Keithley 2001 repair

[beuford]

I picked up a non-working Keithley 2001 today. I've located the service manual and am reviewing repair options. I'm not an EE, but I have an engineering education (Civil Engineering). I've dabbled in electronics as a hobby as a young man, and I have basic solder skills. A repair is not out of my capabilities (I hope - these boards are more complex than I've worked on in the past), but any advise would be greatly appreciated.

I have not tried to power it up, but the owner advised that the power fuse blows on each power-on attempt. I removed the case and found what appears to be leaking electrolytic caps (based on the oily film near them on the power board). I've been searching the forum and have found this to be a common problem with the 2001. Pictures are at the bottom.

Questions:
1) Is there any thing else I should look to replace while I'm inside of the case?
2) Are there specific recommendations for replacement caps?
3) Any suggestions/shortcuts/gotcha's regarding disassembly?
4) Are professional repairs available, and if so can you recommend a shop?
5) What's the add-in board in the rear-accessed slot?
6) I've previously used anhydrous alcohol spray to clean PCBs; is this safe to use on these boards?
7) Any feedback re: the photos?

I would like to get this operating, as it is probably the only chance I'll have to own a Keithley (REALLY good deal on this one).

Thanks,

Beuford

[TiN]

You can reference to my work on these units here, here on forum and higher end sibling here.
Must remove ALL electrolytic caps, clean board VERY thorougly with clean fresh IPA, repair any damages, including broken vias and inner layer traces.

Then troubleshoot and repair the failed tests, the unit has pretty comprehensive self-test. After this you are in looking for somebody with high-end calibrator to adjust the unit up to spec, or for 300-500+$USD for calibration from lab or Tektronix (today Keithley part of Tektronix and co).

Addon card is the scanner board.

For professional repair, I believe your best case is T&M lab or Tektronix, but you will end up paying close to what new unit cost, so whatever your really good deal was, it will not be the one anymore  .

[beuford]

For professional repair, I believe your best case is T&M lab or Tektronix, but you will end up paying close to what new unit cost, so whatever your really good deal was, it will not be the one anymore  .

I dunno - free is still a pretty good deal. Worst case I can always sell it for parts.

It's really too bad that they built the power supply on the analog board. This would be much easier if I could just replace a bad power supply board.

Beuford

[TiN]

Now you need to spend time on it, and if you don't have expertise and equipment to fix such precision meter, it will be all for waste
So I'd be careful with free items, as time is not free. 

[HalFET]

It looks as if it hasn't progressed too far, I think? (Though TiN would be infinitely better at judging this, in my humble opinion. Mine wouldn't be working if it wasn't for him.)

If I had it myself I'd give repairing it a go, since these are the heavy lifters as far as bench DMMs go. But, you could always sell the scanner card for the typical €150-200. Front panel assemblies can net you anywhere from €80-200 depending on how much demand there is in your area at the moment. But, please, only do that if you're 100% sure it's dead and no-one is willing to give it the care and love it deserves.

[beuford]

I followed your links to the manuals; Thanks!!

However, the manual references some drawings (re: "see drawing 2001-051", etc.). The drawings are not part of any of the files. Any idea where I can find the drawings references in the repair manual?

Thanks again,

Beuford

[HalFET]

All files are listed on https://xdevs.com/fix/kei2001/

[beuford]

All files are listed on https://xdevs.com/fix/kei2001/

Funny... I can't find the drawings referenced in the repair manual. Oh well; repairs are going well, anyway. I got the analog board out OK, and there does not appear to be any damage to the board (although I can't see all of the bottom yet). Any idea on how to remove the plastic bottom cover without destroying the mounting pins? There must be some trick for these...

Beuford

[HalFET]

The right area near the heatsinks looks pretty bad, that'll need a good IPA treatment. The centre one it's hard to tell but looks like it's just the one that went on that picture. Think you might have caught it in time!

[Kleinstein]

Besides the capacitor leakage,  CR106 and C107 and area around it (power input to transformer) also look suspicious.

For cleaning the residue from the caps, water can be more efficient than IPA for the first step. It needs care not to contaminate the board in parts that are hard to clean, like below the SMD chips.

