perhaps somebody experienced similare problems and may help.
I am using Aspire preparing and generating gcodes for carving with an SM 350.
For wooden refliefs for example it worked fine so far, but since a few days I am faced with a problem that I don´t understand.
I prepared a model based on a dxf file ( or STL ) for some ribs that I wanted to cut out from 6 mm plywood using a 1 mm flat end flute. After a successsful and error free toolpath simulation in Aspire, I saved the gcode to USB stick using the latest aspire to snapmaker post - processor from github.
The carving starts as expected cutting out the ribs, but after a while Snapmaker makes surprisingly unexpected movements. For example it jumps back into already finished zones , doing toolpath that you can´t find in the gcode and does ad hoc Z steps, so that in consequence the stock material material is damaged and the 1mm flutes are broken.
If I load the same gcode file into NCViewer there isn´t any failure to discover.
I agree with @brent113 it looks like some steps were missed especially as some of the pictures really look as a linear displacement of the cut. Did you leave any tabs on the cutouts? If one of the parts simply gets loose, that’s a good way to get your bit stuck in between somewhere causing it to break or to get the linear modules to skip steps. (And after that your workpiece is basically ruined)
on the one hand it looks like a linear displacement ( pic 1)but the marked areas are showing a weird zig-zag- movement (pic 1 and 2) that is not root caused by linear displacement.
The question is, why the machine randomly is doing toolpaths, that are not defined in the gcode. In pic 2 the tool has moved independendly in a small area going deeper and deeper until it was broken. So all 3 axes were engaged.
The stock material was fixed with the original clamps
Well, I guess just going through each of your questions / statements, here’s what I would say about them.
Without being able to specifically reproduce that in a way that clearly shows it’s an arc problem I’m not inclined to believe there’s something wrong with the way G2 is handled, it appears to be caused by something else.
I don’t agree with that conclusion - once steppers lose their position it’s difficult to predict exactly how the machine will respond to the shock loading that occurs by driving bits into a sidewall of the cut.
My conclusion is because these motions are after-effects after losing steps. More reproducing required to isolate or prove wrong.
Did it go down, or is it going up? Maybe your linear modules overheated, who knows.
Unlikely, but to troubleshoot that you’ll have to engage support.
If your machine is behaving like this frequently your only avenue to get it fixed is to email support, assist them in troubleshooting, and they will identify replacement parts and mail them to you.
As @brent113 already answered extensively. There is no way to know that for sure.
So it was clamped from the side and the parts your cutting out were just loose? I didn4t see any tabs to hold them in place, so these can just start flying around and cause all kinds of problems.
I recently had a little mishap and got on video: things can go wrong fast (the magic starts around 40 seconds in the vide where you see the bed tilting up and down):
In the end the bit might end up anywhere, drill 5mm deep, get stuck trying to move sideways etc.
Thank you for your support. The short video is interesting but this didn’t happen, and although the tab was picked up by the tool, it didn’t have an influence for further toolpath. The tool was moving on its way inclusively the rotating tab. Maybe that Z step was partially blocked .
I will do the job again and come back with further information.
Are you talking about your situation, or the video I posted? Because I can assure you it most certainly did. I’m not going to repeat myself, but have a look at this topic where I shared my little failure and analysis: CNC gone wrong: sharing failure - #3 by brvdboss
Oh, this thread is helpful. Now I understand better what you mean with loosing some steps. My situation was similar: I used a 1mm tool with cutting lenght 6mm but programmed carving depth was 6.3mm in 6 steps. So with the last step the conic section between cutting length and 3.175 shaft was slightly " rubbing" the stock . As I said, I will prepare a proper setup and will do the job again.
