VW Vortex - Volkswagen Forum banner

The complete guide on casting parts

364053 Views 691 Replies 112 Participants Last post by  Brian.G.
Vw parts of course!
Said Id throw this up for the craic, some of you may be interested in the process so I might as well share given the fact that I took pics of the progress and build as I went.
First off, this line of work is next near impossible get info on so alot of it I had to design up myself and choose the suitable materials too along the way.
Casting is now a dying trade and fully automated around the world in controlled environments.
While Im sure nobody will go to the trouble I have here it may rise some interest as to whats involved, the process, materials, procedures, mould making, melting, design, etc.
I had to do this to figure out a few things I needed to know for my bigger smelter Ill be using to melt the alloy for the cylinder head Im designing.
Using this smaller smelter less heat up times, gas, and raw materials can be used in order to get the feel for molten alloy in general.
Plus, its a bit easier than winding up my bigger smelter which can melt approx 9litres of alloy at a time, where as this one can do 1.25 litres.
Im going to be using this to make a few parts I have in mind for a while, DTH throttle bodies and a few trick intake manifolds being just some of the stuff.
Casting is a pretty simple process, but the finished item all depends on the quality of the mould the metal is poured into.
The main mould types ill be using will be of the sand variety, meaning, you make your pattern part you want in timber(Iroko) and you then strike both your sand mould halves off this pattern. The part shape is then formed in the sand and you can pour in your alloy, that way, ending up with the timber pattern shape, just thats its now alloy.
Its a bit more complex than that but Ill try my best to explain all as I go.
Pattern making is what Im good at as my main ''skill'' is cabinetmaking so making very complex accurate pattern parts is not a problem.

Its important to note that this is not the cylinder head thread, this will be covered in another thread on Club Gti
Ok, so onto the first step of a long road, and that is the smelter itself.
A smelter is basically just a round oven where the crucible is placed in order to melt the charge inside it.
The smelter must be well insulated as the heat inside is in excess of 700 degrees.
The smelter is powered by plan ordinary gas you get for your cooker.
Its all pretty basic, now onto the build...
For the body of the smelter something round is needed, you could use a large diameter pipe but I used a gas cylinder, the walls are 3mm steel and there tough by nature.
I filled it with water to expel any gas still left inside.

Next up, off comes the top with my favourite tool[LOL]

A second ring is then cut off the main body to for a locating ring for lid.

The ring is then split and welded to lid around the outside.

The lid now fits snug back onto main body again.

The lid is turned over and I welded some wire lattice in there to hold in the refractory material a bit better.

The refractory mix consists of,
Perlite is a natural volcanic material, it is a natural insulator.
Cement is standard cement.
Sand is normal sharp sand.
Fireclay is the cement fire bricks are made from, kinda like normal cement but with a better heat resistance tolerance.

Mixed up

First the base is poured, approx 80mm high.

Then, I wrapped up a bit of Formica to form the inner circle.

Pouring the walls.

Done and tamped.

Onto the lid
A bit of pipe is placed into where the valve was in order to form a vent hole up through the lining. The pipe is removed when lining is set.

Thats the smelter pretty much done now, so next up, onto the crucible.
The crucible is the cup used to place the metal in that is to be melted.
This is used from start to finish, that is, the raw unmelted alloy is placed in this and stays in here until it melts, the crucible is then removed from the smelter and the alloy poured from it directly into the mould.
So, the material needed for the crucible has to be picked with care, depending on what you are melting, what the part is for, and what properties the finished part has to have all reflect on the crucible material choice.
If poor crucible materials are used some material from the crucible walls can leech out into the molten alloy and effect the alloys properties.
In this case Ill be using stainless steel >316. Its ok for small parts and doesent really have any bad quality's that will effect the final part.
However for the cylinder head, I will be using a pure graphite crucible, graphite poses no threat on the final part qualities, be it strength, structure, or machinable properties.
Obviously On such parts as throttle body's, intakes, and the like, the final properties can afford to be altered a tiny bit as there is no work on these parts as such, and machining of parts is minimal, so, stainless is a perfect choice for crucible material.

The Crucible manufacture.
I got a bit of pipe approx 250mm long with a wall thickness of 5mm, to this I welded a base plate in 6mm.
Rods used are also 316 s/s.

Next, a 'V' is cut for spout

The spout is constructed from two pieces cut to triangle shapes

Spout tacked in place

Finish weld

Next, a ring is needed at the rear in order to tilt vessel in order to pour alloy

Pins are welded to the sides to catch crucible with the pouring tongs

I gave the whole thing a quick sandblast after to remove any oxides or ****.

Sandblasting really works in lifting any dirt or general dirt, a close up of one weld shows surface finish.

It may appear an over kill looking at the scale of the plate thickness, welding, ring, spout, sizes etc, but keep in mind this isent a milk jug and has to with stand very tough conditions for may cycles.

Thats the crucible done, next up the burner tube.
See less See more
1 - 20 of 692 Posts
The burner tube is pretty much where it all happens, air is forced in one end, the other end is inserted into the smelter at the base, gas is introduced into a venturi in pipe near the hot end. The air speeds up as it passes through venturi and carries the gas with it, as it emerges inside smelter the air/gas mix ignites and makes the temperatures required to melt the alloy.
The pipe enters at the side of smelter and has the effect of swirling the flame around the walls, passing around the crucible completely on all sides, and exits out the top vent hole.
With the Formica former now removed you can see the hole where the tube will sit.

The burner tube is approx 1200mm long, 50mm in diameter, and has a 3mm wall, its is of mild steel.
First, 3mm slots are cut all around to form venturi.

These strips left are tapped in to close slots, and once again, they are recut.

Tapped in once more you can now see the crude but effective venturi formed

This is then welded up, its crude as hell and pretty wide cuts had to be filled in either end but its fine for what it has to do

Welder is then turned up to high and a hole blow through to fit a straight fitting to supply gas

Fitting tacked in

Now the end that sticks into smelter has to reduced down a bit too, few more slots a a bit of tapping and thats it done

Fits like so

There now has to be some form of way to vary the air from the blower, My blower is the exhaust air from my workshop vac but, it isent vari-speed, so I had to fit a butterfly to burner tube.
I wanted this as handy and as basic as possible so I got thinking, I made it just like a throttle body butterfly, one toyota wheel brace with a slot cut in it, and a disc 43mm in diameter. Job done. I fitted a valve spring and a welded washer as to make a bit of resistance and hold it at whatever position I wanted.

I also welded a leg to tube to hold free end at correct height, the leg is a bit of scrap from the pile and happened to be just the correct size!

Thats the burn tube done.
Up next, tempering the refractory lining.
I set the smelter aside for a week to air dry.
If it is dried too fast the lining will crack, and thats no good.
After a week or so, a small fire was set inside of timber off cuts and paper, this was repeated for a few evenings in order to dry out lining further.

The smelter is now ready for the first run.
First run.
How the smelter looks after the tempering.

Test fit/run

Made up the tongs too earlier, forgot to take pictures, its pretty simple

A bundle of lit news papers are inserted between crucible and refractory wall, blower is turned on, throttle closed, gas is turned on at low reg setting.
As temperatures rise the news papers are burnt, throttle is opened, gas is increased, burner tube sustains ignition by itself and the whole thing comes to life in a fantastic jet engine like roar!
For the first melt I used a few alternator brackets, some water pump casings, and some other brackets. This is a bit of a messy way of doing it as it creates a lot of dross, or scum on top of molten alloy that has to be skimmed off before pour. This dross is the impurities and dirt on the old parts.
Thankfully I have found a spot in Ireland that sells Ingots off the shelf in 5kg bars, they are LM25 which is the exact alloy I need for the cylinder head as it has all the correct properties I need. It is also Idea for all the other parts.
Just warming up, crucible loaded

8 minutes from ignition and all the alloy has now been melted, what you see on the top is the dross, it looks like grey banana skins and is very bad if it makes its way into mould, its pretty easy spoon it off before pour though.
My optical thermometer told me that it was approx 700 degrees in there after 8mins.

