Metal Casting With Single Shelled PLA Masters

[3DTOPO] does a lot of metal casting (video link, embedded below). That’s obvious by the full and appropriate set of safety gear, a rarity on YouTube.

They had all the equipment to do it the normal way: craft or CNC out a master, produce a drag and a copy, make any necessary cores, and finally; pour the mold. This is a long and tedious process. It has a high rate of error, and there is a parting line.

Another set of methods are the lost ones. With these methods the master is produced out of a material like foam or wax. The master is surrounded by refractory and then melted, burned, or baked out of the mold. Finally the metal is poured in. Theoretically, a perfect reproduction is made without ever having to open the mold.

Their favorite method is a variation called the lost PLA method. In this method the master is 3D printed in PLA. PLA is happy to be burned or baked out of a mold.  However, doing this does offer a chance for the mold to break, and the extra step takes time and fuel.

Recently they’ve discovered that simply printing the model with a single shell lets them pour the metal straight into the mold without the burning step. For most parts, there isn’t enough plastic to cause a serious off-gassing event, though they do recommend wearing the full and appropriate set of safety gear just in case. Video after the break.

21 thoughts on “Metal Casting With Single Shelled PLA Masters

  1. A truly practical way to make things with a 3D printer. Very cool! From the title I was expecting shell casting where the investment is dipped in a ceramic slurry and then fired. The example is a typical sand cast finish. It make me wonder if dipping the PLA master in plaster of paris and then drying it in an oven before ramming up would give a better finish. Given the typical surface of many 3D prints it might not help though. In any case, very well done!

    1. P.S. that casting came out a little rougher than what is possible with the technique. We used green sand that is difficult to keep the ideal water content in, so the sand used for that piece was a bit more wet than what is ideal. We just got some petrobond sand that eliminates this problem. It uses oil instead of water for its green-strength, and doesn’t dry out and need constant conditioning like greensand does. ;)

  2. That must have stunk. I can almost smell it just watching the video. I’ve done some lost foam casts and my sand never smelled right again. But it is a cool technique.

    1. PLA is made from starches extracted from corn, sugar beets, sugarcane, and other sources. Can’t say I’ve ever actually ‘burned’ any PLA, but when it’s being liquified at upwards of 190~220°C and squirted out of a 3D printer’s extruder nozzle, it smells like sugary baked goods or pancake syrup.

      …this is in stark contrast to ABS which, at its worst (some brands don’t really smell, and some are godawful), smells like something along the lines of a tire fire, or an oil stain on the concrete at a gas station. ;P

      1. Agreed completely. At these temps, the smell is even less noticeable I think. But we also use a vent hood and a big fan for plenty of fresh air too.

        I liken the smell of burning ABS and EPS foam to that of death! No sir, I don’t care for it one bit! I just started experimenting with SLA prints – and it is pretty foul too.

    2. As others have mentioned, PLA smells like pancake syrup when it’s heated up to its liquid state. It only gets a little stronger when it gets burned away, and the aroma essentially stays the same. Not unpleasant at all. I actually love 3D printing PLA, makes my apartment smell lovely. ABS on the other hand, smells like oil burning. It has a very sharp odor. I used to print with ABS, the smell of certain rolls would be nauseating. It has its perks, but overall I have had a bad experience with it and prefer the ease of PLA.

      1. Holy crap! This looks fantastic. I actually like the sand roughened surface texture, because the piece still kept all it’s original detail – just adds a certain rustic quality.
        I have carried out some very rudimentary lost PLA casting, with Lead-free Pewter, because it’s so much easier to heat up than copper, iron, or aluminium.
        I made a Pewter Benchy, came out alright but it didn’t manage to form the box behind the cabin properly, left gaps. I was amazed at the level of detail though!
        I used some special plaster called Crystacal Alpha-K, a high detail plaster, cos that’s wot I had lying around. Put it in the oven to melt out the PLA.
        After pouring I then let it cool and dropped the whole thing in water for an hour, then chiselled off the plaster. Came off with quite some resistance, but retained the detail really well, even the print striations!
        I’d love to try this sand technique out, and I’d love to use some ‘better’ metals, but I need a proper foundry and stuffs to do it with.

        Great work on the Big Yell!

        1. Thanks! Sounds cool! I agree with you that the roughened texture looks kind of cool, but its worth noting that finer surface quality has been achieved with this technique. The sand had too much water content than what is ideal for this cast. So if you want that rustic look – add a little bit extra water to the green sand. ;)

  3. Excellent work gentlemen. I once did quite a bid of prototype lost-foam aluminum casting by hand-sculpting the foam. In that process, there is a single sprue and the gas from plastic flows directly into the sand. So, here is the suggestion: instead of using fine sand with clay binder and packing it down (which forces the gas out your secondary chimney) use a coarser sand without binder. Coat the hollow PLA positive with a fine, porous refractory surface coat as is used with industrial lost-foam. This will pick up the detail. Place the assembly in a metal can and vibrate the sand around the PLA. The vibration will settle the sand and make it very firm. The hydrostatic pressure of the metal then keeps the sand from collapsing into the melt during pour.

    1. Thanks Scott. I have thought of trying that, but the problem I foresee with it, is the shell would need to be baked out first, which could cause the plastic to deform, which would cause the thin shell to crumble (don’t ask me how I know this ). I also know that using a water-bonded mold, if it isn’t baked out first, the water content will cause serious surface quality defects. Might be doable with a refractory shell that uses silica instead of a H20 bond I suppose – but for all that its probably cheaper to just use the Lost PLA process – its hard to beat the cost of plaster of paris and sand.

Leave a Reply to James PurcellCancel 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.