So i have a idea for a metal 3D printer and i just want you guys to give me a response of what you think about it.
Basically it’s a normal FDM 3d printer. But the thing that differs is that the bed is cooled and that it uses flux in combination with the metal filament to prevent oxidation.
The proof of concept is to use normal solder because it has flux in it’s core and try to extrude that. Also to have a cooled bed so that the heat can be transferred away quickly otherwise the metal will not properly cool down and form a puddle. The solder is also a good start metal because it’s easy to find and it only needs to be about 300 C for it to melt.
Some problems may be the volume changing on the metal and depending on what metal it is. Also, having a cooled bed might lead to some problems i don’t know of. Although i haven’t seen or heard of metal warping it might happen here i don’t really know.
I have not thought through this at all, it’s just some random idea i had and because i really want to have a metal 3d printer.
Markforged makes the closest analogue to what you’re describing. They have massive software compensation for shrinking. Maintaining bed adhesion while also controlling temperature may be difficult, you may need to look into a sacrificial build plate of some variety. Check out what happens when a drop of solder hits the ground (a cool plane): It doesn’t adhere, and is easy to pick up.
To be a little blunt, if extruding spooled metal out of an FDM machine was so simple, why can’t I buy a device like that from 3DS, Stratasys, etc? The requirements for a metal printer massively exceed that of traditional plastic FFF/FDM machines. If the goal is to simply push something metallic, this can be done with metal-fill filament. If the goal is to create functional parts out of usable and quantifiable alloys (6061 T6, Inconel varieties, 316L, etc), the problem starts to get more expensive to solve. Right now the cheapest option is the Markforged Metal X, but they use a proprietary plastic binder in the filament to ensure model adhesion and shape. The binder gets washed then burned out, and the final part is something like 98% dense metal.
Basically, metal printing is hard. It’s not impossible, and maybe there’s a breakthrough to be had. Just don’t expect to throw reprap hardware at the problem and come away with blazing success.
Yeah i see your point. I’m not going into this thinking it will work. Again, i just like to play around with the idea in my head and figure out why it won’t work and what will work. Thanks for suggesting the Markforged Metal X.
The problems start showing up when you take a usable alloy (aluminum, 6061 T6, for example), and heat it to glass transition. Additives and bits that make the alloy valuable start to burn off, separate, etc. The extruded metal is generally not precisely the same metal you began with.
If you’re OK with "Well, it’s kinda close to 6061 but tbh we don’t really know how every print will come out), then by all means FDM metal might work for you.
Sintering powder with a high power laser seems like a more direct solution to me.
Well, i didn’t even think of trying with any aluminium alloy and definitely not something like steel. Again also not expecting results that are usable or having the metal keep it’s properties. Especially not getting quality prints or anything. I can’t start to imagine what hell awaits if i try to make this reliable.
I just want to see what happens when you try doing metal FDM that doesn’t need to be treated or altered after printing. Again because the cost and to trying to keep it small.
There are a few companies using MIG welders to print. They are pretty messy & very slow, but it’s probably the cheapest form of metal printing, its just a cnc welder, which have been around for decades.
this is actually sort of interesting. The first video you sent was a combination of a welder and a miller. This design of using a welder is probably not the best option for printers that are going to be small or anything.
But extruding metal as a FDM printer and then milling it may be a viable option to get rather good prints actually. Only problem may be that the print may now stick to the bed and thus will get thrown away by the mill.
Maybe if it starts printing of a bed of the same material? Hmm that would maybe work but then you have to waste huge plates of metal each time you are printing. Idk.
Thanks for the reply though, definitely interesting!
Not quite. They use lasers to heat the parent metal, the the powder is deposited into the weld pool.
The problem you would have in trying to extrude metal like a traditional plastic 3d printer is that the previous layer needs to be hot enough for the next layer to stick to it.
It kind of the reverse of plastic printing, the temperature of the filament doesn’t matter, it can be room temperature. It’s the thing you want it to stick to that has to be hot.
TIG welding is a good example. You have to heat the metal to a certain point where you are able to deposit a filler rod into the liquid weld pool. If you don’t reach that temperature on the parent metal, the weld will not stick.
So the only way it would be possible using FDM, would be using the sintering method like K3g suggested, or by have a laser heating the previous layer just as the next layer is being deposited onto it. Kind of like an arcless TIG process I guess. Not something that would be cheap anyway & it would need a fairly advanced control program to do a neat weld.
They do make robotic TIG welders, but the TIG process uses a consumable electrode to create the arc, so you would have to change or adjust it frequently during a print for the best result, which would be kind of awkward.
I see this as looking at the problem backwards.
An engineer won’t want a cheap metal printer unless there is some benefit to the process,
ie: detailed precision or ease of manufacture.
Sometimes the old ways are best. I think the cheapest way to make metal prints would be to do a FDM print using wax and then making a traditional mold from that.
I’ve done Lost-PLA casting, where I burned out the PLA in a kiln and then cast parts in the mold. Works quite nicely. Since PLA is a low temp thermoplastic, wax isn’t really needed.
There’s also a few castable photopolymers out there for SLA casting purposes.
Additionally, but unrelated, I’ve done low-volume (<100) out of SLA and SLS printed molds.