SOLUTION to ELE22EMI 2004 NOVEMBER Exam

Question 1.
(a) What are the name and symbol for the fundamental SI unit of luminous intensity?
Candela (Cd).
(b) Given one example each of a direct, an indirect, and a null measurement.
Direct measurement: measuring the dimensions of A4 paper by placing a ruler beside
each edge of the paper.
Indirect measurement: measuring temperature using a mercury thermometer.
Null measurement: measuring resistance using a Wheatstone bridge.
(c) What kind of AC bridge is suitable for measuring:

(1) High-Q inductors? Hay (or Owens).

(2) Capacitance? Schering.
(3) Low-Q inductors? Maxwell (or Owens).

Question 2.
(a) Give five properties of an ideal operational amplifier.
Infinite gain, bandwidth, input impedance, power supply rails. Zero output impedance,
noise generation, phase shift.
(b) Draw the circuit for an op-amp differentiator.
Rf

Ci
Vi
Vo
+

(c) In a differential op-amp:

(1) Define the input offset voltage. The potential difference between the negative and the
positive inputs (va = v v+ ). [NB: The value of the offset voltage when the output
voltage is zero (vo = 0) is called the zero offset.]
1
 (2) Why does a real op-amp have a nonzero input offset voltage? Because its differential
gain is not infinite and its common-mode gain is nonzero.

(3) Why do its inputs have nonzero currents? Because its input impedance is finite.

Question 3.
(a) Name two applications of isolation amplifiers.

(1) In surgery, separating the currents of the (ungrounded) very low-current sensor
attached to the patient from the biologically dangerous currents of the (grounded)
electrical equipment that processes and displays the sensors signals.

(2) Signal acquisition in high voltage circuits. The isolation amplifier protects the low
voltage electronics from the high voltage signal sources.

(b) What is the unit for:

(1) noise specification? V/ Hz.

(2) voltage drift? V/K or, equivalently, V/ o C.

(c) Describe two (2) situations in which the effect of noise can be reduced. In each case, explain
the noise reduction technique that you would use, and why it works.

(1) Still images with salt-and-pepper noise can be be improved by taking successive pho-
tographs and averaging. The S&P noise is random, so it diminishes when averaged.

(2) Mains power noise can be eliminated by a 50 Hz notch filter. This works if the
desired signals do not include any frequencies near 50 Hz because the notch filter
rejects frequencies immediately near 50 Hz and passes all other frequencies.

Question 4.
(a) Strain gauges:

(1) What is another name for the sensitivity S of a strain gauge?

Gauge factor.

(2) Write the formula for S in terms of the resistance R and the length L of a conductive bar
under longitudinal compression.
R/R
S=
L/L

(b) Thermistors:

page 2 of 12 OVER/
 (1) Write the formula for the resistance R of a thermistor as a function of absolute tempera-
ture T, coefficient , reference temperature T0 and reference resistance R0.
R
= exp[( )]
R T T

(2) If a thermistor has a resistance of 300 k at 20 degrees Celsius, and a coefficient =

4000 K, then at what temperature is its resistance equal to 280 k?
k
= exp [ K( )]
k T + . K
K
T =
ln( ) + .
T = . K = . oC

(c) For a thermocouple, what is the name of the effect whereby a potential applied between the
junctions generates a temperature gradient?
Peltier effect.

Question 5.
(a) In the 74LS163 (see following diagram), what is the function of the control inputs labelled
CLK, CLR, LD, ENP and ENT, and what does it mean if the output RCO is high?
CLK
CLR
LD
ENP
ENT
A QA
B QB
C QC
D QD
RCO

CLK = CLocK, which paces the counter (controls the rate at which it counts up).

CLR = CLeaR, which when 0 sets all Q bits to 0.

LD = LoaD, which copies (A B C D) to (QA QB QC QD).

ENP = ENable Pulse, which when 0 halts the counter by preventing the Q outputs
from changing, thus making the CLK input ineffective.

ENT = ENable Transition, which when 0 compels the RCO output to remain 0.

