Semi-Lost-PLA Casting Method

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Here is a shell process.

This explains how they get the mold to fill when making the shell; they introduce granular sand with binder, vibrate the mold, and then cure it to a hard shell with heat.

Mold adhesive applied at 3:38 to prevent runouts.

Apparently the casting is a thin-walled vase.


 
I am wondering about how the lines could be eliminated from a lost-PLA mold.

What if ceramic mold coat were poured into the lost-PLA shell after it was made, and then baked dry (carefully, since alcohol-based mold coat creates flamable vapors).

In essence, you are coating the mold just as you would spray on ceramic mold coat with a two-piece mold.

You can see the ceramic mold coat being sprayed on at 4:37 in the video below.
Ceramic mold coat does a good job of evening out the interior surface of the mold, but the main reason it is used is to prevent burn-in / burn-on of sand into the surface of the iron (not sure what the correct terminology is).

With ceramic mold coat, you can use a dry paintbrush to dust off any residual sand sticking to the casting, and you get a very smooth casting surface finish that totally eliminates post-casting cleanup.

.



I posted this earlier, but this is the surface finish you get with gray iron and ceramic mold coat, straight out of the mold, with no cleanup.
Clean, bright, shiny, and smooth.
I literally had just flipped this piece over out of the sand mold when this photo was taken.
The slight amount of sand still on this casting can be brushed off with a dry paintbrush.


110217-rImg-7743.jpg
 
Last edited:
Last edited:
Some information about low weight PLA, and a selection of filaments made by "Polymaker".

https://us.polymaker.com/products/polylite-lw-pla

Description​


PolyLite™ LW-PLA is a special foamed PLA designed to print with similar settings as regular PLA but with half the weight.
What is the difference between active foaming and passive foaming?
PolyLite™ LW-PLA is not an active foaming filament, which means it will not foam when extruded from the nozzle depending on the temperature. PolyLite™ LW-PLA is already pre-foamed.
What are the pros and cons between active and passive foaming?
Active foaming:
You need to heavily modify your printing settings depending on the temperature and setup to compensate the foaming expansion of the material when printing.
You need to print at very high temperature to achieve light weight results (~250˚), these high temperatures will create a lot of defects on the print such as stringing and blobs.
At very high temperature, active foaming can achieve lighter print than passive foaming however the print may suffer of serious stringing defects.
Passive foaming:
You do not need to change any settings from your regular PLA settings (slight increase retraction) as PolyLite™ LW-PLA will offer light weight results even when printing at very low temperature (~190˚C).
The lower the temperature the higher quality the print.
Printing and Drying Settings?
Printing Temperature: 190˚C - 210˚C
Bed Temperature: 30˚C - 50˚C
Printing Speed: 25mm/s - 60mm/s
Fan: ON



polymaker.jpg
 
Yes the "foam" type PLA has less material to burn out, someone posted about it here a month or so ago.

Pat I don't see why you say trees of castings cant be done with sand moulding. It is really no different to using a match board with multiple parts on it with all the runners and gates. This is a very commonly used method as it saves time having to mould each individual part in a separate flask, reduces total amount of metal to be melted as you don't have as many risers etc.

I've even made patterns that consist of several items "treed" together so they are quicker for the foundry to cast and less risk of loosing individual small patterns. This is probably a better option for the home foundry as they are not going to be into production quantities that match boards suit.
If I think of a tree i immagine something like this, were parts are combined to larger groups.
tree_.jpg

Different parts can be combines more or less random.
The main advantages would be that the parts can be very close to final dimension.
Some weird shapes are possible that would require fiddling with cores in some sort.

Greetings Timo
 
I did post examples of 3D printed wax results the other day Pat. They were done with a printer that is a bit better than your average hobby one though.

Having seen the castings in the flesh the quality is very high.

https://www.crofittings.co.uk/miscellaneous-items/

A printer that can do small layer heights and has variable height ability would go a long way to reducing the layer lines that get carried over to the casting be that lost PLA or traditional pattern use.
 
This person casts a pretty decent looking V-4, 2-stroke engine crankcase, using the lost foam method.
I think this is an aluminum casting.

I have not seen many hobby lost-foam engine blocks that look this good, especially at this scale.

This engine block would be a real test of the structural integrity of a lost foam casting.
I know commercial engine blocks are cast this way, but they use expanded polystyrene beads, and a very specific coating material.

The ability to CNC some complex foam patterns appears to be driving much of the popularity of this method.

I have seen the patterns and core boxes for a model engine V-8, and they are quite complex, but are reusable.




.
 
I was looking through my 3D models, and wondering if any of them would be a good application for the lost-PLA process.

I can cast pretty much anything I need, in gray iron, using resin-bound sand, and cores of the same material, but there is one item that I recall now that I really need to use the lost-PLA process for, and that is the Frisco Standard (JasonB 3D-generated) helical gears.

I 3D printed these gears that Jason designed, and they mesh perfectly.

My thoughts were to print them cored out, so that after ramming the mold, the the core could be removed, and then a heat gun could be used to soften the remaining rim with the gear teeth, perhaps easing out one tooth at a time.

The intent is not to melt out the PLA, but rather just get it soft enough to allow extraction from the sand without damaging the teeth in the sand mold.

With mold coat, and bound sand, these gears should be usable without modification after casting, after the hole in the center is machined.

I will have to check and see what size gears I actually need for the Frisco Standard I am planning on casting.
I just picked a random size to 3D print these, as I recall.

In my opinion, this will be the acid test of how good the lost-PLA process may be, or how bad it may be.

This is the one item where I have to try some intermediate step towards investment casting, so here we go.

I have never increased the resin density beyond what the instructions recommend, but should this attempt fail, I may try increasing the amount of resin, to create a harder mold.



rImg_9015.jpg
rImg_9020.jpg
 
For the Frisco Standard, my piston would be 2.1" diameter, flywheel would be 11.5" diameter, and the approximate size of the scaled gears is probably in the 1.5" diameter range, so perhaps I did 3D print the above gears to scale.

If I can cast the gears using the lost-PLA process, then the next potential item to try the lost-PLA process on would be the cylinder head, which is a rather convoluted piece.
I could cast the head using bound sand, and that is how the original heads were cast, but if I can find an easier method, I will use that.

I want to cast the holes in the valve guides slightly undersized, and possibly cast the cylinder head bolt holes.
May as well cast as much as possible, and a long cast hole would give an easy path for a reamer to follow.

.

Image250.jpg
Image251.jpg
Image253.jpg
Image255.jpg
Image260.jpg
 
Something to aim at. The diff gears of my 2" traction engine are sand cast and have a very good finish. The larger two gears are 4.2" dia, the pinions 1.6" dia

A7-Diff.jpg
 
I really would advise against putting a reamer down a cast hole and expecting it to be true to a valve seat. Even putting a reamer down a previously drilled hole is not good practice, better to drill, bore and then finally ream as a reamer will simply follow a wonky drilled (or cast) hole.
 
This person casts a pretty decent looking V-4, 2-stroke engine crankcase, using the lost foam method.
I think this is an aluminum casting.

I have not seen many hobby lost-foam engine blocks that look this good, especially at this scale.

This engine block would be a real test of the structural integrity of a lost foam casting.
I know commercial engine blocks are cast this way, but they use expanded polystyrene beads, and a very specific coating material.

The ability to CNC some complex foam patterns appears to be driving much of the popularity of this method.

I have seen the patterns and core boxes for a model engine V-8, and they are quite complex, but are reusable.




.

I have seen some very large engines made with foam. They a chemical bond sand but is was a engine for a ship.

For small engines I would use sleeve engine.
Make a outside pattern then make plaster cast.
Use wax sheets inside the plaster cast for core.

Dave
 
I really would advise against putting a reamer down a cast hole and expecting it to be true to a valve seat. Even putting a reamer down a previously drilled hole is not good practice, better to drill, bore and then finally ream as a reamer will simply follow a wonky drilled (or cast) hole.

The holes would be made with graphite rods, and so potentially somewhat straight and even, but I hear what you are saying.

Perhaps one of those odd drill bits that they use to drill out the center of 3/4" graphite rods when making a pyrometer.
They are suppose to drill very straight without drifting.
.
 
