My switchwire gone crazy

[skyguynca]

I finished my switchwire a few weeks ago. I have about 80 hours of printing without problems. I was using cura but switched to orca about 30 print hours ago. Still no problems.

I had a stepper problem, replaced 2 of them. Did all my test and calibrations again, printed the voron tets cube v.7 and it was good.

I tried to print a jewelry tree for my daughter and about the 12th layer the tool head crashed into the bed and it tried to keep printing as if it was still 12 layers up but was dragging across the bed.

I checked every thing. Z offset, bed level, accel and speeds. Did another cube and a bench all printed fine. Tried to print a simple box from an stl 105mm x 65mm x 125mm and again about 4 layers up it crashed into the bed and tried to continue while scraping across the bed.

I need some help

Thanks
David

 

[Darkstar77]

Which stepper(s) did you replace, and why? Did you tension the belts on X & Z after the stepper replacement? Did you lock tight the grub screws on the drive pulleys for X and Z? It sounds like either belts skipping, or the belt drive gear slipping on the stepper shaft. I suppose it could be klipper config related, but I doubt it. If your run current is too low, you might be losing steps. The sudden drop to the bed you're describing really does sound like a belt drive gear slipping on the stepper shaft to me.

Chances are since this happened after replacing your steppers that this is a mechanical issue. Loose grub screw(s), loose belts, something along those lines. Do the replacement steppers exactly match what was there previously? I am still curious why you had to replace steppers. If they don't, could be just the klipper config, but I don't think that would create the variability you're seeing.

 

[skyguynca]

Which stepper(s) did you replace, and why? Did you tension the belts on X & Z after the stepper replacement? Did you lock tight the grub screws on the drive pulleys for X and Z? It sounds like either belts skipping, or the belt drive gear slipping on the stepper shaft. I suppose it could be klipper config related, but I doubt it. If your run current is too low, you might be losing steps. The sudden drop to the bed you're describing really does sound like a belt drive gear slipping on the stepper shaft to me.

Chances are since this happened after replacing your steppers that this is a mechanical issue. Loose grub screw(s), loose belts, something along those lines. Do the replacement steppers exactly match what was there previously? I am still curious why you had to replace steppers. If they don't, could be just the klipper config, but I don't think that would create the variability you're seeing.

I checked the grub screws and belt tension 3 times when I re-calibrated after each crash. When I say crash that what it is. You hear and see the tool head driven into the bed by the steppers. I am trying cura today to see if it is something that orca started doing when slicing.

I replaced the steppers because they were making a squealing noise like a bushing or bearing was going bad. I had them in my old printer for 5 years. I believe they were just wore out.

If it does it under cura too I will try compiling my firmware again, maybe it got corrupted somehow.

 

[Dave32]

I am trying cura today to see if it is something that orca started doing when slicing.

Rather than switching software, just go back and re-print a gcode you sliced previously and printed successfully before the problem occurred. That would very quickly help identify if it was slicing related or not.

 

[skyguynca]

Rather than switching software, just go back and re-print a gcode you sliced previously and printed successfully before the problem occurred. That would very quickly help identify if it was slicing related or not.

Well sorta did that already. Like in the first post the Voron Cube and the Benchy printed fine. I did a couple of others and they printed fine. Just a few of the STL's I have been trying since the motor change and recalibrations are messing with the print. Those STL's are doing the same thing in Cura and Prusa too. I am comparing the settings to see what is the same and what is different. The difference is not causing it so that can be ruled out right away.

 

[skyguynca]

Ok, so now I know what happened. I have everything up and working again flawlessly. It is totally amazing and so simple. I never tied it to a event that happened some time ago and to some really weird events during prints.
So the simple was my webcam plugged into the pi4. Yep it was causing errors I wasn't seeing because it is running headless. Occasionally the picture would flash or blank out for a second. I thought nothing of it. I also use obico and sometimes due to internet traffic the video stream will freeze when checking on the printer.......well that was mostly because of an error and the Pi reconnecting with the camera but the error still existed. After so many attempts the pi would reboot, so loss of connection to the MCU. That explains the occasional failed prints that sometimes were minutes into printing and some 18 hrs into printing.
These same errors also caused communication errors between Klipper, the firmware and the pi. That was what would cause the crash. The Pi tried to recover and continue printing. I haven't sorted how it caused the head to get driven into the bed by the steppers. However in the logs I can see the z moving down so I know it was commanded but it is not in the gcode.
I finally poured thru so many Klipper and moonraker logs and print logs plus gcode files all day today. Unplugged the camera, did all my calibrations again and been printing the rest of the day no problems at all.
I found the problems when I ssh into the pi and entered "dmesg", it printed out the log and it was long but I found all the errors and finally at the end was the reboot cmd at the same time the MCU lost connection. Alot of the crashes also coincided with error fault with socket errors and memory errors caused by the camera and polling to re-establish connection with the camera.

 

[Darkstar77]

Wow, crazy issue. I presume either the webcam is bad, or it's drawing too much power from the Pi? Glad you're back up and running.

 

[skyguynca]

Wow, crazy issue. I presume either the webcam is bad, or it's drawing too much power from the Pi? Glad you're back up and running.

According to all the linux and like groups, these usb 2.0 cameras have caused alot of issues and errors and there are some with work arounds but no real fixes.