Keithley 196 repair adventure.

[work_smarter]

Good day everyone. - A few things right off the bat. Thank you for providing this resource. Next as a long term web author and webmaster, I DID USE THE SEARCH FUNCTION. I did not find anything that was of help to me specifically for the failure mode of one of my prized, go-to bench meters, the Keithley 196.

Due to time constraints, other factors and a ridiculously low sale price, when my original failed, it seemed best to just purchase a replacement and put the malfunctioning meter on the list of required repairs and calibrations. I have yet to open the meter up, but I thought it might be worth posting about the failure to see if this is a common or known failure mode.

One AM I got up and went into the lab and fired up the 196 for warm up and right out of the gate, in all ranges -it simply reads 6.5 digits of totally random numbers. Connecting test leads to anything in the appropriate range seemed to do little and more likely nothing to the errant erratic display. The numbers change at the meter's sampling frequency. Since the time this happened and I retained a replacement and calibrated it, well fixing my original meter has gone for far too long on the back burner.

So, while yes, the first things I would do is check the voltage rails and regulation for deviation (or insanity in this meter's case) as a starting point and I do have some other notions as far as good paths to follow to try my best to locate the issue(s). Of note, I smelled no magic smoke emitted by the unit at any time.

The simple question is, instead of 3 hours of troubleshooting (and hoping for success) has anyone experienced this type of failure with a Keithley 196? It would save a lot of my overbooked valuable time if this is a common failure and someone has a direction to point me in.

Reality speak is that while I know with enough time spent I can get the unit operational and ready for re-calibration, but that could mean me poking around in all of this spare time I do not have and hope I can find the root cause for the failure. All I know about the time it failed was simply I walked into the lab and turned it on to stabilize it and that a mere 10 or so hours earlier it was in use with no issues before shutdown.

The ESP that I don't have makes me think a voltage regulation issue or perhaps an E/EPROM failure.

Bottom line, has anyone had a Keithley 196 fail in this manner? And if so, what was determined to be the root cause? Thank you all in advance, if this failure has been seen on other 196's and they were repaired, a response could make a huge difference in the time I have to put into this meter - which I really do like a great deal and am low on time to allot to the issue.

PS: All of the self diagnostics seemed to work properly, but that "might be" disputable as that was well over a year ago.

Regards,
Douglas W.

[Vgkid]

1) Check rails,check for ripple
2) since it seems like your input is floating, even with your probes shorted.(do your probes test out good, meaning do they short on another meter)
3)Connect a wire(add a series resistor, somewhat small(current limit in the event of misplacement)) to the negative terminal. Starting at the input work your way back until the reading propery zeroes itself.
Note, I haven't read the users manual, but it worked for my 7065.

[plesa]

For first shot I will suspect LM339 comparators.
https://www.fer.unizg.hr/_download/repository/Keithley196_im.pdf

[work_smarter]

Thank you both vgkid and plesa.

Probes are fine actually moved that pair to the replacement so as not to open another pair at that time.

Good advice and where I was going to start working my way back inward. I have the factory published manual and from both experience and maybe some logic the top of my list was regulation on the rails, or. Vgkid, you could be spot on with the comparators - if any of those toasted, that might produce this. (And thankfully they should be easily replaced if they are the source of the problem.)

Thank you to you both and anyone else that may post. The one thing I will do is document all of what I find and the repair procedure if it is at all challenging I'll post back here once I have success for the next guy. --And Googleing a Keithley 196 and the word random or common faults, failures, etc got me no place at all on Google or here.  Hopefully, I have success and can add my findings to this wonderful community.

Will try and tear into this tomorrow - this has been sitting too long and upsets me to look at it being non-functional! 

[plesa]

On other Keithley gear mostly i needs to replace comparator(s) only. Fault is  similar ( random reading).

[wiss]

On my 192 one of the JFETs started leaking ever so slightly. Check the current through the gate-resistors. In my case I had several mV over the 1meg gate resistor of the AC-volt switch-jfet. In your case there would be a lot more than that.

(edit typo)

[work_smarter]

Thanks to all that have posted. It has been a crazy weekend and I have to run another errand, but late tonight or early AM tomorrow I am going to tear into it. It is a strange sort of irony that I'll use a 196 to diagnose its brother.

Thanks and I'll give a full report.

[TimFox]

My Keithley 196 developed a similar problem, which I was able to "fix" temporarily by thumping the back of the meter up and down and manipulating the rear/front input switch on the rear panel.  I'm not sure which "therapy" was efficacious.  If I leave the unit powered up and stationary, it seems to stay OK, but cycling power or moving the meter can bring the erratic readings back.  I also noticed that multiple "therapies" of this type were sometimes required to get both voltage and resistance readings to stabilize.
My guess is the rear/front input switch, but I haven't been able to prove that.

[greasemonkey]

My 196 started having this exact problem 2 day ago. So I opened it up yesterday and the -15V rail was a bit off.

Long story short C106 (the 620uF capacitor before the -15V regulator) was dead. I changed it and the problem was fixed.

I had been meaning to change all electrolytic capacitors but they beat me to it. I will compile a list of capacitors to be changed and post it (eventually  ).

So if you are having this problem before you go down that long, cold rabbit hole start with the obvious. Voltage rails on connector P17.

[greasemonkey]

As promised here is a list of the capacitors. Dedicated to the lazy folks (like me)  .

Designator	Description	Digi-Key part no.
Display Board
C1	10μF, 25V	493-13463-1-ND
Main Board
C101	10000μF, 25V	493-7297-ND
C102, C104, C107, C108, C127, C140, C147	10μF, 25V	493-13463-1-ND
C103	1500μF, 25V	493-15301-ND
C105, C106	620μF, 35V	P14940-ND
Analog Board
C33	10μF, 25V	493-13463-1-ND

And a video of me replacing them.