Input:
Output:
Have a suggestion?
Is this a LASER etcher? Is the LASER stationary or moving?
If moving; Check the wires that are being moved while etching. Check their connections while they move, especially to the left side. There should be a good "service loop" that lets all mechanical devices move freely, without tension or binding.
Maybe a problem with signed/unsigned integers? Or, maybe integer overflow?
i.e. If you try to go over 32,767 with a signed integer the sign bit flips, and because of the 2's compliment representation, things get screwy.
Sometimes it's not obvious because there can be scaling or other operations and you never "see" the actual value that's causing the problem.
I use lasergrbl and my device moves but I think its a problem about software or sth like this as it always occurs for this shape!
I see. You have tested another shape with good results?
The "top, right" values should be similar to the "top, left" values. Can you "cut" the shape vertically, in half, then mirror the good half? BUT FIRST check @DVDdoug suggestion.
In the upper right corners of the screenshots, I see negative numbers and I'd guess that's a clue. The first one has a negative X value and the second one a negative Y value. And the corruption is different in the images...
I don't know what those numbers represent (current position?) but do they look correct to you?
Are negative coordinates valid? You could have an error either way... Maybe they are supposed to be negative but the software library can't handle it, or maybe they are supposed to positive and they've overflowed to negative????
@DVDdoug No, the current position is the second X and Y and they are true. The problem should be sth else 
something
Make a physical measurement of the broken half... do any values have a relation to the bad offset? Does the "negative" do bad things? Can you arrange the origin to be 0,0 and not the negative values?
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.