Y axis bad when torch fires

[burns cnc art]

ello,
Just finishing up my build and having some issues. I have CandCNC Ethercut Plazpac ,DTHVIV, with Feather Touch purchased this year. Also have the Candcnc computer and software. The problem I'm having is the Y axis runs great up until the torch fires. Then you hear the ratcheting sound. I have tried running it in manual and manually firing the torch with the same results. It runs great until the torch fires. It will run the gcode test pattern I made with no problem as long as I turn the Hypertherm off. swapped the Y with A (slave) at the control box and the problem stayed with the motor.
I have it grounded per the instructions, I have moved grounds , added grounds, and I have everything unhooked now and just using the z axis switch to touch off . I am definitely stuck ! I know its a noise issue but I just can not figure out where. The motors all sound great even running a 600 ipm when the torch is off.

[miller7596]

I had the same issue when I built my table. I ended up having to get a braided copper wire loom and ran all of my wire through the loom, ground the loom only on the end next to control cabinet, this will take any interferences collected and send them to the earth ground. Fixed my problem with the motors jumping when the torch fired. Hope this helps

[acourtjester]

I wonder if you had a typo as you said Y axis at the start of the description and then Z axis later with only it hooked up.
You said when switching motors the problem stayed with the motor. I would say the problem is in the motor itself.
Two other things do you have your plasma torch cable separate from the other wiring, and have you checked the ground at the power panel that the Hypertherm is powered from.

[tcaudle]

one subtle thing another user on our Support Forum found: Never coil up excess torch leads or workclamp leads under or near the controller or the plasma cutter . You plasma leads are giant emitters of noise . Keep then away from running close to and in parallel with other wires that can act as antenna to pump noise back into the control box. The co mingling of grounds between the control side (PC, Monitor/ Control box ) and the plasma side via LOCAL co shared grounds is a conduit for noise.

Another thing you might try is to reverser the STEP polarity (NOT DIR polarity) on the signals. Sometimes that will help make the signals less noise sensitive.

Since it only happens with the torch fired its noise related and there is not enough noise in any shop to effect a motor. It has to be getting inot the low level (logic) side of the control.