Firmware 1.0.4-RC1 for the MMU2 unit  

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Drive error correction and counting-
1.)"the MMU2 unit stops printing and starts flashing continuously with red and green LEDs"- Do you mean all 5 LED's flash red/green?
2.)How does one recover a print if this error state occurs?
3.) The motors seem to be running much faster during an unload. Is this a change in the new firmware as well?
Thanks in advance.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I had issues with the MMU losing track of the filament and the selector would be on the wrong filament position.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

filament unload on tool changes is insanely fast, too fast. there is a high pitch whine as it flies out of the bowden. If there is any hint of resistance, the gear instantly grinds a divot into the filament and it is ruined.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I suspect my bowden lengths got reset and are longer, so that it subtracts more and more from the delay and hits ludicrous speed.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Nope. Those checked out. But when trying to fine tune them, it ground a hole trying to unload at light speed and ruined another length of filament. I'm chewing through this roll faster than if I was actually printing something.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I do not think it uses "bowden lengths" for unload

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I do not think it uses "bowden lengths" for unload

I meant for speed. If you read the firmware, it starts speeding up after it unloads a certain distance. It keeps accelerating until it gets to a point near the end. So if your bowden length is long, it could end up going faster.

But this is not what is going on here. That code didn't change since 1.03 from what I can tell. Anyone else seeing stupid fast unloads with 1.04?

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I do not think it uses "bowden lengths" for unload

I meant for speed. If you read the firmware, it starts speeding up after it unloads a certain distance. It keeps accelerating until it gets to a point near the end. So if your bowden length is long, it could end up going faster.

But this is not what is going on here. That code didn't change since 1.03 from what I can tell. Anyone else seeing stupid fast unloads with 1.04?

Yes! I opened an issue on Github. I like the speed, but the noise is an issue at night. I can hear it from another room. There needs to be a stealth mode.

"Ludicrous Speed" 😆

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I do not think it uses "bowden lengths" for unload

I meant for speed. If you read the firmware, it starts speeding up after it unloads a certain distance. It keeps accelerating until it gets to a point near the end. So if your bowden length is long, it could end up going faster.

But this is not what is going on here. That code didn't change since 1.03 from what I can tell. Anyone else seeing stupid fast unloads with 1.04?

Yes! I opened an issue on Github. I like the speed, but the noise is an issue at night. I can hear it from another room. There needs to be a stealth mode.

"Ludicrous Speed" 😆

but if you cant hear it while your sleeping....how will you know it hasn't jammed? 😀

Re: Firmware 1.0.4-RC1 for the MMU2 unit

How are the load/unloads compared to TheZeroBeast FW? That was a LOT faster and took some months to get used to the sound.

Suggestion: Please keep fail stats available after a print.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Hmm, this firmware changed the position slightly for the selector and how it aligns to the filaments. I can't load filament 1 anymore. Filament 2 is sort of rough as well. The previous firmware didn't have this issue for me.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Hmm, this firmware changed the position slightly for the selector and how it aligns to the filaments. I can't load filament 1 anymore. Filament 2 is sort of rough as well. The previous firmware didn't have this issue for me.

I'm having the same exact issue. Just opened a bug about it on GitHub. https://github.com/prusa3d/MM-control-01/issues/117
Please go comment if you have a GitHub account.

Would you mind removing the FINDA and ball bearing and then shining a light and see if your selector is about 1 mm to far to the right? This is what mine is doing. I can post pics if you'd like.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Looks like someone else also reported it.

https://github.com/prusa3d/MM-control-01/issues/94

Re: Firmware 1.0.4-RC1 for the MMU2 unit

UPDATE: final version 1.0.4 is out https://shop.prusa3d.com/forum/general-discussion-announcements-and-releases-f53/firmware-1-0-4-final-for-the-mmu2-unit-t27891.html

Re: Firmware 1.0.4-RC1 for the MMU2 unit

is the alignment issue on the selector only on the 2.5?

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I loaded this new firmware yesterday on my Mk2.5 with MMU2 (upgrade from the MMU1, which was working well) with, initially mixed results. The good news was that a three colour print of the penguin @ 0.15mm worked flawlessly with no missed layers. The bad news was that it took a lot longer to complete (>12hrs for a <10hr print) than it should have!

