FDM metal printing

[Ryan Carlyle]

So, the biggest problem with running low-melting metal filament (like solder) through a typical FDM 3d printer seems to be that liquid metal doesn't have the right viscosity profile to achieve proper cap zone behavior. The fact that polymers have an exceptionally wide range of viscosities between the glass point and full printing temp is a major contributor to their ability to be extruded by simple filament-pushing printers. The viscous-shear pressure cap at the hot/cold transition zone is pretty stable and effective at keeping molten plastic from squirting up the thermal barrier tube. Whereas metal has a sharp melting point and doesn't develop very good viscosity-pump action in the hot end, so you can't extrude fast or reliably. 

What I'm wondering is whether you could get some "polymer-like" filament performance by selecting an alloy with a wide range between its solidus and liquidus points. 

http://en.wikipedia.org/wiki/Solidus_(chemistry)

http://en.wikipedia.org/wiki/Liquidus

This isn't strictly comparable to the wide melting range of polymers, because metals form a slurry rather than a viscoelastic fluid, but it might exhibit a broad enough viscosity transition to work in an extruder. It looks like there has been some partial success with this approach in the "distant" past: http://blog.reprap.org/2011/06/new-approach-to-printing-metals.html

Manufacturing custom alloys is for the birds, though. I'm wondering if there's an available solder wire that exhibits an acceptable viscosity profile. I think it would be much more reasonable to build a custom hot end for an existing solder size rather than try to manufacture custom wire filament with a consistent diameter. Alternatively, a large solder wire could be drawn down to a standard nozzle size. 

Another issue is metal solutions. Many low-melting alloys will attack and dissolve brass. So traditional hot end materials seem to be out. I'm thinking a PEEK thermal barrier and epoxy-coated nozzle (from Performance3D) should be able to handle liquid metals for extended periods. 

Then there's adhesion. Previous experiments indicate blue tape works fine for a build surface. But layer/layer bonding might be an issue. Normally with solders you need flux to clean off oxide films that prevent proper adhesion. And I don't think liquid metals exhibit the same kind of diffusion-welding action that polymers do... it's possible you will need to extrude material at a high enough temperature to actually melt the surface of the previous layer, and form a true fusion weld. I'm not real sure on this point. An inert atmosphere or flux mist spray might be necessary to get really good bonding, and that explodes the engineering complexity and cost of the system. In a perfect world you'd pick an alloy that doesn't experience much oxidation -- which kind of rules out all the usual tin-based metals.

Finally, cooling -- liquid metal has a very high heat capacity, so you'll probably need a ton of cooling. But on the plus side, the previously-printed layers should have extremely high conduction of heat away into the print. So I would actually expect higher layers to be a lot easier to cool than lower layers. This could also mean vertical walls print well but overhangs/bridges (with minimal contact to previous layers) may not work. 

It's an interesting subject. I'd like to try it out at some point. 

[Ryan Carlyle]

I finally bothered to look up the yield strength of solder, and now I don't even think it's worth bothering with. A regular lead/tin would have about the same strength as ABS. Except more brittle and contains lead. So... not really seeing much appeal at this point. I suspect most low-melting metals will be similar.

Seems like the only real argument for printing solder is making circuit traces.