External offsets - Elements to cut are different heights

08 Jul 2018 06:44 - 08 Jul 2018 06:44 #113789 by rodw
Hi guys, I've seen this before and I'm pretty sure its a configuration issue or work offsets or the like. I've been building a new standard external offsets config using as close as I can to the default interface. There is a standard cut file that comes with the sim that looks like this

But on my new config, it looks like this.

I'm sure its something I've done in the config as the files are identical and changing back to an older config renders it correctly. The machine complains about the Z axis being out of range if I try to cut this.

Any ideas?
Last edit: 08 Jul 2018 06:44 by rodw.
08 Jul 2018 06:56 #113791 by rodw
Strange, I found I was missing some info in my [TRAJ] section of my ini file so I added the last 3 lines to this

#added these lines below
CYCLE_TIME = 0.010

Improved it, and no longer stepped on different levels,but still higher than the material surface by a long way. (The Tool is on the surface)

09 Jul 2018 13:44 #113873 by andypugh
A top height of 79.7 and a pierce height of 3.5 makes me wonder about a mm/inch mix-up somewhere.
09 Jul 2018 14:37 #113877 by rodw

andypugh wrote: A top height of 79.7 and a pierce height of 3.5 makes me wonder about a mm/inch mix-up somewhere.

Quite possibly. This particular gcode file is an included external offsets demo file for metric machines which is just a wrapper around an imperial subroutine. Come to think of it, I wrote the wrapper! I did make a change as we had missed a hardcoded imperial tolerance so maybe I broke it.

Its kinda like the path is rendered at the wrong end of the axis.

It does not explain why the sim my machine is based on displays the file correctly. Are there config settings that impact on a gremlin display?

I opened an old file in a gmoccappy config today which was a nest of 5 parts and it was displaying the same as the stepped screen dump above. I had a good look at both configs and could not see any obvious differences. I ran a dry cut with the plasma off and it cut the nest OK so its more of a render thing.
11 Jul 2018 21:21 #114030 by rodw
I'm pretty sure this is something to do with the probing for the surface as in the sim, this code is commented out. I think there is a solution as I'm sure this has been discussed on the forum but I can't remember where. Grotius is also mentioning it on another thread. forum.linuxcnc.org/38-general-linuxcnc-q...tion?start=20#114022
Time to create page: 0.119 seconds
Powered by Kunena Forum