Another plasma component...

hello! recently I've updated from old 2.8 plasmac version to 2.9 with new axis interface, but i discovered that continues jogging doesn't work at all. o tried fresh new install on new debian wheezy machine from official dev repo, but it also failed.
but jogging with increments works fine. am i doing something wrong? (default axis without plasmac works fine)

just now I've tried gmoccapy but it even doesn't want to start, it behaves like some kind of windows 98 program. and it also happens only in plasmac. vanilla gmoccapy works perfectly fine. i tried this on tree different computers and it all the same.

sevaz wrote:

just now I've tried gmoccapy but it even doesn't want to start, it behaves like some kind of windows 98 program. and it also happens only in plasmac. vanilla gmoccapy works perfectly fine. i tried this on tree different computers and it all the same.

Thats strange. I compiled from 2.9 source today after the latest edits and Gmocappy Plasmac is working for me. But I did not run the configurator.

Are you compiling from source? I've got in the habit of checking the last commit number reported by git pull against this
github.com/LinuxCNC/linuxcnc/commits/master
3ee4401

The issue I had was I did not realise Plasmac was using symbolic links to plasmac.hal. I don't think this is something I've done.

I don't really think this is a good idea. There are some things I need to overwrite on plasmac.hal becasue I have a number of pins plasmac uses already in use. So I lost the edits I made to plasmac.hal a couple of days ago which was a bit annoying.

The other thing is that Plasmac still includes an extra spindle without me ever asking for it so I had to comment it out too.

Thank you for the idea. I'll try to compile linuxcnc master from source today, but actually i thought there is no difference between linuxcnc buildbot repos and self compiled one. Isn't it true?

Ps: forgot to mention, i had to comment two statements about additional spindles in plasmac.hal and in plasmac_parameters.hal just to make it start without linuxcnc errors.

You'd need to compare the buildbot commit somewhere I think to see if its up to date but I'm not sure where to look for that. sometimes Buldbot goes to sleep so it can get behind

phillc54 wrote:

If you updated since the 18th you need to run the Configurator in update mode as mentioned in this post

PlasmaC has used used links for a long time. If you can't connect to existing signals you could either disconnect and reconnect pins in postgui.hal or you could create an equivalent to plasmac.hal and rename it in the ini file.

What is the problem with having an additional spindle enabled?

Yeh I updated after that post but did not realise I was editing the linked file so I lost what I did a couple of days ago. I've got some complexity driving warning lights so it is hard to move it to a hal file thats loaded after Plasmac. Normally I would unlinkp signals and reconfigure it all.

I might look at it again as it really is a pain in the backside on an upgrade. My config predates Plasmac by about 3 years you know.

I added a new hal file this way today for the experimental THCAD based ohmic sensing PCW and I are working on and it took 5 minutes. Testing this today showed promise, but there is a bit more hardware to reconfigure and a component to tweak. Then we should have a more robust ohmic circuit with the ability to adjust ohmic sensitivity. Once complete we might be looking for a bit more real estate on the plasmac GUI.

The spindle issue is the pin being used in plasmac.hal does not exist for me so I keep commenting it out.