Lab 3: The Transformer

ELEC 3105
(last modified Dec. 2008)
Read this lab before your lab period. It is a long lab and requires the full 3 hours to complete.
Divide yourself into groups of 2 (or more if necessary). Someone must have the following roles:
-- experimentalist (taking measurements off the oscilloscope)
-- theorist (doing calculations and organizing the experimental data)
A demo is expected from your lab instructor.

1. Introduction
The objectives of this lab are to investigate:
a)

the relationships between current and voltage in the primary and secondary windings of a
transformer,
b) impedance transformation with a transformer
c) how well a widely-used equivalent circuit describes the behavior of a real transformer
d) how real transformers have a limited frequency range of useful operation.

2. Theory
The starting point for analyzing the transformer is the basic equivalent circuit of Fig. 1. In
this circuit, L1 is the inductance of the primary, L2 , the inductance of the secondary, and M the
mutual inductance between the two coils. Current and voltages in the transformer are described
by
~
~
(1a)
v~1 = L1 d i1 M d i2
dt
dt

~
d
~
v2 = M i1

dt

~
d
L2 i2

dt

(1b)

In the following we will use the symbol v~ to represent a complex voltage, and v to represent the
voltage amplitude. M is related to L1 and L2 by
(2)
M = k L1L2
where k is the fraction of the flux in each turn of coil 1 which also threads each turn of coil 2.
Also, if there are N1 turns in the primary and N 2 turns in the secondary

L1 L2 = a 2

(3)

where a = N1

is the turns ratio of the transformer. In a well-designed iron core transformer,

N2
k is close to 1. In an ideal transformer it is assumed that k = 1 . In this case

v1

v2

N1

N2

=a

(4)

If a load resistance RL is connected across the secondary of the transformer, it can also
be shown that the input impedance Zin = v1 seen "looking into" the primary is given
i1
approximately by
(5)
Z in = a 2 RL
The transformer can therefore provide a very valuable impedance transformation function. It
should be noted that (5) is an approximation which is only valid if L1 >> a 2 RL . In other
words, for impedance transformation to work it is vital that the magnitude of the reactance
associated with the transformer primary winding be much larger than the transformed load
resistance. Under this condition, it can also be shown that the currents in the primary and
secondary are related by
i2 = a
(6)
i1

Fig. 1 Basic transformer equivalent circuit.

It is often convenient to redraw the equivalent circuit of Fig. 1 in the form shown in Fig.
2 below. The ideal transformer in the center of this circuit has the same turns ratio a as the real
transformer, but has perfect flux coupling and infinite inductance in the primary and secondary
coils. The inductance (1 k )L1 is sometimes called the leakage inductance, while kL1 is the
magnetizing inductance. You should be able to show that this circuit gives exactly the same
relationship between voltages and currents at the terminals as the circuit of Fig. 1.

Fig. 2 Alternative transformer equivalent circuit.

Losses in real transformers are often approximately modeled by adding to resistances to
the equivalent circuit of Fig. 2, giving the circuit of Fig. 3. The resistor R1 represents the
resistance of the primary winding, and R2 the resistance of the secondary winding. Resistance
Rc approximately represents losses due to hysteresis and eddy currents in the core.

Fig. 3 Transformer equivalent circuit allowing for losses.

Fig. 3 is still an approximate description of the real transformer. It makes no allowance
for the capacitance between the windings in the primary and the secondary, for the fact that the
resistance of the windings is distributed throughout the coil and cannot be represented by a
single lumped resistor, nor for the dependence of core losses and core permeability on
frequency.
When the secondary windings are shorted, the equivalent circuit for the primary reduces
to that of Fig. 4. This circuit shows that by measuring the input impedance with the secondary
shorted it is possible to determine the winding resistance and the leakage inductance.

Fig. 4 Equivalent circuit with secondary shorted.

Usually we will have Rc and kL1 both much greater than the series combination of
2

R1 + a R2 and (1 k )L1 in which case the equivalent circuit reduces to that of Fig. 5.

Fig. 5 Simplified equivalent circuit with secondary shorted.

When the secondary is open circuit, the equivalent circuit becomes that of Fig,. 6

Fig. 6 Equivalent circuit with secondary open.

2. Equipment and Procedure

The transformer considered in this lab is a typical low cost audio transformer designed
to impedance match all 8 speaker to a transistor amplifier output stage over the frequency
range from approximately 20 Hz to 20 kHz. The core of this transformer consists of "soft" iron
plates laminated to reduce eddy current losses.
To complete the lab, it will be necessary to measure the current flowing in the
primary of the transformer. This will be done by connecting a small resistor in series with the
primary, and measuring the voltage drop across this resistor. The voltage drop will be amplified

by an amplifier board so that an output of 1 V is obtained for 1 mA current. The resistor and
amplifier are integrated in a readout board which should be connected as shown in Fig. 7.

