totse.com | Charging Capacitors

Endotropic Decay · #1 2008-09-07, 03:04

I have 7 or 8 electrolytic capacitors that I've pulled out of old computer PSUs (all are 200v @ 470 uf).

After I got them, I realized I have no idea how to charge them. Seems like every few websites I go say the previous website is wrong, and their way is the correct way.

So, who would know more about capacitors then DIY?

Pretty much, I want a reliable and preferably fast method of charging these capacitors.

How do I charge them anyways?

Do I need to find a transformer that outputs 200v DC?

To charge a 200v capacitor, do I need a charger that outputs exactly 200v?

Anything I may not be aware of?

I don't plan on charging them to their specs, that doesn't seem too safe. I was thinking 150 or so, 175 at most.

wolfy_9005 · #2 2008-09-07, 06:48

It depends. For faster charging you want close to the rated voltage, for slower(but safer), you can always use a smaller voltage or a battery. Ideally you need a way to measure the power into the capacitor. Some transistors will automatically shut off the power at a certain voltage. The recommended way(by some...) is to use a camera flash. Just take it apart, remove the capacitor and replace with your own.

Got pics of your capacitors?

Ill post a pic of a 2.1kv capacitor i found in a microwave...

http://i38.tinypic.com/2lt1y7d.jpg

Since it's the same question, how would i charge this big capacitor? It has 1 pin and 3 pins(ie. 1 on left, 3 on right). Googles provided unsatisfactory results.....(grrrr google cant answer all my questions)

phmeworp · #3 2008-09-07, 14:49

The simplest charger for your application would be just one component (a diode) connected directly to the mains in your house, via a regular power cord. Connect the anode of the diode (rated at least 400 volts --- like 1N4005) to one side of the power cord and the cathode to the (+) terminal of the capacitor. Connect the other power cord lead to the (-) capacitor terminal.

This will charge the capacitor to 160-170 volts. The cap will charge to the peak AC voltage. Peak = RMS x 1.4 (120 x 1.4 = 168).

From there, you could add some simple tricks like a switch to turn the charging on and off, a resistor in series with the diode to lengthen the charge time, and maybe a fuse as well, just in case you do that booboo thingy. For fastest/most efficient charging you could use a bridge rectifier, but you don't need any fancy stuff like transformers, regulators and the like.

Be careful of polarity when connecting the caps (backwards = bang!) and try to not electrocute youself.

phmeworp · #4 2008-09-07, 14:56

Quote:

Originally Posted by wolfy_9005 View Post

Ill post a pic of a 2.1kv capacitor i found in a microwave...

Since it's the same question, how would i charge this big capacitor? It has 1 pin and 3 pins(ie. 1 on left, 3 on right). Googles provided unsatisfactory results.....(grrrr google cant answer all my questions)

That capacitor has just two "terminals" positive and negative. There are three pins on the one terminal to facilitate multiple connections to that one terminal.

protip: Read the lable on that thing. 2+ Kv is nothing to be playing around with. If you have to ask how to charge it, you should just leave it uncharged.

Hung Like Christ · #5 2008-09-07, 23:07

phmeworp:
Haven't seen you posting in awhile.
Your input is always welcomed.

regarding capacitor charging:
It takes place in under a second, based on voltage.
Create the appropriate voltage and you're done, in a flash.

phmeworp · #6 2008-09-07, 23:55

Quote:

Originally Posted by Hung Like Christ View Post

phmeworp:
Haven't seen you posting in awhile.
Your input is always welcomed.

Hey, HLC... glad to know that you remember me. For the record, I check into this forum, and about a dozen others every day, but refrain from posting unless I have something of real value to add that has not already been posted. I have to make the choice of giving away free advice or billing the customer (not going to boast here about my hourly billing rates).

And you are correct about the charging time frames... which is why I alluded to the difference between a half wave (one diode) and a bridge (four diodes) being notable, but not significant... who cares whether it is 16 or 32 milliseconds???

