Marlin firmware issuing Stop

CNC: Lowrider2
Controller: Rambo 1.4a (from Ryan)
Firmware: Latest Marlin DualLR firmware
Host: Repetier-Server (latest)
Post Processor: flyfisher 604 for Fusion 360 (latest) - removed M0 and G92 from gcode

The Drill operation, the 2D Pocket, and the Bore, all executed fine. The 2D Contour operation was cutting out the piece with tabs and had made the first 2 passes, each going down 6mm on a 3/4" piece of MDF. Then it stopped saying the firmware initiated a Stop. How do I debug this? Why might this occur. I could not see or hear a malfunction, and I happened to be watching it. Thank you.

Did the display give you any information about the error? As a first step, I rerun the file in the air (router off) and see if the problem is reproducible. If it is, and if the display or your g-code sender gives you additional information, take a picture and post it here. If the coordinates are being displayed, record them. You may be able to use them to locate the point in the g-code file the problem occurred and check to make sure there’s nothing funny in the g-code file.

I’ve never used it, but in searching for what conditions Marlin might stop, I ran into M999. It be worth a try since it could potentially save a job.

Though I don’t expect it to show anything, it is worth a couple of minutes to check your g-code file in a g-code simulator like this one.

The Lowrider2 no longer has a “Full Graphic Smart Controller, Big” attached to the Rambo 1.4a. It went bad and I replaced it with a 7" touch screen attached to a Raspberry Pi, which in turn is connected to the Rambo 1.4a. No error was shown on the touch screen, or on Repetier-Server’s web interface, merely that the firmware had issued a Stop.

I tried the " G-Code Q’n’dirty toolpath simulator" you suggested. It complained at the beginning that it did not understand M117, but that was at the very beginning and gcode was well along making cuts before the Stop. I don’t remember the exact numbers, but they were all positive. X was between 0 and 20. Y was probably less than 300. Z was between 10 and 20. I ran the Fusion 360 simulator for the gcode, which contained only the 2D Contour operation. It completed successfully. I looked for a simulator that was Marlin specific, but did not find one.

jeffeb3 mentioned my machine is larger than the max values in the firmware, and that caused the problem when homing. I was thinking this was not the problem, as the gcode had already traversed around the piece at least once, and was making its second or third pass (each cutting another 6mm deeper).

I would try rerunning the code in the morning, unfortunately the power company is cutting our neighborhoods power for 8 hours. I will try tonight.

If the problem occurs “in the air”, I could try rebuilding the Marlin firmware with the actual machine values and see whether it resolves the problem. Does Repetier-Server have a plugin that gives better error messages for Marlin firmware? Any other techniques for getting a better error message? I will also look in the gcode Console. Maybe something got dumped there that I did not notice. Thank you for your suggestions. I am SO close.

The machine values won’t affect movements, just homing.

1 Like

Thank you for ruling out the machine dimensions being the problem. This weekend got me past a number of problems. There were lots of red herrings and my mistakes.

I found the Repetier logs and they were filled with networking problems - failures to contact the license server, DNS lookups, removing the WAP and restoring it. It seems likely that the USB is interfering with the Raspberry Pi’s on board Wifi. I temporarily ran a long cable into the garage, but will probably have to use a separate USB Wifi dongle and antenna. This was causing intermittent issues getting Repetier-Server to respond.

I had used a finishing step on my 2D contour. In the cases where a firmware Stop did not occur, but it wasn’t moving, it wasn’t actually stopped. It was just moving really, really, really slowly. I increased its speed 5X, so it is still slow, but moving. I tried to only set only one finishing repetition, but it is doing two anyway. I still have a lot to learn/relearn. I don’t remember making these mistakes with my old MPCNC. But that was not causing the Stop. Still not sure what was.

Finally, for the air test, twice I set the Z position too low, when it was actually at its highest. The gcode, trying to get above the stock, ran into the max limit switch and slipping the coupler, made grinding noises. Setting it 20mm higher avoided that. Silly me. But the air test ran for over an hour, didn’t Stop, so I tried it on the MDF.

The gcode for the 2D contour has now run to completion. No Stop(s). Not sure what fixed it. Still have the wired connection. My current issue is that 1.35mm of the 3/4" MDF was not cut. If I can’t figure it out, I will create a new topic to ask the question. Thank you both for your help.