Jeshua Lacock Presents Lost Shell Sand Casting for PLA 3D Printed Models

Jeshua Lacock is back with a new metal casting technique for 3D models printed in PLA. After receiving feedback that his Lost PLA Casting Technique just 'looked like too much work,' he decided to try another route, and came up with something entirely new called the Lost Shell Sand Casting technique. Also working with PLA, this technique is easier, faster, and more affordable, and the results offer up finely detailed metal parts. The technique also works with ferrous metals. Read more at 3DPrint.com: https://3dprint.com/140523/lost-shell-sand-casting/
https://3dprint.com/wp-content/uploa...ntitled-81.png

way cool. dude !

But why is it a different method ?
Looks pretty standard to me.

Quote:

---

Originally Posted by curious aardvark

way cool. dude !

But why is it a different method ?
Looks pretty standard to me.

---

Thank you!

What is different is that with traditional sand casting the pattern is removed from the sand before the metal is poured. Since it has to be removable without damaging the fragile sand mold, it greatly limits the type of geometry possible (undercuts, cores, etc. become a design limitation).

With this technique, the pattern is left in the sand (allowing practically any geometry to be cast) and burnt out directly by the molten metal.

Hopefully I have answered your question, let me know if you have any other questions.

okay. Yes now you mention it.
In the lost wax/pla process you heat the mould to melt the wax out before adding the metal.

So why are the tubes (can't remember what you called them) for the metal in and gasses out so large ?

And would this process also work with lead ? or does that not get hot enough to vapourise the pla ?

These guys don't seem to have very high standards for their castings. If one of mine came out that poorly, I wouldn't be celebrating; I'd melt it down and try again. Burning the PLA inside the sand mold evidently disrupted the surface significantly, so any crisp detail was lost. I just hope nobody watches this video and hurts themselves badly trying to do it themselves.

Andrew Werby
Juxtamorph.com

Quote:

---

Originally Posted by curious aardvark

okay. Yes now you mention it.
In the lost wax/pla process you heat the mould to melt the wax out before adding the metal.

---

That is correct as well.

Quote:

---

Originally Posted by curious aardvark

So why are the tubes (can't remember what you called them) for the metal in and gasses out so large ?

---

The the pour tube is called a "Sprue". It is both tapered and oversized to help build up as much hydrostatic pressure as possible (also makes it easier to pour into). It also acts as a "Riser" that stays molten long enough to feed the metal as it shrinks as it cools which is called the "Suck".

Quote:

---

Originally Posted by curious aardvark

And would this process also work with lead ? or does that not get hot enough to vapourise the pla ?

---

Let's see; it looks like lead doesn't boil until 3,180F. So you could get it plenty hot enough to work for the process. How much hotter it needs to be over the melting point of 621F would take some experimenting. I imagine if you got it up to around 1000-1200F that seems like is should be adequate.

Quote:

---

Originally Posted by awerby

These guys don't seem to have very high standards for their castings. If one of mine came out that poorly, I wouldn't be celebrating; I'd melt it down and try again. Burning the PLA inside the sand mold evidently disrupted the surface significantly, so any crisp detail was lost. I just hope nobody watches this video and hurts themselves badly trying to do it themselves.

---

For staters they are sand castings - which will never quite have the same detail as investment castings. You are incorrect about any crisp detail being lost - many of the original print lines are evident and you can't expect to do any better than that from a 3D print.

Next, these casts are straight out of the sand - no touch up work at all was done to them. It only takes very minor work to sand things into shape.

Finally, where they came out a little rough it was only because the sand wasn't adequately packed tight enough - but a little learning is to be expected with a whole new technique. These are literally the first casts as we learned the new process. We didn't think of it at the time, but all we needed to do is flip the sand mold over and repack the sand from the other end.

And we take safety very seriously! I pour in a full fire proof proximity suit (with full head protection), we pour in a dry sand box, the molten metal is never lifted above waist level, the building is adequately ventilated, we use proper tools, etc. and so on.

Have to admit for an uncleaned model it looked pretty good to me.

Quote:

---

Originally Posted by 3DTOPO

For staters they are sand castings - which will never quite have the same detail as investment castings. You are incorrect about any crisp detail being lost - many of the original print lines are evident and you can't expect to do any better than that from a 3D print.

[If you were setting out to demonstrate the utility of this process for creating small detailed castings like the one you attempted, you have not succeeded. This process might be okay for rough castings, but since there's no way to introduce a core, it's not going to work too well for most of them either. Solid lumps cast poorly, and some of the problems in that one might be due to shrinkage porosity. If you try it again, use a chill and see if it works any better.]

Next, these casts are straight out of the sand - no touch up work at all was done to them. It only takes very minor work to sand things into shape.

[As someone who's done my share of that "minor work", I'd say melt it down and start over.]

Finally, where they came out a little rough it was only because the sand wasn't adequately packed tight enough - but a little learning is to be expected with a whole new technique. These are literally the first casts as we learned the new process. We didn't think of it at the time, but all we needed to do is flip the sand mold over and repack the sand from the other end.

[I think the process, such as it is, would be improved a lot if you used a refractory coating on the part before packing it in the sand. People who do the similar lost-foam process with aluminum use dry-wall compound, but there are lots of better ones that might stand up to hotter metals.]

And we take safety very seriously! I pour in a full fire proof proximity suit (with full head protection), we pour in a dry sand box, the molten metal is never lifted above waist level, the building is adequately ventilated, we use proper tools, etc. and so on.

---

[Yes, I noticed that. (Nice induction furnace, by the way! Is that home-made?). But I worry about the people who watch You-tube videos and go out to try this at home, in t-shirts and tennis shoes, with the kids all crowding around. You could shake off that "bottle rocket" explosion you got on your first attempt, but it might be a life-changing event for them (and not in a good way).]

Quote:

---

Originally Posted by awerby

[Yes, I noticed that. (Nice induction furnace, by the way! Is that home-made?). But I worry about the people who watch You-tube videos and go out to try this at home, in t-shirts and tennis shoes, with the kids all crowding around. You could shake off that "bottle rocket" explosion you got on your first attempt, but it might be a life-changing event for them (and not in a good way).]

---

It is an informational video of what we do here. We go out of our way to discourage others from trying it at home.

How many of these people that you have in mind, just happen to have an induction furnace laying around that is able to melt a pot full of copper at a flip of a switch?

I could only hope that if someone owns *any* type of professional industrial equipment, that they know how to use it properly and safely before attempting to even turn it on.

Note that I spend 5-minutes going over basic metal safety in this video:
Lost PLA 3D Print to Metal Casting; Complete