You are on page 1of 15

# TESTING OF DC MACHINES

## NACCESITY OF TESTING: The performance characteristics of a dc machine are determined by testing of

the machines. By using these characteristics we can identify the suitability of machine per particular
operation and also the behavior o f the machine under various load conditions. For determination
efficiency of a dc machine it is necessary to have knowledge of the power input and power output
The performance characteristics of a motor include the speed (N), torque (T), input current (I), efficiency
().
DIFFERENT METHODS OF TESTING:
Dc machine can be tested by 3 different methods namely
1. Direct method
2. Indirect method
3. Regenerative method
Direct method: In this method the efficiency and losses of only small machine can be determined. In this
full load is directly applied the input and output is directly measured the ratio of output power to input
power gives the efficiency of a dc motor
EX: BRAKE TEST
It involves dissipation of large amount of energy.tha quantity and lots of energy wasted are very high
1. This method gives the correct value of the efficiency of the machine
2. The temperature rise and commutation condition can be checked on the spot at full load
INDIRECT METHOD
By this method efficiency of shunt and compound dc machine be determined in this method the no load
machines are first measured by a suitable test and then the additional losses on no load are determined
from machine data in order to calculate efficiency
EX: Swinburnes test
Less power is required (power is required to supply losses only) even for very large machine also
Because of no load on the machine the temperature rise and commutation conditions are not observed
REGENERATIVE METHOD
In this method two identical dc machines are coupled both mechanically and electrically one o f the
machine is made to run as a motor and drives the other machine as a generator so the generator feedback
power in to the supply thus the total power drawn from the supply is only for supplying the internal
losses of the machines
EX; Hopkinsons test
1 Even the large machine also tested a small power
2 The machine can also be tested under full load condition d long duration to observe the temperature
rise commutation.
The 1
st
equation is giving an idea about the direct estimation of the efficiency. In this method the
machine is fully loaded and the output is directly measured. This method of measurement is only
applied for the small machines.
The 2
nd
and 3
rd
equations are giving an idea about the indirect estimation of the efficiency. Indirect
method is helpful of determining the efficiency of shunt and compound wound generators. In this
method it is required to determine to determine the losses only. So, power supply is required to supply
For the regenerative method of determining efficiency, it is required to have two identical machines. One
machine is used as motor and drives the other and the other is used as generator and feedback the power
into the supply. Two machines are mechanically coupled. Therefore the losses can be determined
because the internal power drawn is only to supply losses of the two machines.
Except these testes, the insulation test and the test for making the commutation satisfactory is done while
building up the machine
This method is an indirect method of testing a dc machine. It is named after Sir James Swinburne.
Swinburne's test is the most commonly used and simplest method of testing of shunt and compound
wound dc machines which have constant flux. In this test the efficiency of the machine at any load is pre-
determined. We can run the machine as a motor or as a generator. In this method of testing no load
losses are measured separately and eventually we can determine the efficiency.

Swinburne's test:
The circuit connection for Swinburne's test is shown in figure below. The speed of the machine is
adjusted to the rated speed with the help of the shunt regulator R as shown in figure.

V
3-point starter
A
AA
A
A
2

I
F
FF
I
sh

L F A
A
1
1
I
a
V
DC
SUPPLY
+
-
D
P
S
T

s
w
i
t
c
h

V
3-point starter
A
AA
A
A
2

I
o

F
FF
I
sh

L F A
A
1
1
I
ao
V
DC
SUPPLY
+
-
D
P
S
T

s
w
i
t
c
h

Swinburnes Test

Connection Diagram of Swinburne's Test
Calculation of Efficiency When the Machine Is Motoring On Load
Power input = VI
Armature copper loss, PCU = I
2
Ra = (I - Ish)
2
Ra
Constant losses, WC = VI0 -(I0 - Ish)
2
Ra
Total losses = PCU + WC
Efficiency of the motor:

Calculation of Efficiency When the Machine Is Generating On Load
Power input = VI
Armature copper loss, PCU = I
2
Ra = (I + Ish)
2
Ra
Constant losses, WC = VI0 - (I0 - Ish)
2
Ra
Total losses = PCU + WC
Efficiency of the generator:

