Like it or not as we roll into the second half of 2014 four-stroke engines comprise the lion’s share of the motocross landscape. The latest crop of four-stroke MX engines are better than most could ever imagine would be possible. Yet despite all the advances the basic primal gearhead urge in most of us still wants more. So after the obligatory overpriced pipe is purchased and installed, usually with less than satisfying results, a rider is often left with the question, “what really makes power in a four-stroke”?


In the heyday of the two-stroke porting was the key to unlocking the beast hiding within most engines. Porting on the venerable two-stroke is really about the Time/Area relationships of the ports to the engine rather than just improving airflow in a port. Modifying that same Time/Area relationship on our four-strokes is much more involved.


First a little background that will be obvious to some but maybe not others.

Torque is entirely dependent on cylinder filling. Basically how much air fuel mix can you jam in the cylinder during that short window while the intake valve is open? When you look at a dyno sheet the torque graph you are looking at is essentially a cylinder-filling (volumetric efficiency) graph. The torque peak will occur at the maximum cylinder filling point on a per stroke basis. Horsepower peak will occur at max cylinder filling on a time basis (rpm). Four strokes have relatively small valves in relationship to the bore area, and they have a very short time to open and close at high rpm. As a result the acceleration forces on valves can be quite high. Get them open as quick as possible and keep them open just long enough to keep things from backing up, then close as quick as possible. So we need valves small and light enough along with springs stiff enough to control valve float at high rpm, yet big enough to provide an adequate pathway for airflow. Simple right?
If you want to improve cylinder filling you have a number of choices:
- Tune the intake and exhaust waves (lengths and diameters) very sharply to a specific rpm. This works exceptionally well but only over a very narrow rpm range. Outside of that rpm range the wave tuning disrupts cylinder filling and causes a torque trough. Pipes, which can work such wonders on two-stroke engines are considerably less sensitive to exhaust pipe tuning on a four-stroke. So cams and intake length as well as area have a major influence on this. Intake length is determined by cam duration, sonic velocity and rpm target, so it’s a just one part of a bigger system.

- You can increase the TIME available to fill the cylinder by increasing the intake valve open (cam) duration, but it’s not a free lunch. The cylinder filling is a function of the pressure drop caused by piston motion. The piston motion only sucks (cylinder pressure lower than atmospheric) for 180 degrees. After that we have to consider that air is elastic; it can stretch and compress. Consequently, we must open the intake valve earlier and close it later at high rpms to gain additional time for filling the cylinder, which gives us intake cam duration greater than 180 degrees. If cam duration is excessive, we will experience flow reversion at both open and closing of the intake valve because the cylinder pressure is greater than the atmospheric pressure in the intake. Flow direction will go towards the area of lowest pressure. While extra intake duration works fairly well at high rpm, it leaves the valves open too long at lower rpm and again disrupts cylinder filling due to reversion and causes a torque trough as a result. The trick is to balance cam timing and port air speed to maximize cylinder filling but minimize back flow or reversion. At a given target engine speed, it is important to produce the maximum air speed through the ports and past the valve. When a port is too large, the air speed is reduced. When a port is too small, air speed is increased excessively to the point of choking turbulence. The valve to port area ratio is critical to achieving maximum flow at a given engine speed.

- We can increase the valve AREA with larger valves well but you quickly start as running into space and cooling and valve control problems. Larger valves are heavier and harder to control at high rpm. To keep from breaking parts the larger valves will have to be opened and closed more gradually, which limits the open time and tends to defeat the purpose. Larger valves also tend to be subject to shrouding from the cylinder walls which limits their potential flow. But AREA is the key here, specifically valve curtain area, not valve head area. When multiple valves are fitted the area increases considerably but the available valve curtain area does not since emerging air flow will collide with an adjacent valve. Yamaha's initial approach to this problem was to use the 5-valve design, which provides greater valve area at low rpms than a comparable 4-valve design. The smaller lighter valves that come with the 5-valve design also allow them to open the valves faster (most likely the cam profiles will be similar based on the valve mass and target rpms) which provides more flow area over a longer period of time.(Again we are limited by piston motion suction.) Four valve engines like the CRF tend to gain much of that advantage back at higher lifts and higher rpm, and done correctly will surpass the best 5 valve heads in mass flow for the same basic valve area.
You can't instantly open a valve in a four-stroke. The camshaft has an intricately designed opening and closing ramp to ease the valve off its seat and back on again. This cycle has to be repeated hundreds of thousands of times for each hour of running. So as a cam designer you have to balance opening the valve as quickly as possible with high valve acceleration rates versus long term reliability. Keep in mind that until the valve is well off of it's seat, it's mostly just a hindrance to flow, but even a few thousand of lift will provide gas flow if the pressure is great enough as in an exhaust valve. On the intake side there is very little flow past the intake valve opening point because the piston is going the wrong way. A valve leakage of just a few thousands during combustion will turn a good engine into a boat anchor.
Now here’s where the porting issue comes into play. The trickest port in the world can't flow worth a lick till the valve is up off its seat and out of the way. The port area nearest to the valve seat has the greatest impact on how well the port flows especially at low rpms. The shape of the valve head, the shape of the valve seat , the throat port area behind the valve head and the valve exit diffuser shape are critical here. The fluid dynamics concept of attached flow comes into play here.

To expose the maximum valve area for the longest period of time designers try to accelerate the valve off it's seat as quickly as possible. The intake valve can open slowly since the piston motion would produce reversed flow, but it must close quickly in order to prevent more reversed flow. The exhaust valve must open late but rapidly in order to take advantage of cylinder pressure during the initial blowdown phase when cylinder pressure is high. But you can only accelerate the valve so fast before you can no longer control it's mass. If you can't control the valve acceleration reliability suffers in a hurry, as modern four-stroke owners know all too well.

I realize this is spinning off topic fairly quickly, but suffice it to say to you can only really get all these factors to line up over a fairly narrow rpm range. Airflow and valve open time (actually, time curtain area) requirements have a square root relationship so balancing flow and cam timing is critical to building a balanced package. Ports that flow lots of air have different cam requirementsthen ports that flow less but airflow, port velocity, cylinder filling and peak torque range are all intimately tied together. The relationships aren't always obvious or intuitive. Someone once said anything that really matters is invisible.

Porting is a matter making some hard choices between mass flow and port velocity for the available cam and the type/amount of power you want to make. In addition, camshaft indexing to piston motion plays a large part in all this.


Small high velocity ports can make great torque over a fairly wide rpm range but ultimately they will choke mass flow and limit peak power. Larger ports will tend to show greater mass flow numbers and higher peak power but the trade off is generally a lazier port at lower rpm and a subsequent loss of some torque at lower rpm, and possibly a narrower torque range (aka powerband).

Keep in mind this is a HUGE generalization, but hopefully it illustrates the balancing act between cam timing, port flow and rpm.

For a lot of weekend riders looking for a rideable package that delivers a broad spread of power I feel the trade off of some mass flow at high rpm is worthwhile to obtain a wider torque range with a smaller higher velocity port. For someone else with different needs we might go a different route. A lot of engine builders will go with a larger port area and a higher mass flow to make more peak power, but they will usually trade off a lot of low to mid range torque as a result. This tends to make the bike harder to ride and tougher to keep hooked up for everyone but the most skilled riders. One approach isn’t necessarily superior to the other; it’s more a matter of trying to give the rider the package that best matches his needs and requirements. For motocross, low and midrange torque is critical including rapid throttle response. For road racing, mid and high rpm range becomes more useful. This is where data acquisition equipment really some in handy. Being able to record lap data and then come back and check your data collection rpm histogram to see where the engine rpm time is really spent can really help you zero in on where the engine needs to make power for the application. When it comes to porting and four-stroke engine building, one size never fits all. That's a big part of why the shops with the big ads in the magazines will never be able to take the job of real engine builders like Eric Gorr or Ron Hamp. . Those glossy ad shops try to build simple production line engines and sell porting done by low priced unskilled labor, but four-strokes are far too complex with too many closely tied interactions for that approach to work very well. That may have worked in the two-stroke days, but not anymore. ( Truthfully it didn’t really work then either.)


Hopefully buried somewhere in all this run on ASCII you’ve gotten a slightly better understanding of four-stroke porting and basic airflow.