The best news was that the reason for the delay was obvious - the problem occurred during the filament retract from the hot end back to the selector. After the filament started to go back up the bowden tube it was obviously supposed to speed up dramatically, however in my case it actually stopped and the extruder stepper in the MMU2 made a loud noise as it was just skipping steps and not rotating. After a while this stopped and the filament was retracted very slowly into the MMU2 unit, actually stopping and reversing for a bit a couple of times in the process.

At this point it is also worth noting that my printer was one of those that was having problems with alignment of the selector after initialisation with version 1.03 of the driver - it would move quickly to the left and hit the left wall with such a large force that the wall would actually bend over and then stop, still hard against the wall.

Looking at the symptoms I realised that the most likely explanation was that the 12V supply in the mk2.5 is only marginally able to drive the stepper motors in the MMU2 at the higher speeds. I therefore measured the output of my PSU, which was set to about 12.15V and increased it by about 1V to 13.15V (beware, this may also affect/damage the heaters, fans, etc.). When running a test print the improvement was dramatic - the filament retraction was fast and much quieter. The initialisation also behaved totally differently, with the selector moving to the right wall and returning to the correct position (about 1mm from the left hand wall and correctly aligned to slot one).

I suspect that the supply voltages on the Mk2.5 units may be subtly different and also the cable resistances to the MMU2 unit may be a little variable, both of which may affect whether or not this marginal issue is seen. I suggest that it would be a good idea if Prusa and any beta testers, if they don't already, set the PSU voltage a little lower than normal during their testing such that any problems like this (which may only affect a relatively small number of users) are revealed.

The very best news is that the MMU2 that I have had since the end of November last year is now seems to be working reliably for the first time! Initially I could not get it to work for a single print, but I did get much better results after changing the bowden between the MMU2 and the hot end with one having a 2.5mm internal diameter and drilling out its connectors a little.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

I loaded this new firmware yesterday on my Mk2.5 with MMU2 (upgrade from the MMU1, which was working well) with, initially mixed results. The good news was that a three colour print of the penguin @ 0.15mm worked flawlessly with no missed layers. The bad news was that it took a lot longer to complete (>12hrs for a <10hr print) than it should have!

The best news was that the reason for the delay was obvious - the problem occurred during the filament retract from the hot end back to the selector. After the filament started to go back up the bowden tube it was obviously supposed to speed up dramatically, however in my case it actually stopped and the extruder stepper in the MMU2 made a loud noise as it was just skipping steps and not rotating. After a while this stopped and the filament was retracted very slowly into the MMU2 unit, actually stopping and reversing for a bit a couple of times in the process.

At this point it is also worth noting that my printer was one of those that was having problems with alignment of the selector after initialisation with version 1.03 of the driver - it would move quickly to the left and hit the left wall with such a large force that the wall would actually bend over and then stop, still hard against the wall.

Looking at the symptoms I realised that the most likely explanation was that the 12V supply in the mk2.5 is only marginally able to drive the stepper motors in the MMU2 at the higher speeds. I therefore measured the output of my PSU, which was set to about 12.15V and increased it by about 1V to 13.15V (beware, this may also affect/damage the heaters, fans, etc.). When running a test print the improvement was dramatic - the filament retraction was fast and much quieter. The initialisation also behaved totally differently, with the selector moving to the right wall and returning to the correct position (about 1mm from the left hand wall and correctly aligned to slot one).

I suspect that the supply voltages on the Mk2.5 units may be subtly different and also the cable resistances to the MMU2 unit may be a little variable, both of which may affect whether or not this marginal issue is seen. I suggest that it would be a good idea if Prusa and any beta testers, if they don't already, set the PSU voltage a little lower than normal during their testing such that any problems like this (which may only affect a relatively small number of users) are revealed.

The very best news is that the MMU2 that I have had since the end of November last year is now seems to be working reliably for the first time! Initially I could not get it to work for a single print, but I did get much better results after changing the bowden between the MMU2 and the hot end with one having a 2.5mm internal diameter and drilling out its connectors a little.

That is an interesting theory. Did you flash the 1.0.4 Final firmware and see if there was the same issue (without needing to up your voltage). If what you are saying is correct, it might make sense to make a separate power supply for the mmu2 when using it with the mk2.5.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Looks like someone else also reported it.

https://github.com/prusa3d/MM-control-01/issues/94

I was going to try what you asked but the 1.0.4 final firmware fixed this for me. Everything aligns now.

Re: Firmware 1.0.4-RC1 for the MMU2 unit

Does it still have the same horrible sound when unloading?