Table

finishes.

temperature
on

and their

age types

unnecessarily

They

working
poorer

Data showing

whereas
air has
pock-marked
with

makes

thermal

causes

~,, dty conductive

and insulating
,,,:><:
~~;%
%:*7aditional
materials.
These
therefore

strength

be

&s@
S&d in the following
section,
:,, .+; ~
-J$, avoid using grease, manufacturers

$$~+
~.,!,
~~<~i,t,
,,,,Z
~ \,f. ...
~.!.
.s

voids

air

Otherwise

usad in the interface.

intetiace
greaae

the

path.

compounds,

of a compound

redu@,QN~in Yhe

slightly

the surface

t$ermal

a resistivity
of #~$#ximately
1200.Cw~A,~inc&
surfaces
$~~g!vse

dissi-

fill

will

dep@*:@+.&~on

For

minute

to

~S:~$hermal

observed

is being

It is

structure

it becomes

Table
Approximate

Values

for Intetiace

Thermal

in Motorola
Dry

intetiace

othewise
a hole

values

noted
in the

are

subject

to wide

the case temperature

heatsink.

(See

Applications

variation

B for a discussion

of extreme

Test
Torqua
In-Lb

Maaaurementa

dependence
located

of Intetiace

Petiormed

upon

directly

Thermal

Intetiaca

Description

from

Laboratory

by a thermocouple

Package Type and Data

JEDEC
Outknes

Data

Engineering

because

is monitored

Appendix

Resistance

surface

under

Resistance

Thermal

Measurements.

Resistance

Wth

Metal-to-Metal

conditions.

Unless

the die reached

through

)
~),?
~
~,!{+
t~.,,t,,,$
:t:+ ,.-}::
~ , , -~+

~CW)

Insulator

S* !*,

Drv

Lubad

Drv

Lubed

Tvpe

,awt&\:
,,, .m.\w.,.
J~
~:]:
,>,,
,,,.+.
.;,,/!,$\>>
a,

DO-203M,
TO-208AB

TO-21OAA

10-32 Stud
7/16 Hex

15

0.3

0.2

1.6

0.8

3 mil
Mica

DO-203AB,
TD-208

TO-21OAC

1/4-28 Stud
11/16 Hex

25

0.2

01

0.8

0.6

$*
5 ~fik+:
Mi6&Lti$T

DO-208W

Pressfh, 1/2

0.15

0.1

TO.204M

Diamond, Flange

0.5

0.1

1.3

Diamond

1.5

0.5

2.3

2.0

1.3

TO-126

Flange

Thermopad
1/4 x 3/8

Manufacturers

data

shows

an interface
thermal
resistance
better than a metal interface with filled silicone
grease.
Similar
dry conductive
pads

are available

fairly

recent

exist,

have

from

other

development;
not yet become

INSULATION
Since
device

most

construction

lector
lem

of isolating

one.

For lowest

important
to

from

the

ground,

evident.

it is common
the

overall

cornti~n

rath@~&ep

to use an insulator

aw~~~heatsink.

possib~t$,~~wever,

safety

reaW~&~:Jn?tances

where

a heatsink

isolat@p~Wages.

In these

~~!$ackeges,

Heatsink

because

heatsinks~~mte

individual

iconductor

situations

components

importance
two

intetiaces

than

power
stand

with

exist

is used,

thermal
of one

because

chips
wide

heatsink.

used in the past because

and fairlv

Full pak and EMS

tainlv

within,
burden

grease

contact,
and

some

rubber

with

insulators

itvto

fill in most of the metal

for thermal

suffered

from

ones

bv having

grease.

presentiv
imbedded

are also shown,

thermal

has been

widely
voltage

at a low cost but it cergained

under
voids

have
of

first

a few

abil-

reduces

introduced,
Vears

solved

Kapton

because
Their

at the intetiace

When
after

favor

prassure.

thev

in service,

this

problem

or fiberglass.

266

_.__

low

in contrast

high breakdown

have

available
pads

for

to with-

grease.

conformal

cut-through

isolation

resistance,

Mica

it offers

are somewhat

the need

for

material

materials

resistance

be used

Thermafilm

is used

of insulation

thev

The

because
materials,

not

toxic.)

easilv.

of handling,

low thermal

should

Silicone

is of greater

ease

high
the

(It should

of its low cost abilitv

insulating

rsnge

from

the

and

when

choice.

which

flakes

to

semiconductor

and ease of handling

and

of other

TheV covers

materials

is unquestionably

It is a popular

resistance

a metal-to-metal
instead

material

Kapton),

to

in Figure

the interface

of the

oxide

resis-

manufacturers.

one mavconcludethat

high temperatures,

A number

as a
non-

Similar

that

when

applied

are shown

as the dust is highly

polyimide

which

torques

several

it is en expensive

of

With

thermal

interface

arrangement,

beryllium

applications

to mica

Resistance

an insulator

some

are used

such as the MOtOrOla

Thermal

the

and

from

exceeds

insulators

es, c:ntain
the electrical
isolation
material
$:~webv saving the equipment
manufacturer
the
.pf addressing
the isolation
problem.

When

is a filled
(variation

is not

serves

with

is handled,
However,

showing

surfaces.

available

be cut orabraided,

it is best

to several

a,.\.rn%U[

Insulator

best.

requirements,

a chassis

is common

that

surface,

of interface

packages,

greased

are

uneven

reduction

insulators

resistance

power

structure

isolation

of EMI

and

1! 2

2 to 1 and 3 to 1 are typical

(junction
to case).
Referring
to Figure3,

col-

is quite

between

bare
shown

markedly

and TO-220

It is obvious

is a common

be dissipated,

h~k~%k~sem

3, for
those

2 mil
Mica

by Thermallov,

different

(TO-3)

thermal

case; the probwhich

for

2 mil
Mica

16

materials

obtained

TO-204

power

(anoda,

t$the

p~~;#:@ust

always

to~~~~~~$:~$he

t~ii~~~~~acture

th$&%#~, resistance,

entirg

semiconductor

vertical

this termin#~&L@~~round
high

~a

Data

~~t~k~,tance

\/,$>p*r;;:
i..
~,+i ~,,...,>>
. !~?.

outpu$~~le-e

electrically

when

isolate

They

problems,

semiconductorq,&~&:/re

with

or drain)

term

Isolation

N~PStanCe Of between
,,;. grease is used.

CONSIDERATl?&<~&
,k\h.,;;+:>,+,:,.

power

semiconductors

manufacturers.

long

,:t@y

it to provide

2 mil
Mica

,,&y,ab+y
/,, ..Q.
,....,
~,s~,.
..:
4. 3.s~
.*Y: : :-,.33
,. ?,$. ~~\

\
1.2
1.0
,~ :~*:,4 f
8
TO-220AB
Thermowatt
., , +. . *
,..
NOTES, 1. See FIgres 3 ad 4 for add,ti.al data an TO-3 and TO-220 packages.
.%
(/%
:!..~;i>
2. Screw nol ,nsl.led See F,gure 12
+<
.,..{.lt$:}
~i$
formed to the surface roughness
of both the heatsink and
have a hard,
sq:~-q.bs~&ica,
.
semiconductor.

t~~ 3miI
Mica
,. ~:y..? ..

(TO.3)
TO-213AA
(To-66)

$~,l::j,r

0.:$,::

.-

BV

::,

16
[1, [1) Thermal film,

14

002 I 051 thick

121 [2) Mica, .003 ( 08) Ih,ck

.002 (.05) th,ck
[3) [3) W..,
[4, [4) Hard anodized. .020 ( 51) Ihck
(5I [51 Aluminum
oxide, .062 (1.571 th,ck.

12
1

[61 Berylti.m
[7) Bare joint

08

oxide,

no

062 (1 57 I
f,n,sh

[6, [8) Grafoil, .005 (.13) lh, ck .

171 -Grafo81 8s no, an <nsla,l.g
ma,

06

?h,ck.

er,,l

04
02
(81
00123455
MOUNTINGSCREWTOROUE (IN lBS)
,
0

72

145

217

290

t
435

362

INTERFACEPRESSUREIP,,I

3a, TO-204AA
Without

(TO-3)

Thermal

Grease

3d. TO.220
With

Thermal

Grease

.
:igure

3. Intetiace

Th~,rn#

Resistance

*aJ:@YeFundion
-,:

!.?,,> , J,.
Darinq.$iqur%3c

., -

com
masil,
the
for

&~X~220

~~tiese

,,,~il. +,:@entical

!i:}

extremes
clear that
than

insulators

under

conditions.

The

others

of out-of-flat
newer

noted

grease,

resistance

offer

as

that

Ther-

has

about

greased

silicone

Packages

(Data

Table

mica

rubber

performance

carefullv
interface

K-10

pad,

about

2/3

the

interface

which

would

place

pad.

surfaces.

products

Bergquist

1671

be

and TO-220
Torque

Courtesy

2. Thermal

using

Different

Insulating

Materials

Thermalloyl

Resistance

Manufacturer

inau-

of a number

controlled,

of Silicone

Rubber

R*CS (a
3 Mils*

Prod uct

thermal

have

for example,

Since

resistance

been

the tests

is described

resistance

of the

its performance

close

AAVID

also

offers

were

introduced.
Sit

Delta

Bergqu[st

Sil Pad

The

Pad

Bergquist

Sil Pad

Thermally

Thermalsll

Shin-Etsu

TC-30AG

Bergquist

Sil Pad

Chomer(cs

1674

Wake

Delta

field

as having
Pad

Ablastik

Thermal

Thermallov

Thermals!l

Chom-

-tesl
267

ktxl.

400.9
II
400.7
174.9

1000
Wafers
Ill

1671

Chomerics

pad called

K-4

Pad

Sil Pad

1000

173-7

1867

Rubbar

Bergquist

to the

an isolated

Wake field
Stockwell

nearlv

are over 2:1 for the various

materials.
It is also
some of the insulators
are much more tolerant

performed,

erics

it can
without

of manufacturers
2 shows measured

,{k,%,rWTable

,$$~
,*i,,
,.,.
~:~ .,,,

3d,

TO-3
Screw

Pads
R@CS CI

1.5 Mils*

package.

~~ ~;~~ber

$,

and

a ft$l#,~il
icone
rubber,
$. J.,,.,
s?m$
!til@rface
thermal

for TO-204,
of Mounting

re Uev,

al,

on

am flat from Thermally

.790
.752
.742
.735
.660
,664
633
592
5J4
529
500
.440
367
EIR86-101O

1 175
1 4J0
1015
1,205
1 045
1,260
1,060
1 190
.755
.935
,990
1,035
,655

Rubber-Due,

however

heats;nk

therefore

ison.

and

Published

was

data

0.3CW

for

flatness

and other
thermal

AAVID

above

details

be made

to sutiace

fairly

rigid

package.

insulated

with

that

worst

yields

results

of the

typical

insulator,
mance
under

the

having

In order

Data

Thermasil
case

mica,

2 roils, a situation

that

used

used

factor

(7.5

the thermal

ously
ularly

4.

roils)

be

with

so having
roughness

resistance.

finishing.

12

care

to avoid

be

on

&lth~gh
**. . ,.J~\,
req&kre4~~anual

t~m?~s~on
2,

,:,,

)5

~m,

,.,:,

mm

,,,

$ ently

0002

0.004

0006

0008

well

isolated

001

consist

(Energy
8, Case

*.
,>,,,.e$
insulators
Besides

chips

and initial

contact

in a study

of the

impedance
of 1000
first

M~sP:@$%e

hou{Y$:~&r6
on

~?~+~~?flventional

decrease@+~~@$&
confor,~~1

Th-lb

katerials,

from

during

0.74C,W,

mounting

hardware

a reduction

manufacturers
data obtained

performance
widely
which
ASTM
face

is the

in thickness,
need

Appendix

to be controlled,

Committee

except

6 in-lb,

in torque

for

mica

B discusses
At the

D9 is developing

time

have

hex and the

wire

and

soldering

to

a number

available

from

covering

and

integrated

single

packages
has

chip

var-

of parts

the

metal

mounting

into two
the

semi-

an exposed

area

shown on Figure
(TO-258M)
(see

in this

parts which

as

the part. The EMS

Series) Modules,
and Case 388A

contains

circuits

between
baae;

pres-

devices.

can be grouped

insulation

base is used to secure

examples

offerings

category.

The

have a plastic

over-

base,

The

Full

Pak,

3. Petiormance

leasurd
Material

had

With

non-

would

have

which

a standard

Rubber

Thermal

hermalloy

Bare Joint, graased

8e0, greased
Cho-Therm, 1617
Q Pad (non-hnsulated)
Sil-Pad, K-10
Thermasil Ill
Mfca, greased
Sil-Pad 1000
Cho-therm 1674
Thermasil II
S1l-Pad 400
Sil-Pad K-4

varies

the variables
of this

of Silicon

Insulators

per MIL-I-49456
Resistance

(CW

the
The

is not in good agreement.

Table
from two sources.
The relatie

same,

usually
is costly

These

and the mounting

Tested

%d
thermal
resistance.
in c :,?>?,
,, ~.,,,~e interface
.~@:$e
of the difficulties
in controlling
all variables
,,a#@{!ng
tests of interface
thermal
resistance,
data from
~~~~~ent
~$shows

first

at the end

occurred

initial

category

interface

measured

an

11) are

Table

For example,

to 0,7WCM

ROCS

flatness

pad thermal

change

for
The

of unusual

by surface

tt@e is a factor.

droppe#$~~O.9WCN

hours.

200

torque

press@e,

Ch~Th&~@?688

Figure

chips

conventional

Management
806 (lCePAK)

second
mold

~~$M~#number

be~~$$,~~ted

longed
1950s.

which

lead

became

the

the

In the Iateeighties,

parts

of

have

the stud

manufacturers.

insulated
The

of the mounting

rubber

margin

semiconductor

assembly

and

must
testing

and installing

types

the

of multiple
more

The newer
conductor

Silicon

large

since

between

mounting

with-

of these
hi-pot

insulated

stud mount

topofthecase.

as the

categories.

characteristics.

du~$fcause

acturers

effective,

$~:~wtrically

o$@r%{-

<,

and

distan,:~4~,Sf&$ifi-

usual l$$.~i~$sthe
.,%,,?
bu&,Sxcess

packages

Id~s semiconductor

Dust

handling

an

oxide

can.

st.,\
matter
.,.,.. )\i.
or seri - , ;::..,,,,.

systeq

collecting

equlpmd~%manuf

a Iayerq$berylt?um

,,,,s

grease

n::,,@@&$Qr

first to appe~$+~ere

006

occur.
creepage

of the nu~%~~~g,of

mounting,

009

it useless

may

coat-effectivq$n$%~d

,,,

rendering

prototy~..~n~a
,!$ ~)<::\,,,:,$S
.. . ~?.\ .
i$~~,....:jgy,.
Pq~ka&b~

Electrode

accessories

to keep the

of foreign

are not amenable.!~a~~lv~ls,

done

Because

be taken

panicles

insulation

employed,

Insulated

,,

Cost may

a clean assembly
area i~iii@p$ti%nt.
and humidity
also lQw,R~.N#&lation

of the

which

safety

al,,,

grease.

strength.
In addition,
par~:,~.$%,$~
than 300 V are enc0unt~@@f&3.:

shorten

Use of thermal

voltage

should

Therm

is that

dry, will out

a selection,

must

Small

creepage
can

be removed

,.erm,ls,l

mica with

the Insulation,

material

cantly;
Sutiace

the performust

clean.

problems
eign

conductive

flatness

insulators,

surfaces

lowering
its dielectric
when voltages
higher

factors,

,,

all this data

mounted

in making

using

can puncture

resistance

more

spot

mating

stand

!2

from

pads,

Resistance

When

(TO-3)

on Figure

requires

rubber

the commonly

Insulation
insu-

under

a TO-204AA

surface

perform

to be drawn

of silicon

be a determining

note.
rubber

when

the

conclusions
types

so a compar-

in this

Ill to exceed

total

below

is shown

for

The
some

surface

encountered

twice

(3 roils),

for Thermasil

of greased

ROCS

silicone

for

to a

compar-

However,

data

flatness

about

case

in the

shows
psi.

of some

is sensitive

package

500

other

lators

Observe

vulcanized

are not specified

with

resistance

base

available

not included

from

pressures

ison cannot
The

it is only

writing

for inter-

1) From Thermal)oy
EIR S7-1030
2) From Berq,st Data Sheei

measurements,

268

0.033
0.082
0.233

0.263
0.267
0.329
0.400
0.433
0.500
0.533
0.583

Data(1)

lerquist

Data(2

0.008

0009
0.200

0.400
0.300

0.440
0.440

Case 221C,

illustrated

in the second
Parts

in the

metal

first

flange

13, is an example

of parta

Clips
Fast assembly

category

or tab

non-insulated
ing system

in Figure

category,

those

are

counterparts.
where

pressure

molded

type

should

washer,

described

an exposed

mounted

the

However,

aa with

is bearing

be used with
later

with

in this

same

aa their

watts

mounted

any mount-

on plastic,

a conical

a few

AND

the over-

coat

compression

to

Characteristics
the

tools

of fasteners,

to secure

use in mounting
lems

have

CHARACTERISTICS

because

of several

associated

determine

the various

arisen

characteristics

them

packages.

many

choices,

of hardware

asthe

for

Normal
for

split

ring

Iockwashers

power

are discussed

washer

flattensat

about

pounds

is needed

for good

A very

useful

called

a Belleville

shown

in Figure

constant

piece

The

50 pounds,

whereas

heat transfer

of hardware
washer,

is the conical,

range

200/. to BO/0, When

torque

150to

of its physical

deflec-

installing,

until the washer

of

with

must

be spe~lly

andt.a~i
\.t\ ).. .: ~j.
p~~~,~~4*~ch

plastic

t:y~a%:~:yembly,

therm{

il.$~l~t>nce

is~ftek~p

with:a

In order

when it is ,@q:~n~dfor
top oft~@.W.~@
over

usually

the

fastener

high

than

proper
the die.

under

~~~~

because:.
he
3%IJ?.
~ii. . . .
~~t,i~.
.*
~,,.. ;:.. .:.

.1,:.
.:l!f ,.+
screw&$:c~,&Tw
ashers,
atro~~%free

package::.%~

die

through

prassure

and

fastener

ich~ave

at one

readings

(or sync-

for all types

holes.

.Torque

However,

ratings

must be exercised
it from getting

incoR&ri/torque

nuts

system

mounting

torqW&$$&cesaary

dry; the<~~8~~&re
mal g~aa~..~~prevent

to half its

in a low

handled,

Screws

proper

the assem-

depresses

an inte-

thickness

In addition

mounted

nuts) can form

As

a fairly

in the tab

Machine

sometimes
washer.

the

with
result

heatsinks.

clip

interface

package

when

Machine

300

to maintain

TO-220

location
hole

the

popular

lower

Iarger

only

board-

is being

heatsink

and T0-126.

end.

#6

at the interface.

compression

over a wide

tion generally

best choice
Atypical

5, it has the ability

pressure

bler applies

arenotthe

semiconductors,

pressure,

other assembly
methods
pressure
to bear on the

the basic

Hardware

mounting

TO-220

power

Whan

small

dissipators

with

a particular

the clip provides

prob-

next.
Compression

beuaed

proper
for

clips.
the

manufacturers,

higher

conductor
package.
clips are especially

and

suitability

Since

of improper
types

hardware,

their

When

clip may

provide

heat

by several

assembly.

with

dissipated,

or free-standing

aeparate

note.

HARDWARE

being

gral clip, offered

designed
FASTENER

is accomplished

are

result,

apply

when

when using theron the threads

as

Machine

screw

heads

Wbuldwot
directly contact the surface of plastic packages
flat tO PrOi~~s
~~ the screw heads are not Sufficiently
the assembly
Iineto
determine
the proper torque for the . .. ... .
~d~properly
distributed
force. Without
a washer,
crackfastener
used toachieve
50"Adeflection.
)Thewasherwill
:<
J#g of the plastic case may occur.
absorb
any cyclic exoans ion of the Dackaqe,
insulatin~~~.<.e,
original

height.

washer

or

(Tests

other

Conical

mounting

of devices
or when

ingscheme.
the

includes
the

They

packages.

the

are

requiring

plastic

key

variation

to setting

tem Oerat *-7

Nga$.\
successf~}

Self-TaDDina

oft~e

m@Sunt-

screws

are

mount-

process

with

isused{~l~+,

larg:,f~co~tacting
of

semiconductor

be solder*$~@t~!:PC

mount:

screw.(4)

..

bd~.%+,
,,~:a,,
~%, ~.,,
~~;,t+.

6-32

Und~r

will

tc$~~falwasher

of a nut assembl~:~$l$~a
can

up

to

strict control

are used wlththe

device

bv

the

hardware

A new

prior

caused

washers

it as part

patented

be run

materials

changes,
ing force

should

Screws

ca-refully

surface

in the
that

metal

Nut,

may

result.

board

and

they

must

nut.

If a self tapping

machine

must

screw,

could

Sync

When

conditions,

However,

a standard

develop

able

controlled

acceptable.

being

the

be used in a clearance

be used

which

an unaccept-

thermal

metal

tappingprotrusion

threaded;

sheet

process

the

a volcano-like

increase

standard

sheet-metal

during

resistance

screws

are used,

hole to engage
is desired,

roll-forms

the

machine

a speed-

screw

screw

type

threads,

Rivets
Rivets

are not a recommended

plastic

packagea.

package

or EMS

heatsink,

rivets

used.

crimping

force

ferred

steel

over

tivet

mounted

be

cause

because

a gradually

regulated

increasing

Useofsharp

blows

used

is

the

high

of most

less pressure

are

semi-

much

is required

preto set

is improved,

is preferred

pressure

to a

and evenly.

because

rivets

pressure

directly

slowly

deformation

conductivity

of the

flange-mount

press-riveting

be applied

rivet, or eyelet,

An adjustable,

provided

Aluminum

and thermal

The hollow

is being
used

never

could

packages.

for any

metal

must

should
force

conductor
the

over solid rivets,

press

is used such that

is used to pan the eyelet.

could

damage

of the

surface

the semiconductor

die.

Solder

,..

b.:,

be

fastener

a rugged

module
can

Crimping

Pop-riveting

When

Hgure

5. Characteristics

Waahers

Designed

for

of the

Conical

Uae with

Plastic

Until

Compression
Body

Semiconductors

Mounted

the

advent

nique,

solder

was

not considered

power

semiconductors.

to the development
[4) IW

Shakeproof,

St. Charles

Road, Elg,,

IL 60120,

Acceptable

269

soldering

mount

However,

of new

user

packages

methods

assembly

a suitable

include

tech-

faatener

demand

has

for
led

for this application.

conventional

belt-

furnace,

irons,

vapor-phase

It is important
that
exceed the specified
bond

to the case could

has excessive
failure

and

be damaged.

thermal

under

reflow,

infrared

the semiconductor
maximum
(usually

power

resistance

reflow.

To

temperature
not
260C) or the die

A degraded
which

die bond

often

leads

prevent

Another

Adhesives
are available
which
have coe~cients
of
expanaion
compatible
with
copper
and aluminum.(5)
types

are available;

approximately
0.3CW
ferent types are offered:
servicable

systems

serviceable
when

assembly
the

the

solder

low

parts

used

greases

coatings

types

bonding

are

thermal

conformal

strength

Adhesive

mounted
because

with

Plastic

or

used

for

are

part

is rigidly
system

package
imized

characteristic.
fastening
uable

When

system,

to assure

time

and

loss

of contact

is mounted
chip

will flow,

plastic

materials

compression
that

the

temperature

will

widely

in this

parts

of the

form

washers

assembly

cycling.

pressure

but differ

are highly

will

not

loosen

As previously
increase

the

some

minimize
movement.

valwith

Stud Mount
Parts which

thermal

shOw

fastening

TECHNIQUES

classes

techniques.

type are discussed

in following
considerations
follow.

of packages
Details

in use requires

pertaining

sections.

Some

copper

and

by two

metal

or more

screws

parallel

increased

stiff nessik,

the

temperatb~
c .:$
\.: .:*,i
.,, \.
,i.,:$.\!t ,:,,.

tiounting

caab,e Q+#&, >.

- .\$.., ,,.,,
~$<.,$

to kg.)$@t

oversized

grease

Sta~rd

or mount-

stud-mount
errors

classification
with

confined

to application

..

Heat
D~&~hwfi
Through Thermally
Conducive
Adhesives,
EMTAS
>$~:&nfere.ce,
February 1-3, Phoenix, AZ; Soc,ety of
~::~;.r,.9
E.gneers,
One SME Drive, PO aox 930, Dearborn,

general

6b. Isolated

Types

6c. RF Stripline

figure

CASE 244-04
(.28o STUD]

Opposed

6. A Variety

Emitter

(SOE)

of Stud-Mount

270

..
---- .. .

are

>~.:>

(5) RQb&.tBa,$on, Ell,ot Fra. nglass and James P. Moran,

to each

CASE 145A-10
(.5oo STUD)

CASE 145A-09
(.3B0 STUDI

for

non-insulated

TvDe

+:t
*
~!~~teASE 144B-05
.>,
(.380 STUDI

to

allow

, ~,~

Non-Isolated

so

as tem-

tind semiconductor
and

Series

Parts

CASE 305-01
1,204 STUD)

441
,..,+ ,::~*
.$::$\\<.,,.

.~%?~~
!:!. *:*
*#a.
.t\
the,~~,~.~~- ~~

CASE 311-02

6*

th:ff,v

causeq

t~$~~$~~~~%tsink

gradient

pa-~s are generally

of

a:d%,~

surfa+~%k}:occur
\a ~.!,..>s>l
; > >./,

the h~~l$~

f~$i ++~~~j~e

but

also

Bending

component

tern-

coefficient

ia excessive

be damaged.

holes

with

alumlnum,

each

If bending

slip between

in Fig@~y8,

,.:,~$.,:>+

Each of the various

the

a copper
heataink,

bend

is the thermal

through

heatsink

ing pad between

stud-mou~j~d

different

not to mar

will

perature
changes.
Using a highly conductiv~t~~$<~l

3,

resistance.
FASTENING

could
the

2. Allowing

discussed,

interface

when

is that when

by:

that

materials

be taken

which

only

for

fins to provide

process,

plastic

Not

to bend.

1. Mounting

foul

Hardware

Moat

occurring

to an aluminum

results

gradient

conductor

soldering

mounted

different

each component

field-

greases

changes.

temperature

not compatible

and the

must

to be considered

a bimetallic
expansion

has

is attractive

in wave

precautions

factor

perature

interface
thermal
resistance,
Difhigh strength types for non-field-

systems.

case

a 10 mil layer

from

finish.
based

conductive

action

used on aluminum
heatsinks
in a corrosive
many
devices
are nickel- or gold-plated.

Consequently,

to a

cycling,

Adhesives

Highly

galvanic

devices are
atmosphere,

CASE 332-D4
1.380 STUD)

of excessive
heatsink
of the

torque

hole,

hex base which

The only

electrode

less

non-insulated

cised

to avoid

insulation

to the circuit
RF.

transistors
(SOE)

straints

because

age. Special
Iine leads

in the

in the

must

oxide

while

the

entitled

to the

ceramic.

be attached
the stud

hex

the nut is torqued.

stripline

opposed

additional
of the

connections

ceramic

as

be exer-

stress

construction

part

conpack-

to the strip-

so no tension
metal

or shear

intetiace

are

and

Han-

Connecting

Terminals.

)b

The hole edge must be chamfered as show to preven! shear,ng

off the knurled edge of the case durag press-,.
The pressing
force should beapplled
evenly othe Shoulder ring roavo!dt,ltag
or cant(ng of The case ( the hole dur, ng the pres$lng operation
Also, lhe se of a Thermal joint compound will be of considerable
atd. The press(ng force w(fl .aryfrom
250101000
pounds, depending pen the heatsink mater!al
Recommended
harnesses
are
copper-less than 50 on the Rockwell F scale: aluminum-less
than
65 on the Br#nell scale. A heatslnk as th!n as 18 may be used.
bt The $ntedace Ihermal
resistance WIII ncrease
I. d,rect pro
Portion 10 the contact area. A lhtn Chass,s req(res the add,t,o
of a backup plate,

Rgure

Flanae
gory

of parts

as shown

in Figure

exist

the

culties

with

such as the TO-204

between

screws

difficult

to cause

which

good
that

its final

is shown

For

secure

most

mounted

8. A special
must

applications,

according
fixture

to the
meeting

the

press-fit

instructions
the

case
shown

necessary

should

tion.

be

The

in Figure

screws,

the package

other

to alternate

should

or

rivets

guidelines

Fastener

and

Hardware

copper

flange

of the

may

be used.

271

is

of the

After

the

be torqued
steps.
tvpe

(preferred)
be

used

In the previous

to
sec-

Characteristics.
Energy

Management

(EMS) Modules
is very thick. ConsequentIv,
rugged
and indestructible
for all practical

requirements

side

flange

screws

it

IS started

sequential

for a popular

10. Machine

eyelets,
using

two

one

tightening

hardware

in at least

make
mounted

screw
applied.

installation

in Figure

to

or unless

the

diffi-

packages,

unless

is evenlv

the

specification

self-tapping
Fit

practice

pressure

mounting

bowed

before

cate-

base and distance

combine

any warpage

is badly

mount

mounting
mount

holes

are finger-tight

A typical

flange

The rugged

It is therefore
so

smaller

known

(TO-3).

excessively

screws

fit into the flange

9. Few

mounting

on a surface
tightened

part

Package

and

die

extremely

to

8. Press-Fit

Mount

A ~arge variety

Press

CHAMFER

base

same

(BeO)

some

to make

to any
section

but care

stud-mount

of the unique

are applied

mount
the

In addition,

impose

and to mount

discussed
dling

wire.

package

techniques

a stud

mounted

that the leads must

flexible

7.

or tension

be used to hold the pati

emitter

forces

a shear

die.

is to use a nut

in Figure

a berrylium

dictates

with

a warpage

on

are

counterparts,

This requirement
should

They

usually

a threaded

method

packages

applying

layer,

stud

are shown

hardware.

into

may cause

crack the semiconductor

fastening

the details

Insulatad
require

the

practices

may

recommended

and washer;

their

or tapping

Both these

Series

the parts
purposes.

are
No

~~

E*>S

CASE 1, 3, 11
To-204AA
(TO-31

@SE 211-09

CASE 357S-01
CASE 383.01

9a. TO-3 Variations

9b. Plastic Power

Tap

CASE 319.04

CASE 31a.01

-~1~
CASE S07-01

CASE 373-01

CASE 336.03

CASE 333.03
**-

CASE 809.01

CASE 813.01

CASE 812.01

,@,

*m:

>~.t,,~ ~.
CASE 814.01

CASE 819if~~$L.
.}.1.
\ s,
<,:$\:,,$::,
SerieS*:,<e~?.?,%

CASE S16-01

9c. Energy

Management

(Isolated

Baae

CASE 7~-02
9d. RF Stripline

CASE 744A.01

Isolated Output

Plate]

. k&~~,,~,t
,, &:iJ, , ~
, .,:$..!.*,
. . .>.,.
Figure 9. A L~~g&$%dy of Parts fit into the Flange-Mount

(SOE)

Opposed

Emitter

Series

Classification

,*.
<b

special
parts

precautions
to a heatsink.

Some

packages

example,
torque

should

The
sitive

with

@~uJ@s

th~~,t~~~$

tRat the

long,

ceramic
will

heatsink
flanga,

than

The

mora

other

in EB107

a combined

heatsink

Over

haatsink

have

mounting
in

isolated

substrate.

ba used

nearly

pefiect

returned

haa

deep

packages
Tha

to illustrate

Iarga

because
area

problem

are critical
of

Case 368-1
areas.

tha

to RF

internal

(HOG

It is more

in the

sen-

other

RF

instead
BeO

amount
torque,

tha mounting

the initial

areas

have

the

length

roils

the thermal
hole araas,

soma

been
of

with

compound

expand-

mid-section,

result

is applied.

can

under

also

if an excasWith

will squeeze

but will remain

flange
screw

from

of the device

compound

observed.
tha

a tamperatura

intha

A similar

parts
10 roils

moting

if the

upwards

mounting

screws

up to

is preventad

dia.

recom-

recommended.
In

two

the

of thermal

ara

problem,

cases,

the

mounting

that

it bends

and

on recindicates

example,

indentions

the flange

ing in length,

information

washers

straw
of

In such

cracking

during

flat

indicate

of 75C.

the

For

analysis,

excess

is axcessive,

ficient

moat

experience

on the 4-40

is a common

change

sive

PAC)

and

mounting
in

contain

ignored.

failure

torque

occur

than

procedures,

torque

Spring

for

increases

and flange

techniques

sheets

often

Calculations

Many

flatness.

recommendations

are

maximum

torquing

allowabla
baen

data

mounting

is 5 in~lbs,

,, Mounting

Modulas,

the

they

mended

to a rel-

damaga

RF Power

that

sen-

packages

maximum

in datail

Motorola

are

is attached

mechanical

presented
can handle

series)

from flat of 7 to 8 roils which

is commonly
Others must be held to 1.5 roils, which raquires

Specific
devices

(MHW
substrate

to avoid
for

:@rts

procedure,

mounting

davices.

Although

dete[,~yne$#hd
Of&

to proper

power

ommended

,,~rib~g

~~M{#?iOn
%Vallable.

sitive

For

a c.~~@@$BeO)

CQ@:@*ions

these

PoweR$Ja$
package,
Figure 9b, final
\
be a~~l~cd h~st to the center position,

to the fla&~,@,~ofthe

atively
flange

fastening

the

RF power

because

are nece~s~tien
,*:,
,:,
, :iJl,, ,;:.~
spec~~t~~~~htening

suf-

ot of

the center

272

...

--

..--..

the washer
ing

NO, 6 SHEfl METAL SCREWS

hole

holes

IS only

are

bushing
case;

Important

exceeds

0.140

needed
when

the s!ze ot the mount-

[632

to accommodate

the

however,

when

inch

screw

should

to provide

hardware

exceed

a diameter

of 0.250

plastic

body

during

device

connections.

TO-220

packages

packages
.1

combination
screw head

=P

of other

The

lift

applied

to

counter

this

turer

one

of the flange,

deforming

it. De,f$~i$~bs

been

meaaured

between

such

conditions

(enough~$&:~ra~

Another

problem

flange

changes

ature

excursion

around

0.25

p, ~:

..,.,1

arl~s

b>$~use

&ntl@ned,

the ends

internal

witQj$@~$ature.

n~k&$@~

of 23 roils have

the ~n~~and
the

the

occurs.

thickness
7VC

increased
in further

Therefore

CASE 221A.02
(TO.220Aa)

of similar
as

B, Figure
hardware
beam

Bl,)

applies

221a-ol
(TO-220AC)

of the

amount

thermal

cycling

is
of
CASE 314a
(5 PIN TO-220)

CASE 3140

CASE 339

not

only ~~b$?lsk
of structural
damage
but ofien causes
,. ,t$, .$
thg ~s~bly
to loosen which raises the interface
resis,+,a~~~~~se
of compression
hardware
can eliminate
this
,,
,:,:

p~b~$m.
, *!

) :, ~..$$b
Mount
,,,?
$:<,;:,
,:l!.,
<,,.$
The tab mount
class is composed
of a wide array of
..,*.
,,11.:..$,1,
,.
packages
as illustrated
in Figure
11. Mounting
consid.. ,..,){,
,, ..
erations for all varieties
are similar to that for the popular
y~
TO-220
package,
whose
suggested
mounting
arrangements
gular

and hardware
washer

distortion
could

shown

of the

cause

are shown
in Figure

mounting

damage

in Figure
12a

flange;

12. The

is used
excessive

to the semiconductor

CASE 340.01
lTo.21a)

rectan-

CASE 387-01
(TO-254AA)
CASE 388A.01
(TO-25aAA)

CASE a06.02
IIcePAK)

to minimize
distortion
chip.

Uae

figure

of

273

11. Several

Types

of Tab-Mount

is
To

It separates

temper-

tightening

con-

manufac-

which

In addition,

or
the

pressure

under

the mounting
s*f&,@, thus increasing
the effective torque
from the j@~~al $~~ue, With a decrease
in temperature,
the opp~~!$~;,:,~ct

one

cantilever

on the tab.(6)

TO-220

a spacer

ceramic).

For the

results

surface

at least

plastic

Motorola

bushing
to raise
of the plastic.

and others

con-

internal
end.

need

Appendix

t$nde~y,

offers,#~~@

.,.,..

~y$ne

*,whting

w~~~e

moreev~m~?~asure
*> ..,.,
\ \.

~);a ~.%.,

a cha@r

manufact~{e%wy

t%

used
with

ope~~$i~~: %uch

To minimize:~h~$~~r%blem,

TO-22,~F~sk~~
off

is also

to assure that th~t~~

never come~~~$?~ct

spacer and ~.[%ion

above the ~p~,i~$~~ce

popular

struction

distortion

to the pla~~,podpand

have

necnever

during mounti~~.
A
is suggeste~,~~o
,,.L. ,y<:~+,

the driving

tact can result in damage

to the

than

and should

Flange

screw.

Care should be exercised

to drive the mounting
screw
the

connected

clearance
inch.

Larger
insulating

not be larger

possible
if excessive
torque
is used
maximum
torque
of 8 inch-pounds
a 632

lower

is electrically

the holes

essary

using

clearance).
the

Parts

Plastic
a) Preferred Arrangement

b) Alternate Ar,angemet

for Isolated or Non. isolatd

for Isolated Mounting

Mounting.

Screw mua

Swew is at

Semicondtior
Potential. S-=

Case
Hardware is

Ux

site side

44

PAN OR HEX HEAD SCREW

,,) RECTA NGUMR

oping

the chip

to feature

minimum

resistance,

For the Thermopad

size with

mounting
Full

tab

force

is encased

is applied
from

types

packages;
With

in plast~,@.~8,&@ke

to plastic,

the

th&~@nting

pro-

a standard

TO-2~
ari~~s similar to
,~,1,,:..
+ ..&.>)?\
may%:$$eti
to secure these

of fasteners

machine

screws

open. T.@@ , $b
.s, ,?)
is obtain~d,st
,$

that strict attention,@~~%

\\ .\*;\.>. >
~.;: ..
221 C-01)
is similar
ggi~ ~:220

of the Thermopad,

Several

ST:EL

copper

hole

of this construction

(Case

the

differs

st.,\
.,.,.. )\i.
envel - , ;::..,,,,.

The

IS molded

procedure.

Pak

that

surface.

plastlc

the mounting

of a requirement

that

SEMICONDUCTOR

but leaving

the expense

cedure

hTN8sHN

as a mounting

for mounting;

resistance

mounting

WASHER

packages

in this group.

designed

low thermal

The
VA

power

of packages

in thermal

ia exposed

haa a hole

to the

been

sheet

except
B

Pak plastic

(Case 77) parts this is accomplished

by die-bonding
the
silicon chip on one side of a thin copper sheet; the oppo-

Parts Listed Below.

+
632 HEX
HE kO SCREW

have

Full

13 are typical

no compromise

Choose from Parts ti$ted

nelow.

and

in Figure

They

be et Heatsink

Potential. 4-4o Hardware is

Mount

Thermopad

shown

when

Used.

U*.

Body

The

screws,

or eyelets,

e@~@,.6?clips

are preferred.

a ::h~}$b,.~asher

should

be used

which applies the proper f~~@*the

package over a fairly
wide range of deflec~~a.~ an~$distributes
the force over a
fairly

~w
21 REtTANGW*

large

surface

aZ~,@*,&rews

should

able

conicah&t~SheY

Figure+.~4

IS shown

shws

details

in Figure

t~~t~,::~er

mounting

,..J~~@Io~
~J~
,,$Qthreads

3 FM1 WASHER ~
~

t0MPRESS104

WASHER

The

procedure

chosen
sink,

clip,

is much

mal

resistance

occurs

inserted

holds

the

B1 of Appendix
course,

alao

Figure
case,

scr~$~@

15c, ~~$bt%h

mounted
high

a p[~,::~~f~%preader

or cli~,

than

(Caae
with

clips

with

806-02)
isolated

mount
cracking

be used

under

with

the cantilever

screw

mounting.

ia basically
chips. The

parts

as shown

without

bar should

resistance

ca@ be lower

tab. Tab
with

pressure

clip. ln~~ti~k$~thermal
,T%R i*PAK
.T*~ackage

in

package

procedure
should

In situations
direct

contact

provided

sharp

described

for

the

where

a tab

mount

the heatsik,

blows

or impact

shock

package

flat

screw
but
Both

mounting
a conical

ware

as COm Dared

with

a TO-220

afford
which

the ther.

method,

screw

the

should

die.

is shown

die

causes
(See

the

Figure

of a layer

thermal

pad,

15b may

washer

a major

This

of the

consist

in Figure

to the conventional

package

pres-

on top

the

compression

methods

heat.

clip

a screw,

conductive

shown

of

in the

with

for either

under

or a highly

holes
though

while

the

that

15c, one properly

directly

B,) The interface

permits
over

Of

also

should

be

reduction

in hard-

mountina

method

in Figure

15a

the

flange

CASE 77

mount

ITO-225AN
TO-126)
(THERMOPAOI

may

is making

;ASE 221c-01
(Full Pak)

CASE 221 D.01=CASE 340B-03

[Full Pak)
(Full Pakl

be used,

ia avoided,

Figure

..

bears

screw

a elongated
mounting
pre-

package

a eyelet

the clip

340B)

Even

obtained
the same

the

beam

be used.
with

than

that

simplified

slotted

needed.

electrical
a bushing

the

?#~#~lwns
for the TO-220
consequently
apply. In addition,
~~nce two mounting
screws
are required,
the alternate
t~ghtening

used

included.
the metal

be efi~$~efY

into two

is about

grease

insure
base.

in Figure

to lift up somewhat

of thermal

may

lower

because

package

be

As shown

When

mounting,

221 C, 221 D and

is all the hardware

sure

is applied.

to be greatly

TO-220.

Case 77 devices.

minimum
hardware,
chosen to insure that

screw

hole will
the metal

Full Pak, (Case

a standard

*~,..{.

with

the m0untin9
do not contact

mounting

and

force

is required

equipment
Of a SUit-

5,

of mounting

Clip mM&tJng
is fast and requires
hO~*+
t~ clip must be properly

the mounting

not be tightened

with any type ~h:~~a~iven

torque
gun or
which may C~g~~Ip@
impact.
Characteristics

l&sM-OR

13, Plastic

Body-Mount

Packages

i4

..-

,/
T

HEAT SINK
SURFACE

MACHINE SCREW OR
SHEET METAL SCREW

COMPRESSION WASHER

4.40 SCREti

THERMOPAD ?ACKAGE

L*

\
U

,N~~*T,,~

~~s~~

(OPTIONAL,
-\.

14a. Machine

MACHINE OR SPEED
NUT

Screw

Mounting

.!,.

14b.

Eyelet

Mounting

.,:3s

~+

15b.

~~~<k::l 4, Recoin mended

~*;*>?.
(TO-126)
,,$$.. ,$:. . TO-225AA
;?? ,,.~.+,

*J:i,
f$:..\
, .
>,*..,.-.,>,
..3!
.:..

Although
face

mounting
assemblv
shown

manv

mounted,

in Figure

mount

small

have

been

16, for example,

for

parts

have

using

been

packages
surface

developed.

The

will accommodate

surfor

a die

thermal

resis-

tance

to case.

thermal

resis-

iunction

The

Full Pak

DPAK,

has a typical

2C/W

15c. Clip-Mounted

mount

up to 112 roils x 112 roils, and

around

Full Pak

Packagea

footprint

semiconductors

techniques

Arrangements

Thermopad

of the tab
special

power

Mounting

Screw-Mounted

igura

15. Mounting
Compared

275

Arrangements
to a Conventional

for tha

Full Pak as

TO-220

tance

values

The

of the solder

printed

circuit

Standard
good

heatsinks

mal

resistance.

assymtotes
about

As Figure

for the preparation

on such

coating
with

indicate

designed

of 1C~,

type, [7) The substrate

or it can
other

to be compatible

techniques,
insure

no special
that

foil

Tests

run

surface
with

surface

are

age.

where

A plastic

may

and

support

seals

to
sup-

around

be supported

bv

high shock and vibration

and

where

beasshotias
reduce

however,

be firmlv

no heatsink

possible

thermal

it is better

resistance.

to support

for the TO-220

As a

Package

and ot~~% ,. ~k..

thermal

the CASE

a small heatsink;
however,
no definitive
dat~~$i~a~~>le.
have
When using a small heatsink,
it is good ~~~~~~tb

77 (TO-225AA)

the sink rigidly

are

possible

assembly
other

than

profiles

are

mounted

total

of part

but the

package

such that

support

arrangements

arrangement
package,

(TO-127)

for
are

the

@ t%e board

s~lc~~uctor.

sho~.~~~~

Two

Figure

(a) could,~@@~$~d

scheme

has+x$ed

the ~~k

with

CASE 369-03

figure

18.

any

of @r@ll~b)
?,$,:.l$:+>i
,i~
,$,,>..*.,l, .

is more

Mount

D-PAK

Plate, Vertically

Moted

practical

Parts

LEGS

100

18a. Simple

HEATSINK

F
.0

$In applications
the
may

order

where

of a watt

be mounted

average

or so,

with

little

most

power

dissipation

power

semiconductors

or no heatsinking.

The

is on
18b. Commercial

Sink, Horizontally

Mounted

leads
Hgure

(7) Herb Fick, <Thermal Management

of Sfiace MoI Power
Devices,,
PoweFcOnvers(o
and Inte$l( gent Motion, Angus? 19S7,

18. Mathods

of Using Small Heatsinks

Plastic Semiconductor

276

Packages

The

plastic

CASE 369A-04

1S, Sutiace

$.

the pack- >.kj:::j~$,,

not exceeded.

,/
&

is

to increase

depend-

packages

are needed

temperature/time

resistance

must

of cracked

packages

not encountered

practice

is providing

mount

possibility

plastic

are not designed

cases

similar tvpes is offered

bv heatsink accessory
vendg,~%~s~
In many situations,
because
its leads are fairl~s~%g~

to a conventional

mount

the

Manv

packagea

their

leads in applications

general

performance.

precautions

maximum

values

power

used. The Ieadsshould

rnav be an effective

be attached

for improved
and

copper

metal

packages;

to avoid

leads.

stresses

at

or copper

traces,

exact

itself,

to

their

for low thermal

conductor

board

designed

ported
the

of board
occurs

that case to substrate

is in the vicinity

DPAK

inches
returns

one or two ounce

ing upon

Since

make

resistance

have thick aluminum

of printed

a product

heatsink

not

thermal

at 10 square

heatsink
finned

do

the

vibration

that

is overlayed

resistance

of the various
support

the thin foil has a high ther-

of diminishing

A dielectric

resistance

boards

ICN,

heatsink.

inches.

are offered

substrates.

under

as the

17 shows,

2WCN

a point

3 square

Boards

2-ounce

because

to about

although

is well

also serves

Glaas-Epoxy

very

area,

intetiace

board

With

with

Case 77 Thermopad

tYPes, mounting
more practical.
In certain

situations,

is required

sockets

are desirable,

ets for

many

user

specific

details.
to obtain

packages

Sockets

The leads
is

stand

are

semiconductor

being
from

Kelvin

readings

When

packages

axial
may

pull,

wires

and

HANDLING

tabs

to avoid

must

undue

which
of strain

used

for

damage

imposes

maybe

caused

relief

by thermal

should

connections,

to the

axial stress
be devised,

care

should

for

are nec-

across

semi-

as a result
extremes

plastic,

be handled

failure.

dimensions

TERMINALS

mechanical

semiconductor

temperature

and

stress

Change

of thermal
must

and RF stripline

could

in mechanical

cycl!ng

be

cOected

which

over

operating

considered,

packages

Standard

each

have

some

considerations.

Metal

Packages

The

pins

glass

and

to metal

nificant
Wires

may

When

lugs

seals

bending

or solder,

of metal

packaged

are not designed

or stress.

be attached
provided

wires

braided

If abused,

using

are attached

sheet

crimp

ratings

directly

leads are recommended

using
any sig-

19a.r@~~nent

the seals could crack.

sockets,

the data

devices

to handle

connectors

are observed,

to the pins,
in order

flexible

to provide

or

strain

Parts

of a Stud

$~+$,,,~ckage.
Flange
,.~
are Similarly
.~t..u:.,. ~,
~..,
.~,+k.

,,

EMS

Modules

The

screw

terminals

rugged.

Since

of the

EMS

heatsink,
some

the

connection

flexibility.

terminals.

the flange

modules

to

A rigid

Lugs with

the

buss bar should

PRINTED
CIRCUIT
BOARD > #

to@rigid

terminal~;t~:st

allow

nh.~,,be-bolted

tO

are preferr@&{y
,,t\* ,.,, ,,,,
,!.., ,,,
~
, ~,?,k<,,(..
Plastic Packages
.,fj:~,~!..+,
The leads of the plastic packa~~~?~.somewhat
flexible
. ),s+
snd can be reshaped
althoug~~?~{snota
recommended
procedure.

which

In

makes

many

and tab-forming
large

cas+~;:,a

&atsink

lead-bend~$g<o~$cessa
op~~~$~$~iavailable

quantity

orde~.

risk of device

basic con~#erati&8s
lea~;~$~~~rt

should
must

bend},~&:~,n&~the

from

P~yformed

~a~&a-caused
#
~~~~
f~d{~ ending

If, however,

can

leads

remove

;ME,A,

HEAT
Sl&K

SIDE\tlEh
CROSS SECTION

chosen

19b. Typical

lead-

Motorola

PRINTED
CONDUCTOR
PA~ERN

TOP
~lEw

Stud

Type

Mounting

on

SOE Transistor

Method

the users

by bending,
IS done

by the

be observed.

be placed

package,

be

ry. Numerous

/
i

HEAT
SINK
SURFACE
Bg
D FMT
r

braid

,,FuT

;,

look decg~

base is mount~d

-.

,:

,~

tively

Stripline

Packages

Constructed

.:,?,:<>,,
.;3....
,.:$.!
.\$h.......*)

Mount

Mounted

,.,+,.,
,,:

relief.

the

greater

arrangement

method
are

to with-

in this direction

in permanent

a condition

some

are not designed

Force

result

If the mounting

on the Ieads,

Motorola.

connections

voltage

4 pounds

cycling,

sock-

catalogs

of plastic

excessive

device.

developed,

have provided

manufacturers

with

accurate

AND

leads,

special

package

heatsink

terminals.

properly

metal,

of the

available

to consult

CONNECTING

cause

the other

where

Manufacturers

of the

essary

Pins,

in particular
or prototypes

is urged

conductor

With
on top

than

testing

The

devices.

the transistor

between

For forming

user,

When
the

small

several
bending
point

of

quantities

of u~~s?~jpair
of pliers may be used to clamp the leads
~,:,,,.
~ase, while
bending
with the fingers
or another

~t~~~

~.k.d~~{d~pliers.
For production
quantities,
a suitable fixture
}.
~Otild
be made.
,,.?
.<~:,?$}v%The following
rules should be observed
to avoid dam-

$
.I.+~~JU
-.$.
t>++.,<.
,$,+>,7,>,
:s,

age to the package.

1. Aleadbend
radius
for TO-225AA
2.

No twisting

3.

No axial
respect

~ETAL

greater

(CASE
of leads

motion

than

77) and
should

of the lead

l/16

inch inadvisable

~EAT

SINKSURFACE

~,~~

,~~

CROSS SECTION

1/32 inch for TO-220.

be done

should

19c. Flange

at the case.

be allowed

Type

SOE

Transistor

Mounting

Method

Details

SOE Transistors

with

to the case.

Figure

277

19. Mounting

for

be

exercised

to assure

cause

movement

tions.

Highly

viding

strain

that

the
point

lead

of the wire

lead

lead-to-plastic

of the

at the

wires

leads

is restrained

of the

does

CLEANING

not

soldering

exceed

and

26VC

The

for

It is important

pro-

case

for

than

must

not more

BOARDS

1/8 inch from

of the devices.

the reliability
Freon
plastic

solvents
devices,

Hydrocarbons

Freon

or cleaning

affect

age.

not

any solvents
of decreasing

with

than

that

used in the process

inated

and

be soldered;

however,

be applied

greater

provided

plastic

may

temperature,

must

at a distance

the

leads

CIRCUIT

junc-

are good

is permissible,

between

wrapping.

the maximum
seconds

movement

or braided

relief.

Wire-wrapping
the

that
of the

flexible

may

or flux
Alcohol

are generally
since

they

such

as

chemicals

removal
and

satisfactory

for

do not damage
gasoline

cause the encapsulant

do not
unchlor-

and

to swell,

use

the packchlorinated

possibly

dam-

the plas-

tic case.

Stripline
The

energy
and time of application.
if ay packages
are free-standing

Packages
leads

of stripline

into a board

while

sink as shown

packages

normally

the case is recessed

in Figure

19. The

are soldered

to contact

following

This is particula~~~q~~k
without
SU~@O{~aNC

a heat-

rules

should

...

be observed:
1. The device

should

ner as to place
2. The device
direction

should

never

force

towards
device

board

with

the copper

passing

adequate

be mounted

achieved,

in such a man-

leads

is mounted
stud

through

clearance
shear

and

a hole

must
forces

BeO

portion

be provided
from

being

for the

Some

clearance

and the circuit

must
board

be allowed
when

between

the device

BeO

For a full exp}~~~tion

to the

Note,

sinks
6, The

should

before
leads

be properly

its leads

secured

are attached

on stud type

devices

Figure
board

19b

and

shows

heatsink

\:*area,

not be used$b

stud torque
a,$~$;,,
for this purp%
;$

a cross-section
assembly

for

mounting

H is less than

of

the minimum

there

is no possibility

BeO

ceramic

package
cap

to

Beo

are

of tensile

joint.

dimension,

rences

distance,,{r~%e

to the mounting

joint

at th~}~int.

failure when the str&&Jyre

ing the lead. sold~~~#~o@ratl
to stud failurek$$h~,f$~~e

copper

greater

The

Iarger,

a typical

flSN@$~pe

transistors.

stripline
from

the top

joints

surface

exceeds
those

the

package

discussed

AWG,

the following

where

TC

= case temperature

the power
methods

oscilloscope

pictures

for

and current

waveforms,

Graphical

at the varioua

problems

similar

(C)
(C)

resistance

junction-toon the data

(CW)
dissipated
the

in the device

equation

dissipation,
are

His less than the miniof transistor

lead to the

devices

Aver-

computed

Two

graphical

often

(W)

lies

commonly

in

used

integration

and

substitution.

Graphical

type

be

temperature

in applying

to

dimension,

for the stud

(See

be used,

+ R~JC x PD

sheet

as

if distance

can

= junction

PD = power

empirical

case.

case temperature,

TJ

circuit

However,

the
should

case as specified

determining

on

the junction

equation:

H is defined

are avoided.

on

temperature

= Tc

difficulty

of the

It is based

between

point

R@JC = thermal

failure

or

as appli-

a calculation

TJ

Again,

forces

ratings

thermocouple

to determine

printed

tensile

amplifiers

is sufficient.

resistance

junction

Lack

board

applications,

wire

technique

of the

of the flange,

in the package

as #36

The

Application

power

or equivalent

reference

A,) A fine

mounting

~~heatsink
surface.
If distance
mum distance
from the bottom
bottom

in device

RF

temperature

of thermal

is a cap joint

is high enough,

this may result

as po%.er is ~ppiled.
~~?~&?@<?9c shows
~$t~.~stance

limits,

in many

operating

from

~we~%-lhe
device and the heatsink
surface
..?, ,,.,,
occut$@$&Q#Miff
erences between
H and the package

dimen$kn~,cdme

di/dt

the

of contact
will

notably

!Y heated, as might occur duron; while the second is BeO

generated

not average

of its use, see Motorola

a temperature

such
age

is applied
to the
joint. TWO Occurfirst

temperature,

be observed.

junction

Appendix

stud

than

must

concept

and

package,

f~~ceq:!rithe

If, ho~.e%~,t:~?s

conside~bl@force
and th,~$&@?& stud

possible

bottom

~~%gt$fthe

ithe semicycle, the

duty

highly reactive loads, may create severe

conditions
which render the traditional

Fortunately,
the

&ka~Q,?Vpe
of
If

the junction

of thermal
resistance
or transient
thermal
Invalid. In this case, transistor
safe operating

thyristor

cable,
average

of a prin&@~:&?&~t

stripline
device. H IS the distance from th,e m~~$~hce
the printed
circuit board to the D-flat h#%@$$urface.
the lead material

~~*Pts
I@edance

into the heat-

into the circuit.

must

prevent
device
rotation
during
cation, A wrench
flat is provided

been

swtt~hes,~rlvlng
c@$~s$$@Owding

to

the heatsink.
5. The device

A~S69.

Oth&{~,$h~J\cations,

the leads

is secured

or p~@F@n$TlOn

the

has

temperature,
ma$b?i$h>$
limiting
condition.
In this case,
use must be @~~@&transient
thermal
resistance
data.

leads.
4.

mountin9

incu,~?~t~mage

to as&e&.@inwhether
~t.,.-$..,&
bouR~~$~i,,, .

is within

instantaneous

of the
boards,

applied

it is important

meW.~$$

In applications
wherethi~w@rdissipated
conductor
consists
@~}~,l:,es
at a low

circuit

in the circuit

a suitable
without

temperature

in a vertical

in a printed

that

semiconductor

in tension.

the cap,

the

to prevent

Assuming

in such a man-

joints

on the strip

When
header

be mounted

ceramic-to-metal

ner as to apply
3.

never

should
H

sponding

to

Integration
integration

be taken
points

able number

can occur.

current

values

may

with

using
the

are then

of time
are

be

performed

of a complete

cycle

a limit device.

temperature
read from

increments,
multiplied

by

The pictures

stabilized.
each

taking

of the voltage

photo

Correat a suit-

Each pair of voltage

together

to give

and

instanta-

278

.-, .--

--

-----

neous

values

of power.

graph

paper,

the

counted,
along

The

number

and the total

the time

dissipation.

axis.

divided
The

and

are plotted

is the

are available

make

the

the

throwing

curve

under

of squares

average

the

these

test

load

switch

This method
urable,
switching
sting

is baaed

smooth

upon

dc source

arrangement

the load with

substituting
for

is provided

the device

an easily

a complex
which

under

allows

The basic equation

so that

where

q =

written

under

h = heat

transfer

A = area

involved

work

electrical

in terms

of thermal

of temperature
tance,

or power

From

Equation

it easier

= LT/q

(h) depends

used and various

l/hA
upon

factors

Involved

often
be

made

thought

spondsto
ogous

The
using

to form
of

Equation

models

* ~~he
~~~$,may

~~~+2F

in that particd~r

as

a voltage
resistance

model

Al.
equivalent
Kirchoffs

(l). This

(2) and

of heat flow.

electrical

to current

resistance
Figure

between

therm:,~,

value
be

is

N~~g tR&t T could

*@rice

a semicsm~~)
> ${.$::s~!.
~~.i] ,$+,
electric?!is~r~lt
may

corre-

a~&fk~%!~owing

as indicated
be analvzed
equation

by
by

results:

are multiplied

for

minimize

= pOw**~ipatiOn

re,
thermal

resistance

to case),
thermal
resistance

heatsin

components.

(case

k),

Each component

junction

interface

terms.

temperature

thermal

A proper

resistance

its thermal

must

be

is to result,

resistance,

is also

increases

of the heatsink

and

variable;

ROCS,

procedure

can

are normally

it is also

In some

fiers

short-pulse

and

and

localized

be the controlling

resistance

where

factor

current

and temperaspecifications
densitv

is fairlv

such as in RF power

applications,
heating

by orientation

of the semiconductor

of biasing

thermal

applications

asambienttem-

affected

resistance

at conditions

uniform.

is not absolutely

increases

It is a function

Semiconductor

uniform

mounting

efficiency

of the sink. The thermal

will

devt~e$~l!age
>.!,,:~l
* ,+J,
~,
,.F(s ~,,,,
.,,;,,,
>!.$!
,i\. ~~,

ROCS.

perature

ture.

&~&er

nec:~~ry,~~at

the

= junctio~$:mpgratu

for the

The thermal

Law

10 sec~nds.

sigif,~~nt
~o~p~~ed
tothe~ther
the~mal.

constant;

(R); q@Wdp&er
(q) is anal\$,:.,
.
gives f,~~~~p~
basic thermal

for

Law

@~s

values

pD

,+,$,~in i& zed if the lowest

.:i

analogy

the

for about

TJ

of t%w?ndividual

~{$$$p,%:$resistance
the heat transfer

mechanism.
An

resis-

Ro, is
R#

T@

to

as the ratio

1, thermal

tem-

.,.$~,$es~ = heatsink thermal

resistance
(heat.*>,,,
.,.r~.
>.:),.,,
sink to ambient),
,,.:,i::.,
..*\.i
+1 it \$TA = ambient
temperature.
Ji
tk$&#al
resistance
junction
to ambient
is the sum

between

find

dc

,,* .+
, ., ,+,+s
.~~?::,+,
\:.t
- \
,,,..\\\,.,-,... ,,,
~lt.y
~ .,>.
$:$~,
\\!$

$;,:*$to

defined

to

The

case

f$~ction
R~$$=~~fetiace

of heat transfer,
generally

level.

when

ROJC =.$&~~nductor

dissi-

in heat transfer,

resistance,

to power.

The coefficient
anism

where

while

power

constant

position

used

supply,

semiconductor

It is generally

be

device

power

measurement.

steady-state

difference

engineers

power.

connection

coefficient,

temperature
regions

However,

to each

By

the

the normal

at full
the

and current

(1)
transfer

power

approximately

average

as:

rate Of heat
pation (PD)

is thrown

position,

a dc

APPENDIX A
~
RESISTANCE CONCEPTS

q = hAAT

AT

remains

is monitored.

test

supplies

is adjusted

Kelvin

test, until it stabilizes

for heat transfer

is generally

to

of the switch
it operating

to obtain

A
oper-

THERMAL

conditions

is connected

The dc voltage

meas-

waveform.

temperature
to the

to keep

perature
Substitution

switch

pole

supply

calculations.

Case

the

another

power

to pedorm

necessary

in temperature.

on linear

within

by the number

quotient

Oscilloscopes

measurements

results

of squares

current

density

in the semiconductor
in determining

power

ampliis not
chip
han-

APPENDIX
MEASUREMENT
Meaauring

the

interface

aPPears deceptively
ia a thermocouple
mocouple

on the

measuring
the

ing method

also

the results,

not

in good

mocouple

directly
which

the

dependent

upon

material,
be argued

that

Thermal

paths
materials,

clamp
does

allows

pains
The

described
mav

tab or a clip bearing

vielddifferent

values

tance. Another
discrepancy
package
is exposed
to the
should
mated
from

can take

be covered
that

with

Another

uses

the

package

bodv,

These

place.

