Solartron 7075 Fault

[wn1fju]

My Guildline 9577 DMM (equivalent to the Solartron 7075) seems to have developed a fault where if I stick it in ohms mode and give it a high resistance (like 1M), it will slowly count up from about 700K to 1M.  Takes about 5 seconds.  It's almost like there is a capacitor somewhere that is charging.  Yet if I short the input leads, the resistance will instantly go back to zero ohms.  Don't get this kind of behavior in DCV or ACV and the unit passes all the self-tests.  I've checked the current sources and sinks in the ohms section and they seem perfectly stable while the thing is ramping up the ohms measurement.  So I'm guessing the ohms section is OK and the problem is in the input somewhere.  Any ideas?

[Kleinstein]

The DC input amplifier uses chopper stabilization. It could be that part somehow gets into saturation and than needs quite some time to recover. It might be worth looking at the output of the chopper part - normally this should be rather stable.

[wn1fju]

I looked at the chopper and didn't see anything troublesome. 

If I simply put a 1M resistor as the DUT and then monitor the voltage across the resistor with a separate voltmeter, I see the voltage ramp up slowly.  If I switch to a 2M resistor, the voltage immediately goes too high and then ramps even more slowly back down.  So whatever is happening is happening right at the front.  Yet, hooking up a DC P/S or an AC signal generator, the 7075 tracks the source voltages as fast as I can spin the knobs.  Seems odd since I thought a resistance measurement simply develops a voltage across the DUT and then the DMM acts like a DC voltmeter.

The ohms current source (and sink) have resistors in series and I've monitored the voltage across them.  They are perfectly stable while the machine is "counting" up or down, indicating that the current source is stable (1 uV on the highest resistance range).  It is as though the entire ohms current source is floating upwards or downwards with respect to signal ground. 

Can't put my finger on it.  It may have something to do with the guard or bootstrap lines, but I admit that I don't understand them at all.

[Vgkid]

Send "wiss" a pm. I know he has owned a few 7075's. Once i sell my 7065 I think I will be done with the brand.

[wn1fju]

The end of the saga has been fortunately reached!  I thought there was a problem with the ohms circuitry, but I went into DCV mode and put 30V through a 10M resistor into the 7075.  Since at that range, the input impedance is also 10M, I should get a reading of 15VDC.  Which I did, but only after about 20 seconds of ramping down from 30V to 15V.  So that led me to believe that the input impedance of the 7075 was varying, as though a capacitor was somehow in the circuit and charging or discharging. 

And so it was.  A 7406 controlling various relays had a stuck line which forced the ACV circuit to be permanently switched in, hanging right off the input lines regardless of whether I was in DCV or Ohms mode.  Some input/output capacitors in that circuit, aren't there? 

Anyway, a new 7406 fixed all ills.

[wiss]

Thanks Vgkid!

The current source for the ohms unit is floating! That is why there is both a source and a sink.

Many of the reed relays fail on these, I have not found one where it was the driving electronics, yet...