"Old" electrolytics

[joCCe]

Lots of useful information here. Thanks for sharing guys!

Joakim

[tribi9]

Would this work for the caps on a couple of forty something year old can caps? I picked a vintage Japanese tube head that 3 needs weird cap configurations. Could they be reformed this way?

[VacuumVoodoo]

Test 1
- remove from amp
- shake close to ears and listen closely
- if they make rattling sound like maracas they go to disposal bin.
Test 2
If no rattle I try to find original data sheet (sometimes a "Mission Impossible) and check specs for weight.
I use small laboratory scales. If the can in question weighs 10-15% less than spec it goes to disposal bin

If cap passes both tests I'll attempt to re-form it.
I do this only on really valuable vintage equipment when it's a restoration job, otherwise it's replacement with new caps without all this wild dancing and alchemist bs.

[Cliff Schecht]

Can caps tend to use that dry electrolytic that was referred to earlier. This isn't supposed to dry out over time as badly as normal e-caps do but it does lose the oxide layer that forms on the aluminum plates over time. This is what needs to be reformed, if the cap isn't dried out (they are sealed much better than old paper e-caps).

The only way to find out if the caps are good is to reform them while watching the leakage current (measure the current across the 100k limiting resistor). It should drop to the maybe 10-20uA range or less after reforming, if it doesn't go this low then it probably needs replacing.

[Cliff Schecht]

Here's a nice little snippet my dad wrote me on reforming caps a few years back. I've posted it once or twice before, it's great info from someone who knows electrolytic caps!

Here’s how to reform old caps (assuming they’re salvageable):



Take the unformed or old cap and measure its capacitance value with a Fluke DVM. Be CERTAIN that you haven’t charged the cap prior to doing this. It may read fairly close to its rated capacitance value but there’s a good chance that a lot of the dielectric has deformed and the cap will leak a lot of DC current when put into service.



Reforming the cap rebuilds the dielectric layer that the manufacturer established on the cap when it was new. Once reformed, the cap will not only have its rated capacitance but it will also leak very little.



To start: Put a resistor in series with the cap you’re going to reform. Choose the value of resistance to limit the short-circuit current to no more than a couple of milliamps (watch the resistor power rating as well, in case the cap is shorted and stays that way).



Next, hook the cap up to a variable power supply (minding polarity, of course) and slowly bring the voltage up to a value that’s a bit under the rated WVDC of the cap. IF you don’t have a variable high-voltage supply, you can hook the RC setup to a non-variable HV supply and just switch the thing on, although it’s preferable to apply the forming voltage slowly. Leave the cap hooked up to the HV supply for a few minutes and then check the leakage current flowing into the cap (measuring voltage across the limiting resistor and computing the current works fine). The cap’s leakage current shouldn’t be more than a couple of milliamps…some of the older ones may leak up to perhaps 10mA.



After the cap has sat under the charging supply for a while, disconnect the HV supply and then discharge the cap through a low enough resistance to run it down fairly quickly without making a lot of sparks while doing so. Don’t dead-short it to discharge it; it makes a lot of neat sparks and noise but the high surge current is really hard on the internal structure of those older caps. Then hook the cap up to the HV supply again (through the limiting resistor), leave it for a few minutes and then check the leakage current before discharging as above. Repeating this process a few times usually reforms just about any old cap that’s still in usable shape.



After the last forming run, discharge the cap one last time through a resistance and this time, place a shorting wire across the output for several minutes. A lot of electrolytics exhibit enough dielectric absorption to allow them to recharge themselves to significantly high voltages if left unshorted after being charged and that charge can fry your capacitance meter when you go to check the capacitance one last time.



So, after the cap has sat with the short across it for several minutes, remove the short and check the capacitance with your DVM. It should read within its tolerance limits. Since it’s been reformed, it won’t leak much either.



The term “dry” electrolytic is a bit of a misnomer. Modern, polymer dielectric electrolytic are truly dry in that sense but the older aluminum-foil and paper electrolytic all had some liquid in them. They were called “dry” since they didn’t have liquid sloshing about in them. Instead, the electrolyte was typically held in a paste form. If properly stored, some pretty old caps are still usable today. It’s easy enough to check the seals to see if they’ve broken or become embrittled and a capacitance/DC leakage test will also tell you whether they’ve survived or not.



If the electrolytic caps you have were used, then it’s probably best not to try to salvage them. The heat they endured for the years they were in service in tube equipment was usually enough to compromise their integrity or even kill them. NOS parts are worth trying to reform unless they’re really old (WWII or earlier).

[martin manning]

Nice clear discussion from your dad, Cliff; thanks for posting it. This thread should be a sticky!