On Tuesday 05 April 2016 21:53:21 Gene Heskett wrote:

> Greetings all;
>
> Background: Del Dimension 745 using our wheezy iso install, with a
> 5i25 card, driving 3 axis's of steppers, one of the Pico PWM-Servo
> amps running the 1 hp pm-dc OEM spindle.
>
> Having a few to play after getting back from din-din, I went thru the
> ini file and raised the Pgain of the steppers PID's to 4000, set the
> servo thread down to 250 microseconds and fired up LCNC.  Then ran the
> G33.1 peck cycle for about 15 minutes, 3 times, while I attempted to
> get halscope to show me the data for one plunge, but didn't fare that
> well, one plunge takes  more time than the halscope can watch in 1
> sweep.
>
> What I did see indicated that once the moves had started, it was quite
> stable.  So I let that run and hit a couple news sites with iceweasel,
> that, and LCNC seemed not to be suffering because the machine was too
> busy while I watched a news video or 3.
>
> I think I'll leave it that way. ;-)
>
> Cheers, Gene Heskett

However, i do see one point that could be troublesome.  The Z turnaround 
at the bottom of the stroke. It has developed a backlash in the .00675" 
area, the source of which I haven't located yet but am suspecting the Z 
axis thrust bearing as its nut for that is not double nutted.

I do have a limit3 in the spindle drive path intended to soften the 
stress on the motor, and I can further limit the ramp speed presented to 
the PID.S.command input with that, giving Z time to make its backlash 
move without losing a count.  When this occurs, it does not seem to 
affect the end position of the upstroke, but has resulted in its moving 
relative to the soft limit, so that if the g33.1 peck is position 50 
thou below the up soft limit, some time after it has made 20 pecks, each 
time hitting the home switch an indentical amount, but somehow the 
machine counts have been lost and it thinks it has hit the upper soft 
limit while still physically up to .1" below it.

While I can slow the ramp in the ini file and have done so, this also 
expands the overtravel, so what I'd like to do is, just as the spindle 
reverses, I'd like to freeze the output of that limit3 for the few 
milliseconds it takes Z to accomplish its backlash takeup move.

This is not directly doable for 2 reasons, first being that limit3 has 
no "freeze" input, effectively removing it from the servo-thread execute 
loop,  And motion has no backlash move in progress output pin.

I can probably cobble up something to generate a slight pause in the ramp 
at that reversal point, but it really seems like it should only be 
effective until the backlash move has been completed. 

Hints on developing that "backlash move in progress" signal from what we 
do have would be much appreciated.

Thanks all.

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)
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