Hi.
Yesterday I finished building my MK3S+ kit (This is my second MK3). I calibrated it, added sock, PID tuned etc.
It prints just perfectly (no ringing [I greased bearings so I guess it helped]). I added heatsings to all stepper motors to keep them cool (which works super great).
I noticed now, while printing the dog included on the SD card that it went to X:0, Y:0 randomly during print and went back to printing without any failure. I also noticed the same while printing small custom print, it happened once after 2nd layer.
Firmware: 3.9.3.3556
For dog I used file included on SD.
My print, I sliced with Prusa Version 2.3.0+win64
In attachments, I include gcode files.
Why is my printer going to 0,0 while printing and then continuing like nothing happened?
And here is the dog gcode.
You need to zip any files and attach the compressed archive to a message here.
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
Just to confirm: You are NOT using OctoPrint, correct?
He is intelligent, but not experienced. His pattern indicates two dimensional thinking. -- Spock in Star Trek: The Wrath of Khan...
@hooch
What do the Fail Stats say?
Cheerio,
Turn off crash detection. See if that helps.
Statistics shown that it crashed at Y axis. No idea what caused it.
Y axis moves very smooth. Belt tension seems ok (260). Prints came out excellent quality. Steppers are cold (I 2-3 40x40x11 radiators to all of them.
Its false crash detections and I and quite a few others have reported this since installing the 3.9.1 firmware.
It absolutely plagues long prints with big perimeters and trashes what would have otherwise been a flawless print.
Its been raised on the firmware GitHub with no response from the developers. I cannot see anything in the later firmware release notes that say it's been resolved.
Disabling crash detection stops the issue, but is an unacceptable workaround because, when a REAL crash happens; the printer will just smash through it and keep going. This could easily end in severe damage to the printer.
Among the common causes of Y axis 'false' crashes are: (in easy-to-check order)
Objects or wiring obstructing the path.
Debris clogging the Y idler pulley - but you haven't created much debris yet.
Y idler not straight or bearing failing (often due uneven pressure/friction when not straight.)
Damaged or wrongly installed Y belt, inc wrong tension.
Dirty Y axis smooth rods.
Y drive pulley loose or wrongly installed - inc grub screw misaligned.
Y axis smooth bearings distorted by overtight U bolts.
Y axis smooth bearings failing due clumsy installation.
Internal bearing in Y axis stepper failing.
Cheerio,
Even when all the mechanical parts work well and no obstructions while printing, I got y-crash detection on both my MK3S machines. Gave up finding the cause and disabled the crash detection.
I am sure it happened along the way of upgrading to 3.9.1. No failed prints after disabling crash detection.
It would be nice to have a way to lower the sensitivity of crash detection on y-axis, as this is the only one I have problems with.
Even when all the mechanical parts work well and no obstructions while printing, I got y-crash detection on both my MK3S machines. Gave up finding the cause and disabled the crash detection.
I am sure it happened along the way of upgrading to 3.9.1. No failed prints after disabling crash detection.
It would be nice to have a way to lower the sensitivity of crash detection on y-axis, as this is the only one I have problems with.
Agree 100%.
No mechanical issues at all and only started after upgrading the 3.9.1
Does the problem go away by reverting back to an earlier version?
Haven't tried to revert back to an earlier version.
For now I live with the disabled crash-detection, but it is far from optimal.
In the Prusa M-code list, there is: M916 - Set TMC2130 Stallguard sensitivity threshold
It looks like this can be used to control sensitivity for stall and crash detection, but I have no idea on how to use it...
Y axis crashes for no reason, seems not to be a new problem. I found this thread: Y-Axis fake(?) Crash detections - let's debug from 2018.
Does the problem go away by reverting back to an earlier version?
Yes. Going to older firmware version elimintatates this issue 100%. I just finished 7h print with 0 crashes. On current firmware version, same print crashed 6 times.
@hooch
Which version doesn't have the problem?
3.8
Here is link to GitHub Issue: https://github.com/prusa3d/Prusa-Firmware/issues/2939
Looks like there are more people with this issues. And interesting note by one of them. It doesn't crash when using OctoPrint.