Jinx,

This is a hip-shot, but it might be worth investigating:

It might be that when the unit is turned off, the 555 keeps running
for a while off the charged stored in its 470uF supply capacitor.  In
principle it "shouldn't" be doing anything during this period, but my
guess is it could be continuing to run as a long-period astable
multivibrator off the leakage current in pins 2 and 6, putting out
brief pulses at a low rate (a couple per second, perhaps).

As this happens, the FETs are still seeing these pulses as drive
(albeit flakey drive); and they are still connected through the motor
to unswitched +12V.  As the 470uF supply capacitor slowly discharges,
the gate drive pulses to the FETs become weaker and weaker, until a
point is reached where the FETs no longer switch fully on, but instead
are only partially conducting.  When this occurs, these FETs are going
to be dissipating a LOT of power (hundreds or perhaps thousands of
times as much as during normal operation) and perhaps this is what's
frying them.  In other words, they might be burning to death instead
of being electrocuted.

It's also possible that during this wind-down period following turn
off, the 555 output at some point in time simply goes HIGH and stays
there; and from that point until the 555 supply voltage drops below
the Vth of the FETs, they will REALLY be dissipating power.

I haven't done any PWM motor control in years, and way back then only
with commutated SCRs, so I'm not enormously confident of the scenario
I've described; but it sounds borderline plausible and might be worth
checking out.

Having had a couple of beers already this evening, I'd probably better
refrain from proposing a fix.

Hope this helps...

Dave

Jinx wrote...

>A friend of mine has asked me, in desperation, if the list
>has any comments about his circuit.
>
>http://home.clear.net.nz/pages/joecolquitt/12vmotor.html
>
>The problem he's got is that this circuit tested OK on the bench,
>and 2/3rds of the finished units seem to be alright in use, but
>the others fry all 3 FETs short circuit, causing the motors to run
>full tilt, causing a scramble to disconnect the battery. It even
>happened twice this week, on separate occasions, with units
>that were seemingly turned off, then spontaneously burst into
>life a few minutes after being used. I've suggested a varistor
>across the FETs, but that doesn't explain why the units start
>on their own. Any clues ? (personally I'd have made the circuit
>more fail-safe, but he's very cost-conscious. Perhaps to the
>extent of false economy it seems)

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics