Restoration glory of Keithley 2001 DMM

[girish]

Try replacing LM339 which drive those switches. I had bad 339's in two out of 6 units before. Also could be your U510 is leaky/flaky.

Thanks!
Yes, I am about to order some DG211Y s, 411s, and 339s. Its definitely temperature related since the test passes at cold startup.

[pelule]

If there is a temperature depending issue, you may search the critical part using coolant spray.
PeLuLe

[girish]

Replaced U510(DG211Y) today. Self-test have been running for 4 hours and passing. I'll keep it running overnight, but looks like a flaky analog switch which was acting up at elevated temperatures was the issue here.
Thanks!

[girish]

Automatic tests have been passing for 2 :(4+ hrs now. I think the unit is fixed. Now calibration $$$...

[Signal32]

Now calibration $$$...

What have people been quoted for calibrating a 2001 ?
Also any recommendations for a quiet fan replacement ?

[Le_Bassiste]

for my KEI2001, it was (done by TEKTRONIX / KEITHLEY in Germering, Germany)
Cal & Adjust: 340 EUR
Cal Report: 70 EUR
Pick-up and Bring Sevice via UPS 50 EUR
so, that totals to 460 EUR plus 19% VAT = 547.40 EUR. (insert ring of cash register here).
cal docs consist of two reports: as-received and calibrated, with used assets listed.

for my KEI2002, they quoted cal & adjust 700 EUR (sans VAT)

[TiN]

I'd suggest to keep same CFM/speed fan, to keep capacitor life. It also could affect calibration, as temperature condition is important with respect of all parts.
K2002's manual even state that unit need to keep same condition of rear scan-option opening as was during calibration (either covered or open).

Le_Bassiste

Congratz on your cal Is yours A08 firmware or B-series?.
Btw, that funny 20 uV range you saw on K2002 during power on, before trigger is not fake, it's lowest range meter can go with K1801 option.

[Le_Bassiste]

it's actually this one   
so, it's an A08

[TiN]

I was wondering where did the unit end up. Oh well, long live for meter. Would be interesting to know cal data, if you feel sharing.

I might find some testlogs of DCV and OHM tests of it, but that will need effort .

[Le_Bassiste]

here you go... https://xdevs.com/doc/Keithley/2001/K1/cal/KEI2001_CAL.pdf

edit:link updated

[Jay_Diddy_B]

Hi,

I got a shipment of new parts from Keithley to complete my 2001 restoration. I bought my unit for $30.00 Canadian. I did the 're-cap and clean' and electrically it is 100%. Now it is time to fix up the outside. I got the meter so cheap I decided to buy new parts from Tektronix. It took about two weeks to get the parts.


New Handle

My unit was missing the handle



New Rubber Keypad

One of the range arrow keys was broken on mine.




Metal Case

The metal sleeve was missing.



Rear Bezel

This part was missing when I got the unit.



Total Cost around $250.00 delivered to my door.

Regards.

Jay_Diddy_B

[RobK_NL]

The transformer costs 320(!!) euro shipped. awful lot of money for a simple transformer. Its from Tek. If somebody got a good one for less I'm still happy to hear!

Only just saw this and jaw dropped. I paid 179 euro last year.

[Jay_Diddy_B]

Hi,

I have to calibrate my Keithley 2001 post fix.

I have good equipment to perform the calibration including:

Fluke 732A
Datron 1281
Fluke 5101 which can be used as source, monitored by the Datron 1281
ESI 4W Resistance standard RS 925


So I am in pretty good shape.

The first step requires a 4 Wire short, using the Keithley 8610. I found this picture of the 8610:




And I have made this artwork:





Does anybody have an original 8610 or know what the other side of the board looks like?

I have read the Fluke white paper that says a low thermal mass is better.

Link: http://us.flukecal.com/literature/electrical-calibration/watch-out-those-thermoelectric-voltages-cal-lab-journal-reprint

Thanks !!

