Calibrating a HP 431C power meter

[LazyJack]

Hi,

I recently scored a nice HP 431C power meter with cable and 8478B thermistor. (And also a 11527A adaptor, which is not required for this setup.)

I cleaned it, and seem to work fine, I can zero it, reads about right, but i have no idea how accurate it is. There is about 0.5 dB difference between the meter on my 8640B, and the power meter, but the signal gen's level meter is far from precision equipment.

The 431C was last calibrated some 20 years ago. The manual calls for the 8402B calibrator to calibrate it. Which is usually available on eBay, but would be uneconomical to get it because of the shipping cost. And that still doesn't calibrate the thermistor.

Anyway, do you have any idea on how I could calibrate or at least verify the accuracy of this meter? The sensor reads from 100MHz, so I can't just do an RMS measurement. Although, I do have a HP 3406A sampling voltmeter, which in theory can calibrate with its own calibrator, but that has not been calibrated for a long time, and I had to do repairs in it anyway. I do have 300MHz scope, but that is not precision equipment, either.

I don't have an other power meter or a calibrated power RF source.

[LazyJack]

Well, the RMS measurement on the scope (Tek 2440) and the 431C agree, so this indicates that the sig gen's level meter is showing a bit low.
But the original question still stands, is there a DIY way to calibrate this old power meter?

[Tony_G]

That sensor can read from 10MHz to 18GHz - The issue you're going to run into is that the sensor has a calibration table that changes as frequency changes so you really need to characterize it over at least the frequency range you're interested in. I don't think there is any shortcut to that unless you have a source that you already know its values and uncertainties over that range.

You might want to see if your national standards lab will do a thermistor mount calibration (but then you're back at the 431C calibration as it is a DC substitution meter and it'll introduce errors).

Hopefully some others have better suggestions for you.

TonyG

[LazyJack]

Thanks. I do have the cal factor table for the sensor, but of course, that is probably not valid anymore. Since the whole instrument cost 35 bucks, I don't think taking it to a cal lab is worth it. Probably I'll look around if I can do some comparison with a good power meter, that should be sufficient for my needs.