I have to say it was pretty cool, and pouring the molten alloy was a sight to be seen, sadly, I dident get a photo as my hands were full.
I poured it into a very crude normal sand mould, just to get it into a shape to be able to be remelted easily. I also wanted to check cooling times, shrink, etc.

The blank when cooled

Notice the shrink at top once it dried

This shrink has to be taken Into account in the design of all parts, moulds, risers, gates, and all other mould aspects.
Youll see about all them in the next part when I make up some moulds for parts and other bits.
I had trouble getting high grade casing sand but I have some located and ill be picking it up sometime next week along with a crate of ingots and a load of other vital bits incl de-gas tabs.
One last shot of the crucible post pour
It seems to be well up to the job and can ''take the heat''!
See less See more
Next up a run down on the basic casting methods and the various moulds needed to make particular parts.
Simple parts such as brackets and other solid items are pretty easy cast, meaning they only require one pattern in order to make that particular shape.
Slightly more complex parts which have a hole through them such as the flange on the end of a cylinder head require a ''core'' to be installed inside the main mould in order to create the void through the flange.
This core is also made of sand and requires a pattern to make that too.
All parts in or on an engine that fall into the sand casting category use these two types of casting methods.
Depending on part complexity the mould may require multiple patterns to be made up in order to strike the associated sand moulds off them in order to build up the final mould for the part required.
Take two basic parts below as an example>
One is an alternator bracket.
The other a coolant flange.
First up, the bracket, this is a basic part and requires only one pattern, and just the two mould halves, the top half and the bottom half.
The top half of the sand mould is called the cope, and the bottom half is called the drag.
The sand is held inside frames called flasks. There is a flask for each mould half.
Casting sand isn't like the sand your used of, once packed into say a block the size of a concrete block it holds its shape because of the binders added to it.
Its very fine also and takes sharp details pretty easy.
Therefore no support is needed to hold it in shape, a simple frame(flask) is used in order to aid positioning of both mould halves, even though the moulds are open top and bottom the sand will not fall out of its frame, its brilliant stuff.
Anyway, onto the bracket
As you can see its fairly basic,

Now, lets imagine you wanted to make that, you would need an exact replica pattern in order to ''make'' that shape inside the moulds.
Sadly, if you wanted to make a copy of the above part, you cannot use that part as a pattern because you have to factor in shrinkage, as the alloy cools it shrinks back a bit, so, this has to be factored in when making the timber pattern.
If you were to use the alloy bracket as the pattern once your new part had cooled it would actually be slightly smaller that the original...not good.
LM25 is the alloy Ill be using throughout and the approx rate of shrinkage for this material is 1-1.3% from pattern dimensions.
Or approx 3mm shrink on a 300mm long part, enough to throw things a bit.
So lets say I had that part made in timber with the shrink rule factored in and it was ready to use as a pattern, heres how you would go about striking the sand moulds off it in order to create the ''space'' to pour the alloy into.
First off, you can see the line all the way across the centre of the part, this is called the parting line and it is where the top and bottom mould halves come together.
Parting lines exist on alloy items that are cast this way, and also exist on ALL plastic items too as the moulding process is similar, you will see them on may items if you have a closer look.
Parting line

In an ideal world parting lines would not exist, but they do, on very cheap parts parting lines will be more prominent because of lack of accuracy, tolerances, and general sloppiness when it comes to positioning both mould halves exactly in-line and opposite each other. If they are off a bit, the mould halves become offset and this can be seen on the finished part.
Have a look at some cheap kids toys, you'll find them in a second.
Another extremely important feature you have to keep in mind when designing parts is the ability to be able to remove the timber pattern from the moulds once the sand has been packed around it.
If the part had angles that widened as they originated from the parting line more mould design would be needed leading to a more complex setup and a 'multiple sand parts mould build' so that the pattern would be able to be withdrawn from the sand without causing damage to mould.
On a two part mould design its pretty easy to get around, you just have to make sure that the angles never widen as they originate from parting line.
This insures easy pattern removal from the sand mould before pouring.
Below you can see this slight angle present on the bracket, this angle is know as the draft angle, and for sand casting its usually anywhere between 2-6degrees.
See less See more
You'll see why the parting line matters in the next bit where Ill show how a flange is made.
The water flange is a bit more complex in the fact that it has a void in the middle for the coolant to flow through.
Therefore a sand core has to be installed in the mould to prevent the alloy filling up the mould totally and the flange ending up as a solid lump of alloy with no hole through it.
Flange in question is the plastic one found on the end of the 8v head which always likes to leak, Im going to make up a few alloy ones for that reason.
Its easier to describe with a few diagrams first before I move onto the real deal so you'll get an idea of what it looks like in cross-section.
Cross-section photos aren't possible in real life so I hope this helps with the process your going see when your looking at the actual mould and casting photos.
Heres a picture of said flange, Im sure you'll recognise it!

