Question 1 of 5: Series voltage regulator — output voltage, line ripple and efficiency
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Paper format. National Exams, December 2015 — 07-Elec-B5
Advanced Electronics. Three hours, CLOSED BOOK; any non-communicating
calculator permitted. Answer all FIVE (5) questions; all questions
are worth 20 marks each. In schematics ground and chassis are common; op-amps are
ideal and supplies are ±15 V unless stated otherwise. Candidates are urged to
state any interpretation assumptions inside their answer — this solution does
so in the shaded Check callouts.
Reference texts.
A. S. Sedra & K. C. Smith, Microelectronic Circuits, 8th ed. —
Ch. 4 (diodes / shunt and series regulators), Ch. 7 (transistor amplifiers),
Ch. 10 (frequency response, open-circuit time constants), Ch. 11 (feedback),
Ch. 12 (output stages), Ch. 17 (tuned amplifiers).
B. Razavi, Fundamentals of Microelectronics, 2nd ed. — Ch. 8
(bias and small-signal models), Ch. 11 (frequency response, Miller's theorem),
Ch. 12 (feedback topologies).
P. R. Gray, P. J. Hurst, S. H. Lewis & R. G. Meyer, Analysis and Design
of Analog Integrated Circuits, 5th ed. — Ch. 7 (frequency response),
Ch. 8 (feedback).
C. K. Alexander & M. N. O. Sadiku, Fundamentals of Electric Circuits,
7th ed. — Ch. 14 (resonance, bandwidth and quality factor).
Check — how the figures were recovered. Every circuit on this paper carries data that exists only in the drawing (which
branch a component sits in, which transistor terminal the source drives). Where a drawing
is ambiguous or under-specified, the reading used is stated explicitly in a
Check callout inside that question rather than being buried in the
arithmetic.
Question 1: Series voltage regulator — output voltage, line ripple and efficiency (20 marks)
Find. The nominal regulated output VOUT, the peak-to-peak
ripple that a 1 V p-p disturbance on VDD produces at the output, and the
power efficiency η = PL/Psupply.
Q1 — series (linear) regulator. R1 and the zener D1 set the reference VZ; the ideal op amp A1 compares it with the output and drives the Darlington pass pair Q1–Q2.
Approach. Fix the reference by finding the zener current, use
the ideal op amp with 100 % negative feedback to transfer that reference to the
output, treat the ripple as a small signal divided between R1 and the
zener's incremental resistance, then compare load power with the total power drawn
from VDD.
Part (a) — Check that the zener sits at its specified operating point.
The reference branch is R1 from VDD into the zener; the op-amp
input draws no current, so all of that current is zener current:
$$I_Z=\frac{V_{DD}-V_Z}{R_1}=\frac{10-6.7}{3.3\ \text{k}\Omega}=1.00\ \text{mA}$$
which is exactly the current at which the data sheet quotes VZ = 6.7 V. The
reference is therefore 6.7 V with no correction needed.
Close the feedback loop to obtain the output. The inverting input is
tied directly to the output node, so the loop is a unity-gain follower wrapped around
the Darlington pair. An ideal op amp in negative feedback forces
$v_- = v_+$, i.e. $V_{OUT}=V_Z$:
$$\boxed{V_{OUT}=V_Z=6.70\ \text{V}}$$
The two base-emitter drops of Q1 and Q2 (1.4 V in total) are absorbed
inside the loop — the op amp simply raises its output to
$V_{OUT}+2V_{BE}\approx 8.1\ \text{V}$ to supply them, which is well within its
±15 V rails.
Part (b) — Convert the supply ripple into a reference ripple.
For small signals the zener is replaced by its incremental resistance
$r_z = R_Z = 10\ \text{k}\Omega$, which forms a divider with R1 across the
ripple source:
$$\frac{v_z}{v_{dd}}=\frac{r_z}{R_1+r_z}=\frac{10}{3.3+10}=0.752$$
The op amp and pass pair reproduce that reference ripple one-for-one at the output, so
$$\boxed{v_{out(\text{p-p})}=0.752\times 1\ \text{V}=0.75\ \text{V p-p}}$$
State the line regulation that this implies. Expressed as a ratio the
line regulation is $0.752\ \text{V/V}$, i.e. $20\log_{10}(0.752)=-2.5\ \text{dB}$ of
ripple rejection. This is very poor for a regulator, and it is entirely the fault of the
10 kΩ incremental zener resistance quoted for D1: a real 6.7 V zener biased
at 1 mA would show tens of ohms, which would give a rejection near 80 dB. The circuit
topology is sound; the reference device is the weak link, and the calculation is worth
doing precisely because it exposes that.
Part (c) — Compute the delivered load power. With the output
regulated at 6.70 V across 4 Ω,
$$I_L=\frac{V_{OUT}}{R_L}=\frac{6.70}{4}=1.675\ \text{A},\qquad
P_L=V_{OUT}I_L=6.70\times1.675=11.22\ \text{W}$$
Account for every ampere drawn from the supply. The load current is the
emitter current of Q2, so
$I_{B2}=I_L/(\beta+1)=1.675/101=16.58\ \text{mA}$, and that base current is in turn the
emitter current of Q1, giving
$I_{B1}=I_{B2}/(\beta+1)=0.164\ \text{mA}$. Only the two collector currents and the
reference branch are drawn from VDD (the op amp runs from its own rails), and
by KCL these total
$$I_{DD}=I_Z+I_{C1}+I_{C2}=I_Z+I_L-I_{B1}=1.00\ \text{mA}+1.675\ \text{A}-0.164\ \text{mA}=1.676\ \text{A}$$
$$P_{supply}=V_{DD}I_{DD}=10\times1.6758=16.76\ \text{W}$$
Form the efficiency.
$$\eta=\frac{P_L}{P_{supply}}=\frac{11.22}{16.76}=0.6697
\quad\Rightarrow\quad \boxed{\eta=67.0\ \%}$$
The dissipation is dominated by the 3.3 V that the pass pair drops at 1.675 A
(5.53 W), which is the unavoidable price of a linear series regulator: the efficiency
ceiling is $V_{OUT}/V_{DD}=67.0\ \%$ and this design essentially reaches it.
Part
Quantity
Result
(a)
Nominal output voltage VOUT
6.70 V
(b)
Output ripple for 1 V p-p on VDD
0.75 V p-p (line regulation 0.752 V/V, −2.5 dB)
(c)
Load current / load power
1.675 A / 11.22 W
(c)
Supply current / supply power
1.676 A / 16.76 W
(c)
Power efficiency η
67.0 %
Check — assumptions carried in Question 1.
(i) β = 100 is quoted only for Q1; Q2 is taken to have the same
β, which is the usual reading for a monolithic Darlington and in any case changes
IDD by less than 0.01 %. (ii) The ideal op amp is assumed to draw no supply
current, so Psupply counts only the VDD rail. (iii) The Early voltage
VA = 100 V plays no part here: the feedback loop, not ro, sets the
output. (iv) RZ = 10 kΩ is taken literally as the zener's incremental
resistance, as printed.