Glad you fixed it - one day I'll get round to fixing the two broken M752's I have. Not had chance to look at them but I suspect some of the MOSFETs would need replacing.

Bit off topic but...
They're cutting it fine with the MOSFET ratings - both of our M542s use MOSFETs only rated for 55V absolute maximum. Considering they're driving an inductive load and the manual states they can be used on up to 50V I'm surprised the voltage spikes on switching do not exceed 55V. The M752 (75V) and 2m880N (70V) have a much more realistic safety margin as they both use IRF540 MOSFETs which are rated for 100V. I therefore run my M752 drivers on the rated 75V having also checked with oscilloscope that the voltage spikes do not exceed 100V.

Quote Originally Posted by 2e0poz View Post
The part number is SMAJ48A-TR from RS for all those lurkers waiting for it
Ahh so it was a transient voltage suppression diode...I don't see why you would need to consider scraping the mask away to test.

Quote Originally Posted by Robin Hewitt View Post
When the circuit breaks the supply to a coil the coil impedance wants to keep the current flowing. If it has nowhere to go the Volts go through the roof, probably exceeding the breakdown point of the transistors.

The diode carries the excess current back to the top of the coil so it can go round and round until the coil resistance eats it up.
Yep, except the diode in question is across the supply not the individual MOSFETs in the H-bridges, however the principle is the same. I think the difference here is all the energy is immediately dumped into the junction of the diode, not necessarily going round and round. That diode should protect it against unplugging the power when running, but please don't try it!