boosting a supercap?

That's 230 amps incoming at 1.3 volts. You won't find an IC to do that.

John

On a sunny day (Thu, 10 Dec 2009 10:01:05 -0800) it happened snipped-for-privacy@gmail.com wrote in :

Those are impressive capacitors, why not use some in series?

Use more in series. With a switcher, power in = power out / efficiency, that means that input current = output power / (input volts * efficiency)

-- so if you want to take the cap down to 1/2 voltage (1/4 the total energy) you'll be sucking over 250 amps from the thing -- that's more than their biggest one will handle (and have you checked the energy storage yet?).

So:

First, you aren't going to find the switcher that you want in a single IC. I'm not sure there are _any_ 200W single-IC switchers out there, and yours is a special application.

Second, the _biggest_ cap they have on their data sheet doesn't have the discharge rate you need. Since you need more than one, what's the big problem with using more than one in series? It'll lower the current and increase the voltage, both of which will make the switcher design easier.

Pull a Marx: charge them in parallel, use them in series.

John

by the time you mount switch capable of 230A+derating and an inductor rated for 230A plus an extremely low ESR,ESL capacitor bank out the output to catch the 250Ap-p ripple, you're probably better off putting a few if these in 3 or 4 series to make 12V from that voltage.

If the 10 seconds at 200W is a hard requirement a SLA battery or two might be a better fit.

A switched capacitor convertor? :)

Well, if you've always got 12V on the output side (or maybe 11V to detect a dropout condition?), that's more than enough to switch a FET. You could use a UC3843 or similar controller, although you'll need a rather small current sense resistor to use it; maybe you can hack it to use Rds(on), with slope compensation if necessary.

Tim

-- Deep Friar: a very philosophical monk. Website:

Before you get too wound up in the design details, did you check the price of these, compared to say an SLA battery ?

Yes. If it's 1:1, it's a flying capacitor isolator, seldom used for power but actually very interesting. For voltage stepup (and isolation!) it becomes essentially a Marx generator, using ssr's or maybe even real relays.

Reminds me of the Monte Python skit, the one about the British Society For Putting Things On Top Of Other Things.

John

Jan Panteltje a écrit :

Impressive?

I know somebody who worked on replacing a subway accumulator based backup supply with one based on a bank of those supercaps. The whole train can go from one station to the next only on the caps energy.

Sure impressive.

Use multiple phases.

Bunches of them.

On a sunny day (Thu, 10 Dec 2009 22:49:45 +0100) it happened Fred Bartoli wrote in :

You do not need caps if down hill.

On Thu, 10 Dec 2009 10:01:05 -0800, snipped-for-privacy@gmail.com wrote: .. Thanks for all the suggestions.

The voltage characteristics is

V = V0*exp(-t/RC)

solving for C,

C = -t/ln(V/V0)/R

For a regulation of 95% in 10s, by which I mean the voltage will have dropped 95% or 11.4V requires a capacitance of approximately 300F.

Mouser has some some 52F 15V caps for 200$. So your looking at about 1k for your design.

If I were you I'd just use a led acid battery for about 50$. Or better yet, just by yourself a UPS?

p.

Don't be crazy, use the 16 V modules, or maybe the 48 V modules.