Saturday, November 17, 2018

3D Printing Metal Parts Without A Metal Printer

There's no doubt that 3D printing as an industry continues to grow, and the technology is pushing into all sorts of new areas.  Solid Concepts printed the first full metal 1911 in 2013, using a process called Direct Laser Sintering in which a laser heats a metal-bearing powder causing the metal to melt and flow in a layer.
[see video ] [I]in overview, the machine contains a tub of a specially formulated powdered metal, and the laser heat is used to fuse particles of the powder into solid metal. The tub is lowered a small increment (.001"?) and the powder re-leveled, allowing the laser to sinter the next layer.  Layer by layer the part is built up until the final shape is there.  Post-processing - clean up, heat treating, and finishing - is required.
It's cool, but it shouldn't be the only way to get metal gun parts using a 3D printer.

Do you have a class ring from school?  How about other commercial jewelry?  The vast majority of commercial jewelry is made through a process called lost wax casting (overview here), in which a wax model of the jewelry piece is embedded in a ceramic mix which is fired to harden the ceramic (called investment) and then further heated until the wax melts and runs or burns out of the mold - where the wax is "lost" in lost wax casting.  Finally, the metal is cast into the cavity in the ceramic which is identical size and shape to the initial wax model.  The technique is also widely used in manufacturing of many things. 

There are thousands of home hobbyists who cast silver or gold jewelry at home (or in a club or Makerspace environment), and these metals melt at high temperatures.  Sterling melts at 1640 F, while pure silver melts at 1761 F.  Pure gold melts at 1945F, and different karat gold mixes melt at lower temperatures.  By contrast, aluminum melts at 1220 F but iron melts much hotter: 2802.  For any engineering use, you should check the melting point of the alloy, but the point is that the process is certainly compatible with aluminum alloys, and with the right torch (to get that higher melting point), casting steels seems to be within reach.

But what about the mold?  A few minutes of searching found three different videos showing plastics designed for 3d Printing molds that will be cast in metal.

Moldlay filament is used in this video, which features a home made printer designed to be easy to put together.

MachinableWax's Print2Cast printing filament specifically made for metal casting 3d printed models is introduced in this video.

Wrapping up these there is PolyCast™, another filament designed specifically for the metal casting industry; and the video demonstrates what the industrial process looks like.  

I'm deliberately avoiding the subject of green sand casting molds made from wood or other things in the shop; this sort of casting is part of a lot of home shops.  This is just concentrating on  3D printing.  Because of the flail about 3D printed guns.


  1. When I first heard about 3D printing, I thought about this application.

    And, of course, for extremely complex shapes (like engine blocks) the wax or Styrofoam loss part can be made of multiple sections, allowing for all sorts of little channels and such, and then the sections glued together before the ceramic/sand shell is filled. OMC (Johnson and Evinrude outboards) pioneered this process for making engines. Really neat looking at a 300hp motor and seeing the outlines of the expanding foams and the subtle line between sections.)

    1. To be honest, my first thought about getting a 3D printer was to do small castings. It's still a pretty compelling argument for ways to make parts for the itty bitty engines I like to make.

      About 15 years ago, we had a Saturn. I don't recall exactly what part it was (the alternator?), but something in the engine compartment was lost Styrofoam cast, and the metal looked like Styrofoam.

    2. Yep. A US outboard motor company pioneered that technology. Really interesting how they cranked performance up while getting rid of handfuls of extra gaskets, bolts, hoses and such.

  2. Guy in our hackerspace cast wedding rings from 3D printed models. Printer was especially high resolution, using resin intended for jewelry casting use. Did the casting with an established jeweler, then finished ring himself under jeweler's supervision. Prototyped in cheaper silver, then did again in gold.

    There exists a casting material which absorbs microwaves, put in non-food microwave and it gets hot enough to melt jewelry metals. Build mold with a funnel shape on top, molten metal drops into the mold. However, I suspect centrifugal casting has fewer flaws.

    For larger castings, there's a loose powder 3D printer which deposits glue to glue sandcasting sand together. Don't have to put cores in later, just build them in.

    1. 3D printing tech now allows things that either built-up molds or lots of welding or bolting used to accomplish.

      When NASA was looking at rebuilding the mighty Rockedyne F-1 engine (think Saturn V first stage, 5 of those bad boys) they figured that between 3D printed parts and advance CNC manufacturing, they'd be able to reduce parts to the hundreds instead of the tens-of-thousands, reduce overall weight by 10% or more and increase the overall thrust from 1,500,00 lbf of the original by 15% to 1,800,00 lbf (which, of course, they achieved in the F-1a back in the late '60s (but never flown, darned it, they just flew up-rated F-1s, increasing the thrust by 22k lbf per engine or so...)

      3D printing and modern sintering production is what allows SpaceX's Merlin engines to be so darned cheap, powerful and reliable (in comparison to everyone else's old school engines.)

    2. Being a US government project funded with tax money, all that engineering work product, modern CAD files and experimental firing test data, has been released to the public without copyright for use by the people who paid for it. Right? Where's my peace dividend?

  3. This is the way that space habitats will be built, probably by SpaceX.

  4. There's no need for any exotic filament for use in casting, PLA apparently works just fine.

    Just search "lost PLA casting".

    Also, SiG I'm surprised you didn't have a 3d printer already.

    Now is a great time to get one, with the price to performance ratio being so much better than even just a few years ago.

    1. Also, SiG I'm surprised you didn't have a 3d printer already.

      This may sound odd, but I'm somewhat surprised I don't have a 3D printer already, too. I've thought about one a bunch of times, and thought about building one. I have the advantage of being able to make parts for it out of metal so I don't need to bootstrap myself to having a printer.

      At the moment, I'm kind of short of places to put one, as strange as that sounds.

    2. Build? Ha!

      You can't buy all of the parts for one as cheap as an ender3 "kit".

      I understand about the lack of a suitable space, but if you have about a 18"x 18" x 26-28" tall bit of free space you can fit in a printer.

      I have two now a Monoprice Mini v1 and an ender3.

      If you have any questions just email me at my gmail, snarkyposters

    3. The guy whose plans I used for the CNC conversion on my big mill included plans for a 3D printer. It reminds me of the RepRap, but I don't know them well enough to tell if that's what it is.

      One thing my reading has led me to think is that an enclosure is important. I suppose I could make an enclosure for that ender3, just a box to put over it, but when I start thinking like that I quickly go off the rails. I like the look of laser cured prints - the original stereolithography - because they seem to get finer resolution and fewer bumps and blobs on them.

    4. You only need an enclosure if you are printing ABS.

      Everything else doesn't really need it unless you are in a cold climate.

      The quality of the prints from recent FDM printers are amazing compared to even a few years ago. Some prints are so good you almost can't tell that it wasn't injection molded.

      On the SLA printers, the resolution is very very good but the strength isn't. The price of the resins is still very high and there is a lot of post work needed.

      I'll post a few YouTube links, so you can see what I'm talking about.