Modern multi-material printers certainly have their advantages, but all that purging has a way to add up to oodles of waste. Tool-changing printers offer a way to do multi-material prints without the purge waste, but at the cost of complexity. Plastic’s cheap, though, so the logic has been that you could never save enough on materials cost to make up for the added capital cost of a tool-changer — that is, until now.
Currently active on Kickstarter, the Snapmaker U1 promises to change that equation. [Albert] got his hands on a pre-production prototype for a review on 247Printing, and what we see looks promising.
The printer features the ubiquitous 235 mm x 235 mm bed size — pretty much the standard for a printer these days, but quite a lot smaller than the bed of what’s arguably the machine’s closest competition, the tool-changing Prusa XL. On the other hand, at under one thousand US dollars, it’s one quarter the price of Prusa’s top of the line offering. Compared to the XL, it’s faster in every operation, from heating the bed and nozzle to actual printing and even head swapping. That said, as you’d expect from Prusa, the XL comes dialed-in for perfect prints in a way that Snapmaker doesn’t manage — particularly for TPU. You’re also limited to four tool heads, compared to the five supported by the Prusa XL.
The U1 is also faster in multi-material than its price-equivalent competitors from Bambu Lab, up to two to three times shorter print times, depending on the print. It’s worth noting that the actual print speed is comparable, but the Snapmaker takes the lead when you factor in all the time wasted purging and changing filaments.
The assisted spool loading on the sides of the machine uses RFID tags to automatically track the colour and material of Snapmaker filament. That feature seems to take a certain inspiration from the Bambu Labs Mini-AMS, but it is an area [Albert] identifies as needing particular attention from Snapmaker. In the beta configuration he got his hands on, it only loads filament about 50% of the time. One can only imagine the final production models will do better than that!
In spite of that, [Albert] says he’s backing the Kickstarter. Given Snapmaker is an established company — we featured an earlier Snapmaker CNC/Printer/Laser combo machine back in 2021— that’s less of a risk than it could be.
Idk but on my single nozzle dual extruder setup all that purging adds up to about a half gram on smaller prints up to 1.5g on larger prints and that could be improved by making the purge tower smaller. Is anyone wasting 10s of grams per print in purge material?
Well, I don’t own one, but from what I have seen in youtube video, by default, it’s VERY conservative to prevent color mixing, and thus purges a lot (ballpark is ~50%-100% of the object in poops and/or tower.)
You can tune it to purge way less, and there even is a method of “spike creation” that almost removes purging. (basically, you retract a bit the filament and turn off heating. After a small amount of time, you push the filament again and retract it slowly. If correctly calibrated, this maneuver both grab almost all residue of the nozzle, and created a nice spike that won’t clog the tubes when switching. I don’t remember which filament change does it, was it the “Enraged Carrot Feeder”? or the “Box Turtle”? maybe the “Pico MMU”? )
But, yeah, multi material by filament swapping is kinda slow and wasteful out of the box.
“it’s VERY conservative to prevent color mixing, and thus purges a lot”
Im surprised that bambu or some other company hasnt presented a single nozzle color change system that uses a CCD based purge monitor to optimize the transitions
Naw b. I don’t change temperature except just before a tpu retraction otherwise some tpu is left in the nozzle. Changing filaments is just a 15mm retraction at 25mm/s (15mm/s for tpu). There is nearly zero colour mixing and this is solved by printing a ~1.5cm diameter purge tower which is really overkill and the only reason I do so is in case there is a jam (common with tpu) because I can fix it while it is still on the tower.
I really only had to tune the retraction distance and speed to get the nozzle clean.
Well, just today, Bambu Lab teasered their new Vortek system, which shares the induction heating and hotend-changer approach with Bondtech’s INDX. Filament handling is different, though.
https://www.youtube.com/watch?v=rluJj3NEdQA
I guess a system that changes the whole head will be outdated very soon.
