Keithley 2000 and the leaking caps

[max666]

I know this topic has come up before, but since I'm now the proud owner of a Keithley 2000, I'm worried if I should replace those caps:

Since you got the keithley dmm. IMMEDIATELY change all the caps in the power supply. I have seen at  least 5 keithley multimeters 2001 / 2002 with shorted out electrolytics. It fries the motherboard before the fuse pops.... I salvaged the lm199's so i got a few of those laying around.

I was just told by an EE friend i have that : "This is my best translation ..."

Nichicon apparently in 2002 bought some Electrolyte fluid from a "Scam company".
It was used in the HM & HN series of caps. They made the bad caps until 2004.

The problem with the bad electrolyte was that it became conducting over time ,
and when there was voltage applied it "rusted/destroyed" the foil with a bad result.
Sometimes there went thermal runaway in the cap , followed by a bang.

He said they can look ok but still be bad ....

And if they get hot they're bad for sure.

Now the date codes on the nichicon caps are H0409, H0410 and H0452.
However my board

is different than this one, which has obviously wet itself (source: http://www.ebay.at/itm/Keithley-2000-Digital-Multimeter-Tischm ... e05)


more pictures (not sure if you guys prefer fully inserted pictures of linked thumbnails, here in the forum?):

[ManateeMafia]

I think that appearances are not everything. I have had to replace caps in one of my Keithley 2001 that had no outward damage. It was cooking the pcb underneath it though.

I still have a couple of the removed caps and they have H9417 date codes on them.

You can measure the power supply rails and determine if anything is out of tolerance. A scope would help too. I am on the fence for capping old gear. If it isn't broke.... However, Keithley seems to have more than their fair share of bad caps.

[jpb]

I don't know about the K2000 - I've just bought one but it is new (or newish) so I'm not worried about the capacitors.

But when I bought my Advantest off e-bay I changed all the capacitors simply as a precaution. I noticed in the manual that it recommended the these are changed every 5 years.

It turned out to be a more tedious job than I expected as the solder was old and I handn't yet learned the trick of mixing new solder with it. If you have the right tools it is a straight forward job and the cost of the capacitors is not much so it is probably worth doing just for peace of mind.

[Kjelt]

Free electron says the 2001/2002 not the 2000. Not sure if they are also affected? I got two of them, they seem to work fine.
Perhaps only some series in a limited time frame where they used the bad caps? Would be interesting to know what timeframe we are looking for and what the numbers on the caps that are affected are.

[max666]

So I measured the supply rails, they should all be +/-5%, which they are:
+5VD = 4,99 V
+37V = 36,24 V
+15V = 15,41 V
-15V = -15,73 V
+5VRL = 5,07 V

and the same in pictures:
+5VD


+37V


+15V (please excuse the crude differential probe  )


-15V (please excuse the crude differential probe  )


+5VRL (please excuse the crude differential probe  )


And the caps don't get warm, so I guess they are good.

The 2000 series was definitely affected, as proven by the ebay example I linked. But perhaps my unit is new enough and never got the bad caps.
If we weren't talking about a precision measurement gear here, I would whack in new caps no questions asked. But I'm very reluctant on doing it on a 6 1/2 digit meter, which is why I'm asking if I should. (and of course if I don't do it and later on my unit pees itself, I have someone to blame  just kidding)

[tautech]

Not that I own a Rigol, but on your images I can only see DC coupling.
If you want to clearly see/measure ripple you must to use AC coupling.
The on screen measurements of P-P for 5V 280mv and 37V 5.6V seem excessive.

[max666]

Not that I own a Rigol, but on your images I can only see DC coupling.
If you want to clearly see/measure ripple you must to use AC coupling.
The on screen measurements of P-P for 5V 280mv and 37V 5.6V seem excessive.

Fair enough. Let's see if I can do that better. Here's everything AC-coupled:

digital supplies, no BW limit (ground spring connection):


digital supplies, with 20 MHz BW limit (ground spring connection):


analogue supplies, no BW limit (ground leads, all three connected to TP):


analogue supplies, with 20 MHz BW limit (ground leads, all three connected to TP):


As far as I can tell, the digital +37V rail is only used for the display/key board, so I'm not actually worried about the ripple on there, although it's quite prominent.
Regulators are all around 45-55 °C and the caps around 35 °C (with open case, 25 °C ambient).

So what do you guys say, is that looking good? Would you replace the caps?

[tautech]

By your OP images it seems it has a SMPS.
Your last screen shots are shown with a 10 mS timebase setting that might be appropriate for a linear PSU but not for SMPS. Speed up the timebase.
Probably to the 20-50 KHz range.

If the ripple is at SMPS frequency then one would suspect the E-caps are not fully doing their job.
What is the Manual spec for PSU ripple?

[Kjelt]

I would have expected that Nichicon would have a list with bad caps somewhere and a free replacement program.
Anyway, I am going to replace them all preventively. Better safe then sorry and for 20$ total for 2 meters it is not such a big deal.