Powell Technical Briefs - Combined 1-106

01.4TB.
001

Fast Bus Transfer Revised

Powell Industries, Inc.
PO Box 12818
Houston, TX 77217

2005 Powell Industries, Inc. All rights reserved.
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J anuary 22, 2013

Fast bus transfer is normally used for transferring a bus supplying motors to an emergency power source
on failure of the normal source of power. It is essential that this transfer be accomplished with a minimum
of "dead time" to prevent loss of critical motors or damage to the motors on re-energization.

Two schemes of operation are used for fast transfer. In the first scheme, the trip signal to the opening
breaker and the close signal to the closing breaker are given simultaneously. This is called a
simultaneous fast bus transfer and the dead time will typically be 1-3 cycles. However, there is a
possibility of overlap between the two sources, which may lead to the incoming breaker closing into a
fault. This can be prevented by adding a few milliseconds of time delay to the closing signal. In the
second scheme, the closing signal of the second breaker is initiated by a "b" contact of the opening
breaker. This may be either standard "b" contact or a fast "b" contact. This is called a sequential fast bus
transfer and the dead time will typically be 5-7 cycles. Both the schemes require a high speed sync
check relay between the alternate source and the motor bus for phase angle measurement. Make sure
that the V/Hz value does not exceed 1.33 p.u across the alternate source and the motor bus before
closing the alternate source breaker.
We have performed the timing tests on the PowlVac
vacuum circuit breaker to determine fast transfer

dead times. The following table lists the dead times for simultaneous and sequential fast bus transfer
schemes.
Source of Closing Signal
Dead Time, ms
No Arcing With Arcing
Simultaneous Close and Trip
Signals
7.0 - 17.0 (1.0)* - 9.0
Trip Then Close, Using Fast "b"
Contact
53.0 - 63.0 45.0 - 55.0
Trip Then Close, Using Standard
"b" Contact
57.5 - 67.5 49.5 - 59.5
*Possible overlap

Baldwin Bridger, P.E.
Technical Director

This technical brief was originally issued by Baldwin Bridger on April 23, 1990. It has been revised and reissued by Santosh Reddy.

01.4TB.002

Closing and Latching Capability of Medium Voltage Power
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

May 18, 1990 Superseded by PTB #107 (November 15, 2012)

ANSI Standard C37.06-1987, American National Standard for Switchgear - AC High Voltage Circuit
Breakers Rated on a Symmetrical Current Basis - Preferred Ratings and Related Required Capabilities,
includes a column in Table 1 headed Closing and Latching Capability. In older editions of this standard,
the current value in this column was given in rms kiloamperes, and was determined by multiplying the
maximum symmetrical interrupting capability by 1.6. In the 1987 edition, this current is expressed in crest
kiloamperes, and the value is determined by multiplying the maximum symmetrical interrupting capability
by 2.7.

Other standards had previously required the closing and latching current to have a crest value of 2.7
times the maximum symmetrical interrupting current, so the performance required of the circuit breaker
has not really changed. Only the method of stating the requirement has changed. This change was made
to bring the ANSI standard in line with the IEC standard, which also expresses closing and latching
capability in crest amperes.

Since many specification writers will be using older standards, or copying older specifications, we will
probably see both methods of specifying closing and latching current used in specifications for many
years. The following table gives both sets of values.
Rated
Maximum
Voltage
kV, rms
Rated Short
Circuit Current
kA, rms
Nominal
MVA
Closing and Latching Capability per ANSI
C37.06
1979 Edition kA, rms 1987 Edition kA, Crest
4.76 29 250 58 97
4.76 41 350 78 132
8.25 33 500 66 111
15.0 18 500 37 62
15.0 28 750 58 97
15.0 37 1000 77 130

01.4TB.002

Closing and Latching Capability of Medium Voltage Power
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
If the specified value of closing and latching current matches a value from either edition of the
standard, we can assume that a standard breaker is desired. If there is any possibility of
confusion, the specifier should be contacted to determine which basis is being used to specify
the close and latch rating.

Technical Director

01.4TB.003

Capacitance Current Switching Capability of PowlVac

Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J une 7, 1990 Superseded by PTB #108 (November 15, 2012)

We have recently had capacitance current switching tests performed on our "Dash 3" PowlVac
circuit
breakers, using GE interrupters. The results of these tests showed that these breakers are qualified as
definite purpose circuit breakers, in accordance with ANSI Standard C37.06-1987, Table 1A, for both
isolated and back-to-back switching of capacitors.

Table 1 lists the maximum rating of capacitor bank that can be switched by each rating of circuit breaker
when applied in accordance with ANSI/IEEE Standard C37.012-1979. The values in the table were
calculated using a total current multiplier of 1.25 for ungrounded capacitor banks and 1.35 for grounded
banks. These multipliers include allowances for higher than normal voltage, capacitor tolerance, and
harmonic components in the current. See ANSI/IEEE C37.012-4.7.1. When PowlVac
circuit breakers
are used in a back-to-back switching situation, inrush currents and frequencies must be limited to the
values given in Table 1A of ANSI C37.06-1987. This may require the addition of reactance between the
two capacitor banks.
Table 1: Capacitor Bank Switching Capability of " Dash 3" PowlVac
Circuit Breakers
Circuit Breaker Type and
Rating
System Voltage
kV
Maximum Nameplate Rating of Capacitor Bank,
MVAR
Ungrounded Bank Grounded Bank
1200A
Breaker
2000A
Breaker
1200A
Breaker
2000A
Breaker
05PV0250 4.76kV 250MVA
2.4 2.09 3.33 1.94 3.08
4.16 3.63 5.76 3.36 5.34
4.76 4.15 6.60 3.85 6.11
15PV0500 15.0kV 500MVA
11.5 10.04 15.93 9.30 14.75
12.47 10.88 17.28 10.08 16.00
13.2 11.52 18.29 10.67 16.94
13.8 12.05 19.12 11.15 17.71
14.4 12.57 19.95 11.64 18.48

01.4TB.003

Capacitance Current Switching Capability of PowlVac

Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Note: This table does not apply to PowlVac
circuit breakers using Mitsubishi interrupters. We have not

tested those breakers for capacitance current switching capability, but we do have some data from
Mitsubishi that allows us to apply them. Such applications should be referred to me for checking.

Technical Director

01.4TB.004

Umbilical Cord Used on PowlVac
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J uly 28, 1990

Occasionally, customers or prospective customers question our use of a manually-operated control
disconnect ("umbilical cord") on our PowlVac
circuit breakers. Some of the questions asked, and our

answers to them, are:

Q. Why does Powell use an umbilical cord for its control disconnect?

A. The use of the umbilical cord is part of our user-friendly design, which locates all circuit breaker
control accessories in the front of the cell. In addition to the control disconnect, these devices include the
mechanism-operated cell switch (MOC) and the truck-operated cell switch (TOC). In our PowlVac

design, these devices are located where they may be observed by an operator inserting or removing the
circuit breaker, allowing the operator to check alignment and operation when the circuit breaker is
installed. These devices are also available for servicing without removing the circuit breaker from the cell.

Q. Is this design safe?

A. Yes. The umbilical cord's plug mechanism is mechanically interlocked with the circuit breaker to
insure safe operation. Interlocks provided include:
The circuit breaker cannot be inserted into the cell without plugging in the umbilical cord.
Once the circuit breaker racking mechanism has been operated to start the circuit breaker
insertion process, the plug cannot be removed. It is therefore not possible to disconnect the
control circuits of a circuit breaker that is in service.
Unplugging the umbilical cord trips the circuit breaker if it is closed and discharges the closing
spring if it is charged. Since the plug must be removed in order to remove the circuit breaker from
its cell, these interlocks insure that the circuit breaker is open and all energy storage springs are
discharged when the circuit breaker is taken out of the cell.
Q. Why does Powell differ from all other manufacturers in the method of disconnecting the
control connections to the circuit breaker?

A. Powell does not differ from "all other manufacturers". While the umbilical cord design has not been
used frequently in the United States, other American manufacturers have used it. It is also commonly
used in Europe. We chose to use this design because we think it offers superior performance in total.

01.4TB.004

Umbilical Cord Used on PowlVac
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Q. Does the umbilical cord design meet ANSI standards?

A. Yes. This design, including required interlocking, is covered in detail in ANSI/IEEE Standard
C37.20.2-6.2.7. The PowlVac
circuit breaker meets these requirements.

Technical Director

01.4TB.005

Comparison of Porcelain & Cycloaliphatic Epoxy Insulation

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J uly 29, 1990

PowlVac
vacuum circuit breakers and metal-clad switchgear use a primary insulation system of
cycloaliphatic epoxy. This insulation has given excellent results in the eight years since we first
introduced PowlVac
, but we still have customers who request porcelain.

Powell is far from alone in using cycloaliphatic epoxy insulation. The material has been in common use in
Europe for a generation, and other U. S. users include Westinghouse, S&C and Square D. It is especially
interesting to see the first two of these companies using cycloaliphatic epoxy. A few years ago, both were
strong proponents of porcelain insulation.

Although there are many formulations of cycloaliphatic epoxy and a number of varieties of porcelain,
each of which has its own specific qualities and parameters, there are a number of general comparisons
which can be made.

First, in the physical area, the following relationships are typical:
Cycloaliphatic epoxy ("cyclo") weighs less than 70% of porcelain's weight.
The thermal coefficient of expansion of cyclo is 1/20th that of porcelain.
The tensile strength of cyclo is about 11 times that of glazed porcelain.
The compression strength of cyclo is 4 to 6 times that of glazed porcelain.
The flexural strength of cyclo is 16 to 18 times that of glazed porcelain.
The Izod impact strength, unnotched, is about the same as glazed porcelain.
Dimensional and shape control is much easier in cyclos than in porcelain.
While the repairability of cyclos is limited, porcelain is unrepairable.
In the electrical area, you will find:
The dielectric constant of cyclo is only about two-thirds that of porcelain.
The temperature class of porcelain is much higher than that of cyclo, but cyclo mixtures with
temperature classes of 105 C or 130 C are readily available.
The track resistance of cyclo is slightly less than that of porcelain.
The water absorption of cyclo is slightly greater than that of porcelain, but is still in the range of
2/10's of 1%.

01.4TB.005

Comparison of Porcelain & Cycloaliphatic Epoxy Insulation

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Finally, cyclo exhibits excellent resistance to common industrial chemicals, is readily washable, and has
excellent erosion resistance and weathering properties.

In summary, we believe that the excellent physical properties of cyclo make it the insulating material of
choice in spite of some small sacrifice in electrical properties. This is especially true for applications
requiring great strength under severe dynamic loading, such as support insulators in circuit breakers and
switchgear.

Technical Director

01.4TB.006

Effect of Solar Radiation on Outdoor Metal-Enclosed
Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J uly 30, 1990

From time to time we get questions about the rating of outdoor metal-enclosed switchgear which is
exposed to solar radiation. It is fairly obvious to anyone who thinks about it that switchgear sitting out in
the sun gets hotter than switchgear sitting in the same ambient air temperature inside a building where it
has no solar exposure. How should we handle this extra heat?

Metal-enclosed switchgear built to ANSI standards, as is all Powell switchgear, is rated in accordance
with the usual service conditions set forth in those standards. All four of the ANSI product standards we
commonly use (C37.20.1 for low voltage switchgear, C37.20.2 for metal-clad switchgear, C37.20.3 for
interrupter switchgear, and C37.23 for bus duct) include as one of the usual service conditions that the
effect of solar radiation is not significant. Thus, all testing and rating of switchgear ignores the effect of
solar radiation.

When switchgear is installed in a location where solar radiation is significant, there is another ANSI
standard to give guidance in properly applying the switchgear. ANSI/IEEE C37.24-1986, IEEE Guide for
Evaluating the Effect of Solar Radiation on Outdoor Metal-Enclosed Switchgear, gives the information
necessary to allow calculating the derating of the continuous current capability of switchgear exposed to
the sun. This standard is site-specific; the derating depends on the location of the switchgear installation.

As a switchgear manufacturer, we assume that our customers specify switchgear ratings in accordance
with the usual service conditions given in the product standards. We further assume that the specifier will
do the necessary evaluation and either limit his loads or upgrade his ratings to take care of any solar
radiation derating that is needed. If requested, we will be glad to discuss this derating with our
customers, and to assist them with the calculations if necessary, but we should not be expected to
automatically quote a 2000A circuit breaker where a 1200A circuit breaker is specified, just because the
installation is outdoors in Yuma, Arizona.

Technical Director

01.4TB.007

Seismic Testing of PowlVac
Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 29, 1990

We often see specifications that call for switchgear "to be suitable for use in seismic zone X", where X
may be any number from 0 to 4, depending on the location of the final installation of the switchgear.
Unfortunately there is no ANSI standard that defines "suitable for use in seismic zone X". Seismic
requirements for nuclear generating station equipment, which do exist in standards, are not stated in
terms of seismic zones, but are site specific.

ANSI Standard A58.1-1982, Minimum Design Loads for Buildings and Other Structures, gives some
guidance for the seismic loading that various items must withstand, using the basic formula:

where is the lateral force to be designed for,

is the seismic zone coefficient, which varies from 0.125 for Zone 0 to 1 for Zone 4,

is the occupancy factor, which varies from 1 for Category I to 1.5 for Category III,

is the horizontal force factor, which is 0.3 for all machinery in a building,

and is the weight of the equipment.

From basic mechanics, Force =Mass x Acceleration. In the above formula, F
p
is a force. W
p
is a weight,
which is the product of a mass and the acceleration of gravity, or g. It follows that the product of Z, I and
C
p
is a dimensionless coefficient for g. For a worst case situation, where the switchgear is installed in a
critical occupancy in Zone 4, the value of this coefficient is 1 x 1.5 x 0.3, or 0.45. Since seismic testing is
performed in terms of acceleration rather than force applied, the test level for a worst case installation
should be 0.45 g.

The other aspect of suitability is the performance of the equipment under the specified conditions. Here,
we have absolutely no guidance from ANSI standards. Based on past experience and input from various
users, Powell has decided that the following are reasonable criteria for suitability:

01.4TB.007

Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

1) There shall be no structural damage that prevents normal operation of the equipment after the event.

2) No doors or covers shall open during the event.

3) The circuit breakers shall not open or close during the event except on command.

4) The circuit breakers shall not move from the fully connected position during the event.

5) After the event, it shall be possible to open and close the circuit breakers and rack them into and out
of the connected position.

6) Primary and control fuses shall remain in their fuse clips.

7) Transformer rollout drawers shall not come open during the event.

8) After the event, primary circuits shall withstand a 27 kV power frequency withstand test (hipot). The
value of 27 kV is chosen because it is the power frequency withstand voltage specified for field testing of
15 kV metal-clad switchgear.

About four years ago, Powell had samples of PowlVac
metal-clad switchgear tested for the ability to

withstand Zone 4 seismic forces. These samples were single-unit equipments, to give the narrowest
structure possible, and had the heaviest circuit breakers installed in the highest positions in which they
are ever used. They were therefore worst-case seismic samples.

Based on the requirements of ANSI A58.1-1982, we chose to use 0.45 g as the zero period acceleration
(ZPA) value for these tests. The seismic experts at Southwest Research Institute in San Antonio took this
value and developed a required response spectrum (RRS) that peaked at about 1.8 g at 3.5 Hz for
vertical acceleration and about 1.9 g at 2.5 Hz for horizontal acceleration, with a minimum value of 0.45 g
(the ZPA) at frequencies above 32-33 Hz. Full seismic tests were done by Southwest Research Institute
at these values of acceleration.

The eight criteria listed on the previous page were used to judge the performance of the equipment
under seismic test. In addition, the circuit breakers were successfully closed and tripped on command
during the seismic test. Except for a minor problem with the transformer rollout drawer, the equipment
performed as required. The rollout drawer fastening system was reinforced, and the equipment
performed successfully on retest.

01.4TB.007

Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3
Based on these tests, standard PowlVac
metal-clad switchgear is suitable for use in seismic zones 0, 1

and 2. With the addition of holding clips at the transformer rollout drawers, PowlVac
is suitable for use in

zones 3 and 4.

Technical Director

01.4TB.008

Preventing Voltage Feedback in Synchronizing Circuits

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 22, 1990

Many synchronizing schemes use two lamps in series, connected from the incoming voltage source to
the running voltage source. This "dark lamp" synchronizing indication can be used by an operator to
supplement the meter and synchroscope readings to insure synchronism before closing the incoming
circuit breaker.

This scheme, however, can allow energizing of a supposedly dead bus if the synchronizing switch is
accidentally left in the "ON" position. The two lamps will be in series with the secondary of the bus
voltage transformer, and this circuit will be connected across the energized incoming voltage transformer
secondary. The portion of this voltage which appears across the bus voltage transformer will be stepped
up by the ratio of the bus voltage transformer, and this higher voltage will be applied to the switchgear
bus.

To prevent this voltage feedback, a dead bus relay (27B) should be connected in the circuit as shown in
the figure below. For simple synchronizing schemes, where one or more generators are manually
synchronized to a common bus, this circuit with its one 27B relay is satisfactory. For more complex
schemes, involving automatic synchronizing, machine-to-machine synchronizing, or synchronizing to a
utility source, a more complex circuit may be necessary to insure that no voltage feedback circuits exist.
All synchronizing circuits should be reviewed carefully to prevent voltage feedback through the
synchronizing lamps.

Technical Director

01.4TB.009

Fuses for Use in DC Control Circuits

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J anuary 9, 1991 See revised PTB #105 (September 14, 2012)

The majority of control circuits in metal-enclosed switchgear, particularly in metal-clad switchgear, are
supplied from a dc power source. For nearly half a century Powell and other switchgear manufacturers
have used 250-volt cartridge fuses (so-called "Code fuses") to protect these control circuits. Typical fuse
types are Bussmann Type NON and Shawmut Type OT. The application of these fuses to this type of
circuit has been generally successful and has been generally accepted by our customers.

From time to time, however, someone raises the question of the dc rating of these fuses. Bussmann
advises me that the Type NON has been tested successfully for 10 kA interrupting capability at 250 V dc,
which is the rating commonly ascribed to these fuses. Based on this test data, we can safely apply these
fuses to dc control circuits where the short circuit level of the control circuit is 10 kA or less. The typical
control battery used for switchgear can deliver a short circuit current of about 10 times its one-minute
discharge rating, so it would be a very unusual dc control circuit that had a short circuit capability in
excess of 10 kA.

Another question sometimes raised is whether or not these fuses are UL listed for dc applications. The
answer is no. If a fuse with a UL listing for dc use is required, we should use either Fusetron Type FRN-R
or Low-Peak Type LPN-RK. These fuses are dual-element time delay types which may be used in the
same fuse blocks used for Type NON fuses.

Technical Director

01.4TB.010

Transient Recovery Voltage (TRV) Values for Testing
PowlVac
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J anuary 10, 1991 Superseded by PTB #104 (September 14, 2012)

The interrupting performance of any circuit interrupter is affected by the transient recovery voltage
appearing across the first pole to interrupt. Both the absolute value of this voltage and its rate of rise are
important in determining the interrupter's ability to meet its interrupting rating. The required values of
transient recovery voltage are included in ANSI/IEEE C37.06-1987, along with the other ratings of circuit
breakers.

The conventional way of specifying the rate of rise of the transient recovery voltage is to specify the peak
value (E2) and the time required to reach that peak (T2). The rate of rise is then determined by dividing
E2 by T2. The nominal values are those for a full rated short circuit interruption. For lower currents, both
higher peaks and faster times are specified. Table 6 of ANSI/IEEE C37.06-1987 lists the multiplying
factors to be applied to E2 and T2 for interrupting currents below the full rating of a circuit breaker.

Table 1 of ANSI/IEEE C37.06-1987, which gives the preferred ratings of indoor oilless circuit breakers,
such as PowlVac
breakers, calls for E2 to be 1.88 times the breaker's rated maximum voltage for tests
at 100% of the circuit breaker's interrupting rating. Unfortunately, values of T2 are not standardized,
leaving the manufacturer with no guidance on this subject. In order to assign some reasonable value to
T2, Powell decided to use the rate-of-rise values given in Table IIA of IEC Standard 56, interpolating
between the listed values to match the ANSI voltage ratings, and multiplying the rate-of-rise values by E2
to obtain T2. The values obtained by this method were used in the testing of PowlVac
circuit breakers,
and are given in the table below.

01.4TB.010

Transient Recovery Voltage (TRV) Values for Testing
PowlVac
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

PowlVac
Transient Recovery Voltage Test Values

Current % of Interrupter
Rating
Transient Recovery Voltage
Rated Maximum Voltage = 15 kV
Rated Maximum Voltage = 4.76
kV

7 to 13 33.00 29 1137 10.47 19.8 529
20 to 30 31.86 29 1098 10.11 19.8 510
40 to 60 30.17 49 615 9.58 33.1 289
100 28.20 73.6 383 8.95 49.4 181

Technical Director

01.4TB.011

Consequences of Vacuum Interrupter Failure

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 1, 1991

Users and prospective users of vacuum circuit breakers frequently ask us what happens if a vacuum
interrupter fails to interrupt. The short answer to this question is that the interrupter is usually destroyed
and must be replaced. However, this short answer needs some additional comment to be really
informative.

First, failure of a properly applied vacuum interrupter to interrupt a fault current within its rating is a very
rare event. In the 8 years that we have been building PowlVac
vacuum circuit breakers, we have

manufactured over 3200 breakers. Assuming an average of two years in service for these breakers, we
have a history of nearly 20,000 interrupter-years of service. We have never heard of a failure to interrupt
by any of these circuit breakers. We are proud of this history, but, based on industry statistics, we are not
surprised by it.

Second, even if an interrupter does fail, the consequences are not the disastrous burn down that some
people imagine. During some recent design tests of a prototype of a new version of the PowlVac

breaker, we drove an interrupter far past its rated contact life span and had a failure. Photo 1 shows the
failed interrupter. When failure occurred, the internal shield was burned through and the ceramic
envelope, exposed directly to the arc, broke apart. The arc continued for several cycles, until the circuit
was opened by a backup circuit breaker. Aside from the failed interrupter, the only damage to the circuit
breaker was a small area of smoke and burn discoloration on the nearby insulating material. Photo 2
shows this area, which was about 6 inches square. Five minutes with an industrial cleaner and a couple
of paper towels removed all but about one square inch of this discoloration. The remaining area seemed
to be singed, but there was no detectable erosion of the surface of the insulating material. Had this
breaker been in service, it could have been returned to service immediately after replacing the
interrupter.

01.4TB.011

Consequences of Vacuum Interrupter Failure

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Summing up, interrupter failures are rare, and when they do happen, most are not a major disaster.

Photo 1
Failed Vacuum Interrupter
Photo 2
Discolored Insulation at
Failure Location

Technical Director

01.4TB.012

Continuous Current Carrying Capability of Low Voltage
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 4, 1991

Various types of low voltage circuit breakers have differing continuous duty capabilities. Some are rated
to carry 100 percent of their trip rating continuously, while others are rated to carry only 80 percent of
their trip rating continuously. It is important that we understand the difference and apply these breakers
properly.

The general run of molded case circuit breakers in frame sizes of 400 A and below are rated to carry only
80 percent of their rated trip current on a continuous basis. Particularly when these breakers are
mounted close to each other in a panelboard, the extra heat generated by carrying 100 percent of the trip
rating will both lead to false tripping and cause long-term degradation of the insulating material of which
these breakers are made.

On the other hand, all low voltage power circuit breakers and the general run of insulated case circuit
breakers are capable of carrying 100 percent of their trip rating on a continuous basis.

Some confusion can arise when using large molded case circuit breakers, in frame sizes of 600 A and
above. These breakers may be rated either 80 percent or 100 percent, depending on the model and the
manufacturer. As you would expect, the 100% breaker costs considerably more than the 80% breaker.
Some models have both 80% and 100% ratings available. The 100% rated breaker may require a larger
enclosure and/or more ventilation than the 80% rated breaker of the same model.

Please observe the following application rules:

1) Apply MCCB's in 400 A frame size and smaller based on continuous loads of not more than 80% of
the circuit breaker's trip rating. If trip ratings are selected by our customer, assume that they are based
on the 80% load requirement.

2) Apply insulated case breakers and low voltage power circuit breakers based on continuous loads of
not more than 100% of the breaker's trip rating. If trip ratings are selected by our customer, assume that
they are based on the 100% load requirement., Be sure that the insulated case breakers selected are
100% rated.

01.4TB.012

Continuous Current Carrying Capability of Low Voltage
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
3) Apply large molded case circuit breakers based on either the 80% or the 100% rating, making sure
that the breaker selected fits the application, and that adequate space and ventilation is provided for the
breaker chosen. If trip ratings are selected by our customer, be sure that you understand which basis
was used for selection.

Technical Director

01.4TB.013

Future Use of Space in Powell Equipment

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 27, 1991

Powell's switchgear and motor control equipments frequently include space which is not used by active
switching devices, but is available for future use. This space varies in the amount of equipment present,
and is called by many different names. Some of the terms used include space, future, future space,
equipped space, space only, spare, and blank. Unfortunately, there are no industry standards defining
these terms and their use varies widely throughout the industry, so there is often confusion between
specifier and manufacturer or between engineering and shop personnel about what is desired on a
particular job.

In order to minimize the confusion, we have adopted the following terms and descriptions in Powell for
internal use:

Spare - A complete, ready-to-operate unit, including the drawout switching device (circuit breaker or
motor starter) and all required secondary devices, fully wired. A spare differs from an active unit only in
that the spare has no assigned function in the power system.

Fully Equipped Space - A spare without the drawout switching device. Includes all required secondary
devices and wiring, a finished unit door, primary buswork and disconnecting devices, and all cell parts
required for inserting the drawout switching device.

Equipped Space - Includes a door with cutouts for primary switching devices but not for secondary and
control devices, primary disconnecting devices and riser bus connecting them to the main bus, and all
cell parts required for inserting the drawout switching device. No primary or secondary devices are
included, and wiring is minimal.

Blank Space - A blank door, no primary or secondary devices, buswork, wiring, or cell parts required for
inserting the drawout switching device. Steelwork should be done so that the blank space can be
equipped in the field with little or no cutting or welding.

Blank - An area that can never be used for a primary switching device. This area is made unusable by
thermal limitations of the equipment, inability to bus to the area or to maintain proper isolation of bus or
outgoing leads, or some similar problem.

Related to these definitions but somewhat different is Mounting and Wiring for a future device or a
device to be field installed by the user. Mounting and wiring may be furnished in any of the above units or
in an active unit. Mounting and wiring includes the necessary space, physical supports, and primary and
secondary connections to allow easy installation of the future device. This may include temporary
primary and/or secondary connections or jumpers to allow use of the circuit pending the addition of the
future device.

01.4TB.013

Future Use of Space in Powell Equipment

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Where any of these conditions leave openings in the front door or in isolation barriers required by
standards, the opening must be covered by a temporary cover plate.

Technical Director

01.4TB.014

Autotransformer Starting of Motors

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 1, 1991

One of our customers recently experienced failures of two autotransformers used in medium voltage
motor starters. The circuit used was the familiar 3-contactor, 2-coil Korndorfer circuit, which has been
used for many years and appears in textbooks and handbooks on motor control. The primary circuit is
shown below:

An investigation of the failed autotransformers by their manufacturer showed that the failure had been a
surface flashover from the line end of the winding either to another tap of the winding or to a ground
point. There was no damage to the winding or the core, and the autotransformers could be easily
repaired and put back into service.

We consulted with both the autotransformer manufacturer and the manufacturer of the contactors used in
the starter, and found that there had been previous experiences of this problem. The flashovers occurred
because system transients generated during the starting sequence caused an excessive voltage to
appear on the line end of the autotransformer winding. Upon analysis, we found several conditions that
contributed to this problem:
The starter was located at the end of a rather weak supply line.
During the starting sequence, the user switched in a rather large capacitor bank to minimize the
line voltage drop. This bank was switched off automatically, during the starting sequence, when
the voltage recovered to a fixed point.
The autotransformer was set on the 80% tap.
We are uncertain of the setting of the timer used to transfer from the starting connection to the
running connection.

01.4TB.014


PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Although the contactors used in this particular installation were vacuum contactors, the manufacturer
informs us that similar problems have been encountered with both air and vacuum contactors. The type
of contactor used doesn't seem to be a factor in the occurrence of the problem.

Further discussions with our suppliers led to several suggestions to minimize the occurrence of this
problem:
Insulate the transformer connection points, both the taps that are used and the unused taps. This
should be done on all future starters of this type.
Use a lower voltage tap on the autotransformer, such as 65% or 50%, if the motor will accelerate
successfully on these taps.
For induction motors, be sure that the timer that transfers to the running connection is set at a
long enough time so that the motor is fully accelerated before changing to the running connection.
Add an instantaneous current relay to the circuit, set to pick up at about 5 A and drop out just
below that current. This relay will pick up when the motor is started and drop out when it reaches
full speed. Connect the coil of this relay in any phase CT. Use the contact of this relay to bypass
the timing relay contact, insuring that the motor has fully accelerated before the starter is
transferred to the running connection. See the control circuit below. In the future, please include
this relay in all starters of this type.

01.4TB.014


PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3
In extreme cases, it may be necessary to connect intermediate class surge arresters to the line
taps of the two autotransformer coils.

Technical Director

01.4TB.015

Directional Overcurrent and Directional Power Relays

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

May 24, 1991

From time to time we experience some confusion about the difference between directional overcurrent
relays, ANSI device 67, and directional power relays, ANSI device 32. Although there are some
similarities between these two types of relays, they are really very different in both construction and
application.

Directional overcurrent relays (67) respond to excessive current flow in a particular direction in the power
system. The relay typically consists of two elements. One is a directional element, which determines the
direction of current flow with respect to a voltage reference. When this current flow is in the
predetermined trip direction, this directional element enables ("turns on") the other element, which is a
standard overcurrent relay, complete with taps and time dial, as found on a normal non-directional
overcurrent relay. Because these relays are designed to operate on fault currents, the directional unit is
made so that it operates best on a highly lagging current, which is typical of faults in power systems.
Directional overcurrent relays are normally used on incoming line circuit breakers on buses which have
two or more sources. They are connected to trip an incoming line breaker for fault current flow back into
the source, so that a fault on one source is not fed by the other sources. In complex distribution or sub-
transmission networks, these relays may be used to improve coordination of the system.

Directional power relays (32) measure real power , so they operate best at a high power
factor. Various degrees of sensitivity and speed of operation are available in various models of
directional power relays. There are three typical uses of these relays:
Connected to measure power flow into a generator, the relay will operate to trip the generator
breaker if the generator begins to draw power from the system and act as a motor. This is usually
due to loss of prime mover power.
Connected to measure power flow into a transformer from the secondary side, a very sensitive
directional power relay can measure core loss power input to the transformer, detecting loss of
the primary source to the transformer. The transformer can then be disconnected from the
system.
A directional power relay can be used to limit power flow in a circuit. The relay may trip a breaker
or initiate control action to change the system configuration. By using quadrature potential
connections or a phase shifting transformer, these relays can be made to measure vars
. A typical use would be to limit the real or reactive power drawn from a utility source
to a contractual level.

01.4TB.015

Directional Overcurrent and Directional Power Relays

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Neither the functions (67 and 32) nor the actual relays are interchangeable. Be sure to use the function
and the hardware which fit the application.

Technical Director

01.4TB.016

Preventing Condensation in Medium Voltage Motors

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J une 12, 1991

Condensation or other accumulation of moisture can be very damaging to the windings and mechanical
parts of a motor, especially a medium voltage motor. This is not usually a problem for a motor that is
running, as the windings generate enough heat to prevent condensation. When the motor is stopped,
however, supplementary heat is often required to keep the motor dry.

One way of providing the required heat is to install heaters in the motor. Another way is to energize the
motor windings from a low voltage source. The one-line diagram below shows the connections for this
method of heating the windings. This method may be preferable to the use of heaters, as it actually heats
the windings instead of relying on the transmission of heat from a separate heater.

When using this method of heating, several precautions must be observed:

01.4TB.016

Preventing Condensation in Medium Voltage Motors

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
The heating contactor must be a full line voltage contactor, as the motor winding side of this
contactor is energized at line voltage when the motor is running.
The running contactor and the heating contactor must be mechanically and electrically interlocked
so that only one of them can be closed at any time.
There needs to be a time delay between the opening of the running contactor and the closing of
the heating contactor, to allow the residual voltage on the motor to decay before the motor
windings are connected to the low voltage source. Since it is not critical to apply the heating
circuit immediately, it is recommended that this time delay be in the order of 2 to 5 minutes.
Tests show that there is an open circuit time of approximately 75-80 milliseconds when the
running contactor is picked up by a "b" contact of the heating contactor. The user should consider
whether this is an adequate time period to prevent unwanted system problems. If not, a time
delay of a few seconds can be inserted in the pickup circuit of the running contactor to be sure
that the heating contactor has cleared before the motor is energized by the operating voltage.
The voltage applied to the motor windings must be carefully selected to produce the proper
heating. This value must be selected by the user, based on input from the motor manufacturer.

