2

Can someone shed some light in why FDM (Fuse Deposition Modelling) is not possible with metals?

Has anyone attempted any experiments with it?

Trish
  • 20,169
  • 10
  • 43
  • 92
justanu
  • 21
  • 1
  • welding wouldnt have a consistent layer deposition as other posters have tried to joke about. It would be good to hear from more experienced people about the challenges of metal FDM printing, and especially without an arc – Andi Iacob Oct 20 '22 at 16:14
  • I removed some of the superflous tags that have nothing to do with the actual core question here. – Trish Oct 21 '22 at 15:40
  • Although I did post an answer, did you attempt to research? https://i.stack.imgur.com/xhrNW.png – Stanley Oct 28 '22 at 21:58
  • This looks like a posibility: https://phys.org/news/2022-10-scientists-material-plastic-metal.html – Perry Webb Nov 03 '22 at 19:08

5 Answers5

4

Yes, it is possible and it's done all the time. It's more commonly known as welding.

We did TIG welding on our Hyrel printers back in 2017.

Note: I work for Hyrel.

Davo
  • 2,362
  • 1
  • 12
  • 26
  • There are also (possibly experimental one-off) CNC type wire feed welders, commonly known as MIG welding, that are akin to a 3D printer. – fred_dot_u Oct 20 '22 at 15:59
  • @fred_dot_u MIG welding is industry classic, though automation is somewhat in its baby shoes. – Trish Oct 21 '22 at 15:39
  • Is that really a TIG (with a modification that feeds the filler along the axis next to the tungsten tip) or a MIG that uses the wire as an electrode? – Trish Oct 21 '22 at 15:53
  • Our chief engineer told me it's a TIG. I wouldn't know one from the other, personally. – Davo Oct 21 '22 at 15:56
  • I am not an expert myself, and only know that MIG feeds its electrode as filler, while TIG has a permanent electrode - and might work without an electrode melting the stock on both sides of the stock. The two are very similar in construction and even look similar, so I guess your engineer is right. – Trish Oct 21 '22 at 16:34
  • Robots have been welding for decades, but they don't build objects from scratch with welds as far as I know – Kilisi Oct 23 '22 at 10:02
  • @Kilisi https://wonderfulengineering.com/worlds-largest-3d-printed-steel-bridge-made-welding-robots-6-months/ – Davo Oct 24 '22 at 11:55
  • In Polish language, we have this word "napawanie". Literal translation would be "welding on", in contrast to regular welding. And yes, it is sometimes used to build 3d metal shapes from scratch by hand by our artist welders. Don't know if it was ever done in computer controlled manner. – Mołot Oct 27 '22 at 11:19
1

First of all, it is possible in multiple ways:

Computer-controlled-Welding

The most akin to FDM is using an automated MIG-welder, building up weld beads upon weld beads using an arc to melt the metal and keeping the puddle in position by carefully low speeds and coordinated movements. With a little ingenuity, even a TIG could be used, if you manage to feed the filler rod.

Converting a MIG is possible even on a hobbyist scale, as this hackaday 2021 article shows. However, research into something like that is in research since at least 2018, and Hyrel experimented with a TIG setup in 2017.

The items created this way have a rather rough surface, yet they do have upsides: The created metal is easily machineable and can reduce a lot of waste. It might not beat machining from bar stock at the moment, but it can beat steel casting in price, though not in quality. However, automated welding also has the chance to have defects. In some applications, it is used industrially, for example, to manufacture specialized stock for machining with minimal waste or from a special material.

Oh, and it is even rather simple to repurpose 3D welding robots to 3D Weld a gap closed bead to bead, akin to how you'd 3D print in FDM.

Metal Filaments: BMD

There are filaments that contain about 80% of metal powder. These highly abrasive filaments contain a binder that is thermoplastic. You can print on a somewhat modified machine with those. That's called BMD - Bound Metal Deposition. This method is industrial by 2022, and often post-processes the part to remove the binder.

Usually, this post-processing is done in a kiln or sintering oven, melting or cross bonding the metal between layers more than with the plastic binder, and simultaneously at times burning out the binder. This technique exist since at least 2018, as this answer shows.

