Derek

It's most likely the machining of the 40 mm round down to the profile that is the majority of the cost. Not the hardening. If I was doing it I would do all the round work in the lathe and then switch over to the mill with a 4th axis to cut the lobes. I can do this but I can make more money doing what I do best and pay someone who does cam cores all the time. And that costs between $1300.00 and $1800.00 each. My inbox hasn't exactly been overflowing with people who want to buy a head. And the cam prices haven't made it any better. Let's face it. When we were all saying I wish there was a bolt on cross flow twin cam head for the L6 we should have added "and it needs to be under $5000.00 complete with cams valves etc". Because based on the feed back I've received off forum this is the point where the decision is easy. After that for every $1000.00 you add it becomes harder and harder. The $3000-$4000 "just for cams" pretty much killed off the two that were serious. If I pull the plug now at least I'll shut everyone up about a twin cam head for the L6 and the mods can instigate a immediate ban on anyone who mentions it again:) Because a twin cam head for the S30 can't be produced in the numbers that will be purchased by S30 owners for under $8000.00. And that's if you're creative and can scrounge. $10,000.00 or more if you don't want to scrounge and would like to make reasonable power. Even if I based it on an RB30 the additional machining costs would put it on par with the K20. There is a reason the OS Giken is $40,000. This kind of work is expensive. 4 or 5 people who are ready to commit at these price points and it can move forward if not I've accomplished what I set out to do which is to showcase 3D sand printing to my clients. And believe me at this point they're impressed. Derek

Thought I'd do a quick update. The cam situation is a lot tougher than I led myself to believe. Actually it's not tough at all it's just expensive:) Custom made steel billet cams are going to be in the neighborhood of $1300.00 - $1800.00 EACH! And this does not include the final grind. I'm working on a much saner approach using cast iron cores which is what most of the cam cores are made up of. Google "chilled iron cam cores" and you'll learn plenty. The manufacture of these cores are not at all outside the bounds of what I can do. Being a cast item it actually brings it into my sphere of influence. I have 2 foundries that can pour the cores depending on what iron needs to be poured. The general consensus of my "brain trust" is that Ron Iskenderian is the guy to talk to as he is receptive to this kind of crazy talk. He is on vacation for two weeks so unless he's following this thread it will probably be three or more weeks before I can get some answers. So we're basically on hold for a bit. I also revisited why I ultimately chose the K20 over the RB head. The RB head is a lot trickier to machine and cast. It has an internal oil passage that needs to be cored as well as 2 passages that need to be gun drilled the length of the head. They are too small to core as the sand would probably break. Plus there is the matter of machining the cam towers and precision boring the lifter bores. If we were doing a group buy of 10 or 15 heads and getting all the machining done at one time then this would be feasible. Of course if we were doing a group buy of 10 or 15 cam core sets the price would probably come down on those as well. So it goes. Derek

Happy birthday you old fart! Welcome to the club. Wait a minute isn't it tomorrow where you're at? Or is it yesterday where I'm at.

That was awfully sappy...I think I threw up in my mouth a little..

Nope:) But it is what I do for a living. So there's that. The internals are very close to the K20 with extra meat in a few areas. Honda tries to save every gram for production costs as well as vehicle weight. Me, not so much. Thicker combustion chamber, thicker bottom surface, larger radius on the fillets, every head x-rayed, every head impregnated, etc etc. But you certainly bring up a valid point so don't think I'm dismissing you. All I can do is bring everyone along every step of the way and it's a journey we'll take together. The one that goes on my car may be the last and that's ok too! Derek

Head casting pricing. Some will be happy. Some will be sad. And so it goes:) Well I was trying to get a more complete picture for you but unfortunately I don't have cam pricing yet. And I'm not a big enough fish to push. The head casting, valve cover and front cover UN-MACINED is going to be $4500.00. My ESTIMATE for machining the parts to the point that an automotive machinist can finish it is going to be in the $500.00 range. I'm counting on being able to fire and forget on the machining. This should be doable based on the fact I'm printing the cores. If I have to stand over it the whole time it will double for sure. I'll have a handle on that after I machine the first one. Every head casting will be X-rayed and impregnated. Impregnation is a processes where the casting is put under a vacuum and a sealant leaches into the pores. Not in this is the intake manifold. I'll obviously be offering one but I haven't had time to draw it up to get pricing. If you think you are going to want one of these then by all means PM me. The more people we have on board in the beginning the better. Thanks Derek

