SPECTRUM OF DIGITAL SIGNALS

1 Objective
The aim is to graphically portray the fundamental effects on digital signal within a frequency
spectrum and to understand the effects of filters on digital signal. We also aim to understand
the behaviour digital under frequency domain and time domain.

2 Task 1
RMS of a signal

50Hz sin wave

Vpeak=2.275V

Vrms=1.61V
measured with a
digital oscilloscope
Figure1: sine wave 50Hz
Vpeak to peak=4.56
measured with a
digital oscilloscope.

Vrms=1.61V

Multimeter reading Theoretical calculation

1.59Vrms V peak
Vrms = = 1.61V
2

The variation between the calculated and measured RMS values is very small, thus we
approach almost ideal cases. In most instances there is a large variation between calculated
values and measured values because practical components are defective.
 SPECTRUM OF DIGITAL SIGNALS

RMS voltage with zero DC offset

50Hz block wave

Vpeak=2.51V

Vpeak-peak=5.08V

Figure2: 50Hz square wave

Multimeter reading {Vrms=2.72V}

Theoretical computation {Vpeak=Vrms=2.51}, only for square waves.

3 Task 2
-3dB Filter Point

Filter Design

The required bandwidth is 100 kHz; a capacitor of 1.6nf was chosen to meet bandwidth
specifications.

1 1
R= = = 994.7Ω , 1 k Ω was chosen.
2πfc 2π × 1.6 × 10 −9 × 100000

Vout

Figure4: RC low pass filter circuit

 SPECTRUM OF DIGITAL SIGNALS

Table 1

freq(Hz) Vin(V) Vout(V) Vout/Vin log(freq)dB

10 1.5 1.53 1.02 1
20 1.59 1.59 1 1.30103
31 1.58 1.58 1 1.49136169
40 1.61 1.61 1 1.60205999
50 1.6 1.6 1 1.69897
60 1.61 1.61 1 1.77815125
70 1.63 1.63 1 1.84509804
80.9 1.66 1.66 1 1.90794852
90 1.63 1.64 1.006135 1.95424251
100 1.65 1.65 1 2
110 1.67 1.67 1 2.04139269
120 1.68 1.69 1.005952 2.07918125
130 1.62 1.62 1 2.11394335
140 1.61 1.62 1.006211 2.14612804
150 1.66 1.66 1 2.17609126
160 1.66 1.67 1.006024 2.20411998
170 1.61 1.62 1.006211 2.23044892
180 1.6 1.6 1 2.25527251
190 1.63 1.64 1.006135 2.2787536
200 1.64 1.65 1.006098 2.30103
251 1.61 1.62 1.006211 2.39967372
300 1.59 1.6 1.006289 2.47712125
350 1.59 1.6 1.006289 2.54406804
400 1.61 1.62 1.006211 2.60205999
450 1.6 1.61 1.00625 2.65321251
500 1.59 1.6 1.006289 2.69897
600 1.59 1.6 1.006289 2.77815125
700 1.6 1.61 1.00625 2.84509804
800 1.59 1.6 1.006289 2.90308999
900 1.6 1.61 1.00625 2.95424251
1000 1.6 1.61 1.00625 3
1500 1.59 1.6 1.006289 3.17609126
2000 1.59 1.6 1.006289 3.30103
2500 1.59 1.6 1.006289 3.39794001
3000 1.59 1.6 1.006289 3.47712125
3500 1.59 1.6 1.006289 3.54406804
4500 1.59 1.6 1.006289 3.65321251
5500 1.59 1.6 1.006289 3.74036269
6500 1.59 1.6 1.006289 3.81291336
7500 1.59 1.59 1 3.87506126
8500 1.59 1.59 1 3.92941893
9500 1.59 1.59 1 3.97772361
10000 1.59 1.59 1 4
11000 1.59 1.58 0.993711 4.04139269
12000 1.59 1.58 0.993711 4.07918125
13000 1.59 1.58 0.993711 4.11394335
14000 1.59 1.57 0.987421 4.14612804

Katlego Mohlala Electronics 4A01 Page 3

 SPECTRUM OF DIGITAL SIGNALS

15000 1.59 1.57 0.987421 4.17609126

16000 1.58 1.57 0.993671 4.20411998
17000 1.58 1.56 0.987342 4.23044892
18000 1.58 1.56 0.987342 4.25527251
19000 1.58 1.56 0.987342 4.2787536
20000 1.58 1.55 0.981013 4.30103
21000 1.58 1.55 0.981013 4.32221929
22000 1.58 1.54 0.974684 4.34242268
23000 1.58 1.54 0.974684 4.36172784
24000 1.58 1.53 0.968354 4.38021124
25500 1.57 1.52 0.968153 4.40654018
26500 1.58 1.52 0.962025 4.42324587
27500 1.57 1.52 0.968153 4.43933269
28500 1.58 1.51 0.955696 4.45484486
30000 1.57 1.5 0.955414 4.47712125
35000 1.57 1.48 0.942675 4.54406804
45000 1.57 1.42 0.904459 4.65321251
50000 1.56 1.4 0.897436 4.69897
55000 1.56 1.36 0.871795 4.74036269
60000 1.56 1.33 0.852564 4.77815125
65000 1.56 1.29 0.826923 4.81291336
70000 1.55 1.26 0.812903 4.84509804
75000 1.55 1.22 0.787097 4.87506126
80000 1.55 1.19 0.767742 4.90308999
85000 1.55 1.16 0.748387 4.92941893
90000 1.54 1.13 0.733766 4.95424251
95000 1.54 1.1 0.714286 4.97772361
100000 1.51 1.05 0.695364 5
105000 1.54 1.05 0.681818 5.0211893
110000 1.54 1.02 0.662338 5.04139269
115000 1.54 0.992 0.644156 5.06069784
125000 1.54 0.941 0.611039 5.09691001
135000 1.53 0.889 0.581046 5.13033377
145000 1.53 0.856 0.559477 5.161368
165000 1.53 0.778 0.508497 5.21748394

Input

Output

Figure5: input-output waveforms of a sine wave through a filter.

 SPECTRUM OF DIGITAL SIGNALS

Bode plot of trnsfer function vs

dB frequency
6

4
Vout/Vin
3 -3dB Point
Conner frequency
2

0
0 0.2 0.4 0.6 0.8 1 1.2
Log(frequency)(dB)

Figure6: bode plot

Power comparison

Assume a load resistance of 1Ω.

2
V 2 4.12 Vout 3. 2
Pin = = = 16 .81W P out = = = 10.24W
R 1 R 1
The out power is about a sixth of the input power.

Task 3
Spectrum of a Digital Signal

Time domain spectrum at a fundamental frequency of 500 Hz:

2.5V dc
offset

Figure7: Time domain representation of square wave with 2.5V DC offset at 500 Hz.
 SPECTRUM OF DIGITAL SIGNALS

Dc component
at 0Hz

2nd
st
1 Harmonic Harmonic
at at 700 Hz
500 Hz

Figure8: frequency spectrum of 500 Hz square wave with harmonics placed odd multiples of
the fundamental frequency.

Figure9: Frequency spectrum outline of 500 Hz square wave.

With reference to figure8, the pulse harmonics are decreasing due to the fact that a Fourier
transforms sums decreasing pulse amplitude with an increase in frequency.

Knee frequency

t rise1 = 1.656 × 10 −6 s , the rise time measured with digital oscilloscope?

1 1
k nee = = = 192.73kHz
π × Trise π × 1.656 × 10 − 6
 SPECTRUM OF DIGITAL SIGNALS

4 Task 4
Filtering a Digital Signal

Filter design

Fundamental frequency was chosen as 500 Hz.

Capacitance, c=4.7µf, we therefore calculate the required resistance.

1 1
R= = = 63.66Ω ≅ 64Ω , however a resistance of 63.66Ω was not
2πfc 2π (500)( 4.7 × 10 − 6 )
available during the practical therefore we used a 68Ω resistor. We then designed a low pass
RC filter from these parameters.
a)
Rise
Fall time
time tr
tf

Fall
Rise
time tf
time tr

Figure10: filter input output wave forms of the RC filter. Time domain spectra.

b)

figure11 (a) (b)

 SPECTRUM OF DIGITAL SIGNALS

Figure11: (a) depicts the FFT of the input waveform and (b) if the FFT of low pass filter
output. Frequency domain spectra.

With reference to figure11 (b) we can conclude that a low pass filter bypasses high frequency
components and suppresses low frequency components and the dc value remains unchanged.

Table 2

Input Output
Rise time 415.5μs 708μs
RMS
voltage 1.74V 1.04V
Fall time 503.5μs 776.6μs
Vpeak-
peak 4.48V 3.44V

1
k nee − frequency = = 766 Hz , input signal.
πt rise

1
k nee − frequency = = 449.95 Hz , output signal.
πt rise

Verification calculations

Filter rise time:

1 1
t rise = = = 636.6us
πf π × 500

Measured rise time value=415.5µs. This could have been caused by the variation in the
resistance value, we designed for 63.6Ω but we only had 68Ω available, however the
difference is good enough for practical purposes.

t rise1 = 1.656 × 10 −6 s

2 2
Output rise time= (t rise1 + t rise 2 ) = 415.5us which is the same as the measured value.

Katlego Mohlala Electronics 4A01 Page 8

 SPECTRUM OF DIGITAL SIGNALS

5 Task 5
5.1 Effect of bandwidth on RC filters

We cascade two filters as follows:

Figure12: series cascaded filters

Filter1 has a bandwidth of 500Hz and filter2 has a bandwidth of 250Hz .

a) Transfer function
Vout  1  1 
H ( s) = =   
Vin  1 + sR1C1  1 + sR2 C 2 

b)

Figure12: Input and output of cascaded filters. Same analogy in figure10 applies.
 SPECTRUM OF DIGITAL SIGNALS

Input having High frequency

high components fall outside
frequency the bandwidth, other
components frequencies are
bypassed

figure13: FFT (a) input (b) output

c)

Table 3

Input Output
Rise time 3.608μs 760μs
RMS
voltage 2.09V 548mV
Fall time 3.608μs 756μs
Vpeak-
peak 4.56V 1.92V

1 1
k nee − frequency = = = 88kHz
πt rise π × 3.608 × 10 −6

Verification Calculations

Rise time 500Hz filter= 636.6µs

1
Rise time 250Hz filter= = 1.27 ms
π × 250

Conclusion

The practical was succesful we managed to draw a graphical interface into the insight of
filters affect digital signals and how signals are effect by variation in frequency. The -3dB
point was observed and and we saw how the FFT functions relates to spectrum.
 SPECTRUM OF DIGITAL SIGNALS

Katlego Mohlala Electronics 4A01 Page 11