On Wed, 23 Dec 2015 05:09:18 -0500 Gene Heskett <ghesk...@shentel.net> wrote:
> On Tuesday 22 December 2015 21:02:31 John Hasler wrote: > > > Gene Heskett writes: > > > In this case, the size of a capacitor hooked to a square wave > > > source, and I want to know how much charge is transfered for every > > > full cycle of the input square wave. > > > > Zero assuming linear source and load impedence. > > Ahh, but its being fed to a voltage-doubler with schottky diodes, so > there should be a calculatable amount of charge delivered to the > output capacitor Cf, by the pump capacitor Cp, per cycle. With Cp > being of a capacity such that with the usual logic circuits output > impedance, capable of delivering 10 to 20 milliamps into a resistive > load, IOW not a zero impedance source. Cf would be sized such as to > be about 50x Cp, and will be driving an enhancement mode fet which > will turn on a 40 amp SSR once the charge has built up on Cf to make > the fet conduct. > > Do it numerically with a spreadsheet. It takes a while with any reasonably-sized spreadsheet, but you can trust the result, you can see every stage and can spot anything obviously wrong. I once needed to analyse the start-up surge into a large capacitance using a toroidal mains transformer, and given the non-linearity of the rectifiers, I could see no other way. When results can reach hundreds of amps for a millisecond or so, you don't want large piecewise-linear approximations. -- Joe