The main advantages of this test are :
- This test is very convenient and economical as it is required very less power from supply to
perform the test.
- Since constant losses are known, efficiency of Swinburne's test can be pre-determined at any load.
The main disadvantages of this test are :
1. Iron loss is neglected though there is change in iron loss from no load to full load due to armature
reaction.
2. We cannot be sure about the satisfactory commutation on loaded condition because the test is done on
3. We cant measure the temperature rise when the machine is loaded. Power losses can vary with the
temperature.
4. In dc series motors, the Swinburnes test cannot be done to find its efficiency as it is a no load test.
Hopkinson's test is another useful method of testing the efficiency of a dc machine. It is a full load test
and it requires two identical machines which are coupled to each other. One of these two machines is
operated as a generator to supply the mechanical power to the motor and the other is operated as a
motor to drive the generator. For this process of back to back driving the motor and the
generator, Hopkinson's test is also called back-to-back test or regenerative test.
If there are no losses in the machine, then no external power supply would have needed. But due to the
drop in the generator output voltage we need an extra voltage source to supply the proper input voltage
to the motor. Hence, the power drawn from the external supply is therefore used to overcome the
internal losses of the motor-generator set.
Connection Diagram of Hopkinson's Test

Here is circuit connection for the Hopkinson's test shown in figure below. A motor and a generator, both
identical, are coupled together. When the machine is started it is started as motor. The shunt field
resistance of the machine is adjusted so that the motor can run at its rated speed. The generator voltage is
now made equal to the supply voltage by adjusting the shunt field resistance connected across the
generator. This equality of these two voltages of generator and supply is indicated by the voltmeter as it
gives a zero reading at this point connected across the switch. The machine can run at rated speed and at
desired load by varying the field currents of the motor and the generator.

Calculation of Efficiency by Hopkinson's Test
Let, V = supply voltage of the machines.
Then, Motor input = V(I1 + I2)
I1 = The current from the generator
I2 = The current from the external source
And, Generator output = VI1..................(1)
Let, both machines are operating at the same efficiency ''.
Then, Output of motor = x input = x V(I1 + I2)
Input to generator = Output of the motor = X V(I1 + I2)
Output of generator = x input = x [ x V(I1 + I2)] =
2
V(I1 + I2)..................(2)
From equation 1 an 2 we get,
VI1 =
2
V(I1 + I2) or I1 =
2
(I1 + I2)

Now, in case of motor, armature copper loss in the motor = (I1 + I2 - I4)
2
Ra.
Ra is the armature resistance of both motor and generator.
I4 is the shunt field current of the motor.
Shunt field copper loss in the motor will be = VI4
Next, in case of generator armature copper loss in generator = (I1 + I3)
2
Ra
I3 is the shunt field current of the generator.
Shunt field copper loss in the generator = VI3
Now, Power drawn from the external supply = VI2
Therefore, the stray losses in both machines will be
W = VI2 - (I1 + I2 - I4)
2
Ra + VI4 + (I1 + I3)
2
Ra + VI3
Let us assume that the stray losses will be same for both the machines. Then,
Stray loss / machine = W/2
Efficiency of Generator
Total losses in the generator, WG = (I1 + I3)
2
Ra + VI3 + W/2
Generator output = VI1
Then, efficiency of the generator,

Efficiency of Motor
Total losses in the motor, WM = (I1 + I2 - I4)
2
Ra + VI4 + W/2
Motor input = V(I1 + I2)
Then, efficiency of the motor,

