I wonder if any of you saw this 3d printer before:
It prints using powder in a regular crucible, then you put the crucible in a furnace, some metal on top of it. The metal melts and then infuses the powder.
The process seem a bit more involved than regular 3D printing, but it is far easier than regular casting with virtually no shape limitation, so it sounds like a great solution to make complex metal parts at home!
Of course the parts need some post processing to give them a nice finish, but it doesn’t seem much worse than regular sand castings.
Moreover, it seems like the powders and stuff can be purchased for a relatively affordable price (I saw around 50 bucks per kilo).
Unfortunately the machine is not open source, but I don’t think it is really complicated to build one actually.
I was looking at building a metal 3D printer using SLS with powerful lasers, but the cost of the lasers was really prohibitive so I couldn’t really afford it… Here with this technique I don’t see anything being expensive at all !
So I guess after I finish my coreXY printer I will start working on this. Maybe I’m a bit too optimistic but I think there is nothing really impossible to do here for the homegamer, I need to think about a good way to precisely deposit the powders. I think there is no absolute need to use their slicer software, I might be wrong but I believe I could use a regular FDM slicer, in which case I’d just have to create the 3D model embedded in the support material and treat it as a multi material/multi colored one, one color for the model, one other for the support.
Any thought on this matter is very welcome, maybe that could be a fun group project, it could use all the great minds here and maybe we could come up with the holy grail everything is waiting for : a metal 3D printer for the price of a regular FDM printer!
Solid idea, no sintering or major shrinkage…RAD! Not lost casting…but wetting a metal powder after the fact. New to me. Any idea of bond strength between the powder and molten meterial?
What are you thinking for the powder pusher. Screw drive and a funnel? Slap that on an MPCNC and give it a shot. Then you can hold a bunch of weight right on the gantry. Am I naively assuming the screw powder delivery system is the only hiccup, or hardware/thing we haven’t done?
Two color slicing and the purge tower/pile to clear each material.
Seems doable to me…I would love to see it happen. I have to prioritize and get some of my projects wrapped up before I can start any new ones.
I might be wrong, but I’m not even sure they are wetting anything. It seems to me like they just lay some dry sand and some dry metal powder. At least that’s what I plan to do.
I can’t seem to find any good/clear explanation for the working principle actually. From what I understand the metal powder doesn’t even become liquid at any point.
For now I don’t really know how to build the powder pusher, but I just saw this video this morning so I’ll keep thinking about it, don’t expect to find the best solution in such a short time, I need to educate myself a bit first.
In my opinion getting the powder to flow is not the hard part as long as it’s clean and dry, actually what I wonder is how to stop it from flowing. That would be the equivalent of retractions on FDM, you need to stop the flow when going from somewhere to somewhere else. If there could be some easy way to make some kind of micro valve at the very end of the nozzle it would be great.
The plan would be to have one nozzle for each material, to avoid mixing stuff.
Not sure what you mean here, a crucible isn’t very expensive, I suppose you could get a graphite or a ceramic one for a few bucks.
Regarding the process: very cool! It would be interesting to see how the material properties change with a process like that. The video was copper poured into iron, right? So it would not have tensile strength higher than the copper. But it also has a bunch of iron in there, so . Seems like it would be mean to tools
About the nozzle, I think a screw drive, or a auger makes a lot of sense. But it would push it past a threshold, and all the powder past the threshold would end up going all the way out the bottom. I wonder if you would need a sensor to give precise feedback on the volume. Every print is 100% infill.
The metal sand would need to be fine, and well sorted, so each grain is about the same size. I don’t know if that is a typical spec for this kind of thing.
I think a machine with two X carriages would be pretty good at this. An i3 style would have a big pause to lift the nozzle over the edge of the crucible every time. I don’t know for sure, but I don’t think you’d want to shake the crucible on the Y axis. So maybe there is something like a corexy with two Xs and the whole top layer raises and lowers.
The slicer software looked a little goofy, honestly. I am not sure I quite understand all the challenges they have overcome with the software. It looked like they were doing some wider and thinner extrusions. I haven’t done any multi material prints either, so I am not sure how much control you have over the details. Definitely possible though. If we had a machine, someone will probably want to write the software for it.
Are there any problems with patents on this kind of printer?
Unfortunately I don’t think so, I believe that such tiny particles will remain slightly magnetic after you switch off the electromagnet and will quickly end up sticking together. Would have been great though.
Would that be a problem? If you have a screw to apply pressure to the powder then i don’t think it would get clogged in the nozel. Unless you think it would disrupt the print but i feel it would be minor (no more problematic than the issues with filament globbing at the start of a print). But maybe I’m wrong. Another idea could be a rubber tip. (Similar to the rubber valves on cups) the rubber would expand and allow the powder to flow when there is enough pressure behind it. But would close when that pressure is removed.
The rubber tip would also help act as a pressure equalizer. Compensating for small variations in particle size.
I think that would just overcomplicate it. Use two feed hoppers, one small auger in each, feeding into one funnel. No nozzle changing calibrations or anything. I think that would be the benefit of using a CNC frame instead of a printer frame. A CNC frame can handle the weight of two hoppers, pretty easily.
There was a while when I really wanted to be able to make some metal parts using 3D printed methods. What I was looking at at the time was basically making a printer that could print paraffin wax, to use sand to make a lost wax casting. Wax melts are a reasonably low temperature, so it SHOULD be possible to handle. It would be nice if I could make a usable 1.75mm filament of it, then it would be very easy to make it print using a regular 3D printer, but that just doesn’t look feasable.
I’ve done some lost wax casting before, so I thought that it would be a good solution. It is possible to get very good resolution. My problem was actually printing the wax. I looked at pellet hopper extruders and a few other solutions and finally gave up on the idea temporarily. I’ve kind of renewed my interest in the idea, but I don’t think I’m quite ready to make it a reality just yet. Meanwhile solutions like this get put forward, so maybe the problem ends up getting solved for me.
I had a related thought a while ago, if you had copper powder, and it were hot, above the melting point of zinc, and you poured molten zinc into it, maybe, maybe the zinc would dissolve the copper and result in a brass(-ish) casting where the maximum temperature need not be above the melting point of brass. And no zinc fumes. In my mind I was thinking of a more traditional investment style process and not 3D-printed sand. The question is, does there exist a combination of particle size and temperature where the final material is an alloy and not just a zinc casting with lumps of copper?