Color Blending, one of 3D Printing's Worst Fails

Color blending 3d printing seemed like a promising idea... until we tried it.

I'm social!

0:00 Introducing...
01:07 Hopefully Beginnings
03:43 The failings
10:24 What did we Learn?
12:03 RIP

[butre.]

this tech did lead to one thing, the cht nozzle. people noticed pretty early on that these multiplexing hotends could achieve some pretty killer flow rates and just a slightly different implementation gave us one of the biggest game changers in years

[necrokittie]

i forget which big 3d printer youtuber it was but he made a video where he 3d printed his logo into a long serpentine string so the cross section was his logo. he then took that and ran it thru a filament maker to get it to the right shape which puts it thru a hot end.. to get it to 1.75mm and it didn't not mix. then he printed with it and cut a cross section of his printed part and his logo was still there squished down and not mixed... so when you said it wasn't mixing in the nozzle, i wasn't surprised at all.

[dylandreisbach]

Mixing the filaments together to mix color is a great example of good ideas on paper but bad in reality.

Pellet extrusion mixes color pellets with uncolored pellets, that works great. Though it needs a screw and lots of pressure. But surely using filament would work too.

But no, it doesn’t work. Worth a shot though.

[remotepinecone]

I was thinking about this.
If you use white filament and a dye made for plastic it could work.
Or take an ink jet head and hit each voxel after its printed

[ericthecyclist]

I remember years ago seeeing ads for color mixing nozzles, but then I stopped seeing them. Now I understand why.

[andreas.grundler]

Full Color 3D Printing exists, but it is not FDM. Mimaki, for example, sells a resin-based printer that can print with over 10 million colors. A thin layer of white resin is applied as a base and cured. The ink is then applied as with an inkjet printer. The results are amazing but it is a bit more expensive.

[Chad.The.Flornadian]

You know it's gonna be a kickass video when it starts with a bed slinger and fire!

[saddle]

I always thought where they went wrong was not using the infill properly to stop the heat creep. It only has to be the outer wall in the color required, not the whole model.

[naasking]

The problem with with this colour mixing idea is that FDM printing needs the melted plastic to have laminar flow to work well, but good colour mixing needs turbulent flow to evenly distribute the pigments for the right effect. They are just fundamentally incompatible.

[aterxter]

Polymer mixing wouldn't work at all. Just look at cnc kitchen's video about diy volcano cht nozzle, despite there being several strings to cause turbulence in the filament path, the flow stayed laminar due to extreme viscosity and extremely slow speed (the Reynolds number is way too low to have turbulences that would mix). The only option in my opinion would be to have a little motorised impeller in the mixing chamber to forcefully blend the polymers before extruding. An other option would be to have extremely thin and pigment rich filaments to add from all sides to a standard white filament, but that would be extremely difficult to load as you would need at least 5 filaments for every primary color (CMYK). I should try that one day

[sirrola]

An interesting technique could be to have an additional laser engraver which can neutralize specific color pigments by different wavelengths while printing.

[davethebuskeruk]

Really interesting video. I wasn't aware of this printer but I did buy a Geeetech A20M ages ago. That was a 2 in 1 out mixing hot end. Oh boy was that "fun" to get working well-ish. I swapped out the stock extruders for Ender 3 type single gear ones, the hot end assembly had back-flow discs intended to stop the filament pushing back up the heat break, great idea but they just clogged all the time, so I removed those and just made sure I kept filament loaded in both sides. Had to create a custom slicer profile for it based on one I'd found of Thingiverse, then found the purge blocks were so incredibly wasteful so I followed others in designing a purge bucket, then of course it needed custom firmware to work all that. In the end I did wind up with a printer that would print well in one or other loaded colour, it did gradients and mixing ok, and eventually I tuned the colour swaps so it purged into the purge bucket. I just wasn't using it enough to justify the space it was taking up though, so I passed it onto a family member and got another Ender 3 Pro :) It's still running fine and every so often I get a notification that someone has downloaded my profile, firmware etc from Thingiverse so I assume people are still trying with these printers :)

[woodwaker]

Love taps at the end. Looks like an Anet at the end

[biomashed]

This would work, very With a pellet extruder and purging between color changes in order to verify that there is no of the other colors left over. I think a slicer that was designed for a pellet extruder, and then it becomes a multicolor pellet extruder where all the colors at the specific ratio could be mixed together, and the associated pellets that are available, or should I say the extruded selected color that is available can be verified with a flow sensor during print and purge.

[chrisdixon]

I wonder if colour mixing would have worked better if distinct nozzles were used, set at an angle to each other, and came together at a common point, so that you could stop feeding and heating the filaments you aren't currently using

[TheLaXandro]

People printing articulated dragons to sell would love this printer tho. Now you don't need to buy co-extruded filament in limited colours, you can just stock up on any random PLA offcuts and feed the beast. Today's special, strawberry chocolate dragons. Consistent orientation too.
In more practical terms, it seems like a really cool way to make PETG+TPU parts. The two bond together well but not well enough to just print one next to another, co-molded tool handle style, but if you do a few co-extruded layers inbetween you get a strong meshed connection between hard inner and soft outer shell. I played around with the concept with those printed spiral things, they do great at attaching layers of TPU to layers of PETG, but it's not too useful like that.

[zeretube]

I guess, the idea is was not to make a full color model, but to avoid buying a lot of filaments of different color. And it would be limited to tones that always keep all filaments flowing. We will have to wait for next idea to achieve this. I had that idea before 2019, but didn't have resources to make it real.

[ZOMBIEHEADSHOTKILLER]

burning the printer was not needed, even if it wasnt a good one.
There is much more that could be done with it, or its parts.

im gonna give you the benefit of the doubt and assume that it was only exposed to fire for such a short time, that any damage was superficial.

[erobwen]

Maybe they should invent a way to dispense, inject and mix color pigment in powder form directly into a white filament instead.

[greenbuttonpusher_hc]

nhaa...Bambu/Orca can't beat the Q of my old Builder 2 color mix printer. Way to go... :D