From: Jonathan Kirwan
Subject: Re: zero-power toggle circuit; was, how to master electronics
References: <3DC3EFA5.firstname.lastname@example.org> <3DC517EB.email@example.com> <3DC57B4E.firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org>
X-Newsreader: Forte Agent 1.92/32.572
NNTP-Posting-Date: Tue, 12 Nov 2002 18:30:24 GMT
Organization: AT&T Broadband
Date: Tue, 12 Nov 2002 18:30:24 GMT
On 12 Nov 2002 06:56:43 -0800, Winfield Hill
>Jonathan Kirwan wrote...
>> Now, when the switch is pressed, one end of R2 is brought
>> momentarily close to V+ and starts pulling Q1's base closer to
>> V+ and shutoff. But not close enough, from what I can see.
>> With Vbe of about 0.65V and C1's charge under 0.1V potential,
>> the initial current must be around 50uA or so. This just seems
>> too little, since Ib is already some 200uA to 700uA at the time
>> and in *deep* staturation. It looks as though Q1 will stay on.
> A very nice explanation of how the circuit works, thanks!
> You're right of course. I'm sorry - that's what can happen
> when a design passes through several iterations on an ASCII
> computer screen; I should stick to paper for my doodlings!
> Anyway, I overlooked updating the value of R2, it should be
> much smaller.
This was what I was hoping you'd suggest. I also considered the
idea of not operating Q1 with such a 'heavy' base current, in
combination. But reducing R2 is what made the quicker sense to
me. I tried to present my description in just such a way that,
if I hadn't messed up somewhere in it, that would be the
> In fact one could eliminate that resistor
> entirely except for the fact it's not nice to blast the base
> of a transistor with many amps for a microsecond! If we
> choose say 330 ohms, that should suffice, yeilding a 330us
> time constant and a peak base current of 20 to 50mA.
That sounds about right.
> is easy on the transistor and insures that the FET's required
> turn-on gate charge is provided. The 2n4401 can be expected
> to provide at least 400mA of charging current to the FET gate
> for at least 100us, that's over 40000nC of charge. A modern
> IRF540 FET turns fully on with less than 30nC, the original
> version might require 50nC of gate charge. Larger FETs may
> require up to 250nC, so they can be nicely accommodated too.
> I'd like to increase R1 to 470k as well.
> Perhaps Terry can try these new values in his model. Enjoy!
hehe. Thanks for the corrections!