Hello, Anyone caught a bug when carving just stops without any error messages?

Hello,

Anyone caught a bug when carving just stops without any error messages? Seems like server meets a stopping point in a g-code. After I sent a “Resume movement from stopping point” command, carving starts again, but seems that part of g-code loses in this case, because the final carving looks wrong.

I’m sure that there is no any stopping point in my g-code. Carving never stops if i connect my X-carve directly to my Mac. But if i use remote access to Raspberry Pi Zero - i get the stops.

On the Raspberry i use the last version of serial-port-json-server - 1.92.

I tried different g-code, and always the same - no stops by direct connection to mac, and few stops each time by connection to Raspberry.

I tried different settings in the “Gcode Sender Options”, now i use the “Pre-Upload 1,000 Lines Into Buffer” option.

One thing that i noted - raspberry never stops when doing the z-probe work (sometimes it takes 3-4 hours). But it stops during the carving job.

Any ideas?

Screenshot would’ve helped as not sure if you’re on TinyG or Grbl. If you’re on Grbl which is my guess since you mentioned X-carve then there is still an ancient bug that rarely pops up where the Grbl buffer in SPJS could possibly get into a deadlock. This bug used to exist on TinyG buffer but got stamped out 1.5 yrs ago and no reports since.

Hi John, you are right, I’m in Grbl environment. Not sure that screenshot can help you, because there is no any alerts or error messages in the “Serial Port Console”. The carving job just stuck. The spindle continue to spin. But motors stuck. The jog pictogram on the 3D view also stops. The tilda icon - “Resume movement from stopping point” - launches the motors again.

I can start the JSON server in the verbose mode. Maybe there will be some errors. I will do in on this weekend’s.

Oh, no, I just meant had you posted a screenshot we all would’ve been able to see that it was Grbl, and SPJS version, and Grbl version, etc. Yeah, you are seeing that old bug creep up. A grbl user would have to transpose the TinyG buffer in SPJS back to Grbl to finally solve the bug.