34401A ohms drift

[strawberry]

Cleaned front/rear switch but still -20ppm drift. Could that be some relay, current source fault?
VDC stable to 1ppm ,using LM399 ref and probemaster testleads.
What test leads for <1ppm drift

[Kleinstein]

The ohms drift can be quite different in different ranges. With very high resistance (e.g. 20 Meg)  20 ppm drift may not be that bad. The reference should have not much effect in the ohms measurement.

The requirements on the test leads are different depoending on the range: low ohms (e.g. < 1 K) would like low thermal EMF and care handling them, e.g. wait after handling the cables to get back to stable temperature.
With higher resistance like >100 K the isolation qualitiy of the leads can become important. Still I doubt the 34401 would be really good in this range.

The accurcy in ohms range can be low than in DC  volts range. I would not expect better than 1 ppm stability from the 34401 in the ohms range - more like change to a Keithley 2002 or Fluke 8858 for this.

[strawberry]

fixed transistor fault, 1..2ppm drift after warmup.

[floobydust]

Which transistor? Guessing the PNP's Q203-Q210 MMBT6520 or the JFETs?

There was an Engineering Change to the Ohms Current Source after S/N 3146A30500. I'm not sure why.
R207 from 2k37 to 5k62 1% and CR203 from 5.1V to 4.7V zener.

[strawberry]

I guess it is some clamp, placed J112, will order original or close to it
will calibrate with vishay 0.005 resistor
2010 model

[WattsThat]

And the resistor value is ?

[strawberry]

in theory should be 10.0000k but after repair is 10.003k

[strawberry]

resistor vishay Z201T 10.0000k

[Kleinstein]

Interesting that a bad JFET can effect the drift and noise in the ohms mode. The crude function was still working, so the JFET was not totally dead, more like either excessive gate leakage or the threshold dirfte too much to be boarderling (which I would consider not very likely).
One may want to check the protection part ( the high voltage PNPs) too.

The J112 may not be such a bad choice for the ciruit, especially if the controling OP still has enough headroom for control.

[strawberry]

PN4117 less leaky than J112

[Kleinstein]

It depends on the position of the FET. The one for the protection is used as a diode and here low leakage is essentially the only important point. So the PN4117 is good (alternative would be a low leakage diode).

The other FET is for the current regulation and this would need a fet for a higher current. Rethinking this, this one should be a p channel FET - so not the J112.

With the diode used FET in the protection broken, suspicion is high that some of the high voltage PNPs may also be bad. To damage the protection diode would need the PNPs to brake down and this can very well cause a damage there too.

[strawberry]

With PN4117A

[daveyk]

Was the fault Q202, MMBF5461SEL?  I can not find MMBF5461 anywhere right now. 

All the PNPs test okay.

Not sure what Q211 does.  I think it must be used for its gate leakage.