How to Recondition (Reform) Electrolytic Capacitors and Why

[maxbarjr]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Can we see some pictures?

If the liquid really came from the caps... sorry, but they are DONE. No reforming or any voodoo magic can help them. That's because once they vent, moisture and other contaminants will eventually enter the cap and make it go bad over time. Also, with the vent open, the electrolyte will start to dry out, further accelerating the deterioration of the caps. But most importantly, the electrolyte in most caps is HIGHLY CORROSIVE and *will* destroy traces on PCBs.

So there's not much point in messing with these caps, other than charging and discharging them for fun, if you're into that.

Sorry for the late reply. Picture attached.

Thank you momaka.

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

I don't see any liquid/electrolyte on that picture. Just residue from manufacturing and/or dirt from over the years.

[Behemot]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

I think there should be rubber plug in that hole, its pressure sealing. If it is gone, the electrolyte has been in contact with air - it's done.

[Eaglecrest]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

I have lots of stuff that have been sitting unplugged for 3-5 years. Not one has ever failed on me. Even some really old stuff that I found and plugged in that haven't seen power for over 8-10 years worked fine. Again, there's a reason why electrolytic capacitors are called "self-healing" capacitors. Of course, if you do want the caps to last as long as possible, then reforming the capacitors (or plugging in the device and using it) is recommended.

So if I have a stereo receiver that hasn't been used in 10 years, what is the best way to proceed with it?

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

So if I have a stereo receiver that hasn't been used in 10 years, what is the best way to proceed with it?

If it's an oldschool stereo amplifier (1980's or 1990's) with Japanese caps, it probably will be okay to just plug it in. Most of the Japanese caps used in those amps are non-aqueous, and often don't degrade that much in storage.

But if you really wanted to be extra careful with it, you can just attach a 60-100 Watt incandescent light bulb in series with the line plug. This should slow down the startup voltages a tiny bit, giving the caps more time to reform to full voltage. It will also limit the current going through the amp, should something decide to go bad inside (i.e. output stage going out). After 5-10 minutes operation, you can just plug the amp straight into the wall.

And of course, if you really are paranoid, you can take off the capacitors and reform them manually... but I would say that's not necessary at all.

[Per Hansson]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

(TLDR version second paragraph below).

Have been debating weather to post this in the funny shit thread or electronic mistakes thread but I'll go with here as it's "sorta" ontopic anyway
I always wanted a capacitor discharger for work, I've used all sorts of ghetto solutions but something nicer like Mr Carlsons capacitor discharger would be ideal.
Of course we then come back to my lazyness, for example in post #42 I write that I've bought a second digital display for my capacitor reformer, well, it still sits in the parts drawer today
So that got me looking for another solution, and actually my Fluke 113 with it's input resistance of 3kΩ is pretty ideal, however I don't use that for work as it's a useless meter.
My actual work meter, a Fluke 28-II does not have dual input impedance modes.
So I found the Fluke SV225 LoZ adapter, it looks pretty ideal for my purposes, but of course it has PTC's just like a regular meter otherwise it would catch fire if you connect it to the grid indefinitely...

So I wanted to test if a LoZ meter can discharge a large capacitor but of course I then needed a large charged capacitor, on my bench was a used 420v 390uF that seemed good for a first test.
I charged it to 30VDC with my lab power supply, the Fluke 113 could discharge it quickly but of course at 30VDC it had very little stored energy...
So I figured I'd use a servo that was on my bench anyway to charge it to 325VDC, that should equal 20J of energy if my math is not off.
However even with the bench supply's low 30VDC there was a quite big spark, so I thought I'd limit the current using a series resistor to protect the capacitor and servo.
There happened to be a resistor infront of me on the bench so in it went.
I connected her up and BOOM!
WTF just happened? Looked at the servo and it was fine, looked at the capacitor and it was fine too, but charged to only 175VDC.
The resistor you ask? It had a hole blown in it's side
I looked at it's color bands: yellow, violet, black, gold
That's a 47Ω resistor, maybe some of you are seeing were this is going now, if not look at this formula:

Calculations: power draw P = (V^2)/R

325v x 325v / 47Ω = 2247w, through a 1/4w resistor, hmm, I wonder why it blew?
Annyway to make a long story short I charged it with a 470kΩ resistor instead, that was much less entertaining though
The Fluke 113 discharges that in seconds.
I also tested to discharge the servo with it, on it's own it took over 5 minutes to get to single digit voltages, with the Fluke 113? Only 12 seconds

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Of course we then come back to my lazyness...

