Compatibility with Lantek and other high-end software

More
3 years 2 months ago #279392 by santy
My current question is, how can i ensure that when filtering the gcode generated by Lantek, the Kerf width compensation is automatically made like it occurs in QTplasmac Conversational. In image 1, we have the code generated by Conversational for cutting an identical geometry as shown in image 2, wich represents the code generated by Lantek. Te objective is for this filter to adjust the Kerf width compensation, just as it happens in Conversational. Im using the custom_filter.py.

Please Log in to join the conversation.

More
3 years 2 months ago #279423 by santy
I managed to make a py script that converts the code. I print the M190 PXXX with the material number i want, and it runs as a separate program.
of course, the program just adds the right offsets for the kerfs, and replaces the "G" moves and adds the kerf based on hal parameter. All good.

The problem is the G03 moves. Lantek exports this arcs with some error in the arcs and i get: radius to end of arc differs from radius to start linuxcnc


What could be the problem?
Here is a sample of ALMOST zero problems: (The error i get is always the same)
Code:
M66 P3 L3 Q1 M190 P1 G21 G64P0.25 G40 G49 G80 G90 G92.1 G94 G97 (preamble) f#<_hal[plasmac.cut-feed-rate]> G00 X22.5 Y-1186.4 m3 $0 s1 G41.1 D#<_hal[plasmac.kerf-width]> G01 X17.61 Y-1189.88 G03 X26.745 Y-1182.155 I1186.6133363484503 J-1186.6133363484503 G03 X19.02 Y-1191.29 I1186.6133363484503 J-1186.6133363484503 G01 X22.5 Y-1186.4 G40 m5 $0 G00 X22.5 Y-998.9 m3 $0 s1 G41.1 D#<_hal[plasmac.kerf-width]> G01 X17.61 Y-1002.38 G03 X26.745 Y-994.655 I999.1533716101848 J-999.1533716101848 G03 X19.02 Y-1003.79 I999.1533716101848 J-999.1533716101848 G01 X22.5 Y-998.9 G40 m5 $0 G00 X215.535 Y-945.15 m3 $0 s1 G41.1 D#<_hal[plasmac.kerf-width]> G01 X210.645 Y-948.63 G03 X219.775 Y-940.905 I969.4141832699787 J-969.4141832699787 G03 X212.05 Y-950.04 I969.4141832699787 J-969.4141832699787 G01 X215.535 Y-945.15 G40 m5 $0 G00 X215.535 Y-1107.65 m3 $0 s1 G41.1 D#<_hal[plasmac.kerf-width]> G01 X210.645 Y-1111.13 G03 X219.775 Y-1103.405 I1128.4253890820607 J-1128.4253890820607 G03 X212.05 Y-1112.54 I1128.4253890820607 J-1128.4253890820607 G01 X215.535 Y-1107.65 G40 m5 $0

Is there any way to merge the arcs or add a line before/after that will enable me to remove this difference?


I tried to mess with some INI parameters such as:
ARC_BLEND_ENABLE = 1
ARC_BLEND_FALLBACK_ENABLE = 0
ARC_BLEND_OPTIMIZATION_DEPTH = 50
ARC_BLEND_GAP_CYCLES = 4
ARC_BLEND_RAMP_FREQ = 100

is this the right approach? I had zero success (always the same errors)
Isn't there a way to allow more %relative error and just merge the lines?


 

Please Log in to join the conversation.

More
3 years 2 months ago #279424 by phillc54
The ARC_BLEND parameters that you are changing have nothin to do with G2/G3 arcs, they ar for look ahead trajectory planning.

There is a CENTER_ARC_RADIUS_TOLERANCE_MM here:
linuxcnc.org/docs/devel/html/config/ini-...sub:ini:sec:rs274ngc

I tried adjust this earlier with some gcode that you posted but it did not resolve the error.
forum.linuxcnc.org/plasmac/49697-compati...-end-software#276995

 

Please Log in to join the conversation.

More
3 years 2 months ago #279427 by santy
just tried it. changed to 10 no luck

Please Log in to join the conversation.

More
3 years 2 months ago #279429 by phillc54
I converted the file from your earlier post so it would run on an AXIS sim and the same errors happen.
I also ran this file on an online gcode simulator  ncviewer.com/  and it ran fine.
I think that I am unable to help any further. It may be worthwhile posting on the  G&M Codes  section of the forum.

G-code:
Warning: Spoiler!


Error:
Warning: Spoiler!

Please Log in to join the conversation.

More
3 years 2 months ago #279437 by rodw
Does by chance Lantec uses absolute coordinates  and not relative on arcs?

Please Log in to join the conversation.

More
3 years 2 months ago #279438 by phillc54
Looks like that is it, below is with G90.1 at the beginning of the code:

Warning: Spoiler!

Please Log in to join the conversation.

More
3 years 1 month ago #279939 by santy
Hello.
I am still having some errors.
Could you please send your filter? It seems to work without errors.
thank you

Please Log in to join the conversation.

More
3 years 1 month ago #280049 by phillc54
I think this is the correct one:
 

Attachment not found


 

Please Log in to join the conversation.

Time to create page: 0.226 seconds
Powered by Kunena Forum