Keithley 2000 re-cap & Digi-Key selling new old stock (WTH?!)

[Chernobyl]

Today I re-capped my 1996-vintage Keithley 2000 DMM as a precautionary measure (all the electrolytics were OEM). I was a bit surprised to find them all in pristine and near-new condition. There was zero electrolyte leakage, the capacitance values were bang on, and the ESR for each was less than 0.1 ohm. But I didn't want to push my luck so I replaced them proactively with hi-temp (105 degC) and low impedance Nichicon caps from Digi-Key. When I received my order earlier this week I was a bit surprised to see the date code on one cap (Nichicon UPA series) was October 2013! It even came in a bag that had an old label from 2018 (perhaps packaged for another customer and then returned?). More concerningly, there was a small amount of greenish corrosion on the very base of the leads where they connect to the cap internals (electrolyte leakage?). I cleaned the corrosion using IPA, installed the cap, and it seems to be fine -- but I can't help wonder if I should have just left the old unit in place. Probably not, given that it was 26 years old, but the "new" one Digi-Key sold me is already almost 9 years old and had some corrosion. Not very inspiring.

Has anyone else had a similar experience?

Pix
1. Original caps before removal:
2. Corrosion on "new" cap from Digi-Key:
3. Date code on "new" cap:
4. Old label on the shipping bag (new label on the other side):
5. Original caps after removal (note zero electrolyte leakage):

[Zucca]

I would call the Digikey customer center and ask for at least a refund.
Send them also a link to this thread...

Muppets.

[Someone]

Or send them back to their own statement:
https://forum.digikey.com/t/shelf-life-of-electrolytic-capacitors/7946

The current aluminum electrolytic capacitors shelf life is approximately 2 years.

That said, for orders of less than manufacturers standard qty you cant really expect fresh stock. Be happy they sell in singles at all.

[Chernobyl]

I would call the Digikey customer center and ask for at least a refund.

Meh...for a $2.50 part? Not worth it. I was mostly just curious about other folks' experiences.

Be happy they sell in singles at all.

That's absolutely a fair statement. Even after all these years of buying from them it still amazes me that I could order a single component and have it in my hands the next day (from halfway across the continent and crossing the US-CAN border).

[BILLPOD]

Good Morning Chernobyl,  I too have a Keithley 2000 that I got off Ebay in 2018.   And mine has also a 'born-on' date of 1996,
based on serial #.   I also have been pondering on re-capping,
but it has a calibration sticker of 2017, which I didn't want to break, since I thought it might add some value if I wanted to sell it.   I have since fallen in love with it and will  probably keep it
until my dirt nap.    I take solace in that the caps in your unit were in such good shape, so I am not going to break my cal. sticker. Thank you for sharing your experience.  Take care, my
GWN Friend.

[mikeselectricstuff]

if it ain't broke...
if kept cool, electrolytics last for decades. Unless it's a piece of kit with known cap issues, I can see no benefit to recapping.

[Hydron]

Mike - the 200x Keithleys have a rep for catastrophic electrolytic leakage, hence the OP's pre-emptive recapping.

As for electrolytic shelf life, I normally try and play it safe with the soup-can sized ones and reform them before use, but #yolo with the baby ones, never had any fail after being on the shelf for years.

[precaud]

Yeah, I've received lytics with old date codes several times, from both Digikey and the Mouse. The more oddball the value/voltage, the more likely to encounter it. Only in one instance (2.2uF@350V) did the caps measure out of tolerance, with very high ESR.

[Per Hansson]

I was mostly just curious about other folks' experiences.

There is a lengthy thread on the Badcaps forums about this, no real conclusion about what the issue is though:
https://www.badcaps.net/forum/showthread.php?t=82018

[Chernobyl]

There is a lengthy thread on the Badcaps forums about this, no real conclusion about what the issue is though

That's exactly the kind of feedback I was looking for - much appreciated.

[ps]

Today I re-capped my 1996-vintage Keithley 2000 DMM as a precautionary measure (all the electrolytics were OEM). I was a bit surprised to find them all in pristine and near-new condition. There was zero electrolyte leakage, the capacitance values were bang on, and the ESR for each was less than 0.1 ohm.