I doubt you're lazy. You probably just have more important things to do or too many projects, and that's why you don't get to this. I'm like that too, sometimes (though at times I truly am lazy ). That said, I did actually make a somewhat more stand-alone capacitor reformer rather than using my breadboard every time. But when I have only one or two old caps I need to reform, I still do the "quick/lazy" way - i.e. a few jumper cables on the floor connected to a resistor and a power adapter.

I connected her up and BOOM!
WTF just happened? Looked at the servo and it was fine, looked at the capacitor and it was fine too, but charged to only 175VDC.
The resistor you ask? It had a hole blown in it's side
I looked at it's color bands: yellow, violet, black, gold
...
325v x 325v / 47Ω = 2247w, through a 1/4w resistor, hmm, I wonder why it blew?

Well, the power alone wasn't what blew it here. It was the energy stored in the cap.
Indeed as you noted above, the energy stored in that cap was about 20 Joules. For those wondering, energy stored in a capacitor is
E = 0.5 x C x V^2
where E is energy in Joules, C is Farads and V is volts.

Now, we also know (or do we? ) that the energy E is the amount of power P applied in a unit of time t... i.e.:
E = P x t

So 1 Joule = 1 Watt x 1 second

The 47-Ohm resistor above might have started dissipating 2247 Watts of power, but that was only in one instant. As the voltage in the capacitor fell, so did the power the resistor dissipated. But we don't care about that. All we care about is the average energy (heat) it has to get rid of.

With 20 Joules, you can say that the resistor dissipated 2247 Watts in 0.0089 seconds... or 224.7 Watts in 0.089 seconds... or 22.47 Watts in 0.89 seconds... or to get a more meaningful result, 20 Watts in 1 second. And if we keep going further...How about 1 Watt in 20 seconds - energy is still the same. Would a 1/4 Watt resistor burn @ 1 Watt of power in 20 seconds? Or how about 1/2 Watt @ 40 seconds? - I think YES, very likely. Here ElectroBoom explains it too (and it's a good video, BTW )
https://www.youtube.com/watch?v=xjW-isgOijs&t=7m17s

[lookimback]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Nice write up. You could also crack your power adapter open and add a potentiometer to the feedback circuit of the PWM so you could then adjust the duty cycle to get different output voltages.

[Per Hansson]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

momaka, read again, the cap was discharged, I was charging it up through the resistor. (So I could perform my tests).
Otherwise it was right

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Ah, I see.
Well, either way, it doesn't matter whether you try to discharge a cap that has 20 J of energy stored in it or whether you're tying to charge it to that potentially hold that much energy when it's empty.

You could also crack your power adapter open and add a potentiometer to the feedback circuit of the PWM so you could then adjust the duty cycle to get different output voltages.

Good idea. I've thought about modding ATX PSUs, but not one of my power adapters, for some reason. And I do know how to mod these adapters, which is the funny part.

Hmmm. You've given me an excellent idea for a bench power supply now!

[lookimback]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

I had to open one to fix it once and was surprised to see it was just a mini SMPS. Since then, I've modified several, including one from a Nintendo DS which now provides 3.3 volts to a bluetooth module in my Logitech computer speakers. I have a few 3A@
12v ones from old cable boxes which I use when I'm fixing main boards, and you have actually given me the idea to use them
for a bench power supply as well. Not sure if I can get one all the way down to 1.25v, but it's worth a try. Just getting down to 5v would be fine for most of what I do. Maybe the laptop ones could be adjusted up to 24v, but your output current would decrease (I=P/V).

[Eaglecrest]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

But if you really wanted to be extra careful with it, you can just attach a 60-100 Watt incandescent light bulb in series with the line plug.

I am assuming this should be in series on the hot side, is that correct? What is the difference in effect between a 60W and a 100W bulb?

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

I am assuming this should be in series on the hot side, is that correct? What is the difference in effect between a 60W and a 100W bulb?

