Name:

Class:

6 - AM demodulation
Experiment 6 – AM demodulation

Preliminary discussion
If you’ve completed Experiment 4 then you’ve seen what happens when you use a 2kHz sinewave
to amplitude modulate a carrier to produce an AM signal. Importantly, you would have seen a key
characteristic of an AM signal – its envelopes are the same shape as the message (though the
lower envelope is inverted).

Recovering the original message from a modulated carrier is called demodulation and this is the
main purpose of communications and telecommunications receivers. The circuit that is widely
used to demodulate AM signals is called an envelope detector. The block diagram of an envelope
detector is shown in Figure 1 below.

RC
Rectifier
LPF
Recovered
AM signal message

Rectified AM signal

Figure 1

As you can see, the rectifier stage chops the AM signal in half letting only one of its envelopes
through (the upper envelope in this case but the lower envelope is just as good). This signal is
fed to an RC LPF which tracks the peaks of its input. When the input to the RC LPF is a
rectified AM signal, it tracks the signal’s envelope. Importantly, as the envelope is the same
shape as the message, the RC LPF’s output voltage is also the same shape as the message and so
the AM signal is demodulated.

A limitation of envelope detector shown in Figure 1 is that it cannot accurately recover the
message from over-modulated AM signals. To explain, recall that when an AM carrier is over-
modulated the signal’s envelope is no-longer the same shape as the original message. Instead,
the envelope is distorted and so, by definition, this means that the envelope detector must
produce a distorted version of the message.

6-2 © Emona Instruments Experiment 6 – AM demodulation

The experiment
In this experiment you’ll use the Emona Telecoms-Trainer 101 to generate an AM signal by
implementing its mathematical model. Then you’ll set-up an envelope detector using the Rectifier
and RC LPF on the trainer’s Utilities module.

Once done, you’ll connect the AM signal to the envelope detector’s input and compare the
demodulated output to the original message and the AM signal’s envelope. You’ll also observe the
effect that an over-modulated AM signal has on the envelope detector’s output.

Finally, if time permits, you’ll demodulate the AM signal by multiplying it with a local carrier
instead of using an envelope detector.

It should take you about 1 hour to complete Parts A to D of this experiment and another 20
minutes to complete Part E.

Equipment

▪ Emona Telecoms-Trainer 101 (plus power-pack)

▪ Dual channel 20MHz oscilloscope

▪ two Emona Telecoms-Trainer 101 oscilloscope leads

▪ assorted Emona Telecoms-Trainer 101 patch leads

▪ one set of headphones (stereo)

Procedure

Part A – Setting up the AM modulator

To experiment with AM demodulation you need an AM signal. The first part of the experiment
gets you to set one up.

1. Gather a set of the equipment listed above.

2. Set up the scope per the instructions in Experiment 1. Ensure that:

▪ the Trigger Source control is set to the CH1 (or INT) position.
▪ the Mode control is set to the CH1 position.

3. Set the scope’s Channel 1 Input Coupling control to the DC position.

4. Locate the Adder module and turn its G and g controls fully anti-clockwise.

5. Locate the Variable DCV module and turn its DC Voltage control almost fully anti-
clockwise.

Experiment 6 – AM demodulation © Emona Instruments 6-3

6. Connect the set-up shown in Figure 2 below.

Note: Insert the black plugs of the oscilloscope leads into a ground (GND) socket.

Figure 2

This set-up can be represented by the block diagram in Figure 3 below. It generates a 100kHz
carrier that is amplitude modulated by a 2kHz sinewave message.

Figure 3

6-4 © Emona Instruments Experiment 6 – AM demodulation

Note: Although the scope is now displaying the Adder module’s output, it’ll just be a straight
line because the output is 0V.

7. Use the scope’s Channel 1 Vertical Position control to move the trace so that it lines up
with the horizontal line in the middle of the scope’s screen.

