Turning A 3D Printer Into An Injection Molding Machine

Injection molding machines are able to form very detailed plastic parts, simply by squirting plastic into a mold. 3D printers squirt plastic. Why no one thought of using a 3D printer extruder to push plastic into a mold until now is something we’ll never know.

[bfk] has been working on a way to produce very small, very detailed parts for a while now, and realized the extruder of a 3D printer serves most of the functions of an injection molding machine. It takes plastic, melts it, and forces it through an orifice. Whether that plastic goes to a build platform or into a mold is beside the point; but with a simple silicone mold, anyone can replicate extremely small parts with a tool every hackerspace already has.

The tools required are RTV rubber, which is the most popular mold material around. Aside from that, it’s just silicone lubricant, dowels and LEGO to make sprues, and of course something to make a mold from. Once the mold is made, it’s a simple matter of holding the mold up to the nozzle of a printer and extruding a bit of plastic.

The resulting ‘print’ is as detailed as the best prints that will ever come off a resin printer. It’s great for making parts for very small models like [bfk]’s current project, but this technique could be expanded to anything that needs a lot of small plastic parts with tight tolerances.

Video of the process below.

29 thoughts on “Turning A 3D Printer Into An Injection Molding Machine

  1. Sure, it’d work for small items like that, but get much bigger and you risk the plastic cooling before it hits the bottom, and you’re left with gaps and holes in the finished piece.

    1. Is it possible the silicon molds would manage to retain a lot of the heat, allowing for bigger parts?

      A lot of these materials can hold their heat pretty well. I’ve had plenty of failed prints in both ABS and PLA that ended up as a single blob – a blob that was still quite pliable even though it’d been sitting in the open air. I’d imagine that combined with the insulative properties of the mold you could get away with reasonably sized parts.

      As far as size goes – I’m not sure I’d want to use this method for large parts anyhow. Large parts made this way would use a lot of material, or require cleverly design molds to reduce material use. At that point it makes sense to just print the part instead.

  2. Silicon is “a chemical element with symbol Si and atomic number 14. It is a tetravalent metalloid, more reactive than germanium, the metalloid directly below it in the table.”

    Silicone (note the additional ‘e’ on the end) is a family of “polymers that include any inert, synthetic compound made up of repeating units of siloxane, which is a functional group of two silicon atoms and one oxygen atom frequently combined with carbon and/ or hydrogen. They are typically heat-resistant and rubber-like, and are used in sealants, adhesives, lubricants, medicine, cooking utensils, and thermal and electrical insulation. Some common forms include silicone oil, silicone grease, silicone rubber, silicone resin, and silicone caulk.”

    Thanks, Wikipedia… sheesh…

    Oh, and Hackaday — you’re better than this.

    1. Im just going to tack my bitch to yours….
      rtv ELASTOMER not rubber. Rubber comes from plants. SIlicone is NOT a rubber…its an elastomeric polysiloxane.
      Not so sure hackaday is better than this. Before they sold out they couldnt afford decent proofreaders and editors…Not sure what their excuse is now.

      1. “synthetic rubber
        noun
        1. any of several substances similar to natural rubber in properties and uses, produced by the polymerization of an unsaturated hydrocarbon, as butylene or isoprene, or by the copolymerization of such hydrocarbons with styrene, butadiene, or the like.”
        http://dictionary.reference.com/browse/synthetic%20rubber

        This is a totally bog-standard definition and not just a “lol internet dictionary” thing.You could maybe argue that this or that silicone compound isn’t even a synthetic rubber, but you *can’t* seriously claim that the word rubber applies ONLY to the natural from-a-plant stuff, even if that’s the standard usage in the circles you find yourself in.

        I dangle my preposition in ur face.

  3. You’re using the extrusion part of the 3D setup only. It’s a good start, but if you’re serious about small-scale injection molding, a ram-type injection molder can be made of common shop parts (basically a cylinder with some type of heater (often Calrod or similar) to heat resin pellets into a liquid state, then a piston to push it into a mold). This is an example: http://makezine.com/projects/make-41-tinkering-toys/diy-injection-molding/

    Silicone molds will deform under pressure, and the moldings will often cool too slowly to be useful (commercial molds are typically made of plated/polished aluminum with cooling channels to expedite cooling). Conceivably you can use any rigid material if you’re willing to wait a long time for it to cool properly. Once you have the basic assembly, an extrusion orifice will also allow you to produce your own filament for the 3D printer.

    Mold design, cooling problems and attendant concerns have been the stock in trade of plastics engineers for decades.

  4. Save yourself 2000 dollars on a 3D printer and the hassle of clogging nozzles and get a wax injection machine. For the price of a spool of PLA you could have 10KG of injectable wax.

    Silicone mold idea is good. There are better compounds than RTV but it is a good start. Check out Rio Grande supply for both.

    1. Or, you know, shop almost anywhere else. :S

      My little sister is a silversmith, and I keep telling her that most of the tools and supplies she buys from Rio Grande (Basically everything except the actual silver) I could probably find elsewhere for a lot less… Recently, she was looking at this wire/tubing bender, and the Rio Grande price was absurd. I showed her she could get literally the same tool from Micro-Mark at half the price, and even *their* prices are ridiculous on a lot of stuff.

