reforming capacitors / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

reforming capacitors

Move Thread

LAN_403

rp181

Sat Jun 13 2009, 01:38AM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Hello all,
I want to reform some electrolytic capacitors, but I do not have a variac available. How would just charging the capacitors to the rated voltage, and holding it there work? I charged the bank, and it leaks a lot at 400v volts. The leakage lowers at around 350 volts.

OiD

Sat Jun 13 2009, 04:10AM

Registered Member #2150 Joined: Tue Jun 02 2009, 08:33PM
Location:
Posts: 29

Hi rp181,

You could try using a large resistor such as a few light bulbs in series. Try different values until you can maintain a voltage slightly below the maximum voltage.
You would provably have to keep an eye on the voltage. If they stop leaking the voltage could rise above the rated voltage.

You could also use a voltage doubler or multiplier, but the same problem could arise. Over voltage.

If they leak a lot then this method might not be suitable; Charge them to the max voltage, turn of the power, let it drop a bit, turn the power on again and repeat. It's what I do. Get a book or something to do meanwhile and a bit of patience. ;)

Sulaiman

Sat Jun 13 2009, 06:31AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3141

Maintaining electrolytics at their rated voltage will reduce the leakage current
usually to well below the specifications of the manufacturer
e.g. for BHC Aerovox ALS30/31 range (common in UK)
I = (uF) x (Vrated) x 0.006 ... OR ... 10 mA, whichever is smaller.

It's getting the caps to the rated voltage where care is required
ac mains, a rectifier (half- or full-wave), and a mains filament lamp in series
(plus a 1K resistor if you will not be nearby in case of disaster).
Bring each cap up to voltage separately.
(apply ac untill the lamp extinguishes, remove ac,
short across the rectifier - the lamp discharges the cap)
Then construct your capacitor bank, bring up to voltage via the lamp and leave on.

Don't do the initial reforming with the capacitors in a bank because,
- if one is very high leakage you'll have to dismantle the bank to find it
- if one fails at or near full voltage the entire capacitor bank will discharge through it - bad !

Proud Mary

Sat Jun 13 2009, 07:22AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

The life expectancy of an aluminium electrolytic is approximately given by:

Lt =Lr*(Vr/Vo)*2(deltaT/10)

Where:
Lt = operating lifetime at given temperature and voltage (the capacitor core temperature, not just the ambient)
Lr = the lifetime at rated limits
Vr = rated voltage limit
Vo = operating voltage
deltaT = difference between operating temperature rating and capacitor core temperature - Celsius

Dr. Dark Current

Sat Jun 13 2009, 08:29AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

Harry wrote ...

The life expectancy of an aluminium electrolytic is approximately given by:

Lt =Lr*(Vr/Vo)*2(deltaT/10)

so if DeltaT is 0 then life expectancy is 0 too?

Steve Conner

Sat Jun 13 2009, 09:22AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Pretty much, the life expectancy drops like a rock as temperature goes up. Harry's formula may be wrong though: if you plug in the rated voltage and rated temperature (so Vr/Vo = 1 and deltaT = 0) you should get the rated lifespan, which is usually only 2000 hours.

Keeping the core temperature within limits is so important that some of RIFA's inverter grade capacitors have a hole through the middle to fit a temperature probe, which can be hooked up to a shutdown circuit.

Proud Mary

Sat Jun 13 2009, 10:11AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

Steve McConner wrote ...

Harry's formula may be wrong though: if you plug in the rated voltage and rated temperature (so Vr/Vo = 1 and deltaT = 0) you should get the rated lifespan, which is usually only 2000 hours.

How true!

Still, I can always fall back on being an old fossil!

rp181

Sat Jun 13 2009, 04:41PM

Registered Member #1062 Joined: Tue Oct 16 2007, 02:01AM
Location:
Posts: 1529

Thanks for the input. Last night, I already started doing the whole bank (except 1 I accidently left out...). My stupid bank design makes it very hard to put the bus bars back on. I charged to 400 at around 10:00 pm last night, and this morning there was 288 volts. Now I am frequently charging it up, working my way to 450 (My guess of the surge rating). I can already see a big difference in the leak rate.
I can not find a datasheet for these, so I have no idea of specs. They are also surplus (the alltronics ones), so they have been sitting un used for a long time (still guessing here). Should have reformed when i got them....

EDIT: I think I am done. The leak at 419 volts is about the same as 280 before reforming. I charged back up to 450, and will let it sit for around 10 more hours, and maybe over night, for about 20 hours.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob