PWM output on 5i25, seeming discrepancy in pins

More
14 Jan 2024 05:00 #290610 by shasse
I am attempting to get a PWM output from a Mesa 5i25 card have a firmware with 3 stepgens, 1PWM and 1 encoder. I'm running LinuxCNC 2.8 and am using PNCConf to configure everything so far. Running "halcmd show all | grep pwm" shows mappings like the following:

hm2_5i25.0.pwmgen.00.out0.invert_output hm2_5i25.0.gpio.007.invert_output

which makes me thing that the PWMGen is on output 007 (DB25 pin 17), but I am not seeing output on that pin with a (physical) oscilloscope. I also still see output 007 showing up in pncconf as a configurable GPIO. PNCConf UI only allows my to pick "PWM Gen-P" as the second I/O in the list, after 000 and before "Step-Gen-A". I thought this implied the PWM would be on I/O 001, DB25 pin 14.

I am planing to try an update to LinuxCNC 2.9, but I am hoping someone can help me sort out what seems to me to be a discrepancy.

Thanks!

ScottĀ 

Please Log in or Create an account to join the conversation.

More
14 Jan 2024 05:10 #290612 by shasse
Adding some detail, I did add the PWM after I had already configured some I/O, and reading the docs (linuxcnc.org/docs/html/man/man9/hostmot2.9.html#pwmgen), it says that the PWMGen will take 3 I/O pins. If I did not have that many (or specifically the ones the pwmgen expects) available when I added the pwmgen could that cause when I am seeing?

Please Log in or Create an account to join the conversation.

More
14 Jan 2024 05:17 #290613 by PCW
Is this with a running LinuxCNC configuration?

Is the PWMgen enabled and have some non-0 or fullscale
value setting?

The GPIO input pins of secondary output functions
like PWM pins are still be available (but no output pins
or is_output parameters should be present)

Please Log in or Create an account to join the conversation.

More
14 Jan 2024 18:01 #290679 by shasse
I looking at this to answer your question I realized what the issue was. Enabling the spindle via Axis set the PWM value to 1.0. I thought this was on a scale of zero to 1, but it was actually on a scale of (in my case currently) 0 to 2000. So I was getting a PWM pulse, just way less than I was expecting. When I set the spindle speed via S 2000 and then started it via M3 then the PWM waveform was what I expected.

Thanks for the "fullscale value" prompt that helped me figure this out!

Scott

Please Log in or Create an account to join the conversation.

Moderators: cmorley
Time to create page: 0.148 seconds
Powered by Kunena Forum