Measurement of Flow

P M V Subbarao
Professor
Mechanical Engineering Department
An Essential Requirement in CV
Based Industrial Appliances.
Mathematics of Flow Rate

The Scalar Product of two vectors, namely velocity and

area..
Important characteristics of the dot product

The first point to note about the definition is that the

coordinate system does not enter the definition.

The second point to note is that because cos=cos(-), the

order is not important, that is, the scalar product is
commutative.

The scalar product is also distributive.

ll these !ualities help in development of a sin"le

instrument to measure the scalar product.
Types of Flow Measurement Technolo"ies

#ariable rea $rotameters%

Rotatin" #ane $paddle & turbine%

Positive 'isplacement

'ifferential Pressure

#orte( Sheddin"

Thermal 'ispersion

Ma"netic Ma"netic

Thermal Mass

)oriolis Mass

*ltrasonic
Some Facts About Variable Area Flowmeters

)alled +float type float

type,, +rotameter--, or
+variable area, flowmeters.

.y far the most common

specified, purchased, and
installed flowmeter in the
world
Variable Area Flowmeters

Fluid flow moves the float

upward a"ainst "ravity.

Float will find e!uilibrium when

area around float "enerates
enou"h dra" e!ual to wei"ht /
buoyancy.

Some types have a "uide rod to

0eep float stable.

1ow )ost $pricin" usually starts

2 345%

Simple Reliable 'esi"n

)an Measure 1i!uid or 6as

Flows

Tolerates 'irty 1i!uids or Solids

in 1i!uid
Measuring Principles of Variable Area Flowmeters

Flow Rate nalysis.

The forces actin" on the bob lead to e!uilibrium between7

the wei"ht of the bob
b
gV
b
actin" downwards
the buoyancy force gV
b
and
the dra" force F
d
actin" upwards.
8here V
b
is the volume and

b
is the density of the bob,
is the density of the fluid, and

g is the "ravitational acceleration7

d b b b
F gV gV + =

The dra" force results from the flow field surroundin" the bob
and particularly from the wa0e of the bob.

In flow analyses based on similarity principles, these influences

are accounted for by empirical coefficient C
1
or C
T
in the dra"
law for7
9 9
Flow, Turbulent U D C F
b T d
=
U D C F
b L d
= Flow, 1aminar
The volume flow rate throu"h the rotameter is7
8here m is the open area ratio, defined as7
nd D is the tube diameter at the hei"ht of the bob.
d b b b
F gV gV + =
where the parameter is defined in terms of a constant
K :V
b
;D
<
b
characteristic of the shape of the bob7
for turbulent flow7
for laminar flow7
8ith either laminar or turbulent flow throu"h the rotameter, the
flow rate is proportional to m.
If the cross/sectional area of the tube is made to increase linearly
with len"th, i.e.,
then since the cone an"le
of the tube is small,
and the flow rate is directly
proportional to the hei"ht h
of the bob.
h
Similarity nalysis.

The basic scalin" parameter for flow is the Reynolds number,

defined as7
where U
I=
is the velocity at the rotameter inlet, and the tube diameter
D is represented by its value at the inlet, e!ual to the bob diameter D
b
.

Throu"h the Reynolds number re"imes of laminar or turbulent flow,

and particularly important for the rotameter flow re"imes with stron"
or wea0 viscosity dependence can be distin"uished.

It has been found to be practical for rotameters to use an alternative

characteristic number, the Ruppel number, defined as7
where m
b
:
b
D
<
b
is the mass of the bob.
.y combinin" >!uations, the mass flow throu"h the rotameter
can be written as7
The relationship between the Ruppel number and the Reynolds
number7
The advanta"e of the Ruppel number is its independence of the flow rate.
Since the Ruppel number contains only fluid properties and the mass and
the density of the bob, it is a constant for a particular instrument.
'esi"n )harts for 1aminar Rotameters
'esi"n )harts for Turbulent Rotameters
$?% >nd fittin" @ flan"e
shownA
$9% flowmeter bodyA
$<% rotation pic0up @
ma"netic, reluctancetype
shownA
$B% permanent ma"netA
$4% pic0up cold wound on
pole pieceA
$C% rotor bladeA
$D% rotor hubA
$E% Rotor shaft bearin" @
Fournal type shownA
$G% rotor shaftA
$?5% diffuser support and flow
strai"htenerA
$??% diffuserA
$12% flow conditionin" plate
$dotted% @ optional with
some meters.
Theory

There are two approaches described in the current

literature for analyHin" a(ial turbine performance.

The first approach describes the fluid drivin" tor!ue in

terms of momentum e(chan"e, while the second describes
it in terms of aerodynamic lift via airfoil theory.

The former approach has the advanta"e that it readily

produces analytical results describin" basic operation,
some of which have not appeared via airfoil analysis.

The latter approach has the advanta"e that it allows more

complete descriptions usin" fewer appro(imations.

Iowever, it is mathematically intensive and leads rapidly

into computer/"enerated solutions.
>liminatin" the time dimension from the left/hand/side !uantity
reduces it to the number of rotor rotations per unit fluid volume,
which is essentially the flowmeter K factor specified by most
manufacturers.

