PRUSAs MMU2 not working

That's it, the PRUSA's multi material unit does not work. No matter what you read in the forums. I guess they are all infested of PRUSA's agents. It will not work.

Of course, the device does all it is supposed to do. It has three steppers and they all do move as programmed exactly what they have to move when they are required to do so. Yet the machine doesn´t work. This all might seem odd for someone who hasn't tried this device. So where's the problem? Well, the problem is related to filament tips, which is something that any 3d printing fan has never ever thought about unless he has one of these machines.

PRUSA's multicolor printing system uses just one extruder. This implies that filaments are continuosly getting in and out of the hotend. A new character appears in the scene, and it's called stringing. The filament tip, on its way out of the hot end, drags molten material in the shape of strings, globs or hairs. This debris end up jamming the system and the print fails, usually skipping layers.

PRUSA's attitude facing this problem has been all wrong. Their solution is settings adjustments. There are something like 11 parameters in PRUSA's FW to shape the tip. They include ramming sequences, retraction speeds, and cooling passes. A process so impossible to fathom that PRUSA's people have been unable so far to produce a minimal guide about tip shaping. In any case, theoretically, you should be able to program these 11 parameters so you always get perfect tips that do not dirty your system.

Anybody can imagine that there's a random factor about this dipping the filament into the hotend. Pretending to obtain ALWAYS a perfect tip has proven impossible, and just one single misload will cause a 30 hour and one thousand filament change print work to fail, so system's reliability is close to zero. PRUSA just denies it and say that their system is workable and issues don't exist. Just adjust better your settings, you sucker!!.

The truth is that PRUSA's device is just poorly engineered. Their firmware does not use an optical filament sensor installed in the machine, and anybody would say they did not put a minute of thinking about tips rather than their odd settings solution. But the device is not a complete pile of poo. It can be easily re-fitted into a workable device changing the firmware and installing a device called TheFrog (https://www.cgtrader.com/3d-print-models/hobby-diy/other/the-frog-the-prusas-mmu2-hairless-extruder-fix) that I designed and you can find here in CGTrader.

A week after posting TheFrog in the PRUSA's shop's forum (it had already been in CGTrader for about a few weeks then) PRUSA (that had denied until that moment that there was any issue about the MMU2) has released their MK3s extruder, which features 4 characteristics present in TheFrog: a gate, an indirect sensor mod, a ball bearing and magnets action.

I am not going to wail about being copied. That's life. The one thing that I think it's open to discussion is if PRUSA's solution will really work, and we can talk about it until they finally send it for free to our homes (a veiled confession that their MMU2 does not wortk), and from past experiences I'd say that will at least be three or four months.

This is PRUSA's design for the MK3s (https://www.prusaprinters.org/original-prusa-i3-mk3s-and-mmu2s-release-sl1-and-powder-coated-sheets-update/)

I must say that I have my doubts it this will work because I tried something alike and it didn't. This is my MK6 design.

I'm sorry but it looks like the system does not let me upload a picture (either that or I have posted the same picture about five times)

Similarities: Hinged gate, magnet action

Differences: Single gate /Double gate - inverted position of the hinge - a bearing to receive the tip and open the gate vs tip pushing open the gate on the way out.

Why my MK6 design did not work: The rotating gate produced a tweezer effect, trapping material. After a few hundred tool changes, this ended jamming the system. Cleaning was a mess.

This is why I suspect PRUSA's solution is half baked again and might not work as expected. Probably there are thousands of engineers in CGTrader (and possibly a few of them even own an MMU2). So, as a mind game, we have several months to speculate if PRUSA's solution will or will not work. I am really interested in the views of other mechanical akin minds. Thank you