MK3: New Firmware + New Slic3rPE = Disappointing Results
Yesterday I updated my MK3 from firmware version 3.1.3-245 to 3.8.1-2869, and Slic3rPE from a few months old (don’t have the version) to 2.1.0. The results are a disappointment.
For about a year and a half, I’ve had this MK3 printer that gave fantastic results. After this update I’m left with a run of the mill printer no better than a Prusa clone. Here’s what I’m seeing:
1-Prior to the update, the only real noise from the printer was a vibration in the PS on large Z moves (mostly homing). X and Y moves were virtually silent. Now on any medium to large X and Y move, the steppers buzz. Especially noticeable on diagonal moves.
2-Slicing leaves ridges in every other layer that weren’t there before. On the attached picture (taken at about the 4th or 5th layer, the 3 y-shaped pieces all have the same ridges. The next layer, which is printed at about 90 degrees to the previous is OK, and then the subsequent layer has ridges again. On the final top layer, the ridges still show through from layer “top-1”.
3-Prior to the update, the first layer was printed at a low speed. After the update the first layer speed appears to be much faster and there are artifacts in the layer as viewed from the top (which do get covered by the next layer). Note that after the print is finished and the visible part of the first layer is examined, there isn’t any visible issue.
4-As the printer fills in an area, it used to get to the end of a movement and then reverse smoothly. It now appears move “jerky”. Maybe I’m seeing something that isn’t there, maybe it’s just printing faster than before, but there’s definitely something different.
5-I’ve noticed that the bed leveling is different, and don’t know why. It used to test 9 points that were evenly spaced across the print bed. Now, the whole pattern seems to be shifted left, with the leftmost point of each row just at the edge of bed, and the rightmost point spaced away from the right edge of the bed maybe 1”+. This doesn’t appear to be causing any problems, but I’m not sure why it’s different.
Items #1 and #2 are the real issue. The others are just things I’ve noticed.
Any idea why this is happening and where my great MK3 printer went?
Print in Progress:
Do some elimination.
Roll back to your previous firmware and see what the effect is. Make sure you did not use the MK3S firmware by mistake.
Same for PrusaSlicer.
I checked and I did use MK3 firmware: prusa3d_fw_MK3_3_8_1_2869.hex
I guess I'll try rolling back the firmware first - but I have to believe others have run the latest FW and Slicer. I can't be the first to notice this.
I'm also 99% sure I used the correct Slicer download, but would the MK3 Prusa Slicer even be different from the MK3S Slicer?
PrusaSlicer is the same for all machines.
And I would say - don't roll both back at the same time.
Try to identify which one is affecting you.
#1: Looks like your PS settings are different now. Try to slow down during the print by 5% and check if your printer get silent again. Sometimes it's just resonate at certain speeds.
#2: Looks like over extrusion to me. As the part is very thin most likely your first layer is too low and the filament is building up to the top without having any buffer/space to release the over extruded filament.
My suggestions would be to do factory reset (with the new firmware) and redo all calibrations.
Let me second Nikolai's suggestion about factory reset and clear storage.
That should really always be the first step if you upgrade firmware and get odd results.
Thirded here. I've had a couple of oddities updating firmware and do a full factory reset if anything seems amiss. It's annoying repeating the calibration steps, but it beats time and money lost if prints aren't successful.
With a minor firmware revision, you can generally get away without a factory reset, but a major revision 3.1 to 3.8 you MUST do a factory reset.
The EEPROM storage locations move and you end up with garbage data where there should be some other value.
A factory reset, sets the EEPROM values in the locations for the firmware to the correct default values for the firmware it is run on.
Yes it's a PITA, but necessary for trouble free printing. The 8bit EINSY is really being pushed to it's limits and simply does not have enough memory space for everything, so as features are added, others are dropped (Some are legacy and no longer in use, but the code remained. Recently there has been a purge of this code, clearing out memory to be re-used by newer features and other sections of code have been rewritten to make better use of the limited memory available.).
Well, I took the advice above and started with a process of elimination.
First, I tried the same print with different filament (Prusament in this case) - just to be sure that the older filament I was using didn't contribute to the problem. Same exact results.
Next, since I had another computer with a much older copy of Slic3rPE, I re-sliced the print and generated new g-code. Same exact results. So, this eliminated the new version of PrusaSlicer from the issue - the problems weren't caused by the updated slicer. I also reviewed all of the individual settings between the old slicer and the new version, and surprisingly, they haven't changed. So I can't explain why the first layer is printing faster, and why change of directions appears to be more "violent" - if that makes sense.
Next, since the consensus above was that the latest version of firmware really required a factory reset, I bit the bullet and did that. Factory reset, new calibration, the works. Same exact results.
I did localize the new noise to the Y axis, and it's kind of a cross between a buzz and a hum, and definitely the stepper motor stepping noise - if that makes sense. Doesn't appear to be resonance. I don't think changing the print speed will correct it, since the noise occurs on the first and subsequent layers (with a different buzz/hum), and the first layer prints at a slower speed than the other layers.
I'm going to live with it as-is for a while, rather than back out the firmware and go through the factory reset/calibration again. And I'll see if the print quality issue continues, or is maybe just an artifact of slicing this particular set of parts.
Thanks for all of your suggestions - if anyone has something else to try, let me know.