Technical Director

01.4TB.017

Ground Lead Disconnectors on Distribution-Class Surge
Arresters

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J uly 18, 1991

Many current models of zinc oxide distribution or riser pole arresters come equipped with ground lead
disconnectors. This is a device which is mounted on the ground end of the arrester and which looks
about like a small hockey puck. The enclosure is black, blue or green plastic, a couple of inches in
diameter and an inch or so tall.

The normal failure mode of these arresters is a short circuit to ground, causing ground fault current to
flow. This current will cause the arrester body to fail if it is not stopped quickly. The first function of the
ground lead disconnector is to disconnect the ground lead of the surge arrester in case of an internal
failure of the arrester, preventing explosive failure of the arrester body. The ground lead disconnector
contains a cartridge in series with a gap. The gap is shunted by a resistor. As the current rises, the
voltage across the gap increases until the gap flashes over, creating an arc which ignites the cartridge,
blowing the ground lead free.

The ground lead disconnector is not a fault current interrupter. The arc drawn by the ground lead as it
separates from the body of the arrester may or may not go out on its own. If it does not go out, a circuit
breaker, recloser or fuse must operate to extinguish the arc. The ground lead disconnector is expected to
create a gap which will not reignite when power is reapplied to the circuit, but the gap which will be
created is a function of the length and flexibility of the ground lead.

The second function of the ground lead is to give a visible indication of arrester failure for arresters
mounted on overhead distribution lines. If a lineman sees an arrester with its ground lead hanging in
midair, he knows that he has a failure which must be replaced.

These explosive ground lead disconnectors are not suitable for use in metal-enclosed equipment.
We do not want the explosion and subsequent uncontrolled arc inside equipment, where the clearances
are not nearly as great as on overhead lines, and where secondary damage from the arc is much more
likely to occur. The visible indication function of the disconnector is useless if the device is mounted
within an enclosed equipment.

All surge arresters used in Powell's equipments should be of the type without ground lead disconnectors.
If a user requests that we include a surge arrester with a ground lead disconnector, we should offer an
equivalent model without the disconnector.

Technical Director

01.4TB.018

Operating Times of PowlVac
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J uly 19, 1991

We are frequently asked about the actual operating times of PowlVac
circuit breakers. The following

values may be used in application studies for these circuit breakers.

Closing Time

For all current production models of PowlVac
circuit breakers, the time from energizing the closing coil

with rated control voltage until the primary contacts touch is 80 milliseconds or less. Typical values are in
the 44 to 45 millisecond range.

Opening Time

Opening times vary with the model of PowlVac
breaker, as shown in the following table. All times are

from energizing of the trip coil with rated control voltage until the primary contacts part.
Breaker Model " Dash 2" " Dash 3"
Vacuum Interrupter Mitsubishi General Electric
Opening Time, milliseconds
Design Limits
Typical Test Values
25-35
26 or 27
40-50
48 or 49
"S" (asymmetry) Factor 1.2 1.1

All of these breakers are rated 5 cycles interrupting time in accordance with the preferred ratings found in
Table 1 of ANSI C37.06-1987, even though they may be faster. The "Dash 2" breaker, in particular, is
very nearly a 3 cycle breaker.

Technical Director

01.4TB.019

Use of PowlVac
Circuit Breakers for Continuous Currents

Above 3000 Amperes

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 26, 1991

In accordance with ANSI/IEEE Standard C37.06, the highest continuous current rating of our standard
line of PowlVac
circuit breakers is 3000 A. For systems that require continuous current ratings above
3000 A, we can offer two possible solutions.

First, we can offer our standard 3000 A circuit breaker with cooling fans. We have a design that has been
successfully tested at 3750 A, and the results of that test indicate that the fan-cooled breaker may be
applied at 4000 A without overheating. This design requires a unit somewhat wider than the standard 36-
inch switchgear unit to include the necessary air ducts. The standard fan control equipment includes a
current-actuated control to start the fans at about 2500 A and an alarm circuit which uses air flow
switches to detect and alarm loss of cooling air at currents above this level. A completely redundant
second set of fans can be furnished if desired. Fan cooling is our preferred method of obtaining higher
continuous current ratings.

A second method of providing for high continuous currents is to parallel two circuit breakers. Using this
approach, we can provide for continuous currents of about 3500 A by paralleling two 2000 A breakers
and about 5000 A by paralleling two 3000 A breakers. When breakers are paralleled, the interrupting
rating is neither increased nor decreased. Precise timing in closing or opening the two paralleled
breakers is not critical, as whichever breaker closes first can carry the continuous current for the few
milliseconds until the second breaker closes, and the last breaker to open has the capability of
interrupting the full fault current. Paralleling of breakers does require special circuitry to balance the
currents between the two breakers and individual overcurrent protection for each breaker as well as
combined overcurrent protection for the entire circuit. Main bus construction must also be very carefully
balanced to insure equal impedance in both legs of the circuit. Parallel breakers should only be used for
a user who refuses to use fan cooled circuit breakers.

Regardless of which breaker uprating method is used, special attention must be given to the design of
any portions of the switchgear bus which are rated over 3000 A. If the main bus exceeds 3000 A,
standard PowlVac
bus cannot be used, and the required special bus design limits the switchgear to
one-high construction.

Technical Director

01.4TB.020

Application of Dummy Circuit Breakers in Metal-Clad
Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 27, 1991

Dummy circuit breakers are used in metal-clad switchgear to provide a method of disconnecting and
isolating a circuit or circuits without using a circuit breaker. A common use of a dummy circuit breaker is
as a temporary connection in a switchgear cell where a circuit breaker will be installed as part of a
planned future expansion. Another use might be to isolate one end of a tie bus or cable from a
switchgear bus.

Because a dummy circuit breaker is really a set of three jumper bars mounted on a breaker carriage, it
has absolutely no current interrupting rating. If an attempt is made to withdraw the dummy circuit breaker
with current flowing, arcing will occur at the primary disconnect fingers. This may result in operator injury,
equipment damage, or both. Therefore, dummy circuit breakers normally are interlocked with other
switching devices so that the dummy cannot be withdrawn until the other devices are opened, insuring
that no current is flowing in the dummy.

A particular application that can be troublesome is isolating a tie cable that has been opened by a circuit
breaker at the other end. If the cable is still attached to an energized bus through the dummy breaker,
cable charging current will flow through the dummy. It only takes a few hundred feet of 15 kV cable to
draw a charging current of as much as half an amp. This highly capacitive current is difficult to interrupt.
It is recommended that the interlocking for any circuit involving power cable and a dummy circuit breaker
be arranged so that the cable is completely deenergized before the dummy circuit breaker is removed to
isolate the cable.

Deenergizing the unloaded bus of a lineup of metal-clad switchgear by withdrawing a dummy circuit
breaker is an acceptable application. The limited length and very low capacitance of a switchgear bus
structure keeps the charging current low enough to be successfully interrupted by withdrawing a dummy
circuit breaker.

Technical Director

01.4TB.021

Switching Capability of Rollout or Tiltout Carriages

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 3, 1991

We are often asked about the switching capability of the rollout or tiltout carriages used in medium
voltage switchgear to mount voltage transformers, small control power transformers, and fuses for larger
control power transformers. This question usually takes the form "How large a CPT can you handle with
fuses mounted in a rollout or tiltout?"

There is no industry standard to measure this switching capability, and no test data is available to certify
this performance. The switching capability will vary with the details of the design, and to some extent will
depend on the operator, since the speed of opening a rollout or tiltout depends on the individual opening
the device.

Within these restraints, however, our experience with 5 kV and 15 kV equipments over the years has led
us to adopt the following limits:
Voltage transformers: A set of three wye connected VT's or two open delta connected VT's can
be switched with a rollout or tiltout without any interlocking of the secondary circuit.
Control power transformers: A CPT up to 50 kVA single phase or 75 kVA three phase can be
switched with a rollout or tiltout provided the carriage is interlocked so that the CPT must be
unloaded before opening the primary device. The CPT may be mounted on the rollout or tiltout, or
the rollout or tiltout may contain only the fuses for a stationary mounted CPT. Larger CPT's must
be switched with some other mechanism, such as a load break disconnect switch.
Capacitors: Rollouts or tiltouts must not be used to switch capacitors.

Any other application should be reviewed by Powell's engineering department.

Technical Director

01.4TB.022

Short Circuit Currents Crest, rms Symmetrical and rms
Asymmetrical

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 4, 1991

The figure below shows a typical short circuit current wave form and defines the various component parts
of this wave. At the moment of initiation of a short circuit the ac current wave, which is normally
symmetrical about the zero axis BX is offset by some value, creating a waveform which is symmetrical
about another axis, CC'. The degree of asymmetry is a function of several variables, including the
parameters of the power system up to the point of the short circuit and the point on the ac wave at which
the short circuit was initiated. In a 3-phase circuit, there is usually one phase which is offset significantly
more than the other two phases.

It is convenient to analyze this asymmetrical waveform as consisting of a symmetrical ac wave
superimposed on a dc current. CC' represents the dc current, and the value of that current at any instant
is represented by the ordinate of CC'. The dc component of the current normally decays rapidly, and
reaches an insignificant value within 0.1 s in most power systems. The rate of decay is a function of the
system parameters. When the initial value of the dc current is equal to the initial peak value of the ac
current, the resulting waveform is said to be fully offset, or to have a 100% dc component. It is possible,
in some power systems, to have an offset in excess of 100%, which may result in a waveform that has
no current zeros for one or more cycles of the ac power frequency.

The ac component of the short circuit current will also decay, at a rate dependant on the system
parameters. In general, the closer the fault is to generators or other large rotating machinery, the faster
the decay will be.

01.4TB.022

Short Circuit Currents Crest, rms Symmetrical and rms
Asymmetrical

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
In the figure, I
MC
is the crest, or peak, value of the short circuit current. It is the maximum instantaneous
current in the major loop of the first cycle of short circuit current.

The rms symmetrical value of the short circuit current at any instant, such as EE', is the rms value of the
ac portion of the current wave. Its value is equal to , and it is shown graphically by the distance
from CC' to DD'. The rms asymmetrical value of the short circuit current is the rms value of the combined
ac and dc waves, and it is calculated by the formula:

Technical Director

01.4TB.023

Using Design Tests to Qualify Several Ratings of Equipment

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 5, 1991

The many variations in construction and ratings encountered in the typical switchgear or motor control
product line make the planning of design and conformance test programs quite complex at times. Of
course, it is possible to run every test on every possible rating of equipment but such an extensive
program is very expensive and is seldom required to fully document the performance of a product line.

The ANSI standards for switchgear recognize this complexity and provide for the qualification of a piece
of equipment for all lower ratings provided test results show it to be qualified for the highest rating for
which it is used. Some of the conformance test standards in the ANSI C37.50 series discuss the
principles of testing to qualify multiple ratings. These standards also give guidance in the grouping of
equipment ratings for testing.

A typical example of qualifying multiple ratings by a single test is the bus structure used in PowlVac

metal-clad switchgear. This bus structure is the same for all voltage and short circuit ratings, varying only
for continuous current ratings. To demonstrate the momentary and short-time current ratings of this bus
structure, tests are performed on the bus with the lowest continuous current rating, 1200 A, which uses
the smallest, weakest bars of any continuous current rating of PowlVac
bus. The tests are performed at

the maximum momentary current, 132 kA crest, and the maximum short-time current, 49 kA rms,
required for any rating of PowlVac
switchgear. It is fairly obvious that passing these tests qualifies the

1200 A bus for this rating and for all lower momentary and short-time current ratings. What may not be
quite so obvious is that successful tests on the 1200 A bus also qualify higher continuous current ratings,
such as 2000 A and 3000 A. These higher bus ratings are covered because they use larger bus bars,
which are mechanically stronger and which have greater thermal capacity than the bus bars used in the
1200 A bus.

The grouping of ratings and the selection of which rating to test requires a thorough knowledge not only
of the standards but also of the particular product line being tested. The grouping of ratings may differ for
different tests. It also may differ for different products, or different manufacturers offerings in the same
product line. The example given in the previous paragraph is true for PowlVac
switchgear, but may not

necessarily be true for other manufacturers' similar products.

Although Powell and many other manufacturers have used these principles in performing their design
tests for many years, not everyone in the industry understands the concept. To aid in this understanding,
all future Powell test reports will document the additional ratings covered by any test.

Technical Director

01.4TB.024

Sizing Bus Bars in Switchgear and Motor Control

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 7, 1992

We occasionally get questions about how we select the size of bus bar for various continuous current
ratings in Powell equipments. The answer is that we use temperature rise as the basic criterion. All of the
ANSI, IEEE and NEMA standards for switchgear and motor control have requirements for the maximum
operating temperature of various parts of the equipment. For bus bars, the requirement is generally for a
temperature rise of no more than 65C, although this may vary for different classes of equipment. These
requirements are designed to prevent overheating the insulation supporting and enclosing the bus bars,
since excessive temperature shortens the life of the insulation.

A number of factors affect the temperature rise of bus bars. Some of the major ones are:
Size and material (copper or aluminum) of the bus bar.
Whether the bar is insulated. Surprisingly, a bus bar covered with insulation generally runs cooler
than an equivalent bare bus bar, because the usually darker color of the insulating material is a
better radiator of heat than the shiny surface of a bare bus bar.
Size and material (magnetic or non-magnetic) of the enclosure around the bus.
Flow of ventilating air past the bus bars or the bus enclosure.
Proximity of other conductors and other heat-producing devices.

The complex interaction of these and other factors makes it nearly impossible to calculate temperature
rise, and leads to the requirement in all applicable standards for continuous current tests to determine
the temperature rise of a bus design.

Specifications will sometimes call for bus sized by current density, a favorite requirement being 1000 A
per square inch for copper bus. This may be a good way to choose bus sizes for the mythical "single
conductor in free air", but it isn't a satisfactory way to design buswork in practical equipments. Consider
the following chart, based on bus sizes used in our PowlVac
metal-clad switchgear:
Switchgear Bus Rating 1200 A 2000 A 3000 A
Number of bus bars per phase 1 1 2
Size of bus bar, inches 1/4 x 4 1/2 x 6 1/2 x 6
Cross section area of bus, square inches 1 3 6
Current density, amps per square inch 1200 667 500
Maximum temperature rise, from test data 60C 59.7C 59.5C

01.4TB.024

Sizing Bus Bars in Switchgear and Motor Control

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

The last line of the chart shows that the temperature rises of the three bus ratings are almost identical in
spite of the 2.4:1 ratio of the current densities.

Technical Director

01.4TB.025

Application of Metal-Enclosed Switchgear at High Altitude

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 11, 1992

Both low- and medium-voltage metal-enclosed switchgear and the circuit breakers used in these
equipments depend on air for both cooling and insulation. At high altitudes, the less dense air is less
efficient both as in insulator and as a heat transfer medium. Because of this, the ANSI standards require
derating when these equipments are used at high altitudes. The following tables show the altitude
correction factors taken from the ANSI standards.
Low Voltage Switchgear and Breakers
Altitude (ft)* Voltage Current
6600 (2000 m) (and below) 1.00 1.00
8500 (2600 m) 0.95 0.99
13,000 (3900 m) 0.80 0.96
Medium Voltage Switchgear and Breakers
Altitude (ft)* Voltage Current
3300 (1000 m) (and below) 1.00 1.00
5000 (1500 m) 0.95 0.99
10,000 (3000 m) 0.80 0.96
* Intermediate values may be obtained by interpolation.

You will notice that there are different altitudes given for low voltage and medium voltage. I have never
been able to get a reasonable answer as to why this is true, and I understand that the committee
responsible for the standards is reviewing these values with the idea of reconciling them.

In all cases, the current correction factor is applied to the continuous current rating of the switchgear and
the circuit breakers. This does not usually present a problem, as we seldom design a system with load
currents over 95% of the equipment rating. The current derating does not apply to interrupting current or
any of the other high-current ratings of the breakers.

01.4TB.025

Application of Metal-Enclosed Switchgear at High Altitude

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
For low voltage equipments, the voltage correction factor applies to the low frequency withstand (hipot)
rating of both the breaker and the equipment. It also applies to the rated maximum voltage of the circuit
breaker. When derating the rated maximum voltage, the short circuit rating of the circuit breaker cannot
exceed the rating at the voltage before derating. For instance, if a breaker is used on a 480 V system, as
most of those in Powell equipment are, with a 0.95 rating factor the short circuit rating at 480 V may be
used, since the rated maximum voltage for that system nominal voltage is 508 V, and 0.95 x 508 is 482.6
V, slightly above the 480 V service voltage. However, if this same system required a 0.80 rating factor,
the breaker short circuit rating at 600 V must be used, since 0.80 x 508 is only 406 V, less than the
service voltage, but 0.80 x 635 is 508 V, comfortably above the service voltage.

For medium voltage equipments, the voltage correction factor applies to the low frequency withstand
(hipot) rating and the impulse withstand (BIL) rating of both the breaker and the equipment. It also
applies to the rated maximum voltage of the circuit breaker unless a sealed interrupter, such as a
vacuum interrupter, is used. The use of surge arresters to protect the equipment should be considered
for all such high altitude installations.

Technical Director

01.4TB.026

Voltage Ratings of Surge Arresters

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 13, 1992

Surge arresters (formerly known as lightning arresters) are applied to electrical power distribution
systems to protect the equipment and the circuits from damaging overvoltages caused by lightning or
other surges. It is important that surge arresters of the correct voltage rating be used. The proper voltage
rating depends on the system line-to-line voltage, the method of system grounding, and the type of surge
arrester used.

Older designs of surge arresters generally consist of silicon carbide resistor blocks in series with air
gaps. These arresters carry no current in the normal state. Each arrester model has a single voltage
rating. For solidly (effectively) grounded systems, the next higher arrester rating above the system line-
to-neutral voltage is used. For resistance grounded or ungrounded systems, a ground fault on one phase
can raise the other two phases to line-to-line voltage above ground, so the next higher rating above the
system line-to-line voltage is used. Except for a few special conditions, application seems quite simple.

About a decade ago, the metal oxide surge arrester was introduced to the industry. It consists of a
number of blocks of a variable resistance material, usually zinc oxide, with no gaps. It does carry some
slight current at all times. It has many advantages as a surge protector, but it is somewhat more
complicated to apply correctly. Instead of one voltage rating, it has three: a nominal voltage, a maximum
continuous operating voltage, and a one-second temporary overvoltage capability. Although there is a
slight variation with the nominal rating, the maximum continuous operating voltage is about 85% of the
nominal rating and the one-second temporary overvoltage capability is about 120% of the nominal rating.
For times other than one second, the temporary overvoltage capability is established by curves supplied
by the surge arrester vendor. Care must be taken to avoid overstressing the arrester.

As an example, let's consider a 13.8 kV system. For a solidly grounded system, the continuous operating
voltage is 13,800 divided by the square root of 3, or 7970 V. This is above the MCOV of 7,650 V for an
arrester rated 9 kV. Depending on the value and expected duration of system overvoltages, it may be
necessary to use a 10 kV arrester with an MCOV of 8.4 kV or a 12 kV arrester with an MCOV of 10.2 kV.
For an ungrounded 13.8 kV system, the 12.7 kV MCOV of a 15 kV arrester is not adequate. It is
necessary to use an 18 kV arrester with an MCOV of 15.3 kV. Finally, for a resistance-grounded 13.8 kV
system, the choice will be between arresters rated 12 kV, 15 kV and 18 kV, depending on the time
needed to relay ground faults off the system.

Technical Director

01.4TB.027

Testing of Switchgear and Motor Control Equipment

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 14, 1992

Although each particular product line is governed by its own industry standards, switchgear and motor
control equipment of the types built by Powell are generally subject to three major categories of tests. As
defined in ANSI/IEEE C37.20.2-1987 for Metal-Clad and Station-Type Cubicle Switchgear, these
categories are:

Design Tests: Tests made by the manufacturer to determine the adequacy of the design of a particular
type, style or model of equipment or its component parts to meet its assigned ratings and to operate
satisfactorily under normal service conditions or under special service conditions if specified, and may be
used to demonstrate compliance with the applicable standards of the industry.

Production Tests: Tests made for quality control by the manufacturer on every device or on
representative samples, or on parts, or materials required to verify during production that the product
meets the design specifications and applicable standards.

Conformance Tests: Conformance tests demonstrate compliance with the applicable standards. The
test specimen is normally subjected to all planned production tests prior to the initiation of the
conformance test program.

Typical design tests for equipment and circuit breakers will include continuous current (heat runs),
momentary and short time current, low-frequency withstand (hipot), impulse withstand (BIL) for medium-
voltage equipment, and mechanical tests to demonstrate the effectiveness of interlocks. In addition,
circuit breakers are subjected to a series of interrupting tests to demonstrate their ability to interrupt
currents of various magnitudes, operational life tests, and several types of timing tests. Many of these
tests are somewhat destructive, and therefore they are run on manufacturer's prototypes, not on
production equipment which is supplied to customers.

Conformance tests generally include certain of the design tests, chosen to demonstrate compliance with
the standards. These tests are frequently used for third-party certification of a design.

Production tests include hipot to demonstrate insulation integrity and mechanical and control circuit tests
to demonstrate proper operation. In addition, circuit breakers receive timing tests to show proper closing
and opening speed. Records of these tests, which Powell furnishes to customers on request, can be
used as baseline data for future maintenance programs.

01.4TB.027

Testing of Switchgear and Motor Control Equipment

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Each type of test, and each test within a given type, has a particular part to play in the overall process of
producing quality equipment properly rated for a user's needs. No single test demonstrates the proper
design and operation of switchgear or motor control equipment. It takes a combination of tests to do the
job properly.

Technical Director

01.4TB.028

Short Circuit Current Levels Used to Test Various Types of
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 25, 1992

When applying interrupters of various types, it is important that we understand the meaning of the
interrupting rating given to these devices. Consider, for instance, the methods of making interrupting
tests on various types of circuit breakers. As the breakers get smaller and less costly, the test methods in
the industry standards generally get less demanding.

When testing the interrupting capability of a high-voltage (over 1000V) circuit breaker, the current
measured is the actual fault current flowing through the circuit breaker at the moment of the interruption.
To rate a breaker of this class as a 25kA interrupter, it must actually interrupt 25kA. Momentary and
short-time current requirements of the switchgear are also based on actual current flowing during the
test. The reference standards are ANSI/IEEE C37.04, C37.06 and C37.09 for the circuit breakers and
C37.20.2 for the switchgear.

For low voltage circuit breakers, this requirement changes to rating by prospective current. The test
terminals of the laboratory source are short-circuited, as indicated by point A in Figure 1, and the
required current flow is established. That short circuit is then removed and the equipment to be tested is
connected to the test source. A short circuit is then applied to the equipment and the test made. The
location of the short varies with the type of circuit breaker or equipment being tested:

Figure 1: Fault Locations for Testing Low Voltage Equipments
(A) Low Voltage Power Circuit Breakers
(B) Molded Case Circuit Breakers
(C) Low Voltage Motor Control Centers

01.4TB.028

Short Circuit Current Levels Used to Test Various Types of
Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
For a low voltage power circuit breaker, the fault is placed at the load terminals of the breaker, at
point B in Figure 1. The reference standards are ANSI/IEEE C37.13 for the circuit breakers and
C37.20.1 for the switchgear equipment.
For a molded case circuit breaker, the fault is also placed at the load terminals of the circuit
breaker, at point C in Figure 1. However, 4 feet of appropriately-sized conductor may be included
between the test station terminals and the line terminals of the circuit breaker under test. The
reference standard is UL 489.
For a typical combination motor starter unit in a motor control center, the fault is placed at the end
of 4 feet of appropriately-sized conductor connected to the load terminals of the starter unit, at
point D in Figure 1. The reference standard is UL 845.

At each step of this chain, impedance is added to the test circuit, reducing the actual fault current the
circuit breaker is required to interrupt. Several papers presented at recent IEEE conferences have raised
questions about the adequacy of equipment certified to some of these test standards to interrupt all
possible faults downstream of the circuit breaker. At least two IEEE subcommittees are discussing this
matter.

Technical Director

01.4TB.029

Interchangeability of Drawout Circuit Breakers in
Switchgear Assemblies

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 28, 1992

One of the key features of switchgear assemblies using drawout circuit breakers is the interchangeability
of circuit breakers within an assembly. This feature allows users to make use of spare circuit breakers to
replace circuit breakers which must be taken out of service for maintenance, minimizing down time when
a circuit breaker problem occurs.

The ANSI standard for Metal Enclosed Low Voltage Power Circuit Breaker Switchgear, ANSI/IEEE
C37.20.1, addresses interchangeability in 6.11.4. This section requires that "All removable elements of
the same type and rating on a given assembly shall be physically interchangeable in the corresponding
stationary housings. This need not include electrical interchangeability of electrical control circuits."
Switchgear of this type and the circuit breakers used in it typically have mechanical interference
mechanisms for breakers of the same physical size but of different ratings. These mechanisms typically
prevent interchanging breakers if either the frame size (maximum continuous current rating) or the
interrupting rating differ. Trip device characteristics and ratings and electrical accessories available on
this class of circuit breaker are so numerous and changeable that no attempt is made to prevent
interchangeability of breakers with differences in these features.

The ANSI standard for Metal Clad Switchgear, ANSI/IEEE C37.20.2, addresses interchangeability in
6.2.5. This section requires that "All removable elements of the same type and rating on a given
assembly shall be physically and electrically interchangeable. Removable elements not of the same type
and rating shall not be interchangeable." Since the breakers used in this class of switchgear are not
provided with variable trip devices or very many optional electrical features, this is seldom a problem.

Occasionally, however, a user desires to have some electrical accessory on some but not all breakers of
a given rating in a particular assembly. Most commonly, this is an undervoltage device, which may be
required on motor feeder breakers but not on other breakers of the same rating. When this occurs, the
easiest way to solve the problem is to furnish the required modification on all breakers of that rating in
the assembly. If this is not acceptable to the user, it may be necessary to make specific modifications to
the control circuitry of the breaker with the accessory to prevent breaker interchangeability.

01.4TB.029

Interchangeability of Drawout Circuit Breakers in
Switchgear Assemblies

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
As part of the standard design of our PowlVac
circuit breakers, we provide interference mechanisms

which prevent a breaker with a lower rating from being used in a cell with a higher rating, but allow a
higher-rated breaker to be used in a lower-rated cell. While this feature is not strictly in accordance with
the ANSI requirements, it allows users to minimize the number of spare circuit breakers required to
replace all breakers in the assembly without using any breaker in a cell where it would not meet the
needs of that circuit.

Technical Director

01.4TB.030

Static Relays and Meters

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 16, 1992

In recent years, we have seen a decided trend toward the use of metering and protective devices using
solid-state components, particularly microprocessor-based digital devices. These devices are often used
instead of the conventional electro-mechanical relays or analog meters. While we have no formal
statistics, looking at the equipment manufactured by Powell I estimate that static devices are used for
about 40-50% of the relays and meters we provide to our customers.

Some of the reasons for this shift to static devices include:
In general, static devices are more accurate and more repeatable than the equivalent electro-
mechanical devices.
A single static device can perform the functions of many electro-mechanical devices.
Equivalent functions can often be obtained at lower cost, particularly if a multi-function device is
used.
Some functions or operating characteristics which are not possible with available electro-
mechanical devices can be done with static devices.
Static devices can be provided with communications capability which is not available in electro-
mechanical devices.
Static devices can be made highly resistant to corrosive or dirty atmospheres.

However, not every engineer is happy with the idea of relying on static devices for protection functions.
Some of the reasons are:
Possibility of total failure of the protective system due to failure of one component on the critical
path, such as a common power supply.
Long-term familiarity and satisfactory experience with electro-mechanical devices.
Lack of service capabilities for static devices.
Existing company standards.
Concerns about possible failure in adverse environments, both physical and electrical.

01.4TB.030

Static Relays and Meters

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
As time goes on, many of the shortcomings of static devices available 10 or more years ago have been
overcome by further development of solid-state components and better packaging. New static products
are arriving in the marketplace with great regularity, and their capabilities are constantly being expanded.
On the other hand, little if any development work is being done on electro-mechanical relays and meters.
Long-term, I expect that the balance between static and electro-mechanical devices will shift to 90-95%
static. I would expect this to happen in about the next 10 years, but don't hold me to the timing.

Technical Director

Effects of Harmonics on Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 18, 1992

Harmonics in power systems is the current hot topic in power systems engineering. The technical
programs of two recent IEEE conferences I attended were full of papers on the subject. The trade press
carries article after article on harmonics. The recent survey of the readers of these Technical Briefs
mentioned harmonics more than any other topic as a suggestion for a subject to be covered. However,
very little if any of this flood of information concerns the effects on switchgear. This PTB will try to fill that
gap.

First, I have not heard any reports of adverse effects of harmonics on switchgear itself. Switchgear does
not appear to be subject to the serious additional heating effects which harmonics can cause in
transformers or rotating machinery. However, there are a few points which bear watching.
A highly distorted voltage wave can have a peak value much greater than the peak value of a
sine wave of the same rms value. The repeated application of a higher peak value may result in
premature failure of the insulating systems in the switchgear.
Since the effective resistance of a conductor goes up as frequency rises, a current wave rich in
harmonics may cause greater heating in the switchgear power conductors than a sine wave of
the same rms value. However, since we seldom load switchgear to its maximum continuous
current rating, this effect will probably not be a problem.
Circuit breakers interrupt current flow at a current zero. A badly distorted current wave may
contain current zeroes at locations other than the normal zero of the fundamental sine wave. If
present, these spurious current zeroes could cause premature interruption and restrike during a
circuit breaker opening operating.
The effect of harmonics on the operation of protective and metering systems is not fully
documented. Thermal devices, such as fuses or thermal elements in circuit breakers, are
inherently rms sensing, as are many modern solid-state protective packages, but the heating
effect of a harmonic-rich current on the protected load may require a different level of protection.
Other relays may be designed to operate on certain selected frequencies and may thus operate
incorrectly in the presence of a significant level of harmonics.
Fortunately, high fault currents are generally not severely distorted, as the limiting impedance is
that of the power system, and a level of harmonic current which may be quite significant when
compared to load current is much less significant when compared to fault current.

Technical Director

Replacing Older Medium Voltage Circuit Breakers with
Vacuum Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 19, 1992

Many older installations of metal-clad switchgear are candidates for modernizing or upgrading. The
interrupting duty may have grown past the interrupting capacity of the existing circuit breakers,
maintenance costs may be getting out of hand, or circuit breaker parts may no longer be available.

Modernization may be accomplished by either of two methods, conversion of the existing equipment or
replacement of the circuit breaker with a new, modern vacuum breaker. In some cases, modernization
may involve conversion of the switchgear equipment and replacement of the circuit breaker.
Modernization may also involve an increase in one or more ratings of the switchgear and/or circuit
breaker. Powell offers a series of PowlVac
circuit breakers, both conversions and replacement

breakers, to aid in modernization.

In any case, users should insist that modernization be done in accordance with applicable industry
standards to insure that the modernized switchgear meets the required ratings. For replacement circuit
breakers, this is fairly simple. The breakers should be tested in accordance with ANSI/IEEE C37.09-1979
and ANSI/IEEE C37.20.2-1987, just like any other new circuit breaker that is used in metal-clad
switchgear. The only caution is that certain tests, including continuous current, momentary current, and
BIL, should be performed in a switchgear cell of the basic design in which the breaker will be used. All of
these tests involve interaction between the cell and the circuit breaker. For instance, we have performed
full wave impulse (BIL) tests where both the breaker alone and the cell alone passed the test, but the
combination would not pass, requiring additional work to the breaker to achieve the desired result.

A new ANSI standard, ANSI/IEEE C37.59-1991, IEEE Standard Requirements for Conversion of Power
Switchgear Assemblies, covers conversion of both switchgear equipments and power circuit breakers.
This standard specifies the design and testing requirements for conversion, including uprating, of existing
equipment. Requirements for new documentation, including nameplates and instructions, are included.
Any user considering converting low-voltage or medium-voltage switchgear equipment should review this
standard and insist that vendors conform to it.

Powell, through our Powell Apparatus Service Division, offers both conversions and replacement
breakers meeting these standards.

