Thank you again for replies. We swapped the magnet board with another functioning 253, this did not fix our instability. We removed the source and all connections seem normal to tight (Huiming, do you recall which connection you had trouble with?), as do the connection inside the source at the feed throughs, as do the connections outside the source via the hex nut wire fittings. The power coming into the instrument is also very stable. I have reduced all vibration sensitivity to undetectable (by removing pumps and padding metal flexible tubing). The instability continues.
Yesterday we observed ~10 V variability in the accelerating voltage power supply while on another 253, the variability was more like 0.5 V (that is 10 V and 0.5 V variability on 10 kV). We also measured the current through to the first plate (extraction) which should be nominally zero, correct? We were able to detect a slight current at about 200 uA. Is that normal? The current through the high voltage to the emission controller board was effectively zero (< 1 uA). However, upon reconnecting after measuring current, the suspect 253 accelerating voltage power supply exhibited 0.5 V variability on 10 kV (just as the normally functioning 253 did). We had apparently solved the issue? A trace of intensity showed variation approaching normal. I ran zeros all night and at some point in the middle of the night, the instability returned. This morning, the instability persists.
1) Has anyone measured the accelerating voltage power supply and made note of the variability on a MAT253? Should I expect 0.5 V or is 10 V variation normal as well? Note - while watching the meter show 0.5 V variation, the IsoDat software shows 10s of V variation...so I assume the software is wrong.
2) Has anyone measured the current through any of the source plates? Am I correct in assuming these should all be zero? What is zero?
Any other thoughts?