My machine randomly started stalling out. It seems to be around specific types of Gcode, but not always in the same place. It seems to like doing it after a recent switch it work offset, like from a G55 to a G56. It’ll get through that command. It’ll get through the first move to be over the new part. And then it’ll get a command like
G0 G43 Z0.6 H2
H2 is already in the spindle from before. It will make it down to Z0.6 and then just stall! Sit there spinning and doing nothing! Sometimes it will randomly start back up again and go on with its day just to stop again later at a similar spot. Sometimes the system will get tired of waiting and throw an error that flashes by too quickly to read or snap a picture of. And then it will just shut down. After shutdown a pure black screen will read
Linux Mint 17.3 Rosa tormachpcnc ttyl
Tormachpcnc login: _
It seems to be happening more and more frequently to the point I can’t get any work done now. Weirdest part is it started doing it on a file that I’ve run literally dozens of times over months and months. Now it does it elsewhere too.
One more detail. I forgot to mention that when it does this the cycle start buttons light is blinking! Like it does when waiting on a tool change or is being free held. But clicking it or seemingly any other buttons doesn’t seem to help.
It is a PCNC770 that last time I ran these files was on a PP 1.x version. However, in between then and now I did go to PP 2.x. There was a fairly large gap between now and when I last ran this file that I know didn’t use to do this! Don’t remember which PP2 version, think just the first one but could check.
Is this a machine that runs regularly every week, month?
I would check connections. Pull motor plugs and reconnect and loosen and wiggle all the interface cables.
It is not a production facility machine. Weekend warrior type thing. That doesn’t mean you’re not right, something g could be loose! My main concern with that theory is why a loose wire would specifically hate G43 H# related codes, as far as I can tell, exclusively.
It’s not a requirement but pp2.x uses more resources and it can bog down an old controller. You tool height change line betting the trigger is curious though.
Definitely disabling the cameras will impact memory.
Do you have this issue when running conversational programs?
Why are you only changing the tool height and not the whole tool?
No idea. That is what fusion360 spit out. I have a handful of paths that’s run tool 2, then it swaps over to a new work offset, runs them all again, does that up through 4 work offsets, and then does the same thing with another tool. That line I wrote above is like line 3 in the first path of the new work offset g57.
I did remember this morning though that I have seen it stop at other places than just tool changes. It has stopped twice I can now remember where it was just randomly running in the middle of a path.
Cameras? What cameras? Well, at least I don’t have that to worry about.
As for memory. I was thinking about it. If my stoppages were due to memory issues, then why would it blink the green light on the right right of the cycle start button? That almost sounds to me like the machine stopped, and for good reason, on purpose and is letting me know that.
When I get time to test it, this is where I’m trying first. Seems like something is shaking lose or something and it stops as a safety precaution, tells me something is up, and I just am not fixing it so it doesn’t go away.
Found some time today to be able to work through this and for the life of me cannot find a single thread on how to go backwards to PathPilot 1.x. Any help in the right direction would be greatly appreciated as at this point the machine is just sitting idle with parts waiting. Pray for me that it doesn’t require buying another USB and shipping that doubles its cost just to undo the work of the last USB I bought.
Open a command window and dump out the logs and errors. They should give you some more information as to what’s going on under the hood. Since it doesn’t seem to be a common problem, it’s likely related to your specific machine.
I think the PP logs are in the same place as Linux CNC, but check that if you can’t find them.
It also won’t hurt to put it on a network and upgrade to the latest PP 2.0 as that will get wiped if you downgrade anyway (then take it off the network as an additional debugging step)