MMU2 Everything could happen.... an incredible workbench to do practice... but not prints...  

MMU2 Everything could happen.... an incredible workbench to do practice... but not prints...

Since I installed the MMU2S I had the impression to have lost the reliability of the printer, but it is not true. It happened before.

Here is what happened.

1. One week before installing the MMU2,  on a long print with many retractions, my filament jammed into the extruder. A head of the filament was not able to pass through the cooler PTFE tube diameter and stays in the mid aluminium cavity blocking the extruder.
   I had to dismount everything to remove the filament, and got the opportunity to do the MMU2S Extruder upgrade (from I3Mk3).
2. Two prints later the same thing happened again, this time the nozzle was completely clogged. I had to replace the nozzle.
3. Two or three prints later again a head of the filament was not able to pass through the cooler PTFE tube diameter and stay inside the PTFE tube itself blocking the extruder.
   I had to dismount everything again, from the extruder motor, to cooler and PTFE, in order to remove it.
   This time I decided to replace the inside PTFE that has 1.85mm inside diameter, with a larger one of 2.2mm inside diameter.
   I verified that all the culprit filaments were all able to pass through the new PTFE tube prior reassembly all. Then I performed a PINDA alignment + an XYZ alignment + First layer calibration
4. Then I made some print with no issues.
   I decided that was time to install the MMU2S. I installed it without the 5 PTFE tubes and buffer. Spools just on a bar on the wall over the printer.
5. After few successful prints in MMU2-Single, I got two failure at the end of a print.
   The filament head was able to pass through the PTFE cooler and Extruder, but jammed inside the PTFE tube from Extruder to MMU. The 2.0 mm diameter was too small for the filament head.
   Bloody head 😡 , I replaced the PTFE tube from Extruder to MMU (2.0mm internal diameter), with a 3.0mm internal diameter. I verified that all the filament heads passed also through this tube.
6. Several prints, with no issues, where I learned that the PLA filaments are not all the same, even from the same brands.
   In particular, the COLOUR : some colours do not attach to PEI sheet. Black adhere well on PEI, Orange not at all, Blu not so much, Red adhere well and so on, and of course this vary from manufacturer to manufacturer.
   With some filaments I need to use hair spray on the PEI or the filament just slip away with zero adherence.

7. I tried the first multicolor print and the blade moved down and jammed completely the MMU.
   I had to dismount the right MMU2 motor and the shaft in order to be able to extract and reinstall the blade. 😡 

8. Learnt that, I made my first real multicolor print with no issue. Just luck.... 🙃 
9. I installed the PTFE tubes and the buffer. Since then, the printer was not able to complete a single print (or even to start... 😱 )
10. At the first test the blade dropped again and jammed completely the MMU.
    I had to dismount MMU2 motor and shaft again in order to extract the blade. 😡 This time I did not reinstall it.

11. I tried a new print. The filament load procedure failed! 
    The MMU idler was able to select the slot, but then the pulley rotated, the bearing was still, and the pulley just ground the filament in small powder. Then retryes, failures, red leds.... It was clearly a matter of friction that was blocking the filament, since it was working before I installed the PTFE buffer tubes.
    I was not able to find a suitable position of the PTFE buffer tubes to reduce the friction. Even a small change or simply a rotation of the inside filament could cause too friction inside the tubes, and the MMU could not move the filament. I could manually the filament, but I feel the strong force I had to apply.
    Unfortunately at home I have no space, and I can't put the printer and buffer as best they want, so the buffer is in its position behind the printer and the spool are on the left. But in this position the friction inside the tubes is too high.
12. New filament load attempt, and the MMU2 seems crazy. I discover by chance that the FINDA sensor is 1 without filament inserted.
    I clean everything, but there is no plastic, no residues, and the ball is down. 😕 I dismount the FINDA and remove it.
    The FINDA sensor is still in state 1 with the FINDA outside in the air.... 😱 
    I turn off and on the printer and the FINDA comes back to 0. Then I had to reinstall and recalibrate the FINDA. 😡 
    This is a FIRMWARE or BOARD ISSUE.

13. I understood that IT WILL NEVER PRINT as it is, and it is clear that since it was working in a no-friction situation without the PTFE tubes, I see two options:
    - remove the buffer
    - modify the tube system to remove the friction.
    So I decided to replace the five PTFE  2.0mm diameter buffer tubes, with 5 PTFE 3mm inside diameter tubes.
    
14. I performed a First Layer calibration, and finally with the new PTFE tubes the loading filament procedure come back to work. It loaded the filament and made two prints.
    But at the second filament unloading, the filament jammed AGAIN inside the extruder... 😭  
    ...and the IR sensor was state 0 with filament inside... 😱 😱 (apparently...😎 )
    This was a new case: when I dismounted the IR door, I saw that the filament head had a hook shape,  it was out of the PTFE of the cooler, the pulley was empty, but due to the shape it was blocked at the last 5mm before to fit into the last hole near the IR area.
    So it was correct that the IR state was 0, since the filament started just over the pulley and in that position there is no filament on the pulley that moves the IR bar. It was blocked just before exit from the extruder. I could only cut it, no other way to remove it.

Moreover I did not mention two situations of firmware bugs where after a filament unloading, the menu was showing only few options.
I had to reset to revert to normal.

What else will happen tomorrow ? 

I think that everyone else with less patience would have given up with the MMU and the extruder jams. This was a good opportunity to do a lot of practice. 
Moreover the first layer calibration does not plan to extract and round the filament head for the presence of a MMU. I bet it.
This task is done normally by the PrusaSlicer filament settings, while during the first layer calibration the retraction/unloading is handled by the firmware. How Firmware consider the number of unload moves ? No idea

I will update this saga.
Regards