ERIK,

IIRC, you once said that two circuits are not isolated from one another if they're powered from the same source. That's certainly true.

Now, It's been suggested that what he needs to do is to lower the input impedance of his inputs. That makes sense. Kai recommended adding some capacitance, which, of course, will reduce rise/fall times. All of this makes perfect sense. None of this should be difficult. If the interconnection between supply GND and circuit GND is made at the far (remote) end of the cable, then it probably should be made at both ends.

My personal preference, in cases like this, is to AC-couple GND and earth. I don't like direct paths from ESD-loaded fingers to my circuit GND. ESD isn't the issue here, but it could be that coupling between the AC motor supply and the circuit GND is affecting things. Lowering the impedance of his inputs will reduce susceptibility to coupled noise, but it won't help with the noise induced in the GND. Lowering the impedance of the GND back to the supply will help.

He said he had a "blown-out" '245. Those are, IIRC, pretty robust circuits and shouldn't fail from typical stresses. There must be something really major going on, and I don't think it's come to the surface yet.

Perhaps the resistance of the circuit-GND-to-supply-GND connection is a factor, but it will be nearly zero if it's connected at both ends of the cable, won't it?

RE