reforming electrolytic capacitors

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Think of a 3300 uF reservoir cap measuring only about 100 uF. The capacitors that were mounted close to the heatsink had degraded most.These caps were significantly younger than 40 years, but they all required replacement.
I agree that heat is the ennemy for electrolytics. Some degrade much faster than other, it depends from the manufacturer. In case of doubt, a conservative measure is to replace electrolytics older than 20 years.
 
and the good ones are also extremely expensive
I laugh only when i read that they do not form or reform but it is just the ear that adapts to the sound
I am listening to a little nad c352
It was left unused for 6 months
At the beginning I was very perplexed about the sound ... after some 100hrs it is quite better
The only possible cause are the caps ... not any other components
These electrolytic caps are a nightmare
They change with time like no other component
Regards,
gino
 
An electrolytic will generally reform in under 24 hours or not at all. 100hrs changes are more likely to be your ears/brain getting used to the sound. There may be some minor changes in leakage over a longer period but these should not affect sound. If leakage is an issue then an electrolytic should not be used at that point in the circuit.

6 months powered off will require a few minutes reforming at most.
 
Last edited:
They change with time like no other component
I could name a few counterexamples: beloved tubes for one, carb comp resistors, paper capacitors (the dodgy ones) mylar capacitors (the ones that copied the dodgy metalization processes from the dodgy paper caps).... etc

In short, I'd trust better a well used and well (over)heated Siemens axial E-cap from the early seventies than many other components, older, younger or contemporary: even Ge transistors of that time can have unpleasant surprises in store
 
Just today I started refurbishing a tube power supply. Actually, two identical units. First, I took the capacitors out and measured them. To my surprise, the reservoir caps of one unit all measured 55uF per section, where the can says 50 uF. And they pass the leakage test. These caps are 60 years old and well used! With the other unit, I was not so lucky, I had to replace all electrolytics, because they degraded. One section even seemed to have vanished entirely (measuring only 300 nF).

The units are full of carbon-comp resistors. Many of them are out of spec and have to go. Luckily, they are all very accessible. The last components that give me a headache are the selenium rectifiers. I am suspicious that they have a bit more voltage drop than they should, and that they will overheat in time due to this. When that happens, the result will be very unpleasant.

All in all, the result will be a set of almost fully rebuilt power supplies. I had hoped to get away with it a bit cheaper, but then I can not ensure safety and reliability.
 
Last edited:
An electrolytic will generally reform in under 24 hours or not at all.
100hrs changes are more likely to be your ears/brain getting used to the sound.
There may be some minor changes in leakage over a longer period but these should not affect sound.
If leakage is an issue then an electrolytic should not be used at that point in the circuit.
6 months powered off will require a few minutes reforming at most.

Thank you very much for your valuable reply
I did not know about the 24 hours only. This is nice (I do not like to wait)
I strongly think i will replace those two caps with something new
Thanks again
Kind regards,
gino
 
I could name a few counterexamples: beloved tubes for one, carb comp resistors, paper capacitors (the dodgy ones) mylar capacitors (the ones that copied the dodgy metalization processes from the dodgy paper caps).... etc
In short, I'd trust better a well used and well (over)heated Siemens axial E-cap from the early seventies than many other components, older, younger or contemporary: even Ge transistors of that time can have unpleasant surprises in store

Hi ! the problem with aging is that caharacteristics are time dependant
I do not feel comfortable with a sound changing with time passing
I find it disorienting
Unfortunately caps are caps :eek:
thanks and regards,
gino
 
Hello,
I have some caps out of a Hafler 9505 that are impossible to find new. Does anyone know of this reforming service here in the USA? I would like a pro to do this service or KaBoom!!! 500C223U100CJ2B 3 5/8" tall to fit the top cover like new, no mods.
 
Last edited:
You can re-form a cap pretty easily. The basic idea is to drive the cap at its rated DC voltage through a somewhat high valued resistor, perhaps 10KΩ to 100kΩ, depending on the size of the cap. A resistor and a bench supply are all you need.

However, re-forming will not replace missing electrolyte, or undo corrosion caused by chlorine ions and the like. Re-forming is great for brand new caps that have sat on the shelf for a long time, but it does not cure all possible ills.
 
Just today I started refurbishing a tube power supply. Actually, two identical units. First, I took the capacitors out and measured them. To my surprise, the reservoir caps of one unit all measured 55uF per section, where the can says 50 uF. And they pass the leakage test. These caps are 60 years old and well used! With the other unit, I was not so lucky, I had to replace all electrolytics, because they degraded. One section even seemed to have vanished entirely (measuring only 300 nF).
.

Good caps will last a very long time if they are powered up frequently, leave them fallow for a few years and all bets are off. I have a Heathkit 0-400v power supply I built back in 1963 that still works fine and all those electrolytics are the originals.

I would replace any selenium rectifier i came across, they never were very good. Just becarful the reduced voltage drop of a silicon rectifier doesn't have unintended consequences.
 
Well, I recently got my hands on three NOS orange colored Siemens screw type capacitors manufactured in 1977 (10 000uF, 100V). They were never used, no signs of screw marks and I took them from their original box.
They needed a thorough reformation, I kept them for 72 hours at limited current in order to get acceptably low leakage current.
Now they measure and perform excellent, they have an ESR around 15mOhms and the capacitance is within specs.
 
Never used and stored as specified, electrolytics will have a very long shelf life.
But will NEED to be reformed slowly to achieve their specified parameters.

I am not surprised that high quality 40 year old electrolytics still meet specification after reforming.

BTW
what was the leakage current down to?
I found the Leakage <0.02CV to 0.03CV can be bettered by at least 100fold and some achieve ~1000fold reduction.
 
I use a Radiometer IM-6 insulation tester for reforming electrolytic capacitors. This instrument can give 1 to 1000V in 1V steps, and can measure pA range leakage or Teraohms. If the capacitor has 10Mohm under nominal voltage, I stop formatting. I increase the Voltage in 10V steps, wait until read 1Mohm before increase by 10V. Sometimes it takes days. Better than applying full nominal voltage right away.
 
I mess around repairing and restoring vintage valve radios and high power amps here in the UK and never use a variac, preferring to follow the practices used by those who have been doing it longer than me. That is, you hit them with full voltage straight off and see what happens. About 80% of the sets I power up seem either to be ok with the original power supply electrolytics left in, or else they fail within an hour of use. This is of course with high-voltage types (250 volts and more and I've never actually measured the capacitance. (I dont have a meter for that either). Just my experience with dozens of the dusty old relics. Steve. :drink:

This has been my experience too.
I own an Eico capacitor checker/reformer and while it does work
the capacitor is never like new, shorting them was what convinced me.
Even after the capacitor has reformed when shorted the spark is anemic compared to a fresh cap.
All that said I still use my variarc to power up old stuff but I am just doing slow surge and hoping the light bulb goes off by full voltage.
 
stevekendal said:
I mess around repairing and restoring vintage valve radios and high power amps here in the UK and never use a variac, preferring to follow the practices used by those who have been doing it longer than me. That is, you hit them with full voltage straight off and see what happens. About 80% of the sets I power up seem either to be ok with the original power supply electrolytics left in, or else they fail within an hour of use.
He should have said "those who have been doing it longer than me yet still haven't learnt how to do it properly". Hitting an old cap with full voltage will certainly test whether it needs reforming. Unfortunately it can be a destructive test: those which needed reforming will fail (possibly explosively) even though most of them could have been reformed if this was done properly.

Note that the correct method does not require a variac, but a suitable resistor and a voltmeter. This is because it is current which needs to be controlled, not voltage.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.