Forming Cap's question..

[Notorious]

Hi guy's.. I'm bussy with reforming my capacitors junk box and I'm not sure if I'm doing it right..
What I did was hook 4k7 5w in series with 50000uf 50v cap, set transformer voltage (slowly) to 48v assuming that voltag after rectifiing is going to be about 5-10v higher. Now from what I understood was that after some time voltage over resistor should read zerro after cap was fully formed.. Am I right in my assumption? And if there is a formula for different voltage/capacity/resisance relation would you guy's like to share it with me? Thanx in advance..
P.S. I did search the forum but couldn't find anything related to my question..

[dhaen]

Hi,

I saw this post come in, and I'm surprised no one answered. (I forgot it.. )
There are varying opinions on this, but here's mine:
Caps do not significantly de-polarise until they are 20 years old +.
After that time, it is safer to apply a reduced voltage until the leakage current is "tamed". This can take a few hours.
Perhaps if you intended to use an old-ish high voltage cap near the limit of it's rating, more care would be merited.

[Notorious]

Thanx Dhaen..
Yes I was surprised as well, but than you can't expect from people to know everything now can you..
Anyway I went in wide world of Google search, and found lot of different aproaches to "tackle the problem"
For high voltage cap's 450-600v apply the rated voltage through a 5W resistor. Anything from 20K-50K will do. Watch voltage drop ( volt meter parallel with resistor) when there is a large voltage drop (more than 20% of the applied voltage), there must be a significant current flow through the capacitor. So when there is such flow, something must be wrong. For low voltage 15-100v I found that 10k 3w resistor does great job. Of course theorie mentioned above stays the same for low voltage cap's as well.
Yet somebody else came up with formula, Voltage, Resistance and Time for caps that are idle for 30 years.
Quote; Set the power supply to 450V since that's the capacitor's rated voltage. Time T will be (5min+30 years x 12 months x 1 min/month=365 min) just over 6 hours. Maximum charging current occurs when the full supply voltage appears across the resistor. R=V/I=450V/10mA=45K ohms. Power rating for the resistor will be selected for maximum-current conditions as V x I = 450 x 0.01 = 4.5W. I'll use a 47K 5-watt

So there you go guy's... I hope you find that info usefull, and that we all learned something today...

[dhaen]

Have you found significant numbers of de-polarized caps?
What was the age, and the time out of service, of those found?

[Notorious]

Nope... But I bought variac and have ton's of nice cap's at home so I wanted to "do it right"..
Now that I read some nice articles here, I have "urge" to do something with all that stuff that costed me lot of money, and at the moment only collecting dust..

[Zero Cool]

What about for caps that are rated for 75 to 250V any suggestions on resistor values or charge time?


Zero

[Zero Cool]

Also suggestion for how do i test for leakage. and how much leakage is too much??

IE: if i have some 10,000uF 75 VDC caps and i reform them and test for leakage, how would i know if there bad....

Obviously if they explode there bad but i meant how would i tell if there is too much leakage to be usable.


Zero

[SY]

What you want is to limit the current during the forming process. The exact value isn't critical. If you want to keep the initial current down to, say, 10 ma, you can calculate a series resistor knowing the input voltage.

For example, I want to form some 10,000 uF/100V caps. My supply should be at 100V, so I calculate a series resistor R = V/I = 100/.01 = 10K ohm. Power initially is P = IV = (.01)(100) = 1W.

For leakage, it's best to look at manufacturers' specs. If you've got significantly more than spec, toss the cap.

Caps can do nasty things like explode or vent hot fluids. Use safety precautions like shields, eye protection, and gloves.