Fundamentals of Intake System Design

I have bad news for you. It takes HP to suck the combustion air into the engine. This HP is subtracted from what you get out of the e-shaft. That HP also burns fuel reducing the MPG you are going to get. The intake system is an aerodynamic device just like any other aerodynamic device such as an airplane. The difference is it is inside out. The air flows through a pipe rather than around an object. All the same rules apply. You want to reduce the surface area the air is applied to as it could be going up to 400 MPH and skin friction causes drag. Any bends or sharp right angle turns the air is forced to make reduce the amount of air going into the engine. It takes energy (pressure) to overcome turbulence caused by bends, ridges, restrictions, etc, in the air path.

Rotary engine side ports are a good example of sharp right angle turns. Since there are at least four side ports in a two rotor engine you can see the problem. Peripheral port engines have no sharp right angle turns and there are only two runners. Four, 1.5 inch outside diameter runners, have 50% more surface area and skin friction drag than two, two inch outside diameter runners if my arithmetic is right. Probably more as the air velocity could be higher. The drag inside the pipe is proportional to the air velocity squared. This increases the pumping losses and fuel burn and decreases the HP out of the eccentric shaft.

Mazda spent a huge amount of money back in the early 1980's experimenting with different size intake pipes to determine the maximum amount of air they could get into the engine and therefore the maximum amount of HP they could get at any RPM. What they came up with was a single pipe per rotor about two inches in outside diameter and about a one and 7/8th inch inside diameter. The hole in the rotor housing was about two inch OD. It was called a P-port for peripheral intake port engine. Up until the RX8 engine in 2004 all Mazda rotaries had peripheral exhaust ports. The RX8 engine has both side intake and side exhaust ports done for emissions reasons. Not the most efficient configuration at aircraft power levels. A p-port intake system is easier to build as there are no flanges bolted to the side of the block. Recently we have discovered that stainless steel p-port tubes can be merely TIG welded to the rotor housing steel liner. The article about it is on this web site. See for a how-to. Use the slide throttle design on the web site for EFI or install a small plenum and a Holley 500 CFM carburetor for rock bottom cost. See

The high power RX8 six port/runner engine is even more of a nightmare for our purposes. Mazda had to do it to get more torque at 2000 and 3000 RPM for car use. High torque at low and medium RPMs is useless for a geared aircraft engine. The four port/runner low power RX8 engine is worse as the breathing is becoming limited over about 6000 to 7000 RPM. Mazda has done away with this engine starting in 2007.

Unlike a car engine the rotary aircraft engine runs at a high continuous RPM therefore the intake design is entirely different.

Don't get hung up on low RPM. Forget everything you know about piston engine RPM. The rotary is a different animal. It is more akin to an electric motor than a piston engine. There are no parts that jerk back and fourth, fatigue and crack. Forget the 2.17:1 planetary gear box. It rotates the prop in the wrong direction anyway. It is ancient history. Go for the 2.8:1 gear box or even a 3:1 belt drive. Also go for a constant speed prop if you want a lot of MPG as you can slow the engine to 4500 or 5000 RPM cutting engine friction and improving the BSFC.

Also you want to decrease the velocity of the air flowing through the intake pipes to decrease the skin friction drag. The organ pipe tuning effect takes advantage of a resonant system and the inertia of air. Any resonant device will keep vibrating with little input of additional energy. Think back to the days when you were a child and your friend was pushing you on a swing. As you swung back and forth your friend had to give you a little push now and then to keep you swinging. No matter how hard he pushed you always swung back and forth in the same amount of time. You just went higher. You were part of a classic resonant system. Pendulum clocks do the same thing and were invented hundreds of years ago.

The same thing happens in an engine intake system. At a given length and diameter of intake pipe there will be one RPM where the energy required to keep the air flowing back and forth is a minimum. That will be where you find the torque peak. In other words the best volumetric efficiency. VE for short. VE is defined as a ratio between the volume of air getting in the engine divided by the displacement of the engine. When these two numbers are equal you have 100% VE. Most side port rotaries are about 100%. A p-port rotary VE is about 120%. An efficient engine is an easy breathing engine. Surprisingly the diameter of the pipe also has a small effect on the resonant frequency. I am not sure why but I think it has to do with the average speed of the air flowing through the pipe.

Here is some data from the Mazda 4 rotor Lemans winning engine. The torque readings are half the Lemans engine readings as we are dealing with a 2 rotor here and not a 4 rotor. The lowest fuel burn per HP generated always occurs at the torque peak as that is where the engine takes in the most amount of air for each revolution of the engine. The VE is optimum. In other words it is the most efficient operating point for the engine. The throttle is wide open so there is minimum aerodynamic drag from the throttle.

Some throttle types are lower drag than other throttle types. The slide throttle is the lowest drag. Some modern carburetors use slide throttles. Butterfly throttles are used on older carburetors and give less HP.

Air planes fly on HP and not torque. HP is a function of torque times RPM divided by a constant so the HP can increase despite a drop in torque. The prop torque is whatever the gearbox gear ratio deems it to be. With a 2.85:1 gear ratio the torque at the prop will be the engine torque times 2.85 or around 600 foot pounds for a p-port engine. This is the HP curve of the 4 rotor Lemans engine. Divide by 2 for a 2 rotor of course.

This engine had a trombone intake system were the length of the intake pipe could be continuously changed as the engine accelerated up through the RPM range. This HP curve reflects that fact. This is unnecessary for our purposes as the aircraft engine runs 99% of the time at one RPM. Pick an RPM where you want to maximise the torque and minimise the fuel burn. Then pick a pipe length that gives you peak torque at that RPM. The complication of a trombone intake system is unnecessary for the vast majority of aircraft applications.

What exactly is going on inside the intake pipe in terms of pressure waves? Mazda instrumented a p-port engine and measured the intake manifold pressure near the port as a function of e-shaft angle. Here is their data.

We did something similar using modern technology. We used a low cost HyTek data acquisition system costing $85 and two FreeScale low cost pressure sensors costing $10 each to sample the data 1000 times per second which was the response time of the low cost pressure sensors.

We measured the pressure in the intake runners on the back side of the slide throttle as a function of e-shaft RPM for both rotors simultaneously.

The vast quantity of data was sent to a notebook computer using the USB port and then plotted with a spread sheet program.

Since Mazda optimized the design of the p-port nobody has obtained more HP at any given RPM than Mazda. Fortunately Mazda published the data in an SAE paper number 920309. You can down load this paper in our tech paper download section.

Mistral Ltd spent a lot of dyno time and money on the side port intake and they got only 190 HP at 6000 rpm or so. No getting around it. Putting it another way the max torque on the side port engine is about 160 foot pounds maximum no matter what you do. It is, after all, only a 160 cubic inch engine. Since HP is Torque times RPM this limited the HP. Airplanes fly on HP and not torque.

People erroneously think the RX8 engine is a magic 240 HP engine all by itself regardless of the intake. It WILL generate 240 HP at 8500 RPM with the stock intake system. That is 28 HP per 1000 RPM. So 7500 RPM is 212 HP and 6500 RPM is 184 HP. Unfortunately the stock intake system is too high to fit in most airplanes. The high end HP is sacrificed for low RPM torque. With a p-port the engine would easily generate 300 HP at its current red line of 9000 RPM.

Since 7500 RPM is the max RPM you are going to see due to prop tip velocity limitations with a 2.85:1 gear box and a reasonable diameter prop 212 HP is about all your going to get with the stock or perfect custom intake system using the side ports. Unfortunately the stock system is too bulky and too high so people have to cut it down to fit under the typical cowl. If you do cut the length of the runners down all bets are off. Most attempts result in well under 200 HP. You just wind up with too many warts in the intake system. Badly done it will be about 150 HP.

The p-port on the other hand is an easy 215 HP at 7500 RPM no matter how you mess up the simple runners and the throttles. 200 foot pounds of torque is no problem. Moving the torque peak up by shortening the length of the runners results in a lot more HP at the higher RPMs. 300 HP at 9500 RPM with 2 inch diameter runners and a slide throttle is no problem at all. We demonstrated 240 HP at 6500 RPM on the dyno using two inch OD runners 24 inches long on a 13B 2nd gen engine and with a Weber carburetor no less.

Ironically there is no BSFC advantage with the side port. Probably on the contrary as the pumping losses are higher with the side ports and multiple runners. The best BSFC always occurs at torque peak RPM values. In other words that is the best cruise RPM for good MPG.

Why a slide throttle? As I said above the intake systems is an aerodynamic device. Anything you put in the air stream like a butterfly valve or venturi is going to cut the amount of air getting in the engine. Less air less HP. When the slide throttle is wide open, which is 99% of the time over 8500 feet, the p-port intake system is perfectly streamlined. Here is a Porsche animation of what hppens in an intake runner with a butterfly valve.

.....Porche anaimation of butterfly valve in intake runner. Presented at the 2008 World Motorsport Symposium in Oxford UK.

This also affects the BSFC due to increased pumping losses.

Paul Lamar 11/24/2007