The Merkur Encyclopedia

4 Valve Cylinder Head

by Rick Byrnes

This has been asked enoughh times that it finally deserves an answer, and a little bit of discussion. I will also offer some opinions that may or may not be very popular on

In chronological order

Zakspeed developed a pure racing 4 valve, direct acting bucket follower cylinder head that bolts on the 2.3L OHC architecture. This, I think is when they were doing some F1 engines in the late 70's. By production standards it was not appropriate for use and had a very short expected life. When one designs true competetion components, you begin with a set of assumptions, one of which is the expected life expectancy with some margin of safety. An example, my engines for land speed racing are defined as a life expectancy of 2 hours at design conditions.

Hank Dertian, an old friend of mine at Ford, was working on the turbo program and actually had a program to redo this head to be appropriate for use in a production application. Hank is a good engineer, and with the expertise of a number of other engineers and technicians, sort of bootlegged this head design into several samples to be built into test engines. I know of only 1 engine completed. It was in a grey XR4Ti and presented to Ford management in hopes of creating wants and needs for this type of engine. The car drove like it should (production), and went like stink.....No I never drove it. Just saw the thing once. The difficulty is that in the mid 80's the 5.0 Mustang had come alive, and Ford management already had an embarassment on their hands with the SVO mustang, that would kick the 5.0's ass in almost every venue, and they were not too interested in a really high tech engine. Rembmber this is the mid 80's and the 4.6L 2 valve and 4 valve had not yet even become a chief engineers dream. I have no idea of what happened to the car or engine.

During the same period of time I was slightly involved with another cylinder head design that was being proposed for the Ranger truck program. It was a twin cam 2 valve with direct acting bucket followers, that had pent roof chambers, and the potential to become a 4 valve with few changes. The cost and complexity were higher than the cast iron "D" port head, but it showed great promise. The design work was completed, and ready to go off and make prototype parts. Concurrent with this there was a group of engineers at advanced engineering working on the helical port dual plug engine that was reported to be the best thing since sex...
A design review was conducted and a decision made to go ahead withe the dual plug and scrap the design for the twin cam parts. Now since we are so ISO perfect, nothing remains.....So now we have an engine that still does not give us the power the dual plug guys promised, but oh we do have Distributorless ignition system....oh boy....can you detect a note of sarcasm. Actually you should not. Again we need to remember this was in 1983/84 and things were driven by promises and politics and cost. People promise certain performance and decisions are made. Once that decesion is carved in stone it is almost impossible to go back because of the millions of dollars in investment. I'll bet that back then, the new design head, which would have required a completely new cylinder head maching and assembly line, would have cost Ford at least $20 Million.
So none of these parts were ever made. Again remember this was 14 years ago and we did not have the analytical tools that could do a good job of predicting performance like we do now. It is a whole new world of engine design. However in a way I see us making some of the same mistakes. but that is another story...boy I really am ready to retire.

Now some time in the mid to late 80's Duaine Esslinger had mentioned to me that he heard of a 4 valve head being done in Argentina. This is just after I had begun traveling to Taubate Brasil, and I queeried through my friends in Taube and San Paulo. They thru the network could not learn of any such head at Ford or even outside, perhaps aftermarked. This was never verfied and no one in North America ever saw any parts. At least not in Dearborn or Southern California.

The 4.6L 4 valve Modular engine was born, and everyone who studied it knew the design of the cylinder heads was quite nice. Especially for a mass produced peice. Valve train geometry was good, with a quiet hydraulic system and roller rocker arms that have been working quite well in other applications. It was decided by Ford to initiate a 4 valve head for the 2.3L OHC. The design assumptions included utilizing the 4.6L design intent and the head looked sort of like a copy of the modular head. Thats ok, especially if it works. This was at the beginning of the Full Service Supplier initiative and Roush Technologies was given the job of completing prototype components. Yes, the same Jack Roush as Roush Racing does a tremendous amount of engineering and manufacturing for the Big three in the Detroit area. How is it we think he could have built the racing deal into a profit making business. Not without some real money making deals around. Especially before Winston Cup. At any rate, they produced some really nice castings, developed the valve train and produced a number of fully functional parts. (I don't know how many). At the beginning of the program it was rumored that Roush absorbed part of the head tooling costs for the rights to produce special parts for their own use. (now do you call this foresight or what).
Ford for whatever reasons cancelled the program. All advocates of 4 cylinder power in Ford just fraked out. This was our last chance to make the OHC a high tech engine. One vehicle was built, a normally aspirated Mustang and I was told by some friends that it drove really well, and met all expectations, but again the investment was really high.
Ford truck office though came in and said they would like to work with the design for a potential Ranger truck enhancement. One engineer was the project manager. He virtually moved everything to Batten Engineering and proceeded to make new castings with one of the intake runners for each cylinder smaller, to be used as a primary runner for high velocity and the hopes of a high torque engine at a lower engine speed. The goal was hororable, but this was a bootleg program being run by people with out the background or discipline of the engine engineering group, and it eventually failed. Again there were some parts made but I did not see any. KEEP IN MIND THESE WERE ALL N/A ENGINE PACKAGES. NOT TURBOCHARGED.
A turbo package in aluminum needs much thicker sections in the casting.

Of all these iterations, I know of only 1 part that is still in captivity outside of Ford. I have no Idea of how it occured, but if there is one, there are probably more.

