Keithley 237 intermittent startup issue

[Spike]

Hey all,

I just grabbed one of these units off eBay, advertised as “used, powers up, as-is.”  I asked the seller whether it actually worked and he said that it does and was pulled from a working environment, but some time ago.

Needless to say, when I got the unit it was exhibiting a problem from the start (pun intended).  Namely that initially when I pressed the power switch, about half the time it would start up without issue, passing self tests and getting to standby mode.  The other half of the time the fan would come on (I understand this is a 115V AC fan on its own transformer tap?) but the display would remain blank.  No errors, no other indicators, just remained blank.  I have not opened the unit yet because I want to be sure I’m going to hang on to it, but it feels like a mechanical issue rather than an electrical issue because lifting an edge of the unit about a centimeter and setting it down firmly resolves the issue sometimes, but usually only for 1 startup.  At this point it almost never powers up without some jostling as described above.

Once started the unit seems to check out based on the limited testing I have done.  it sources voltage fairly accurately on all ranges, including the resonant converter, right up to the limit.  It also seems to source current accurately through the full range of compliance voltages as well.  It may or may not be within calibration limits but if it isn’t it’s very close.

So my question is, has anyone experienced a similar issues with one of these units?  Does anyone know if there are common issues of this sort with these units.  Yes, I will crack it open at some point and have a look but if anyone has a clue about what this might be before that, it would make my decision easier.

In case anyone is wondering why I’m hesitant to just crack it open it is because the seller added some tamper resistant tape to the case between the time it was photographed and when I received it.  Presumably this was to protect himself against scammers who buy stuff and remove parts??  In any case if I decide it’s not worth my trouble I don’t want to complicate things by being accused of trying to scam him.

[D2236]

Had a similar problem with a 237. I removed all the socketed IC's, including  EEproms and CPU from their sockets, sprayed the sockets with a good quality switch cleaner, and reinserted everything again. After this, I had no more problems.
Ask the seller if you can open it to check, or ship it back. They may agree for you to have a look.

[Kean]

Intermittent issues like this can certainly appear only after shipping.  It probably was working fine when last used, and very likely an IC socket or connector as D2236 mentioned.
Unfortunately such issues can also be a pain to track down as they may come and go randomly, so you need to weigh up the pros and cons of keeping vs returning.  As suggested, discuss with the seller.
If you got it at a good price and when it does start up properly it also seems to function correctly, then I'd probably hang onto it.  You will likely want to open it up anyway to replace any older capacitors, and any ongoing maintenance.

[Spike]

Thanks, guys.  Once I saw that my measurements were close to spot-on I think had decided to keep it.  I just didn’t know it yet.  Because I feel like the issue is mechanical, and because it is intermittent I do worry that tracking it down might be a pain but I guess I’ll take a chance.  If anyone has advice on diagnosing such issues I’m all ears.

As it turns out, once I decided to keep it I purchased a second parts unit from the same seller (I didn’t mention previously that it was local pickup) for what I consider an extremely good price.  It is complete and intact and said to blow fuses.  That’s a repair for another day, though.  For now I’ll focus on trying to track down this intermittent issue.

As an aside, the “working” unit is firmware revision A06 and the parts unit is rev A10.  Is there any dependency on PCB revision for this firmware or can I upgrade at will?  Also, are there any sites that have release notes for these revisions?

[David Hess]

I usually start with the power supplies and electrolytic capacitors when test equipment is old, but anything mechanical, especially edge wipe sockets from Texas Instruments, is suspect.

[Spike]

Of course now that I’ve got it opened up I cannot provoke the behavior at all.  Sigh.

[Kean]

Of course now that I’ve got it opened up I cannot provoke the behavior at all.  Sigh.

Yay!  Fixed!   (for now...)

If you really want it faulty again, you can try...
- Heat or freeze spray
- Drop from 1 inch exponentially increasing distances above bench

[Spike]

Yay!  Fixed!   (for now...)

Indeed, for now.  Right up until I button it up, probably.

I’ve poked, prodded, tapped and flexed.  Nothing so far.

[David Hess]

I would at least reseat any connectors and IC sockets.  Does anything have an IC socket?