Old electrolytic caps will fail on ESR long before they fall out of spec on capacitance. I wouldn't bother testing old caps unless you include an ESR test.

Not on capacitors that haven't been used for 36 years. The caustic electrolyte very slowly etches away the oxide layer dielectric - capacitance increases as the dielectric gets thinner, but the thinner layer has lower breakdown voltage which is the potential problem that is concerning the TS.

The brand mentioned probably had pretty good ESR for those days, and won't have deteriorated as much as many cheaper brands.

A dodgy incomplete electrolyte formula stolen and shared among various Chinese manufacturers gave everyone a perception of ESR risks that doesn't apply to all makes.

 

Good brands can last decades but only at room temperatures at 30C the life expectation is already degrading while in a chilly warehouse they will be fresh after 40 years

 

Not on capacitors that haven't been used for 36 years.

I may have misread his intention. I thought he was referring to other capacitors he had. I already weighed in that I would have zero concern about his unused PSU caps.

 

Make sure it isn't getting hot - the dielectric is an oxide layer formed on one of the plates, the caustic electrolyte very slowly etches it away. Leakage current in normal use is what maintains the thickness of the oxide layer. The thin oxide in a neglected capacitor can result in a dangerous amount of leakage current before you even reach the rated voltage - enough leakage can cause temperature and a head of steam that bursts the can.

If you need to re form the oxide; you need a resistor that limits the current under what causes heating and bring it up to rated voltage over about 24h.

I'm worried about the leakage current. Looks like I need to measure it properly. As an initial test I reformed the cap and then charged it to max voltage of 24v(cap says 25V and max surge of 35V) and kep meeasuring it for sometime, it did hold the charge for quite long and dropped only gradually.

 

Just another thought... Remember, there are an awful lot of vintage electronics from that era still working away. As mentioned, check for heat, then know if you see a bulged can or smell a fishy odor, an electrolytic is probably giving up the ghost.

If caps simply went bad after 30 years, I'd be plumb out of test equipment!

Guess I'm lucky there bro. The cap seems to be in perfect condition from outside.

 

Good brands can last decades but only at room temperatures at 30C the life expectation is already degrading while in a chilly warehouse they will be fresh after 40 years

This was kinda stored mostly in an air conditioned store roome throughout.

 

As said above, capacitance should stay constant because the cap has the same number of square inches of metalized foil all wound up in a coil.

The issues is that the caustic electrolyte can dissolve more and more if the metal over time. Even if unused. This makes the metal foil thin. Thin foil causes two problems. 1) the resistance back to the leads increases and 2) related to (1), the max heat dissipation goes down.

The good news is that applying DC voltage can reverse (redeposit) metal. And you device is a DC power supply with caps that are huge and do not see a lot of voltage change, even at full load.

I have an HP power supply built in 1975 (42-years old) and the caps are fine. I use it every time I turn on the main power at my workbench.

 

If they have an oscilloscope and function generator at their disposal, yes. I had to rely on a more complicated circuit. Same basic idea though.

As said above, capacitance should stay constant because the cap has the same number of square inches of metalized foil all wound up in a coil.

The issues is that the caustic electrolyte can dissolve more and more if the metal over time. Even if unused. This makes the metal foil thin. Thin foil causes two problems. 1) the resistance back to the leads increases and 2) related to (1), the max heat dissipation goes down.

The good news is that applying DC voltage can reverse (redeposit) metal. And you device is a DC power supply with caps that are huge and do not see a lot of voltage change, even at full load.

I have an HP power supply built in 1975 (42-years old) and the caps are fine. I use it every time I turn on the main power at my workbench.

Wow. 42 years! That great. So , there is hope for me that my cap maybe good too.

 

I'm worried about the leakage current. Looks like I need to measure it properly. As an initial test I reformed the cap and then charged it to max voltage of 24v(cap says 25V and max surge of 35V) and kep meeasuring it for sometime, it did hold the charge for quite long and dropped only gradually.

A long stored aluminium electrolytic could have thinned oxide layer. The worst case scenario is that it could break down at less than rated voltage and go dead short.

Leakage current will certainly start to rise at a lower voltage and could overheat the capacitor - but only leakage current can re form the oxide layer.

The usual remedy is to bring it up to rated voltage over about 24h, a series resistor limits the leakage current to a safe value - if you've already powered it up and it didn't go bang; you're probably in the clear.

 

"The usual remedy is to bring it up to rated voltage over about 24h, a series resistor limits the leakage current to a safe value - if you've already powered it up and it didn't go bang; you're probably in the clear.

I did try reforming the plates like you mentioned above and only then did I screw it back to the PSU. The voltage across the cap is usually 17.5V. So,I hope its working fine then.As a test I did try charging an old SLA battery with it yesterday at 7.7V, 1.2A and it was working fine.