Re the MMU2, I really enjoyed tracking down the bolts that fell out and rolled around the floor everytime I opened the MMU2 up to fix something ... glad they finally fixed that after 5 years. And then there was the known 5V issue (that they fixed with that little board included in the MMU3), but that can also be fixed on the MMU2 by rerouting the 5V line to bypass the dodgy diode ... Just a shame that Prusa didn't do anything about that for 5 years (and they knew about it from endless support calls) and that it was finally fixed by some users ... Prusa do make some great printers ... and then they sell complete garbage like the MMU2 that is only fit for hard core hobbyists to spend endless hours fixing ... all for roughly the price of a Bambu AMX ... I am amazed there was never a class action over the MMU2's ...
The 5 year wait for the 5V fix - which you have to pay more for - is poor service. I wasn't plagued with under volt issues like some people were but based on the anecdotal evidence of RUclips and reddit comments it seems pervasive.
Are you kidding me? I literally thought there was something wrong with the PCB components, so I bought a new board, and then I still had the same issue. So I guess I was right after all. The MMU would continue to error out and give the flashing lights of death. That ticks me off. The only advantage for Prusa has over Bambu is they do have reliable machines even though they have GARBAGE customer service... that and the Czech republic is not controlled by the Chinese communist party.
@@AsaMakes It seems like both the original PCB for me and the replacement PCB both have issues. Not sure if they are both 5V issues, but they both would not work for me.
A couple of days ago I changed my order for the MMU2 upgrade for MK4 to the one to the MK3S+ instead. Thanks for the video, it looks like it'll be pretty easy to put together 😊
Thanks for watching! The upgrade is pretty straightforward - the only difficult part is loosening the extruder body to add the 'chimney' filament sensor to the top of the extruder. And if you already have disassembled or partially disassembled an extruder before it will be easy.
off-topic but this is some of the weird moments when different hobbies on the internet mix up, like here my go-to bread baking channel commenting on a random 3D printing hardware video 🙃
The main reason not to buy the MMU3 is that Prusa did a poor job supporting the MMU2. It’s nice to see them moving forward, however, it just seems that an upgrade could have been done much sooner.
Guess it'll be in stock in about a week or so, but if someone is particularly impatient until then: I ordered one without the printed parts and even though it showed two weeks lead time, it shipped the same day.
I already purchased the MMU3 upgrade, just watching for fun. Prusa lead times have been pretty good for me (in the US), I received the kit about a week after ordering. I agree on the spool holders, they're not great. What really sucks about them is that if your spools have any defects on the edge, they won't work. Most spools have slots along the edge to stick the filament in for storage. I have several spools where those slots were dented in during shipping. Catastrophic failure when trying to use them.
I have Mk4 in prusa lack table enclosure (that was for mmu2s). Now I'm building the MMU3 and priting new buffer holders and spool holders to fit the lack enclosure. I hope for the best, thanks for the video
I guess i am one of the lucky ones with the mmu2s and not haveing any issues. Decided to get the upgrade kit cuz the loading of filaments looks so much better in addition to others ive talked to experience with the mmu3
Yeah the filament load and buffer upgrades alone are a huge benefit for the MMU3, and indeed you seem to be one of the lucky ones! It took me a lot of effort to get MMU2s dialed in
I have the X1C combo alongside my MK3s MMU2. The AMS never gets used because of its complete inability to print TPU. Meanwhile the MMU2 is chugging along printing it all day.
I have 12 Prusa printers only 1 is a mini. To your point, the price and lead times for Prusa are almost prohibitive. I just got a P1S combo. Stunned that I got it in days due to US stocking, prices are great and reliability is amazing. The P1S runs nonstop and I need more multi color capability. Currently I'm looking at another P1S combo because of the Prusa cost and lead time.
How is the print farm management running the two different brands? I really like to be able to run the same g code on all my printers Thanks for watching!
Really hasn't been a problem or learning curve. With the variety of Prusa's I have that is problem more of a challenge. I print to P1S mostly from the slicer and the complaints about the P1P/P1S interface is just white noise. I use only .3mf's from Fusion 360. Good luck with your MMU3 and share any updates you have. Thanks @@AsaMakes
Yes this is more correct I got the impression from the blog posts that it works with worse reliability than the MMU2/MMU3 with the Mk3 Either way, it isn't available today
Do they have the cutter up and running on the 3? It seems that is part of the magic juice on the bambu machines is cutting the tips so they don't end up with any stringies to get hung up any where or trip a sensor or accident.
Oh I didn't make these they're one of the common MMU multi color test prints on Printables.com That being said in Prusa Slicer you can use a paint tool to assign different colors to your print. I've used this for my MMU3 quite a bit and it works well and is intuitive Thanks for watching!
Actually the MMU2 had three issues: The 5V-issue, the mechanical stability and the filament tip on unload (the special heatbreak does not result in much improved filament tip without a lot of tuning the ramping profile). The MMU3 fixes the first two issues. The third one can be solved on the mk3s+ by using a Revo Six instead of the original hotend. The integrated Nozzle/Heatbreak combo leaves a much nicer filament tip on unload. Trianglelabs Unified Nozzle might do the trick also, for a much more reasonable price. However, on the mk4 the Nextruder nozzle is even better than the Revo. Practicly no tuning of Ramping Profiles needed. But with the MK4s you need to "downgrade" to a standard flow nozzle. If you already have a MMU2, you should do the upgrade. The two mechanical/electrical parts (the metal drum holder and the smal PCB) you can get for a fraction of the upgrade price on AliExpress. And a TUN hotend is about 25€ (if you have an mk3s).
My biggest problem is the MMUs can not be accommodated with a Lack enclosure; I do not have the option of having the MMU3 in my lack enclosure. So, I will not buy an MMU again. It's horrible when converting the price to Canadian Funds and then getting nailed with at least $30 customs charges on the purchase; just not affordable.