3/4" that's a bit big for a valve stem though;)

Known as core drills, usually 3-flutes

https://www.drill-service.co.uk/pro.../dcore-hss-core-drill-3-flute-straight-shank/

But if you ar egoing to put a drill down it before reaming might as well just go with a standard drill bit into solid metal, at least you will be able to get one easy enough in reaming sizes. Does the graphite rod come 8thou or so under the size of hole you want to ream or will you have to machine it to diameter? Even then you would need to setup the cast hole to run perfectly true so the valve seating could be machined true to that.
 
Last edited by a moderator:
The 3/4" graphite rod, or perhaps 1/2" is what is used as a sheath for an aluminum or brass/bronze pyrometer thermocouple.
The graphite protects the thermocouple for an immersion type pyrometer.

The drill bits I saw (if I remember correctly) where just drill rod with a round end that had half of the end cut off, ie: 1/2 of the drill rod protruded forward at the tip.
The foundry guys were able to drill straight through a 12" piece of graphite rod, and so I am guessing it would work in a cored hole too.
Apparently the drill rod shaft guides the bit straight.

The valve guilde holes will not be very large, but I can't recall the exact size.
The graphite rod diameter would be slightly smaller than the finished valve guide hole.

Perhaps even a round broach to get things started.

Once the guide has been drilled and reamed, then a cutter mounted on a shaft would be inserted into the valve guide hole, and the valve seat trued up.

Graphite rods come in many sizes, and can be used for core material to make pretty accurate holes, especially with bound sand where they would be rigidly supports in an exact spot.

A short core print for the graphite rod could be 3D printed, but it would probably be better to insert a piece of wood dowel rod into the pattern where the graphite rods will go (as a core print).


graphite-rod.jpg
 
Last edited:
Yep as I thought they come in nominal sizes not about 8thou or so under that you would want for reaming a 3/16" or 1/4" valve guide hole.

Seems like a lot of extra work and cost for what gain? Sourcing and possibly sizing graphite rod. Special drill and/or broach, valve seat cutting tool.

When it could all be done on the lathe with the usual set of drill bits, reamer needed in either case and the valve seat simply cut with a boring bar with the top slide set over.
 
I think they call the "D" bits.
Simple to make, just use a cutoff disc in a dremel.
The chips would go down the hole in front of the cutting edge.
This would get it a little under (maybe), and then a final ream.

I saw someone make a valve seat cutter here not too long ago, but I forget where.
Very simple affair; with perhaps one carbide cutter set at a 45 degree angle.
I seem to recall hand-rotating one of these when rebuilding lawnmower engines.

The seat would be cast into the casting at 45 degrees, so it would just need a cleanup.
With the cutter fastened to a rod the size of the valve guild hole, you would cut the seat exactly eccentric to the valve rod.

I have wide assortment of graphite rods already.
I use them for various foundry uses, since they withstand molten iron temperatures very well.

I think the D-drill bit would work, and thus the broach would not be needed.
I think a round broach could be made from drill rod, if that were needed.

Seems like you would have to bolt an L-bracket onto the lathe faceplate, and bolt the head to that.

For the headbolt holes, the diameter would be some nominal amount larger than the bolt diameter, and those holes would not be critical.

The bound sand method is quite accurate.
You can pick up a stray hair in the sand, and mirror that into the iron casting.

.
 
I've made a a reasonable number of D bits. Mill them to half diaand angle the end then harden and a final lick with a diamond stone, why would you need a dremel on material that is easily cut/machined? You would need to take the drill rod down to reaming size.

If I can cast the gears using the lost-PLA process, then the next potential item to try the lost-PLA process on would be the cylinder head,

I thought you were talking of lost PLA for this one, now you are talking of placing the rods into bound sand? Could probably place them more accurately in the lost PLA method than in bound sand. Push rods into holes in the 3D pattern with some sticking out each end and dip/invest. Then when you burn out the rods should be held by the surrounding coating but would be worth a trial before wasting a head pattern. You would still need to set the "cast" hole to run true to the lathe axis as a D bit down a cast hole will want to wander if the hole is not set true simply because it can flex and will try to follow the path of least resistance eg the out of true hole.

I've never found the need for a valve seating tool and don't seem to get the problems of valve leaking that I see here a lot often by people using some form of set cutter/finisher. Again worth a simple test before wasting a nicely smoothed head pattern

Still seems a lot of extra work and I thought your aim was to cut down time?
 
Back
Top