plasma torch switch relay for 5i25 / G540

The combo of tying the grounds together and inverting that ouput did the trick!

spindle off (default LinuxCNC startup condition or after M05):

IN1 = 2.68 V
relay NO = open
relay NC = closed
LED on relay module = very dim
Could hear relay click off with M05 command.

spindle on (M03, no speed specified):

IN1 = 44.6 mV
relay NO = closed
relay NC = open
LED = on, bright
Could hear relay relay click on.

When I closed LinuxCNC, IN1 went to 1.41 volts and the relay clicked on just like with an M03 command. So I'll have to remember to wait to turn the plasma on until after starting LinuxCNC and turn it off before closing LinuxCNC so I don‘t end up piercing extra holes in things.

I wonder if I should set the jumpers on the relay module to active high mode and uninvert that output pin on the DB25. Thinking that might solve the issue of the relay turning on when Linux CNC is closed. Will have to try that tomorrow...

No love with the active high configuration. The output pin would go high when the spindle was commanded on, but the 2.68 V is just not enough to drive the opto-isolator and switch the relay. So back to active low configuration. I think I'm going to put a SPST switch on one of the torch switch lines coming from the plasma. That way, I can "arm" it right before running a job and then "disarm" it right after the job finishes. Perhaps a better long-term solution would be to get a 3 / 3.3 V relay module and see if the active high configuration works with that. It looks like they make similar relay modules with that voltage and little power supply boards where you plug in a mini/micro USB charger and get 3.3 V as one of the outputs.

Also, a note on the M03 command not needing a speed specified: this does work, but only after specifying a speed with the first M03 command issued. I just did "M03 S100" and after that it would work with plain "M03" commands.

As a rule, plasma is the last thing to turn on and the first thing to turn off.

Absolutely. Best and safest practice.


Rick G

I'll have to see what opto is on the board and dig up a datasheet or just take some measurements with 5V driving. You'd want the same amount of current flowing through the left side of the opto with 3V as you would with 5V so you don't over-current the opto or the LED downstream. Like Tommy said, that R1 could be lowered to get more current with the lower driving voltage. But not too much (current)...

Not too much

Took some measurements today:

VCC = 5.08 V
V_R1 = 2.14 V
V_F = 1.10 V (anode to cathode of the input side of the opto)
V_LED = 1.84 V (the SMD indicator LED)
i_F = 2.14 mA

V_R1 + V_F + V_LED adds up to VCC, which is good. If I get rid of R1 (jumper over it), then V_F + V_LED = 2.94 V, which is not too far off from the 2.68 V that I was getting from the active high output on the DB25 pin. Hopefully it'll still provide enough forward current to make the opto function correctly. If proportional (2.68 / 2.94 * i_F_original), new forward current should be about 1.95 mA.