Ah, fortunately there is no customs charge for MMU3 or the 2-3 upgrade kit in the states. Shipping is expensive from Prague but I'm hoping overall prices go down as Prusa bought the US supplier Printed Solid. I'm surprised about the Lack comparability thought - I would think you could run the MMU on the outside of the Lack enclosure and run the PTFE tube through to the extruder on the inside
What are you talking about there are many "mods" to add the MMU to the Lack cabinet "mod". Sounds as if you should do a bit of research. Also its quite easy to addon pretty much anything to the lack cabinets. Heck I out my Ender 3 inside one to print some ABS for my voron v0 and V2.4 parts.
Not to my knowledge, but you don't need to load up 5 colors/materials. 99% of the time I run my MMU3s with 3 colors of PLA and leave material #4 and #5 empty in the MMU. Thanks for watching!
Does this system need a filament cutter in the extruder? Also can someone explain why this doesn’t need to poop out filament like the Bambu AMS? Or maybe I missed it and it does poop. I see the concept is essentially the same although different method and form factor.
With the MMU (2 and 3) there is no filament cutter in the extruder. The way it works is the filament is loaded in and out of the extruder by a boden style drive coming from the MMU itself. Instead of the filament poop from Bambu the previous material or color is just purged into a wipe tower. The wipe tower you can see in the time lapses next to the prints - it's just a rectangle that's the same height as the print. When there is no color change the wipe tower is just built up to maintain the same height as the print. When there is a color change the previous color (material) is purged in the wipe tower.
I believe the approach Prusa uses is to pull back the filament slightly while the hotend is at a certain temperature and then bump the filament down against the inside of the nozzle to shape the tip so it can be withdrawn without jamming. It seems like a really finicky way to try to do it that might vary a lot between types of filament.
@@AsaMakes Thanks that makes sense. But I’m confused why Bambu feels the need to poop out filament when Prusa doesn’t. They both have a tower they print (Bambu’s Prime Tower and Prusa’s Purge Tower) but Bambu also poops out a lot of filament by default. Does the Prusa have the same waste but it’s in the tower instead of pooping it out? Or is the Prusa prone to color contamination? No expectation to answer all my questions, I’m just thinking out loud. Merry Christmas and thanks for the reply!
@@Dave_the_Dave That does seem finicky and that everything has to go exactly right for this to work. But seeing this video it does seem to work at least for the uploader. Cheers!
@@c0mputerIn the Bambu X1C the filament is cut just above the extruder assembly (nozle + colling part). I suppose the do a certain amount of retraction before cutting the filament. But there is almost the full length of the hotend assembly with current color in it, that has to be extruded first, than there is an amount of the new color filament that is used to clean the nozzle so to have a print with the wanted new color. This system is less prone to problems since the clean filament tip, but, and it is a big but, needs a lot of waste to work. Generated waste is proportional to the number of filament change per print. For some 3D printed objects it could be 2, 3 or even 4 times the weight of the printed object. It is a hidden cost that many don't talk about, or don't want to talk about. If You want to print a lot in multicolor, than a multihead printer is economically a better option. Just make a simulation with the slicers of the objects You think to print and You'll get good numbers of the waste of filament, and associated waste of money. Has also to be taken in the ewuation that color change is a slow process in a single head 3d printer. As always all depends on what, how much, with which materials, ... one wants to print.
Looking at the prices to upgrade my mk3s to a 3.5 and order a new Mmu (do not have one currently) I’m looking at 600 shipped. I might as well sell my mk3s and buy a bamboo labs already assembled. Probably cost less as well
If multi material or multi color printing is your priority I tend to agree. I haven't tried out a P1S or X1 yet but I hear quite good things. I prioritize commonality so I'd rather keep my fleet all Prusa to simplify my workflow (for now). Thanks for watching! Let me know what you end up going with.
Orderd my mmu3s upgrade kit for mk3s+ yesterday evening, already have the dpd transfer code so should arive in about 6 day's. Are you going to test the Ultimulti vs prusa mmu3s also?
I've tried multi- material prints with PETG and PLA for easy support material removal, and I've done multi color ASA prints. I haven't tried TPU, PC, or Nylon yet. My only concern would be with TPU I imagine the MMU load and unload might struggle with the flexible material. Thanks for watching!
Wow, that equipment, though. Takes up way too much room. Im absolutely spoiled with the convince of the ams and the look. Prusa really needs to up their game. Printing their parts for printers and mmu looks cheap now a days.
Yeah I agree - one of my main complaints is the overall packaging! If I didn't have a fleet of Prusas already and was starting over I'd probably go with P1S + AMS
@@AsaMakes Filament change on the MMU3 is faster than on the AMS and even the AMS light. I sliced a test print (4 cubes side by side in different colours) both in Bambu Studio for the A1 mini and Prusa Slicer for the mk3s+ with MMU3. Estimated printing time including filament for the Prusa was just 75% of the Bambulab. Also waste on the Prusa was only about 25% compared to Bambu. Can not wait for the MMU3 for the mk4, as this not only increases printing speed and quality, but with the very long heatbreak of the nextruder also should allow faster change due to better filament tip and even more relibility. So unless you buy a toolchanger for multicolour, the Prusa solution actually is the better solution for experienced users if you want full multicolour. However, the AMS is the more beginner friendly and compact. Also if you mostly use the AMS for colouring only the bottom or top layers, use it for horizontal support interface layers or just for changing to the next spool after using up the previous spool, neither waste nor changing time matters much, so the AMS is the better solution here.
Yes it is, though I haven't personally tried it There's a post on Prusa's website: help.prusa3d.com/article/mmu2s-and-soluble-materials-pva-bvoh_162860 (Written for the MMU2 but also relevant for the MMU3) Thanks for watching!
I was hoping for a little bit more. For instance, do the materials need to be the same or can you for instance, print a TPU overmold on a nylon part? If so, how does that work with wipe into part or infill? My primary interest is printing the support material out of something easier to remove. How does that work? What material might that be? It looks like you're probably using PLA, so your spools are sitting out. I have an enclosure because I usually print ASA. What parts should go inside the enclosure and which parts should stay out?