RCO = 1 means that ENT is 1 and the Q bits are all 1 so they represent the hex
number F (decimal 15), and the next number will be all 0. RCO acts as a carry-out
from the 4-bit counter and is intended as a carry-in to the next higher stage.
page 3 of 12 OVER/
(b) Using a 74LS163, draw the circuit for a counter with counting sequence 4, 5, . . . , 12.
+5 V

Rpu

CLOCK CLK
CLR
LD
ENP
ENT Q0
A QA Q1
B QB Q2
C QC Q3 /CNT12
D QD
RCO

(c) Using as many 74LS163 ICs as necessary, draw the circuit diagram for an eight-bit counter.

CLOCK CLK
/RESET CLR
/LOAD LD
CNTEN ENP
ENT
D0 A QA Q0
D1 B QB Q1
D2 C QC Q2
D3 D QD Q3
RCO RCO4

CLK
CLR
LD
ENP
ENT
D4 A QA Q4
D5 B QB Q5
D6 C QC Q6
D7 D QD Q7
RCO RCO8

Question 6.
(a) For the simple Schmitt trigger in the next diagram, assume that the maximum and minimum
output voltages are +12V and -12V, and that R1 = 3 k and R2 = 6 k. Calculate the input
voltages at which the transitions occur.

page 4 of 12 OVER/
 Vo
+12V
Vi
Vo ?
+ Vi
R1 ?
R2
12V

v+ R1
=
vo R1 + R2

so the upper and lower thresholds for v+ can be found:

vo = +12V v+ = +4V

vo = 12V v+ = 4V

Therefore if vo = +12V and vi exceeds +4V, then vo will swing to -12V; conversely, if
vo = 12V and vi drops below -4V, then vo will swing to +12V.
(b) State the sampling theorem.
A bandlimited signal may be reconstructed exactly from a sample if the sampling rate is
at least twice the signals bandwidth.
(c) An eight-bit DAC has a minimum voltage of zero volts (for the binary input zero) and a
maximum voltage of 5.10 V. What binary number produces the output voltage 3.20 V?
Eight bits implies a digital range of 0 to 28 1 = 255. Thus each increment corresponds
to a voltage step of 5.100.00
2550 V = 0.02 V = 20 mV. Each digital number n corresponds to a
voltage of 0.00 V +n 20 mV = 20n mV. Conversely, a voltage v corresponds to a number
v
n = 20mV . Thus 3.20 V which equals 3200 mV corresponds to n = 320020 = 160.

Question 7.
(a) Describe amplitude modulation (AM).
In AM, the modulated carriers voltage amplitude is a linear function of the signal ampli-
tude:
cm (t) = C0 [1 + s(t)] cos(ct)

If the signal s(t) is a sinusoid, then:

cm (t) = C0 [1 + m cos(st)] cos(ct)

page 5 of 12 OVER/
(b) Describe frequency modulation (FM).
In FM, the modulated carriers frequency is a linear function of the signal amplitude:
t
cm (t) = C0 cos{[ct + s(t1)dt1}
0

In the simple case of a sinusoidal signal, s(t) = cos(st), the formula for the modulated
carrier voltage evaluates to:

cm (t) = C0 cos{[c + sin(st)]t}
s

(c) Quote the formula for the modulation index m of an FM system, and name all the variables
in the formula.
This applies to a sinusoidal signal s(t) :
f
m= =
fm s

where fm is the modulating frequency which is the instantaneous frequency fs = s /2 of

the signal s(t), and f is the maximum frequency deviation which is the difference between
the instantaneous frequencies of the modulated and unmodulated carriers.
(d) Define the following terms:

(1) rise time;

Rise time is the interval during which a pulses rising edge increases from 10% to
90% of its amplitude.

(2) ringing.
Ringing is the (unwanted) oscillation of a circuit under test (CUT) that can superim-
pose on its normal frequency response when the input of the CUT is frequency-swept
too quickly.

Question 8.
(a) Name the three (3) functional parts of a cathode-ray tube (CRT).

(1) Electron gun.

(2) Deflection plates.

(3) Phosphor screen.

page 6 of 12 OVER/
(b) Give the formula relating the oscilloscope bandwidth BW to the rise time tr .
0.35
BW =
tr

(c) Draw the waveform produced by a sweep generator. Label each of the sections and signifi-
cant points of the waveform.

Sweep voltage
Beam at far right
of screen

Vmax
Sweep ramp Retrace

Vmin

Beam at far left Wait for trigger

of screen
t
sweep interval TS TR retrace interval

(d) Name two (2) advantages of sampling oscilloscopes over standard oscilloscopes.

