Feature Request

Well that is bizarre, you did do a git pull?

What version does it report?

phillc54 wrote: Well that is bizarre, you did do a git pull?

What version does it report?

I actually didn't since last I added the two lines today. I just changed repositories as reported.

I can try that... maybe that's the culprit.

0.165

0.165 is the latest

Once you clone a git repository, the git pull command syncs everything. If you checkout another branch, it syncs your source to that branch so you should not have any problems that require starting again.

If say you wanted to try another repository say one of mine, you could clone my repository github.com/rodw-au/linuxcnc into a new folder say , linuxcnc-rodw

Then you might want to checkout my arc-radius branch ( its a bit behind) and make sure you use the correct rip-environment script. Then running Linuxcnc from the command prompt should open the right version of the software.Using different rip-environment scripts should swap between versions.

This pull request outlines what this branch is trying to do but the devs want some changes.
github.com/LinuxCNC/linuxcnc/pull/900

I will wait until 2.8 is done and dusted cos Andy has been guiding me in the background.

So I hit it with a big hammer.

Blew away the linuxcnc-dev folder contents, git cloned it back in there, ran the stuffs, compiled, and it's back to Torch Enabled and Disabled. It's odd to me that nothing I did fixed that. I tried every combination I could think of. Git still baffles me. It probably wanted to hang onto that file since I edited it, I'm not sure how to tell it to let it go though.

Rod, that sounds interesting, I may try to have a second RIP version of your branch and check it out.

At any rate, I've got it back to "normal" now. I'm now afraid to change it from enabled to on and disabled to off lol. It's going to irk me as to why it did that.

snowgoer540 wrote: Dang, good eye man! Guess that's an ohmic card all the way to the right... looks to be of their own creation.

As it turns out, we were both wrong. Watched their install video today out of curiosity. This is the card they point you to to do the ohmic sensitivity adjustment, which they do with a pot screw, and it's also the THC.

I'm trying to help a friend diagnose his table.

He swears his config tab says v0.169. However his <machine_name>_config.cfg says v0.161. Is that possible?

He's having some odd issues, and honestly I wonder if his setup is hosed up. Scribing doesnt work anymore (he says Plasmac strips it out of the g-code). Previously Ohmic3 wouldnt work when Ohmic2 did, but now even Ohmic2 doesnt work, etc.

Was just trying to start simple first. He has really terrible internet service. I'm wondering if getting rid of the build bot version, and putting him on a RIP 2.8 version wouldn't be a better idea.

I attached some of his files to have a quick look, I dont see anything out of place, and the table was working fine previously.

I won't get a chance to look at the config until tomorrow (it is Jameson time)

He swears his config tab says v0.169. However his <machine_name>_config.cfg says v0.161. Is that possible?

I wouldn't think so, as <machine_name>_config.cfg is where the config tab reads the version number. Maybe he has a mix of configs???

First up get him to open a terminal and run man plasmac which will show a version number.

If it is v0.169 get him to do a manual upgrade from:
/usr/share/doc/linuxcnc/examples/sample-configs/by_machine/plasmac/configurator.py
that should at least make sure all files are in sync.

If it is not v0.169 get him to do the sudo apt update then then sudo apt dist-upgrade

phillc54 wrote: (it is Jameson time)

Have one for me!

Thanks, I will have him check those things out. That gives us a place to start, I suspect his files got out of sync as well.

Ok, so working through some of this. He's definitely on version .169.

Learning more about the problems, I think the Ohmic problem is a loose wire (he recently transported the table).

However, the scribing problem is, I believe, due to plasmac_G-Code.py stripping out the M3 $1 S1 commands. I confirmed with a test file he sent me that my machine does the same thing. I looked and the G-Code.py was updated 21 days ago, in my brief look I didnt see anything that would have broken it, but mostly I think I should stick to testing only :laugh:

I'm guessing either not many people scribe, or those who do don't update. I attached a file for your convenience.