Technical Director

Partial Differential Relaying

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 16, 1992

"Partial differential" relaying is a form of overcurrent relaying frequently used to detect main bus
overcurrent faults and to back up feeder overcurrent relaying. The basic circuit is shown in the one-line
diagram. Note that this is a double-ended substation, with two main breakers and a tie breaker. The
partial differential relaying concept cannot be used on a straight radial distribution system.

True bus differential relaying compares all currents entering and leaving a switchgear bus. Within the
limits of the accuracy of the CT's and the relays, true bus differential relaying will detect all faults on the
protected bus. Since all currents are taken into account, the relays can be very fast. However, bus
differential relaying provides no backup to the feeder overcurrent relaying, so additional overcurrent
relays are required on main and tie breakers to provide this backup function. Also, high speed bus
differential relaying can be quite expensive, and many switchgear users do not feel that it is economically
justified.

Partial differential relaying sums the currents entering or leaving a switchgear bus through main and tie
breakers. If a fault exists on the protected bus, the currents will add in the relays, but if fault current is
flowing through the bus to a fault on another bus, the currents will subtract and the relays will not
respond. If the fault is on a feeder, the partial differential relays will act as backup to the feeder
overcurrent relays.

Partial Differential Relaying

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Similar protection can be obtained by using separate overcurrent relays on each main and tie circuit
breaker. However, proper coordination of the overcurrent protection requires that the tie breaker relays
coordinate with the feeder relays and that the main breaker relays coordinate with the tie breaker relays,
for a total of three steps of relaying at this bus. Using the partial differential circuit, however, eliminates
one step of coordination, since the same relays serve both the main and the tie breakers without
compromising coordination. This reduces the time delay required for the main breaker relays and
improves the chances of getting good coordination with upstream relays, which are often on the utility
system serving the substation. This improved coordination is the principal benefit of partial differential
relaying.

Technical Director

Polarity Markings on Instrument Transformers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 17, 1992

All instrument transformers are provided with polarity marks on certain primary and secondary terminals.
In ANSI/IEEE C57.13, American National Standard Requirements for Instrument Transformers, polarity
is defined as follows:
Polarity. The designation of the relative instantaneous directions of the currents entering
the primary terminals and leaving the secondary terminals during most of each half cycle.

NOTE: Primary and secondary terminals are said to have the same polarity when, at a
given instant during most of each half cycle, the current enters the identified, similarly
marked primary terminal and leaves the identified, similarly marked secondary terminal in
the same direction as though the two terminals formed a continuous circuit.
This definition relates only to the relative instantaneous direction of current flow, not an absolute
direction. Since the direction of current flow reverses 120 times per second in a 60 Hz circuit, no mark
can identify the absolute direction of current flow.

Different kinds of switchgear mount instrument transformers, particularly current transformers, in different
ways. Outdoor circuit breakers have bushing CT's mounted with the polarity mark toward the line end of
the primary bushing and the non-polarity end of the winding toward the breaker contacts. This
arrangement is required by the ANSI and NEMA standards for that class of circuit breaker. Most circuits
shown in instruction literature by relay and meter manufacturers are based on this arrangement of CT's.

In metal-enclosed switchgear with drawout circuit breakers, however, the CT's are part of the equipment,
not the circuit breaker. In many cases it is more convenient to mount the CT's with the polarity mark
pointing toward the breaker. In the typical drawout switchgear enclosure, mounting the CT with polarity
mark toward the breaker also means that the CT nameplate is visible, which is desirable.

From time to time, some users have expressed concerns that reversing the polarity marks from the
arrangement shown in the relay or meter instructions would lead to improper operation of the device.
This is not true. Relays or meters will work properly regardless of the direction of the polarity marks on
the CT's if the connections are made properly. Great care must be taken in making these connections,
especially for such things as differential relays on delta-wye transformers, but proper operation is not a
function of which way the polarity marks point.

Technical Director

Setting of Targets on Electro-Mechanical Protective Relays

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

January 14, 1993

Almost all protective relays used in switchgear are equipped with targets to indicate which protective
function caused the circuit breaker to trip. For the most common varieties of electro-mechanical relays,
these targets take the form of small auxiliary relays which are actuated by the current flowing through the
relay contacts and the circuit breaker trip coil. This auxiliary relay provides a visual, manually-reset,
indication of operation. In addition, it often includes contacts which bypass the main contacts of the relay,
relieving these contacts of the duty of carrying trip current.

Many of these target units have tapped coils, with the choice of two pickup currents, most commonly 0.2
A and 2.0 A. Proper setting of these target coils is often neglected. The current setting must be low
enough to insure actuation under all fault conditions, but not so low that the voltage drop across the
target coil is excessive. This is not always an easy choice. Some of the reasons for this dilemma are:
Trip currents of modern medium voltage circuit breakers may be much lower than those of older
breakers. For instance, at 125 V dc, typical trip currents are now in the 3A range, with some
breakers drawing only a little over 1 A.
Trip times are faster than in the past. Figure 1 is the trace of the trip current of a typical PowlVac

circuit breaker. You will see that the duration of the trip current is about 41 or 42 ms, or 2
cycles on a 60 Hz basis. The target relay takes about 1 cycle to pick up.

Figure 1. Typical Circuit Breaker Trip Current
Trip current rises slowly, barely reaching its nominal rating before being cut off, as seen in Figure
1.
Some faults may cause the closure of two or more relay contacts at the same time, paralleling the
target coils of these relays and reducing the current through each one. One manufacturer
recommends expecting two relays to close simultaneously on time delay relays and three on
instantaneous relays with target coils. (Note: The instantaneous elements on most 50/51 relays
have mechanical targets, and are not a problem.) Thus, the selected tap on the target coil should
operate properly on no more than half of the rated trip current of the breaker.

Setting of Targets on Electro-Mechanical Protective Relays

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
For a list of trip currents and recommended relay target settings for PowlVac
circuit breakers see

Table 1.
Table 1 - Trip Currents and Suggested Relay Target Taps for PowlVac
Circuit Breakers
Trip Voltage
PowlVac
-0,-2,-5
PV-26
PowlVac
-3,-4,-6
Trip Current Target Tap Trip Current Target Tap
24 V dc 16.9 A 2.0 A 16.9 A 2.0 A
48 V dc 10.2 A 2.0 A 3.5 A 0.2 A
125 V dc 3.2 A 0.2 A 1.3 A 0.2 A
250 V dc 1.7 A 0.2 A 0.7 A 0.2 A
120 V ac 3.4 A 0.2 A 3.4 A 0.2 A
240 V ac 1.7 A 0.2 A * *
Capacitor Trip 2.2 A 0.2 A 0.6 A 0.2 A
Consult factory if this rating is required.

Technical Director

Epoxy Bus Bar Insulation

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 4, 1993

Powell has recently installed a new process line to coat bus bars with an epoxy insulating material.
Currently, this insulation is being used for bus in metal-clad switchgear equipments rated above 15 kV,
and for other jobs with special requirements. As more production capability comes on line, its use will be
extended to additional equipments. The process consists of preheating the copper bus bar, dipping it in a
liquid epoxy mixture, removing the coating from areas where it isn't wanted (contact surfaces, etc.), and
curing the coated bar at a high temperature. The coated bars have passed the tests required by
ANSI/IEEE C37.20.2-1987, including the 5.2.1.3 Test for Bus-Bar Insulation and the 5.2.7 Flame-
Resistance Tests for Applied Insulation.

While this process is new to Powell, similar materials and processes have been in use in the industry for
20 years or more. It is also similar to the fluid bed application of epoxy to bus bars, which has been used
on selected Powell equipments for the past several years. We chose the liquid dip process over the
fluidized bed process both because the final coating is more rugged and because the process itself is
less subject to interference from the ambient conditions of the factory floor.

Some of the advantages of the epoxy dip process are:
The insulation is extremely rugged and has excellent dielectric properties.
Conductors of any size and shape can be coated with equal ease. Using extruded rigid tubing
limits the choice of conductor cross-sections to those for which the extruded tubing is available,
which may not be the optimum size from a current-carrying or electric field standpoint.
Already-bent bars can be coated, eliminating the need for tape or boots at bends.
Unlike some heat-shrink tubings, the insulation conforms to the bar at all angles and bends. Heat-
shrink tubing may pull away from the bar at the inside of a bend.
Since the coating conforms to the surface of the bus, there is no possibility of setting up a partial-
discharge cell between the surface of the bus and the insulation. This is especially important at
voltages above 15 kV.

We believe that this new insulating process is another step in our process of continuous improvement of
Powell switchgear.

Technical Director

Testing for Loss of Vacuum in Vacuum Interrupters

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 5, 1993

In the 30 years or so that they have been used in circuit breakers, vacuum interrupters have proved to be
extremely rugged, reliable devices. However, like any man-made device, they can fail. A frequently-
asked question is "How can I tell if my vacuum interrupter has lost vacuum?"

Modern vacuum interrupters are evacuated to a pressure on the order of 10
-7
Torr. A Torr is the pressure
exerted by 1 mm of mercury, or 1/760 of a standard atmosphere. 10
-7
Torr is approximately equal to the
pressure of the atmosphere on the surface of the moon. Testing by manufacturers of vacuum interrupters
shows that the vacuum interrupter will still interrupt its rated interrupting current at a pressure <= 10
-3

Torr, but at pressure above 10
-3
Torr is interrupting capability falls off. Thus, a successful loss of vacuum
test must detect the difference between pressures above and below 10
-3
Torr.

An ideal loss of vacuum test would be an on-line test that constantly monitored the pressure within the
vacuum interrupter and alarmed when this critical pressure was approached. Unfortunately, no such
device is presently commercially available.

The present state of the art leads Powell to recommend the use of a high voltage test across the
contacts of an open vacuum interrupter. This test should be at least 25 kV, 50/60 Hz, or the equivalent dc
voltage, 35 kV
(1)
. A breakdown on this test indicates a loss of vacuum in the interrupter. If dc is used in
this test, it must be supplied from a full-wave rectifier. The use of a half-wave in the power supply can
lead to applying excessive dc voltage to the vacuum interrupter. In all cases, the manufacturer's
instruction book for the circuit breaker should be consulted for proper procedure and cautions before
making the test.

At least one switchgear manufacturer is stating that loss of vacuum can be detected by disconnecting the
movable contact of the vacuum interrupter from the mechanism and manually pulling on the movable
contact. Atmospheric pressure (760 Torr) on the exposed side of the movable contact pushes the contact
closed with a force proportional to the area of the contact. This force is about 40 pounds for a typical
vacuum interrupter used in our PowlVac
( circuit breakers. If there is a complete loss of vacuum, this

40-pound force disappears, and a pull test will certainly be effective. However, partial loss of vacuum in
the 10
-2
Torr pressure range puts the vacuum interrupter at risk of not performing properly, but decreases
the 40-pound force by only a fraction of an ounce. This minor change in contact loading would hardly be
detectable by a good force gauge, much less by feel. Powell therefore does not recommend this pull test
as a satisfactory loss of vacuum test.

(1) revised 6-4-93

Technical Director

01.4TB.038

Using Switchgear at Frequencies Other Than 60Hz

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 21, 1993

All ANSI circuit breaker and switchgear equipment standards specify a rated frequency of 60 Hz, and all
Powell switchgear is designed for and tested at 60 Hz. The nameplates of PowlVac
circuit breakers
carry a rated frequency of 60 Hz.

However, many applications of Powell switchgear are made at other frequencies, particularly 50 Hz, and
users may question the suitability of the circuit breakers and equipment at other frequency. The following
paragraphs discuss the effect of frequency on various major characteristics of switchgear.

BIL: Since the BIL rating is the ability to withstand a dc impulse, power system frequency has no effect
on the BIL rating of switchgear equipment.

Power Frequency Withstand: Although this rating is best demonstrated by a test at rated frequency,
ANSI/IEEE C37.09 allows the test to be made at rated frequency 20%. This allows 50 Hz tests to
establish a 60 Hz rating, and vice versa. Since the voltage level is the same, regardless of frequency,
and the rate-of-rise of the voltage increases with frequency, the higher frequency test is slightly more
severe than the lower frequency test. Thus, an equipment which is tested at 60 Hz is satisfactory at any
lower frequency.

Continuous Current: The ac resistance of a conductor, and thus the heat loss at a given current,
increases with frequency. The ANSI standards require continuous current test to be made at a frequency
no lower than the rated frequency. The standards actually allow the assigning of a higher continuous
current rating at very low frequencies, such as 25 Hz.

Momentary, Short Time and Interrupting Currents: Again, ANSI/IEEE C37.09 allows these tests to be
made at rated frequency (20%, so tests at either 50 Hz or 60 Hz can cover both frequencies. For lower
frequencies, it may be necessary to derate the interrupting capacity of vacuum interrupters. One of
Powell's vacuum interrupter suppliers has tested at various frequencies and suggests the following
derating formula:

01.4TB.038

Using Switchgear at Frequencies Other Than 60Hz

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Where

=power system frequency, 16 2/3Hz <=f <50 Hz

=short circuit current at frequency f

=short circuit current at 60 Hz

Summarizing, switchgear equipment and power circuit breakers rated 60 Hz may be applied at 50 Hz
without changing ratings. At lower frequencies, some derating of interrupting capacity may be required,
and some increase of continuous current rating may be possible. Of course, devices in the switchgear,
such as relays, instruments and instrument transformers, must be checked for application at the power
system frequency.

Technical Director

Motor Branch Fault Short-Circuit Protection

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 22, 1993

I have received several questions about ratings and settings of short-circuit protective devices used in
motor branch circuits. Generally, the devices in question are instantaneous-trip molded case circuit
breakers of the motor circuit protector type, but some of the questions have related to low voltage power
circuit breakers. The problem is that some motors, will not start satisfactorily with these devices set at the
maximum setting allowed under 430-52 of the 1993 National Electrical Code.

High efficiency motors typically have lower full load currents than lower efficiency motors of the same
rating, but do not necessarily have reduced starting currents. They therefore have starting currents that
are higher multiples of full load current than those of the lower efficiency motors. Since all of the
protective device ratings in the NEC are expressed as percentages of full load current of the motor, a
high efficiency motor is more likely to have starting difficulties due to current inrush.

430-52 allows an instantaneous trip circuit breaker (MCP) to be set at up to 1300% of the motor full load
current. Although the NEC does not specifically say so, this 1300% is generally interpreted as being the
maximum setting of a separately adjustable instantaneous element on a power circuit breaker or a
molded case circuit breaker having that feature.

So what can be done to take care of the problem? As of now, I know of no good answer, but there are
several possibilities:
Use a thermal-magnetic circuit breaker rather than an MCP. NEC 430-52, Exception No. 2C,
allows an inverse time circuit breaker rated up to 400% of the motor full load current to be used
for motors with less than 100A full load current. Small circuit breakers, up to 150A rating, have
fixed instantaneous settings, with minimum pickup varying from 7.5 to 12 times rated current, so
the instantaneous pickup can be from 30 to 48 time the motor full load current, rather than the 13
times required for an MCP. You have degraded the protection of the circuit, but you have met the
Code.
A fine print note in NEC 430-52 allows an instantaneous trip circuit breaker to include a damping
means to accommodate a transient motor inrush current without nuisance tripping of the circuit
breaker. If you can find one of these, you can use it, but so far as we know, no such device is
commercially available.
Use a fused switch instead of a circuit breaker to supply your motor circuit, and find a fuse that
meets the requirements of 430-52 and still allows you to start the motor.
Ignore the NEC. Not a recommended move, but I understand that it is being done.

Technical Director

01.4TB.040

Temperature of Cable Terminations and Cable
Compartments in Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 23, 1993

One of the important features of almost all low-voltage and medium-voltage switchgear is provision for
terminating power cables. Power cables are sensitive to the temperature of the surrounding air, and must
be chosen carefully to operate in the various environments in which they are installed. A single run of
power cable may pass through several different environments, and the effect of each environment on the
cable selection must be considered. One of these environments may be within the switchgear cable
area.

In metal-enclosed switchgear, power cables usually terminate on buswork, not directly on the terminals
of the main switching device. This is in contrast to panelboard, switchboard and motor control center
construction, where power cables may terminate on the terminals of molded-case circuit breakers or
starters. The allowable temperature rise of the connections to insulated cables and the allowable
temperature of the air surrounding these cables is given in the ANSI switchgear standards, ANSI/IEEE
C37.20.1 for low-voltage switchgear, ANSI/IEEE C37.20.2 for metal-clad switchgear, and ANSI/IEEE
C37.20.3 for metal-enclosed interrupter switchgear.

All three of these standards require the same temperature for these features. 4.5.5 of each of these
standards limits the temperature of the air surrounding insulated power cables to 65C, when the
switchgear assembly is equipped with devices having the maximum current rating for which the
assembly is designed, is carrying rated continuous current, and is in an ambient temperature of 40C.
Table 4 of each standard limits the temperature rise of silver or tin-surfaced connections to insulated
cables to 45C, or a total temperature of 85C. The tests to demonstrate conformance with these limiting
temperature rises require including appropriate sizes and lengths of power cables in the continuous
current path.

When cables are connected to metal-enclosed switchgear, the cable selection must take into account the
air and terminal temperatures encountered in the switchgear.

Technical Director

Plating of Contact Surfaces in Switchgear and Circuit
Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

June 3, 1993

The various ANSI standards covering metal-enclosed switchgear and circuit breakers used in this
switchgear prescribe temperature rise limits for various parts of the equipment. Included are limitations
for buses with plated and unplated joints in switchgear, and for plated and unplated contacts and
connections in circuit breakers. In all cases the temperature rise allowed is considerably higher for plated
connections than it is for unplated connections. Typically, the limit for unplated copper connections is
30C rise, while the limit for plated connections is 65C rise. The higher temperature rise is allowed for
plated connections because plated copper does not oxidize nearly as rapidly as bare copper. Copper
oxide is not a good conductor, and once the oxide forms, the resistance and the temperature rise of the
conductor usually increase rapidly. Since limiting the temperature rise is 30C would require
manufacturers to double the amount of copper used, joints are almost universally plated.

The two materials commonly used for plating are silver and tin. The standard for high voltage circuit
breakers speaks of "silver, silver alloy, or equivalent" surfaces, with "equivalent" being undefined. This
standard was last revised in 1979. The low-voltage and medium-voltage switchgear standards, revised in
1987, speak of "silver surfaced, tin surfaced, or equivalent" connections.

Which material is better, silver or tin? At Powell, we generally use silver, particularly for sliding contacts.
Silver plating is harder than tin plating, and withstands the stress of a moving joint, such as a hinge point
or a primary disconnect, better than tin plating. However, tin plating is superior in certain industrial
atmospheres, such as those containing hydrogen sulfide. On request, Powell will provide tin plating on
the connections of the bus bars in equipments. For various technical and manufacturing reasons, it is not
practical to substitute tin for silver on surfaces within circuit breakers, or on circuit breaker primary
disconnects. If the atmosphere attacks silver surfaces, they should be coated with contact lubricating
grease to prevent corrosion problems.

Technical Director

01.4TB.042

Momentary Rating and Construction of Bus in
Metal-Enclosed Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J une 4, 1993

The bus in metal-enclosed power switchgear is required by ANSI standards to have a momentary current
rating equivalent to the momentary rating of the circuit breaker, switch or fuse used in the particular
switchgear equipment. In the case of low voltage switchgear, where the circuit breaker momentary rating
may vary with the breaker frame size, a particular assembly is required to have the momentary rating of
the smallest frame size of circuit breaker used in the assembly. Thus, if the circuit breakers or other
switching devices are properly applied, the equipment momentary will match both the needs of the
system and the rating of the switching device. Where bus duct is used in conjunction with metal-enclosed
switchgear, the bus duct momentary rating should match that of the switchgear.

Momentary ratings are normally proved by high current testing. The bus must withstand the test without
any permanent deformation of the bus bar, or if there is permanent deformation, it must not be sufficient
to prevent the equipment from passing its standard dielectric tests (hipot and impulse). There also must
be no breakage of the bus supports.

There are a number of design variables that enter into the ability of the bus structure to withstand a
momentary current. The forces involved are quite high. For a bus consisting of one " x 4" bar per
phase, on 6" phase centers, a 50 kA current creates a force on each bus bar of nearly 17,000 pounds
per foot of length. Both the bus supports and the bars themselves must withstand this force. The force is
directly proportional to the square of the current and inversely proportional to the phase spacing, so
moving the bus bars apart decreases the force. The force is also affected to a small degree by the
dimensions and shape of the bus bars.

The ability of the bars to withstand the force created by the momentary current is a function of the size,
shape, and material of the bar and the arrangement of the bars with respect to each other, i.e., flat-to-flat
or edge-to-edge. In mechanical terms, these determine the section modulus of the bars in the phase-to-
phase direction. The deflection of the bars is also affected by the length of the span, or the distance
between supports. The strength of the bus support structure is determined by the material and
configuration of the supports and the distance between them.

The standards require test to demonstrate the momentary rating of the bus, and most bus structures
within switchgear are somewhat complex and not subject to easy analysis, so we seldom use
calculations for the design of bus structures. However, for relatively simple bus configurations, such as
bus duct, it may be easy to modify a design based on previous test data. For instance, if phases are
spread apart further than they were in the tested sample, the allowable increase in spacing between
supports can be easily calculated.

Technical Director

Device Function Numbers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

July 30, 1993

The system of device function numbers used in switchgear schematic and connection diagrams is
documented in ANSI/IEEE C37.2, IEEE Standard Electrical Power System Device Function Numbers.
This system is over half a century old. (No, I haven't been using it quite that long myself, but I've seen it
used on drawings dating back to the 1930's.) However, like any standard, it is revised from time to time.
The latest revision is dated 1991. The issue before that was 1987, and the one before that was 1979.
Listed below are some of the changes made in the last two revisions.
7
In the 1979 edition, device 7 was an anode circuit breaker, a device
frequently used in mercury arc rectifier equipments but no longer seen in
this day of solid state rectifiers. In the 1987 edition, device 7 was
"Reserved for future application." In the 1991 edition, device 7 is used for a
rate-of-rise current relay. For many years, the description of device 50,
which is commonly thought of as an instantaneous overcurrent relay,
included the rate-of-rise feature. This feature is no longer part of device
50's description.
11
In 1979, device 11 was reserved for future application. Beginning with the
1987 edition, device 11 became a multifunction device. This is defined as a
device with three or more important functions. Typical use would be for a
multifunction motor protective relay. When device 11 is used, the functions
included in it should be defined in the drawing legend.
24
In 1979, device 24 was reserved for future application. Beginning with the
1987 edition, device 24 became a volts per hertz relay. These relays are
typically used to protect large generators from overvoltage during sub-
synchronous operation, and are seldom seen in Powell switchgear. I
mention it because there are still those among us who remember the use
of 24 for a bus tie circuit breaker. In the standard, the proper number for a
bus tie breaker has been 52BT since some time in the 1950's, but we still
see the designation 24 from time to time.
47
Up through 1979, device 47 was a phase-sequence voltage relay. Starting
with the 1987 edition, this definition was expanded to read phase-sequence
or phase-balance voltage relay, and the description specifically lists
negative phase sequence overvoltage as one of its applications.
50 See device 7.
82
Up through 1987, device 82 is described as a dc reclosing relay. In the
1991 edition the description is expanded to read dc load-measuring
reclosing relay.
Some other changes were made, but these are the ones most likely to affect switchgear for utility and
industrial distribution systems.

Technical Director

Preparing Foundations for Indoor Installation of Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 2, 1993

Nearly every manufacturer of switchgear and motor control equipment will tell users that the equipment
must be installed on a level foundation. If the equipment includes drawout circuit breakers or contactors,
most manufacturers will recommend that steel channels or rails be imbedded in the floor to provide
support and a level surface. These recommendations lead inevitably to two questions: "How level is
level?" and "Do I really have to install the floor steel?" The quick answers to these are "Damn flat!" and
"No, but you'll be sorry if you don't."

Some smaller equipments, like low voltage motor control centers and switchboards, either come with
their own built-in base channels or are not terribly sensitive to slightly uneven floors. For larger
equipments, however, a level floor is absolutely necessary to maintain the critical alignment of drawout
devices. Manufacturers have been rather shy about putting a tolerance on "level" over the years, but the
lore in the industry seems to be that a one-eighth inch slope, evenly spread over the front-to-back or the
end-to-end dimension of the lineup, is tolerable. For indoor equipments without built-in bases,
maintaining such a tolerance almost certainly requires carefully installed floor channels.

Once you decide that floor steel is required, here are a few cautions about using it:
Be sure to locate the channels where the manufacturer shows them. Normally, each lineup has a
channel near the front of the gear and another near the rear of the gear. These are usually
located under the bolt-down holes in the equipment, so their location is important. Frequently, in
deep switchgear, such as PowlVac
metal-clad switchgear, a third channel is shown somewhere

between the first two. Often, this channel does not match any bolt-down holes, so its location may
not seem to be as important as the location of the other two channels. However, the manufacturer
may have located this channel under some feature of the design which need good support, so its
location may be as important as the locations of the other two channels.
The concrete needs to be no higher than the floor steel. If the circuit breakers roll out on the floor,
the floor on the drawout side of the switchgear needs to be flush with the top of the floor steel so
the breakers will roll in and out smoothly.
Once a level surface is established by the floor channels, be sure that the equipment sits flush on
the surface of the channels. This may seem elementary, but I have seen installations where one
side of a 36" unit was flush with the channel and the other side was " above the channel. All the
effort and expense put into the level floor channels was negated by a poor installation of the
equipment on the channels.

Technical Director

MVA Interrupting Rating of Circuit Breakers Used in
Metal-Clad Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 22, 1993

Modern medium voltage circuit breakers used in metal-clad switchgear have no MVA interrupting rating.
Undoubtedly, this statement will come as a surprise to some readers of this PTB. Although it is quite
common for all of us to talk about 500 MVA circuit breakers or 1000 MVA circuit breakers, this rating
does not appear anywhere in the ANSI standards applying to these breakers, nor does it appear
anywhere on the nameplate of the breakers.

A little history is in order. The first ANSI standards covering circuit breakers, including those for use in
metal-clad switchgear, were developed about 50 years ago. Under these standards, interrupting ratings
were based on the total current interrupted, including the dc component. These "total current" standards
included ANSI C37.4 through C37.9 and C37.12. In the rating structure used in these standards, MVA
was included, and those of us who date back to that era got used to referring to breakers by their MVA
rating.

In 1964, a new series of ANSI standards were first published. These standards used symmetrical, rather
than total, current as the basis for interrupting rating. These new standards no longer referred to MVA in
their rating structure. The interrupting rating in these standards is expressed kiloamperes. After a couple
of decades of development, these standards now include six documents:
ANSI/IEEE C37.04-1979 Rating Structure
ANSI C37.06-1987 Preferred Ratings
ANSI/IEEE C37.09-1979 Test Procedure
ANSI/IEEE C37.010-1979 Application Guide - General
ANSI/IEEE C37.011-1979 Application Guide - Transient Recovery Voltage
ANSI/IEEE C37.012-1979 Application Guide - Capacitance Current Switching

However, so that manufacturers would not have to retest all their breakers, certain equivalences were
established, and for a few years the preferred rating tables carried "nominal" MVA ratings for
"identification". This last appeared in the 1971 edition of ANSI C37.06, but was missing from the 1979
edition. Unfortunately, we have continued to use these identifications informally, and sometimes we get
wrapped around the axle about just what they mean, particularly when applied to circuit breakers used at
a voltage considerably less than their rated maximum voltage, such as 4760 V breakers used at 2400 V.
The chart below compares the nominal MVA ratings to the actual MVA ratings calculated using the rated
interrupting currents established in the current standards.

MVA Interrupting Rating of Circuit Breakers Used in

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Rated
Maximum
Voltage
kV, rms
Rated Short
Circuit
Current
kA, rms
Nominal
MVA
System
Operating
Voltage
kV, rms
Interrupting Current
@ Operating Voltage
kA
Actual MVA @
Operating
Voltage
(1)

MVA
4.76 29 250
4.76 29 239
4.16 33 238
2.4 36 150
4.76 41 350
4.76 41 338
4.16 47 338
2.4 49 204
8.25 33 500
8.25 33 472
7.2 38 472
6.6 41 469
15.0 18 500
15.0 18 468
13.8 20 478
11.5 23 458
15.0 28 750
15.0 28 727
13.8 30 717
11.5 36 717
15.0 37 1000
15.0 37 961
13.8 40 956
11.5 48 956
(1) Slight variations in MVA may be due to rounding of interrupting current values.

If you are interested in the development of these standards, a good history of these standards appears in
the forewords of the various documents.

Technical Director

Significance of K factor in Circuit Breaker Ratings

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 23, 1993

The most basic interrupting ratings of a high voltage circuit breaker are a rated short-circuit current, I,
and a rated maximum voltage, V, at which the breaker must interrupt I. This is a simple concept, widely
understood. However, these breakers also have a rated voltage range factor, K, which is not so widely
understood. (Note: This K factor has absolutely nothing to do with the K factor used to determining the
suitability of a transformer for use in a circuit with high harmonic levels.)

In many types of circuit breakers the physics of arc interruption are such that a given circuit breaker can
interrupt a higher current at a lower voltage. In order to take advantage of this capability in the application
of circuit breakers, the K factor was introduced into the ANSI standards for circuit breakers. The K factor
is a dimensionless number which defines the range of voltage over which the interrupting current
increases. The rated maximum voltage divided by the K factor, or V/K, gives a voltage below which no
increase in interrupting current is required.

The K factor also defines the magnitude of the increased interrupting requirement. The current which
must be interrupted as voltage V/K is KI. Between V and V/K, the current increase is proportional to the
voltage decrease, and may be calculated by the formula:
Required symmetrical current interrupting capability =
rated short circuit current x (rated maximum voltage/operating
voltage)

This formula yields a constant MVA interrupting rating, equal to the square root of three times VI,
between V and V/K, and a constant current interrupting rating, equal to KI, at voltages below V/K.

Two things need to be said about the K factor. First, the concept agrees with the physical reality of oil-
blast and air-magnetic circuit breakers. Breakers using these technologies really do have higher
interrupting ability at lower voltages, and assigning a K factor other than one allows a wider application of
a given circuit breaker. However, circuit breakers using vacuum or SF
6
puffer interrupters are essentially
constant current interrupters up to a limiting maximum voltage, so a K factor other than one does not
match the physical attributes of circuit breakers using these technologies. In the latest (1987) edition of
ANSI C37.06, K has been set to 1.0 for all circuit breakers except indoor oilless circuit breakers, which
are the breakers used in metal-clad switchgear.

Significance of K factor in Circuit Breaker Ratings

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Second, for a known system voltage, the K factor is unimportant. On any given system, if the voltage
decreases, the available short circuit current will also decrease, not increase. If a circuit breaker is
properly applied at the maximum system voltage, it will have the necessary short circuit capability for any
lower voltage on that system.

Technical Director

X/R Ratio

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

November 17, 1993

What is X/R ("X over R") ratio, and why is it important? First, X/R ratio is simply the ratio of the system
reactance to the system resistance, looking back to the power source from any point in a power circuit,
assuming that a short circuit is applied to the system at that point. It is another way of stating the power
factor of the source system. Mathematically, if Power Factor = cos theta , then X/R = tan theta. Note that
this is the power factor of the system up to that point. It has absolutely no relationship to the power factor
of any load on the system. Since generators, transformers and transmission lines are generally quite
highly inductive, the X/R ratio is generally significantly above unity in any utility or industrial power
system.

Why is the X/R ratio important? Its importance is that it affects the level of short circuit current a circuit
breaker is required to interrupt. When a short circuit occurs, the rms value of the symmetrical fault
current is determined by the system source voltage and the total system impedance to the point of fault.
However, almost all faults involve significant asymmetry in at least one phase. This asymmetry is treated
in analysis as a dc component, which must be combined with the ac symmetrical component to give a
new current value, the rms asymmetrical value. It is the value of the rms asymmetrical current at the
moment of contact part which a circuit breaker must interrupt. See PTB 22 for further information.

This dc component of a fault current decays rather rapidly, reaching an insignificant value in a matter of 3
to 5 cycles of the power frequency. However, this rate of decay is determined by the X/R ratio of the
circuit at the point of fault. The higher the X/R ratio, the more slowly the dc component decays.

Circuit breakers are tested using power sources with an X/R ratio prescribed by industry standards. For
power circuit breakers, both low and high voltage, the ANSI standards require this X/R ratio to be 6.6 or
higher, corresponding to a power factor of 15% or less. For a given level of symmetrical fault current and
a given circuit breaker contact part time, this X/R ratio establishes the value of asymmetrical fault current
the breaker is required to interrupt. A higher X/R ratio, with its slower decay rate, will result in a higher
asymmetrical fault current at contact part time. If the X/R ratio is too high, the asymmetrical fault current
may exceed the breaker's interrupting capability.

