Agilent U1252B DCV & Resistor measurement not working properly

[hlj.pycad]

I got two U1252 multimeter, one is U1252A, the other is U1252B. Their are almost identical and has the same feature as U1253A/B except it is LCD monitor while U1253A/B has a fanccy OLED, and BOTH of my two U1252 have problem with measuring DCV & Resistor when most of the other features are fine.

I record it to a video to demostrate what is like now, here are my observations:
1. with probes open, the resistor reading is not "OL", it's tens or hundreds Mega ohm. Measuring resistor more than 1M will notice obvious drift.
2. it's highly huminity sensistive, a few second's of hair-dryer blowing will make the reading to "OL", and everything seem more accurate than before, but it will slowly drift back to prior status.
3. DCV is similar, the reading drift and highly senstive to the hair-dryer.
4. DC mV is very accurate and not drifting. So I believe it about the attenuation/amplification?
5. There are 2 peice of PCB, and I tried apply the hair-dryer seperated, the buttom one is suspicion for the problem, it's much more sensitive to the hair-dryer.
6. I have two of these multimeters and BOTH of them have the same problem, so I think it's kind of common problem for this model, maybe there are people've run into and solved it, but there are not more useful result by Google "U1252A problem".

Thanks in advanced.

[jchw4]

Did they experience any battery leaks? If they did I would suggest searching for "Fluke 189 battery leak" here on eevblog forum and you will get plenty of information on cleaning the contaminated PCB.

[tooki]

2. it's highly huminity sensistive, a few second's of hair-dryer blowing will make the reading to "OL", and everything seem more accurate than before, but it will slowly drift back to prior status.
3. DCV is similar, the reading drift and highly senstive to the hair-dryer.

When you say you’re using a hair dryer, are you blowing cold air or hot air? Most hair dryers only blow warm or hot air, and it seems to me FAR more likely that it’s the heat, not a drying effect, causing the behavior you observe.

Another possibility is that it’s neither heat nor humidity causing the effect, but static electricity. I have two U1252B myself and I know they’re quite sensitive to electric fields even when the case is closed, so with the case open you could be creating static charge right on the board, which then dissipates. (And if you’ve left the filter setting turned on, that will also make the displayed value fall slower.)

[hlj.pycad]

Warm air indeed. At first, I was thinking warming the PCB up cause it, but I also tried puffing air by mouth to it, the reading will instantly drop significantly, which I forgot to put it in the video. So warm moisture has oppisite effect to warm dry air, I believe huminity take a greater part in it, again, just an observation, no technical evident by now.