(1) Much greater bandwidth.

(2) Digital storage of waveforms.

Question 9.
(a) What is the approximate capacitance of a one metre (1 m) length of small-diameter coaxial
cable?
100 pF.
(b) If a CRO has an input impedance of 1 M in parallel with 30 pF, and is connected to 90 cm
of coaxial cable to a signal generator producing a 1 MHz sine wave, then what is the capacitive
impedance of the combination of CRO and cable?
Ctotal = CCRO +Ccable = 30 pF +90 pF = 120 pF. Zcap = 1/ jCtotal = -j /(2 1 MHz 120
pF) =-j 1326.
(c) If the signal generator in part (b) has a source impedance of 600 , then what is the ratio of
the voltage measured by the CRO to the nominal voltage produced by the signal generator? (The
nominal voltage may be taken as being the same as the signal generators Thevenin equivalent
voltage.)
Ignoring capacitive effects, the ratio is 1M / 1.0006 M = 0.9994.
page 7 of 12 OVER/
(d) What is the purpose of the variable capacitor in a times-ten probe?
To adjust the probes high-frequency compensation.
(e) What is flashover?
The arcing of electric current over the body of a probe when attempting to measure a very
high voltage.

Question 10.
(a) In a spectrum analyser, what is the function of the voltage controlled oscillator (VCO)?
The VCO produces the swept frequency that provides the frequency range represented by
the horizontal scale of the spectrum analysers display.
(b) Wave analysers are used to measure what three (3) types of quantity?

(1) Amplitudes of frequency components.

(2) Signal amplitudes in the presence of noise and interference.

(3) Energy of a signal in a specified frequency band.

(c) Define total harmonic distortion.

Eh VRMS (harmonics)
THD = =
Ef VRMS (fundamental)

where Eh = the energy (RMS voltage) due to the harmonics produced in the distortion by
the nonlinearity of the amplifier, and E f = the energy (RMS voltage) of the fundamental.
(d) Describe how a distortion analyser works.
A sinusoid from an oscillator is fed into the Amplifier Under Test (A.U.T.) and the ampli-
fiers output voltage is read on an RMS voltmeter. This measures the total output energy
E. The output of the amplifier is switched to a notch filter centred on the oscillators fre-
quency, and the filters output is read on the RMS voltmeter. This measures the energy
of the harmonics Eh . The ratio of these two voltages gives the THD of the amplifier at the
oscillators frequency and amplitude.
(e) Name five (5) instruments contained in an audio analyser.

(1) RMS voltmeter.

(2) DC voltmeter.

(3) Low-distortion audo oscillator.

(4) Signal-to-noise (S/N) ratio meter.

(5) Frequency counter.

page 8 of 12 OVER/
Question 11.
(a) Define frequency response.
Frequency response is the ratio of a circuits output voltage to its sinusoidal input voltage
plotted against the frequency of the input.
(b) What is intermodulation distortion?
Intermodulation distortion is the amplitude modulation of one signal by another due to
nonlinearity of the system in which both signal simultaneously occur. This distortion is
measured by IMD = ba a where a is the distance between the centres of the upper and
lower envelope waveforms, and b is the peak-to-peak voltage of the full envelope.
(c) Define slew rate.
An amplifiers slew rate Sr is the maximum rate of change of the output voltage, ( dEdtout )max ,
during the transition from the low voltage rail to the high voltage rail.
(d) Define full-power bandwidth.
Full-power bandwidth is (high) frequency at which an open-loop amplifiers gain falls to
unity (1) while it is delivering its full rated output power.
(e) Define input offset voltage.
Input offset voltage is the voltage between the inverting and noninverting inputs of a dif-
ferential amplifier when its output voltage is zero.

Question 12.
(a) State the formula for the total impedance of the RLC series resonant tank circuit.
1
ZRLC = R + jL +
jC

(b) Give the formula from the notes for the bandwidth of a tank circuit.
1 R
B=
2 L

(c) In a tuned transformer, there is a magnetic coupling between the primary and secondary
windings.

(1) If this coupling is large, what effect does it have on the Q of the circuit?
The quality factor Q decreases.

