MMU3 Release was is it good for?  

RE:

1. Yes. There is no understating this -- YES. Total number of "tool changes" (that's what the slicer refers to all filament changes) is the major time factor for total print time. As long as everything runs smoothly, there's several actions that have to happen every time a filament change happens:

• Nozzle is brought to the purge tower and performs a pre-unload purge with cooling moves
• Filament is unloaded from the nozzle [1]
• MMU selector is moved to new filament [2]
• Filament is loaded to the nozzle [3]
• Post-load purge is performed

5 major steps between each actual model print operation. If everything is running smoothly, and you are using the standard 70mm purge volumes for load/unload, then this normally takes somewhere around 45 seconds. For a 2-color print with 250 layers needing dual color (50mm for 0.2mm layer height), this adds 3 hours to the total print time. On the extreme end, for a 5-color print with 1000 layers (200mm) with full color density throughout (5000 tool changes), you're talking almost 3 days. Just from the tool changes.

Myself, my longest print was a 3-color print that was 180mm tall. 72h total print time, even with a 0.6mm nozzle.

... and that's when it's running smoothly. The 3 main issues I marked above are very, very important.

[1] Filament can get stuck on unload (and possibly break). Silk filament is the biggest culprit for getting stuck, as the tip shaping routine doesn't exactly work correctly for it in my experience. There's also potential issues with the FINDA, or grinding of the filament caused by improper MMU idler bearing tension.

[2] Fault during selector movement. There's nothing in the MMU2S to detect if there's any issue moving the selector, so if there's a piece of filament still present (either a stringy tip, or faulty FINDA reading), it'll just jam itself over to the next position with the filament in between (this is the worst recovery scenario, btw -- requires disassembly of the selector rod section to fix).

[3] Fault during filament load. Either the filament isn't getting pushed all the way by the MMU, there's too much friction in the PTFE tubes, there's too much tension in the buffer system, there's a bad filament sensor reading, or possibly ALL of these at once. From personal experience, this is the most frequent failure point, and even with recent firmware updates (prior to MMU3), it has only served to fix the blobbing on failure (after 3 attempts, the nozzle is moved to the back left corner away from the purge tower, so if it takes a while for human interaction to happen, all the oozing is happening away from anywhere it can cause damage and complete catastrophe)

These events are the main reason I'm very excited for the MMU3, mainly because it's supposed to communicate what the failure actually is instead of having to diagnose the situation based on blinking lights and prior experience of past failures. I expect failures to happen, but being able to immediately go down the right path to fix things can mean the difference between completing the print or throwing away 70+ hours of colorful plastic mass.

Oh, I guess the main reason I bring this up -- every failure adds to print time. I know this sounds silly to mention, but it can mean the difference between getting a project done in a few hours vs having to babysit a monster through every free moment of your weekend.

2. Soluble material. It's possible. PVA is the main material of choice for PLA. The biggest concern is making sure it is bone dry -- it's basically gluestick in filament form, so any moisture it comes in contact with will start to turn it sticky. It also doesn't like sticking to the print bed by itself, so remember to use gluestick first. There's also the matter of the purge tower, which is its own logistical problem that deserves its own thread to discuss at length.

It's possible though. It only takes a few things:

• Near zero friction in the full PTFE line
• Near zero tension pulling from the spool
• No bend radius lower than the breaking point of the filament
• Temperature dialed in perfectly to prevent stringing on tip shaping
• Dry storage and handling from the moment you receive the spool
• The patience of a saint to actually dial everything in and deal with all the failures along the way

Other (stronger) support materials will be a lot easier to work with, since the problem of breakage is no longer a concern. PVB, BVOH, breakable supports, etc -- all of these have no issues on the printing side. HOWEVER -- there is still the need for a purge tower, so if there's no adhesion at all between the support material and the model material, you're going to have a bad time (unless, as mentioned before, that issue is solved -- and again... that's a discussion for a different thread).

As far as the print time concern -- yes that is definitely a thing, and can absolutely destroy layer adhesion.

Dedicated extruders to minimize tool change time is the way to go for soluble supports. It's the main reason I want the XL.

Yes, lots of extra time - but that's not *your* time, you can get on with life while it chugs along.

2nd, the reason that I have an interest in the MMU3 is to print with water-soluble (or other material) supports

It rather depends on why you are using soluble support: If you just want smooth overhang surfaces on a figurine or trinket then it's a reasonable target.

If it's a functional part or you are using solubles to seperate print-in-place mechanisms then you need excessive purges, even tiny amounts of solubles carried over into a working part dramatically reduce strength.  This can double filament useage.

Cheerio,

RE: MMU3 Release was is it good for?

It is the latter. I'm a ME and print lots of prototype parts and I'm having a particularly difficult time with ABS supports. 

RE: MMU3 Release was is it good for?

Thanks for your thorough take on MMU3.  It confirms that it is not for me right now. If Prusa offered a dual-extruder machine or add-on, I purchase it in a heartbeat.

Mark

RE: MMU3 Release was is it good for?

You mean other than the XL?

RE: MMU3 Release was is it good for?

Regarding #2... it is possible to only have the support interface layers be soluble, which will potentially help with speed of printing as the rest of the supports are made with what ever filament is being run at that time.

https://help.prusa3d.com/article/mmu2s-and-soluble-materials-pva-bvoh_162860

Posted by: @mbiasotti

Okay, so I see that the release of the MMU3 for Mark 3S+ is is ready to ship.  I have a few questions for those in the know.

