Circles Still Not Closing Porperly

I’m looking to follow up on a thread that was closed. This is an issue that I’ve been dealing with since November.

@LightBurn your attention is sincerely appreciated

I’ve looked at the files, but I can’t see any reason at all that RDWorks would perform better than LightBurn with this. The files you sent, when simulated, are run in a different order, but the actual cutting path data is the same - it’s just the order that changes, and they look worse than ours, not better, when simmed. I don’t have any good way to debug or test this, since I have no machines that exhibit this behavior.

Is there anything I can do to aide, assist, or otherwise?

Short of sending me your machine, I can’t think of anything. To be clear, it only works when you use the ‘backlash repay’ setting in RDWorks, correct? Otherwise the output is similar? You also sent me a file that had very small circles inside larger ones. Do you get the same misalignments if you exclude the small circles?

lol, if you were anywhere near me, I’d consider sending you the unit for sure hahahaha

Yes. My shapes only cut perfectly when the ‘backlash repay’ setting in RDWorks is enabled. Otherwise, output is very similar.

‘Hide backlash’ in Lightburn half works - meaning it clearly changes the starting point, and does create “perfect” shapes, but not on 100% of the job.

And yes, with or without the small holes the results are the same

I’ll have to see if I can figure out what RD’s ordering rules are, and why theirs actually works. Like I said, when simulated, their version looks worse than mine.

This reply to your original post was comprehensive and thorough

It’s clear to the poster, and to me, that your machine is not true. All the software compensation in the world cannot alter that, no matter how much you can try and mask it.

The machine needs to be trued and aligned before you make tuning adjustments in software.

I’ve gone through all of those steps. Intensely. After all the aligning and adjusting, the results remained the same. I’ve spent countless hours and have had others investigate in person as well. Yet, here we are and here we remain.

I cannot stress enough how much I appreciate your time.

You probably need to enlist the help of a mechanical engineer to true the machine.

Logic dictates that you are using a tried-and-true controller with tried and true software and getting unsatisfactory results, it’s a problem with your machine. All you can do is to try and mask the problem, but it won’t go away.

Start with the legs the machine is standing on - flat and level- and work your way up.

I use a digital inclinometer to start with to ensure the bed and the rails are level. There’s no particular reason to be level, except the earth makes for a useful datum. The key is that they are level relative to each other.

Another useful tool is a digital square, to measure real angles.

Of course you could use engineers equivalents, but they tend to be much more expensive

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.