Went and picked up the casting yesterday. Looks great. No surprises. I checked the shrink rate when I got back and my edumicated guess appears to have been good. I was off by .015" or so on the total length which for this size casting is not too bad. keep in mind trying to measure a casting is like trying to measure a rock. Just to give you an idea how cool 3D sand printing is here is a shot of the part number and name. Unfortunately that's the only one you'll see like that as that area is going to be where the new water outlets will be. Not a lot of clear exterior area for branding! Met with the shop yesterday that will be doing the automotive end of the machining. He didn't see any obvious issues which was good news. I should have some rough pricing soon. I'm waiting for cam core prices. That will give you a more complete picture. So the next step is to make a plastic mockup of the head. I'll machine it as far as I need to to feel comfortable with the strategy. Which at this point will be most of the way. The worst part is I only have one 4th axis and it's making money for the next few days so I'm on severe hold I hate when work gets in the way of fun.

I believe you are correct. Tony pointed this out to me but it was too late in the process as the molds were already printed. Version 2 will have access to the water jacket above each exhaust port. If you want to get to it you mill through the head. This way if you have a high heat turbo build you can exit the water at each port tied to a single tube with thermostat. Personally I won't need it for my NA build but it will be there for those that need it. The OS Giken is this way. Yea that would suck. I thought about that a few times. I hope they are actually getting the backup service now.

Exactly. The combustion chambers, ports, water jacket and the front and back of the head are printed in one piece. If you look at the bottom picture of the casting you will see a tab on the front and a tab on ...well half a tab on the back. It short ran. Anyway on the front tab you see 2 pins. Those are hardened ground steel dowel pins that are cast in place. There should be 2 in the back, and that will be fixed on the next ones. Those pins will slide into a fixture on my rotary table. I'll then probe various points on the head to find the center of the casting. You then average out from there. I do this pin trick on all the larger castings I do. Makes life little easier for the machinest. You know how cranky they can get. When I modeled in the cam tower and rocker, the valves that I had modeled in previously didn't line up with the rocker. I tilted the valves .5 degree pivoting at the bottom of the valve guide and it lined up. So the difference between lining up and not lining up is half a degree. Now I could drill from the top like the factory and wherever the valve seat ends up well thats where it ends up. If I had a head to scrap this would be a simple fix as I will get the angle right after a couple tries. Trying to get it right on the first hole will be tricky.

Larger gear means more teeth which means new gears up top to keep the ratio correct. Also the honda chain is a smaller profile than the L6/KA chain so there might be a strength issue.

The crank gear was too small to fit over the snout. single chain was my preferred route but there was just no way to make it happen. At least not with off the shelf components.

Pretty much everything. Think of it as the ultimate Vtec killer. I still have things to work out with the cam towers and oil flow. Technically the cam gears would be in the back of the motor if all things were equal to the K20. Since I'm having custom cams made this won't be an issue. Of course since it's custom anything is possible:)

Yea if I was doing this for a customer I would tell them to budget for four heads. One for the foundry to get right. One for the machinist and one for the unknown. If your lucky you might hit on #3. There is a lot of good and bad in the 3D printing field. I just wish they hadn't settled on mesh as the modeling standard. I can be quite cavalier when I'm modeling for CAM. who cares about gaps as long as the smallest ball mill won't drop into it.

You can also just use all exhaust rockers as they aren't Vtech. As far as machining goes there is one point that may be vexing for people (me) who have never machined a raw head casting. Getting the tops of the valves to line up with the rockers. I modeled in a cam tower and two rocker arms. The models are based on high res scans so they are accurate. I based the location of the tops of the valves from one of the heads I purchased. As you can see in this screen grab there isn't a lot of room for error as far as the angle of attack of the rocker to the valve top. This has been my biggest worry. I imagine you only get one chance to get the angles right on drilling the valve guide bores. I'm not totally sure if the head casting is any good or not. Castings shrink as they cool and you factor that into the design. The problem is at the level I work at it's guesswork. The more cores a casting has the harder it is to predict the shrink rate because the hard sand reduces the shrink. So the reality is the casting may be too long or too short. It may also fail when I have it X-rayed. If the casting is junk then I have 12 tries to determine the right angle. If the casting is good then I'll probably machine a plastic test head that matches the actual casting and practice on that. If anyone has any experience on this kind of valve work I'd love to hear from you:)

Hi I wasn't. I was merely using it as an example of the high end of things. Not the market I'm going after. This project is as much about proof of concept as anything else. Patternless casting has arrived. It seems expensive, but for this type of project it's ideal. I'm not the only one doing this. Ford racing bought a few of the 3D sand printers and are using them in their head development. There is a guy reproducing Ferrari heads that are NLA, and I'm sure there are more. I do a lot of motorsports work. This head will open up some of my clients eyes even further as to the possibilities. I'm doing this head for the Z because it interests me. And I have a car to put it in! I mentioned in the post that the nissan and 3000 were going to be machined in. The actual castings will be blank.

