SEPIC Capacitor

Tony (remove the "_" to reply by email)

Digikey have a few of them available...

The datasheet recommends tantalum or ceramic.

The RMS current rating is based on the heat produced as the current passes through the internal series resistance of the cap and the temperature rise that heat produces escaping from the cap. If you can't find an RMS current rating on a data sheet, it is pretty hypothetical.

Does anyone know how can I figure out the RMS current capabilities of any given capacitor??

Tony (remove the "_" to reply by email)

Thanks again for you reply, you have been very helpful to me both now and in the past! I do have another question though...

Say I have a capacitor with a D.F of 2.5% and have calculated an Xc of 1.6R, ignoring temperature, can I work out the ESR as, 0.025 x 1.6 = 40mR and then from that the current from I = sqrt(P / ESR)?

Regards,

AJ

That is the general idea. But there are two problems.

You don't know that the D.F. is is entirely because of ESR. There are several other loss mechanisms that can contribute to it. For that reason, this calculation may be conservative. In other words, ESR is probably lower than this calculation predicts.

The big problem is coming up with a value for P. If the ESR is distributed throughout the capacitor, it will have one value. If it is concentrated in one or two internal current pinch points (often the connection point of the external leads with the plate structure) that involve only a percent or two of the capacitor's volume, then those points will be hot spots that have a lot of thermal resistance between them and the bulk of the capacitor's mass and surface, and P will be lower.

So the concept, while useful in understanding how ESR limits the RMS current rating for a capacitor does not do much to help you quantify the current rating if all you know is the capacitance and D.F.

Normally RMS current is only used for life time calculations. The more RMS current (calculated as the RMS value of the AC current in the capacitor), the more heating, the shorter the lifetime.

A rule of thumb is like this : every 10°Celcius hotter is halving the life. When using a capacitor rated 2000Hours at rated current and 105°C, you can expect 2000*8 = 16000 hours at 75°C.

Get the correct specified allowable ripple current from the datasheet. For most Low ESR type capacitors, it is specified for other frequencies than for standard capacitors. (for example 100kHz instead of 50/60 Hz). For these types of capacitors, the ESR value will be specified also in the datasheet.

I don't think you can rely on the dissipation factor. Use the specified values for ripple current and life time.

Stefaan

----- Original Message ----- From: "AJ" Newsgroups: sci.electronics.design Sent: Wednesday, November 02, 2005 12:04 PM Subject: Re: SEPIC Capacitor

any

passes

rise

in

then

any

passes

rise

in

then