The merits of this test are
1. This test requires very small power compared to full-load power of the motor-generator coupled
system. That is why it is economical.
2. Temperature rise and commutation can be observed and maintained in the limit because this test is
3. Change in iron loss due to flux distortion can be taken into account due to the advantage of its full
The demerits of this test are
1. It is difficult to find two identical machines needed for Hopkinson's test.
2. Both machines cannot be loaded equally all the time.
3. It is not possible to get separate iron losses for the two machines though they are different because of
their excitations.
4. It is difficult to operate the machines at rated speed because field currents vary widely.
TESTING OF DC SERIES MACHINES:
Small series machines can be tested by break test similar to shunt machines, but the large serier
machines cannot be tested by Swinburnes test in the same way as shunt machines, because series motors
cannot be run on no-load due to dangerous high speed. In view of these, the field test is quite suitable dc
series machines. At the same time, there is no difficulty in having two similar dc series machines,
because series motors are used for traction purpose and are, therefore usually available in pairs.
However, the accuracy of this test depends on accuracy with which the motor input and generator
output are measured.
In this test two similar dc machines are required. These two machines are mechanically coupled
together and their fields are connected in series, as illustrated in fig. in order to make iron losses in both
machines equal. One of the machines operates as a motor and drives the other machine operating as a
l
R is directly connected to the armature without any switch.
The motor is started in the usual manner and the output of the generator is dissipated in the
) 1
(V across the motor terminals is kept equal to its rated value.
Obviously, the supply voltage (V) will be equal to ) (
1
V plus the voltage drop across the field winding of
the generator. The hot resistances of the various windings are measured by voltmeter ammeter or other
)
(
L
R is varied till the ammeter ) (
1
A shows the full load current of the
motor. Now the readings of various measuring instruments are noted.
Let supply voltage = reading of voltmeter V =V volts
Motor input current = reading of ammeter
1
A =
1
I
Terminal voltage of generator = reading of voltmeter
2
V =
2
V
2 2
I A =
Armature resistance of each machine =
a
R
Series field resistance of each machine =
se
R
Input to the whole set =
1
VI
Output =
2 2
I V
Total losses of the set,
2 2 1
I V VI P
T
=
Series field and armature copper losses of motor = ) (
2
1 se a
R R I +
Series field and armature copper losses of generator = ) (
2
1 se a
R R I +
Total copper losses of the set,
a se a c
R I R R I P
2
2
2
1
) 2 ( + + =
Stray power losses for the set =
c T
P P
Stray power losses per machine,
2
c T
s
P P
P

=
Motor efficiency:
Motor input =
1 1
I V
Motor losses =
s se a
P R R I + + ) (
2
1

Motor output =
s se a
P R R I I V + ) (
2
1 1 1

Motor efficiency,
1 1
2
1 1 1
) (
I V
P R R I I V
s se a
m
+
= q
Generator efficiency:
Generator output =
2 2
I V
Generator losses =
s se a
P R I R I + +
2
1
2
2

Generator input =
s se a
P R I R I I V + + +
2
1
2
2 2 2

Generator efficiency,
s se a
g
P R I R I I V
I V
+ + +
=
2
1
2
2 2 2
2 2
q

1. A relatively small error in the measurement of the input to motor or output of generator
may cause a relatively large error in the computed efficiency.
2. Total power supplied to the set is wasted.

RETARDATION TEST:
This test is also known as running down test. It is used for finding out the stray losses of
shunt wound dc machines.
In this method of testing of dc machines, machine under is speeded up slightly above its
normal speed and supply to the armature is cutoff. Consequently the armature slows down and
its kinetic energy is utilized to meet the rotational losses (stray losses).
Kinetic energy of the armature =
2
2
1
e J
Where J is the moment of inertia of armature and e is the angular speed at this instant.
Rotational losses
s
P = rate of change of kinetic energy= )
2
1
(
2
e J
dt
d
=
dt
d
J
e
e
Since
60
2 N t
e = ,where N is the speed of armature in rpm at that instant.
Rotational losses,
dt
dN
JN
N
dt
d N
J P
s
2
)
60
2
( )
60
2
( )
60
2
(
t t t
= =
Hence to determine stray losses, the value of moment of inertia (J) and rate of change of
speed
dt
dN
must be known.
The value of J and
dt
dN
are determined as follows.
(1). DETERMINATION OF
dT
dN
:
The machine under test is connected, as shown in fig. the voltmeter V connected across the
armature shows the instantaneous back emf of the motor. Since back emf of the motor is directly
proportional to speed( ) N E
b
o , the voltmeter can suitably calibrate to indicate the speed. When the
supply of armature is cutoff, the speed of motor decreases.
The speed or readings of voltmeter are noted at different intervals of time and a curve is drawn
between speed and time, as shown in fig. from any point P lying on speed time curve , tangent is
drawn meeting the X axis and Y axis at points A and B respectively , then

onds OAin
inrpm OB
dT
dN
sec
=
(2). DETERMINATION OF MOMENT OF INERTIA:
First speed time curve (slowing down curve) is plotted with armature alone as discussed above.
Then a flywheel of known moment of inertia ( )
1
J is keyed to the shaft and speed time curve is plotted
again. From the above two curves, the value of
|
|
.
|

\
|
|
|
.
|

\
|
2 1
dt
dN
and
dt
dN
are determined. Since the moment of
inertias in the second case is more and losses in both the cases are almost same, the value of
|
|
.
|

\
|
1
dt
dN
will be
more than value of
|
|
.
|

\
|
2
dt
dN
.
Hence; in the first case, rotational losses,
|
|
.
|

\
|
|
.
|

\
|
=
1
2
60
2
dt
dN
JN P
s
t

In second case rotational losses, ( )
|
|
.
|

\
|
+
|
.
|

\
|
=
2
1
2
60
2
dt
dN
N J J P
s
t

Comparing above two expressions we have
( )
|
|
.
|

\
|
|
.
|

\
|
=
|
|
.
|

\
|
+
|
.
|

\
|
1
2
2
1
2
60
2
60
2
dt
dN
JN N
dt
dN
J J
t t

( )
|
|
.
|

\
|
=
|
|
.
|

\
|
+ =>
1 2
1
dt
dN
J
dt
dN
J J
|
|
.
|

\
|
=
|
|
.
|

\
|
=>
2
1
2 1
dt
dN
J
dt
dN
dt
dN
J
|
|
.
|

\
|

|
|
.
|

\
|
= =>
2 1
2
1
dt
dN
dt
dN
dt
dN
J
J
Substituting values of
|
|
.
|

\
|
|
|
.
|

\
|
2 1
1
,
dt
dN
and
dt
dN
J in the expression the value of inertia of armature J can be
determined.
Second method:
In this method, first time is noted for the machine slow down by 5%. Let it be
1
t in seconds.
Then an additional load known as retarding torque, mechanical or electrical is applied and again the
time taken by the machine to slow down by 5% in noted, say
2
is applied by throwing the double throw switch
2
S over terminals 1 , 2 just after disconnecting the
armature from the supply mains. Thus a non inductive resistance R will be in the motor circuit and now
kinetic energy of the armature will supply the power to the load resistance R in addition to meet the
rotational losses. The additional losses say P in non inductive resistance will be equal to the product of
Now rotational losses,
1
2
60
2
dt
dN
JN P
s
|
.
|

\
|
=
t

And
2
2
60
2
'
dt
dN
JN P P
s s
|
.
|

\
|
= +
t

Dividing the above two expressions, we get
|
|
.
|

\
|
|
|
.
|

\
|
=
+
1
2
'
dt
dN
dt
dN
P
P P
s
s s

1 1
2
'
dt
dN
P
dt
dN
dt
dN
P
s s
=
|
|
.
|

\
|
=>
|
|
.
|

\
|

|
|
.
|

\
|
= =>
1 2
1
'
dt
dN
dt
dN
dt
dN
P P
s s

If the change of speed in both these cases is kept same and time taken is noted, let it be
2 1
andt t seconds
respectively, then
|
|
.
|

\
|

=
2 1
2
'
t t
t
P P
s s

SWINBURNS TEST:
It is the simplest indirect method of testing of dc machines in which losses are measured
separately and efficiency at any desired load pre determined. The machine is run as a motor on no-load;
the excitation is adjusted to give the normal speed and normal voltage applied across the motor
terminals. The connection diagram for determining no load losses of dc shunt machine is given in fig.
The resistance of field winding and armature are measured at room temperature (at 15 ) by
voltmeter-ammeter method. Since the windings of machines would be hot in those normal operating
conditions, the probable of hot resistances are computed from the measured cold values for a
temperature rise 40 C. From the data so obtained, different losses are computed and the efficiencies are
determined.
Let supply voltage, motor input current and shunt field current measured by voltmeter V,
ammeters
2 1
andA A connected in the circuit V ,
sh
andI I
0
respectively.
Since no-load mechanical output of machine is zero, whole of the input on no-load is used to
supply the internal losses of the machines i.e. shunt field copper loss, armature copper loss and stray
power losses in shunt wound machine and in case of compound wound machine input is also used to
supply field copper losses in addition to shunt field, armature and stray power losses.
SEPERATION OF LOSSES:
THEORY: in this experiment a dc machine is run on no-load as a motor. A resister R is connected in
series with the armature. The purpose of this resistance is achieved the desired speed. It is also used for
starting of dc machine as a motor. A resistance

## is connected in series with the field windings of the

machine. The purpose of this resistance is to adjust the excitation level. A schematic circuit is shown in
fig. but, we are interested in knowing the following losses separately.
(1). Frictional losses
(2). Wind age losses
(3). Hysteresis losses
(4). Eddy current losses
The total of all the above four losses can be determined by deducting the armature copper losses
(

) from the observed input to the armature of motor in reference fig. we can write,
The input to the armature =

## be represented by P. obviously, P would

represent the total of the above mentioned losses.
P= (frictional losses + wind age losses)+(hysteresis losses + eddy current losses)
N= speed of the motor
We, also know that
Frictional losses
Wind age lose

