MMU2S keeps printing even if IR sensor goes to zero
I'm on the latest firmware (3.8.1 / 1.0.6). I've had nothing but issues with the MMU2S. If I bypass it by removing the power cable and feed in single filament, it works absolutely perfectly (I've run it through 20+ hr prints) so I know my printer is correctly tuned. I have gone through nearly every buffer mod imaginable including auto rewind spools and it's just not behaving correctly.
I blame my set up though and how I am routing filament through the drybox on the lower lack table all the way up to the MMU2S unit. I'm sure it doesn't like the tension / routing. But until I figure another solution out for that, another issue is really annoying me: that is, if the IR sensor triggers to 0 during a print (e.g. due to tangled filament), it still keeps printing. This is very annoying because it will just print in the air for hours. I'd expect it to immediately stop and ask to load filament if the sensor is tripped (like it does with a normal MK3S setup). FINDA correctly shows 1 since filament is still detected through there so is it only using that value and ignoring the IR sensor? If so, why?
Thanks, sorry if this was already asked. I could not find it with the search terms I used.
RE: MMU2S keeps printing even if IR sensor goes to zero
You're describing a really interesting case. Regular run-out logic for MMU is triggered from FINDA because MMU is pulling the filament out. Once triggered, the extruder is pulling the filament up so MMU gears can grep the filament and pull.
The only case where I can see (under correct setup conditions) your described situation is happening, is when the filament breaks between FINDA and IR sensor. Inside the bowden tube it's very unlikely and I haven't seen this cases before, but anyway. FINDA would still see the filament but the IR sensor would trigger once the broken part is used. The printer could stop in this case and you would require to dismantle the extruder to get this piece out (or maybe you're lucky and can push it through).
Feel free to open up a feature request on github for this case. But I would recommend to check your setup and IR sensor. This cases shouldn't happen under normal conditions.
Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram
RE: MMU2S keeps printing even if IR sensor goes to zero
Thanks for the reply. So once the IR sensor is triggered at the start of filament load (after triggering the FINDA of course), it is never checked during a print? I can see a lot of issues with that. In my case, I do believe extruder is clean but what is happening is that friction / tension in the drybox path to the MMU2S is causing the extruder motor to not be able to pull it properly. Here a more descriptive use case for clarity: it is printing normally, it gets to a point where there is a tangle in the spool, extruder can no longer pull the filament; however, since FINDA is still triggered by the already loaded filament, it will just keep printing (in this case, with no filament extruded).
I'm surprised I am the only one who has seen this. Basically, if there is any friction in the filament path whatsoever I can see this happening, unless I am misunderstanding this mechanism.
RE: MMU2S keeps printing even if IR sensor goes to zero
In case of no filament movement either from heat creep/jam/clogg or tangled filament, you should still have both sensors triggered because filament is inside/loaded. Neither of the sensors can detect this. Why do you think IR sensor could handle that?
Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram
RE: MMU2S keeps printing even if IR sensor goes to zero
Clearly, I would not expect any sensor to handle that ... but in my case, the IR sensor IS being tripped back to zero. My guess is that it is extruding whatever is remaining in the hotend, cannot pull it anymore, and then goes to zero (correctly, because the extruder "door" is no longer pushed "open"), but the printer happily goes on printing since the FINDA is still triggered. This leaves huge gaps in the print because it thinks nothing went wrong. For a case like filament run out, it makes sense to check the FINDA since filament is going through there first, but in this case, it's not sufficient.
I guess my question is: is this normal behavior? When FINDA and IR are both triggered (printing like normal), and IR goes to zero without an unload, should it keep printing? Obviously my set up could be better not to be causing this, but the MK3S without the MMU2S handled this just fine: any time IR is tripped, it asks to unload/load filament.
RE: MMU2S keeps printing even if IR sensor goes to zero
Sorry, I can't follow you. The IR sensor sits in the chimney above the gears. In case of tangled filament, extruder gears will grind down the spot and that's it. Filament will stay in place above, so the IR sensor should be still triggered (because filament is inside). I've seen false IR sensor display though. Maybe there is a firmware bug. Ideally you could mod the sensor with LED, so you can see the true value.
Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram
RE: MMU2S keeps printing even if IR sensor goes to zero
I can confirm that this is the behavior of the unit. I agree with nikolai-r that I it seems unlikely for real-world problems to frequently have a problem with this behavior.
On the other hand, suppose the IR sensor is calibrated fairly well, but every once and a while hits 0 momentarily even though filament is present if the print head is moving back and forth very rapidly. (Yes I have seen this.) If the behavior was as you wish, this would cause extra print stops. (On the other hand, it might get people to calibrate their IR sensor more carefully, which would be a good thing!)
RE: MMU2S keeps printing even if IR sensor goes to zero
Sorry, I can't follow you. The IR sensor sits in the chimney above the gears. In case of tangled filament, extruder gears will grind down the spot and that's it. Filament will stay in place above, so the IR sensor should be still triggered (because filament is inside). I've seen false IR sensor display though. Maybe there is a firmware bug. Ideally you could mod the sensor with LED, so you can see the true value.
This is not right. If the grinding is severe enough, the sensor will be zero because it's attached to the door and the door moves in as the filament at the gears thins out. In the MMU it senses the door tilt, not the filament itself.
RE: MMU2S keeps printing even if IR sensor goes to zero
I can confirm I have seen this failure myself.
Even if it's a rare failure, it's the correct behavior. If the IR sensor doesn't see filament, the print should stop. Can you describe a case where the IR sensor doesn't see filament and should keep going?
RE: MMU2S keeps printing even if IR sensor goes to zero
Related note - this feature is slated for 3.10 firmware!
RE: MMU2S keeps printing even if IR sensor goes to zero
Related note - this feature is slated for 3.10 firmware!
Great news! Thanks! It is a very annoying problem where the print keeps going but should stop.