To avoid

cause

The

thermocouple

Because
mOntig

the die is

with

}*
+
.~,:$,
.\\+,,.$
under$$~x~~e

directlv

the bottom
is close

the

ofk@%~,~#&case.

to th~i~e~pn

base reached,

molded

the

two

reqi@-

~oh$het

opportionofthe
screw.

position.
Locations

Consi~@~.,k~s~uation

are generally

depicted

in the

not

figure.

the qn~
a~aof direct contact
is around
the
sqm~~$~~rlv
all the heat travels
horizontally

the,td~k$$~om

the

Since

M$$~~@tconstant

die

for

values

$*thermocouple

a given

point

contact

test

setup,

the

location

warmer.

resistance
calculated

decrease

case

caae

Conseis hotter

the

values

as the

become

for

area.

thermal

resistance

increase

readings

reference

to the

junction-to-sink

thermal

cak-to-sink
of

hole
is

b~~wthemounting

issold~pto
,.
~~~&ee

~~aton-to-case

meth-

top

blind

quentlVl$be
tefiperature
atthe
JEDEC location
., ,!..,
than ~~~h~~~ermaliov
location and the Motorola

which

the

t~oug%ka

bodyf:o~~:shermocouple
,.~,> .:,i\$\.

locatioo

molded

ad

temperature

Thus the choice

temperature

is quite

important.
There
the

[t h~~e~*sti-

are

ca~a$~ii~~urred
,,,:.:...,,;r;.) ,
,> \.\t:\...t$:
mqasQk:&&6t
discrep-

between

heatsink,

the

conduction

the

mica
path

heatsink.
acrosa

If

tightening

instead,

where
a

the screw

will

a small

from

with

not bow

the

drop

mounting

at the EIA location

temperature

mailov

location

thermocouple
will

at the

EIA

location

small.

The

Motorola

The

EIA location

peratureon
ratings

belower

butclose

as the

lateral

location

will

ischosen

the case,

Itis

are supposed

the placement

ous and Ieaves

the semiconductor

be

Ther-

flow

is generallv

be coolest,

toobtain

UnfortunateIV,

will

atthe

to the temperature

heat

thehighesttem-

of significance

to be based

to the

will exist

package

base so that the thermocouple

The

heat

mica

the

hottest,

the

primarv

temperature

is

and

the package;

The

of the

the

grease

package

the die through

dimension

between

different

waaher

be deformed.

is from

vertical

relationship
are

mica

semiconductor

will

In this case,
the

the

temperatures

situation.

installed

measure

examples

thermocouple

previous

because

onthis

power

reference

point.

of the thermocouple

is tedi-

in a condition

unfit for

sale.
The

Motorola

temperature
case

is making

heatsink
Figure
Heatsink

B1. JEDEC TO-220

Showing

and Liting

Various

Cauaed

Package Mounted
Thermocouple

bv Pressure

to

fabricated,

Locations

and

at One End

does

not easily

280

location

ischosen

of the case at a point

contact

to accommodate
this

method

not mark
accessible

toobtain
where,

to the heatsink,
the

the device.

the

Once

this

the

the special
has

to production

However,

to the user.

highest

hopefully,

thermocouple

Iendsitself

~?..,<::,,,

oft~~s>+ ,t~).$,
.>s~

are shown~

along

as a

st.,\

lessens

a package

an insu-

such

To

areslightlv

the screw

locations
is

pres-

slightlv.

evaluate

in ROCS
for

with

location
the

between

shunt

through

oftheth~$~~~~$bleto

is inevitable

contact

Temperaturesat

this, th,ap%f~ae

foam.

bar under

in place.

the same.

for interf acethermal

insulating

is the placement

a screw

Packages

location

JEDEC

is e$en~tter.

packages,

TO-22O

Motorola

through

it can

can occur if the top o~~~l

ambient
air where
r~<d~~$(on

a 15 to 200/. error
significant

accurately

Motorola

thermocouple

tab

in tests

fixture

either

this source.

ancies

screw

all possible

surface

c. The ThermalloV

and

because

end where

mounting

of the package

drilled

path,

some

on the plastic

and

the other

the

mounting

the

but some
Three

swaged

oriented,

to secure

with

surface

values

to be used

to more

previously,

area

probably

conduction

be mounted

and convection

screws

Motorola

arrangement

Aa

contact

lifting,

b. The

package

material

off

Consider

B1. The

Use of a spreader

into intimate

ther-

the screws,
yields

lift

temperature.

in Figure

reached through
a hole in the heatsink, Thethe$@~~le
is held in place by a spring which forces thet~:~o~~uple

screws,

a TO-3

path

to insulate

in order

not provides

TO-220,
odsoften

head

on

insulation

this is application

insulation
lated

shunt

resistance

loy takes

conduction

also

This

for the

screw

thermal

through

ear

Iow

MI L-I-49456

forintetiace

place

with

shown

contact,

a. The

are

and

to

structure.

forintefiacether-

in place

flange

heatsink.

are artificially

the

fasten-

influence

methods

concave.

finish

in this Appendix.

package
take

the

located
improve

manufacturers

Fastening

may

from

to

ly, values

by different

case

package

sure at one end cauaes

placement

a significant

are considered

conduction

example,

The

RESISTANCE

semiconductor

TO-220

and

and

In addition,

have

Consequent

fasten ing the test

thermal

flatnesa

THERMAL
the

and

sutiaces

on the surfaces.

can

agreement.

of applying
the

by surface

presented

locations

When

R~cs
needed
a ther-

R@CS is proportionalto

be a factor.

thermocouples

resistance

means

between

of pressure

may

mal

for

area

is affected

and the amount

upon

anda

However,

of contact

consequently

of the

heatsink,

INTERFACE

resistance

simple. All thats apparently

on the semiconductor
case,

DC power.

amount

thermal

OF

been

testing

location

is

The Thermally

location

sen by equipment

is convenient

manufacturers.

ishes the case and may yield

for a TO-220
mal

package

grease

compared
often

used

usually

and

no

to the
with

a few

this

package,

watts.

When

values

semiconductors

becoming

becomes

semiconductor
tureruse
Another

This

of

some

reference

dissipation

the

better

when
is

ple near

its center.
under

A good

power
the

dif-

case

that

the

thermal

equipment

and

This

ideal

then

temper-

point

1 milfinch,

easy

is to also
while

test

testing

If the junction-to-case
as insulators

the case

International
Magnolia

are

reference

has a finish

for

use

but

Staver

Shore,

$!8,;~~

Berquist,
n esota

,@OO E@{na Industrial

SM?8$Q30% 1

Minneapolis,
(612)

16 Flagstone

Long

Thermally,
Blvd.,

~.$:!{f

Ch~m~,~ic,,

Company,

Drive,

Hudson,

Min-

Lane,

835-2322

New

Hamp-

1-BOO-633-WOO

New

Texas

~$~f~~latively

t#~,~,

&%ried,

Corporation,
California

Corporation,

Nebraska

6S601

135 West

41-51

Saxon

117W

Engineering,

West

Valley

1044,

Wakefield,

View

243-4321

Columbus,
(402)

Inc.,

Bay

(51 6) 666-8000

2021

Box

S49-24S1

Avenue,

(214)
PO.

etc.,

91S02

75234

Tran-tec

Olwo

York

Ot

is sag:~$ctw?.

Inc., P.O. Box 34S29,

Dallas,

Wakefield

281

Island,

Inc.,

has

:/.t
c~\@<,of

junct:m-~icase

(213)
The

interface

int:<f~$o~hermal

values

Research
Burbank,

the

and is therefore

in the

for

changed,

point

Electronic
Boulevard,

to

confidence

is

thermocou-

includes

conditions

grease

and the heat-

has an imbedded

improve

resistance

(thermal

package

reference

nearly

to

reference

constant
reference

way

resistance.

manufac-

point.

of establishing

washer

is flat to within

63 ~-inch,

resistance

however,

and

washer

than

copper

thesemiconductor

application-oriented.
It is also
become
widely
accepted,

specified

higher

a soft

between

sink. The

ther-

dissipaters

to the

utilizes

used)

it is important

available,
and

is small

power
of

atures

up to 1C~
without

of heat

since

manufacturer

EIAmethod

error

compared

significant

the same

differing

resistance

cho-

italsoblem-

to a heatsink

insulator.

thermal

junction-to-case
ference

results

mounted

and is often

However,

~-27~

Massachusetts
(617)

245-5900

PACXAGE

INDEX

PREFACE
When

the JEDEC

lines

started

tially

whenever

package

from

non-related
turers

assigned

attempt

system

to

relate

into a category,

1968

to

part

to

time
Can

new,

or manufacan

industry

based

new

on their

heights

with

the TO-204

family.

inal

outline

established

the

The

packages

are

caught

general

TO-3

physical

re-registered

the venerable
differ

60milpins,

for

TO-204AE

JEDEC

in which

were

to the

hae60

new

convey

TO-3

system

The

the
mil
well.

has many

available

andwithout

numbers

on very
still

anditia

with

lugs.

as

TO-204AA

TO-3

having

variation.

30,40,

lttsnow
40

mil

for

the

su~cient

old

parts

that

the

meaning

$*,>>
.tk~km~,,

totheori@,t:/:~

pins

pins, for example.

It aeema

50, and

class!fled{n

conforms

communication.

appearance.
Differences
between
specific packages
in a
category
are denoted
bysutix
lettera. The older package

new

permitted.

For example

variations

eithera

by JEDEC
the

establish

minor

designations

out-

sequen-

description.

to ease confusion,
in late

As

standards,

issued

appended

assigned

wished

industry
was

for package

were

standard,

these

via some

In an attempt
present

industry

number

would

standard

system

numbers

manufacturers

as an

developed

registration

in 1957,

while t~e --:,

,-r..<~~*,(}\,,,
,{7:,

really,~#ti?
DO-~$~Q.,&,.~nd

for$enma~.tierbal
,,,)$ $.:,;\ .(C.J*
.,>.,

i~i%$
.!.

282

..--

Mounting
Very High
Prepared
Heige

by

O. Granberg

RF Engineering
Advanced

Products

RF power
MRF155,
01,

semiconductors

MRF156

whereas

MRF176G
these

Group

and

such

MRF430

MRF141G,
use

Case

are high

power

(both

devices

large

amount

physical

area.

For such

special

attention

must

be paid

as well

as the finish

equal

to or better

face flatness
heat

sink

can

should

beyond

perature
under

normal

150 W and

sink

flange

surface.
sur-

with

lower

The

thermag

The

heat spreader
for case

extend

the

data

above

hardware

~@:.

sc~~~~r~a>

the

amount

t$e

with

split

the

limits

of

been

all

operated

measured

insulators

operating

is taken

Into

and

implies

that

case temperature,

that.

.. .
.-...,..
AJ.,
*, <A>
%. ;- ~
T+$
j,,,

copper

would

torque

indicate

flange

an inch w[th

be an even
that

Increases

but

lower

if the mounting

screws
!n length

expand

mid

section,

well

as the dice.

of the flange

the length

a temperature

cracking

are torqued

It must

increases

change
but

the

dissipation
would

temperature
in the

rating.
limit

sheets.

within
150C

will

with

thousands

too tight,
bend

the

device

can

is true

but at a

case

good

temperature

to not higher

Oxide

be noted
temperature.

that

of
case,

In the

insulators

as

the thickness
For the excur-

283

CASE 36e.01

be

engito

than twice

the flange

upwards

or the
storage

which

if found

of 75C. In such

It

these

at 25:C

The

However

the

and the

data

of the case 368-01

of two

Beryllium

also

choice

can
with

dissipation

specifications.

in excess

cannot

better

cen-

The same

can be kept

its f~%;,,~j,~~riess produces

or lower

t-he

deflections

pcW~t{ebgalnst
the heat sink. The split lock washers
are
.~g~~~lewith
various spring tensions.
Bell type cOmpres.$:@ S~~ washers
,*
W!th 5 in.-lbs.
~. ~i:$,y..!
>~:?
Calculations
,,t$l>!:.,,t.<

flanae
on

RF devices,

account.

and the die temperature

sufficient

The
hole

the flange

to be operated

70-8VC

for

of the

its type,

given

practices

torque

tom--

torque,

Depending

junction

usually

neering

enough

mountina

is less severe.

indicated

derated

center

between

Oxide

flat-washe~f,
ofi~$?ch
the latter should be in immediate
contact @l&.the
Flanges top surface.
Experiments
have
~,.
shown
~%~~rely
compressing
the split lock washer
to

much

screw

by such conditions.
mounted

are

at that

copper,

heat sink, it can

as the heat sink

the initial

and

if the case temperature

temperature

of case 375than

out of the mounting

the

to

mounting

nt of thermal

sufficient

flange

problem

factor

type
flange

manner.

be able for the device

derating

lock- and

of

amo;

compound

created

dissipation

should

actually
follow
them.
torque
on the #4 size

in. -lbs. along

have

the ini-

exact

durina

under

a similar

The maximum

on the

procedures

remain

Beryllium

levels

of

with

from

coefficient

excessive

along

it in

with

thicker

lWC

torque

roils,

m0ufltin9

is difficult

expansion

if an

corners

happen

total

mounting

ver&%&@igners
,re:&mkended

the

0.25

the

of increase

a copper or aluminum
during a cooling cycle

will

of 23 mile

and the bot-

information

on

of thermal

terand

A tem-

dissipations

effective

COm DOUnd will squeeze

but

of

Th&copper-tungsten

IS applied

areas.

type

5C and

d~~~

device

deforming

d?~ temper-

(n caseJS @$%

sh~t31~qEtaln

scr@~$~$*-5

and

i$!.pound

1.0

on t@t~~~:ice

involved+~

is appr&>~@ly
<~+. ~, 1*
con}Q*@Bnd

as well$las

measure

., ~~pthe

>%

at leas~.O.5*

increq~i,~~k~~

Howeve~&ti&whount

ma$q~~al contracts.
....* De?Brmation
can also occur

s~~+d+o

375-0:

is around

fu:$b~a,t$~ghtening

but if @$uk~&@on
be +ti:$~~~bent

conductivity
sufficient.

in

01 has :+,$Q&!&wer

material

of the mounting

depending

operating

these devices,
The maximum
mounting

heat

thus

tial value.

300 W respecti,~:l~~+;%

Although
subject

should

the mounting

of the

to the

of 0.25

of devices

between

center

within

above,f~~~ount

results

but in that case a coD&%~>\... thermal

of dissipation

calculations

screws,

of

transistors,

and for the mounting

be used.

edges,

and
All

dissipated
power

of material

which

results

high

mentioned

368-

W), which

can be considered

and

the flange

and the amount

ature

copper

thickness

for 368-01

below).

at least a thermal

such as aluminum,

a minimum

0.375

tom

than

be made

spreader

have

have

f 0.0005

conductivity
heat

and flatness

should

MRF175G

of heat

sion

MRF154,

in Case

shown

(200-600

a small

material

housed

MRF151G,

375-01

in an abnormally

The

as MRF153,

are

CASE 375.01

284

--. .--..-...

_.

ANI044
The MC1378 A Monolithic
Video Synchronizer

Composite

./.,

Prepared by Geoffrey Perhns

INTRODUCTION
,;?.

The

MC1378

made

between

and

a locally

the

necessary

color

designed

remote

to enable

composite

controlled

color

RGB source

synchronizing

an interface
video

of video,

circuits,

plus

to be

PDI

sources

color

MC1377

and

be found

the MC1378
in the

a detailed

in AN932,

The

and MC1377

MC1378

rature

encoding

It contains

horizontal

sync

a complete

PLL. The

4 MHz

the

accuracy

externally

burst

are

from

is very

similar

used

discussion

of this

subject

decoding

flag

differences

and

between

encoding

sections

and color

subcarrier

determined

adjustable,

operate

circuitry

major

color

digitally

the

MC1378

are that
quad-

and

are

not

is designed

to

phase

dete~?r

and

compares

eight

t~~~~?st

gate

th~.S%unt&wn.

within

the6+wWfor

video

si@3~*&urst

gating

abou,~.th~;<$~ps

burst

-..

VCO

stage

divider

is also

burst

flag

is used

clamping

is 4 AS wide

flag

signal,

other

signals

chroma
is also

should

,k~~hori%entalfrequencv; e.g. TDA3301/3.

and

is centered

gate

that

by

extensively

and

Burst

devices

to form

by 256

gate

and

hor-

internal

divided

and

8urst
gate

out &.@~n 5 to drive

4 MHz

is internally

freq~e~~%e

to develop

the

the MC137&s

cq~?w$~~the

VCO$ISfia~

~;f

t~

TTL sync fed into qifi@.~Jth

to horizontal

NTSC/PAL

to the

is a digital

izontal

encoder.

The
can

was

fed

be locked

Phase

detectors

a 5 V supplv.

,.m,
3 & 4 are not activelv used in the LOCAL MODE
but
The MC1378
contains
all the necessarv
circuitrv to lock ~s ~D~sets
an arbitrarv
oscillator
phase to the two eleca com Duter to a remote
color com Dosite video sourcw. ~:+ .&onic
ohase shifters,
:1.
and to switch
between
the remote
and the Iocallv @#%,R~
In the PAL mode the R-Y modulator
is phase inverted
crated
using

signals

to create

an additional

overlavs
when

in RGB

operated

remote
can

can

directlv

1 AND

the

No external
functio~

it is~pgi&l

intW~,yS:

number

frequencv.
at pins

could

the

video

signals
frequencv,

This

10 and

be used

zontal

displav

phase

as

basic
mode

the

the

and

when
the

subcarrier
The clock
wilhin

the

forms

overlav

oscillator

as

signals

at pin 27.

graphics

must

will

25)

the

svstem
the

B-Y

phase

is disabled

can

should

NTSC

bv

be reset

PAL
at this

be set

or PAL

low

in the

encoded

RGB

MODE

separator

to

svnc

its svnc

pin

24

to

signal

detect

being

and

going

computers

the

divided

encoded

host

heat

hori-

the

result,

285

down

svstem

The

on

4X

range

color
color

to
and

burst

the

The
VCO

separated
using

into

subcarrier

sync

PD1,

svstem

to

bv 256

to

in PD2

with

at pin 40, The output

clock VCO,

14 to 36 MHz

Horizontal
from

is divided

is compared

requirement.

sig-

timing

svnc is sep-

svnc.

H svnc signal

active

fed to the vertical

4 MHz

4 MHz

this

from

fed

all

Composite

and then

of PD2 is used to IockthesVstem

can

are

or remote

provide

vertical

The

frequencv

of which

signal

fed out to the graphics

generator.

the TTL negative

A separate

to the

into

detectors

video

svnc is used to Iockthe

horizontal

crystal

or beating

phase

to the host computer.

lock

to

all
valid

the remote

being

be related

be fed

the vertical

divided

mode

An external

from

the

dot crawl

circuit

flip-flop

information

of subcarrier

not

and unpleasant

ident

arated

a multiple
that

R-Y and

at an arbitrarv

to pin 29. If a particular

(pin

to view

remote

PD5.

composite

signal at pin 8 and

down by
is divided

11 as a reference.

disadvantage

the
PAL

enable

MODE

nal

svs-

oscillator

a PLL using

to both

runs

connected

LOCAL

except

PD5, is used to com-

4X subcarrier

if it is not

subcarrier

mav

mode
diode

is required,

In the

the

1 V RGB

graphics

crystal

detector,

is sent

modes,

separately.

them

local

subcarrier

flag

PAL flip-flop

pin.
The

is to encode

phase

running

subcarrier

the

and

sync

{~+~..?~$heappropriate

frequency

to describe
in the

4X

~%~ f&@r with the returning

.~.+a~.~~.~{rol the clock oscillator.

clock

local

an external

a burst

The

REMOTE

or PAL and to drive

the

balanced

p~twoowfree

oscillator

in the

the

MODE

in two

of ~:MC1378

clo@}@~@

and

BL@CK;~IAGRAM)

vid+~;$~~uired

a refek~x}
A~~~~:

..t,?t,.<<
\\\
.$>.,*.,,,
,. ,,.,
. ,,
!,:,

with,@u~~~~@ding

bv line

modulators.

quali,ty~,~~b~perienced

MCIW&~,~~&tes

local and remote,

signals

to its output

line

~~t

tK%~~C1378,

mod&,, ~a-&es

DESCRIPT~Q:kJ;t~~~AL
,,. .... :-M.\,*/

(SEE FIGURE
Because

Because
locked

video.

airn#ltan@us

sophisticated,,:~~~k$$
~:?b$%,,\>

in less

SYSTEM

tems

remotelv

no loss in picture

happen

main

in composite

the TDA3301/3,

be created.

in the

signal

re-encoding,

overlavs

device,

the frequency

depending

The

svstem

frequencv

upon
clock

within

is
the

pin 40,

the

remote

signal

oscillator

using

is used
PD3

to lock

which

is

r,.,

Burst Gate Out

n
I

I
12

I
I

r Corn;.
Sync
In

Lx

I
I
I
.

7/26

figure

1. MC1378

Appli~tion

Schematic

XTAL Spacifimtiorrs
Freouencv

Ior
I Mode:

I 14,318180

MHz

I 17.734475

MHz

MC1378

-w
cPAL

AA

I Fdametal

Frequency
at 25C

Tolerance

~. ~pM Max

] Frequency
Temperature
Tolerance Aflfo
O-70C

CER RES
BU};;P:;TE (

40 PPM Max

~.

H0R12 SYNCIN

39

COMPSYNCOU1

,..,.

NTSC

lM

100k

Load
Capacitance

18 pF

Vcc
m,

:~

18k

9
0.01

E
lofi

2.7k

IM

Vcc

100k

For no,sy remote video signals

lakes

longer),

.5ethese

(color lock
valesfore#ther

NTSC or PAL,

S,,

,.

Remote

Mode

14321773

MHz

@ 3584.43MHz
500 mVP-P

.}
Horizontal
Pull-in ITYP)

Ceramic

Resonators

For MURATA

with
of the

burst

gate.

Iocallv

BV using

generated

PD4and

composite

525 Line CSA4.03MTF102

I 47D

18k

680

10.0221

625 Line CSA4.00MTF102

I 1.8k

68o

100221

the two

burst

tor

made

and

amplitudes

Iocally

equal

generated

comparing

the

generated

video

the

falls

from

signal

below

adjusted

burst

bv the

SimilarIV,

287

amplitude

being

bvchanging

= 300mVp-p),

equal.

chroma

turned

a predetermined

470 kQ kills at about

essentially

a variable

kill detector,

the remote

are made

47P

are compared

using

encoder
section with the same aubcarrier
reference
used
to lock PD3, thesubcarrier
phases of both the local and
signals

CA

The absolute

burst

Rg

I -400,-400

is detected
gated

RA

I +400,-400

in the

~\.?l

CK

Resonator

10-20

ACC

amplifier

path.
of the remote
thechroma
off when

level.

thavalue

in the ACC detecgain

mVp-p

kill level
resistor

remote

burst

onlv

Iocallv

the remote

The

of the

signal
of the

burst
can

be

at pin 31.
(normal

In the
burst

PAL

mode

the

is compared

flip-flop.
ident

Ifan

is not

flop

error

ident

with

the

the

black

local

operated

by the overlay

be used

that

burst

from

phases,

(sync,

local

(SEE

BLOCK
MODE

specific

to weak

signal

14,3MHz
LORC
COULO BE
vARIABLE

switch

to create

of the timing
from

either

purposes,
remote,

OPERATION

such

17.78 MHz
17.90 MHz
LORC

or VCR

compares

the

internal

horizontal

the

4 MHz

VCO

with

derived

from

the

master

clock

from

The

PLL formed

host

computers

Iocka the internal

used

in LOCAL

MODE.

used

in LOCAL

MODE.

active,

setting

between

chroma

phase

but

providing

an

the subcarrier
of the

locks

the

host

the

locally

to the

master

clock

VCO

(divided

host

computer)

the

four

subcarrier

crystal

oscillator.

the system

the output

composite

within

becomes

phase-shift

and

generated

frequency

timing

The

video.
down

to

to

we

crystal

standard

in

,p~h

the.,~t~~~

.,3. ,<},.,.,,i,,.
x,.,
,,,~ ,$.>,
..!..
,..
.->.:Ti$.,;
~,],

MODE.
REMOTE

sync

computer,

signal

arbitrary

reference

subcarrier
times

frequency

Horizontal

horizontal

PD2 not

the

signal,

PD3 not

COULD BE
VARIABLE

SUMMARY

DIAGRAM)

from

Iator

ampli-

pin 25 can

to the other
portions

derived

PD5

burst

are compatible

fed into

for

DETECTOR

LOCAL

PD4

on remote

etc. ) can be selected

due

flip-

reduction.

PHASE

PD1

Even

sources

reduction

jitter

signal

burst,

or remote

as noise

one

local

the

the fast video

signal

video.

Typical Master Clock Frequency

Configurations

2.

is killed.

states

signals,

the

Figure

If a continuous

or no burst

signal

enable

in composite

waveforms

tape

local

remote

to switch

overlays
the

and

the PAL

ident,

and in the case of PAL, ident

between

remote

from

remote

processor.

levels,

of the

indicating

i.e. fixed

in the

ident

line signal

the

ident

is detected,

the chroma

Because

of the

half

is detected,

using

error

tudes,

the

compatible

is reset

signal,

phase

with

MODE

and Iocksthe
intern al@N,d&k$ed down
remote
hd?~~~~~? Sync. ItiS

PD1 compares

4 MHz VCO to the incoming

fast

acting

to follow

PD2 locks the

the

internal

from

the

horizontal

host

source
clock

phase

frequency,$divi~~~by

the outgoing
incoming

lp~

PD5 ~op.~~
,}.$$>.
~~:,
:*,

$fferent
values
that

the

video

in picture

incoming

stables
shifter

to assure

has the same

phase

that

drive

as the

into

the

dows

the one

typical
mica
stability

right

Clock

in the standard
the circuit

except

cen be standard

a Master

shows

frequencies.

capacitors

in Figure
delays

input

delays

position

D, in AN932)

on the

blanking

MC1378.

aspect

ratio

Tl,

may

monopulses

T2,

such that the inserted

and

to

is the picture

3. Six-one-shot

and

TEST FIXTURES

TO SIMULATE

Sometimes

problems

the

0.001

MC1378

frequency
schematic.

It is recommended
and tem-

WF coupling

MC1378
is tried

can be solved

to

T3,

picture

and
win-

be adjusted.

using

in the total

fixture

are shown

in Figures

modated

Clock,
Hz PAL.

one

for 525/60

Other

by changing

clock

A COMPUTER
be avoided

a computer

using

to be fixed

625/50

can

to

details
Master

caps

ceramics,

major
the

application

problems

and component

be used for accuracy

for

shown

Appendix
or vice-versa

MODE.

shown

clock

bars

variable

overlay

size,

that can be used with the RGB color

color

circuit

create

T4 are variable

at PD3,

require

IN PICTURE

(see schemadc,

connecting

tothe

for

silver

perature
which

from

figure

to

~o~;~j~f;~~EQUENCY

applications

}i,:wny

insert

locks the crystal

phese

test fixture

bar generator

four

in REMOTE

,.~~&AL
:;,.!;;,?,l.
+!..
@Y$, T

Another

returning

@~d}burst

re~~&ke~~&rst

PICTURE

etc.

by comparing
H sync

~~~bwhhich

remote
signal bursh\
~.
PD4 _ controls~~~~~~?nal

The

f14$F~t~$#ns,
@fil\~tor

signal(:J~$@K@
..\,, },!

computer,

PD3 a gated
oscillator

VCR
master

a test fixture.

the fixture
system.
4 and

Hz NTSC

the divide

and

the

The major

leaving

Two

5, Both

frequencies

if, before

system,

types

smeller
of test

use a 36 MHz
the other

can

for

be accom-

ratios.

288

.-.

e---3.-.Rcture
----In

Raure

Ktiure

HORIZ.SYNC INPUT
FROM TEST FIXTURE

+5V

FROM MC1378

R3
0.05
_

005

m
lT41R4c4120wsl
I

T = RC 10a.2
logn2

= 69

P
T3

T2

T4

EXAMPLE OF FORMAT
.$,.
.

The Master
ered

using

Clock

a shoti

is amplified

ers

the

accordingly,
output

with

going

down

t:

?@:$@ch

MC7WLS221

pin 2,

or 4.43

,,,&3$k~,?at

pin

10 so that

correct

frequency

NTSC

MHz).

the

MHz

pin 8. This
burst

signal

frequency

should
at pin

5.

Switch

to REMOTE

6.

Adjust

the

correct

subcarrier

now

at

reads

+5

to within

V).

9 to

give

the

50 Hz at pin 20

Feed

1 V p-p composite
and

27, The

pin 25.

composite

color

incorrectly

as

burst

color
sync

will

video

into

should

now

be absent

if the

pin 24, Color

appear

at pin

100 k pot was

adjusted.

MC1378 APPLICATION UPDATE: CLOCK

OSCILLATOR ALIGNMENT

MHz

pin

and
3.58

1 =
at pin

capacitor

= 4.4333619

Reconnect

(pin

frequency

burst

at pin 20 is the

into

MODE

100 k potentiometer

as in #2.
7.

frequency,

frequency

~ 10 kHz.

phase

waveform

5 PS wide

variable

feeding

this

the correct

10mV

10 Hz.

until the meter

(PAL)

the

PAL

pin 40

has

Pin 2 waveform

and the twO sche-

appearing

MHz
signal

Measure

Clock Oscillator
or 4,43

burst

to within

3.579545

Disconnect
4,43

the

the coil to give

oscillator

sYnc fre-

the

subcarrier

adjust

the

it to give

5 (use

to feed the MC1378

possible

of accurate
trigger,

When

adjust

at pin

TTL

oh],~as a guide.
v,><:i;

2.<~:1~9
a source
~$,:
~
?* oscilloscope

oscillator,

frequency

trigger).

adjust

an RC network.

M~f@
S~T-UP PROCEDURE USING
~ t:.<.i,..
$~s,>;:3 TEST FIXTURE
, ~$:tt
. h:,,..
N
..j.i,.,
h..
>,\
1. S~~,W to LOCAL MODE (pin 1 = O V). Ground

...}~i,;:.
:.,,.$
..
\.\.l;i, ~.$>.>
::,7,.~~~.,
3.
1

MHz

This

horizontal

produces
signal

first divid-

in the 4 MHz

horizontal

as a scope

hex

locked,

MODE.

roun~:d:@%&ng

are u~@

va~J,:$~~,:$are

are show~

~~~wz

correct

followed

fae&S&,

in the,&~&V

If a coil is used
the

clock

the .,~&~4L~4

shunt

horiz@~*{~~c

at pin 40. Other

of 50 Q coq~?,~he
of

4.

378 ca~ be @liv-

the first di:@&.,,~he

PD5

and

the MCI

section

frequency
to drive

A one-shot

negative
matics

one

dividers

quency.

(<12)

to drive

is reduced

Additional

length

connected
stage

bring

from

using

buffer-inverter
by another

outDut

MHz
the

be phase-locked

Two

(3.58/

adjust

the

(NTSC)

signal

new

and

Iocked

loop.

Figure

remote

to the

ance

20.

289

are shown

range

promise

to

circuits

pull-in

between
modes.

speed

in Figure

of the

6b shows

Clock

a circuit

the characteristics
Both

on the alignment

circuits
of the

6 to improve
Oscillator

the

phase-

that

has no com-

in both

the local and

allow

a much

Clock

Oscillator.

wider

toler-

Rgure 4. Test Hrzture,NTSC 525/W Hz

4+5V

.
I

b7.16

I .

4CP

W URECnY ACROSS
MNS 7 TO 14.

Ik

m,

OMCP

Ik

MW

OWCP

. . . b?

OucP

.
. . . -.

OL

SN74LS05

+SV
7005

?k
2,2 k ,

290

. ...
.....

-- .-,.

Hgure 6. Clock Oxillator Akgnment

Rgure 7. RGBI to RGB Converter

PIN37
0,01 WF

220 PF
= 1V/. ~

5.6 k

2Nw1
,,,k~ow

LOCAUREMOTE

(al

RGBITTLTORGB,
Figure

7 shows

personal

a circuit

MC1378.

Ifthe

coupling

capacitors
The
clean

very

signal.

to

circuit

MC1378.

with

I V ANALOG CONVERSION

computer

be

(b)

to interface

the

RGB

is used

to

IC4 is used

RGBI

output

inputs

the values

of

the

shown,

no

to the RGB inputs

of the

to the 390 Q resistors

prevent

TTLdiaplay

with

are required

+ 5 volt supply

a TTL

analog

interference

to simulate

on

brownto

should

the

encoded

be compatible

monitors.

USING THE MC1378 IN CONJUNCTION WITH

THE TDA3301/3 FOR OVERLAYS IN BOTH RGB
AND COMPOSITE VIDEO
ilxi
In some
posite
MC1378

application

overlav

not only

color

The

TDA3301/3
MC1378

output

and corn-

situations

locked

of the

pin

Because

at

27.

buffer

ble the burst

gate

but

also f&.
, ..

svstem

operating

the

as sho~,

shown

in

Noninverting Buffer, Level Changer

+12V

2NW2
FROM PIN 5
OF MC1378
4 Vp-p

8, to ena-

BURST GATE
%

TDA3301/3

2.2 k 12 VP-P n
BURST GATS

15 k

2,2 k

of the MC1378

onlv when
must

TO PIN 27
OFTOA33013

2NM01

of the

the

lk

at pin

slice level

op~~qg,$+t
pin 38

Th@,*Wh&@Foutput

8.

a small

~,p?gura

A vertical~mfor

in the REMOT~~M~~

Figure

of the

otkf~~%vel

to exce~}$~~,wquired

can be

is applied.

app@#~Jng
pul~~$[:d$t

the TDA3~l~Sd~ectlV,

is used,
pulse

TDA3301/3.

clamping

MC1378

Ss

tb~.i:+l.,

to the rem~~~k

computer,

the sandcastle

is too low to drive

a valid

video

be inverted

figure 9. VerticalOutput Inverter

as

9!; If ~ continuous
vertical
pulse
is
, ... +,->..,,
so tha~.~hS.@ti~ut
clamps of the TDA3301/3
are

required
alwavs

base

graphics

to drive

noninverting

signal

as a time

for the

gate

be used

at the

RGB overlay
In these

decoder.

burst

5 can

both

are required.

can be used

source,
the

video

video

Figure

oper~ti

n&~@focked

50/60

Hz oscil Iator

to be

use~~this
cbuld consist of a MC1455
.,) ,,,,.
circuit. Ifs ~~$~al
pulse is produced
bv the
puter:,~rw:c$
in kQ.~L
Mq#*

source,
MODE,

found

it should

be used

an alternative

to drive

<W ~%
overlav
fast
~,.
+ J+..t.
~ ..T@3301/3
operate

the

source

wil I have
tvpe

timer

microcom-

instead.

When

of vertical

svnc

Figur~ 10. OverlaVInput Inverterand DelaV

TDA3301/3.

video

switches

in the

opposite

in the
sense

MC1378
to each

and
other.

TO PIN 23 OF

$ira,:Y$Therefore
an inverter
must be used between
pin 25 of
.
~,,i;~,.
, ~$>.t
and pin 23 of the TDA3301/3.
,,>-s. .,>,:* the MC137B
,,+$. \
The delav produced
bv the use of a delav line in the
~:$.
,,.
luminance
path of the MC1378
must be compensated
bv
using

a similar

delav

in Figure

10. The

between

the

connected

in the overlav

RGB

MC1378

as shown

inputs
and

the

in Figure

enable

are ess&ntiaiiV
TDA3301/3,

line as shown
compatible
and

can

:-N;!:
be

11.

291

MC1378

figure 11. RGB Input Connection

Hgure 12.3.68 MHz Chroma Trap

10 pH

10wF

14 ~

0 ~

TO
MC1378 15H

0 F

25

To
TDA330113
1 Vpp

0 W26

10 pF

16-

TW&pN37

24 Sl;C~ YYEE
TOA3301)3
OATASHEETI

INPuTS
%
RGB

390 PF

390 PF

figure 13. RGB Output Blanking Circuit

(One of Three Required Shown)
+12V
POSITIVE
COMPOSITE

5V ~

2NW1

BLANKING

TO

22k

TO

GREFN CKT BLUE CKT

NON-ADJUSTABLE

B-Y

PIN 20

NT3C CONNEUION

I
I RED FEEDBACK
I
[ NOTE: Se, feedback
Pots ad
/ brightness
,.,,.1
{or correct
~ b~~~al
.,0.,
_______

3.58 MHz

shown

chroma

in Fiaure

the TDA3301/3
A circuit
with

blanking

data

could

Figure

and

in Figure

Normally
applications

13. The

Figure

edges

is nat

it may

ne$$,$~J#

delay

the offending
nance

subcarrier
MC1378
artifacts

crass

bad ~@$~;~tempting

to record

wad~

the

eK~:,&,~+On

Because
aging,
to

pAL

the cross

operating

notch

filter

in the LOCAL

MOOE

can

-w

of

10k

Of lumi-

can be espe-

method

with

composite
be

type

video

signal

input

at the

composite

35 dB at 4.43

Figure

video

are better

16 will

bv an additional

292

+ X

47k

about

frequencies

present

2N4402

33 k

when

22 k

2N4401

75 n

is equally

and its pack-

10 k

22 pF

REMOTE
VIOEOIN
YT

inter-

troublesome

a video

Improved Remote Wdeo Input

IsolationUrcuit

16.

A+5V

in the

some

on consumer

in the device

remote

output

on fast

chroma-horizontal

The

limitations

talk from

video

MC1378
PIN22

1.2k

pictures,

Or NTSC.

of certain

composite

effects

notch

problem

cially

random.

more

Figure

ta remave

VCRs

leav(~$~Q:#kmes

for

at the expense

calor

on playback

but

noninterlaced

a$$~~pie

band~(~$,%e

91PF
~
3.58MHz 1 =
22 PH
_
-

*1. FOR443MH2,
C=68DF
L=18#H

RGB

,.

~%<~j$fhe

q~$~,a~br

the

part

app$~~~%t

OELAY

1.1 k

1000

be rQqut(~d.

es~ec?$~~,:in

15 shows

luminance

400 ns

MC1378
PIN 17

a 75 ~~a~i

5 V c~,wg~e

.,,.. ${,. :4.}~{+

MC1378 SUBCARRIERNOT@ F@~ER
Cross color can cause ~#~&*?R~rainbow
luminance

is

see
\:ii;,;

o$~~$ c~?cuit
.?. \.., ~;,
$~~~..%,,,,
for balanciS~.,&~:$.58
MHz

leakage

this

driving

I
1
,

input

information.

by using

a method

demodulator

outputs,

luminance

filtering

4,

exacting

the
aeneral

be developed

14 shawa
MHz

for

mare

sheet,

is shown

in Figure

or 4.43

trap

12. For

for blanking,

1 V p-p

shown

NOTE The c,rc.<t shown with!.

the donti
1,. must & dpl,ca.
ted for blue and for green,
_______

imprave
20

input,

10 pF

Typically,
better

and

than

60 dB, The circuit

dB.

TO PIN24
MC1378

the crass

MHz

the isolation

FROM
LOCAU
REMOTE

at 3.58

talk

MHz.

shown

in the LOCAL

is

Low
in

MODE

NTSC LUMINANCE COMB FILTER

MC1378
To avoid
color

loss of luminance

artifacts,

(see Figure
agement
scope
only

a simple

bandwidth

comb

filter

17). For 625 line PAL,

has

of this
effective

to

be

made

application

which
note.

on interlaced

Noninterlaced

signals

a more

The

comb

become

arran-

worse

added

is

signals.
with

this

input,

amplitude

a small

the

filter

horizontal

line

The

in NTSC

be bevond

NTSC

and

delav

removing

complex

would

color

could

while

can be used

pins

1 and

2, on noninterlaced

and

phase

adjustments

amount
at the

(550

mVp-p)

output

of the

adjustments

are trimmed

22,

this

BV using

artifacts

are

of 3.58

MHz

400 ns delav

for minimum

technique,

removed

virtuallv
without

signals.

are made
line.

The

subcarrier
all the

loss

when

subcarrier

of

ia
two

at pin

cross

color

lumi~nce

,>
NTSC

DECODER

Figure
the
notch
this

COMB

18 shows

luminance

a circuit

as shown,

FOR THE

similar

bandwidth

in the luminance
filter,

FILTER
bv

channel
is onlv

luminance

T~301;3

to Figur&~7t&$mprove
remoyjw

&

3.58

of t&t~D~%01/3.
app$$~~~$o

MHz

NTSC.

respectively

laced

video.

amplitude

Again,
Both

and chrominance

Iuma

nance

to

The

setup

and phase

are combed

remove

is accomplished

for minimum

output.

.f,y.>,\\
:.;>..?+.

,<?+,
.%~pfigure
,;>Z
,,.>, !,\:...$
I

18, NTSC Decoder Comb Filter for the

TDA3301/3

;.

IJ

ADJSFORNULLF
;22
CHROMAIN LUMA CHANNEL.

=002
b

WIRE:36G [0,1mm)
34 TURNS

293

!.

colored

2:Y;fifi:~E
3.579545MHz
63556 MS

of chroma
artifacts

bv adjusting

subcarrierat

and

in interthe

the lumi -

Figure 19. Carrier Balarteaof Color Modulators

CARRIER BALANCE OF COLOR MODULATORS

+5V

Certain applications require perfect carrier balance of

the color modulators. This is simply realized in Figure 19.
The two 100 k potentiometers should be adjusted with a
black signal for minimum subcarrier at the video output.

lM
100k

MC1378
PIN 12

R.Y CARRIER BAMNCE

+~
+5V
lM
100 k

MC1378
PIN 13

B.Y CARRIER BAWNCE

k~

,,\ ,,
Figure 20. Printed ~rcuit Board Lavout
:$)

tircult

Side Pattern (not full elze)

294

:...}.>
\

_.-~
--

MC1378
Pin 1

Local-O

Volts,

Remote-5

3 Vdc Approximate

4 MHz,

Distorted

4 V, 4 ps Wde

NTSC-O

Ground

3.58

MHz

Signal

14,32/1

7.73

MHz,

Local

150-300

RGB

17

Inverted

Luma

Chroma

Output

19

3.4 Vdc
Chroma

mVp-p

MHz

Probe

Sine

or Square

Wave

but otherwise

Signals,

These

Sinewave

Low

for

are Analog

1 Vp-p

3.58/4.43

Input

(Scope

for

MHz

3.58/4.43

3.3 Vdc
Invetied

23

3.5 Vdc

Black,

Inputs,

100A Color

with

High

24

Remote

Video

Overlav

Enable

Bars

Noise

MHz,

Burst

0.5 VP-p,

Burst

100 mVp-p,

Chroma

1000/. Color

Input

1 Vp-p,

Bars

Input;

Low

Negative

Composite

28

VCC

29

PAL

Video

In PAL

RGB,

High

Waveform

DC 0.5 V

30

2.7 Vdc Approximate

31

DC 0.6 V with
36 MHz

100 mV Vertical

200 mVp-p.

mav

36 MHz,
become

+ 5 Vdc

VCC

37

2.2 Vdc Approxi~te@ee

38

Local

Composite
Ve.~&#

3g

Composjte~WTTL

40

Horizo~~~?@%NC
,.,,. ,..l..

s~$~a,~

Input

$,YNC

Output
Output
TTL

with

300 mVpp,

in LOCAL

Le~~$@

for

Horizontal

at the,,s~fbw.
.i$.:~;::,,,

V Black)
Chroma

Color

Bars

-, ,, *$., .
?t,~,fi~,,
..,i<
\
,.,,.
.

..,?6,.,$ ....
Ri~$l~~~ti
Color

Application

Sw

Input

at Vertica({~~~.:
~,.{~ ~.!$.J
,.~
.iai.,
\*,!:>*.

Clipped.

at Black

appear

t$t~.

.t~$~fve
/,,, .$*
.~:>., ..&%
~y:.+
.).:$1 .*,S

~!gg~?and
.,,.,,T
,,\*A
., ..,,..,~,

36

in NTSC)

Difficult

Problems
Output

Used

All Blanking

Rern~~e,$gnal

1.4 V

~~,,,,

Pin (Not

Stepped

In NTSC

~tx~t
.1 .$ :
-,+~,\,, ..
,5PY<?;$,$p
. :.,:,\ ~:.
>~.
S.>!
.,
,. ~~.i

PLL)

Beat

. :J?\\.
,>.a:\i ~*$

Output

Identification

Mode

300 W~h~_~,,?800A
!:$
:(t~.$
,$*; *..,.,b
300 mVp-p,
10OO&,~@r
bars
~~~
ii ~:
~s $$,.
~~>~x,,.~~+
f: .>.>
V Black),
.\:\*,

(0.9 White/l.4

+ 5 Vdc

Remote

Shows

disturb

100 mVp-p,

SYNC

Encoded

= Approximately

27

Output

in Local

RMI

(1.8 V White/2.8

Harmonics,

Ground

Clock

PLL)

on RGB will

Approximate

Threshold

35

will

for Color.

so anv

Approximate
Luma

from

Probe

Amplitudes)

22

Ground

,*\
~.. .
..\,.\..,..,
,{.<.,
*\, t...
,F:Y >+,
,,,)\ .
~ . $t.
.,>.~i.
. .,,,s.+>
, ,~:~
,i- 4.,>,
..-$$

the Horizontal

Sianal

Approximate

21

34

disturb

unimportant

Amplitudes)

(Approximate

32/33

will

Color

Signal

(Approximate
20

mVp-p
Subcarrier,

Approximate

1 Vp-p

18

26

(Oscilloscope

to Horizontal

300-800

and

and Vertical.

25

Locked

14.32/17.73

3.5 Vdc

14/15/16

Note)

Wave

Signal
Pulse

MHti4.43

Distorted

12/13

Application
Sine

V/PAL-Open

Remote

10
11

mVp-p

4 MHz

WAVEFORMS

Volts

(See

200-300

EXPECTED

Unkilled,

Open

Also

4.2 Vdc Approximate

Conventional

Circuit

same

Oscilloscope

Approximate.

Scope

Problem

When

Probe

When
as with

used
32/33

Color

because

Killed

of Grounding

at Lower

Frequencies

the

at 36 MHz

Note)
MODE

in REMOTE

TTL

MODE

Negative
TTL

Negative

Negative
Nagative

296

.-

...

-.

APPENDIX
DIRECTORY
California

Crystal

OF

COMPONENT

MANUFACTURERS

(800) 333-9825

Laboratories

crystals
Coilcraft
1102 Silver
Cary,

(31 2) 639-6400
Lake

Road

coils

IL 60013

COmtec

(602)

526-4123

crystals
Fox Electronics

(81 3) 693-0099

297

298

._

ANI047
Electrical Characteristics of the
CR2424 and CR2425 CRT
Amplifiers
BV Dan

Brafion

Therefore,
case

The
pair

transistor

circuit

of the

DESCRIPTION

CRT

of complementary

DC stacked

across

is connected

at high

becomes

active.

All four

transistors

plasma

nitride

mounted

on

spreaders

(active

which

the transistor

the

x ,055
serve

flange

a thermal

resistance

(flange)

Junction
power

power

temperatures
dissipation

dissipated

amplifier

and

are

~:?Blp~e,
t~~istor

in eac~~~~:~tor

he~t

r$sults

i;%~tie

..~~!:
?.>,,..
t? Case I
f,.
Case 1[
Case Ill
Case IV
Case V
Ganeral

voltage

source

charges

CRT
HYBRID

c
T+

in
figure

active

is known.

is a function

if
The

of the

Then

the

through

1, Electrostatic

energv

a resistor

R = 10 ohms

inside

Discharge

the

to the CRT

and

C =

Simulator

capacitor
hybrid,

is discharged

Test

150 pF were

CASE

@
NPN

the

capacitor

pD(W)

pD(W)

075

075
1,6
0.2
0.8
1.0

<0.1
<0,1
<0.1

<0,1
<0.1
<0,1

0.2
1,6

1, UNPROTECTED:

The CRT hvbrid

pD{W]

VCC = 60 V; load = a.5 PF

of

1,

B
pD(W)

White level, output = 5.0 Vdc

SQ wave input f = 60 Hz; output = 40 Vp.p
SQ wave input 7.5 ns pixel; output = 40 Vp.p

conditions

used,

failed

Electrostatic

ground
ground,

during

charge

is used
which

at 2500

Discharge

period,

to prevent

could

damage

volts,

Because

Simulator
a 0.01

output
the

output

is connected
KF

of the

DC
CRT

vc~ = 60 V

0.2
1,6

1
VFAIL = 2500V

figure

2. Circuit

for Case

of
to

blocking
hvbrid

hybrid.

No connection to input pin 1; output = 30 Vdc

Black level; output = 55 Vdc

conditions:

up

(OUTPUT)

be computed

co&
fl&S as listed in Table
. ss<,,,
~,il:
,..*..,...
1. ~$n~tor
Power
Dissipation
...
h<
Q~
Q~
,y@;$3
.lTJ... .,, PNP
NPN
PNP
.... *$*.:
*
A
B

0.2
1.6
0.8
1.0

variable

tvpical)
four

operating

c=
!*,*,
@se al
x8tq~&&a.yv

th~~~$~:ldered

Table

Designation
Tvpe
Class of **n
.-.\\ i\.,\\
Cased ,l$,.,.

of

amplifiers

heat

ea~~~~~i~
.\ 1>...\ I .
.:,, ,.h..:~<-

can,

over

tvpe

~~

~,
~
,\.\. .*..
~$,,$$t~

fl~
,.$J, frdh
thin-f ilm Su~trate

for e~h

driver

eac~$$~~

copper

max~%$c!~att

for

transistors.
the

her-

This str$$t~c~

of 3~C/watt

to case

in Mot@~:&RThybrid

The

for operatio~%~~bWC.
At 150C junction
tern.!J,,,
MTTFs~~s
aWlndividual
transistor
chio is

e$#i@,:.@.
~~$ilu;.

identical

plated

orcasel

rise

5WC.

fre-

metal lizatio

the

,2:.*,
~1:}
~,,*>*
, ,,,,

have

the alumina
(heatsink

to the back side of the substrate.

for junction

mid

transistors

gold

to maximize

die through

to the aluminum

These
inch

used

$ 33+@

source

Class A pair drives

gold

is 1.6 watts

PNP device

current

bipolar)

junc$~ti*~p~perature

of a

followers,

areas),

passivation,
,055

top

case

A stages

DESCRIPTION

(silicon

geometries

Class

at DC through

B emitter

THERMAL
izontal

The

source

frequencies,

Class

consists

emitter

This complementary

complementary

peratures,

amplifiers

the 60 V supplv.

as a current

quencies;

driver

common

worse

(flange)

is rated

CIRCUIT

.,,t;,, p

.,s,%

CRICUIT AND THERMAL DESCRIPTION

OF CR2424 AND CR2425 CRT DRIVER
HYBRID AMPLIFIERS

to

CASE 2, PROTECmON RESISTOR:

A protection
resistor of 47 Q is connected

between

E.D.S.

volts.

and

hybrid.

a 0,01 pF blotting
discharging

The

hybrid

capacitor

failed

CASE

at 4500

is used to prevent

4, BYPASS

AO.1

the

and

Again

the hybrid

CAPACITOR:

pFbypass

resistor.

at 15,000

capacitor

wasadded

Inthiscase,

failure

along

of the

with

hybrid

diode

occurred

volts.

to ground.

0.1
#F

Rgure

CASE

3, PROTECTION

A protection
of the

3.

hybrid

istics for this

DIODE:

diode

(1S583

occurred
diode

Grcuit for Case 2

are

Hitachi)

at 9500
listed

volts.
in Table

was

added.

Electrical

Failure
CONCL~W
!$r}.>iit~

character-

2 below.
Obviously
tion
diode

the circui{.,~

to the

hybrid

should

ca&4

a~~~~~r.

offered

The

be E%ti$sc

the best

bypass

protec-

and

capacitor

losetohybridVcc

node

as

possible,
and ~f~and
.&ads on the bypass capacitor
and
hybrid shOul@~~S&~#to
carry surge current to inSUre the
best

