Reforming Electrolytic Capacitors?

[rock soderstrom]

Hi guys, I have a bag full of axial capacitors 25uF/70V which unfortunately all show between 33-50uf capacitance. ESR is between 0.3-1 ohm, Vloss between 1 to 2%. All measured with my cheap Chinese component tester.

What should I do? Throw the caps in the trash can or reform them?

How do you do this and what are your experiences?

 

[rock soderstrom]

ok, I found a thread about exactly this topic.

https://groupdiy.com/threads/reforming-electrolytic-capacitors.64581/
Whats your opinion? Reform or trash bin?

 

[trobbins]

Do you just throw parts/equipment in the bin because you can't be bothered to make any effort?
Imho, check a sample batch for leakage current, and parameter spread.

 

[JohnRoberts]

I would bin them, but it might be educational to try to reform them...

Slowly with a current limiting resistance in series..

Budget some time.. measure your results

Then bin them...

JR

 

[Ciro Marzio]

In the past years usual tolerance in elyt caps was -0+70%.
If they do not show significant leakeage, IMHO they could be worth using.
ROE caps are often sought after for their rich and mellow tone

 

[rock soderstrom]

Do you just throw parts/equipment in the bin because you can't be bothered to make any effort?

Huh, what are you talking about? I asked a question in an interational forum in a language I barely speak, at the same time I'm researching the topic and building an experimental setup to randomly test capacitors that are probably over 40 years old. This does not look like "you can't be bothered to make any effort?" to me.
Had a bad day?

I would bin them, but it might be educational to try to reform them...

Slowly with a current limiting resistance in series..

Budget some time.. measure your results

Thanks for feedback.
That's exactly what I'm doing right now, very interesting "scientific" project , my measurement results so far raise some new questions.

From the point of view of a professional developer, the whole thing understandably makes no sense. As a manufacturer I would not think about it for a second. The risk of serious follow-up costs and loss of reputation cannot be justified.

For me as a curious do-it-yourselfer, it's a different story. A strangely exciting thing to waste some of my time.

ROE caps are often sought after for their rich and mellow tone

They also look very nice!. They are still "made in West Germany".
Long time ago but I still remember very well.

Then bin them...

I don't want to publish the final results of my "research" yet, but that's what it looks like at the moment....

 

[trobbins]

Huh, what are you talking about? I asked a question in an interational forum in a language I barely speak, at the same time I'm researching the topic and building an experimental setup to randomly test capacitors that are probably over 40 years old. This does not look like "you can't be bothered to make any effort?" to me.

Your first post suggested you would be happy to put them in the bin - presumably if others said to bin them.

Cap detail was pretty much just a photo - no assessment of maker or model or age or history of 'the bag', or part quantity or physical size. Have you checked if equivalent rated modern parts are much smaller and have better specs like ripple current? That's the lack of effort I was referring to.

I suggested you test a sample batch for leakage current, as your other testing does not show any abnormal values that would immediately suggest 'the bin'. If you sample a statistically significant number, and all have a leakage below a level comparable to a modern cap (or an age equivalent spec if you can identify such), then that would provide confidence that the caps are 'ok'. That is by default 'reforming', but imho reforming is not the technical outcome you should be focussing on per se.

Not all e-caps are born the same, and for many applications not all NOS e-caps should be binned once they get to 20 or 30 years old - certainly caution is required.

 

[rock soderstrom]

Your first post suggested you would be happy to put them in the bin - presumably if others said to bin them.

Nonsense, reread my first post and you will see that I am just asking the forum a question. I am not happy to bin a bag full of good looking caps at all.

Cap detail was pretty much just a photo - no assessment of maker or model or age or history of 'the bag', or part quantity or physical size

Yeah, so you got the same Info as me. A bag full of 25uf/70V caps, build by Roederstein in West Germany. What's your problem?

Have you checked if equivalent rated modern parts are much smaller and have better specs like ripple current? That's the lack of effort I was referring to

Again nonsense, modern axial caps are not much smaller, show me one proof.
I'm here for fun on the forum, stop harassing me or I'll get in a bad mood.

I suggested you test a sample batch for leakage current, as your other testing does not show any abnormal values that would immediately suggest 'the bin'. If you sample a statistically significant number, and all have a leakage below a level comparable to a modern cap (or an age equivalent spec if you can identify such), then that would provide confidence that the caps are 'ok'. That is by default 'reforming', but imho reforming is not the technical outcome you should be focussing on per se.

Not all e-caps are born the same, and for many applications not all NOS e-caps should be binned once they get to 20 or 30 years old - certainly caution is required.

You have no idea what I'm doing here, just stop it. I'm not in the mood for your lecturing chatter.

 

[rock soderstrom]

Okay, let's move on.

I worked on several of the capacitors with the procedure suggested by JR and got the following results:

At first it ran well, after a few hours the capacity dropped towards the nominal value, on average about 30uf.
I monitored the whole process with current and voltage meters. Everything was as expected.

The problem was, after a short time the capacity rose again slowly continuously to the first measured value.

Ok, it seems the caps need to be worked on even longer. Lets do it.

I have then reformed the caps overnight further.

The result was disastrous.

More than half of the capacitors excreted an aggressive liquid and had no measurable capacity anymore.
There was an amazing amount of liquid coming out of the capacitors, I was really surprised.