And here it is in a simple drawing

And a cross-section drawing, the light line showing the water way inside, the black bit in the middle being the void that has to be hollow.

So, keeping that in mind its easy to see that a mould is needed to cast the outside shape, and a sand core is also needed for waterway in order to install inside the parent mould in order to ''keep out'' the alloy from that bit.
Onto how its done, a timber copy of the flange is turned on the lathe in timber, the timber blank used to make the copy is first glued from two pieces of timber with a piece of paper separating the two pieces.
That way once turned the timber flange is easy enough to split in halves along its centre line.
The flange is then turned up on the lathe but with the addition of a spigot either end, like so shown below.(The timber pattern is solid)

Next, a copy of the internal void, or waterway inside flange is turned up, again, with the extra length or spigots either end.

The flange pattern is then split along the paper glue line and is mounted up into the mould boxes(flasks)
Both halves are separated with a 3mm thick sheet of steel separating the top and bottom flask also, the halves locate on this sheet using positioning pins show in red, that way both halves are exactly opposite each other.

The top and bottom flasks are then filled and packed with casting sand.

The top flask is then lifted off carefully, this is where the draft angles of the part are important so that the part does not interlock in the sand.
The part is then lifted out of the sand contained in the bottom flask.
Putting the two flasks back together again the void can be seen that the pattern has left.

Next, the inner timber pattern for the core shape(shown below again) is transferred to to the same shape but this time made in sand.

This transfer from timber to sand is done using a core box, ill cover this later.
With the mould open again the sand core is now installed inside the parent mould, its now clear why the spigots are need on the timber flange pattern, the sand core hangs or sits on these inside the main mould.

A sprue and riser are formed in the top mould in order to feed the mould with the molten alloy, a few 1mm vent holes are also made using a bit of wire.
The sprue is used to fill mould and also acts as a reservoir for the mould as the mould fills and also feeds the mould after the pour is complete and as the part cools if any shrink occurs.
The riser is not needed on some pours, its function is to indicate when the mould is full as the molten alloy rises up this meaning the mould is filled and pouring can stop.
A ''gate'' is also made at this time from the base of sprue into mould cavity, its size and cross-section depends on part complexity, wall thickness, and material being poured.

The mould is then filled

Once solid the mould is knocked out and the part removed from the sand, the sand can be re-used many times.
This sounds like a long process but its very quick, the mould and core can all be made very fast once you have patterns to strike them off.
For more complex cores of a bigger area a special sand can be used which cures when Co2 gas is passed through it, a core 600mm long can be made this way then and can remain unsupported without collapse inside the main mould.
Once the part is removed the sprue and riser remain on it, it would look something like below

These are then cut off and ground at the area where the gate was feeding part.
The gate location on the bracket can be seen clearly inside the red circle
See less See more
So, you can now see how a core is used to create a void within the part to be made.
Granted the water flange is pretty simple as thats open either end and the core can be ''hung'' inside the mould pretty easy.
But on an intake manifold or other part such as a cylinder head where the core is totally contained a different method has to be used.
On the water flange it was possible to hang the core from either end as the space inside exited out either end so that made core positioning pretty basic.
There are other ways to hang internal cores inside moulds on parts that have no holes through them to the outside.
Take the manifold below, the plenum chamber for all intensive purposes is a sealed box, therefore the core used inside it at the time of pouring has to be suspended some way, and in the correct location in order for all the plenum walls, and runners to be of the correct thickness/position.

