Description
After upgrading our 36 MK3S+ printer farm to 3.14 yesterday, I started ~20 printers today and about 8 of them failed to start their next print with a Z-axis calibration error. When I investigated, I saw that the plates were on fine (usually the cause of this). Allowing the printer to home the Z up top and retry the G80 would not solve the issue, even with a couple printer resets. Most of them failed on the first 10 or so points, but one or two of them made it all the way to the back line before failing, again all with properly installed plates.
I've managed to resolve each of these by running Calibration > Z Calibration. All 8 printers resolved their symptoms immediately after running the Z Cal, but interesting that the automatic Z calibration that's kicked of when G80 finds a problem does NOT solve the problem, it had to be from the Calibration menu.
These GCodes mostly came from 2.7.4 PrusaSlicer and do have the parameters for G80 showing the bounding box of the print. My printers all have 7x7 MBL turned on. They are all running OctoPrint from a RPI4 with a plugin sending M79 pings every 20s to the printer so we can utilize the 'Set Ready' (thanks for this!) feature for our internal print job mgmt system
These GCodes are the same ones we print every day and haven't been re-sliced in a few months.
Here's the start gcode for one of the prints that was acting up. Note that Y is a negative value, is this expected? This print takes up a large portion of the plate and has a skirt that comes near the limits of Y.
M862.3 P "MK3S" ; printer model check
M862.1 P0.8 ; nozzle diameter check
G90 ; use absolute coordinates
M83 ; extruder relative mode
M104 S170 ; warm nozzle to idle
M140 S90 ; set bed temp
M190 S90 ; wait for bed temp
M109 S170 ; wait for extruder idle temp
G28 W ; home all without mesh bed level
G80 X22.0967 Y-0.203449 W205.807 H210.407 ; mesh bed levelling
G1 Z2 F720 ; raise Z so ooze doesnt bind to nozzle
M104 S240 ; set extruder temp
M109 S240 ; wait for extruder temp
Activity