Filament falls out of the mmu on filament change during multicolor print
 
Notifications
Clear all

Filament falls out of the mmu on filament change during multicolor print  

  RSS
joe-5
(@joe-5)
Eminent Member
Filament falls out of the mmu on filament change during multicolor print

I'm having an issue where the print starts fine, but when it comes time for a filament swap, the filament being swapped out actually falls out of the unit, rather than just being unloaded. I thought this had to be an idler tension issue, but my screws are a couple turns below "flush" with the part, which is tighter than the guide recommends. Could it be something else?

Posted : 26/07/2021 4:05 pm
Baklin
(@baklin)
Reputable Member
title

It is not the idler that holds the filament in the MMU.

That is done by the PTFE tubes at the back of the MMU. The PTFE tubes make a little bent when they are clamped by the mmu2-rear-PTFE-holder. This gives enough friction to keep the filament from falling out.

If you use a larger ID tube there or some kind of rewind spoolholder it can get pulled out to far.

Posted : 26/07/2021 5:33 pm
joe-5
(@joe-5)
Eminent Member
Topic starter answered:
title

Oof, support told me it WAS the idler that held the filament in! That's great to know. I have 3mm ID tubes on the back.. and gravity auto-rewinders.. so I suppose I ought to swap back to the stock PTFE tubes and see how it goes. Thanks a bunch.

Posted : 26/07/2021 6:00 pm
Baklin
(@baklin)
Reputable Member
title

I think you can use the 3mm PTFE tubes, but you have to make a bent somewhere. If the filament doesn't have any resistance in the tubes that gravity spool holders are known to pull out the filament some times.

 

Strange that support told you that the idler is holding the filament.

You can check that easily. With all filaments loaded you can probably pull out four of them. My experience is that the filament to the left of the one that is selected is held by the idler, that others are only held by the resistance in the tubes. This is just because of the way the idler is designed and rotated during use.

Posted : 26/07/2021 6:49 pm
joe-5
(@joe-5)
Eminent Member
Topic starter answered:
title

Thanks again. Out of curiosity, what's your filament management solution? Are using the stock buffer and tubes?

Posted : 26/07/2021 7:36 pm
Baklin
(@baklin)
Reputable Member
RE: title

I have tried different buffers and spoolholders. The rewind spoolholders make it easy on my setup, but I fine them not reliable enough.

And I have no space behind my printer and want to have the spools above it. That is why I'm now using this buffer with some modifications.

All with 2 mm ID PTFE tubes. I have bought some 3 mm, but never installed it.

With no space behind the printer it is a little difficult to get the tubes like I want it. But it has worked without problems so far.

What I like about a buffer is that it doesn't have moving parts and there really isn't anything that can go wrong. With the rewind spoolholders I have seen a few times that the filament retracted faster than the spool could rewind and the filament would end up next to the spool and get stuck. That only happened on some spools and only when they where almost full.

 

This is an older picture of my setup. I have made some changes, but most of it is still like this.

 

Posted : 26/07/2021 9:19 pm
joe-5
(@joe-5)
Eminent Member
Topic starter answered:
RE: title

Wow, that looks absolutely terrific! Well done.

I may have to try a buffer. I swapped the stock PTFE tubes back on a couple of my filaments and I'm still running into the same issues. When it swaps, it ejects the filament altogether probably 25% of the time. Then I have to intervene and manually reload it.

In this picture I have the stock tubes on the white and purple filament. It didn't help the issue, but I'm wondering if there's something I'm missing regarding the "bend" in the tube? The white is easiest to see--just wonder if anything is glaring. FWIW, I've installed the M10 PTFE holder passthrough adapter on the back (seemed to be highly recommended), but I haven't seen anything that would lead me to believe this could be causing the issue.

Many thanks.

This post was modified 2 months ago by joe-5
Posted : 27/07/2021 2:23 am
Baklin
(@baklin)
Reputable Member
title

This sounds like the gravity spoolholders are pulling to hard on the filament.

The bent that I mentioned in the original Prusa design is in the mmu2-rear-PTFE-holder. You can see it in this picture of that part. There is a little kink in the channel for the PTFE tube that adds some resistance to the filament. That is normally enough to stop the filament falling out.

But a rewind spoolholder can be pulling to hard if the rest of the PTFE tubes have very little resistance.

 

 

The part that you use with the m10 fittings (I use that part too) has an almost straight channel (and less resistance)

 

I would think it is a combination of things that give you problems. The modified PTFE holder with m10 fittings, the almost straight PTFE tubes between the MMU and the spoolholders, the 3 mm ID tubes and maybe the almost full filament spools. A heavy full spool pulls harder on the filament.

 

Adding a bent in the tubes between the MMU and the spoolholders could add just enough resistance for the filament staying in place.

 

You could try putting some thing under the front of the spoolholders. By lifting up the front you make the track that the spool runs  up and down on less steep. That should reduce the forces.

 

 

Posted : 27/07/2021 6:03 pm
ssill2
(@ssill2)
Noble Member
no gravity

I tried those gravity rewind spool holders but had a similar problem.  I also hated the clunking sound.

I think Baklin is correct, the M10 PTFE replacement for that part is 100% required upgrade.  The other thing that works really really well for me is horizontal mounting of the MMU.

Here's the arrangement that works for me now.  Shorter PTFE tubes coming out the back and feeding the filament from sunlu dryer boxes.  No buffer in the middle other than enough room for slack to safely absorb retracted filament.

I just completed two prints mmu prints of a light saber that were 1700 filament changes with 1 intervention on the first and zero on the second, not to mention several prints before that that were in the hundreds of filament changes.

Posted : 31/07/2021 1:50 pm
Share: