Kelvin Varley will not balance

[kkayser]

I have a Fluke Kelvin Varley that will not balance on 3 positions of the "A" decade.  When I attempt to get a null, the voltage changes but gets nowhere near zero at any setting of the pot.  We pulled one pot, and it is ok.  All three positions demonstrate the same malfunction.  I figure it is a resistor or connection, likely the same on all 3 positions.  Since all three positions have failed in the same manner, it could be a weak spot in the design or manufacture.  The 3 positions that cannot be nulled are as accurate as any other position: 1ppm measured with a 732B and 3458A.  There are a few positions on the entire divider that are over 1ppm

Does anyone have any ideas where I might look for the problem?

[Bill158]

I have a Fluke Kelvin Varley that will not balance on 3 positions of the "A" decade.  When I attempt to get a null, the voltage changes but gets nowhere near zero at any setting of the pot.  We pulled one pot, and it is ok.  All three positions demonstrate the same malfunction.  I figure it is a resistor or connection, likely the same on all 3 positions.  Since all three positions have failed in the same manner, it could be a weak spot in the design or manufacture.  The 3 positions that cannot be nulled are as accurate as any other position: 1ppm measured with a 732B and 3458A.  There are a few positions on the entire divider that are over 1ppm

Does anyone have any ideas where I might look for the problem?

Try following this thread from the "volt-nuts" forum.

https://www.febo.com/pipermail/volt-nuts/2015-September/004477.html

The resistors in the oil bath are well matched for temperature coefficents, but not for absolute values.  This is why there are "factory selected" trim resistors in the adjustment circuits.  As explained in this thread the resistors in the oil bath can drift and some have experienced this drift.
Bill