      That’s why I always save specialty suppliers as a last resort; they like to charge a premium for everything. (Though a wax injection machine *is* probably something you’d have to get from Rio Grande or another jeweler’s supplier.) My sister never shuts up about wanting to get into casting… Heh.

  5. Plastic injection molds can withstand normal shots of 40,000+ PSI. Silicone isn’t going to hold up to this for anything even remotely large in size. Nice idea though for smaller parts.

  6. Injection molding is not “simply squirting plastic”, pressure and speed are the critical factor there.
    What the point here? Getting burnt fingers, and weak parts? I’d rather use epoxy, way stronger, less dangerous…

  7. >Why no one thought of using a 3D printer extruder to push plastic into a mold until now is something we’ll never know.

    Possibly because, one, arguably the biggest benefit of a 3D printer is _not_ having to deal with injection molding for 1-off parts, and, two, injection molds normally operate at way higher pressures than a 3D printer can manage.

    I mean, at this point, 90% of the printer isn’t even being utilized. Not much benefit over a resin casting or a bare injector. Worse, it’s only good for replication, as, if your printer is unable to manage a suitable resolution/surface finish to produce the part in the first place, it’s going to be unable to produce a master to make a mold from.

    Don’t mean to hate, but this is REALLY niche; parts to small for a 3D printer, that you already have a suitable master of, that can be made with a 2-piece mold, that isn’t such a shape that the silicone will deform when casting. Not to mention that the biggest benefit of injection molding (easy mass-production) is negated since you have to sit there with the mold for every piece you make…

    1. Think about this, you print the thing, you clean it, you sand it, make a mold and now with injection you have a fine final product. When modeling repeating stuff or detailed stuff which needs lots of cleaning this will be very good.

  8. What’s the point of reinventing the wheel? The 3d printer already print the piece without human need so.. it’s suposed to be simple, easy, fast, cheap, etc.

    And without need of molds, because if you come into consideration with all required things to produce with molds.. guess it is better to buy a pro machine instead of a 3d printer. In my mind, a 3d printer is to create a few of each diferent parts with easy, if you want to mass produce something a 3d print will never be the fast etc etc…

    industrie produces a lot, faster for a few decades and they dont use 3d printers.. so i assume it is because it would not be efficient with mass production.

    Although it would be nice, and i would like to see here an easy way of melting plastic to liquid state and use it to produce this parts. This way we could use and recycle a lot of plastic garbage in house.

    1. except a 3d printer is NOT fast, cheap or easy. well it may be easy but not the other two. the material compared to plastic pellets is 10 to 50 times more expensive, and the TIME is even Worse!!! 3d printers are slow, if you compare the two and animals, it is like comparing a lion on crack to a snail high on good pot, and the lion runs the race 50 times before the snail makes it one time.

      If you need to make 500 of the same part, a I M, can do it in a day, a 3d Printer, would take WEEKS to do the same amount. And this guys idea of holding the mold to the print head, well that would take a week with no sleep, because you have to manually hold it, wait for it to fill (while burning your hands) take part out of mold, hold it up to print head again, and rinse and repeat a I M you dont have to do any of that.

  9. Lots of ups and downs with this process, been working with this same idea for a while now.
    The Pictures are quite bad, but Im willing to bet a few things .
    -The surface finish is probably much better, like a lot better than any 3d printed part.
    -Strength of the part will be much better, and not have de laminating issues like with 3d printing.

    One of the ideas I had for this method, would be to inject a shot into a a void infill cavity to give it way more rigidity.

    just some commentary
    plastic retains heat 30x more than metal
    lol @ 15 second injection time thats so long

  10. I had the enjoyment of running a large aluminum die casting machine (injection molding) for a year.

    I don’t know much about plastic. But with aluminum, the shots are forced in with tons of force in mS shot times. The clamping force of the machine on the die cores was in the thousands of tons to keep all that molten metal in the die cavity and not have it “flashing out” all over a poor operator. Shot pressure and speed were critical to make sure the part was fully formed and had uniform density throughout. A weak shot would mean “poorfill” where the part would have chunks missing or be malformed and “porosity” where a cross section of metal will have sponge like bubbles in it.

    If your dies were too cold = stuck metal

    If your dies were too hot = stuck metal

    Not enough die lube = stuck metal

    Too much die lube = bad parts

    Polished steel dies, cooling channels, die lube robots, nearly getting set on fire a few times…

    Ahhhh memories.

  11. I really like this idea. If you wanted to go for something a bit bigger what do you think about having an aluminium mould and using the heat bed (or just put it in the oven for a bit) to warm it up so that the PLA will take longer to solidify?

  12. I worked at a plastic injection molding company in Minnesota. Heated screw pushed the plastic into a huge block of steel heated and cooled at the same time. When starting they tell you not to work about getting stuck in the machine. Have seen pressure hoses break ribs and throw people. Tons of pressure.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.