Love it. I think I want my member name changed from "That Horny Z Guy" to "That Derek Fellow"

Cheap as compared to the OS Giken:) I think that comes in around $40,000.00.

All Honda. At least the intake inlet and the exhaust outlet. What happens in between is my design. I could but in reality it's a Vtech killer. I suppose if someone was inclined they could make it a Vtech but that is a hill I'm not climbing! No problem prodding! I'm going with custom Crower cams. This is not as big a deal as is it used to be. Thank you technology:) Here is the really neat thing about 3D sand printing. If you want intake on the left and exhaust on the right all I need to do is 3 mouse clicks and the entire mold package is mirrored. Sometimes cost goes beyond the price of the components. The RB has cam saddles that are part of the head. This requires line boring. The K20 has bolt on cam towers. This reduced the cost of machining greatly. That was the tipping point for me. I don't own a line boring machine:)

Hey I just noticed that I failed to mention that ALL of the foundry sand for the mold was 3D printed. not just the cores. There are no patterns except for some exterior gating pieces. This is pretty cutting edge stuff. I've been woking with the process for a few years now and it's really amazing. Any changes I need to make to V2 are done digitally which is nice.

Thanks. The intake will be custom and will be flanged to accommodate the DOCE pattern. I think the port spacing on the DOCE pattern will work so I can use 6 individual Jenvy ITBs. My plan for the exhaust is a little more convoluted. L6 block on engine stand, Mount P90 on block, bolt on header and fixture it to the side of the block. Remove header and p90. Saw flange off of header. Mount new head on block and return header to fixture. Manipulate tubes with heat and fabrication to match new flange and weld. This build really is about how cheap. Although the Jenvy ITBs are a luxury, I still want them! I could go with OBX stuff and may have to if I bust the budget.

Hi Ray. Since I'm using the Honda K20 valves and valve train I felt it was best to stay as close to their design as possible. We decided against the RB as a donor because the K20 just had so much stuff easily available. This build is going to be the budget build. I bought 2 K20A3 heads used. That will give me enough parts to do one head plus some spares. Since I'm not using the Vtech option I needed 4 additional exhaust rockers. Derek

Well you know..Idle hands. I'm finally going to use the .040 motor I bought from Tom. It's going to be the mule for this head.

This is all Tony D’s fault:) Hi everyone. Some of you know me from my DIY EFI manifold project http://forums.hybridz.org/topic/63445-making-my-own-efi-intake-the-first-casting/ Well now I’m tackling a DOHC head. Some history: I get a PM from Tony D in September of 2012 about the Goerz-Paeco DOHC L6 Head coming up for air.http://forums.hybridz.org/index.php/topic/109116-dohc-l6-was-somebody-looking-for-the-goerz-paeco-dohc-l6-head/?&gopid=1020596 Which led to a brief discussion about what would be involved in producing a head. And that was the end of it... Or so you thought. Tony and I started trading emails about the feasibility of producing it and what kind of heads we could use as a donor for the valve train. Tony suggested bringing Jeff P on board because of his work with cooling the L6 head. Tony and I kicked around RB and KA motors for a while and then Tony suggested checking out the Honda K20. Bingo. Removable cam towers, factory roller rockers and a plethora of aftermarket parts. I called one oh my clients that is big in the import drag market and procured an old K20 head. I did some quick measurements and decided that it was probably workable. And so the journey began. The idea is to use the cam towers and valve train from a K20. I grabbed a quick scan with my white light scanner of the exterior. It’s dirty because I’m only using it as a reference scan to design against. I imported the scan into my solids program and then proceed to replace the mesh with solids. Jeff provided me with a great CAD drawing of his head gasket to work from. I decided to take a module approach to model the head. I made a single cylinder complete and the copied and offset it for the other cylinders. I then pulled the solids together to create a single model. At this point I was at a standstill until I could figure out what to do about the timing chain. Jeff and I had a long conversation about pros an cons of various designs. The preference would be a single chain so I ordered up a K20 set to see if it was workable. NOPE. Lower timing gear is too small to bore out to fit the L6 crank. Back to the drawing board. I decided to give the KA set a try. I decided the Altima set would be a good choice as it’s only a single row chain up top. With the roller rockers I don’t think you need a double row. I did a rough mockup on a piece of wood mounting all the pieces. Then I machined a chunk of plastic so I could actually mount it to the block. You can see I’m trying to incorporate the KA guides and lower tensioner but that won’t happen without a new lower timing cover. Here is the lower section with stock modified L6 components. The only tricky bit is I need to remove one link from the chain. I think this is a workable solution. Now I could finish modeling the front and back of the head. Next the front timing cover. The water outlet exits the head in the front and makes a 90 degree turn through the timing cover where the thermostat housing will mount. Now the valve cover. The K20 cam towers kind of dictated the overall size but I was still able to get the styling I was looking for. I was trying to have the flavor of the S20. I’m still tweaking the design but so far I’m pretty happy. The “NISSAN” and “3000” will be machined in so they are kind of placeholders for now. After talking with Jeff about his testing on coolant flow I decided to get the most out of the 3D sand printing process as I could. I designed a water jacket that would be pretty hard to duplicate traditionally. There is a lot of surface area so hopefully there will be large improvements in cooling. Here is a nice shot of the intake and exhaust cores, water core, and upper tensioner. You can see where the cross drilling will go to supply oil to the head, tensioner and idler gear lube. There will have to be additional oil supplied to the last three cam towers as the single feed in the front won’t cut it. Here is a cross section of the head. The cut plane is through the center of one of the intake valves so the ports look a bit off. There is machine allowance on the bottom so the combustion chamber is a little larger than it will be after surfacing. Well like I said in the original Goerz-Paeco post I think this is a very viable approach to producing DOHC head. Should make for an interesting conversation. Derek Oh and one more thing. What good would a thread about casting a head be without....... / / / / / / / / / / / / / / / / / / / An actual casting. We poured this Monday and I just got these pics from the foundry today. I haven't seen the head personally but we're pretty sure it is good enough to move forward with. Still lots to do but you can't imagine how happy I am to get the first one right. As long as things proceed as planned I will be selling these. There is still a lot to figure out. I have the foundry working up some pricing for me and I'll post that as soon as I can. Thanks for looking Derek EDIT: I now have a blog about this head on my site. I'm going to continue to post here but the blog is a condensed version of what you see here. There is a FAQ on the left side bar that has a lot of data. If you are new to this build you may want to come up to speed there and then start following it here as there is a lot of great back and forth that isn't on the blog. http://www.datsunworks.com/Blog/

