Click to Print

Technical Advices > Cylinder Heads

Cylinder Heads

5.1 Types of cylinder heads

There are both OEM type and aftermarket cylinder heads that can be used in combination with the various types of Fiat/A112 blocks. The head bolt pattern is the same, with the exception that some models used 10 x 1.25mm head bolts, while the majority used 9 x 1.25mm head bolts. It should be noted that Fiat is probably the ONLY car company to use 9 x 1.25mm hardware extensively.

OEM Heads

Fiat 600 and 600D - The early Fiat 600 head is distinquished by 8mm rocker stand studs and 6mm rocker arm adjusting screws. The later 600D heads had 10mm rocker arm stand studs and 7mm adjusters, as the early 6mm ones did not stand up well to the rigors of competition.

Photo – 600D head combustion chamber

Both 600 and 600D heads were used for most of the early Abarth coachbuilt automobiles. The head featured a deep “bathtub” type combustion chamber that required a piston with a “kidney” shaped dome on the piston to get the compression up. The valve sizes that would fit into the head were also limited, although at the time this was dictated by FIA rules and homologation papers. As engine displacements grew to 982cc, as for the 1000TC motor, this proved a real challenge.

Fiat 850 - The Fiat 850 head configuration was quite different. Still using a 2-valve side-by-side arrangement, the chamber was now much more open, with a good squish area to increase the combustion chamber turbulence as the piston came up to TDC. This same chamber configuration continued on and was eventually used for the A112 motor, with only minor modifications around the intake valve to assist flow.

850/A112 Combustion Chamber – Note that the chamber is not yet completely finished.

Abarth TCR and OTR - While considered and OEM head, the TCR and OTR heads did not look anything like the Fiat 600/600D or 850 heads. While the head bolt pattern is still the same, that is about where the similarity ends.

The general design of the head is a scaled down version of the head designed by Aurelia Lampredi for the Fiat 2300 saloon, and later adopted for the Abarth 2400 couple. It is still push rod operated, but with a hemispherical combustion chamber. Valves are operated by rocker arms, however intake and exhaust valves have separate rocker arm shafts. The TCR/OTR head is usually equipped with either dual Solex or Weber carburetors. The exhaust manifold is also different, as the port spacing and location does not coincide with either the Fiat 600/600D or Fiat 850 heads.

Aftermarket Heads

Vizza Motorsports

Vizza Motorsports Cylinder head

This head is a modified Fiat or Autobianchi standard head, and is sold by several people including Vizza Motorsports. The head has been welded up and then individual ports machined for each cylinder. As the ports enter the top of the head, the intake has to make a rather abrupt turn so that the dual Weber carburetors sit at the correct angle. I have also seen this head with two 40mm DCNF downdraft carburetors, and in this case the entry into the head is pretty much straight down. I have not had any experience with this head, but I would imagine the downdraft version would be more effective that the side draft version.


Astech Cylinder Head

This head was designed in the USA and produced for a number of years. It uses a log-runner intake manifold design for a single 40 or 45 Weber DCOE carburetor. The head is no longer in production. Given the design of the intake manifold, this type of installation would provide good torque and horsepower in the range from 4000-7000 RPM.


Designed by Paul Swenson, this head has been in constant production since 1968. There was one revision of the casting models in the 80s to slightly change the head so that it would work on the A112 blocks,which had a different bore spacing. Scuderia Topolino is pleased to carry on the production of this head. The chief characteristics of this head are improved squish area, a “high angle intake port and revised cooling layout.

Here are two implementations of the 8P head. The first is a more conventional version with twin 40DCOE Weber carburetors. The second is a racing implementation using 4 Keihin FCR single barrel carburetors. In this case the carburetors sit at a 50 degree angle and the approach to the intake valve is very direct. Even with the Weber carburetors the curvature of the intake runner is very gradual, which is the reason for its good performance.

5.2 Head Layout and Modification

Valve size - It is possible to put larger valves in 850, A112, PBS and Vizza heads. I am not sure about the Astech head, as I have not worked on one. The maximum valve size is 32mm intake and 28mm exhaust. Using these sizes does pose some interesting problems. First, the intake and exhaust seats must be “nested” in one another (the exhaust is cut into the intake) and this can significantly weaken the seat area of the head. Second, by using such a large intake valve, the edge of the valve is shrouded by the edge of the combustion chamber and does not unshroud itself until the valves is nearly completely open. This causes a disruption in the flow of the valve. Third, using valve this large means that they “nearly” touch. If anything goes remotely out of sync, the valves could end up touching and this could be disastrous. The only way to rectify this is to move both the intake and the exhaust valve in the head. This would mean installing off-center valve guides, welding up the seat area, and cutting new seat pockets. This is a large, and exacting, job and not for the faint of heart. At Scuderia Topolino we consider 31 and 27mm valves for intake and exhaust respectively, to be the maximum size that we feel comfortable installing. Even then we go to some extra effort, on fully race ported heads, to take the combustion chamber wall out as far as possible. The shrouding of the valve is thereby kept to a minimum.

