Author Topic: Repair and calibration of Keithley 237 (236/238) (Read 9532 times)

balage · « **Reply #25 on:** January 12, 2020, 12:39:53 pm »

I have tried the same way with Agilent Ebay GPIB converter. I have tried LabVIEW and the communication expert software. Only once I've succeeded to put the instrument into remote mode and was able to read the measurement data. But since that after several attempts I was not able to put the instrument to remote.

Possibly can you please tell/send me the exact method/procedure/message with which you did it?

I am waiting for a National Instruments GPIB USB converter, just to avoid the Agilent one.

rolkinas · « **Reply #26 on:** January 12, 2020, 08:35:13 pm »

I also have bad luck with Ebay Agilent interface + Labview combo.
On only one of many computer it works more or less reliably , but all settings are the same in other maschines.
If you have Agilent USB-GPIB adapter it come with driver+software. It can be downloaded from keysight also.
In Agilent instrument detection software is option to send individual commands to instruments.
I used that way, simply inputing commands from service manual calibration chapter.
C0X {enter}
C1X {enter}
and so on

balage · « **Reply #27 on:** January 19, 2020, 09:54:24 pm »

F...in' freakin' bastard 82357B... I've spent so much time with the Agilent one with no success.

It took two clicks to communicate with the Keithley using the NI GPIB converter.

Anyway... I have a 82357B converter for sale. Anybody?

I decided to calibrate the 237 regardless the faulty resonant converter. I want to feel success after all this trouble.

MiDi · « **Reply #28 on:** January 19, 2020, 10:56:44 pm »

Have not reread all answers, so forgive me if this is already known:
There is a topic about similar problems: High voltage troubleshooting of Keithley 237
The issue with limited voltage output with installed short between low to gnd: Keitley 236 Source Meter - Teardown and Review

balage · « **Reply #29 on:** January 23, 2020, 07:54:34 pm »

Finally the problem is identified and solved! Here is the procedure I've been through.

So the symptom was that sourcing below -350 Volts the resonant high-voltage converter became loud and the output voltage dropped. I have changed the output filter capacitors (orange ones) and the dummy load at the output. It has not solved anything. In fact one of the ceramic caps was blown. After many hours spent on an incorrect direction, I have put a 120kohm dummy load on both rails of the converter. It has sourced the 10mAmps perfectly. Converter was locked out as a problem-source.

Output stage was the next step. The voltage dropped on the Q3 & Q4 seemed good with an output voltage of -50 V. Each of them dropped abound 400 V. When the output was raised lower than -170 V the voltages dropped on Q3 & 4 were fluctuating from 250 to 350 V. Below -350 V the converter started the loud oscillation and the voltages dropped on the FETs were zero. I have changed Q3, Q4, Q10 and Q19 but with no success.

After all this I have stared monitoring the -150V rail on C19. It was -170 volts actually. Going down vith the source voltage this rail was decreasing down to around -400 volts!! CR4 is a high voltage diode with the purpose to separate the high voltage rail and the -150 V rail. It seemed the diode is not acting this way. I have changed it to a big ugly one and everything is fine.

AFAIK there were not any 237 related problems here like this one. I hope it will help somebody one day.

A fellow member help me a lot solving this issue. Thanks z01z!

If anybody needs an MTP2P50 FET or MJE350 or MJE340: it is available here at the local dealer just PM me. They are geniue.

Next step is calibration. Also I would like to pimp my 237, with changing the high voltage wires to nice, flexible silicone ones, replace the 150kohm resistors near the supply to higher power rated ones, etc. Thanks for all your help!

balage · « **Reply #30 on:** January 24, 2020, 05:08:35 pm »

I have just made a huge step to have a calibrated 237 that is working perfectly I am now stopped again.

Something is wrong with the calibration. When I send the command C0X, the instrument enters calibration mode. I know this because in case I put the 237 back to local and I push any of the buttons it is written on the display that "in cal".

But from this point the instrument is not responsing the commands. It is not doing anything (at least I cannot see) for C1X command. Also when I send C4X it is not outputting +1V. It is not sending any characters back at all.

Can anybody say if their instrument is acting the same way? Should the instrument source +1V by the command or should I set it manually? It seemed me it should do itself. Also, Does it sending back any acknowledgement for the cal commands?

rolkinas · « **Reply #31 on:** January 26, 2020, 03:06:46 pm »

instrument should output corresponding calibration voltages itself. Is unit in "operate" mode during calibration?

balage · « **Reply #32 on:** January 26, 2020, 04:55:28 pm »

Yes, it is in operation...

Should it send back any acknowledge message?

balage · « **Reply #33 on:** January 29, 2020, 08:30:36 pm »

I tried to calibrate the low current measurements.

To test if it is doing anything during calibration, I have applied 0,6nA to the 237 instead of 0,9nA. When I sent the cal command it hasn't done anything, as the applied 0,6nA remained on the screen. In this case the 237 should have tought the 0,6nA is the 0,9nA.

Isn't it?

MiDi · « **Reply #34 on:** February 03, 2020, 09:20:16 am »

I have similar problems in calibration mode.
At least when C1X does not take ~30s there is an error.
The errors can be read by SPOLL command followed by evaluation of the status from UnX command.
A good source how the cal procedure should be done is the example calibration program in QBasic at end of cal chapter.
It is rather complex, so it will take quite a while to implement this in python.
Updates will follow...

balage · « **Reply #35 on:** February 03, 2020, 02:52:28 pm »

And how can you see if it has executed the C1X command? I cannot see anything changed on the display.