Since the X/R ratio of a system is inherent in its design and cannot be changed easily, what can be done
about a high X/R? The approach is taken by the ANSI standards is to establish multiplying factors for the
symmetrical fault current. These factors vary with the system X/R ratio and the speed of the circuit
breaker. For high voltage circuit breakers (over 1000 V), information about the values and use of these
factors is found in 5 of ANSI/IEEE C37.010-1979. For low voltage power circuit breakers, this
information is found in 10 of ANSI/IEEE C37.13-1990.

Technical Director

Temperature Rating of Conductors Connected to Molded
Case Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 2, 1993

Books can be written, and probably have been written, on the subject of proper sizing of conductors to
meet all the various requirements of the National Electrical Code. This PTB will address one small facet
of that overall problem: the temperature rating of conductors connected to the terminals of molded case
circuit breakers.

The NEC, as a general principle, requires most material used in electrical systems to be listed by a third-
part certifier, such as UL, and to be installed in accordance with the conditions attached to that listing. In
UL's Electrical Construction Materials Directory ("Green Book") we find the following conditions attached
to the listing of circuit breakers:
2. Circuit breakers with a current rating of 125 amperes or less are marked as being
suitable for 60C, 75C only, or 60/75C rated conductors. It is acceptable to use
conductors with a higher insulation rating, if the ampacity is based on the conductor
temperature rating marked on the breaker.

3. Circuit breakers rated 125 amperes or less and marked suitable for use with 75C rated
conductors are intended for field use with 75C rated conductors at full 75C ampacity
only when the circuit breaker is installed in a circuit breaker enclosure or individually
mounted in an industrial control panel with no other component next to it, unless the end
use equipment (panelboard, switchboard, service equipment, power outlet, etc.) is also
marked suitable for use with conductors rated 75C.

4. A circuit breaker with a current rating of more than 125 amperes is suitable for use with
conductors rated 75C.
In view of these rules, you might ask why anyone would want to use conductors with a higher
temperature rating than the breaker rating, when these higher rated conductors are presumably more
expensive than lower rated conductors. Outside of the possibility of convenience (the 90C wire was
laying around doing nothing), you may find that other derating factors applying elsewhere in the
conductor run will reduce the allowable ampacity so that the 60C or 75C rating at the terminal is met
without difficulty.

As an example, a #4/0 AWG copper conductor with 90C insulation has an ampacity of 260 A per Table
310-16 of the NEC. However, if you connect this conductor to a molded case circuit breaker terminal, its
ampacity is limited to the ampacity of the same size conductor with 75C insulation, or 230 A. The MCCB
terminal temperature rating is the limiting factor in this conductor application. Now, let's look at a circuit
with two of these conductors per phase in parallel, with all size conductors run in the same conduit. Note
8 to Table 310-16 requires an adjustment of the ampacity to 80% of the ampacity listed in the table when
there are 4 to 6 current carrying conductors in one raceway.

Temperature Rating of Conductors Connected to Molded
Case Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
The two 90C cables now have an ampacity of 2 x 260 x 0.8 = 416 A. At the MCCB terminals, the
allowable current is twice the 75C rating of the cable, or 460 A. Now the MCCB terminal temperature is
no longer the limiting factor, and the use of the 90C insulation is advantageous. If 75C insulation were
used, the wire size would have to be 250 kcmil to carry 408 A, and the conduit size would have to be
increased from 2" to 3".

Technical Director

Industry Standards Covering Powell Products

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 3, 1993

All Powell products are designed and tested to conform to applicable industry standards. The following is
a list of the principal standards applying to each of these products. In general, each of these standards
includes a list of reference standards which further define the details of products, test methods, etc.

The major standards organizations whose standards are referenced below are:
ANSI: American National Standards Institute

IEEE: Institute of Electrical and Electronics Engineers

NEMA: National Electrical Manufacturers Association

UL: Underwriters Laboratories

NFPA: National Fire Protection Association
PowlVac
Vacuum Circuit Breakers

ANSI/IEEE
C37.04-1979
IEEE Standard Rating Structure for AC High-Voltage Circuit
Breakers Rated on a Symmetrical Current Basis
ANSI C37.06-
1987
AC High-Voltage Circuit Breakers Rated on a Symmetrical Current
Basis -Preferred Ratings and Related Required Capabilities
ANSI/IEEE
C37.09-1979
IEEE Standard Test Procedure for AC High-Voltage Circuit
Breakers Rated on a Symmetrical Current Basis
ANSI C37.54-
1987
Indoor Alternating-Current High-Voltage Circuit Breakers Applied
as Removable Elements in Metal-Enclosed Switchgear
Assemblies - Conformance Test Procedures
PowlVac
ANSI/IEEE C37.20.2-
1987
IEEE Standard for Metal-Clad Switchgear and Station-Type
Cubicle Switchgear
ANSI C37.55-1989
Metal-Clad Switchgear Assemblies - Conformance Test
Procedures


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Medium-Voltage Switch-and-Fuse Equipment
ANSI/IEEE C37.20.3-
1987
IEEE Standard for Metal-Enclosed Interrupter
Switchgear
Low-Voltage Power Circuit Breakers
ANSI/IEEE
C37.13-1990
IEEE Standard for Low-Voltage AC Power Circuit Breakers Used
in Enclosures
ANSI C37.16-
1988
Low-Voltage Power Circuit Breakers and AC Power Circuit
Protectors - Preferred Ratings, Related Requirements, and
Application Recommendations
ANSI C37.17-
1979
American National Standard for Trip Devices for AC and General-
Purpose DC Low-Voltage Power Circuit Breakers
ANSI C37.50-
1979
Low-Voltage AC Power Circuit Breakers Used in Enclosures - Test
Procedures
Low-Voltage Power Circuit Breaker Switchgear
ANSI/IEEE
C37.20.1-1987
IEEE Standard for Metal-Enclosed Low-Voltage Power Circuit -
Breaker Switchgear
ANSI C37.51-1989
Metal-Enclosed Low-Voltage Power Circuit-Breaker Switchgear
Assemblies - Conformance Test Procedures
UL 1558-1984 Metal-Enclosed Low Voltage Power Circuit-Breaker Switchgear
Bus Duct
ANSI/IEEE C37.23-
1987
IEEE Standard for Metal-Enclosed Bus and Calculating
Losses in Isolated-Phase Bus
Medium-Voltage Motor Control
NEMA ICS
2-1988
Industrial Control Devices, Controllers and Assemblies, Part ICS 2-324,
AC General-Purpose Medium Voltage Contactors and Class E
Controllers, 50 and 60 Hertz


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3
Low-Voltage Motor Control Centers
NEMA ICS 2-
1988
Industrial Control Devices, Controllers and Assemblies, Part ICS 2-
322, AC Motor Control Centers
UL 845-1988 Motor Control Centers
PCR
Power Control Rooms

There are no industry standards for these buildings themselves. Individual Powell-
built equipments installed in PCR
buildings are built per the standards listed

above. Other equipments and devices installed in PCR
buildings are normally

manufactured per the applicable industry standards for the particular item. The only
industry standard that applies to the complete assembly is ANSI/NFPA 70-1993:
National Electrical Code. All electrical work within and on the outside walls of the
PCR
is done per the NEC.

Technical Director

NEC Article 384 Switchboards and Panelboards

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 10, 1993

Every three years the National Electrical Code is revised, and a cottage industry springs up to inform all
of us about the changes made in the new version of the Code. However, sometimes when you are
looking for the intent for the detailed meaning of a particular section of the Code, it is just as useful to
know what proposed changes were not made, and why the code-making panel rejected the proposal.
This information appears in the NFPA Technical Committee Report (TCR), which is issued during each
code cycle. In this document, each proposal for revision of the NEC is reproduced, along with the panel's
action and the panel's reason for taking that action.

In the 1992 TCR, which preceded the 1993 revision of the NEC, there were two interesting rejected
proposals with regard to 384-4 - Installation. The first of these suggested adding the following text:
"Where water sprinklers are in a position where they can deliver water to service
equipment or service bus duct, then that service equipment and service bus duct shall be
rated as raintight."
The code-making panel unanimously rejected that proposal, with the following comment:
"Experience has proven that if a fire activates sprinklers, the sprinklers, if properly
installed and maintained, provide effective protection with virtually no hazard to personnel
and with no measurable increase in damage to the equipment as compared with the
damage done by heat, flame, smoke and the manual hose streams."
The other proposal suggested removing the words "foreign to the electrical equipment" from the text of
the section. The proposer's substantiation was:
"This particular sentence in the 1990 NEC as written can be interpreted as permitting the
installation of a dry type, floor mounted transformer below a panelboard. The transformer
is equipment not foreign to the electrical equipment (panelboard), thus permitting the
installation below the panelboard."
The code-making panel also unanimously rejected the proposal, with the comment:
"Such a transformer not being foreign to electrical equipment is permitted in the dedicated
panelboard space provided it does not intrude into the 110-16 work space."

NEC Article 384 Switchboards and Panelboards

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Figure 1 - Panelboard/Transformer Arrangements (DS* is panelboard's Dedicated Space)

Figure 1 shows three examples of a transformer located near a panelboard. View 1, with the transformer
mounted below the panelboard, is not permitted, as the transformer intrudes into the work space required
by 110-16 of the NEC.

Views 2 and 3 are both installations permitted by the NEC. In View 2, the transformer is located above
the work space, which only extends up to 6 feet from the floor. In View 3, the panelboard has been
mounted away from the wall so that the transformer does not extend into the work space.

Note that the "dedicated space" above and below the panelboard is the width and depth of the
panelboard and extends from the floor to a point 25 feet above the floor, or to the structural ceiling,
whichever is lower. A typical hung ceiling of drop-in panels is not considered to be a structural ceiling for
this purpose.

Technical Director

Arc Resistant Metal-Clad Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

June 2, 1994

Powell has recently announced the availability of arc-resistant metal-clad switchgear, Type
PowlVac-AR
. Arc resistant switchgear has been available for some time in Europe and Canada, and is
now being specified by some users in the United States. Arc resistant switchgear is designed to minimize
the likelihood of injury to a person standing nearby, in the event that an internal arc occurs in the
switchgear. This equipment is sometimes erroneously referred to as "arc-proof", which is incorrect.
"Arc-proof" would imply that an internal arc could not occur; arc-resistant simply means that the
equipment resists the effect of the arc in the unlikely event that one does occur.

Arc resistant equipment must resist two phenomena which accompany an internal arc-increased internal
pressure and burnthrough. Resistance to both of these events is demonstrated up to a height of 2
meters. Typically, arc resistant switchgear is equipped with panels, located above the 2-meter level,
which blow open under the increased internal pressure caused by the internal arc. The lower parts of the
switchgear are reinforced to prevent opening or burnthrough of the enclosure parts.

At the present time, no U.S. standards exist for arc-resistant switchgear, but the IEEE Switchgear
Committee is working on one. Until it is available, the two generally recognized standards are
International Electrotechnical Commission (IEC) Standard 298, Appendix AA, and Electrical and
Electronic Manufacturers Association of Canada (EEMAC) Standard G14-1. These two standards differ
somewhat, and the major differences are shown in Table 1. Powell has designed and tested Type B
switchgear in accordance with the EEMAC standard, which we consider to be more stringent.

Both standards require similar tests to qualify a design. The following criteria must be met during the two
tests (which may be combined into one test if desired):
Properly secured doors, covers, etc., do not open.
Parts which may cause a hazard do not fly off.
Arcing does not cause a hole in the surfaces covered by the type under test.
Black cotton cloth indicators are placed around the equipment under test, up to a height of 2
meters, wherever flame or hot gases might escape. These indicators must not ignite.
The grounding connection must remain effective.
When applying arc resistant switchgear, several items need to be kept in mind:
The arc-resistant rating is at a particular voltage and short-circuit current level. These ratings
should match the ratings of the switchgear and circuit breakers, and the requirements of the
system.
The area of exposure should be considered. See descriptions in Table 1.

Arc Resistant Metal-Clad Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Quoting Note 4 of the EEMAC standard, "The overpressure in the electrical room caused by arcing due
to an internal fault in the switchgear and the ejection of gases from pressure relief devices should be
taken into consideration in the design of the building." At high arc currents, this can be a very dramatic
event.
Attribute IEC 298, Appendix AA EEMAC G14-1
Accessibility:
Type A
Restricted to authorized
personnel only. Test Front. Test
sides which are readily
accessible.
Arc-resistant construction at the
front only.
Accessibility:
Type B
Unrestricted accessibility,
including general public. Test
all accessible sides.
front, back and sides.
Accessibility:
Type C
--------------------------
front, back and sides, and
between compartments within
the same cell or between cells.
Duration of
Tests: Pressure
Withstand
0.1 second (100 ms) 160 ms
Duration of
Tests:
Burnthrough
1 second 1 second
Indicator
Location
Type A: 30 com from unit
Type B: 10 cm from unit
10 cm from unit (all types)
Indicator
Material Weight
Type A: 150 grams per square
meter
Type B: 40 grams per square
meter
150 grams per square meter (all
types)

Technical Director

Wire Fill in Seal Fittings

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

June 14, 1994

NEC requirements for wire fill in seal fittings are different from the wire fill requirements in conduit. In
conduits enclosing 3 or more conductors (except for lead-covered conductors), conductors may occupy
40% of the area of the conduit. See Table 1 of Chapter 9. The fill tables in Chapter 9 of the NEC, which
give the number of conductors of specific types and sizes allowed in various sizes of conduit, are based
on this 40% fill.

For seal fittings, the fill may not exceed 25%. This is not specifically stated in the code, but is backed into
by the requirement that listed items be used in accordance with their listing, and UL lists seal fittings for
25% fill. This is not unreasonable, as the packing material used in a seal fitting must be packed around
each wire, separating them so that the poured sealing compound will effectively seal each wire. The UL
requirements are found in UL Standard 886, Outlet Boxes and Fittings for use in Hazardous (Classified)
Locations.

When installing seal fittings, be sure that the 25% fill is not exceeded. A table of permissible ills by wire
size and seal fitting size is found in UL 886, Table 35.1. UL also requires manufacturers of listed seal
fittings to include the 25% fill requirement in their installation instruction. In some cases, this may mean
using a seal fitting a size or two larger than the conduit size, with reducers to allow the conduit to fit. Any
reducer or other fitting used must also comply with the listing requirements for use in a classified area.
For very short runs of conduit, it may be easier to use an oversized conduit so that the seal fitting is not
overloaded.

Technical Director

Note: Thanks to Crouse-Hinds for the information calling industry attention to this situation.

Hardware for Bus Connections

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

July 5, 1994

What kind of hardware should be used for bus joints in metal-enclosed switchgear? While there may be
several acceptable answers, for half a century or so the workhorse of the industry has been the Grade 5
carbon steel bolt, or, more properly, hex head cap screw. Each bolt is installed with two flat washers, a
split-ring lock washer, and a hex nut. Zinc plated to retard corrosion and installed with the proper torque,
this hardware has a long history of satisfactory performance with both copper and aluminum bus bars.
The most common size used is 1/2-13, but 5/8-11 hardware is used for some very large joints and sizes
down to 1/4-20 are used for smaller jobs, such as fastening terminals for small wire sizes.

Figure 1 -Proper Bolt Assembly

Proper assembly of the hardware is vital to a low-resistance joint. The hardware should be assembled as
shown in Figure 1, with the flat washers next to the bus bars on both sides of the joint and the lock
washer under the nut. The bolt should be long enough that a minimum of two full threads extend out of
the nut when the bolt is tightened. For the 1/2-13 size, use a bolt one inch longer than the combined
thickness of the bus bars being bolted together. Other bolt sizes may take longer or shorter bolts to
compensate for differences in the thickness of the nuts and washers used. The bolt should not be longer
than necessary, either, as extra bolt length usually decreases the clearance from the bolt end to the
nearest other phase or to ground.

Proper bolt torque is vital to a good joint. Grade 5 hardware is used to allow high installation torque.
Torques to be used for various size bolts are shown in the table. This information is given on a label
installed in Powell switchgear near bolted field connections. Do not overtorque the bolts. Excessive
torque can stretch the bolt past its elastic limit and cause failure. I've seen bolts which were barely half
their original diameter in the middle removed from bus joints.
Bolt Size 5/8 - 11 1/2 - 13 3/8 - 16 1/4 - 20
Torque, pounds-feet 55-70 35-50 20-30 5-7

Hardware for Bus Connections

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
How about other types of hardware? Some users specify aluminum hardware for aluminum bus, bronze
(usually Everdur) hardware for copper bus, or stainless steel for either bus material. There are two
reasons usually given for this requirement. The first is the inhibition of corrosion cells where dissimilar
metals made contact. This may be a valid reason for joints that are exposed to the weather, such as
open buswork, or for installation in contaminated atmospheres. However, for the usual metal-enclosed
switchgear, where all the bus joints are inside the enclosure and are expected to be warm and dry, the
special hardware is usually not necessary.

The other reason for specifying hardware of a material similar to the bus bar is concern that differential
expansion between the bus and the hardware may lead to loosening of the joint. For copper bus, which
is used in almost all Powell switchgear, the difference between the expansion of the bus and the
expansion of steel hardware is on the order of 0.0004 inch per inch of joint thickness, or 1 part in 2,500
over the entire 65C allowable temperature rise. Offsetting any advantage of better-matched expansion
characteristics, however, is the difficulty of finding high-strength hardware made of these alloys. Lower
hardware strength may require reduced torque levels on the joints.

One other piece of hardware frequently requested is the spring washer, or Bellville washer. This washer
is used to replace the split-ring lock washer, and is intended to compensate (within limits) for the
differential expansion of the bus material and the hardware. In our experience, it may be of some value
when aluminum bus is used with steel hardware, but is generally unnecessary when the bus is copper.

While Powell will be glad to furnish special bus joint hardware when our customers specify it, in our
experience it is not needed for the usual installation of metal-enclosed switchgear or control equipment.
There are many types of equipment, which have been in service for 50 years, or more using carbon steel
bus joint hardware.

Technical Director

Arc Resistant Switchgear Construction or Arc Detection
Devices?

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

November 16, 1994

In PTB 51, I discussed the arc resistant construction which is now available in Powell's PowlVac
metal-
clad switchgear. As noted in the PTB, arc-resistant switchgear must resist two phenomena - increased
internal pressure and burnthrough. These two phenomena operate in two different time frames.
Burnthrough is a relatively long-time event. The test to prove resistance to burnthrough is of 1 second
duration.

Internal pressure, on the other hand, reaches its peak in 5 to 10 milliseconds after the arc begins. Any
damage that is going to be done by the pressure wave will happen in that time frame. For instance, in
one of our first tests of a preliminary arc resistant design, the front door of a breaker cell was launched
across the test cell at that time, digging a divot out of the concrete wall on the other side of the test cell.

Several manufacturers are now offering optical arcing sensors as an alternative to arc resistant
construction. These sensors will detect an arc within the switchgear and signal a backup breaker to trip,
clearing the arcing fault. However, consider the timing involved. Assuming that the sensor itself has a
zero time of response to the presence of an arc, the backup breaker must still open and interrupt before
the arc is extinguished. The fastest circuit breakers commonly used in metal-clad switchgear are 3-cycle
breakers. This breaker takes 3 cycles, or 50 milliseconds on a 60 Hz system, to interrupt a circuit once its
trip coil is energized. This time of operation will give good protection against burnthrough, but by this time
the internal pressure in the switchgear has long since reached its peak, and any physical damage that
pressure is going to cause has already happened.

Optical arc sensors may offer excellent protection to gas-insulated substations, where the principal
problem is protection against burnthrough. However, no detection system can protect metal-clad
switchgear of normal, unreinforced construction against damage from arc-generated pressure because
the backup breaker cannot open fast enough to prevent this damage. If resistance to damage due to arc-
generated pressure is desired, the arc-resistant design of metal-clad switchgear must be used.

Technical Director

Useful Life of Metal-Enclosed Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 8, 1994

We are frequently asked how long we expect metal-enclosed switchgear to last in service; what is the
design life of our products. Other than the very complex procedures used in the nuclear industry, there is
absolutely nothing in any industry standard that addresses this question, for the very good reason that
there is no reasonable way to test the life of a piece of switchgear other than to put it into service and see
how long it lasts. However, for the over 40 years that I've been in the industry, most manufacturers have
rather consistently answered this question with an estimate of 30 to 40 years of useful service life. There
are numerous installations of metal-enclosed switchgear which have been in service for 40 years or
more.

Of course, to get this kind of service out of switchgear, there are a few ground rules you have to follow.
Here are some of them.
Install it properly. Make sure the foundation is level, and that the equipment has been set level
on the foundation and properly secured. Check alignment of all disconnects and interlocks on
drawout circuit breakers to make sure that they fit properly into the cell. Be sure all bus splices
are assembled properly, with the right size splice plates and the right size and number of bolts,
properly torqued, and insulated if insulation is required. Connect cables carefully. Check out all
control wire connections.
Don't overheat it. Heat is one of the two great enemies of electrical insulation. Don't overload the
equipment in service. Make sure that ventilation is adequate, and that any filters or other
ventilating openings in the switchgear are clean and that air flow is not restricted. If artificial
cooling is required to keep the ambient temperature within limits (usually a maximum of 40C), be
sure it operates properly.
Keep it dry. Moisture is the other great enemy of insulation. Make sure there are no leaks that
will allow rain or other moisture to dampen the switchgear. For outdoor installations and indoor
installations in damp climates, space heaters are recommended to deter condensation.
Take care of it. The equipment should be inspected, cleaned, and lubricated on a regular
schedule, and after any traumatic event, such as interrupting a major fault.

Treat your switchgear well, and it will last you a long time. Abuse it or neglect it, and it will die young.

Technical Director

Control Wire and Wire Markers in Switchgear and Motor
Control

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 19, 1994

ANSI standards for metal-enclosed switchgear require the use of either Type TBS or Type SIS wire for
control wire in this type of equipment. Further, the wire is required to have a minimum size of #14 AWG,
and to be flexible (41 strand or more) where it is connected across a hinge. Powell's standard is to use
Type SIS flexible wire for all switchgear wiring. The wire will be #14 AWG unless a larger wire is required
by the circuit or is specified by the purchaser.

Switchgear control wire has traditionally been colored gray, although this is not called for by any industry
standard. Powell's standard is to use gray wire except for ground wires, which are green. Other colors
may be used if specified by the purchaser, but elaborate color coding can be rather costly, as wire
manufacturers require sizeable runs of special items such as unusual colors.

Special wire types can also be used when required, but are also subject to special charges for non-
standard wire mill runs.

Wire markers, or wire tags, are used in most equipment we produce, but are not required by standards.
Therefore, they must be specified by the purchaser when desired. When wire markers are used, Powell's
standard is a white tube marked with black characters. The wire marker is marked with the wire number
shown on the schematic and wiring diagrams.

Special wire markers can be furnished on request. Some special features requested from time to time
include special sleeve materials, special colors, heat shrink installation, and special wire designations or
numbering. All of these can be furnished, but at a cost.

Technical Director

Ratings of Medium Voltage Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

December 22, 1994

Based on questions I receive, there appears to be quite a bit of confusion in the industry about the
meaning of various voltage and current ratings applied to circuit breakers used in metal-clad switchgear.
This PTB will attempt to clarify some of the more commonly discussed ratings, as these ratings appear in
the ANSI circuit breaker standards.

First, there is no such thing as an MVA rating for a circuit breaker. See PTB 45 for a detailed discussion
of MVA. Other important ratings are as follows:

Rated Maximum Voltage V: The highest rms (root mean square) voltage for which the circuit breaker is
designed, and the upper limit for operation.

Rated Voltage Range Factor K: The ratio of the rated maximum voltage to the lower limit of the range
of operating voltage in which the required symmetrical and asymmetrical interrupting capabilities vary in
inverse proportion to operating voltage. For more on K, see PTB 46.

Rated Short Circuit Current I: The highest value of rms symmetrical current which the circuit breaker is
required to interrupt at rated maximum voltage.

Maximum Symmetrical Interrupting Capability: The highest value of rms symmetrical current which
the circuit breaker is required to interrupt. It is required to interrupt this current at a voltage of V/K and at
any lower voltage. Numerically, this current is equal to KI. Interrupting capabilities at voltages between V
and V/K are calculated by a formula given in PTB 46.

Rated Short Time Current: This is the rms value of the current which the circuit breaker is required to
carry for 3 seconds. It is not an interrupting rating; the breaker is not required to interrupt this current until
it has cooled down to operating temperature. Numerically, this current is equal to KI, the maximum
symmetrical interrupting capability.

Closing and Latching Capability: This is the peak, or crest, current that the circuit breaker must be
capable of making and immediately thereafter, latching. Numerically, this current is equal to 2.7KI. The
circuit breaker must also be able to withstand this same value of current in the closed position as a part
of the short time current test. This capability is sometimes referred to as the "momentary current" rating,
although this term does not appear in the ANSI standards.

Ratings of Medium Voltage Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Since interrupting ratings vary with voltage, it is absolutely imperative that purchaser and supplier
communicate clearly about the voltage at which a specified interrupting rating applies.

Technical Director

Rating of Ground Bus in Metal-Enclosed Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

January 17, 1995

All metal-enclosed switchgear built to ANSI standards is required to include a ground bus to connect
together all items in the switchgear that are required to be grounded. This includes such things as the
enclosures, circuit breaker frames, CT and VT secondary circuits, relay and instrument cases, etc. The
purpose of the ground bus is to keep these items at a common potential under normal conditions and to
carry ground fault current when a ground fault occurs.

The ground bus is rated for short-circuit and short-time duty. For metal-clad switchgear, ANSI/IEEE
C37.20.2, 6.12, requires the ground bus to carry the rated short-time current of the switchgear for 2
seconds. For low voltage metal-enclosed switchgear, ANSI/IEEE C37.20.1, 6.1.2, requires the ground
bus to carry the rated short-time current of the switchgear for 0.5 second. Tests to demonstrate these
ratings are included in the conformance test standards, ANSI C37.51 for low voltage switchgear and
ANSI C37.55 for metal-clad switchgear.

Ground bus is not expected to carry any continuous current. If a 4-wire system with line-to-neutral loads
is in use, a neutral bus may be needed to carry any current unbalance. This is quite common in low
voltage switchgear, and much less common but not unknown in metal-clad switchgear.

None of the ANSI standards require a continuous current rating for ground bus, and none of these
standards describe any test to demonstrate such a rating. While a ground bus, like any conductor, will
carry a certain amount of current continuously without damage, just how much and under what
circumstances is not defined. Therefore, specifying a continuous current rating for ground bus has little or
no meaning, and should be avoided.

Technical Director

Bus Spacings in Metal-Enclosed Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

January 24, 1995

From time to time we are asked what bus spacings are required by ANSI standards for switchgear.
Those who ask are frequently surprised by the answer: None. ANSI switchgear standards are generally
performance standards. Dielectric tests, power frequency withstand for all voltages and impulse
withstand for medium voltage, are specified in the standards. The design must pass these tests. How a
manufacturer designs equipment to meet the requirements is up to the manufacturer. If you can place
bare conductors 1/2" apart and meet the test requirements for 15kV equipment, that is fine. And before
you conclude that I'm being ridiculous, remember that we do this every day in vacuum interrupters.

When considering bus spacings, two dimensions are important. The first is clearance, or the distance
through air between conductors of opposite polarity or between an energized conductor and ground. The
second is surface creepage, or the distance across an insulating surface. The distances are measured
from metal to metal, and vary with voltage and also with whether or not the conductors are insulated.
Phase-to-phase and phase-to-ground dimensions are the same because switchgear used on
ungrounded or impedance grounded systems will have phase to phase voltage between the unfaulted
phases and ground during a ground fault condition.

It is not possible to test every configuration of bus used in switchgear, so every manufacturer has a
working guide of dimensions to be used for configurations that aren't tested. Remember that these are
dimensions used within metal-enclosed switchgear equipments. They do not apply for overhead lines,
pole-top hardware, outdoor substation construction, etc. The dimensions used by different manufacturers
may differ a bit, but they are usually pretty consistent. The following table shows some of the more
common dimensions we use at Powell.
Voltages Air Clearance Surface Clearance
Rated
Maximum
Low-Frequency
Withstand
Impulse
Withstand
Insulated
Conductors
Bare
Conductors
Insulated
Conductors
Bare
Conductors
635 V 2.2 kV N/A N/A 1" N/A 2"
4.76 kV 19 kV 60 kV 2" 3 1/2" 3" 5"
15 kV 36 kV 95 kV 3" 6" 5" 7"
27 kV 60 kV 125 kV 6" 9" 9" 14"
38 kV 80 kV 150 kV 7 1/2" 10 1/2" 11" 17"

Technical Director

01.4TB.060

Switching Small Currents with Vacuum Circuit Breakers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J anuary 30, 1995

We occasionally see a specification which requires that the medium voltage circuit breakers that are
supplied should be capable of switching very small currents, in the range of a few percent of the
breaker's continuous current rating. Generally speaking, switching of low currents is not a problem for
vacuum circuit breakers.

Air-magnetic circuit breakers, which have been successfully used for many years in medium voltage
switchgear, do have some problems with switching low currents. These breakers depend on a magnetic
field generated by the current flowing through auxiliary coils in the interrupter circuit to drive the arc into
the arc chute for successful interruption. When the current is very low, only a small field is generated,
and the necessary arc movement may not be obtained. Almost all varieties of air-magnetic circuit
breakers are equipped with a "puffer" device, a small air piston driven by the opening of the breaker. This
piston is equipped with a nozzle that is directed at the area where the arc is struck by the opening of the
breaker, so that the arc is literally blown into the arc chute. This device is needed on air-magnetic circuit
breakers to insure interruption of small currents.

At least in part because of this problem with the interruption of small currents, the ANSI standard for
circuit breaker rating allows the interrupting time for the interrupting of current below 25% of the required
asymmetrical interrupting capability to exceed the rated interrupting time by as much at 50%.

Vacuum breakers, on the other hand, handle small currents with ease. We have tested one model of
PowlVac
circuit breaker at inductive currents as low as 25-30 A, and some other models at capacitive
currents as low as half an amp. Load current switching tests have been made on all of our present
production models of PowlVac
circuit breaker at inductive current levels in the 250-260 A range. All of

these tests were successful, and the interruptions took no longer than high current interruptions. The
oscillograms indicated no distress in the interruptions. As far as we are concerned, interruption of low
currents is a non-problem for PowlVac
circuit breakers.

Technical Director

Overcurrent Definitions

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 7, 1995

There are several terms that we use to name abnormal current in an electric power system. Although
these terms are sometimes used interchangeably, they really aren't interchangeable. Recently, I came
across a set of definitions that made a lot of sense to me, so I'm passing them along to you. These are
taken from a couple of standards of the International Electrotechnical Commission, IEC 439-1 and IEC
947-1.
Overcurrent: A current exceeding the rated current.
Short circuit: The accidental or intentional connection, by a relatively low resistance or
impedance, of two or more points in a circuit which are normal at different voltages.
Short circuit current: An overcurrent resulting from a short circuit due to a fault or an incorrect
connection in an electric circuit.
Overload: Operating conditions in an electrically undamaged circuit which cause an overcurrent.
Overload current: An overcurrent occurring in an electrically undamaged circuit.
Fault current: A current resulting from an insulation failure or the bridging of insulation.

Note several relationships among these various currents:
An overload current is always an overcurrent, but not all overcurrents are overload currents.
An overload is not a fault.
A short circuit current is both a fault current and an overcurrent. However, not all fault currents
are short circuit currents. Also, not all overcurrents are short circuit currents.
A fault current is not necessarily an overcurrent. Under some fault conditions, the fault current
may be much less than the rated current. A typical example is a ground fault current on a high-
resistance grounded system. This current may be only an amp or two, compared to a rated
current of up to several thousand amps.

Technical Director

Choosing Current Transformers for Relaying Use

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

March 28, 1995

Current transformers used for relaying under high current conditions must have a relay rating. This rating
is expressed as the letter "C" or the letter "T" followed by a number. The number represents the voltage
output at the CT terminals with 20 times normal secondary current flowing (100A for a typical 5 A
secondary CT), without exceeding 10% ratio error. Standard values are 10, 20, 50, 100, 200 and 400.
The "C" or "T" describes the method used for determining this voltage. The voltage for CT's with a "T"
designation has been determined by test, the usual method for wound primary CT's.