8. While watching the Adder module’s output on the scope, turn its g control until the DC
level is 1V.

Tip: It is probably most convenient to do this with the scope’s Ch-1 Vertical Attenuation
control on the 0.5V/div setting.

9. While watching the Adder module’s output on the scope, turn its G control to obtain a
1Vp-p sinewave.

Tip: You may need to adjust the scope’s Timebase control to view two or so cycles of the
signal.

10. Set the scope’s Mode control to the DUAL position to view the AM signal out of the
Multiplier module as well as the message signal.

11. Measure the AM signal’s depth of modulation.

Tip: If you’re not sure how to do this, see Experiment 4 (page 4-14).

If you’ve connected the set-up and adjusted the controls correctly, the AM signal has a depth
of modulation of about 0.5.

Ask the instructor to check

your work before continuing.

Experiment 6 – AM demodulation © Emona Instruments 6-5

Part B – Recovering the message using an envelope detector

12. Modify the set-up as shown in Figure 4 below.

Remember: Dotted lines show leads already in place.

Figure 4

The additions to the set-up in Figure 4 can be represented by the block diagram in Figure 5
below. As you can see, it’s the envelope detector explained in the preliminary discussion.

To Ch.2

AM Demodulated
Rectifier RC LPF
signal AM signal

Figure 5

6-6 © Emona Instruments Experiment 6 – AM demodulation

13. Adjust the scope’s Vertical Attenuation controls to appropriate settings for the signals.

14. Draw the two waveforms to scale in the space provided below leaving room to draw a third
waveform.

Tip: Draw the message signal in the upper third of the graph and the rectified AM signal
in the middle third.

15. Disconnect the scope’s Channel 2 input from the Rectifier’s output and connect it to the
RC LPF’s output instead.

16. Draw the demodulated AM signal to scale in the space that you left on the graph paper.

Experiment 6 – AM demodulation © Emona Instruments 6-7

 Question 1
What is the relationship between the original message signal and the recovered message?

They’re the same (not withstanding the phase shift).

Ask the instructor to check

your work before continuing.

Part C – Investigating the message’s amplitude on the recovered message

17. Vary the message signal’s amplitude up and down a little (by turning the Adder module’s G
control left and right a little) while watching the demodulated signal.

Question 2
What is the relationship between the amplitude of the two message signals?

The bigger the original message signal’s amplitude, the bigger the recovered message

signal’s amplitude.

18. Slowly increase the message signal’s amplitude to maximum while watching the
demodulated signal.

Question 3
What do you think causes the heavy distortion of the demodulated signal? Tip: If you’re
not sure, connect the scope’s Channel 1 input to the AM modulator’s output and set its
Trigger Source control to the CH2 position.

Over-modulation

Question 4
Why does over-modulation cause the distortion?

Over-modulation distorts the AM signal’s envelope which upsets the operation of the

envelope detector.

6-8 © Emona Instruments Experiment 6 – AM demodulation

 Ask the instructor to check
your work before continuing.

Part D – Transmitting and recovering speech using AM

This experiment has set up an AM communication system to “transmit” a message that is a 2kHz
sinewave. The next part of the experiment lets you use the set-up to modulate, transmit,
demodulate and listen to speech.

19. If you moved the scope’s Channel 1 input to help you answer Question 4, reconnect it to
the Adder module’s output and return the scope’s Trigger Source control to the CH1
position.

20. Adjust the message signal’s amplitude back to 1Vp-p (by turning the Adder module’s G
control anti-clockwise).

21. Modify the set-up as shown in Figure 6 below.

Figure 6

Experiment 6 – AM demodulation © Emona Instruments 6-9

22. Set the scope’s Timebase control to the 2ms/div position.

23. Turn the Buffer module’s Gain control fully anti-clockwise.

24. Without wearing the headphones, plug them into the Buffer module’s headphone socket.

25. Put the headphones on.

26. As you perform the next step, set the Buffer module’s Gain control to a comfortable
sound level.

27. Talk, sing or hum while watching the scope’s display and listening on the headphones.

Ask the instructor to check

your work before continuing.

