You can see that the raster and the engraving have an offset that grows as it moves from one image to another. 1st one is okay, second has a 1mm gap, third 2-3mm, etc.
So as it works it is getting worse. If you run them one by one it will all be good.
I tried to use K40 Whisperer, MeerK40t, different versions, but with the same result.
I disabled the rapid moves, tried different options but nothing helped.
I also tightened all the belts, but without any luck.
I called the machine supplier, and he is certain that it is a software issue because the offset is always good on the first one and grows as the job goes on and always behaves the same, not sporadically. I couldnât find anything that can help.
I am wondering if you have any suggestions or had this issue before. Any help or suggestions would help.
Itâs based on the order in which they were printed, and not their location on the bed?
(I sure would have guessed laser out of alignment from bed position.)
Um, one more thought on position-based problem. Have you done a test shot straight out of the tube for which TEM the laser is emitting? If so, can you share that picture?
Yes, the position on the board is not relevant. It started in the upper left corner, moved to the right and then down. And it is progressively works as it moves from part to part.
Also, I tried creating a Ludo Board, and it is also out of whack
As you can see, the whole raster is on a constant offset from the engraving.
It seems as it is a time based error if we look at the ludo board, since it was doing all the raster and then the engraving. Following the logic of first job is good, it should have turned out nice.
Have you done a test shot straight out of the tube for which TEM the laser is emitting? If so, can you share that picture?
Can you elaborate? I havenât done that test. But it seems like the over and alignment is okay over the whole surface.
The position being irrelevant (that is, itâs not always going left-right, top-bottom) means that my idea doesnât work.
Iâve seen pictures on this site of test shots from tubes in degenerative TEM with asymmetric output that could make otherwise acceptable alignment fade, and if iteration had been a proxy for location it might have made sense.
I would have guessed belts, but you said youâve addressed that already.
Um. Is there any chance that you have a limit switch that has moved slightly, and you are hitting the hard limits of travel on the other side and forcing either a belt skip or lost step? Is your power supply dipping on the motor supply voltage, reducing torque and causing lost steps?
I think it doesnât touch the limit switches.
I wanted to try to find some test to see if the Y axis motor is losing steps or something like that.
But the constant offset for one, lets say group, is whats breaking me. It is too much rules to say that something is broken, I would expect it to be random.
As for the power supply, Iâm not sure how to test that safely. Do you have a process for that?
Motor voltage power supply, not the laser voltage. Should be 24V or less, which is considered low-voltage DC. The control board for the steppers should have a 24V (or possibly 12V?) header, and you should be able to measure that voltage with a normal VOM/multimeter.
I am not coming up with any likely answers, so Iâm kind of trying to help think through the unlikely answers.
Iâm not a laser guy, but if your machine runs off of GRBL, have you checked the calibration of the steps per mm for your axes? The first burn would be OK and the second if off by some percentage, and then the 3rd is moreâŚetc.
I donât know what a K40 looks like. My VG-L7x laser engraver is one of those that has an X track that is cantilevered on a single Y track. The tracks themselves are 20x20mm aluminum extrusions commonly seen in 3D printers. Three V wheels fitted to the X axis carriage (which travels along the Y track) and the Laser carriage (which travels to and from the Y track along the X track) pinch the tracks and keep everything rigid and the Y and Y axis perpendicular. One time I noticed that my squares and rectangles were coming out as trapeziums. The problem was that the V-rollers where not âpinchingâ the Y-track tightly enough, causing backlash.
The distance of backlash in such cases is clearly dependent on the position of the laser along the X track. Whatever the design of your own engraver is, there should be no backlash caused by inertia in any direction, and itâs a simple matter to check that there is no unintended movement on your laser head by just trying to rock it gently in all directions while holding it between your thumb and forefinger (when everything is off, of course) . It took a lot of trouble shooting before I realized what was going on, as many of us tend to assume that these instruments do not need regular checking and adjustment and that things never work themselves loose.
In your case, the backlash may be intermittent and Y-position dependent???
I hope that your problem is as simple as that and that it can be solved just as easily as mine.
I was looking at the image again. Your order of printing is left to right and top to bottom?
If so then the problem can definitely be a loose carriages along the horizontal tracks. The worst print is on the bottom row and especially towards the bottom right - The weight of the laser exerts the most leverage on the horizontal track that runs along the top of the prints when its printing the bottom row, and if things are loose anywhere, well, I hope that you see what I mean. If the order of etching is along the top, left to right then the bottom row - that might explain why you perceive the problem to be âtime relatedâ. I suggest check that none of the there is no backlash at several places along the track. Its possible that the wheels run tight at one end of the track but are loose at the other end. Note that even a small amount of play can make a huge difference at the opposite end of the track. Those suckers need to be tight!
Thank you both for the suggestions. But I am not sure it will be the problem in those. The machine uses profiled guide rails that have no slop and I tightened all the belts, so it would be hard for it to skip there, and expecially between 2 coasters and not somewhere while doing one.
If you check the bottom part of the first pic, you can see 3 âhousesâ they were used as a test. I ran them to combine engraving and raster and they turned out great. Also with the small messy geometric shapes.
It looks like the job size is impacting the ammount of the offset, like the error is accumulating in some way
My gut feeling is that this must be a missed step problem.
If it happens âbetweenâ individual sections then it is happening during long high speed travel moves, maybe the driver or stepper misses some steps at very high travel speeds. But the short travels within each section are not getting fast enough for it to cut in..
Try lowering your maximum travel speed on the affected axis, for GRBL see this.
Some members are suggesting misteps etc and adjusting the steps per mm. But I am assuming that your machine was working alright at some stage? If so I suggest you try some basic trouble shooting stepsâŚ
What have you changed since your machine last worked OK?
Perhaps its the file containing the design?
Go back and change that
Maybe its the software that you used to general the GRBL files? Go back and change that.
If you changed the number of steps per mm then go back and change that etc etc.
Maybe you flashed with new firmware?
Use the Engineering trouble shooting principle of 2 and 1.
If you something goes wrong try to reproduce it ( rule of 2 ) when trying to fix it change only one thing at a time.
After all that, the only things left is that something is broken, the software needs re-installing, the controller board or one of the motor drivers may need replacing, one or more of the motors may be skipping, and so on.