It was the same with my Keythley 2000. The electrolytic capacitors were like new, both from optical appearance and from electrical performance. It seems that Keythley once installed a poor batch of capacitors, which led to their bad reputation, but in general their components are not that bad.

When I received my order earlier this week I was a bit surprised to see the date code on one cap (Nichicon UPA series) was October 2013! It even came in a bag that had an old label from 2018 (...)

Usually, the data sheet of the component includes an specification of the shelf life. E.g. 2000 hours at 105°C, which corresponds to 15 years at 40°C. This is in addition to the specified service life (e.g. another 5000 hours at 105°C or 2000 hours at 125°C). This means that nine years is not what we would expect from a major distributor, but provided the components have been stored properly, this is not really a problem.

[voltsandjolts]

if it ain't broke...
if kept cool, electrolytics last for decades. Unless it's a piece of kit with known cap issues, I can see no benefit to recapping.

I recapped my trusty decades-old K2000, although my impression is that other models in the 2000 series are more susceptible to cap leakage.
IMHO the benefit is (a low cost) peace of mind that the instrument is safe guarded for the next couple of decades.
A leak may not be noticed until the pcb is badly damaged.

[thm_w]

https://www.eevblog.com/forum/chat/digikey-sold-me-8-years-old-electrolytic-caps!/

[taste_tester]

I stopped ordering from
Digikey because I have received MANY very out of date
parts in the last few years out of not that many orders.
The only time in my life I ever bought "new" rheostats
from a good brand (Panasonic) showed up DOA from Digikey.
Once upon a time I actually used to prefer them to Mouser.
Now I just simply won't order from them unless they have something I can't get elsewhere.
But nowadays if Mouser doesn't have what I need, I'll probably be trying Jameco and TME before Digikey.

if it ain't broke...
if kept cool, electrolytics last for decades. Unless it's a piece of kit with known cap issues, I can see no benefit to recapping.

I recapped my trusty decades-old K2000, although my impression is that other models in the 2000 series are more susceptible to cap leakage.
IMHO the benefit is (a low cost) peace of mind that the instrument is safe guarded for the next couple of decades.
A leak may not be noticed until the pcb is badly damaged.

"if kept cool, electrolytics last...." Yeah well tell that to all the folks who brought me a DVD player, Game Console or Set Top Box which cooked itself to death in an enclosed TV stand with no ventilation and the customer refuses to replace said TV stand? Or keep a fully loaded rack powered on 24/7 with no spacing between the rack units to let them air out.  Yes I agree that if kept cool (and the caps used are properly selected for their application) they last.

I emphasize this part of the quote.
As someone who has replaced more failing or aging capacitors than could count, I have seen it all with capacitors...

-ones that pissed out and electrolytic corrosion ate the PCB, (PCB looks fine from top, but flip over and its all burned and black gunk)
-ones that looked like the top was about to burst, yet still worked to some degree (always fun to see people bring in something like "i keep cycling power until it works"
-ones where the top DID burst and leave scorch marks against the chassis
-ones that looked fine until removed and were found to be bulging on bottom and testing bad
-ones that test fine in terms of capacity out of circuit, but under load capacitance tanks (testing at multiple frequencies is very helpful here)
-ones that seem in spec "enough" but for whatever reason are too unstable on the power rail for sensitive electronics to work correctly (AC ripple for instance)
-ones that looked fine but don't even measure as a capacitor; LCR meter thinks its a resistor(!!!) - this was from a 35 year old switch mode supply

If you only repair your own things, not broken things other people bring you, I would also be more skeptical of recapping when it is not known to be the needed solution.  Something important to consider is diagnosing whether the capacitor is the cause of the failure or whether it failed due to another part of the circuit going bad.

Another thing to consider is many times people will allow something to start "acting funny" then wait for it to fail to repair it. I think the repair and diagnosis should have began when it was acting funny and not fully failed. After all if you put in the same effort to test the individual components out of circuit then you will find the problem component anyway before it totally failed.

On my own possessions, I will do preventative full recap of the crucial components, usually on the power section. I am frequently repairing items which the capacitors are clearly already failed and often times they are no name caps or not from reputable brands. I have seen and been brought too many things where preventative maintenance would have actually prevented the failure. Now at this point I simply don't wait for them to die first.