TECH LETTER $ 4

MEASUREMENT OF OUTPUT IMPEDANCE OF A CONSTANT VOLTAGE POWER SUPPLY

HARRISON LABORATORIES
DIVISION OF HEWLETT-PACKARD COMPANY
c 100 Locust Avenue
Berkeley Heights, New Jersey 07922

c
TECH LETTER # 4 December 2 2 , 1964

MEASUREMENT 0F OUTPUT IMPEDANCE OF A CONSTANT VOLTAGE POWER SUPPLY -I

INTRODUCTION

Any load d e v i c e connected to a power supply seldom is of s u c h a nature as t o

require a constant flow of power from the supply. Since the load d e v i c e usually con-
sists of active as well as passive d e v i c e s , the current drawn from t h e supply c o n s i s t s
of a n AC component superimposed o n t h e DC output current. (See Figure 1.) This AC
component is usually not sinusoidal i n nature, but c a n , by Fourier expansion, b e
broken down into purely sinusoidal components. For purposes of describing how con-
s t a n t the output voltage of a power supply c a n remain in spite of load current varia-
tions, it therefore becomes useful to specify the output impedance of t h e power supply
over a wide band of frequencies.

THEORY

This output impedance is defined as

where EAC is t h e sinusoidal voltage a c r o s s the power supply output terminals a s a r e s u l t

of the sinusoidal current, IAC, flowing through t h e s e r i e s loop c o n s i s t i n g of power sup-
ply and the load device.

T o measure the output impedance of a power supply a t a n y frequency, therefore,

it is n e c e s s a r y to draw a sinusoidal current from the power supply and measure the AC
component of voltage which r e s u l t s a c r o s s t h e output terminals of the supply. Divid-
ing this AC voltage by the AC component of the load current y i e l d s t h e v a l u e of the
output impedance of the power supply a t the frequency of the sinewave load. A cir-
c u i t suggestive of t h e method t o be employed for measuring power supply AC output
impedance is shown i n Figure 2 . This circuit is not a practical o n e , but it does s e r v e
to illustrate the measurement principle involved, A signal current, IAC, is c a u s e d to
flow through the output terminals of the power supply and a current monitoring r e s i s t o r
(RI) in s e r i e s . The output impedance of the supply is then:
 However, there a r e still several shortcomings t=>the c i r c u i t shown i n Figure 2 .
!se include:

(1) The signal generator m a y be incapable of withstanding the DC voltage

oss the power supply output terminals. A blocking capacitor is therefore d e s i r -
a b l e between t h e power supply and the signal generator.

(2) If a capacitor is added, t h e power supply is unloaded at DC. This is

corrected by addition of RL a s i n Figure 3 . RL is much greater than Z o a n d there-
fore d o e s not affect t h e measurement of 20.

(3) Because of the low AC load impedance, t h e o s c i l l a t o r may be mismatched,

with resulting waveform distortion. The high frequency components of t h i s d i s t o r t e d
wave will r e s u l t i n a larger voltage a c r o s s the power supply due to t h e higher imped-
a n c e a t higher frequencies. A matching r e s i s t o r (RM) may b e n e c e s s a r y .

(4) A t low frequencies, t h e output impedance of t h e power supply is l o w

and t h e current which is required i n order to obtain a readable voltage a c r o s s ter-
minals AA' may be very large (e.9. a s s u m e a power supply output impedance of
1 milliohm. In order to get a 1 millivolt signal for presentation on a n oscilloscope,
it is n e c e s s a r y t o feed a 1 ampere current IAC through terminals AA'). Thus, a power
amplifier with low output impedance may b e n e c e s s a r y .

(5) Since the AC current which will b e required i n order to generate a read-
a b l e s i g n a l a c r o s s AA' is large, t h e power supply would b e called upon to accept
c (rather than deliver) high v a l u e s of current, i f not properly DC loaded. However, a
DC power supply is e s s e n t i a l l y a unilateral d e v i c e . I t can conduct current i n o n e
direction only. Any attempt to force current i n t h e r e v e r s e direction through t h e
power supply terminals during half of the AC cycle will simply r e s u l t in t h e output
capacitor of the power supply being charged to a v o l t a g e higher than t h e normal out-
put voltage.

Similarly, i f too much peak current is drawn from the supply, t h e current l i m i t
circuit might b e activated. Therefore, IDC and IAC must b e correctly a d j u s t e d .

PROCEDURE

A b a s i c equipment arrangement is shown i n Figure 3 . Below is the b a s i c pro-

cedure to b e followed.

