Hello Guest it is April 19, 2024, 04:13:38 AM

Author Topic: x axis drift  (Read 3491 times)

0 Members and 1 Guest are viewing this topic.

x axis drift
« on: May 18, 2007, 07:56:09 PM »
There is likely a post about this, but I'm too tired to search for it.  My router is losing its position on the x axis only. It seems totally arbitrary, but the x axis will shift to negative, by as much as .25 inches. So far I've destroyed a couple of really nice pieces of wood.
Run the g code once, and things go fine. Next time - whamo, I get gouges where I really don't want them. Shucks, I want to cry.
I generated new g codes, checked tuning on my G320's (gain and damping) nothing to tweak that I can see. I clean the linear bearings, lubricate racks, talk nice and offer up animal sacrifices....
Sometimes, but not always, I have noticed a hesitation when the x axis is cutting to the positive direction. Rapid movement, it seems, is not affected and works smooth as silk.
This problem appears to have happened since upgrading to version 2 of Mach 3.
I don't think it's noise, because I'm running with shielded cables and bleeds. Upgraded grounds, yadah yadah, (I had another post about this boar's nest)



Any guidance out there, this situation has effectively shut me down.
Re: x axis drift
« Reply #1 on: May 18, 2007, 08:22:20 PM »
What do you have for a breakout board?
Fixing problems one post at a time ;)

www.newfangledsolutions.com
www.machsupport.com
Re: x axis drift
« Reply #2 on: May 18, 2007, 09:35:54 PM »
Bob Campbells' breakout
4x Gecko drive g320
Ametek motors
Sheilded cable harnesses to the encoders
US digital 500 cpr encoders
Windows 2000 on a fairly new pentium 4

Re: x axis drift
« Reply #3 on: May 18, 2007, 11:50:04 PM »
That is all good stuff ... and  it is only the x axis hmmm is the X axis your longest cable?

Try setting your accel to about half and see if you still have the problem
Thanks
Brian
Fixing problems one post at a time ;)

www.newfangledsolutions.com
www.machsupport.com
Re: x axis drift
« Reply #4 on: May 21, 2007, 10:08:05 PM »
I went a little mental with my fix.
I cut the accel as suggested.
Slowed down cutting feed.
I also had some line drivers on hand from US Digital since my encoder cables are rather long, (duh- I forgot I had them). So I made a new pigtail for the x motor, with line driver at the encoder and ferrite cores on the cable harness.
(The A  axis cable is the longest and it is slaved to the Y, but there are no apparent problems there.)
The result?
Smoothed motion, in fact I'm going to put new pigtails all the way around. However I still lose up to .25 inches somewhere.
It seems when I load the g code fresh, I get a nice, clean cut. Rewind and execute the program again.....problems.

I am perplexed by the perplexity of the situation.

Do you wanna hear about the fun I'm having with Bobcad 2007....? Arghhh.