Prusia i3 Mk3 Multi Material @ MRRF 2018  

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

I'm guessing that's part of what the cutter is for. It would make sense to trim off 1-2cm of filament on every change in order to have an exact known length of filament in the system and also make sure there are no blobs on the end that might get in the way of a clean feed. Having an exact endpoint also means it doesn't need to depend on the filament sensor to get the right amount into the extruder.

To put this in context, the Palette+ filament splicer builds 130mm of purge into every filament change by default. Compared to this, snipping off the ends is nothing--it will still have 75% less waste than the Palette.

First part about trimming filament, makes sense they might do that. I never heard them specifically say that, but, that is something I would honestly expect to POSSIBLY see.
I'm not sure where you get 130mm of purge for Palette... but... no. It's not nearly that much. They had a *BUNCH* of pallet. (almost as many as mk3) Beautiful system. WAY WAY WAY over engineered. I was terrified of them becuase of their complexity... Playing with one hands on, They are built to last. That's also why they cost 800$... more than the mk3 itself.

I have a Palette+ with a fair amount of use. I can assure you that the default settings are in fact for it to purge 130mm of filament on every splice. And yes, this is a lot.

You can adjust this down, but the long purge is necessary because the Palette can get a little out of sync with the print. The longer the purge, the more tolerance you have for slop in the system. I turned my purge down to 50mm, but only after I really had my calibration dialed in.

And yes, the Palette is just as expensive as the MK3. On the other hand, one Palette can be used with multiple printers, which wasn't possible with the original MMU. When my MMU arrives, I have two other (non-Prusa) printers I can move the Palette to.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

With using one extruder there will always be a need to purge out the previous color. Prusa is working on a way to utilize some of this purge as infill when possible. The purge tower as I understand it always has to be maintained at the height of the print to avoid collision when a purge is needed, but the size and amount of infill can be varied. In theory, your purge tower could be almost hollow except for the layers you actually need to change material on. If I got this wrong, someone please clarify.

This is exactly my understanding. I'm not sure on HOW they are going to do the purge tower. But I would simply assume what you said is what they are looking into, or something that has the same "idea", and just different execution.

I have a Palette+ with a fair amount of use. I can assure you that the default settings are in fact for it to purge 130mm of filament on every splice. And yes, this is a lot.

You can adjust this down, but the long purge is necessary because the Palette can get a little out of sync with the print. The longer the purge, the more tolerance you have for slop in the system. I turned my purge down to 50mm, but only after I really had my calibration dialed in.

And yes, the Palette is just as expensive as the MK3. On the other hand, one Palette can be used with multiple printers, which wasn't possible with the original MMU. When my MMU arrives, I have two other (non-Prusa) printers I can move the Palette to.

Yeah, I need to acquire the money to order a MMU for the mk3. I'm 100% set on getting one now.

130mm is absolutely crazy. That's 1/2500 of the ENTIRE SPOOL.

printing a " tall" multimaterial print, with 2 materials of filament...

Your purge block would weigh 1.2kg.

With 4 materials per layer, would yeild 2.4kg of filament in the purge block... That's 2 ENTIRE SPOOLS to print a 4 color cube 200mm tall... That maybe weighs 25g.

Or about 96% waste?!?!?!?

I would *NEVER* under *ANY* circumstance buy a Pallete if that's true.
That's absolutely CRAZY.

Each purge would weigh almost .4g?! With my .25 nozzle, that purge would literally take around 2 mintues!!!!

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

The filament sensor on the multimaterial is on the "filament selector/cutter/nozzle" thing.

After looking at the photos of the MMU2 i am somewhat in doubt if there is filament sensor in the moving carriage. The reason for this is that none of the photos shows wires being routed from this moving head. If there was in filament sensor and if this is complete and working version of the system, we should be able to see these wires.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

The filament sensor on the multimaterial is on the "filament selector/cutter/nozzle" thing.

After looking at the photos of the MMU2 i am somewhat in doubt if there is filament sensor in the moving carriage. The reason for this is that none of the photos shows wires being routed from this moving head. If there was in filament sensor and if this is complete and working version of the system, we should be able to see these wires.

I mentioned 5 filament sensors and got shut down VERY quickly, they said there is only one. If there's only one, there's only one place it can be.

No one said they were functional. I will say there was some filament particles in one of the gears however. So they may actually be functional.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Each purge would weigh almost .4g?! With my .25 nozzle, that purge would literally take around 2 mintues!!!!

A little closer to 0.3g by my math, but yes it's a lot of purge.

There's some things you can do to reduce this. First and foremost, don't print with really thin layers. When doing multicolor I usually print with 0.25mm layers, which cuts the number of transitions way back. Second, don't print really tall objects. The Palette+ is best suited for things like keychains, badges, phone cases, picture frames, etc., which are relatively flat.

You can also reduce the amount of purge with careful design. For example, I design phone cases with just a few layers of color on the bottom (which is the back of the case), and then a single color for the rest of the body. So there's a lot of purge on the first few layers, but none after that so the purge/print ratio is pretty good. It also makes sense where you have large volumes of different colors to print them separately and assemble later (the old fashioned way). Then you save the multicolor printing for details (text, logos, etc.) which would be hard to assemble by hand.

Edited to add:
Those rules for minimizing purge may sound limiting, but it's possible to get some really sharp looking results. I've got some photos on my Etsy shop; all the items were done with my MK3 and Palette+: https://www.etsy.com/shop/BogusLakeDesigns

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

I’m confused. Is this available now or not? People are saying they already ordered one but it is not available on the website (not to me anyway). Where are we getting pictures and info?

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

I’m confused. Is this available now or not? People are saying they already ordered one but it is not available on the website (not to me anyway). Where are we getting pictures and info?

The MK3 MMU is listed on the website under "Printer Upgrades." It used to be available for preorder, but they turned off the "Buy" button at some point.

A bunch of people, including myself, preordered the MK3 MMU last fall. Those preorders are still active, and we're expecting a small refund once they set the new pricing for the MMU2 because we've been told the MMU2 will be a little cheaper.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

A little closer to 0.3g by my math, but yes it's a lot of purge.

There's some things you can do to reduce this. First and foremost, don't print with really thin layers. When doing multicolor I usually print with 0.25mm layers, which cuts the number of transitions way back. Second, don't print really tall objects. The Palette+ is best suited for things like keychains, badges, phone cases, picture frames, etc., which are relatively flat.

You can also reduce the amount of purge with careful design. For example, I design phone cases with just a few layers of color on the bottom (which is the back of the case), and then a single color for the rest of the body. So there's a lot of purge on the first few layers, but none after that so the purge/print ratio is pretty good. It also makes sense where you have large volumes of different colors to print them separately and assemble later (the old fashioned way). Then you save the multicolor printing for details (text, logos, etc.) which would be hard to assemble by hand.

Edited to add:
Those rules for minimizing purge may sound limiting, but it's possible to get some really sharp looking results. I've got some photos on my Etsy shop; all the items were done with my MK3 and Palette+: https://www.etsy.com/shop/BogusLakeDesigns

I'm not saying it don't have it's uses, it's just quite "extreme" on filament waste. Of all the video's I've seen reviewing it there is VERY little mention of it. (I did notice the purge towers were big, but it's hard to compare a MMU for mk2, vs Pallette+ purge block size, from two different reviewers, on different printers, when only watching a video. That's why I'm amazed)

I could have sworn One of the Pallete+ reviews I watched said it was like 25-50% LESS purge than the mk2's MMU. I never asked at MRRF2018, becuase I just assumed what I knew as fact. I am seriously mind blown. (Still in a production enviroment, where you are selling stuff for 10x more than material cost, it's totally worth it, as the material is honestly cheap) What I like about the Pallete+ is the ability to do the "semi-blend", by just changing the colors every ___ mm, without purgeing. Yeilds some AWESOME results.

I’m confused. Is this available now or not? People are saying they already ordered one but it is not available on the website (not to me anyway). Where are we getting pictures and info?

It was avaliable for pre-order since almost when the mk3 was first avaliable. I'm guessing they took it off the website somewhere around the same time they announced the re-design. They have not determined the new price so it makes sense they won't accept orders, as they will have to refund some of the price of what everyone has already purchased, and they likely don't want to do any more of those than they have to, I'm sure they are expensive to do. I'm sure that once the price is decided/announced, they will put it back up for pre-order.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Thank you for posting up the photos. I 've ordered one last fall and have been waiting since then.

What I can make out of the photo (I could be very wrong):

One motor to turn the eccentric bearings to feed and retract different filaments. A clever design I have to say.
Another motor for cutting filament. (i am not so sure about this)
The third motor is to move the feeding tube to different position according to filament selected. Although a tube is used, it is still a direct feeding mechanism.

Looking forwards to seeing it getting refined. Hope it worths the wait.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

If there’s a cutter in the MMU, it opens the possibility of printing filament with an embedded continuous fiber (like the MARKFORGED printer).

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Thank you for posting up the photos. I 've ordered one last fall and have been waiting since then.

What I can make out of the photo (I could be very wrong):

One motor to turn the eccentric bearings to feed and retract different filaments. A clever design I have to say.
Another motor for cutting filament. (i am not so sure about this)
The third motor is to move the feeding tube to different position according to filament selected. Although a tube is used, it is still a direct feeding mechanism.

Looking forwards to seeing it getting refined. Hope it worths the wait.