Intake port short side radius – Probably the most important area of the any of the heads, more so for the Fiat/A112 heads than the PBS 8P, is how you treat the “short side radius”. As the intake charge enters the port, and subsequently flows around the valve head, the general idea is to induce a high helix swirl pattern. In fact there are probably several of these patterns being generated.

All too often we get carried away with making the hole in the head bigger without considering the short side radius. You want to leave the radius as large as possible, and make it smooth while removing the minimum of material. It is already a very short radius, so the last thing that you wish to do is lower the floor of the common chamber that feeds all four inlet valves. It would take a great deal of work, but it could be that there is advantage to be gained from actually raising the floor of the chamber slightly to that a larger short side radius is maintained.

Here is a good cross-section of a typical intake port. The one on the right has a larger physical port, but the one on the left will actually flow air more efficiently. The secret lies in the port floor and the short side radius. In the port on the left the floor is raised slightly, and the radius approaching the valve is larger and more rounded. The top of the port will still flow more air, but the ratio of flow between the top and the bottom is much smaller.

In the case of the PBS 8-P head, because of the 45 degree angle of the intake runner the short side radius is very well formed and the flow around the valve is almost equal.

The exhaust port on all of the heads, Fiat/A112/PBS is much the same . The exhaust port also has a very abrupt short side radius. However, because we are not simply relying on atmospheric pressure for getting the burnt charge out of the cylinder, the effect is much less pronounced. You still want to make sure that the flow radius is a smooth as possible, without any abrupt changes. For the most part the port at the exhaust manifold face is almost too big. Generally you want an exhaust valve that is between 83-86% of the intake valve. This makes the 27mm exhaust valve just right, as the actual valve seat diameter will be around 26.5mm. The remainder of the port is somewhat rectangular, ending at the exhaust manifold in a round shape that should be slightly smaller than the internal diameter of the primary tubes. This slight step will provide a small amount of anti-reversion and aid in exhaust tuning.

The Bowl – The next most important area is the shape of the bowl. This area can have a large impact on the swirl of the fuel charge as it enters the cylinder. Increased swirl generally means higher efficiency and increase flow. The disturbed characteristic of a swirling air mass also means that the fuel will stay in suspension better for improved combustion efficiency.

The bowl area also includes the portion of the valve guide that protrudes into the intake port. Some head experts like to remove this portion of the guide, while others like to leave a small amount of material there and use it as a small “air director” to get the air to assume a swirl attitude before it passes the valve seat.

The bowl on the exhaust port is no less important and should provide the best possible exit path. Due to cooling considerations, this may not always be optimum.

There is not set formula for how a bowl should look. It is one of those “black art” areas, and only a flow bench will partially answer the question of where to remove material. There are some things that we know work, but the last detail is often just a matter of trial and error.

Head seat configuration – If we can achieve a 83-86% intake/exhaust size ratio, then the next thing to consider is the seat angles in the head. The principal seating surface can be either 45 or 55 degrees.

Above you will find a typical intake port.

I then use a 45-55 degrees seat angle with a bottom cut to blend into the bowl area and a top cut to aid flow into the cylinder. If you go for a 55 degree seat, then you will have to run Beryllium or AMPCO copper seats, or you may have transfer of material from the seat to the valve surface, affecting the seal of the valve. For configurations where the intake valve may be shrouded this higher angle may be of some help, as it assists flow into the cylinder.

For the exhaust port I generally use a 45 degree seat, principally for reliability.

Valve seat and valve head configuration – This is a very complex interplay of components, theory, and practical experience. Each “expert” will have his or her ideas of what works.

The following link is a very good technical explanation of what you should try and achieve in modifying any head. While it is written around typical American production head work, much of it is directly applicable to Fiat heads.

Once you get on to this site, click on the Tech Articles link and go to the last item “Head Porting Principals”.