Fig. 7 Circuit configuration for measurement of v1 and i1 .

Fig. 8 Experimental setup with major components

One side of the transformer primary is connected to the plug marked "IN2" on the
amplifier board, while the plug labeled "10 mA" is connected to ground. The other side of the

transformer primary is connected to the Wavetech, and also to the plug labeled "10V" on the
amplifier board. The output labeled Vout1 on the board gives v1 , while the output labeled
Vout 2 is proportional to i1 , with a scaling factor of 1 V per 1 mA.
How to use the Oscilloscope:

When measuring the traces on the oscilloscope it is important to maximize the trace area
to ensure as accurate a measurement as possible.
When taking a reading off the time axis between two signals, take the centre of the
signals if they are fuzzy.
For maximum accuracy, take the peak to peak voltage values (Vpp) instead of
amplitudes. As long as you are consistent in using peak to peak values, it makes no
difference in the calculations, since they are always used as ratios.
When taking a peak to peak reading off the voltage axis, take the reading from the topmost point to the bottom-most point, even if the signal is fuzzy.
At low frequencies such as 20Hz, switch from ALT to CHOP under vertical mode to
make the signal more clear.
One division is a square length and each tick is 0.2 (or 20%) of a division. So if Ch.1 is
set to 1V/div, and you read a peak to peak value of 1.2 divisions (one square and a tick
length), then the peak to peak voltage is 1V/div 1.2 divisions = 1.2Vpp.

To find the input impedance of the transformer, we will need to measure the phase angle
between v1 and i1 . This can be done by displaying both the voltage and current waveforms on
the oscilloscope and setting the time base to the longest value for which a half-period T/2 of both
waveforms is visible. Letting to be the time difference between the zero crossing of the voltage
waveform and the zero crossing of the current waveform, we have
t
(7)
= o 360 o
T
This technique is illustrated in Fig. 9.

Fig. 9 Measuring phase difference between v1 and i1 - voltage lags current in this case.

For this method to work, it is essential that the zero levels of the two
traces be accurately set!!

In an ideal inductor, current lags voltage by 90o. We will define this to be a positive
phase angle. If current leads voltage (the case in a capacitor), the phase angle is negative.
Throughout this lab, it is important to note whether current leads voltage or vice versa.
For an accurate reading, adjust the volts/div and sec/div until your display looks
somewhat like the right display in Fig 10.

Fig. 10 Side-by-side example of maximizing display settings to improve accuracy of results

The amplifier used in the measurement of the primary current introduces a phase shift
between its input and output, particularly for f > 10 kHz. To prevent this phase shift from
interfering with measurements, a second unity gain buffer amplifier is provided to measure v1 as
shown in Fig. 7.

3. Measurements and Calculations

Carry out the following measurements on the audio transformer. To do this, connect the output
of the Wavetech directly to the oscilloscope. Once the desired voltage has been set, you are then
to connect the output to the primary coil as is shown in Fig. 7.
a)

Using a digital multimeter, measure the transformers primary and secondary coil
resistances, R1 and R2, as shown in Fig. 3. Make sure the transformer is not connected to
anything. (1 mark)

b) In this experiment, we measure the turns ratio, a, of the audio transformer, assuming its
ideal. Set the Wavetech generator frequency to 1 kHz and use the low output. Set the
voltage amplitude to be v1 0.2V , or 0.4V peak to peak. To compute the turns ratio,

a=

v1

, use the oscilloscope to measure the voltage across the primary coil (v1 ) and across
v2
the secondary coil (v2 ) . Do not use the amplifier board yet. Use the external trigger feature
7

of the oscilloscope, and connect the external trigger input to the digital reference output of
the Wavetech. Compare to the turns ratio of the transformer specified by the manufacturer,
which is 7.9 : 1. (1 mark)
c)

Introduction to this question: In this experiment you will compute the input impedance
and its components (resistance and reactance) with the secondary coil short circuited. Using
the set-up shown in Fig. 7, you will measure the current and voltage in the primary to
calculate the ratio v1 , which gives the input impedance. Note that this will be complex as
i1
we have the effect of a resistance and an inductance connected in series, as shown in Fig. 5.
The resistance is R1 + R2 a 2 and the inductance is (1 k )L1 .
Procedure: Connect the transformer as shown in Fig. 7, short circuit the secondary coil,
and turn on the power supply to the amplifier so that the current flowing in the primary can
be measured. Set f = 1 kHz and v1 0.1V on the wavetech generator. Measure v1 and i1
and the phase angle between them. Remember that the output labeled "VOUT1" on the
board gives v1 , while the output labeled "VOUT2" is proportional to i1 , with a scaling factor
of 1 V per 1 mA. See Fig. 9 for instructions on how to measure the phase angle. Dont
forget to zero the levels of both signals.
Given Z in is represented as a resistance Rs in series with an inductance Ls the following is
true
v
and
(8)
Rs = 1 cos( )
Ls = v1 i sin ( )
i
1
1
Compute Rs and Ls . From Fig. 5, we should have Rs = R1 + R2 a 2 and Ls = 2(1 k )L1.
Compute R1 + R2 a 2 from the results of parts (a) and (b) and compare with the measured
value of Rs . Compute (1 k )L1 . Dont forget to show all of your calculations! (5 marks)