Part E – The mathematics of AM demodulation

AM demodulation can be understood mathematically because it is uses multiplication to
reproduce the original message. To explain, recall that when two pure sinewaves are multiplied
together (a mathematical process that necessarily involves some trigonometry that is not shown
here) the result gives two completely new sinewaves:

▪ One with a frequency equal to the sum of the two signals’ frequencies

▪ One with a frequency equal to the difference between the two signals’ frequencies

The envelope detector works because the rectifier is a device that multiplies all signals on its
one input with each other. Ordinarily, this is a nuisance but not for applications like AM
demodulation. Recall that an AM signal consists of a carrier, the carrier plus the message and
the carrier minus the message. So, when an AM signal is connected to a rectifier’s input,
mathematically the rectifier’s cross multiplication of all of its sinewaves looks like:

Rectifier’s output = carrier × (carrier + message) × (carrier – message)

6-10 © Emona Instruments Experiment 6 – AM demodulation

If the message signal used to generate the AM signal is a simple sinewave then, when the
equation above is solved, the rectifier outputs six sinewaves at the following frequencies:

▪ Carrier + (carrier + message)

▪ Carrier + (carrier - message)

▪ (carrier + message) + (carrier - message)
▪ Carrier - (carrier + message) which simplifies to just the message

▪ Carrier - (carrier - message) which also simplifies to just the message

▪ (carrier + message) - (carrier - message)

To make this a little more meaningful, let’s do an example with numbers. The AM modulator that
you set up at the beginning of this experiment uses a 100kHz carrier and a 2kHz message (with
a DC component). So, the resulting AM signal consists of three sinewaves: one at 100kHz,
another at 102kHz and a third at 98kHz. Table 1 below shows what happens when these
sinewaves are cross-multiplied by the rectifier.

Table 1 100kHz×102kHz 100kHz×98kHz 98kHz×102kHz

Sum 202kHz 198kHz 200kHz

Difference 2kHz 2kHz 4kHz

Notice that two of the sinewaves are at the message frequency. In other words, the message
has been recovered! And, as the two messages are in phase, they simply add together to make a
single bigger message.

Importantly, we don’t want the other non-message sinewaves so, to reject them but keep the
message, the rectifier’s output is sent to a low-pass filter (which explains why a low-pass filter
can double as a peak detector). Ideally, the filter’s output will only consist of the message
signal. The chances of this can be improved by making the carrier’s frequency much higher than
the highest frequency in the message. This in turn makes the frequency of the “summed” signals
much higher and easier for the low-pass filter to reject.

[As an aside, the 4kHz sinewave that was generated would pass through the low-pass filter as
well and be present on its output along with the 2kHz signal. This is inconvenient as it is a signal
that was not present in the original message. Luckily, as the signal was generated by multiplying
the sidebands, its amplitude is much lower than the recovered message and can be ignored.]

An almost identical mathematical process can be modelled using the Emona Telecoms-Trainer
101’s Multiplier module. However, instead of multiplying the AM signal’s sinewaves with each
other (the Multiplier module doesn’t do this), it must be multiplied with an additional pure
100kHz sinewave. The next part of this experiment lets you demodulate an AM signal this way.

Experiment 6 – AM demodulation © Emona Instruments 6-11

28. Modify the set-up to return it to just an AM modulator with a 2kHz sinewave for the
message as shown in Figure 7 below.

Figure 7

29. Check that the message signal’s amplitude is 1Vp-p. If not, adjust the Adder module’s G
control until it is.

Tip: You’ll need to return the scope’s Timebase control to its earlier setting if you’ve not
done so already.

