What's new
What's new

How to cast engine manifold

Wingit

Aluminum
Joined
Oct 1, 2006
Location
Wisconsin
I have some old engines with rusted up manifolds and would like to know if it is feasible to have new ones cast. As far as I know there are no reproductions being made. There is a foundry near my town I want to check with but would like to know more about the process and how difficult it is to make the pattern. I don't understand the core pattern and placement. These are 4 and six cylinder engines with one piece intake and exhaust manifolds from about 18 to 24 inches long overall.
Thanks
Wingit
 
That's like asking, "I need a camshaft, can somebody tell me how to make one?"

A foundry familiarization course in a college or technical school used to take at least one semester. You could check on Amazon or ebay and buy some textbooks. There are foundry owners and patternmakers on here; they'll probably quote upwards of 15K to make one part from your sample, if that gives you an idea of what you're asking.
 
You say these are one piece intake and exhaust manifolds, are they for tractors? Not that it matters. The first thing you need to is find a competent pattern shop. Somebody that can take your rusty manifold and make a drawing and then make the pattern and core box. Next off to the foundry You want to get the foundry involved early in the process. They can give valuable input into making a better casting before you pore any iron. After the castings are made they need to be machined.

You are looking at thousands and thousands of dollars to make you first manifold the second one will be much cheaper.
 
A brilliant patternmaker friend made 3 patterns and cores for me for $15,000. He made about $20/hr doing it.
 
The issue is core. To duplicate the inside you would have section an old one and make a drawing. Then you will have to design the mold for core support points (core prints and chaplets). Alternately, you could make a drawing of the outside of the casting and then assume a wall thickness and make a core drawing that. The danger with this is that in the development of the engine, various sections may be of different thickness to control temperature and expansion.

Sounds like a project.

Tom
 
So you make foam manifold and bury it in sand - don't you get a solid - not hollow - manifold?

Or do you figure out how to make a hollow foam manifold and get the sand core inside?

Or does your solid foam manifold have core prints and a regular core box?

:D

Don't even need that. Find someone who can do lost foam in cast iron and make your own pattern out of polystyrene.

edit : Looky here :

Lost Foam Casting
 
So you make foam manifold and bury it in sand - don't you get a solid - not hollow - manifold?

Or do you figure out how to make a hollow foam manifold and get the sand core inside?

Or does your solid foam manifold have core prints and a regular core box?

:D

Lost foam casting works surprisingly well for one-offs with a loose sand core:

Automotive Water Neck | The Home Foundry

This casting is more inlet manifold related but has no core to speak of:
Carburetor Plenum for Automotive Induction System | The Home Foundry
 
The cores and patterns really are the easy part. It's knowing where to put the chillers, risers, and gates that will get you a good part. When I set-up for parts in a foundry, we had one part that was cast 6 times before it was right. Cast iron is a lot easier, but when you have a complex part like that, you'll need to X-ray the part after it's cleaned. That is unless you've got an extremely talented foundry engineer.
 
I built a schedule 10 SS exhaust manifold for a tractor... looked like it belonged on a teenagers honda civic, but it solved the constant cracking problem the farmer had, much cheaper than trying to cast something.
 
Why even bother to cast manifolds? You can make your own exhaust manifold with a welder and grinder and it will work. The intake for a 4 or 6 cylinder engine isn't that difficult either. All you need are basic hand tools and time. How much rust can possibly be on a manifold to have it stop working? Is it rusted in two? Have a hole?

Now if you've got one of them Buicks with a plastic manifold that has failed and don't want to pay $3000 to have the dealer fix it, that is something else.
 
Why even bother to cast manifolds?
I ASSumed when he said "no reproductions are being made" that this was for something collectible and he wanted it to look the same. :)

Otherwise, you're right. Especially the exhaust, just a flat plate and some tubing, really easy to make your own headers.
 
Wingit- taking your Q at face value, here is an example of a core that would be similar in concept, but simpler, to what you would need to develop. If you are planning to do it yourself, I say go for it. You could probably have the first iteration done in a week or so of full time work. The foundry itself will supply (& charge you for) the "rigging" which is how the iron gets into the mould and flows smoothly for a complete, void free fill. Given your intimation that this is probably a hobby or collector application, I suspect Emmanual King/Cattail Foundry could do it with something close.
But again, one step at a time.

