Restoration glory of Keithley 2001 DMM

[TiN]

Btw, there is another thread about 2000 series meters (2000,2015, 2010) - https://www.eevblog.com/forum/repair/repair-old-keithley-2000-teardown-and-fix/ , perhaps you can take few photos of rust and muck, and we can take it from there if it needs repair.

2000/2015 are very different meters to 2001/2002, PCB and topology wise.
You can compare both 2001 and 2000 from my notes & photos.

DCV readings jumping all over the place on 0.2,2,20V range (or 0.1,1,10V in your case) are completely normal, as on this ranges meter have high impedance and just picking up change from air/statics. You can short input with 10meg resistor and readout should be around zero.

[eas]

I'm not sure if the 2015 has the same built in self-test as the 2000 (the 2700 doesn't), but if it does, run it. You'll need some way of shorting all 4 inputs together for some of the tests.

[TiN]

That is normal, nothing wrong in this perspective.

[bson]

DS1245YK installed.   The TL866A doesn't support this chip and using the 1230 profile to initialize it didn't work for me, perhaps because the extra address lines weren't grounded but left floating and in my case floated high.  So I used the secret menu to initialize it instead, which worked great.  I assume the secret menu is the actual procedure used by Keithley themselves...  Nice to finally have the instrument back in use!

[TiN]

Review of 2001M internals is public now, sorry all for delay, took a while to get it all posted:



Article on photo click

Hope you enjoy the efforts 

P.S. I don't own unit, photos are shared to me by it's owner to be used in article for our reference.

[Zucca]

Sorry I was on vacation cruising with my GS1200 in Europe. Meanwhile you guys were having fun here...

Review of 2001M internals is public now, sorry all for delay, took a while to get it all posted:

and here the bomb...

For those who want upgrade their 2001 to 2001M, I tried to flash usual Model 2001 with digital board revision K (two ROM chips) with this firmware 2001M B17, and everything went well. Meter was detected as 2001M and all functions were operation correctly. No recalibration was required.

What a bless your article... I enjoyed it so much! Thanks TiN  O0

[dkozel]

There are two on Ebay at the moment for just under $400 US, listed as for parts.
http://www.ebay.com/itm/Keithley-2001-Parts-unit-/400988421081

I asked for photos and they sent these:
https://drive.google.com/folderview?id=0B5SWO8HGjYzSU2gwRG54MEx0Ylk&usp=sharing

Looks like one is missing a case and both were knocked around a bit. So sad. They didn't send any interior photos so I'm hesitant to go with them when I know that there are other deals out there.

Thanks for all the resources! I think I'm going to go with a Keithley 2000 instead as those come up more often in good condition under $300.

[eas]

Given how much that place asks for working equipment, I'd guess that those parts units have been stripped to the bone.

There have been some working (but not necessarily well) 2701s that have gone for ~$750 or so in the last few months. You should have no trouble getting a working 2000 for under $300. Actually, check the for-sale forum here. Timb had one he was selling.

[dkozel]

Given how they've only supplied the bare minimum of what I've asked for in each email I'm guessing you're right as well. They know the value of it in parts and have probably kitted it out, or at least verified that it isn't easily fixable.

I'll put a post up in the wanted forum, see what comes up. I'm looking for my first bench meter, having been using a Fluke 189 for the past four years. Computer control is a must which limits the options somewhat but thankfully all these high end meters come with RS232 and one day I'll get into the GPIB game.

[timb]

Yeah, I've still got my 2000 for sale. I just got a best offer on eBay for $250, so if anyone wants it here for that price contact me within 24 hours. I'd rather sell it off eBay and save the fees.


Sent from my Smartphone

[macboy]

Given how they've only supplied the bare minimum of what I've asked for in each email I'm guessing you're right as well. They know the value of it in parts and have probably kitted it out, or at least verified that it isn't easily fixable.

I'll put a post up in the wanted forum, see what comes up. I'm looking for my first bench meter, having been using a Fluke 189 for the past four years. Computer control is a must which limits the options somewhat but thankfully all these high end meters come with RS232 and one day I'll get into the GPIB game.

The Keithley 2001 has what appears to be an RS-232 port (a DB-9 connector) but it is definitely not an RS-232 port. That is a digital I/O port, with one input and four outputs. They can be read or controlled by GPIB commands and can be tied into some functions (high/low tests). The 2001 can be controlled remotely by GPIB only.

[Vgkid]

One of those k2001 looks hollow...

[dkozel]

macboy: That would have been a surprise if I bought one! The 2000 does have RS232, so another reason it's probably a better match at this point I suppose.

Vgkid: It's not hollow, but it's missing it's case so you can see straight through the bottom because it doesn't have the optional add-in card.

[Vgkid]

Makes sense, must be a full sized card.

[Macbeth]

macboy: That would have been a surprise if I bought one! The 2000 does have RS232, so another reason it's probably a better match at this point I suppose.

Yes, I have a 2000 and a 2015 and both work perfectly using the RS232 port with a cheap USB serial cable. All SCPI commands are available directly through a terminal, or using VISA drivers.

