Suck, Squish, Bang, Blow: Part 4

On this installment, we will peek into the inner reaches of modified cylinder heads to find out just what is involved in modifying one. There is a lot of work to be done and many choices you can make. Part four of our series is intended to help you understand some of the more common tricks so you can make the appropriate decisions when it comes to modifying your head for maximum power.

Achieving good airflow through the cylinder head is essential for creating big horsepower. How does increasing airflow get you more power? To start, an engine is basically an air pump. To create the most power possible, it is important to get the maximum amount of fresh air into and burnt exhaust out of the engine with the least amount of mechanical effort. Excess intake and exhaust port restriction makes more work for the engine--work that is not used to turn the wheels. This extra work is called pumping loss.

Imagine sucking a thick milkshake through a small straw. This takes a lot of sucking effort on your part. Now drink that same milkshake through a larger straw; it's easier isn't it? The extra effort you are expending to drink that milkshake through the smaller straw is a pumping loss. The more work it takes to get the various gases into and out of your engine, the less power will be available to spin the crank and the wheels.

Another major factor is volumetric efficiency. Volumetric efficiency is the percent of an engine's displacement that is filled on each intake stroke. To make things easy to understand, let's imagine that you have an engine that displaces 1000 cc. If the engine can take in 800 cc of air on the intake stroke, the engine has 80-percent volumetric efficiency. Eighty-percent volumetric efficiency is typical for a modern, production-based, naturally aspirated engine.

The factory, especially with modern import engines, usually calculates port size and configuration carefully to optimize volume and geometry in stock configuration. However, when an engine is modified for more power, the flow demands of the engine also change. Air intakes, headers, exhaust systems and cams that promote higher rpm operation also demand more airflow. More flow often requires bigger ports with more cross-sectional area. Imagine drinking that thick milkshake through a straw twice the diameter. It's much easier, right? That is the main effect of porting; reducing pumping losses and increasing volumetric efficiency by reducing restriction.

Sometimes the design of the head and the port layout itself is compromised to make the head fit for a lower hood height or even bending around head bolts, water passages or pushrods. It is only recently that engine designers, specifically the designers of high-output import engines designed the head first around the ports for best flow and good volumetric efficiency. Domestic engines seem to place the importance of the location for almost every component before the ports. The designers of the Big Three seem to place the ports wherever there is room in the casting, considering their position last in the design process, long after all the other parts of the head have been located. Partly this is because many of these engines were designed a long time ago and the manufacturers are trying to milk the life out of the old, tired designs in order to maximize profits.

In cylinder head porting, generally, the intake and the exhaust ports of the head are carefully reshaped by hand. The ports are enlarged, straightened and streamlined to reduce any pumping-loss induced restriction and to reduce turbulence to increase flow velocity as much as possible.

For the most part, the ports are straightened out using a die grinder and a carbide bit to a line of sight configuration. Straightening gets rid of turbulence-producing bends in the port. Like almost everything in life, there are a few exceptions to this rule, but for the most part this is a good generalization. Porting involves extensive hand finishing with the aforementioned die grinders to remove tooling cuts, sand casting pits, and the usual lumps and bumps made by the mass-production tooling of the factory. Sometimes even CNC machines are used to rough port heads for popular models of engine. Even though it is highly beneficial for power production, heads are usually not ported by the factory, as it requires too much labor-intensive handwork by a highly skilled artisan, which makes it very expensive.

Porting, as with most high-performance modifications, has its limitations. It is possible to make the ports too big. An amateur cylinder head tuner armed with an air-powered die grinder can simply hog the ports out, making them as big as the Holland tunnel. Big ports can flow big numbers, but big flow numbers alone will not make big horsepower. Larger ports have a lower flow velocity given the same flow demand. An air column of a given mass at a lower velocity has less inertia, and less potential energy, possibly negating any ram charging effect. The ram effect is critical for obtaining complete cylinder filling at low rpm and is just as critical for maximizing volumetric efficiency at high rpm.