First, you need to make a pattern on a matchplate that would form the outside of the form you want to duplicate. For CI, it also needs to be 1/8" per foot larger overall since the hot iron will shrink. Look at this simple pattern for an aircraft stick grip - notice that there are some extra nubbins on the plate. The keystone projects on the actual patters are called "core prints". they will make a hollow shape in the mould that extra-long ends of the cores will be place in, to support them. The gray shape in the middle is part of the "rigging" which the foundry installed on my pattern, to form a "pouring basin" to act as a reservoir and flow guide for the metal.

smt-pattern1.jpg


This is the other side (bottom, or drag side) showing how the foundry added runners as part of the rigging, to get the metal smoothly into the backside of the handles. Visualize how the pouring basin and runners connect to provide a void for moving the molten metal.

smt-pattern2.jpg


This is a shell core box, to make the cores. Your manifold would not need complex keystone shapes because it has enough protusions through the open ends of the manifold to support it from twisting. Round core prints added onto the ends of the patter runners would be fine. Notice the little round projections on the handle at the top and back - these are more core prints that support the core in the mold and keep it from twisting or floating when the metal is poured. You might have to come up with a few locations on the manifold to include something similar, and then maybe put pipe plugs in or weld shut on the finished. If that is objectionable and you can't find any "accessorY" holes in the original, there are other ways around this. Also, some old manifold shapes were such that just supporting the core at all the openings was more than adequate. You should have a talk with the foundry and ask.

The blue part is what the actual core will look like.

smt-pattern3.jpg


I use Freeman's (foundry supply) repro products to duplicate parts and patterns, and to make test prints.
This is a test taken from the pattern, and from the core box, to make sure they match up and don't have any crush when the foundry makes the real deal.

Everything blue represents sand (or no-bake). The voids will all be metal, including some protrusions from the runners, which the foundry will bandsaw and rough grind off (process known as "snagging") before dumping my parts in a crate.

smt-pattern4.jpg


2 versions of the product after boring. You can see how I positioned the small core prints to take advantage of holes that would need to be added anyway.

smt-pattern7.jpg


smt-pattern5.jpg


If you have to pay to have the work done, it is probably not cost effective.
If you wanted to do it yourself as a project, it is very accessible.
Read a bunch of books, then take a sample to the foundry and ask them how they want it mounted on what size matchplate.
Also ask them what process they prefer to use for cores, so you can make the appropriate core boxes. (You have to read ahead so you understand what they are telling you when you ask for advice. They won't actually tell you how to build things except maybe a few pointers. But they will define what process they intend to use, which will define how you make the patterns a core boxes.)

Go to the Foreman's site and watch all the videos about their products. You could fill and smooth an original and about build it up long enough just to duplicate the exterior with their products (not cheap, though, you can't cheap out and be successful). However, if your manifold is over a foot long, you probably have to bite the bullet and carve a new oversize pattern. At some length building up an original won't work because the port centers will shrink out of line.

Your core box would not need to be metal, these days no-bake products work as well as the old shell cores. Shell cores are cheaper for production multiples, though. Paste up sand cores are still viable at small foundries, but will cost several multiples in price compared to more modern processes.

I have more photos of using Reprothance for other casting products, but enough for now, you may decide it does not interest you.

smt
 
Bear in mind that casting manifolds and many other components was mainly done for reasons of cost on high volume production .......the always high cost of patterns and tooling offset over the numbers required.

So;- unless its an ultra rare exceedingly high value, originality is everything application, fabrication is the way to go, ..further more a skilled fabricator can make something like a manifold ALMOST look like a casting, ...and still come out $000's to the good.
 
now if you really want a kind of good idea of how to do it
go the a casting form AlloyAvenue network
there are guy there who have done similar to what you want
in their back yards.

now you would have a hard time doing a part of the size you are
talking about in a back yard furnace that much iron is just too heavy
 
Possible error above -

I may have described which side is the drag incorrectly.
I'm no longer sure which side was actually up when poured. though probably opposite what I described. There could be "logical" reasons based on head, and on the flow of clean metal just before it enters the part cavity, to do it more or less either way. Reservoir and pouring well down seem to make more sense. The metal accummulates there, the turbulence & velocity of pouring settles down so the flow does not scour the walls, and then when it attains runner height, flows evenly, smoothly, and steadily without voids into the mould at a uniform temperature. The sprue (shaped by the foundry, into the pouring well with a tapered tool, when they make the mold) provides the head.

What bothers me about the scenario described above, and the reason I originally descibed the runner side as the drag, is that castings are usually bottom filled so the crap floats to the top (& preferably into a riser that will be cut off). Parts were from some 25 years ago, I sorry about my confusion on that point.

However while it can influence how you position a part on the matchboard, the foundry figures this all out and places the rigging accordingly. On some rigging, they may offer you the option to add it, if they think you fully understand the details. However that service was usually negligible cost for the things I've had cast and if the foundry does it, it is on them that it flows and chills correctly.

smt
 








 
Back
Top