Even better, Keithley actually have an IVI-COM driver for download (most IVI drivers I find are IVI-C which are not user friendly unless you have LabView or something). IVI-COM and IVI.NET you can very easily use in VB, Excel, C#, etc.  (Mind you, just using SCPI with the VISA library is no big deal either) 

[eas]

There is a hardware and software flaw that lead to annoying drop-outs in RS-232 communication in many Keithley 2000s. The solution is to use a cable or adapter that leaves the RTS pin unconnected.

[TheSteve]

Received my TL866CS (now converted to a TL866A) today, read the 2001 B09 firmware and uploaded to the FTP site.

[TiN]

Added to collection.
Thanks!

[TheSteve]

Hah, installed 2001M B17, ran the self test which has never failed since the last repair and I now have faults again.

309.1 - Amps 200 uA range (this one isn't constant)
310.1 - Amps protect - Amps protection
412.1 ACA switch - AC amps switching error

Service manual gives good detail on them, I am now off to check CR305 and CR309 before anything else.

Edit - sure enough CR305 has an open diode in it. Now to find a replacement...

[Jaap]

Had 310.1 Amps Protect and 412.1 ACA Switch - AC Amp Switching Error errors too one year ago.
Also open diode in CR305. Original one here was a PE05 from EDI (Electronic Devices Inc.)  -  50V 5A
Replaced it with KBL04  -  400V 4A. Leg distance is slightly different though. Then all was ok again. Success

[Magnum]

First time 2001 user here. Bought one on ebay for a good price.
Received the unit, self test ok. Opened it up to check the production date, closed it and made some measurements.
Then I had approx 7V on ACV measurements with nothing connected. With AC coupling the 7V stayed even with shorting plug inserted. With AC+DC coupling it read 0V. Self test gave one error (forgot which one). Opened it up again and found relay K503 stuck closed. Took it out and AC measurements are ok, but self test throws 407.1 Front end, 407.2 Front end, 407.3 Front end, 410.1 TRMS, 411.1 TRMS, 411.2 TRMS as TheSteve had before.
I don't have the original relay available, so I used another reed relay. Error messages are gone except 408.6 /100 Freq Comp. But this error is not always there. I hope it will be gone when I receive the original relay.

[TheSteve]

I replaced CR305 today - all self tests are passing just fine again.

I did find one area in which the 2001 still has an issue though - measuring resistances over 20 meg ohms it varies a lot and is sometimes in the ball park but not enough to really trust it. This occurs in both the 200 Meg and 1 G ohm range. My test is a 22 meg ohm resistor which I know is fine. I am not the first person to work on this 2001 but the board looks very clean and it has been carefully cleaned in areas with IPA. Before and after cleaning there was no difference. I did notice with the cover removed and the analog shield removed if I blow gently under the A/D board I can consistently make the resistance drop down. I don't believe there is any problem with the A/D board itself, if I measure the voltage on pin 3 (A/D in) it varies when I see the incorrect values.

Any thoughts on how to futher troubleshoot this? It is something I can live with but it will always be in the back of my mind.

[TheSteve]

I replaced CR305 today - all self tests are passing just fine again.

I did find one area in which the 2001 still has an issue though - measuring resistances over 20 meg ohms it varies a lot and is sometimes in the ball park but not enough to really trust it. This occurs in both the 200 Meg and 1 G ohm range. My test is a 22 meg ohm resistor which I know is fine. I am not the first person to work on this 2001 but the board looks very clean and it has been carefully cleaned in areas with IPA. Before and after cleaning there was no difference. I did notice with the cover removed and the analog shield removed if I blow gently under the A/D board I can consistently make the resistance drop down. I don't believe there is any problem with the A/D board itself, if I measure the voltage on pin 3 (A/D in) it varies when I see the incorrect values.

Any thoughts on how to further troubleshoot this? It is something I can live with but it will always be in the back of my mind.

Well after messing around I suspect the problem is leakage/contamination in the front/rear switch - part S300. Has anyone removed one to clean it before or can I get a replacement easily from Digikey/Mouser?

[Magnum]

First time 2001 user here. Bought one on ebay for a good price.
Received the unit, self test ok. Opened it up to check the production date, closed it and made some measurements.
Then I had approx 7V on ACV measurements with nothing connected. With AC coupling the 7V stayed even with shorting plug inserted. With AC+DC coupling it read 0V. Self test gave one error (forgot which one). Opened it up again and found relay K503 stuck closed. Took it out and AC measurements are ok, but self test throws 407.1 Front end, 407.2 Front end, 407.3 Front end, 410.1 TRMS, 411.1 TRMS, 411.2 TRMS as TheSteve had before.
I don't have the original relay available, so I used another reed relay. Error messages are gone except 408.6 /100 Freq Comp. But this error is not always there. I hope it will be gone when I receive the original relay.

Unit is fixed. Received the original relay (thanks TiN), but that didn't cure the problem. Q518 was shot, IMHO because K503 was stuck closed before.
What I didn't understand was that test 409.6 /750 Freq Comp. made no problems at all, as this is the same test procedure (except U526 switched on) which needs Q518, too.
Don't care why as it is error free now.

[TheSteve]

Finally received my Dallas DS1245Y from a seller in china, it took 2 months but it arrived and is working. It had data stored in it so I assume it is a used pull. I blanked it using my TL866A and then used the secret menu in the 2001 to enable it. Once enabled I read it it out again and see it stored the serial # and another 65 bytes after that with what looks like a two byte checksum at the end.