Arc interpretation error? (G3 / G2)

Started by borisw37 on Jan 18, 2017 in TinyG Support. 1 reply, last activity Jan 18, 2017.

borisw37 Jan 18, 2017 #

I've been using tinyG w/ ChilliPeppr for a few weeks now running code generated by HSM works.
Yesterday I had a "tool crash" luckily just a gouge into soft wood.
I traced it to the following lines of gcode:

N12 G0 X48.224 Y-49.76
N13 Z5.
N14 Z2.
N15 G1 Z0.625 F300.
N16 G18 G2 X48.807 Z-0.008 R0.635
N17 G3 X54.236 Z-5.715 R5.715 F400.
N18 G1 X54.237 Z-6.35
....

The N15 G3 Command caused a "concave" movement of the tool as opposed to convex.
I've attached two images:
one from ChilliPeppr showing the concave arc one from a G-code viewer showing the convex arc (how it should be)

Discrimiator g-code viewer

Is this a bug in tinyG, is it a toolpath generation or post processing bug? Toolpath simulates correctly in HSMworks.
*The same "crash" occured on the other side of the circular part (180 degrees around), other G2 and G3 commands seem to function just fine.

Thank you,

Boris.

cmcgrath5035Staff Jan 18, 2017 #

I'll assume your tinyG is running Firmware 440.20.
It is the best at this, but there are still bug(s) related to arcs

Possible short term solutions
1. I can't tell for sure if your Gcode is mm or inch. If it is inch, try regenerating in mm; mm always works better. TinyG is native mm, so inch Gcode gets a lot of translation manipulation.
Recognizing the rather large numbers 48, 54, etc, I am guessing you are already mm (or have a really big machine)
2. To get your job done, try to generate the Gcode without arcs. I'm not sure if HSM has that option. You will get a Gcode file with arcs replaced by short G1 moves.

This topic is from the original forum (January 2017). Your reply starts a new discussion linked to it.

Different problem? Ask a new question