Thanks John,,

All cables, including the short fruns from the BOB to the steppers, the cables out to the motors and the returns for the proximity sensors are screened and grounded at a common point. The false trips I get are for a limit switch, the same problem I got when I first switched over from microswitches to proximity sensors. I ended up fixing this by using relays as isolators so that there was a dead short to ground for the BOB inputs at all times except when the limit is triggered. The BOB itself sits inside it's own screened box. I haven't been able to check if it's the BOB circuits that trigger or the noise is going direct into the parallel cable and upsetting the controlling PC.

Investigations with an osilloscope make me sure the problem is leakage from the VFD mains wiring. The scope shows very spikey noise all over the earthed case of the controller, the PSUs inside it, the BOB supply rails and all the terminals on it as well when the VFD is running. Changing the timescale shows each pulse to be a burst of several cycles of about 6MHz. This could be a fast edge triggering ringing somewhere but I can't delve that deep. Using a more remote mains socket for the VFD fixes the fault, though I have not measured the change in noise inside the controller. Further investigation of the cause would be an interesting project but I don't have time for that, I just wan't the machine working! Fingers crossed the mains filter, possibly with some filtering on the earth lead as well, will sort it out.

Have you been following the thread 'Why Star Ground?' at all? There's been quite a lot of discussion on there over recent days.

Kit