Great questions! Wish I had time to really do a deep dive... I'll answer where I can The materials can be different, for example I'll regularly do PLA supports on PETG parts which makes support removal much easier as the PLA doesn't adhere well to the PLA. I haven't tried TPU in the MMU2 nor 3 - the TPU parts I've needed have all been single material prints. Honestly I'd be a little worried about the stiffness of TPU filament loading and unloading reliably. It would be worth checking Google how well TPU specifically works with the MMUs. Mostly I print PLA but I've done dozens of multi color prints with ASA and the MMU2 before without any issues. There are soluble support filaments as well you could definitely use along with ASA. I keep my PLA spools in a small lab oven when not printing + run a large dehumidifier so I just print with them out in the open above my enclosures hanging from a rod. With my setup the MMUs are inside the enclosure while the spools are outside. If I was worried I would build a small upper enclosure for the spools like the Rep Box. I haven't had any issues impacting the MMU2 when I print ABS and ASA and the temperature inside the enclosure is high. Though I've only done 100 hours or less of ASA/ABS printing on my printers with the MMUs. I Hope this was helpful and thanks for watching!
@@AsaMakes There is a "VarioShore" TPU, wich is quite stiff while cold but the hardness transforms the moment it is molten in the nozzle. Depending on nozzle temperature hardness changes - as higher the printing temperature as softer the print gets. While soft material can not be used in a filament changer (as while changing filament any MMU works like a bowden extruder), VarioShore works on bowden extruders and I saw a video of it being used on the AMS. So it might be usable on the MMU3 also. However, one of the main issues of any filament changer is the quality of the filament tip on unload. While the bambulab printers use cut&poo to get an always perfect filament tip, the MMU3 uses ramping on unload (wich improves a lot if you replace the V6 hotend by a Revo Six or maybe Trianglelabs TUN with integrated Nozzle/Heatbreak combo). So the already molten plastic in the nozzle might not be the best for ramping; cut&poo might be the better aproach here. However, while I haven't done it yet (I just got the A1 mini and just upgraded my mk3s+ to mk4) it might be worth a try.
Interesting! The TPUs I have are quite soft - I'll check out VarioShore if I need to do some multi material print with it I also have a preorder for the XL 5 tool head which I'm super excited about, no idea when it will ship though....
You can, but be aware that some polymers do not mix with each other. That can be a good or bad thing. The good, you can use one polymer as structure and another as support material with near-perfect layer separation. The bad, when polymers do not mix well, it is super-hard to purge out the previous polymer. It will often stick to the wall and will coat the surface of the new polymer for quite some time, resulting in really bad layer adhesion (because of bad mixing and bonding). Others have done some tests which polymers do mix and adhere well or bad. Best example is PLA and PETG that make support removal a breeze. Unfortunately, when purging from PLA to PETG, and vice versa, you should increase the purge volume to make sure that the layer adhesion is not compromised.
@@leeroyjenkins7915 Definitly. Thats one advantage of not using cut&poo - you unload as much from the nozzle/heatbreak as possible on unload, reducing the poo/purge needed. However, if you use multimaterial a lot, you should think about a toolchanger. There are quite nice projects for Ratrig VCore 3.1 and Voron Trident. Of course, Idex might be also an alternative. Even an Idex bedslinger might be fastet than a bambulab ifor full colour prints. Also I think two ERCF for an Idex might be interesting - changing filament on the toolhead not in use at the moment. This would reduce printing time a lot over all... If you want do multimaterial instead multicolour, you still could use the seperate toolheads for two different materials. I think there will be quite interesting improvements in the future.
A filament jam at the PTFE adapter on the filament selector itself, I've never seen it jam there before when my MMUs were MMU2s The filament got stuck on a lip inside the PTFE adapter and I just had to remove the PTFE adapter to get the filament loose, put it back, and continue the print. Thanks for watching!
Lol, no instead of a poop chute the MMU will make a wipe tower on the build plate - in the video you can see the rectangular wipe tower being printed next to the prints
@@AsaMakes The wipe tower is bigger on the MMU3, but the cut&poo produces more waste. The AMS is more than 3 times as wastefull in default settings. Also after cutting, the part of the filament left in the nozzle/heatbreak can only be pooed, not be used to print infill or for another object, as you can not use retractions on the cut filament. On the MMU3 with waste2infill ore waste2object you can reduce the amound of waste a lot. Still a prime tower is needed, but only to eqalice the preasure inside the hotend after filament change. So very smal.
@@Zyghqwyv only for the purge tower, not instead of poo. Poo is the filament left inside the hotend after cutting. As it is cut, it can not be retracted anymore. Theoreticly you could use it for an object - but without being able to retract, this is bad quality and therefore waste anyways.
I prefer to buy my multimaterial solutions from companies that don't take three versions and MANY MANY years to get them working. While the current price is reasonable, add in the price of a MMU2 and a MMU1 for the people like me that wasted many hours on those past products. What is that saying about fool me once, fool me twice...
A fair point - it's definitely painful when the customer (you in this case) are also the technology development testers. I do intend to get an X1C + AMS when I can get one on the used market. Or if BL comes out with a large format printer before my Prusa XL 5 tool head order ships I'll probably cancel my XL order I really like having the fleet of four Mk3s+ so I can just send the same g-code to any of the printers to make my workflow as simple as possible. Even having the one Mk4 is a bit disruptive (though I'm very glad to have the Mk4). For this reason I'll stick with Prusa for the majority of my small print farm for the time being. Thanks for watching and commenting!
I personally kinda feel the opposite. Prusa sent me the upgrade from MMU2 to MMU2s for free, then it was only about $100 shipped for the upgrade to MMU3. Meanwhile they've fully supported community mods and improvements. While the out of box experience might not be as seamless, I much prefer to have a product from a research company with a long history of supporting and upgrading devices (which are open source too) than a complex product I will be unable to repair or upgrade in the future at the discretion of a closed-source company.
What I am really wondering is why you wasted this much filament on those sheep. That purge block obviously would have been the same if you printed them all on one plate. Did you just want to test the reliability? Lol
Yes wasteful but printing them individually does more filament changes, and I give them away as little gifts for Etsy sales For proper testing the torture test is much better than the sheep Thanks for watching!