If we look a bit closer things become apparent, notice the frost plugs used to close the holes that the core has been suspended on.
Its the very same principal as the flange core, but a different shape.
Spigots are also incorporated on the plenum core which are needed to hold the core in the correct place and at the correct height.
These holes formed by the location spigots are then closed off after the part is finished by inserting frost plugs into them.

Or, in the case of the other end of this manifold a breather pipe has been inserted into the spigot hole. This solved the problem that end and saved the use of a plug as the tapping was needed for the brake booster line.

If we look at other parts that have enclosed cores such as the manifold the spigot positions used to carry cores can be located too
Take a look at the 16v cylinder head below, a core is needed to form the inner waterways around valves and chambers, while this core is pretty complex the basics still apply.

And a picture of a water way core for use in a cylinder head(not the core for the above head)

You can see the extra spigots on both ends needed to locate it in the main moulds, these spigot holes left behind after casting sometimes become waterways either end of head, or can be capped off. Both are done on the 16v VW head as shown below
Spigot/core carrier hole right hand end, milled flat for water flange

And the hole the left hand end, machined, and capped with a frost plug

The same can also be seen on the engine block itself, here the three spigot holes needed to carry the water jacket core are capped off with frost plugs on the rear of the block

A cutaway view of water jacket and a guide as to what shape the jacket core would have been

And another showing frost plug removed

And an example of what a core needed for water jackets would look like, this is from a 6cyl but the principal is the same, check out the spigots on it needed to position it in the parent mould

Another feature of these spigots on the cores is to help expel any hot gasses that build up inside the core at the time of pouring. These hot gases escape out these, through an air hole in parent mould and out to atmosphere.
If these were not there, the gas could build up and rupture the casting from the inside out before it had fully cooled.
All cores are also given a gas impermeable wash before being installed, this helps direct internal gas out to the areas of the core where the spigots are, these spigot areas are left untreated so the gas can escape.
You'll notice above the wash is kept off these areas.
To give you a rough idea of what an engine block mould would look like heres a picture of a partially constructed one below, granted its done as a sketch but you can see the holes along the face of the red sand mould where the water jacket core would hang once assembled.

You can easily see that a more complex part requires multiple sand moulds and cores to be integrated together in order to form the final mould.
Anyway, more about cores and how they are made later on.

Modified by chippievw at 3:27 PM 11-26-2009
See less See more
Re: (chippievw)

This is some pretty awesome stuff. wish had the ability to do this kind of stuff. I would love to cast my own 16v head being able to design the intake and exhaust ports the way i wanted them.
See less See more

Awesome work and great thread! I did a very small amount of casting in my metal shop class in junior high, but nothing complex or requiring dimensional precision. Very cool process, and I'd have loved to learn more about it. I look forward to following this thread and seeing your progress http://****************.com/smile/emthup.gif
See less See more
Re: The complete guide on casting parts (chippievw)

great post!
Re: The complete guide on casting parts (Pf3il)

Excellent Thread!
I have found a few places in the New Jersey area which are willing to entertain the idea of custom casting a one-off, or low quantity run of pieces... Most places are only interested in mass quantity runs.
Here's a cool thread from not too long ago: http://forums.vwvortex.com/zerothread?id=4420125
@ Jetta boy, Indeed I did, and a sound guy he is too.
Thanks for the kind words to everyone else, glad your finding it to be of interest!
Next up a run-down on the flasks and the parting plate.
They are pretty basic, but vital non the less.
You could make these very easy if you wanted to, later I'll be making a bigger set for an Intake manifold but these ones are perfect for small castings.
I picked these up years ago at a car boot sale with the intent of doing all this, hard to believe Im only getting round to it now.
The guy I bought them off thought they were grates from a stove............I told him ''Yes, they must be'' and handed him 20.
The drawings I did above pretty much sum them up but a few pictures and a quick run down on them wont go astray.
They are made of alloy, very strong, and have two integrated positioning pins to locate the top flask exactly onto the bottom flask.
The precision of these pins is vital in a good casting, if there is play, the top half wont locate on the bottom half exactly and a step at the parting line in the finished part will be very noticeable.
Luckily for me, these are quality items and no more than .25mm play exists, just enough to make removal of top half easy.
The guide pins do need a bit of a shine up and oiling but thats pretty easy to do.
Here they are below, in the assembled position, notice the large corrugations in the walls of the flasks to hold sand in place, remember, these are open top and bottom.

And in the open position.

A view from the side in the closed position.

And again from the side in the open position.

A closer look at the guide pins, one side has a round pin, the other a hex pin, this way the top can only go on the bottom one way.

And the hex pin, slightly larger in size than the round pin. The hex pin socket is also adjustable clockwise or anti-clockwise so any play can be compensated for.

See less See more
Next, onto the parting plate and how its made.
Originally I was going to make this from steel, but given the fact that steel isn't transparent it could make taking photos or explaining stuff awkward, so, I went with some 12mm acrylic sheet.
Once a size was taken I cut it out on the table saw, I left a bit of extra all round the edges to make it easier to grip it when removing it.

Chop chop.

After that I gave all edges a run with the electric planer, they were a bit rough after the saw and Im going to be handling this plate a fair bit.

And a bevel to remove sharp corners.

Job done.

Now onto the holes, these have to be pretty exact too in order for the plate to slip down onto guide pins.

Transfer measurements to plate and scribe them out.

Bore guide holes

Like so.

Next up, 3mm bit in the router to inscribe the word ''up'' on plate, that way installing it upside down by mistake will never happen.
You could use a marker but, I doubt it would stay on there too long.

Parting plate installed, mould in the closed position.

And a view from the side.
Drawing I did earlier below it showing proposed set-up as shown now for real in photo(Pattern omitted).

(more to follow)
See less See more
Re: (chippievw)

Outstanding stuff!!
One question though, what's the purpose of the venturi in the burner tube? Several years ago, I made a similar burner but without a venturi it worked great. I've since switched to a naturally aspirated burner.
Re: (ABA Scirocco)

This is so cool! I can't wait to see the finished parts http://****************.com/smile/emthup.gif
See less See more
Re: The complete guide on casting parts (chippievw)

Quote, originally posted by chippievw »
Its important to note that this is not the cylinder head thread, this will be covered in another thread on Club Gti

One more thing, when you start that thread, would you please post a link to it here?
See less See more
@ABA I added the venturi for two reasons, the main one being that when the gas reg is removed from cylinder just before shut down a small suction is created by the passing air where gas enters, this then completely clears all gas from the length of hose. And second, I was afraid that gas inserted into a 2'' full bore pipe would not mix well enough with the passing air, so, a reduction at the venturi lessens cross-section and increases velocity of the passing air @ fitting.
Hope this makes some sense!
Ill be sure to insert link when the time comes.
Re: (chippievw)

That does make sense, I don't think it's necessary but it's at least prudent.
It's one heck of an ambitious project you've set out for yourself, I'm really looking forward to seeing it, especially keen to see how you make the core boxes for the coolant passages etc.
Hmmmm... I seem to recall that very complex shapes, such as the cylinder heads and manifolds, may also be made by the various "lost X" casting methods.
I've seen an example of a gear-drive transfer case being cast from aluminum using a foam template left in the sand, which evaporated away through an increased number of risers and vents as the metal was poured. Would that not be easier than attempting to create such complex, intricate plugs?
With the disadvantage, of course, of being only usable once.
All manufacturers Im currently aware of use the sand method to make their engine blocks and heads, this incls Bmw, Vw, and Ferrari. F1 also uses sand. Manifolds are also made using sand, as you can see from my photoabove and the rundown on it.
Im sure you could use foam for other stuff and Im not dissing it, BUT you do need a template in foam for each pour that would either have to be rapid proto-typed, cnc'd, on cut by hand. With sand you only need one.
Hope this helps explain why I chose sand.
Thanks ,
1 - 20 of 692 Posts