@thomas0815
Just for my understanding: You generated gcode with another postprocessor and tried it on your SM (2?) - according to the chosen postprocessor, the âdialectâ of the generated gcode doesnât have to (proably wonât) fit the dialect of the controller used in the SM. Either that could lead to the observed ârandomâ movement or a simple variance in the origins.
As long, as I use a Linux, I donât know exactly, where to replace the faulty file. But, I had similar error messages at first. After double checking and carefully copying the relevant file (snapmaker_freecad_post.py) again, freecad finally âacceptedâ the new postprocessor.
Nevertheless, in the end the output of your postprocessed gcode should look something like this handcoded testing script with an additional header:
Header (modified for testing)
;Exported by FreeCAD for Snapmaker
;Post Processor: Snapmaker with mod. b0.1
;Output Time:2020-11-23 15:28:40.848902
Working GCode:
G0 Z25.00 F7500
G21
M3 S0
;beginn operation: drilling
G1 X0.00 Y0.00 Z20.00 F6000
G1 X0.00 Y0.00 Z-13.00 F150
G1 X0.00 Y0.00 Z20.00 F6000
G1 X180.00 Y0.00 Z20.00 F6000
G1 X180.00 Y0.00 Z-13.00 F150
G1 X180.00 Y0.00 Z20.00 F6000
G1 X180.00 Y145.00 Z20.00 F6000
G1 X180.00 Y145.00 Z-13.00 F150
G1 X180.00 Y145.00 Z20.00 F6000
G1 X0.00 Y145.00 Z20.00 F6000
G1 X0.00 Y145.00 Z-13.00 F150
G1 X0.00 Y145.00 Z20.00 F6000
G0 X0.00 Y0.00 Z150.00 F6000
;finish operation: Drilling
M5
I used it to test drilling 4 mounting holes with the proper grid of the good âwasteboardâ. The speed settings (FâŠ) are roughly guessed, as I didnât calculate the proper feed and cut rates as I should. The ones above used for testing with a 4,5 mm drill in 12mm multiplex worked acceptable. The downstep on the Z axis with 13mm was a bit deep. 12,5mm will be enough. I wouldnât set the speedrates for the dive into the material higher. REMEMBER to span an additional wastable wasteboard below the workpiece, as long as you donât want to add four 0,5mm blind holes to your âgoodâ one.
@Streupfeffer
linuxcnc: I share your interpretation. In addition, the changes made according to the template doesnât seem to be extensive. In addition, the circle movements G2 and G3 are processed faulty.
.
.
.
.
Aunt Edit said:
Brent pointed me to the fact, that SM adopted the marlin language for their controller. Itâs worth a look if you want to get deeper into understanding the proper gcode for cnc milling.
explanation for the variance of origins:
freecad will use the origin plane as origin for the path calculated in gcode. If you first extrude your pad eg. 15mm ABOVE the origin in freecad and calculate a milling path from it, you have to set your machine origin on the lower side of your workpiece. Then you will see the Snap going up (same in thomas picture) and milling the stock downwards layer by layer back to the set origin of your machine.
Alternatively I used to construct my pads downwards from the origin plane. Iâm setting the origin of my machine on top of the workpeace.