6-12 © Emona Instruments Experiment 6 – AM demodulation

30. Modify the set-up as shown in Figure 8 below.

ADDER MULTIPLIER UTILITIES

COMPARATOR
REF

DC
X
AC

DC
Y IN OUT
AC
RECTIFIER
G kXY
MULTIPLIER
A
DIODE & RC LPF

X DC
RC LPF
g

B GA+gB Y DC kXY

VARIABLE MASTER
DCV SIGNALS

+5V

VDC

GND

VDC

100kHz
SINE
100kHz
COS
SPEECH
100kHz
DIGITAL
8kHz
A EXOR DIGITAL
2kHz
DIGITAL
2kHz
SINE
B A+B

Figure 8

The additions to the set-up can be represented by the block diagram in Figure 9 below. The
Multiplier module models the mathematical basis of AM demodulation and the RC Low-pass filter
on the Utilities module picks out the message while rejecting the other sinewaves generated.

Figure 9

Experiment 6 – AM demodulation © Emona Instruments 6-13

31. Compare the Multiplier module’s output with the Rectifier’s output that you drew earlier
(see page 6-7).

Question 5
Given the AM signal (which consists of 100kHz, 102kHz and 98kHz sinewaves) is being
multiplied by a 100kHz sinewave:

A) How many sinewaves are present in the Multiplier module’s output?

B) What are their frequencies?

A) Five.

B) One at 198kHz, one at 200kHz, a third at 202kHz and two sinewaves at 2kHz.

These latter two are in-phase and combine to form one larger one. So technically, you

could argue that there are only four. (Note: The 4kHz sinewave isn’t generated.)

32. Disconnect the scope’s Channel 2 input from the Multiplier module’s output and connect it
to the RC LPF’s output instead.

33. Compare the RC LPF’s output with the message and the output RC LPF’s that you drew
earlier (see page 6-7).

Ask the instructor to check

your work before continuing.

A common misconception about AM is that, once the signal is over-modulated, it’s impossible to
recover the message. However, when the AM signal is generated using an ideal or near-ideal
modulator (like Figure 3) this is only true for the envelope detector.

The AM demodulation method being implemented in this part of the experiment (called product
detection – though it is more accurate to call it product demodulation) doesn’t suffer from this
problem as it’s not designed to recover the message by tracking one of the AM signal’s
envelopes. The final part of this experiment demonstrates this.

6-14 © Emona Instruments Experiment 6 – AM demodulation

34. Connect the scope’s Channel 1 input to the AM modulator’s output.

35. Set the scope’s Trigger Source control to the CH 2 position.

36. Slowly increase the message signal’s amplitude to produce a near 100% modulated AM
signal by adjusting the Adder module’s G control.

Note: Resize the AM and demodulated message signals on the scope’s screen as
necessary.

37. Slowly increase the message signal’s amplitude to produce an AM signal that is modulated
by more than 100% while paying close attention to the demodulated message signal.

Note: Notice that the demodulated message signal is not distorted even though the
carrier is over-modulated.

As an aside, the commercial implementation of AM modulation commonly involves a Class C

amplifier for efficiency (that is, to minimise power losses). When a Class C amplifier is operated
at depths of modulation above 100% the circuit’s operation no-longer corresponds with the
model of an AM modulator in Figure 3. Importantly, in addition to producing an envelope that is
not the same as the original message, the over-modulated Class C circuit produces extra
frequency components in the spectrum. This means that neither the envelope detector nor the
product demodulator can reproduce the message without distortion.

Ask the instructor to check

your work before finishing.

Experiment 6 – AM demodulation © Emona Instruments 6-15

MATLAB code:
clear all;
clc;
close all;

fs = 1e5;
t = 0:1/fs:0.1;
Ac = 1;
fc = 1e3;
Am = 0.5;
fm = 100;

message_signal = Am * sin(2 * pi * fm * t);

figure;
subplot(2, 1, 1);
plot(t, message_signal, 'b');
title('Message Signal (Time Domain)');
xlabel('Time (s)');
ylabel('Amplitude');
grid on;

