From: Winfield Hill
Subject: Re: zero-power toggle circuit; was, how to master electronics
Date: 14 Nov 2002 06:02:47 -0800
Organization: Rowland Institute
References: <3DC517EB.email@example.com> <3DC57B4E.firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com>
X-Newsreader: Direct Read News 4.10
>"Bill Bowden" <**firstname.lastname@example.org*> wrote:
>> Looks similar to a debounced MOSFET toggle switch I did
>> last year and Win helped me fix it. But it's not zero power and
>> uses a NPN and one less resistor. Came out to 470K and 220 ohms.
>> The extra 10uF and 10K and diode ensures the thing powers up
>> with the MOSFET off.
> Thanks Bill, another nice one for my files. BTW, I see both that
> circuit and the 'Single Transistor Relay Toggle Circuit' directly
> below it sport 2-pole buttons; guess you must have some surplus stock?
> As I've occasionally mentioned here, the garage light toggle circuit I
> built three years ago based on your 'CMOS Toggle Flip Flop Using Push
> Button' is still in daily use.
> I improvised each of my momentary contact push buttons (at three
> convenient locations) from a hinged 1" x 4" piece of wood, with a
> couple of drawing pins (US = thumbtacks) as contacts, plus a small
> spring from the junk box. So I can toggle the lights by a brief tap,
> if necessary with a wrist or elbow when carrying stuff.
Yeah, the extra switches on Bill's drawing were to display the concept
of using this circuit with several remotely-located pushbuttons.
Here's a repost of a circuit for use where a single contact to ground
is desired for the pushbutton, such as in an auto or truck. It also
features an optional reset button. Of course any toggle circuit can
be adapted for grounded pushbuttons if one used opto-FETs. :-)
. ,---------------------+---+-----(O) +13
. | | |
. 1.0M 100k_|_
. | | /_\
. +-------, ,--330k---+ |
. ,------+ any \ / +---+-----(O) SW
. | Q1 | npn / |
. | 2n c / \ |--' Q2
. | 4401 b --+-' '--+--|| IRF540
. R1 | e | | |--,S
. 4.7M | 10k 1k |
. | gnd | | +---------(O) COM
. | | +13 | |
. C1 +-------o o--' | | gnd
. 0.22uF | ,-|<|--2.2k--' |
. === | optoFET |
. | | |
. gnd (O) (O)
. /toggle /reset
This circuit is not zero power either, but it only takes 16uA when off.
Bill's circuit could be recast for lower power consumption as well. Two
issues come to mind, sensitivity to false triggering with long-distance
pushbutton wiring (the optoFET solves that problem) and slow FET turnon
due to a 1M gate pullup.
For example, an IRF540 has say 1600pF of input capacitance and is spec'd
at 50nC gate charge, so that's 6ms delay for full turnon, with 3ms for
the drain-voltage fall time portion. This isn't a problem. For example
say the load was a 75W 12V lamp, drawing 20A when cold. The FET spends
3ms going from 12V drop to 0V, dissipating less than say 200W dropping to
0 in 3ms, or under 250mJ during the turnon, which won't bother it at all.
Note, the FET always turns off quickly, because Q1 pulls down with about
250mA of sinking capability until C1 is discharged (dropping to 5mA).
This is a useful property when driving high-current inductive loads such
as motors, where the full current continues to flow through the FET, with
up to 13V drop, until it's fully off.