Right but I thought the initial issue was that the switch wasn’t triggering. Wouldn’t noise cause it to trigger at the wrong time, rather than not trigger at all?
That’s what I thought but if the machine is not in a homing cycle and i don’t have the switches set as limit switches there would be no impact on machine operation. In the above post I meant to say 0.7 volts peak to peak. I really don’t know why the switch is not triggering since i would think that going from 5 volts with the switch open to zero volts (ground) when it is closed would be a significant enough change to cause the homing to trigger in the grbl software. In any case I can’t get the homing to trigger at all.
I think the last thing to try might be taking the switch out of the equation completely. Could you attach a bit of wire to the appropriate gshield z-axis pin, start your homing sequence, and then touch the wire to ground? I agree with @paulkaplan that noise on the line seems unlikely to cause a false negative.
I have already done that. To check this I disconnected my signal and ground wires from the limit switch and used them to short directly to the pins on the gshield. I first checked the signal wire directly to the ground pin and the homing signal did not trigger. I next checked the ground wire directly to the signal pin and the homing sequence initiated properly. I have replaced the switch in case it was bad but unless I have two bad switches in a row it did not make a difference. It is interesting that putting the signal wire on the ground pin did not activate the homing response. Still baffled.
I was dissatisfied with the connection I was getting using the provided black header to connect to the gshield’s pins, so I ended up soldering my wires on there directly. I guess maybe that’s the next-next step: instead of just directly touching the test wires together, try putting a switch in there. Maybe your z-axis limit switch wiring became frayed, or there’s an intermittent signal due to a bad connection at the shield?
I had the noise issue with one switch. But it always read as closed not open. You could see the noise on the scope, and adding a cap cleaned it up, but the voltage dropped to about 4.7 which was low enough to be read as closed. Running a new line for the switch and not putting it in the chain with the other wires fixed it. I still don’t know the source of the noise, as it would be there when the steppers were not moving and the spindle was off.
Gremlins, obviously. Or it was haunted. Possibly haunted gremlins.
1 Like
Must be the power supply then? Maybe adding a choke to the switch cable or twisting the wires would help.
I finally got my homing switches to work. Installed shielded cable and that did the trick along with changing to a tinyg board.