Charging Capacitors / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Charging Capacitors

1 2 next

Move Thread

LAN_403

Shaun

Mon Jun 04 2007, 12:08AM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

This mainly deals with electrolytic caps, but I guess it could be applied to any type. Everyone always says NEVER EVER overvoltage electrolytics, as they can pop like firecrackers. So lets say I had the setup shown here:

The power supply would be 400 V, and the capacitor would be a 350 V electrolytic. Suppose it began charging, and while watching the voltmeter, it would slowly rise from zero to 400, at which point the capacitor would probably be damaged in some way. My question is, suppose the switch was closed, shorting out the capacitor, before the voltmeter read more than 350 V. Would this still be damaging/dangerous for the capacitor (or me)? I have scoured the internet, and all I can find are warnings that the voltage across a capacitor should never exceed the rating.

...

Mon Jun 04 2007, 12:28AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

That would work, however when you closed the switch it (the switch) would explode. You need to put a resistor in series with it (say 40 ohms, which would limit the current to 10amps) that is sized to dissipate all of the energy stored in the bank.

Shaun

Mon Jun 04 2007, 01:32AM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

Thanks, really there would be a load in there and the switch would actually be an SCR. This problem came to me while building a coilgun, you can see how. What if your power supply was of a rediculosly higher voltage, like 10,000 V or something? Would this setup still be feasible?

...

Mon Jun 04 2007, 02:28AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

This would work with a supply of a million volts, assuming it can give enough current to charge the caps at a resonable speed.

Part Scavenger

Mon Jun 04 2007, 02:39AM

Registered Member #79 Joined: Thu Feb 09 2006, 11:35AM
Location: Arkansas
Posts: 673

If you are really going to build this thing, put a resistor in between the source and the capacitor. If you don't the capacitor could either:

1. Charge so fast you can't stop it before it overvolts
2. Blow the breaker/fuse

Just choose how fast you want it to charge and use the RC time constant to determine the ohmage of the resistor.

R*C*4 = 98%*Vin

Ohms, Farads, & Volts.

Shaun

Mon Jun 04 2007, 02:53AM

Registered Member #690 Joined: Tue May 08 2007, 03:47AM
Location: New Jersey, USA
Posts: 616

Would the extra volts cause it to charge faster? Like if I hooked it up to my flyback, which is around 15KV @ a few ma, would this charge faster than a 350 V supply @ a few ma?

Bjørn

Mon Jun 04 2007, 03:09AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

No, it will follow Ohms law. If you connect a flyback to the capacitor, the capacitor will short out the flyback and the voltage will drop to about 0 V. The voltage will slowly rise until the leakage current in the capacitor is the same as the maximum current of the flyback, or until the capacitor fails.

Simon

Mon Jun 04 2007, 06:52AM

Registered Member #32 Joined: Sat Feb 04 2006, 08:58AM
Location: Australia
Posts: 549

That's right, but with a much more powerful high voltage supply that can keep up the current, yes charging to a low voltage is faster if you use a much higher input voltage and stop at the low voltage.

Weaker power supplies (at least qualitatively) are better imagined as constant current devices rather than constant voltage devices. Voltage rises linearly on a capacitor with constant current.

Danielle

Mon Jun 04 2007, 04:36PM

Registered Member #632 Joined: Mon Apr 09 2007, 01:09AM
Location:
Posts: 267

Ive overvoltage a very large lytic and it already had cracks in it the cap sparked and smelled but after a week it still works and now I tested it to full voltage and it still works. Is this because I didn't fully destroy it or because it can self repair? This happened when my boost converters regulator failed and the voltage started climbing faster than I could react.

Bert

Mon Jun 04 2007, 05:33PM

Registered Member #118 Joined: Fri Feb 10 2006, 05:35AM
Location: Woodridge, Illinois, USA
Posts: 72

Rocketry76 wrote ...

Ive overvoltage a very large lytic and it already had cracks in it the cap sparked and smelled but after a week it still works and now I tested it to full voltage and it still works. Is this because I didn't fully destroy it or because it can self repair? This happened when my boost converters regulator failed and the voltage started climbing faster than I could react.

To a limited extent low voltage electrolytic capacitors can "self-heal" a dielectric puncture IF major damage was not done to the affected plate area during the failure. In this case, self healing is simply the "re-forming" of the aluminum (or tantalum) oxide dielectric layer (an electrolytic process, similar to anodization). However, with high voltage electrolytics, overvolting is often catastrophic since the stored energy in the cap typically vaporizes (and then arcs) the surrounding area where the defect originated. The sudden increase in gas pressure can cause the capacitor to vent hot gases or steam, or even explode. The failing section of your particular cap may have benificially become internally disconnected when was physical damaged during the earlier overvolting. It may have left the remaining portion of the cap functional but with less capacitance. More often, the capacitor simply fails, sometimes quite spectacularly.

Some specially designed metalized film capacitors can also self-heal. However, in this case, specially designed areas in the thin metalization act as fuses, isolating a defective section. Each time this occurs, the overall capacitance of the device declines slightly since the active plate area has effectively been reduced. These caps reach an end of life when the capacitance has declined by 5% or so.

1 2 next

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