(2) What happens to the frequency response curve if the coupling is very large?
The curve forms a trough between two broad peaks.

(d) What is a notch filter?

A filter that passes all frequencies except for a very narrow band about one frequency (the
notch frequency).
page 9 of 12 OVER/
Question 13.
(a) What company devised the ATE bus architecture on which the GPIB was based? What was
their bus called?
Hewlett-Packard. The Hewlett-Packard Interface Bus (HPIB).
(b) Name the two GPIB data rates, and state how many kilobytes per second (kB/s) each oper-
ates at.
Unrestricted rate (up to 250 kB/s) and Restricted rate (over 250 kB/s).
(c) Of what does a minimal GPIB system consist?
One talker, one listener, no controller.
(d) What are the three (3) functions of a controller in a GPIB system?

(1) Designate the talkers and the listeners.

(2) Command actions within devices.
(3) Is usually a computer (or possibly a programmable calculator).

(e) Name four classes of commands or address codes used by the GPIB.
Four (4) of:

(1) Addressed Commands.

(2) Universal Commands.
(3) Listen Addresses.
(4) Talk Addresses.
(5) Secondary Addresses or Commands.

Question 14.
(a) Name the three (3) classes of noise source given in the notes.

(1) Transmitted.
(2) Intrinsic.
(3) Interference.

(b) How does pink noise differ from white noise?

White noise is wideband; pink noise is more intense at low frequencies.
(c) What causes shot noise?
Small local fluctuations in DC current.
(d) Electrically coupled interference voltage increases with what four (4) quantities?
Four (4) of:
page 10 of 12 OVER/
 (1) Proximity.

(2) Capacitance between source and pickup circuit.

(3) Frequency of source.

(4) Amplitude of source.

(5) Input impedance of pickup circuit.

(e) List four (4) everyday sources of electromagnetic interference.

Four (4) of:

(1) Radio and TV stations.

(2) Radio and TV receivers.

(3) Remote garage door openers.

(4) Remote controls for TV receivers.

(5) Unterminated, active power sockets.

(6) Fluorescent lamps.

(7) Car ignitions.

(8) The sun, Jupiter, and the galaxy.

Question 15.
(a) State the international standard color for each of the three (3) lines in an AC power cord.

(1) Active = brown.

(2) Neutral = blue.

(3) Ground = green with yellow (diagonal) stripes.

(b) Describe a situation that may give rise to a leakage current inside an instrument.
If a break occurs in the insulation of the neutral wire inside an instrument with a grounded
chassis, then the neutral and the chassis form a leakage capacitance through which AC
may flow from the neutral to the chassis, and hence as a leakage current through the
ground wire.
(c) List four (4) possible causes of a ground loop.

(1) Potential differences between two connections to different points of a ground plane.

(2) Inductive (magnetic) coupling.

page 11 of 12 OVER/
 (3) Capacitive (electric) coupling between an electrical system and ground.

(4) Common-mode noise.

(d) Give a possible solution to the problem of crosstalk.

Completely shield the circuits for the mutually interfering channels, and use separate
power supplies. (However, this is costly, and not always feasible.)

Question 16.
(a) Name and describe two (2) properties of real amplifiers that give rise to distortion.
Two (2) of:

(1) Finite bandwidth (gain depends on frequency).

(2) Nonlinearity (gain depends on amplitude).

(3) Frequency-dependent phase.

(4) Amplitude-dependent phase.

(5) Finite power supply.

(b) What is hiss noise?

Radio frequency (RF) noise that is evenly distributed over a wide frequency band. (In
other words, its RF white noise.)
(c) If an electric shield has an opening in the shape of a cylinder with a diameter of 2 cm and a
sleeve of length 8 cm, then what is the attenuation
q of an interference waveform of wavelength
60 cm that enters that opening? [Hint: 54.5( Lc ) 1 ( c )2 , where c = r. ]
Given data: r = d/2 = 1 cm = 0.01 m; L = 8 cm = 0.08 m; = 60 cm = 0.60 m.
Derived data: c = 0.01 m.
Attenuation is: r
0.08 0.01 2
54.5( ) 1( ) = 138.6
0.01 0.60

(d) What might you use to eliminate the interference caused by the 50 Hz hum from the mains
power?
A notch filter centred at 50 Hz.

page 12 of 12 END.