Jay_Diddy_B

[saturnin]

I bought 4-wire short from Fluke: 884X-Short (http://www.newark.com/fluke/884x-short/4-wire-short/dp/55M2652). It is nearly identical design to Keithley 8610. There is nothing particularly interesting on the other side

[hugo]

I bought my unit for $30.00 Canadian.

I thought the bottom price is 300 USD for an used and abused Keithley 2001 ... congrats to you

[TiN]

If one is patient, it's sometimes possible to get them for 100-200$ abused with missing parts. 

Seems it's a calibration week for everyone, don't mind if I join too?

[Jay_Diddy_B]

I bought 4-wire short from Fluke: 884X-Short (http://www.newark.com/fluke/884x-short/4-wire-short/dp/55M2652). It is nearly identical design to Keithley 8610. There is nothing particularly interesting on the other side

Thank you for taking the time to post the pictures.

Regards,

Jay_Diddy_B

[TheSteve]

If one is patient, it's sometimes possible to get them for 100-200$ abused with missing parts. 

Seems it's a calibration week for everyone, don't mind if I join too?

Seems you'll never really know what the real voltage is...   

[Le_Bassiste]

but now you know that "42" was just part of the correct answer!
(to a voltnut, that is)

[lukier]

Hi,

recently I managed to grab 2001 for parts/repair really cheaply. It seems the unit had a long history, manufactured around 1992, A06 firmware, RevK analog board and RevJ digital. Someone already replaced most of the caps on the analog board around 2004/2007 (these were the dates on the caps). I replaced some of them anyway and replaced all on the digital board (these were original and leaking a bit). All the power rails seem OK.

However, my problem is with the ADC board. It is also unusual - it seems Keithley didn't have 7.68 MHz oscillator, so they developed a 2001-132-02A mini board with XTAL, HCU04 & misc to generate the clock. It was failing all the 200.x tests and in consequence some other tests as well.

I've checked the front end switches and muxes and these seemed to work ok, providing on AD_IN 0V, 7V, 10V or 1.75V*5 in various diagnostic modes. Then I noticed that R857 (sits between -15V and NE5534) is damaged, it got burned pretty badly, leaving tar on the PCB and came off the PCB in two parts. I replaced that with a new 100 Ohms and hoped that was the problem, but unfortunately no I did some more probing, checked Q806 (as this was the issue for ManateeMafia) etc, and then I must have slipped the probe perhaps (I don't see any other explanation), because the Altera CPLD no longer works. It receives clock and data on SCLK/DATA_IN but DATA_OUT remains silent. Now built-in tests hang on 200.x.

I know it's a long shot, but maybe one of the forum members has a spare 2001 ADC board they could part with (for a reasonable price)?

[TiN]

Preparing some 2001 for upcoming sale out 

Initial calibration and check on 10V from Time 9823 MFC.



Guardband - Both K2002's and 3458A down below, absolute tolerance +/-1.5ppm (Both 2002 calibrated 25/08/2016 versus <30day 732B).
Blue charts - 3 different K2001's.
Dashed brown line - ambient temperature.

Clearly visible ADC performance difference between different league meters 

[Kjelt]

Wow those 2002s play in a different league. How does the 2000 hold up against the 2001 ?

[TiN]

I don't have 2000 anymore, it found a new home while ago

But I had similar 10V test on 2002+2000+3458 before. Datalog here.

[maxwell3e10]

I have a K2001 that at some point started to show "Overflow ADC" on all ranges and functions. On xdevs I found Illya mentioned replacing U315 optocoupler to fix this. Should I check something else before trying to get this chip?

I have an old TLP582 optocoupler on my board in a large package. It seems one can also get them in a DIP-5 package. Would that be OK?

[maxwell3e10]

I found another Keithley 2001 to swap the ADC card with and it is the problem with the ADC card, not opto-couplers.

My ADC card has digital signals coming in and out on most pins, so its not dead.
I don't know what the digital data should look like, though. For example DATA_OUT looks a little erratic, not a clear serial pattern.