Recent work by a working group of the IEEE Power Systems Relaying Committee has shown that low-
ratio CT's applied to systems with high fault current levels may saturate under fault conditions and cause
improper or no operation of the overcurrent relays. Extreme saturation of the CT's can result in very
narrow pulses of secondary current, only a few electrical degrees wide. Although the magnitude of the
pulse may be above the pickup setting of the relay, the current does not flow for long enough at a time to
operate the relay.

To be sure that the CT's do not saturate, the relay rating of the CT should be twice the voltage necessary
to drive the maximum fault current through the connected burden. For example, assume a bus with 24
kA available fault current and a CT rated 400/5 A (80:1 ratio). Divide 24,000 A by the CT ratio, 80, to
determine the secondary current, 300 A, under maximum fault conditions. If the secondary burden is
0.33 ohms, the voltage required is 300x0.33, or 100 V. A CT with a relay rating of C200 or T200 should
be chosen for this application.

Where low-ratio CT's are needed on a system with a high available fault current, it may be very difficult to
find a CT with the necessary relay rating. Several possible ways to improve the situation are to use
higher ratio CT's, to use higher accuracy CT's, to use lower burden relays, to divide the burden between
two sets of CT's, or to use separate instantaneous overcurrent relays connected to a separate set of
high-ratio CT's.

If you want to review the entire report of the PSRC working group, you can find it published in two
different IEEE Transactions. The report title is "Relay Performance with Low Ratio CT's and High Fault
Currents". It may be found in IEEE Transactions on Power Delivery, Vol. 8, No. 3, July 1993, pp. 884-
897, and in IEEE Transactions on Industry Applications, Vol. 31, No. 2, March/April 1995, pp. 392-404.

Technical Director

Capacitance Graded Bushings

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

June 12, 1995

Bushings or similar structures are used in medium voltage switchgear to carry a conductor from one side
of a grounded barrier to the other side. Typical uses are roof entrance bushings, to bring conductors from
outside the switchgear through the roof to the switchgear interior; primary disconnect spouts, to bring the
primary connections of a circuit breaker from the bus or cable compartment to the circuit breaker
compartment; and main bus penetration from one unit of switchgear to the next unit.

In its simplest form, a bushing consists of a conductor, some surrounding insulation, and a mounting
provision of some kind. If the voltage isn't too high, and the configuration of the conductor, the insulation,
and the ground plane isn't too bad, a bushing with suitable insulation characteristics and long life can be
made with only these basic ingredients.

However, sometimes the spacing is very tight, or the ground plane has sharp corners or protruding
hardware so that the local stress on the insulation is excessive. In this case, a capacitance graded
bushing may be the answer. In a capacitance graded bushing, layers of conductive or semi-conductive
material are placed in the thickness of the insulation in such a manner that these layers serve as a built-
in capacitor and the stress on the insulation is equalized and controlled. Proper use of capacitance
grading can control both through-insulation and over-surface stress. This technique has long been used
in medium voltage switchgear. For instance, GE's vertical lift Magne-Blast circuit breakers, first
manufactured nearly 60 years ago, use capacitance graded top-mounted bushings to control the stress
where the bushings penetrate the top frame of the circuit breaker.

Capacitance Graded Bushings

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
The figure above, shows a section view of the capacitance graded primary disconnect spout used in
Powell's PV System 38
switchgear. Each line in the thickness of the insulation represents a conductive

layer. The innermost layer is connected to the primary conductor, and the outermost layer is connected
to ground. In addition to controlling the stress in the insulation, this particular construction provides an
area around the spout shielded from the electrical field, for mounting current transformers.

Technical Director

Load Switching Capability of NEMA General Purpose
Contactors

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

June 22, 1995

NEMA general purpose contactors, of the type generally used in motor control centers, have a number of
different load switching ratings, depending on the type of load being switched, the duty cycle of the load,
and the switching sequence being used. The table below shows the ratings assigned to size 1 through
size 5 for use on 60 Hz, 480 V systems, with motors rated 460 V.
NEMA
Size of
Contactor
Motor
HP Full
Voltage
Starting,
Non-
plugging
and Non-
jogging
duty
Peak
Circuit-
Closing
Amps
(Inrush)
rms Amperes
Transformer
Primary
Switching
kVA
Continuous
Current
Switching Rating
Inrush
<=
20x
Full
Load
Inrush
>20x
<=40X
Full
Load
Tungsten
Lamps
(300V
max.)
Resistance
Loads &
Discharge
Lamps
1 10 288 27 15 27 8.5 4.3
2 25 483 45 30 45 14 7.2
3 50 947 90 60 90 28 14
4 100 1581 135 120 135 47 23
5 200 3163 270 240 270 94 47

These values are taken from tables in NEMA Standard ICS2-1988. Other tables in this standard give
ratings for other sizes of contactors, other voltages and frequencies, single phase, other motor duties
and starting methods, etc.

Load Switching Capability of NEMA General Purpose
Contactors

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
In most cases, the rating limit is established by the contactor's ability to close a certain peak current.
Loads with high inrush characteristics require oversized contactors to handle the inrush current without
damage. For instance, consider a 15 kVA, 480 V, 3 phase transformer, which has a full load current of 18
A. This is well within the continuous current rating of a size 1 contactor, but switching the primary of this
transformer required a size 3 contactor if the inrush current is <= 20 x full load, and a size 4 contactor if
the inrush current is between 20 x and 40 x full load.

Technical Director

Starting Synchronous Motors

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

June 29, 1995

Starting synchronous motors is a two-step process. The first step is to accelerate the motor to near
synchronous speed. Most synchronous motors are equipped with a squirrel cage winding on the rotor for
starting purposes. This differs from an induction motor squirrel cage in that it is not rated to carry load,
but only to assist in starting. The synchronous motor is started as a squirrel cage induction motor, and
can use any starting method that is used on a squirrel cage induction motor: full voltage,
autotransformer, reactor, capacitor, etc. Protection required for the motor's stator and this first stage of its
starting sequence is the same as required for an induction motor of similar size and starting method.

Once the motor reaches near synchronous speed, usually defined as about 95% speed, d.c. is applied to
the rotor's field winding, pulling the rotor into step at 100% of synchronous speed. This is the second
step of the starting process. During the acceleration stage, the field winding is shorted through a field
discharge resistor, which is removed from the circuit before the d.c. is applied.

There are two basic types of fields used on synchronous motors, brush type and brushless. For either
type, much detailed information about the motor's characteristics is required for proper application of
starting equipment. On brush type motors, which are an older design, the two ends of the field winding
are brought to slip rings, which are contacted by brushes, giving this type its name. D.c. from an external
supply is applied to the field using external switching devices, usually a field contactor. The external
supply may be a rotating exciter, either shaft-driven or a separate m-g set; an excitation bus; or the
output of a solid-state excitation package. Speed is sensed by measuring the frequency of the induced
a.c. current which flows in the field winding and the field discharge resistor. This frequency decreases as
the motor speed increases. When an appropriate speed is reached, the control closes the field contactor,
opening the discharge resistor circuit and closing the main field circuit. The control may be electro-
mechanical or solid-state, and includes protection against loss of field and incomplete starting sequence.

Brushless motors have the armature of an a.c. generator, a rectifier package, the field discharge resistor,
and a solid-state sensing and switching package mounted on the rotor of the synchronous motor. The
a.c. generator serves as the exciter, and its field winding is stationary, and requires no slip rings or
brushes. A small d.c. supply is required for this field winding, and is usually supplied by a solid-state
package. Controls are also usually solid-state. Protective relays are unable to measure the d.c. field of
the synchronous motor directly, so loss of field protection is provided by relays looking at the power
factor of the motor, which changes suddenly when field is lost.

Technical Director

Use of Auxiliary Current Transformers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

November 3, 1995

There are several applications for auxiliary current transformers in switchgear, including matching ratios
of main CTs, stepping up the secondary current of main CTs for greater sensitivity in metering or
relaying, and combining several main CT secondary circuits for totalized metering. Since makers of
auxiliary CTs will provide almost any ratio you might need, these devices can be a very useful tool in
complex or specialized CT circuits. However, their use may create a burden problem for the main CTs if
care is not used in the application of auxiliary CTs.

A typical auxiliary CT circuit is shown in the figure above. The auxiliary CT secondary burden, , is
reflected into the main CT secondary circuit by the square of the current ratios, per the following formula:

where is the total burden on the main CT and is the burden of the auxiliary CT itself.

From this equation you can see that using an auxiliary CT to step the main CT secondary current up
increases the burden on the main CT by the square of the step-up ratio, whereas using the auxiliary CT
to step the main CT secondary current down decreases the burden by the square of the step-down ratio.
From this, it would seem that, if you are matching two main CT ratios by using an auxiliary CT, it would
always be preferable to step down the higher main CT secondary current, which is the current from the
main CT with the lower ratio.

Use of Auxiliary Current Transformers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Unfortunately, it is not always that simple. A couple of other considerations come into play:
The higher ratio CTs may have a higher burden capability than the lower ratio CTs, and thus be
capable of handling a larger burden with the same accuracy.
The lower current resulting from the step-down transformation may require a lower relay setting to
achieve the desired sensitivity, and this lower setting may greatly increase the relay burden.

As always, the best solution is to do a complete burden calculation to see which way, step-up or step-
down, gives the best results. "Best results" means that all of the devices, including the main CTs, the
auxiliary CTs, and the relays or meters connected in the circuit, are operating within their capabilities,
and that the overall circuit will give the performance desired under all operating conditions.

Technical Director

Using Latched Contactors in Medium Voltage Motor Control
Centers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

November 13, 1995

The motor starting contactors used in medium voltage motor control are usually held closed by an
operating coil, which is energized to close the contactor. When the coil is deenergized, the contactor
opens. Since the contactor's coils is usually energized from a control power transformer connected to the
primary circuit of the starter, this design provides automatic undervoltage protection for the motor.

For loads other than motors, however, it is sometimes desirable to maintain the circuit during an
undervoltage condition. This is commonly done for transformer feeders originating in the medium voltage
MCC. To handle this type of circuit, latched contactors are available. When closed by its operating coil, a
latched contactor will remain closed even if the closing coil is deenergized. The latched contactor is
opened by energizing a trip coil, something like a circuit breaker is opened. Latched contactors may be
equipped with manual closing and/or tripping operators if desired. These may be in addition to or in place
of the operating coils, leading to quite a few possible combinations of operators.

Several cautions are in order when latched contactors are used:
For non-motor loads, such as transformers or capacitors, the motor starting current-limiting fuses
should be replaced with general purpose current-limiting fuses. These fuses may stand alone as
the overcurrent protection, or they may be supplemented with overcurrent relays for better
overload protection.
Since automatic undervoltage protection disappears when a latched contactor is used, separate
undervoltage relays must be provided if undervoltage protection is needed.
There is a major difference between the control circuit for a latched contactor and the control
circuit for a circuit breaker. The latched contactor has no anti-pump feature. If a latched contactor
is presented with simultaneous, maintained close and trip signals, it will cycle closed and open
until one of the signals is removed or until the contactor destroys itself. A circuit breaker, on the
other hand, will close once and open once, then remain open until the closing signal is removed
and reestablished. Control and interlocking circuits used with latched contactors should be
investigated very carefully to make sure that there is not a circuit that could result in damage to
the contactors.

Properly applied latched contactors are useful devices, but they don't work just like circuit breakers. Be
sure you understand the differences and take them into account when using latched contactors.

Technical Director

Instantaneous Ground Fault Relays (50GS) and
Zero-Sequence CTs

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

November 28, 1995

In PTB 62, I discussed the problem of low-ratio CTs used on systems with high fault current, and I
mentioned the IEEE working group report on this subject. Because of the emphasis in this report on
making sure that CTs do not saturate, a number of people have become concerned about the operation
of instantaneous ground fault relays connected to zero-sequence, or core balance, CTs. Because of this
concern, Powell recently ran a series of tests to check the operation of typical CT-relay combinations.
Two different relays were tested with each of two CTs. The relays were the GE HFC and the ABB IT.
Electro-mechanical relays were chosen for the test because their higher burden places a greater load on
the CTs. The CTs used were both made by ITI. The first Model 141-500, 50/5, C10 accuracy. The
second was Model 143-500, 50/5, C20 accuracy. The test results are given in the table below.
Relay Under
Test
Relay Pickup
Setting
Current
Transformer
Primary Pickup
Current
Time to trip at indicated current
(ms)
Pickup 600 A 1200 A 1800 A
HFC 0.5 A
141-500 15.8 A 39.3 34.85 24.05 36.9
143-500 13.6 A 36.35 33.75 27.05 31.95
IT 0.15 A
141-500 6.5 A 95.2 61.3 67.7 72.6
143-500 5.4 A 132.4 56.3 60.0 68.9

Both of these relays operated correctly and reliably with both CTs. However, we also tested a third relay,
the ABB ITH, a high dropout version of the IT. We found that this relay was not reliable in this service. It
picked up at quite low values, and operated well with primary currents up to about 150 A. At the higher
currents, 600 A and up, it chattered quite badly and did not close its contacts for long enough to operate
a circuit breaker. Asking around, I found that this relay had been recommended for 50 GS service some
years ago, but its manufacturer (Westinghouse at that time) changed the recommendation when the
chattering problem was discovered. Based on this information and the tests, Powell strongly
recommends that the ITH relay not be used as a 50GS relay.

Summarizing, both the HFC and the IT work quite well at primary ground fault currents up to 1800 A,
even though the CTs are badly saturated at that current level. This circuit, with these CTs and relays,
should not be used on solidly grounded systems with high ground fault current. For these systems,
residually-connected relays should be used, or the zero-sequence CTs should have higher ratios.

Technical Director

Wound Rotor Induction Motors and Starters

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

January 5, 1996

The garden variety of induction motor used in industry is the squirrel cage induction motor. Many of the
characteristics of these motors, such as starting inrush current, starting torque, and slip, are fixed.
Although a motor designer can vary these characteristics at the design stage, once the motor is
manufactured these characteristics are fixed.

The wound rotor induction motor has a rotor winding that is not short-circuited on the rotor, like a squirrel
cage motor, but is brought to slip rings so that the rotor circuit can be modified by inserting external
resistance. This added resistance can accomplish two things:
Since the starting torque developed in an induction motor by a given starting current is
proportional to the rotor resistance, adding resistance during the starting cycle can increase the
starting torque, or lower the starting current for the same torque. By decreasing the rotor
resistance as the motor accelerates, the torque can be controlled to provide a smooth
acceleration of the load, sometimes called a "soft start". This can be very helpful for loads such
as cranes, hoists, and loaded conveyors.
Since the slip of an induction motor increases as the rotor resistance increases, some speed
control is possible using a wound rotor motor and varying the rotor resistance while running.

Controllers for wound rotor motors include for the stator the same protection and switching functions that
are used for squirrel cage motors. For the rotor, a multi-step resistor and switching means for that
resistor are required. For a soft start application, the resistor is switched out of the circuit step by step as
the motor accelerates. Once it reaches full speed, the resistor is completely shorted out, and the motor
runs like a squirrel cage motor. The number of steps varies, depending on the motor size and the starting
characteristics desired, but generally falls in the range of 3 to 7 steps. The switching may be done by a
manual drum switch or by a series of contactors activated by timers. For speed control, some resistance
remains in the rotor circuit for all speeds except maximum speed. The running resistor may be further
steps of fixed resistance, like the starting resistors, or a continuously variable resistance, like a liquid
rheostat.

Wound Rotor Induction Motors and Starters

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Wound rotor motors have some drawbacks. Both the motor and the controller are more complex and
cost more than a similar squirrel cage motor and its controller. The added running resistance in the rotor
circuit of the motor on speed control increases losses, and thus operating costs, and the maximum
speed is limited to something less than synchronous speed for the number of poles in the motor.
Because of these limitations, many users now turn to squirrel cage motors and variable frequency drives
to meet their needs for both soft start and speed control.

Technical Director

01.4TB.070

The Importance of Transient Recovery Voltage

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J anuary 10, 1996 Superseded by PTB #104 (September 14, 2012)

In PTB 10 there is some information about transient recovery voltage (TRV) and some data about the
TRV values used by Powell in testing PowlVac
circuit breakers. I'll try to answer some additional recent

questions about TRV in this PTB.

Q. Just what is transient recovery voltage:

A. As circuit breaker contacts part during the interrupting process, an arc is created between them. The
contact-to-contact voltage is the arc voltage, typically about 600V in a 15kV circuit. When the current
passes through zero, conduction ceases, and the contact-to-contact voltage moves toward the difference
in system voltages on the two sides of the open circuit breaker. This change in voltage is the transient
recovery voltage.

Q. How is TRV determined?

A. TRV is an inherent function of the parameters of a power system, and of the location and magnitude
of the fault being interrupted. Since the circuit breaker is part of the system, it may have some minor
effect on the TRV, but this effect is usually unimportant.

Q. Why should I care about TRV?

A. TRV withstand capability is a rating of a circuit breaker, just like continuos current, maximum voltage,
and interrupting current. If the TRV withstand rating is exceeded by the system TRV, the circuit breaker
may fail to perform properly.

Q. How do I know what my TRV will be?

A. This may take a computer-based system study. Computer programs designed for power system
analysis can usually calculate TRV at selected points on the system.

Q. What TRV will my breaker withstand?

A. ANSI Standard C37.06 defines the requirements. The rating is stated in terms of a peak voltage, E
2
,
and a time to reach that voltage, T
2
, at full rated short circuit current for a fault at the terminals of the
breaker. For lower values of short circuit current, the voltage is higher and the timer is shorter. The curve
between O,O and E
2
,T
2
is defined as a "1-cosine" curve, shown visually in the figure below. For circuit
breakers used in metal-clad switchgear, E
2
is required to be 1.88 times the breaker's rated maximum
voltage. The present standard does not specify T
2
, but a proposed revision lists values from 50s for
breakers rated 4.76kV to 125s for breakers rated 38kV. Both E
2
and T
2
increase as the breaker's rated
voltage increases. Breakers rated 121kV and above also use a different curve shape.

01.4TB.070

The Importance of Transient Recovery Voltage

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Technical Director

Starting Methods for Large Medium Voltage Motors

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 9, 1996

Most ac motors, both synchronous and induction, are started "across the line"; that is, the starting
contactor or circuit breaker connects the 3-phase motor winding directly to the 3-phase power supply.
However, when motors are large with respect to the capacity of the power system it is often necessary to
use a starting method that reduces the impact on the power system caused by starting the large motor.
Several methods of accomplishing this task are available. This reduced-impact starting is frequently
referred to as reduced voltage starting because most of the common methods involve applying a
reduced voltage to the motor winding. Some of the more common methods are described below.
Autotransformer: An autotransformer is connected between the power source and the motor
during the starting period. Motor starting autotransformers usually have taps that apply 80%,
65%, or 50% of the line voltage to the motor to start. Line current is reduced by the square of the
tap; that is, using the 80% tap on the autotransformer requires only 64% of the across the line
starting current. Starting torque is also 64% on the 80% tap. Unless otherwise requested, the
autotransformer will be the medium duty type, allowing 3 starts, followed by an hour's rest before
repeating the three starts. Modern starters use the closed transition switching sequence, in which
the autotransformer winding is converted briefly to a series reactor near the end of the starting
sequence, then shorted out. This sequence requires three switching devices (circuit breakers or
contactors).
Reactor: A reactor is connected in series with the motor. These reactors usually have taps that
apply 80%, 65%, or 50% of the line voltage to the motor to start. Line current is reduced to the tap
value; that is, using the 80% tap on the reactor requires 80% of the across the line starting
current. Starting torque is reduced by the square of the tap, and is 64$ on the 80% tap. The
reactor may be placed on the line side of the motor or in the neutral. Reactor start requires only
two switching devices.
Capacitor: A bank of capacitors is connected in parallel with the motor during starting, canceling
out the large reactive current drawn by the motor on starting. The motor thinks it is seeing a full
voltage start, while the power system thinks it is seeing a running motor. The capacitors are
removed from the circuit as the motor reaches running speed. Two switching devices are
required.
Wound Rotor: For induction motors only. See PTB 69 for further information.
Solid-state Drives: A wide variety of starting and speed control performance can be obtained
through the use of modern solid-state drives.

Technical Director

Open Circuit Protectors for Current Transformers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 10, 1996

Current transformers should never be operated with an open secondary circuit. If the secondary of a CT
carrying primary current is open circuited, a high voltage can be developed across the CT terminals.
Depending of the characteristics of the particular CT, this voltage may be several hundred volts if the
primary current is high enough. This voltage may be dangerous to personnel servicing the equipment
and damaging to the CT itself or to devices connected to the CT. To guard against this possibility,
devices used in CT secondary circuits are designed to prevent open circuits. For instance, ammeter
switches have overlapping contacts so that the circuit is never opened as the ammeter is transferred
from phase to phase, and drawout relay cases have shorting contacts in current circuits so that the CT
circuits are shorted before the relay coil is removed from the circuit. Also, CT secondary circuits are often
wired to special terminal blocks which allow a short circuit to be placed on the CT secondary if it is
necessary to service the secondary circuit.

For those users who wish even further protection against the possibility of an accidental open circuit in a
CT secondary, open circuit protectors are available. These are useful especially where the CT secondary
leads are long and subject to possible damage, such as in substations where CT leads from high voltage
circuit breakers or transformers may runs as much as several hundred feet to reach the secondary
devices. Some users require these devices in metal-enclosed switchgear, but only about 5% or less of
the switchgear we build has these protectors.

There are two basic types of protectors available - variable resistance and electronic. They both work by
limiting the CT secondary voltage, but in very different ways. The variable resistance type carries enough
current to limit the voltage across the protector, much in the manner of an MOV or a surge arrester.
However, in order to protect itself, this device includes a heater element in series with the variable
resistor and a bimetallic contact which will short out the CT secondary before the variable resistor
element is damaged. The electronic type monitors instantaneous voltage, and shorts the CT secondary
through an SCR if the instantaneous voltage exceeds the set point. This short remains for the rest of the
half-cycle of the voltage wave on which it occurs, and is then removed. Each half cycle is separately
monitored and acted upon.

Both types of protector come in several voltage classes. It is very important when applying these
protectors that the proper voltage class be chosen, and coordinated with the operation of the relays at
maximum fault current. This is especially important for CTs connected to high impedance bus differential
relays, which are normally voltage actuated. The protector must operate for a true open circuit, but must
not operate to short circuit the CTs under fault conditions, preventing desired relay operation.

Technical Director

Metal-Clad Switchgear or Metal-Enclosed Switchgear: Which
is it?

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 23, 1996

Would you be surprised if I said it might be both? Metal-clad switchgear is one of three common types of
metal-enclosed switchgear, as defined in ANSI standards. Before I confuse you further, a definition is in
order.

Metal-enclosed power switchgear: A switchgear assembly completely enclosed on all sides and top
with sheet metal (except for ventilating openings and inspection windows) containing primary power
circuit switching or interrupting devices, or both, with buses and connections. The assembly may include
control and auxiliary devices. Access to the interior of the enclosure is provided by doors or removable
covers, or both.

This definition appears in all three of the ANSI/IEEE standards covering the types of metal-enclosed
power switchgear: C37.20.1, Metal-Enclosed Low Voltage Power Circuit Breaker Switchgear; C37.20.2,
Metal-Clad Switchgear and C37.20.3, Metal-Enclosed Interrupter Switchgear. Let's look at some of the
things that define each of these types.

Metal-enclosed low voltage power circuit breaker switchgear is obviously for use on low voltage systems.
The maximum ratings in C37.20.1 are 635 V for ac switchgear and up to 3200 V for dc switchgear. The
interrupting device is a low voltage power circuit breaker, either withdrawable or stationary. It may be
either manually or electrically operated, fused or unfused. Each circuit breaker is enclosed in a grounded
metal compartment, but other isolation, such as between buses and cable connections, is not required.
Shutters are not required over bus connections when drawout breakers are used, and bare bus is
standard. While some of these features may be available as options (Powell can provide all of them),
they are not required by the standard. Under the ANSI standards, low voltage switchgear cannot be
considered metal-clad.

Metal-clad switchgear requires that the main switching and interrupting device be drawout. It may be
either a circuit breaker (usual) or a load-break interrupter switch (unusual). Circuit breakers are always
electrically operated. Extensive barriering, shutters over the primary circuit elements when the interrupter
is withdrawn, and insulation-covered bus are all required. The minimum rated voltage listed in C37.20.2
is 4.76 kV. This rating is commonly used on 2400 V ac circuits, but is seldom used at lower voltages.
Ratings are available up to 38 kV. Metal-clad is usually considered the top of the line of medium-voltage
metal-enclosed switchgear.

Metal-Clad Switchgear or Metal-Enclosed Switchgear: Which
is it?

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Metal-enclosed interrupter switchgear covers the same voltage range as metal-clad, but is of simpler
construction. The main switching device is an interrupter switch, usually load-break, which may be
stationary (usual) or drawout (unusual). Overcurrent protection is usually provided by fuses. Bare bus is
standard and required barriering is minimal. Although electrical operators are available, the switches are
usually manually operated.

Technical Director

Enclosures for Metal-Enclosed Switchgear and Motor Control

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

July 12, 1996

The enclosures which are a part of metal-enclosed switchgear and motor control equipment furnished by
Powell serve two major purposes. First, they protect users of this equipment from injury by preventing
access to live parts by foreign objects, including parts of the human body. Second, they protect the
working parts of the equipment from the effects of the environmental conditions in which the equipment is
installed. However, while performing these two duties the enclosures also allow reasonable access to
parts for operation and maintenance, and must allow ventilation adequate to keep the equipment within
its temperature limitations while operating.

There are three major sources of enclosure information applicable to metal-enclosed switchgear and
motor control: ANSI/IEEE, NEMA, and IEC standards.

ANSI/IEEE C37 Series: C37.20.1, C37.20.2, and C37.20.3 are the basic standards for metal-enclosed
switchgear. These standards recognize only two types of enclosure, indoor and outdoor.

NEMA 250: This standard lists numerous varieties of enclosures for electrical equipment, but applies
only to equipment rated 2000 V maximum. It is the source of the definitions for the familiar NEMA 1,
NEMA 3R, NEMA 4, etc., enclosures.

IEC 529: This standard also lists numerous varieties of enclosures for electrical equipment, and there is
no voltage limitation. It is often referred to as the "IP Code", because each enclosure type number is
preceded by the letters "IP" (for International Protection).

IEC 694: This standard covers high-voltage switchgear and controlgear. It refers to IEC 529, but
specifically limits high-voltage switchgear and controlgear enclosures to enclosure types with no degree
of protection against harmful ingress of water (second characteristic numeral X). Outdoor equipment
uses the suffix W.

Each standard defines its various enclosure types and prescribes appropriate tests to demonstrate the
effectiveness of the enclosure. Unfortunately, there is no exact equivalence between the enclosure types
in the various standards, but the following table shows a comparison of the closest types in each
standard.

Enclosures for Metal-Enclosed Switchgear and Motor Control

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
ANSI/IEEE C37 Series NEMA 250 IEC 529 IEC 694
Indoor NEMA 1 IP20 IP2X
Outdoor NEMA 3R IP24 IP2XW

In addition to the standard indoor and outdoor enclosures, Powell offers arc resistant indoor enclosures
for PowlVac
metal-clad switchgear and weather-proof Power Control Room (PCR
) enclosures for all

types of equipment.

Technical Director

Overlap and Bolting of Bus Connections

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

July 29, 1996

From time to time questions arise about bus overlap dimensions and the number of bolts required in a
bus joint. As pointed out in PTB 24, the factor which determines the adequacy of bus work is the
temperature rise. If the temperature rise is less than the limits given by the relevant product standard, the
bus work and its joints are satisfactory. If the temperature goes too high, no amount of overlap or number
of bolts makes the joint okay.

When two bus bars are bolted together, the current transfer from one bar to the other takes place at a
number of locations where microscopic projections on the surface of the two bars are deformed by the
pressure of the bolted joint. The great majority of these points of deformation take place under the
hardware. If you project a line outward a 45 from the edge of the flat washer through the thickness of
the bus bar, as shown by the dotted line in figure 1, you define a surface area where effective bar-to-bar
conduction takes place. This is shown by the heavy line in Figure 1. As long as the overlap of the two
bus bars is sufficient to cover this area, the bolted joint will be effective, and additional overlap area is of
little importance.

Figure 1. Bolted Bus Connection

In order to create the pressure necessary to deform the micro projections on the bus bars, the bolt torque
must be rather high. Proper torques are given in PTB 53. To achieve high torques and large contact
areas, it is preferable to use a few large bolts rather than many small ones. For most main buses in
switchgear and motor control, we prefer to us 1/2" bolts. It is our experience, based on numerous
temperature rise tests, that 1/2" bolts make a very satisfactory joint for bus rated up to 4000 A. Since
most joints are made by bolting a bus bar to a splice plate or a riser bus, and then bolting the next
section of bus to this same splice plate or riser bus, the complete bus-to-bus splice will normally have
four bolts.

Copper bus bars are normally plated with silver, tin, or nickel at all joints in order to prevent the formation
of copper oxide in the joint. Copper oxide is a semiconductor, and its presence leads to increased
resistance and high temperature in the joint.

Technical Director

Bus Duct Enclosure Material

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 19, 1997

Bus ducts built by Powell are metal-enclosed, in accordance with ANSI/IEEE Standard C37.23. The
metal enclosure of the bus duct, like the metal enclosure of switchgear units, performs several functions.
First, it protects personnel from the hazard of contact with energized electrical conductors. Second, it
provides mechanical support for the conductors and their insulation system. And third, it protects the
conductors from mechanical damage and environmental distress.

The first material we think of when talking about metal enclosures is steel ordinary, every-day hot-
rolled carbon steel. However, steel has some drawbacks as an enclosure material. Most importantly, it is
magnetic, and the magnetic field surrounding the conductors induces circulating currents in the
enclosure. These currents have a couple of ill effects. They represent energy losses as they flow through
the resistance of the steel. This, in turn, raises the temperature of the enclosure, reducing its
effectiveness in dissipating the heat generated in the bus bars, making the entire bus duct run hotter.
This effectively reduces the continuous current rating of the bus duct, since the continuous current rating
is determined by the temperature rise of both the conductors and the enclosure.

The usual method of reducing these magnetic effects is to make at least one side of the bus duct
enclosure of a non-magnetic material, often aluminum. Powells usual practice is to make the top cover
of bus ducts rated over 1200 A of aluminum. At some very high ratings, the entire bus duct enclosure
may be made of aluminum.

Another drawback of steel is that, even with a good paint job on galvanized steel, it may not stand up
well in some harsh environments. Powell routinely uses galvanized steel for all outdoor bus duct
enclosures. To a lesser degree, aluminum may also have environmental problems. If steel is not suitable,
but aluminum is, an all-aluminum bus duct enclosure is preferred.

However, if neither galvanized carbon steel nor aluminum is suitable for the environment, a third
alternative is to make the bus duct enclosure of stainless steel, which withstands bad environments well
and is also non-magnetic. The drawback to using stainless steel is its cost. The basic raw material is
much more costly than either carbon steel or aluminum, and considerably greater effort is required to
fabricate it. If the circumstances require it, Powell will furnish bus duct with stainless steel enclosures, at
an appropriate price adder. Because of this increased price, a stainless steel bus duct enclosure should
not be specified unless there is a real need for the material.

Technical Director Emerititus

Insulation of Bus Joints

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 20, 1997

One of the defining features of metal-clad switchgear built according to ANSI/IEEE Standard C37.20.2 is
that all power circuit bus work, including both factory-made and field-made joints, is covered with
insulating material. The insulated bus, including the joints, must pass a power-frequency voltage test for
one minute at rated voltage. This voltage is applied between the insulated conductor and an electrode,
usually a foil wrapping, on the surface of the insulation. The insulation decreases the likelihood of
accidentally starting a fault with a misplaced tool (or body part). If a fault does occur, due to a break in
the insulating system, the fault is usually kept from traveling the length of the bus because the insulation
keeps the arc rooted in one place. And finally, the insulation allows reduced spacing between the bus
bars for a given B.I.L. rating, allowing more compact switchgear.

However, taking advantage of this smaller size equipment requires insulation of all joints, both in the
main bus and at cable or bus duct terminations in the switchgear. Over the years a number of methods of
insulating joints have been used, including the following:

1. Taping. This is probably the oldest method of insulating a joint, and it is still used in some cases. It is
slow, often taking several hours per joint, and a good joint is highly dependant on the skill and care of the
person doing the taping. However, some joints are so complex that taping is about the only way to
insulate them.

2. Asphalt-filled plastic boxes. 45 to 50 years ago, many main bus joints were insulated by enclosing
them in hard plastic boxes and pouring the box full of an asphaltic compound. This was a messy
process, and the materials used were neither flame retardant nor track resistant. It was especially messy
when a joint had to be opened and cleaned for some reason. The method was abandoned before 1960
for new equipment, but there are probably many of these joints still in service in older switchgear.