Incomplete cylinder filling at low rpm causes an engine to have poor bottom-end power and throttle response. Symptoms of over-porting are an engine with a soggy bottom-end or an engine that only makes horsepower in a narrow, few-hundred rpm band at the top-end, accompanied by a rough, lumpy idle.

In engines with carburetors or throttle-body fuel injection, oversize ports with low velocity and stagnant flow can cause poor fuel atomization with its attendant bogs and stumbles. Typically, an engine with overly big ports, a high-performance camshaft and a carburetor will barely run at low rpm. Going too big in the ports can also mechanically weaken your head to the point where it flexes, blowing head gaskets frequently or even cracking.

The main trick to effective head porting is making the "straw" big enough to feed your thirst, but not making it so big that you can't suck hard enough to bring the milkshake to your mouth! Truly effective porting is an arcane combination of craftsmanship, artwork and science. There are no hard and fast engineering rules that can be applied to all cylinder heads; every type of cylinder head will respond a little differently.

To truly figure out the optimal pattern for porting a particular cylinder head, a head porter must spend many hours on a flowbench to find out what tricks will eke out the best flow numbers from a head, mostly by trial and error. A flowbench is a machine that can measure a head's airflow capabilities. A good cylinder head specialist tries to shape the port to get the maximum flow with the minimal amount of enlargement, keeping the velocity a main priority. They will also try to get equal flow with each port so each cylinder will get the same charge of air. Most good cylinder head tuners have their own closely guarded shaping secrets for finding the magical diametric combination of low port volume, high-flow velocity and high overall flow. In NASCAR, CART, NHRA and F-1 competition, port shape is one of a team's most highly guarded secrets.

As stated in the previous paragraphs, line of sight porting almost always works better than stock; however, it is not always the best possible way to port a head. There are some exceptions to the line of sight rule that are counter intuitive and must be proven on a flow bench or dyno because they don't always work. Engines with big bores and short strokes with valves located close together often prefer to have the port shape biased toward the cylinder walls. This prevents the intake charge from blowing right out the exhaust on overlap.

Engines with a shallow included valve angle often want the port floor to have a hump near the inside radius of the valve seat to help point the airflow toward the bottom of the cylinder so the incoming charge won't get blown out the exhaust. Heads with valves near the periphery of the cylinder often want the ports biased toward the center of the bore to reduce the cylinder walls shrouding of the airflow out of the ports. These strategies cannot be predicted by using magic engineering formulas or rules of thumb. This is where dyno testing and hours on the flow bench will separate the men from the boys. This is where art can surpass science.

Another important aspect of headwork is the valve job. Believe it or not, more than 50 percent of the flow gains generated by good headwork can be found in the valve job. Factory valve jobs often involve only cutting the seat i.e., the 45-degree surface to which the valves seal. Sometimes there are additional rough cuts made to the port to ease the air's approach angle to the valve.

This is why factory valve jobs are usually either called one or two angle, one angle being just the seat surface, two angle being a seat surface plus a smoothing throat cut. In a cost-conscious, mass-production environment, there is simply no time or money to spend on small details like multi-angle, precision valve jobs. If more flow is deemed necessary, it is usually cheaper for engineers to spec a slightly larger valve. Wide seating surfaces are more forgiving to mass production mismatches, giving good sealing for the life of the car even with a significant misalignment of the seating surfaces. As valves and seats wear, they tend to sink lower into the seats creating seating surface mismatches and valve shrouding (partial blockage or restriction caused by objects in close proximity of the valve seat; imagine having the valve having to lift itself out of a deep crater before it can start to flow). Wide seating surfaces tend to last a little longer when the valves sink under wear.

High-performance valve jobs have three-angled cuts, with one machined surface on each side of the valve seat cut. First there is the throat cut which is typically around 60 to 70 degrees. This helps ease the air's transition to the 45-degree seat cut. The second cut is the 45-degree seat cut, which is the surface that the valve actually seals against. On a high-performance valve job, the seat cut is narrowed down to the minimum needed to seal the valve reliably and the seat's contact patch is matched accurately to the corresponding seat cut on the valve.