@@AsaMakes I did some mods years ago when I bought it, but failure rate is at about 90%. Did the firmware update now and run into extruder problems😬. It looks I might have to upgrade to MK3S+ (still have MK3S only) with MMU3. I really like the concept with multicolor, but so far it was just pain
@@markusv311I was running into mmu mcu errors continuously. Prusa had a design flaw with the mmu2 board After the lack of accountability and no help from support, I bought the PD daughter board from Ali express instead of buying the upgrade. It is such a difference. The mmu3 uses the same board but they include a daughter board to address the design flaw.
Well you can buy an X1C +AMS combo and have it in a few days instead of waiting 6 months to a year for an XL multi head or waiting for the Mk4 version of MMU3 to ship (few weeks or months?). So if you need something today I fully understand the Bambu approach. If I saw a deeply discounted X1C + AMS combo on the used market I would probably buy it for testing. But for now I vastly prefer to simplify my workflow and stock with Prusa Mk3s/Mk4/XL. I want to reduce, as much as possible, the different types of printers and configurations
Yeah for new buyers A1 with AMS is such a better price point It's not meant to be enclosed so there are some drawbacks but the price is excellent. For my normal output for Etsy I'll stick to Prusa for now since I don't want more types of printers in my little farm. I like to be able to post the same g code to all printers
Theyre lagging so far behind the competitors it's not even funny. An xl wirh a tool changer is the best for waste reduction but not worth the price. They got complacent and itll take a massive price drop to retain repeat customers. I know they have great customer support but they need to balance it properly with price and performance. 90 percent of their install base isnt following them into the manufacturing space if they ever get their print farms finished on time. There are still alot of giant concessions they are asking of their customers. There's gonna come a point where people are going to have to admit they love stifled innovation with a giant apple-like tax befause they're fanboys. It's not a knock against prusa, if i was into mmu id have a tool changer. Wasting plastic like bambu does is unacceptable. I only print functional stuff for my electronics lab though so i am viewing it from my use case. An mk4 kit costs 2k in Canada so i couldnt justify it. I get why people like them though. Theyre reliable machines with great print quality.
Wow I did not realize how expensive a Mk4 kit is in Canada, shipped to the States a kit is right around 1,000 USD which is still really pricey compared to the A1 (well... once they figure out the bed heater cable fire risk issue). But anyways, the state of things as I see it: - Prusa has high quality, high reliability for single color printing on the Mk4 - Prusa is not cost competitive if buying new - Multi color printing Prusa has fallen behind Bambu I'm really excited for my XL to arrive in a few days but the cost is far too high to justify for 99% of people/use cases. And I think it's highly likely that Bambu launches an XL competitor within the next 18 months. Prusa is definitely innovating on their flagship products and their accessories but they don't appear to prioritize cost reduction for customers. If I didn't already have 4x Mk3s and one Mk4 I would probably buy a Bambu X1C. I have enough capacity for my Etsy store, channel projects, and personal projects and I don't expect to expand more in the near term. I definitely don't plan to replace the fleet with Bambu printers as the Prusas I have work great and I know how to maintain them. Anyyyyways, thanks for watching!
They're still working on the MK4 - MMU3 reliability and it isn't currently available (unless you know something new) "we have decided to not start shipping the MMU3 for the MK4 until we’re sure the reliability is hitting the goals we set. It is extremely hard to estimate when this will happen." blog.prusa3d.com/development-diary-update-on-mmu3-for-the-mk4_85043/ Thanks for watching!
Fortunately for the entire 3D printing market they're still putting out good new product development (nextruder, XL tool changer), but they need to close the gap on price aggressively. I think there's very little that will push most consumers to get a Mk3 or Mk4 when the A1 is getting good reviews for less money. Hopefully the overall competition forces Prusa to innovate more and also push down prices. We'll see... Thanks for watching!
Re the MMU2, I really enjoyed tracking down the bolts that fell out and rolled around the floor everytime I opened the MMU2 up to fix something ... glad they finally fixed that after 5 years. And then there was the known 5V issue (that they fixed with that little board included in the MMU3), but that can also be fixed on the MMU2 by rerouting the 5V line to bypass the dodgy diode ... Just a shame that Prusa didn't do anything about that for 5 years (and they knew about it from endless support calls) and that it was finally fixed by some users ... Prusa do make some great printers ... and then they sell complete garbage like the MMU2 that is only fit for hard core hobbyists to spend endless hours fixing ... all for roughly the price of a Bambu AMX ... I am amazed there was never a class action over the MMU2's ...
The 5 year wait for the 5V fix - which you have to pay more for - is poor service. I wasn't plagued with under volt issues like some people were but based on the anecdotal evidence of RUclips and reddit comments it seems pervasive.
LOL< Yes I hated those bolts too
Are you kidding me? I literally thought there was something wrong with the PCB components, so I bought a new board, and then I still had the same issue. So I guess I was right after all. The MMU would continue to error out and give the flashing lights of death. That ticks me off.
The only advantage for Prusa has over Bambu is they do have reliable machines even though they have GARBAGE customer service... that and the Czech republic is not controlled by the Chinese communist party.
@@AsaMakes It seems like both the original PCB for me and the replacement PCB both have issues. Not sure if they are both 5V issues, but they both would not work for me.
Wow super unlucky... is it working now at least with a 3rd replacement?
A couple of days ago I changed my order for the MMU2 upgrade for MK4 to the one to the MK3S+ instead. Thanks for the video, it looks like it'll be pretty easy to put together 😊
Thanks for watching! The upgrade is pretty straightforward - the only difficult part is loosening the extruder body to add the 'chimney' filament sensor to the top of the extruder. And if you already have disassembled or partially disassembled an extruder before it will be easy.
@@AsaMakes Thanks. I use the kits so I can both assemble and disassmble the printer 😁
Perfect you should have no problem then
off-topic but this is some of the weird moments when different hobbies on the internet mix up, like here my go-to bread baking channel commenting on a random 3D printing hardware video 🙃
The main reason not to buy the MMU3 is that Prusa did a poor job supporting the MMU2. It’s nice to see them moving forward, however, it just seems that an upgrade could have been done much sooner.