d) Introduction to this question: The purpose of this experiment is to measure the eddy
current losses (Rc) and kL1. For this we open circuit the secondary and hence, Fig. 6 comes
into effect. As in the previous experiment, you shall measure v1 and i1. From that you can
estimate the effective series resistance and the inductance of the circuit in Fig. 6 Rs and Ls
as before.
From Fig. 6 you will see that these values depend upon R1, (1-k)L1, Rc and kL1. But we
already know the values of R1 (part a) and (1-k)L1 (half the value of the inductance found in
(part c)), so to find the remaining unknowns, (Rc and kL1), you must use relationship (10),
below, and then use the values of Rs, Ls, R1 and (1-k)L1 to compute Rc and kL1 .
Procedure: Open circuit the secondary, and set f = 1 kHz and v1 0.2V . Record v1 , i1
and and re-calculate Rs and Ls from (8). To determine Rc and kL1 in the equivalent circuit

of Fig. 6, we need to account for the winding resistance R1 and leakage inductance (1 k )L1 .
Rs and Ls are defined as

and
(9)
Rs = Rs R1
Ls = Ls (1 k )L1
Additionally,

and
(10)
Rc = Z s / cos( )
kL1 = Z s / sin( )
where
2

Z s = Rs + Ls

(11)

and

(12)
tan( ) = Ls Rs
Compute Rc and kL1 . Using kL1 determined here, and (1 k )L1 found in part c), estimate
k (you have 2 equations and 2 unknowns). Dont forget to show all of your calculations! (6
marks)
e)

Repeat the measurement described in part d) at frequencies of 20 Hz, 100 Hz and 10 kHz.
Remember to keep track if the voltage is lagging or leading the current, which tells you if
the reactance of the primary is capacitive or inductive (see the Experiment and Procedure
section). Also remember to switch to CHOP for low frequencies, instead of ALT. Use the
below table arrangement to record your data (which includes Rc and kL1 ). Comment on the
ability of the equivalent circuit of Fig. 3 to accurately represent the behavior of the
transformer over this frequency range. Speculate on why Rc and kL1 appear to depend on
frequency. (4 marks)
f(hz)

f)

v1 (V)

i1 (A)

t0 (s)

T (s)

()

Rc ()

kL1 (H)

Repeat the measurement of v1 , i1 and at f = 100 kHz. Is the reactance seen looking into
the primary now inductive or capacitive? Suggest an explanation for your observation.
(3 marks)
F (Hz)

v1 (V)

i1 (A)

t0 (s)

T (s)

()

g) Connect a 10 - 12 load resistor across the secondary of the transformer. Set v1 0.2V ,
and measure v1 , i1 and at frequencies of 20 Hz, 1 kHz, 10 kHz and 100 kHz.
Representing Z in as a resistor R p in parallel with an inductance L p the following is true.

v
and
(13)
R p = 1 / cos( )
L p = v1 i / sin ( )
i
1
1
Construct a table showing R p and L p as functions of frequency. For an ideal transformer
we would have Rp = a 2 RL (see equation (5)) and L p would be infinite.
Note that experiment (g) (and (h) to follow) is a repetition of the previous experiments, with
2 changes:
1. the 10 - 12 ohm load on the secondary winding.
2. the resulting equivalent circuit looking into the source v1 is as a resistance and an
inductance in parallel, as opposed to serial, as has been the case in previous
experiments.
With v1 and i1, you will measure the resistance and inductance connected in parallel. Use
relationships (13) to measure the values of Rp and Lp. Compare the behavior of the real
transformer at different frequencies with the ideal model described in (5) and speculate why
your results do not satisfy the simple model, especially at low frequencies. The currents and
the voltages distort at some value of v1 as you increase it. Note that point and explain why.
Hint: What do you remember of inductors, when it comes to DC currents and very high
frequency AC currents? (8 marks)
F (Hz)

v1 (V)

i1 (A)

t0 (s)

T (s)

()

|Rp| ()

|Lp| (H)

h) Leave the 10 - 12 load resistor in place across the secondary and set f = 20 Hz. Change
the Wavetech output from low to high so that you have 20 Vpp output range instead of
2 Vpp. Increase v1 , until the i1 waveform starts to distort. Note the approximate values of
v1 and i1 at which the distortion begins. Make a sketch of the current waveform. Also
make a sketch of the v2 waveform under these conditions. Give a brief explanation for the
distortion. (2 marks)

10