In the ideal situation, the meter response is perfectly linear and

determined only by "eometry.

In some flowmeter desi"ns, the rotor blades are helically twisted to

improve efficiency.

This is especially true of blades with lar"e radius ratios, $R;a%.

If the flow velocity profile is assumed to be flat, then the blade

an"le in this case can be described by tan : )onstant J r.

This is sometimes called the +ideal, helical blade.

In practice, there are instead a number of rotor retardin" tor!ues of

varyin" relative ma"nitudes.

*nder steady flow, the rotor assumes a speed that satisfies the
followin" e!uilibrium7
The difference between the actual rotor speed, r, and the ideal
rotor speed, r
i
, is the rotor slip velocity due to the combined
effect of all the rotor retardin" tor!ues , and as a result of
which the fluid velocity vector is deflected throu"h an e(it or
swirl an"le, .
'enotin" the radius variable by r, and e!uatin" the total rate of
chan"e of an"ular momentum of the fluid passin" throu"h the
rotor to the retardin" tor!ue, one obtains7
N
T
is the total retardin" tor!ue
Industrial )orrelations for Frictional 1osses
Electromagnetic Flowmeters

Magnetic flometers have been widely used in industry for many

years.

*nli0e many other types of flowmeters, they offer true

noninvasive measurements.

They are easy to install and use to the e(tent that e(istin" pipes in
a process can be turned into meters simply by addin" e(ternal
electrodes and suitable ma"nets.

They can measure reverse flows and are insensitive to viscosity,

density, and flow disturbances.

!lectromagnetic flometers can rapidly respond to flow chan"es

and they are linear devices for a wide ran"e of measurements.

s in the case of many electric devices, the underlyin" principle

of the electroma"netic flowmeter is Faraday-s law of
electroma"netic induction.

The induced volta"es in an electroma"netic flowmeter are

linearly proportional to the mean velocity of li!uids or to the
volumetric flow rates.

s is the case in many applications, if the pipe walls are made

from nonconductin" elements, then the induced volta"e is
independent of the properties of the fluid.

The accuracy of these meters can be as low as 5.94K and, in most

applications, an accuracy of ?K is used.

t worst, 4K accuracy is obtained in some difficult applications

where impurities of li!uids and the contact resistances of the
electrodes are inferior as in the case of low/purity sodium li!uid
solutions.

Faradays !aw of "nduction

This law states that if a conductor of len"th l $m% is movin" with a

velocity " $m;s
L?
%, perpendicular to a ma"netic field of flu( density
# $Tesla%, then the induced volta"e e across the ends of conductor
can be e(pressed by7
#l" e =
The velocity of the conductor is
proportional to the mean flow velocity
of the li!uid.
Ience, the induced volta"e becomes7
#D" e =
#D" e =
" D $" %
9
B

= =
D
#%
e

B
=
#ltrasonic Flowmeters

There are various types of ultrasonic flowmeters in use for

dischar"e measurement7

$?% Transit time& This is today-s state/of/the/art technolo"y and

most widely used type.

This type of ultrasonic flowmeter ma0es use of the difference

in the time for a sonic pulse to travel a fi(ed distance.

First a"ainst the flow and then in the direction of flow.

Transmit time flowmeters are sensitive to suspended solids or

air bubbles in the fluid.

$9% Do''ler& This type is more popular and less e(pensive, but
is not considered as accurate as the transit time flowmeter.

It ma0es use of the 'oppler fre!uency shift caused by sound

reflected or scattered from suspensions in the flow path and is
therefore more complementary than competitive to transit time
flowmeters.
Principle of transit time flowmeters.
Transit Time Flowmeter

Principle of $peration

The acoustic method of dischar"e measurement is based on the

fact that the propa"ation velocity of an acoustic wave and the
flow velocity are summed vectorially.

This type of flowmeter measures the difference in transit times

between two ultrasonic pulses transmitted upstream t
9?
and
downstream t
?9
across the flow.

If there are no transverse flow components in the conduit, these

two transmit times of acoustic pulses are "iven by7
Since the transducers are "enerally used both as transmitters and
receivers, the difference in travel time can be determined with the
same pair of transducers.
Thus, the mean a(ial velocity alon" the path is "iven by7
>(ample

The followin" e(ample shows the demands on the time

measurement techni!ue7

ssume a closed conduit with diameter D : ?45 mm, an"le :

C5M, flow velocity : ? m;s, and water temperature :95M).

This results in transmit times of about ??C s and a time

difference
t :t
?9
L t
9?
on the order of DE ns.

To achieve an accuracy of ?K of the correspondin" full/scale

ran"e, t has to be measured with a resolution of at least ?55 ps
$?J?5
L?5
s%.

Standard time measurement techni!ues are not able to meet such

re!uirements so that special techni!ues must be applied.

'i"ital timers with the state/of/the Lart Micro computers will

ma0e it possible to measure these time difference.
Point #elocity Measurement

Pitot Probe nemometry 7 Potential Flow Theory &

.ernoulli-s Theory .

Thermal nemometry 7 =ewton-s 1aw of )oolin".

1aser nemometry7 'oppler Theory.