UNIVERSITY OF LONDON

BEng and MEng EXAMINATIONS 1999

PART II

for Internal Students of the Imperial College of Science Technology and Medicine
This examination is also taken for the relevant examination for the Associateship

ELECTRONICS, INSTRUMENTATION and CONTROL

for Mechanical Engineers
Thursday, 10th June: 14.00 to 17.00

This paper contains THIRTEEN questions, EIGHT in Part A and FIVE in Part B.
Attempt ALL questions in Part A and THREE questions in Part B.
Part A accounts for one-third of the total marks for the paper and all questions in Part B carry
equal marks.
A handbook of Data and Formulae is provided.

This is a CLOSED BOOK examination

PART A
Enter your answers in the spaces provided ON THIS PAPER, and submit the paper at the end of the
examination.
SHOW ALL WORKING, either on this paper or in your answer book.

Candidate’s Number: …………………………

+12 V

1k

vL
100

0V

Figure A1

A1. The silicon transistor in the emitter follower stage of Fig. A1 has a current gain hfe of 100.

(i) Determine the input voltage and current if vL = 5V.

……………… mA, …………… V

(ii) Estimate the power dissipation in the transistor under these conditions: …………… W

© University of London Turn over

Page 1 of 10
(iii) At the same value of input voltage, a second 100 Ω resistor is connected in parallel with
the first. What is the new value of vL?
…………… V

Page 2 of 10
 100 k

100 k
vo
+ P1 P2 100 k
10 V
100 k

0V

Figure A2

A2. Potentiometers P1 and P2 in the circuit of Fig. A2 act as position transducers. P1 is at mid-
position, P2 at zero (i.e. minimum output voltage). What will be the output voltage vo if:

(i) Both P1 and P2 act as ideal analogue position transducers?

…………… V

(ii) Each potentiometer has a total (i.e. end-to-end) resistance of 20 kΩ?

…………… V

log gain, phase

dB shift
90°
G=

1 rad/s -90°

Figure A3

A3. When the feedback loop within the system of Fig. A3 is disconnected, the system has the
Bode plot shown on the right: its gain increases at 6 dB/octave, and its phase shift is 90°.
d
(i) Show the open-loop transfer operator G, as a function of D = , on the block diagram.
dt

(ii) Determine the transfer operator of the closed-loop system: G′ = ………………………

(iii) Clearly show a corner plot, on the Bode axes, of the closed-loop gain response. You
need not show the phase response.

Turn over
Page 3 of 10
(iv) What is the exact gain and phase shift of the closed-loop system at a frequency of 2 rad
s−1?
…………… dB, …………… degrees

Page 4 of 10
 1µ
a
c
b
F
100 k
q
d

2.2 k

0V

Figure A4

A4. The piezoelectric force transducer shown in Fig. A4 has a sensitivity of 5 pC N−1.

(i) A force of F = 1 kN is applied. Determine voltages at points a, b, c and d.

a: ……………… V.
b: ……………… V.
c: ……………… V.
d: ……………… V.

(ii) Determine the magnitude of the current in the detector stage feedback loop when the
force increases at 1 kN s−1.
……………… A

(iii) What is the maximum rate of change of load which could be detected by a type 741 op-
amp?

……………… kN s−1

(iv) What is the voltage gain of the second op-amp stage?

……………… dB.

A5. A load cell is used to measure mechanical vibration. The load cell signal is amplified with a
gain of 60 dB to give an output of 2V peak to peak. This output is input to an analogue-digital
convertor (ADC) whose range is −5V to +5V.

(i) What is the peak to peak amplitude of the load cell signal?
……………… V pp

(ii) If the amplified signal contains noise at a level of 20 mV pp, what is the signal-to-noise
ratio of the load cell signal?

……………… dB

Turn over
Page 5 of 10
(iii) What is the lowest ADC resolution, in bits, at which noise will just begin to appear in
the ADC output?
……………… bits

Page 6 of 10
 3 iy , mA
x
2

1
v xy, V
0
1 2 3 4
-1

-2
y
-3

Figure A6

A6 Device T is a 4 V, 4 kΩ Thévenin source. Device N is a 2 mA, 1 kΩ Norton source.

(i) T is connected between terminals x and y in Fig. A6, tested, and removed. N is then
tested in the same way. In each case iy >0 when x and y are short-circuited. On the
‘volt-amp’ axes of Fig. A6, sketch and label both measured characteristics.

(ii) T and N are now connected in parallel between x and y. In the box, sketch a circuit
diagram of this combined source, and on the axes sketch its output characteristic.

(iii) While iy = 0 , an ‘internal’ current still flows between T and N. Treating N as a load
connected to source T, indicate its operating point on the volt-amp axes.

− τ1D F

1
v −
τ D
2

Figure A7

A7 Fig. A7 shows the block diagram of an electro-mechanical system, in terms of the transfer
operators of its components. The input is a voltage v, the output is a force F.
(i) Determine the overall transfer operator of the system of Fig. A7:

…………………………………………………………

(ii) Write down the transfer operator of an electronic system which, if connected in cascade
to the input, would provide exact frequency response compensation:

…………………………………………………………
Turn over
Page 7 of 10
(iii) What phase shift does the compensating system introduce at:

Low frequencies? ………………………… High frequencies? …………………………

Page 8 of 10
A8 A DC motor behaves as a linear first-order system with a time constant of 1 second. When an
armature voltage of 1V is applied, the steady-state motor speed is 100 rpm.
(i) The motor is at rest and the armature voltage is instantaneously increased from 0 V to 10
V. What is the speed 2 seconds later?
……………… rpm

(ii) While the motor is running at its steady speed, the voltage is instantaneously changed
from 10V to −10 V. What is the speed after a further 2 seconds?
……………… rpm

The motor is built into a closed-loop speed control system using simple proportional feedback.
When a voltage of 1V is applied to the system input, the steady-state motor speed is now only
50 rpm.

(iii) The motor is at rest and the armature voltage is instantaneously increased from 0 V to 10
V. What is the speed 2 seconds later?
……………… rpm

Turn over
Page 9 of 10
PART B
Enter your answers IN AN ANSWER BOOK.

B1. Fig. B1 shows an arrangement by which a 100 kΩ potentiometer controls the power delivered
to a resistive load RL using pulse width modulation. An external signal generator provides a
triangular wave at 1 kHz with 10V peak-peak amplitude and zero offset. A 741 type op-amp
is used as a comparator. The base resistance Rb is low enough and the transistor gain high
enough to ensure saturation in the ‘on’ state. A diode protects the transistor base from
negative base-emitter voltages which might damage it.
+15 V

R R v
2 L L

100 k
R
b

R
1

0V
−15 V

Figure B1

(i) Sketch, showing significant values, op-amp output voltage vs. time and vL vs. time for (a) v+ =
−3 V and (b) v+ = +3 V. Hence explain how the potentiometer controls the average voltage
v L applied to RL.

(ii) Derive or write down an expression for the non-inverting input voltage v+ in terms of R1, R2
and the position 0<λ <1 of the potentiometer (λ = 0 meaning that the non-inverting input is
connected to R1 and λ = 1 that it is connected to R2). Assume the supply voltages to be exact.

(iii) Calculate values of R1 and R2 so that v L just begins to increase from zero at λ = 0.1 and
reaches its maximum value at λ = 0.9.

(iv) Sketch an alternative arrangement to that of Fig. B1, using all of the same components except
R1, R2 and the signal generator, which could maintain vL constant and directly equal to a
potentiometer voltage of a little over 0V to a little under 15V. What is the principal
disadvantage of this option compared to pulse-width modulation?

Page 10 of 10
B2. An RS Reflective Opto Switch, type 307-913, consists of an ‘input’ infra-red emitting diode
and an ‘output’ phototransistor housed in the same package. It acts as a transducer for the
proportion of infra-red radiation reflected by a surface, as shown schematically in Fig. QB2.
The infra-red emitting diode is electrically similar to an LED, but has a slightly higher cut-in
voltage of 1.7 V. The phototransistor is electrically similar to an npn transistor, but has no
electrical base connection.
+15 V

distance R1
x

R2

10 kž
reflective 10 kž
surface
10 kž vo
vx

307-913

0V

-5 V

Figure B2

If diode current ia and surface properties remain constant, the 307-913 can be used to detect
the distance x of the transducer from the surface. For a standard surface, a graph on the 307-
913 data sheet shows output current ic vs. distance x, at ia = 40 mA (the maximum permitted
diode current) and phototransistor emitter-collector voltage vce = 5.0 V. The table below
shows values extracted from this graph:

Position, x (mm) 4.0 5.0 6.0 7.0 8.0

Collector current, ic (µA) 1250 800 550 400 290

In the proposed distance detector circuit of Fig. B2, op-amp supply voltages are ±15 V.

(i) Choose an E12 value for R1, explaining your choice.

(ii) Choose an E12 value for R2 so that vx changes by as near as possible to 4V as the distance x
increases from 4 to 8 mm. Explain your method, and explain how the first stage maintains vce
= 5.0 V whatever collector current flows.

(iii) For the chosen value of R2, tabulate vx and vo for x = 4.0, 5.0, 6.0, 7.0 and 8.0 mm.