A very fair point, they did a poor job with the MMU2 and it took too long for a fix to be available
I can number on one hand how many successful prints I had on my mmu2. Its gathering dust here somewhere.
Guess it'll be in stock in about a week or so, but if someone is particularly impatient until then: I ordered one without the printed parts and even though it showed two weeks lead time, it shipped the same day.
I already purchased the MMU3 upgrade, just watching for fun. Prusa lead times have been pretty good for me (in the US), I received the kit about a week after ordering. I agree on the spool holders, they're not great. What really sucks about them is that if your spools have any defects on the edge, they won't work. Most spools have slots along the edge to stick the filament in for storage. I have several spools where those slots were dented in during shipping. Catastrophic failure when trying to use them.
Thanks for watching!!
I have Mk4 in prusa lack table enclosure (that was for mmu2s). Now I'm building the MMU3 and priting new buffer holders and spool holders to fit the lack enclosure. I hope for the best, thanks for the video
Good luck! My two MMU3s are still working great
Great information. Thank you for the objective breakdown of the product.
Thank you for watching!
MMU2 fiasco is still fresh in my memory... saving for the 5 toolhead XL
Mine should ship in March/April. So excited!
By building the 5 toolhead just proves any MMU version will always be junk. One head for each filament.
The XL 5-toolhead is a game changer for me
I still use my two Mk3s w/ MMU3 very regularly but the XL multi head clearly outperforms
For the spool holders, maybe you could fashion something heavy to insert into the center of the spool, to hold it down when it runs low on filament.
Great review thanks!
Forgot to add, great review.
I guess i am one of the lucky ones with the mmu2s and not haveing any issues. Decided to get the upgrade kit cuz the loading of filaments looks so much better in addition to others ive talked to experience with the mmu3
Yeah the filament load and buffer upgrades alone are a huge benefit for the MMU3, and indeed you seem to be one of the lucky ones! It took me a lot of effort to get MMU2s dialed in
You are a rare one indeed. Total dis satisfaction with those units.
Looks pretty good! Would rather stick with my AMS and X1 from Bambu Labs though.
I definitely also want an X1C with AMS but can't currently justify the cost... maybe if my RUclips channel takes off ;-)
Thanks for watching!
I have a X1 "Non Carbon". I snagged one before bambu discontinued them.@@AsaMakes
I have the X1C combo alongside my MK3s MMU2. The AMS never gets used because of its complete inability to print TPU. Meanwhile the MMU2 is chugging along printing it all day.
I don't print TPU. If I printed TPU i would just print a y-splitter and call it good. The MK4's do look pretty solid.@@kirche5
I have 12 Prusa printers only 1 is a mini. To your point, the price and lead times for Prusa are almost prohibitive. I just got a P1S combo. Stunned that I got it in days due to US stocking, prices are great and reliability is amazing. The P1S runs nonstop and I need more multi color capability. Currently I'm looking at another P1S combo because of the Prusa cost and lead time.
How is the print farm management running the two different brands? I really like to be able to run the same g code on all my printers
Thanks for watching!
Really hasn't been a problem or learning curve. With the variety of Prusa's I have that is problem more of a challenge. I print to P1S mostly from the slicer and the complaints about the P1P/P1S interface is just white noise. I use only .3mf's from Fusion 360. Good luck with your MMU3 and share any updates you have. Thanks @@AsaMakes
I still can't believe that the MMU doesn't work on the MK4 yet.
Yeah it's really a gap in the MK4 capability
I believe its not "not work" but more like "cant compete" with others for current state.
Yes this is more correct
I got the impression from the blog posts that it works with worse reliability than the MMU2/MMU3 with the Mk3
Either way, it isn't available today
And thanks for watching!
Do they have the cutter up and running on the 3? It seems that is part of the magic juice on the bambu machines is cutting the tips so they don't end up with any stringies to get hung up any where or trip a sensor or accident.
Yes there's a blade on the filament selector that trims off any stringing (similar setup as MMU2)
Sir: HOW did you come up with those sheep files, such that some parts come out white, but other parts come out black? What wizardry is THAT?
Oh I didn't make these they're one of the common MMU multi color test prints on Printables.com
That being said in Prusa Slicer you can use a paint tool to assign different colors to your print. I've used this for my MMU3 quite a bit and it works well and is intuitive
Thanks for watching!
my MMU1 was amazing, but my MMU2 I never got working properly no matter how much i tried. Not bothering with a MMU3.
Actually the MMU2 had three issues: The 5V-issue, the mechanical stability and the filament tip on unload (the special heatbreak does not result in much improved filament tip without a lot of tuning the ramping profile).
The MMU3 fixes the first two issues. The third one can be solved on the mk3s+ by using a Revo Six instead of the original hotend. The integrated Nozzle/Heatbreak combo leaves a much nicer filament tip on unload.
Trianglelabs Unified Nozzle might do the trick also, for a much more reasonable price.
However, on the mk4 the Nextruder nozzle is even better than the Revo. Practicly no tuning of Ramping Profiles needed. But with the MK4s you need to "downgrade" to a standard flow nozzle.
If you already have a MMU2, you should do the upgrade. The two mechanical/electrical parts (the metal drum holder and the smal PCB) you can get for a fraction of the upgrade price on AliExpress. And a TUN hotend is about 25€ (if you have an mk3s).
My biggest problem is the MMUs can not be accommodated with a Lack enclosure; I do not have the option of having the MMU3 in my lack enclosure. So, I will not buy an MMU again. It's horrible when converting the price to Canadian Funds and then getting nailed with at least $30 customs charges on the purchase; just not affordable.
Ah, fortunately there is no customs charge for MMU3 or the 2-3 upgrade kit in the states. Shipping is expensive from Prague but I'm hoping overall prices go down as Prusa bought the US supplier Printed Solid.
