OpenServo.com

[dsandber]

Am I correct that a firmware error could cause current to "shoot-through" the h-bridge and release the hexfets' souls?

I'm choosing h-bridges/FETs for the next-gen controller for my robot, and was considering using the same components as the OpenServo rather than the currently used Allegro 3959 all-in-one chips... I know the OpenServo's parts were selected to minimize PCB size, but I thought I should check regardless...

Thanks!

-Dan

[guru]

It would be *possible* for a firmware error to "release the fets souls". This would occur if the error happened to set the h-bridge pins to output and also output an active signal on both of them. There is probably a pretty low chance of this happening and I dont read about it happening with open servo.

By default during power-on the pins are set as input and thus high-impedance and thus hbridge is off, this would be the usual things that happens on a firmware error. If you fiddle with the programming of the openservo regarding the PWM output, a bug could cause both outputs to be on. The PWM code is pretty mature so this is probably unlikely to occur anymore either.

Basically, what I am saying is you should be safe.

C

[ginge]

I only let the soul out of one fet while testing the v3 code

Incidentally that is where you need to be careful. If you use a bridged based on the version 3 design, you could get to a situation where all bridges are always on, even if the code says they shouldn't be, Be sure to enter some small delay between bridge transistions to avoid the fets locking into an unknown "not high and not low, but between" state. This gives the fets time to discharge before they are switched, and a clear dischange path from the motor through the fets.

Barry
_________________
http://www.headfuzz.co.uk/
http://www.robotfuzz.co.uk/