prot~qtion.x~

NOTE@*~~,ode,
to besd%.that
drj+,~~$$put

Dz, should
large

be added

negative

surges

if there

may

is reason

reach

the video

pen,

:+i$::+!<

~~,
~~~6RMANCE
,.

Typical

driver

CHARACTERISTICS

bandwidth

areshown

and

rise and

in Figures

fall times

6through

of the

CRT

10.

.,.,, ,.
ELECTRICAL CHAR~@~RIS~CS
,. ,,,. ,:,
.},. , :.,$t}.~
1:~
.,>,, !.:>$ia
,:., . ...,
Forward

~,$~>

Reverf$~u{~~t
,,l~., ~~..,)-.,.
:
R$~&,~$Recoverv
~.-,
@+,klass

Time

(TA = 25-C)
Limit
Symbol

Test Condition

Unit
Min

Max

VF

IF = 100mA

1.0

IR

VR = 220 V

1.0

WA

trr

tF=iR=30mA
RL = 50fi, irr = 0.1 IR

80

ns

Seallng Codilion,

,:,,

300

-.

..

_---
._.
_,--.-

__,., ._

Rise and Fall Times

versus Output Swing
(Under Regular Oparation
Vcc = 60 V, Load =

Figure 6. Bandwidth versus Output Load, CL

Vo = 20 Vpp
CL

Vo

BW (MHz)
175

(PF]

18

Vvp

BW (MHz)
145
145

172
166
160
150
140

8.5
10
12
15

= 40

Figure

7.

output
Swing (V)
50

140
130
120
100

tr [nsl
2.6

and Overshoot
Voltage
Condition
8.5 pF)
Overshoot
(V)
Leading
Traifing
1,2
2.5

tf ins)
2.4

VIOEOOUTPUTSWINGIVOLTSP.PI

Figure 8.
Hgure

9. Rise and

Fall Times

versus

and

Bandwidth

Loads

CL (PF)
6.0 pF
8.5 pF
10 pF

A. CL = B.5 PF Vcc

= 70 V

12pF
15 pF
18 pF

B.CI

=15

DFVCC=70V

t, (ns)

tf (ns)

BW

133

3.5

2.5

105

50 V Swing

3,6

2.8

83

55 V Swing

(Standard

10

12

14

16

18

CL, OUTPUTLOAO [PF)

301

condition
40 V Swing

21

C. CL = B.5 p VCC = 60 V VOUT

(MHz) I

2.6

Operating

= 40 V Swing

Conditions)

tr (nsl

tf (nsl

SW (MHz)

2.5

2.0

142

All standard teat conditions, except add R.@

302

,-

..... ._,_________________

AN106I
Application Note
REFLECTING ON
:):,, <.:$**.?

This appkcation note deecri~s introductory transmission line characterization, analysis, an~x%ation.
Over the past couple of years, microprocessors and digital 16gic in general have se~-~~wntial
increases in tine drive capabikfy. This increase has fostered the current logic and mi-~@sor
speeds
readily available t~ay. The relatively qui~ rise and fall time of todays digital d~vi@+@akes an understanding of transmission lines and their effects on system rehability a neces+~jj}$

$b?.i~
,}>!,,
....,.,,
\.

TRANSMISSION

t.:.
,,,:.
, .,
$%
+3:
.:*?>
1>
:t+;:
~>..

LINE CHARACTERIZATION

When discussing transmission lines one should reflect on the fol,@.~in@Sefinition. A transmission line is
two or mre conductors separated by some insulating mediurniiw~

carry a signal. At first glance this

seems rather trivial, but upon closer examination one finds

,:, ~
,(,,h$@Of physical nuances which make the
transmission fine a Wphistcated element to describe, %@mch
are:
).:*.

1. Line resistance present in any non-ideal c@,cfor

2. Line conductance ((l/R) = G) present @a~ non-ideal insulating medium resulting in leakage
\,,,,t,:,
:!..,>*%:.?~
currents.
*;*,.?,,>.i:
3. Line inductance present in any curre~car~ing
conductor undergoing a change in magnetic flux.
4, The line capacitance prese$$~e~een the two conductors separated by the insulating medium.
m...
Figure 1 shows the tine underd~~~;fi.
Thecircuitconsists of two series elements (Z + L) and two shunt
,,,$.
.:);,,}*
~?,.
,,:
.,,b,
elements (C+ G).

,,f~,:.,
.$, *::,....

!,

Figure 1. Transmission

Line Circuit

~. t,.,

......
,,,: ~~st,<
:!,..
,s:.\.,
~,.<~.+a
.:~:.l,
,,,,l~~.-....
..>,.
......
::~:<>
>.,;

dur discussion will be primarily concerned wtih C + L, because these elements are the f requency dependent components of the tine (neglecting skin effect). For frequencies above approximately 100 kHz, ~,
the characteristic impedance of the fine, is equal to the square root of UC and is independent of line
length. The propagation constant (t@)or time delay constant is the square root of L*C, and is a function
of tine length, Zo is of particular importance to our diwussion

because when you match this impedance

to the load, you reduce the effects of transmission impafled to both the source and the load.

303

TRANSMISSION
Refletibns onaline

LINE REFLECTIONS
arecaused byamismatch inim~dance

beWeenthe hneand the load. Ifall the

,tt,

~werdehvered to the line is absorbed by the bad then there will be no reflected power back at the source,,,.~:, ~~
side of the line. This principle ofpowerconsewation isthecornerstone ofthis application note. Ref4-v~
to Figure 2astheequations are discussed. Theequation describes theratio ofabsorbed Powe~4@o%&&*'
,,.,*:1,>
.:$V
:$~.[
~,:.*+
fleeted power based on the ratb of line to load impedance.
,~$$.)
.(, ~<1}
~e,
~<
J.~.\2,\-\\$::h{
.:,!, .>,
~~,.,.,,
,\,,?
.,.!
.-,,.

Flgure2.

Transmissi&Line&

~,.~;f,
~~.:...,:,
.~~

A.V**,$,*:$
.:3,>:,,,,1
:l.%..

The current delivered to the load is IL= I,w - l~FL(iri$#nt current minus reflected current), while the load
votiage is, VL=VINC+V~FL (incidentvoltag@plus$eflectedvokage). Weneedtofind anequationthatreIates incidentvokage to reflected vokag%~~~efore notingthat the Ioadcurrent IL= (VF-VRFL)/~ (incident
votiage minus reflected voBage dvid~;~s$e
characteristic impedance) we can see the following rela.$,.,,:?...
.4 .\:.
..,,..
tionship.
#$?,,?%,
VINC + Va,,s.
.JWK~,FL
(1)
ZL $$,!$..
:.$%
*
/;/,$.$\jl
Solving for VltiRFL ~:),)h+$
.,$..
$>>*.,.>,
\,,..,
,, .*.
(2)

304

_-..

...--.

~=z.-&
Zs+z~
Onecanseethatthere

(5)

arethreedistind

~ssiMlities

which require inspedion.

First, thesituation,~$$$~

the load impedance equals the line impedance (ZL= ZO)and p~= O(no refledions - a properly t-d
line); semnd, where the bad impedawe
generating a refletion

is greater than the line impedance (ZL > &) and ~~$dtive,
,,x~,:,
t*
whose polarity matches that of the incident vo~age, and, finall~,~~~the
load

impedance is less than the line impedana

opposite tothatof

theincidentvohage.

and p~ is negative, generating a refledi@~@e

i :.{;.,
.,.,
Le~stake acloser look atthelasttwo@~$+$
.. s:),<f~
,,>,

Assume that Z~=4Z0, and thatthe soume impedance

= loadvoltage

(see Figure 3).

.,J.:,

~lartiy

is

= lineimpedance.

~s~$:source vokage, and V~

\ .~.,,,;i..~
,..*.),, >.2

{.,

Thusatt = Oavoka&~@~:of
l/2(V) (because Z~and Zoformavoltage
divider on V) begins totravel
downthe hneand~$@$at
Z~onet@ orpropagation delay later. Whenthe wave eKounters the bad
impedance mi$~~$a
refleded wave equal in magnitude to (V12T0.6 is reflected back toward the
,.}!.;
.
source, and a$lve$ at the source again one ~ later. This Muses the voltage at the source to rise therefore
a$,~.Y.,s:
creating t~e b~~~s~ overshoot mndfiion.
,f ?..+,<+.
:2$$,,,$!,
SJ~ce?@%ource and hne impedame are matched no further reflections are generated and the line has
~,$,~~ed
~}e**

its steady state mndtion.

See F~ure 4.

2@
Figure 4. Voftage versus Time Plot of 2L = 4 * & and Z~ = &

305

Thenext scenario iswhen~<

~. Forttiscase

assume the following conditions. ZL=~4and

Z~=~

See Figure 5.

?:
Attimet=

Oavotiagewaveequal

inmagnitude tol/2Vbegins~~~~%own

theline arriving atthe load

onedelaytimelater.
Theimpedance mismatch generates a&fle&dwave
equal inmagnitude tothere.:~+b
fleeted wave discussed in the first example, but opposi~~$n ~larity. At time 2t@this wave reaches the
source andsums with theexisting voltage present f{~~,ti@et=O (V/2),
reducing its value to V~/5or
((V/2 )(4.6) +V/2). This istheclassic undershoQ~i@n.
See Figure6.
~8-.

$.

t}

F,@ti~S$Voltage
~.i, i
,,,{it.,+,,~:,.:~
~,, ,$

versus Tfme Plotof

ZL= Z~4and

Z$=Z0

At this poie~,~etied
to reflect on one of the equations described earher. The equation states that VL =
VINc+V&~;;~& can see this holds true asnoted lnthepreceding
examples, where VLand V~ either
incr@~~,~Tdecreased with corres~nding mismatches in impedance.
,, ~t>,

>..,
*t%

~;;.,+,,,.

.$.! .3, >\~

wFkAmlCE
DIAGRAM
\\f}?,.~
i.,,..?s
Ii..,,,,.,.,
The lattice diagram permits a network
$~~

l~{$<i,::
.>,..!,}

to be checked quickly for balance (match). The diagram is essen-

tially a fwo-line graph with corresponding source and load impedance, connected by a reflection diagonal
with a period of 2t@(twice the line delay time). This diagonal is used to represent the ref Iected voltages
magnitude, See Figure 7.

306

..

..

-- .

SOURCE VOLTAGE (VS)

t=2
vti12 = Q Vfill
t=4
vr4=Pvr3

307

LOAD VOLTAGE (V~

p~_ 75-50

pL -3900-50

(VL )

(v, )

7.5 t 50

3900 t 50

=0.974

= -0.739
t=o

vs =2.18V, t = 2

Vs=l.86V,

t=4

vs = 2.ogv, t = 6

Vs=l.93V,

t=8

vs = 2.05V, t = 10
t=ll!vL=l.72v
v~=l.98v,

t=12
t= 13, VL =2.18V

v~ =2.02V, t = 14

tgram for ZS=

,!:

Transmission

7.5Q,

Zo=50Qand

ZL=3.90

~i.

$?hf~pes
~,*?+:*W
*

There are ess~i#ly

two types of transmission lines; the microstrip and the stripline. The microstrip is

shown in.@~re lo. It consists of a mndudorseparated

!!.~<.
.t.
\@&.
3,$,>),
~*
,*:J.t+<:,

from the ground plane on one side by a dieledric,

308

The charaderistic

impedanw

of a one ounce line configured

as a microstrip on &l

GROUND PLANE
DIELECTRIC
GROUND PLANE

309

O fiber glass is:

Loaded Transmission

Line Propagation Delay and Impedance

As stated earlier the unloaded propagation of a microstrip line is:

~

= 1.I07J0.475ER

.>,.,
.!. ..tp:~.
~,+.
.,/..
~t.,li~..$*.
\*-\/\}
.*
,, !.,::,.,~<,,

+ 0.67 ns/ft

This delay increases with capacitive loading. The increase is equal to =

where CDisthe ~~~~~

uted capacitance and COis the intrinsic capacitance of the line. COis obtained from Figure 3~$$

~@#r-

ence 1, or akernatively it can be calculated as CO= ~flo. For the micro strip described abQk:~@~hickness (h) of 0.062 in, and signal trace width of 0.015 in, CO= 15 pf. Assuming this line i$,lo%~ti with five
v
10 pf loads the loaded propagation delay bemmes:
~P,,{<\,:{:$
~

~w
...$>..>.,.
.
,t* .....}
$!;,,*~>
(1.73ns) ~1 + 50/15 = 3/60 ns/ft
t,:$~,,,..:,>,,).
.s:
... ...
~.;.$:-,~.:>:.r.
?.,,
The loaded line impedance 20 = 20/j=
= 116.6/2.08 =<:~~~i
.,,.,.,>+...
.:
For the stripline di=ussed

a~ve there is a corresponding inc&#~&,~$ tp and ~.

.!:,,;>~.

The loaded

delay

,,..:

propagation

20 = 60/~_

b~ = 2,2 ~-,

= 28.8 Q.

It is apparent that capacitive loading increases t@e>opagation

...!.
,.I!:i,,.
pedance.
.*>,.J.l>y.
,,
.~,,
1,,
\ {:,.:~>,>.::t.i.

TRANSMISSION

~:.::!

S\q.57 ns/ft, while the loaded

,,t.?a.+},,.
..~;:
~ b{,>,,>+
- !,.,.
,.,,
, ,.~>\t,
. \.,*
,,r$:pi

impedance

delay of the line while decreasing its im-

LINE TEMM~NATION
.\,.. .i)..,
~.
,,.
~~?l:.,>
\ ~.,!J

No discussion about transmj$~~.[hes

terminate a line. Essenti~ .,i\,,:,
t%k
*:..:.*,.

would be complete without examining the techniques to ProPerlY

are three (3) methods which can be employed. They are:

1) Unterminat@.1~.~kontrolling
,.,..{$*

board parameters to match line and load impedance).

2) Series4tQ[&&~$on.
~,~+,
3) Pa$al#!.~@ination.
w;., i$
<,,,l:?;+,,,+?*+

Untq:$~ated
Line Method
,,:,),
,(\ . !<.
inVOlVe5controlling the
~ ~rnethod

length of the line such that any reflections caused by the load are
,~,.?+,,,
,$,f?...dbso~ed by the rise and fall time, t, and tf of the driving gate. For this method to be effective the propaga~:~t+ tion delay (loaded delay) of the hne must be short relative to t, and t,. This allows the reflected wave to
N*?. . $\.
$:,,.~.*,
sum with the rising or falhng driving gate waveform. If four times the propagation delay of the line is less
~j:
,+:
than or equal to t, or ~, then minimal ringing (overshoot, undershoot) will be observed. Specifications for
t, and t, for various logic families are readily available. Knowing these times one can set the mmimum

310

._.,____

tine length such that the lines &<= tr/4, For distributed bads that are stubbed, the length of the stub
should be set to minimize any ref Iections. At, /~ ratio greater than 8:1 should suffice.

,,,.
,,,.
,.,.,..
....,~>.

Series Termination

...7....

In series termination a resistance is inserted between the driving gate output and the line. The ,~wd
output impedance of the driving gate plus the added series resistance is selected to equal t~.jw
pedance of the line. Sncs the input impedance of the driven gate is much greater than~&.@
ring. Basically this termination configuration will ring once and reach steady state wit&n2~fknd
loading, (lumped loading) is the only method of loading that is recommended for t~~&6f

im-

line will
of tine

termination.

This is because any distributed load on the tine sees a vokage equal to v/2 until ~~$~ state. This condition could violate the valid V(Hor Vl~ specification of these gates. Clearlyi&ibuted
bads are to be
avoided. Receivers at the end of the line will not experience this condiJ~~:&~e
incident voltage and
the reflected voltage add together to equal the load vokage (VL)one.@$$@r the signal is assetied.
,\
~~!:;$.:?
,:,\.~>J
..

~..:.,L}J.
,&j*
...,,.~$
.k,~,
J,>.:..
,?-.<
k.

Parallel Termination

In the parallel termination method two resistors are plac@ a~~~e end of the line. One resistor from the
line to ground, and the other from the fine to VCC. Th&Q~$~l&~combination of these resistors is set to be
equal to the loaded impedance of the fine. For ex~ple?%~
of the fine is equal to 50 Q, then the parallel
*? .
combination of both resistors should equal 50,9$o~.ttis
method of termination requires more drive current. The driver selected must be able to h~n,~et~e additional load placed upon it by the added parallel
load. Also it is apparent that this method~{($~@nation

consumes power even in the steady state, as an

additional current path has been set up beWeen Vw and ground,

i~~,:.$,
..:;.j.~,.
~f \.,,\

Upon wmpleting

the pa,~$@~n

transmission line pro~l~~:i~~r

..:,
.,.
~s:~**.h

for our new project, we begin to peruse our schematics for possible
the purposes of our discussion assume the following configuration:
5 volts
microstrip

ER= 4.7,
Fast ~L

G1 Ofiber glass, 1 oz copper,

0.015, t =0.0015, h = 0.062

configuration,
W =

(dtive and receive side of line)

F241 buffer
2 ns (for 50pf lumped load)
5 ( input capacitance= 5pf/load)
(lIL= 600@, /lH = 100 P)
Distributed loads approximately every 2 in. for a
total trace length of 10 in.

311

Procedure
1. Calculate thehnes characteristic impedance(~).
ZO = sameasexample
described earlier = 116Q.

.141.
,..,+,::~*
.$::,?
<.,,.
,.$., .\\>
,~~~
..,A:\,~t
,,
t@ = 1.017 J0.475ER +0.67 = 1.73 ns/ff
>?>,
.+$.;!,,
.>
y,, \*ii
~~,ap,<t}:
\
3. Calculate thelines intrinsic capactiance (CO).
>*:4.,%tt:\,d*
Co= ~~
expressed as nffi
\.y. .,
~..\\~.:..
?~J&!.:
CO=(l.73n#ft)/116
=15pfM
=1.25 pffin.910 in. =12.5pf
.,{$:
$+>..,.
,,*
4. Calculate the loaded fine impedance (~)
:<$, ..%,.,
*,+,:):,
ZO = 1164-5
= 67 Q
... ~,~.:i
., .$.;..,,$...,
\,.:~:
,,?,.
5. Calculate the fines foaded propagation delay (tpd)
,*,,,
~.Y !
.1>
,~(~<.yt~
#<yy
, ,&b,
tw = 1.734.,+$\~
.~ s
.$.<>.,!.
tx = 3.0 ns/ft = 0.25 ns~n. 10in. = 2.5 ns >> t,/4
,*<,
~,$,,>
,:
y. *
.,,,*.-st,,,.
2. Calculate unloaded propagation delay (t@).

As described earlier, since the baded propagation delay of the lin~ &c~~s t,j4, we will have to terminate
.x,.,
,\\.*
the line. The loads are not lumped at the end of the tine, they a~gstr~uted.
As explained earlier, serieS
termination cannot be used because of the possible thre@ld
v%lations. For this example we will use
parallel termination. The parallel resistor combination ~#,$,&Wosen to match the loaded impedance of
the line. Noting the drive current of the F241, (IoL= ~~~:rn$ IOH= 15 ma), we can set the source current
,~i
.:~?.,
,\,;q,,
resistor equal to:
~$*,,, .
.+

312

1.

.>
2. Solve for the maimum

stub le@h (x)

313

314

External-Sync

By S.K. Tong

Power Supply

with

and K.T. Cheng

.,.,
. #$.
..\>\

ABSTRACT
This

paper

flyback
monitor,
from
izontal

the

power

[n order

to

the switching

matically

are

enables

the power

ranges,

90130

the

ripple

design

and

components
the-art
transistor

monitors

(e.g. 8Q~@~;S}

ing noise

beat w~~~~

emitter

MJE18004

and

to

the

power

become

SUPPIV

critical

sup-

at two

Vat.

power
recently

supply.

input

bulk

o:,~h~~~~~bn.

~&&~:.

MOC81O2

hiw-~f.%tion

?~~~~rformance.
t.l ~,~.:.
3;,

have

with
can

interferences
the

For
are

different

in IBM
from

switching

power

supplv

three

CGA,

EGA

and

three

developed
scanning
as shown
The

HV

Focus
HV

315

Multi-Sync

Color

Monitot

popular

of computer
VGA

display
modes

scanning

fre-

kHz. Hence,

in this

note

frequencies
in Fiqure

specifications

be

displav

and

15,7 kHz to 31.5

monitor,

d.c. outputs.

became

modes

resolutions

to the horizontal
color

monitors

PCs. The

horizontal

ranging

multi-svnc

of Modern

oper-

monitors
or TVS rnst
frequencv.

to several

examples,

used

1. Block Uagram

of

random

switching

are:

YNC

figure

of
and

resolution

because

asynchronous

color

adapt

quencies,

vides

Nowa-

switch-

frequency

the horizontal

the horizontal

multi-sync

synch ronize

monitors

Asynchronous

monitor

the high-definition

they

displays.
modes

has
color

also deflect the electron

beams and blur their
controlled
positions, Thus, the switching
power

because

h~
\*s
..
~,.<!\.i
,.!,+
i/{$t, .$
~:<+;.. ..L..Y
*, .&,\,.\
col$~,.xsp$ay
increases,

bv the

supplies

scanning

undesirable

color

pul~x:+*2rated
atld~<and
,:f~reclselv

higher).

It affects

h[~$-~{iRltion

The
%PPIV for
introduce~$~+~ .~ts~nchronous
The state-~f~;j+i,.q,:
Recently,

opto-isolator

these

the

power

regu-

However,

of high-definition

horizontal

creating

the linear

light weight,

onJ~~l@wlution

?~~~o~

switch.

bipolar

in its features

jitter

input

It can min-

effects

the color

input-voltage

epi-collector

of modern
for

use

power

1. INTRODUCTION
resolution

of the power

of the

on the

horcolor

to operate

requirement
how

of the
of the

universal

supply

utilized.

As the

frequency

to 32 kHz)

r~place

generate@~~~~itching

adverse

(15

in a low-cost

perforated

EM1/RFl

su~~~:s

effic~~~.$~d

interference

stresses

demonstrates

the

to high

can be auto-

Vac or 180-260

current
the

due

screen

a new

adaptor

lators

color

supply

line and load regulations

Also,

90 W

power

the

at the fixed

voltage
imize

a multi-sync

switching

the power

frequency

excellent.

capacitors

of a low-cost

for

minimize

synchronize
The

design

supply

noise,

scanning

monitor.
ply

describes

switching

daya,

HIGHRESOLUTION
MULTISYNCH
COLOROISPLAY

can

the
be

of the

1. It uro-

outputs
+ 110 V 0.7 A

for HV, RGB dtivers

+ 12 V

0.3 A

for auxilary

+5V

0.2 A

for logic

and deflection

Vac or 180260

Vac

performance

power

supplv

clusions

use.

ICS.

Inputs
90-130

4, the

50/60

and

further

are discussed.

include

a aummarv

supplvand

the future

converters

suitable

improvements

of the

In the last section,

the con-

of the design

developments
for

of the power

of switching

multi-svnc

power

monitors.

Hz

Power
90 W with

overload

COnvesiOn

protection

Eticiency

JOO/.at full load

Minimum
Others
External

synchronization

32 kHz) which
modern

color

obtained

bv

with

are regarded
monitors.

d.c. isolation

power

The two

post-regulators

supply

(15 kHz to

standards

for

outputs

are

low-voltage

of the

+ 15 V and

+8

1.It

In Figure

2, the

block

SUPPIV, according
the

input

filter,

rectification

It mainly

circuit,

universal

select

the

order

and

nificantly

reduce

capacitors,

V range,

works

due

to wide

input-voltage
The current

the

range
ripple

circuit

the ttiac

range

universal
onlv

for

more,

the

higher

for the same

output

90

voltage

stresses

Some

adaptor

handle

In section
reviewed,
voltage

2, the
whereas

adaptor

the

is given

design

5.

Good

6.

The.~6J&~pg

controller

the noise

is

immunitv

sens;@~. circ&~t.
roll-QW:~~&Pacteristic

:@&~j,@
t~. ,,,.

if single-ended

g!oss~~gulation.
dutv

i%~aaiikuiarlv

of the

of feedback

power

cOn-

circuits.

[1 1

configuration

is

[1]

cvcle can be greater

important

for multi-sync

than

500/.. This

monitor

power

previous
the ,f:~&

are US,~~$~

:1,High

VCC)

2.

if~e~qs~llv

caRa+~Q?S:#?
.%?+t-~~
.**, ~~,w
fl~ba~
&nverter

o~t%$~~~versal

The

3.

low

2. Block

Oiagram

large

EM1/RFl

Higher

ripple
screen

the

in section

flvback

result in high

and voltage

clamp.

transformer

causes

and flux fringe.

appearing

output

intefierence.

supplv

horizontal

this point

currents

windings

in the

current

greater

the power

input-

switch,

air gap

higher

cause
is

and peak transformer

in power

produces

RM~

losses

cOnverters:,$#u~&&PY

in secti~
3$ Then,
*i,~\*,,.\?,:,,..,
i:
~<1

Figure

Simplified

can sig-

capacitors
(at

of the

the

rectifica{l::h~!j~

of smoothing

design

4.

~FSntmode

It can-~$~$~ve

simplifi@/~$&esign

used.

bridge

power

line.

verter

in

th,ej~~:r
SUPPIY are
h$,\$3,3,
..* ,.
i:i@~@!I$nt
because the cur-

fed to th~

triangular.

The

of

on the power

of VCC.

ripple

value

operation

of the current

a voltage
off and

size and ~tpu$:~hoke.

cost

of the two smoothing

bv simple
W

forms
turns

mode

Single-pole

3.

is continuously

This design

ripples

triac

and

waveform

strictly

VCC in between

the triac

of thesmoothing

amperes

controls

as normal,

the current

without

several

the

Ci, and the switching

transistor(s)
designs

and

d.c. voltage

V range,

circuit

can automatically

rectification

In 180-260

rectification

rent

adaptor

adaptor

range

the rectified

the whole

doubler

input-voltage

Current

the

converter.

input-voltage

to provide

2.

Besides

parts

transformer

densitv

lowered.

power

is shown.

of three

universal

input-voltage

200 to 370 V. In 90-130

fired

of the switching

consists

the

and the 90 W flvback

The

diagram

to the specifications,

has-smaller

power

inputs.

voltage

The switching

is designed

frequencv.

becomes

in output

ripple

The

capacitors
which

in synchronization
adverse

mav

frequency
effect

of
with

due

to

less significant.

of Switched-MOde

Power

SUPPIV

for Multi-Sync

Monitor

316

.-

4.

Transformer

and snubber

netic energy

stored

released.

This

previous

designs.

Wth

the

cross

phenomenon

found

for

90 W

control

effectiveness,

topology

power

are minimized
is employed

and

It is par-

in this power

2.

line ripple

Eliminate

the

continuous
power
3,

possible

mode.

supply

(8 Vo/8VCC

double-pole

This would

under

Discontinuous
current

rejection

some

mode

mode

critical

flyback

operation

instability

power

problem
transistor.

in

suitable

working

Assuming

power

that

duty cycle
the

primary
st.,\

are constant,

(Energy

.%,* a::$~,
,:;. ,,.,$,,~,!: (z)

law)

(Faradays

= conduction
I/fs

70 Y., the

is how to determine

Lp and maximum

= Lplpk/tc

T =

is about

128.8W

and input

wheretc

= O)

= Lplpk2fs/2

VCC

and it is aasumed

efficiency

Pi is,

inductance

inductance

supply

characteristics

cause

power

= 90/0.7

D of the

care-

into account

conversion

the following

primary

converter

through

ia taken

typical

input

Then,

because:

1.inherent

the

total

in the

size,

is selected.

switching

Disadvantages
(see later).

Current-mode

that

core ia completely

can be often

of cost

flyback

suitable

application.
ful design

If the efficiency

ring after the mag-

Pin

considerations

regulations,

ticularly

capacitor

in the magnetic

law}

., ~+
of theswi$~.~~l~~~$

time

= switching

period.

~l:;.*$.:i

of the

conditions.

topology
due

has excellent

to

Isrge

current

amplitude.
4.

Synchronization
greatly

is easier

affecting

to

the converter

implement

The current

without

petiormances

wavefor~~~$~~n

(1) or (2), then the pri~a~y

and cir-

in Figure

IWuctance

4. put ipk intO

is calculated

to be,

cuit configuration.
5.

Simple

andlowcost

trollerl

C is available.

UC3842A3843A,
used

in the

operation

Motorola

power

supply

.Thefeedback

is through

lowest

a new

OF FLYBACK
value

current

mode
the

secondary
Motorola

con-

control
current

IC, is
mode

side to primary
opto-isolator.
.$ lpk%32kHz
= (0,6847)(3.215)
,fJ\L:.>,}
,Y,$}h a~
Dmaxat32kHz
= 0.4/0.6847

TRANSFORMER

of VCC

is assumed

50 V below the rectified low-line

255 V), and the highest value

the fly back converter

current-mode

to perform
from

MOC81O2,

2.2 DESIGN
The

ascommercial

shown

to be 200V,

peak voltsge
is about

in Figure

i.e$,

$For

(180 x 1.d~~{...:

the

mode

f[yback

converter

side oftransformer

3 should

less

than

such

switching

fall

0.416,

and

the

other

relation

VCC=

isslightly

the

remaining

non-ideal

calculate
is used

waslumDedt0

= 32kHzand

to secondary

capacitances,

etc. To

power

discontinuous

0.4, which

because
stray

rise times,

oower(90W)

(3), atfs

aetto

inductances,

inductances,

the outout

in
respect

is used to compensate

time

as leakage

ondary

D =td/Tis

(10,584)

switching

from

= 0.584

operating

at 32 kHz, the duty cycle with

370 V, Ther~~a&
op,er~~

= 2.2A

ities
finite

the

sec-

again.

+11 OVOutDut,
200V,

I
Figure
(Discontinuous

3. Flyback

Converter

Inductor-Current

Figure
Mode)

317

4. Switching

Waveforms

of Flyback

Converter

If

P.

= 90W

where

P.

= VolpkD/2

net output

V. = output
Ipk, = peak
D
Hence,

= td/T

lDk

Then,

voltage

Ls(ll

diode

ratio

Np

two

windings

rectifiers

low-voltage

of turns

cores

The

of turns

are found

Ae is 130.65

10)

=
=

Ns(8)

= 6

where

N~(15)
N~(E)

maximum

172

66,67)/(0.25
Irms

X 124.15)
X 3.215

(0.4/3)1/2

= 1.17 A

into

(4)
inductance)

of L~(llO).

factors

working

for the

Th>yare

EE40

are just around

flux

density

Bmax

cross-sectional

mm2.

tc)l(Bmax

X 130.65)
Ns(l

Ipk. At fs = 15 kHz, Ipk

The turn

to be suitable

For EE40 core, the effective

= (Vcc

taken

= 2.22

core, The spacing

Ns(15)

are

of Lp (primarv

core and ETD39

Np

wire

are,

outputs.

of the fly back transformer,

area

required

in the following.

= 0.273

= (200 X 0.4X

have,

implementation
both.

18.2/66,67

The

listed

= 2.2 PH.

output

= number

is set to 0.25T.

mm2,

are also

= 6.9 WH

[Lp/Ls(l10)]l/2

= number

magnetic

0.4 for

is 124.15

to,
10)

N~(lIo)
Two

of the

for the two

Np/Ns(l

where

of other

O) (9/1 11)2

n is equal

n =

Ae

winding

of +11 OV winding

(16/111)2

drops

consideration

D
Np

we

core,

of each

mH

= Ls(11o)

Ls(8)
The

3),

12.5ps

into (l)or(2),

the inductances

Ls(15)

+11 OV windings

of

to Figure

= inductance
= 0.334

And,

current

= 0.4(referred

lpk,

ETD39

Irms value is equal tO (D/3)1/2

6 A and td = 18.2 W, hence,

+11OV

of

inductor

= 4.1 Aandtd

substitute

Ls(11o)

For

gauges

power

Ae)

(200 X 0,4 X 66.67)/(0.25

163

163/2.22

= 73

ideal ii,&, *~e

sh:,w~ iri~igure

11

u~a?$~~~~.g.
= number
= number

of turns
of turns

of Ls(15),

and

?d~@~

of Ls(E).

transformer
10. To meet
VDE,

should

refer

UL,

CSA,

construction
diagram
is
with the world safety regetc. ) for

to corresponding

the

transformer,

regulation

books

and$f4),
b..

J Ml
+

I ._PT

uC3842A
.,,..

c,L.!,..._lyMiP4N,o

,7A CS13
, U
R,

c >8

.8V

ROOD
.. .

Rgure

5a. Current-Mode

IF

Controller

and Sync

Grcuit

RE

RB1
h

for MTP4N90

(MOSFET)

318

If the output
-1OV

1+

10)

R8

lB

used

output

in the power

supply,

R110

is cascaded

put ripple

voltage.

frequency

for that

2.3 DESIGN
The
tifiers
tifier

OF OUTPUT

following

the values

Controller

(Bipolar

as shown
MU R140

describe

capacitors

in Fiaure

is chosen

how

Two

J@W@f

~$$O.&.AWe

~rk$~volt,:$,+i~,,:,::~...
+>,. -%,.$,
an~J@g\:aW
used

R15

~~t,he.~~?ee

to

outputs.

disaipa~~~~~$.,s
ea~,,o
,,,~ytto~

are ~ow~<#

Iower

the ot-

Ftgure 8, The corner

at + 1 l~t~$,utput

is about

outp~?~~~oltage

6.2 kHz and

is,

(15/6.2)4]1/%,
w> \**,!
OfB84
;> .,\.{\

V (peak-to-peak).

Circuit

ultrafast

output

rec-

recoverv

rec-

for D1 10 due to its fast recovery

power

flyb~~b$o~r

to determine

and to select

3, The

current

Transistor)

CIRCUITS

paragraphs

of output

and Sync

Juntiion

1/[1

currents

maximum

currents

with

They

the approximate

5b. Current-Mode

R1 10,

is set to 5.6 kQ and

LC filter

ripple

by ESR and ESL

two of 22 YFAQ,$~~F

ripple

Their

load

A} is so large

to be connected

caused

AS a result,

maximum

minimum

(1.55

individual

ripple

capacitors.

each with

maintain

current

are needed

to Iowertheir

age is 160 Vdc.

The dummy
resistors

OTHERS ARE SAME AS IN FIGuRE 5(aI

for MJE18004

ripple

capacitors

in order

of the output

is set to lA of VO, i.e. 1 V,

X 18.2/CO(l10)

KF

output

or more

capacitors

a~
Rgure

5.13

= 46.68
the

two

voltage

and the additional

01
MJE18004

*CB

tipple

1 = 0.5x

in parallel
Ic

uC3SA3A

Co(l

that

MOTOROU

However,

1N474QA

Vo

8V0

switches

supply.

theq~~te-o~-the-art
,$htro&wced

They

petiorated

in 1988.

The

are

considered

are TMOS

power

emitter

new

seties

~~
simplifies
the design of driving
$F,J,
:~.,exfremelv fast switching
transitions.

for the
FETs,

bipolar

of Motorola
circuits

and

transistors
TMOS

and provides

These MOSFETS
can
The maximurn~
$operate
in the MHz range. In this power aupplv, although
+ 370, n = 277 V, s@ ??it,:$$
*.,, > y+,%
the switching
frequencv
is relatively
low, it still provides
400 V device is selected,
The average
current of Dl~~~sI..w
several advantages
such as simple drive circuit, less sup0.7A maximum,
D15 and D8areschottky
diodes, MBRIW
plv current for the MOS driver, fast switching
times which
and 1 N5819 respectively,
because schottkv r~tifid$
are
result in less energv
loss at switching
transitions,
and
,+*?lQ*
more suitable for low voltage
outputs,
hence a smaller value of snubber capacitor
Cl (1000 pF)
During td, the output voltage
rises fro~.{i{s
minimum
,,,<~r~..;~t:{
is required,
Since the maximum
drain voltage
of Ml is
>.,
value to its peak,
near 850 V (see later), and the peak drain current is 3,2 A,
time

(75

reverse

ns),

voltage

reliability

and

of this diode

low

cost,

ia 110

MTP4N90
the

is selected

approximate

3.2]2

x 4 =

load.

The

power

5,5 W at f5 =

power

that

To
design

of the flyback

native

for

1000 V, the

is well

power

the maximum

improvement
8JT

collector

is chosen

current

cost and

its lower

performance

is quite

transistors.
which

For

the

face three

dispersion

of device
several

collector
Since

the
much,

speed

emitter
various

from

below

power
(e.g.

which
and control

of BJT affecta

= 250 mA
features

MOSFET.
transistors,
BU508),

and

hFE

hours.

MJE18004
of device

uses,

these
times,

degradaThe

epi-

improve

characteristics.

the device

structures

Its

bipolar

long switching

operating

its

70 kHz

the previous

characteristics,

emitter

the

diffused

problems:

Ml

is 5 A and

in Japan

thousand

technologies

the switching

used

major

tions

after

different

for

two important

loss than

triple

are still widelv

devices

power

the

is above

= 70 ns and tsi = 0.6 ps at IC = 2 A, Ibl

are

the

an alter-

V(8R)CES

for switchers

5 V) [6]. Another

of

familv,

also provides

MJE18004

are excellent

= 200 V and full

below

voltage

continuous

is [(0,4/3)1/2

bv the device.

supply

device.

and VRE(OH)

verv

15 kHz, VCC

switching

breakdown

times

in Ml

perforated-emitter

a new
its

switching

319

the

introduced

[51. Thus,

4 Q rDS(on)

Iosa

dissipation

demonstrate

because

with

can be dissipated

newlv

(tfi

for Ml,

conduction

performance

have evolved.

With

Motorola
ture,

SWITCHMODE

the

panied

speed

Ill, with
RBSOA

by the increased

technology).
emitter
the

and

For

die size (about

the

perforated

is interleaved

emitter

speed

perimeter

switching

emitter
are

this

emitter

structure,

thus,

this

to area

ratio.

That

means

can be fabricated

the

3, a dissipated

Its function
Ml

is to reduce

at turn-off

also called

begins

to the snubber

higher

of VDS,

VDS

so the
be

snubber

Cl

Is product

area

is particularly
transistors.

perforated

certain

emitter

bipolar

can

be chosen

the

fall-time

current

for

of

power

old

transistors,

to be as low

power

transistor

includes

it is difficult to calculate
an optimum
requires
the fall-time
information

oftransistor

be completely

dissipated

on.

in the

However,

h cannot

supply,

oscillation
the energy
charged

to the

previous
function
circuit.

that

operations
onant

This

should
the

Note

ratio

Ml

any standard

text

severe

stray

RI

sea~~n

LP-CI, ~~~~~t
res-

R2 =

ance

19.67

4, a high-voltage

by the

oscillation

transformer.

spike

of leakage

The

betwe;n

t;me

(point

The

because
VCC.

lpk2 f~ is COn -

At 15 kHz, the

pri -

of M7 is limited

then,
=

W)

IS limited

BASE

at20V,
The

to 950 V and R2 = 33.8 k~

AND

The circuit
maximum

EXTERNAL

Iessthan

gate

voltage

configuration

current-sense
R~is 0,93,2

0.5 W, Three

connected

tance.

lC selected
MTP4N90,

SYNC

spikes,

= 0,28 Qwith

corner

is used

in Figure,

voltage

on

pin

3 of

the current
power

sen-

dissipation

W) and one 2.2 Q (1 4 W)

to obtain

the

required

(1 k~ and 470 pF) is added

The

UC3842A

it is operated

is shown
[9]. Hence,

1 Q (14

in parallel

A RC filter

is the
UC3842A

because

(CS)

is 0.9 V (mlnimuml

sing resistor
are

control

For MOSFET,
sufficient

UC3842A

frequency

resis-

to kill

of the filter

the

is 339

kHz.

by

To be able to synchronize

induct-

ener;;

DRIVE

current-mode

to provide

in the

A) in VDS

magnetic
A a;d

point

X 15 k X [1 + 244:(850-370-244)]

Vspk

or UC3843A.

voltage

caused

and the leakage

discharge

and

If the peak voltage

k~ (11,7

must

have

a free-running

the simplification

In Figure

between

CIRCUITS

[2].

caused
the

of Dllo

level

= 754H,

2.5 CONTROL,

during the conduction

time Of Ml. But, due
oscillation
caused by CI, Lp and RI! the resis-

capacitance

into effect

(7)

$~~sl
should have a power dissipation
of 3 W.
~$~ Another
RC snubber
of 180 () and 470 pF used
the junction

to that

loss due to L3 is essentially

frequencies

power

For MJE18004,

charge r~t~~f.a~atturn-on.
So, a standard value of2,4 kQ
is use~,
$~aximum
power dissipation
of R1 is equal
,,.%
to C&VC~~fiax)2
f~(max)/2
= 2.2 W, for complete
dis-

the stray

is si;ilar

(9.95 W). Practical values of 20 kQ (10 W) and 33 k~ (10 W)

are used for MTP4N90
and MJE18004,
respectively.

~q
,. -.
$
~1 L@. z~,;?~;,,
.
?.>

is to damp

result

(850 - 370)2/R2

control
From

sinusoid~+o$$~ktion.

The power

switching

to O.15 mH/2

ratio

supply

On

inductance
which stores energy
L3, and the leakage inductances

side only come

0.5 X 75 p X 3.22

the tww~p~~sing
bq~~~~. *?8
~

above

to 850 V for MTP4N90,

has a~t~~

is s@?~~~#&
undershoot
below VCC
\,\~
:\!
will result.
~-y,..t
~.,\.pr.\
,
Thus,
,, .*S.$$
1 = 0.5 x R1 wJl@$/l,66m)l~2
or R1 = 2,58 kf)
.,.<.\ i-.
In practice, @RaWer
value of R1 will increase the dis-

power

side.

~J@~~~ inductance
was measured
to be 0.15 mH with major
~~cdndary
winding
(110 V output) short-circuited
at zero
.$ bias current.
It is about one-tenth
of LP. So, L3 is equal

on linear control%yst~~s,
.% ~...,) ..
>+
/
= m
2

inductance
~,?~{imary
,$. .:.k;~, *

althop%~$e

$:~~{~ssame

:+. \

c@~&$.@$CI
~$~=tray

that

sa~~a~

turns
power

For a ~~~~~~
cm

nvo)

leakage

A and &?~~{igure4,

should

is often

for the

be considered.
damping

of the damped

If the damping

asswitc,~~
.. :..i,
the mi@Qr :\*&?>
. ...>:!,:.*\ ..,.
7X,* ~~>.
..\\l,.

.~&&%$2
t$~,.$ +!:.
f \{:,,>*,:*
,.,.,.-..-~>,w
~ ,,.;,>
~i,~ ,,,$
~!.~~,~
,,\>!
= ts,p$~

law,

Vcc

shown in [3],$b@~$~@ge
to bedissip~~@s~erely

value of CI
without
the

because

between

the envelope

ratio

Faradays

= (Vspk

$*,
Of .tkyk$$,,

v01ta9e

inductance
neglecting

t$pk
fs lpk/2

nvo

), 1/fs

L3

the spike voltage.

an input

pF. As

flyback

the resistor

with

RC snubber,

substitution,

capacitance

the switch

phenomenon

Therefore,

a compromise

circuit,

Damping

that

it acts as a damper

Then,

converter

slow

is caused
by Lp and Cl occurs when
in the magnetic
core is completely
dis-

loads.

designs.

have

for the leak-

in data

in Cl at turn-off

in RI when

of

of power

the snubber

output

discontinuous-mode

always

which
stored

stored

clamp

con-

effect

source

in limiting

nVO and the leakage

From power relation,

of the shunt

in the sewnd~ay

the charge

snubber

approximately
ing inductor.

where

of COs~ [21,[31.

Theoretically,

the

FETs and

given

[Co~s),
which

shunt

Ipk L3/(Vspk

This

and

as 1000

sheets

effect

theeffect

the

3, a voltage

a boost

and from

rate

cross-over

of TMOS

flyback

current,

can be the dominant

in Figure

L3 lpk2 fs/2

switch

value.

than

for C2R2

D1 within

(during

inductor

It is actually

effect

It is

the increasing

the advents

capacitance

the power

acceptable

important

With

of VDS.

the diode

down

to

is shown,

turns off, the induc-

from

through

slows

limited

slope

Ml

discontinuous-mode

peak

inductance

effective

loss of the transistor

When

the

ageinductance
limits the spike voltage
to a designated
value, VsPk, In [3], it points out that voltage clamp is more

and lowers

snubber

rising

to commutate

capacitor
capacitor

bipolar

the

limiter.

tfi. The snubber

can

the power

by limiting

the dV/dt

tor current

time)

RC turn-off

greater

loss. As shown

increases

the cost.
In Figure

Since

has

leakage

in a smaller

frequencies

period.

vefier

125A of standard

base,

the operating

strucaccom-

by the

transistor

die size. It improves

hollow

improvements

is

lation

in

instead

B represents

of the design

in UC3842A,
of a resistor

12 in UC3842A

externally,
frequency

a constant
RT. Since

provides

the power
below

and operation
current

source

the internal

a discharging

For

of the oscil11 is used

current

current

supply

15 kHz,

source

of 8,4 mA,

320

-.._

.- -... . ..-------

the dead
mined

time

t2 and switching

frequencv

immunitv

can be deter-

error

as follows.

and

amplifier

stability.

Since

is from

1.4 (two

+ 0.3 x 3) typicaliv
ll=CT;

and12ll=

(I2 > 11)

CT;

voltage

the

[9], and

of EA), the voltage

output

diode

voltage

drops)

Ve is equal

Ve across

of the

to 4.1 V (1.4

to (5 output

ROPE is from

0.9 to

3.6 V.
12-ll_tl

In the

(9)

11

-~

T=tl+t2=l/f~

tl

gular

voltage

of the oscillator

is 1,6 V. The time

and t2 are the rise and fall times

waveforms

inductance,

(VCT),

other

Due

parasitic

32 kHz, the dead time

running

frequencv

to

the

effect

and snubber

t2 ia set to 6-8

is assumed

of the trianof

ps. Then,

or 11 = 0.756

mA

The

current

constant

= 0,036

source

PNP transistor

about

200, The current

RE =
and

IB3

Since

voltage
llq

1.32

= 0,756

VB3

5 x RB2/(RBI
RB11RB2
RB1
The

= 0.91,

ulated

= 3.3 V,

sw-itching

frequency

source

current

circuit

when

the

single

high pulse

falling
the

50-100

di~~(l

built

edge

its,:$&@W40n
. {..e\
t::~*,#, ?

a$Q&ti@t:U2.

of ~ri8\~$$ise
r@istor

tim<~~$+~~rmal

The mini~~~
vo~?age drop
2,8 Oi#\~,$<$7 V because
2.8 V~~,~$

r~spect

Then,
capacitor

RC at a verv

fast

value

~:a%

are

rate

The feedback
with

scheme
comparator

5
1.2 to

of the input

Varv the

is selected
(linear

detects

and amplifies

the error

of

opto-coupler

MOC81O2.

the

amplifier

(EA) in UC3842A

on

com-

Besides

the

Pulse

as follows.
error

signal,
The

and drives
gain

The

is set to unitv

than

one

for better

noise

connection

minimized

above

the

capacitance,

of capacitively

coupled

is selected

due

For the

of If is chosen

bv

has

been

coplanar

die
end-

other.

The

which

minimizes

noise

that

result

is a
the

is injected

to its moderate

[11], Then,

lowest

CTR

bv

(from

two extreme

If delivered

to be 0.5 to 20 mA.
If ranga,

ia 3,6

V/O.5

value,

i.e. 0,5 approximately.

current

x 20 mA)

is considered,

mA

The

cases

bv TL431,

it

For the highest

i.e. 20 mA, the value

360

Q, if CTR
Then,

Of ROPE

is at the

nearly

by the design

an

Since
values

the value

point

is directlv

of theoutputcapacitora

be placed
forms

Thus,

= 35o Q. A 330 Q resistor

feedback

terminal

lowest

whole

with

ranges

ROPE eqUal

before

the

additional

the internal
of Rx and

output

taken

double-pole
reference

Rv Ithe

voltage
voltage

Rv)

321

= 2.5 or Rx/Rv

= 43

in practice.

from

CO(IIO).
LC filter

to be
LED

of ROPD is (8 1)

is used

the

positive

This point
because

in the

the

feedback

of TL431
divider)

be Rx = 142 kfl and Rv = 3.3 k{) because,

the LED
error

specifiof CTR.

the noise susceptibility

pF coupled

considered.

V/20

TL431

of the

Their
values

to 3600,
The practical value for ROPE is selected
3g0 Q, For the determination
of ROpD, the maximum

A voltage

amplifier)

base

minimizing

of CTR and If are covered

voltcapac-

PeriOd.

features.
window

should
provide
sufficient
coupled
current
to develop
a
minimum
voltage of O.9Von
ROPE. The operating
current

,.~~a~$-parallel
BE diode, 1N4148 is to prevent the BE junc,:;,
t~n from possible avalanche
breakdown
if the amplitude
ark :* :{f Vsync ia above 5 V,
.F;
*! ~:,
it is also possible to combine
the avnc circuit into the
$~,+**, -.~~
.i$ ,.s .,.
constant
current source bv injecting
the svnc signal into
,ii~:
.
the base of the current source transistor.
reference

supplies,

puts the LEO and phototransistor

limit of the selected

[9], and the saturation

can

internal

0.73 to 1.17 at IF = 10 mA)

= 47 Q

0.5 V. The choices

%fid BE resistance

not onlv

but high petiormance

power

ia further

which

MOC8102

of RC

on RC ia approximately
VCT swings between

to ground

about

that low deg-

achievable

the optoiaolator.

once

timing

devices

rather

range
X 0,756)

&s@!~OO

a~% @~.d~fs

simple

in future

effectively

0.2

amount

one-

can be calculat,e@e~~:t,$(5 -26-

reliable,

Noise

to-end,
mere

appears.

The

each

placement,

CUrrent

It is active

rate).

Also,

problem.

reg-

bv the

of 2to$ps~#rgesthe
chargi#~

steps
burn-in

basia, but also that fliers

can be
ore,
the opto-isolator
can be regarded

be used

eliminated,

N5294),

short-c~,:$,&Jt:d$.The

is actually

@e diWctly

synch ronization.,~s~~~~&ed

ttiggeting

CT through

is aro~~d

11 can

regulating

gate-source

current

external

(about

aa &~@W-cost,

atreas

This means

are now

output

~}~,o degradation
of CTR, the new MOC81OX
series opto~,,,
are specifically
designed for switching
power
,., ~~$ohplerthat

at pinch-off.

The

degradation.

to

to improve

processing

of accelerated

ch%~xristica

light

efforts

LEO wafer

5*Nburs

sho~e~ro

~one~~to

their

is
In

decided

of LED

c-~rated

cert~i$n~~i~@l

an aver~$~~~~an)
elirn;@~,@~,%eref

= 3.3

current

bv Motorola
output

They

provides
two additional
+) ai... ...~supplies
for RF and CT are 1.2 k~ and 3*@;#*
cations include tightlv controlled

constant

unu-

Operation

industrv-x$~pblem

control

testing

RB2 = 39 kQ

is a JFET with

[DSS

ahot

+ RB2)

valuea

The

replaced
which

and

the

Optoe~e$~~@~~s

and eventuailv,

have,

eventually

Motorolas

resolve

= 4 @,
0.7

free-running

13 kHz.

we

that

~~~ot

degradations

degradation.

is Set to 4 V since

is 3 V. Then,

@~al@

1000 houw

degradation
at If = 25 mA.

RB1 and RB2 is aaaumed

voltage

conditions

after

in:~{W-

Past in#ti~$<@LED

$&?~@

is

using

of 2N3906

of the .LED

increase

the diode.

accelerated

mechanisms

radation

= 20 k~ and

andthe

gain

to the

within

under

current
for CTR

approximately
50A. A tvpical p~,e.pge
40/. after 105 hours normal ~~er~n
1987,

= 0,515

Dractical

data

from
cause

sual. Of even more concern was the~$~~~~fi.a? the population also contained
fliers units:~~,~q,~~
infant mor-

if the free-

kO

m~200

= 5

recombination

suffered
The main

in efficiency

due

burn-in

YF

current

through

of VCT

charge

reduction

opto-coupler

a 150/. to 20% degradation

11 is implemented

Q3. The

to be 20 x IB3, and the emitter

the peak

the

tality

and CT

is the

within

at fs =

to be 12.5 kHz, tlfl

have

(CTR) degradation.

leakage

circuits

12 11 0.91
=
11
1 0.91

single

opto-couplers

ratio

degradation

The hysteresis
periods

past

transfer

must
filter
loop.

ia 2.5 V, the

are chosen
110 Rv~(Rx

to
+

The

gate

drive

to minimize

the base drive

in

current-mode
minimize

are

the power

imum

to

hFE value
=

0.32

base current
voltage

= (10
base

0.95

drive

(2wCBRB)

capacitor

< fs(mi)/2,
of MJE18004will

because

the

9 V. Other
[7] (e.g.,

optimum
how

to use base

As shown

in Figures

3 and
The

the transformer
diode

ulate

the

MTP4N90,
current
18 turns

is about

an extra

smoothing

an unobvious

effect

transients

for

BE

power
rating

IC1.
Thus,

the turn-

control

and

a kic~~ ~f$ r,~r

Application

kicti

to be

,$$but not

k@n&@r.

s}~{er

RA is :4

Figure

6(a)

power

supply.

reference

conducts

small

FEEDBACK

The

supply,

is the

feedback

system

which

signal,

If

UC3842A 38(3A
.INTERNAL DIVIDER Q

f,
e

OPE
= 390 OHMS

-13
D

EA

= ,63,1
,WD

-Swp)

Power

Supply

&

Figure

6a. Approximate

d.c. and Low-Frequency

322

a,c. Model

of the

Flyback

the last
loop.

of the flyback
is the

The H-block

CTR
= %IAPPR OX,

= 303mS

1 -

short

internal

is 2.5 V = 10/., and

OPTOCOUPLER

be turned

power

FORWARD GAIN BLOCK (G)

1 ROOD

will

for a very

diagram

of the

in the TL431,

to the feedback

the

the auxiliary
at the emitter

all thecomponentvalues

to design

input

the

LOOP

the gain-block

voltage

Then,

power,

of almost
piece

is
can

is connected

quickly.

and the transistor

very

emitter

CA above

is fed from
d.c, voltage

transistor

shows

is compared

Sin~:t$$/$ur-

charges

for the flyback

least

Its

its collector

of J~~,842&3843A

the~$jc~

tha

~~~<~he

re~~~~orforthe

re~k~~~nd

kick:

threshold

voltage

t~,,~$~b}
supply

When

the

d~configurations

is on the s$~{~~

circuitry.

+ Vcc.

+$s.:+:sri~
~~~$~

of CAtO

time

using

~~~,i.

ar~rt~~:~~

a NPN high volta~~~~wa~$$ower

of the

an~~sissipates
> ~.::$.:$
.,?~.~.,:+,.
2.6 @o~G
THE
~1>..
~+:*+,>.,:
,3r~~:<Metermination

CA is for filtering+bufi~<~

of its capacitance

to

mains,

small

and the rel-

izethevalue

Motorola

the

time

DA and RA, A

NA is chosen

start

of the kick~r%q~&&},

is used to regUC3842A

be tied
to a.c.

larger

volta~<h.
from a
~,13e,\
,,, deffved
by a resistor cy,h~~~~
across + Vcc

as output

off. Thus,

cir-

is delivered

5 V for regulation,

capacitor

of the primary

For

of

much
to the

reference

biased

base

The
is due

respectively.

as a simple

The

the

can be fed from

of UC3843A

power for the con~~~~~~rcb!try

windings
(NA), w~!ch~;$esthe

in

control

diode
the

start-up

is about

is 20 V and the total supply

20 to 50 mA.

to provide

the

connected

oper-

above

NA and RA for MJE18004

to minim
long

switching

of CA are 3,3 pF for

case

voltage

is actually

circuit,

values

hys-

of CA must

be kept

the power

latter

10 Q (1 W)

initial

must

UC3843A.

current.

in p~evious

regarded

1/

can be found

NA through

voltage

voltage

to 47 Q, Th

and

breakdown

primary

required

voltage

the supply

zener

that

to improve

5, the

winding

of appropriate
supply

R8 is,

circuits

inductor

control

(1.2 W)
by

base

starter

transistor
V

transistor).

is self-supplied.

zener

current

be determined

in the

to avoid

described
kick

for

of the supply
and

KF and

before

#F

It is also possible

output
is0,95

2200

used

large

13 turns

IB is

is used prac-

of BE junction

drive

atively

at full

source

1 MF. Note

voltage

base

of power

from

for

control

current

not have avalanche

off operation
cuitry

resistor

CB can

breakdown

hykterasis

of MJE18004

= 20 Q

i.e. C8

junction

value

2 V at an output

of base

2)/0,35

capacitance

the

voltage

The practical

and

functions

saturation

on the

of VBE(~at)

the value
-

UC3842A

that the min-

of 0.35A

drop

is about

And the value

[61. Therefore,

transistor

point,

the transformer.

hFE

base

threshold

has a voltage

the capacitance

to maintain
supply

which

of the tight

minimum

lower

has

UC3842N3843A
lockout,

i.e. the

and mean-

Other

typical

because

the

larger
[91, the

circuit,

enough

ations,

to

controller

in under-volt

be large

of the

10 V in order

it is assumed

is 10 partly

of UC3843A

of 0,35A.

to

voltage.

The

A to maintain

From

transistor

The

UC3842A.

voltage

of UC3842A,

of supply

Then,

load. A slightly
tically.

instead

rent-mode
teresis

as that for MOSFET,

supply

loss in base drive

is 14 [6], and thus,

in manufacture.
3.2/10

The

10 Q resistor

But for MJE18004,

is lowered

is used

ON threshold

MJE18004

Rg

5(b).

controller

identical

of a seties

problem.

is not as simple

Figure

UC3843A

a lower

consists
ring

circuit

It is shown

while,

circuit

the gate

is purely

a voltage
value

divider

formed

in this block

difference

or error

amplifier

signal

in TL431,

compensation
grating

+ 3.3)

is then

which

by the

error

externally.

The

to consist

Cf and a resistor

Rs = 0.28 Q

Rf. Thus,

we

IGOI = 229 or 47.2

have,

tranaform

operator

effects
1,

(jw for sinusoidal

analysia),

The capacitance
all stability
gain

G is known

The

which

supply

with

V.

capacitance

[1],

RL(max)

The

at no

MTP4N90),

The

the

of a d.c.
network

at l;(floRL),

where

Co ia the

maximum

load

resistance

32

kHz and

current

was

VCC

measured

measure-

TO simulate

200

sisting

V (for

x 0.06)

equivalent

total

MTP4N90),

the capacitances

lumped

+ 110 V output,

Co

= [(11OV)

Hence,

the

power

supply

lowest

of the output
transfer

(66 PF)

capacitors

The
vidual

Wp

2tip

gain

13.8

elemental

blocks
converts

current

internal

one-third

dat~ shael~,

g%n

.;WO
..,, $&t&*$$
Ipk
.> ::,
mi~~ Thus,
4$;L*\t, .
~,:v:.i;
~>~
Go
\,,
.-

all equivalent

can-determine

dynamic

a.c, oains

the

performance.

val~e
Since

of the converter

of Cf for
there

optimum

ia merely

one

are the resis-

the resistor

ROPE which

current

of UC3842A;3843A
configuration
R~ which

of !he

power

intO

the non-linear
Ic, the

(the

f(k)

minus

1000

of the OP amp),
relates

stage

VC to

which

Ipk,

includes

stage

can be

1pk2 s

=
v

- (RODE/RODD)
=

~cTR)

3 R~
of d.c. gain

by substituting
32 kHz and
known

(14)

CfRf)

~Lp RL fs
;~

The value

when

The

is,

fly&*~*$

of TL431

the sign$~~Ql@The
d.c. gain of the power
dire$t~+@~i}$d
from the power reiatiOn.
$FY$&~LP

amplifier

the G:.@/ock #as a

coupled

sen%~%esistor

final ly<t$e

(8 V}

of the

th$,&&@@$oltage

d~~da~

<,
hf~uit
\-. - *\,,
.::

con-

6(a).

2.2 Hz. If the ESR and E#~

a voltage$~~.$~~~the

current

and

fp

in Figur&.~t@~.*They

sign is due to t,Q~~.e+&hg

the

pF)

:<~~ufitwe

block

in Figure

11 Ho).

~%e~knowina

are

relation,

for the,.~~~~,@OC8102,

CTR (0.65 to 4.5 from

(for

circuits

an additional

is placed

~~,
,: ~+t.
~$}$
(11)
i\iy?;+.:.
\ ~~.~
,. ... .
rad s.
,!F..> .\~
, *V5
G is s~$~~~d
into its indi-

block

tor ROPD which

generates

by charge

are neglected,

Wp)

forward

the diode

output

frequency

[1] as,

Go(l-s

where

capacitance

at three

+ (15 V) (330

corner

= ,,1~~~,

whe@~,~=w#(l

output

is approximately

function

G=

= 1 kQ

and

the @Q~~~~&~}[13J,

of 9/(1 11 *~V8hly

zero freque~:of
the error
.,/! ,:..,,,/:

to be 0,06A

wf

to

=~~57;

discontinuous.
consists

by experimental

= 1102(200

F@r the

st.,\
bl~~~~:

for over-

and thus,
RL(max)

additional

i.e. 9/(1 11 Hat

and RL is the total load resistance

f~

input

for

converter

equivalent

load,

by the
A-block,

the forward

roll-off

is approximated

ments

when

a single-pole

has a pole frequency

with

condition.

model

flyback

enhancement
in parallel

k~,

once

the worst
a,c.

cascaded

total output
at

under

current-injected
block

= 3,23

of Cf can be determined

power

low-frequency

mode
gain

+ Ry)

value

of the

are:

loop-gain
netted

Rf = RxRy/(Rx

dB

It is observed
that a local feedback
occurs in the TL431
outDut circuit and the LED of the ooto-couoler.
Ita end

(lo)

s ~ ~aplace

= 330 Q CTR = 1 (for MOC81O2)

Lp = 1,66 mH,

we have,

of an inte-

sC<Rf

where

ROPE = 390 Q ROPD

= Ho. The

amplified

is chosen

the gain

Ry, thus

= 0.0227

is compensated

network

capacitor

A=L

by Rx and

is 3,3/(142

the

~RL LP fs
\/

Go can be determined

parameters

under

RL = 1 k~ (including
value

of

(12)

Go

is highest.

worst
-8V

analytically
case,

and
On

+15

i.e. f~

V rails),

substituting

figure

the

6b, Bode

Plot

of the

fs = 32 kHz and

parameters,

323

Flyback

Converter

No Load

at

H,

parameter
mum

that

can

condition

we set the

minimum

guarantee
placed

the

30/45

Wf

= 4.64

gain

initial

phase

Cf =

loop

loop

That
beyond

only

one

opti-

can

be satisfied,

gain

to

means
Wp

diode 1 N5953A
(1 W) is connected
put rail. If abnormally
high voltage

12W to

Wf should

aPPears
On this
form a permanent

be

such

or,

can

Wp

down

slope

of the

is 450/decade
shifi

of

phase

and

of the flyback

the

PI controller

con-

(64)]

power

supply
in

Since

has an

1.355

at fs

Figure

for

the

overall

= 32 kHz and

6(b),

with

the

Ho

= 229

Wp

(f)

3.

Phase
The

= 0.2 Ioglo

(f)

= Arg[A

unity

phase

gain

margin

the

phase

the

gain

transient

[f) x G x Ho]

(f) x G XI

bandwidth
is about

plot

than

HOI

62.

40 Hz (at fT) and

But,

the

value

of

O dB.

dominant

value

128 at wf,

It determine$

the

nearly

Under

all

responses.

anode

OPTIONS
normal

should
not
itor circuits

circumstances,

the

output

example

can

appears),

an optional

figure

mow:
.@i&N:Y~

Gate

wi\#h

the

ci:+c&

L@,

an~<,.vlh~f~;

prote@{~i~~&&n-

output(s).:%wwn

or

UC38~~~43A
~;.+t>~
,$$,~.
st:~{~,~t~~nsients,

to th~.kumt-mode

by a
a soft-

controller.

td~rovide

a rather

d.c. #$~#&.?oltage

narrower

at either

low

bridge
range

or high

of

range

Vac or 160-260
has been shown

Vat. A simin Figure 2,

and @Qe~&tied
circuits
are shown
in Figure
TQe~~olt~ge
range
selection
is performed

7(a) and (b).

by an over-

while
The

and

simple

bridge

rectified

out~ut

the

adaptor

is

supplied

from

At low range, the triac is fired CO.

and a voltage
doubler
informed,
rectification
voltage

is retained
(Vcc)

range

at hiqh

range.

is fro-m

200-to

370 Vdc.

high-vohagsz,~~~r
,, -~:!;-\,?,*,

7a, Negative

voltages

rectifiers
circuit

pr~~tp~~>$?

be fou~@&:tg~
i:. \~.7+
.$~ ,J,:i*~t,,

~~@e
PumP circuit.
tenuously
bytheadaptor,

voltage

exceed
150 V, But, as protection
for the
[it would
generate
X-raV if extremely

voltage

control

be added

:{o~,~~detector
2.7 OTHER

Short

of input v~lta~~.,
i.e. 90-130
plified
~;$~~it block diagram

where

low

with

maxi~~*+l~&

can be fed to the

and
mav

operated

over-power

UNIVERSAL lNPU~,V@TAGE ADAPTOR

~>,c<
,.,$::j.t,.>
adaptor
is used with
The universal i~~t.~@~tage

rectified

in

to the

power

protection.

are short-circuited,

optocoupler.

circuit

is

circuit

in the

overload)

in secondary

signal(s)

rectification

is about

is its lowest

is greater
load

I A

installed

be required
just

the output

13.8

= 0.0227

Gain

Then,

will be zapped
to
better OVP circuits

converter

outputs

the 110 V outcontinuously

O V shutdown

is inherently

to be damaged.

best

Typical

may

it

be delivered

To improve

Go

cost.

power

will

erally

unit

control,

currents.

start

1 + Swp

flyback

diode
Other
and

(not

power

+ 3,57

where

Go

which

output

are likely

mini-

following

jw

option

across
(>150V)

circuit

short-circuit

if the

Motorola

206.4

L+~=
sCfRf

higher

the

foldback

A(f)=

with

But,

#F

Plots

rail, the zener

short-circuit.

crowbar

current-mode

equations.

G=

is

tected,

of 1,5 #F is used.

is shown

be used

supply

90. Then,

k) (3.57)

as SCR

Another

= 64 rad/s

of the

load

of the

stability.

value

gain

mum

phase

phase)

decade

1/[(3.23

A practical

or

relative

Wp

the

verter

i.e. Cf, and

gain

= 0.667

100.667

because

be varied,

(either

[Triac)

Current

Preferred

324

_ .

3.1 ADVANTAGES

OF USING

INPUT-VOLTAGE
Three

,.,:$~i?$$.the
worst
case, VCC(pk)
= 180 x 1.414 = 255 V,
..!:
,?
,?,,JVCC(min)
= 200 V, Pin = 128.6 Wandfi
= 50 Hz since
...,.,, i. ,,.
voltage
of the flvback power supply
the Iowf est worting

UNIVERSAL

ADAPTOR

advantages

are

gained

by using

the

input-voltage
adaptor.
They are:
1, smaller
ripple current
in the smoothing

univWj~*3

bulk

is 200 V, and thlefrequencV

,.,.
ca@aci-

ofinput

voltage

is from

Cin
The

= 205.6~F
time

rectifiers,

period
is given

ta,

the

conduction

time

of the

bv,

CC(pkl

3.2 DETAILS
To

OF CIRCUIT,.R~/G~

select

capacitors
Sketches

a suitab~e/~apW~ance
Cin,

for

the:$p~~~ltage

of voltag~$~~

current

the

input

bulk

at VCC

is considered.

ripples

are

shown

in

Figure 7(c) an~i~~~~r

the following
analvais.
Figure 7(c)
is for norm~l
br~~~$rectification,
while Figure 7(d) is for
voltage da~~.kr.
For ai~~~~,~?dge
is re@~J~~~he
$,{,..,
po~:fi$~, e~%tion.
x~~+ +

rectification,
capacitance
It appliea

the ripple

vOltage

of Cin aa follows,
provided

that

8VCC

from

ta is much

the
less

lcaplpkl

(16)
.&

,,

,~

lAC ONLY
and
where

~VCC

= Vcc(pk)

Vcc(pk)

VCC(min)

peak

voltage

voltage

VCC(min)
fin

1.414X

inPUt

(rms),

= 10westvOltage

= frequency

at VCC

of input

50Hz

to 60 Hz. There ?fore,

at VCC,
voltage.

Figure

325

7c. Waveforms

of Bridge

Rectification

bridge

capacitance

is at its minimum,

but

the

voltage

on the

other capacitor
is at half way between
peak and minimum
voltages,
VC(Pk) and VC(min)
respectively.
The value Of
VC(min)

can redetermined

as follows.

vCCiokl

~
l
<

Vcc

Vc(,k)
,---

/
t

[
,;-

(,1

--(;2-

v~>

Icap

(AC ONLY]

.- ..y,~,>

.m.

> =

,<2,.tyq

figure

7d. Waveforms

of Voltage

Doubler

~dm

ta=

1.86A

$+tA$,J@
POwer SUPPIY is designed
~$~~~ranges,
the latter case defines
,~

cOa-l
[=1

ripple

tiveness

current.

In order

of theuniversal

current

and voltage

to be, with

Ci

to operate
at both
the relevant
maxi-

to demonstrate

input-voltage

assuming

=330

YF, Vin

[1272

nodoubler
=90

the

adaptor,

effec-

theripple

precalculated

Vacand

Pin=

128.6W

at 50 Hz,

2T$

Vcc(min)
6VCC

127

23.3

for high
ta

128.6/160 X165 P)11;2

=

103.7

= 23.3V

V(comparedwith55V

range)

= 4.4 ms

!Cap(pk)
= 6.5
Icap(rms)
Such

a large

switching
mance,

= 3 A (nearlY dOuble
voltage
doubler).
ripple

voltage

transistor

at VCC will

and will

especially

the

Of the value

with

greatly

degrade

stress

the overall

conversion

the

perfor-

efficiency

and

to be 1 N5398,

a 1.5A

regulation,
The bridge
device
**$>,

.\.*>391.47

0.9 x 128.6/90
A

q,

a,:,~:~~~that

the

with

a.c.

component

contributed

by the

Ci,

powered
up

angular

t.,,$$xwftching
:~additional

cn,

With

frequency)
small-valued

Vcc.

thetwocapacitors
voltage.

Note

at instantaneous

to Figure

7(d),

forthe

arealternatively
that

whenever
minimum

voltage

charged

the rectified
VCC(mi),

doubler,

thevoltage

average

line input

1 W resistors,

to discharge

the input

instead

controller
the

start-up

of the

It provides
control

and drive
circuitry

current

circuit

is still

is limited

capacitor

(the

starting

to

is

in parallel

is connected

ground

the

current

after

to start-

inverted

tri-

current

for

at initial

power-

not

the

self-supplied.

approximately

2 to

4.6 mA,

to peak iine

voltage

end

sign).

when

The

reference

are selected

highest

= 1,3 A. The two

are used

current-mode

ripple current
is filtered
by the
capacitor
(0.1 #F) connected

the

off, Note that one of them

at one

~~~~~fid
OPeration
of the fly back converter
is negligible.
~,,y~~~~ assumption
holds
because
the high-frequency

across

rectifiers

because

The

VCC is

level

of one

room

inrush

input

current

by the thermistor
temperature

and

which

is limited

to an acceptable

has a resistance

1 Q after

heated

of 5 fi al

up,

326

,_

_-._.

MAC229A8
universal

haa been

found

input-voltage

suitable

adpator

for the triac

becauae

To

in the

first

points:
1.

It is a sensitive
mum

gate

foroperation

gate current
sipation

device

requirement

Its breakdown

3.

ensures

lower

25 V/Ks,

rate

the accurate

exceeds

loss in the

drop

across

MT1

MC3423

and

ia originally
circuit,

MT2

due

for overvolt

applicable

because

of

the

terminal

of the

at which

thetriac

the divider

of

2,6

input

ratio

internal

universal
working

in better

Thus,

in Figure

7(a)

to

if the trip

off is set to 135 Vacor

one
point

At high line, high

191 Vdc,

value

is,

sumption

before

noise

and resistor

10 Q resist:~~~$

R2)

R2 = 30 k~.
current

tripping

source

and

connected

start-up

values

has a time

is connected,
SCR

the charge

pump,

delay

[15].

the supply

Hence,

voltage

stable

the sensing

can operate

If ?%%power

supply

is started

260 Vat,

during

the falling

a resistlV~

and th$@/,~Will

.Ci~%~O &

and:$

s~af?~signal

input,j@f.~~,@23

are

to6.2
diode
also

-400 I
0

Figure

7e. Worst

Case

to limit

the

Consideration

for

IN4148

for

added

MT1

bulk

input

adap-

at high line, high range,

edge

of input

instant.

MC3423

the next

will

This

capacitor,

but

(VCC)

Therefore,

without
yet

not

cycle,

double

effect

not only

produces

for the flyback

to serve

of triacfor

nearly

and

Then,

to nearly

CG is connected

of the triac

turn-on

voltage,

harmful
also

voltage,
tO abOut
operation

had

positive

be stressed

rating.

voltage

terminal

1. to delavthe

two

abnor-

converter,
to the gate

purposes:

a quarter

of one

cvcle.
2. to increase

the

(> 200 V/~s)

in the

I
T

1
T

T
3T

for the

until

voltage

the

high

in a small
snd

+ 0.7 =

capacitor

inrush

input-voltage

cycle of input
since

bulk capacitor

damages
really

CG,

not be tripped

of its normal

remain

negative

capacitor

the lower

V Of supply

filtering

gate

would

is used to clamp

of the cr?wktifenser

in the remaining
the

,~}.

@;~W~?~40

a 6.2 V zen&<~$~@

operation

divider
the

frOm

of the universal

.,@.

of appt~f:?y?.

with,~~(ma~)

.>~.t! ,,.

MC3423

C is used

con-

ng.

$pn~ the SUPPIY v01ta9e Of MC3423

is char9ed
are 50 nF a~.l~,<~~,i~ V, the triac will be turned on for the doubler

through

(MCR102

Off.

with

power

= 0.37 W. The

of the capacitor.~

at pin 4 to ground
which

tran -

0.8 A). When the input voltage

is detected.,t~<be
above
the trip point, the SCR is fired to shunt al?y~:~coming
from

serl~s

:,hi~~p*-Wti

maximum

is 7 x 0.053

,,,:~~~=

SCR. It results

controlled

The practical

to a small-power

clamping

the

prin~~le,

tor, ~@i&@~&ant
point which has not yet been considered
is t~.~azard
of severe overvoltage
at VCC during start-

(pin 4) can provide

the crowbar

immunity

650 Ks. The output

7 V for

less than

balance

= 370 V and the maximum

The

of the ~~}~~aptor

So fap%~&esign
and

560 kQ, respectively,

voltage

voltage

.*;$

r@@,,~p

of I is 53 mAi~~.~?~z,

current

of the adaptor.

The

is much

charge

and is

the

internal

is connected

is turned

constant

delay

current

circuit

(Vp), from

that

crowbar

= 72,5

= 2,2 MO

divider,

V which

comparator.

191 X R2/(Rl

or R11R2

i mate

Assuming

(100 #F).

voltage

[13].

similar

voltage

sients

pump

capacitor

of input

is

voltage

in the

reference

a time

charge

time.

pump

(1 N4001 ), a

voltage

to its low

[14], It has a temperature-compensated

The

diodes

at operation.

designed

adaptor

of off-state

condition

R1

fall

charge-pump

all input

of MAC229A8

device

but it is also

input-voltage

the

of the charge

of two

the rise time

duting

on the

for

C, and a smoothing

peak of input voltage

of rise

operation

power

2,6

drop

the capacitance

value
principle

It consists

during

discharged

dis-

suitable

capacitor

C is charged

C.

is 600 V, which

Low

sensing

the power

considered.

coupling

maxi-

Ill [13]. The small

C, the working

limits.

Guaranteed

4,

and will

capacitor

voltage

IGT of 10 mA

Iland

will minimize

in the adaptor

voltage

with

quadrantsl,

of the charge-pump
2.

calculate

capacitor

of the following

Universal

327

..

and

dV/dt

blocking

hence,

capabiiitv

the overall

of the

svstem

T=1661m$

Input-Voltage

Adaptor

(Negative

Gate

Current)

triac

reliability

4 x 1N5398

o
330pF

5D.I1

Q1fiF

2wv- ~~~ p -

400V
W-130 VAC

33 @F

OR 180-260
VAC

2WV7
KMG

T
-

50,60Hz

+ v~~

(200-370
Vdcl

47 k
lW

1N59%A

39k

1N5MA

lW
I

I
I

NOT INCLUDEDIN
~

THFDEMOBOARD
~-

--

----

~700py)
--

11

~
J

MAC229A8

1N4001
lti

10
T2
3I

Yr MCR102

12~F

Iw ~F
6V

lk

~.

20 k

1K2
d

I 1{
(Q3A1

{Q2A1

142k

H
1~,

330

MOC81O2 *
1F5
.-

3K3
,0.

1 N4740A
, ,7

Start-up
R2

Rgure

8. Complete

Grcuit

20 v

10V

20 k

33 k (1OWI

Schematic

of 90 W Off-the-tine

TL431CP

__-_J

I
I

Power

HEATSINK

Supply

328

-.,-___...

___I_,

*, .I

.i

,Q

.::

.:;

.t!v
iix

..____

_-

..-.--

_I_.

TRANSFORMER

CONSTRUCTION

OMGRAM

80TTOM VIEW
I

Ls(11oI

.
0

*
AWG #22

Lp = l,5t01,75mH

N,,5),~iNs(8:RlMAR,.To.

Np = 172
AWG #23

.$;$:,>

N$(81 = 7
0
LA

WINOING AREA

Ls{151

A,,,%,,,,,,,)

AWG #26

BOBBIN

4mm APPROX

Figure

it is least

sensitive

MTP4N90

the

adaptor
can

the flexibility
adaptor
the

lower

the

the

power

needed.

Therefore,

becomes

optional,

ponent

layouts

and the

main

board

the

9. The

construction

shown

in Figure

ductor

components

this

printed

universal

of power

used

supply

may

can

in this

be

the

for

increasa

the

approach,
circuit

the

board

power

not

adaptor
and

com-

in Figura

transformer

all Motorola

-11

is

semiah.

to the

MUR180
M8R160
1N4735A

62 V

1 N4740A

10 V

1 N4747A

20 V

1N5953A

150

1N5956A

200 V

(for

MJElaO04)

(for

MTP4N90)

V lootlonal)

Con-

figure

lower

has

it hashigher

:~~~it
and
in
~OSFETS
can

conduction

power

losses

the switching
compete
with

conduction

loss

in the

transitions.
advanced

at

(700A),

is74,2/o
= 360V.
loss

than

base

drive

This
is whv
BJT even with

relatively

low

switching

frequencv,
The

maximum

imately

150mV

ripple

as predicted

observed

to bestable

currents.

The dynamic

with

an overshoot

VCC

= 200 V, from

Also

in Figure
no load

and

undershoot

tory,
this

will

Typical

in slower

experimental

switching

frequencies.

Also,

of the

90 W off-the-lina

20 V

isquitesatisfacbv increasing
loop,

waveforms

currents,.

input

Figure

power

disturbances

But,

responses.

switching

1),

power

is about

feedback

transient
load

Figure

Of the

reduced

Cfinthe

of

15.7 kHz and

(see

which

range

satisfactory,

overshoot

can be further

capacitance

result

atf~

large-signal
30V,

0,2 /~of

antire

respOsas

The

is over

The overshoot
integrating

the

is also

8V

forverv

to full-load,

is approx-

2,3, The power

to full-load

transient

areintroduced

the

over

response

half-load

12, the

output

is Iess than

in section

of Ieas than

from
the

at 110V
which

voltaga,

load

supply

voltage

(Desk-to-Desk)

11, at different

expected

Alt#&~~~~JE18004

in Figure

1
1
1
1

rails iswithin

manufactured,

the best efficiency

15.7 kHz and VCC

higher

2, Line
0,50A) for

@&@,r?S.70Aa t10(110)
= 0,7 A,fs = 15.7kHz
,>p
=~WVfor
MTP4N90;
whereas
for the bipolar

VCC

1
2
1
4
1
2
2

than

forothertwo

efficie@,?$,\$~se

supplvis

1
1

RESULTS

in Table

(better

properlv

the output

1
1
1
1

~@latl&n

AND

power, ~?%nsiskr
MJE18004,
at IQL{l~~:=
0.7 A, fs =

supply,

Qty.

@~~?Fent

transf:w!~

~~4&~0,

,@

are

OVoutput.

andthebest

adaptor(s)

are shown
power

is

regulation

version

normal

adaptor

load

and

MEASU&&F+S
<,. t;;
,,*,
a\+~l,+~W~rtzed

measurements

10/0, if the

input-voltage

input-voltage

1 lists

and

orunpluggedfrom
affacting

of the

10. Table

universal

and

if the

supply

diagram

the

,~?t...$,
..:
*it,,,.i.:,
,$.!...:*
~,~,..t
~,,

Construtiion

D.C.

),,,< f.? ~,~

,,,.
s. ,.$* ~$.
~.>:t%.+ ~~.~
,:..i,?:.,,i~,
!,.<s~~
.~
.

TOK PST.39

4.2 EXPERIMENTAL

circuits

approach

The universal
removed

using

of the

cost

supplv,

The

board

power

without

power

have

a modular

system

can be simply

supplv
of

Drive

Transformer

shown.

to

Then,

of manufacture.

board

operation

because

range.

Fiyback

10.

mode.

are also

is unnecessary

at one

adaptor

in this

MJE18004

it

input-voltage
only

to noise

and

Sometimes,
used

,+
,,$,

FERRITE CORE TDK ETD.39 H7C4

Ig 1-

AIR GAP 1~

for

=::{;

I
+/

are shown
voltages

13 shows

the

and
photo

supplv,

5, CONCLUSION
A low-cost

90 W flyback

synchronization
multi

-svnc

color

The

power

and

is found

multi-sync
both
greatly

and

supply

monitor

mains,

affecting

been

has excellent
monitors
i.e.

thasvstem

line

in the
orNs.

90-130

supply

with

input-voltage

has

to besuitable
color

a.c.

power

universal

load

application
Also,

V or
cost

discussed
and

and

external

adaptor

for

in detail,
regulation
of low-cost

it can operate

180260

V, without

performance,

at

Table

2. Performance

of 90 W Off-the-tine
MTP4N90

10 (11OV)

v)

(8.0v)

Fl@ack

Power

SUPPIV

[MOSFETI
Efficiency

f~

Ii

0.2
0.5
0.7

110.1
110.0
109,9

16.01
16.23
16.31

8.88
9.05
9.10

15.7
15.7
15,7

0.12
0.26
0.35

300
300
300

0.7
0.7

109.9
109,9

16.32
16.30

9.10
9.10

15.7
15.7

0.55
0.29

200
360

0.2
0.5
0.7

110.1
110,0
110.0

1599
16.19
16.25

8.88
9.03
9.08

25.0
25.0
25.0

0.13
0.26
0.35

300
300
300

0.7
0.7

110.0
109.9

16,26
16.25

9.07
9.08

25.0
25.0

0.53
0.29

200
360

, $;,
t \$..
,

:$,.L?2,6
,t 73.7

0.2
0.5
0.7

110,1
110.0
110.0

15.98
16.17
16.23

6.88
9.03
9.07

32.0
32.0
32,0

0.13
0.26
0.35

3061< +{ ~
~oo f)%...

56.5
70.5
73.3

0.7
0.7

110.0
110.0

16.24
16.23

9.07
9.07

32.0
32.0

0.53
0.30

kHz

(15 v)

18.0 V)

Vo(llov)

(15

MJE18004
Io(llov)

V.

(11OV)

:.;@@,

f*

0.2

110.8

14.41

8.82

15.7

0.7
0.7

110.6
110,6

14.73
14.83

9.06
9.11

\s.7

110,8
110.8
110,7

14,44
14.70
14.78

8.83
902

0.7
0.7

110.7
1107

14.77
14.78

9.08
$~!
909,,,+
f

0.2
0.5
0.7

110,8
1108
110.7

07
0.7

110.7
1107

A
Ripple voltage

v
at 110 V Otpt

14.43
14.68
14.75

,)<: 8.83
.+:, :~ol
,<). , :,~, 9.07

14.7*)
$ L;
9.07
14?%>.< $$ :
9.08
~~.+..>;.>
v
,,,,. .\.tt.
1~.~pp
~, V~~ = 300 V,
IS abow?N.
~+

.*:. ,

157

$Q!&,\.
<:2% JR3

72.6
71.3
v

0/.

Efficiecv

054

360

.,.
74,2

300
300
300

56.8
68.4
71.8

--

25.0
25.0

0.53
0.30

200
360

73.1
71.8

32.0
32.0
32.o

0.13
0.27
0.36

300
300
300

56.5
68.4
71.8

32.0
32.0

0.54
0.30

200
360

71.8
7~.a

0/.

kHz
f~

,,
:..
69.9 y{<~i :,st
7363: ~?. *, ~;5 if)
~~

Vcc

,:4* <it ~yz

,;8,
, I

61.2
70.5
73,3

.1,, .

,$<:~$:
~l,i *. ,..
A
,, \,t.
,>.. , ~...$.*
.,i,:,$..l!.J\*
$.t$,
,,*q,&
,

(Bipolar)

0.2
0.5
0.7

Vcc

kHz

&

10

0.7

A.

,..

~ :kt,~
.?*
,:.

*>*

figure

11. Experimental

Oscillograms
C.)

5,

CH2

,x,

,0,,

,35)

VER1

fl

#
1,

25 kHz

Vcc
3W v
10 05A

*c

MJC18W

u;+

CH2gnd
cHl9M-

._

CH>

Figure

ha.

Kev Waveforms

VCC

at fs = 25 kHz and

figure

= 300 V (for MTP4N90)

llb.
VCC

337

VCE

FREO

and VBE

= 300 V (for

25126 kHz

at fs = 25 kHz and
MJE19004)

Figure

CHI
C,

?0 V
lM V

Figure

1 lc. VDS
VCC

CHI
,.,

8SV

and VGS

Oscillograms

VERI

at fs =

15.7 kHz and

= 360 V (for MTP4N90)

mv
w

,0.,

11. Experimental

,,

,0.s

8%V

vER:

H1d~
figure

1 lf.
Vcc

:,

70

CH?

100 v

CH2

VDS

FREO = 25

and VGS

WO
XU
at f~ = 25 kHz and

= 200 V (for MTP4N90)

,0.,

!22

ViR-

CH19d
c>

Figure

llg.

VDS

VCC

and VGS

= 360 V (for

Figure

at f~ = 32 kHz and

llh.

VDS

VCC

MTP4N90)

332

FRCO

32051 ,,

and VGS

at fs = 32 kHz and

= 200 V (for MTP4N90)

Hgure
CH1

CM?

10v
mv

IM .s

703 mY

12. Large-Sgnal

Transient

Load

VERT

CHI

,0 V

5V

Responses

,m .,

703 .V

VER1

NO 10AO TO FULL LOAD

wLF.LOm

10 FuLL.LOAD

cH2gd,

~::!:!:w::!!

CHlgnd

CH2gnd - -

CH2 fREQ = 1 $@! Hz

figure

12a. For MTP4N90,

From

at f~ =
CH1

,oV

ml

?0 v

No Load

to Full-Load

15.7 kHz.
CHI

,mm,

414V

VER7

CH2

1,

,57kHz

Vcc
2M v
M,E >am,

CH2gn6.

4!:!:

Figure

12c.

For MJE1800$(,
atf~f~

%Q$No

Load

to Full-Load

Rgure

12d.

1~,.~~Hz,

For MJE18004,
atfs

CH1

10V

CH2

5,

From
=

A IN.,

Half-Load

to Full-Load

15.7 kHz.
IO]mv

VER1

cdc~
CH2

Figure

12e.

For MTP4N90,
atf~

FREO = 1920Hz

From

No Load to Fu11-Load

figure

= 32 kHz.

12f.

For MTP4N90,

From

Half-Load

at f~ = 32 kHz.

333

to Full-Load

*3,
1j
_.,
i ;2
!:i

:a

AN1092
Driving High Capacitance DRAMs
In An ECL System

st.,
\
.,.,..
.)\i.
,::;
~~.,?,i.
*.,>
,.y,,~,i:>,.ii.
.:,
.:!~.}i:+
..
.:$

~,.,
* .,,:

INTRODUCTION
In present day computer/controller systems where speed
,0A2D0
and efficiency are of the utmost impotiance, system designers
areusing mixed technology in their designs toachievethenecessary speed, power, cost and processing capability desired in
high speed data processing systems.
lmol
The logic type most applicable to the high speed function of
J,i
such a system is Emitter Coupled Logic (ECL). Motorolas
~.:.:,,.
IGNDoI,,
10K, 10H, and ECLinPS devices make it possible to operate
$>,>
..,<:<. .1
with clock rates up to 1 GHz. However there are sections of a
,,:- ,,
system where ECL speeds are not necessary, For example, in
~i$:nl
, ,:},...,
the area of bulk memory that is not awessed every clock cycle
,? ~t,~~
,}, ?>%:,:$
a large CMOS DRAM is lass costly, uses less power and takes
.$:,
~
,T$$
up less board space per bit than an ECL memory. Now, since
.
..:,\
ECL and CMOS are of different Iogicformsandtheir signal lev,<>,\,r\::tx
,.,$1,
els are not compatible there needs to be a level translation to
,>,,.,?s,
.!,.,
enable the two logic families to be used together. The Motorola
-+ ..$
MCI OH/100H660 4-BIT ECL-~L LOAD REDUCING DRAM . !s,.>.$
,\.,,,.:.*\
DRIVER was designed for this purpose, The H660 is shown r%;,.
s-.
>+kt,
.,,.,>
.*
a simplified typical system application in Figure 1,
D2This paper will explain the features that were desi~~ed i@O
the H660 and how to apply them in a mixed technol@~5ystem
to obtain the best performance versus power rati~.,, <~
i,,,.
,{
,X!,$*,,
.,\~.., .,*

F
DQ

Q1A

EN

OGND1
Q1B

F
F
DQ

Q2A

OQ

OGND2
Q2B

EN

OW23

Q3A

EN

0GND3
Q3B

MC1 0/100H660 Logic Diagram

o\. ~; -*
~~o$~tch
highly capacitive loads at speeds of a few nano~j$y~~+@ds, the device must supply a large amount of current to
~{,k.iwrge
the lines then it must sink this current to discharge
,,,,.,
~~
,.,.,~~them.
This fast switching on an unterminated line can result in
$l#J* a substantial amount of over shoot and ringing.
$
To eliminate the overshoot and ringing, a small value series
resistor (Rs) can be placed at the driver. Figure4 shows an

FEATURES

OF THE H660 DRAM DRIVER

The H660 translates the ECL signal to a TTL level suitable

for driving DRAM memories with high input capacitance.
The input impedance of the 660 varies with frequency, at 10
MHz it is typically about 150 to 250 ohms and goes down to
about 50 to 60 ohms at 200 MHz as shown in Rgure 2,

application of the H660 with a series resistor.

335

A latch is added to provide the capability for a memory controller to propagate new addresses to different banks without
having to wait for the address timing constraints to be satisfied
from aprevious memory operation. Forsystem \mplementations where this IS acceptable, the user has the capablity to
keep the latch open, thus having the part act as an address
translator/buffer, with minimal pefiormance impact dueto the
k
additional propagation delay incurred from the internal latch. .{::,J>A%
The latch is controlled with an already existing ECL ley.ek,a,+~
DRAM timing signal.

200

&
~
100-

IMPEDANCE
I

0+
0

1
,

100
FREQUENCY,

200

MHz

figure 2. Input Impedance

For every ECLinput there aretwooutput lines, each capable
of driving 300 pF. Assuming 5to 15 pFcapacltance perpinfor 1
megabit of DRAM, each output pa[r could be connected to 40
megabits of DRAM,
The H660 has a totem type TTL output stage with no IOS
limiting resistor. The output IsInk capability is 48 mA. The output transistors are driven differentially with a dual phase splitter from the translator, this assures that both the output totem
transistors will never be turned on at the same time therefore,
with no load, Iccdynamic power remains constant over frequency. We recommend a m[nimum load of 100 pF the graph
in Figure 3 was made using a special input signal just to show
that the typical internal TTL current glitch IS not present. The
dual phase splitter is a unique method, patented by Motorola,
.
..,..J

..

50 !
o

100
CAPACITANCE,

200

300

PF

Hgure 5. MCI O/100H660 ICC versus CL, RS

RS=O
RS=5
Rs=1o
Rs=20

10

I
I
2
1b
CAPACITANCE,

s ~~fgure 3. Icc versus Frequency

.,..,
<t:
,,...
,<$$+:?l
with 300 pF and No Load
,.:: ,
$.,[f.>e~.?.
.F,+, ~~e dtivers are arranged to be used in pairs, each output
~~~kis associated with aground pin and every two pairs with
:mne VCCT pin. The internal logic VCC and ground pins are
separate from the output VCC and ground pins, this keeps the
noise from the high current output from feed(ng back to the internal logic. If there IS ever a need to use only two data lines,
power would be needed only on half of the device.

300

200
PF

Hgure 6. TPD versus CL, RS

Another important benefit of the sefies resistor is that (t reduces the device dynamic current by shaping the waveform,
giving it slower rising and falling edges. The slower edge rate
eliminates the overshoot and ringing that are associated with
very fast signal edges on unterminated printed circuit card
traces. Also, the slower edge allows for longer circutt traces to
be used without the need to be terminated. As shown in the
waveforms in Figure 7, the signal can be shaped to meet many
system requirements.

336

.
_____________________

155.~0ns

N5.600ns

OS
255.600

I
1

Ch, 1
= 800.0mVolts/dlv
Ch. 2
= 2.000volt#dv
=1 00 nsltiv
Tlmebse
Ch. 2 Parametem
Fall Ime
= 8.8850 ns
P-P volts
= 78750 Volts

Nse T(me= 10,7652 ns

Preshoot =24.460/.

Ofiset
onset
Delav
+ W!~th
Overshoot

= -3.185 Volts
= 6.375 Volts
= 155.600 ns
= 47.4560 ns
=4.0810/.

b
H660

WAVEFORM
,

155600 ns
I

,,*

SCOPE
5

/
300 pF

..,,

,;.$.,> $$
..

205.600 ns

255.600 ns

SCOPE
20 Q

H660

,r$

,-<> . . . . .. .

=*;,
,%,

~mekse
=10.0nsldv
Ch. 2 Parameters
Fall ~me
= 14,9356 ns
P.P volts
= 3.6875 Volts

Rise T(me= 13.9250 ns

Preshoot =3.6360/.

Offset
Offset
Oelay
+ Wdth
Overshmt
WAVEFORM
Rgure

= -3.185 Volts
= 6.375 Volts
= 155.600 ns
= 46.2754 nS
=3.836%
3

7,

337

300PF

There are times when the overshoot is desirable as when

driving CMOS memories requiring arailto rail input signal.
When the load is capacitive with no pull up or pull down resis-

--

--[

---

--

tor, the output will go all the way to each rail and will not discharge in a cycle time period. An example of this phenomena is
shown in figure8.

---

---

Ch. 4
Tlmebase
Vmarkerl

= 2.00 mVolts/div
= 20.0 nsldv
= 0.0000volts

Oflset=t.912 Volts
Vmsrker2

=6.1250 Volts

Delay =16.0000n5
Delta V = 6.1250 Volts

,..

P
,,,.,, , .~p,

.,..$$,,
,,,

When driving a resistive load it is seen on the chart in Figure

9thatthe VOH ievelremains somewhat constant over iOH
loads that are over the device rating,

*,

As a precaut~~n~~y @&, if an output is being used without

the series re~~Wa*&\f it becomes shorted to ground while in
a high sta~~, it~~$ource over 700 mA and in a short period of
time thw%~j,ce will be destroyed.
Aft@We @oper memory addresses are selected and the
TJ,b~~at~Js transferred from memory the data is then trans~~ab~k
to ECL by use of a TTL to ECL translator such as a
~~&Ybla MC1 OH/100 H600, 602 nine bit TTLto ECLtranslator
J or a MC10124 or MC10H124, 4 bit translator.
,?,J

CONCLUSION
Mixed technology systems are becoming very popular
where system designers must optimize system performance
while keeping overall system cosVpower in line.
This application note described the MC1 OH/100H600 4-BIT
ECL-TTL LOAD REDUCING DRAM DRIVER and some application techniques that can result in an improvement in system
performance and rellatility.

338

ANII06
st.,
\
.,.,..
....)\i.
.........
..
,.!*\.
,! ..:!,
.>
?,,*> ,>!.+

Considerations
in Using The MHW801 and
MHW851 Series RF Power Modules

.,%&,\\
:$,,
;,.$?
....:, ~?t~:,
.,).%,,
, .~;,,,,..-..
ii
~$$l:,v.ii:$t
~t+..!ts
,,,,
~*,..+< (.~,
.~~!
*:*
, i

Norm Dye and Mike Shields

RF Products Division

by

INTRODUCTION
The

MHW801/851

primarily

for applisatlons

module

IS frequency

called
the

AMPS;

the

European

compatible
the

-4

with

compatible

models

Other

The

Ser+es

of modules

MHW801,

only

mounting
flange

IS the

system

are considered

to be sufiace

mount

modules

Introduced

beyond

the

Intended

talned

amount

networks

used

show]ng

and

VCOnt

,~odul,@

Note,

modu~

@%

CW

the

IS

of applica~(~;~OTmatlOn
&$j&*

as func~,:,i$f
,.
;*

sheet.

lines

is to amplify

CW

signals

that

are biased

if tha signal

as amplitude

be

is at least

of the module

the last two stages

will result

such

input

that case

temperature

by the same

operation.

contains

modulation.

How-

The

thermal

long pulse

could

be much

trains
higher

can no longer

constant
time

10 micro-seconds

ately

operation.
output

cu wes

difference

constant

which

wtth low duty

of

says

cycles,

than that predicted

be tied
used for

the

die

IS

that for moderdie temperature

from

have not been characterized

It is to be assumed

power

mode

VSWR

and output

however,

The modules

Also

and decoupling

efficiency,
frequency:

output

connec-

should

CW measure-

ments.

IS COn-

pedormance

suda~:~~ont,

as [*\,Yof

data

information,

aPPrOXIMately

attachm:m$~&~~%eatsink

th&.@bdule

module

from

layout

The first two stages

to die temperature

Th~&@W851

Th@#~~~bxtends

in the test fi~~~;~~plcal

power

by attaching

similar

the

lines

the ground

board

the

)$hyer, it is possible to operate the module in less than a CW

,.~\+*.*>,
;: condition.
In a pulse mode of operation,
any duty cycle up
dis- ,+ ~,.
,).\ !%,.,$~~
to 1000/. should create no problems
provided the peak power
th%~@{~$
does not exceed the rated CW output power of the module.
the{k~
~

appreciable

modules.

that

is soldered

of the

It IS intended

MHW801

the

feedback

and this

becomes

cl~~~%lrd,

isolatm&@put
il.\ .
,>.,,..;
*

.~~ amphude

in signal

Remember

sub~~~A#$ground

ar~.bl%,@-~lass
A; however,
,3, ,,$t.
C%
C. Significant
distotiion

data

MHW851

case

that

t~$~,~~ult

~p~bllifibs.

flanget@$@ti*n

to external

Normali@W

of mOuntinQ*~~jW

dlagrarn,~~

parameters

and output

The

MHW801/MHW85.$$,

are a block

any
and

of mounting.

substrate/cap

In the

Included

and

to offer

method

to be by means

A significant

In the

not extend

solder.

were

and

will be accomplished

heatsink

conventional

MHW801

module

are fr~~~~modulated.

all

the

of the circuit

50 dB.

and

and meet

on

back

such

NTACS

operation,

fotih

design.

with

more

of

consequent

tion

with

NMT
the

with

to the module

system

are identical

substrate/cap

to a heatsink

called

set

-1

is frequency

with

frequency

flange

does

the PCB

to the

module

specifications

the flange

beyond

-3

In the

The

compatible

compatible

than

difference

radios.

the American

and MHW851

electrical

sheet.

tance

the

the Scandanav[an

of the MHW801

general

with

system;

are designed

portable

IS frequency

is frequency

in Japan.

modules

In cellular

low-level Impedances

prevent

of power

-2 module

TACS

module

system

the

Series

can

of operation;

first consulting

power

voltage

temperature.

be obtained
however,

greater

from

the

than

module

because

of concern

as well

as maximum

power

rated

CW

in a pulse

this is not recommended

the factoy

swings

for pulse

that

without

for maximum

die temperature.

,.*,.

~.~,
.,
.-,,$
1,

GEN$5~~:L*ECTRlCAL
Mo~,j~<r&matched
lnp~,@~~put.
,. :;*,\<
~ih$%$ransmltter

Thus

their application

poflion

NOISE CHARACTERISTICS

CONSIDERATIONS

to an impedance

of 50 ohms
m a sub-system

of a portable

radio

One

for both

parameter

is nolae.

such

radios

IS relatively

where

that

neously.

to

capable

t- the module be de-coupled

by means of by-pass
capacitors
::+, ~,,,.>.
1> ~,>,,&,
:,0,,,
and/or chokes
to prevent
bias and power supply
circuity
\.,
contributing
to circuit instabilities
(spurious
oscillations).
It is

teristics

:w&S~,~~@htforward.
+t<+$~puld

recommended
the

However,

be observed.

decoupling

Second,

First,

there

that the module

Information

grounding

of the

are certain

[t IS tmpotiant

user

pay careful

presented
module

precautions
that

In the

should

be

DC inputs

attention
data

radio

which

Today,

of the
is normally

modules

radios

are

and recerving
are concerned

transmitter
45 MHz

Motorola

cellular

frequently

of power

modules

and receiving

cellular

performance

In dupiexed

339

transmitting

manufacturers

tee noise

to

of power
applications

of transmltfing

this reason,

to

sheet.

adequate

Most

in the
above

has begun
of modules
radios.

not specified
have

been

in

did notoccursimulta-

duplexed,

i.e, they

at the same

time.

are
Thus

about

the noise

charac-

receive

frequency

band,

the transmit

frequency.

to characterize
designed

For

and guaran-

ptimarily

for use

Noise

power

guaranteed
is

for

the

MHW801

in a 30 kHz

shown visually in Figure

for two widely

power
modules

only.

~fferent

by as

is that the small

much

45

Seties

of the

signal

(noise)

modules

The

is

MHz above fo. This

1. Note that the noise

temperatures

A characteristic

is approximately

/851

bandwidth,

Figure

IS specified

3 dB

as

the

2. Several

order. Hrst,
as close

to kTB

MHW801/851

generator

uses

Series

control

voltage

(VCont)

sized
is

decreased.

source

source

must

a cavity oscillator

be extremely

does

not. If this

to make

noise

measurements,

filter which will

(or Wavetek

type of signal
it is

+:
E
g

E
2

%
0

TRANSMIT SIGNAL
(80+940 MHz)
~

30kHz RECEIVE BAND

-.- /
+5 dBm

MAXIMUMNOISE POWER

45 MHz
D

FREQuENcY
Figure

1. Noise

Power

In Receive

Band

,$,7
?.

1,HP 71000 SPECTRUM ANALYZER

J. HP POWER METER
K. HP POWER METER
L 10 dB INTERNAL ATTEN
M, 6d8 PAD
Figure

2. Block

Magram For Sdeband Noise Measurement

340

synthe-

2520A)

necessa~

pout

signal

the require-

generator

reject any signals

in

are in

low noise,

HP8614A

and satisfies

the output frequency.

is shown

On the other hand, a frequency

such as the HP8656

generator

bandpass

measurements

about the block diagram

noise as possible. The

ment of low noise.

gain of the amplifier

for noise

comments

the signal

and for rated output

35 dB at rated output power but increases

as

block diagram

45

signal
%>.
,..,+ ,,b:~*
is us~,~~~~:

to a~$ a$$.

MHz,~~~$

x . .!,, \i
:\.,t,:.:,,$
,:\*~
*,\<

<$:<.!~ .
.,>.. .*,, .,:3*
,.~~,
:.+:.\...
~:,k,.+ :,/,
.{.:4,
$,

Remember that any noise at the input of the MHW801/851

istics, efficiency and harmonica will degrade at reduced
Series module is amplified by approximately 35 dB. This
output power. As output power is reduced, the class C
noise amplification should not be confused with internally
stages of the module operate further and further from their
optimum load line resulting in significantly poorer efficiency.
generated noise which could be caused by a high stage noise
As their operation approaches the more non-linear region of
figure or by regeneration in one of the module stages, neither
the transistor transfer function, noise will likely increase and
of which is a factor in the MHW801/851 Series module
harmonics will increase with respect to carrier power. G~,:rdesign.
ally these degradations in performance are not seri,A~@,
Second, it is essential that the module be terminated in
cause they are relative to carrier power level, F~&~@ple,
a circulator which will prevent out-of-band impedances of the
e~ciency becomes much less at output po~~$&y&ls Of
subsequent RF network from affecting the stability (and, thus
100 mW but current drain is much lower Q*R f6~the case
noise) of the module. Third, care must be taken to prevent
;5,4>.,..,,
of 2 watts of output power, so this is gen,~~~&~,@}considered
the carrier frequency from saturating the input stages of the
~~t:t,t:.:~
r>
a problem in radio applications.
spectrum analyzer used to measure the noise level. Again,
it is crit!cal in accurate noise measurements to be certain
Other circuit considerations extey$~%%~ module that are
sometimes overlooked are sa~:ce *~@~ load impedances,
that the sensitivity of the spectrum analyzer be at least 10 dB
Note that the stability of the.?~~ple IS guaranteed only for
better than the noise level being measured.
Normally to accomplish this it is necessary to reduce the
source VSWRS of 3:1 a@J@~%SWRs of 6:1. Frequently
resolution bandwidth (RBW) of the spectrum analyzer to
the load for the modu~k~$the transmit portion of a duplex
filter, The out-of-bqpd~tince
presented by the filter Can
30 kHz and set the video filter to 100 Hz bandwidth. The
manufacturer (Hewlett Packard) of the spectrum analyzer
affect the stability &$~*module.
The impedance reflected
recommends a video bandwidth 100times less than the RBW
to the modula~peri$s
on the length of transmission line
between tq~~~~~b and the filter thereby causing line length
for best noise averaging.
to be :? ~,ddll~~al circuit consideration. it should be rememThe filters, H and G (in Figure 2) are stagger tuned to
bere~ ~~a~~e MHW801 /851 Series of modules are not
obtain adequate selectivity for rejecting the carrier frequency
at the input to the spectrum analyzer. Obviously a single filter
:~c~ti{jonally
stable for all load and source impedances.
can be used if it has a reiection level of aoDroximatelv 60 dB,
:~~ut-of-band impedances of filters result in siqniticantly hiqh
45 MHz away from the bandpass of the filter. The actual ~$~, Y*Rs
at out-of-band frequencies. If these i~pedances a~e
rejection needed depends on whatever is required to preve~~ ~].. &flecled to the module such that the module is terminated
,..$i,
,>-.~:
t$
In impedances that lead to regions of instability, the module
saturation of the spectrum analyzer by the carrier sign@@,.<$+\$,
,:
will oscillate.
.,s,,
.
.,
Input power to the module can va~ from a low value of
,,,:, .*S
,,,::,.
GAIN CONTROL
O mW to a recommended maximum of 3 mW. Input powers
~~.t,.
The data sheet recommends gain control b~l~::~g
input
greater than 3 mW are not recommended because of the
power at O dBm and varying the co:~$~li:~~~e,
Output
potential damage that might result from overdriving the two
power versus control voltage is shown~,~e+~ical
characfinal class C stages in the module. Overdrive results in
teristics of the data sheet. Gain cofl%$~~~~ MHW801/851
excessive power dissipation particularly for the simultaneous
Series module is obtained by con{w~~he
bias tOthe ClaSS
condition of maximum supply voltage of 7.5 volts. Overdriving
A input stage as opposed to q~er:~~ules
that controls the
the final Class C stages can also lead to circuit instabilities
voltage to Class C driver me>~~e
benefit of this method
because of changing impedances. Likewise, supply voltages
of control is sigmficantly..,1,
greater than 7.5 volts should not be applied to the module
-,~~~ico~rol
,>.}!,,,...,. current and a Iowerslope
of the output power ~rs~
control voltage cuwe.
for the same reasons of overdissipafion and potential instabilIt is possible t~9@.n?hWbutput power from the module by
ities.
controlling inpu[-~w:~
:<.,:.~ with the control voltage maintained
at a fixed JSVEI (generally maximum). This is somewhat
MOUNTING CONSIDERATIONS
intuitive; dv,@k
,{:, \.},..
~. a major benefit of this method for power
GENERAL
out c~~~j~a~
not be obvious. This benefit is the best noise
In mounting power modules, consideration must be given
pefi@?~ri&& of the module because the small signal gain
~,!.
.. ~...
to heat dissipation and grounding. Motorola specifies the
j,$:.w$xlmately 3 dB less at high control voltage as comrange of case temperatures over whtch the module will
~ p~fk to low control voltage. Other important factors such
perform safely. The upper temperature is determined by
.<q*$y+*w
b..,., stability, input VSWR, harmonics, efficiency and load
thermal resistances between each die and the case with the
~.*e~hsm
mismatch are essentially unaffected by the method of output
guideline that die temperature will be maintained below
$~.+t,i
, **
power control.
;,:,,.. .>
200C, which is considered a safe temperature for silicon
v
transistors, All the user has to do is provide sufficient heat
OTHER CIRCUIT CONSIDERATIONS
sinking for the module to be certain that the flange of the
Performance of the module at less than rated output power
module does not exceed the maximum operating temperais sometimes of significance in typical module applications.
ture rating. The maximum power dissipated by the module
Regardless of output power control, the noise charactercan be determined by determining the maximum DC power

341

input
the

less

output
the

the

RF power

maximum
power

rated

Maximum

minimum

efficiency

and

function

to determine
the

then

the

rated

dissipation

2 Watts).

This

of the

module

that the r!se in flange

the temperature

only

the

small

amount

thermal

and the
using

heatsink

at the mounting

temperature

at the ends

screws

(at center

of the flange

of flange)

should

ing

the module

screws

only

to external

should

be

adequate

oscillations

provided

the

excessively

resistive

as a result

the nickel
oxide)

plated

flange.

!s resistive

ground

paths

ground

Nickel

and

between

circuitry

to prevent

contact

does

of nickel

oxide

(unlike

copper

Its formation

can

lead

the module

and

external

ment to a heatsink
should

above

inch-pounds.
for the

MHW851

output

power

spurious

permissible

but

excessive

MHW801

When

solderlng,

prevent

any

temperature
such

as 520/.

recommended

should

are

the

pati

designed

be attached

Flatness
less

primary

module

than

165C.

In and

480/.

because

Keep

n mind that the internal

been

achieved

which

Ilquifles

using

Sn

this

flange

with

(along

from

with

62A

placed

to

the

flux)

heatsink

recommends

the

$$.$~~vent

W&~~:#~$~

heatsink

t,: $~ta~$l ?he

MHW801/851
circuit

yM%oth51

Series
either

for

modules.

module
concave

that

be flat to
should

which
that

believes

be distorted

or convex

be

module

substrate
bending

Motorola

can

modules.

Series

board

requirements

IS
an

is much

power
surface

easy

substrate

Of

usin9

MHW851s

a dimension

o@~$m]c

low

it is Motorolas

higher

O.OO$,.J$~t$~T

th,@ st~ent

LS~&@e&

(the~a~$&tact

Use o~{he~?grease

requir~~~~
~~~vj~

.{$~Q~*?:

are
that

as much

without

damage

~t$~~$rnodule.

is

~cause

bending

Unnecessary

h~

scribed

20/0 Ag sa~~d#i$/Y

flange

because
Intimate

it IS not,,,.
,a&~&
... ~.
it not be u@.@3<Q

that

and th~~+d,:.f
than

is not rec~rneq~

Since

than

as ,~i~O,~$ches

1500C.

of the module

screws

greater

+ or -

negates

solder

below

Pb and

If the module

be

no

be taken

IS constructed,:,~::@T&*ptlnted

the

solder.

achieving

R type

Iiquifies

construction

on

should

A low temperature

solder

36/0 Sn,

at 179180C.

ears

consideration

of the

greater

without

sealed,
potentially

for attach- ~\\~,

~,>.!~~mkt
!:,,

The cutouts at eac~

,:i,,~

grease

module

must

that

of the

critical

Motorola

circuits.

to a heatslnk

with

440

amount

to the heatsink,

amount.

relatively

modules

They

on

with ears

of screws.

not require

care

recommendation

and silver

to an

Series

surface

to intermittent

by means

does

the flange

on

have flanges

The use of thermal

mount-

Series

flange.

into the circuit~

will accommodate

be torqued

within
MHW801

as stated

to be immersed

$*,>>

indicate

forming

Also,

module

The part is not hermetically

penetrate

type modules

end of the flange

not become

oxide

the

Series

module

not exceed

through

the end result,

results.

MHW851

1 @c.
Grounding

system.

could

MHW851

be-

provided

the MHW851

being

do not permit

by

of power
contact

Ilquids

disastrous

or

divided

sheet,

in a flux removal

subtract

MHW801

(2 Watts

by minimal

Calculations

to a heatslnk

either
watts

relatively

be dissipated

the flange

for

is 2.44

electrically

data

and

modules

in the application.
attached

way

is to divide

power.

power

normally

tween

by the

Series

.45 minus
can

Another

to be dissipated

output

MHW851

output.

power

in EB107

RF Power

i~,~i@&

to

circuit

board

about

<Mounting

Modules.

ic substrate

requirements

wor~

modules
substrate

This
and

are

relaxed,

tightening

Considerations

EB was wr[tten
does

modules

as

for ceram-

in total

as the

342

to printed

MHW801/851

Series.

de-

for Motorola

primarily

not apply
such

it !s also

sequence

-_._.,._

Understanding
by Norman

RF Data Sheet Parameters

E. Dye

RF Products

Division

the transistor

INTRODUCTION
Data sheets

are often

the capability

the sole source

and characteristics

Iarly true of unique

by equipment
circuit

des!gner

on the data

often

cannot

sheet

many

it is important

Th!s

parameters
tics.

speak

fully

a common

by the

from

RF

about

ratings

mto 5 basic

RF

what

Translators,

ampkfler

to functional

?~~l

module

are their

dures

used

set forth;

slgn!ficance.

to obtain

the

background

A brief

description

Impedance

Importance

Translators,

i.t~,

data

of test

and

circuits

4) Power

->~eslgned

thery~dat~

and

is given to help und,$rA&~.@w

~t:\
Ikneanty requirement~::,
e,:
~
,$.,
~s,\ ....

noise

and

varied

Many

because
essary)

wafer/d!e

result

of material

sired

impurities

sources
..yv

Bas!cally

beta)

and

gain,

of breakdown~@*$?
capacttan~e%b

leakage

.w~le

the

ruggedness,,$~s~flgure,

dlstorbon,
either

an

functional

thermal

created
These

hFE (DC

specs

dO nOt fall cleanlY

resistance

and power

!. {W.-
.:~..
..

des!gn

do not pose

sodium)

tend to change

in

leakage

specific

current

for commercial

that has

IS critical

and RF voltages

that appear

across

the various

IS essential

by the DC
junctions

complete

of

343

render

reverse

temperature

be exceeded

that

can be dlff!cult,

voltage

on the one hand

such
pipes

are constant
appl{cat!on

reliability

prob-

or where

device

which

device

between

bias

specification.

(primarily

test[ng.

patilcularly

current

useless
sources

requ[re

conslderabons

a leakage

by channels

is one reason

bias) and burn-in

applications

that

the

environments

Mllltery

created

in the oxide

and can lead to increases

Distinguishing

in

that the

WIII NOT

be

Inadequacies.
that

term

currents

with ttme

age current

It IS important

a transistor

related

for a patilcular
long

Some

problems;

can

processing
currents

contaminants

aPPllSatlOn

with the supply

which

ionic

application.

compatible

select

leakage

by mobile

application.

voltages

hand,

reduced

voltages

engineer

breakdown

and/or

as a
unde-

processing.

or they

In leakage

unnec-

andlor

rellab[llty

to

specified

arise

can be material

dislocations

so. They

in an intended

breakdown

currents

currents

wafer

cur-

informa-

(and frequently

are potential

defects

result

as

are Ilkely

and also more

during

currents

and

such

junction

ImperfectIons

acceptable

On the other

dissipa-

its .~W~&ding
called thermal
characteristics.
Figure 1 is
x t .,i~
twk~ge
of a typical RF POwer data sheet showin9
DC and
un@lonal specs
$i,,,, ,;<
,~R critical pafl of selecting
a transistor
IS choosing
one that

available

enter

currents

In

lems.

mtO

can .be.ett~
DC or AC. Thus, we will treat the spec
..: !$ $t.!
of th~r~?? Y@istance
as a special specdication
and g(ve it

~~}~$%as

mask

func-

of transistors

biased

Leakage

and tf tnltially

WIII remain

tion

\
$S~

that

by mask
sources

with time

cover

and ZOut, S-parameters,

etc.,~her%%haracterletlcs

cate$~.$fn$e

currents,

defects,

faults

potiolIos

breakdown)

in excessive
losses.

are not. Leakage

as stacking

ar~t~~{a@erized
by two types of
.. ..... :,
DC and func[io~al.
~be
DC spece consist (by

parameters:
def]ntbon)

most

,*,

RF transistors

with

product

do not have leakage

yteld

of leakage

entwined

voltages

as reverse

specif!catlon

can result

resls-

reasons

50 volts.

to avalanche

m their

and

a transistor.

fam!lles

supply
and

(defined

transistors
they

have

as

devices

by material

Most

specified

prior

such

2). It is for these

are intimately

28 volts

currents

that occur

voltage

determined

[ndustrv

6 to

of the superior

In making

characteristics.

at

typically

applications

high

by

a 12 volt

28 volt supplles

take advantage

(Figure

transistor

from

voltage,

ground

voltages

tO be Met

commonly

of

voltages

12.5 volts,

Leakage

t!ve

IS

is ela~at~;

l~~~r

are largely

use

b~&d~down

operate

volt supplies.

depths

for

7.5 volts,

be more

of the pro%-

information

considerations

@#Rges

oower

rents

~s~{~

X%#sttcs

performance

thekF

characterls-

nd~~a~~

4Q mr
-$, ,:}
w~fally

tlvity ,@&j~&tion

IS

1) DC Specifications,

Power

station

th~~ ~~~$own

and

sections:

3) Low

radios

WJQ::24 &.50

breakdo@

the

devrce

rad!os

appllc~~~~~~e

electrom~,

operate

~t
~;
,
.>. ~i::$>
,,r,,...
Modules and 5) Linear Modules.
Comments are made abouj *\,

...?,.. :*,
critical specifications,
about how values are determined
an~,:i,+
2) Power

base

peflormance

m them-

i.e., what

his

avionics

medical

for

des!gner.

transistor

maximum

It IS dlvlded

language,

says

circu!t

reviews

are unique

poflable

9 volts;

hand~a~

at frequen~~~~@Ztlves

Mobile

source;

wh[le

he

And

that the user and the manufactur-

manufacturer

paper

the

w!th the factory,

information.

of the Specificabons

er of RF products

understood

talk direclly

Because

and on the other

the gain

permit

the transistor.

that are used

world.

for his dev!ce

selves.

semiconductor

the

that

about

This IS pati!cu-

devices

all over

RF devices,
Thus

of a product.

RF semiconductor

designers

relies

of information

Ilmtt

for

of leak-

dev[ces
HTRB

for
(h!gh

However,

even

where

battery

drain

dictate

Iimitattons,

be included

It

In any

ELECTRICAL

CHARACTERISTICS

(Tc = 25C unless othewlse

noted.)

Characteristic

Symbol

Min

Collector-Eminer Breakdown Voltage (Ic = 20 mAdc, IB = O)

V(BR)CEO

16

Collector-Emitter Breakdown Voltage (Ic =20 mAdc, VBE = O)

V(BR)CES

36

V(BR)EBO

4.0

Typ

Max

Unit

Vdc

Vdc .<.,,
*
v~%,,,, &..
\ ,.. ?,,.

I OFF CHARACTERISTICS

Em!ner-Base

Breakdown

Voltage (IE = 5.0 mAdc, IC = 0)

ICES

Collector Cutoff Current (VCE = 15 Vdc, VBE = 0, TC = 25C)

$m&*~
.!. .>,.,
~,, ..
>,,,!
*, ..-,,~i,

10

ON CHARACTERISTICS
DC Current Gain (Ic = 40 Adc, VCE = 5.0 Vdc)
DYNAMIC

$ii.~, ..
~,:.~,.:,i,,.,...,.:$.,
. . . ...

CHARACTERISTICS

Output Capacitance
(VcB= 12.5 Vdc, lE=O,
FUNCTIONAL

f= 1 OMHZ]

TESTS

Common-Em!tter
Ampllfler Power Ga[n
(VCC = 125 Vdc. Pout = 45 W lC(Max)

Input Power
(Vcc = 125 Vdc, Pout =45 W f = 470 MHz)
Collector Efflc]ency
(VCC = 125 Vdc. Pout =45 W, lC(Max)

4.s;,.3 4 54
\f,\ 1~., ;R+
, ,\\ ~i~!. \?&I
,;\~
13
in ,*.>ia
v
>.
.,. ,.*,...
,.,.. c+,
55
60
~,y,~...
:,f,
,,:>,.$., , $
*
sy.t:.
No Degradation
,
GPe

= 5.8 Adc, f = 47o MHz)

= 58 Adc, f = 470 MHz)

Load Mfsmatch Stress

(VCC = 16 Vdc. P,n = Note 1, f = 470 MHz. VSWR = 201, All Phase Angles)
Ser#es Equivalent Input Impedance
[VCC = 125 Vdc Pout = 45 W. f = 470 MHz)

,,?

.!
+,*
. ...
~.t., ..,~td+.
,~,. ,,,}$:,
..JZ:.,..s

Series Equivalent Output Impedance

(VCC = 125 Vdc Pm,,, =45 W. f = 470 MHz)

ZOL*

dB

15

Watts

~/o

n Output Power

I 12+,2.8]

test 8, amOm P!15hed ( the standard

ohms

tesl

00PING OENSITYCB (cm-3)

Figure

2. The Effect of Curvature

and Reaiativity

on Breakdown

Voltage

344

- ____

,,

DC parameters

such as hFE and COb (Output CapaCltaflCe)

need little comment.

Iy unimpotiant

Typically, for RF devices,

because

the des!red frequency

the functional

of operation

that DC beta is related

WIII track

DC

Generally

RF device

Iimlts placed
poslfion

beta

IS specified.

to AC beta (Figure
paflicularly

at

Primarily

of gain at

Note, though,

3). Functional

lower

manufacturers

on hFE.

hFE IS relatlve-

parameter

RF

gain

frequencies.

do not like to have tight

the reasons

that jusbfy this

are:

a) Lack of correlation

with RF peflormance

b) Difficulty

m wafer

c) Other

!n control

device

functional

processing

manufacturing

performance

constraints

dictated

specs which preclude

by

tight Ilmds

fOr hFE.
A good rule of thumb for hFE IS to set a maximum-to-minimum

ratio of not less than 3 and not more

mlnlmum

hFE value

functional

determined

than 4 with the

by an acceptable

margin

In

gain.

The value

of V(B~~&j&
sometimes
misunderstood.
Its
i. ,.,
can approaq@.@@<$Ven
equal the supply voltage rating
..,,,. ,$,:,i
of the transistor. Th@~uestlon
naturally arises as to how such

value
40
POWER GAIN

a low volta~~~~
V(BR)C@@,~#~e

30

be used

(n practical

breakdown

voltage

applications.

First,

of the collector-base

junct:~n$lus$h>
forward drop across the base-emlffer
]unc,.,.,,\\,..
tlon$~,~
Wb base open, and It is never encountered
in

10 I

amoti%s
.,.

where

;&~lffer.That

\\

the base

IS at or near

is to say, most amplifiers

#`'JkQ$~fieVUSea

1
5

1
10

1
20

I
100

50

involves

1
200

FREOUENCY,MEGACYCLES

the current

freqUeflCy
,,, IT

.~,*,*:*~\

Output

size

capacitance
parasitrc
metal

. . $. ~~~.l~,~.
cOmparisOn

(base

area)

IS created

capacitance

provide$~~+~)ority

by the b:s$~.mfbr
arlsingfr~~:w~k$

not vary. Also,

voltage

atwhich

(FtgureJfl)

in cox~~~ev[ces,

exists~~$~~eferred

transistor

V~Cra%yi.e.,

capacitance

WIII

apacltance

wtll

one should

a~l~oc$pacltance

standard

ture

pads and other top

~}k~parasltlcc

note

mum

the

at Motorola

die

case

temperature,

ture

of the device
(PO)

IS measured
(see Figure

#.~~:*
,2,, .J~xlmum
ratings (shown foratyplcal
RF power transistor
.;,.;$,
k.~Flgure
5) tand to be the most frequently
misunderstood
~@r~
p,,+. Kl:$
,.],~..i>..a.a.
.:)*
.>>

group ofdevice
vo/tages
values

Ratings

fo~ardand

In question

down voltages,
junctions

potentially

meets

then voltages
to reach

destructive

thespeclf!ed

m!nimum

less than the mmrmum

reverse

current

voltages.

bias breakdown

Ievels

lnan

easily

to compare

determine

!t operates

max!mum

Normally

several

case tempera-

Thedletemperatu

readings

PD

of 25:C)

at or near

using an tnfra-red

of the die and an average

ress-

assumlngamaxlmum

case temperature

RFclrcuit.

simultaneously

devices,

of thermal

done by monltormg
while

everyone

the rating of maxi-

rated
re(TJ)

microscope
as small as

are[aken

over

value IS used to specify

lJ

formax/mum/unct/on

for breakdown

that case temperaprovldlng

6) wh!ch has a spot size resolution

1 mil in diameter.
the sutiace
It is true

simply reflect the minimum

set forth in the DC specs

thedevtce

cause

specifications.

arestraight

However,

stated

tlJC is normally

(Tc)

output power

*.:},\,..

can

is closely
maximum

a kmd of figure of merit

KnOwlng fiJCand

one

as

at RF

at DC
(PD)

Actually

w[th a determination

on the previously

Measuring

etc.

number

((IJc).

tool wtth which

rat!ng begins

[based

12.5 vO1tsfOr 12.5 v01ttranslst0rs

and 28 v~i@r
28 volt transistors,
+.!t .:+.,,,
~ >.
~

drss/pa//on

resistance

at 25C.

PD IS a useful

tance d]eto

is the

than the value

power

at the val~ie In a s]milar manner,

The

V(BR)CEO

and Increases

It IS based on the assumption

is maintained

arrives

tsspecif led. No Industry

voltage

greater

PD is m reality a fictdious

and not

Second.

the ValUe Of V(BR)CEO

ratlngfor

with thermal

because

Of OutPut
junction

of the dte. Remem~~}?~a&junction

vary with voltage

associated

ISan excellent m~~~yhpf

capacitance

of device

Thus

is always

Themaximum

V(BR)CES.

gain of the transistor

Increases

frequencies

>~max

of the

!0WVal"e0freSiStaflCeSuchthatthebreakdow"

i~~o:,$$$~value~f Interest approaches

02

the ootenttal

have the base shorted

[f

break-

Apoorly

result

hot

40-50C

WIII not

that temperatures

tO20C.
In

Likewise,

in hot spots

with the

properly

that can result.

345

spot

destgned
(worst

case)

poor die bonds

but these

designed

over

are

a dle WIII vary

dle(lmproper

could

temperatures

that

(see Flgure7)

can result

not normal

and assembled

typically

ballasting)

characterletics

transistor

die,

va~

of a

The RF Line
NPN Silicon
RF Power Transistor
designed

for 12,5 Volt Volt UHF large-signal

and commercial

FM equipment

. Guaranteed

operating

440, 470, 512 MHz

Outout

Power

Mtn[mum

amplifier

applicabons

in industrial

to 520 MHz.

12.5 Volt Characteristics

m

= 50 Wafts

Gain

Efficiency
lRL=l

m,,,.

= 5.2 dB @ 440,

470 MHz

= 550A @ 44o, 47o MHZ

OdB

. Characterized

with Series

Equivalent

Large-Signal

Impedance

Parameters

from

400 to

520 MHz
. Built-in
. Triple

Matching

Network

Ion Implanted

. Implanted
. Slllcon

Emttter

Nitride

for Broadband

for More
Ballast

Operation

C~nslstent

Characteristics
, .; \

Resistors

r.$}*l.\f,
,,!3?>\ *.:*>.

Passlvated

. 10OO/. Tested

for Load

15.5 Vdc,

Mismatch

Stress

at all Phase

Angles

with 20:1 VSWR

J
~

@$ $~

2.0 dB Overdrive

*;f {?

,:..,
. .,:~.
Collector-Emitter

Voltage

Voltage

Collector-Current

165

Vdc

VCES

36

Vdc

vEBO

40

Vdc

Ic

12

Adc

135
0.77

Wlc

,,.,..,,. ,.,
,.y .:~:;:~
,.,~
.<,
., ~.. , ~f+

Oerate abOJe 25C

Storage Temperature

?;.*,,.~:,.,$+
\,, ., .,, *
;, ~.<T>:},
.
!~J, ..>,,,
... .

Range

CHARACTERISTICS

Ch,:\~~r:$ic
Thermal

Value

CEO

Continuous

Total Devfce Dlss,patlon @ Tc = 25C

THERMAL

Symbol

.:.,.,<,
~),. : :,..,,
.,,,.,.,..,,
,.,.
\v
.. .

Collector-Em! tier Voltage

Emitter-Base

,.~:,
,:

$;.

Rating

Resistance,

Junctton to Ca&%,

Unit

watts

Tstg

-65 to +150

Symbol

Max

Unit

l)JC

1.3

cm

,T v.
Fi~ure 5. Maximum
Ratinga
:!.
~,.r.+J:*\t>
k,?,{.
a*,\
<:J
$J.y$,.
By measuring
T~.~~~d TJ Along with PO and Pin bOth
s, ~..~y .,!s
DC and RF on~p~,$~alculate
HJC from the formula
eJC
= (TJ -

TC)/(~l~.&$~O).

translstor,,~$~~~~
ic

= l$~~$~ri(RF)

= k~~$ze$)l(l
k$f~~{a!
$~~.~~
@rid

reasons

realrze

hot a worst
ValUe

= 10 W

o + {12.5

First, thermai
we

Typical

TJ = 130C;

of 8JC,

PO

x 12} -

dictate

for

= 50C;
(RF)

And,

third,

a realistlc

value

Motorolas

practice

aPProxlmatelY

25A higher

~JC

value

be placed

RF Power

surements
case

power

= 12.5 w

= 50 W. Thus

Increases

PD. Generally,

RF

30)= 80/80= 1OCM.

is NOT

number.

an
VCC

a conservative

resistance

Tc = 25C

case

values
TC

of a Typical

descrtbed

Illustrated,

Now

a few

Reliability
(gold)

Second,

by using

a conservative

IS determined
is to publish

than that determined

TJ is

for maximum
6JC numbers
by the mea-

In the preceding

a value
words

are

system

(die top

TJ max is determined,
pD

of 6JC

considerations

In order
dictate

metal

paragraphs,

or for the

= 1.25Cm,
about

a safe

and

along

with

= (TJ

(max)

wire)
a value

die

temperature.

value

for an all Au

to be 200C.

Once

for HJC, maximum

IS simply

PD

with temperature

realist(c).

Tranaiator

Specifying
necessity
25C.

The

Max/mum
jectlve

(max)

maximum

to derate

maximum

deratlng

factor

co//ector

maximum

PD

TC

25 C)/f;JC.
= 25C

PD for any value

IS simply

current

rating

for

(Ic)

the

data

to

the

of TC above

reciprocal

is probably

on the transistor

leads

of flJc!

the most
sheets.

subIt has

../_ ,.

* *._1_-

transmission

coefficients

a 50 ohm

system.

the

reflection

input

VSWR

by the

tikewise,
clent

the device

Figure

t3.

coefficient

equation

IS directly

VSWR

related

is also

the

on data sheets

the

= (1 + ISIII)

VSWR.

gain

as lnsetion

of the

/ (1 -

Gain.

gain
are

of the device
50

be acheived

impedances

(which

coefi-

ohms)

ohms

to 50

is the

larger

func-

greater

It IS referred

the linear
is the

useful

to

ohms.

are

when

An

by matching

ISIII).

transfer

Note, however,

is the power
impedances
always

input

ISZ1 12, which

device.

of

to

reflection

of the input-to-output

power

into

magnitude

related

of the output

to output

of the magnitude

is embedded

ISI11,

is directly

1S22[, the magnitude

s~uare
tlon,

when

See

almost

device,

the

always

the

and

load

in gain

can

input and output

different

of matching

lower

to use

source

the devices

by means

need

the

improvement

from

neWorks.

impedances

matching

50
The

and the

networks

to achieve

ga[n.

that IS21 12

INPuT
1

A1

Zo

LINEAR
WO-PORT
BI
91

,,,

,:..~i~.$,,\\

Another

gain Specification

is called
Gam

Associated

IS GNF.

matched

on

some

Unilateral

Gain.

GU

IS the gain

max

Input

and

power
GU

transfer

max
GU

Simply
which

max

expect,

%@Slstor

when

the

thi:~#!he

<*s&hm

~w~,~

(1 -

ga{n increased

matching

RF.t~~~~8Wer

Chart,

then

stable

and the device

by a factor

minimum

by

But

able

data

sheet

maximum

Daram#~&rS;klated
to noise fiaure
NFm,n IS defined as the
rni~ms~
@ise figure that can be achieved with the transistor.
*..w.
. ,gve
th!s NF requires
source
Impedance
matching

reflection

.~~~i~

IS usually

~~%!mum

gain.

?s always

gain

and

A typical

Noise

Smith

Figure

Chart

that

contours

plot

required
noise

to

Chart

which

tvDicallv
show;ng

be

drawn

are

is uncondition-

are outside

IS considered

outside

the concontained

to

must

be

concern

the normal

the Smith

<conditionally
himself

with

frequency

range

includes

Noise

Parameters,

input reflection

figure.

Its symbol

if you

match

is r.

or sometimes

rOpt,

you are likely to have far less gain than

is achiev-

gain

The

is normally

coe~cient

for

value

WIII

to achieve

reflection

transistor.

this

a value

coe~cient

of input

by the

Another
which

then,
tool

plot of constant

can

circles

then the device

designer

are circles

wtthin

Input

reflection

called

rMS,

maximum

coefficient
while

gain

the

for

output

is normally

called

rML,

ach!eve

amplifier,

gain and NF. A useful

IS a Smith

conditions

transistor

of a low

between

compromise

ooeratina

from

des!gn

a compromise

to aid in this

soecific

different
The

several

as

sheet

noise

coe~cient
noise

to

used

device

pafllcularly

remember

low

led

are

Chart,

gam

of the gain circles

be given for the optimum

again

If the

contours

complete

of operabon.

the output.

transistors

ampl!flersk.~~~%has

the

14, These

or partially

Chart.

If poflions

lnstabtl!tles,

1S2212)}.

the input and increased

which.~~s~ts

Smith

If the data

IS1112

in Figure
totally

wtth[n the Smith

ally stable.

.~4%&rs:

= ISZ112 ~f~l+~

stated.

Many

b~~e

of the

entirely

conjugatqj~.wa@hed
for maximum
,,+.l...
.. <e+!.,,
S12 = O.~ ~h~ E-an derive a value fOr

scattering

represents

a factor

IS GU max,@~~@~tght

is shown

are either

fines

of the $~~~~~hen

are

(and

using

which

Yet anoth~~+~~-~erm
is
,l,,,\.
,,.
It IS,,,~w
Maximum

and

achievable

contours

data $~et>.,.

used for @*!&

gain

figure,

sheets

Ifs symbol

output

the

noise

data

on low power

The symbol

It IS simply

for mlntmum

shown

shown

Gain.

important

is given

symbol

Sometimes

in tabular

ized

ohms

to

50

sign! flcance

for

noise

the

determines

form,

In whtch

of rn can
no{se

parameter

NF

resistance

IS expressed

you may

see this value

case

be seen

figure

IS noise

Rn and

in ohms.

it is designated

in the formula

of a transistor

normalrn.

below
for

any

The

which
source

bias and freauencv.

constant

gam

and

reflection

NF

coefficient

[-s [f the

three

noise

parameters

350

~~+,,

+i50

Vfl

VCE=6V
l~=3mA
f= 2000 MHz

-150

(A) F = 2 GHz

NFmln,
f!gure)
the

rn and r.