Listen I'll be the first to admit this is still and will probably be a pipe dream. But as technology advanced things that were previously absurd become feasible. The only way i would consider getting involved in this is if it used off the shelf components. That just makes sense. Also keep in mind what I'm pushing is being done right now by ford, chevy etcetera. This is how they are prototyping their stuff. When ever I quote out manifold or head work this is how I propose it. By printing the cores you can keep tweaking things until you get it the way you want it. Then after you have a workable design in both performance and cast ability you commit to traditional core boxes for production. By 3D sand printing the cores the combustion chamber and ports can be tailored to each head. As long as it fits inside the outer shell your good to go. The thing I like about this design is it looks right for the car. So if you can design an exterior patten that looks period correct and custom print the internals what's not to like! Derek

Maybe because I'm a pattern maker the casting seems to be the least of the problems. Cams,timing chain and gears, lifters etc are all going to be harder to come up with. Also people qualified to bring a raw head casting to a finished product probably aren't looking for any hobby projects. And thanks to Tony instead of concentrating on the bid I should be working on i've given the casting approach some more thought. And I really think this is feasible. Acquire a copy of the prints. Develop the 3D model. Carve a traditional cope and drag pattern for air set sand for the exterior of the head. This will cut down enormously on the cost of the 3D printed sand and shipping. Have the water cores and air passage cores 3D printed. The benefits to printing the cores are huge. first and foremost your not constrained by the limitations of a traditional core box. Since each core is printed separately any changes are just a matter of redrawing the model. Also because it's a continuous 3D print there are no part lines and the sand is super smooth. To put this into perspective I would design the intake, exhaust and combustion chamber as one continuous chunk of sand. The finish and precision of this method can only be matched by investment casting. Also this gives you tons of freedom to individually design each heads internals. Here is a shot of a sample of some sand I had printed. I was doing some restoration work for the King Lunalilo Tomb in Hawaii and was curious how the sand would look. I sent them a STL file of the crown scan and they printed this. I was really blown away by the detail it picked up. So you can see the possibilities for this. First step get the prints. Second step find a machinist. Preferably one that want's a twin cam L6 head Third step source cams and assorted hardware. Fourth step develop 3D model for casting. (me) Fifth step have machining model and prints developed from casting model with proper tolerances and such. Step six patterns and castings (me) Step seven Machine castings So fill in all the steps except for four and six and this could happen. There's makers and takers in a deal like this. Makers donate labor takers donate cash. Derek