5.3 Valve Lift and Flow

One of the best engine technicians was Smokey Yunick of stock car racing fame. He concluded from air flow observations that he made that to open a valve more than 33% of the valve diameter may not be productive due to the additional parasitic loss generated. Therefore for a 31mm valve, the maximum valve lift should be 10.9mm.

I know from my own tests on the flow bench that the “incremental” gains in air flow over 11mm of lift are indeed small. However the real advantage to higher lifts lies in the fact that in order to achieve these lifts a more aggressive cam grind must be employed. In doing so we increase the “area under the lift curve” in general, and because the lobes on the can are more aggressive, we increase the “rate of lift”. This improves low lift flow.

Our biggest camshafts have 12.4mm of lift at the valve, and as long as low friction components are used in the valve train, and sliding friction losses are reduced to a minimum, additional benefit will result from the increased gross valve lift and rate of lift.

5.4 Head Gaskets

There are all types of head gaskets available and each has properties that are preferred for specific applications, but all depend on having an absolutely flat, clean and smooth head and block surface. This means a flat surface, at 90 degrees to the bore, with an RA (Average Surface Roughness) of between 14 and 20.

If you have a road car, with 10:1 compression, flat top pistons and a road camshaft, then any standard head gasket will be sufficient for your needs, PROVIDING you use the appropriate grade of fuel to prevent detonation. Note: You will read a great deal more about the subject of detonation in the technical sections and the FAQ section, as it is a little understood, and often neglected, area of engine tuning.

One step up would be to use the Spesso Competition head gasket. This is 1.6mm thick (uncompressed) and compresses to approx 1.2mm (0.048 inch). These head gaskets have special silicone beads around certain oil and water areas for improved sealing. In addition the “fire rings” are general made of stainless steel.

If you have a competition engine up to 12.5:1 compression, with a more aggressive camshaft with higher "dynamic compression" then you should consider the Spesso Competition or a Scuderia Topolino solid copper head gasket. The copper head gasket is 0.043 inch (1.09mm) thick. Care should be taken that the copper head gasket is properly softened (annealed) and coated with sealant on both sides before used.

The next step up is to use an Multi-Layer-Steel gasket, such as the one made for Scuderia Topolino by Cometic.

This gasket consists of a minimum of upper and lower stainless steel layers with a number of intermediate layers. In this way the gasket can be varied according the squish area requirements of the engine. The standard thickness for this gasket is 0.035 inch (0.9mm). The upper and lower surfaces are coated with a rubber material and not additional coatings or sealers are required.

Finally, if you have a very high compression motor (13.5:1 computed compression) with a very aggressive camshaft and high dynamic compression (10:1 or more), then all of the above options may eventually fail. In this instance you have an additional option.

The ultimate head gasket is a combination of two gaskets. It consists of a special MLS gasket with a separate 0.031 outside diameter stainless steel, pressurized o-ring, added as a pressure sealant, to contain the combustion pressures.

The "Helicoflex" o-ring is hermetically sealed during manufacture and has an internal gas pressure to maintain a positive contact against the block and head. As the ring heats up the pressure inside the ring increased, resulting in a higher pressure seal. The diagram above shows the combination of the two gaskets.

This combination of head gasket materials gives the best possible combination. The MLS head gasket acts as both a media sealer for oil and water, and a peripheral stop for the gas filled o-ring. The pressure filled o-ring acts as a super efficient pressure barrier for the combustion process. Another company that implemented this pressure ring technology is Coopers, who provided this technology for the Cosworth turbo-charged, FI motors.

If the piston is installed with "0" deck height, then the squish distance would only be 0.035 inch (.98mm). Given that connecting rods so stretch up to about 0.015 inch (.38mm), this would leave only 0.023 (.6mm) piston to head distance at high RPM. This would be the minimum piston to head distance for a 1050 motor, that I would recommend.

Postscript - Below are some photos of a head gasket that was destroyed by detonation. The detonation was caused by trying to run an engine with high dynamic compression(9.8:1) on 100 octane fuel. It probably would not have made any difference as to what type of head gasket was used, as even with a pressure ring the end result would have either been the same, or the next weakest link (probably the pistons) would have failed.

Three of the gasket fire rings have been deformed into a "teardrop" shape. This is a classic end result of detonation. In detonation the cylinder pressures would have spiked to over 3000 PSI. This amount of pressure would have lifted the head, releasing the clamping force on the head gasket and simply pushing it aside. The only reason the fourth cylinder did not suffer the same fate was that the car had already stopped running.

Specialist Services
Calendar and Race Reports