[Solved] MMU2S failing to load filament to nozzle
After over 50 MMU2s "Failed" errors, prints that will not start and half finished prints, loads of research on the topic, checking tip shapes, reducing resistance and so on, I found found two root causes. The main problem is the MMU2S filament positions I use. With 5 colors loaded, I select filament position in slicer according to the color loaded at that position. When using 2 and 4 only, the filament will not even load, and if by some miracle it does start, the print fails part way through. Using position 1 and 2, no problem. It works like a charm. I can print single color from any position with no trouble. I know it's not the hardware. I have a new machine, 3S with MMU2s and have the latest firmware and drivers loaded. I removed the tubes from the filament buffer and hand fed the filaments. Absolute minimum resistance in the filament path. Many fixes suggested have nothing to do with the root cause. The MMU2S has a software problem with certain position changes or the order of change. I could even be the positions selected and which order they start. This becomes a huge problem when using 5 colors. At some point, the position combination or order could cause the MMU2S to fail to load. I need way to much testing to find which positions, or order, or combinations, cause MMU2S fail to load errors. I'll be paying close attention to that the next time I have a fail to load. If you're still with me, the second problem I found, if you have a filament eject and pull out some, cut it off and restart, the filament buffer no longer has extra filament in it. It can be straight from the spool, through the buffer and out to the MMU. That can add a ton of resistance to the filament that the MMU2S doesn't like. We spend a lot of effort trying to make sure there's low resistance at every passage but those can be very small compared to loosing the buffer in the filament buffer. You may think the filament buffer is to keep the filament neat and orderly when retracted from the nozzle. It does do that but more importantly, it's a buffer to the resistance caused from pulling the filament, directly from the spool, through the buffer and tubes, into the MMU. When you load filament or cut off ejected filament, make sure to push some filament from the spool into the buffer. The first time the MMU tries to load that filament to the nozzle, the MMU will thank you. I need to put things back together and continue testing but at least I'm starting to have some successful multi color prints. One tip! if you think your MMU2S has problems, try printing the two color sheep from PRUSA's sample prints. If you can print that, try printing a small single color print from each of the 5 positions on the MMU. If you can do that, chances are your MMU hardware is not the problem.
Update! I took the same stl file and creat a few GCODE files using combination of MMU positions in PRUSA slicer. Position 2,4 worked as well as the others. I can assume my GCODE file on 2,4 that failed was a corupted file. Everthing is working great. If I'm having trouble with MMU fails, I'll replace the GCODE file or go to a differetnt part file to see if I still have MMU problems.