Calibrating an AC meter

[Norm Dresner]

Let's assume that I don't want to spend a fortune but I want to try to
calibrate my AC (True RMS) voltmeter -- at least at some level of accuracy
which I may never know exactly anyway but I'd really like to get within a
percent or so. This precludes purchasing calibration standards and implies
that I'm going to do it "on the cheap".

Let's assume I have a DC "calibration" voltage source which is manufacturer
spec'd to about 5-10 PPM so I can calibrate a DC meter to a reasonable
degree of accuracy.

Can I calibrate the AC meter (a Fluke 8050 "true rms" meter) using a square
wave of known duty cycle and controllable frequency (so it's comfortably
within the meter's band)? If so, I could create a square wave generator
with CMOS logic and by measuring the high and low levels (I assume I can
easily switch the circuit to a known DC state or to oscillate) and duty
cycle and calculating what the RMS voltage should be? If so, then it should
be "easy" to create an AC signal with known RMS voltage.

If I can't assume that making just measurements of DC voltage level and duty
cycle of a square wave are sufficient to create a "calibrated" AC source,
what else would I need to measure or do?

TIA
Norm

 

[Robert Baer]

Let's assume that I don't want to spend a fortune but I want to try to
calibrate my AC (True RMS) voltmeter -- at least at some level of accuracy
which I may never know exactly anyway but I'd really like to get within a
percent or so. This precludes purchasing calibration standards and implies
that I'm going to do it "on the cheap".

Let's assume I have a DC "calibration" voltage source which is manufacturer
spec'd to about 5-10 PPM so I can calibrate a DC meter to a reasonable
degree of accuracy.

Can I calibrate the AC meter (a Fluke 8050 "true rms" meter) using a square
wave of known duty cycle and controllable frequency (so it's comfortably
within the meter's band)? If so, I could create a square wave generator
with CMOS logic and by measuring the high and low levels (I assume I can
easily switch the circuit to a known DC state or to oscillate) and duty
cycle and calculating what the RMS voltage should be? If so, then it should
be "easy" to create an AC signal with known RMS voltage.

If I can't assume that making just measurements of DC voltage level and duty
cycle of a square wave are sufficient to create a "calibrated" AC source,
what else would I need to measure or do?

TIA
Norm

If and only if the meter actually *measures* true RMS, as compared
with the majority that measure an average and read RMS, then
theoretically the waveform makes no difference, and the squaer wave is
the easiest and simplist approach.
However, that being said, one can determine the RMS value of a
(reasonably pure) sine wave by measuring its peak accurately.
One simple minded way is to use a diode and capacitor to get the peak
voltage, and not load it at all when measuring; adjust a DC reference
onone end of your meter with the other end being onthe capacitor (ie:
the "slide-back" method).
Another way is to use an operational amplifier to make an ideal diode
/ peak detector.

 

[Jim Yanik]

Let's assume that I don't want to spend a fortune but I want to try to
calibrate my AC (True RMS) voltmeter -- at least at some level of
accuracy which I may never know exactly anyway but I'd really like to
get within a percent or so. This precludes purchasing calibration
standards and implies that I'm going to do it "on the cheap".

Let's assume I have a DC "calibration" voltage source which is
manufacturer spec'd to about 5-10 PPM so I can calibrate a DC meter to
a reasonable degree of accuracy.

Can I calibrate the AC meter (a Fluke 8050 "true rms" meter) using a
square wave of known duty cycle and controllable frequency (so it's
comfortably within the meter's band)? If so, I could create a square
wave generator with CMOS logic and by measuring the high and low
levels (I assume I can easily switch the circuit to a known DC state
or to oscillate) and duty cycle and calculating what the RMS voltage
should be? If so, then it should be "easy" to create an AC signal
with known RMS voltage.

If I can't assume that making just measurements of DC voltage level
and duty cycle of a square wave are sufficient to create a
"calibrated" AC source, what else would I need to measure or do?

TIA
Norm

You ought to call some cal labs and get a price quote on calibration.
Often it's not that bad if there are no repairs needed.