(iv) In principle, R2 could be replaced by a network of components whose effect would be to make
vo = x − 4

Turn over
Page 11 of 10
where vo is in volts and x in millimetres. Write down, in the form of a table, the volt-amp
characteristic which this network would be required to have.

Page 12 of 10
B3. Fig. B3 shows a circuit diagram for analogue stages within an instrumentation system. The
input is connected directly to a tachogenerator of high output impedance. The tachogenerator
speed can only vary slowly, but its output signal vi contains high-frequency noise. The system
output vo is fed to an analogue-digital convertor.

220 k 1 µF

10 k
vi 22 k
10 k vo

0V 220 k

Figure B3

(i) Draw a block diagram of the circuit in Fig. B3. Indicate individual op-amp stages by name
and transfer operator, and state the value of the time constant.

(ii) Show, by considering arrangement as a negative-feedback system, that it has an overall

transfer operator
1
,
(1 + τ ′D)
and state the value of the modified time constant τ ′ .

(iii) Explain what purpose the circuit of Fig. B3 serves in the system described.

(iv) Draw the complete circuit diagram of a cascade system having the same transfer operator and
the same input resistance as the circuit of Fig. B3. Identify each stage by name and transfer
operator and show typical component values.

(v) Explain briefly why the system of Fig. B3, in which a feedback loop contains two op-amps, is
more likely to suffer from instability at high frequencies than the equivalent cascade system.

Turn over
Page 13 of 10
B4. A servo actuated by a permanent-magnet DC motor unit is used to control angular position θ.
When the motor unit is first tested separately in open-loop configuration, it appears to behave
as a simple first-order low pass device: on increasing the armature voltage va instantaneously
from zero to a constant 10 V, the speed increases towards 25 rad s−1 with a time constant of
0.8 s. When a torque is applied to reduce the speed to zero at the same va, the armature current
increases to 12.5 A.

(i) Determine the EMF constant Ke, armature resistance Ra and effective polar moment of inertia J
of the motor gearbox unit. Friction may be neglected.

In the position control servo, the motor is driven by an error amplifier with a variable gain A . θ is
measured by a rotary potentiometer with a sensitivity KP = 10V rad−1, which provides negative
feedback to a difference amplifier stage at the error amplifier input. The demand signal input is vi.

(ii) Draw a block diagram representing the servo as a system having vi as input and θ as output.
Show transfer operators for each stage, and indicate the position at which va is measured. By
applying a general expression for the transfer operator of a negative feedback loop, derive
expressions for the natural frequency ωn and damping factor ζ of the system response.

(iii) Calculate the demand voltage vi needed to command a final steady-state rotation of +45°.

(iv) Calculate the error amp gain required to achieve optimum response to a step function. The
error gain is set to this value whilst the initial position θ remains stationary at zero. Calculate
the initial armature voltage va and current ia immediately after vi is changed instantaneously to
demand a position of +45°.

(v) The position is now restored to zero and the error amp gain increased to make ζ = 0.1.
Estimate, by inspection of the second-order response curves provided in the handbook of Data
and Formulae, the armature current as the motor is instantaneously stationary at the limit of
the first overshoot.

Page 14 of 10
B5. A closed loop negative-feedback control system whose gain and phase margins are low can be
improved by reducing the steady-state gain of the feedback loop. A better method is to
introduce a phase compensation stage having the transfer operator
(1+ τ 1D)
,
(1 + τ 2D)
where τ1 and τ2 are time constants, into the loop. This corrects the loop phase shift over a
relatively narrow range of frequencies, without changing the steady-state loop gain.

A closed-loop negative-feedback system has the open-loop frequency response tabulated

below, giving it a gain margin of 11.4 dB and a phase margin of 28°. This is judged to be too
low.

Frequency (rad s−1) 1.0 2.0 5.0 10.0

Loop gain (dB) 6.2 -2.3 -16.2 -27.8
Loop phase shift (°) -134 -159 -191 -225

(i) Using your knowledge of a simple second-order position control system, briefly explain the
disadvantage of stabilising the system by reducing the steady-state loop gain.

The present system is now modified by introducing a phase compensation stage, having
τ 1 = 1 s and τ 2 = 0.5 s , into the feedback loop.

(ii) Explain how the Bode plot of a transfer operator (1 + τ 1 D) can be deduced from that of a
1
transfer operator .
(1 + τ 1D)
(iii) Using the frequency response data provided in the Data and Formulae handbook, tabulate the
gain and phase response of the phase compensation stage at the frequencies shown.

(iv) Hence calculate the frequency response of the compensated loop at the frequencies shown and
estimate the new gain and phase margins. Linear interpolation may be used; a full Bode plot
is not required.

Page 15 of 10