On Tuesday 01 December 2020 18:11:39 Gene Heskett wrote:

[and snipped]

But I would like to be able to try at_pid but can't. I think the reason 
it failed after the first and only move it made, was because it killed 
the psu stopping at the end of that move, and the led on the psu was out 
of sight so that failure was not noticed until much later when I had 
re-installed the regular pid and restarted linuxcnc.

To facilitate testing, it needs to have a test setting that limits its 
speed and accel to something the hardware can actually do, but I wasn't 
able to get it to do anything until I had set it for very large moves, 
at which point it ran the motor wide open for about 2 minutes, and 
jumped off the ladder step stopping, jerking the motor wire out of the 
driver connection as it did so.

So can it be patched to also accept a maxtunespeed setting? And do 
something without having to give it an effort setting that is equ to at 
least a full turn of the bs-1 when it is finally assembled?

I now have some settings that are "in the ball park" placed in the ini 
file and its acting decently. But I would like to be able to use those 
settings as a starting point for the autotune mode.

For starters, the manpage, which seems to be the only docs we have on it, 
needs to describe the halscope presets, so the improvements it makes can 
be made easier to see.

I am more than willing to bleed a little to make this at_pid module more 
usable for everyone. As is, it seems bent on self-destruction of the 
hardware its controlling.

A maxtunespeed option would be a great first step in that direction.  And 
accepting a much smaller "effort" setting would be another step in the 
right direction, as it should only need to run far enough to reach the 
cruising speed that maxtunespeed allows, which can be achieved in this 
case in about 1 longish second.

To that end, I'd suggest terminating that "cycles" test move when the 
encoder velocity says its no longer accelerating or that its too fast 
and windup is occurring.  And log why it stopped that move to the 
terminal that launched LinuxCNC as the fix is different in each case. 

This also should be noted in the man page as educational input to the 
person doing the testing. In the case of windup, an automatic 20% 
reduction in the maxtunespeed, repeated on a per tune cycle basis until 
windup is tolerable, and logged to the terminal (or to a file) seems 
like a good idea.

Discuss?

Thank you.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page <http://geneslinuxbox.net:6309/gene>


_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to