Mesa 6i24 - high stepgen speed, axis go crazy

More
13 Mar 2016 20:26 #71583 by PCW
For the stepgen, say .5 mm or 20 mills for ferror and 1/2 of that for min_ferror
would not fault unless something is really wrong but would catch gross setup errors

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

More
14 Mar 2016 19:10 - 14 Mar 2016 19:22 #71650 by yaqwsx
I looked into the commit that fixed P value in PncConf and the bug caused only improper values when the servo period was other than 1 ms (which is not my case). So I think editing of the P value shouldn't help (however I cannot test it right now, as I am abroad for a week and a half).

EDIT: My bad, I missed a word; I use 1 ms servo period
Last edit: 14 Mar 2016 19:22 by yaqwsx.

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

More
14 Mar 2016 19:14 - 14 Mar 2016 19:45 #71651 by PCW
There is no question that editing the P value will help
unless you need to tweak things with pncconf there should
be no issue with simply setting the proper value in the INI file

Pretty sure that unless you are using master, pnnconf will have the wrong PID value
( This is easily checked/fixed by editing the .INI file )
Last edit: 14 Mar 2016 19:45 by PCW.

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

More
25 Mar 2016 13:33 #72131 by yaqwsx
Thank all of you for your help - I missed one commit that fixed the bug. I upgraded to the 2.7 branch and it works perfect now.

However there's one more strange thing - when I invert step pin, no pulses to the stepper driver are actualy made. When the pin is not inverted, everything works (however there might be disalignment in step/dir signal so it is possible I loose a step every time I change direction).

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

More
25 Mar 2016 15:01 #72133 by PCW
if you invert the step pin you completely change the step timing (step length and stepspace) , so its
rather expected that it may not work without changing these timings at the same time

Also , as you mention, since the active step edge changes if you invert the step polarity, you may be violating
the direction setup and hold times (which are measured from the leading edge of the step pulse)

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

More
26 Mar 2016 09:00 #72175 by yaqwsx
That's not true - in my setup the signals are inverted (becouse of optocoupler), so step is active low and simple invert of the output pin is what I need.

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

More
26 Mar 2016 14:23 - 26 Mar 2016 14:26 #72182 by PCW
Actually what I said _is_ true, you _do_ change all of the timings the step drive sees
by changing the step polarity

Note that if you are driving active low ( Common anode) OPTOs directly from the 6I24 outputs
you will either have to tie the common to 3.3V or if the COMMON anodes are at 5V, set the output to
open drain mode to fully turn off the OPTOS when the outputs are high
Last edit: 26 Mar 2016 14:26 by PCW.

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

Moderators: cmorley
Time to create page: 0.074 seconds
Powered by Kunena Forum