Re: Switch debouncing and related stuff (like CPU reset circuits)

[Sergey <skis...@gmail.com>]

Wow, that's a lot of replies. Thank you all! :-)

The comment in Intel ATX specs regarding switch debouncing is about
the /PS_ON signal itself. It is quite possible to use a toggle switch
connected directly between /PS_ON and the ground for controlling ATX
power supply. I wanted something that will work with a push button
that normally comes with an ATX case.

I was considering using a simple micro (e.g. Microchip PIC12Fxxx), but
it looks like a bit of overkill, and it is one more component to
program. My current solution is a bit more complicated (74x14 + 74x74
+ a few discrete components), but looks more like 80's one.

I did try using a 555 timer as single shot multivibrator for
debouncing, but it has one problem - the 74x74 itself needs to be
reset to a known state (so that by default, after connecting the
power, ATX power supply will be off), otherwise it ends up being on or
off randomly each time power is applied (well, maybe not randomly, but
theoretically it could end up in any state). Since 74x74 doesn't have
Schmitt triggers on /SET and /RESET inputs, it is not correct to
connect an RC delay reset circuit directly to these inputs (while most
likely it will work, sometimes it might not), so I needed to use an
external Schmitt trigger for that anyway. So I ended up using a 74x14
for that, and for power switch debouncing.

I've seen some microcontroller supervisory chips before (I think P112
uses one of these), but they are mostly overkill in our designs. Also
they don't solve the contact bounce problem for the power on switch.

Regarding John's circuit - most probably it will work, but I'd connect
a resistor between the switch and the ground to reduce the current.
The problem with it (and with any RS flip flop debouncing circuit) is
that it requires a SPDT switch (which I don't have).

Thanks,
Sergey