EEVBLOG 121GW voltage readings slowly increasing and LowZ mode not working

[sojhalb]

Hi,

I have a 121GW multimeter which started exhibiting some problems namely: the voltage readings slowly increase over time, I used a Keithley power supply at work set to 3.6V and the readings slowly creep up reaching 3.7V after a few minutes, the Low Z mode just reads 0V and appears to just not be working altogether.

I have opened the meter up to inspect and didn't see anything that was obviously visually wrong.

I have emailed Dave about the problem and haven't received a response, so I'm reaching out here to see if any one has ideas as to what the culprit is.

Thanks,

Sojhal

[Kleinstein]

A power supply may not be the very best test source. It would not be the first power supply to show some drift. At least make sure the supply is warmed up before the test, so one would only see the power on phase of the DMM and not the source.

It the current mode still working OK ?  E.g. use a good quality (sufficient size or low TC) series resistor to create a current of some 2 mA or 200 µA. First lett the source settle (resistor warm up) with a short and only than use the meter.

A point to test would be the input resistance: it this still 10 M in the normal voltage mode and a reasonable value (could vary quite a bit from a PTC) for the low Z mode.
There may be a fusible resistor or ferrite at the input that has failed (e.g. cracked).  AFAIR there was some issue in the early versions with a part placed in an area with high stress and thus susceptable to mechical damage.

[sojhalb]

Hi Kleinstein,

I forgot to mention that the supply was on for several hours before and was in use at that specific voltage for other tasks so it was sufficiently warm and I also confirmed with another multimeter that the power supply itself was not drifting upwards

Regarding the current test, I set my supply to 10V to feed a 8.10 kΩ resistor, and let it sit for 2 hours, afterwards the resistor measured 8.11 kΩ, to draw about 1.23 mA of current. I then put the 121GW in series with the resistor in current mode and it measured 1.39 mA at the start 1.43 mA after 1 min, 1.65 mA after 2 minutes, 1.74 mA after 3 min, 1.79 mA after 4 min, and 2.94 mA after pulling out the probes from the multimeter.

I confirmed with another meter that the current was in fact exactly 1.23 mA.

I then checked the signal path through the mA range, namely R43, SHUNT, FB2, R126, R46, R106, FB3 and R104 (refer to the 121GW circuit diagram).  All the resistors measured their correct values and the ferrite beads measured shorts.

I then measured the input resistance in the volts mode and it started out at like 2 MΩ and then dropped to like 500 kΩ, I did the same test with my other meter the BM235 and it measured like 4 MΩ and then later dropped to like 1 MΩ so I'm not sure if I did it right or not. 

Thanks,

Sojhal

[sojhalb]

I should add that I also measured the impedance of the LowZ mode and it measured to be 2.88 kΩ and that I measured the turn on voltage of BD1 and it was 0.00V or a dead short (so I'm beginning to suspect that).

Thanks,

Sojhal

[sojhalb]

I take back the bridge rectifier measurement, its actually aproximatley 1.07V, I believe I was accidently touching the 0.01 Ω current shunt with one of the probes earlier.

[Kleinstein]

So much drift in the amps range is unusual. It seems to be even worse than votlage. This points to a problem that is not just at the voltage input, more like a problem with the main chip / reference.

2.8 K in the low Z mode is reasonable. The mormal mode should be more like 10 M , maybe a little lower if the MOVs are about to blow up when they see mains.

[sojhalb]

Yea that was my fear, the hycon chipset doesnt seem easy to get hold of