Still zero successful prints  

Re: Still zero successful prints

I got a couple of successful single-color prints and then actually got a successful MMU print:

Re: Still zero successful prints

That looks like only one color change per filament. Meaning, you didn't reload the same color more than once, correct? That should work just fine. It is the subsequent loads that are the problem.

Re: Still zero successful prints

That looks like only one color change per filament. Meaning, you didn't reload the same color more than once, correct? That should work just fine. It is the subsequent loads that are the problem.

There's a few, but not many.

Of course the next print was an "airprint" and only one filament. Whee.

Re: Still zero successful prints

https://www.thingiverse.com/thing:3184278

I use this for testing. About an hour - 90 minutes in 5 colors.

Re: Still zero successful prints

I, and others, have made suggestions about tip-presenting, in order to tune the prints.

But there's not even interest in basic functionality, it's really hard to push features. Not having to reset the machine 10-15 times per actual print would be a huge upgrade - how do you balance those needs?

Apparently by doing something else entirely.

Re: Still zero successful prints

Mine still zero success print without jam. Problem with this MMU2.0 upgrade is reliability and high jam rate. This MM2.0 can't be left totally unattended during print since needs frequent jam clearing. I might uninstall this unit for the meantime while waiting solid solution for this problem and luckily for me that I ordered Mosaic Palette 2.0, expected to be shipped this November, to substitute.

Probably I will try to fix my MMU2.0 once the solution became available. Currently I have not time to experiment on my MMU and I have already wasted long hours and material on it.

Re: Still zero successful prints

What filament are you using? What slic3r settings for temp and advanced settings (load/unload/ram)?

For me, the keys to success are:
- debug one thing at a time
- patience

Pick your most common fail and describe it here. We'll work on that.

Paul, great response and discussion for troubleshooting the MMU. I'll be using your comments during my build after I get back from my anniversary trip. What are your comments for those that would like to print the MMU 2.0 parts in another color? My main problem that cautions me is that no matter what Slicer program I use the 20mm cube I print is off in one dimension by +0.8mm.
I would think this would have a huge impact on functionality.

Re: Still zero successful prints

Paul, great response and discussion for troubleshooting the MMU. I'll be using your comments during my build after I get back from my anniversary trip. What are your comments for those that would like to print the MMU 2.0 parts in another color? My main problem that cautions me is that no matter what Slicer program I use the 20mm cube I print is off in one dimension by +0.8mm.
I would think this would have a huge impact on functionality.

Use the parts that ship with the printer first. Once you have the MMU2 successfully working, then you can consider swapping parts.

I will say, however, if you see such a significant size delta on larger parts, I would debug that at some point. If you slice the Prusa parts on Prusa Slic3r with their recommended settings and print with PETG/ABS as they recommend, you should get the same sizes they get. That is how they are printing them, after all. If you don't, you might want to ensure your firmware settings are correct.

BTW, a quick check is to print one of the parts they ship you, and then line them up and make sure they are the same size. It's more important that the pieces you print are the size Prusa expects and/or gets on their print farm than getting an exact absolute size.

Re: Still zero successful prints

I too have removed the MMU as I have yet to have anything other than a jam occur. I really like the MK3 but the MMU is not ready for prime time. I've had better luck with cheap Chinese printers that this MMU piece of dung.

Maybe after about six months we will have something that works. VERY DISAPPOINTED after such a long wait.

Re: Still zero successful prints

Paul,
My big problem is without feedback about "this is normal" or "this is a bug", it's REALLY hard to know what you need to change.

My printer prints at 35% speed most of the time.

I was able to modify the g-code to put M220 S100's all throughout it, to get around it.

It just doesn't seem like a good idea. But I have to because Slic3r PE puts M220 S35's every 25 lines or so.

I've posted on GitHUB for MMU, GitHUB for Slic3r PE, GitHub for the mk3, in response to firmware/software releases here on the forums, and as its own thread.

I mean, short of sending a guy to my house to ignore me personally, I don't think PR could do less to be helpful. I've made progress figuring out where it comes from, but the community won't benefit because they can't fix the issue if they ignore it.

Re: Still zero successful prints

I am making progress. My best is almost 80 filament changes before it shits the bed. Just a few more tweaks...

Re: Still zero successful prints

I also had lot of issue with filament change success rate. The frustration level was on very high level. 👿

Finally the reason for this was that this two screws - see attached picture - were slightly tightened (1-2mm) to strong. After loosing the two screws slightly to be even with the body (up to 0,5mm protude out of the body) all the issues were gone. Based on my experience it should be highlighted in the manual to be very important. And in case of filament change issues it should be checked first … before trying to tune unnecessary the custom filament length setup in the hidden menu what lot of users are doing.

After the mentioned change (loose the screws slightly) i had hundrets of successful filament changes.
😀 😀 😀
I hope this hint will also help other users.

Link to the manual:
https://manual.prusa3d.com/Guide/2.+Pulley+body+assembly/755#s12429

Re: Still zero successful prints

Yes, I had that same issue initially. I think it’s unfortunate design that the system is so sensitive to screw position (really spring tension) such that one turn of the screw changes pass to fail. I have fixed that as you did.

Now all my problems are due to stringing which have the following failure modes:

1. String on tip causes unload to fail by blocking Finda
=> This is mostly resolved with improved slic3r profile

2. String on tip folds over causing load jam in tube
=> Also resolved as above

3. Fine detached strings (I call hairs) build up in finda well, eventually blocking ball and cause unload fail
=> Not solved but made less frequent

4. Same hairs build up along tube or in extruder and eventually clump and jam filament load
=> I haven’t seen this since I lowered the cooling tube section

In contrast I find the load / unload of mmu to work reliably otherwise and I have not had any issues with extruder taking over grabbing filament once I tuned the length. So if I can solve hairs I think I’m good.

Re: Still zero successful prints

...
I hope this hint will also help other users.
...

Not really. Please post a picture how deep are they now. Without knowing how tight it was previously for you and how much you've loosen it, it's not very helpful.
Same thing applies to the manual. It lack of this information. Some people are saying the screw heads should be flush with the plastic body.

Re: Still zero successful prints

Slic3r PE puts M220 S35's every 25 lines or so.

Abraham, are you sure there's nothing in any of the G-code sections causing this? I have not seen this behaviour before.

Can you zip up you Slic3r profiles and also a sliced G-code and I will have a look.

Peter

Re: Still zero successful prints

I am making progress. My best is almost 80 filament changes before it shits the bed. Just a few more tweaks...

Do you want to try printing the attached Gustav? This is something I have (at last) successfully printed.

Uses E1 to E4, all PLA. Basic purge, so ignore colour bleeding. Slight modification to an earlier upload which would not print on my printer.

Peter

Re: Still zero successful prints

I am making progress. My best is almost 80 filament changes before it shits the bed. Just a few more tweaks...

Do you want to try printing the attached Gustav? This is something I have (at last) successfully printed.

Uses E1 to E4, all PLA. Basic purge, so ignore colour bleeding. Slight modification to an earlier upload which would not print on my printer.

Peter

Will give it a try. Thanks. Is this from KISS?

Re: Still zero successful prints

Here is a part I made to help the situation:

https://www.thingiverse.com/thing:3199061

Re: Still zero successful prints

Is this from KISS?

Only the best 😉

Peter

Re: Still zero successful prints

This is the closest I've come to getting a successful print. I only had to open the MMU once to clear an issue. Looks like only 2 missed layers.