1. Not really right
2. Wrong.
3. Yes.. kinda

1. The motor on the eccentric bearings simply rotates the free spinning bearings and at the right angle, it will press the filament tightly against the shaft with 5 extruder gears.
2. This is the "Extruder" motor, this turns the ENTIRE shaft with 5 extruder gears on it. Only the filament that is "pressed" against by motor 1, will be fed. (the other filaments will sit on the gears, but unable to feed due to nothing "binding" them into the feed gear.
3. The selector carriage. This is what moves the bowden tube to whichever filament is desired. The cutter is on this assembly, and it will "cut" the filament by moving... They were "hush" about when and why it will cut the filament.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Probably need to do the cut after retracting filament, don't want to feed a blob/stringy thing back in when it's time.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Probably need to do the cut after retracting filament, don't want to feed a blob/stringy thing back in when it's time.

This is exactly my thought. I honestly think this is what it's for.
I know for a fact there will be a TRAY for "cuttings" as well. I can't recall if I mentioned that... whoops.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Ok, so that tray is new info. But i still dont get where that cutting and dropping of a blob should happen, i would've guessed the blade sits on top of the bowden tube, so the cut off part (e.g. the blob from the retraction) would remain inside the tube with no way to drop it off into a tray.

For what you guys talked about to work you would have to retract such a blob beyond the end of the bowden tube, e.g. move the tube a bit in the opposite direction of where the filament you want next sits, move the blob back down a little into free air and then move the tube with a blade attached facing outside (not inside the tube) all the way to the filament you want to switch to, thereby cutting the blob while passing it.

The biggest issue i got with this solution is the massive amount of retraction you'd need (heatbreak to end of bowden) and the remaining issue of retraction problems with the blob. But I guess the last issue might be a non issue, because from what i read the old MMU did mostly jam when trying to reinsert a blob and not when retracting one through the splitter, so no issue pulling it out but rather pushing it back in.

I'm getting more and more curious here 😀

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

2:40 start

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

2:40 start

The engineer made mention of "cut the filament if it don't work the first time" I assumed that was a "concept" rather than "plan", so I did not want to repeat it. I guess I was mistaken.

Also, Just because I found this funny...

For the first 10 seconds you see "Today's 3d Print" talking to someone... invisible... behind tom.
Shortly after you see tom try to clobber me... I seriously think he would have taken me down. He had NO idea I was there. lol!

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Just saw that video yesterday and this seems to imply indeed, that the new MMU is going to retract through the whole tube, otherwise I can't see the need to cut "bad" ends.

I hope they can positively surprise me, but this solution seems to be more complicated, than it needs to be with being capable of cutting filament AND both a direct and bowden extruder motor for pulling/pushing filament every possible way you want from both sides of the tube.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Is there a reason we have not yet seen this thing running on an actual machine yet? hmm?

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Is there a reason we have not yet seen this thing running on an actual machine yet? hmm?

Ummm, because it's not reliable yet?

Given that they're not going to start shipping them for months, my assumption is that it's still under heavy development and there's still kinks to work out.

Re: Prusia i3 Mk3 MuliMaterial @ MRRF 2018

Is there a reason we have not yet seen this thing running on an actual machine yet? hmm?

Ummm, because it's not reliable yet?

Given that they're not going to start shipping them for months, my assumption is that it's still under heavy development and there's still kinks to work out.

MMU will start shipping next month.

I saw the units in person, one of the units had filament ground up in the gear.

I'm guessing there is no software for it yet. It will take a MASSIVE modification to the mk3's firmware. (As the MMU runs it's own arduino and firmware, and digitally talks to the Einsy) So there's going to be a fair chunk of programming that needs to go into this. They never said this, but they did beat around the bush by taking every software question I asked and responded with a whole bunch of "cool ideas" that they MAY do. (That's why I didn't repeat the way the cutter works. I thought it was a "concept" not a "plan")

Once they get these running on a machine, they will be ready to ship.

LITERALLY all the unit does it "auto load" filament. It's kinda a surprising stupid/cheap/efficient system!

My understanding operation.

Filament #1 is loaded. Call for Filament 3 to be loaded.
Bondtech on printer AND MMU both "retract" on filament #1, once past optical sensor on X carriage, speed up retraction, and retract until MMU filament sensor no longer detects filament.
Move FIlament selector from Filament #1 to Filament #3. rotate filament coast roller tube to select filament #3.
Start feeding Filament #3. MMU Filament sensor detects movement, and distance starts counting. X Carriage Filament sensor detects filament, and bondtech gears start "loading" filament. Once "engaged", dis-engage feed gear in MMU. Start priming nozzle, and print purge block.

Mechinically, there's almost NOTHING complex going on.

Software on the other hand... That's actually quite a complex little program there, and until they finalized the mechanical design, working on the software would have been a bad idea, as a small mechanical design bug, could mean re-writing a HUGE chunk of the program.