Mk3s prints perfectly and then loses calibration and mind
Just finished my MK3s today, and after some fiddling and calibrating, managed to get two excellent quality prints using the included filament. Jumping right into another print, when I started my 3rd print, the initial 9-point calibration sequence was "too far" back on the y-axis points, and even bottomed out at the far end and crashed. It repeated this crash until I hit the reset button.
Thinking it was a calibration issue, I went to re-calibrate XYZ, and now something on the Y-axis is fouled.
- Self Test is OK
- Auto Home -- Goes to the right X/Z axis, but Y is about 5mm closer to the front than should be
- XYZ Calibrate fails with "Bed Calibration Point not Found"
- PINDA sensor is working as expected, based on sensor info (and LED flicking off as it gets almost to the heatbed)
- Y-Axis seems to have a full range of motion using Move Axis (-4.0 - 212.0) without anything blocking
- Y-axis belt tension is 270
- As far as I can tell, the X-Axis is unimpeded, the bundle coming out of the back of the extruder rests on the Einsy Case at 0.0
- I am getting some garbage characters in the display sometimes while testing these results (weird ASCII characters)
- Calibrate Z ends in "Bed Leveling Failed Sensor Didn't Trigger" -- and you can see the HB flex as the nozzle pushes down on it!
I don't know what else to check -- as I said, it worked perfectly for two prints and lost its mind.
the Mk3S does not have any Physical end stops on the X and Y axis,
it relies on Stalling the motor against the end of travel, to set Virtual Endstops... which works well most of the time.
however if the printer finds unexpected resistance to movement then the end stop position can be mis calculated.
try lubricating your smooth rods with a modest film of grease from the tube provided by Prusa (don't add a lot, because the linear bearing dust seals will simply sweep the excess to the ends of travel. )
move the X and Y carriages back and forth a couple of times, grease again, and move the carriages again, then re try your printer.
hopefully this will provide sufficient lubrication for a short period of use. (If all is well set a reminder to lubricate the linear bearings properly at the next opportunity, no point ripping a working printer to pieces at this stage.... )
Below is a link to the Prusa Rods and Bearings Maintenance page!
If you haven't already done so, please set up 7x7 mesh bed levelling using the LCD menu
Thanks very much Joan -- I suspect you are correct. After I finished my post, I spent a couple hours fiddling around with the printer. Eventually, everything just started working again, and I was more confused than when I started.
The one thing I noticed was that the heatbed thermistor wire had come untucked from somewhere, and seemed to be catching the Y-Axis of the bed. I suspect this is where everything started to go wrong. Then, after the update to the virtual end stops, nothing was calibrated correctly anymore. I've since tucked the thermistor wire away better, and hopefully won't have this problem again.
I've also lubed all of the bearings and rods, and setup 7x7, because that can only help. Is there a prescribed method for resetting the virtual endstops, should something like this happen again?
Glad you got sorted...
The Virtual endstops update at the beginning of every print as part of the Auto Home procedure.
Protect your Thermistor wires at all times, they are the most easily damaged part of the extruder
This saved my life 🙂 After successfully printeng my first 3 objects, the printer stopped with a "Crash detected" message... No way to print any anymore, and it seemed the whole X-axis was kind of "shifted" to the right... I tried to recalibrate, even did a factory reset, to no avail... It would always fail to self-test the X axis. In the end it passed the test, but would stop at the XYZ calibration as it didn't go sufficiently to the left... Then I came across this post and I simply pout some drops of lubricant on the rods. And presto, everything went smooth again.
Sometimes the answer is simpler than the problem!