MRF942

NF

data

= NFmin

The

(the source

are known

locus

Typical

sheet

+ {4rn
of points

resistance
noise

are shown

Irs

-,150

r012}

for a g[ven

AREA

OF INSTABILITY

for minimum

parameters
[n Figure

/ {(1 -

from

15

,~e

d~ln

..,: %s
~ ;@mith
3.

c!rcle

being

a point);

thus,

by choosing

of NF one can Plot a seres

of noise
Chart. Incidentally,
rn can be measured

Kqh::nolse
figure
[1 + rqj2JL:~,w
above,

lrs12)

NF turns

noise

taken

for

r~

= O and

apply[ng

the

different

Crcles on the
by measuring
equation

stated

out to be a c~~

.-..,

..>1
$v>)A param~-~
found on most
.><.<,.<,..
common%
%(L$8 the current
sy~~~%~.,

Somebmes

QY~,j%yse

tt IS generally

,+:,f~u~cy

current

RF low power

data

gain-bandwidth

sheets

product.

It is referred

to as the cutoff

thought

to be the product

gam and the frequency

at which

is

things

The

frequenof low

the current

frequencies.

dev[ce

designer,

emitter

and base dlffusrons

other

h!gh f~ mean

factors

decreased

and it means

being

equal.

spacings
shallower

are

more

complete

shown

in Figure
begin

dlfflcult

to achieve

max!mum

Another
Class A)
device
Gajn

between

ITO

351

power

transistor

collector

current.

17) will increase

to decrease

!nformatlon

achieve

diffusions

low

a plot of f~ versus

useful

To the

RF

include

then

s. Q$Ifi becomes
umty. While th!s IS not prec{aely
true (see
.,m
...<..
.
, ~:~gure
16), !t IS close enough for practical
purposes
And It
,\>>,,,.,
IS true that f~ is an excellent
figure-of -mer!t which becomes
.,::,:*, <::>
>, :.\\,&,
useful m comparing
devices for gain and noise figure capabili:;+f,
,,
ty. High values of ft are normally
requtred to achieve higher
ga!n at higher

wh!ch

n mak[ng

an

RF transistor

Its

about
device

the optimum

Compress[on

thereby

current

associated

has to do whh the degree

Po[nt

it IS sometimes

will

a cuwe

(as

flatten

and

reveallng

w!th whtch to

gain

IS I!near. Most common

as

sheet

current,

as IC increases

group of characteristics
transistors

with

data
Such

with I!near (or

to which the

are terms such as PO. 1 dB

and 3rd
called).

Order
More

Intercept

Point

WIII be said

(or

about

+_hfeo

\h{el

~1

. t

hfed,~

ldBGAN

COMPRESSION POINT

--------

SLOPE REGION

1,0

GAIN

hfeJ2

;_:_

2.0

EXTRAPOUTED

L_:~T

ii
Is

zf~

I
fT

I
f,

-f

= MAGNITUDE OF SMALL. SIGNAL COMMONEMITTER (CE)SHORT-CIRCUIT (SC) CURRENT

GAIN,hfe

WHERE Ihfel

hteo

fB

= LOW.FREOUENCYVALUE OFhfe
= 3dBCUTOFF FREQUENCY FOR CE, SC
CURRENT GAIN

IT

= TRANSITION FREQUENCY =lhfel. fMEAs

WHERE fMEAs =FREQuENCY OF MEASUREMENT

(NOTE:2slhfels
fl

~]

FREOUENCY AT WHICH Ihfel= 1

Figure

16. Small
varsus

Signal Current
Frequency

Gain

.:L....
W@ULES

LINE*

Let~#~~%~ow
tion~iefi~untered
A~~~&A

that

Intended

-to
c&patible

w!th

using

a push-pull

common

emitter

stage

Clrcu!t configuration

[ndust~standard
are

Iosses

in Figure

24

to boos~

volts

RF

with

of frequency

the frequency

@llecWCurrent

<::,.~~

a
a

A basic

are

from

packaged

in Figure

signals

m an

20. Because

that

have

of coaxial

been

cable

(the

thelrgamcharac-

are very

These

impotiant.

of slope and flatness

band of interest.

the difference

that

19, Most operate

frequency),

are defined by the specifications

are

in which

base stage,

(n long lengths

[ncrease

as a function

Remember

and

shown

the

by the losses

of which

terlstlcs

IC COLLECTO#%UR&NT (mA)
.*.$\:,x~..,f.Y
17, Gain-~5d@dth
Product versus

of

They

of gain each

of 2 transistors

a common

configuration

used

attenuated

drives

is shown

voltage

stages

configuration.

Se-

characteristics

requirements

of 2 or more

cons]stmg

specifica-

applications

appliCatiOnSare

nemork

cascode
IS one

some

to specific

gain and distofiion

cable

stage

they

unique

consisting

cascode

a standard

Characteristics

examine

forcabletelevlsion

the

amplifiers

and

are

have the desired

+: knear

Figure

Functional

toamplitiers

Slope

is defined

over

simply

as

in gain at the high and low end of the frequency

non-l lnearltles
about

L}near

thatPO,

an&s~{@~on
measurements
In the sectton
.,..
?~~&rs;
however,
suffice It to be said now

1 @.,@@

power,,~&, WCfii?he

~ompression
input

power

PoinYis

simply

hasagaln

band of the ampkfier.

asthedevlation

the output

associated

gain wh]ch Isdetermined

with

loss funct!on, Motorola

it that+s 1 ~&less
than the low power gam In other words,
?,s.
th?~~~-~:ls
begmnmg to go into saturation
which is a
c%~kfi

where

increases

in

$w.@ses
in output power. The
?~lllustrated
lnFlgure
18,
The

impotiance

dynamic
ofdynamtc

accepted

]n a linear

range

power

of the

range

to realize

(high-to-low)

variations

but spec{fies

the flatness

CompressIon

PoinV

as the Iimlt of non-lineatity

and

Iowpoweramplifler.
imposed

leads

one

that
to

On the low
by noise,

In the Industry

and on the

MHz

352

ingainacross

and

overall

system

have bandwidths

WIII accommodate

designers

of a CAW

6MHz

slgnala).

band,

because
have

gain level.

of channels

requires

cable

the peak-to-valley

the frequency

band requirements

color TV

from anldeal
universal

as a PIUS, mmus quantity

by the number

Each channel

bya

measures

that cable television

conventional

end

normally

of adjusting

system.

the

mtheband)

theoretically

[t IS assumed

determmed

IS

on the other hand, IS defined

the capability

The frequency

ampl]fier

is the limit

fall

concept of gam compression

of the 1 dB Gain

that this IS generally

IS tolerable

Input

Flatness,

(atanyfrequency

used

amplifier

bandwidth

Currently
extending

(to handle

ava!lable
from

up to 77 channels,

are

!n the CATV
models

40 to 55o
the center

considerlnq
we can

the first three

F(x)

= CIX

where

Flsthe

Hgure

19. Basic

CATV

i e., make

the assumption

+ C3X3,

+ C2X2
output

signal

C2 and C3 are constants

(gain)

terms,

write

for the first,

andxisthe

input

that represent

second

and third

s[gnal.

the transfer
order

Cl,

function

terms.

Amplifier

IWERwAmS)

~gure

frequencies

of which

frequency

CATV

must handle

aredetermlned

many

to the s]gnal
being

(lMD),

Composite

Triple

Beat

what

terms

mean,

these

dlstotiion

signal
gam
under

signal

If we

apply

In general

th~~~~~,ef

need

except

transistor

knear.

if one

kQ ~dtel~- transfer function

for a transistor
ampllfier,
a
~ -~! .,,.+$
ty~~~$.,~~~>.output
cuwe for which IS shown m Figure 21,

and

~:%?~f!nd

or

the region

,.: by;squared

terms,

output

is propoflional

to the

tal increases
I.e., the

m input

output

mathematically
an amplrfler
terms.

is best

is propotilonal
rigorous

would

However,

represented

include

to the

analysls

cube

of the

of the transfer

an mfmlte

an excellent

by cubed

divide

number

approximation

funcbon

of higher

pansion/compression

of

the amplitudes
that

by

353

exist

there

[n this

orlglnal

tr!ple

compo-

beat terms

reveals

add[tlonal

frequencies

enhance
The
are

gain

of fl,

amplitude
such

f2

(expansion)
of these

that

we

can

self-expan-

cross-expanslonlcompression.
terms

terms

compoare called

are frequency

examination

of a single
have

of two frequencies.
simple

( these

are called

terms

involving

the frequen-

frequency

into two categories

and

by the amplitude

the terms

are also

(compressIon].

of terms

2nd

at three times

frequencies

both

compression

the group

mmed

of all
order

at the

Self-expansion/compressIon

order

is obtained

gam
and

sionlcompresslon

terms,
Input.

reduce

are

called

there

can

the

by x2 terms

are

Llkewlse,

A close

terms

created

frequencies

But in addltton

components

on a

f3. These

difference

IS not alll

analyzed
at two t!mes

and difference

+, f2 +,- f3. These

f3. These

expansion

le., where
,,,,.+$i?~<,,$uare of the input. And the region near saturation,
,.,,:,
,
. ~Ni.,& the ampllf{er produces
less Incremental
output for mcremen:>,));\
Vi.+..>,.,~!*
.> !..*,F
~:~,.,.
?.

this

and

x, x2 and

Itnear aMpliflCatlOn

components

components

f3. And

and

IMD).

amplttude

near zero to be one best represented

I e., the

f2 and

at fl

And

were

f2 and

function

x2 when

f3. Also

components.

at sum

nents

even

In signal

The

+ Acost,}3t.

revolving

the pedec~,

at sums

of fl,

transfer

terms

involving

f2 and

(n frequency

3rd order

for a constant

not petiectly

of fl,

of fl,

nents

The output

are,

result

6 MHz.)

+ Acos(,,2t

to the

many

Terms

components

X3 result

signal

represent

mtermodulation

understand

amplifiers

Input

approaching

Acost,~lt

we WIII find

basis

combinations

to be sa[d about

ampllfier,

as the input

circumstances,

signal

and

equals

signal.

frequencies

cies
Imear

factO{,,*UnfO&flately,

frequency

IS the quality

to better

~fe~words
.? .,.,,.
~fr ~\-

thts

F(x),

of the input

Dls&~.~$generally
speci,7.,..:, ,,
- ,.*~
@,..QMer
Interrnodulation
s.,,,..,!+

(CJ8)>~J,#rder

lets con-~~.~tiectly
\.:?,,
IS exactly~e&me

dlstoflion

(XMD)

over a bandwidth

x then

x3. The ,x terms

~~s&&mpl!fiers

MoQyl~n!bistortlon

spread

calculate

and they

lkne~r,%d~s

and, thus, t~~pti%

ways

Cross

TV sI@~!s

simultaneously,

The more

viawed.

In 3 conventional

Dlatofllon

must ampli~

channels
linear.

of the TVpicture

Flrst,

by industry

~.! .
-(!?+
:a:$quen;es
$::~,~.!.
.{},
input
stan&~@

,,,.>

ampkflers

be extremely

that IS added
fled

Package

allocations.

Because

must

20. Standard
CATV
(Caae 71 4-04)

example

have

amplitudes

frequency

while

amplitudes

determined

A summary
IS given

detercross-exby

of the terms

in Table

1,

Table
Terms

in Output
Signal

FIRST ORDER
kiAcos

1.

40

for Three

Frequency
T

at Input

COMPONENTS

a + klBcosb

+ klc

Linear Ampltflcatlon

cosc

SECOND ORDER DISTORTION

k~A2/2 + k~B2/2 + k~C2/2

COMPONENTS
3 DC components

k2AB cos(a+,b) + k2AC cos(a+,c)

k2BC cos(b+,c)

+ 6 Sum & Difference

k~A2/2 cos2a + k~B2/2 cos2b +

k2C2/2cos2c

Beats

3-2nd Harmonic

Component

THIRD OROER DISTORTION COMPONENTS

k3A3/4 c0s3(a)+ k~B3/4 cos3(b) +
3.3rd Harmon!c
k3C3/4 C0S3 (c)

Component:

3k~A2B/4 cos(2a+ b) + 3k3A2C/4

c0s(2a+,c) +
3kSB2A,4 cos(2b+,a) + 3kqB2 C/4
:os(2b+,c) +
3k~C2W4 cos(2c+.a) + 3k3C2B/4
cos[2c+,b)
3k3ABC/2

12 Intermodulat!on

Beats

4 Ttiple Beat Components

COS(a+.b+ C)

3k~A3/4 cos(a) + 3k3B3/4 cos(b) +

3k3C3/4 COS(C)

3 Self CompressIon (k~ IS +)

or Self Expansion (k3 is )

.,:,.,,..
,

3k~AB2/2 cos(a) + 3k~AC2/2 cos(a) + 6 Cross Compression (k3 is

3k~BA2/2 cos(b) + 3k3BC2/2 cos(b) + or Cross Expansion (k~ is )
3k3CA212 COS(C)+ 3k3CB2/2 COS(C)
Before

gotng

Into an explanation

on I/near amplifiers
to rev!ew

tudes

such as CATV

a concept

mathematically

called

signal

level,

tudes

while

that

level

are

Hence,

on a log-log

scale

(or dB/dB

corresponding

mental

responses

responses
a slope

and 2nd order

3rd

order

distoflton,
change
signal
This

p> determine

of the

a@~j:<

input,~ndh

of 1 and ~NW.&n
fundamental
,.,+.\.., .,f*..
of 2, ~,at ~ to say, for 2nd

order

results

level

d!s?;~o~

however,
-.:
,.l,.

:~~,,2

d~change

gr~wll~

in Figure

d% ~~yrd
<. .,<:
tb
Q#}&t\evel
,,\ .:>}.
shoul~~tiproye
to 40
to -80

(90

in s19nal has improved

point

(or 3rd) order

each

other.

on the

plot

response
It IS also

where

wh{ch the level ofd!stotiion

For example,

]f at our point

would

and

Change
in

values

of

dlstorfton
dBc.

the intercept
the

have

even In this situation

and

a 10

distortion

rasponse

of stgnal
equal

lines

and

With

by

this

sheets,

practice

at

Cmsa

level.

will

the 2nd order

examine

Order

these

Intermodulatlon

one must
level.

(expressed

Both

in Figure

the

and 3rd

22 using

Note,

circuits

spurious

responses,

the concept

of Intercept

to

from

one

used

etc.

that

However,

po!nt IS still valid;

is to specify

Information,

a 2nd

Intercept

lets

on many

turn

RF

Order

stated,

at a bme.

Distofllon

First,

(lMD).

order
point.

specific
amplifier

telev!slon

it IS common

lntermodulation

and Composite

to

linear

are for use in cable

previously

Second

distoflion

are

the 3rd order

listed

Distoflion

the

also, that

are still 1, 2 and 3 respectively

be done

as

is
3rd

Intercept

point for 2nd and 3rd order

unless

and

level

2nd order

examples,

Then

in dBc)

as the 2nd order

illustrated

If the amplifiers

to specify

point,
signal

dBc,

background

Modulation

means

is 40

different

systems,

typically

carrier,

distotiion

specifications

distribution

intercept

the lnltial signal

of measurement,

data

needs

point

distotilon

2nd

(hypothetical)

all that

intercept

by the

order

of the responses

can

governed

signal

value

even-order

dlstofl!on

2nd order

same

leVe[

If the

In the preceding

In general

you

example,

are

will

but 3rd order

we see that

points

re-

of Intercept

For

IS the same

of

order

Signal

any

Intercept

+ 20 dB = 30 dBm.

assumed

suppress

[f we

3rd

point

Intercept

the slopes

Thus

the

a 1 dB

dlstoflion.

level.

2nd

for

which

or

IS the value

and

at a known

of the distotilon

signal

order

the Wo straight

value

to the

dlstotiion

of fundamental

that

add

in a 1 dB

:$i$gb~
and 3rd order distoflion
has Improved
by 20 dB.
~ Now for intercept
po!nt. We define the intercept
poin~
%s the

half the value

results

is at 60

dBm].

take

then

dBm

point,

the

3rd order

dlstotiton

10

intercept

stated,

to determine

dBm

of operation

60 dB below

3rd order

then 2nd order

dBc (50dBm)
dBc

relationships

intercept

the cuwes

In a region

than

greater

IS 10 dBm;

of the concept
the

dtstofllon

mathematical

or 10 dBm

in 3rd order

of

IMDs

point

distotilon

order

to 1 O dBm,

are

order

2nd order

IS at 30

beauty
know

and

IS O dBm,

change

The
you

value

+30

fundamental

once

order

22, Using

22 that

the

level

is 40 dB above

cross.

you

+10 dBm

fundamental

at which

the

measure

a slope

po!nt

in Figure

IS that

Mkewise

of the r~,:-*~Funda-

and the signal

intercept

~te

provided

o~<~?.w~nd
order
,..
of 2 and t~:3r&qr&er
responses

22, If thq@,$~uFlevel

~$kease

of the inRut

IS a slope

in 2nd

wi~~j@~a&

Pwnt

WIII be a str~WFll,T&wtth

a 1 dB chang~+m,+$w

level

have ampli-

have

that the differen@b*&n

IS a slope

IS shown

Figure

scale)

WIII have a slope

of 3. Note

cube

2n@$&\$er

w!&!
$&nal
,%t,h,.,+
+:~x:e
lines

that a plot of ea:h(~,~~~e

to the order
WIII have

products

4~kdBc

+30;@~<i

It IS appropriate

to the square

to the

the

petiormed

products

dlstotilon

propodlonal

distotiion,ds

point, It can be shown

d!stoflion

It can be concluded

tests

amplifiers.

propotiional

3rd order

a slope

of the

intercept

that 2nd order

that are directly

-Ecp--Y--

COMMENTS

Triple
consider

Remember

Distotion,
Beat.

We

Second
these

are

354

unwanted
two

signals

frequencies
a

created

by the sums

present

in the

specified

at

channels

MO for input

the resulting
are

given

distotiion

standardized

that typically
distoflion

results.

listed

Iookmg

the 3rd channel.

that

would

tlon

64

if the

Intercept
an

for a VOut

function.

of the

presence

caused

by the

carriers,
each

The

Then

and the presence

amplltude

er levels

available

are shown

In Figure

for

Vout

an IMD

of

,@*;:< ~ ;
I ndustr!e

of

the

generated

by

i which

wave

consists

of

modulation

RXt2

dletotilon
present

invQ,~?~$~.,4Re
~et ~~he

one

~$~:ad

is measured,tx?:i,,

detector

relationships

from

such

measurement
output

F, F1,
level

level

levels

are

along

Fm,
with

6 dB below
frequency

then

F is
F2

exist

level

F, FI,

F2 and Fm

the frequency

In the

and

Fm

and power

DIN test,

K
g
m

dBc

%
g
I

POUT

I
I

FZ

of F,

Frequency

F1

dtfferent

F2 = F 12 MHz

25 !Ilustrates

that

the

+ dBc

dBc
triple

F1 and

FM

between

F1 = F 18 MHz;

relationships

normal

F2 and

that,

to a level

= F + F2 - F1. Figure

,$..

the

a 40

from

level 60 dB below the level of F, and FI

adjusted

are as follows:

beat measurement

If we call the four frequencies

?*Distotion
IS measured
at the
,.,. ..*~\:
~QW$,. N relationships
(used by Motorola)

analyzer.

,+

F2 are

that

signals

the stan -

DIN45004B

to produce

difference

is the fact

to:a distotiion

d%

triple

adjusted

Wd!tional

requtred

and

IS #45004B.

as t~~.

in the XM~%~~$
..),

23.

Ie~.@,.An

beat mess-Ys
,.,,b,,\ ,$,
for the t~?e~$d~blnmg

modulation

is removed

(AM)

carrter
of

Standard)

the s!gna%~ware

distoflton

number

(Germa~,%~~*hl

for C~~,~rn$tflers
...$.3..
case of ~ti~]~hannel

is a special

test is a measurement

square

Norm

dard that appltes

input-output

of a large

of residual

Figure 24. Frequen~~!~~&er

Relationships:~{,~~

etc,

result

measurement

1006

,, e.\
..,,t\;T,,<:.,

that led

if the specifica-

expect

is

in
IMD

(or weaker),

terms

modulation

Matrix

appear

on an unmodulated

modulation

and frequency

that

amplifiers

the XMD

actual

the

channels

the 2nd order

(XMD)

in the

with

of crest.

two

per channel,

contribution

carrier

consists
first

Max.

dBmV

Distonlon

channel

commercially

would

= +50

with

an

dB

one

of modulation

modulated

kHz.

68

distotilon

%
~

stronger

of

In general,

m a manner

the 1:1 relationship

then

E
g
~

for the 2nd order

In other words,

non-llnearity

modulating
15.75

were

combinations

products

and cross-expansion

order

transfer

at

Point.
IMD

Modulation

third

In the

to predict

signals

involves
to measure

but selected
condition

measurement

to remember

cross-compression
the

case

for the distotiion

per channel,

dB Max

Cross

indust~

If one wishes

guarantees

+46 dBmV

one

(unmodulated)

occur

Order

and

and

a worst

it is only necessary
to a2nd

level

The channel

of any

is normally

output

An actual

signals

I MD

frequencies

frequency.

in the

gtves

ing output
and

signal

and differences

amplifier.

DISTORTION
NOISE FLOOR

FM

.
FREQUENCY
F,= F-18 MHZ
F~=F-t2MHz
F~=F+F*-F,
Figure 25. Frequency
Power
Relationships
for DIN45004B

,.?k~
?=
\.}t,\$.
?U,,&@thout modulation.
Then one channel frequency
is removed
:s
.+$:,,.>
c.~.,> end the presence
of signal at that frequency
IS measured.
~II,},.
..,).
The signals existing in the off channel are a result of triple
.{;*%J.,
.?.?$
.::,,,.:!. .
beats (the m!xing of 3 signals)
among
the host of carrier
~:~
frequencies
that are present
In the amplifier,
A graphical
representation
European
tional
DIN

of the

test

cable television

specification
test.

CTB

DIN

for

is shown

systems

m Figure

usually

linear

amplifiers

is a German

standard

Invoke

which
meaning

Linear
tions

amplifiers

aimed

generally

have

will

frequencies.
test

with

Basically
power

24.

signal.

Relative

an addl-

Figure

26,

is called

Thermal

the

Power

Deutsche

355

It IS another

levels

and

power

levels

resistance

modules

at television
another

appltca-

test

3rd order

intermodulatlon

frequencies

ratmgsof

described

transmitter

dlstotiion

that

s!mulate

and frequencies
CATV

modules

in the next sect!on)

revolving

a TV

are shown

In

(as well

as

are, perhaps,

overall gain and mechan!cai

form factor suitable for a particu-

lar application.

[-8 dBJ

VISION

CARRIER

Power

SOUNO
(-7 dB 0,)
CARRIER (-10 dB)

sloEBANo

modules

most significant

SIGNAL -1 6 B)

IS affected
JY

for mobile

unique specifications

factors

and potiable

is that of stabillty.

and

by the

005

01

02

0.5

remain

stable

greater

the range of values

the more

the

AF,FREQUENCY [MHz)
26. 3rd Order

IMD Test for TV

degree

value,

under

stable

several

heat sources

it IS necessary
to the

of the

temperature

output

mlillwatis

modules

such

(75

for

tant

MHW51

doubler

I!near

ohms

(and

power)

matched

CATV

22A

such

for

modules

are

connections

modules

amplifiers

discrete

means

Typically

guaranteed

under

modules

that yo~.s%@J~~

inputs

:~~

in a 2-way

~~or-

Figure

common
that

assembly
ances.

are

consist
emitter,

Class

Efficiency

The

heat sinking

w!th nominally

Selection

of several

the proper operating

crease

Stability

stages

for

for some

Involves
frequency

choosing

eff!clency

(output

power

consumed
output

the module

biastng

and

Overlooking
results

for

these

stages

In an atifically

Most power

by

er slump

In terms

temperature;

ble decrease
output.
other

low level

conditions

Generally
thermal

one having

stage

range, output power,

Some

others

by

It IS generally

efftclency

than

the

DC

will delevels.

output

power

for module

power

use)

effrclency.

Include a cuwe
modules

stage.

they

of output

specify this powoutput

at a stated

state the maximum

(in dB)

referenced

to note the temperature

applied

net

at lower power

the

of a mln[mum

It is impotiant

judgement

integrated

(and

data sheets

in power

the

Input power)

Includes the current supplled

high value

module

maximum

because

other

currents

IS to divide

by the module.

power

specifica-

radio applications.

mmus

IS operated

power versus temperature.

of amplification

in a hybrid

to specify

at rated

when

Important

for portable

Be careful that the specification

the transmtt

level for radiation

way

In RF power

speclfled

that the

50 ohm RF mptit and output imped-

voltage,

Affecting

an Increasingly

m modules

the total DC power

Characteristics

C except

A) combmed

of a module

Factors

IS becoming

correct

increase

IS

be designed

used to amplify

Class

27. External

tion pati[cularly

the

if the fixture

connectors)

radio to the desired

They

f@j~@@r

of Input and OutPut

Functional

are generally

lmped~.:dwl.,

rtant, for accuracy,

(llh@and

MODULES
modules

(usually
stages

and then

bas!s.

to 10

decouple

a~pquate

involvi:,$~~~%~~$~es

then it IS extrems$yi~ti~

antenna.

phase

~~r$~~uction
., ....::

by

to exhibit return los&@+rWter

than 35 dB. Motorola modifies
ill. .,
the RF connectQrSqs@
in the fixture and, then, calibrates
.~>.:.\3,
;i\,
their fixtures, tal~e~re
that the fixture does not Introduce
,**. +t?..
errors in memr~g
module return loss.
~/.>~ ~\\
.*?*:~ ~~x,
...

;$e

during the pre:wd.titlon

but not tested

85.

ohms

test (D,U$~~l@s~ever,

Input and output netw~~s

{$:~~~ls

of loadi*&%R
(the larger the
:* ,, es,.
specif!@?t%are
generally eva-

Stabll!ty

thle

tO@:.fi~,@S
fOr
~ >,.?
R~th@~@C
power
$,,:, ~+.,~.i,
[f ~3~@Ortaflt If YOU

of making

test fixture

increase

have

50

being tested,

I[nes and that you provi~e

,,~~w

IS
as

Basically

DC power

R@

l~o~ance

b~$~lerated

Actual

are generall~~$~$

own

VSWR,

cap

case

consumed

system

and

devices.

simply

but can

build your

for Ilnear

which

IS at most a few

of heat.

to standard

to the module

fOr the de,ce

Of panlcul~~

mismatch

the operating

as the MHW51

test circuits and fixtures

than

the mO~~/~ ~~

of 0p~4@lQflri+,%e

rating, Actual power

modules

in the form

to be cefiam
conditions

runs in the range of 5 wafls for conventional

[hat are

ampllflers),

,,,,,st.,
,,\ ~

Input

heat smklng

operating

that most of the power

as the

for a power

Because

the

rating of the device.

CATV

IS d!sstpated

power dissipation

that

by considering

current

have

of RF

for which stabilit~a$,,~~~teed,

within the amplif)er,

sufftc[ent

its maximum

of most

wh!ch means

outputs

such

can be determined

the module

watts

ampltfier

and the max!mum

RF power

the amplifiers

that are contained

never exceeds

dissipation
voltage

Because

for the user to provide

case

external

by the value

by the stated value

the better).

luated thoroughly
by their absence.

typical

is the module.

of load

evidenced

conspicuous

impedances,

amount

each factor must be considered

i

Figure

of a module

but also by many

such as load and source

voltage

also have
One of the

s!gnal.*.,> , ::*..*,
External factors influencing stabllrty are highlighted in Figu~
,+i.<~:
~ ,$
27. Combinations
of these factors over a range Of ValUe$~$,,w

0.01 002

The stability

not only by Its design

of supply

radios

related to their applicabons.

to the specification

permissi-

to rated

power

range

and the

before

passing

on this specification,
power modules,

reslatance
modules,

specified

there

would

like linear modules,

do not have

from die to heatsink.

For multiple

need

to be a specific

thermal

356

resistance from heatsink to each die. Thermal design of the

module will take care of internal temperature rises provided
the user adheres to the mwimum rating attached to the
operating case temperature range. This is an extremely
important specification, particularly at the high temperature
end because of two factors. First, exceeding the maximum
case temperature can result in die temperatures that exceed
200C. This, in turn, will lead as a minimum to decreased
operating life and as a maximum to catastrophic failure as
a result of thermal runaway destroying the die. Second,
hybrid modules have components that are normally attached
to a c[rcuit board and the crrcuit board attached to the flange
with a low temperature solder which may become Ilqu}d at
temperatures as low as 125C. Again, the power to be
dissipated can be determmed by considering the RF output
power and the minimum efficiency of the module. For example, for the MHW607, output power is 7 watts and input power
is 1 mW efficiency is 40% minimum. Thus the DC power
input must be 7/0.4 =17.5 watts. It follows that power dissipation would be 17.57 = 10.5 watts worst case.
Storage temperature maximum values are also Impoflant
as a result of the melting temperatures of solder used in
assembly of the modules. Another factor is the epoxy seal
used to attach the cover to the flange. It IS a material stmilar
to that used in attaching caps for discrete transistors and,

ICONT.
- 13Qk:A~+~6NT.=gv
Figure 28. ?M,.Power
versus Gain
,z=.of
Volta9e

dynamic range control of output power. Most modules pro~~

gain control by adjusting the gain of one (or two) S@9eS

01
o

for

by means of changing the voltage applled to \~} stage(s).

Usually the control IS to vary the collector @~t~~J applied
to an Intermediate stage. A maximum vti~~
stated on
the data sheet to Iimlt the control vol~$+~$?k?a safe value.
This form of gain control is quite se$~~~.$dsmall changes
m control voltage as IS evidenced

ti~w+~mg
:\:A.

which

base-to-emitter

uses

&,cb$

ally class A) as i~~&$<~,#in

significance

in p~ta~~s

While B@.@~t,@

Figure

because

the control,?@two~J&xternal

stages
29

IS to vary

the

(which are gener-

Thts is of particular

of the power

diselpated

In

to the module

on most data sheets,

it IS always

possible

DATA SHEETS

to com&o~$&*utput
power of the module by controlling the
, $<$.... ,,
R~i~~ut
B?bnal. Normally thle is done by means of a PIN

4
2
3
vcont GAINcoNTRoLvoLTAGE~dc)
l~oN~ = 100$LA@ VCONT, = 4 V

distance from the transmit frequency. Caut\on must be taken

in making measurements of nolee power. Because the levels
are generally very low (--85 dBm), one must be assured of
a frequency source driving the module that has extremely
low noise. Any noise on the Input signal is ampllfied by the
module and cannot be discerned from noise generated withfn
the module. Another precaution is to be sure that the noise
floor of the spectrum analyzer used to measure the nolee
power IS at least fO dB below the level to be measured

the output power

16ss current

volta@:%,{N@)~nput

Figure 29. Output Power versus

Control Voltage

versus control voltage curvSss~&~Wded for the user (an

example
isshown in FI@~$fi<$T.
An alternative
COntrOl
procedure

I
1

OF THE FUTURE

World class data sheets in the next fewyearswtll tend to

provide more and more Information about characteristics of
the RF device; information that will be directly applicable by
the engineer in using the device. Semiconductor manufacturers such as Motorola WIII provide statletlcal data about
parameters showing mean values and sigma deviations. For
discrete devices, there WIII be additional data for computer
aided circuit design such as SPICE constants. The use of
typ!cal values will become more widespread; and, the availabillty ofstatlatlcal data and the major efforts to make more
consistent products (six-sfgmaquahty) will increase the usefulness of these values.

q(o~~~enuator.
Controlling the RF input signal allows the
.. . l.)
rntqule to Opefate at Optimum gam condibons regardless of

:$:t,@tput power. Under these conditions, the module will pro,,.~,i.,.,

,->
?t},$w,~duce less sideband noise, patilcularly for small values of
output power, when compared to the situation that arises from
i.,.
J?l?
.....3,;~,
s,-,
gain control by gam reductton within the module.
>..,>.
,~
No+se produced by a power module becomes significant
in a duplexed radio in the frequency band of the received
signal (see Figure 30). A spec!flcat!on becommg more promrnent, therefore, m power modules IS one that controls the
maximum no]se power in a specified frequency band a given

357

SUMMARY

Understanding

33 dBm

mean

can be a major

about

selecting

application.
sheet

45dBm

tors

Power

are

dlficult
device

FREQUENCY
30. Noise

more

in Receive

efficient

should

contact

manufacturers

the

cfrcuit

358

emphasized
from

the

It is hoped

and

has been

thts

effort

WIti help

with ~Fw~spf~~t?bnd

all inforrn##~~~@ut

explained

RF

fii~:$)&aper.

The ,Q{~,:

manufa@u~?@r
*,, ..,, . .
it IS appro@~@.%st

de~gn

Nor

engineer

more

detailed

if not all current

plifiers

purpose

wherever

have

exten-

of assisting
assistance

st.,\
~<,::k,,

th,s+..,:$~

*>; *>i:y
made

device

whWve@and
.<?. *
::?~.t>.~:x
;, .+;.>.{:,
,,!, ~{.~ .

has
man-

and use of RF semi~o{:,~~}

of RF transisto~>~.a~
.* ,,3
supped for~p*ress

designer

data

and

semiconductor

semiconductor
Not

they

unique

amplifiers

effective.

in a data sheet.
the

what

es he goes

for his specific

some

and

his selection

complex

whenever

sive applications

needed,

mean

characterize.

characteristics

and

designer

RF semiconductors

and the amplifiers

unusually
to fully

Information

make

can all be covered

Band

specifications

to the circuit

has

point-of-view.

The RF transistor

Figure

paper
these

engineer

ductors

asset

of RF transistors

what

ufacturers
circuit

sheet

and using

This

parameters

explained
TRANSMITSIGNAL

data

is

AN1122
Running the MC44802A PLL Circuit

.>,

.,+.
.,t;$~
,<>\.~h.,
,$y,,.,
.?, .>.,\?
.. t.\b,.,
.\,,,
.2.,......,>
~i~ $i~.
,~.
~
.,...,).\.{.
;:>
,,$$
,\{~ .,:Lt
for MCU cent{.+t<$.:-+
.!

Prepared by
Paul Brownlee/Unear Application
Bipolar Analog IC Division
INTRODUCTION

The MC44802A is the PLL portion of a tuning circuit

intended for applications involving television, FM radio, and
Set-Top converters up to 1.3 GHz. Coupled with a VCO and
mixer, a complete tuning circuit can be formed. The tuning
frequency is controlled through an MCU serial interface (12C).
As noted in the MC44802A data sheet, an MCU is recommended for sending the serial control bytes. This application
note describes combining
an MC68HC11 E9 with an
MC44802A in a tuner design. The information IS sufficiently
general however, that most any MCU could be used for this
function. Those with a limited background in the use and
programming of MCUS will find the information adequately
detailed to permit a thorough understanding.

12c

intetiace

Selectable +8 prescaler a@a+$5:bit divider accept frequencies up to 1.3 GHq.

ts
Programmable refer$w;$ivider.
Phase/f requenc~,~%rator
output can be set to high
impedance for~$&%J,ri~,
Op amp prov~*,@?ect tuning voltage output (0.3 V !0
.:iy+.,;;t:%>
30 v).
Seve+@~,ogr$hmabie output buffers (1O mA, 12 V) for
band%@Wing, etc.
@~~$%@tions for 62.5 kHz, reference frequency and the
#* *g*mmable
divider which are useful for system de.
:%~~gging.
,.
,
.!.*:$\..
..,, - Figure 1 shows a simplified block diagram of the
,.i$~ MC44802A. The 12C Bus receiver IS a central block that
A Look at the MC44802A
.,\\~$,\!3,
,, J
controls the HF prescaler, 15-bit divider, the oscillator (typ.
The MC44802A is manufactured using Motorolas high..<a2t16i,\.::$:t.
*. ?.
4.0 MHz crystal) reference divider, and the output buffers.
density b!polar MOSAIC process. It features:
fi+. ~
.,

Hgure 1. MC44802A Simplified Block Diagram

359

transfer is initiated by a master and acknowledged by a slave

dev!ce. Each slave is assigned a unique address, allowing
multiple 12C devices to be connected to a single bus. An example of a data transfer is shown in Figure 2.

The 12C Bus

The 12C (Inter-integrated Circuit) Bus required by the
MC44802 is a serial transfer process using two wires for data
and clock (SDA serial data, SCL serial clock). Each

STOP

START
IDLE

; ;

ADDRESS

OATA

rl

FINAL SUE

,.

Figure 2. Complete

II

.,

~: t+~}

,,+~\*>;\.<.i
,, ..,.p

Data Transfer Process

the master terminates the t~a~~r$~nd generates a stop

Referring to Figure 2:
\..:tt,:::.:
~ ~
Idle When there are no transfers taking place on the bus,
Condition.
.,. ,.
\J$.
SDA and SCL idle high.
The Microcontroller ?JJ
Starf-A
master initiates a data transfer by pulling SDA low
The MC U chosen.~$~~~~~ data transfer must have a Serial
while maintaimng SCL in the high state. At this time all
port with the follo~~;g]ti~racteristics:
slave devices on the bus are listening for their address.
Two-linQ@,$~&and data, with open drain (collector)
Address The first byte is sent to select a slave device(s).
-.+\\:
.
Outp\yts ..~
Slaves that have read and wntecapabilities have aunlque
8-@Wransfer
buffer
address for each. Upon completion of an address trans~
12&Mterface or l/O setial lines capable of emulating
mission, the master must leave the data line high and
$!2C$rotocol (Idle, StaflStop conditions and ACK pulse).
create the ACK clock pulse. The slave device is to ac.+&~ ,Wmicrocontroller
examples are the MC68HC1 1 or
knowledge by pulling the data line to a stable low state bew % HC05 families.
fore the end of the ACK pulse. From this point until a Stop
Condition ISgenerated, only the selected slave(s) device
.

A SAMPLE SYSTEM
. ,:
w.
IS acttve.
Overview
,.*:\L
Data The transfer continues with data bytes sentm thW,,
The remainder of this application note is devoted to describsame manner as the address byte. An ackno,#$~s
ing a sample MC44802A system. From a high level view this
required at the end of each byte (except the l@~~b~,The
system is simple (see Figure 3). Whenever the push button is
master indicates the last data byte bK:l~d,,~Q
the
pressed the circuit responds by changing the tuning frequenacknowledge (low) blt rather than leavW~;~Q~high for
~., .,.,.,!:*
cy, and provides a display indicating the frequency. The followslave acknowledge
.LF\\,+:,+
ing paragraphs describe this system which was built and
Stop The master creates a Stop ~n~$lon
by sending
tested to demonstrate the functionality of the MC44802A. hcluded are descriptions of each segment of this system PLL
tuning circuit, MCU control, user interface and LED displays.

,2~

12

*S* h:+
*,,<.,
,.-,~,:.,.
~\\ ,~,1,,+,,>
..
..~:

3- Mgit
Frequency
Display

Hgure 3. Stmpiified

Block Diagram of the Video Frequency

Controller

360

-.._.._..-_

PLL Tuning ~rcuit

Implementation

The MC44802A works with an MC1 648 voltage controlled

oscillator (VCO). to form a Phase-Locked Loop (see Figure
4). The MC1648 requires an external parallel tank circuit
consisting of an inductor (L) and capacitors (Cv and Cx).
Varactor diodes (Cv) are used in this case to provide a voltage variable capacitance for the VCO. The MCI 648 may be
operated from a +5.0 or 5.2 Vdc supply, depending upon
system requirements (+5.0 V in this case). Its maximum
frequency is typically 225 MHz.
The VCO output is connected through a capacitor to lhe

phase detector input of the MC44802A, With the feedback

network (G(s)) the MC44802A produces a stable voltage
input to the tank circuit. A general purpose open collector
output buffer (B2, Pin 9) is used in this application to switch
a capacitor (Cx) in and out of the tank circuit. When that output buffer is switched low (by writing a1 to it), the pin,diode
(Dl ) conducts making Cx part of the tank circuit (Cx#&$~~@~).
When the output buffer is open D1 does not con~:~::t~~~eby
presenting a h!gh impedance to Cx, making it i~e~~ctJ~. The
tank circuifs capacitance

is then Cv/2,
: it:.{
* .~;.$
e,, $+%
<s ., @w:

MV209

51k

47UF

Cu

0.2 PH i

Cv

001

MV209

LI

1.

Cx
15pF
Tanktircuit
____J

Fosc

D1
MMVB3401L

Figure 4. Sample PLL Tuning Circuit

..:~:~~..k~usInterface in five bytes as shown in Rgure 5, Communica\./.

~ tion of the data is covered in the section describing MCU
.li~,*$,$::..\t..
..::
Implementation.
.,?

CO Control Info
BA Band Info.
Frequency Info, O N14 N13N12N11 N1O N9
FL Frequency Info
N7 N6
N5 N4 N3
N2 N1

FM

Figure 5. MC44802A 12C Byte Detinitiona

361

N8
NO

Referring to Figure 5:
CA 12C chip address for the MC44802A, $C2 (fixed
internally).
CO Sets up the 4.0 MHz oscillator divider ratio (Rl, RO),
prescaler (P), test outputs (R2, R3) and phase comparator output state (R2, R6, T) according to Figure 6.
BA Each band buffer (Pins 7-13) can be set to active low
by writing a 1 to it.
FM, FL These two bytes set the tuning frequency. Their
relationship with frequency (at Pin 4) dependson whether
or not the prescaler IS enabled, and the setting of the reference division ratio:
Fout x Divider ratio

N=

Ro

RI

Oivider Ratio
2048
t 024
512

00
01

10
11
?,,,,,,

256
,,,,,,,,,,,,

Prascaler

,,

(prescaler disabled)

Fcrystal

or: N =

Fout x Divider ratio

(Prescaler enabled)

Fcrystal x 8
A hexadecimal representation of N at FM and FL sets the
tuning frequency (Fout).
Per Figure 5, the address is sent and followed by CO, BA
and/or FM. FL. Control and frequency byte pairs are distinguished in the first bit (1 for control. O for frequency). Therefore, it is not necessary to always send 5 bytes. A data transfer
could consist of CA-CO-BA. or CA-FM-FL. The following example describes the five hex control bytes required to instruct
the circuit to tune to VHF Channel 2 (101 MHz):
1) $C2(1f00001 O) This is the MC44802A address. The
first byte of all MC44802A transmissions must be $C2.
2) $88(1000 1000) R2, R6 and Tare set to 000to indicate
normal operation. P=Oenables the internal prescaler. R1,
RO=OOsets the divider ra!io to 2048 which gives the
greatest frequency resolution in the < 5f 2 MHz re~~+n.
R3 is optionally set high to output a 62.5 kHz test,g~%j:
, ,&,.
at Pln 10 (B4).

.,4X

~- t,) .

FOut (MHz)

%2 $80

3) $04 (0000 0100) Sets band buffer B2 ~Wn~[#~h

q;&.@
thereby disabling Cx.
,Ti,.,.;t ,.?.
4) and 5) $1940 (OOOf f 001. 01000000),,~~~~fhe
given
prescaler and divider values, the fre~~+~h$~defined by
N = Fout/l 5,625 Hz. For 101 MH~, ..,:~\

506
$08
$OC
$12

$40
5C0
$80
5C0

$19
$19

$00
$40

::
35
50
75
100
101 (ch2)

Dsplsy

FM

FL Fout (MHz) DsPlay

None

$IA

5C0

107 (ch3)

None

$1 C
$lE
$20

540
$CO
$40

113 (ch4)
123 (ch5)
129 (ch6)
150
170
200

None
None
075
090
C02

$25
52A

$80
$aO

532

$00

C03
C04
C05
C06
150
170
None

P=O, RO=R1=O

Figure 7. Sampla Frequency

Control

Bytes

sample MC44802A interface to this MCU. A full listing of the

code is included in the Appendix. An HC1l program is written
without line numbers. Tha code shown is the program.ls~ version created by the assembler which inserts the line numbers
and machine code.

Note that this ISnot a uri?~~e~lution to getting 101 MHz out

of the c!rcuit since a~~~tentcombination
of prescaler sefllng,
divider ratio and N~o~~&used.
Figure 7 shows atableoffrequency contro~&~~@~t~~, FL) used in this application note. In
all cases them?k~a~prescaler ISenabled, and the divider ratio

Pin Descriptions
Note that only the HCII pins used in this exercise areshown
in Figure 8. Many of the 1/0 pins can be configured for different
functions throughout the execution of a program. This IS noted
by pins labeled namel/name2. The names in bold indicate the
functions used. They will be referred to by their functional
name from here foward and are briefly explained below. Refer
to the Appendix for code lines.
/C3 (Input Capture 3) is an edge triggered ;nterrupt pm
that can be configured for rising, falling, or both edges. It
iscorrfiguredto respond to rising edges (code lines 70 and
71). All controller output changes are initiated at this pin.

)>,
,Y,

:!:
+>

WFL

M:~&:<*@mentation
,., wj~otorola
MC68HCf 1E9 has the required characterls~~{~+~or generating 12c transfers. It is equipped with parallel
~nd serial l/O ports, timers, a pulse accumulator, an ND converter system and expansion capability for mult!ple MPU systems. Each of these functions must be set-up and activated in
user-programmed software to be part of the system. This allows the user to be concerned with only applicable functions,
What follows are hardware and software descriptions for the

362

.-.

.
T

Pushbutton
Switch

59

60

a
57

58

PAoIIC3

34

MCU
MC68HC1 1 E9

Figure 8. MCU I*mekfation

:?*?*
SPI (Serial Peripheral Interface) Pins:
#5&$SOhWare
DeacriDfion
MOS/ (Master Out Slave In) ISthe serial output line us~,~$.+., ~
The software is written in two functional blocks a main
..
for 12C data communication with the PLL chip. The v}
program and an interrupt service routine (ISR), The main protroller IS configured as the master device in this exercls+$
gram sets up the MCU potis and control registers. It then goes
This Ilne is referred to as PORT D, bit 3 whe~.the $~1 is
into a Iowpowerstopped state until an interrupt isinitlated. The
disabled (the SPI ISenabled only during a sQ#~{ransfer).
interrupt service routine creates the required serial and paralIt is essential that this reconfigured as an ~@n main Outlel output signals, and then returns control to the main program
put (an external pullup is used) whep ~~@mming the
which waits for another interrupt.
SPI Control Register (SPCR, coda~~$s j~21 23). This
allows the slave device (the PLL) @!i@wledge
by PullThe interrupt structure provides flexibility for expansion of
~p*\t$\
.e~i.,
ing the data line low.
*,, ~JLi?:,,+
th!s system. Other functions can be easily added to the ma!n
M/SO (Master In Slave Qut)~%Kerial
input line to the
program without affecting performance of the serial interface.
controller. Ted to MOS~+S~~sa
hi-directional data port
But for this exerctse, the main program IS kept simple. It sets
permitting the MCU ta~ead}the acknowledge pulse.
up memory address references (lines 2038), parallel Port A
SCK _ is the clock &@$~W 12C protocol, It is referred to
(lines 4648), parallel Port B (line 51) and the interrupt control
as PORT D, b~~,4$~.h:,#the SPI is disabled.
(lines 66-73). The main program then goes into its low power
SS is a slav@$e&fl\ne
that must be tied high (reactive)
wait state. It does nothing until control is transferred to the ISR.
to set the MC&>~&+themaster.
,>***
An ISR flow diagram is included as Figure 10 for clarification.
* .,-:.,,*$
pofl A pi~f~,a!+
The followlng program was written under the assumption
pA~:>$$general l/O pin. It must be configured as input
that eventually the system will be run as a stand alone. Thus,
~,,~tput
depending on the desired function. It is conflthe setial bytes pertaining to tuning requirements must be
tiY+:+~g*d here as an output (code lines 4W8) to drive a bit
stored in the MCU EEPROM. To avoid program modification
,,
% the seven segment display (in conjunction with
each time such requirements change, data space has been al;$,
,::t:>,Itl.,,~j PA6PA4)
located for this function beginning at Iocatlon B700, The pro.]~ ~~$,~,
~ PA6 to PA4 are fixed direction output pins also used for
gram requires a specific data format while maintaining appli,,~?,
h+.,
,,t:\,
~
the seven segment displays.
cation flexibility.
i,.
?..\,..
..\,z,
....
The f!rst requested transfer will output bytes starting at locaPort B Pins:
tion B700. Transmission continues until a null data byte ($00)
PB7 to PBO are fixed direction output pins used for the
is encountered (which is not outputted). The two bytes followseven segment displays.
ing contain the display Information. Transmissions of this forSTRB is an enable line that provides an active low pulse
mat should follow consecutively as desired with another null
each time new data is written to Port B. This is used to
after the last display value. The program will then reset the
latch data into the display decoders.

363

data pointer to B700. Figure 9 shows the frequency data space

for the sample system. It contains bytes for vaflous frequencies from 75 MHz to 170 MHz. Band switching is done between
the 90 MHz and 101 MHz (VHF Channel 2) frequency values.

Note that this example contains WO tive-bfie transmissions

and the remainder are three-byte transmissions. Three-byte
transmissions are useful as unchanged control, and band information need not be repeated. The displays will cycle
through 075, 090, C02, C03, C04, COS, C06,1 50, and
170 which is a mix of frequency (in MHz) and VHF channel
displays. The lower four-bits of the first display value (set to f)

B700>c2880412
cOOOOf75c2161000
0f90c288
B710>OI 194000cf
02c21acOOOcf03c2
1c4000
B720> cf 04 C2 le CO00 cf 05 C2 204000 cf 06 C2 25
B730>80001f
50c22a80001
f70Wffffffffff
Figure 9. Sample System Control Data

Create 12C
Idle Condition

Clock In

I,@

*Q
.. .

L
,

...

e.
i

Yes

End
01 Data

Space?

N.

Reset Pointer to
Top of Data Space

Figure 10. ISR Flow Diagram

364

.,. - ...

.. .

I
I
I
I
I
I

Q
I

{,. .,.<
\

.~,?:
.. ~,(,1)

The M68HC11 EVBU (Universel Evaluation Board) pros~bler.

This program will generate the necessary object
vides a friendly environment for developing an HC11 system.
+*code and, if desired, a listing file. The object file (xxxx.list) is
Programming is a three step process which includes w:@&
then downloaded into the HC11:
software, assembling it, and downloading it to the MC@~,.. ~:
program,list listing file,
.*:.*j,,.*;)\
-., . ~
program.s19 file to be downloaded.

Downloading/Debugging
Performance of the software and hardware should be evaluated with the help of apersonalcomputer (Macintosh ora PC
compatible) and a terminal emulation package such as Freeterm or Kermit. This program allows communication between
the EVBU and computer.

366

....

APPENDIX
0001
0002
0003
0004
0005
0006
0007
0008
0009
0010
0011
0012
0013
0014
0015
0016
0017
0018
0019
00200000
00210004
00220002
00230026
00240008
00250028
0026 O02a
00270029
00280009
00290022
00300023
00310021
0032
00330000
00340001
0035
00360000
0037 OOe2
0038 OOe3
0039
0040
0041
0042 b600

2 Program Listing

Motorola SPS Bipolar Analog IC Division

Written by Paul Brownlee
This M68HC11 code provides control bytes to operate
an MC44802A (Motorola PLL Tuning Circuit) via 12Cprotocol
The bytes are to be determined by the user and placed in
memory stating with location B700 (see technical data sheet
for control byte information).

.
.
.
.
.
.
,
.

st.,\
.,.,... )\i.
.. .....,,..*
..,,:.
~t. *?},.
,....<,~,.
k ~:b.t.?~
,,,~,,.~t:m
,.*.~<,,,%
*,,.,.~
,<.1,
~.,,< ,.s>
....
.*;:,,J .*\~~
%....
~,.:,,/.,1.*.~.+:,>
.~.+b}a
~.;.
>,,,
.<...

Communication is achieved using the HC1 1s Synchronous Serial

Peripheral Interface (SPI) to generate both the clock.and data
signals. The main program is a short monitor loop. Output is
implemented as an interrupt service routine for the edge
triggered interrupt IC3. Thus, the location to the routine,
B640, must be entered in user RAM as a jump destination for the IC3, ~,
:~,?tt,
.$:,
service routine. The interrupt can then be implemented as a
,C:.,, %,..:,
~!.:~:,~
simple debounced switch.
~,,.+
,,<r$ -~.
~.\,,
:::; *,..,
,,,h,~i,...
.,,., ~:,i~
REFERENCED TO X-OFFSET ($1OOO)
PORTA
EQU
PORT ~~~A
REGISTER
$00
PORTB
EQU
P@h~ B DATA REGISTER
$04
Plot
EQU
$02
:,,,@~~~!fEL
1/0 CONTROL
PACTL
EQU
$26
i @~LXEACCCNTRL
REG (pORT A)
PORTD
EQU
508 q+P>&ft%~RT D DATA REGISTER
SPCR
EQU
$38 e~% SPI CONTROL REGISTER
EQU
SPDR
SPI DATA REGISTER
,$*
EQU
SPSR
.,,.=?$
29,
SPI STATUS REGISTER
DDRD
PORT D DATA DIRECTION REGISTER
EQU .,$,, *9
TMSK1
EQ~ t>>*$22
REGISTER FOR INPUT CAPTURE ENABLE
E~&:~r)J$ $23
TFLG1
REGISTER FOR INPUT CAPTURE STATUS
EQU*7!>
TCTL2
REGISTER FOR INPUT CAPTURE CONTROL
$21
REFERENCWTOV-OFFSET
(STARTS AT$B700)
DATA
.>~:f?$,\EQU
DATA SPACE (REL DATA POINTER)
$00
NEXTD
~{,;;~: EQU
NEXT DATA BYTE POINTER
$01
. REF<~+~<~~TOOOOO
EQU
RAM LOC FOR CONTROL DATA
YSTQiti~;$*
50000
iq~~~::<.~
THE LOCATION FOR IC3 JUMP INST
EQU
5E2
EQU
5E3
LOC. TO PLACE THE JMP ADX
&BbRr
?;,, +?.~
\*\,..:a
~?;,
J.

.,..

LDX
PORT ASET-UP

#$1 000

(FOR HIGH
LDAA
ORAA
STAA

BASE FOR CONTROL REGISTERS

ORDER 7SEG DISPLAY OUTPUT)

PACTL,X
SET PORTA, BIT 7 TO
#$80
AN OUTPUT PORT
PACTL, X

PORT B SET-UP (FOR 2 LOW ORDER 7 SEG DISPLAY OUTPUTS)

BCLR
Ploc,x 5FF
SIMPLE HANDSHAKE MODE
TEST OUTPUTS
LDAA
STAA
LDAA

#$AF
PORTA,X
#$BC

367

PUT AN R IN THE HIGH

. HEX DIGIT
AND A BC IN THE LOW

APPENDIX
0057
0058
0059
0060
0061
0062
0063
0064
0065
0066
0067
0068
0069
0070
0071
0072
0073

b612 a704

2 Program Listing (continued)

PORTB,X

STAA

b614 18 ce b700
b618 18 df 00

LDY
STY

#$B700
YSTOR

b61 b 8e 00 ff

INITIALIZE USER STACK POINTER

LDS
#$FF

b61e667e
b62097 e2
b622 CC b640
b625 dd e3
b62786 01
b629a721
b62b 1C 2201
b62e Oe

INTERRUPT PREPARATIONS
LDA
STAA
LDD
STD
LDAA
STAA
BSET
CLI

#$7E
IC3JMP
#$B640
IC3JMP1
#$o 1
TCTL2,X
TMSK1,X$OI

.
,

FOR LED DISPLAYS

SET MEMORY POINTER

.
.
.
.
.
.

0074
0075
. MAIN PROGRAM DO NOTHING LOOP
0076 b62f
0077 b62f 01
0078 b630 cf
0079 b631 20 fc
0080
0081
0082
0083
0084 b640
0085
...
,
0086 b640
START
~F*A
* ::0
0087 b64086 64
0088 b642 18 ce 03 e8
OUTERD
:kDh,>
#1 000
DELAY FOR SOFTWARE
0089 b646 1809
0090 b64826 fc
0091 b64a 4a
0092 b64b 26 f5
0093
0094 b64d 18 de 00
0095 b650 1C 0810
.,. >.\
0096
~,:,:,:l,~u~\*,,,,<
0097
}$i ~YHE REMAINING LOOP IS EXECUTED AS MANY TIMES AS THERE ARE
,,.3J,,K.\*8By
TEsT0
BE Outputted.
lTSTARTS AT B700(0RwHEREvER
IT LEFT
0098
~,:,,+
OFF ON PREVIOUS interrupt
HANDLED) AND"OUTPUTS UNTiLA NULL
0099
,,.
0100
BYTE (OO)lSFOUND (OOISNOT OUTPUTTED) .THENEXTTWO
BYTES
,+
8s $,
0101
ARE DISPLAYED AND THE POINTER UPDATED.
?\ .:?,i,=,,.i
~$.~,,
~$.
0102
.
0103 b653 .,~m~:
LOOP
EQU
0104 b6~+l&~~00
LDAB
DATA,Y
LOAD THE PRESENT BYTE
0105 @5&~ 2840
BCLR
SPCR,X $40
DISABLESPI
0106 ~5~1C0808
PORTD,X $08
BSET
SET DBIT3HIGH
(IDLE)
~$~~b~>~8638
LDAA
#$38
SS=1 , SCK=MOSI=l
i,,$O~b%5ea709
DDRD,X
STAA
~/~*Q~ b660 Cl C2
CMPB
#$c2
CHECK DATA TO SEE IF A
l@l10b66226 03
BNE
NOSTART
. START CONDITION IS REQ
0111
. (IF FIRST DATA BYTE)
0112
..2,

368

. .. .. .. --------

APPENDIX 2 Program Listing (continued)

0113
0114
0115
0116
0117
0118
0119
0120
0121
0122
0123
0124
0125
0126
0127
0128
0129
0130
0131
0132
0133
0134
0135
0136
0137
0138
0139
0140
0141
0142
0143
0144
0145
0146
0147
0148
0149
0150
0151
0152

Thisse menttransfers abytefrom the HC1ls SPl

tothel 2 Cperlpheral. Upon Entry, data isin AmB,
* w_start istheentry point forsending astaR bit,
nostati istheentry point fortransferring data
without a start condition.

b664
b664 ld0808
b687
b66786 73
b669a728
b66b lC0808
b66ee72a
b670a629
b6722a fc
b6741d
b677 a6
b67984
b67b a7

W_START

WAIT

0810
28
bf
28

BCLR
NOSTART
LDAA
STAA
BSET
STAB
LDAA
BPL

EQU
PORTD,X $08
EQU
#$73
SPCR,X
.
PORTD,X $08
SPDR,X
SPSR,X
WAIT

BCLR
LDAA
ANDA
STAA

PORTD,X $10
SPCR,X
#$BF
SPCR,X

b893 1808
b695 1808
b697 Id 02 ff
b69a18a600
b69da700
b69f 1808
b6al 18a600
b6a4 a7 04

,,.4,.
;$

::~~,N~AR SpE, DISABLE SPI

%3~AUSES PD4 (SDA) TO GO HIGH
TEST NEXT BYTE, IFOO
SLAVE GENRTS ACK (LOW)

b87d 186d 01
b68026 3a
b682 ld0808
b6851c 0810
b68821 f8
b68ald 0810
b88dlc08
10
b890 lC 0808

$:.\,+
..~t:f:g,\$.t*
~.,,,:!.,>,
- ,..,,,,
\ !+..).,
.~.:j+., >,j$
ENABLE SPI (SPE+j~ :@~~ER
CP0L=CPHA=O~>,8k~Aw=CLW32
RETuRNPD3$Q$9J&
sTATE
. WRITEDA~~*%~::,
. WAIT FOR$~,~:DF XMISSION
IF NOT :WAIT:<
!>\\:,:ii
,
,. ~\..* ,~+~
4@A~kCLK
(PD4) LOW
;:$~~E&TE ACK PULSE
.

START CONDITION
.

LO_ACK

ELSE, CLEAR ACK BIT

GEN ACK CLOCK
INSURE PULSE WIDTH
CLOCK LOW
GEN STOP
CONDITION

PNT TO FREQ VALU

SIMPLE HANDSHAKE MODE
LOAD MSB OF FREQ VAL
AND OUTPUT IT
MOVE POINTER
LOAD 2 LS DIGITS
AND OUTPUT THOSE
POINT TO NEXT GROUP
SAVE NEW POINTER
CHECK FOR LAST GROUP
IF NOT, KEEP YSTOR
ELSE RESET POINTER
TO TOP OF DATA

LDAA
STAA
INY
LDAA
STAA

DATA,Y
PORTB,X

INY
STY
TST
BNE
LDY
STY

YSTOR
DATA,Y
MODATA
#$B700
YSTOR

Ploc,x $ff
DATA,Y
PO RTA,X

LDAA
STAA
RTI

#$ol
TFLG1 ,X

BSET
LDAA
ANDA
BNE
BRN

PORTD,X $10
PORTD,X
#$04
ERROR
Hl_ACK

369

CLEAR INTERRUPT
STOP SERVICE OF OUTPUT

GENERATE ACK CLOCK

CHECK FOR SLAVE ACK
BEING A LOW BIT 3
IF NOT, BRANCH TO ERROR
EN5URE CLK PULSE WIDTH

APPENDIX
0171b6c71d
0810
0172 b6ca 1808
0173 b8cc2085
0174
0175 b6ce 86 ee
0176 b6d0a700
0177 b6d2a704
0178 b6d4 7e b6 bO
0179
0180

2 Program Usting (continued)

PORTD,X $10

BCLR
INY
BRA
ERROR

BCLR4, PORTD
POINT TO NEXT DATA BYTE

LOOP

LDAA
STAA
STAA
JMP

370

AN1207
The MC1 45170 in Basic HF and VHF Oscillators

,*\

~....
..\,.\..,..
,{.<.,
*\, .,...
\~:v >+,
~ ,t,)\ .
before being fed to other sections of the radio. ~~~~~
output
can be directly used in computers and oth~~~,~~~~equipment.
The output of a VCO or VCM is typicall~$o~~~,
as shown,
As shown in Figure 2, the MC1451~&oQ?alns a reference
oscillator, reference counter (R Co*J;~,@~NCM
counter (N
Counter), and phase detector. A ~We,~etailed block diagram
is shown in the data sheet. ,a,,$~i e@
<>..
.:~~
HF,~&~ESIZER

Prepared by David Babin and Mark Clark

Phase-locked loop (PLL) frequency synthesizers are commonly found in communication gear today. The carrier oscillator
in a transmitter and local oscillator (LO) in a receiver are where
PLL frequency synthesizers are utilized. In some cellular
phones, a synthesizer can also be used to generate 90 MHz for
an offset loop, In addition, synthesizers can be used in computers and other digital systems to create different clocks which
are synchronized to a master clock.
The MC145170 is available to address some of these
applications. The frequency capa~lity of the MC145170 is very
broad from a few hertz to 160 MHz.

