MESA 5i25/ 7i76 connect to M542T stepper driver

More
07 Nov 2016 00:31 - 07 Nov 2016 00:32 #82524 by cts1085
I am having an issue getting my new 7i76 to connect to my driver - when connecting the driver to a parallel BOB - it works fine. I added the mesa config lines from pncconf to my existing hal/ini files for the "B" axis and connected the driver to the stepgen.03 outputs (gnd + Pul+ / Dir+) and the driver is not moving when i try to jog the "b" axis.

the driver is a M54T from www.omc-stepperonline.com

I have included my ini/hal files - any suggestions would be appreciated.

I am converting from a plain parallel bob to mesa and am doing it bit by bit - this axis will be the first.
Thank you!
Attachments:
Last edit: 07 Nov 2016 00:32 by cts1085.

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

More
07 Nov 2016 00:46 #82525 by PCW
I didn't notice any obvious errors in the hal/ini files
but the M542T manual says this:

For reliable response, pulse width should be longer than 1.5μs


So you might try setting the step pulse width to 2 or 3 usec

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

More
07 Nov 2016 08:47 #82529 by rodw
Hmm, I was looking at the manual for my DM542A which seems to be the same controller. I noticed that in some BOB installs on CNCzone guys mentioned they had to connect ground to the step and direction. Perhaps thats why I never got the drives working on a BOB with LCNC but I've driven them from an Arduino. The wiring is documented in the instructions and I made a mental note to be prepared for problems when I wire them to my Mesa 7I76E in the next week or so. I'm wondering if the Mesa sides with the Arduino or with Mach3 BOB.

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

More
07 Nov 2016 15:57 - 07 Nov 2016 15:59 #82548 by cts1085
ok - so i increased the steplen/stephold to 2500 - no change. (BTW 1800 worked on the parallel BOB)

I also tried various ways to connect the 7i76 to the driver:

1) 7i76 pul-/pul+ --> Driver step-/step+
2) 7i76 GND/pul+ --> driver step-/step+
3) 7i76 pul-/Vcc --> Driver step-/step+

(Did the same with dir as well)

No difference. the UI will jog the "B" axis just fine - nothing happens on the motor end.

Also - tried to invert the signals as well - hal/ini attached...
setp    hm2_5i25.0.gpio.006.invert_output true
setp    hm2_5i25.0.gpio.007.invert_output true

Any other troubleshooting steps???
Attachments:
Last edit: 07 Nov 2016 15:59 by cts1085.

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

More
07 Nov 2016 16:10 - 07 Nov 2016 16:56 #82549 by PCW
You might check the B direction pin to see if it changes when jogging different directions
just in case there might be some hal issue

Also you might try a longer step length (parallel port step lengths are often a few usec longer than programmed due to access time issues)

Another thing to verify is that the 7I76 5V supply is within specs
Last edit: 07 Nov 2016 16:56 by PCW.

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

More
07 Nov 2016 17:31 #82556 by cts1085
ok - so increasing the steplen/stephold to 3500 from 1800 worked. Does this mean that I will need to increase the steplen/hold for my other drivers as well? I guess I will find out.

Note: for others - it works with simple wiring - Pul-/Pul+ -> Step-/Step+ and Dir-/Dir+ -> Dir-/Dir+ - and with the increased steplen/hold times i get a consistent stepper movements.

Thank you for the suggestions!
The following user(s) said Thank You: rodw

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

More
09 Nov 2016 16:31 #82612 by andypugh

Also you might try a longer step length (parallel port step lengths are often a few usec longer than programmed due to access time issues)


A parallel port step will either be one whole base thread cycle (if reset is not used), typically 20uS or the length of time that the base thread takes to execute (variable) if reset is used. Or pretty-much as-programmed if reset is used and the step length is set longer than the base thread execution time.
But, the point is, parallel port step time can only really be measured with a 'scope (and not Halscope) but will only be at least as long as programmed.

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

Moderators: cmorley
Time to create page: 0.090 seconds
Powered by Kunena Forum