Due to wide production tolerances on a mass-production engine, the seat cut ends up being much wider than actually needed. For a high-performance valve job on a four-valve motor, the seat cut is usually made to be 0.040 inches or so on the intake side and 0.050 inches on the exhaust valve. The exhaust is made slightly wider to have more contact area so the exhaust valve can be cooled better by conducting heat into the head. With the valve seat width at a minimum and the seats matched to the valve, the incoming air or outgoing exhaust has less restriction due to the unshrouding effect the adjacent cuts produce. Simply put, as the valve moves away from the seat, more area for flow is exposed sooner. The third and final cut is called the top cut. The top cut is generally about 20 to 30 degrees and is made immediately after the seat. The top cut also helps reduce valve shrouding of the airflow past the valve (or before as in the case of the exhaust valve) as the valve starts to lift off of the seat.

Sometimes a head tuner will make a five-angle valve job, adding additional, shallower angled cuts for the entrance and exit to make the airflow path to the seating surface even smoother.

Some head tuners feel the very best valve jobs are radius valve jobs. A radius valve job is a five-angle valve job where the two angles adjacent to the valve-seating surface are hand blended together into the port wall and combustion chamber for a totally smooth transition from seat to port to combustion chamber. These valve jobs are very labor intensive and hence expensive. Some head artisans firmly believe (and have flow bench proof also) that this blending actually hurts flow, another example of art meeting science. Another trick to increasing flow is to have an approximately 30-degree backcut on the valves away from the seating surface. This also aids in unshrouding the valve to airflow as the valve starts to open. This is especially important at low lifts.

Another trick is to reduce the diameter of the valve stem directly under the head of the valve. The valve's stem is turned down a few millimeters, effectively shrinking the stem so it blocks less of the port. One should be careful not to overdo this, as overzealous grinding can weaken the valve stem. The head of the valve can be swirl polished to further aid flow. The swirl pattern reduces the thickness of the boundary air layer on the valve, reducing surface drag much like the dimples on a golf ball.

Airflow at low-valve lift is critical in making lots of horsepower because the valves spend more dwell time at low lifts and near their opening and closing points than at full lift. In simple terms, the valve is only fully open once per intake or exhaust cycle, but is near its seat twice, since the valve has to both open and close. That is the primary reason why a simple valve job can make so much of a difference in flow and horsepower.

Unlike most high-performance mods where bottom-end performance is robbed to get better high-end power, valve jobs fall into the free horsepower category. A high-performance valve job can increase power throughout the powerband with no sacrifices or drawbacks of any sort except cost and perhaps long-term durability, as the narrowed seats can tend to wear a little faster. Street high-performance valve jobs usually have a little wider seat cut for longer life, although some head tuners dispute that narrow seats wear more quickly. Modern DOHC multi-valve engines usually have hardened valve seats for unleaded gas use and run lower valve spring seat pressures than old-school big V8s, so valve seat wear is lower and seat width is not as critical for long-term durability anyway.

Sometimes larger valves can be installed in an engine to increase flow. However, the addition of larger valves should be checked on the flowbench when doing head R&D because larger valves can sometimes show a lesser flow due to an increase in valve shrouding. Heads with valves close to the periphery of the cylinder or engines with deep combustion chambers and small bores are prone to situation. This is because a larger valve can put the edge of the valve in close proximity to the combustion chamber wall, blocking the flow of air from the port or exhaust flow out of the port.

Good tools are important to make a good valve job. Many head specialists use a drill-like fixture with stones of differing conicity to grind the various valve seat angles. A good valve job can be done like this, but it can be labor intensive and difficult to do an equal and consistent job from valve to valve. One sign of a good head specialist is the use of a Serdi Valve grinding machine. A Serdi is an expensive ($50,000) machine that makes precise, repeatable multi-angle valve jobs in a snap. Serdi machines pilot off of the valve guide hole with a floating power head to ensure precise alignment of the cut seat with the valve's axis. The depth of the angle cuts is also easily controlled. A Serdi machine makes the creation of a good valve job easy.

