Ok, now what are the differences between a Keithley 195 and a 195A?

[wkb]

Maybe I am too dim or the Google-fu has left me but: what are the differences between a Keithley 195 and a 195A?

Reason for asking: I recently picked up a 195 with a Uncal error, and most info I can find is for the 195A. Basically I wonder if info that I can find on 195A applies to 195 as well.

I hope to be able to resolve the error and then add the current / AC option.

tia,
Wilko

[alm]

Looking at the difference the -A revision adds in other Keithley DMMs, like Keithley 175 vs 175A, I expect the changes to be pretty minor, but possibly some circuit changes with more modern parts. I'd expect the specs to be pretty close, but the schematics may or may not match.

There's a user manual with schematics for the Keithley 195 right here, so I'm not sure what information you're finding for the 195A that you can't find for the 195?

I have a 1983 paper catalog that describes the 195. Comparing it to to the 195A in the 1984 catalog it looks like the 195A added temperature measurement and better AC voltage/current specs. Let me know if you want me to scan those pages, but it seems to me like you should be able to find all relevant information by studying the manuals.

[Swainster]

I've got a couple of 195As, both on my todo list. I bought the first one for fun, but it had slowly wandering readings (i.e. a few 10s of counts over the course of a day). As a TEA sufferer, I naturally bought another one to compare it to... and found that it wandered even worse . Despite this, I still dont think that this is normal behaviour, just that they are suffering from the same affliction. To be fair, the casing of the second one had been smashed to bit so it wasn't in best of condition.

Anyway, the point of this waffling is that the second one arrived with the uncal error, and a quick subsitution showed that the xicor X2212D NVRAM was at fault. A "new" chip from eBay solved the uncal problem (but not the wandering readings).

BTW, at least on my units, the 5V regulator runs really hot. I'd be interested to know if your 195 is the same.

[wkb]

@Aim: Well, my 195 has a little daughterboard holding the Xicor 221 NovRAM. I cannot find that in the 195 manual/schematics. I wonder if that was a ECO fix for something? Does the 195A have that etc.
I suspect the 2212 has died, there are 2x 2212 on the slowboat from China heading this way, hopefully that can help me cure the Uncal error. Not sure if this is a common failure?
As for the AC & current specs: as far as I can tell the Option 1950 is the same one used for the 195 and 195A? So this puzzles me a bit.

@Swainster: you mean the 5V regulator for the digital section on the heatsink? Yes, that is quite hot to the touch.

best,
Wilko

[Swainster]

Yes, the 195A NVRAM is the same - on a little daughter board.

I find that the heatsink attached to the 5V regulator is barely touchable, although it's the bridge rectifier next to it which has actually 'cooked' (discoloured) the PCB. Despite this high temp, the rectifier seems to be working fine.

[alm]

As for the AC & current specs: as far as I can tell the Option 1950 is the same one used for the 195 and 195A? So this puzzles me a bit.

There's a couple of options here. Maybe there was a change to the later production options, but that they didn't change the model. Or maybe the change is in the main unit, for example a firmware change. Or it could be that there was no change at all, and the improved spec is the result of having more data available on the stability of the units, like HP did with the 3456A. But in the case of HP we know this because they described it in HP Journal. I'm not aware of such a resource for Keithley, so we have to assume the lower accuracy for the 195.

[wkb]

As for the AC & current specs: as far as I can tell the Option 1950 is the same one used for the 195 and 195A? So this puzzles me a bit.

There's a couple of options here. Maybe there was a change to the later production options, but that they didn't change the model. Or maybe the change is in the main unit, for example a firmware change. Or it could be that there was no change at all, and the improved spec is the result of having more data available on the stability of the units, like HP did with the 3456A. But in the case of HP we know this because they described it in HP Journal. I'm not aware of such a resource for Keithley, so we have to assume the lower accuracy for the 195.

Well, yes, all possible. Beter stability components maybe. First I need to get the Uncal error out of the way, only once done I will investigate the OSHW Option 1950 board.

thanks!

[wkb]

I've got a couple of 195As, both on my todo list. I bought the first one for fun, but it had slowly wandering readings (i.e. a few 10s of counts over the course of a day). As a TEA sufferer, I naturally bought another one to compare it to... and found that it wandered even worse . Despite this, I still dont think that this is normal behaviour, just that they are suffering from the same affliction. To be fair, the casing of the second one had been smashed to bit so it wasn't in best of condition.

Anyway, the point of this waffling is that the second one arrived with the uncal error, and a quick subsitution showed that the xicor X2212D NVRAM was at fault. A "new" chip from eBay solved the uncal problem (but not the wandering readings).

BTW, at least on my units, the 5V regulator runs really hot. I'd be interested to know if your 195 is the same.

Was resolving your Uncal error accomplished by simply swapping the X2212 or was it more involved? Obviously you would need to run at least a partial calibration to get halfway sensible cal data stored. And, I presume, to get an OK checksum in the 2212.

Wilko

[Swainster]

I think saving anything to the NVRAM gets rid of the uncal message - I certainly haven't done any adjustment on it yet. Just to be clear, the uncal message is gone but as expected the original calibration data is still lost. I'm still looking for a suitable AC source before I can start doing any adjustments.

[wkb]

I think saving anything to the NVRAM gets rid of the uncal message - I certainly haven't done any adjustment on it yet. Just to be clear, the uncal message is gone but as expected the original calibration data is still lost. I'm still looking for a suitable AC source before I can start doing any adjustments.

Yes, obviously the original cal data is gone forever.  On my unit I cannot get rid of the Uncal message, tried to cal some DC ranges using my Hp3478A as a comparison DMM! so I assume the 2212 is a goner.

OK.. anxiously awaiting the slowboat from China here..