% Compute and plot the frequency spectrum of the message signal

message_fft = abs(fftshift(fft(message_signal)));
f = linspace(-fs/2, fs/2, length(message_fft));
subplot(2, 1, 2);
plot(f, message_fft / max(message_fft), 'r');
title('Message Signal (Frequency Domain)');
xlabel('Frequency (Hz)');
ylabel('Normalized Magnitude');
grid on;

% Generate carrier signal

carrier_signal = Ac * cos(2 * pi * fc * t);

% Plot the carrier signal

figure;
subplot(2, 1, 1);
plot(t, carrier_signal, 'b');
title('Carrier Signal (Time Domain)');
xlabel('Time (s)');
ylabel('Amplitude');
grid on;

% Compute and plot the frequency spectrum of the carrier signal

carrier_fft = abs(fftshift(fft(carrier_signal)));
subplot(2, 1, 2);
plot(f, carrier_fft / max(carrier_fft), 'r');
title('Carrier Signal (Frequency Domain)');
xlabel('Frequency (Hz)');
ylabel('Normalized Magnitude');
grid on;

% Amplitude modulation
modulated_signal = (1 + message_signal) .* carrier_signal;

6-16 © Emona Instruments Experiment 6 – AM demodulation

% Plot the modulated signal
figure;
subplot(2, 1, 1);
plot(t, modulated_signal, 'b');
title('Modulated Signal (Time Domain)');
xlabel('Time (s)');
ylabel('Amplitude');
grid on;

% Compute and plot the frequency spectrum of the modulated signal

modulated_fft = abs(fftshift(fft(modulated_signal)));
subplot(2, 1, 2);
plot(f, modulated_fft / max(modulated_fft), 'r');
title('Modulated Signal (Frequency Domain)');
xlabel('Frequency (Hz)');
ylabel('Normalized Magnitude');
grid on;

% Manual envelope detection using absolute value and low-pass filtering

rectified_signal = abs(modulated_signal); % Full-wave rectification
disp('Done')
% Design a low-pass filter
fc_lp = 200;
[b, a] = butter(2, fc_lp/(fs/2)); % 6th-order Butterworth low-pass filter
envelope = filter(b, a, rectified_signal); % Apply low-pass filter

% Plot the demodulated signal (envelope)

figure;
subplot(2, 1, 1);
plot(t, envelope, 'b');
title('Demodulated Signal (Envelope, Time Domain)');
xlabel('Time (s)');
ylabel('Amplitude');
grid on;

% Compute and plot the frequency spectrum of the demodulated signal

demodulated_fft = abs(fftshift(fft(envelope)));
subplot(2, 1, 2);
plot(f, demodulated_fft / max(demodulated_fft), 'r');
title('Demodulated Signal (Frequency Domain)');
xlabel('Frequency (Hz)');
ylabel('Normalized Magnitude');
grid on;

Experiment 6 – AM demodulation © Emona Instruments 6-17

6-18 © Emona Instruments Experiment 6 – AM demodulation
Experiment 6 – AM demodulation © Emona Instruments 6-19
Simulink model:

Components:
• Signal generator
• Constant
• Sum
• Product
• Analog filter design
• Zero order hold
• DC blocker
• Scope

Procedure:
• Set message signal frequency to 100Hz and add a DC offset of 2.
• Then product with carrier signal frequency of 5000Hz.
• The output signal would be an Amplitude Modulated signal.
• To demodulate it in a coherent method, first produce the modulated signal with the
same carrier frequency.
• Pass the output through a low pass filter.
• To eliminate the DC offset, the signal needs to be sampled and passed through a DC
blocker.

Output:
Modulated:

6-20 © Emona Instruments Experiment 6 – AM demodulation

Demodulated:

Report Question for Matlab and SIMULINK:

• Write a MATLAB code for coherent detection of AM modulated signal.
• Design a SIMULINK system to demodulate an amplitude modulated signal using a
non-coherent method.

Experiment 6 – AM demodulation © Emona Instruments 6-21