I think the Vortek is not the same thing, first of all, the filament need to be unloaded, and loaded it back through a single tube, which means the the bambulab one has to:
Unload the filament
Park the nozzle
Load the new nozzle
Load the new filament
were a real multi tool will only swap the tool.
bear in mind, all of this would be for only one colour change, so, every time it’s 20/30s, sure they address the poop problem but seems like they either:
Didn’t understand the assignment
Didn’t want to looks like they are copying someone (which is worse)
What’s the point to have the fastest printer if you waste time like this?
unloading and loading of filament can be happening while the parking and loading of the hotend is going on. While the full hotend swap is surely faster, I cant imagine this Vortek process being any slower than their current poop method is.
We both know they will not do both at the same time.
I bet it will be as slow, that’s why they focused the video on the poop problem.
Neither of us knows anything. Unless one of us works for Bambu or has the system on loan for testing we are just speculating.
Youre a naysayer forecasting fail.
Im predicting, perhaps optimistically, that the total time, whether simultaneous or not, will not exceed current poop times.
Only time will tell which of us is correct.
It looks like the Vortek is an upgrade for the H2D/H2S systems. Glad its not an entirely new printer only thing.
It has 7 nozzles but While they say ” you can print with up to 24 colors / materials at the same time while Studio is smartly minimizing the amount of purged materials.” I wonder how much control/override the user has over studio’s decision making process.
Also, it would be cool If they could come up with an internal “clean station:” that heated and used some sort of plunger mechanism to prepurge a nozzle that was going to be used with a different filament on its next cycle.
Hmm to get this one or to get a Prusa Core with better local support…
Prusa has just teased a multimaterial for the core.
I love the idea of a tool changer, but not with just different hotends. If you got 4 tools, maybe do FDM printer head, laser engraver, vinyl cutter and rotary tool for cutting wood/steel. I don’t really see a point in 4 different heads besides maybe different nozzle sizes. .4, .6, .8, 1.0 mm could work, but I don’t think i’d ever use that. Different tool heads for different things would be cool.
Unfortunately the chassis needs to be strong for CNC cutting. Now you’ve made it more expensive for people who just want 3D printing. Plus it’s a compromise, so it’s awful for CNC anyway.
Can we fix it? No. We can’t.
The sort of motion system you want for a 3D printer is very different to the kind you need for a router/mill. Laser and vinyl cutting are between these extrema, but this seems like a case where optimising for one compromises all of the others.
I understand the reason for changing hotends in this instance is to remove the “filament poop” of a filament switching system.
Bambu offer combo printer/laser/vinyl/plotter units (No rotary tool though) – the H2D & new H2S
From what the reviews were saying, the H2S has some limitations which kind of cripple it for laser work- the power supply can only support a 10 watt laser, unlike the 10/40 watt H2D.
Snapmaker got started with that kind of multi-machine, though head-swaps IIRC were manual. The last link in the article is to a review of one.
Bambu have just released details of their vortek system, it looks like a step up from what snapmaker are doing. https://www.youtube.com/watch?v=rluJj3NEdQA
“The printer features the ubiquitous 235 mm x 235 mm bed size” Actually the build volume is 270 mm cube.
To Clarify that is the snapmaker machines build volume.
For some reason this comment has been made in response to a mention of Bambu’s Vortek system which is an upgrade for the H2S/H2D which have build volumes of 340320340 mm³ and 350320325 mm³ respectively, How much of that area is sacrificed in the upgrade has yet to be announced.
The article says the U1 bed size is 235mm x 235mm, but the video (@ 6:01) says 270mm x 270mm. Did I misunderstand something?
I really debated joining the Kickstarter for this. I was almost going to, but then I saw videos about Bondtech’s INDX system, which has some advantages. Now seeing the video about Bambu’s upcoming system leads me to believe I’m okay with being patient and waiting to see what the market will offer. I currently use my 3D printer only very occasionally to make practical things, so I’m not in a big hurry to replace it.
So, with this thing, what Open Source software would you use to create the multi-material models?
Blender, export as 3mf
You can also use their “mesh painter” tool within their slicer on your obj/stl files
OpenSCAD obviously!