3. Flexible boots. Boots, molded to fit the shape of the joint, are the most common method of joint
insulation in switchgear up to 15 kV. The great majority of these boots are molded of PVC, but other
materials have been used. The boots can be installed quickly and easily, removed readily, and reused.
At voltages above 15 kV, however, it is very difficult to get these boots to pass the test for bus bar
insulation required by the standard.

4. Heat and cold shrink materials. There are a number of heat shrink and cold shrink products
available. These generally do a good job on straight in-line joints, but can be difficult to use on joints with
a more complex shape.

Insulation of Bus Joints

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
5. Poured joints. For our PV System 38
switchgear, Powell has developed a system of poured joints,

using removable molds and a clear urethane insulating material. This joint insulation method does meet
the bus bar insulation test at 38 kV, and provides for visual inspection of the bus joint.

Please be sure to insulate the bus joints when you install your metal-clad switchgear.

Technical Director Emerititus

Circuit Breaker Trip Defeat Switch

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 1, 1997 (corrected)

The Circuit Breaker Trip Defeat Switch allows the owner to take the functional testing of the protective
relay scheme and extend it all the way to the trip coil without opening the circuit breaker. Thus confirming
that the relays output contacts will indeed pick-up the lockout relays and that the lockout contacts will
energize the trip coil of the circuit breaker. This action combined with a healthy coil monitor device or a
healthy coil light tests the circuit to the final element without opening the circuit breaker.

With the increased reliability of relaying and the pressure for extending the maintenance intervals: more
equipment owners are asking for the capability to do final element functional testing on the circuit
breakers feeding loads that they can not afford to de-energize on a regular basis. As the failure rate of
relays goes down, the equipment owner would prefer to functionally check the protective relay scheme
and setpoint yearly, and then pull the relays for a full calibration check every second or third year.

The Circuit Breaker Trip Defeat Switch requires a two position control switch (95 device). Position one
is the normal trip mode (see sketch). In this position the trip coil is in the circuit. The second position is
the trip defeated mode. In the trip defeated position the circuit breakers trip coil is cut out of the circuit
and replaced with an indicating light and a dropping resistor. The dropping resistor is sized to make the
circuit draw the same amount of current as the trip coil. Due to the critical nature of defeating the trip coil
a separate contact should be wired to the substation annunciator and to an amber light mounted in the
front of the breaker panel to indicate when the breaker is in the trip defeated mode.

With the switch in the trip defeated position the technician can function the relays -- tripping lockouts
and see that the lockout would have energized the trip coil via the indicating light. Once the testing is
complete the technician confirms that the lockouts are reset, the trip light is de-energized then transfer
the Circuit Breaker Trip Defeat Switch back to the normal mode.

Now for my note as a circuit breaker manufacturer. This scheme allows a much more comprehensive
check of the protection system than we have done in the past, but dont forget the circuit breaker, Yes,
we have tested the protective logic all the way to the trip coil but the breaker has just gone another year
without functioning and the grease without exercise hardens over time. Granted in an air-conditioned
substation the interval between maintenance can be extended but not indefinitely.

If we can help with this or any other topic please don't hesitate to call.

Circuit Breaker Trip Defeat Switch

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Jim Bowen
Technical Director

Ferroresonance in Voltage Transformer (VT) Circuits

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 1, 1997 (corrected)

In the last couple of months we have received several different questions as to what is ferroresonance in
a VT circuit, when does it occur and how do we protect against it.

Ferroresonance can occur when the primary of a voltage transformer is connected line to ground in an
ungrounded circuit. This configuration results in the magnetizing reactance of the VT being in a parallel
loop with the coupling capacitance to ground of the system (see Figure 1).

Figure 1

The coupling capacitance is primarily made up of the capacitance of the system dielectric between the
phase conductor and ground. The value of the voltage transformers magnetizing reactance varies as a
function of the amount of flux going through the iron. This results in an LC circuit and requires only a
simple voltage transient to excite the resonant frequency. Once the ringing begins the voltage across the
individual components of magnetizing reactance and coupling capacitance can reach high levels and the
ringing can go undamped if the voltage transformer is lightly loaded. The loading of the VT has a very
important part to play in limiting the magnitude of current in the oscillation circuit since the resistance of
the load will act as a current divider and send a portion of the current to ground. This graph from the
IEEE Red Book shows the impact of load on the magnitude of the current in the ringing circuit (see
Figure 2).


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Figure 2

During the oscillation, the current can drive the magnetizing force to saturate the VT. When the VT is
saturated, the reactance to ground will diminish and the current to ground through the primary of the VT
will go high. At the end of the sinusoid the VT will drop out of saturation, but with a low loss system the
stored charge remains relatively high across the system coupling capacitance. As the polarity of the
sinusoid changes the process repeats itself. The current surges, through the VT primary during the
periods of saturation, can be much greater than full load rating but not approaching fault current levels,
making it very difficult for the fuses on the primary of the VT to interrupt. Thus current surging may result
in a blown VT fuse but often results in a shorted VT.

To keep the resonance magnitude down, the secondary side of the VT circuit can be artificially loaded.
There are two common methods of loading used to minimize the effects of ferroresonance. One is to
install the VTs with their secondary windings connected in a broken delta and with a resistor completing
the broken delta circuit. The watts of the resistor should equal 50% of the VA of a single VT.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3
The second and most popular method is to put a resistor across the secondary of each VT. The rule of
thumb from several old references is that the resistive loading should range between the VA required to
excite the core at no load and 50% of the thermal rating of the VT. For specific VTs, the manufacturer
can recommend a precise value of resistance.

Due to the varying frequency of the transient and the magnetizing reactance this is not a problem that
occurs in every system or even every time a voltage transformer is connected to ground on an
ungrounded system. If the resonant frequency of the LC circuit is excited the swamping resistor will
dampen the ringing to prevent long term effects.

If we can be of help on this or any other topic please don't hesitate to call.

Jim Bowen
Technical Director

Switchgear in a Sulfur Rich Environment

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 27, 1997

Sulfur rich environments such as those found in paper and refinery processes have a special problem
with the silver plating in Switchgear and Motor Control Centers. Silver plating is used throughout
switchgear because of the superior conductivity and longevity. The silver is found on the bus, in the
circuit breaker, in protective relays, auxiliary relays, control switches, and test switches. In and around
process units with sulfur present it is very common to open up the switchgear and see fine black hairs
commonly referred to as whiskers growing from any and every silver plated surface, - be it a relay
contact or a breaker stab.

These whiskers and the black tarnish forming under them are silver sulfide. The whiskers are semi
conducting and the tarnish is highly resistive. The silver sulfide tarnish grows in areas of the switchgear
where the highest concentrations of sulfur is exposed to heat and since the hottest areas are bus joints
and sliding contacts, such as the bus stabs, this is not good. The tarnish at the splices and sliding
contacts result in a high resistance connection, which produces more heat, and in turn accelerates the
tarnishing and the growth of the semi conducting whiskers. This death spiral continues on until you clean
the surfaces or the whiskers get long enough to reach a ground plane.

How does the sulfur get in?

The sulfur from the process combines with air to form Hydrogen Sulfide (H
2
S) with some Sulfur Dioxide
(SO
2
) and Sulfur Trioxide ( SO
3
) by-products. ISA Standard 70.01 defines a harsh environment as a
concentration of 50 ppb (parts per billion) of H
2
S or 300 ppb of SO
2
and SO
3
. Your nose can give you
some idea whether sulfur is present. The odor threshold is down around 8 ppb for the H
2
S. But the best
way to detect sulfur is by surveying the existing gear for the problems described above and in the case of
grass roots facilities discussing with your environmental people the types of fugitive emissions expected
from the new units.

To combat the effects of sulfur on switchgear there are a couple of steps that can be taken. These
include:

Chemical Filtration

One of the most effective ways of combating the whiskers is to filter the H
2
S, SO
2
and SO
3
out of the air
in the switchgear room. If the Switchgear and MCCs are going into a separate Power Control Room the
air conditioning can be fitted with activated carbon filters. These filters are housed in a separate box
about the size of one of the air conditioning units and reduce the ISA harsh environment levels such as
the 50 ppb of H
2
S to an H
2
S concentration of 3 ppb and SO
2
and SO
3
to 10 ppb. The unit does require
some maintenance and the filters have to be changed approximately once a year.

Switchgear in a Sulfur Rich Environment

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

The next level of protection is to fit individual active carbon filters over all switchgear louvers. These
individual filters are disposable and can do a good job of filtering whatever portion of the air that goes
through the louvers. While the individual filters do have a significant impact on how often the switchgear
has to be maintained it will not protect the equipment in the instrument compartment or cubicles where
there is not a definite air entry, so their overall effectiveness is limited.

Exposure Avoidance

The other strategy commonly used is to minimize the amount of silver exposed to the sulfur by specifying
tin plated bus in place of the silver and gold plated contacts or hermetically sealed relays where
available. Conducting grease applied at any sliding contact points can help to seal out the sulfur and
stiffer springs to improve the wipe can also assist in fighting the problem. In the areas where we replaced
the silver with either tin or gold we have eliminated the chance for the whiskers to grow. But there are
difficulties with this strategy, many control switches, test switches and protective relay contacts are not
available with gold plated contacts. And tin has a couple of problems such as galling and softness that
make it less than ideal for sliding contact applications. See PTB 41 "Plating of Contact Surfaces in
Switchgear and Circuit Breakers".

It is well worth while to evaluate the cost of these different options on any job where H
2
S or SO
2
and SO
3

may be present. For those, who have experienced these problems first hand you know that this is more
that just a shorter maintenance interval. The whiskers are not only a threat to the power circuit they can
be rather insidious in affecting the functionality of the protective scheme.

If we can help with this in any way please give us a call.

Jim Bowen
Technical Director

Direct Control of Motor Contactors via PLCs and
Distributive Control Systems

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 28, 1997

As we automate the process systems more and more we are looking for direct control of the motor
starters with PLCs and Distributive Control Systems (DCS). A commonly asked question is whether the
interposing relay situated between the two systems is really needed or not. I have gone through and
looked at the more common systems and put together this technical brief to highlight some of the
concerns involving the interposing relay.

Digital Outputs Modules

The electro-mechanical and the solid state (triac) output modules are commonly available for either
PLCs or DCSs. Both of these digital output modules are listed in the vendor literature as having a 120
Vac rating of such and such current. Two amps seem to be the most popular for the electro-mechanical
value and amp for the solid state output. The thing I had difficulty finding was the expected power
factor of the load. A power factor of 0.5 seemed to be the most common although I did find one
manufacturer who based his rating on a power factor of 1.0. The expected power factor obviously
changes from one manufacturer to the next for both PLC and DCS. The normal contactor coil has a
power factor of 0.1, at this level the output rating should be derated by 50%. This is without the
consideration of the wire impedance connecting the two systems.

In addition to confirming the make and break capability, the application engineer should realize that the
turn off of the standard starter results in an inductive kick that sends a significant voltage spike though
the system. The output relay ends up with contact pitting and the contactor coil receives a steep fronted
voltage spike which shortens the life of the insulation. The spike can be limited by installing a snubber
circuit across the output contacts. For solid state outputs the snubber is a 0.1 microfrad capacitor and a
100 ohm resister series together. The cap should have about a 400 Vdc rating for a 120 volt control
system. For the electromechanical output internal protection is usually provided for loads up to 1 amp
after that you need to provide it separately. In most cases the manufacturer has a kit available as an
option.

The triac outputs add an additional level of possible problem due to the triacs leakage current during turn
off and forward bias requirements. The forward bias means that the triac must maintain some minimum
current flow to keep the device in the on state. The leakage current is a small quantity of current that
trickles through the device even after turn off. This few milli amps will be in parallel with the distributed
wire capacitance of the wire between the control system and the MCC to make the turn off a significant
problem. This appears to make the triac an undesirable combination with the standard contactor.

Direct Control of Motor Contactors via PLCs and
Distributive Control Systems

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
Smart Motor Control

When using many of the new smart motor control centers the control circuit is no longer interrupting the
coil current directly, if it is wired properly! So the concerns relative to interrupting a large inductive load
are eliminated. The electro-mechanical relay output is less likely to have problems getting the two
systems to match-up.

The reliability of the triac output module will allow it to do more operations successfully than the electro-
mechanical, but since the coil is not operated directly, you may have to add a swamping resistor in
parallel with the MCC input to make sure the circuit draws sufficient current to keep the triac forward
biased. In addition, the high relay input impedance of the smart MCC and the distributed wire
capacitance may begin to play a part in the functionality of the stop circuit. The longer the length of the
control circuit the higher the leakage current and the more likely resistance may need to be added to
assure that the circuit will recognize a stop command.

A separate concern about doing away with the interposing relay you may want to discuss with the end
user is the safety aspects of bringing 120 volt control power from the MCC starter drawers into the PLC
or DCS. The way most plant treat the insides of their PLC and DCS is as though there is no voltage level
greater that 50 volts to ground. At 50 volts and below OSHA Standard 1910 and NFPA 70Es Electrical
Safe Work Practices & Working Space Requirements need not apply. But by bringing the 110 Vac into
the control system the rules identified in OSHA 1910 are applicable. In some cases this can have a
tremendous impact on the cost of doing work.

Based on the legwork I have done the interposing relay still makes good sense until the confidence in the
twisted pair communication allows us direct communication between the smart MCC and the PLC or
DCS, with a stop station in the field next to the load.

If we can help with this or any other topic please do not hesitate to call.

Jim Bowen
Technical Director

Physical Installations of Surge Arresters

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 16, 1998

Surge arresters are commonly used in switchgear today to clamp transient voltages at levels below the
allowable BIL of the equipment being fed and the switchgear feeding it. The arrester plays a special part
for motors and generators in extending the equipment life under certain system configurations. Metal
Oxide Varistors (MOV) are commonly used in switchgear and medium voltage motor control centers (MV
MCC). The metal oxide arresters offer smooth turn-on and turn-off at lower surge current levels than the
older valve type silicon-carbide technology. To install MOV surge arresters in switchgear and MV MCC
there are several physical aspects to be considered to assure reliable operation. The most significant of
these are ambient temperature, the lead length of the conductor connecting the arrester to the
equipment being protected, and the spacing between the arrester in two separate phases.

Ambient temperature The Metal Oxide Varistors are rated for a -40C to 40C ambient with a
temporary maximum air temperature of 60C. As the temperature of the device increases -- the leakage
current continues to climb. The increase in leakage current causes the surge arrester temperature to
exceed the temperature rating of the MOV. The internal resistance of the surge arrester begins to drop
off and the device turn-on voltage is lowered. If the temperature exceeds the thermal capability of the
arrester, thermal runaway occurs and the arrester turns on at the system operating voltage and will not
turn off, thus resulting in a line to ground fault.

Cable connection tail length It is critical that the length of the leads connecting the surge arrester to
the protected equipment and the arrester to the ground bus must be minimized. The wave continues to
travel down the system past the surge arrester near the speed of light at the original surge voltage until
the arrester turns on to clamp the voltage at the arrester discharge voltage. The voltage drop of the lead
wire connecting the arrester to the equipment being protected and the arrester to the ground plane
adversely effects the discharge voltage of the surge arrester. A rule of thumb is that the arrester
discharge voltage is decreased by 1.6 kV/ft. for every foot of lead length between the phase conductor
and the ground bus. The decreased discharge voltage becomes more important in systems above 15 kV
where BIL coordination margin is reduced below the 20% protection margin recommended by ANSI.

Skirt to skirt distances The surge arrester is made up of two dielectric materials in parallel. The outer
surface is typically a polymer insulating material that creates a voltage gradient along the length of the
arrester. This surface distance is commonly referred to as the creepage distance. The second part is the
inner portion of the arrester, the MOV disk, an insulator at nominal voltage and a conductor at higher
voltages. To prevent conduction across the surface (tracking), adequate phase to phase and phase to
ground spacing must be maintained. A skirt to skirt spacing of 1 inch/30 kV of BIL is recommended.

Physical Installations of Surge Arresters

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Conductor to conductor distances -- In additional to skirt to skirt spacing, it is important that phase
spacing be maintained between the conductors throughout the lineup. The conductor connected to the
lug on top of the arrester is at line potential and must maintain the same phase spacing as the bus phase
conductors (see PTB #59 for distances). If the switchgear spacing is based on insulated phase
distances, the lug connection on top of the arrester must be booted. The boot fit should barely cover the
first skirt ring, and should form a continuous path from the conductor to the first ring of the arrester. To
keep from "shorting out" the creepage distance, the boot may not make contact with more than the first
skirt.

These physical considerations play an important part in assuring that the arresters are able to function
properly.

Jim Bowen
Technical Director

Additional Safety Features

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 3, 1998

There are several optional features that our clients specify to enhance the overall safety of the
switchgear for the individual electrical technician. This PTB will begin a series highlighting some of these
features so that others might benefit from the collective experience.

Three of these topics that were discussed at the last PCIC Safety Workshop were:

Shutter labels:
The simplest enhancements to add to a switchgear line-up are shutter labels. The shutter label indicates
to the technician performing testing or grounding on a vertical section the destination of the top and
bottom stabs. The shutter is the moveable guard that drops in front of the breaker stabs as the breaker is
racked to the disconnected position. The labels are decals mounted on the shutters in front of the circuit
breaker stabs. The labels identify whether the stabs are "Load Side", "Line Side" "Bus Side A" or "Bus
Side B".

Tasks often require an electrical technician to open the shutters on an energized cell. Whether the
shutters need to be opened to megger a motor feeder or to insert a ground and test device it is important
that the technician be given visual confirmation of which set of stabs are energized. Yes, the safe-work
practice requires that the stabs be checked for voltage prior to hooking up the test equipment. But, this
simple label offers a valuable confirmation to the technician in the field that has proven to be effective.

Shutter locks:
The shutter mechanism is the last level of protection between the stabs and a person doing work in the
cell. By padlocking the shutter closed you protect technicians from mistakenly opening a shutter on an
energized set of stabs. Our existing shutter mechanisms have a set of holes to allow the shutters to be
padlocked in the closed position. We also have an optional design that brings a bar from the shutter
mechanism to the very front of the cell. This extension design allows the shutter to be the primary point
of Lock and Tag out.

Once again this is something that is covered by the plants safe-work practices. Every safe-work practice
says assume everything is energized before you touch a conductor. But we have had another case here
in the Gulf Coast region just recently of a individual getting electrocuted on an energized stab while doing
preventative maintenance. The lead technician was performing preventative maintenance on a
secondary selective system. He had performed the proper isolation and lock out procedure. As planned,
he had left a load side CPT energized via a down stream emergency generator to provide station service
power for the shut down. The technician was going down the line up cleaning all the breaker stabs when
he mistakenly went into the cubicle with the load side stabs energized and was killed when he came in
contact with the stabs.

Additional Safety Features

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Because of other work going on, the group required access to the cubicle so they had to be able to leave
the cubicle door unlocked. A simple lock and tag on that particular set of shutters would have prevented
the technicians mistake. There is a pair of 3/8" holes through the moving and fixed portion of the shutter
mechanism that permit the locking of the shutter. This locking mechanism also proves to be useful with
any main-tie-main system. The shutter lock is the best system available for protecting people when the
switchgear has a tie cubicle and half of the system is out of service for maintenance. The shutter lock is
also a very effective point for locking out the breaker and cell.

Cell locks:
The most discussed topic when drafting a site Lock-out and Tag-out Procedure is where to place the
locks on metal-clad switchgear. Locking out the cell is replacing locking out the circuit breaker due to the
increased safety. Locking out the cell assures that a spare breaker cannot be racked in and mistakenly
energize downstream loads. A cell lock allows full access to the breaker out of the cell on the floor for
maintenance purposes while people continue to work under their lock-out and tag-out on downstream
loads. The cell lock absolutely prevents any breaker from being racked onto the stabs.

In all cases the shutter labels, shutter locks, and cell locks can play an important part in how the
switchgear is operated. Every site has different skill levels and site procedures that determine when and
if these features should be incorporated into the site safety program.

Jim Bowen
Technical Director

01.4TB.084

Capacitor Trip Unit

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

November 17, 1998

A capacitor trip unit is a prepackaged module that supplies power for tripping an AC controlled circuit
breaker with discrete relays following the loss of the AC control voltage. DC control utilizing a charger
and battery bank is the more reliable method of supplying tripping power but, in installations of only one
or two circuit breakers, sometimes it is difficult to justify the higher cost of the battery system.

The capacitor unit has a blocking diode to maintain the storage capacitor charged at the peak AC
voltage. In case of loss of AC the blocking diode prevents the capacitor from discharging due to
upstream loads. The standard product holds sufficient charge to trip the breaker for 12 seconds after loss
of AC voltage. The capacitor trip units are also available in a battery-assisted model. This model protects
against losing power for time period up to 2 days by having a small gel cell battery support the voltage.
Due to the long charge retention time these unit are usually supplied with a toggle switch to disconnect
the unit and discharge the capacitor to allow trip circuit maintenance.

Once the load is connected, the stored 30 watt-seconds of energy dissipates very quickly. The load is
typically either a lockout relay or a circuit breaker trip coil. One capacitor trip unit should be provided per
coil load. For example if you have two lockout relays and a trip coil -- this circuit requires three of the cap
trip units. The unit can not support indication lights, healthy monitoring relays or any other such load
continuous loads, as they would drain the energy stored in the capacitor when the source voltage is lost.

There are no set points to the cap trip unit and most of the designs do not permit monitoring relays to
warn that the capacitor is still working. It can hold a charge for a surprisingly long period of time,
therefore it is important that the user have a written procedure for discharging the unit by jumpering the
capacitor prior to working on the control circuit to prevent an electric shock hazard.

The most effective way of discharging the capacitor is to utilize the unit energy to open the circuit
breaker. To accomplish this, disconnect the AC control power and operate the circuit breaker via the cap
trip unit. In so doing not only do you partially discharge the capacitor but you also get a functional test as
a bonus. The capacitor may still have some residual charge and needs to be discharged prior to touching
any conductor in the trip circuit. Discharge the additional stored energy by installing a jumper with a
resistor in series. The battery assisted units have a disconnect switch wired into the front of the unit.

Obviously for you to disconnect the AC you need a disconnecting means for the AC. This can be the pull-
apart fuse block, the circuit breaker panel or a local knife switch depending on how the circuit is set up.

01.4TB.084

Capacitor Trip Unit

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2
If site safety requirements make it difficult for you to install the jumper on a circuit that may be in excess
of 50 volts to ground you may decide to have a push button installed to discharge the remaining
capacitor charge. I do not recommend a maintained switch, as there is too great of a chance of
energizing the circuit with the capacitor trip circuit shorted.

In closing remember there is a limited life to the battery assisted capacitor trip units. With either type of
device it is critical that you include this device into the site maintenance plan.

J im Bowen
Technical Director

The Application of 600 Volt Class Current Transformers in
Medium Voltage Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 10, 1999

ANSI Medium Voltage Switchgear commonly uses 600-Volt Class window type current transformers in
medium voltage circuits. For many years manufacturers have installed 600-volt class current
transformers in medium voltage switchgear. The preferred location for the current transformers is the
circuit breaker spouts.

To certify the dielectric capability of the equipment the switchgear type tests are performed with the
maximum number of current transformers mounted on the circuit breakers cells feed through bushings.
A typical configuration of the switchgear with the smallest air gaps is then subjected to the series of Basic
Impulse Levels Tests required by ANSI. The success of the BIL testing in the area around the circuit
breaker feed through bushings relies on the following combination of insulation:
The solid dielectric of the circuit breakers feed through bushings
The air gap between the conductor and spout
The air gap between the spout and the case of the current transformer
The 600 volt class insulation of the current transformer
This series combination of various insulating mediums provides an insulating system that goes through
the dielectric testing along with the entire switchgear system. During the power frequency test (high
potential) and the impulse test required in ANSI C37.20.2, it is proven that the system dielectric strength
exceeds the nominal BIL rating for the voltage class.

Bus Mounted Current Transformers
Occasionally the clients current transformer requirements exceed the physical mounting capacity of the
circuit breaker spouts. In these cases it becomes necessary to add additional window type current
transformers. These additional current transformers are mounted in the cable compartment around the
run-back bus.

Since these are non-standard components, they do not get included in the switchgear BIL testing. To
maintain the BIL of the equipment an industry rule of thumb for both factory and field installations
requiring bus mounted window type current transformers is to allow a 1 inch air gap between the
insulated bus and the current transformer housing.

Powell went through the process of testing the 1-inch air gap rule of thumb with various bus insulation
systems that we use in the manufacture of the switchgear. We have established an internal matrix that
applies to the various insulating systems, the associated standard current transformer and a special
"increased dielectric" current transformer. The table below indicates what configurations were acceptable
when tested for 95kV BIL.

The Application of 600 Volt Class Current Transformers in
Medium Voltage Switchgear

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Table of BIL Tests on 6.5 Inch Bus Mounted Window Type Current Transformer
Insulation
Material
Number of Bus
Bars per Phase
and Size (inches)
Standard Current
Transformer
Increased Dielectric
Current Transformer
Hipoxy-2000
(1) 1/2 x 3 Passed Passed

(1) 1/4 x 4 Failed @ 82kV Passed
GE-Noryl (1) 1/4 x 4 No Test Passed
(2) 3/4 x 4 No Test Failed @ 94kV
Scotch BBI-4A
(3) 3/4 x 4
(sandwiched)
No Test Passed
Note: Standard current transformers would be suitable for all switchgear designs requiring 60kV.

Jim Bowen
Technical Director

01.4TB.086

Altitude De-rating of Fuses, Surge Arresters and Potential
Transformers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 11, 1999 See revised PTB #106 (September 20, 2012)

As a dielectric dry air works satisfactorily. However as the altitude at which we apply our electrical
equipment increases, the effectiveness of the dielectric property decreases. An increased altitude also
decreases the continuous current carrying capability of electrical equipment due to the reduced thermal
conductivity of the lower density air.

In the past we have discussed the altitude de-rating factors for switchgear (see PTB 25). The IEEE
Standard for Metal Clad Switchgear (C37.20.2) explains that switchgear assemblies, which depend on
air as an insulating medium, will have a lower dielectric withstand capability when operating at altitudes
above 3300 feet. This Powell Technical Brief will explain how altitude de-rating effects the installation of
medium voltage fuses, surge arresters, and voltage transformers.

Fuse De-rating
The IEEE Standard for Conditions and Definitions for High Voltage Fuses, Distribution Enclosed Single
Pole Switches, Fuse Disconnecting Switches, and Accessories (C37.40 section 2.3) offers de-rating
factors for high voltage fuses. The de-rating is a function of the ability of the fuse to clear a fault and
achieve a dielectric across the fuse element gap faster than the system voltage across the fuse can be
established. This de-rating would be applicable for non-sealed fuses like expulsion fuses and current
limiting fuses, where the outside air is the insulating means used to isolate the line and faulted load.
Since the dielectric strength of air is reduced as the altitude is increased, the fuse has to be de-rated in
accordance with the published chart. (see Table)

The thinner air also results in lower thermal conductivity, which requires a de-rating of the continuous
current. To compensate for the reduced thermal conductivity the standard allows for a reduction in either
the maximum ambient temperature or the continuous current rating, but not both.
Altitude in Feet
Dielectric
Strength
Rated
Continuous
Current
Ambient
Temperature
0 to 3,300 1.00 1.00 1.00
3,301 to 5,000 0.95 0.99 0.98
5,001 to 10,000 0.80 0.96 0.92
10,001 to 16,000 0.65 0.92 0.848

When the dielectric strength of the fuse is de-rated, the application engineer is often required to choose a
fuse of higher voltage class. This approach may result in higher current chopping and associated voltage
transients. An alternate solution is the hermetically sealed fuse, which does not require de-rating of the
internal operating mechanism. Note that hermetically sealed fuses are both expensive and may have a
long delivery time. Care must also be taken to confirm that the outer creepage distance is acceptable
within the de-rated values.

01.4TB.086

Transformers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Surge Arrester De-rating
There are two types of surge arresters in common use today -- the expulsion-type arrester and the metal
oxide arrester. The expulsion-type arresters have an internal system that depends on ambient air as a
dielectric and, as such, must be de-rated as a function of altitude just as fuses. The metal oxide surge
arresters are sealed and outside air does not play a part in the isolating process, so the surge arrester
does not require de-rating for the internals to operate properly. However, air also plays a part in the
external surface creepage distance of the housing. The outer surface has a certain creepage distance
that prevents the surge arrester housing from tracking across the surface of the arrester. Powell
Technical Brief #59 describes the acceptable surface creepage distance for various impulse levels given
insulated conductors. The insulated conductor distances can only be used if an insulating boot properly
covers the ferrule at the top of the arrester.

Care must be exercised when applying the altitude de-rating factor. To maintain the desired discharge
level of the arrester and the required creep distance, it may be necessary to have special arresters
manufactured with a standard duty cycle arrester in a housing with higher creep levels.

A special caution when applying de-rated higher voltage current limiting fuses. Each fuse has a
characteristic current chop that results in a maximum allowable voltage transient of three times the rated
fuse voltage. The transients that result from applying a higher voltage class fuse due to altitude de-rating
can often result in the surge arrester conducting during a fuse interruption.

This should be avoided for line side station class and intermediate class arresters as the conducting
impedance of this arrester is low enough that conducting during a fault will cause serious damage.
Distribution class arresters have a high enough conducting impedance that this is not a problem.

01.4TB.086

Transformers

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3

Voltage Transformer De-rating
When voltage transformers are applied at higher altitudes, ANSI C57.13, section 4.3, identifies the de-
rating factors. This de-rating requires use of a higher voltage class transformer with the appropriate
primary voltage and the ratio to maintain the desired secondary voltage.

For example: Given a line-up of switchgear operating at 4.16kV at an altitude of 10,000 feet, the
appropriate choice is a voltage transformer with a primary voltage of 4200 Volts and a secondary of 120
Volts. The transformer would have an insulation class of 8.7kV with a BIL at sea level of 75kV. When the
80% de-rating factor is applied due to the altitude, this VT has a BIL of 60kV.

This Powell Technical Brief should clarify some of the concerns expressed by customers on recently
shipped substations installed at higher altitudes. If I can be of further assistance please do not hesitate to
call.

J im Bowen
Technical Director

Current Transformer Grounding

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

May 12, 2000

With the new PowlVac-AT Model ATSB outdoor substation circuit breakers recently introduced by Powell
Electrical Manufacturing Company, several questions should be addressed regarding the ground point
for the Current Transformer (CT) circuits. This Powell Technical Brief investigates the preferred ground
location of typical CT circuits such as transformer and bus differential relays.

IEEE standard C57.13.3 serves as the ANSI guide to standardize instrument transformer grounding
practices. The grounding of current transformers is important to both safety and the proper operation of
the protective relays. To assure the safe and reliable operation, the neutral of the current transformer
secondary should have a single ground location for each circuit. The single ground is irrespective of the
number of current transformers or the chosen grounding location. Utilizing a single ground eliminates the
risk of redundant ground loops and associated problems.

During normal operation more than one ground on a CT circuit is not an obvious problem other than the
difficulties it may cause during testing. However during a fault condition, multiple grounds allow a
different ground potential rise for each current transformer. The result is a significant current flow through
the CT circuit that is not representative of the primary current. This ground loop typically creates a
potential across the operating coil of the differential relay causing the relay to pick up as though a fault
exists in the relay's protective zone. Tripping a differential relay due to a fault external to the zone of
protection, is one of the more popular nuisance trips. These nuisance trips may not only shut down the
load, but may require a maintenance crew to spend days in testing to determine that no real problem
exists in the differential zone. Further, the actual problem may go uncovered until the system is re-
energized into the original fault.

Figure 1
To demonstrate what happens with a second ground on the current transformer circuit, Figure 1 shows a
typical differential relay with two current transformers. The recommended method of grounding is to
install a single ground point at the first point of application (switchboard or relay panel) of the current
transformer secondary circuit.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Figure 2

A substation circuit breaker should have the wye or delta connections made-up at the CT terminal blocks
in the equipment. If the protective relay is mounted in the breaker the wye connection is grounded at the
circuit breaker. If the protective relays are mounted in a separate building, then the wye connection is in
the breaker but is grounded at the relay house. The same applies for current transformers in the
transformer tank. The CTs are grounded at the place where the metering or relaying is located (see
Figure 2).