I'm surprised about the Lack comparability thought - I would think you could run the MMU on the outside of the Lack enclosure and run the PTFE tube through to the extruder on the inside
What are you talking about there are many "mods" to add the MMU to the Lack cabinet "mod". Sounds as if you should do a bit of research. Also its quite easy to addon pretty much anything to the lack cabinets. Heck I out my Ender 3 inside one to print some ABS for my voron v0 and V2.4 parts.
I have a mmu3 on my lack enclosure. Plenty of mods that make it work.
oh wow, quite a bulky mmu system. would never fit in my room. i need to think about this.
Is there a 3 color MMU instead?
Not to my knowledge, but you don't need to load up 5 colors/materials. 99% of the time I run my MMU3s with 3 colors of PLA and leave material #4 and #5 empty in the MMU.
Thanks for watching!
Does this system need a filament cutter in the extruder? Also can someone explain why this doesn’t need to poop out filament like the Bambu AMS? Or maybe I missed it and it does poop. I see the concept is essentially the same although different method and form factor.
With the MMU (2 and 3) there is no filament cutter in the extruder. The way it works is the filament is loaded in and out of the extruder by a boden style drive coming from the MMU itself. Instead of the filament poop from Bambu the previous material or color is just purged into a wipe tower. The wipe tower you can see in the time lapses next to the prints - it's just a rectangle that's the same height as the print. When there is no color change the wipe tower is just built up to maintain the same height as the print. When there is a color change the previous color (material) is purged in the wipe tower.
I believe the approach Prusa uses is to pull back the filament slightly while the hotend is at a certain temperature and then bump the filament down against the inside of the nozzle to shape the tip so it can be withdrawn without jamming. It seems like a really finicky way to try to do it that might vary a lot between types of filament.
@@AsaMakes Thanks that makes sense. But I’m confused why Bambu feels the need to poop out filament when Prusa doesn’t. They both have a tower they print (Bambu’s Prime Tower and Prusa’s Purge Tower) but Bambu also poops out a lot of filament by default. Does the Prusa have the same waste but it’s in the tower instead of pooping it out? Or is the Prusa prone to color contamination?
No expectation to answer all my questions, I’m just thinking out loud. Merry Christmas and thanks for the reply!
@@Dave_the_Dave That does seem finicky and that everything has to go exactly right for this to work. But seeing this video it does seem to work at least for the uploader. Cheers!
@@c0mputerIn the Bambu X1C the filament is cut just above the extruder assembly (nozle + colling part). I suppose the do a certain amount of retraction before cutting the filament. But there is almost the full length of the hotend assembly with current color in it, that has to be extruded first, than there is an amount of the new color filament that is used to clean the nozzle so to have a print with the wanted new color.
This system is less prone to problems since the clean filament tip, but, and it is a big but, needs a lot of waste to work.
Generated waste is proportional to the number of filament change per print.
For some 3D printed objects it could be 2, 3 or even 4 times the weight of the printed object.
It is a hidden cost that many don't talk about, or don't want to talk about.
If You want to print a lot in multicolor, than a multihead printer is economically a better option. Just make a simulation with the slicers of the objects You think to print and You'll get good numbers of the waste of filament, and associated waste of money.
Has also to be taken in the ewuation that color change is a slow process in a single head 3d printer.
As always all depends on what, how much, with which materials, ... one wants to print.
Looking at the prices to upgrade my mk3s to a 3.5 and order a new Mmu (do not have one currently) I’m looking at 600 shipped. I might as well sell my mk3s and buy a bamboo labs already assembled. Probably cost less as well
If multi material or multi color printing is your priority I tend to agree. I haven't tried out a P1S or X1 yet but I hear quite good things.
I prioritize commonality so I'd rather keep my fleet all Prusa to simplify my workflow (for now).
Thanks for watching! Let me know what you end up going with.
Thank you
Thank you for watching!
Well done video.
Thank you! And thanks for watching!
Orderd my mmu3s upgrade kit for mk3s+ yesterday evening, already have the dpd transfer code so should arive in about 6 day's. Are you going to test the Ultimulti vs prusa mmu3s also?
Oh wow I've never heard of the ultimulti - I've got some reading up to do!
So what multiple materials will it handle? I don't recall you mentioning what materials you tested. You mention colors but I'm not interested in that.
I've tried multi- material prints with PETG and PLA for easy support material removal, and I've done multi color ASA prints. I haven't tried TPU, PC, or Nylon yet. My only concern would be with TPU I imagine the MMU load and unload might struggle with the flexible material.
Thanks for watching!
Wow, that equipment, though. Takes up way too much room. Im absolutely spoiled with the convince of the ams and the look. Prusa really needs to up their game. Printing their parts for printers and mmu looks cheap now a days.
Yeah I agree - one of my main complaints is the overall packaging! If I didn't have a fleet of Prusas already and was starting over I'd probably go with P1S + AMS
@@AsaMakes Filament change on the MMU3 is faster than on the AMS and even the AMS light. I sliced a test print (4 cubes side by side in different colours) both in Bambu Studio for the A1 mini and Prusa Slicer for the mk3s+ with MMU3. Estimated printing time including filament for the Prusa was just 75% of the Bambulab. Also waste on the Prusa was only about 25% compared to Bambu. Can not wait for the MMU3 for the mk4, as this not only increases printing speed and quality, but with the very long heatbreak of the nextruder also should allow faster change due to better filament tip and even more relibility.
So unless you buy a toolchanger for multicolour, the Prusa solution actually is the better solution for experienced users if you want full multicolour. However, the AMS is the more beginner friendly and compact. Also if you mostly use the AMS for colouring only the bottom or top layers, use it for horizontal support interface layers or just for changing to the next spool after using up the previous spool, neither waste nor changing time matters much, so the AMS is the better solution here.
They eat their own dog food.
id rather buy bambuu but very good video👍
Would this be suitable for PVA supports?
Yes it is, though I haven't personally tried it
There's a post on Prusa's website: help.prusa3d.com/article/mmu2s-and-soluble-materials-pva-bvoh_162860
(Written for the MMU2 but also relevant for the MMU3)
Thanks for watching!