The exact valve angles used by cylinder porters, like the geometry of the ports, are a closely guarded secret of many head artisans and racing teams. The angles and techniques listed here will vary from one cylinder head specialist to another. To successfully modify the cylinder head on many late-model (1996 and later), OBDII-equipped engines, extremes in port size must be avoided, as huge ports and valves can cause such a significant reduction in intake flow velocity. This can cause backflow in the ports to occur more easily during the overlap period of the four-stroke cycle. This backflow, or reversion, can cause several different problems. Reversion's symptoms are more pronounced at idle and low rpm due to the already low port velocities under these conditions. Severe reversion can play havoc with airflow meter readings. If the airflow meter voltage fluctuates out of the window of operation allowed by OBDII, a failure code will be triggered. Reversion caused by big, low-velocity ports can make the loping idle of a high-performance camshaft more pronounced. This can cause irregular readings from the crank angle sensor. The OBDII system will see this for what it is-a misfire-and will trigger a failure code.

A loping idle can also cause an engine to run too rich. This can cause the O2 sensor to read at too high of a voltage at idle which will also trigger a failure code. Stored error codes can be a problem when you go to get an emissions test, depending on how totalitarian the emissions laws are in your state. In some areas, a stored code means you fail the test no matter how clean your car runs!

To correctly modify the head of an OBDII-equipped engine, close attention must be applied to keep the overall port volume small. No huge, tunnel-like ports here. Since simply increasing the overall diameter of the port is out, careful re-contouring of the port to improve flow without sacrificing velocity is essential. More flow can be picked up through the careful re-cutting of the valves and valve seats without risking error code triggering.

Some head specialists also modify the quench zones of the cylinder head's combustion chamber. The quench zones are the flat areas of the cylinder head where the piston comes in close proximity to TDC. Pentroof DOHC cylinder heads typically have four quench zones at the ends of the combustion chamber. Quench zones promote more complete burning and reduce the likeliness of detonation by increasing turbulence of the fuel air mixture as the piston comes to TDC by squishing the fuel air mixture toward the sparkplug and away from the end zones of the combustion chamber. This reduces the amount of fuel-air mixture near the ends of the combustion chamber where it does not completely burn (thus being wasted) by pushing or squishing it toward the centrally located sparkplug where it can easily be ignited. When heads have additional quench area, they normally need less timing advance to make power. Thus a skillful tuner can tune the engine to be further from the detonation threshold, making the engine more reliable.

The quench zones can be welded, milled and reshaped by hand to make them bigger, shaping the combustion chamber like a cloverleaf instead of the stock pentroof rectangle. This reduces the combustion chamber volume, increasing compression as well as making the quench zone more effective. This can also make the combustion chamber less likely to promote engine-damaging detonation because the turbulent air/fuel mixture squished by the bigger quench zones burns completely and smoothly. AutoPowerDesign's Stage 6 modification is a good example of a modified quench zone. In fact, AutoPowerDesign pioneered this technique for modifying four valve heads.

The power gains produced by headwork can be significant. Late model import four cylinder DOHC four-valve engines as found on Acuras, Hondas, Nissans, Toyotas and Mitsubishis have pretty good modern heads to begin with and usually only experience 10-20 hp of power gain with even extensive headwork. Domestic models can sometimes experience up to 50 hp gains and older pushrod V8s can get over 100 more hp with good headwork due to their poorly engineered port layouts and rough manufacturing techniques.

We hope this primer has done much to explain to you what your head expert is or should be doing with your hard-earned money. It is an attempt to explain some of the more general and common techniques used to modify cylinder heads. Headwork is a step that only a few hardcore enthusiasts venture into, but if you are on a mission to have an edge on the majority, it is a critical step you must make in the trail toward serious power.

While your head guru may agree or disagree with some of the specs listed here, the specs outlined will give you some guidelines to let you know what your head guy will be doing so you can ask the appropriate questions or judge good work from bad. If in doubt, ask for a flow chart from similar work or ask for references from satisfied customers. Although most head artisans are usually reluctant to give you the exact specs, they are usually more than happy to show you the flow bench results of their work.

Special thanks to BCE Racing heads (909) 944-8097 for supplying photographic samples.