Most of the multi color printers out there are AMS/MMU or similar, and there are many DIY options, like Armored Turtle or ECF.
They are an evolutionary dead end. Slow, wasteful, expensive to run.
The Prusa XL, or the Snapmaker U1 are the future direction.
Also a good CoreXY machine like vorons/sovols/ratrigs/VZ, etc can be upgraded with the Bondtech INDX tool changer.
We are talking 5x lower print times, 5x lower material costs.
There is going to be a glut of used Bambus and other multi material unit printers, when print farms unload them, since the tool changers will massively boost their bottom line.
Comments?
There is little reason to abandon a perfectly good older machine that still works just fine all because a newer technology shows up. That’s wasteful of money and material in itself.
I doubt there will be a glut of used Bambu printers being dumped by print farms either. They will continue to use those “old and obsolete” printers until they wear out enough to be replaced at the scheduled time. At which point it could very well make sense to swap them out with better tech. As long as that tech works first time, every time-- which hasn’t been demonstrated yet. It’s one of the major reason many print farms use Bambu printers and not Vorons or Crealty printers.
I may agree in a DIY setting, but for a business it’s a sunken cost fallacy. I’ve owned 2 businesses where productive equipment were integral. Replacing machinery where there is a substantial improvement does make absolute business sense. Also, if you allow a machine to wear out you are shooting yourself in the foot. It’s much better to unload them in the second hand market, and renew. I haven’t owned a print farm, but I’m certain that amortization is fairly quick.
Question is, how long is the expected lifetime of a consumer-grade FFM printer in one of these settings anyway? My bet is that it’s only a couple of years—certainly no more than five. Short enough that many businesses would be able to afford to wait for the expected end-of-life before replacing their old printers, even if there’s new tech out that might make some difference to their bottom line. At worst, the older printers will still be perfectly good and competitive at producing single-material prints, since I don’t think an MMU you don’t use incurs much of a penalty.
If you are actually using it a lot, yeah, definitely.
But a hobbyist that wants to print with support interfaces, or occasionally do some small multicolour prints, or just wants the ability to swap between PLA and PETG without material swaps, they are still pretty great and inexpensive solutions you can bolt on as a simple upgrade.
I kinda view them more as a spool holder upgrade than a proper printer one. And some you can actually swap between printers.
Most MMUs are in the 300€ range. For that money you’ll soon be able to buy an INDX tool changer from Bondtech.
No Brainer.
Anycubic offers them as an add-on for their printers for under ~120€ when purchased with the printer. That’s a reasonable price. But buying them on their own the ~200€ price is a bit much
Is that compatible with other brands? I believe most MMUs are only compatible with their own printers. That’s another advantage of INDX, and other toolchangers I’m sure are in the pipeline. I’m going to be installing an INDX in my heavily modded Ender 5 Plus, and if I ever decide to sell it, or retire it, I know I’ll probably be able to carry it over to another printer.
Memory Management Unit was the first thing that came to my mind when I read that title.
I don’t know what that says about me.
They have their place. If you only do multicolor prints rarely, but change materials between prints a lot, that’s where they excel.
I have both an MMU (Prusa MK3S + MMU 2) and a toolchanger (very custom Voron 2.4 with Tapchanger), and the MMU gets used plenty to swap filament between prints. I look at my toolchanger as being for color prints, and I usually keep 6 colors of PETG on it. My MMU gets used more as the functional printer with all the engineering filaments on it like TPU, PC, ABS, PA. I rarely have to change filament rolls with this setup.
I am also looking at building one of these Swapper3Ds, which should prevent all the waste from printing multiple colors with the MMU.
The swapper may reduce poop, but it looks like it won’t reduce print times.
Also, it looks like it has a ton of failure points. Tool changers are much simpler and faster.
I have no need of this tool changer myself nor am I selling them. But that Swapper IS the OG tool changer. Industry has been using that turret style tool changer since automatic tool changers have been a thing. So there is no new idea involved in this design, just the application.
Those turrets are dead reliable on machining centers and seemingly never wear out. At least I’ve never had to repair one, even on 30 year old machines. To be fair, those are made from hardened steels parts. And not the plastic of the Swapper. Still the design IS well proven on production floors across the planet.
Fun side note: If you look at the photo on their web site, the turret shows 25 slots. But you can’t load 25 tools, you can only load 24. Because you always need 1 empty slot in the turret to make the first tool change. The 25th tool is already loaded in the print head.
