Do DC caps have an AC rating?

Sorta kinda, but really no. And "AC" cap is a misnomer, although "AC rated" is spot-on.

I hope that helps.

Oh hell, I'll expand:

All capacitors have a maximum voltage that they can withstand, from which the ratings are derived. "AC" ratings are based on the maximum expected instantaneous voltage that a cap is expected to see when the line voltage is at the rated value. "DC" ratings are really instantaneous ratings, and are more "real", if you will.

Those are electrolytic capacitors. You can tell by the name, and by the small size for the capacitance. Electrolytic capacitors are essentially _really crappy_ diodes* with a lot of junction capacitance and with electrolytic goo (salts, water, and some sort of gell) as one of the terminals. So they have _two_ behaviors to instantaneous voltages: the desired one, where the 'diode' is reverse biased and you're using it as a cap, and the other one, where the 'diode' is conducting, the electrolytic is boiling, and the cap is about to undergo a steam explosion.

So you may not want to use them with AC, because within the first 1/60th of a second they'll be fixing to explode -- and there's not much useful work you can get out of an RPC in 1/60th of a second.

• In fact, you can take the same metals and chemical goo that you use to make electrolytics, and make electrolytic diodes instead. They were popular in the early 1900's before vacuum and mercury rectifiers came down enough in cost to be feasible.

Very educational.

What if you connect two electrolytic capacitors in series, but with polarities reversed?

+-...-+

Then the pair would not be able to conduct current in either direction, but would be able to accept the charge?

AFAIK that's what's done to make "non-polar electrolytic" caps. But I'd check before I did it.

Note, too, that electrolytic caps generally have problems with too much AC current, and they are often not clearly rated for their current handling ability, although things have gotten better in the last decade or so.

DC rating means exactly that. Yes, they'll pass AC, but they're electrolytics built for DC PS filtering use. Probably are 2500 uF +-

50% or so. The more amps out of the power supply, the more capacitance is needed. If you hook them up to AC, you'll get a big boom as the electrolyte turns to steam. The dielectric on the "plates" of the capacitor is the oxide film formed from the action of the electrolyte on the aluminum foil. If you put AC on them, this gets stripped and things get hot.

If they're more than a few years old, they're probably useless without reforming. I use a Sprague cap test meter for this, runs up to 450v and also reads leakage current. I usually start at about 1/4 of the rated voltage and increase as the leakage current goes down, usually takes 3-4 hours per step. If the current never goes down as time goes on, it's a dud, time to go to the next one. Electrolytics are fungible items, need to be changed out every so often as ESR and leakage current increase. NOS aluminum caps may well be junk, they don't age well even just sitting there.

AC caps either have oil-impregnated paper or solid insulator materials for dielectric. If you need AC caps, buy AC-rated caps, DC electrolytics aren't going to do it for you.

Stan

Yes. DC allows a kind of chemistry (electrolytic capacitor) to make a thin insulating layer, which gets destroyed by AC or the 'wrong' polarity.

It takes two of 'em in series, with two steering diodes, to replace an AC capacitor. The capacitance of the series pair is HALF the capacitance of the individual capacitors.

Yep, that's the benefit of the electrolytic type; the insulator part of the capacitor is more compact than one can safely construct plastic or paper or ceramic equivalents. 350 VDC means two in series will hold off about 250 VAC (because the peaks of 250 VAC are actually higher than

250V).

Using these in a rotary phase converter, they would be only connected briefly, for starting; otherwise, you'd need to worry about the ripple- current rating as well as voltage and polarity. Capacitors can overheat and blow up...

Not good for RPC's. That type of capacitor is only good for DC and have to have the polarity correct. Install them with the wrong polarity and they are likely to get hot and explode. Putting AC on them means the polarity is wrong half the time, so again likely to explode. By likely I mean almost certain to explode.

=20 Dan

DC rating indicates that measuring from anode to cathode the voltage can be 0 to +(rated volts), but that the voltage anode to cathode can not be allowed to go negative.

In other words, DC indicates the allowable polarity/bias.

Using it as an AC cap will result in a short life, I believe, how short depending on the actual type (electrolytic, etc). The capacitance may not even be the same biased in reverse.

Reverse biasing an electrolytic can lead to near-instant dramatic failure as an electrochemical reaction (in some) destroys the internal plate-plate insulator.

Dave

Aluminum and tantalum electrolytics inadvertently installed backwards may take a while to vent or explode. I've encountered a batch of dipped tantalums that were marked backwards and ALL installed wrong. The applied voltage was considerably lower than the rating and they failed fairly quietly, blowing a chip out of the side with a muffled POP. Electrolytics can also grenade and fill the air with caustic- covered aluminum confetti.

Two connected back to back for AC should be bypassed with diodes, or better just buy AC-rated ones.

DC caps have an AC ripple current related to permissible heating from their Equivalent Series Resistance (ESR). Caps meant for linear power supplies are cheaper and have a lower ripple current rating than ones designed for switchers.

jsw

If the steering diodes are correctly installed, the circuit consists, during up-cycles, of one capacitor and one diode; during down-cycles, the other capacitor and other diode. So the harmonic-mean series-capacitance formula, C = 1/(1/C1 + 1/C2), doesn't apply. That is, if capacitance of each capacitor is C, the effective capacitance is C, not C/2.

[snip rest]

Non-polarized capacitors also have a max VDC rating, which means the max voltage can go from -(rated volts) to +(rated volts). Of course, the capacitors he's describing would be polarized.