Table of Contents
Advertisement
d.
FROM 50, 000 MEGOHMS TO 500, 000 MEGOHMS.
To determine the value of an unknown resistance between
50, 000 megohms and 500, 000 megohms, proceed as
follows:
(1) Perform preliminary procedure as stated
in
para-
graph 2-3.
(2) Set RANGE switch to 500 and NULL switch to 0. 1.
(3) Connect unknown resistance between input ter-
minals.
(4) Adjust voltage readout dials for a convenient
meter deflection.
(5) The resistance in megohms may be calculated by
substituting the meter reading
in
volts {Em, 0 to 0. 1 volt
on bottom scale) and the voltage readout dial setting {E)
into the following equation:
Rx
=
10 ~ megohms
Em
2-9.
NOTES ON MEASURING VOLTAGES
a. ADJUSTMENT OF 500 VOLT SUPPLY.
The 500
volt
DC
reference supply may be adjusted {paragraph
2-3, step c) at any time deemed necessary without heed
to the position of the switches and without removing any
input or output connections. However, until the instru-
ment has warmed up to an equilibrium temperature
{about 1/2 hour),
it
should be adjusted prior to each
specific measurement for best accuracy. When making
prolonged measurements, allow one hour warmup time
to insure that 500 volt reference supply does not shift
during the final warmup phase.
b. GROUND LOOP PRECAUTIONS. Ground loop cur-
rents should be avoided to assure accuracy when making
measurements. Potential differences are often found at
different points on power system grounds. When this is
the case, current may flow from the power system ground
through the 825A and the equipment under measurement
and back to the power system ground. To avoid this when
system being measured is grounded, do not connect 825A
input binding posts to chassis ground post.
c.
RECOMMENDED NULL RANGES.
Certain null
ranges are recommended for use with each setting of the
RANGE switch for the following reasons.
With the
RANGE switch at 500 volts, the last voltage readout dial
(E) changes the reference voltage in steps of 0. 01 volt.
Therefore, the unknown voltage would have to be an exact
multiple of 0.
01
volt if a null is to be obtained on the 0. 1,
0. 01 or 0. 001 volt NULL range. Furthermore, it is un-
likely that an unknown voltage of a few hundred volts will
be stable within 10 millivolts.
Finally, the regulation
of the reference supply is approximately 0. 0025% or
2-6
±0. 0125 volt for a 10% change in line
v~ltage.
Although
this is more than adequate when the instrument is used in
the recommended way, a badly fluctuating line voltage
may cause the 825A to meter the regulation of its own
500 volt reference supply.
For example, when meas-
uring 500 volts a line change of 10% may cause the 500
volt reference supply to change as much as 12.5 mv. Al-
though this is small, the 825A will indicate full scale for
a change as little as 100 mv,
10
mv, or 1 mv if attempt-
ing to use the 0. 1, 0. 01, or 0. 001 null ranges.
d. EFFECT OF AC COMPONENTS.
An
AC component
of several times the unknown
DC
may be present on the
unknown and the 825A will always indicate well within the
specifications for frequencies over a few hundred cycles.
An
AC component may have an adverse effect if it is of a
low frequency or if it has a frequency that is a multiple
or submultiple of the chopper frequency.
A
double sec-
tion low pass filter (R201, C201, R202, and C202) is
used at the input of the null detector to reduce any AC
present on the
OC
being measured.
At
lower frequencies,
this low pass filter is less effective and the magnitude of
the AC component is more significant.
If
this frequency
is below 100 cycles, the accuracy may no longer be with
specifications. For example, a 60 cycle AC voltage that
is
1~
of the input voltage will cause an error of approxi-
mately 0. 01% which is well within the specifications. At
low fre'-luencies other than 60 cycles, a larger AC com-
ponent can be tolerated. Also, since the input attenua-
tion is less for the more sensitive null settings, the accu-
racy may
be
effected only on the more sensitive null set-
tings. When the frequency is very close to a multiple or
submultiple of the chopper frequency (approximately 94
cycles), the meter needle will oscillate at the difference
frequency.
If
AC components that affect the accuracy are
ever encountered, additional filtering will be required.
For an AC of a single frequency, a twin-T filter is effec-
tive and
has
the advantage of low total series resistance.
For an AC variable frequency, an ordinary low pass filter
may be used. In either case, high quality capacitors of
high leakage resistance should be used.
e. MEASUREMENT OF NEGATfVE VOLTAGES.
Be-
cause of a polarity switch, voltages which are negative
with respect to ground as well as the more commonly en-
countered positive voltages may be measured with equal
facility.
If
the upper input post is connected to the metal
case, either at the 825A or at the source under measure-
ment, the accuracy of the voltmeter may be reduced.
However, with the polarity switch, the upper input post
never
has
to
be
connected to ground.
If
the unknown vol-
tage is grounded, always connect the grounded side to the
lower input post (middle post) and use the polarity switch
to obtain the proper result.
Advertisement
Table of Contents