I was hoping for a little bit more. For instance, do the materials need to be the same or can you for instance, print a TPU overmold on a nylon part? If so, how does that work with wipe into part or infill? My primary interest is printing the support material out of something easier to remove. How does that work? What material might that be? It looks like you're probably using PLA, so your spools are sitting out. I have an enclosure because I usually print ASA. What parts should go inside the enclosure and which parts should stay out?
Great questions! Wish I had time to really do a deep dive... I'll answer where I can
The materials can be different, for example I'll regularly do PLA supports on PETG parts which makes support removal much easier as the PLA doesn't adhere well to the PLA. I haven't tried TPU in the MMU2 nor 3 - the TPU parts I've needed have all been single material prints. Honestly I'd be a little worried about the stiffness of TPU filament loading and unloading reliably. It would be worth checking Google how well TPU specifically works with the MMUs.
Mostly I print PLA but I've done dozens of multi color prints with ASA and the MMU2 before without any issues. There are soluble support filaments as well you could definitely use along with ASA.
I keep my PLA spools in a small lab oven when not printing + run a large dehumidifier so I just print with them out in the open above my enclosures hanging from a rod.
With my setup the MMUs are inside the enclosure while the spools are outside. If I was worried I would build a small upper enclosure for the spools like the Rep Box.
I haven't had any issues impacting the MMU2 when I print ABS and ASA and the temperature inside the enclosure is high. Though I've only done 100 hours or less of ASA/ABS printing on my printers with the MMUs.
I Hope this was helpful and thanks for watching!
@@AsaMakes There is a "VarioShore" TPU, wich is quite stiff while cold but the hardness transforms the moment it is molten in the nozzle. Depending on nozzle temperature hardness changes - as higher the printing temperature as softer the print gets.
While soft material can not be used in a filament changer (as while changing filament any MMU works like a bowden extruder), VarioShore works on bowden extruders and I saw a video of it being used on the AMS. So it might be usable on the MMU3 also.
However, one of the main issues of any filament changer is the quality of the filament tip on unload. While the bambulab printers use cut&poo to get an always perfect filament tip, the MMU3 uses ramping on unload (wich improves a lot if you replace the V6 hotend by a Revo Six or maybe Trianglelabs TUN with integrated Nozzle/Heatbreak combo). So the already molten plastic in the nozzle might not be the best for ramping; cut&poo might be the better aproach here.
However, while I haven't done it yet (I just got the A1 mini and just upgraded my mk3s+ to mk4) it might be worth a try.
Interesting! The TPUs I have are quite soft - I'll check out VarioShore if I need to do some multi material print with it
I also have a preorder for the XL 5 tool head which I'm super excited about, no idea when it will ship though....
You can, but be aware that some polymers do not mix with each other. That can be a good or bad thing.
The good, you can use one polymer as structure and another as support material with near-perfect layer separation. The bad, when polymers do not mix well, it is super-hard to purge out the previous polymer. It will often stick to the wall and will coat the surface of the new polymer for quite some time, resulting in really bad layer adhesion (because of bad mixing and bonding).
Others have done some tests which polymers do mix and adhere well or bad. Best example is PLA and PETG that make support removal a breeze. Unfortunately, when purging from PLA to PETG, and vice versa, you should increase the purge volume to make sure that the layer adhesion is not compromised.
@@leeroyjenkins7915 Definitly. Thats one advantage of not using cut&poo - you unload as much from the nozzle/heatbreak as possible on unload, reducing the poo/purge needed. However, if you use multimaterial a lot, you should think about a toolchanger. There are quite nice projects for Ratrig VCore 3.1 and Voron Trident. Of course, Idex might be also an alternative. Even an Idex bedslinger might be fastet than a bambulab ifor full colour prints. Also I think two ERCF for an Idex might be interesting - changing filament on the toolhead not in use at the moment. This would reduce printing time a lot over all... If you want do multimaterial instead multicolour, you still could use the seperate toolheads for two different materials.
I think there will be quite interesting improvements in the future.
What was the failure you encountered?
A filament jam at the PTFE adapter on the filament selector itself, I've never seen it jam there before when my MMUs were MMU2s
The filament got stuck on a lip inside the PTFE adapter and I just had to remove the PTFE adapter to get the filament loose, put it back, and continue the print.
Thanks for watching!
does this not poop like the bambu ams?
Lol, no instead of a poop chute the MMU will make a wipe tower on the build plate - in the video you can see the rectangular wipe tower being printed next to the prints
It's more wasteful in terms of filament waste than the Bambu AMS
@@AsaMakes The wipe tower is bigger on the MMU3, but the cut&poo produces more waste. The AMS is more than 3 times as wastefull in default settings. Also after cutting, the part of the filament left in the nozzle/heatbreak can only be pooed, not be used to print infill or for another object, as you can not use retractions on the cut filament. On the MMU3 with waste2infill ore waste2object you can reduce the amound of waste a lot. Still a prime tower is needed, but only to eqalice the preasure inside the hotend after filament change. So very smal.
@@oleurgast730 you can purge to object and infill for the AMS
@@Zyghqwyv only for the purge tower, not instead of poo. Poo is the filament left inside the hotend after cutting. As it is cut, it can not be retracted anymore. Theoreticly you could use it for an object - but without being able to retract, this is bad quality and therefore waste anyways.
I prefer to buy my multimaterial solutions from companies that don't take three versions and MANY MANY years to get them working. While the current price is reasonable, add in the price of a MMU2 and a MMU1 for the people like me that wasted many hours on those past products. What is that saying about fool me once, fool me twice...
A fair point - it's definitely painful when the customer (you in this case) are also the technology development testers.
I do intend to get an X1C + AMS when I can get one on the used market. Or if BL comes out with a large format printer before my Prusa XL 5 tool head order ships I'll probably cancel my XL order
I really like having the fleet of four Mk3s+ so I can just send the same g-code to any of the printers to make my workflow as simple as possible. Even having the one Mk4 is a bit disruptive (though I'm very glad to have the Mk4). For this reason I'll stick with Prusa for the majority of my small print farm for the time being.
