User-Agent: Mozilla/5.0 (Windows; U; Win95; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0
X-Accept-Language: en-us, en
Subject: Re: zero-power toggle circuit; was, how to master electronics
References: <3DD39D90.firstname.lastname@example.org> <email@example.com> <3DD90A12.firstname.lastname@example.org>
Date: Tue, 19 Nov 2002 13:05:11 GMT
NNTP-Posting-Date: Tue, 19 Nov 2002 08:05:11 EST
Organization: Road Runner
Terry Pinnell wrote:
> But I'm still hoping to hear from circuit's sponsor, Win, or original
> author, Wafer. Particularly interested to know whether Wafer built it
> *only* with that specified low-power MOSFET, BS110?
> I had actually tried R1 = 10K, and C1 = 1uF with the IRF9513 and it
> still doesn't work. Just tried both combined, and that doesn't
> simulate either. Also, with some MOSFET types load is on at power-up,
> while with others it's off.
> So, whatever inherent flaws in circuit itself, I still suspect my
> technique or CM's models may also be implicated.
> That's why I'm hoping one of the Spice experts might try it too.
> Terry Pinnell
> Hobbyist, West Sussex, UK
OK. For the record, I actually also designed a version that used a PNP
in place of the P-ch MOSFET. It worked well, but of course takes a
measurable amount more current to keep the circuit on. My very original
circuit was designed using only bipolars, and was for 3.6V operation. I
replaced a slide switch in an old cordless phone. That was the whole
genesis of the the zero power thing in the first place. The bipolar
version was a bit touchy to get right because of the .6V threshold, so I
used a diode to keep the thing off with a charge under about 1V or so.
The 2N7000 gave much more of a voltage difference between the off and on
states, because the gate won't hold the voltage down like a bipolar's
base will. As I noted in an earlier post, some 2N7000's gate threshold
voltages are inconsistent, but the ones I have from Motorola are quite
close to each other, making it easy to get good results. I think the
version with the P-ch MOSFET worked down to about 6.5-ish volts. I
believe I tried up to about twelve, but since the idea was for 9V, going
up higher was not too important. I made the voltage divider off the
output (with the 220K) such that the turn-on and turn-off
characteristics were about equal. That is, it would take about the same
time for the circuit to prepare to switch on as it did to switch off. To
test this, I simply pressed the switch faster and faster until it
favored one state over the other, then adjusted the resistor value to
even it out, until it switched one way just as easily as the other, not
favoring one state too noticeably. Very scientific, ah? I don't care if
it doesn't check out in some computer model. I built it, and it worked.
I know that if I stick with the one manufacturer of 2N7000's, there
shouldn't be a problem. I did not test with high current P-ch MOSFET's.
Again, 9V operation. I recently ran into a postit-note with a similar
circuit I drew up for switching a power transformer. I think it's the
same as the phone one (only bipolars), so maybe I'll try that and post