[TiN]

Horrid electrolyte spills everywhere. So procedure is standard:

[beuford]

Ok, so ready to work on the board. I'm using a 3M anti-stat mat & wrist strap, digital temp control Weller soldering iron, and fine copper braid for wicking solder away from PC board, an anti-stat solder "sucker", and an old Fluke meter if needed. However, I have a question or two (remember, this was a hobby for me in the past; I'm not an experienced EE):

1) The manual suggests using water to clean the boards, spot-cleaning with alcohol. Because of the mess made by the caps, I'd rather blast the board with either spray anhydrous isopropyl alcohol or a flux cleaner. I have the stuff shown in the picture below; has anyone used any of these products on these boards (without anything melting)?

(BTW - the capacitor residue looks very oily... should it be water-soluble?)

2) The manual calls out to use "OA" solder; I have RA. Other than water vs. isopropyl cleanup, any significant issues with RA over OA?

3) Any particular recommendations for the caps? Or can I use the Fry's NTE products (actually marked "Jackson") as replacements? These all appear to be Nichicon VZ(M) type caps.

Beuford

[HalFET]

Depends on who you ask. Keithley suggests methanol sometimes as well in service manuals for other instruments (electrometers and SMUs). But TiN said IPA before, so think it's ok.

Overall dry IPA is mostly an issue for connectors. Most electronic components themselves withstand it. Just be sure to put it in a solvent drying oven after the fact, or let it sit in a dry spot for a day.

We also often put boards in dry IPA in ultrasonic cleaners if it's for high impedance applications, but don't even try that unless you have a good fume hood. But I'm not sure that'd be a good idea with an old board like this.

[beuford]

Thank you for the recommendation, but that's a bit more work than I'm willing to put into this project. I am hoping to get it running by replacing the caps. If it doesn't work then I'll debug further. If it takes work much beyond that, though, then I'll sell it for parts and save my money for one in working condition.

The board actual looks quite good; no (obvious) signs of broken/melted traces or excessive heat. All the messy parts of the board appear to be only leaked electrolyte and a bit of dust. I have high hopes that it will work once everything is cleaned and the caps replaced.

Thanks again,

Beuford

[TiN]

That "only electrolyte" is exactly the problem. It is highly corrosive and conductive, causing all that analog circuitry around to fail miserably. And since many parts in red area powered by floating 72V power doubler sh1t get nasty real bad if there is even little bit of electrolyte left under some chip. Removing all parts and replacing stuff is _quicker_ than doing "usual repair" approach in this meter. Read Keithley 2002 repair thread from nikonoid in repair section to learn the right way and how much effort it took him, on meter which is far less corroded than yours. If you don't trust that and looking for easy way, well, you got wrong meter, and just sell it before destroying further by power on attempts I had eight of these meters before and spent two years fixing them till I got five working good and sold em for sake of better meters.

[Kleinstein]

There is at least one capacitor and rectifier that look bad too. To blow the fuses it takes quite a lot of leakage. Also with non working caps some other parts could be damaged easily, especially if the power supply delivered too much voltage  - worst case this could still be a unit beyond repair.

For the sensitive parts, it only takes minute leakage (e.g. pA range) to degrade the performance. One can not even see the small amount of electrolyte to cause this. Especially cleaning under SMD parts works best with unsoldering them. The board is way to large to take the chance hunting for a droplet left under a chip.

For cleaning, I would suggest something like deionized water or a 50/50 mixture with IPA  for the first few passed, as the electrolyte tends to be better soluble in water than IPA. It's also no problem to use plenty of water. Cleaning with pure IPA (no need to be super water free at first) is than a second step.
Just spaying with IPA is likely not enough. I would plan for a gallon of water and maybe 1 L of IPA. The spray may be for the last tweaks or removing flux for single parts changed later.

[macboy]

There is at least one capacitor and rectifier that look bad too. To blow the fuses it takes quite a lot of leakage. Also with non working caps some other parts could be damaged easily, especially if the power supply delivered too much voltage  - worst case this could still be a unit beyond repair.

