Nope, fast PWM works just fine in momentary drivers… I use it all the time.
When programming PWM values of 0 or 255 in any PWM mode, you should check for spurious PWM pulses using an oscilloscope. Figuring out what the PWM generator does at those value by reading the data sheet can be confusing. I always disable the PWM generator and manually set the output bit high or low when using those values… but that takes some extra code space that you may not have.
It works fine, with whatever code JonnyC uses for the sleep function. If it turns off and changes modes and doesn't do anything noticeably wrong, is there anything else to worry about?
Like what? Beyond battery drain I don’t see what the concern could be. As texaspyro said, as long as the scope shows a real “off” state then it should be OK.
Sort of… PWM’ing of AMC7135 chips by pulsing the Vdd pin is not actually specified in the data sheet. There is no guarantee it will work at all! There can be a lot of variation in how they behave. With the narrow pulses needed to do moon modes, some chips turn on… some don’t. And it varies with temperature, voltage, age, sex, hair color, etc.