The basic informi~#$~~ired

for desigmng a stable highfrequency PLL f~~$$~y
synthesizer is the frequencies
required, tuni,~~esol~on,
Iocktime, and overshoot. FortheexADVANTAGES
ampledesig~~~~re
3, the frequencies needed are 9.20 MHz
Frequency synthesizers, such as the MC145170, use digital
to 12.l~M~g.The
resolution (usually the same as the fredividers which can be placed under MCU control. Usually, all
que~~ $~~~or channel spacing) is 230 kHz. The lock t]me is
that is required to change frequencies is to change the divide
8 ~t~da
maximum overshoot of approximately 150A is tarratio of the N Counter. Tuning in less than a millisecond IS
,t~ted:~or
purposes of this example, lock is considered to be
achievable.
%-en the frequency is within about 17. of the final value.
The MC145170 can generate many frequencies based on ~i,wt~ .~.
the accuracy of a single reference source. For example, they,,
.....
,.;&F SYNTHESIZER LOW-PASS FILTER
reference can be a low-cost basic crystal oscillator or a temDe%., .
ature-compensated c~stal oscillator (TCXO). Therefore, ~~~>:
In this design, assume a square wave output is acceptable.
tuning accuracies can be achieved. Boosting of the referen~@
To generate a square wave,aMC1658 VCM chip is chosen, Per
.,>~
,,\,.
frequency by 100x or more is achievable.
the transfer characteristic given In the data sheet, the MC1 658

.,,.><.,

transfer funct!on, KVCM, is approximately 1 x 108 radians/

second/volt. The loading presented by the MC1 658 control Input is large; the maximum input current is 350 vA, Therefore, an
active low-pass filter is used so that loading does not affect the
filters response. See Figure 3. In the filter, a 2N7002 FET is
chosen because it has ve~ high transconductance (80 mmhos)
and low input leakage (100 nA).

ELEMENTS

IN THE LOOP**;, *:i

t:.~:~,x.,.,.,
~,.
The components used in the PLL freg~n~:,~yhthesizer
of
Figure 1 are the MC145170 PLL chip,)&<%~,~ filter, and voltage-controlled
oscillator (VCO). ~,~~~mes
a voltagecontrolled multivibrator (VCM) is &#,R@l~ce of the VCO. The
output of a VCM is a square ~},~fid
is usually integrated
,,.\~t. .~.>~1~~

DIVIOE
VALUE

REFERENCE
COUNTER
[RCOUNTER)

REFERENCE
OSCILLATOR

fR
PHASE
OETECTOR

FROM
%
vCONCM

VCONCM
COUNTER
I (NCOUNTER)
I
1

fv

MULTIPLYING
VALUE
figure

1. PLL Frequency Synthesizer

Figure 2. Oetail of the MC145170

371

TO
LOW.PASS
FILTER

+5V

+5V

+5V

0.01 NF
MC145170

DATAIN

1.8 MQ

ENABLE

MCU

CLOCK

1 MQ

MC74HCU04
510 a
LOW-PASSFILTER
.l..k\\$

In order
Counter,

to calculate

follow

frequency;

the

average

this procedure.

this

is (12.19

diwde

First,

+ 9.2)/2

value

determtne

= 10.695

for the

~+~,

*>*
:;....;<i.$i*J,

Figure 3. HF Synthesizer

the average

MHz

or approxi-

mately 10.7 MHz. Next, divide this frequency by the resolution:

10.7 MHti230
kHz = about 47.
Next, reference appl!catlon note AN535 (see book
DL1 30/D Rev 1). Theactivefilter chosen takes the form shown
in Figure

9 of the application

single-ended

phase

The

phase

detector

K.=

VDD/4n.

note.

detector

This

output

associated

For a supply

filter

is used with the

of the MC145170,

pDq#~

with

a ,~~i~~:.

PDOut

has

of 5 V, this is 5/4n = 0.39~.~<{~:

shown m Figure 4. Id ~~:ve

about 150/0overshoot, a damping factor of 0.8,&rW~@This
causes frequency to settle to within 10/. at o~,,~!$$~?:
The Information up to th[s point is as fo~~~$.
,Y>%;,
fref = 230 kHz
:,>.. ~..:
fVCM = 9.2 to 12.19 MHz; th~kvis 10.7 MHz,

The systems

step response

is

o
m

0.5
0.4

factor

= 0.8

lock time

0.3

~~.>,,i.,

t = ~,<~~~~tling

to within

0.2

10/0, Wnt = 5.5

01

Ko or Kp =10.~8tiad.
fk

on

note,

equation

61,

on

= 5.51t

0
0

5.;::;::@,tzs.

Equq~Q&:&$/& R1C= (KP Kv)lmnz N

.$}.
$+ ):~
= (0.398 X 1 X 108)/687.52X

=(2

X 0.8)/(1

A standard

value

X 10-6X

2.0

3,04.0

5.0

6.0

7.0

8.0

9.0

10

11 12

13

14

ant
47

figure

4. Type 2 Second Order Step Response

HF SYNTHESIZER PROGRAMMING

%1 = 1.79/C which is 1.79 MQ, The standard value of 1.8 MQ

is used for R1.
Equat!on 63 is R2 = (2c)/C On
= 2.33

1.0

Programming the MC145170 is straighffomard. The three

registers may be programmed in a b~e-oriented fashion. The
registers retain their values as long as power is applied. Thus,
usually both the C and R Registers are programmed just once,
right after power up.

687.5)

kQ.

for R2 of 2.4 kQ is utilized.

372

The C Register, which configures the device, is programmed with $CO (1 b~e). This sets the phase detector to
the proper polarity and activates PDout. This also turns off the
unused outputs. The phase detector polarity is determined by
the filter and the VCM. For this example, the MC1 658 data
sheet shows that a higher voltage level is needed if speed is
to be increased. However, the low-pass filter invetis the signal
from the phase detector (due to the active element configuration), Therefore, the programming of the polarity for the phase
detector means that the POL bit must be a1 .
The R Register is programmed for a divide value that
results in the proper frequency at the phase detector reference
input. In this case, 230 kHz is needed. Therefore, with the
4.6 MHz source shown in Figure 3, the R Register needs a
value of $000014 (3 bytes, 20 in decimal).
The N Register determines the frequency tuned. Tuning
9.2 MHz requires the proper value for N to multiply up the
reference of 230 kHz to 9,2 MHz. This is 40 decimal. For
12,19 MHz, the value is 53 decimal. To tune over the range,
change the value in the N Register within the range of 40 to 53
with a 2-byte transfer. Table 1 shows the possible frequencies.

VHF SYNTHESIZER
The MCf45170 maybe used in VHF designs, also. The
range for this next example is 140to 160 MHz in 100 kHz
increments,
VHF SYNTHESIZER LOW-PASS FILTER
To illustrate design with the doubled-ended phase de~ctor,
the $R and $V outputs are used. This reqUireS an OP~~~Ql
amphfier, as shown in Figure 5. From thedesign,@t,@~ri8s
shown inthe MC145170data sheet, the follo~\@@sWns
are used:
i: $?

Table 1. The HF Oscillator Frequencies

N Value

40

Frequency, MHz
9.20

The control voltage range on the input to the VCO is picked to

be2tof OV.
The average frequency = (140+ 160)/2= 150 MHz. Therefore, the average N = 1500.
The above choices for < and On are rules of thumb that are
a good design statiing point. A larger @nvalue results in faster
loop lock times and higher reference frequency VCO
stdebands for similar sideband filtering. (See Advanced
Considerations.)
Choosing Cl to be 4700 pF, R1 is calculated from the
rearranged expression for on as:
EXTRA

~},

R,_ K@KVCO_ (0.796V/rad)(l,57

xi07rad/s~)

,:.T

Clw~N

,)C*>,,~:*. .. *

When the H~~sd@tor was built, the proper frequencies

could not be,~ned?~$~e output of the MC1 658 was examined
with an o:ax~,wpe
and the switching edges were discovered
to be rag~e~,,%at is, the output did not appear to be a square
wav?~~*,*ean
transitions.
, %$~
input of the MC145170is sensitive to500mV p-p
:,,wa!~%ndthe
ragged edges were being amplified and
F cQ,O%teddownbythe NCounter. Therefore, theedgesneeded
~$+.~~aning up. One method would have been to add a low-pass
~,,i...~+:>:
..<,
..,..~,{,filter
between the MC1658 and MC145170. However, be..,.<
cause an additlenal buffer was needed elsewhere in the cir+&*$:.\
,..?*
it,.,~..~,+.. cuit, an MC74HCU04inverf erwasused inplace of the filter.
*
$$
This inverters frequency response is low enough to clean up
the ragged edges. That is, filtering of the ragged edges
occurred, andtheoutput
had smoother transitions. As mentioned previously, one of the elements in the invetier package
was used to buffer the output of the VCM before feeding it to
the outside world. See Figure 3.

(4700 pF)(12,566rad/s)2(1500)
=11.23kQ

(6)

Therefore, chose anll kQstandard value resistor.

R2 is determined from:
2~
(2)(0.707)
R2=
OnCl=
(12,566) (4700pF)

= 23.94

~ or
2.4 kQ (standard value)

(7)

VHF SYNTHESIZER EXTRA FILTERING

For more demanding applications, extra filtering is sometimes added. This reduces the VCO sidebands caused by a
small amount of the reference frequency feeding through the
filter. One form ofthisfiltering consists ofspiffing Rl intowo
resistors; each resistor is one-half the value of Rl, as indicated by R1/2 in Figure 5. Capacitors CC are added from the

373

4700 pF

2.4 kQ
4 x 5.6 kQ
2x1500PF

+5V
1 MHz
YI

20 pF

.,$!.

100 pF
~

OUTPUT

midpoints

to ground to further

The value of CC is chosen

added networ~does
bandwidth

filter the reference

not significantly

mB.

affect t@~<&al
\:*,..y., , ..
,~ ,,a.\
*:.\*.

The rule of thumb for an initial value,~~ ~~,~~)

where

mRC isthe

choose mRCto
original

filter.

seties

~~$#&.

As above, let mRC =257,600

loop

than C5 and

be neglected for this calculation.

R1 waspreviously

chosen

rad/s be the cutoff of thlsfilter.

to bel

OkQ. Therefore,

( R1 ORC),

filter cutoff freqqw~$$.~+~oodvalue

be 10x OB, soasto

&t S$nificantly
~,.,,i:
{a,,,,$i:r,
:}
..< ~:>:.

w!th the rest of the circutt) is much smaller

can therefore

so that the corner fre.q&.&y.B~this

is to

C5=L

impact the

(257,600)

ORCR14=

(11)
(10kQ)

= 388 PF = 390 PF

THE VARACTOR
The MV2115

wasselectedforits

capacitance can bechangedfrom

reverse

bias swing

resentative

tuning rafioof 2.6to

1. The

49.1 pFto127.7pF

overa

of 2 to 30 volts. Contact your Motorola

for information

regarding

the MV2115

rep-

varactor

diode,
For example,

,*,

=1383
There

be selected

to ensure

affect the loop bandwidth.

nated by R14

Nominal

CR=

Capacitance

fR = Frequency

pF=1500pF

is also a filter formed at the input to the VCO. Again,

this should

three parameters

CT=

For this example,

with C5. The

CR=

that it does not significantly

capacitance

are considered.

capacitance
ratio

ratio
Cvmin
()
Cmax

~X

P
(12)

Vmin

the filter is domi-

of the varactors

where

(in

p = the capacitance

exponent

374

...-

..

......

fmax =
(13)

log(2 .6) = plog(l 5)

(14)

p = iog(2.6)/log(15) = 0.3528

(15)

,0

100 pF=

0.3528
(16)

Tv

100 pF

=1.382
Cvmax

100 pF

= 72.4 PF
1.382
Solving for Cvmin:

Vco.
Cvmin
49.1 pF
Cumin = (2.6) (49,1 pF)

Cumin = 127.7 pF

\y~:\4~3Y,\..

The mf;~~$er
is programmed for a divide value that
(17)
deliv@ tk@per
frequency at the phase detector reference
input}~~,~his case, 100 kHz is needed. Therefore, with the
,,,~&Hz &rystal shown, the R Register needs a value of
,t,&$@OOA (3 bytes, 10 in decimal).
/,
*$sThe NReglster determines the frequency tuned. To tune
~~,~-++,J~40MHz, the value required for N to multiply up the reference

THE VCO

value is 1600 decimal. To tune over the range, simply change

the value in the N Ragister with a 2-byte transfer

For convenience. the MC1 648 VCO is selected. The td~ina-

ADVANCED
(18)

As shown in Figure @@@~ta sheet, the VCO tank circuit

is comprised af two ~ra~rs
and an inductor. Typically, a
single varactar ~~~k~%fised in either a series or parallel
configuration. Ho~v.&, thesecand varactor hasa two-fold
purpose. fi~~?~~~
the~b kQ isolating impedance is left in PlaCe,
the varac~~.a~$!n series for a smaller capacitance. Secand,
the aq%~~,ri?tor
acts to eliminate distortion due to the tank
vol~~;w$h~figing.
:JJ*f@~re.
with the two varactors in series, Cdmax =
~sCd~ti2,
The shunt capacitance (inpu!pluS eX!ernal CapaCi*: ta@e)is symbolized byCs.
.jJj~.*?
:..:+$,N
-?,%?::
3.\l$\~ Therefore, solving for the inductance:
>,:,$, \..,:
1
$~,.
\*:S.:.
=19.9 nH=20nH
(20)
L=
~1>
>,.
(2timin)2(CdmaX + Cs)
.\.,
The Q of the inductor should be more than 100 for best performance.
1
2x[(19.9 nH)(69.85 pF)]O.5

= 135 MHz

CONSIDERATIONS

The circuit of Figure 5 may not function at very-high temperature. The reason is that the MC145170 is guaranteed to
a maximum frequency of 160 MHz at 85C. Therefore, there
is no margin for overshoot (reference Figure4) at high temperatura. Thare are two possible solutions: (1) maintain the ambient temperature at Iessthan 60C, or (2) limit the tuning to less
than 160 MHz.
Operational amplifiers are usually too noisy for critical applications, Therefore, if an active element is required in the integrator, ane or more discrete transistors are utilized. These
may be FETs or bipolar devices. However, active filter elements are not needed if the VCO loading is not severe, such
asisencountered
wlthmost discrete VCOdeslgns Because
active elements add noise, same performance parameters
are improved if they are not used. On the other hand, an active
filter can beusedto
scale upthe VCO control voltage. For
example, to tune a wide range, the control voltage may have
to range up to 10 V. Fora 5 V PLL output, this would be scaled
by 2x vla use of active elements.
Some applications have requirements that must be met in
the areas of phase noise and reference suppression. These
parameters are in conflict with fast lock times. That is, as lock
times are reduced, reference suppression becomes more difficult. Both reference suppression and phase noise are advanced areas that are covered in several publications. As an
example, consider that the VCO input voltage range for the
above VHF loop was merely picked to be 8 V. Advanced

(19)

fmin =

(22)

Again, programming the three r&,~t#sof the MC1 45170

is straightforward. Also, usually @jh th&:~ and the R Registers
are programmed only oncek,w~~wer
up.
The C Register confi~~te$~
device and is programmed
with$OO (1 byte) .Thi3Wt@
phase detector to the correct
polarity and activat~~~~~~and ov outputs while tUrnin9 off
the other outputs. L~;t%HF
oscillator, the phase detector
polarity is dewned
by how the filter is hooked up and the

Solving for Cvmax:

2,6=

= 173 MHz

The frequency ratio IS1,5to 1 and is impacted by the tuning

range of the MV2115varactor diode used inthe tank circuit,
Therefore, the required range of 140 to 160 MHz is not limited
by this VCO design.
A PCboard should be used to obtain favorable results~Jth
this VHF circuit. The lead lengths in the tank circuit sh~w
kept short to minimize parasitic inductance. The l~~~<#@the
trace from the VCO output to the PLL input sho~~.~~pt
as
short as possible. In addition, use ofsurfac@~W~~tc omponents is recommended to help minim iz:,,+r5~. ,@
,> ,l,s~~
VHF SYNTHESIZER PROGRAMMl~G$+S
4
,,<...,,+. J..*..

Using the nominal capacitance of 100 pF at 4 volts:

Cvmax
()

1
2n[(19,9 nH)(42.2 PF)]O.5

(21)

375

techniques demand a trade off between this voltage range and

the spectral purity of the VCO output. This is because the
lower the control voltage range, the more sensitive the VCO
is to noise coming into its control input,
A VCO IC may not offer enough performance for some
applications. Therefore, the VCO may have to be designed
from discrete components,
Hgure 6 shows the petiormance of the VHF Oscillator

prototype on a spectrum analyzer. Note that the reference

sidebanda appear at 100 kHz as expected, andare50dB
down.

REFERENCES
CMOS Application-Specific Standard ICS, book DLI 301D,
Motorola, 1990, MCI 45170 data sheet and AN535 application
note.

$*,>>
::kyk...:$:,,

t.>:

fi
E
g
~
o
u
:
%
~

1
/
)

:\Y.*.

.!3,

/
I\.
100kHz

CENTER
= 150MHz,SPAN=25Q&#;
~\\,,
,,.,),tt.,,\.
4 ,.
.J..
~gure 6. VHF Oscillat~@@#mance
.<,N.::.,$::.$
:,1},,
~,..
),

100kHz

376

_....-

. ---

.-.

Thermal

Distortion

In Video

Prepared by: Cutis Gong

Mtiorola RF Products ~vision
Torrance. CA

,,*.
~J

.,.

~:~,,

,..,>:*,
..\>,,
,*,?
,t!,

\$\t,,,

b.

g.
e,

.* *, s..~
~,;.,,/> \ ~,,.
,$e*<.\.

ABSTRACT

signal, and can be explained

$.$,,,.,,
.,
<{,,

using~l~d}~

2. Notice

transition from black to white (fro~+hlg~~]tage

Thermal
video

distotion

amplifiers.

instantaneous
problem
known

power

goes

in many

distotion

changes

occurs

it leads

This Application

is, what cauaes

thermal

It and how to compensate

there

stages,

If the
effect

Note will discuss

shows Upon the dlsplay?~~:woh

are

to a visual

distortion,

the video signal is below the s@~ifd

high resolution
when

in the transistor

uncompensated,

as smearmg.

smearing

is a problem

Thermal

after the

to low voltage),

white level. This signal

brighter than white. The

signal does eventual ly,~J~@to$t%e white level; but unb it does,

the display will appea~~rlqf~r
\ y.+<tj:y~ \ ..

what

than it should be.

WHAT ,Q4vd~S THERMAL

*,\.&}:,:.+,,T<\f

how to measure

the problem.

Thetr#H$$~%ofavideo
insta~~ti~,~,~~power

WHAT IS SMEARING?

DISTORTION?

ampliflerareofien
changes

because

swin~,$:~aficularly

on transitions

~Werc%ngescause

changes

subject t0lar9e

of the large voltage

from black to white.

inihetransistor's

These

junctlontem-

,, ti~~ture.
Due to the transistors thermal time constant, which
Smear+** tNe amount of time it takes something to heat up or cool
block IS <
fast enough. It
displayed on an all white background.
Referring to F!gure 1$$~+,,, #own, the transistor cant change temperature
Is this thermal time constant and the fact that VEE of a translaboth Sections a. and b. should be the same brightness. W*XN
Smearing

is best explained

ing, or ghosting,

there

is a smeating

Sect[on

is most

problem,

a. This problem

by using an example.

noticeable

when

Section

is related

a black

b. will b: brighter

to the droop

torchanges

th~k%

o~.the @tieo

with temperature,

-2

mVPC,

that causes thermal

distotion,

<\l>

Figure

Figure

377

BLACK
LEVEL

WHITE
LEVEL

2.

Hgure

3 shows a simple example

plain the thermal

VBE
swing

does

not change

swing

in power

that can be used to exIn the ideal case,

the transistor.

resistance
affects

concept.

with temperature,

of 107 mW across

and a thermal
power

dstortion

of 300 CM,

the output

of 107 mW would

there

is a power

we can see

TE a change

TJ, that the junction

neously.

Using the 107 mW

in the real case.

create

temperature

where

how this

(A change

of *1 .6C.)

temperature

does

This is a result of the thermal

mV/C,

we

can

calculate

VE,

distotion

of the sauare

calculate

IE, and Vo.

This

VBE;

wave.

about the normal junction

Ideal Caae

1.1

+6o V
1

h
~,

lK

vlN

Vo (VOLTS)

Vo

+
VBE
-

v~E (VOLTS)

Vo [VOLTS)
30.32
29.68

20.32
19,68 m

Hgure

378

3,

on the plot of

not change

instanta-

time constant.
from

example

vlN (vOLTS)

Notice

there

clearly

Using

we

shows

Mn
the
.,,.

MEASURING

THE DISTORTION

Making an accurate measurement of the distortion can

be diticult. The oscilloscope must have enough vertical offset
to enable the edge to be viewed with a reasonable scale.
Often, flatness measurements in the 100 mV to 200 mV
range must be measured on a 1 Voltidiv scale. In this case,
the accuracy is not good. Another issue that must be considered is scope performance at maximum offsets. When a
scope is operating at a maximum offset, it may introduce
some of its own distortion. Check with the manufacturer.

HOW TO COMPENSATE

THE PROBLEM

There is no real standard on how small the distortion must

be. Several years ago a 10/. flatness was acceptable
(400 mV for a 40 V swing). On todays high resolution displays, this is clearly unacceptable. A flatness of 200 mV for
a 40 V swing will cause noticeable smearing problems. Some
designers believe a 50 mV flatness is required, but anything

less than 100 mV is generally acceptable. Flatness of

50 mV 100 mV for a 40 V swing is very difficult to measure.
The effect of thermal distortion can be compensated. The
Motorola CR2424 is used as an example to show some of
the compensation techniques that can be utilized. The output
waveform, when there is a distortion problem, appears as
a signal with excassive mid and high frequency gain.,,.~$+
signal would be flat If this excessive gain were eli~h%#&+~One way of doing this is to use a series RC netwOr~a~*?#dback from the output to the input. The CR2424 *,,,..
~a~~~>%ternal
compensation network which noticeably im~{:w$]$he flatness. Unfortunately, this is only a first @$f dpensation
network and doesnt eliminate all pro~~d%t~~
flatneSS can
be futiher improved by adding an ext~~na~$ohpensation net~~,.
,,;,
work consisting of a 150 pF ca~a~~~orahd a 200 kQ resistor.
Figure 4 shows the flatness @t&@~2424
without the inter.. .,,:,,,:$.
nal compensation netwo:~~~]esigure
5 shows the flatness
with the internal networ~~Jei&e
considerable improvement
in the flatness of tmf~~~
waveform when the complete
CR2424, includj.~ i($,$nternal compensation network, is
,,-~rr
used,
>J, ...i ,.~p

Fi~+w$. QR2424 Without Compensation

;~ >+3
\*

figure 5. CR2424 Wth Internal Compensation

379

Figure
network.

6 shows
The

the

effect

improvement

seen on the CRT. Additional

may

be added

scopes,

where

networks

are

There

to fuflher
flatness

of an external

may

seem

external
improve

is ve~

compensation

small,

but

compensation
the

important,

flatness.

it can

IS not flat. This can be seen in Figures

be

this problem

neWorks

from

In oscillo-

as many

as ten

work.

figure

5 and 6. On the display,

up as a gray area right after the transition

to white.

can be corrected

used,

IS another

black

shows

This

is a frequency

by adding

7 shows

the

response

an additional
circuit

and

input

a photo

issue

peaking

of the actual

waveform.
flatness

issue, The first 0.5 VS of the pulse

o
s

50

215

Figure

380

..-

7, CR2424

With

Modified

Input

and
net-

Network

may want to adiust the compensation network (by changing

the capacitor) to optimize the flatness ats different contrast
level (voltage swing) on the display.

When using the external compensation network techniques as previously described, there are several precautions
that must be taken. The first precaution is that thermal
distortion is dependent on signal swing. The distotion
Improves with smaller signsl swings bacause the power

Another area of precaution is the 215 (2 input peaking

resistor. Snce the CR2424 is a feedbsck amplifier, the gain
is determined by the input peating resistor and the feedback
network. The previously mentioned compensation ne~~.~

changes are less. The 200 kQ and 150 pF RC compensation

network wss optimized for a 40 V signsl swing. For smaller
signal swings, the compensation network tends to overcompensate causing the flatness to slope in the opposite
direction, i.e., the smearing would appear darker than white
instead of brighter than white. In this case, the CRT designer

>.$ t>

*3,

were optimized for a 215 Q input resistor. If the re~~?~,as

changed, the CR2424 would have a different ~a& a~ the
compensation networks would no longer b~L,&$~$>&d.
,, :*,~?:.,\h

381

382

______

Using SPICE to Analyz~(~@e

Effects of Board Layq,ti~on
System Skew Whe~@&signing
With the MC10/1OQ@640
Family of Ciock!,p~Ivers
\
.,,?.i.,!,t:

This application note illustrates the cornp/exifies of

board layout influences on the fota/ skew of a system
when designjng with the MC f OH/100H64x farni/y of
clock ddvera. Transmission line theo~ and tha various
termination techniques are discussed. The note a/so
presents guide/jnes to assist designers in ana/yzjng
their board layouts and /oading schemes us;ng Sf/CE
simulations to predict and minimize the tots/ skew of a
system.

383

Using SPICE to Analyze the Effects of Board Layout

on System Skew When Designing With the H640
Family of Clock Drivers
will concentrate on illustrating the relationship of capacitive
..
~A;,,,
loading versus propagation delay and the relationships ;~y,, ~~
dependence on board layout and termination techniq~~.~~;.
Capacitive loading refers to both device output loadinQ:\~~~~$
Transmission line loading. When the interconne~~n~~
short (less than 4.5) the capacitive loading is -~~,~tiy:the
*$,.,.
output of the driving device and the propagatl~~de{ay can
be predicted by assuming a lumped load,i~?%~$~tput Of
the device. This is referred to as deyJ:e &~J@t loading.

Objective

The objective of this note is to illustrate the complexities

of board layout influences on the total skew of a system when
designing with the H64x series of clock distribution chips.
The note will present some guidelines to assist designers
in using SPICE to analyze their board layouts and loading
schemes to predict and minimize the total skew of a system.
The MCI OH/1 OOH64X series of devices are ECL/~L
translating clock drivers designed for systems requiring vey
low skew clock distribution. Skew is most often specified in
terms of Output to Outpu? skew and Part to Part skew.
Output to Output skew refers to the maximum variation in
propagation delay between similar paths of a single device.
Part to Part skew refers to the maximum propagation delay
difference between similar paths on different devices being
driven by the same inputs. The H64x serias skew
specifications are specified based on equal capacitive
loading of all outputs. Since skew is a measurement of
propagation delay, and propagation delay is dependent on
capacitive loading, opttmum skew performance can only be
achieved when all outputs are loaded equally.
In many designs the clock will need to be routed to a
number of receiving gates at different locations in the system.
For the system designer,
skew measured
at these
destinations is a foramost concern. Skew between receivi~
gates IS a measurement of the maximum variatiow~,,
propagation delay between the driving gate an.~,,~:h
receiving gate. This implies that the designer mu~k-~~~~~!y
be concerned with Output to Outpufl and Pa:,@~#~:J$kew,
but also with the propagation delay along .@~~j~& of the
signal. Propagation delay is a function ~J~$~@& voltage,
ambient temperature, and capacitivet~a~g
(CL). Since
propagation delay is dapendent on ~~p&W~tage, which can
vary significantly from board t,W~~t.@skew
between ICS
on a single board WIII be mua, figper than skew between
ICS on different boards ,$~~~~i~~irates the advantage of
placing ICS with tight s~ew~#rements
on the same power
plane. Assuming th~s~ecommon power plane is used and
that tha temperatu?~~$,went over the board is minimal, the
SUPPIY
voltag~+~$
ambient temperature will affect the

However, when the line length exce~~$~y,

the capacitive
loading is seen by the transmis:J?R:4w * opposed to the
a,.,* ,
output device. This w!II be refe~~t?i
as transmission line
Ioadlng. For the case Of ~:~~fiission
line loading,
propagation delay predicti~~~%st
be based on the Tpd
versus CL relationshi~J~vedYor
a desired line length and
termination technt~~fi$,,*Y
propagation delay versus CL
characteristics of~n l@iand a transmission line are different,
therefore tt i~,ti~iwugh
to simply ensure equal CLs on
all clock Q@th#*~, minimize skew.
The Nits
of thla note are applicable to the entire H64x
seri~~~, ~L/TTL
translating devices, although only the
o~t~ s~tion of the H64f is modeled as the driving section
,~~~<#~nalysis
circuit. The ESD protection circuitry and
~ckage model circuity were included on the output of the
,$$driving device and the input of the receiving device to more
accurately model real in-line circuits. The package model
circuit~ simulates the effects of the device packaging. In all
cases, the input clock to the analysis circuit IS a 25 MHz
ECL level input (+3.15 V to +4.15 V) with 1 ns rise and fall
times. Propagation delay is measured from the 500/. level
of the Input clock to the 1.5 V level of the TTL output at the
receiving gata.
Transmission

Line Conceptsl,

2,3

For high speed systems, the interactions between wiring

and circuity are most easily determined by treating the
interconnections as transmission lines. A brief and simplified
review of transmission line theory and termination techniques
will be presented before discussing the effects of termination
techniques on propagation delay. For a mora detailed
discussion of Transmission Line Theoy, refer to The
Motorola MECLTM System Design Handbook.1

propagation ~~~~ of all outputs in relatively the same

mannerPa~%t~s
should have minimal effect on skew.
Props tlo~ bblay due to capacitive loading, however, may
q
..
v~~~,o~~btput
to output significantly affecting skew, This
va~l~~&# !s due to the dependence of capacitive loading on
~&#
layout, termination technique, and fanout.
$~ To realtze minimal skew at the receiving gates, the
designers goal is to design for equal propagation delays on
all paths carrying the clock signal. The remainder of this note

Impedance:
The conductors (interconnect
trace and the AC ground plane) that interconnect a pair of
circuits have distributed series inductance and distributed
capacitance between them, and thus constitute a transmission line. When these distributed parameters are constant
over a length of line, the line is said to have a characteristic

Cftaracteriatic

384

of the receiving gate is large relative to the fine characteriafic

impedance, therefore pL is approximately equal to 1. A large
poaiffve reflection occurs resulting in overshoot. The
reflected signal reaches point A at time 2TD, and a farge
negative reflection results because the output im~dance of
the driver gate is much less than the characteristic
impadance of the line. In this caae the reflection coefficient
is negative. The signal is re-reflacted back toward ~,~d
arriving at 3TD, resulting in undershoot at P** :~ttie
impetus in restricting interconnect lengths is t~~pf@e
the
effects of overshoot and undershoot. A haw~@~6f
thumb
is: to hmit the undershoot to 15Y. of th~, W*
swing, the
two way tine delay should be Iesa th<~@e;*e
time of the
pulse. Thus the maximum lengthl~~~~p~+~eterminad using
.;>.
.\<> ,..
the following equation:
.,,,,,.
.i.-

impedance, ZO. ZO is the ratio of transient voltage to

transient current passing by a point on the Kne when a signal
change =ure.
The relationsNp between the distributed
parameters, characferietic impedance, and tranaient voltage
and current is expressed as
v/1 =20

=Jm

Eq 2.1

where LO = inductance per unit length and CO = capacitance

per unit length. ZO is expressed in Ohms, LO in Henries,
and Co in Farads.
Propagation Velocity: Propagation velocity can also be
expressed in terms of Co and Lo:
v =

1/ (LO/CO)

Eq 2.2

Termination and Reflection: When a signal travels down

a transmission line, if the terminating resistance (RT)
matches the line im~dance, tha ratio of voltage to current
travelng along the line is matched by the ratio of voltage
to current which must prevail at RT. From the viewpoint of
the driving device, no adjustment of output current is
required. If the tine is not terminated in its characteristic
impedance the signal propagating down the line is partially
reflected back to the source. The maanitude of the reflected
it%-Figure2.la. Block Diagram of Unterminated Line
voltage signal is governed by the load reflection coefficient, .,$$,,
. .,

where:

RS = Source Impedance
ZO = Characteristic Impedance of th~[ne

,?,

were derived from SPICE simulations for 10 and 20 pF loads,

a maximum overshoot of 40Y0, and a maximum undershoot
of 200A. S!mulafion results indicate for a 50 ohm fine driving
a 10 pF load, a stub length of less than 5 inches (assuming
Tpd = 0.18 ns/inch) will limit the overshoot to leas than 400/.,
and the undershoot to within 20Y. of the logic swing. When
the load is increased to 20 PF the maximum line length is
4.5 inches. The results are shown in Figures 2.1 b and 2,1 c.
To minimize undershoot the series termination or parallel AC
termination technique should be used.

J\

The reflected signal continues to be refle~~~Ween

the
source and load impedances and is ~~,~~~pd with each
paswge over the transmission lin%$~~a~tput
response
aPPears as a damPed oacillation~@,~![wlly
approaching
the steady state value. This ph~,~,wa
is referred to as
ringing. Ringing has an ady,q~,:~ct
on noise margin. To
minimize ringing, three ~,si~.f~rmination
techniques are
..*.
k,\...~\..
,,!Y:.
~
available:
.,.2:!,,

Unte~@*dLi
,,<~%.2.l

nes
a illuatratas an unterminated transmission line.

~$
.e he reflection coefficient at the load is of opposite
5 ty to that at the source, the signal will be reflected back
;. p~a
,~$~~,~~ forth over the transmission line with the polarity changing
~~: after each reflection from the source impedance. Thus, steps
l~~;:,~t.$x
*.
.,. .
aPpear at the inpUt to the receiving gate. When the driver
,.4
.$
gate delivers a full ~L swing, the signal propagates from
point A arriving at point B a time TD later. At point B, the
signal is reflected as a function of pL. The input impedance

Figure 2.1 b. H64x Driving a 10 pF Load over

an Unterminated Lfne

385

impedance of the driving device was obtained by extrati!ng

the VoL versus 10L and the VoH versus IOH cuwes (refer
to Figures 2.2b and 2.2c), The output impedance of the
device is equal to the slope of the curves, which can be
calculated to be approximately 8 Q. This value was verified
using SPICE simulations. RSer, in Figure 2.2a was varied
from 10 Q to 50 Q in 10 Q increments and the signal was
$:.\,+
monitored at the input to the recewmg gate (refer to Figure .~~~,,;.<k
2.2d). Minimal undershoot and overshoot occurred when tp~$>?

1.

\
4.T UNE -O.98 V

o
/

4.W UNE

-1.01V
1

resistance of the output driving circuit was assumed @ ~+::$

10 Q, This value closely agrees with the 8 Q ValUe m@B$~~~~
in the lab, So, the value of Rser should be Set t~~~z~~) Q
,J\N:
:
for a matched series termination.

5.~ UNE: -1.04 V

20

40

Series termination is useful when the int@5~&&iingths

are long or impedance discontinuitie~ ei~x .,tin the line.
,.,:+
Another advantage of using series ~@tion
is that the
signal travels down the line at ~t$$.~~$lude,
minimizing
problems associated with cros~~a~~
EM Radiation. The

TIME

Rgure 2.lc. H64x Driving a 20 pF Load over

an Unterminated Line

drawbacks of this techniqu~~,~~,$;hofold.

First, is the
possibility of a two step si~?%ppearing
when the driven
inputs are far from the~w of ftie transmission line. Second,
seties terminationt$~$~~lted
use in ~L
interconnect
schemes due to t~e v&&e drop across Rser in the low state,
Any voltage Q~~#%&s Rser will reduce noise margin (NM)
at the rec~ive$(;~his is illustrated below by calculating the
NML of.@~~L dtiver/ receiver pair, using data book values

Series Termination
Series damping is a techmque rn which a termination
resistance is placed between the driver and the transmission
Ime with no termination resistance placed at the receiving
end of the line. Series termination, illustrated in Figure 2.2a,
ZO = 50 OHMS

To= 1.8 ns
f

l~Q~and
.}, 4,,
::
,\,..
~4,..
,,>,:.~
.$1$.
.:,!s.,.,
t,~~L:
NML

Wevo
AB
SERIESTERMINATEDTRANSMISSION LINE

!.

a special
termination
gate (Ro)
resulting in
is

~,/1

VOL max - [VIL max + IIL (Rser)]

= 0.8 V - [0.5 V + 0.4 mA (40 Q)]

= 0.284

However, when driving CMOS inputs, which pull vey Iitile

input current, vey little NM is lost due to the series
termination resistor, Thus, series termination is a viable
termination technique when driving CMOS gates.

~..$.,.\

case of series damping in which t~~~~~.~f the

resistor (R~er) and the output i~~~&*tof
the
is equal to the line charact~fl~~>tipedance,
minimum undershoot and.~ye~oot,
.,
*,..> ,.~!
,..<
Eq 2.5

SWEEP
VOLTAGE
vsrsus

With series terminatlon,,~~~?{h?output

of the driver gate
switches, a change in vd@~e$deNa V, occurs at the input
~.l{,},..
..
to the transmission Iiw:

0~
-lo
-20
-30
-40

IL

For a rn:at~e~$efies termination: Rser + RO = 20, thus

V = V~~~J~&an
incident wave of half amplitude travelS

-50

t~~~smission line. Since the transmission line is

at the receiving end, the reflection coefficient
,.o~:$~f6ad
is approximately unity; therefore causing the
?~&~e
to double at the receiving end, When the reflected
wave arrives at the source It is completely absorbed by the
series resistor since the impedance matches the characteristic impedance
of the transmission
line, The output

IOL

xl
\
\
\
\

down $:

u~k*Y~IMd

SWEEP VOLTAGEM

Figure 2.2b. VOL veraua IOL for H64x Series

386

___

DERIVATION

SWEEP VOLTAGE VOWSIOH

OFT@

vereus

CL RELATIONSHIPS

Once the designer has chosen a termination technique,

the relationship of TPd versus CL for the SpeCifiCapphcation

4.5

should be derived. It is suggested that the derivation be

performed through simulations using the H64x Clock Driver
I/0
Spice Model Kit. A guideline
for deriving,.the
relationships, TPd versus CL is presented through e~$~
for each termination technique discussed.
,~{.
*: ,! .\\,
In deriving the relationships
necessa~,~~w<~edict
propagation delay a reference for Tpd i~~~~lished
by
finding the propagation delay of the H& f% R@put driving
circuit. To measure TPd of the outp~:ti~$rig
gate Usin! I

3.5 ~

.
\

2,5 ~
o

15

10

20

25

30

OUTPUT HIGH (M)

SPICE, the analysis circuit show&~61&~&e

*.>f+\\
,/
~.,..
,1, .,<
,(,

Figure 2.2c. VOH versue IOH for H64x Series

I

3.1 is used,

UR
w
g2
g

, ~

S*,{.
<.$

HS4iDRIWNG 1 GATE OVER VERY SHORT UNE

25

75

50
TIME

Figure 2.2d. Seriss Terminated Transmission U%

Output for R5er = 10,30,40, and 50.@,:,),,
.*<:,, J~,
Psrallal AC Termlnetion

,$
>,
1{*, ~e*~

S*.
*W
.,,.,,,,.,,.,
~<{:+

.\?.,l,t
~t,l ,,

Parallel AC Termination, shown in Fi@rp~,~~ should be

used when the abihty to drive dist~Q*-%@ads
or when
driving hea~ DC ~L
loads is3%{~w&.
Unkke series
termination, the parallel AC ter~~~h$scheme
features an
f? *.
undisto~ed waveform alongv~ ,b+$*
f~
t,<. length of the line. In
parallel AC termination, tma}efiewmg end is terminated to
a voltage through a reW~@#Yin
series with a capacitance
(CT). The value of ~ i#equal to the line characteristic
,<$..... . . .?
impedance. As a:~~~e$~f%umb CT = 10+TD@, where TD
is the delay :~ th~ Wsmission
line, When the termination
resistance @~@es the line impedance, no refletiton Wcurs

In this circuit, the output, driving gate is dtiving one gate

over a very short line (<< 1~, When the interconnect line
length is this short the SPICE transmission line model is
not needed to simulate the interconnect line; and the
propagation delay due to the interconnect line length can
be assumed to be negligible. The propagation delay is
measured from the 50A voltage level of the input signal to
the 1.5 V level of the ~L output and can be expressed as
follows:
Tpd(model) = T@(outPut gate) + ATpd(l gate load) Eq 3.1
Through a SPICE simulation Tpd[model) was measured to
be 2.76 ns. Rewriting the equation above to solve for
Tpd(output gaIe), the equation becomes:

because a%,t~:~nergy is absorbed by the termination. The

parall~t~~a~]ermination scheme consumes no DC current
w~th}~~ts
in either state,

Tpd(output gate)= 2.76 ns ATpd(l gate load)

Eq 3.2

To solve for Tpd(output gate),

the Tpd due to the capacitive
loading of 1 gate is needed. This relationship will also be
very useful in finding propagation delay contributed by
fanout. By using the same circuit as above and incrementing
the number of receiving gate inputs, measurements Of Tpd
are taken for each increment in the number of receiving gates
in order to develop a relationship between Fanout versus
Propagation Delay (ATpd/A#
Of Gates).

387

The

following

measurements

were

Table

#ofGates

e
15

plotted

this

relationship,

modified

by adding

receiving

gate,

value
2.88
3.02
3.2
3.46
3.64
4.01

3.2

and

circuit

capacitor

of the

data

load capacitor
delay

is summarized

Figure

3,3a,

in Figure
in parallel

of the propagation

of CL, The

in Table

the

load

the value

and measurements

T@ H-L (ns)

2.76
2.82
2.93
3.02
3.15
3.99

1
2
4

and

3.1

T@L-H(ns)

determine

taken:

3.1

was

with

the

was

varied

taken

for each

and shown

in a plot

respectively.

Table

3.2

below
~

2.5 ~
o

10

15

FANOUT
figure

The

value

3.2. Fanout

versus

of A (Tpd)/A

(# of gates)

finding the slope of the Fanout

3,2,

A (Tpd)/A

(#

can be calculated

versus Tpd Cuwe.

gates)

of

Line

Tpd fOr a Shoti

can

be

by

~$~

From fi9Ure

measured

to ,,,~:
..+,, ~i?,,

approximately:

3.3a. CL versus Propagation Delay

for Short Mne

Figure

<h+
From

this

respect

data

the

to the change

change

in propagation

in CL was calculated

of the output driver to capacitive

shofl
(CL)

per

gate

delay/gate
Note,Tpd
but,

(output

merely

common
derived
used

the

propagation

to all of the H64x

in the following

as actual

series

gate) isnotthepro&ga~ridelay

and

to base

In real @~:~@
eacQ+&~f$
od;~w..

skew

capacitive
determine

the

loads

line lengths

to achieve

by

either

on appropriate

between
loading

two
and

the capacitive

paths,

lines
a

are

4.5

If the dasigner
relationship
is

than

for
or

taking

= 2.85

propagation

delay

for compensation.

needed

in Figure

the

Output

4.5

line

the

load

ratio

of

pF/gate

value.

no

sees

longer

of capacitive

loading

(output

Delay

line

line

but instead

loading).

versus

tine

that for line lengths

gate

Delay

increases

curve

The
the

line, The

sharply

output

as the

to

value. SO, for accurate

lines are greater

than

whe~

load

gate

simulations
4.5,

line

greater

rolls off and approaches

the output

at

sees

the

end

gate
of

only the load

line and thus, Tpd approaches

To

are

leas than

For line lengths

rolloff occurs
capacitive

linearly

SPICE

Length

Notice

load) increases.

lengths

are no longer

loading)

(transmission
gate

that

3.3b.

the Delay

a constant

demand

of the gate

of Output

the transmission

between

388

by

ns/pF)

constraints

the effects

by the

transmission

knows

load needed

layout

length (or capacitive

on

loading

length

4.5,

seen

shown

be realizable
and fanout

line

ns/gate)/(0.02

at the output

simulations

devices.

to compensate

adding

= (0.057

exceed

termination

symmetrical

will need

calculated

to delay/CL.

board

seen

H64x

on which

of the H64x

be

Eq 3.5
When

not be

for the

and

it will not always

equal

designer

loading

fanout,

dalay

In attempting

Output

,;a~n~~metrlcal
@p&citive

length,

designs,
to have

should

only as a reference

eff~t~$o~~e
~~:..\\ !1,,
on, ~.
bagation

for the,. ms~$f

delay

can

with

for an unterminated

ns/pF. The capacitive

circuit~

and the values

CL curves

of Magation

are de~{@:~.~e

output

$&is value

T&v$us

valu~

the

technique

Me

s.~r~~,

CL/gate

of the H64x,

@~~8Vthe

loading

line was found to be 0.02

delay

and the sensitivity

of Tpd versus

the line should

the
of

a cOnStant
CL when

be modeled

as

a tranemieeion
propagation

tine and the effect

delay

of capacitive

loading

on

in Table

3.3

transmission

re-evaluated.

along

with

fine ~th

measurements

taken

for

= 75 ohms

Table 3.3

2.5

CL (p~
~
g

the

2.3

T@ (nS), ~

= 50

(ns), ~

T~

4.3

= 75

4.27

2.1
3
a

1.9

1.7

1.5

*
g

1.3

::
0

10

8
UNE LE~GTH (In)

12

Figure 3.3b. TPd versus Line Length

Using

the

SPICE

termination

model

techniques

line model chosen

simulator

will

be

developed

termination

line,

The

is available
delay

and

with

in the SPICE

of 0.18

relationships

ca~acitive

three

transmission

line length and termination

along

dalay

transmission

a propagation

between

propagation

for this exarcise

and assumes

Relationships

of

will be examined.

nsfinch.

technique
between

loading

for

type.

CASE 1: UNTERMINATED

The

analysis

insetiing
circuit

circuit,

a transmission

and the

hung in parallel

receiving

TRANSMISSION
LIN@~P
,.,:$
,>,$
.$,:
4$.
.
in Figure
3.1, w~:rnWfied
by
line

gate

between

{h$~sut

circuit., ~j~p#Mtor,

with the receiving

Qas:{wr
*>~*:,:~.\
.*~..
h :+:!

20

40

60

80

100

CL @q

Figure 3.5. CL versus Tpd for Unterminated Line

driving
CL,

was
A comparison

to Figure 3.4).

between

that output Ioadng

a nonlinear

change

versus

CL

Tpd

curve

>4.5
which

for predicting

versus

3.3a

CL curves

as ZO increases.
demonstrates

shows

This imPlieS

should

use the Tpd

to transmission

delay.

for unterminated

Figure

line

3.5 shows

transmission

the ATpd/ACL

This is dua to the fact C050

an advantage

3.5

CL curves.

corresponds

propagation

Figure

line loading produces

the designer

with ZO of 50 Q and 75 Q. Notice,

This

and

in the TPd versus

that, for line Iangths

loading,

Figure

versus transmission

of using

lines

increases
Q > Co 75 Q.

lines

with lower

Zo.
CASE 2: SERIES TERMINATED
TRANSMISSION
LINE

,ine
The

.,~t.tw,

~k,
*!
K.+.i,.,.

-.,:,,1,
,,3 ~.~
!,.

TO determine

a relationship

the Unterminated

transmission

varied
were

and measurements
taken

for each

value

between

Tpd

VerSUS CL fOr

line, the capacitive

of propagation
of CL. The

delay

results

analysis

inserting
and

load was

the

parallel

at the load

a series

389

in

resistor

between

transmission

line.

with the receiving

in Figure

are tabulatad

circuit,

3,6,

Figure

3.4,

was

the output

A capacitor,

CL,

gate. The resulting

modified
driving
was

by

circuit

hung

in

circuit is shown

q:

4.15V G

3.15V -

SERIES TERMINATED uNE

Figure 3.6. Simulation Grcuit

Series Terminated Line

First, 20 was set to a common value of 50 ohms and the

I/ne length was set to 1V, which translates to a line delay,
TD, of 1.8 ns. Wth CL set to O pF and measuring the
propagation delay at the output of the transmission line, the
accuracy of the transmission line models TD can be
confirmed by comparing this measurement to the measured
vlaue Of Tpd at the input of the transmission line. The

..,

CL relationship is not only w&by

line length, but also,
by the termination tech~~ue c~sen by the designer. Using
series termination ,,,~.,i~,<
p%dxs
a significant decrease in
,.
undershoot and q*b~&t.The tradeoff is an increase in
ATpd/A
CL. N,l@lc~~@n when the gate is unloaded, the

Eq 3,6

Plugging measured values into this equation for the above

circuit:
TD = 4.69 ns 2.9 ns = 1.79 ns

series termin*
,t,,,

. .*

i&

line is slower than the unterminated Ime.

~:?.

Eq 3.7

and we see it is very close to the predicted delay of (O.18

nsflnch) 10 inches = 1.8 ns.
To determine a relationship between Tpd versus CL fOr
matched series termination, the capacitive load was vatied
and measurements of propagation delay at the load W:A,,
taken for each value of CL. The results are tabulated Wo@\St
along with the measurements taken for a transmi$,#&).#&e

\ ... ..

,.,l.

Comparing these results to t~y~$~s~k

obtained for an
unterminated line, it can be ~~~
that the Tpd versus

equation for the measured TD is:

TD = Tpdout Tpdin.

@k?!
\~,J*,
:,~,~

~~$~~~ original circuit was modified

by inserting
a
,,,$trahsmission line between the output driving circuit and the
receiving gate circuit. The circuit is shown in figure 3.8. For
Parallel AC Termination the matching nework is a shunt
resistor (RT) in series with a capacitor (CT) to ground, placed
at the output of the transmission line. From transmission line
theoy, the Parallel AC Termination technique requires that
the resistance of RT match the characteristic impedance of
the transmission
line (20) for optimum performance
(minimum undershoot and overshoot and minimum propagation delay). Also as a rule of thumb the optimum CT can be
calculated as, CT = 10TD/Zo, where TD = the delay of the
transmission line and Zo is the characteristic impedance of
the transmission line.

,n,-,gp*ti

1.8.$ delayof the transmission

Itne.

piOflin$&L ~kus Tpd, refer to Figure 3.7, the relationship

bet-R-Q#*and
Tpd is found to be a linear equatiOn,
when
th&*ation

is matched, that can be expressed as follows:

*
PARALELACTERMINATION
WITHLUMPEO
LOAO

slope; 20
y-intercept: TD + delay of output circuit

FigUre 3.8. Simulation

Circuit

Parallel AC Termination

390

WNh ZO set to 50 ohms and TD set to 1.8 ns, RT and

CT were calculated as W ohms and 380 pF, respectively.
CL was varied and propagation delay measurements
recorded at each value of CL. Nefi, 20 W* set to 75 ohms
and TD to 1.8 ns. Values of RT and CT were recalculated
for these conditions and set to 75 ohms and 240 pF,
respectively. Again CL was varied and propagation delay
monitored, The results are tabuleted in Table 3.5.
Table 3.5
CL (pF)

TX (nS), ZO = SO

o
5
10

5.01
5.17
5.32
5,47
5,63
5,75
5.W
8,00
6.14
6.26
6,38
6.50
6.61
6,85
7.54

15
20
25
30
35
40
45
50
55
60
70
100
.Note

TX

TX

(na), 20 = 76
4.86

overshoot, however, it causes a positive finear shift in the

T@ versus CL curve. Series termination caused an increase
in AT@/ACL of approximately 0.01 rrdpF for a transmission
line with a ZO of 50 Q As a result, propagation delays for
series terminated lines quickly pass those of Parallel AC
terminated tines as capacitive load is increased. The tradeoff
in choosing Parallel AC termination over series termi~ation
$,!:.,
is that Parallel AC termination requires an e~ra ~~r,
CT, in each matching network. Comparing the,~~~;~rsus
CL curves for @ = 50 Q and 75 Q in Figurq,~tf!tw.@ seen
that, as w= the case in the other eXamplw@th&@Ypd/ACL
f$;<,e.,:>~,,,:i,
.\~
increases as ZO increases.
,.~!:
.>,,.{.,
.1..:,,
*;)..

5.32
5.73
~

6.06
6.4
6.71
6.95
7,28
8.03

includes the 1.8 ns delay of the transmission

Hne.

-vi**t,.
. ~.~~..!:.;
vt?A*j,

The original cir~~~~~

modified by inserting three
separate transmi<~okh~between
the output driving circuit
$,,,t,.it+
and the receiving ~~@ Circuit. The sum of the time delay of
the three tr@~~Jssion lines being 1.8 ns, to be consistent
with the<,~j~i{a~en
for the other termination techniques.
C?pa~I~e l~dk are placed at the end of eech transmission
linexT@~*fallel AC matching ne~ork is placed at the end
,@$th~~st transmission line. The circuit is shown in Figure
$+&l o.

plOfling
CL versus T~d results in the
Figure 3.9.

PARALLEL
ACTERMINATION
WITHDISTRIBUTED
LOAD
Figure 3.10. Simulation Circuiti Parallel
AC Termination

..~~~.g.

CL verssTpd:

Table 3.8

Parallel AC Termlatio

,, ~, ~~omparison of the results of the Parallel AC termination

\/?:
$Y& %%heme versus the unterminated scheme illustrates almost
:.>>,.?
~+~
.>no increase
in ATPd/A CL, However, the propagation delay
*:*..;,
,+,,~,:>
,,.,i,
for a Parallel AC terminated line driving a O pF load is greater
.$)tj,
?~
!

than that for an unterminated line or a series terminated tine

driving O pF. So, choosing Parallel AC termination over an
unterminated line significantly decreases undershoot and

.Note.

391

T@

CL (pF)

Tpd L-H (nS)

o
15
25
30
4s
60
90

5.01
5.35
5.54
5.6a
5.98
6.29
6.S4

!ncludes the 1.8 ns delay of the transmi~,on

I$ne

determining the relationship, Tpd versus TD fOr each

termination technique when CL = O, the designer could
determine the y-intercept of that termiflatiOfl
techniques ~pd

veraus CL curve for a desired line length,

+a

T@

versusUNE LENGTHFOR DIFFERENTTERMINATIONS

:,

0
Hgure

20
3.11.

40

CL versus
Parallel

60

Tpd:

80

Distributed

Load:

AC Termination

T@ L-HversusCLOAOFOR MATCHEDSERIES TERMINATION

10

,>

Figure

3.l&.:~$ld

T-on
,.:.~rs.-t..
s:.
:{,
;1

*rsus

Tpd) versus

Technique

~*,,~~,!

By settin~$%~ &@%citive load to O pF for each tYPe of

6

termination, a%s:,varying the line length only; this type of

relationa~~~s established. The results are shown in Figure
3.1 $:@~
t% designer knows the length of the transmission
li~:~Q$~he
termination technique, a T~d Versus bne

5
4

$-N

3
0

20

40

60

60

100

~,,> ~$1,
set to 50 ohms and TD set to 1.8 Q~>.&sw the
.?:*, .$,.,.9
equations above RT and CT were calcula~~,;$s:$~ Q and
360 pF, respectively. Total CL was vari~$~~:~ropagation
delay measurements recorded at ea+i,v$~e
of CL. The
results are tabulated in Table 3.6. :~~oqhv
CL versus Tpd
gives the relationship shown i~,~*&@~.11,
Data has now been derive~~or $e relationship between
capacitive loading versus Q~W#6n
delay for the following
termination techniqu::~ un~~inated
transmission lines,
series termination, ~,~aralldi
AC termination. To generalize
these results for any$~@k&nnect line length, the relationship
of (CL versus ~$~~~~rsus Line Length must be evaluated.
With 20

chafl

can

be used

to determine

the y-intercept

of

t~~appropriate termination schemes Tpd versus CL CUWe.

these values have been derived using only the output
&Note,
section of the HW1 driving the input section of the H645.
Therefore
these propagation
delay values are not
representative of actual delays of the H64x and are derived
here only to show the relationship of TPd versus TD. It iS
suggested that the designer derive the Tpd versus TD curve
with CL = O pF for thair specific application, using the H64x
Clock Driver 1/0 Spice Model Mt.
Summary

The MC1 OH/100H~x series ECWL

translating clock
drivers are ideal devices for systems requiring very low skew
clock disttibufion. Optimum skew performance from the H64x
series requires equal capacitive Ioadmg on each output, To
minimize skew in a system not only requires minimal output
to outpu~ skew and part to pati skew, but also requires
equal propagation delay along all paths carving the clock
signal. Perhaps the most accurate technique of obtaining
equal propagation delay along all paths is to add trace to
the lines with shofler propagation delays. However, this is
a trial and error method and does not always provide a
feasible solution due to size constraints of the board. Another
technique of obtaining equal propagation delays on each
path is to add capacitive loading on paths with shorter
propagation delaya. This method requires an understanding

Using the se}~:~rmination

circuit configuration, the delay
(line lefl~~~~@ k transmission line is varied from TD = 0.9
ns to ~ = +.%ns to TD = 2.7 ns. At each line length setting
a ,;*
VMs
Tpd
curve was extracted. The results are
,}:uM&~zed
in the plot, Figure 3.12.
~~,~$~~~ce, changing the length of the transmission line
~erely causes a vertical shift of the Series Terminations Tpd
versus CL cuwe. This will be found true for the unterminated
and the parallel AC termination schemes as well. So, by

392

.-

,.__

finally series termination. The tradeoff in choosing terminated

lines versus unterminated lines is, of course, minimized
undershoot for an increaae in ATPd/ACL. The tradeoff in
choosing parallel AC termination versus series termination
is an increase in the number of parfa for a decrease in
AT@/A CL. Since the values in this note have been derived
for the specific case of the output section of the H641 ~+ving
the input section of an H645, the values of propagati~ay
are not representative of actual delays of the ~~x~ries
of devices. Also, due to SPICE simulator J&ti~ns
of
.>:.. ~.$
*>.
accuracy, delays are not exact and should,@;u,S~ to predict
relative differences only. For these rea~ri~~fi~
designer is

of T@ versus CL relationships. As shown in tMs note, TX

versus CL relationships are dependent on fine length,
termination technique, and the characteristic impedance of
the trsnsmiesion line. If tine lengths are less than 4.5,
propagation delay can be predicted by assuming a lumped
capacitive load at the output of the driving device. When lines
exceed 4.5 the capacitive load is no longer seen by the
output driving device, but is instead seen by the transmission
line. A tifferent Tpd vemus CL relationship eXiStS for the
transmission line than the output device. The transmission
line TPd versus CL relationship is dependent on termination
technique and line characteristic impedance. The dependence on termination technique is important at line lengths
greater than 4.5 because at these lengths undershoot
becomes significant enough (20% of logic swing for a 20
pF load) to necessitate some sort of termination scheme to
minimize its adverse effects. Relationships of TPd VerSUSCL

encouraged to use
Mr to derive the
minimize skew for
H64x Clock Driver
representative.

were derived and compared for three termination schemes:

the unterminated line, the series terminated line, and the
parallel AC terminated line. All Tpd versus CL CUWeSWerO
derived for transmission lines with TD = 1.8 ns and Zo =
50 Q and 75 Q. For all three termination schemes, increasing
the characteristic
impedance of the transmission
line
produces an increase in the ATpd/AcL relationship. Of the
three termination techniques the unterminated line had the
smallest ATpd /A CL, followed by parallel AC terminatiOfl, and

the H64x Clink E~~Y,@ Spice Model

relationships ~~~w
to predict and
their pati$,ula~.k&#tem. To obtain the
..1/0 Spi@Wdel
Mr contact a Motorola
~,;~.<~i,i
,..,:s.
.. m.. .
,.
h

Motorola ln@$
?~Q3.
.~,,,.
..../ Stock C;de HB205R1/D.
2Motoro@~~~PSTM Data Book, Motorola Inc., 1991. Stock
Cod~,\S~~@l
s~a~~~d
.$~~~Or

393

FAST

ID.
Applications Handbook. Fairchild

~orporation,

1987.

SeMICOfl-

394

....-

.....
,.-.. .....

,..,
,:,

$,

;!~,

)5>

,Ja~s ~p(icaf;on note prov;des fhe SPICE information

, ~ce~av
to accurafe/y model system interconnect
+Sk#%ns for designs which ut;lize the translator circuits
*~$@.fhe MC 10H600 tamjly. The note irrcfudes information
$Ofl the H600, H601, H602, H603, H604, H605, H606 and
~~ H607 trans/afors.

395

MCI O/1OOH600 Translator

Family 1/0 SPICE Modelling

Objective

ESD

protection

shown
With

the

difficulty

impedance
boards

PC boards

the ability

to a board
The

of this

information
H602,

models

H604,

ddven

H605,

includes

may

Internal

included

characteristics

as

H606

The

remainder

input

and

of these

of the input and output

structures

little

output

input

analysis.

on

stages
which

the

The

and

H600

is represented

Figure

6. These

on the output,
schematic

the

translators

The

as

H601

represent
single
TTL

input

inputs,

modeled

structure,

the H601s

ECLST
Gate

H60xfamily

the exception
be

with

can

be

using

H606s

the

should

can be represented
TRIW

be modeled

scQ~@$:,$provided

Table

1. Device

Pati Type
H600
H601,N+:

:;;~$T

~ ;M~4

Cross

i#puts

..... ,
.L{N&Ae

H6Q3><?,,,,$
$*,,.