MiDi · « **Reply #36 on:** February 03, 2020, 04:00:17 pm »

The display seems rather unhelpful for cal procedure.
The time it takes for C1X will delay following readings over GPIB e.g. SPOLL, read UnX.
E.g. if there is no short installed, there is an error and no delay.

balage · « **Reply #37 on:** February 03, 2020, 05:53:21 pm »

In the meantime it seems I was able to calibrate the low current measure range. During calibration I was checking the U7 and U1 status words. U7 was always updated with the current step of calibration. Also when the last step was executed, the instrument displayed "temporary cal" as it is supposed.

I have not saved the cal data tough. But after a restart it showed a bit different measured values than with the temporary calibration.

So I have gained of lot of skills about the 237 in the recent few weeks.

MiDi · « **Reply #38 on:** March 07, 2020, 07:06:35 pm »

Now i have three good reasons to develop (semi) automatic cal program for whole SMU family

balage · « **Reply #39 on:** March 07, 2020, 08:21:55 pm »

That is really nice!

Are all of them working well?

I was thinking how such a semi automated program should look like. I would program it in LabVIEW, and pop-up windows would show the instructions what voltages/currents should be applied to the 236/7/8.

MiDi · « **Reply #40 on:** March 08, 2020, 08:29:31 am »

It seems I had much luck with the purchase of 236 & 238

Just needed to replace both blown fuses and after thorough inspection, they turned out just fine

Quick check of the ranges revealed no surprises, maybe it just is out of cal on a couple of ranges.

Now they are burning in and hopefully I have some time for cal program.
It will be python based for linux-gpib and more or less a port from the example qbasic program in the manuals.

Edit: I plan to implement full performance verification, cal procedure (adjustment), reports and interface for extension of instruments e.g. different DMMs and calibrators.

garrettm · « **Reply #41 on:** April 17, 2020, 02:47:25 am »

@balage

Would you happen to know what the part number of CR1, CR2, CR4 and CR6 is? Keithley says they are "HV-15" but I haven't had any luck finding a datasheet for this part.

I recently picked up a 237 parts unit that turns on and noticed that the output board had already been "repaired". (Oh great...) The previous owner replaced the afore mentioned diodes with DG1-827 (mfd by General Semiconductor), but it actually looks like Q19 and maybe a few more P-chan / PNP transistors are the real fault. (The output is stuck at around -90 V to -140 V, depending on range / programmed value.) I've got the new output transistors ready to be installed, but before I turn the unit on again, I'd like to make sure the diodes that were swapped out will actually work and not cause the output board to fail after I go through the trouble of repairing it.

MiDi · « **Reply #42 on:** April 17, 2020, 09:07:32 am »

Seems the diodes are unmarked and special selected with red dots.
Output Mosfets are marked Motorola R251 and R315.

garrettm · « **Reply #43 on:** April 17, 2020, 09:11:43 am »

Well, I looked into the DG1 diodes, and they are rated for 1.5kV peak repetitive. So I figure they are actually okay. That said, you always need to check if the guy before you actually knew what they were doing when swapping out parts. I've seen some shenanigans that still to this day make me question every little detail when something has been worked on previously. But enough of my ranting.

The fix turned out to almost be too easy. It really was just Q19, a lowly MJE350 PNP that shorted out collector to emitter. The 26.1 ohm resistor in series with the emitter got pretty toasty, but still measures spot on. So I’m going to leave it alone for now. The two 2.2 megohm 1/2 watt resistors in the resonant converter also look like they got pretty hot (enough to discolor the PCB and turn the blue resistor bodies green-brown). I haven’t tested outputting over 190 V yet, since I’ve been using the ole Keithley 617 electrometer as an initial sanity test before I delve into recalibrating the 237.

Over all the 237 looks to be a nice upgrade over my 220 current source. Better programming resolution and accuracy when sourcing current and infinitely better voltage compliance. Admittedly, the 220 does have larger output ranges below 100mA, 19995 digits vs 10000. So it’s not really a direct comparison. But the internal picoammeter/electrometer more than makes up for that.

All-in-all, a good buy at 250 USD.

balage · « **Reply #44 on:** April 19, 2020, 07:18:45 pm »

Hi garrettm,

Sorry for my late answer. I have removed one of the diodes you are asking and it was labeled PD76. Nothing can be found on the web.

I have temporary replaced it with a huge 6kV diode, but I have bought GP10Y-E3/54 from Farnell. This will be the replacement for me.

That was a really nice bought for you!

garrettm · « **Reply #45 on:** April 20, 2020, 08:11:41 am »

Thanks balage.

It's odd Keithley used such obscure diodes, but at least Vishay (who bought out General Semi) carries 1500 / 1600 V diodes that work as replacements.

balage · « **Reply #46 on:** April 20, 2020, 08:21:08 am »

But most of the parts are available nowadays as well. Like the big orange ceramic caps.

Quote from: balage on January 23, 2020, 07:54:34 pm

If anybody needs an MTP2P50 FET or MJE350 or MJE340: it is available here at the local dealer just PM me. They are geniue.

If you happen to wish stock replacement FETs, I can buy and send them to you. Just sayin'.


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Repair and calibration of Keithley 237 (236/238) (Read 9532 times)

Share me