Thanks for watching and commenting!
I personally kinda feel the opposite. Prusa sent me the upgrade from MMU2 to MMU2s for free, then it was only about $100 shipped for the upgrade to MMU3. Meanwhile they've fully supported community mods and improvements. While the out of box experience might not be as seamless, I much prefer to have a product from a research company with a long history of supporting and upgrading devices (which are open source too) than a complex product I will be unable to repair or upgrade in the future at the discretion of a closed-source company.
What I am really wondering is why you wasted this much filament on those sheep. That purge block obviously would have been the same if you printed them all on one plate. Did you just want to test the reliability? Lol
Yes wasteful but printing them individually does more filament changes, and I give them away as little gifts for Etsy sales
For proper testing the torture test is much better than the sheep
Thanks for watching!
MMU2S sucks like hell, hope the 3 irons it out. The concept itself is cool
Yeah definitely a huge improvement over the stock MMU2, though if you do have an MMU2 there are some mods you can do to really improve it
@@AsaMakes I did some mods years ago when I bought it, but failure rate is at about 90%. Did the firmware update now and run into extruder problems😬. It looks I might have to upgrade to MK3S+ (still have MK3S only) with MMU3. I really like the concept with multicolor, but so far it was just pain
@@markusv311I was running into mmu mcu errors continuously. Prusa had a design flaw with the mmu2 board After the lack of accountability and no help from support, I bought the PD daughter board from Ali express instead of buying the upgrade. It is such a difference. The mmu3 uses the same board but they include a daughter board to address the design flaw.
Why would someone buy a crappy SUPER wasteful Bambu with an AMS when this can do 5 colors with no waste?
And thanks for watching!
Well you can buy an X1C +AMS combo and have it in a few days instead of waiting 6 months to a year for an XL multi head or waiting for the Mk4 version of MMU3 to ship (few weeks or months?). So if you need something today I fully understand the Bambu approach.
If I saw a deeply discounted X1C + AMS combo on the used market I would probably buy it for testing. But for now I vastly prefer to simplify my workflow and stock with Prusa Mk3s/Mk4/XL. I want to reduce, as much as possible, the different types of printers and configurations
@@AsaMakes It took 3 weeks for my Bambu printer to come. And at half the price, the mmu i can wait a little for
Oh wow I'm surprised it took that long. Last I was looking they had stock at Microcenter which fortunately I live close enough to
@@AsaMakes Yeah theres a Mcirocenter in my state but its 4.5hr drive. So i just waited. 9hrs round trip for a printer , huh uh lol
well just buy a bambu A1 with lite AMS and your good rip prusa
Yeah for new buyers A1 with AMS is such a better price point
It's not meant to be enclosed so there are some drawbacks but the price is excellent.
For my normal output for Etsy I'll stick to Prusa for now since I don't want more types of printers in my little farm. I like to be able to post the same g code to all printers
And thanks for watching!
Theyre lagging so far behind the competitors it's not even funny. An xl wirh a tool changer is the best for waste reduction but not worth the price. They got complacent and itll take a massive price drop to retain repeat customers. I know they have great customer support but they need to balance it properly with price and performance. 90 percent of their install base isnt following them into the manufacturing space if they ever get their print farms finished on time. There are still alot of giant concessions they are asking of their customers. There's gonna come a point where people are going to have to admit they love stifled innovation with a giant apple-like tax befause they're fanboys. It's not a knock against prusa, if i was into mmu id have a tool changer. Wasting plastic like bambu does is unacceptable. I only print functional stuff for my electronics lab though so i am viewing it from my use case. An mk4 kit costs 2k in Canada so i couldnt justify it. I get why people like them though. Theyre reliable machines with great print quality.
Wow I did not realize how expensive a Mk4 kit is in Canada, shipped to the States a kit is right around 1,000 USD which is still really pricey compared to the A1 (well... once they figure out the bed heater cable fire risk issue). But anyways, the state of things as I see it:
- Prusa has high quality, high reliability for single color printing on the Mk4
- Prusa is not cost competitive if buying new
- Multi color printing Prusa has fallen behind Bambu
I'm really excited for my XL to arrive in a few days but the cost is far too high to justify for 99% of people/use cases. And I think it's highly likely that Bambu launches an XL competitor within the next 18 months.
Prusa is definitely innovating on their flagship products and their accessories but they don't appear to prioritize cost reduction for customers.
If I didn't already have 4x Mk3s and one Mk4 I would probably buy a Bambu X1C. I have enough capacity for my Etsy store, channel projects, and personal projects and I don't expect to expand more in the near term. I definitely don't plan to replace the fleet with Bambu printers as the Prusas I have work great and I know how to maintain them.
Anyyyyways, thanks for watching!
You printed 12 multicolored sheep, one by one? That's what I call wasteful.
Yep, not the best stress test but definitely easy
I give them away as little Etsy freebies
It's compatible with the mk4 lol
They're still working on the MK4 - MMU3 reliability and it isn't currently available (unless you know something new)
"we have decided to not start shipping the MMU3 for the MK4 until we’re sure the reliability is hitting the goals we set. It is extremely hard to estimate when this will happen."
blog.prusa3d.com/development-diary-update-on-mmu3-for-the-mk4_85043/
Thanks for watching!
anyone else find the npc movement a little jarring
Lol
I loved it. :) @@AsaMakes
Prusa went from pioneers to over-priced real quick. I love my mk3s, but damn.... little to no features that make it worth the price.
Fortunately for the entire 3D printing market they're still putting out good new product development (nextruder, XL tool changer), but they need to close the gap on price aggressively. I think there's very little that will push most consumers to get a Mk3 or Mk4 when the A1 is getting good reviews for less money.
Hopefully the overall competition forces Prusa to innovate more and also push down prices. We'll see...
Thanks for watching!
An absolute nightmare of equipment, compared to the Bambulab AMS
I plan to get a P1S AMS combo or an X1C AMS, just waiting for a cheap one to show up on the used market.
It's so bulky!!!