For the sensitive parts, it only takes minute leakage (e.g. pA range) to degrade the performance. One can not even see the small amount of electrolyte to cause this. Especially cleaning under SMD parts works best with unsoldering them. The board is way to large to take the chance hunting for a droplet left under a chip.

For cleaning, I would suggest something like deionized water or a 50/50 mixture with IPA  for the first few passed, as the electrolyte tends to be better soluble in water than IPA. It's also no problem to use plenty of water. Cleaning with pure IPA (no need to be super water free at first) is than a second step.
Just spaying with IPA is likely not enough. I would plan for a gallon of water and maybe 1 L of IPA. The spray may be for the last tweaks or removing flux for single parts changed later.

I agree with everything here, especially the cleaning process: distilled water, then distilled water + IPA, then cheap 99% IPA, then if you want, the 99.9% tech grade stuff. Don't waste that on a first pass. Tip: at least in Canada, Costco sells four 500 mL bottles of 99% IPA for <$10. Get a big pack of toothbrushes to use with it while there 

I'd also probably be inclined to remove all components within at least 1 inch of all capacitors, and 1 inch of any visible electrolyte. If you see any evidence that the electrolyte started to migrate under the solder mask (i.e. the mask is damaged above traces), then you might need to go much further.

I have four of these. One had bad caps that did not leak but had high ESR causing power supply failure. I was lucky.

[beuford]

Thank you all for your suggestions.

While I appreciate all of the input, the reality is that the device is not worth the effort demanded by your recommendations. In addition, I don't have the resources/equipment to to do extensive surface mount removal/re-installation (nor the inclination).

That being said, the boards were originally washed with water after manufacture to remove flux/etc., which means the flux and other contaminants were ionic in nature (water soluble), as is supposedly the leaked capacitor electrolyte. Under the circumstances, washing thoroughly with water and properly drying the board should provide acceptable results.

I'll follow the recommendation to use increasing concentrations of IPA if needed to remove the electrolyte, but so far it does seem to be water-soluble.  I'll also use low-pressure dry air to remove as much contaminant as possible from hidden spaces, dry the board in a drying chamber, and inspect the board for damage but in the end that is all that is reasonable given the cost/value of the meter.

I have high confidence that the board will survive. I'll let you know in a week or two when I'm done. Otherwise, someone will acquire a new source of parts...

Beuford

[MadTux]

Why do you want to clean it before recapping? Caps will soon leak new electrolyte.

 Standard procedure: Remove all caps, clean with IPA, Methanol, deion H2O....., inspection for bad PCB, solder back new caps, clean flux, test and find remaining problems.

[beuford]

Why do you want to clean it before recapping? Caps will soon leak new electrolyte.

 Standard procedure: Remove all caps, clean with IPA, Methanol, deion H2O....., inspection for bad PCB, solder back new caps, clean flux, test and find remaining problems.

That's what I'm going to do (except I'm going to try water 1st, then mixtures of IPA). Sorry if I was not clear before.

Caps shipped from Mouser today.

Beuford

[Kleinstein]

Before powering on, one should test a few more parts: One cap at the power input side looks bad. So does one of the rectifiers. It might also be a good idea to check the voltage regulators for a dead short.

There are some parts where it is likely easier to remove them for cleaning.  With suitable tools / experience this would be quite a lot of the board. At least the power supply section (not much left one the caps are gone) should be checked for broken lines / vias. Here the highest level of corrosion is expected, and a broken line can cause damaging high output voltages. Once you see the damage one could still decide to remove more parts.

When cleaning the board still populated, it might be a good idea to avoid dirty water flowing toward still clean / sensitive  areas. At least the first passes should be more in a careful way with sponge and brush on an inclined board, not a full bath.

[beuford]

When cleaning the board still populated, it might be a good idea to avoid dirty water flowing toward still clean / sensitive  areas. At least the first passes should be more in a careful way with sponge and brush on an inclined board, not a full bath.

That's what I did. Had to resort to IPA spray in the end, though.

Board components/solder/some traces do show corrosion, though. Also, cleaning process removed some markings (not by dissolving; the just peeled off of the board). I'll check continuity for traces where possible, but difficult to determine if components are affected.

Oh, well; it is what it is.

Beuford