Figure 3
Note: For a fault, external to the protective zone, insufficient voltage develops across the
operating relay to pick up the coil.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3
For an external fault (see Figure 3), this allows the current flowing through the current transformer on
the line side of the protected zone and the current flowing through the load side current transformers to
develop a voltage of opposite polarities. The result is a voltage of very small magnitude across the
operating coil of the relay. The relay coil does will not operate without sufficient applied voltage.
In the case of a fault internal to the protective zone, the voltage developed by the current transformers is
of the same polarity. The magnitude of voltage drop across the operating coil is sufficient to operate the
relay (see Figure 4).

Figure 4
Note: With a fault in the protective zone, sufficient voltage develops across the operating relay
coil to pick up the coil.
In the final example (Figure 5) there is a second ground is on a current transformer mounted near where
a ground fault occurs. If the fault creates a ground potential rise of 100 volts, then the protective relay will
experience sufficient voltage across the operating coil to cause the relay to nuisance trip even though the
fault was outside the fault zone.

Just as with any other event there is an exception to this standard. Many of the new multifunction relays
(ABB, Schweitzer, GE/Multilin, and Basler) are designed to connect all current transformers coming into
the relay in a wye connection. Each wye has to be grounded. The most desirable way to do this is to bus
the wye points together at the relay panel and have a single conductor to ground, to make certain the
relay has but one ground potential.

As you can see, there are significant considerations in the proper grounding of current transformer
circuits.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 4

Figure 5
Note: With a second ground at a remote location, the voltage across the operating coil is
sufficient to result in a miss-operation for an external fault and cause a nuisance trip situation.

Jim Bowen
Technical Director

Standard Voltage Ranges and Ratings

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

May 8, 2001

Standard equipment voltage ratings and the associated tolerance limits are established by ANSI
Standard C84.1 for electrical systems from 100 volts through 230kV. The question is often asked, How
do established equipment ratings relate to utilization voltage? This Powell Technical Brief explains
equipment voltage ratings, where they come from and how they are related to the utilization voltage.

The voltage variation of a distribution system as a function of the actual load and the impedance
between the source point and the point of voltage measurement are well understood. C84.1 identifies
several of definitions necessary to understand the equipments voltage ratings and the systems
operating ranges.

These definitions include:
System Voltage The voltage bounded by the step-up and step-down transformer
voltage, e.g. 240V, 480V, and 600V.
Maximum System Voltage The highest voltage at which the system will operate under
normal conditions. It is the greatest voltage for which the equipment is designed to operate
continuously without de-rating of other values such as short circuit rating.
Utilization Voltage The voltage at the terminals of the equipment, e.g. 230V, 460V, and
575V.
Service Voltage The voltage at the utility, or source of supply, boundary.
The attached chart identifies the allowable voltage ranges in per unit values with a base voltage of 120V
per unit. The chart shows two different ranges of voltage. The Range A portion of the chart illustrates the
range over which voltage systems are designed to operate under normal conditions. The Range B
portion of the chart is the allowable level of overshoot and droop that will happen while trying to regulate
the system to the Range A values. It is the intent of ANSI C84.1 that operator intervention will
compensate for extended operation at voltages outside Range A boundaries.

Range B also defines the range of voltage variation within which equipment must be designed to operate
satisfactorily. The equipment nameplates vary in which voltage is used as Rated Voltage. For low
voltage power circuit breakers and the metal-clad switchgear the voltage rating is the maximum system
voltage as required by ANSI C37.12 and C37.20.2. For insulated case circuit breakers and molded case
circuit breakers NEMA ICS-1 allows either the utilization voltage or the nominal system voltage to appear
on the nameplate.

For example, a motor control center will carry a nameplate listing the nominal system voltage, while the
close-connected switchgear carries a nameplate listing maximum system voltage, and the motor is rated
based on the utilization voltage as required in NEMA Standard MG1.

Why the differences? The differences in the Standards are set to match up to the way the equipment fits
into a system design. Switchgear is often operated at close to maximum voltage, since transformers are
tapped to maintain the utilization voltage high in order to increase motor torque in the field. The motor
control center can be close coupled to the switchgear or remotely located, so the same design may have
a utilization voltage approaching either the maximum service voltage or the system voltage.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

An example of how the chart works:

For a 480V system the maximum voltage will be proportional to the new systems nominal system voltage
by the ratio of the maximum per unit voltage (127 per unit) to the nominal voltage (120 per unit).

X

127 per unit
Maximum Voltage Rating =

=

480 120 per unit
Maximum Voltage Rating for 480V is 508V.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3

Therefore, the 508V will appear on the nameplate of the low voltage power circuit breaker as the
maximum voltage. The motor control center will list the system voltage of 480V. The motors connected
to the motor control center will list the utilization voltage of 460V.

The difference between minimum service and minimum utilization voltages is the intended voltage drop
within the wiring system. This difference is greater for services greater than that of 600Vac which allows
for a transformer voltage drop between service voltage and utilization equipment.

The Range B utilization voltage limits for 6900Vac and 13800Vac are 90% and 110% of the voltage
rating of the standard motor and thus vary slightly from the chart.

I hope this helps to clarify the different voltage ratings. Please do not hesitate to contact me if I can be of
further help.

Jim Bowen
Technical Director

The New Medium Voltage Circuit Breaker Interrupting
Ratings Based on K Factors of 1

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

July 31, 2001

The ANSI Standard for Medium Voltage Circuit Breaker Test Procedures, IEEE C37.09 was revised in
1999. This standard defines the short circuit tests required to certify a medium voltage circuit breakers
interrupting rating. In the past the interrupting rating changed as a function of the voltage at which the
breaker was applied. The new standards are an effort to recognize that modern vacuum and SF
6

interrupting technologies more closely represent a constant current interrupting device, independent of
the nominal system voltage.

The standards evolved from the 1945 revision when breakers were rated based on interrupting MVA (see
figure below). In 1968 the standards established varying k factors to adjust interrupting rating as a
function of the voltage. The k factor reflects the performance of the oil and air interrupting technologies
available at that time. This practice ended with the 1999 change, where circuit breaker k factors were all
set at to a value of one. The k factor of 1 results in all medium voltage breakers, tested to the 1999
version of C37.09, having a constant interrupting rating irrespective of nominal system voltage. The
changes in interrupting rating at the system voltage can be seen in the graphs below.

Using the 1968 standards, the 250MVA breaker had a 29kA interrupting rating if applied at 4.76kV while
the same breaker applied at 4.16kV had a 33kA interrupting rating and when applied at 3.85kV it had a
35.5kA rating. Using the 1999 standards, a newly certified 36kA breaker with a k factor of 1 will retain the
36kA interrupting rating independent of the applied voltage.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

Important Note:
Circuit breaker short circuit interrupting rating is the symmetrical RMS current at the time power
contacts part.

These new rating structures do not change the rating of circuit breakers certified prior to the 1999
revision. They only impact breakers that are certified to the new testing standards. The new 1.0 k factor
ratings will simplify breaker application and align with the long-standing IEC nomenclature. It is important
that we do not make the mistake of trying to apply a short circuit rating that varies as a function of the
voltage to circuit breakers certified to have a k factor of 1.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3

Jim Bowen
Technical Director

Asymmetrical Interrupting Current Rating of Medium
Voltage Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 15, 2001

As discussed in the last Powell Technical Brief, the 1999 revision of ANSI Standard C37.04 and C37.09
changed several definitions relating to the rating structure of medium voltage circuit breakers. The
earlier revision of the circuit breaker standards utilized the S factor as a multiplying factor that defined
the total current a breaker was rated to interrupt at contact part. The newest revision of C37.04 replaced
the S factor with the %dc as the method of evaluating the asymmetrical current offset. The %dc is
utilized to calculate the total interrupting current.

The asymmetrical current is an important component of the total fault current. It is critical that not only the
symmetrical interrupting current, but also the circuit breaker total interrupting rating be greater than the
system worst-case available fault condition. In the past, we have been able to pay very little attention to
the possibility of a system x/r ratio higher than the nominal value of 17 and the resulting total current.
Today, with more generation being installed, the momentary rating and the total current capability play a
greater role in the sizing of equipment because local generation increases system x/r ratio.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

The maximum fault current occurs during the first loop of sinusoidal current after the instant of fault
initiation. An asymmetrical offset containing a dc component of as much as 160% of the symmetrical
current can be impressed on top of the symmetrical ac current in one or two of the three phase currents.
This dc component quickly decays and the fault current is greatly reduced in magnitude by the time the
contacts part on the circuit breaker (see figure on structure of an asymmetrical current wave). The speed
at which this dc component decays is a function of the system x/r ratio. A high x/r ratio means a greater
system inductance will dominate the fault current and slow the attenuation of the dc offset.

The %dc that circuit breakers are certified to interrupt, is based on the contact part time and a standard
x/r decrement curve. (Refer to Figure 1 above). The combination of the contact part time and the nominal
x/r value, results in the maximum value for % dc that the circuit breaker must interrupt. The nominal x/r of
17 coincides well with the typical 60 Hz industrial substation and utilities distribution systems.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3

The %dc is then used to compute the total interrupting current of the circuit breaker at the moment of
contact part. The following equation shows how this total current is computed.

Utilizing the chart and this formula we can compute the total current. An illustration will help clarify the
calculation. To find the asymmetrical interrupting capability of a 36kA, 3-cycle rated breaker with a
published opening time of 25msec, a contact part time of 33msec is used. The contact part time includes
cycle of minimum relaying time added to the opening time of the breaker.

Using the 33msec contact part time of our sample, we find the breaker is capable of interrupting the 36kA
symmetrical current with a 50% dc component riding on top of the symmetrical current when these
values are plugged into the formula above, the total rms current is 44kA.

Since, this breaker is certified as a 3-cycle breaker it is certified to interrupt a total current of 44kA at time
from 3 cycles to 2 seconds. Note that if a 5-cycle breaker was certified, the contact part time is 50 msec.
The total interrupting current rating would be 40kA, whether it clears in 5-cycles or 2 seconds.

Where does this all become significant? With more and more generation being installed we find that the
system x/r ratio plays a much more significant part in applying equipment properly. In many generator
bus cases the equipment may have to be oversized to handle the higher level of total current or the
tripping of the breaker may have to be delayed a few cycles to allow the dc to decay to an acceptable
level.

Jim Bowen
Technical Director

Condensation in Switchgear and Anti-Condensation Heaters

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 25, 2003

Condensation or dew occurs when the temperature of still air falls to the point the air can no longer hold
the water vapor suspended in it as a gas. The resulting liquid that condenses combines with dust and
other contaminates. When condensation occurs in switchgear there are very few surfaces where the
result is not detrimental. When deposited on the surface of an insulator the semi-conductive film results
in a shorting out of a portion of the creep path. The reduction in impedance of the insulators allows an
increase in leakage current and results in tracking.

In addition to the problems with this semi-conduction film, the moisture can also be absorbed by many
types of insulators. This is not as great a problem as it once was due to improved insulating products and
finishing techniques. It is still considered a best practice to seal all raw edges of insulators with an anti-
track sealer to minimize moisture absorption.

The condensation occurs when warm moist still air comes in contact with a cold surface. The cold
surface chills the air to below the dew point and condensation is the result. To control the condensate
problem, a common approach is to maintain an elevated temperature relative to the ambient. Typically,
condensation is only considered a problem if the internal relative humidity is 65% or greater. Experience
has found that raising the internal temperature by 10C alleviates the problem. The 10C temperature
increase assures air movement and, in most cases, prevents the switchgear from approaching the dew
point.

Outdoor, thermally non insulated equipment is most susceptible to condensation. When the outside
panel of the equipment is exposed to colder air the chilled panels may cause the moisture to condense
on the interior surfaces. Several design considerations are relevant to minimizing this condensation
possibility.

To control the entrance of moist air, ventilation should only be used when necessary to keep the
temperature rise below the acceptable design limits.

Space heaters are used to raise the temperature in the enclosure by supplementing the heating
naturally present in energized equipment. The induced eddy current heating in the steel panels of
the switchgear caused by the current flow and the resistive heating resulting from the impedance
of the current path combine to create heat inside the enclosure.

The heating created by load currents will be adequate as long as there is a significant current
flowing. A 500 amp load current flowing through a vertical section of 1200 amp rated equipment
will result in an internal temperature rise of 11C. But, a current flow of only 200 amps, results in a
temperature rise of less than 2C and space heaters must be used to maintain dew point control.

Special care also must be used in cable compartments with outdoor ventilation, vapor barriers in
bus duct, and for outdoor switchgear. In all of these cases the internal temperature close to the
sheet metal is greatly influenced by changes in external temperature.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 2

For further reading I extracted the following information from the web to provide a better understanding of
the definition of dew point and relative humidity.

What is Dew Point and Relative Humidity?

The warmer the air, the more water vapor it can hold. Dew point is a measure of how much water vapor
is actually in the air. Relative humidity is a measure of the amount of water in the air compared with the
amount of water the air could hold at the measurement temperature. To see how this works, lets use the
chart below which is adapted from Meteorology Today by C. Donald Ahrens, published by West
Publishing.

Air Temperature in C Water Vapor Air Can Hold at This Temperature
30 degrees 30 grams per cubic meter of air

These numbers, which apply to air at sea level pressure, are based on measurements over the years.
They are basic physical facts.

Now, lets see how dew point and relative humidity work. Imagine that at 3:00 pm you measure the air
temperature at 30C and you measure its humidity at 9 grams per cubic meter of air. What would happen
if this air cooled to 10C with no water vapor being added or taken away? As is cools to 10C, the air
becomes saturated; that is, it cannot hold any more water vapor than 9 grams per cubic meter. Cool the
air even a tiny bit more and its water vapor will begin condensing to form dew. Back at 3:00 pm, when
the measurements were taken, we could say that the airs dew point was 10C. That is, if this particular
air were cooled to 10C at ground level, its humidity would begin condensing to form dew.

How about relative humidity? At 3:00 pm the air has 9 grams of water vapor per cubic meter of air. We
divide 9 by 30 and multiply by 100 to get a relative humidity of 30%. The air actually has 30% of the
water vapor it could hold at its current temperature. Cool the air to 20C. Now we divide 9 grams of water
vapor, the vapor actually in the air, by 17, the amount of vapor it could hold at its new temperature, and
multiply by 100 to get a relative humidity of approximately 53%. Finally, when the air cools to 10C, we
divide 9 by 9 and multiply by 100 to get a relative humidity of 100%. The air now has as much vapor as it
can hold at its new temperature.


PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

page 3

What does all this mean to Powell switchgear designs?

At Powell we recommend these guidelines for minimizing the possibility of condensation:

We recommend installation of two 125 watt space heaters in each of the cable and the circuit
breaker compartments for:
o 27kV and 38kV indoor switchgear
o 5kV and 15kV outdoor switchgear
o 15kV indoor switchgear with current transformer ratios of 200:5 and smaller mounted on
the insulation spouts. This is to compensate for lack of heat and air flow due to low
operating current and the lower associated eddy current and conductor resistance
heating effects.
We recommend installation of one 125 watt space heater in each of the cable and circuit breaker
compartments for indoor 5kV and 15kV switchgear.
We recommend thermostatic control to be included on all space heater circuits with the heaters
set to turn off at 60C +/- 5C.
We recommend a second set of space heaters be added to that already recommended and set
for operation at 10C +/- 5C for applications operating for extended periods in climates below
0C.
We recommend installation of space heaters within 2 feet of the outdoor side of any bus duct
vapor barrier to prevent condensation when the bus duct is not energized.
We recommend epoxy or porcelain standoff insulators for outdoor equipment due to the low
water absorption nature of the material.
We recommend that clients specify an ammeter in each space heater circuit to monitor circuit
health and, if a smart substation is purchased, include transducers to monitor each space heater
circuit.
We recommend the use of enclosures designed to minimize the chance of condensation falling
or dripping onto current carrying parts.

Jim Bowen
Technical Director

Arc Resistant Switchgear Tested to ANSI C37.20.7 and
NFPA70E

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

The intent of the 2004 version of NFPA70E, Standard for Electrical Safety in the Work Place, is to
minimize the at-risk procedures used by operators of electrical equipment. The first step is to minimize
risk by having operators perform work with the equipment only in an electrically safe condition, the
second step is to design the hazard out of the normal work procedures, and the third step is to rely on
Personal Protective Equipment to minimize the risk to the individual performing the task.

Arc resistant switchgear can assist with the first step effort by providing enhanced safety conditions when
the operator task involves energized equipment and possible exposure to an arcing fault. Tasks in this
category include:

Racking a medium voltage circuit breaker to or from the bus connected position
Racking a VT or CPT roll-out to or from the bus connected position
Opening and closing a circuit breaker
Calibrating and troubleshooting devices within the instrument compartment

The purpose of arc resistant switchgear certified to ANSI C37.20.7 is to eliminate the risk from the arc
blast and the by-products (heat, pressure, shrapnel, and molten copper) during normal tasks performed
on the equipment. During arc fault design tests the energy release by an arcing fault is monitored by
mounting racks of a black cotton material in panels covering the surface of the switchgear. This material
is similar to 4.5oz/yd untreated t-shirt material identified as Hazard/Risk Category 0 per NFPA70E Table
130.7(c) (11). The panels are mounted at 3.9 inches from all possible seams and one of the many
acceptance criteria of ANSI C37.20.7 is that none of the cotton indicators ignites during or following a
test.

While the focus of NFPA70E is the heat from the arc in medium voltage switchgear, it is the pressure
wave associated with the arc fault that dictates the design of the switchgear. The switchgear designed for
arc resistant protection requires heavy reinforcing of the entire structure.

In conclusion, arc resistant switchgear designs the hazard out of the tasks and reduces the level of risk
for normal tasks to a zone 0 category. The result is a reduced need for PPE. The design focus of arc
resistant switchgear is to provide the necessary enhanced safety features while requiring no addition
maintenance, calibration, or final element tests to assure functionality.

Jim Bowen
Technical Director

Lubrication Failures in Circuit Breakers

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

July 29, 2007

Statistics published for a broad spectrum of circuit breakers indicate that a significant number fail due to
malfunctions in the operating speed of their mechanical components. Frequently, these operating
mechanism failures can be traced to poor maintenance. This Technical Brief explains the role lubrication
plays in circuit breaker operations and reviews maintenance best practices to help ensure their
mechanical reliability.

The condition of circuit breaker lubrication is a primary factor of the overall ability of a circuit breaker to
successfully interrupt fault current. If a breaker is designed to interrupt a fault in five cycles and the
lubrication system is preventing the breaker from operating at the speed at which the breaker was
designed, the arc energy in the interrupter may exceed acceptable design limits. Due to reduced
operating speeds, a circuit breaker may exceed the maximum acceptable arcing times obtained during
its original design tests. If the thermal limit of the arcing contact design is exceeded due to these
lengthened arcing times, catastrophic failure may occur.

Given circuit breakers long intervals of dormancy, it is advisable to test them periodically. There are
several commercially-available methods to test the speed of circuit breakers to ensure they are operating
at proper design speed when opening them for the first time in each maintenance cycle. These
first-shot methods of determining the operating capacity of circuit breakers after long periods of latency
are valuable tools in determining the quality of aged lubricants.

The required interval and type of maintenance of any circuit breaker is dependent on the cleanliness of
the mechanism and the condition of the lubrication. The control of dust, dirt, humidity, and contamination
all play a part and must be considered in designing a maintenance program.

When re-lubricating circuit breakers, it is important that the load bearing surfaces be properly cleaned
and re-lubricated per the manufacturers recommendations. There are dangers in mixing lubrications,
such as utilizing the wrong lubrication so that a chemical bond is formed that does not have the lubricity
of the grease originally designed and tested for the equipment. The use of aerosol or other inappropriate
lubricants can slow a breaker down to the point that it will never successfully interrupt fault current, even
though it may break load current repeatedly.

Please contact Powell @ 713.944.6900 or info@powellind.com if you have questions or need assistance
with information contained in this Technical Brief.

Jim Bowen
Technical Director

Arc Flash Boundaries and Additional Precautions when
Installing Arc Resistant Switchgear
Part 1 Plenum and Exhaust Ducts

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

January 9, 2009

The focus of arc resistant switchgear design and testing is to minimize the exposure of an operator
performing normal duties to the by-products of the arc fault. This is typically accomplished by a
reinforced switchgear design and fast acting protective schemes. These methodologies reduce the
duration of the fault and prevent energy released during the fault from escaping into the designated
protected areas around the switchgear.

The test guides commonly used for evaluating switchgear performance during an arc fault are designed
to monitor the energy released into the accessible areas immediately surrounding the switchgear. This
energy release is evaluated by observing the escaping gas with burn indicators and by physically
observing any mechanical hazards such as falling debris, doors blown open from overpressure, or
ejected parts. There is no procedure for evaluation of other hazards such as the thermo-acoustic wave
or the release of toxic gases.

To minimize the potential hazards that are not specifically evaluated by the test procedures it is
recommended that the switchgear employ a plenum and ducting assembly that can safely contain the
thermo-acoustic wave and exhaust the gases to a controlled location. This location may be an isolated
area within the building, but it is preferable to duct the gases outside of the building.

The duct should be of sufficient size and physical structure to withstand the pressure from the exiting gas
and also designed to minimize turbulent flow such that it does not create any back pressures that could
cause the switchgear to fail during an arcing fault (this must be proven as part of the arc fault testing).

Good design practice would also provide for the duct to slope away from the switchgear and be fitted
with a weatherproof external cover to prevent any ingress of moisture. Additionally, in applications where
the duct is exposed to cold, it may be necessary to equip the duct with a space heater to prevent
condensation.

Michael Wactor
Technical Liaison Research and Development

Part 2 Exhaust Duct Gas

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

January 9, 2009

When a plenum and ducting system is utilized to control the by-products of an arcing fault several site
considerations must be made with respect to where the gas will be vented. Because the gas release is
at an elevated temperature and pressure and contains metallic particles and toxic components, the
location of the vent must be isolated from personnel and sensitive equipment. Further, when the gas is
vented into the building the structural capability of the building to withstand the overpressure and the
effects of the smoke and toxic gas in a closed area must also be evaluated.

Due to the circuitous route the arc blast takes through the switchgear, the plenum, the duct system, a
large part of the fault energy is attenuated inside the switchgear system. There is still a significant
amount of energy that leaves the exhaust port and, depending on the speed of the protective scheme; a
significant level of toxic gas and smoke may also be released.

Pressure measurement in the switchgear during testing reveals significant drops in value as the gas
works its way to the exit point of the duct. A typical 63kA three phase fault in a circuit breaker
compartment sees an average peak pressure in that compartment of 12 psi. That pressure is reduced to
approximately 2 psi when it reaches the exhaust duct external cover. At this point the gas has traveled
at least 8 feet and the drop in pressure reflects the loss of energy that occurs.

The gas temperature also drops exponentially as it moves from the source. The temperature of the fault
gases exiting the duct in the typical test sample are around 200C and will continue to cool as it moves
from the vent. Unfortunately, the toxic nature of this gas does not decrease based on distance traveled
and for that reason it is preferred to vent the gases outside of the building.

Additional precautions, beyond those concerned with the temperature of the gases, will be required when
the gas is released in a confined space or building. These precautions include evaluation of the effect of
an overpressure on the structure and the availability of light and fresh air.

The isolation area for the exhaust duct to vent into can be visualized as a 5 foot diameter cylinder around
the centerline of the exhaust duct that extends 7 feet out from the exhaust duct opening. While the
actual flow of gas resembles an ellipsoid within the borders of the described cylinder and will vary in
intensity based on fault current level and proximity to the vent opening, using the cylinder dimensions as
the borders of the isolation area provides a simple and effective method to define the space. This space
must be free of personnel, sensitive equipment, and physical impediments to gas flow when the
equipment is energized.

Part 2 Exhaust Duct Gas

PO Box 12818
Houston, TX 77217
Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

In almost every case, a duct exiting a building will be approximately 10 feet above grade. It is
recommended that sensitive equipment that could be placed directly below the exhaust path and not
extend into the isolation area be relocated to assure that metallic particles carried by the gas flow not
drop into the equipment.

Michael Wactor

01.4TB.096

Use of a Plenum in Low Voltage Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

J anuary 21, 2009

Many factors combine to determine the intensity of an electrical arcing fault in switchgear. The arc
energy is determined primarily by the short-circuit current level and the arc voltage, which is related to
the length of arc. Unlike Medium-Voltage (MV) equipment, where conductor spacings typically range
from 2 to 10 inches, Low Voltage (LV) equipment spacings between conductors and between conductors
and ground are very small typically less than 3 inches in the switchgear and even less within
components. Therefore the voltage drop of an arc stretching across the spacing in LV equipment is
small. This affects the arc behavior in two ways.

The first effect is that the arc voltage determines the sustainability of the arc fault. It is generally
accepted that for an electrical arc in open air there is approximately 100V/inch of arc length. Applying
this rule to MV equipment where a typical standoff insulator for a 15kV system is 6 inches tall; an arc
fault bridging that distance would produce an arc voltage of approximately 600V. This is a relatively
insignificant percentage of the total system voltage. Therefore, because the voltage drop across the arc
is low relative to the system voltage, the arc will sustain until the source is removed or the combustible
material in the arc path is consumed.

Applying this rule to LV equipment where a typical standoff insulator is 1.375 inches tall; an arc fault
bridging that distance would produce an arc voltage of approximately 140V. This is a very significant
percentage of the total system voltage. Therefore, because the voltage drop across the arc is very high
relative to the system voltage, the arc will have difficulty sustaining once the ionizable material in the
immediate area of the arc is consumed.

Controlling the duration of the arc is one of the methods utilized to minimize the hazardous effects of an
arcing fault. If the fault can be stopped before the short-circuit current reaches a significant level, the
rapid overpressure phenomenon of the arc fault can be avoided. Even if the initial rapid overpressure
cannot be stopped, reducing the duration of the fault greatly reduces the outflow of fault byproducts.
Reducing the duration can also prevent switchgear structural failure by impacting both burn through and
slow pressure build up from lower fault currents.

The second effect is that the arc voltage (arc length) determines the arc energy. Longer arc lengths
produce higher arc voltages and therefore contribute to higher levels of arc energy. Using the values in
the previous example; the MV arc measures roughly 600V while the LV arc measures roughly 140V, it is
immediately obvious that a fault at exactly the same current level produces a fault energy more than 3
times greater in the MV equipment than in the LV equipment simply because to the voltage contribution
to the energy calculation. Higher energy provides more heat to expand the gases and more energy to
vaporize components and thereby produces a larger explosive condition.

01.4TB.096

Use of a Plenum in Low Voltage Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Taking these two factors into account when designing Low-Voltage arc resistant switchgear allows for
significant design differences between an LV design and an MV design and yet the same level of
personnel protection may be met. Through inventive use of phase spacing, insulation, protective devices
and venting, a Low-Voltage design capable of meeting all the requirements of IEEE C37.20.7 can be
attained without the extensive bracing, plenums, and duct work often seen in Medium-Voltage
Switchgear.

Powell Arc Resistant Low-Voltage switchgear designs have been thoroughly tested to the requirements
of IEEE C37.20.7 without a plenum or ductwork. All of these tests were performed in a minimum sized
room and ceiling simulation to assure proper application evaluation. Because of the innovative design,
the release of arc fault byproducts through the top venting areas is both low in volume and made at
pressures so low that a plenum above the switchgear is unnecessary.

Michael Wactor

01.4TB.097

The Debate Over Type C Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

April 7, 2010

A great deal of confusion exists with regard to the actual protection provided when an arc fault rating is
applied to switchgear or motor control equipment. There are several different test procedures currently
used to evaluate the ability of equipment to withstand the effects of internal arcing faults. These test
procedures are IEC 62271-200 Annex A, IEEE C37.20.7, and EEMAC G14-1. All of these documents
have differences in procedures and requirements and one can debate which provides a more complete
or appropriate evaluation methodology. Despite all of their differences they share one common
component that is critical to understanding the level of protection provided; all three of these evaluate the
equipment based on preventing the release of fault gases into designated areas and all three monitor
this effect by using cotton burn indicators. None of the documents address operator injury due to the
release of Thermo-acoustic shock waves, the risks associated with the release of toxic gases (vaporized
metal, burning insulation and paint caused by the intense heat of the arc), sound pressure, or exposure
to electrical shock. Further, the evaluation provided is based on exposure to survivable or curable
burns; not on being injury-free.

The original internal arc fault test appeared in IEC 298 Appendix AA in the mid-1970s (the new IEC
designation is 62271-200). Two levels of protection were offered, based solely on who could access the
equipment; qualified personnel or the general public. The concept of compartment-to-compartment
protection was introduced in the EEMAC document in 1987 and designated as Accessibility Type C. In
2001, the IEEE document, C37.20.7 was published and included a suffix C designation in its annex as a
means of providing an IEEE test that covered the EEMAC design. It was specifically left out of the main
body of the IEEE document and not shown as a preferred rating because it is often misinterpreted to
imply a level of protection with open doors or with parts of the power circuit exposed. The compartment-
to-compartment rating of Type C or suffix C has, through time and overzealous marketing strategies, led
users to believe that working inside energized equipment is safe because the Type (or suffix) C test for
arc fault was passed.

The 2009 edition of NFPA 70E clearly states that the user may take advantage of applying qualified Arc
Resistant (using IEEE C37.20.7) equipment when evaluating the Hazard/Risk Category for certain
operations. Specifically, the document cites activities performed on Arc Resistant Type 1 or 2 (for
clearing times of <0.5 sec with a perspective fault current not to exceed the arc resistant rating of the
equipment) such as circuit breaker operation (opening or closing) and insertion or removal (racking) of
the circuit breaker with the doors closed can be classified as Zone 0. The document further states that
these same operations with the door open, raise the level to Zone 4. A 50% reduction in the Hazard
Zone is offered for activities in the designated control circuit (instrument) compartment (120V or below)
for arc resistant Metal-Clad gear when the circuits are energized, taking the level from Zone 4 to Zone 2.
Reading between the lines here, the control compartment must be isolated from the power circuits to
meet the Metal-Clad requirements. The document still indicates a level of risk that requires PPE once
the control compartment door is opened.

01.4TB.097

The Debate Over Type C Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

There is no distinction made for any of the suffix ratings given in IEEE C37.20.7 Annex A. In other
words, NFPA 70E does not recognize arc resistant ratings claiming protection between adjacent
compartments, so specifying any level beyond the base rating of Type 1 or Type 2 does not gain any
advantage with respect to incident energy mitigation.

All of the ratings and protective schemes discussed in the test procedures and in NFPA70E simply
reduce the level of injury or reduce the likelihood of injury. None of these documents imply that it is safe
to work on energized equipment. There is no additional personnel protection provided in specifying a
rating beyond the Type 1 or Type 2 Accessibility found in the IEEE C37.20.7 test guide. This is not to
say that there is no value in specifying a suffix B or suffix C rating, but it does require a bit more
understanding of what the expected goals are for the protection scheme. These ratings may provide
advantages in limiting internal damage to the equipment and reduced repair times in the event of a fault
when properly coordinated with other protective devices.

Michael Wactor, P.E.
Technical Director Research and Development

01.4TB.098

PPE and Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

October 07, 2010

A common misconception made when considering the application of Arc Resistant Switchgear is that it
completely eliminates the need for PPE. All of the arc fault test procedures and requirements commonly
accepted in North America share a similar evaluation method based on monitoring gas release using
cotton burn indicators. High temperature gas may be released from the equipment, but not in a sufficient
quantity to cause ignition of the indicators in the specified time. Successful testing literally means that
under the conditions of a full short-circuit level 3-phase arc fault, the untreated cotton indicators (cloth
weight of 150g/cm
2
) placed 4 from the gear will not ignite and that the equipment does not break
open/burn open or eject parts for a fault duration indicated by the nameplate (typically 0.5s). The
implication being that if the indicators do not burn, an operator in the same area will not be burned.

The evaluation measures exposure to survivable or curable burns verses life-threatening or incurable
burns; it does not confirm an injury-free condition. Most AR gear designs do not actually reduce the
Incident Energy; they simply redirect the hazardous by-products away from the areas where the operator
might be working. Essentially this means the AR design acts like PPE; providing an effective barrier
between the operator and the fault. The extent of this barrier is discussed in the 2009 edition of NFPA
70E. Activities that require opening the equipment or operating the equipment outside the established
rating for arc fault mitigation will expose the user to hazardous conditions. It is therefore wise to consider
some level of PPE when working around or on equipment, even if it carries an Arc Resistant rating.