(1) The v a l u e of the load r e s i s t o r (RL) is c h o s e n so t h a t the current drawn

through it a t the DC output voltage will b e equal to or greater than t h e peak v a l u e of
the s i g n a l current, IAC, and so t h a t t h e maximum instantaneous sum of IAC and t h e
DC current is less than the s e t t i n g of t h e current l i m i t or the rating of the supply. An
optimum v a l u e of load current is one half t h e rated current.

(2)

WWW.HPARCHIVE.COM
c --
 (2) S i n c e the measurement of output impedance c a n b e no more a c c u r a t e
than t h e tolerance of RI, it is important t o know the exact v a l u e of RI and to be
c e r t a i n that its impedance is constant over the band of frequencies measured. One
convenient way of obtaining a s u i t a b l e 1 ohm resistor for RI is to t a k e t e n 1 watt
Allen-Bradley 10 ohm r e s i s t o r s and put them all i n parallel, twisting their pigtails rl
together tightly. The resulting 1 ohm r e s i s t o r will be flat within 10% over a fre-
quency band up to (and possibly beyond) 1 megacycle.

(3) RM should be c h o s e n so t h a t in conjunction with the output imped-

a n c e of t h e power supply, RI, and the impedance of t h e coupling capacitor, C, the
total impedance into which t h e oscillator (or amplifier) looks is equal to the nominal
output impedance of the oscillator (or amplifier).

For low and medium frequency measurements, H-P Model 4301A includes a
s u i t a b l e oscillator and power amplifier in one convenient package.

(4) The blocking capacitor, C, should be a n electrolytic c h o s e n to have a

v a l u e large enough so its impedance will be small compared to t h e impedance of
the oscillator (or amplifier, i f included) at the l o w e s t frequency of measurement.

(5) I t is e s s e n t i a l t h a t the wires BB' be connected directly a c r o s s the

monitoring r e s i s t o r RI so t h a t the voltage measured a c r o s s BB' includes only the
voltage drop of t h e r e s i s t o r RI and not t h e voltage drop a c r o s s RI plus some amount
of l e a d s connected to it.

(6) A l s o , when measuring the voltage a c r o s s AA' , it is e s s e n t i a l t h a t the

monitoring d e v i c e be connected directly a c r o s s the output terminals of t h e supply. 9
Thus, the measurement of t h e voltage a c r o s s AA' is valid and d o e s not include the
effect of any IR drop i n t h e l e a d s carrying t h e signal current to the power supply
terminals.

(7) The monitoring device used should be a n o s c i l l o s c o p e with a minimum of

1 millivolt per centimeter sensitivity. When' using the o s c i l l o s c o p e , it is a d v i s a b l e
to che'Lk the w a v e s h a p e of the voltage a c r o s s AA' and BB' for the presence of undue
amounts of waveform distortion. Furthermore, a large v a l u e of 60 cycle component
a c r o s s AA' should b e considered a s indicative of ground problems i n the measurement
s e t u p and should be resolved before t h e measurement proceeds. A voltmeter is not
s u i t a b l e a s a monitoring device s i n c e it may give erroneous r e s u l t s due to normal
120 c y c l e ripple o n t h e output of the supply.

(8) The output voltage setting of the power supply does not materially
affect measurement of output impedance; it c a n be set a t a n y convenient v a l u e .

(9) The current l i m i t control of t h e power supply should be set to its maxi-
mum value.

(10) IAC should be held t o a peak-to-peak v a l u e of less than the current

rating of the supply or twice the DC current through RL, whichever is smaller.

(11) The equation, Z o = RI (x) is u s e d t o determine t h e output imped-

3
a n c e of t h e power supply.
 (12) I t is suggested t h a t t h e impedance measurements made should be plotted
on log-log paper. Normally, one would expect the output impedance t o be e s s e n t i -
a l l y r e s i s t i v e (flat with frequency) a t low and medium frequencies and inductive
(rising a t 6 DB per octave or 20 DB per/decade) at high frequencies. I t is also pos-
sible t h a t shorter inductive regions may exist in the output impedance characteristic
a t midfrequencies (see Harrison Laboratories' s p e c s h e e t s for typcial output imped-
a n c e c h a r a c t e r i s t i c s .)

WWW.HPARCHIVE.COM
--
, , _ I _ _
 -
- EDC EO

T
POWER SUPPLY LOAD DEVICE I
Figure 1. Power Supply Feeding Active Load Device

Superimposed on desired DC output voltage EDC is a n AC component, E A C ~

resulting from interaction of AC component of load current IAC with power
supply output impedance Zo.

k
POWER SUPPLY OSCILLATOR

Figure 2. Circuit Principle Involved in Measurement of

Power Supply AC Output Impedance
 rl

-- u

8
w
0

I
m
m a?'
8

u
w 4
W

M
4
d

WWW.HPARCHIVE.COM
- -_ _o__ --
 WWW. HPARCHIVE.COM
.-- - -_ --