Sometimes when I’m carving, my “U”s come out sideways, and it’s usually the second time the letter is used in the carving. I’m using ESTLCAM and RepitierHost, and both of them show what I’m trying to carve (Dr.Ausenhus) correctly. This behavior isn’t consistent, so I have no idea what causes it. It also happened with the second “t” in my engraving of “enter at your own risk”.
What the hell is happening here? It’s kind of funny when it happens, but I’m stumped as to what’s going wrong.
My rig is a stock MPCNC with the dewalt router, using a RAMPS 1.6 with tmc2208 drivers. Workpiece is a pine 4x4, and also occurred on a piece of walnut I was carving earlier. Bed is a MicroJig fixture tabletop that I’m pretty proud of, but I need to make the MPCNC taller so that I can properly use it lol.
Also, please ignore the fact that the crown is in the background of the ESTLCAM. I can’t figure out how to make new files in the program, so I continually edit the same test crown file like a caveman.
Possibly your Y motor driver is overheating and when thermal protection kicks in, it stops driving the motors. But the controller is oblivious and the X motion is still happening.
If you energize the motors and leave it for a while (motor drivers see the same current when stationary) can you tell if the Y axis is not being held firmly? If this is happening then reducing the current on the Y axis would fix it.
Interesting theory! As far as I can tell, the Y-axis driver isn’t warmer than the others, and this issue occurred before when I was using the DVR drivers. The Y-axis seems to work while it is malfunctioning, as it still moves up and down while it is engraving the screwed up “u”.
I also use the same board to 3d print with, and for big prints take as long as 22 hours for me. This issue doesn’t seem to occur when printing, as I don’t skip or lose steps in my prints.
Together, these facts make me think it’s something happening with the CAM step or with RepetierHost; I’ll try using Simplify3D or another program to run my gcode and see if it still happens
That’s pretty hard to believe. Estlcam will show you what the Gcode says.
What I was reading sounded weird, that it was just the one letter that was messed up, and the rest of the text was OK. but it certainly looks as though it goes wrong, then does not recover. This points to something else going on, and not the CAM. It also appears that you aborted the cut (reasonably so) before allowing it to complete, so some speculation on my part is necessary.
X and Z appear to be OK, with it just no longer responding reasonably in Y.
Overheating is one possibility, another is a poor wiring connection to the motors. A poor crimp anywhere in the chain can cause problems. Sometimes, it happens at a certain position, where the crimped connector is just at the wrong angle. (This is more likely if you are using a series wiring kit since it seems both motors in Y are showing problems.)
So some things to try.
Carve the same Gcode into some scrap, and see if it does exactly the same thing more than once. If it’s the Gcode that is wrong, and not an issue with your machine, then the machine will repeat exactly the same mistakes no matter how many times you run the Gcode. (That’s what CNC machines are best for, is repeating exactly the same instructions.) Note that this still won’t 100% rule out a bad crimp or wiring if it happens at the same X coordinates, but even then, I’d expect some differences in the carving.
Try running the Gcode in the air, and watch that it appears reasonable. This can be more difficult with longer jobs, or ones that have a lot of pocketing, but these jobs don’t look like they do.
Thanks for the tip! The wiring issue sounds probable. I’ll check it when I get home and report back!
Also, I wonder if we could make a visual symptom-diagnosis page for CNC noobs like me, similar to the troubleshooting page by Simplify3D has for 3D prints. It was super helpful for me when I was learning to print and it would be cool if we had one for carving. I’d be happy to contribute as I find new ways to screw up!
It appears the issue happens every time. I was cutting vinyl today and though it is hard to see, both times I executed the gcode, it produced the sideways error. Once I edited the file on ESTLCAM to be smaller, it no longer occurred
I still think that this is more likely to be wiring than CAM.
It looks like it’s happening at a similar position, but the 2 cuts do not look like the actual error is identical. Again it appears that you aborted the cuts at different points, so a direct comparison isn’t possible, but even with that, there appear to be differences in the error from what I can see. (It is difficult tp be sure though.
Again, if the CAM is wrong, but the machine is good, it will do the exact same thing.
If you still have the original CAM, try starting it with a different X starting point. Let it try to cut something cheap, like regular paper. Let it make it’s error, then move the starting point diagonally, and let it run again. Compare the two.
Or, if you like, send me the Gcode, and I’ll try it on my machine. If my machine also messes it up, then you’re sure that you’ve got a CAM problem. I might even get an idea of what’s going wrong if I can see it happen.
I haven’t gotten around to trying estlcam yet, but it can’t be that hard to start a new project. I’d be wondering if there wasn’t some artifact I was pulling in that led to the problem. Someone here should be able to guide you through how to start with a clean estlcam slate.