You can't really see it in this photo, but the second photo shows how large parts of the electrolyte clotted along the measuring cable onto my old laptop. Here the result.
Ouch, learning through pain!


The white trace shows where the cable with the electrolyte touched the laptop.

Conclusion: My short career as a researcher is over, these caps are going in the trash.

 

[JohnRoberts]

Out of curiosity what value voltage, and what value current limit resistance did you use?

Good to learn that they were incontinent before using them in a circuit (sorry abut the laptop).

JR

 

[rock soderstrom]

Out of curiosity what value voltage, and what value current limit resistance did you use?

33k/ 2Watt resistor with 30V from my bench power supply.

The charging current was inconspicuously small and the capacitors did not get warm either. Everything went as described elsewhere on the WWW.(until the moment when the electrolyte leaked out.)

Good to learn that they were incontinent before using them in a circuit (sorry abut the laptop)

Yep, indeed a good thing. This electrolyte would destroy a PCB within a short time. I am still surprised how much liquid leaked out of these small capacitors!

 

[JohnRoberts]

1mA of maximum inrush current, and much less if the capcitors charged up to anywhere near 30V should not cause enough heating to boil off electrolyte.

Were the capacitors moved from a cold ambient into a warmer environment?

Modern caps use a rubber bung (end plug) typically with a pressure relief. That old school cap looks like the end is phenolic with the metal case crimped shut around it.

I would have taken one apart to dissect by now (over the decades I have dissected many electrolytic caps to troubleshoot). You should be able to cut back the crimped end and bend it open. The guts should then pull right out.

===

Or just bin them and wash your hands.

JR

 

[cyrano]

I've always reformed caps with minutes of charging and tens of minutes discharging. Only, I don't know why. Someone somewhere told me that was the "procedure". And obviously that's what I went with.

I guess the analogy with lead/acid batteries somehow convinced me?

 

[Whoops]

What should I do? Throw the caps in the trash can or reform them?

How do you do this and what are your experiences?

I totally understand your question and situation, I also had the same dilemma in the past.
I had quite a nice stock of NOS axial Phillips caps, the same type as used in old Neve modules.

Actually the thread linked here on reforming, was done by me at the time I had the dilemma.
https://groupdiy.com/threads/reforming-electrolytic-capacitors.64581/
At first as a learning exercise I still reformed a few caps. It took a long time for each cap and I did a small setup for it, but I had many many caps to go...

Then... I don't mind using vintage/old components but to the exception of Electrolytic caps.
It's my personal opinion that modern Electrolytic caps are miles and miles better than anything that was done in the past, so after reforming a few caps I was still not confident enough to use something that old when I could buy for cheap a Panasonic cap that will make me much more confident about it's performance and reliability.
Not having confidence in a component makes it being not useful to me, reformed or not. This was my final conclusion

So I didn't Bin them, but instead I sold all of them for half the price here in the Black Market to people that feel they are useful to them and I was pretty happy to get rid of them, get rid of the dilemma and have some extra bucks

My opinion is still that Vintage Electrolytic caps are crap

 

[jokeramik]

IMHO Date of production is week 25 in 1970. They are over 50 years old. I don´t use such old components for my projects. There is no guarantee for long life working performance or reliability.
Best regards!
jokeramik

 

[rock soderstrom]

Actually the thread linked here on reforming, was done by me at the time I had the dilemma.

Yes, I saw your thread immediately after creating this thread and that actually cleared up all the technical questions.

Not having confidence in a component makes it being not useful to me, reformed or not. This was my final conclusion

I think confidence is the key word. Of course, I knew the problems of old electrolytic caps for a long time, but on the other hand, you read on the Internet again and again about successful reformation of old capacitors. This and the technical data of the described capacitors have made me curious. These caps would fit wonderfully as cathode capacitors in many DIY guitar amps.

After my short excursion into "vintage capacitor research", I am cured by the results. I definitely have no confidence in these capacitors and would never install them from now on.

Thanks for feedback.

IMHO Date of production is week 25 in 1970. They are over 50 years old. I don´t use such old components for my projects. There is no guarantee for long life working performance or reliability.

They are old, over 50 years is possible (but the 25/70 means 25uF/70V). I think they are from the 80ies. I totally agree with you for reasons now known.

Or just bin them and wash your hands.

That is exactly what I did

 

[trobbins]

IMHO Date of production is week 25 in 1970.

Is there a datecode system for that vintage of Roederstein e-caps? The 'HSF' seems to be the product series, based on an ebay item (5x Elko Condenser of Roe Series Eb, 4.7 μF, 100 V audio capacitor, NOS | eBay)

It looks like a 'B' marking next to the ROE label on one of the caps, but otherwise there is not much to go on.

 

[jokeramik]

Formatting electrolytics is useful for high voltage types e.g. for tube amplifiersand for longer non-used large capacitors e.g. slit-foils.

Best regards!
jokeramik

 

[jokeramik]

They are old, over 50 years is possible (but the 25/70 means 25uF/70V). I think they are from the 80ies. I totally agree with you for reasons now known.

Sorry for my mistake...
Best regards!
jokeramik

 

[leadbreath]

Naahh bru stick them in an emvelope and send they my way...
I will pm u my address