[bald9eagle]

Yeah I bought it from them. Vectric.com. Downloaded straight off the website and emailed a registration code.

[bald9eagle]

Alan...right now I can't post the code. What I know is that it is doing the same thing on every job. I can create new code and it does something similar. I has posted another question asking if you can have backlash on a RP system. I thought my motors weren't synced. I had some lag in my slaved motor but found the setscrew on my gear was loose.

[Spk64]

If steppers are stalling or couplings slipping.
Running same code back to back will show that your zero has moved.
If the second pass retraces its steps exactly then some thing else is going on.


Steve

[bald9eagle]

I'm downloading pics right now Spk...I just ran out and did a test run.

[mactec54]

bald9eagle

Just a long shot, does your I &J in the General Configuration, look like the photo, if it does not, you need to change it to how this photo is

[bald9eagle]

[bald9eagle]

Don't know about I and J....I'm a noob so that means little to me.

[bald9eagle]

The pic above is two consecutive runs of the same circle. You see the consistency of where the circle stops. Had it been mechanical I would have expected some deviations.

[bald9eagle]

Another pic of messed up circles. Again this was the same program run twice and it hit exactly the same spots on both runs.

[ger21]

Whichever axis moves up and down, something is probably loose.

[mactec54]

bald9eagle

Looks like you were trying to cut the cut the Apple logo

The I & J setting is important, but I doubt that this is the problem, it seems your axes are not setup correct, either steps/per, or machine alignment is out of whack

To check the I & J look in the General configuration settings in Mach, it is in the middle of the page

[bald9eagle]

I would have thought the same thing Gerry if not for repeatibility. Is it possible to be that consistent with loose parts?

[bald9eagle]

Mac....no doubt that my gantry is out of square. My issue is what is causing it to do so. What has me stumped is that it ran fine early on but is doing this now without me changing anything. To be honest it has gotten much, much worse.

[Spk64]

Looks like there is some backlash somewhere. Appears to be a lot.
Backlash can repeat the results on same program.
What type drive system do you have?
Ball screws? Rack and pinion? Other?


Steve

[wizard]

Soooo.....it ran fine for some time but now decides to mess up? What am I missing?

This is a Windows machine right? I've seen all sort of issues with copy protected software on Windows. Thing work fine on Monday and go to hell on Tuesday.

[bald9eagle]

RP. Backlash was my first thought. I actually could see one side move before the other but don't know how to fix it. I did a little with the backlash stuff in Mach but didn't know what I doing. I didn't know if backlash could be had in a RP. I had only read about it with ballscrews.

[bald9eagle]

That's what someone else hinted at Wiz except he thought it was a Vectric issue.

[ger21]

You don't want to use backlash compensation for that. It's almost certainly a mechanical issue. Go over it with a fine tooth comb. Something is not right.

[Spk64]

If you post some pics of the drive system we might be able to help.
If you are seeing some movement from one side and not the other, that will create some bad geometry.


Steve

[Spk64]

Agree with Ger21. Need to fix the mechanical problem, do not try to compensate with software.


Steve