Author Topic: Reforming aluminum electrolytic capacitors  (Read 1677 times)

0 Members and 1 Guest are viewing this topic.

Offline innkeeper

  • Supporter
  • ****
  • Posts: 567
  • Country: us
Reforming aluminum electrolytic capacitors
« on: October 16, 2017, 11:43:56 pm »
Looking for comments or improvement recommendations on this... i've been reforming some caps recently and came down to some simple guidelines which have been working well for me.

Maybe this topic exists, but couln't find one dedicated to this topic, just random mentions of reforming.
I became interested in the topic as i acquired quite a cache of old electrolytic i want to make use of.
I've read a few documents on reforming caps, and the methods very, the most common is simply putting a 1k resistor in series with the cap and hooking it ti a supply set to the working voltage of the cap for 30min.

reading deeper the methods all had a few things in common...there is a desire to keep the current limited in the beginning of the reform as to not overheat the capacitor and to ramp up the voltage as it reforms.. typical max current mentioned is 10ma but more often then not it was not mentioned. from all i can gather. though, not sure that rule applies universally. it seems like a good rule of thumb.

Also i've seen various methods mentioned for higher voltage capacitors., higher voltage like being above 50v. these methods all seem to be about the same things, keeping the current down and ramping up the voltage.
these methods include using higher series resistance and lowering the applied voltage and then stepping it up.

As i have been reforming, I've not seen where the internal resistance becomes better, or the measured capacity increases after reforming compered to the initial unreformed measurements. it does prevent them from failing when put into use though. I'm not saying it will not help internal series resistance or capacity, its just has not been my experience.

What i've been doing is simple and I've been sticking to 3 simple rules.
1. keep the reform current under 10ma using a series resistance.
2. work your way up to 100% working voltage. [ NOTE: if necessary lower the supply voltage initially but series resistance usually accomplishes this on its own. ]
3. once a stable leakage current is reached at 100% working voltage, leave powered for an additional 30 min.

so in example....this is how i may apply the rules for a given cap.
1. when I start with a resistance and voltage  that will limit the short circuit current to a maximum of 10ma - personalty i've been sticking to under 5ma. and I work my way to  a 1k series resistance and 100% of the working voltage. so for a 50v cap i start with a 10k series resistance. and once the current drops down to like .5ma i'll change to a 1k resistor.
1a. the series resistance lets to voltage on the cap slowly rise as the current drops.
1b.  i have a current limited supply, but do not use the current limit feature as the way to limit the current to the cap. i use in the resistance. as using the current limit alone with no resistance would put a high voltage on the cap immediately, which would break my rule 2
1c if a capacitor is 50V or under i start with the working voltage applied to the series resistance. for over 50V if not then i'll increase the voltage in 3 stages 50% 75% and 100% not that i count use the resistance alone, its just easier to deal with in my setup and also increasing the voltage is mentioned as a methodology for high voltage electrolytics in a few documents i''ve read.
2. once i'm at 1k resistance, and 100% working voltage, i'll wait until the once the leakage current becomes stable, at that point i'll leave it connected for an additional 30 min.


oh and... i never let this process run unattended.

any comments are appreciated. help me fine tune the rules :)





Hobbyist and a retired engineer and possibly a test equipment addict, though, searching for the equipment to test for that.
 

Offline dexters_lab

  • Supporter
  • ****
  • Posts: 1849
  • Country: gb
    • DextersLab2013
Re: Reforming aluminum electrolytic capacitors
« Reply #1 on: October 17, 2017, 07:29:59 am »
so long as you current limit the input to the cap and slowly bring the voltage up it should work ok

i used a bench power supply, initially set to very low current (few mA) and brought the voltage up from zero to 25% rated voltage and leave it for 10 mins, then take it to 50%... rinse and repeat until your at it's max rating (i in fact took it a couple of v over) and then leave it for 30 mins or so

Using a bench PSU in CV/CC also allows you to see what is going on too, it's really interesting doing it for the first time on a very old cap, each time you charge/discharge you can really see the difference in capacitance, charge time and leakage change. I did my experiments on a big 43,000uf 30v cap from 1974 which i am sure hadn't seen any voltage in an estimated 15 years, probably more, it's large capacity really allowed me to see what was going on because everything happens so much slower.