Trish
  • 20,169
  • 10
  • 43
  • 92
  • I concede. You have a better answer although I will keep mine here. I think this method could be greatly improved by shaping the electric field for better control of the deposition. Heliarc, which actually uses argon, could be used when the metal requires. – Perry Webb Oct 21 '22 at 16:10
  • @PerryWebb nothing wrong with your answer, and it is a good primer about the idea - and it discusses points I don't mention (by design, because I don't want to discuss MIG/TIG too deep) – Trish Oct 21 '22 at 16:11
0

For many metals you would need to run the hotend around 1000 °C. Aluminum melts at a lower temperature but needs to be in an inert atmosphere, such as argon. Solder melts at the right temperature, but tends to stick to most metal nozzles. It would start dissolving a brass nozzle thus enlarging the nozzle opening. Lockheed has a titanium alloy printer that melts powder with a laser. I would assume they need an inert atmosphere since titanium reacts with nitrogen as well as oxygen.

Perry Webb
  • 2,158
  • 1
  • 8
  • 18
  • I am actually interested if it is possible to melt titanium and slowly push it through 0.1 mm hole to get a constant stream of molten metal. Assuming I can get enough temperature to build a small titanium pool. – justanu Oct 20 '22 at 16:30
  • Pure titanium has a melting point of 1668 °C https://www.webelements.com/titanium/ – Perry Webb Oct 20 '22 at 17:31
  • nah, you don't run the hotend at all at that temperature, you use an electric arc of that temperature and have no classic nozzle at all. you go the MIG route. – Trish Oct 21 '22 at 15:29
  • @Trish thus you are still running the hotend at that temperature. You just changed the geometry and ditched the nozzle. You are mentioning methods to greatly increase the temperature gradient. . – Perry Webb Oct 21 '22 at 15:59
  • true true. the argon stream in MIG or TIG keeps the actual nose (cup?) cool enough. – Trish Oct 21 '22 at 16:09
0

It is completely possible, there are 3d FDM printers that could do that with the Ultrafuse 316L. You could print filament that has metal in it, you just need a special nozzle for that. The pure metal-printing printers use an industrial heater/extruder that can reach up to 230 °C.

agarza
  • 1,334
  • 2
  • 10
  • 30
Stanley
  • 65
  • 7
  • Printing with metal clay or, as you suggest, metal-doped polymer is not the same as printing with metal. – Davo Oct 27 '22 at 15:48
0

I have tried some stuff. It depends what problem you're trying to solve.

You can get metal infused filaments, both aluminium and copper infused filaments print fine. But since there's plastic as well you don't get the nice properties like conductivity.

If you just want some of the properties then electrolysis might be the better choice of technology.

If you want robust parts then as far as I know the traditional methods are the best as the metals properties are constant or controlled.

Something like cnc welding layers on layers wouldn't make sense in terms of cost and predictability I would think. You'd need too much heat and it's unlikely to be as strong as traditional metal forming. It has been used for a long time to spot weld, but not build up layers as far as I know.

Kilisi
  • 1,207
  • 2
  • 11
  • CNC MIG welding and milling is cheaper than casting steel. – Trish Oct 21 '22 at 15:30
  • @Trish how do you work that out? I can cast steel in my backyard if I want with wood that grows on trees... but you'd need a CNC mill, cnc tig welder, software, and other infrastructure.... none of which grow on trees. – Kilisi Oct 22 '22 at 02:23
  • Casting properly functional steel for structural parts for industrial functions is hard, and requires oxygen as well as skin-cooking temperatures. Steel, not aluminium or even iron. Using weld bead to build up missing material on a broken part (e.g. welding over a hole or filling a gap) is a classic fabrication method. The main problem with casting steel is the mold. If you can cast proper steel in your backyard, I'd like to have your backyard kiln - because the best I manage is liquid iron. – Trish Oct 22 '22 at 08:28
  • @Trish no special kiln need, just a fan to force airflow... steel is just the addition of some carbon... I'm familiar with filling holes and building up welds... lots of post processing. It has it's uses but I wouldn't build anything from scratch with just welds. – Kilisi Oct 22 '22 at 08:34
  • nothing larger than about 5x5x5 cm I'd say at the current tech – Trish Oct 22 '22 at 08:40
  • @Trish no idea, I wouldn't attempt it, just weld plate steel together or something instead – Kilisi Oct 22 '22 at 08:44