1st, if you a doing a multi-color print, expect print times to be greatly increased?  I'm assuming this because it takes extra time to change out filaments to the hot end.
2nd, the reason that I have an interest in the MMU3 is to print with water-soluble (or other material) supports but my assumption is that the MMU3 is NOT a good solution for this since the print time would dramatically increase as supports are built, in many cases, in between print layers etc.  Perhaps a better solution would be a dual extruder if the intention is to do water-soluble supports...

Would appreciate the community's thoughts

Thanks 

Mark

RE:

It seems one thing its NOT going to be good for, is printing of materials with soluble supports.

According to January Dev Diary release, "Both internal and external tests with PLA are overwhelmingly positive. We have spent the last couple of weeks fine-tuning PETG, and we are also getting good results. These are the two materials the MMU3 will officially support. If you need to print other polymers, it is likely possible, but we want to clearly state that only PLA and PETG are the materials we have extensively tested and officially support."

Which, it seems to me is a HUGE disappointment.  Despite 9 months waiting for MMU3 for Mk4, we see the specification being pulled, and it seems, 9 months wasnt enough for them to test with ASA, PC Blend, PVB, HIPS, and other filaments and soluble materials, most of them sold in the eShop and/or made by Prusa themselves.  Seems to me like they really dropped the ball on this one.

RE: MMU3 Release was is it good for?

I would'n say it's not possible. Prusa just is not makeing tests with hundrets of different support material. It's all about tip control. And I'm waiting for the improved "cold plug" tips for the MMU3 @ MK4

RE:

Yeah sure. But at least they can easily test materials they make themselves directly or sell in their eShop. 

Ignoring the use case of using mmu for printing of models with supports (soluble or not) is skin to ignoring half of its market. But it’s clear we have waited 9 months for release of mmu3/mk4 only to find they have pulled the product specification. its a decidedly negative and lazy approach. 

RE: MMU3 Release was is it good for?

"Not being good for", and not being officially supported yet are two very different things. One good thing about this community, someone will try it, and will come up with the appropriate settings for it to work.

RE: MMU3 Release was is it good for? Failing to read the market.

If PRUSA manufactured a dual extrusion MK4 that printed nicely with support materials, I'd buy two today.  I'm more interested in the size of MK4 than the XL. I don't need swappable tool heads to print functional objects.  The printer market is developing in a way PRUSA didn't predict well.  At first the notion of multi-color prints was a perception regarding aesthetics rather mechanical function.  So PRUSA focused on this because a large portion of their user base are hobbyist  printing playful figurines and the like. Machines geared more toward engineering and manufacturing have always had higher price tags that hobbyist won't play with.  Well the printing environment has changed. Dual extrusion machines that print a variety of materials are showing up at more reasonable prices, and all types of users are interested in these machines. If you're a figurine person, a clean model from a print with removable support material is ideal. Due to the fact the model will likely be painted by handed and look amazing compared to five different colors of PLA. 

What I want is a dual extrusion machine that prints a variety of materials that are strong and stable with success over and over again in a timely manner.  The XL could be just the machine, but not exactly. Why wait for print head changes when one extruder can be riding along right next to the other.  Hobbyist may suggest the XL is expensive, whereas someone interested in manufacturing would say it's very reasonable.  As I'm waiting for my MMU3 and learning about all of its problems and limitations, I've started to look around.  I started the  upgrade to MK3.5 with an MMU3, and if this worked out, I was going to purchase a  MK4 with an MMU3 soon after.  Now I'm thinking maybe I'd be better off with Lulzbot Taz Pro dual extruder, or something similar.  Yep, it's more money, though a solid machine.

I certainly won't buy anything made in China. No f-ing way. I like PRUSA is made in a democratic county, and democracy is my baseline for country of printer origin.  Lulzbot, Airwolf, Makergear, and Ultimaker are all catching my eye.  So if the MMU3 is only good for multi-colored PLA trinkets, I'll be pissed.  Feeling . . . it was nice knowing ya. PRUSA please figure out the MMU3 to work with support materials.  

RE: MMU3 Release was is it good for?

What you say has some merit.
I agree on the need for a dual extruder machine that can handle dissimilar materials, one used for support structures. I need that too, and the perceived need is greater than for multi-colour.
I have never been asked for a multicolour print, but have an ongoing need to print models with supports that do not cause artefacts on the model.  No matter if its just a different material that does not stick overly strong;ly to the main material, or for soluble support material.

A Prusa XL with dual extruders could fit the bill, or an IDEX printer same same.

To have that capability in a 250x210x210 machine would be great, but I also have need of a lager print volume from time to time.  At this moment, I cannot justify the extra cost of an XL, but bought another MK4, making 3.   I have an existing MMU3 carried over from a disassembled MK3S+, and another MMU3 coming.  But bitterly disappointed that it cannot handle TPU.  No better than Bambu AMS then.  

Again with the XL, I am disappointed in the continuing lack of an enclosure solution.  Yes there are 3rd party solutions, but they are not ideal, and more trouble in the assembly.
So, I remain dissatisfied in this area. 
But the MK4's are still soldiering on, printing well.  But needing nurturing from time to time, with layer shifting, nozzle ozzing, nozzle clogs etc going on.
It takes all of my accumulated expertise to keep them on track.
The mk4 is a big improvement over the prior MK3S+, but its still got some way to go to be a complete foolproof solution.

Do I think adding of a dual extrusion capability to the MK4, replacing the Nextruder will fit the needs ?  Well lets just get the Nextruder working better first.
I think it will be less in development lag to make a MK4 with two extruders (including the tool changer) than to develop yet another Extruder.