T~~<@~t

,.t,g~

4~~~u*T~

differential

supply

su~jes.
~~&~~e

means

~~b A and AN
HIGH level at
t .25V and the
with a v01ta9e

protection

~~~$$esented

by the

dev[ce,

circulty
FPSO09E

however,

are represented

both the

by one structure

input

Notice

the ECL

IS represented

as

Figure

in Pad Cell on the

by the ECL

15, and includes

Input Pad Cell

the 50KQ

input Pull down

as is used on the H600

also

ttitstate

should

requires

buffer

one

inputs,

TRI

be set at 1.1 V, while

be driven

by the H601

bias

output

supply,

and TRIB.

the TRI

buffer.
CBIAS,

The
and

The

CBIAS

and TRIB

Inputs

ECL Input structure

SIX ~er~t

The

1/0

butput

mentioned

Output

of

should
from

represent

swings

be forced

is represented
should

of Figure

ESD

protection

by the schematic
be driven

of VCC to VCC - 0.85V.

to f .lV

3.

scheme

The CBIAS

The H603

of

deferentially

and the ENA input should

VCC 0.85 to VCC 10.85V.

same

l/O buffers,

gate

8. The IN and INB inputs

with voltage

buff ers

series

H603

Figure

again

input

be driven
uses the

as the H600.

Reference
~L

Inputs

H601 1/0

H606 ~L

Inputs

H602 ECL Inputs

H601 ECL Inputs

TTLST, DO-De

None

None

None

None

TTLOE

D&D8

None

None

ECLOE

LEN, RESET

D&D8

None

None

None

None

All Inputs

None

None

None

None

None

RESET, CLK, CLKN

CLKT, DWD5

None

None

None

None

All Inputs

None

None

None

None

None

H606

CLK, CLKN, RESET

None

None

CLKT, D&D5

None

None

H607

All Inputs

None

None

None

None

None

H605

Of

dev~:~ich

~9t&~tWESD

circuit~

VCC - 1.75V.

1/0 buffer

should

%ta

the H602

of the

for the ,gbo~

OUtput$$~~@,~;:*~is

OUtpu#J~@t~~tic

and ground
differentially

to2.4V.

this

l/O Structure

H601

u*~:~~~601

ECL Input stru~~~~d

buffer~~8r,)~@*QH60x

rest

and d@:~,J:<:~<
~:$,),\
~*\ ~,:
~..i~
$:,>$:.,,,

inputs&n&,&~H602s

output

The

~:t.,,,,
are
,\$Jp:<+
appropriate

*
method ,e*:J%;t.

and the ESD protection

circuityforthe
ECL Input. The
ESD structure is used on the output buffer secbon of the

iwit,

b~~ae

respectively.

schematics

$ H601

of the ~,@J$

represent
translators.

all of the

TTL

ECL i nput structure,

subclrcu!t

to

$f_@?
**6

CLK~

H606

the exception

the H601

all of the

used

oftranslatorchips.

of the

ECL TRIand

input, which
structure,

which

can be used to represent

All of the ECL inputs

Input

inputs,

structure

with

should

structure.
ECL

all of the ~Oforthe

~L

wh!ch

schematics

H600

davicesaredual

inpw$t~~,ciftuit~

basic

the same

the

devices

shown
in the?&Ol
1/0 Schematic
of Figure 7. The H601
requires ~wingl&@nded
input, IN which should be driven from

s~,~~~,#of
ten

utilize
by

Input and outp,@ @&&

affect the 1/0 characteristics

Overview
are

illustrated
on the
This hierarchical

1 ~+t$~i
*, ( t+
!.,t.?~
....+*\~\$
is ai~ %~~{
supply

VCC ~~$~

There

The

schemabcs

of Figure

is

will introduce

dewces,

Schematic

buffer

is used to help simplify

H602

buffer

swing from -2.OV

of

a SPICE

in the schematics

for the H60x

schematics.

models.

l/O

VCC -0.65V
and the LOW level e~a$~~~c
B and BN inputs should be driv~j@{/fltially

1/0

amount

In addibon

referenced

and

and Ioglc ctrcuity

impact

representation

the buffw

they require +5V, 5.2V

inputs should be dtiven

and package

of the

of this document

well as the other structures

on

H600,

translator

and add a significant

simulation

set for the devices

prov!ded.

the

package

on the

with

analysis

or H607

regulators

have

user

dflving

waveshape

bias

they

of the device

bme to the standard

parameter

the

model

or

schematics

affect

waveforms.

not

various

basic. SPICE

the

and

boxes

by the subcircuits
schematic
sheet,

schematic

logic designers.

circuit~

in shaded

modelled
subcircuit

those

prior to committing

is to provide

as well as ESD protection

which

output
are

behavior

being

controlled

of reworking

for high speed

to petiorm
traces

ch!ps. The packet

structures

circuit

document

H603,

highspeed

and the expense

is essential

interconnect

H601,

designing

to model

layout

purpose

enough
the

in

Kit

396

.
---

.---.

The H604 and H606 utilize the same output buffer. This
buffer is represented by the H604 Output Schematic of
Figure 11. The IN and INB inputs should be driven differentially
with voltage swings from VCC -0,85 to VCC - 10.85V. Note,
the ESD protection circuitry is the same as the H600.

Table

2. Input

Schematic

ECL Input

Figure 12 represents the schematic for the output buffer

utilized by the H605, The IN and INB inputs should be driven
tifferentiallyfrom VCC-O.85to VCC 10.85V, while CBIAS is
forced to 1.1V. Again, the same ESD protection scheme is
used as on the H600.

and Bias
Input

Levels
Level

V88
Vcs

Vcc - 1.3V
VEE + 1.3V

H600, H602

NAN

vcc - o.e5V tO Vcc

output

B/8N
Vcs

vcc - 2.OV to Vcc

VEE + 1.3V

- 1.25V

- 2.4V

$,1

H601 1/0

The H607 output buffer is represented by the schematic of

Rgure 13. The IN and lNBinputsshould be driven differentially
from VCC to VCC 1.8V. The ESD protection circuitry is
the same.

H603 Output

Two input structures can represent most of the inputs for the
H60x family of translators, one for TTL inputs and one for ECL
H605 Output
INIINB
inputs, The exceptions were discussed previously and the
~F..various Inputs and appropriate input models are summarized
V(
in Table 1. For the dual supply devices with ECL inputs the
H604, H606
VCC and the VEE on the typical ECL input gates should be
output
tied to ground and 5.2V respectively, All input pins should
1
have both a package model and ESD protection circuitry
H607 Outm$.. ~G~NB
vcc tO VCc - o.e5v
connected to them. ForTTLinputsthe ESD protectloncircuity
IS represented by the FPSO09E schematic of F!gure 15, For
ECL inputs the ESD protection circuitry is represented along
with a 50KQ input pull down resistors part of the ECL In Pad
~f~: important to properly apply the power supply voltages to
Cell represented in Figure 15. The Package Model of figure
s<:~ccUTately model these circuits. This section will explain the
15 is self explanatory, the parasitic values provided are worst f
@wer supply terminology used on the l/O buffer schematics
case numbers. The package capacitance combines with th~t:~~and how to properly apply these supplies with the appropriate
parasitic transistor capacitance of the input device and ~@&r.
package model.
ESD circuitry to cornpnse the load capacitance of the inpu~~
The various input buffer ESD circuits are outlined i~~gur#15,
Table 3. Powar Pin Descriotiona
notice that the ECL inputs utilize a different stru~~,than
the
TTL inputs and outputs, The typical ECL {Q~<sC%ematic
Power
Description
represents a single ended ECL Input, ,~~&~reference
EVCC
EVCC IS the most pos!t!vesurzplvfor the ECL
should be tied to VCC - 1.3V and the V@,,@$, $hould be tied
Input gate (+5V forthe H607 aid ground
to VEE + 1.3V. To simulate a differenti~~~k$hput
one simply
for H60WH606)
I
connects the complimentary i~~~,,~~ the VBB
side
vEE
VEE is the mostnegatcvesupplyfor an ECLgate.
of the input gate along wit,~e %$~%sociated ESD and
For the H607 it IS equal to ground, for the
package model. The differ:~akwt
does not use the VBB
H600-H606 (t ISequal to 5.2V
switching reference.
:\. .>
,..:~
:.it,.i?\,,,.\r
TVCCI
I InternalVcc for TTL circultrv
I
For all of the inp@ a~ output buffer schematics the
GNDI
I
Internal
,round
forTTL
cfrcultrv
I
resistors should ~@~#&Mulated
as simple SPICE resistors.
Because these ?~st$ts are realized by a diffusion Step in
Table 3 lists the voftaae supDlles referenced on the l/O
wafer procem,g tti& are parasitic capacitances associated
schematics alona with ~ description of each. The kev to
with each.i%~~ybcircuit schematic is shown for the resistors
properly simulatm~ these power ;upplies is in the application
in the Q:~t@J~odeY
schematic of Figure 15, The value of
of the package model. Because the output buffers, to a varying
eac~~bwit
resistor is one half the value given on the top
degree, share VCC and ground pins, a@ustments need to be
l~~e~&&tic
and the parasitic capacitance ISmodel led by a
made to get a more accurate model if all of the outputs are not
t.~I~e back biased to VCC, Also note that the resistor
simulated at the same time. If for example a single output is to
! te@&rature coefficient (TC) values for both the resistor
\:>,; ~.
be simulated the package model for the TVCCI and TGNDf
,JF~J *dbcircuit and the resistors in the device subcircutts are
supplies should be scaled based on the number of outputs
*!:&< $: provided. For modelling at nominal temperatures only, these
which normally share the supplies. If the simulated output
<$*.*, **
,,$~)>$,i. TCS can be omitted. If however modelling will be petiormed
normally shares its supplies with two other outputs the
~$y
at the temperature extremes the TC information should
.,
package inductance would be tripled to simulate the same
be included.
inductive glltch seen on the power pin in an ectual application.
Table 2 is provided to summarize the various internal
The capacitive value for the package model is not as ctitical
voltage swings and bias levels required to run the appropriate
and thus can be left alone. This method will allow users to
SPICE simulations.
more accurately model an autput behavior without resorting to

397

be to manipulate the generic netlists. If, however the netlists

are desired or questions arise about the contents of this
document the user can contact an ECL applications engineer
for assistance,

more accurately model an output behavior without resorting to

more complicated and lengthy simulations. The internal power
and ground pins are ail powered through a single pin and are
basically static, as a result no adjustments are needed for the
package models on these supplies. Table 4 outlines the
internal power distribution for the H60x translators, this
information can be used to determine the scaling factors for
the package inductance for the output buffers. To use the table
simply identify the output in question and divide the number of
outputs in the group by the number of power pins for that
group, this will give the multiplication factor for the inductance.
Summary
The information included in this kit should provide the user
with all of the information necessa~ to do SPICE level system
interconnect modelling. The schematic information provided
in this document is available in netlist form through EMAIL or
an IBM or Macintosh disk. However with todays advanced
design tools it will probably be a simpler task to enter the
schematics in a gwd schematic capture package than it will

398

PKG

NI

FPM1

DS1
GRW1
m
2
TC = 4.4sE4,

2,78E+
REPI
15.2
1

RW
13.7
D3uD&l
n
N2

399

PKG
IN
I

mli

VEE

Hgure

4. H602

ECL

Input

Gate

400

figure

5. H6@

ml;put

Gate

v=
Hgure

6. H600,

H602

401

Output

Gate

PKG
IN D

ECLIN
Pm CEU

CBIAS

IN

INB

,,.,.
.,.. .

Hgure

8. H6D3 Output

402

Gate

FPN025X

Rgure 9. HW1 Output Subcircuits

403

NI

DS1
WN05

RT
5,4

4t
P
~

N2a

Q1
PNN05A
PN::5B

1 r

D2
DSUB2N05

OSUfll N05
TC = 4.45E-4, 2,78 E-6

N1

-.,,,
RX
2

R4
7.89

DSI
GR139

A
d
TC = 4.45E-4, 2.78 E-6

REPI
15,3

REXT
22.9

DSUtiSO03

N2

FPSO03

NI

R1
19,1

DS1
FP025X

N2

PN:;5X

TC = 4,45E-4, 2,78 E-6

,:.

FPN025

FPN025X

figure

10. H603

Output

Subcircuits

404

-..-....
-_

b
N3

Hgure

1ss

VEE

figure

12. H605

Output

405

Gate

1-

g Q:;:14

PKG
OUT

Q17
IPN13
~
ESD

406

NI
o
DS1
WN05

RT
5.4
4)
Q1
PNN05A

P
*

N2~

PNt05B

TC = 4.45E.4, 2.78E.6

1 L
DSU%N05

DSUVIN05

QPNN05M

=
n
i3

o
i4

,..
NI
?
RI
19.1

DSI
FP025X

QI
PN025X

N2

DSU~025X

TC=4.45E-4,278E-6

FPN025X

N3

Ni

R1
19.5

0s1
W108

08

*DS;l

Figure

N3
14. H607

Output

407

Subcircuits

RPKG2

RPKG1

1
T

?;?

Package
(28-lead

Model
PLCC)

408

(1S=1 .0578E-12

.MODEL WN05 D

RS=37.6

N=l ,044 TT=IOPS

CJO=141 .75FF VJ=.4 M= 33

EG=.69 XTI=3 FC=.5 BV=30)

(IS=l E-16 RS=O N=l H=500PS
O

.MOOEL OSUBSI14
+

CJO=2.75PF

EG=I

VJ=.51 M=.24

115 XTI=3 FC=.5 BV=35)

(IS=2.52E-12
RS=l.35

.MOOEL QPS114 O
+
CJO=2. IPF VJ=.4 M=.33
+

N=l .044 TT=1OPS

EG=.69 XTI=3 FC=.5 BV=30)

.MOOEL OSUB025X

O
(CJO=284FF VJ=.51 M=.24)
(IS=4.32E-17
BF=I13 NF=I .006 VAF=30 IKF=1OA
.MOOEL PN025X NPN
+
ISE=O NE= I BR=5 NR=I XCJC=. I VAR=IOO

st.,\
.,.,... )\i.
, ::;~~.,?,i.
*.,>
,.y ,,~,i:>,.ii.
,.
.:, : !
..
,,,,* .:&/y:*

f. ,,,,~$~+.

,1,, ... .,7

>,,.\.,+>, .11..
$?:7
+
CJE=193FF VJE=.9 MJE=.4
~e.~lt$,j+<~+
~$$

,.,,., ~!,. ,
+
CJC=I 58FF VJC=.53 MJC=.37
!$~,,,.,*<:
,.
TF=40P XTF=O ~F=l
00 ITF=5.7MA PTF=O
+
.:,
-.,
- ~:,.
+
TR=200P XTB=l .51 EG=l ,115 XTI=5 FC=O.5 )
+q&
?
i,,
~?$,
.MOOEL FP025X O
(1S=1 .08E-13 RS=4S3 N=l .044 ~=1 OPS
,,p. . ,,!M\\,,
,,
+
CJO=90FF VJ=.4 M=.33
J;:>,,t,@
,,
..:.,.,.:,,.,.
+
.~?,:x
EG=.69 XTI=3 FC=.5 BV=30)
... -.tt
i;:::~~<?.(.?.::,.?,
.MOOEL 0SUB025 O
(CJO=284FF VJ= 51 M=.24)
J,\.
.MOOEL PN025 NPN
(IS=2.45E-17
BF=113 NF=I .008 VAF=30 IKF=IOA
,..
i
~~:~
>.:,
+
ISE=O NE=l BR=5 NR=l XCJC=.1 VAR=IOO
,.$,$:i\*,>*.
+
IKR=l MA ISC=2.66E-16
NC=l RB=193 RBM=89
~,,f,i
..{**,.,>,.
t
~~
RE=O RC=62
+
~ R*.
t
,6
,.il?~,.l.
>,,.~~
+
CJE=123FF VJE=.9 MJE=.4
:>,l.>:~>~
+
CJC=I08FF VJC=.53 MJC=.37
..
?,
,.*?
+
+
TF=40P XTF=O VTF=l 00 ITF=5,7MA PTF=O
... -.$,,
+
TR=200P XTB=l .51 EG=l .115 XTI=5 FC=O 5 )
,,.
~.
+

IKR=1O.65MA

RE=O RC=35,2

.MODEL FP025 O
+

ISC=4.66E-16

(IS=l .4E-13

NC=l

RB=175 RBM=65

RS=52 N=l 044 ~=1

CJO=I17FF VJ=.4 M=.33

EG=.69 XTI=3 FC=.5 BV=30)

OPS:i]
$:i..:1><,,
*,2~
~...
~f~v..$>.(,
*.<

.MOOEL 0SUB139 O
(CJO=2.12PF VJ=.51 M=.24)
a+.
.MODEL PN139 NPN
(IS=l.03E-16
BF=113 NF31.00&VAF=30
ISE=O NE= I BR=5 NR=l XCJC=.1 V&=100
+
+

IKR=4.4MA

RE=O RC=641

ISC=l

+
+

CJE=493FF

22E-16

NC= I R~=l#kRBM=47
L,. +..
~ ,~~,..,:..!>~.~,>,

VJE=.9 MJE=.4 *IF 7J$,

CJC=244FF VJC=.53 MJ~:~,,,#

TF=40P XTF=O VTF=IQO ~~%.7MA

TR=200P

,MOOELGR1390

PTF=O

XTB=l 5%1$~$$4$95 XTI=5 FC=O.5 )

(lS+$&-Y$;~=10
N=l .044 TT=1OPS

CJO=68FF

EG=.69 XT/~.FC#5

v&*

w
BV=30)

.MOOEL GRSO03 O
** /~**8.27E-14
RS=53 N=l .044 ~=1
+
CJO<,:\~\V}&.4
M=.33
+
E@:6~*$?l=3
FC=.5 BV=30)
(IS=l E-16 RS=O N=l ~=500PS
.MOOEL OSUPSOh~;
+

:J,QJO=f27FF
::/~$,~@=l.115

IKF=IOA

OPS

VJ=.51 M=.24
XTI=3 FC=.5 BV=35)
(CJO=1O6FF VJ=.51 M=,24)

:MOR%&@,u~ogE
o
(IS=3.92E-16
BF=113 NF=I ,008 VAF=30 IKF=1OA
.M@\@
@W09E NPN
..)~ k.
ISE=O NE= I BR=5 NR=I XCJC=.I VAR=IOO
*,t$~\+:.
.\ ~,
.~i~
IKR=.3MA ISC=4.25E15 NC= I RB=165 RBM=39
.,.\!#,.
+ ~:.:\
.>>

RE=O RC=3.9
+.:.
,,$~$?$$3+*.$
CJE=I .37PF VJE=.9 MJE=.4
*,,a,*y
,i?*\<4, ..,.
,. .,x..
!;:,,,,
.*,
.

+
+
+

CJC=609FF VJC= 53 MJC=.37

TF=40P XTF=O WF=I 00 ITF=I .64MA PTF=O
TR=200P XTB=I.51 EG=I .115 XTI=5 FC=O.5 )

,MOOEL GRO09E O
+
+

(IS=5.4E-13

RS=9.57

N=l.044

TT=1OPS

CJO=663FF VJ=.4 M=.33

EG=,69 XTI=3 FC=.5 BV=30)

409

.MODEL DSUBI 08 D
.MODEL PNI08

(CJO=163FF VJ=.51 M=.24)

(1S=1 .75E-17 BF=113 NF=l.008

NPN

+
+

ISE=O NE= I BR=5 NR=l

XCJC=.1 VAR=1OO

IKR=.75MA

NC= I RB=638.8

RE=O RC=87

ISC=l.9E16

VAF=30 IKF=IOA

RBM=222

CJE=90.6FF

+
+

CJC=50.3FF VJC=.53 MJC=.37

TF=40P XTF=O VTF=l 00 ITF=4.I MA PTF=O
TR=200p XTB=I.51 EG=l.115 XTI=5 FC=O.5 )

+
.MOOEL WI08

VJE=.9 MJE=.4

(IS=5. IE-13

RS=56.6 N=l .044 TT=IOPS

CJO=66.3FF

EG=.69 XTI=3 FC=.5 BV=30)

ECL Transistor

st.,\
.,.,... )\i.
, ::;~~.,?,i.
*.,>
,.y ,,~,i:>,.ii.
,.
.:, : !
..
,,,,* .:&/y:*

VJ=.4 M=.33

f. ,,,,~$~+.

,1,, ... .,7

,. s,, .,+,!..
~+,~,t .
~:f~>?.::i,
\i:>/\*.
$.~,,e .{*.
~,!.: ,,C
. ,.,

Models

.MODEL T0511 NPN

+

IS=21.16E16

ISE=250E-I

ISC=95.62E-18

EG=I ,11

CJE=67.7E-15
CJC=99.5E15

VJE=I .037 MJE=.5718 NF=I.000

VJC=.603 MJC=.266 NE=2.000

+
+
+

BF=I12
8

BR=5.108

RB=52.7

RBM=O
RC=26.33

RE=l.533
IRB=O

IKF=.0213 VAF=41.8
IKR=53E5

NC=l .141

,., h,
~,~~:t..r:i:,
:!, ..+
,,i ~~:.>,
,,$.. ~,,+. $\\:$*
*:$,* ,.(,,..,
.>, .,J@>
.,

VAR=3.766

NR=.997
XTI=4.7
XTB=l .15

CJS=I 52 E-I 5 VJS=.5052 MJS=.3465 TR=9.92E-9

PTF=20
TF=35E-I 2
XTF=225
VTF=I .67 ITF=,00808 XCJC=,069 FC=.8

.MODEL TPNP2 PNP

+

IS=7.69E17

CJC=I

BF=5 BR=l RB=164 RC=56 CJE=.086E12

4E-12

MODEL T0611 NPN

+
IS=33,33E16
BF=114,5

BR=2.029

ISE=I

RBM=O

ISC=184.7E18

OE-15

RE=I

333

IKF=.0338 VAF=42.7

~,,
.><{.
,.>5
,>:.), ,
.... .
+
CJE=99.3E-15
VJE=I .037 MJE= 5716 NF=l.000
XTI=4,7
,+ .?
+
CJC=124,4E-15
VJC=.603 MJC=.266 NE=2,000
XTB=l ,15
.$3>kh ,$
\*\,-.,.*,..,
+
CJS=170,4E15 VJS=.5052 MJS=.3465 TR=9,92E-9
PTF=40
tb~$.
?,.. ~.
+
TF=35E-I 2
XTF=2.25 VTF= 1,67 ITF=.00806 XCJC=.089 FC=,6 ..:,-*.
fk~~
.MODEL T12B1 NPN
+

IS=5.7E-17

ISE=2.4E-15

RB=56,6

EG=l .11

BF=I13

RC=22.86

BR=I

RB=170

IRB=O

$~:$$
*3:..i]
.
.$A::J:..>>P
,>)(:,!$,
.
~.,+.-,~.$
!,.
[!
!?
,!$: .${n ,,>
,,.,:*:.

IKR= 115 VAR=3.665

NC= I.085
NR=.995

116 RE=I .25

RBM=170

IRB=l

IKF=,0828

~F=4+$
...
lKR=j2W{.~AR=3,6

7E-3

ISC=l .OIE18

EG=I .11

RC=133

+
+

CJE=15E-15

VJE=.656

MJE=.273

NF=I ,000 ,$

CJC=27e-15

VJC=.603

MJC=.369

NE=2.000J$+,#

CJS=I 01 E1 5

MJS=,259

TR=5E&.,

TF=39E-12
MODEL T5406

+
+

IS=3.3E16
RC=23.6

VJS=.429

XTFf=3
NPN
BF=II 3

NC= I .028

VTF=l .4

NR=I ,Q%9 ~1=3

?.i}.
~?~.<,s}
.,:,
.$?,

RE= 833 CJE=.495E12$~JC:i%E1

:-.

~,!.:~

lTF=.00@::@/&.620
1,
.
~s:~J
RB=86.6 BR:L: .<?j&:$

FC=,005

2 CJS=.576E-I

410

~~.
/,

AN I 404

ECLinPSTM circuit

at Nonstandard

411

Perf~$@~nCe
VIH i,~~vels

ECLinPS Circuit Petiormance at Non-Standard VIH Levels

VIHmex and the ECLlnPS Family
As previously mentioned the MOSAIC IIF process allows
for ECLinPS devices to operate at V]Hmax levels somewhat ~~~~,i
higher than those specfid
in the databcok, however t~~ *>$
exact value of VIH for which saturation problems will o~ur~~:~
varies from device to device and even among differenli~.~~+~
for a given devke. This variation is a resuk of the diffe&.~~@@
mnfiguratbns
used on the various inputs @~~&#PS
!\~<\$,>,
devi~s.
0..
,,$,r>i)!..!:;:1$

The easiest way to define an acceptable*~~#ax

for each
device in the family is to define at what ~~:hetipm
transistor
wiil saturate and specify for each i &J~?t the worst case
input transistor mllectorvoftage$l
% %~~lth
:
this information
designers will be able to determ~$~$a
part by part, input by
input basis what input vohaq~,{~~~~~!ll be acceptable fOrtheir
.,,, ~
appticat ion.
i).
.$??::4.,
,..~
* *,$>,,
~
Simulation Res~&$,~v
> >,+. \?<\
The input,<.sq[~~n
phenomenon was characterized
through SPIC@:$],mulations and the results will be reported in
the follo~,@text:Fors implic~ of simulation a buffer similar to
the E~& W% used.
Since the outputs of this buffer drive off
chi~~th~~ifimax
performance of this structure will be worse
tti~~~fipical
input structure, Both a 10OK and a 10H style

When interfacing ECtirrPS devices to various other

technobgies times arise where the the input voltages do not
meet the specification Iimks otilined in the ECLinPS data
book. The pur~se
of this document is to explain the
consequences of driving an ECUnPS device with an input
vohage HIGH level (VIH) which does not meet the maximum
vokage specified in the ECLinPS Databook.
The resuhs outlined in this document shoufd not be viewed
as guarantees by Motorola but rather as representative
information from which the reader can base des~n decisions.
It is up to the reader to assess the risks of implementing the
non-standard interface and deciding i that level of risk is
acceptable for the system design. Motorolas guarantee on
VIH will continue to & the sp=ification standards established
for the 10H and 10OK ECL technobgies.

The upper end of the VIH specof an ECLinPS, or any other

ECL, input is limited by saturation affects of the input
transistor. Figure 1 below illustrates a typical ECL input
(excluding pulldown resistors and ESD structures); the
structure is a basic dfierential amplifier configuration. Wth a
logic HIGH level asserted at the input the collector of that
transistor will be pulled down below the VCC rail by the gate
current passing through the collector load resistor. T@
vohage at the mllector of the input transistor (VC) wik~
dependent on the gate current and the size of the m~to$$
. .,..x.~,
.2~$..
~,:*
load resistor associated with the input gate.

{S*
we~e analyzed to note any discrepancies between the
*standards.
As expected the simulation results showed no
~ difference in the saturation susceptibilky of a 10OK versus a
10H style buffer. Therefore the simulation results of only the
10OK style buffer will be presented to minimize redundancy of
information,
The following text will refer to Figures 48 in the Wpendix of
this document. Rgures&8 are graphical plots of the input and
output waveforms of an El 22 style buffer (structure similar to
that of Figure l) forvarDus VIHlevels. V(in) represents the
input vokage while V(q) and V(qb) represent the output
vokages. The V(vbb) line was included for measurement
purposes on~ and will be ignored.

*,,:.(,

Figure 4 represents the standard operation of the device

as a standard VIH input was used. Note that in this rendition
the propagation delays measure in the 21&225ps range and
the IINH was 42,5PA. The IINH of this device is simply a
measure of the base current of the input transistor when that
transistor is mrtducting current. We will be monitoring both of
these conditions as well as any degradation in the output
waveforms as a sign of the input transistor becoming
saturated. As can be seen in Figures 5 and 6 none of the
parameters change for VIH levels of up to +.4V. With a
collector voksge, VC, of-l .OV these VIHS corres~nd to a
collector base forward bias of 600mV. As the VIH of the input
moves closer to VCC, F~ures 7 and 8, three phenomena stati
to occur: the IINH increases, the delays increase and
significant changes occur to the omput bw level of the 06 pin.

..

-@~&%~ut VIH increases towards VCC the ~llector base

ju~,~fibfthe
input transistor becomes forward Mased; as this
~@Frd
bias condition increases the transistor will move into
~e saturation region. The value of VCB at which the transistor
begins to saturate is process dependent and will vary from
Iogicfamilyto Iogicfamily. Fotiunetelythe MOSAIC Ill process
used to implement the ECLinPSfamily incor~rates a deep n+
mllector doping. This deep collector helps to mitigate the
effects of saturation of transistors by requiring a larger
collector-base forward bias to enter the seturatbn region.

In Figure 7 the IINH of the input transistor has more than

doubled from the standard level. This increase in base
current leads to an increase in the VOL level as the collector

412

..

...

beforathay are fed intothedifferential amplifier input gata. The

switching referance is also shiftad down by one dioda drop to
remain centered in the input swing. Obviously this input
structure will reprasent the tist caae in the araa of extanded
VIHmex performance. In fact this type of input structure will
albw for input vokages even several hundred millivoks above
the VCC rtil. This characteristk makas these typa davices
idaal for interfacing with differential oscillators whose o~uts
Ia& any DC offsat. In the emitter folbwer structura th&.*g
factor will be the saturation of tha emitter fol~~i+~vim
whose collector is at VCC. From the prav~~t,sl.@tilation
results this would suggast a maximum VI~,;~+~P~~V.
,,,. ,1

current must reduce to maintain the anstant emitter current.

As the mllector current reduces, the IR drop across the
collector bad resistor redwee, thus raising the VOL level on
the QB output. Afthough the VoL level has shifted the overall
propagation delay has remained essentially unchanged.
Finally, when the input is switched all the way up to VCC the
VOL level no longer remains in SPW as the input base current
has jumped to almost 1ma and there has been significant
degradation in tha high-bw propagatbrr delay. h is a~arent
that for this rendition an El 22 style buffer will not ~rform
adequately for most systams.
From this informatbn it can ba wncluded that for a
collecfor-basa forward bias of s600mV thare will ba no
adverse mndtions on the performanof the device. Tha
performanm starts to degrade wfih further forward bias until at
a forward bias voftage of =1 .OV tha davice will fail both its DC
and AC specifiiat ions.

ln~t
ECLinPS Input Structures
There are four basic input structures which will affect the
VIHmax performanm of ECLinPS devicas. The four structures
are as follows: an internal buffer, an external buffar, an emitter
follower inp~ buffer ati a series getad emitter follower input.

,:l{~#M
2. Emftter Follower Input Structure
...
The internal buffers are input structures whose otiputs
,~~ TheJ~arias gate emitter folbwer input will represent the
drive other qates internal to the device, the vohage swings of
worst @se situation for a 100E device. Figure 3
the input transistor collectors (VC) on thase de~cas w~l be .,,,: ~oluta
400mV. An external buffer is ona in whichtha ordputs are ~d~ wt~epresents a series gate emitter follower input for a 10E and a
q OOE davice, From this figura k is apparent that the bwer
etiernal to the chip. Because of the relatively Iarga base dr!ve.~+1,,1.
switching level (B input level) is going to be much more
of the output emitter followar for thase structures the$:.ti$+~
susceptible to VIHmax for tha 10OE device than tha 10E
voltage will typically be a couple hundred milivoks lower th~
device. Tha two diode drops used for the 10E device is not
for the internal buffar. Note that bacause of the I@er @Btput
possible for a 100 Edeviceduetothe
smaller VFF
swings of a 10Edevke, a 10E style external buw~ill
require
vokage of a
10OE device.
a VIHmax input level more near the specifie~q$ld& Both of
these structures are similar to that ptiur@@.~Ure
1.
To summarize tha external gate will represant the worst
.*Y ,,\,..
,x,
.,,~..
case VIHmex stiuation for a 10E device while the series gate
The third and fourth structures ar~i:~,~hat
different in
emitter folbwer case will represent worst casa for a 10OE
design than the first two. F@ur~~&,~~~$frates an emittar
device. In either stuation the standard emitter followar will
follower input structure. For the b~~~yerfollower
input the
allow the most leeway for non-standard VIHmax peflormance.
input voltages are dro~ad Xn~@Wiional
VBE (=800mV)
,T}$.>,:::.,..,,~!.
1DEStmcturs

Vcc
T

VBB

Figure 3. EmMer Follower Serise Gate Input Struoture

413

Conclusbns
Simulations show that forward bias Iavels of <6~mV on the
input transistor will keep the input transistor in the active
reghn end tha performance of ths device will not be
mmpromised. This forward bias voltage can be increased
with varying degrees of performance degradation to levels
somewhat higher than 600mV. Initial effects will be an
~~+,\
increasa in the IINH current and a dwraaee In the output VOL .:.$~.,ir~~
level on tha QB output of the input gate. As the forward M@$~:F
increases further the propagatbn delays through the dq~e-~?
?. .,
will be adverse~ affectad.
Lt, ?{$
: \ ~,..
.,,,,,.
:\.
... ,
The following example will outtine ths use of th$$~bl~:~the
appendix to analyze the ~tential performance~$ <.,$esign
using non-datiard
VIH levels. If a desi@S@lXfor
the
10E1 12 and the 10E416 to be driven by ~$+:~~@ut
signal a
designer would want to know if the~wo
&vices WOUM
perform to specifications under th~&~*itions.
From the
table the worst case mllector voh~~~~~.~ould be 1 .05V and
O.OVrespectively. Subtracting *:*:+#ues
from +.2V yietis
forward bias vokages of -~~~d
-200mV respectively.
From this information the de~ner WOUMmnclude that the
10E416 will function @~fio pr~lems however the IoE112
muld suffer perfo[@~Wdagradation
under these same
~,.\t:t
conditions.
3..<, ,,,;
.:.1.
, i+,:.,%.ht.,
The devbetirmatton
contained in the appendix of this
documen&Will ~vide
desgners with all of the information
necass~fltQ
evaluate the input transistor forward bias
co~l$~
f~? all of tha ECLinPS devices for different input
vq~,~S@,Wth these numbers and the information provided in

When driving ECtirrPS devices wth other than standard

input levels there is another phenomena that should be
considered; name~ eff-s
of non-centarad swtching
references on the AC performsof a device. For
non-standard inputwkages the mtipoint of the vokaga swing
may notcorrespond to the internal VBB switching referenm. If
this istha @se the resuftingAC variation should be iwluded in
the evaluation of a desgn.
An input vohage swing not centered about the swtchiW
raferanca will exhibit a delay skaw between the two inputedge
transfiions. Tha size ofthis skew will bedepandent on boththe
vokage offset of the referenw voltage and the midpoint of the
inputswiW and the slew rate of the inWt as it passes through
the threshoti region. As an example for the case in which the
VIH =+.5V and the VIL remains et -1.7Vthe midpointof the
swiW will be at t .tVversus a-1.32V VBB raference. Wfih a
typical slew rate of 1ps/mV for ECLinPS typ edge ratas the
rising input edge delay will be 220ps bnger than normal and
the falling edge delay will be 220ps faster. This results in a
440ps skaw &tween the two input transitions that would not
be seen for an Heal switching reference,
The only means of mrrecting this skew is to bwer the VIL
level to recenter the swing or provide a different switching
referenca for the device. The latter can be accomplished by
buffering the signal with a dfierential input device with one
inpt tied to an externally generated switching reference,
Raising the VIL level is not recommend~ due to the obvious
loss of bw end noise margin accompsnid by any such shti:

.~~~@8cument designers will be able to make informed

Wslons
abut their designs to meet the performance
,,,pde& red at an a~eptable level of risk.

414

-.....

Appendix
Input

Mb

INT

E016

All

E101

All

EF

DM

EXT

E104/107

Dnb

SG

All

INT

Dn

Em

Elll
E112

ENI

INT

E116

All

EXT

E122

All

En

E131

INT

Wr

..

SG

E141

All

INT

E142

All

INT

E143

All

INT

Elm

Dn

EXT
INT

Wr
E151

All

INT

E154

All

INT

E155

All

I NT

+.80

El%

All

INT

~.80

E157

Dn

EXT

4.90

SEL

INT

4.80

E158

Dn

EXT

+.90

SEL

INT

~.60

EIW

R, CLK

SG

1 .20

INT

4.80

-r
E163

All

INT

4.80

E1S4

All

INT

+.80

E166

All

INT

4.80

E167

All

INT

+.80

E171

All

INT

+.80

El 75

All

+.80

E195

All

4.80

E196

All

4.80

E212

NI

4.m

E241

All

E258

All

E3w

,~b

E337

%:,,.;
All :

4.80
.C,
:

4.80
INT

4.80

INT

+.80
0.00

EF

E404

,,

E416

*+,,All

EF

All

INT

4.80

All

INT

4.80

All

INT

~.80

Dn

EF

SEL

INT

late; EXT:

twal

Gate; EF

0.00

0.00
4.90
L

SG= Sanes

415

4.80
@td

II

It

4.%

4.5

Rgure
5. Input and Output Waveforms for VIH -.5
(VoL = -1 .8; TPD++ = 204ps; TPD-- = 207ps; llNH = 4.4pA)

416

V(IN)

v(Q)

TME

F@ure 7. Input and Output Waveforms for VIH = 4.3

(VOL = -1 .8; TpD++ = 196ps; Tp& - = 198ps; IINH = 114.8wA)

417

418

----

.<
.~+.

d:
q~~%ing
dA&~@ *
:Q*(~fino@ies.
k$$.
,,

419

/icdion mteprovides information on system

ECL &k
txhnomies
for rducing
skew over the ahernsfffe CMOS and ~L

ECL Clock Distribution Techniques

lNTRoDucTfoN

for TTL and CMOS devices. Bemuse of the near zero duty
cycle skew of a dtierential ECL device the output-to-outpw
$:.\,+
skew WIII generalfy be larger. The omput-to-utput skew IS .~~~,,;.<:
important in systems where either a si~le devi~ can prov~~r&~,~
all of the necessary clwks or for the first level device$$:fi~ o
nested clwk distributiontree. In these two situations@ @&:~
parameter of importance will be the relative ~sk$a~h
output with respect to the other outputs on th~~@. S#ce
these outputswillall see the same environme$@~~#mss
~tiit~ns
the skew will be sgntiicantl~.J.eS~ lharr the
propagation delay windows specified in @e sf@ard devbe
,~~>.
~~~j
~
data sheet.
:!:r{\.&t,
:.t
,. B
,t~

The ever increasi~ performance requirements of todays

systems has plamd an even greater emphasis on the design
of bw skew CM generefbn and distributionnetworks. Cbck
skew, the difference in time between simultaneous C*
transitions within a system, is a mapr com~nent of the
mnstr~nts whchform the upper bound for the system cti
frquenq. Reductions in system cti
skew allow designers
to increase the petiormance of their designs wkhout having to
resort to more mmpfiatad archtactures or more mstly, faster
logic. ECL @bt*no@iesoffera
numbarof advantagaefor
reducing system CW skew over the affernattie CWS ad
~L techno~ies.

V,,,.P
SKEW DEFINMONS
The skew introdud by @ic devkes mn be divided into
three parts: duty cycle skew, output-to+ utput skew ad
part-to~rf skew. Depending on the s~fb
epplkatbn, each
of the three mmponents Mn k of qud or over<ding
imprtance.

;*

The duty qcle skew is a measure of the dfierence between

the TPLH and TPHL propagatbn delays (figure 1). Because
differen~s in TPLH and TPHL will resuff in pulse wtith
distordon the duty qde skew is eometimea referred to qg
pulse skew. Duty cycb skew is important in applcati~~,i
where timing operetonsmron
hth edges or when the~ty
cycle of the cbck sgnal is crtiical. The later is a<,wm
requirement when driving the ckk
inputs of~~@:J&ad
,,k~:,
-.$:.,.,,,.
mcroprmeseors.

outp*b@Guf
.*...!., . .

,,,%*,

qJPafl-to-Part Skew
The part-to-part skew specification is by far the most difficuff
performance aapect of a devi~ to minimize. Because the
part-to-part skew is dependent on both process variatbrrs and
variations in the environment the resultant spectiicetwn is
sgnificantiy Iargerthanforthe other two components of skew.
Many times a vendor will provide subsets of part-to-part skew
spacifkations
based
on non-varying environmental
renditions. Care should be taken in reading data sheets to
fully understand the cotiitions under which the spactiied
limits are guaranteed. Hthe part-to-part ~ew is spacfied and
isdtierent than the spacfied propagatbn delay window forthe
device one can be assured there are constraints on the
part-to-part skew specification.
Power supply and temperature variadons are mapr
mntributors to variations in propagation delays of silicon
devims. &nstraints on these two parameters are mmmonly
seen in pan-to-part skew specifications. Aftbugh there are
sfiuatbns where the power supplyvariations muld be ignored,
if is difficult for this author to perceive of a realistic system
whose devices are all under idandcal thermal renditions. Hot
spots on boards or cabinets, interruption in air flow and
variations in IC density of a board all lead to thermal gradients
wthin a system. These thermal gradients will guarantee that
devices in various pans of the system are utier dflerant
junction temperature mndit!ons. Ahhough it is unlikely that a
designer will need the entire mmmerclaltemperature range, a
prtbn of this ra~e will need to be considered. Therefore, a

skew

,:k.

o-output skew is defined as the differewe between

~;&agatbn
delays of all the outputs of a devke. A key
q$streint on this measurement is the requirement that the
ohpuf transfibns are identical, therefore if the skew between
all edges produced by a devke
is important the
out~-tmutput
skew WOUMneed to be added to the duty
qcle skew to get the total system skew. Ty@cally the
out~t-tmoutput skaw will be smaller than the duty cycle skew

420

part-to-pert skew spedfied for a single temperature is of Iile

use, eapecial~ i the temperature mefflcient of the
propagation delay is relefive~ large.

inherent differences between the TpLH and TpHL delays in

addtion to the problems with non+entered swtching
thresholds. In devirns spectibal~ dasgnsd to minimtie thffi
parameter k generally cannot be guaranteed to anything le~
than 1ns.

For designs whose cbck distribution netwoks tie on a

single board which utihzes power and ground planes an
assumption of non-va~ing power supplies would be a va~i
assumpt.bn and a specification hmitfor a single power supply
WOUB be valu~a.
H, however, various pi-s
of the total
distribution tree will be on different bards wthin a system
there is a very real wssibil~ that each devbe willsee dfierent
power supp~ bvels. In this case a s~fi=tbn
timkfor a fixed
VCC willbe inadquateforthe desgnofthesystem. Meal&the
data sheets for CM
distri~on
devtes shouti include
information which will allow designers to taibr the skew
ewrwettins
of the davice to their application environment.

SYS~

The m~orcontributorsto outp~-toautput skew is IC layout

and pa&age chobe. Differences in internal paths and paths
throughthe padsage generally can be minimized rsgar~ss of
tb sificontechnobgy ufitized at the die level, ther%~~QL
devkes offer less of an advantage in this area t~a @VQ!her
skew parameters. CMOS and UL output perf~r@R~ is tied
cbsely to the power supply levels and the st,~~]~~ytie power
busses wfihin the ctip. Cbck distribut~~~q+ ~ definitbn
always swfich simuhaneous~, thu~:,,~a*g
sgn~iant
disturbances on the intarnal wweh.~%
To alleviate this
problam mutiple Wwer and grou~ pl$sare utilied on TTL
and CMOS CM distributbnd~~6&&owever
even wkh this
stratagy TTL ad CMOS c~,~ttibution
devbes are hmiad
to 500PS - 700ps outR@k~@@ut skew guarantees. With
differential ECLoutpu~&@li~e
t any noise is generated and
muded onto the inf~.p~,-er
supphes. This mupled with
the faster propagawblays
of the output buffers produces
output-to-out~eke~
on ECL cbck chips as bw as 50Ps.

ADVANTAGES OF ECL

skew Raductl*s

ECL devias provide superior performance in all three

Two a&8~%CL
cloti devims will lead to sign~cantly
areas of skew over their ~L or C~S
mmpetitors. A skew
smaller {~a~$$:part skews than their CMOS and ~
reducing m=hanism mmmon to dl skew parameters is the
~~~~m+?aster
propagation delays and delay insensitivity
faster propagstkm delays of ECL devices. Since, to some
to e~onmental
variations. Variations in Drooaaation delavs
etient, all skew represent a percentage of the typical delays
,:.~th pr&ess are typically going to be bas& on; percentage
faster deteys will usual~ mean smaller skws. ECL devices,
,.
o$~befypical delay of the device, Assuming this percentage is
especially c~k distributmndevices, can be operated m ether +~+<$..
~~oing to be approximate& ~uivalent between ECL, ~L and
single+rsded or differential modes. To minimize the skew ~f:.
> *#-CMOS processes, the faster the device the smaller the delav
these dsvbes the differential mde of oosration shoutd ~..
varistbhs. Because state-of-the-art ECL devkes are at least 5
times faster than TTL and CMOS davices, the expected delay
variatbn wouti be one fiih those of CMOS and ~L devices
without even mnsidering environmental dependencies.
The propagation delays of an ECL device are insenskive to
variations in ~wer supply while CMOS and UL device
propagation delays vary sgnificsntly with changes in this
parameter. Across temperature the psr~ntsge variation for
all technologies is comparable, however, again the fester
propagatbn delays of ECL will reduce the magnfiude of the
variatbn. Figure 4 on the folbwing page represents
norma~isd propagation delay versus temperature and power
su~y for the three technologies.

,$

--

~~yb ,

:3 ,:,

mMY&~i,.-

\$:$\]
*Y,;:,*

~~~:w+
FKuro *J~~
Induoad Duty Cycfa Skew
,4,
*
>..

Low Impadan@ Line Driving

ECL o@,~sb,@ers ioherent~ show very little dtierence

bstwew TQ~#~nd TPHL delays. What dtiere~es one does
sw,@~*%ainly
to switching referen~ levels which are ~t
~a~,,~tered
in the input swing (see Figure 3). For worst
,.o~s~iching
reference levels the pulse skew of an ECL
\ d~ti
will still be kss then 300ps. Hthe ECL device is used
.:5 a~erential~
the varisfiin in the switching reference will not
&:rImpact
the duty cycle skew as it is not usd. In this case the
t,,,. $.
**$!%,
.. ... pulse skew will be Isesthan 50ps and can general~ be ignored
~.}
.......
in all but the h~hest performance desgns. The pro~em of
,?j
generating &s
which are cep~le of meeting the duty cycle
requirements of the most advan~
mcroprocessrs, WOUM
be a trivial task Kdifferential ECL compatible CH inputswere
used. ~
and CWS cbck drivers on the other had have

The cbck requirements of todays systems necessitate an

almost exclusive use of mntrolled impedan~ intermnnsct. In
the past this requirement was urrque to tha ~rformance
levels as~iated
with ECL technobgies, and in fact
precluded its use in all but the hghest performance systems.
However the hgh performance CMOS end ~L
dti
distributionchips now require care in the design and layout of
PC boards to optimize their parforman~, with this criieria
estebhshed the m@ration from these technobgies to ECL is
simplfied. In fad, the diffku~ies involved in designing with
these sbwe~ technobgies in a mntrolled im@ame
environment may even enhance the potential of using ECL
devkes as they are ideally suited to the task.

421

1.W
1.U
1.M
1.02
1,01
1.00
O.B
0.9s
0.97
0.s
0.%

OH

O.w

1,02

1.M

1.10
.$.\*.{:
$.%
.. .\\, $,.<$
: ..:,,

F@ure 4. TpD vs Srrvlronmental Condition Comparison

It is true that dflerential

Tha bw impedanca outputs and hgh im@ance inputs of

an ECL device ara deal for driving 50Q to 130Q controlld
impedanca tranamissbn lines. The specified drivi~
impedance of ECL is 5~ however this value is used only for
convanianca de
dua to the 50Q impedance of most
commonly used measurement equipment. Utilizing hgher
impedance fines will reduce the power dissipated by the
tarminsfion rasistore and thus shou~ be considered in power
sanahive designs. The major drawback of hgher impadanca
Iinas (dalays more d~ndent
on capactive bading) may not
bean issue in the point to point interconnect scheme ganarally
used in bw skew cbck distribution dasigns.

mora S@flab

to be routed on the PC boar~::$~%unately with the wkfe data

and address buses ,:#~\Qays
des~ns tha CIO* hnas
represant a small f$w
&the total interconn~. The final
choi~ as to whettir o~dt to use differential interconnect Kes
in the level of #e,@~ormanu
necessary for the dasign. k
should& not$~@atalthough single-ended ECL provides lass
ettracfiv~$~~w ~rforman~
than differential ECL, it does
provi~q~,sl~fi~ntly
better performance than equivalent
CM% SW ~L functions.
.-~+.,
i...,.,,.,
g>i~;-,
.~ml

.,.}::.:

t.:~ ~~:,,

The device ekaw minimization aspects of dfierep~q

*
have already been di~usead howaverthere ara O$~~g6Y&eM
Ieval advantages that shouti be mentioned. w~~t~~~bhk
lines are distributed over bng distances tha ,%fl@the
tine
and the variatbns in power su@y upset th@,Q~JWlationship
between inpti vohagas and swfichin~ t~e$rds.
Because
differential intermnnect carries ~& ~w~hing threshoti
informetbn from the source t~Y@~~,:~@ the relationship
~tween tha two is less hkaly t~ k~hangsd. In addition for
IOW Iinas the smaller swin~~@{*@L
devke produce much
Iowar levels of croes-ta~a%~een
adjacent tines and
minimizes EMI redist~ttmrn%e PC board.

/
1

L-

,.,

-_d

Cb
~
Figure 5.1

There is a cost a&$@

with fully dflerential ECL, mora
pins for equivai~~.t~~bns
and more intermnnect to k laid
on a ~Peal@~~~d~mowdd
PC board. Tha first issue is
real~ a ,,,~+pw~
for cbck distribution devices. The
output-t~~u[~
and duty cycle *aw
ara very much
depsad~:W@ quiet internal power suppties. Tharafore the
pi@<&~med for the amphmentary outputsWOUWotherwise
.~<$~~ be used as power SUWW pins, thus functbnality is
Waity gained for an equivalent pin munt as the inversion
fhdbn
is sfeo available on a differential devbe. The
praeence of the inverfed s~nal wuld be invaluable for a
desgn which docks bth off the positive and negative edgas.
Figure 5 shows a mathod of obtaining very bw skew (40ps)
180 shifted two phase clocks.

till

I
I

,.+)

Difforentbl Interconnti

,.$<

,~~$- ~~
,
in@&k&Wrequires

W Shifted

Two Phase Clocks

USING ECL WITH POSITfVE SUPPLIES

It is hard to argua wth the cbck distributbn advantages of
ECL presenlsd thus far, but it may be argued that except for all
ECL designs it is too costly to include ECL davices in the
distributiontree. This claim is ba%d on tha assumption that at
least two extra @wer su~lies ara required; the negative VEE
SUWIYand the negative VTT termination voltage. Fortunately
hth these assumptions are false. PECL (Postive ECL) is an
acronym which describes using ECL devices wih a pceitive
rather than negative power supply. h is important to
understand that all ECL devicas are also PECL devbes. By

422

using ECL devices es PECL dev.on a A voR supply and

inco~tating termination techn~ues which do not require a
SSperefe te~inatbn Voftege (series termination, thevenin
equivalent) ECL can be incorporated in a C~S
or ~L
design wth no added met.
me reason for the choke of negative power suppties se
stand=d for ECL h due to the fact that all of the outpti Ieveh
and internal swbhing bias Ieveb are referenced to the VCC
rail. ftisgeneral~ easiarto keep the gmundsquietar and equal
potential throughout a syetam than it is with a pcwer supp~.
Secausethe DCparamatersare referenced totheVCCrail any
dmturbencesorvokagedr~
seen on VCCwilltranslate 1:1 to
the output and internal reference levels. For this reason whan
communicating with PECL between two boards it is
recommended that onw dtierentid intermnn~ be ud. By
using dfierenfial interconnect VCC variations within the
Swfied range Will not in any way affect the performance of
the devbe.

problem of generating runt pulses whan an asynchronous

disabia is usd. The device also provides a muxed dock input
for inm~rating a high speed syetemcbck and a bwer speed
test or -n
dock within the same distribution tree. The
ECUnPS Ell 1 devim is used to receive the s@nals from the
-Iana
and distribute t on the card. The worst =se skew
between all 54 cbcks in this shuatbrr would be 27*s
assuming that dl the bads and sgnal tr-es are equahzed.

finally mentioning ECL to a CMOS dasQner invariably

conjuras up visions of space heaters as their perceptmnof
ECL is hgh pcwer. Ahhough it is true that the static power of
ECL is hqher than for CMOS tha dynamic power differences
betwean the technobgies narrows as the frequency
increases. As can be scan in Fqure 6 at frequencies es bw as
20 MHz the per gala pcwer of ECL is actually leas than for
C~S.
Since CM distributbn devices are never static t
does not make sense to compare the pwerdissipation of the
two techno~as
in a static environment.
~}?$~ed Technology Distribution Natworks
~.

.:..;
,J.!,

.$

!,,

.:<*.,

b<~:,a,~
~;$,

Building do& networks in ~L and CMOS svstems can be

a little more complicated as there are more akernatives
available. For simple one level distribution trees fanout
devices like the MECL t 0H6& 1:9 ~
to ~Lfanout tree can
be used. However es the numbr of levels of fanout increases
the addition of ECL devices in an other wise ~L or CMS
system becomes attractive. In Fgure 8 on the ne~ page an
Ell 1 device is mmbined with a MECL H641 device to produce
81 ~L level cbcks. Analyzing the skew between the 81
cbcks yietis a worst =se skew, albwing for the full
temperature and VCC range variation, of 1.25ns. Under deal
situations, novariatbn intemperature or VCC supply,the *ew
WOUMbe only 750ps. When mmpared wkh distributkm trees
utitizing only m
or CMOS technobgies these numkrs
represent =50Y. improvement, more if the environmental
condtions vary to any degree. For a 50 MHz ckk the total
skew between the 81 ~L cbcks is less than 8.5Y~of the clock
wriod, thus providing the desgner extra margin for layout
induced skew to meet the overall skew budget of the desgn.

+)

Many desgners have already realizd the benefts of ECL

ckk distributbn trees and thus are implementing them in
their desgns. Furthermore where they have the capstilky, i.e.
ASICS, they are building their VLSI circuits wth ECL
ampatible cbck inputs. Unfotiunately other standard VLSI
circuits such as mkroprocesmrs, m~roprocessor suppoti
chips and memory stillclingto~Lor CMOS clock inputs.As a
resuk many systems need both ECL and ~L chks wkhin the
same system. Unlike the dtuation outfined in Figure 8 the ECL
levels are not merely intermediate sgnals but rather are
drivingthe cbck inputsof the bgic. As a resuk the ECL edges
needtobe matched withthernedges
as pictured in Fgure 9.

;$S8

Clock distrhutiin in a ECL system is a relatively trivial

matter. figure 7 illu~rates a two level cbk distributiontree
whch produces nine dtierantial ECL ctis
on six dflerent
cards. The ECLinPS E211 devke givas the flexibi~i of
disa~ing each of the cards individual~. In addtion the
synchronous registered enables will disatie the device onk
when the clock is already in the LOW state, thus avoiding the

423

~- ___

1-

Figure 8. ECL to RL

Clook Distrlbut&n

~-y

used. The value of the delay element WUM be a best guess

estimate of the dtierences in the two propagatmn delays. ft is
hgh~ unhkely that the temperature coefficients of the
propagettin delays of the ECL devkes, ~L dev~es and
delay devbes WOUUh equal. Akhough these problems will
add skew to the system, the resuhant total skew of the
distribution network will be less than t no ECL chips
st.,\
were used.
~.. ,:>::,,
*.,> $:{*.
,.y ,,${~J
PLL Based Cl*k Drlvere

A potential solution for the problem omlined in F~@~?#

the use of phase hked bop basal c&k distri -Wipe.
Because these devices feedba& an output ~ 8.,@&i to a
reference cbck input the delay differen,~~,~~ben
the
varbus technology output buffers will ~lell~inated.
One
might believe that with all of the eup@ria~~@?rounding the
performance of PLL based CW diswn
devkes that the
need for any ECL in the distrib~b@$&@ will be eliminated.
However when analyzed furfh#j@
opposfie appears to be
c\..t~!; P
the case.
~.
.~,>
,F.:i.{\~<
For a single board des~:~h
a one level distribution
system there obviously~ino ntied for ECL, When, however, a
mukiple board syst~~i:~++aquired where nested levels of
devices are nee#&,~
once again hmes
useful, One
ma~r aspd~f
~$$-part
skew for PLL based cbck chips
ofien overb+is
the dependence on the skew of the various
referenc%eloc&being
locked to. h can be seen in Figure 10
the s@@,
part-to-part skew of the devba would necessarily
ne~~.~
a~ded tolhe reference clock skew to get the overall
s&MQf @e CIOA tree, From the arguments presented eartier
\&&~.,<.$,c~
.,~~~w
will be minimized t the reference cbck is distributed
Ih,~CL. k has not been shown es of yet where a PLL based
.<.E~L cbk distrib~ion chip can provide the skew pedormarrce
of the simple fanout buffer. Fmm a system standpoint the
buffer tyW circuits are much easier to design wfih and thus
given equivalent performance would represent the best
akernative. The extra features provided by PLL based chips
muld all be reahzed i they were used in only the final stage of
the distribution tree,
Unfotiunately none of the PLL based devices available
today feature differential ECL compatible reference cbck
inputs. Look for BiCMOS based PLL clock devices from
Mlorola in the near future. There will be a family of devices
featuring varbus technology mmpatible inputs and outptis to
allow for the building of prmisely algned cbck trees based on
ekher ECL, ~L
or CMOS (or a mixture of all three)
mmDatible levels.

424

._-._

@rrcluslon

F@ure 10. Syetm

me bst way to m=imize the performof any

sy~hronous system is to spend the entire CM
perti
performing value added oprstbns. ~vbusfy any Prtbn of
the wriod ~nt
idle due to dd
skew limits the
potential petiormance of the system. Using ECL twhm~y
devkes in cHdistributbn
netwo~s will minimize all a-s
of skew and thus mssimtie the performanm of a SY~m.
Unforturrste~ the VLSI wofld ia nol yet ECL clti -~
that the benefts of a totalfyECL based diatribution~~~c~not
be reahzed for many systems. However there ~,e,~~~s
of
inmrporsting ECL into the intermediate Iev& ~h~ tree to
sgnfcantly reduce the overall skew. In ~fhe system
designers
can utifize their new f@@ Mowledge to
inmrporate ECL mmpatible cti~,,e~;w
VLSI *ips of
whch they have mntrol while at ~
~me time pressuring
other VLSI vendors in ting thq~ame=
that future designs
mn enpy fullythe advant~e~~,~:tributing
d@s with ECL.
~,
~,,w::t
,,:
...
.
.i.
,... ,.

Skew For PLL Clwk Distributors

425

426

EB27A
Get 300 watts PEP Linear
Across 2 to 30MHz from this
Push-Pull Amplifier

,,..~
.,.~,~..
,\,,,,,.~

,.&Tdowwith

amp~ier

is sufficient

only for shoti te-

Wriods

under forcd

air coohng.)

Because of its excellent load and line voltage regulating

capabilities, an integrated circuit bias regulator is used
in the amplifier. The MPCIOOO, originally described in
this bulletin. consisted of a MC1723 chip and a built-in
pass transistor. The manufacture of this device has been
discontinued however, and the board lay-out was modified to inco~rate the above two in separate packages.
The load regulation typically measures less than 2Tc at
cument levels up to 0.5 A, which assumes an h= of 40
for the RF power devices, The board sufiace provides
a sufficient heal sink for the 2N5990 pass transistor, but
a separate heat dissipator, such as Thermally 6107 can
be added if necessag. With the component values
shown. the bias is adjustable from 0.4 to 0,8 volts.

This bulletinsu~}w~ sufficient information [o build a

push-pull ~~wr amplifier for 300 watts of PEP or CW
output, p&~~&$,@$ross
the 2- to 30-MHz band. One of
Motq~*J~~new high-power transistors developed for
s~n~~-~eband, MRF422, is used in this application.
,),+, ..:,
e,$<>~Q*.
,
~, L$?all transistors in its family of devices, MRF422
.$~$,t@mbines single-chip constmctimr that is advancing the
:,.~.)!.!,s,
.3,
,>\ state-of-the-afi, and improved packaging to accommo.+-~:$.f.
..
$?.:l,~~.
~~~ date the low collector efficiencies encountered in class
,+},
B operation. Rated maximum output power is 150 watts
CW or PEP with intemodulatimr distortion specd at
30 dB maximum, 33 dB typical. Although not recommended, a saturated power Ievef of 240- to 250-W
is achievable. Maximum allowable dissipation is 300 W
at 25C.
427

Trandormer Construction
Gain flatness over the band is achieved using base input
networks R, Cj and R2C3 and negative feedback through
R3 and Rd. The networks represent a series reactance of
0.69 ohms at 30 MHz rising to 1.48 ohms at 2 MHz.
A single-turn winding in the collector choke provides a
low-impedance negative feedback source, thus R3 and
Rd determine the amount. The reactance of C4 reduces
feedback al high frequencies with the result that feedback
increases an average of 4 dB per octave at decreasing
frequency.
For continuous operation at full power CW. it is recommended that heat sink compound, such as Dow Coming #340, be applied between the board surface and.,R3
and Ri, and if possible have air circulating over ~%~
::. -,?,
of the circuit board as well.
. ,~:..$
,,,.a,,:...~,
.

mented together for easier constmction. The primary is

made of Y~copper braid. through which three turns of
No, 16 Teflon@ insulated w]re are threaded for the
seconda~,

Term!nal
Feedth,

Pin Sand
ough$

Feedthroah
51..0

Eyelets

Offs

A.

428

.-.. .-

..-

.-

--------

3M-Wati

Linear Amp~ier

Schematic Diagram

-*

Q1,

Q2

T1,

T2,

MRF422,
T3

All capacitors
are .hiDs
Uio.

Q3-2N5990

See texl
except

.lecYrolvt!cs

and C16

Carbide

or Varadye

type

1813

ad

1225,

size 18 or 14, or equivalent

Table 1. Output
harmonic
contents,
measured
at 300-W CW (all test data
taken using a tuned output,
narrow
band sif Iai source).
2nd

@Registered trademark of DuPont

((

f (Mhz)

3rd

below

4th
he cal

5th
sr)

30.0

-38

25

-34

48

20,0

-33

13

43

-45

15,0

-50

lo

-51

-47

7.50

40

-30

55

47

4.0

-37

22

-55

37

2.0

36

-18

-45

-37

PCB, chips capacitors, transformers TI. Tz. Tj, and femite beads are available from:
COMMUNICATIONS
CONCEPTS. 2M8 N. Aragon Ave.. Kettering. Ohio 45420.
Telephone: (5 13) 294-8425.

429

430

T*.w
!p.i .diii
cj*rp;:>~iii
:*;>,i<:
il r:r*,ilii
I?l:~cii;d

i(iY;rr*.,s:ih
r(i

;I

.lil
iii

kc

il~illf~C

Ljr:\Ld
ii:\

~:wi~d.
L:llli~P: l:lililL:i:
s I\ XdiiLL,\!
1,) 3 \\ i ..!I$ (>\_I ?!,&I ,,: ~\.:ny,~-il~~,

!-/:)
iJ>i*

*I:%\
I\l :lr

.:i;. :.i,.
*,~\t*:-

trid !,I, i:tw !iidl iI< ~i?:~L~i~

:h:b:iii: H.I:CJ 1: *:,i;ii:.
,~iC,Wii, iirppci, /,! s.t!Lrw,! i3 211~.I,:; \2,I ail!4 *ia>
dwi
ill: /i pd~rm.mt
n:, ii* i.,arw fliJ> ::!.?I hiiir: iir:&yd ;&!I t!ir j!i,i: W!i it:ii; b C!i.. INi,Eii \t.,dtw~,l!lr: TId?:Sl;lii h.r:,o;!i tl I!!; i:iclill.
i\p!Ld:i
.I 1L
bn.iril, Illll?i,irZ 2~s:::ra- Lhi !Y iiiO,,,i ii<.,1 ;::lou,~
rile tltd :liv.i/ hii :3i?i.\: :I,di I!li. lh,il &.+I

EB59
Predict Frequency Accuracy for

Crystal oscillators are used when it is necessary to gener.

Crystals used wi~~,$*060/61
devices must meet the
ate a precise and highly stable signal, Such circuits typirequirements _
in their data sheet. Since these
devices ot~~te ?? the frequency that protides the low.
tally provide this stdble signal at a frequency close to
the resonant frequeocy (either parallel or series) of their
est impe&e&(series
resonance) between pins 5 and 6, a
/>:,1,>,*
crystal. However, circuit components and other factors
cryst~~ m~ttiot exhibit a spurious response resulting in
iW$q,&X values near or less than the desired series
external to the crystal influence the crysta~s natural
.~,i.,
,.
resonance to some degree, an effect often referred to as
.$ re~{mce impedance, In the evaluations discuswd here,
~SJi. standard commercial crystals with fO.0025Yocdibratimr
pulling or warping. A discussion of the variation in
,.<!
,.
crystal frequency as a function of differing ICS*, temper- ~,:~:i,.
*
.Wlerance, fundamental mode, were used with the
ature, and dc supply voltage is presented in this bullet~ ~, $ MC12060/61 devices. Measured series resonance fre.
series inductance (Ls) and resistance (Rs) values are
presented in Table I.

,.,,,

*Specifically, the Motorola MC12060/12560

and
MC12061/12561 integrated circuits which are designed
for use with an external fundamental series rewnant
crystal. Specified operating frequency range is 100 kHz
and 2 MHz to 20 MHz
to 2 MHz for the 12060/12560
for the 12061/12561. Complementa~ sine wave, com-

\:\.!,t\

As shown by the equivalent circuit 9$%?% 1, crystals

behave as open circuits to dc. Fd.~
ti~nals below a
$;. ~..,,$V
crystals series resonant freque.~~~.~
t~crystal
exhibits
,\,,,,~\ v,\\,i
increases, the series
a capacitive reactance. As fr~~,~~
resonance of CS and LS i%.keah~?d.The crystal then appears as a low value ~~st&xs~s, shunted by a small
above series resonance,
capacitance, Co. ~)J{~~~~cies
the CS, LS comb~at]~~ appears as an inductive react....... .$i>.
... ,
ante. As fre @efi@YHtreases even higher, the inductive
%,,. $
reactance &ro~+<@entuaIly equalling the capacitive reactance ,@f,lQ~.~fis is the high impedance, parallel reso,. .;,... ,.
nant ff~w$~~~ for the crystal. .Mthough the separation
in ~~&&VCy between series and parallel resonance varies
~~~~ffirent
crystals, series resonance will typically
rw,*~c& several hundred Hertz to a few kilohertz below
,,,

TABLE I
Crvstal
Series
Resonant

(MHz)
2.500025
8.079977
13.411100
18.749563
19,999528

F=

FIGURE

1 CrvSal

Equivalent

Frequency

Circuit
I

433

(kHz)
100.002
200.012
500.031
999.985
2000.032

Parameters
Equiv.

Series

Resistance
(Ohms)

38.0
8.4
6.9
12.5
9.2

497
509
995
380
96

RS

Equiv.

Series

Inductance

(mH)
274.0
17.6
7.0
2,9

9857
2629
526

LS

elementary ECL, and sin~e ended TTL outputs are

avaflable. Complete technical specifications for these
ICS can be found on the detice data sheet. Additional
applications information is avaflable in Motorola appli.
cation note ~-756 and engineering bulletin EB-60.
MC 12060/61

MHz. Table 111shows the variation in pull on the same

crystal resulting from the use of different MC12060 and
MC12061 devices.

Figure 3 gives the frequency shiftcaused by the

,,. ~:t.,,,,
..t~>,<
,\$Jp:<+
MC12560/61 devices operating over their temperature
..,.bfa~,
.
range of -55 C to +125C. Similar results can be ex:
$,..*
petted for the MC12060/61 devices over their specif~?,.~.~>
range of OC to +75C. Data was taken with the c ~kwa
a
.
at a constant temperature of approximately .X*33QJ8
isolate the effect of temperature on the IQj-@@ the
~,,.~!.:~,, .~<~i
....
cumes are normalized, one must add t% a~rbpriate
room temperature value (see Table 11)~.o o%Wfrr the net
frequency pull at a specific temp::~:,w
For example,
the MC12561 device operating ,w~,i~he nominal 8.08
MHz crystal would exhibit a.,&$~,~ of approximately
.~)k\.

Performance

The circuit elements in an oscillator environment have

an effect on the fundamental resonant frequency of a
crystal. To measure the influence of the MC12060/61
devices, tests were made using the circuit of Figure 2.
Frequency measurements were taken at the sine wave
output (pin 2 or pin 3), the 680 ohm resistor making it
possible to drive a 50 ohm load. bboratory quantities
of the ICS were tested, consequently some variation in
results could be expected if a production run cross section were evaluated.

SIGNAL
OUTPUT

I
<,,

,V
,t~: ~p+>.$

The measured pull of ,i$~et$~l 2060/6 1 devices On a

-40 - 11 = -51 PPM at +125C. The curves show a
is shown in Table II
small temperature dependence at lower frequencies that
crystals series reso~nt ~~uency
for room temper,a&$S#o<eratiorr. Resonant frequency is
increases significantly above midband. Although not
L, ~
plotted, over the -55C to t85C range MC12560 at 2
always reduce.d,:~$e@Tfectbecoming more pronounced
.?:..\:,l,
i
with ticre~~~ @rating frequency. Where minimum
MHz and MC12561 at 18.75 MHz changed from +155 to
pull is r~gu$~$:the MC12061 rather than the MC12060
-275 and from +7 to -4S PPM respectively, referenced
to +25C,
shoulg<b~!i$dfis]dered for use at or slightly below 2.0
a.
g, ~~+
<$,/$~::V.:$
TABLE II
,4.
.?. e,
.\~,;>..
ICS
Crystal Frequency Pull In Percent For MC12060/61
.,+ \ .X*

@.$..:,/.,,,.,
:}.
.?,

DEVICE
NOMINAL
FREQUENCY
CRYSTAL

MC12060

CRYSTAL

0.100

MC12061

0,200

0,500

1.00

2.00

2,50

8.08

13.41

0.0005

a.oo12

4.0040

4,03

0,0002

4.004

0.0

PULL

IN

PERCENT
.LESS

18.75

20.0

0,03

0.05

(MHz)

THAN

1 Hz, MEASUREMENT

LIMITED

BY

RESOLUTl

434

ON

OF TEST

EOUIPMENT.

0.500

4.0
10.0
82,5

2
10
166

1.000
2.000

pacitor and its effect on increasing frequency. Therefore,

if onfy a amaUincrease in frequency is required, the trim
capacitor value may become unreasonably large. To
assure a suitable value for the capacitor, it may be neceamry to specify the cryatd frequency lower than the
actual desired operating frequency. The pulhg effect of
the ICS will norm~y be much less then that of the trim
capacitor and therefore the crystal can simply be spWified such that the series combination of cWstal m~$AJ,
capacitor is in series resonance at the dedred @t#%
frequency. If it is also desired to account fo~~$,~]kcts
of the ICS, this may be approximated by .~,$,@%.~g tie
MC12060 to add 266 ~ and the M$l~lA>l 6 @ in
.~i,.t, ~~~
\\,.t\.,:,~>
series with the c~staL
k a typical example, a~ume th@&~~l 2061 is to be

M C12081
0.8

2.50
8.08

110
485
1765

13.6
36.2
93.6

13.41
18.75
Leas thn 1 Hz, Mmsurement Iimitd

by rwlut

ion of teSI

Wuipment.

10

o
z
%

a
..,:/

10
:(.!,
NOTES:

2.
20

FREOU$~~.@FT
AR E@P~kD

8ECOMES SIGNIFICANTLY
WORSE
ABOVE MID-FREQUENCY
RANGE.

WHENDEVICES

100
TEMPERATURE,

130

~ >~,,

l.~q>\ ,.>

gbk 4 provides plots of frequency pull as a function

$thange in dc supply for the MC12060/61 devices.

used with a nominal 8 MHz csystal having an equivalent

series inductance LS = 17.6 mH. Figure 5 shows the
equivalent circuit. With no CTNM added, the lC will
lower the crystals resonant frequency by approximately

~
~!,~,. *.J..\.{,.>

t$..~t?v Msign Example

f~t,
,{:i.*:\.+c,m
.%
The ICs are designed to pull the crystals natural series
\,,,*,.
resonant frequency lower.. If desired, this permits a trim
f]
capacitor to be inserted in series with the crystal to set
the oscillator on frequency. Since this trim capacitor
is approximately in wries with CS of the crystal, there is
an inverse relationship between the vahre of the trim ca-

~(17.6 + 0.0016)/17.6 or 0,0045%. Use of a 10 pF trim

capacitor would place a net impedance in series with the
crystal of jtiLIC - j l/mCTRIM = -j 1.909 x 103. This
corresponds to an equivalent capacitance in series with
1
= 2n x 8 X 106 X 1.909 X 103 =
he crystal f CEQUIV

435

m.

10

Af = O FOR MC12061
CHANGED ilO%,

OPERATING

WITH 2,5 MHz CRYSTAL

AND Voc

200 k HZ

o
t
MC12060
1

o
MC12061

AT 18.7 MHz -

20
I

10.42

pF. The crystal

should

~.~.:,i
~~,

@* sp%iffed

requiring an undesirably
large value of 247 pF. The
CTRIM value can approach infinity if the crystal calibratimr tolerance allows the crystal to be series resonant
on the high side of 8 MHz.

so that the

crystal reactance and that of C~~~~

are in series reso,.4$.?,
. ~ ,,
nance at the operating freq~~nc$i(&.OOO MHz). In effect,
this requires a crystal=w~te~$eries
resonance slightly
below 8.000 MHz so t~~<~$ precisely 8.000 MHz it presents a ~ imped?e@>+whl~h equals the -j impedance SUPplied by CTRI~{~$~~@S with LIC, i.e., CEQLIIV.
$,
-:.::*.r.\\.:,l

A similar procedure can be followed for the MC12060

device. In this case the approximation
LIC = 266 PH is
used. The calculated frequency pull for the nominal 500
kHz, 1 MHz and 2 MHz crystals described in Table I is
then -0.0013.
-0.0051
and -0.0253
percent respectively. This agrees closely with the measured values of
-0.0012, -0.0040, and -0.03 percent given in Table 11.

If the crfi{~~$riot
resonant below 8 MHz as suggested,
but ~g.~~ M exactly 8 MHz, CTRIM must then be
cho~h ~~$esonate with an LIC inductance of 1.6 #H
,,,,$ ,.:,,l,,,
j!
,$:,:s.
-,~:.
,, -~..
.,, ~$.,..>..

436

__

R1

L3
===

L1
c1

T1

RF
1

.w~=Fqc

4:1
m

Cl

=-

C2, C8 -27PF
C324PF

Ch,P
Chip

15PF

capacitors

Style

Chip

C4-

L1 0.15
L2

C6, C? 50 pF Chip

L3 0,15

#H

1 Turn

C1O10PFChio

L4 Ferroxc.
01

C12

150PF

Ch,p

C13

270PF

Ch,p

C14,

C16

680

C15 1.OPF

T1,

Z1 M icrostr@ &}ne
i, . ~.
80~>ti&$%&:x

Molded

Choke

AwG,

Molded

1/8,,

with

Ferroxcbe

o grod

ed

of coil
Z2 -

ID

T2 -

be VK.200.1914B

50 V Tatalm

2,25 inches

FIGURE

3 2N6439

(57,15

t,,> ,::
*$!,
225

m ,1, w

L X 5.715
Line
L X 225

Coax (Type
mm) long

UT25)
,,$:,

60 Watt B.ildin$&qck

,,$.**?
.is~. :>;.,.. ,
,,\ ~:, l?:$,
.,q
bii @oard
+.

13.97
0.031,,

LX
mm

LX3.175mm

(0.787

mm w

125mil~w

mm]

W
Glass Teflon

,, = 2.56

%+*.,,*
~\i,$:!,
....
225-400

550mils

mm W

milsw

~~:~~,$+a mm L X 5.?15
Z3, h~,:- Microstrip
Line

watt
25 Q Subminiature

~&.~%+
W$:$trjp
\\~,*ils

Choke

2N6439

R1 -70~2

PF Feedfhru

Co,

BAG

#56-590-65/4B
#22

#H

C1l

DFCh,o

ace 100 mi] TDK.ACr

FC282

Bead

C5, C9-30PFCh,p

-5,1

-==
All Chip

63PFCh,p

MHz

@ -01875 C o~ecm)

I
FIGu R E 4 Construction

Details of the 4:1 Unbalanced

to Unbalanced

Trantiormers

438

.-....-

AMPLIFIER PERFORMANCE

439

~~-.

...___

~--~^I.

.-

. .l..-~_--l-.~~~-~~~.-~-~--~.l.l.,l_,~~
------I

:
!

1
I

2lFtir ;a,
tie_--

~-q-----+--

l(.a

4&...\<
tw

;.;,,,:

zx

!
!
8
/

---_lllll_._l.

_ _

_.

_ ._.,_ __..^ .-..-__ II

..~.-~-.-.-

- __I ..__ ___ ~.-1-11-

.l..

.-l...

_-~-.,-

. ..--._ .~ -~-...

_ -.- _.__.._;~_;

The input and output impedances of the transistor

are matched to 50 ohms by double section low pass
networks. The networks are designed to provide about
3% 1 dB power bandwidth while maintaining a collector efficiency of approximately
3070. There is one
tuning adjustment in the amplifier Cl in the output
network. Ceramic chip capacitors, C2 and C3, are used
for DC blocking andpower supply decoupling, Additional low frequency decoupling is provided by capacitors C4 and C5, Refer to Figure 3 for a 1:1 photomaster
of the circuit boards.

Amplifier

~00

05

1 0,

1
FIGURE

3 Circuit

Photomaster

FIGURE

Assembly

The circuit boards are mounted on a 3,127 x 1.87Y

aluminum block. A 0.062 deep and 0.260
x 0.750
tide slot is milled in the heat sink as shown in
Figure 4.
The transistor mounts in the slot with two 4-40
screws. An alternate approach that would eliminate
the need for milling is the laminated structure shown
in Figure 5.
$:.\,+
Using the laminated assembly, the transistor is .~~~,;.<:
mounted on the surface of the block and 0.062 alumitt.~ .$~~
num shim stock is sandwiched between the block ap~, i $
the circuit bosrds. Connector mounting plate$,l~k<~
required if SMA type connectors are used for}@#@
input and output. The SMA connectors can b#T~taR&d
directly to the block if the milled approq$~~$:dsed.
Either method results in the same perfo~~a~e$or
this
1-watt design. The laminated struct~~ Hwver,
may
not be suitable for higher power d~ws.
@ith higher
power levels the transistor imped@&~$are lower. The
RF ground impedance throu@ &&~~minated metal
may be sufficiently high to ~,@~i&*gain and stability.
This point emphasizes th~~$~~that
the successful
design of RF amplifier: is ~$~~rident not onlyon attention to electrical coRsl&ratlons,
but to the physical
construction as w$$JJ]@iTe construction related parasitic cannot be @ta~}~nored
at medium frequencies,
they can p@q5wIous
problems
at microwave
frequencies. ~.fg recommended that the following constructio~@hri~~ues
be followed when building this
amplifi$~kt~efer
to Figure 6 for the component

4 Amplifier

442

Heat Sink

l_l-___l.~

~.____--.

~-.---

!
i ,// ,l.l Rlhrisl

_.^ _ .-..- .

~--

---._-I

111_--------__,

_I~-

r7
jj

~-

---

1
--- irMr* bxr / ii*,*!
[jr,* ) / IV. iii ,!,/

FIGURE 7 Pedorrnance Cuwes

Pefiormance

Data
I
I
VCC= 24Vdc

Amplifier tuneup is accomplished by adjusting Cl

for maximum output power tith minimum collmtor
cument. The amplifier till tune from 2.25 b 2.35 GHz
while maintaining an input VSWR of less than 2:1.
Typical
performance
curves appear in Figure 7.
Figures 7a and 7b show performance
with the
amplifier retuned
for each frequency.
Figure 7C
shows performance
without re-tuning. Note from
Figure 7C that the instan~neous
1 dB bandwidth is
approximately 70 MHz tith the amplifier tuned to a
center frequency of 2.3 GHz.

1,4

1,3

2,25 GHz -

2.30~H;

I .2

. q

<

2.35 GHz

1.1

10

0,9

/
1

0.8

120

100

80

160

140

P,n, INPUT POWER (mW)

FiGURE7a-

1
1
P, = 150 mW

-1

1.4

F
/

1.3
1.2

1,4
VCC = 24 Vdc
1,2 -pi=,50mw

I
2.25 GHz

1.5

2,30 GHz

1.0

z
~

0.8

t~

0.6

0.4 -

<
k

>

1,0

35

z
5
0

1.1

Power

Pout
\

2.35 GHz-

Output Power ver$uslnput

30
/

25

- 20

3.1

09

VSWR

2.1
a

vcc, suppLY

2.25

28

24

20

2.30

2.35

f, FflEaUENCY(GHz)

VOLTAGE (VOLTS)

FIGURE 7b Output Pmer

1:1

versus Supply Voltige

FlGURE7c

NOTE

444

Output Power. Efficiency and

VSWR verss Frequency

The MRF2N1

lsoe.f

olP.1

as #dbcar&

powers

8famnly

.f2GHzpwer

lranslstorswlth

below

MRF2001

1 w

MRF2005

5 W

MRF2M3

3 W

MRF201O

10 w

RF

*F

--

;a-*_ \~

_ _

Design Considerations
The lower frequencies (136-160 MHz) are serviced
by a design utilizing low-cost dipped silver mica
capacitors. For a broadband response in the higher
frequencies; (160-174 MHz), low inductance, ceramic
chip capacitors are used.
Ease of assembly, repeatability and fast economical construction received the utmost consideration
in the design of this amplifier.
TO-220 devices
result in a low profile circuit which minimizes the
volume occupied by the amplifier. Additionally, the
MRF262 transistor used in the output stage is a
rugged device, able to tolerate high load SWR
conditions.
Maximum
use of printed inductors
assures good repeatability.
Both amplifiers utilize stagger tuned networks to
enhance bandwidth.
Additionally,
each design
retains excellent gain and stability characteristics
when narrow banded. All of these merits are
attributed
to optimum
device
gain and the
reasonably
high inter-stage
impedance
levels
incurred at these power levels,

At frequencies beyond 100 MHz, dipped silver mica

capacitors generally become inductive, and do so
with a high degree ofunpredic
tability.
This
phenomenon
is also dependent upon component
value and becomes
more pronounced
with an
increase in frequency. (Ret 1, 2, 3). To maintain
predictable performance beyond 160 MHz, a second
layout featuring ceramic chip capacitors is offered
(Figure 3,6, 7). The design of these capacitors allows
them to remain capacitive beyond the VHF frequencies. Maintaining
the bandwidth of 160-174 MHz
with this circuit board, the networks become Iossy
and power output suffers
slightly.
Variable
capacitors may make this condition more tolerable
and can be installed in the input and interstate
networks. In some cases the ease of adjustment and
added flexibility
would justify the added cost of
the variable capacitors.
Performance
Normally, this amplifier will not require tuning
provided that components are as described and are
positioned as shown on Figure5 and 7. If an accurate
method of measuring power is available, a quick
check of amplifier performance can be accomplished
by comparing its parameters with the performance
data of Figures
8 through
11. Drive must be
maintained at 220 mW (f20 mW)and VCC held to 12.5
Vdcto accurately reproduce the overall response
noted here. Allow some degree of tolerance (1OYO)in
output power to account for differences inherent in
component values and transistor performance. To
assure broad band per forma nce and tailored
frequency response, the amplifier should be checked
using a swept frequency
generator
capable of
200-300 mW output. Tuning for maximum power out
and minimum reflected power at band centers will
not necessarily
provide a broadband
response.
Figures 8 through 11 graphically depict typical levels
of performance achieved with this amplifier. Either
version is stable into higher than 3:1 VSWR load
mismatch at all phase angles. The output device is
tolerant of short term operation into an open or short
circuit load at full drive.
Harmonic content of a 150 MHz signal at the output of the dipped silver mica version is illustrated in
Figure 12. The2nd harmonicas approximately -50 dB
with respect to the fundamental.
This level of
performance cannot be maintained across the entire
band, therefore, some additional
filtering of the
output signal will be required to meet more stringent
requirements.
With the amplifier mounted on aluminum stock,
2.0 x 8.5 and 0.090 thick, a 25% duty cycle(l min on,
4 min of~ produced a temperature of 500C (122F)
after two hours of operation. A 50% duty cycle (1 min
on, I min offl raised this temperature to 60C (140F)
and full key down operation
caused a stabilized
temperature of 80C (176F). All temperatures were
measured on the heat sink at the final device with
output power maintained at 15 watts. One can safely
assume that a panel on the outside edge (i. e., backside) of a transceiver could be successfully used as a
heat sink for this amplifier,

Circuit Description
The amplifier has two stages and uses 5.0 W and
15 W rated transistors to accomplish
the desired
gain and power output. Two stage transmission line
Chebyshev
networks
accomplish
coupling
and
impedance transformation
at tbe input and output.
Nominal impedance levels are 50 ohms, while the
interstate network transforms device impedances
directly. Values for the reactive elements of these
networks
were almost
entirely
generated
by
computer aided design. Although the interstate
network is straight forward in design, it required
some modification
and refinement
of computer
generated values to achieve the final results and
accommodate available component values.
Construction
The amplifier is assembled on double-sided G-IO
fiberglass
board with 1 oz. copper cladding. The
format is 2.0 x 3.5 and a photomask is provided
(Figure 13). Some method of electrically connecting
the upper and lower ground plane is required. Eyelets or plated through holes are recommended, but
alternative measures such as short pieces of wire
soldered to both planes can be used successfully.
Failure to provide an adequate or consistent ground
plane may result in poor RF performance, instability,
and unpredictable tuning. The reverse side of the
board retains all copper and forms the ground plane.
Component
placement
andtherecommended
position of grounding eyelets is shown in Figures 13,
5, and 7. All component leads are positioned and
soldered above the board. There are no through
connections
other than grounding
points. This
facilitates
component
positioning,
replacement,
and accessibility.
The transistors are fitted into a
0.4 by 0.65 opening in the board and are installed
directly against the heat sink. A coating of heat
sink compound such as Dow Corning 340 between
each device and the heat sink improves thermal
contact and helps prevent power slump.

446

cl 2

RF
Outpul

L4
z
u

C1O
L

L5

35, ca 43 pF
C612PF
C7, C9 50 pF

L1 -L5 Pr!nted Inductor

L3 1 25,#18AWG,
l-1/2
Q1 MRF260

FIGURE

5 Component

Trs,

9/64i

Placement.

448

136-160

RFC4 O 15 pH Molded Choke

RFC5, RFC6 vK200-4a
8 Bead, Ferroxcube 56-590 -65/3B

MHz Amplifier

RFC6

RFC5

F~

L1

RF

L4

output

C8

L5
L3

C7

T
5/8- Hairotn
LmP,#liAWG

L1 -L5

Printed

L3 5/8

MRF260

Q2

MRF262

RFC1,

RFC2

through

7 Component

Placement,

449

Inductor
AWG

Bead

O 15 #H

RFC4

10 PH Molded

160-174

W,re

2 Turns
Ferr)le

RFC3
RFC5,

FIGURE

=18

Q1

Molded

RFC6

VK200-4B

MHz

Amplifier

formed

#26

Enameled

Ferroxcbe
Choke
Choke

,nlo

ha(rp(n

loop

Wire

56-590

-65;3B

FIGURE

8 Power
136-160

Output

versus

Frequency,

MHz Amplifier

18
Z
E
~

16
14

s- 12
:

10

: 60
>
0. 6.o
.
L- 4.0
2.0
136

140

144
152
148
f, FREOUENCY
(MHz)

I 56

160

>i~

FIGURE 10 Power Gain and Input VSWR

versus Freaencv.
136-160
MHz Amolifier

versus

450

Freaencv,

160-174

MHz A%nlI#Ier

EB93

Desi@

Considerations

turns (closewound) on a temporary 1/8 inch form and

then separating the individual tnrns by 0.020 inch. An
Xacto number 11 knife blade was used for this punose
and provides the correct turns spacing. The 10&ohm
isolation resistors, R1 and R2, must be noninductive
and carbon composition resistors proved to be entirely
adequate. In a properly tuned and balanced amplifier
these resistors should remain fairly cool to the touch
during normal operation. Each ampfifier and coupler
...
.,
.
.
input and output port lS designed to be termmated mto.~~~,~$
50-ohms to facilitate testing into a 500hm system~~. >$
A PCB bridge (Figures 3 and 9) is used to ca~~~l~~
of the dc feed circuitry. It acts as a continuatiq~o%:@~
mound plane and enhances circuit stabili&:&gtfd
copper (0.027 inch) and double-sided
ciiiwit~bsard
were used as a construction medium an&$~_
in performance was noted with eithefl~a~rral.
Initial alignment is accomplip~d
~y?driving the
amplifier with a 5 watt CW soti$~,>~t approximately
160 MHz. The applied volt~?a~tat
12.5 volts and
the variable capacitors, G$$$~~,$$5, are adjusted in an
alternating manner to pFQ*<#&aximum output power.
Full drive (15 watts) is-~~~~pplied
and the capacitor
adjustments are r~~ted~At
this point, the circuitry
should be deliy&~~*B& watts or more to the 50-ohm
load with the$5~hkts
input. After the final adjustments arem:~~a *#e isolation resistOr temperature in
either cou$~ should be relatively cool to the touch and
the i~~t V*R
should beat a minimum. Beat results
wi~~:]e~tained if the transistors are beta matched
(#O%~pnor to installing them in the circuit.

Experimental work with 90 (quadrature) couplsrs

proved unsuitable for this application. Generally, they
are sensitive to mismatch and tend to create instability
and loss of power when used in an amplifier. In-phase
(Wilkinson) couplers provide an adequate solution to
this problem. (Ret I) They are relatively insensitive
to phase changes
and offer good bandwidth
characteristic CS.
Printed transmission
lines for the frequency
of
interest can become somewhat
cumbersome
on
standard circuit board material. Therefore, lumped
reactance (Ll, 2, 9, 10 and Cl, 2,3, 14, 15, 16, Figure5)
are used to simulate 70.7 ohm I/4 wave transmission
lines, the main element in the couplers. This approach
not only conserves board space, but protides a means
to compensate
for small variations
in associated
component values.
Microstrip
techniques
are incorporated
in the
amplifier networks to balance RF performance and
promote reproducibility.
Because of the lower circulating cuments and reduced component heating in
the collector circuitry of low-powered stages, smaller
capacitors can be used in the networks at that point
than would be required for a single-ended 60 watt
design. Separating the major heat producing devices
to two areas on the heatsink produces a more even
heat transfer to the ambient air. The combined amplifier presented here has good harmonic suppression
(Figure 8). A low-pass filtering effect is noticeable with
the Wilkinson combiners.
Construction and Alignment
A 1:1 photo mask of the circuit is provided in Figure 9,
and doubl~sided G-10 fiberglass board with two-oun@~
,. .s.:..
copper cladding is recommended for construction. ~.~e *j.
mound points are indicated on the PCB phot:,m~~,,
The inductors required for the splitteri$~~.bl~r
are constructed by winding the appropriat~~.%:@~
Of

.,,

This amplifier has been extensively tested for ruggedness and reproducibility. The 15 watt input level
makes it compatible with the EB-90 two-stage VHF
amplifier as a driver. Together they form a chain
requiring 200 mW of input power for a 60 watt or
more output.

.$

References
1. Lawrence R. Laveller; TWO Phased Transistors
Shortchange Class C Amps, Microwaves, Pg. 4854, February, 1978.

2. Ernest J. Wilkinson; An N-Way

Divider,
PGM TT Transactions,
January, 1960.
452

Hybrid Power
pg. 116-118,

453

455

456

_..

.-

..

EBI07
Mounting Considerations for
Motorola RF Power Modules

,.

~A;,,,
;Jj+:.p,s,:$%

,$t~~ ~ii.b
... ,$~\
~,
$.;.>
$:.. . ,:...
~;~,. >,.,
$>.*
::,,

;J,;l

,),,~.
,$ ,g

~::j

,,;,

.>.

INTRODUCTION

..

,:+>

~J*
,.,,
~:*>\:y
,/.
i
:4Y<: *,.
.~:..,&.~

neerlng Laboratory. GEGM~,a selected to do this work

The packa~ng used for standard Motorola RF Power
because they have dQ~,~{*sive
work in the area of
mdules consists of a copper flange on which the sublaminate stresses a@4av@hvailable
several proven comstrates are soldered and a non-conductive cover which
puter pro~ams wh~$~~$ly directly to this problem. The
is either of a snap-on or epoxy attichd
desi~
The
assi~ed task w~~~bvide
an estimate of the maximum
ceramic substrates are either 96% alumina (Al ~03),
amount of i~~{ial ~w (curvature) in the mounting flange
99.5~c alumina or 99% Beryllium oxide (BeO ). These subwhich wa~$~~l
subsequently cause the ceramic substrate .~.~~~uie
in the final assembled state. For the
strates are attached to the copper flange using either
lead-tin or indlum based soft solders. Typical liquidus
res~lt~of *1s analysis, see Table 1.
,~,,,,,!>
wmpratures of these soldem are in the 149C to 16YC
$.$~
range.
~~M@NTING
CONSIDERATIONS
The pu~ose of this papr is to present the mechanical
F*$, The theoretical analysis shows that some of the refactirs which should be considered in mounting these ,,,F+:. s~onsibility for proper mounting rests on the user. Proper
mdules in equipment.
..~!{%~$,~i
consideration should be given to the following items:
.,
,t.!<,
S>.*
..
1. Flatness of the mounting area must be such that
~~~:?,.
.-
..,
~?.,>
WJOR
MOUNTING
FACTORS
.:,..
the final mounting of the module will not bend the flange
There are three major considerations in mo:ntin~ an
beyond the llmits given in Table 1.
RF power module. First, the flange is used,f& the RF
2. Attention must be given to surface finish and cleanelectrical Wound reference. Typical indum~$,n}b~of the
liness of the mounting surface. For instance, if one
connection pins used on these modules ~s.~*1.18
nano mounts the module with thermal compound and uses a
henries per inch or 1,8 nanohenries #r.$i@.,mils. Since
dirty work area which allows 3 to 5 mil particles to be
at 800 MHz a nanohenry has abouk.~ti;~~~s reactance.
present in the compound, a failure mode can be produced.
.*:> *:
it lS easy w see that it would @
#$&st impossible to
3. Another consideration :s the movement of mater]al
. .. ..,:.?,,
achieve a low reactance ~oun~t~~~flgh the use of Pins
around tapped or punched holes. A tapped or punched
alone. %cond, the coppe~afl~g$$prov] des the thermal
hole which leaves a burr on the mounting surface can
path for the removal of&e @&produced
In the active
lead to failure modes.
devices present in the,+}~hus,
proper thermal han4. In addition, rigidity of the mounting surface and its
dling must be consi~pre$ in mounting the module. Fimaterial should be considered, For instance, the copper
nally, we must @*~W?he
mechanical stresses placed
flange on an aluminum heatsink will result in a bion the mdule &$i%t&mount]ng techniques used. Here
metallic system which can create a bending problem.
we cons~d:.~~t~essek placed on the leads and bnding or
Consideration of the direction ofrlbs in a heatsink should
twlst]ng ~$~ei~ountlng
flange which would cause cebe made to maximize stiffness in the direction of bending
ramic &a*r*.
or adequate thickness of the heatsink must be provided
f~.:+:::,+?
:y.,,i,
to control bending.
t, ...
It lS not deslrabie to mechanically constrain the ends
NQ~&E
FLANGE
FLATNESS
? &qT]ng the processing of the module, consideration has
of the module so that no slip IS possible between the
,,~~$
~,,to:~ ~ven to the various stresses produced Through
module flange and lts mounting surface If the ends are
~,: ,+~.,:,~,~fialysls
of these stresses and the materials used we can
constrained and the temperature differential between
!?*,*.
amlve at the maximum allowable flange bending which
the module and the heatsink is significant, there can be
J!:\+.*.\
.,, si
,? ,.4xcan be tolerated from a mechanical standpalnt. In deenough bending of the module flange to break the ce.~$:
temlning the allowable flange flatness conditions, both
ramic. An example calculation is shown below to demanalytical and empirical analyses were performed.
onstrate this problem.
AWeement between both of these analyses was very
Assume that the ends of the flange are constrained at
good. The theoretical analysls was pefiormed by Motothe centerline of the mounting holes. (2.4 inches for
rola Government Electronics Group, Mechanical EngiMHW612MMHW710/MHW720
series modules). Assume

457

.=

.-

that the module is mounted on a machined aluminum

heatsink.
Thermal expansion coefficients in #incMincNC
Aluminum 25 x 10-6
COppar 17 x 10-6
L = 2.4 inches
For a reasonable approximation assume the thermally
induced bending creates an isosceles triangle as shown
in Fi~re 1.
FIGURE 1

Assume that the module flange changes temperature

from 25C to 50C and the heatsink changes temperature
from 25C to 3VC in the same time (obviously the heat
input to the system comes from the copper flange more
on this later).
Heataink A L (aluminum) = 2.4 x 5C x 25 x 10-6
= 0.0003
Flange A L (copper) = 2.4x 25C x 17 x 10-6
= 0.00102
So length ABC = 2.40102, AB = 1.20051
length AC = 2.4003, AD = 1.20015
And AB2 = ADZ Y BDZ
BD = VAB2 AD2
.$

So BD = 0.029397 inches which far exceeds the s,

..:$,>.l,,}}$
allowable flange bend.
.* ,,,j,\ .
., ;. ?:,
This analysis also points out the advantagq~:~~~e~~ng
the heatsink and the flange at lowest po~$:h$~.smPerature differential through the use of th~i~$~~S3>onducting compounds between the surfaces. .~l,+;$?
For instance, in the example giv~#a~bN@?with an aluminudcopper system, the copper @n$k will remain in
tension at any temperature ~So*~$%e temperature at
which the system was co:t~~f~$~$$!as 10ng aS the temperature ratio between tw, ~~~tsink and flange is kept
less than the ratio of k~e t~~ptial expansion coefficients
or 25/17, Incident al$~<$fii,$assumes that the heat input
source to the system..ow%ates in the copper flange. This
situatiOn pOi?~~~M$:$he folly in some types of temperature cyclin< t~ti~. ,For instance, if the aluminum/copper syst:~l~$~%ra]ned
at 25C and is uniformly heated
to say S2&$>the copper remains in tension if the
sys@~ ?&...&oled below 25C, tbe copper wi 11 go into
coti~$kion.
This is exactly the opposite situation ob$~~~ when the heat input tothe system comes from the
*p&r flange.
The above is a rather elementary analysis of the thermal effects on the module! heatsink system. Many other
factors are involved such as relative strengths of the
materials involved, bending of the mounting screws and
so forth.

What should be derived from this discussion is that

the desi~ of the mounting for the module/heatiink system is not a simple one and should not be done in a casual
manner.
Our recommendation is that a mock version of the
system be constructed early in the equipment design and
thermal cycling performed both with external heat input
to the system and with heat input to the system from the
module. This is a very effective analog computer and
direct measurements of the flange~eatsink deflections .,~,~.,,
can be made. In this manner the actual expected flan$~+tw
excursions can be compared to the recommended m~.!-,s?k~
mum flange bending to determine whether the deai~k~~~
adequate. Incidentally, the recommended maxi@#~~&flection values given in Table 1 have a saf$w .~~&~%of
approximately 2. That is, the deflection re~wi~@ @ crack
the ceramic is approximately twice ~~~%$given.
Table 1 includes data showing the eWp]R$@ deflections
required to fracture a ceramic boW~;~ the module.
5. We strongly recommend t~~$~${& a good thermal
compound between the mouns~. sti~ace. Sufficient material must be used to fill al$N~ @hich may be present.
We have not been able tdw<~
any mechanical problem
with excess compouq.~as f~g as there is a path for the
excess material to &&:@s
the module is tightened down
with the moun,~~~~~~ews. At this point it should be
pointed out ~at,.u:,&ss both the module flange and the
heatsink w~~ap~ed
to absolute gauge block flatness,
there wJ$4al&$ys be a significant air gap between areas
of the$~~$ge and the heatsink. Since it is obviously not
pr~~l~l td achieve a lapped surface of this quality, this
~o~o+mf the mounting problem resolves to one of me<$@~al
rather than thermal considerations, As an
t~slde, some of the Motorola modules also have machined
~\ surfaces which may be oxidized to some de~ee. Infrared
thern]o~aphy of the active die was performed to see if
there was any thermal deWadation due to this oxide
layer and no de~adation could be found. This has also
been found true on lapped discrete transistor flange
mount parts.
Several manufacturers of thermally conductive heatsink compound exist. We have used products from Wakefield and Dow Corning with success.
MOUNTING
HARDWARE
Obviously an ideal mounting hardware scheme would
be one in which the clamping pressure remained constant
with age. One way of achieving this is through the use
of conical washers one trade name is Belleville washers. Another possibility is wavy washers. Proper selection of mounting hardware and torque is also necessary. We recommend the following mounting hardware
sizes and torques:
4-40
3 iwlb
632 5 irr/lb
8-32
5 itilb

TIGHTENING

SEQUENCE

A very important factor h be considered in mounting

the module is the proper torquing sequence, The personnel involved in mounting the modules should be given
careful instruction and their procedures monitored at
regrdar intervals. Since the flanges are punched from a

458

-..

--

roll of material, there can sometimes be a small <roll-up

at the end of the mounting flange. If one considers what
can happen if the mounting hardware were tightened
completely at one end first, it is easy to see that the other
end could be lifted off the mounting surface well in
excess of the allowable flange bending tolerance.
This should be avoided by first lightly alternately
snubbing down the mounting hardware finger-tight,
Next, the hardware can be torqued to its final specification again in at least two sequential steps,

on the leads, even as the fixture wears. Motorolasspecification for lead pull in shear and peel are 908 gm shear
and 454 gm peel for BeO boards and 1500 gm shear and
750 gm peel for alumina boards. Modules from PC86, 90,
and 91 product lines use BeO boards. Modules from the
PC87, PC1031ine useonealumina
andone BeO board.
PC41, PC64, and PC104 use alumina boards,

DEFLUXING
Thesemodules are designed to be manually s~:red
into an assembly, The modules have a silicon&;\p.*at
over the active die, MOS capacitors, and,@f$@@@e resistors, The die coat used will not withst~$$:~~ normal
flu removal fluids and severe reiiabi~~.p~~~ems could
be incurred if the flux removal fl,@&~~-solder fluxes
penetrate the inside of the modul~f~~ ~~$bmmend a flux
activity of no more than R ort~w~used.

THE IMPORTANCE OF THIS TORQUING

SEQUENCE CANNOT BE STRESSED
TOO HIGHLY
LEADS
The leads used on the stmdard Motorola RF Power
Modules are of either tinned copper, gold or silver plawd
KOVAR, or pure silver strap, typically 5 to 10 roils thick
and 15 to 20 roils wide. The leads are intended for making
electrical connections to the modules only and are not
intended to support the module at any time in the assembly process, Consideration should be given to the
stresses which may occur during mounting or testing.
Poorly desi~ed test fixtures can create lead stresses far
above those encountered intheend-use
equipment. It is
recommended that the fixture be deeigned so the leads
are always clamped after the flange is clamped and the
tolerances be such that an upward force is never placed

~<,:<,,,
,+:.::
CONCLUSION

,;.....~}.
,,,

MAXIMUM
RECOMMENDED
DEFLECTION COMBINED
HEATSINK & FLANGE
CONVEX
CONCAVE

,~

..%

constr.ct!o.

of PCM

o.rd-BsOc

.rriert

with total BeO output

ranskstorc onstr.ct!ons

board.

459

OUTGOING

OA SPEC. (MAX)

CONVEX

CONCAVE

0.008
0,008
0.007

0.010
0.010
0.0085

0.005
0.005
0.003

0.005

0.008
0,0015

0.010
0.002
0.0085
0.002
0.004
0,005

0.005
0.001
0.003
0,001

0.007
0.0015
0.0035
0.004

'~~&.~-.ha.9edt
Oal.minsb
<~&#W$has
this construction

In mounting RFp:~)~~+ules,
the following major
areas should be co~dew
.~,~,
.:,.. .
1. Heatsink fla~~gf:~
2. usether%kx}ound
eliminate dirt or~itin
the co~poi~
or on mounting surfaces, use an adequ4~$.~$~,Ountto fill gaps.
3. ~W,~,
modules down in an alternate manner
?$n$$r.tight before final torquing,
:#:+.W&~hreful with defluxing operations.
.+..: ~~JConsider lead stresws, both in mounting and testing.

imilartoPC4l

in February,

0.0015
0.002

1983

411 product

0.005
0.005
0,005
0.002
0,005
0,002
0,0025
0.003

with dale code .883 and

460

EB411

A Digital Video Prototyping System

,,,.

By Aldo Giardine B.Eng (Hens) AMIEE

Consumer Seament
Motorola lnc.,-Semiconductor Products Sector
Geneva
1. INTRODWION
$::1,.,<,k,
!

.\,\

~.$$k,.

. \\:\:

This Engineering Bulletin describes a Digital Video

The focus here i~~~~ %~e functionality of the
Prototyping System (DVPS) that has been developed
combination of ?tw@ove
components and
using Motorolas latest multimedia devices, together
development s@*;~Reference
Section lists&t&
with a PC-basedFieldProgrammableGateArray (FPGA)
sheets an~~~se%manuals containing detailed
descript~@,~$$fi& information on their use.
development system. It is designed to provide a fast
~:?+.
.:!
and effective means of prototypinganddemonstrating
Th~~,~k$as been successfully used to implement
digitalvideo processingfunctions.Afunction developed
~~o W,Y. subsystems, namely, a Picture-imPicture
in this way may later be fully integrated as an ASIC
L%:ocessorand a 4:3t&l 6:9 Picture Processor.Those
device for use in a consumer en~roduct.
:w:]suhsystems are described briefly below.
,.:
s\:\,.
,$~.:(>
,,~~~

2. MOT~O~

,:,::,

.)

DEWCES USED

.tk.

The DVPS takes advantage of s~v~l ~ersatile

multime~a devices, that are liste@~%~. They are
,,,;.,..,
used as a means of generat~$~~,~l~ktaldata from
virtually any analogue video %.~~~yand providing a
means of displaying th~ k,~~~Rg analogue video

b) MC44250(4). This triple Wt Analogu&t&Digital

Converter provides black4evel clamping for either
RGBorYUVsignals. These aretypicallya,c. coupled
into the device from the M C4401 1 which provides
the appropriate clamping pulse, but may equally
come from any other suitable video source.

signals on a consumer.~,,~$~,#, after the digital signal

processing function ~n~prototyped.
\ t..~$
.:,s
,,:}
..
a) MC4401 11,T* is~e multimedia derivative of the
MC44001$~~X[&~@erforms the function of a
Multis$@n~#
(PAL/S ECAM/NTSC)
Chroma
Dec~~
witfi a selection between RGB or YUV
ou~~$~Wals. The M~4011 also generates a T.V.
. m~edclockfordigital
sampfingandsubsequent
:,~~~essing of the output signals.The Iatterfunction
.?
t* $S also available separately in the form of the
*:,
+~.,
.+O$ MC441 45(3]. The output stages of the MC44011
t
*<,:
~,: ....>.,....
are designed to drive the inputs of the MC44250
:..:$;
: ..
t:/\\
.\..,.,s:.
directly.
,:,,,?.,
.,
*,

c) MC44200(5). This is the counterpart to the

MC44250, a triple 8-bit Digital-to-Analogue
Converter for RGB or YUV. It features differential
current source outputs *signed to drive 75Q
loads with 0,7Vpp.
Other devices used include the MC68HC05B6(6)
(8-bit MCU with onboard EEPROM), the MC14576(7)
(Dual Video OpAmp) and some standard CMOS
logic.

461

3. FPGA DEVELOPMENT SYSTEM

The digital processing elementef the DVP$ consists
of one or more FPGA devkes. Th&e comprise of user
Configurable Logic Blocks (CLBS) and 1/0 Blocks
(IOBS)that, together with programmable interconnect,
allow most memory control and simple digital video
processing circuits to be implwented. successfully.
The configuration data is stored in in15rnal RAM. The
reprogrammable nature of FPGAs makes debugging
and development a relatively straightforward process.
The logic mpacity of the FPGA devices ranges from
1,200 up to 20,000 equivalent gates, with between 58
and 240 user-programmable 1/0s, which is ample for
most applications. Their toggle frequency ranges
between 50 and 125MHz, and the devices come in a
range of package types.
The front-end to
Schematic Capture
Library & Interface
Schematic
files

the development
system is a
Package{B),together with the FPGA
runn;ng on a Personal Computer,
are processed
by the FPGA

Development System(g) to produce a graphical file

representing the configuration of the FPGA This file
may be manually edited for routing optimisation before
the final binary file is generated. Programming of the
FPGA devices maybe carried out in one of two wavs:

$
.$\,.\..,..
?:{.<.,
.,...
a) The binary file mav be directly downloaded from the \~:v *\,
>+,,
host computer serial port to a powered device in a S, $
.,>.:1.
matter of seconds. This is the most approPriatetLQr$,.,:*
the debugging and development stage, as i~l.t~~.
.,,>+.,
circuit design changes intoa quick andeas~#,~6~Ss
of device reconfiguration. It may be a~;3~pt& as
making an alteration to the schema~~ d~~$m and
recompiling the design. As long ~Jt~ device Pinout is unaltered, no rewiring i~~$shary,
b) When a design has mature~~~~.~further
changes
are expected, the bina~w,~a~
be programmed
into a serial or paral~.~kPRw or EPROM. This is
addressed bv the ~~k+. device itself to perform
automatic self~n~$~?ation of its RAM as part of
the power-4~$~@kce.

.,,

,:

out~p*s.
The PC download mble connection is
Figure 1 is a block diagram of the DVPS environment.
*.,,,..
\,:,:~
m~wp%rectly to the digital card(s) for configuration of
The rack connects together the input cerd, the digital
t,Qe$>GA(s). A Video source is connected to the front
card(s) and the output card through a backplane. The
external controller board also connects to the backplan,~~,, of the input mrd; the outputs for connection to the final
to pedorm initialiwtion and control of the input @ %!, display are taken from the front of the output card.
~:.>
.,~*,
.,\.}.~~~i,.

made to the appropriate device datasheet

detail on application circuit diagrams

The following three sections describe each card and

its functions in more detail. Reference should be

for more

5. INPUT CARD

u~n

;OR
rLL

H
wu

Vli

B.@TE

,.?

The input card accepts various ~pes a~~~~?~ signal

sources from which it generates threwmwide
data
streams A T.V. Iinelocked clock#~.sp;ko 42 MHz is
also generated on this mrdforu$$~w~tal
processing
of the data. Refer to figure ~$~~;~$%gram of the card.
.,.,. .:.,\ .
Four BNCconnectors~$We$W~t of the card constitute
the inputs. The fir~t~~~~~s composite video of any
standard, or a co@os+&e sync signal accompanied by
either RGB o[$~~~~nals
on the other three inputs.
The desir@ inphp~onfiguration is selectable through

determined by the division ratio set in the FPGA and

isalwaysan integral multiple of the T,V, line frequency,
Normally this would be chosen to be 27MHz, so that
the video signals are sampled at 13.5MHz, as
recommended by CCIR Rec. 601 (lO).
The three signals from the MC44011 are a.c. coupled
to the MC44250 inputs for black level clamping to the
appropriate levels before conversion. YUV or RGB
clamping modes are selectable through a jumper
setting. The RGB-mode clamps the back-porch of the
signals to the bottom of the ADC input ranges, while
the YUV-mode clamps the U and V signals to the
middle of the ranges, leaving the Y clamped to the
bottom of its ADC range. A burst-gate pulse is
generated bytheMC44011 to activate the d.c. clamps
in the MC44250 at the correct time.

iumPer,:~~Kl,$gs on the card. These signals are

proc~;,~$~y
the MC44011 to perform chroma
d~~~~~:and RGB matrixing where necessary.
~::,,,.~
~t>.,
4~@ T.V. Iin&sync pulse from the signal source acts as
:,<~$$+,g reference for the Iineiocked PLL that synthesises
~,?;<
:J*JFthe clock on-board the MC44011, After suitable
t.,\ +$,
,$*..
::,,.,,
buffering, the clock is output from this card for division
?(
,,,.
k.,
,..
down to line frequency by a counter in the FPGA on
.:
the digital card. AT.V. Iin&rate signal is returned from
that card to the phase/frequency comparator to
complete the loop in the MC4401 1. The exact
frequency of the synthesised clock is, therefore,

The three 8-bh data streams resulting from the

conversion are registered and buffered before bel ng
output to the digital card via the backplane.
Further details and circuit diagrams are given in an
application note on video capture(l l).
463

6. OUTPUT MRD

4.<,>

,,\.$,::x\
.* .:,

The output card, illustrated in figure 3, receives three

8-bit data streams from the digital card. Its purpose is
to convert these into the analogue domain for display
on a consumer T.V. or RGB monitor. A Iin&lockgd
clock is also available from this card for applica@~:
where the~ptureanddisplay
clock rates arep~~{~~ll~
~it: .>i., ~.*
different and/or unrelated.
*.<\\.. -~,,.
.>.:~.l?,
~,,.,,{.
Access to the output signals is maXC{~:%~ERITEL

..!l
T~~cl~k

$~~,.
synthesis PLL on this card uses the same

.l?@,@#~eas on the input card, usinga separate counter

~,fhe FPGA on the digital card, if the division ratio or
; Iinefrequency reference is different from that used by
the former.
Data from the backplane is first registered by tri-state
buffers before being input into the MC44200 for
conversion, In the YUV-mode, the m.s. b.s of the U
and V data lines are pulle&up by resistors while all
other bits are pulle~own at the inputs of the converter,
This ensures that the video of any standard, or a
composite sync signal accompanied by either RGB or
YUV signals on the analogue outputs from the DACS,
are correctly set for black-level when the input buffers
are set in tri-state mode. The MC4401 1 may then
clamp these d.c. levels from the MC44200 and hence
provide the right levels for the receiver to display true
black,

(sCART) Connection(l 2) provi~i~j@.@B

signals,
together with a fast+ommuta~~ &t$,wtching signal.
The MC4401 1 device on this ~<d @ayalso be used to
perform the matrixing ~s~,~signals
to RGB, if
needed, ConfiguratioN,@i.~&~ard for the signal types
being processed is s&~~$@
Jumpers, as on the input
card, When co~,~~ted to a T.V. set, the composite
video signal co~l~ft~m
the receiver via the PERITEL
connectiom&~~>e used as a reference to which the
*., it.
display<~,~t:Wnthesised by the local MC44011 may
be l~&~&&This video signal is also output onto a BNC
.CQ~~QCtbr
/.\.,,t ,&M.]
.. for use as a source for the input card, if the
@*@tion
requires it.

The RGB signals from the MC44011 are buffered to

drive the PERITEL socket with 0,7VPP at 75Q, using
MC1 4576 Dual Wdeo OpAmps,

464

....-

--

7. DIGITAL CARD

PC mN.

.>,

*J..
~+.~
Q ,,,. ~+:.

.,?t,

~:t~! ~, . <~,,
..*lr
tg.}t,

.$:>..

Invariably, the processing on the digital card involves

video data storage that requires memory external to
the FPGA. Therefore, FIFO or Dual Port RAM devices
would normally also be included on this card, close to
the FPGAdevice. Figure 4depictsatypical configuration
for this card,

At the centre of the DVPS is~~~afcard.

There may
be one or several of these, ~~~~n
the input card and
the output card, depe@@$@ the complexity of the
digital processing re@lrr
by the application being
implemented, T~.~;~,G~ currently employed in the
DVPS is the ,@1K@S042PC8&l
00 device. This is a
4,200 equiv%~$%ate FPGA with a toggle rate of
10OMH~$i@an 84-pin PLCC package,
.,.,~::,~,,.
?~ ,.:.
t)it>
$:,l:;
In j~~<wplest form the digital mrd consists of the
~.~%,device and the means to load its configuration
,,k~~.
This entails provision for a parallel or serial
,h }, ;?~OM anWor a connection for the PCserial download
,.y.,. ..\~,>.:~
*%*!$~
* cable.

As the function of the 1/0 pins is programmable and

application dependent, connections to and from the
FPGA pins are made using wirewrap, once the pinout
has been defined.

#b,
*
~$
.,.4
~:::-:
t....;.
~,

465

8. RACK AND BAWPLANE

Here, two external 5V power supply units are used for
the digital and analogue sections respectively. Al
power and ground fines are kept separate on the
backplaneand on the inputandoutput cards. One starpoint connection is made between the groundsat the .t~$~~~.
power supplies to avoid loops between the cards iy*.~
.,+.
*i<.
.,-
the rack.
:::i!..
,> .<,,q.>
.::,~<.i.
\ ,,,;v+.,j..
,.$}>
,+,~,\,>.,7,*..
9. CONTROUER BOARD
.!.$,i.d

,,$tl
~,,?,<
.,;,,,,,.+
necessary because the two identi@[ ~&44011
The two MC44011 devices on the input and output
devices, bvdefinition, have the sarnf,$~C-~&address,
cards must be initialised and controlled via software
so that although they may shar~)$~$we
clock, two
using the two wire 12C-BUSProtocol(l3]. For this
separate data lines are need@~&o*intain seParate
purpose, an MCU Controller Board(14)built around
control using the 12C-Bus.$/$$&ol
from a single
the MC68HC05B6 &bit microcontroller, is used with
... ,:,>
.~a$y,
controller board.
the DVPS.
~>i,,,
*!.ii

A 3U-high rack houses the DVPS, providing the

mechanical structure for, and the interconnections
between, the input card, the digital card(s) and the
output card. All the cards described earlier are
constructed in the form of extendedaurocards
(220mm x 10Omm) and use 64-way edge connectors
for plugging into the wir~wrapped DVPS backplane.

The controller bo~$f~~~iso be used to manage the

vertical picture $~rn~~ control, as the processing
requiremen~~$$~~$~ne rate lie within the limits of the
4MHz M,@U. k.
using three wires to the DVPS backplane. This is
,,>..31.
......
~:,
Y.i.,:
,, *...
10. SYSTEM lMPLEMENT~~N:EMMPLES
,,. .>..>
ta$~

This isa stan&alone board with a builtin keyboard and

an Wigit 7%egment displav. It operates in a pseudo
12c-Bsmode of communication with the two devices,

:?,Y:S$*

As mentioned in the introduction, the DVPS has

proved to be useful as a flexible proto~ping piatfo$~
for digitalvideo processingfunctions. This sectio~~w:l
briefly describe WO projects in which the DV&~~s
successfully emploved to implement fu~~~e~ for
demonstration to equipment manufact@,$F~
\:..!s:5<:1
\ \J.:,
.~,:
,,

a) Pidure-in-Pidure

Pr_k,

,,*,+ ~%<%
. i?
6,

Tk second digital card contains dual port vide&RAM

rwhich stores the data received from the first digital
card. The FPGA on this card generates all the control
signals neceswry to access the memory, as well as
performing the divider function for the PLLs on the
inputand output cards. Its function is to write the data
into the memory at 4. 5Msample#s usinga clock that
is iocked to the inserted (PIP) channel, and reading it

Three digital cards were us,@,w#er

with the input
card and the output card @te&late a multistandard
PIP function.The YUV,@~Y&#coming from the input
card are standar& inmndent
by virtue of the

..
The first d,@e~~&&dcontainsan FPGA to perform data
reducti~,@,$@ltiplexing the U and V ~mples as a
mea@%f~b&mpling these channelsand so reducing
th&,@,&andwidth bv 3%. This takes advantage of
X&that
the colour~fference signalseach occupy
~w
half the bandwidth of the luminance signal. The
}WO resulting b~+wide data streams W, UM are then
decimated bya factor of nine, usinga twdimensional
median-filter. The output data represents the inserted
picture but at one third the original height and width.

out again at 13.5 Mwmples/s using a clock that is

locked to the background (MAIN) channel. These two
channels may, of course, be asynchronous to each
other, hence the requirement of the two separate PLL
clock sources.
Data read out of the memory is output to the third
digitalcard. Here, the FPGA demultiplexes the colourdifference samples and interpolates them to
reconstitute three byt~wide data streams together
with the luminance samples. The YUV data is then
converted to analogue signalsand matrixed to RGB bv
the output card.
Here, the MCU controller board was successfully
used to perform control of the twoM~4011
devices,
while also providing vertical timing control and row
addressing for the video memory accesses.

466

b) 43-te169

used are therefore derived from the PLLs fundamental

clock frequency of 40.5MHz.

Hdum Prmer

A single digital card between the input card and the

output card was sufficient to implement a processor
to correctly displaya 4:3 aspect-ratio pictureona169
aspect+atio T.V. tube. Here too, the processor is
multistandard by virtue of the MC44011 on the input
card.
The geometric correction is achieved by writing video
data into FIFO memories at 10. 125MHz and raadng
out the data at 13.5MHz. As the r=d and write clocks
are essentially related and both locked to the video
derived from the T.V. receiver, only one PLL divider
needs to be implemented in the FPGA. All the clocks

Using the same principleas in the previous example,

thetwocolourd!fference data streams are multiplexed
into one by the FPGA before being written into the
memories along with the luminance samplest?~~he
MC441 40(15, PAL and SECAM delay-line,,,f-s
are also performed by this prototype usi@ a~@~her
FIFO memory, hence replacingthe for,~~Mce
in a
system using the MC44001. T@~&}$\:]end, the
multiplexed colourdifference sa@~~s;*?e processed
acrosscon%cutive pairsof kn~:;~*
demultiplexing
into separate data stream~,a@~,@nd being output for
conversion and matrixi~hy the output card.
,,.x$~$~,,
,,ti.~,,,*$*,
>~..
,,.
.,. .!>

.,.
Copies of the Motorola datasheets and application note listed belq$!%o be obtained from Motorola product
,$s.?~?ri,iyit
Marketing, 31023 Toulouse Cedex, France.
(1)

MC44011 Chroma Processor& Pixel Clock Generatorx: ,-~C44011

%,

Advance Information

(2)

MC44001 Chroma And Deflection Processor

(3)

MC44145 Pixel Clock Generator

Product Preview
Product Preview

(4)

MC44250 Ttiple Video ADC

(5)

MC44200 Triple Vdeo DAC

(6)

MC68HC05B6 &bit Microcontroller

u)

MCI 4576 Dual Video O&emp

(8)
(9)

~.:$

~ ?f. ,,<!{:),.

~* !>.
.<:.x.~
.$>
~ ..!:<
~$$,$k,
*
...*,:,*!.
.>.N,
,:+>,
~{:;<!
\,.i:*
.\\/+a
-

M~4000
M~4145

M C44250D Data Sheet

M~4200

Product Preview

MC68HC05B6D

Data Sheet

- MC14576 Advance Information

- User Manual

OrCAD/SDT Schematic Capture!, $~

.:*.y,yt .,
FPGA Developmen~~~,P

User Manual

XILINX

(1O) CCIR Recommendationi$~ *2,*

$
(11) Video Capture Applic~WDf
the MC4401 O & MW250
,*$. *,,
~?..
(12) Peritel Connectio~ + ~~

- Specification of Standard

(13) Philips 12C-bu~~~*sl

- Specification of StanMrd

- Application Note
- Specification of Standard

(14) MCU Con~@?Wrd

- Ma4cTRBDol

(15) MC441.@lWl
,,.,:.,
%It,~f:
....
.,,

- MC44140 Advance Information

DekY4ine

467

468

lnforma$f6n

469

470

Additional information relevant to Radio, RF and Vjdeo applications

may be found in the following Motorola documents, available through
your Franchised Distributor by quoting the appropriate reference.
ANlo51/D

Transmission

BR3471D

Bipolar Logic Circuits Quality&

BR470fD

Motorola Discretes The Complete Solution (Rev. 1)

Line Effects in PCB Applications

BR475/D

Advanced Logic Functions

ReliaMlity

BR904/D

MlL-Processed

BR923/D

Communications, Power & Signal Technologies

September-December
1993

~,
,,,~,,:.{,,

\~iFf
..,.,:.J
>..*, .,*\j
:; . +~~
,~.~,1,].,.4,$
,~~>~:>
$?,
.>.
,P*
,),.?
.,,,,,.
,),
...
,.,.>~
, ..>~.,s,
~s.r ,}
,$.J
~.!/~ ,,$,,~-,

Devices: Technical Data

BRI 330/D

Group, Relia~li@J~:,#
fs\

... ..>,..
,:):,<:t
Military Analog Lineup
.,.:,~~..
\.<k
&
\,t~f ,,~.li,\,i,
Coming Through Loud and Clear
,,,~.~t.yjy
,,,..
Linear Integrated Circuits: New Product Calendar$J~W~@ 1994
<:$,.
~~?)!:
.
ECLinPS Lite Single Gate ECL Devices
?::+,,

BR13321D

Logic Integrated Circuits Division: New Pro@~~$$~qlendar Second Quarter, 1994

BRI 3331D

Low Skew Clock Drivers & Programma~~~

BR13341D

BR9241D
BRl130/D
BR1305/D

Grcuits (Rev. 3)

BR1409/D

High Performance Frequency Contr@~@:@s

,~,.;
~..,~
ECL300 Logic Array
,> ,J:,.,

BR14151D

Military Telecom Special Fun@&s

BR1418/D

Military Analog, Telecom, at~~edal

BR1429/D

Wideband Linear Ampliff~~%@AW,

BRE3781D

UnitPAK Packaging

BRE5041D

Electronic Tuning A~@re~s Systems

DLl10/D
DLlll/D

RF Device D@V~(Rev.5, 7994)

**X
Bipolar PQ$>$,,~hsistor
Data (Rev.6,

DL1 22/D

MEC~.@~y~~~ata

DLI 26/D
DL1281D
DL140/D
DL1451D
DL148/D

Report,

(Rev. f)

?:.\

Functions Fact Sheet, June 1992

CRT Drivers, General Pur~se

~~~:{

1992)

(Rev. 5, 1993)

S~#~%$~@bl Transistors, FETs and Diodes Device Data (Rev. 4)

~l$~~wnd

Interface Integrated Orcuits (2 volume set, Rev. 4, 1993)

.:: ~ph Performance ECL Data - ECLinPS and ECLinPS Lite (Rev. 2, 1993)
,.l.,>:.,.
~1, ..~ilitary MECL Family Data
,3\i.+),?+.+,$?
i Discrete Military Operations Data
Rectifier Device Data (Rev, 1. Rep/aces DL725/D)
Power Applications Manual (Rev. 7)
Communications

Applications

FET Applications

Manual

MECLSystem

Manual (Rev. f)

Design Handbook (Rev. 1)

RF Products Selector Guide &Cross

Reference 1994

timmercial
Plus and Mil/Aero Application
(Rev.5, f993)
SCANSWITCH

~}~

lC&Discrete

(Rev. 71, 1994)

Selector Guide

Selector Guide (Rev. 7,7990)

Mixed Signal Solutions from MOS Digital-Analog

- Quarter 1, 1994

integrated Circutis Division

SG2701D

Discrete Semiconductor

SG365/D

Low Skew Clock Drivers and Program mable Delay Circuits (Rev. 2)

SG366/D

~L,

Cross Reference Guide lgg2

ECL, CMOS and Special Logic Circuits Selector Guide (Rev.3,

471

7993)

Additional Information (continued)

...

SG370D

Discrete Sutiace Mount Selector Guide (Rev. 1, f 994)

SGE1lZD

Cross Reference for NEC-to-Motorola

TB3261D

Radio Frequency Transistors: Principles and Practical Applications

(Dye and Granberg, 1993)

RF Transistors

__

...

rz.,

=.-...,