After treatment it was right within spec for capacity and ESR

http://dexterslab2013.blogspot.co.uk/2015/12/reforming-electrolytic-capacitors.html
"A common mistake people make when trying to design something completely foolproof is to underestimate the ingenuity of complete fools." - Douglas Adams
https://www.youtube.com/user/DextersLab2013
http://dexterslab2013.blogspot.co.uk/
 

Online BravoV

  • Super Contributor
  • ***
  • Posts: 7087
  • Country: 00
  • +++ ATH1
Re: Reforming aluminum electrolytic capacitors
« Reply #2 on: October 17, 2017, 08:47:05 am »
Regarding leakage, although you mentioned the common aluminium electrolytic caps, just fyi, I have done using Fluke 287 logging DMM that has 10 Mega Ohm internal resistance at voltage mode, here is the leakage chart below, its the forming result for about 1.5 hours from a NOS polymer cap OSCON at it's rated voltage, at the chart reading, Y axis -> 1 volt equivalent to -> 0.1uA or -> Volt / (10M Ohm DMM internal resistance).

Suggesting you test for much more than 30 mins, as sometimes the leakage start to form an interesting slope that like this below (look at 30 mins mark).




More detail on the test setup here -> Measure Capacitor Leakage with DMM




Online Zero999

  • Super Contributor
  • ***
  • Posts: 14580
  • Country: gb
  • 0999
Re: Reforming aluminum electrolytic capacitors
« Reply #3 on: October 17, 2017, 12:29:48 pm »
I thought electrolytic capacitors are formed at higher voltages, than the rated values, so reforming at 110% to 125% of the rated voltage, is probably a good idea.
 

Offline innkeeper

  • Supporter
  • ****
  • Posts: 567
  • Country: us
Re: Reforming aluminum electrolytic capacitors
« Reply #4 on: October 19, 2017, 12:02:33 am »
I thought electrolytic capacitors are formed at higher voltages, than the rated values, so reforming at 110% to 125% of the rated voltage, is probably a good idea.

i've not read much on the inital forming process.

every manufacturer document i've gotten my hands on, and, one military one all state reforming at the working voltage though. i'm assuming theirs a reason for this. i can only guess they don't want you to exceed the breakdown voltage and damage the cap.
Hobbyist and a retired engineer and possibly a test equipment addict, though, searching for the equipment to test for that.
 

Offline innkeeper

  • Supporter
  • ****
  • Posts: 567
  • Country: us
Re: Reforming aluminum electrolytic capacitors
« Reply #5 on: October 19, 2017, 12:17:58 am »
Suggesting you test for much more than 30 mins, as sometimes the leakage start to form an interesting slope that like this below (look at 30 mins mark).
....

More detail on the test setup here -> Measure Capacitor Leakage with DMM

that was an interesting read.

when i get some time, I'll have to set up some logging, and do some leakage curves on some of the caps and see how it plays out. I added the 30 min to what most manufacturers recommended as a good measure approach.  From what i gather, the manufacturers consider the cap reformed once the leakage current stops decreasing.

what i have done for a post reforming test for an electrolytic is leave them in a charged state overnight, and check the voltages in the morning. its non scientific, but lets me know if i have one that is self discharging quickly. in theory if the leakage current is low, it shoun't self discharge. but it is like my final test before i use the cap.

I'll see if i can get the same kind of curve you show on an electrolytic when reforming.

Hobbyist and a retired engineer and possibly a test equipment addict, though, searching for the equipment to test for that.
 

Offline innkeeper

  • Supporter
  • ****
  • Posts: 567
  • Country: us
Re: Reforming aluminum electrolytic capacitors
« Reply #6 on: October 19, 2017, 12:26:23 am »
so long as you current limit the input to the cap and slowly bring the voltage up it should work ok

i used a bench power supply, initially set to very low current (few mA) and brought the voltage up from zero to 25% rated voltage and leave it for 10 mins, then take it to 50%... rinse and repeat until your at it's max rating (i in fact took it a couple of v over) and then leave it for 30 mins or so

http://dexterslab2013.blogspot.co.uk/2015/12/reforming-electrolytic-capacitors.html

very cool, and a nice writeup by the way. 
the way you approached it seems very safe.  as you point out, keeping the current low, and raising the voltage over time.
seems somewhat similar to how i treat high voltage caps.
Hobbyist and a retired engineer and possibly a test equipment addict, though, searching for the equipment to test for that.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf