Unintentional diagonal printing leading to tool changer crash
Hello, everyone!
I got the Prusa XL 5T in January and have had a lot of success printing multi color and multi material prints, but last week, a strange problem started.
When printing 2 colors or 2 materials, after the first two mm of the print, the rest of the print is vertically diagonal - there's a small layer shift at each layer. This error is cumulative until after about 30 mm of printing, tool change fails. After the crash is detected, the printer prompts me to reset the tool. After I do so, it homes and resumes printing - but now there is a very large layer offset, because the cumulative error is gone after that. Both the print and the wipe tower are affected!
Image of the print and the slicer:
Notice: First 2 mm are ok, then diagonal printing starts, then after the crash, the top layers are offset, because the printer re-homed into the correct position.
Reprinting produces the EXCAT same result!
If I rotate the the part by 90° and keep the wipe tower oriented the same way, the part is warped in the same direction, but the purge tower is now warped 90° from the direction it was warped before. This implies that the distortion is rotated locally with the part.
I did not let this print continue too long because I didn't want to crash the extruder again.
For some reason, this seems to be material profile related. The part is a combination of PLA and TPU. For PLA, I'm using Prusament PLA profile and for TPU I'm using the Ultrafuse TPU-95A profile. Both of the profiles were downloaded by Prusa slicer and were not modified in any way.
What's even more bizarre is that I've been printing with this profile combo for a while. There's a decent amount of stringing on the TPU, but orienting my parts a certain way functionally solved this problem, so I never bothered tweaking the temperatures.
Reslicing and reprinting the part using Prusament PLA and Generic Flex profiles seems to produce a normal print without the warping. Printing in PLA - PLA combo also produces expected result. This is why I know it's profile related.
Things I've tried that have failed to solve the issue:
- Printing on heads 2 and 4.
- Printing on heads 1 and 3.
- Firmware 6.1.3+7898
- Firmware 6.2.4
- Firmware 6.2.5
- Printing PLA and TPU on two different heads.
- Dock calibration.
- Moving the print into a different location on the bed.
- Rotating the print produces warping of part in the same direction, but purge tower in 90° offset from the original direction.
None of these have resolved the issue since I started debugging.
This started about a week ago and the only thing I did was upgrade from firmware 6.1.3+7898 to firmware 6.2.4. I made no other hardware changes or any calibrations. After noticing the problem, I upgraded firmware to 6.2.5, then back down to 6.1.3.
Any help would be appreciated. If required, I can provide the stl models and Prusa slicer project.
Thanks for your help,
Janez Štefulj
RE: Unintentional diagonal printing leading to tool changer crash
I've only come across something like this in the firmware alpha version that added phase stepping and turning phase stepping off was the fix. Personally I've left phase stepping off in favor of crash detection, but that's personal preference.
If I were in your shoes I would upgrade to the newest firmware version and redo all calibrations. If that isn't enough to fix, try turning phase stepping off. And if even that doesn't work then I'd turn to Prusa support if someone else doesn't have better advice.
XL-5T, MK3S MMU3 || Printing with multiple nozzle-sizes: Official Guide, Unofficial || PrusaSlicer Fork || Other advanced slicer tactics || TPU
Please show us the underside of the failed prints.
Cheerio,