When the official Ford rocker arm 4 valve program failed, Roush, while having rights and now virtually owning the tooling (remember, he paid green money), modified this tooling to enhance cooling and strength for a turbocharged racing engine in the likes of the GTO program they were in the middle of running the XR4 in SCCA Trans Am racing. Roumors had it that the 2 valve program, based on the Esslinger head, made 745 HP and 545 lbs ft when ready to race and more than 1000 HP with no restriction on turbochargers. There was some dyno testing with the new head and I heard figures of roughly 100 HP more in any application, under the same conditions of course. Unfortunately the head never made it into competetion. Again being a company leaning to V8 engines it was decided by Ford to change to Mustangs and 5.0L type engines. End of program.
I have seen 0ne of these heads and they are beautiful. I would give up my first born for one complete system, but alas, NONE are available. Now don't go calling Roush trying to buy this stuff. Jack will NOT sell it. He has established a museum and the parts they currently have are to be retained. I think it is a good idea. Otherwise how do you capture the history and character of an organization.

Now, why has no one in the aftermarket produced a 4 valve cylinder head. Lets start with some assumptions
Bolt on 2.3L OHC Block with no modifications to block.
Unique intake porting requiring a new fabricated intake manifold.
Unique exhaust porting requiring a new fabricated manifold.
A new fabricated cam cover.

Now lets look at the current Esslinger head.
Capable of making lots of power.

Roush 745 @8500 RPM

Byrnes 650 @ 6300 RPM

It is a matter of what turbo and how much air flow and how much pressure and how much charge air cooling. But we already know that dont we. But these packages are really not street deals. I know 3 people that have done the Esslinger head, two on the street. Each is not at optimum but both will run faster than both guys want to. Dyno tests on one show around 400 usable horsepower. How much more do we want? How much more than that which a modified iron head can we afford.
Lets look at cost.

Iron head. MP big valve with porting and Nicks .500 cam.
Complete ready to run how much? I'm not sure, but guess $1200.00
Lower intake porting (not extrude hone) 150.00
Upper intake modification 100.00

TOTAL $1450.00

Both units require really good air charge cooling so we wont even talk about that. Both require head stud kits and should require upgrades to the block crank, rods etc etc, but again that is the same for both applications.


Esslinger head. Now dont get me wrong. If you want to make maximum HP this is the peice you need, and anyone going to it I will help all I can, but on the street, few people "need" it.

Head casting $1300.00
Cam Kit 800.00
Fabricated Intake Manif, machining and fabrication. 1500.00
Exh Manifold (you need this to flow more so you can use
the increased air flow capability of the head. 1000.00
Misc hardware 250.00

TOTAL $3850.00

Now, knowing that each of the guys that have done this swap have spent at least $5000 my figures are conservative to say the least. They are on the east coast so labour may be more.

Now If someone in the aftermarket were to design a 4 valve head, they would end up selling if for what the market would bear, and attempt to recoup the $ 100 K tooling cost, much less the cost of design, development, and testing to make sure the part is durable enough to withstand whatever we are going to do with it. I estimate that cost to be at least another 150 K, so here we are with .250 M invested and we cannot machine the thing. OK lets go out and buy a used CNC machining center. I'll just take a wild guess at $100 to 150 K depending on age and condition, but.....I cant sell junk now so I cannot go out and buy a worn out machine. If someone has the wherewithall to do this project, they want to make a perfect part. Now unless I design this thing to use the same valves that someone uses in production, (Ford did), I'll have to pay tooling costs, but hey we are aftermarked and we are good at applying someone elses technology to our deal. OK so we have lets call it $400,000.00 invested, and have not yet made a part. Part of our development consisted of models and flow bench work that will predict flow at valve lifts under negative pressures, but we have nothing to tell us what happens at boosted contitions. So we have no idea how the head is going to act on an engine until we make some parts and run them on dyno. Lets say we order 6 heads from the foundry. The first time a foundry pours a new part they waste a lot just learning where the air bubles will form and how the gas is vented and how the part cools and freezes, so my experience with the iron SVO block says we would probably loose 20 to 30 parts before having good usable castings with good wall thickness. This is a development process in its self. So lets say that the foundry wants $250.00 for each good casting, but you have to pay some way for all that scrap. I don't know how the cost is passed down, but I know it is, but lets say that you expect to sell 100 parts for the lifetime of the project. All of the scrap must be amortised over those 100 parts. My estimate is that the casting will end up at $500.00 each. So using that, lets walk thru the cost So investement amortization will be 400 K plus another 50 K for castings
so 450 K over 100 parts is $450.00 each
Casting $ 500.00
Machining 250.00
Valve seats If you want max power we need berelium/copper 500.00
Valves, Inconnel X750 Exh SS Intake 500.00
Camshafts: billet special grind 2000.00
Valve Springs Special small diameter High tech for high speed 200.00
Rocker Arms. Billet steel or cast steel
(production 4.6L rockers are stampings & lack the strength
for high lift high load applications) 1600.00
Cam cover Prototype fabricated Sheet metal 500.00
Pistons forged, to accomodate Pent roof chamber 400.00
Tooling amortization 450.00
Intake manifold 2 runners & 2 injectors per bore 2000.00
Exhaust manifold 2 runners & double the pipes 2000.00

We are now at a little more than $10,000.00 for the parts necessary to make this thing run. Both systems will require an aftermarket computer system, a completely re plumbed fuel system, the tallent and skills that few of us have to calibrate this little puppy. It will be capable of producing well over 1000 HP, but then we dont have a bottom end like Roush did to withstand the power and we will end up running over the crankshaft. (thats something I worry about at 650 HP) and I use a billet crank with billet steel main caps, but to support this power you need what is called a girdle. If you could do a girdle it would be another $2000 to $4000.