Sorry, but I live in a world where the difference of 0.005" or a half of a mircrosecond means the difference sucess and failure.

Chat with you again soon???

Endotropic Decay · #7 2008-09-08, 00:09

Quote:

Originally Posted by phmeworp View Post

The simplest charger for your application would be just one component (a diode) connected directly to the mains in your house, via a regular power cord. Connect the anode of the diode (rated at least 400 volts --- like 1N4005) to one side of the power cord and the cathode to the (+) terminal of the capacitor. Connect the other power cord lead to the (-) capacitor terminal.

This will charge the capacitor to 160-170 volts. The cap will charge to the peak AC voltage. Peak = RMS x 1.4 (120 x 1.4 = 168).

From there, you could add some simple tricks like a switch to turn the charging on and off, a resistor in series with the diode to lengthen the charge time, and maybe a fuse as well, just in case you do that booboo thingy. For fastest/most efficient charging you could use a bridge rectifier, but you don't need any fancy stuff like transformers, regulators and the like.

Be careful of polarity when connecting the caps (backwards = bang!) and try to not electrocute youself.

Wow, thanks, about the best answer I could have hoped for.

I have a few bridge rectifiers from computer PSUs, I might sit down tonight and think of a circuit that might work.

I do have a question or two though:

Quote:

This will charge the capacitor to 160-170 volts. The cap will charge to the peak AC voltage.

So, it'll charge to there and stop completely?

Would a multimeter connected to the capacitor terminals during charging be the correct way to measure how charged the capacitor is, or would I have to turn the charger off for that?

Some questions I ask might be kind of stupid, I'm pretty sure of the answer, but I don't want to take any chances, I'm kind of new to this.

EDIT:

http://img242.imageshack.us/my.php?image=picture001jl7.jpg

http://img337.imageshack.us/my.php?image=picture002yv8.jpg

and what is missing/wrong with this?

http://img87.imageshack.us/my.php?image=testax1.png

phmeworp · #8 2008-09-08, 03:26

Quote:

Originally Posted by Endotropic Decay View Post

Wow, thanks, about the best answer I could have hoped for.

I have a few bridge rectifiers from computer PSUs, I might sit down tonight and think of a circuit that might work.

I do have a question or two though:

So, it'll charge to there and stop completely?

Yes. There is no way for the capacitor to charge to a voltage higher than the maximum peak voltage applied.

Quote:

Would a multimeter connected to the capacitor terminals during charging be the correct way to measure how charged the capacitor is, or would I have to turn the charger off for that?

Your multimeter will be able to read the ultimate charge voltage, but that will be reached within just a few miliseconds (20 to maybe 100) of applying power... way to fast to observe without the use of an oscilloscope. And without a decent load, that charge will remain for a good while.

Quote:

Some questions I ask might be kind of stupid, I'm pretty sure of the answer, but I don't want to take any chances, I'm kind of new to this.

Your circuit looks good, but if you are going to be putting that many caps in parallel, I would tend to suggest putting a resistor in series with the switch/diode... just keep the inrush current through the diode in a reasonable range. I'd probably start with something in the 470 ohm range. (for techincal referance, a number I just pulled out of my ass)

Also, give some serious consideration to investing a few cents in a fuse (1/2 amp slo-blow) just for peace of mind.

ArgonPlasma2000 · #9 2008-09-08, 05:23

The resistor is a good idea. If you got enough capacitance you could easilly blow breakers every time you wanted to charge them since the resistance at 0V is nearly zero. The resistor won't have any effect on how high a voltage you can get into the capacitor since impedance increases towards infinity as the capacitor becomes fully charged.

Endotropic Decay · #10 2008-09-08, 22:38

How many watts would you people suggest the resistor to be?

at the moment I have a 560 ohm 3 watt, and a 330 ohm 2 watt.

Going to radio shack tomorrow to pick up a fuse, might as well get a resistor that won't overheat too.