Correct, hot side.
Any bulb above 50 Watts should do really, but I generally recommend above 60W. The higher the power rating of bulb, the more current that can be drawn by the device (both at start-up and continuously). 60W will probably allow you to run the amp with a small load. 100 Watts bulb should allow you to crank it up a bit more. Not much of a difference otherwise. I've even used 40 Watt bulbs in a pinch, too - works with devices that have a lower idle power draw.

[Eaglecrest]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Momaka, am I correct in understanding that the light bulb places some additional resistance on the device so as to charge caps more slowly and thus limiting them blowing out? It that is the case then the 100 W bulb will give the greatest protection?

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Momaka, am I correct in understanding that the light bulb places some additional resistance on the device so as to charge caps more slowly and thus limiting them blowing out?

Yes, sort of.

It that is the case then the 100 W bulb will give the greatest protection?

No.
Lower Wattage = slower turn on and greater protection
Higher Wattage = faster turn on, less protection, but allows device to use more power.

With that said, you can't go as low as you like. Some devices will simply not work when the input voltage drops too much (i.e. when using a lower Wattage light bulb). So that's why I suggest 60-100W, as that will work on most amplifiers without trouble. I myself often get lazy and use my 40W soldering iron (it's a simple plug-in iron with a heating element, so it's pretty much the same as using a 40W incandescent light bulb). But I've encountered switch-mode PSUs that won't run with that.

[Eaglecrest]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Further to this discussion, it appears that for anything with electrolytic caps that haven't been used in some years would benefit from first powering on with 60 watt bulb. How about tube devices? Would they also benefit from this?

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Further to this discussion, it appears that for anything with electrolytic caps that haven't been used in some years would benefit from first powering on with 60 watt bulb. How about tube devices? Would they also benefit from this?

The series incandescent bulb trick is well-known among restoration techs as an alternative to a variac (and in some cases, even better than a variac, since the bulb limits current too).

[socketa]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Would anyone here be able to say:
1. What is the function if the diode in the original post?
2.

Not only that, but a leaky cap (as in, capacitor with a high leakage current… not one that is physically leaking electrolyte) will usually trick your ESR meter to show lower ESR than what the capacitor may have. So if the cap has gone high ESR, your meter may not show it and you might end up putting a faulty cap back in service. To avoid this, check the capacitance of the cap. If it is higher than 20% of its specified capacitance, it is likely leaky and it is time to reform it. If you don't have an ESR or capacitance meter (like me ), then definitely reform it so there won't be any doubts.

Will this also work for new old stock capacitors that are less than 20% of their specified capacitance?

[momaka]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

Would anyone here be able to say:
1. What is the function if the diode in the original post?

It's just for simple reverse polarity circuit protection. That is, if I was to plug in my power supply in backwards by accident (that is negative to positive and positive to negative), the diode would create a short-circuit across the power supply. This will prevent backward voltage feeding into the capacitors and making them go bad.

Since I use mostly SMPS PSUs for my projects, using a diode like that is OK, because the SMPS will simply just power-cycle until I fix the polarity mistake (that is, if the power supply is not too powerful - otherwise the diode would short out permanently.).

2.
Will this also work for new old stock capacitors that are less than 20% of their specified capacitance?

Oh yes, it will work - as in, allow you to plug the NOS caps back into a circuit and not have them blow up or short-out due to oxide layer thinning too much over time.

Reconditioning, however, will _NOT_ restore the original capacitance of the NOS caps. Nor will it restore bad ESR. Reconditioning is simply to "play it safe" before you insert NOS caps in a circuit. And if you ever buy caps and decide to store them, reforming them once in a while (once every 1-2 years.) may help slow down the deterioration rate.

[socketa]

Re: How to Recondition (Reform) Electrolytic Capacitors and Why

i can see that it's far from ideal (but, perhaps, more practical) to reform the caps collectively in circuit; so how about, in respect to PC motherboards, removing the CPU and then connecting a resistor (which would be, for example, 10Kohm X n, where n= number of VRM input caps) in series with the VRM input caps and applying 16V to it, and after that (after changing the resistor value, and voltage divider, accordingly), to the VRM output caps?
Again, far from ideal, but better than firing up a board that hasn't being used for several years with no reforming
Would applying 16V to the high side, and then 6.3V to the low side, damage the FETS?