Valve flow, an intro

By Sean Kelly

In modern internal combustion engines, poppet-style vales are used on both the intake and exhaust ports. The number of valves, angle relative to head face, size, material, and face angles are all factors that contribute to the flow of the port. Forced induction motors, in particular, face additional realms of complication associated with port flow. In addition to the direct effect of valve design on port flow, many other discrete factors play into the efficiency of the port. These include: port volume, heat transfer rates, bowl shape, seat angles, valve stem diameter, surface quality, fueling efficiencies, cylinder head temperature, port matching, shroud area, and numerous ratios involving all these factors and the physical characteristics of the combustion chamber. To analyze every one of these in detail would be a monstrous task, and not the intent of the following prose. Here we will begin to analyze the physical build of the valve and its effect on the flow of the port.

An engine is not a static machine; it has dynamic volume, pressure, temperature, turbulence, vale lift, piston speeds and charge temperatures. To get a workable system, it is easiest to design all of these factors towards one operating point, a datum of operating conditions that corresponds with the desired torque needs. Mechanically, the highest point of efficiency loss in an engine is the valves. Maximizing this point in the machine can lead to great gains in overall engine efficiency. Poppet valves are used because of the low cost of manufacturing, good heat handling, good sealing properties and simple design, which is easily exploited. Generally, the inlet port on a motor is holds 44-48% of the cylinder bore and is circular, or close to it. The size of the vales corresponds to this area, and they should be sized to allow the flow needed for the power desired by the motor, and no larger. Valves that are too small choke the motor and reduce air flow, which reduces the fuelling, which in turn reduces the torque output of the motor. The primary enemy of valve design is the mechanical delay in the valve-train to reach full flow, which occurs at full lift.

Lift is categorized in three steps. Low lift is designated when the flow over the valve is attached to the valve seat and valve head in the form of a surface jet. In medium lift, the flow begins to lift off of these surfaces and in high lift the flow forms a free jet off of the valve and seat surfaces. Specifics of these conditions affect the turbulence and flow characteristics around the valve in response to the mechanical design of the port, bowl, seat, valve head and shroud.

Intake valve flow:

To analyze the flow of the intake valve in and of itself, you consider the valve's curtain area and discharge coefficient at a particular lift value. These factors combine in the following equation:



The coefficient of discharge must be found experimentally, or estimated using a finite analysis effort. This factor decreases with lift due to the separation of the flow jets from the seat and valve surfaces, which occupies, effectively, less cross-sectional area of the valve curtain. A common Cd is 0.6. Once the Cd and valve lift values are determined, the effective flow area can be found as a function of crank angle.

A rarefaction wave travels upstream from the intake valve when the valve is opened. Upon interacting with a change in cross-sectional area, such as a surge tank or the end of an intake runner, a compression wave is sent back down the intake path to the valve. This is called the "ram-air effect", and can be identified as an increase in the local density at the intake valve.

By experiment, it was found that timing this compression wave with a crank angle of pi/2 (90 degrees) can result in a substantial increase in volumetric efficiency. This corresponds with the maximum piston speed; and the timing of this wave can be analyzed by examining the length of the path that the wave travels.

Velocity is distance over time, and the length of the waves travel is twice the intake length (2L). The time is a function of engine speed (N). Solving for length, you arrive at:


where c is the speed of sound. This factor is a product of temperature (Kelvin), ideal gas constant R, and specific heat ratio k, which is Cp/Cv, typically referred to as gamma. The ideal gas constant is derived from the universal gas constant

p * v = R * T And if we have a constant pressure process, then: p * deltav = R * deltaT

The speed of sound is: c = (k*R*T)(1/2)

Exhaust valve flow

The exhaust cycle is divided into two steps for simplification of analysis. First is blowdown, which is modeled as a constant-volume process; next is the exhaust stroke, which is modeled as a constant-pressure process. When the exhaust valve opens, the cylinder pressures are high enough to induce sonic flow across the valve. Hence, the blowdown process is rapid. During the exhaust stroke, the piston does work on the exiting gasses and, pending conditions in the exhaust manifold, either even the pressure across the valve or produce a pressure wave. The events of this process can be analyzed further pending exhaust system design.

A rarefaction wave similar to the one described for the intake track occurs in the exhaust manifold as well. Assuming the system is not a constant-pressure turbocharging application, tuning the exhaust runner length is similar to the intake process, except that the ideal value for piston location when the wave returns is 120 crank degrees. At this point, scavenging potential is at a maximum. The length of the pipe can be found with:


the temperature readings should be taken at the exhaust port, before heat transfer and changes in volume can affect the reading.

Valve timing

The intake valve needs to be open before the top dead center of the piston position to allow time for it to reach maximum lift. As stated before the key downfall in valve efficiency is the dead time spent at lower lifts. Opening the valve early not only helps eliminate some of this dead time, but also allows for some interaction between the intake and exhaust valve flows. Opening the valve too early will do several things: first, during the upstroke, exhaust gasses are leaving at a relatively low pressure compared to compression pressures, opening the valve may induce backflow of burnt gasses and disturb the pressure differential across the exhaust valve, reducing the efficiency of that process. Secondly, allowing the pressure differential across the intake valve to change too early can limit the flow across the valve at higher lift. Lastly, if the exhaust pressure is too high, burnt gasses can enter the intake manifold, which can cause carbon buildup, increase inlet temperatures, and displace some volume of fueled mixture entering the combustion chamber.

The exhaust valve opens soon before bottom dead center to allow time for the exhaust vale to open, increasing the effective area to allow more time for the products of combustion to escape. The engine speed effects how early the exhaust valve is opened. As piston speed increases (engine speed increases), the valve opens later.

At the end of the exhaust stroke there can be a period of vale overlap, at which time both intake and exhaust valves are open. During this time, inertia of the burnt end gasses leaving the combustion chamber can help cylinder filling by inducing more flow across the intake valve, and, as both intake and exhaust valves are nearly closed, high pressure differentials can produce sonic flow across either valve. If the overlap is not carefully considered, reversion of burnt end-gasses can occur, which will displace the incoming fueled mixture, raise temperatures, and reduce power. Too much overlap can also result in some of the unburnt mixture escaping into the exhaust manifold, which will increase emissions and reduce the volume of fueled mixture in the combustion chamber, which reduces torque output, among other detriments occurring from the fueled mixture entering the exhaust system.