MV equipment meeting the test requirements of C37.20.7 is recognized in the 2009 edition of NFPA 70E
as providing protection from arc energy when the appropriate doors and covers are closed and the
clearing times of the circuit protection devices and the prospective fault current are coordinated with the
arc resistant rating of the equipment. Refer to NFPA 70E section 130.7(c)(9) on page 33 of the 2009
edition. In this table, the Hazard/Risk Category is identified for performing specific functions on Arc
Resistant Switchgear. The Hazard/Risk Category defines the appropriate PPE. The advantage of AR
gear is a significant reduction in PPE required by NFPA 70E for the identified activities with MV
switchgear. Activities such as circuit breaker operation (opening or closing) and insertion or removal
(racking) of the circuit breaker with the doors closed can be classified as Hazard/Risk Category (HRC) 0
for qualified MV switchgear. The recommended protective clothing characteristics for HRC 0 are Non-
melting, flammable materials (i.e., untreated cotton, wool, rayon, silk or blends of these materials) with a
fabric weight of at least 4.5 oz/yd. There is no minimum arc rating of PPE required. The protective
clothing recommended is a long sleeve shirt and long pants. Additionally, PPE recommended is safety
glasses or goggles, hearing protection and leather gloves (electrically rated if required).

Arc resistant equipment carrying an IEEE C37.20.7 suffix B rating (Type 1B or 2B) has been tested with
the instrument compartment door open. The burn indicators are placed 4 outside the opening of the
door to simulate personnel standing in the open doorway. In a successful test the indicators do not
ignite, demonstrating that the energy released through the open instrument compartment due to a
primary circuit fault is equivalent to the protection provided by closed the doors and covers of the primary
circuit compartments. Inside the instrument compartment, the energy level may be higher. This test

01.4TB.098

PPE and Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

does not imply that it is safe to work inside the energized equipment as the test does not evaluate faults
on components within the compartment. When the voltage in the instrument compartment is higher than
120V, the compartment is considered by NFPA 70E to be an HRC 2 area. A similar argument can be
made for equipment carrying a suffix C rating (Type 2C). NFPA 70E makes no distinction between any
of the suffix ratings found in IEEE C37.20.7 Annex A. In other words, NFPA 70E does not recognize arc
resistant ratings claiming protection between adjacent compartments.

Unfortunately, AR LV gear is not yet recognized by NFPA 70E. Powell has used calorimeters in addition
to the IEEE C37.20.7 required cotton indicators for many of our LV equipment arc tests. Our observation
is that a successful test never measured more that 2 cal/cm
2
. While this measurement is not in
accordance with any test guide and does not provide 100% coverage of the exposed surfaces, the
results indicate performance in the 0 Hazard/Risk Category.

For these reasons, it is recommended that the scope of work and risks be thoroughly reviewed before
any activity requiring opening a door or removing a panel on energized equipment. Recognizing that an
internal arcing fault may release up to 2 cal/cm in HRC 0 and that arc fault risks exist in the LV circuits of
the designated instrument compartments, Powell recommends HRC 1 clothing and PPE as the minimum
for any work on energized equipment.


01.4TB.099

What Defines an Instrument Compartment in Low Voltage
Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 02, 2011

Arc resistant equipment carrying an IEEE C37.20.7 suffix B rating (Type 1B or 2B) has been tested with
the instrument compartment door open during the fault. The burn indicators are placed 4 outside the
opening of the door to simulate personnel standing in the open doorway. In a successful test, the
indicators do not ignite; demonstrating that the level of energy released through the open instrument
compartment due to a primary circuit fault meets the same requirements applied to the closed doors and
covers of the primary circuit compartments.

For Medium Voltage equipment, the Instrument Compartment is very easily defined as the compartment
where the low voltage control devices are located. Components found in that compartment are powered
by a source that is isolated and protected from the main power circuit, and generally at a much lower
voltage. Monitoring circuits are typically fed through isolation devices and are overload protected.

This is not the case with Low Voltage equipment. Where do you draw the line between what will be
considered LV control verses LV power circuits when they are often the same circuit? Monitoring circuits
connect directly to the power circuit and while they often contain protective devices like fuses, the
connection on the line side of the fuse is capable of full short-circuit current levels because it is directly
connected to the power circuit.

Bringing a connection from the power circuit into any compartment makes that compartment part of the
power circuit, unless the connection is both physically isolated and electrically protected with a fast-
acting interruption device. Without such protection, it has the potential to see the maximum available
incident energy for the power circuit during a fault. Opening an Instrument Compartment door with
internal components connected directly to the Power Circuit has the same potential risks as opening a
Cable Compartment door should an arc fault occur in that compartment. The only fault limiting device
would be the protection found on the incoming breaker or switch. The ampacity rating of the connecting
cable may play a role in the fault behavior because the cable will rapidly melt away from the bus under a
short-circuit. This may reduce the arcing time in the compartment, however, reducing the connection
cable size should not be considered designed protection.

Recognized fault clearing devices such as fuses and circuit-breakers can provide the required protection
for this connection and their performance in fault conditions can be evaluated using the C37.20.7 test
guide. Where such protection is utilized, the user should verify that the location of the device clearly
moves the full fault connection point outside of the compartment to be protected and that the
compartment containing the protective device was tested on the unprotected side to prove the integrity of
that compartment.

01.4TB.099

What Defines an Instrument Compartment in Low Voltage
Arc Resistant Switchgear

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Clearly, to apply a suffix B rating to LV equipment requires more than just physical separation from the
Power Circuit; it requires electrical isolation and coordination of protection when protective devices are
required.

As an example; when a Voltage Transformer (VT) is placed in an accessible compartment it usually has
a fuse in series with it for protection. This fuse does nothing to prevent/protect against faults that occur
on the Line Side terminal, so the full Short-Circuit rating of the bus is available at that terminal.
Therefore, when evaluating this compartment for arcing faults the compartment must be capable of
withstanding the maximum short-circuit allowed by the incoming breaker and protection. That also
means when accessing this device to change a fuse or setting on the transformer, the operator is
exposed to the full incident energy level. This is NOT an instrument compartment even though it
contains an instrument transformer/control device.

Mechanically isolating the VT compartment and feeding the VT via a cable having fuse protection outside
the compartment provides the protection afforded by the fuse rating. This compartment must be tested
in accordance with C37.20.7 for both the peak let-through current of the fuse and the maximum current
(from the fuse melt-time curves) at the rated arcing duration assigned. Performing this test series with
the door open will evaluate the compartment for the suffix B rating. Successful completion of the tests
means that the compartment is B rated when configured with the fuse indicated in the testing. The fuse
designation must be included as part of the nameplate information for that compartment. Changing fuse
type or rating invalidates the tests and the assigned rating.


01.4TB.100

Revisiting the Application of Arc Resistant Switchgear
Advancements in Mitigation Techniques

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 8, 2012

In 1994 Powell introduced PVAR, an arc resistant version of Metal-Clad switchgear. Almost immediately
questions arose concerning the need for such equipment when high-speed detection devices capable of
identifying characteristics of an arcing fault were readily available. In Powell Technical Brief #54, Mr.
Baldwin Bridger explained the relationship between fault timing and protective circuit operation, showing
detection equipment, regardless of how fast it operates, still calls upon a circuit breaker to interrupt the
fault and even the fastest circuit breakers still operated outside the accepted timing threshold for
equipment failure.
So, seventeen years later has anything changed? Have we improved sensing and protection to a level
that standard switchgear can be made arc resistant with retrofit-able detection equipment?

Figure 1 shows graphically the relationship of internal arcing short-circuit current and over-pressure in
the enclosure compared to the response times of various protective devices. The colored bar graphs
below the current and pressure curves depict the expected response times of different protective devices
and the associated time to complete circuit interruption. Much has been said about how fast the new
arc detection systems work; some devices claiming to operate in 1ms. It is significant to note for devices
tripping a circuit breaker, completely removing the operating time of the detection devices from the
equation still puts the opening time nearly two cycles after the point where traditional switchgear designs
begin to rupture.

Looking at Figure 1, the pressure peak is reached just after the crest of the first cycle of current. Is
this relationship related to the arc fault or the switchgear? The answer is both. The physics of an arcing
fault are pretty simple; the heat of the arc superheats the surrounding air and vaporizes the material it
comes in contact with, thus raising the pressure inside the enclosure. It stands to reason that larger
volumes take longer to reach critical pressure. More robust designs could withstand higher pressure and
survive longer events. Unfortunately the variations in the volume and mechanical strength of typical
switchgear verses the speed of expansion seen in a typical arcing fault offer little, if any, observable
improvement in mechanical performance. The construction requirements of typical switchgear
consistently begin to mechanically fail around 5 to 15 milliseconds into the fault event.

Arc resistant switchgear is designed to open pressure relief vents at the lowest pressure possible to
minimize the damage from overpressure on the structure and doors. The fault gases are directed to
controlled areas away from the operator. The profile of the pressure wave for a fault in arc resistant gear
looks nearly the same as that of standard equipment, but with arc resistant switchgear the pressure is
relieved through intentional openings, not from rupturing a door or panel. It is interesting to note the
pressure profile changes very little even when the relief vents are pre-opened.

01.4TB.100


PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

- The red bar indicates the typical response time of current limiting fuse operating in the current limiting mode.
- The yellow bar represents the allowable range of operation for a sensor/relay detection device. Most of these devices operate on the very low
end of this time band.
- The green bar represents the allowable relay delay fromANSI (1/2 cycle).
- The blue-gray bar represents the operating time of a typical 3-cycle MV circuit breaker.

Figure 1

From these observations, it is clear that pressure relief devices alone will not elevate standard
switchgear performance to a level capable of surviving an internal arc fault. The fault current must be
stopped before the open arc can superheat the surrounding air. For the volume and construction found
in typical switchgear, the time is appears to be less than cycle. Unfortunately, the -cycle circuit
breaker is not available and current limiting fuses are limited in the higher current and voltage ratings.

Obviously, not every arcing fault occurs as a 3-phase instantaneous event. Slower evolving incidents at
lower currents may burn for some period of time before they escalate into the catastrophic explosion
captured in test videos and for such events, fast acting mitigation utilizing a circuit breaker as the
interrupting method become viable options for protection schemes. The 3-cycle breaker offers
acceptable operation times when coupled with equipment capable of surviving the fault for that period of
time.

01.4TB.100


PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Evaluation of this equipment construction mitigation speed relationship requires not only consideration
for the explosive nature of the fault, but the potential fault locations and material thickness surrounding
those locations must also be evaluated. Does the protection scheme limit the fault current to a level that
will not erode the structure prior to the circuit interruption? Some form of additional structural integrity,
such as that found in qualified arc resistant switchgear, is necessary to address potential burn-through
issues.

There are new devices and new applications of older concepts being offered to mitigate arcing faults
utilizing fast acting devices to short-circuit the current ahead of the fault or to shunt the current to another
more controllable location. These concepts have their own special list of issues to consider with regard
to their impact on both connected equipment and the overall process being fed by the equipment.
Transformers and rotating equipment are especially sensitive to mechanical damage to the windings
during the bolted short-circuit created by some of these high-speed fault mitigating devices and this must
be considered when selecting a fault protection methodology.

So, seventeen years later, despite major improvements in the both sensing techniques and speed of
detection for the fault detectors, the impact to the total time required to interrupt the fault may still be
insignificant. Certainly, any time you can shave off the fault duration decreases the total equipment
damage and lessens the exposure risks to an operator; so the use of detection devices remains a very
good practice. Its just not the total solution. Until these devices become capable of sensing and
removing the fault in less than cycle, the switchgear must exhibit the characteristics identified as arc
resistant by a recognized test guide for the fault levels available on the circuit where it is applied to allow
the mitigation time be effective.


01.4TB.101

Vacuum Circuit Breaker Maintenance After Fault
Interruption

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

February 24, 2012

Routine maintenance is essential to the health of a circuit breaker. Various factors such as dust,
corrosive gases, and humidity influence the maintenance interval. The manufacturers instruction guide
for the equipment will provide details for routine maintenance and a recommended frequency based on
environmental conditions, time in service, and level of service. It is further recommended that an
inspection be performed as soon as is practical after a circuit breaker interrupts a fault at or near its
short-circuit rating. Load current or rated continuous current operations should not be confused with high
current fault interruptions.
Damage from severe mechanical stresses and metalizing of the interior of the vacuum interrupter may
occur during fault interruption. The most reliable indication of the stress a circuit breaker has been
subjected to during a fault interruption is found in the protective relay memory or in automatic
oscillograph records. Deterioration of the circuit breaker contacts may be assumed to be proportional to
the energy dissipated in the breaker during the interruption.
The energy dissipated is approximately proportional to the current and the duration of contact arcing.
However, the parting of contacts is not always evident on the oscillographs, making a physical inspection
of the circuit breaker and switchgear necessary.
It is recommended that the standard maintenance service be performed in accordance with the
manufacturers instructions. As a minimum, the following items should be inspected and the indicated
tests be performed. Any deviation from the manufacturers specific functional/operating parameters is
cause to replace the damaged item with a new item from the original manufacturer.
De-energize the equipment and follow all safety practices required at the site.
Physical inspection of the Circuit Breaker
1. Primary disconnecting devices.
Look for signs of pitting, discoloration from over-heating, and contaminated lubricant at the
contact surfaces and for any damage to the physical structure and any insulation.
2. Measure the Primary Contact resistance, open contact gap, and wipe or overtravel on the circuit
breaker.
3. Follow the recommended procedures in the manufacturers Instruction Bulletin.3Perform dielectric
tests on the circuit breaker.
Test the vacuum interrupter tubes for vacuum integrity
Test the entire circuit breaker insulation system by performing and AC High Potential test.
If insulation resistance testing is part of the normal maintenance records, perform this test for
comparison to previous measurements.

01.4TB.101

Vacuum Circuit Breaker Maintenance After Fault
Interruption

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

4. Perform timing tests.
Operational tests at nominal control voltage that can be compared to the operating range
indicated in the Instruction Bulletin or against as received Factory or Field test records.
5. Perform any other routine maintenance necessary.
Verify hardware torque
Lubricate per the Instruction Bulletin.

Physical inspection of the Switchgear
1. Primary disconnecting devices.
Look for signs of pitting, discoloration from over-heating, and contaminated lubricant at the
contact surfaces and for any damage to the physical structure and any insulation.
2. Primary bus, cables and connections.
Look for signs of discoloration from over-heating at connection points, any damage to the
circuit support structure and insulators, and bus or cable insulation.

If all is well the circuit breaker may be returned to service. If there is any deviation from the
manufacturers specifications discovered, the circuit breaker should be locked-out per the site operation
and safety requirements until full maintenance and repair can be performed.


01.4TB.102

Arc Fault Exhaust Ducts and Cold Weather Installations

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

May 11, 2012

Most ANSI rated switchgear is rated for operation in a temperature range between -30C and +40C. In
the case of indoor equipment it is hard to imagine ever encountering such extremes, however it is
possible. One specific case where it is a common occurrence involves the installation of Arc Resistant
Switchgear fitted with a plenum and exhaust duct in locations where temperature extremes are normal.
The protective cover at the end of the exhaust duct is exposed to the hostile weather conditions and it
must be treated differently than the exterior walls of the building where the equipment resides.

Examining the two extreme conditions, exposure to the upper limit does not create much concern. The
cover will get hot and radiate some of that heat into the duct and plenum air. The additional heat build-
up is outside the normal air flow path for cooling the equipment and should not present a significant
operating temperature change. Should it be a concern, the solution is simply to have air conditioning in
the building and adjust the thermostat accordingly. Assuming the air conditioning also de-humidifies the
air, condensation should not be an issue.

The cold extreme creates two potential problems. The most commonly discussed issue is the potential
for condensation. The simple solution for condensation alone would be to insulate the cover. The
second, less obvious, potential problem is ice build-up on the exterior that prevents the cover from
opening. If the cover fails to open, the results could be catastrophic for the switchgear inside the
building. In this case, insulation on the vent cover is not recommended so the cover can absorb heat
from the room and from the switchgear to help prevent ice from forming on the exterior. While this can
solve the potential ice build-up issue, the now un-insulated cover is again vulnerable to condensation.

Condensation may occur at many different temperatures, but it is most often discussed when the
installation location is in extreme cold. Many controllable factors influence the development of
condensation. Assuming the gear in a building, is the building heated? Is the gear energized? If so, at
what level is it operating? Energized gear will produce some heat which will rise into the plenum and
help keep the area dry. This heat will find its way into the vent duct and help keep the duct air dry as
well. How much heat gets to the duct depends on how hot the gear is running and how warm the room
is. Additional heaters may be required and are an absolute necessity if the gear is not energized for
extended periods of time.

What damage will condensation cause on this assembly? If the duct is properly angled away from the
switchgear any damage caused by condensation will be confined to the bottom of the vent cover. In the
worst case scenario, a build-up of condensation over a time period sufficient to oxidize the metal vent
cover would only result in the cover tearing away from the building if it is ever called upon to operate. It
will not cause a failure to open.

01.4TB.102

Arc Fault Exhaust Ducts and Cold Weather Installations

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Adding additional heat to the exhaust duct is a simple process. The typical switchgear heating element
can be placed at the end of the duct neat the vent cover and wired back to control power in the
designated low voltage instrument compartment. A thermostat is not mandatory. The air temperature in
the duct need only be warm enough to keep the vent cover surface above freezing. Because the duct is
away from the main plenum and switchgear and is a relatively small and confined space, the heater
element may be run at voltage, or even voltage, and left on constantly. As previously stated, the
additional heat will not adversely affect the equipment thermal performance when the ambient
temperature rises above freezing.

While outside the scope of evaluation for arc resistant switchgear testing, the user should be aware of
the potential for the heater element to become a projectile in the event of an arcing event. Placing a
securely mounted guard of expanded metal around the element will prevent the possibility of ejecting the
heater element when exhausting the by-products of an arcing event.


01.4TB.103

Applying Switchgear Utilizing Sealed Interrupting Devices at
High Altitudes

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

August 1, 2012

Basis of Altitude Related Corrections

Air density (temperature and pressure) varies from one location to another. Because it is a major factor
in the dielectric strength of air, most electrical equipment is tested using the standard reference
atmosphere, also referred to as the normal temperature and pressure at sea level (NTP*) to obtain
consistency in test results. Correction factors may then be used to apply the design in other locations.
Switchgear that depends on air for an insulating and cooling medium will have a higher temperature rise
and a lower dielectric withstand capability when operated at altitudes above sea level.

Mathematically, corrections to these ratings apply immediately as you elevate from sea level, however
historically air insulated equipment has shown enough design margin to be applied at altitudes up to
1000m (3300ft) with no correction factors applied. At elevations above 1000m, the rated power
frequency withstand voltage, the lightning impulse withstand voltage, and the rated continuous current for
the assemblies should be reviewed and may require adjustment by the altitude correction factors found
in applicable design standard. It is important to note that for any insulation not exposed to ambient
atmospheric pressure, the dielectric characteristics are not affected by changes in altitude and no special
precautions need to be taken. An example of this type of internal insulation can be found on a sealed
vacuum interrupter.

Corrections for continuous current

Corrections to the rated continuous current based on altitude can be quite confusing. De-rating factors
are provided in many equipment standards containing application advice that seems contradictory. It is
stated in the high voltage circuit breaker application guide that de-rating circuit breakers for altitude is
generally thought to be unnecessary because the ambient temperature at altitudes above 1000m is
significantly lower than the normal 40C maximum. This is not necessarily correct for indoor rated
equipment. The continuous current capability of any device is based on the ambient air temperature
surrounding the device; indoor equipment may be in a significantly warmer environment that the outside
air, even at higher altitudes. For indoor equipment, the de-rating assessment can only be made based
on the equipment room temperature, not the environmental conditions outside the building.

* NTP is defined as temperature =20C, atmospheric pressure =101.3kPa (14.7psi), humidity =11 g/m
3
.

01.4TB.103

Applying Switchgear Utilizing Sealed Interrupting Devices at
High Altitudes

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Corrections for di electric withstand

The dielectric capabilities of all switchgear components using air as the primary insulating medium will be
affected by the changes in air pressure experienced at higher altitudes. Specifically, the Basic Impulse
Level (BIL) and Power Frequency Withstand ratings must be reduced as altitude increases. While some
components such as sealed interrupters are unaffected internally by decreased air pressure, the external
surface insulation of these devices may experience similar reductions in capability. In applications above
1000 m the use of surge arresters to keep transients voltages below the adjusted levels should be
considered.

Surge arrester selection should be coordinated such that:
1. The maximum Discharge Voltage of the arrester is approximately 20% lower than the de-
rated BIL for the equipment.
2. The Maximum Continuous Operating Voltage (MCOV) for the arrester is equal to or greater
than the actual, not a de-rated value, maximum continuous system voltage for the equipment.

Corrections for rated voltage

Air pressure affects the dielectric capability of the air and therefore affects the interrupting capability of
contacts in open air. The air pressure in a sealed interrupter, such as a vacuum interrupter, is unaffected
by changes in air pressure and will not be affected by changes in altitude. It is not necessary to de-rate
the maximum voltage rating for devices using sealed interrupters based on elevation.

Refer to IEEE C37.100.1 Annex B for a detailed explanation of how the de-rating factors were obtained
and applied.

Michael Wactor

01.4TB.104

Transient Recovery Voltage Update

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 14, 2012

Powell Technical Briefs PTB #10 and PTB #70 discussed transient recovery voltage (TRV) and provided
data for the TRV values used by Powell in testing PowlVac
circuit breakers. The methodology for

defining TRV has been changed since those documents were written. This document will explain the
new methodology. The basic definitions found in PTB#10 and PTB#70 are still correct and the test
values given are still applicable to equipment of that vintage.

What has changed with the latest revisions to IEEE Standard C37.06 is the definition of the TRV
requirements. The rating is stated in terms of a peak reference voltage, U
c
(previously defined as E
2
),
and a time to reach that voltage, t
3
(previously T
2
), at full rated short circuit current for a fault at the
terminals of the breaker. For lower values of short circuit current, the voltage is higher and the time is
shorter. The curve between (0, 0) and (U
c
, t
3
) is defined as "one-minus-cosine (1-cosine) curve, shown
visually in the figure below. For circuit breakers used in metal-clad switchgear, U
c
is required to be
between 1.7 and 2.6 times the breaker's rated maximum voltage depending on the test duty. The present
standard lists the U
c
, and t
3
values in table 2 and 3 according to the test duty requirement represented by
the two-parameter method. Both U
c
and t
3
increase as the breakers rated voltage increases. Breakers
rated 100kV and above use a different curve shape. The figure below shows the two parameter method
with reference line and delay line with respect to 1-cosine TRV curve. The old symbols E2 and T2 are
now replaced with new symbols U
c
, U, t
d,
t and t
3
.

U
c
peak reference voltage in kV,
t
3
time to reach U
c
in microseconds
U reference voltage in kV,
t time to reach U in microseconds,
t
d
delay line in microseconds

01.4TB.104

Transient Recovery Voltage Update

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

The interrupting performance of any circuit interrupter is affected by the transient recovery voltage
appearing across the first pole to interrupt. Both the absolute value of this voltage and its rate of rise are
important in determining the interrupter's ability to meet its interrupting rating. The required values of
transient recovery voltage are included in IEEE C37.06, along with the other ratings of circuit breakers.

The conventional way of specifying the rate of rise of the transient recovery voltage (RRRV) is to specify
the peak value (U
c
) and the time required to reach that peak (t
3
). The rate of rise of transient recovery
voltage is then determined by dividing U
c
by t
3
. The nominal values are those for a full rated short circuit
interruption. For lower currents, both higher peaks and faster times are specified. Table 2 and 3 of IEEE
C37.06-2009 lists the preferred ratings for U
c
, t
3
and RRRV for class S1 circuit breakers rated below
100kV for cable systems.

Samala Santosh Reddy, P.E. Michael Wactor, P.E.
R&D Engineer Technical Director CPD

01.4TB.105

Fuses for Use in DC Control Circuits - Revised

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 14, 2012

The information provided in the Powell Technical Brief #9 regarding the different types and ratings of
fuses has changed. This document will address the new types and ratings of fuses and also answer
some common questions.

The majority of control circuits in metal- enclosed switchgear, particularly in metal-clad switchgear are
supplied from a dc control power source. The most common dc control voltages applied in medium
voltage switchgear are 125V and 250V. For the fuse to interrupt safely, Powell thoroughly reviews the
fuse performance data to make sure that it meets the requirements in a particular dc control circuit before
applying these fuses in the switchgear. The performance data is provided by the fuse manufacturer by
designing and testing the fuses per the applicable standards. Most of the fuse manufacturers test their
fuses per UL (Underwriters Laboratories) and MSHA (Mine Safety and Health Administration) standards.
To select the correct fuse rating for a control circuit, we need to consider the maximum dc voltage
withstand, continuous current and fault current interrupting rating.

Powell uses Amp-Trap midget fast-acting ATM type fuses in dc control circuits. These fuses are
manufactured by Mersen (formerly Ferraz Shawmut) and are rated up to 600V dc with a maximum fault
current interrupting capability of 100kA. However these fuses are used in conjunction with UltraSafe
USM type fuse holders. The application of these fuses in this type of circuit has been exceptionally
successful and accepted by a wide range of customers.

Following are some common questions about dc fuses

1. What is the dc rating of these fuses?

A. These fuses are successfully tested for 100kA interrupting capability at voltage rating of 600V dc.
A typical control battery used for switchgear can deliver a short-time current of about 10 times its
one-minute discharge rating, so it would be a very unusual dc control circuit that had a short-
circuit capability in excess of 100kA.

2. Are these fuses UL listed for dc applications?

A. The ATM fuses provided for dc applications are UL listed to standard 248-14 with current ratings
up to 30A.

01.4TB.105

Fuses for Use in DC Control Circuits - Revised

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

3. Can ac rated fuse be applied in dc control circuits?

A. In general, the ac rated fuses should not be applied in dc control circuits unless the applicable dc
ratings are listed by the fuse manufacturer on their data sheet. The fuses we provide are tested
for both ac and dc rating and can be applied within the ratings listed in the fuse manufacturers
data sheet.


01.4TB.106

Altitude De-rating of Components Placed in the Switchgear
Enclosure

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

September 20, 2012

As a dielectric dry air works satisfactorily. However as the altitude at which we apply our electrical
equipment increases, the effectiveness of the dielectric property decreases. An increased altitude also
decreases the continuous current carrying capability of electrical equipment due to the reduced thermal
conductivity of the lower density air. In the past, we have discussed the altitude de-rating factors for
switchgear (see PTB #25). This Powell Technical Brief will explain how altitude de-rating effects the
installation of medium voltage fuses, surge arresters, current transformers, voltage transformers, control
power transformers and relays.

Fuse De-rating
The fuse de-rating is a function of the ability of the fuse to clear a fault and achieve a dielectric across the
fuse element gap faster than the system voltage across the fuse can be established. This de-rating
would be applicable for non-sealed fuses like expulsion fuses and current limiting fuses, where the
outside air is the insulating means used to isolate the line and faulted load. Any fuse that depends on air
at atmospheric pressure for its insulating and cooling medium will have a higher temperature rise and a
lower dielectric withstand and need to be de-rated for altitude exceeding 3300 ft per the IEEE Std
C37.40, section 2.3, Table 1.

When the dielectric strength of the fuse is de-rated, the application engineer is often required to choose a
fuse of higher voltage class. The fuse interruption at higher voltages will result in higher arc voltages than
the maximum fuse voltage rating. The maximum permissible overvoltages for various rated maximum
voltages are listed in ANSI C37.46, Table 5. However, most of the fuse manufacturers design for the
peak arc voltage to be around three times that of the rated maximum fuse voltage. The peak of the arc
voltage depends on the voltage applied across the fuse and the peak arc voltage should not exceed the
system and equipment insulation levels (BIL).

An alternate solution is the hermetically sealed fuse, which does not require de-rating of the internal
operating mechanism. Note that hermetically sealed fuses are both expensive and may have a long
delivery time. Care must also be taken to confirm that the outer creepage distance is acceptable within
the de-rated values.

Surge Arrester De-rating
The most common type of surge arrester that we use today is the gapless metal oxide arrester. These
arresters are sealed and the outside air plays a part in the external surface creepage distance of the
housing. The outer surface has a certain creepage distance that prevents the surge arrester housing
from tracking across the surface of the arrester. Powell Technical Brief #59 describes the acceptable
surface creepage distance for various voltage levels. The insulated conductor distances can only be
used if an insulating boot properly covers the ferrule at the top of the arrester.

01.4TB.106

Enclosure

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Standard arresters are designed to operate at altitudes that do not exceed 6000 ft. Special arresters are
required for altitude of 6001 12000 ft and for altitudes of 12001 18000 ft. It may also be necessary to
have special arresters manufactured with a standard duty cycle arrester in a housing with higher creep
levels to maintain the desired discharge level of the arrester and the required creep distance.

A special caution when applying de-rated higher voltage current limiting fuses. Each fuse has a
characteristic such that the peak arc voltage generated would be around three times that of the rated
maximum fuse voltage. The transients that result from applying a higher voltage class fuse due to
altitude de-rating can often result in the surge arrester conducting during a fuse interruption. This should
be avoided for line side station class and intermediate class arresters as the conducting impedance of
these arresters is so low that when conducted during a fault will cause serious damage. So consider
installing distribution class arresters for this application which have higher conducting impedance.

Current Transformer De-rating
When installing CTs at higher altitudes, the dielectric strength and the thermal current rating factor are
affected due to reduced air density. The dielectric strength rating (BIL) needs to be de-rated as per IEEE
Std C57.13, section 4.2, Table 1. This de-rating may require the use of higher insulation level current
transformer with the appropriate primary current and the ratio to maintain the desired secondary current.
The thermal current rating factor de-rating is required only when the temperature limits are exceeded
while carrying the continuous current. The thermal current rating factor needs to be de-rated by 0.3% for
every 330 ft.

Voltage and Control Power Transformer De-rating
The basic impulse insulation level (BIL) and the thermal volt-ampere (VA) rating of the VT and CPT are
affected at higher altitudes. The BIL rating of the VT and CPT needs to be de-rated as per IEEE Std
C57.13, section 4.2, Table 1. This de-rating may require the use of a higher voltage class transformer
with the appropriate primary voltage and the ratio to maintain the desired secondary voltage. Always
consult with the manufacturer before applying the VT and CPT at higher altitudes (greater than 3300 ft)
for de-rated value of thermal VA rating and the iron core wattage loss of the transformer.

For example: Given a line-up of switchgear operating at 4.16kV at an altitude of 10,000 feet, the
appropriate choice is a voltage transformer with a primary voltage of 4200 Volts and a secondary of 120
Volts. The transformer would have a nominal voltage class of 8.7kV with a BIL at sea level of 75kV.
When the 80% de-rating factor is applied due to the altitude, this VT has a BIL of 60kV.

01.4TB.106

Enclosure

PO Box 12818
Houston, TX 77217

Tel: 713.944.6900 Fax: 713.947.4453
www.powellind.com
info@powellind.com

Relay De-rating
The application of relay at altitude higher than 5000 ft is considered as an unusual service condition and
need to be de-rated for dielectric strength and maximum ambient temperature of the relay as per IEEE
Std C37.90, section 4.2.1.1 , 4.2.1.2 and Table 1, Table 2. However, consult with the relay manufacturer
for your specific application.


Powell Technical Briefs - Combined 1-106

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Powell Technical Briefs - Combined 1-106

Uploaded by

Copyright:

Available Formats

01.4TB.

vacuum circuit breaker to determine fast transfer

circuit breakers using Mitsubishi interrupters. We have not

circuit breakers. Some of the questions asked, and our

circuit breaker meets these requirements.

, but we still have customers who request porcelain.

metal-clad switchgear tested for the ability to

metal-clad switchgear is suitable for use in seismic zones 0, 1

is suitable for use in

Transient Recovery Voltage Test Values

vacuum circuit breakers, we have

circuit breakers. The following

circuit breakers, the time from energizing the closing coil

breaker, as shown in the following table. All times are

Circuit Breakers for Continuous Currents

bus. The tests are performed at

switchgear. It is fairly obvious that passing these tests qualifies the

switchgear, but may not

circuit breakers, we provide interference mechanisms

circuit breakers, both conversions and replacement

circuit breakers see

( circuit breakers. If there is a complete loss of vacuum, this

metal-clad switchgear, a third channel is shown somewhere

Vacuum Circuit Breakers

Power Control Rooms

buildings are built per the standards listed

buildings are normally

is done per the NEC.

circuit breaker at inductive current levels in the 250-260 A range. All of

switchgear. Each line in the thickness of the insulation represents a conductive

circuit breakers. I'll try to answer some additional recent

metal-clad switchgear and weather-proof Power Control Room (PCR

) enclosures for all

switchgear, Powell has developed a system of poured joints,

(1) 1/2 x 3 Passed Passed

circuit breakers. The methodology for

You might also like