Hysteresis losses

The resistance R is kept at maximum value at the instant of starting of the motor. Also, it is
preferable of keep

## is adjusted to give suitable value of

the field current

the resistance R is reduced gradually, so that the motor picks up speed. Now, the
experimental set is ready for taking down the readings. The value of R is chosen so as to get suitable
running conditions. The value of armature voltage
a
V and armature current
a
I are recorded. These
observations are repeated for various values of R and therefore the speed N. the calculations for
separation of losses can be done as follows.
Under these running conditions, there is no useful output from the machine. The power fed to the
circuit is being consumed in the resistance R and
f
R field winding, armature winding, friction and wind
age losses and iron losses (comprising of hysteresis and eddy current losses). At present our interest
doesnt lie in knowing the losses R,
f
R field winding and armature winding.
Eddy current losses
2 2 2 2 2
Dn n B f B = = o o
Thus, combind equations we can write
2 2
CnDn Bn An P + + =

( ) ( )n D B C A
n
p
+ + + =
|
.
|

\
|

This equation tells that the graph between P/n and n would be a straight line with slope =(B+D) and
(A+C) as the intercept of p/n axis. These facts have been depicted in fig. This curve can help us in
determining the values of (A+C) and (B+D). In other words, we have achieved the separation of losses
proportional to n and those proportional to
2
n . It may be sufficient for many practical purposes.
However, if it is still required to separately determine all the four types of the losses, it can be done by
performing another similar experiment at a reduced excitation. Here, we are aiming to determine all
constants A, B, C and D separately. This reduced excitation can be obtained by increasing the resistance
of the field circuit ( )
f
R . The second experiment can be give data for drawing another curve similar to the
previous one. By changing the excitation level, we change the magnetic flux in the machine. From
equations, we know that the frictional losses dont depend up on the flux of the machine. But, the
hysteresis loss and the eddy current loss depend up on the magnetic flux. This means, that under the
conditions of the second experiment, the constant A and V remain unchanged, but the constants C and D
will change. Let C and D be their new values. The equation can be written as
( ) ( )n D B C A
n
p
+ + + =
|
.
|

\
|

Now, we can determine the values (A+C) and (B+D) by using the dotted curve of the fig. from
the values of (A+C), (B+D) and (A+C), (B+D) we can determine the quantities (C-C) and (D-D). we
know from equation
6 . 1 6 . 1
o| oB C
Therefore,
6 . 1
' '
|
|
.
|

\
|
=
|
.
|

\
|
|
|
C
C

Also using equation we can write,
2
' '
|
|
.
|

\
|
=
|
.
|

\
|
|
|
D
D

We know in a dc machine
|
.
|

\
|
A
ZNP
E
b
|
o
o|
b
E
From equations we can written as
6 . 1
' '
|
|
.
|

\
|
=
b
b
E
E
C
C

2
' '
|
|
.
|

\
|
=
b
b
E
E
D
D

Further, we can write
a a b
R I V E =

a a b
R I V E ' ' ' =
Combining equations we can write,
6 . 1
' ' '
|
|
.
|

\
|

=
|
.
|

\
|
a a
a a
R I V
R I V
C
C

2
' ' '
|
|
.
|

\
|

=
a a
a a
R I V
R I V
D
D

Using these relations we can thus determine the values of all the four constants namely A, B, C
and D which in turn can given us the values of all the four types of losses at any given speed.
PROCEDURE:
1. Connect the circuit as shown in figure
2. Keep R to its maximum value and switch on the dc supply
3. The motor starts running at low speed. Reduce the resistance R to minimum, so that motor
starts running at its full speed
4. Take down the readings of voltmeter, both the ammeters and the speed. Record these
observations as depicted in table
5. Change the value of resistance R several times and take the observations as described in
step 4
6. Increase the value of
f
R and repeat the steps 4 and 6
7. Bring the resistance R to its maximum and switch of the dc supply.
OBSERVATIONS AND CALCULATIONS: the observations of this experiment may be recorded as
depicted in table1

A similar table should be prepared for recording observations at the reduced excitations. The
calculations for A, B, C and D constants should be done as explained under theory.
Precautions:
1. All the connections should be tight
2. The resistance R connected in series with the armature of the dc machine should be kept at
its maximum value before on the dc supply
3. The zero settings of all the meters used should be checked before connecting them in the
circuit.