Calibration; Calibration Schedule; Kelvin-Varley Divider; Verification - Keithley 660A Instruction Manual

CALIBRATION
MODEL 660A DIFFERENTIAL VOLTMETER
I
Circuit
Fig. Refer to
Control Desig.
Ref.
Paranrauh
/(
0.5-volt Hum Oscillator 500-volt 50-volt 5-volt eter Balance Calibrate Calibrate Calibrate Calibrate Calibrate
R1048 R2007 R3025 R3032 R3035 R3038 R1043
20 20 20 20 20 18 18 5-7 5-5 6-4 6-4 6-6 6-4 6-4
TABLE 6. Model 660A Internal Controls. The Table lists all internal controls, the fig-
ure picturing the location and the paragraph describing the adjustment,
6-2.
CALIBRATION SCHEDULE.
Recalibrate the Model 660A yearly.
ca?tibration (paragraph 6-4).
This normally means performing the voltage range
The other verifications need not be done.
b.
Verify the Kelvin-Varley divider (paragraph 6-3) only if trouble is suspected in iti
or if components in the divider have been replaced.
Also make the voltage range calibra-
tion after verifying the divider accuracy.
c.
Check the reference voltage supply stability (paragraph 6-5) only if trouble is
suspected in the supply or if some of its components have been replaced.
6-3. KELVIN-VARLEY DIVIDER VERIFICATION.
a. There is no in-field calibration for the Kelvin-Varley divider; its accuracy can
only be verified. The divider accuracy depends upon matching resistors and switches. At
manufacture,
each resistor within the first two Reference Voltage switches, 53003 and
S3004, is matched to 0.005%. The resistors in the switches are checked as a set to an
accuracy of better than 0.005%.
Individual resistors cannot be replaced without rematching
the string
at
Keithley Instruments,
b.
Kelvin-Varley Divider Accuracy Verification Procedures.
1.
Use the Model 662 Differential Voltmeter or equipment with better limit of error
to match the Model 660A under test. Connect the wiper arm of Switch S3007 of the in-
strument under test to the HI terminal of the Model 662.
Connect both LO terminals.
Be careful of high voltages when working within the Model 660A.
up to 800
volts dc is present at various points.
2.
Set the dials to random settings on both instruments.
Settings should match to
?0.02%. This procedure, however, only indicates the Kelvin-Varley divider accuracy.
The errors of the two instruments may be additive,
causing a false verification.
The
most accurate way is to use standard procedures for checking a Kelvin-Varley divider
or to return the Model 660A to Keithley Instruments, Inc., for checking.
22 1065R