As far speed goes, it’s not like you are actually making the tool changes manually. The turret is still much faster than you and the AMS/MMU systems that are common today. Is it as fast as the Prusa XL or Snapmkaer U1? Nope. But speed matters a lot less than being dead reliable. And the Snapmaker or the Bondtech still remain unproven in that regard. The Prusa XL has been proven to be pretty damn reliable, if expensive. If you want a low cost entry, then the AMS/MMU is the proven system.
My personal opinion about the Swapper is-- a cool try, but this isn’t the correct application for this design. It’s too big and the tech isn’t hobbyist friendly except for a tiny handful of users. And it’s an added cost to the money already spent for a filament changer.
The industrial ones may be tough, but the one in the video looks janky and the abundance of printed parts does not inspire confidence in its longevity. Also, it is an addon to an MMU, making the whole multi color set up closer to $500, and limited to bowden, which precludes flexibles. Once a tool has been swapped it must be heated to the proper temp for the filament.
I’m not intimately knowledgeable with subtractive CNC, but I own and have used a lot a hand router, so I’m familiar with the business end of the things. Reliability is not going to be the same with a dry tool or toolhead, than with an oozy nozzle.
The system is for bedslingers only, which are inferior to cartesian or CoreXY cube printers.
The INDX is way more capable, simpler (thus likely more reliable) made by Bondtech, which has a proven reliability and performance record.
The only scenario where I see this being equal or superior to a toolhead changer is, well, none.
I did say that for as reliable as I know turrets to be, (decades of run time 10 to 24 hours a day), this is not the proper application for one. Like you, I don’t find the plastic parts to be very comforting.
The idea, despite being misguided, is an attempt to keep those perfectly good older bed slingers out of landfills by giving them extra capabilities to extend their life cycles.
Corexy IS faster and if you are a hobbyist buying your first or perhaps second printer I would recommend one and they are an excellent choice. But Bambu, Prusa, Soval, and Crealty still sell bed slingers for a reason.
Same here. Printing something like say, a cow, with frequent swaps would be wasteful, but I’ll do parts with 1-2 color swaps. It’s mostly nice as you said though to have multiples “locked and loaded” to do a 1 filament print.
If you have a print with supports, having a support filament that adds little extra time, compared to an MMU, is much more than nice. Being able to use rigid and flexible prints reliably in the same print is more than nice. From what I have read, flexibles in MMUs are prone to failure.
So I have a FlashForge AD5X with the MMU. It worked amazing out of the box, including flawlessly doing some TPU. They actually mentioned the MMU was designed with TPU in mind. That being said: I have been struggling with basic PLA, even after swapping to nozzle that has run only PLA (even though I only ran <10g of TPU through it). I am still new to a lot of this, and don’t feel experienced enough to fault the hardware. What I can say though is it does seem folks are specifically improving the ability of MMUs to handle flexibles. A big reason I got it was to be able to do ABS parts with TPU gaskets. Ask me in a few months.
I looked quite heavily into the swapper3d, but there doesn’t seem to be much development or adoption there.
Bontech indx is the future.
Looks like they’re using some bullshit proprietary nozzle design…not to mention bullshit proprietary hardware platform.
You have to. The only way it works is with a completely new heating system which requires a completely new nozzle design. Do you not know anything about induction heating? And a bullshit proprietary hardware platform that 100% works with klipper and will be copied by Chinese manufacturers, in fact Bambu already copied it.
Things have to change for them to get better.
You have to. The only way it works is with a completely new heating system which requires a completely new nozzle design.
Yes a new design is partly required, we agree. But there is zero need for them to do everything they can to enforce vendor lock-in other than them insisting on that bullshit.
Please list the design choices that were unnecessary, why they were unnecessary, and how you would design it differently to avoid “vendor lock in.”
By open sourcing electrical specs, hardware specs, design specs, communication specs…so that anyone can make 1:1 drop-in compatible equipment for any single piece of the system.
Name one bontech nozzle, hotend, or extruder you cannot buy a clone of. Also it’s not even released.
There’s not a 3D printer manufacturer who does what you’re asking, which is to hand over years and 100s of thousands of dollars of research and development over to other manufacturers to duplicate. Because you’re not making these nozzles at home, you’re not building an induction heater at home. Go ahead and try and claim Prusa still does. Even Prusa’s latest overpriced beta release printer, that is based on 10 year old tech, still isn’t 100% open source.
Yes
Prusa XL is legit good.
They don’t even handle multiple materials well. Changing between high and low temperature materials can clog the nozzle. Changing between materials that don’t stick to each other will make a weak part unless you purge a lot of filament between changes.
The only good uses for them is automatically swapping to a new roll of filament when the first one runs out or printing parts with a small number of color changes.
