An internal combustion engine invented by Felix Wankel. A triangular rotor spins in a cylinder. Three combustion chambers are created which in turn draw, compress and ignite a fuel-air mixture. More simple, more powerful and more efficient than conventional reciprocating engines. Used in a few cars and airplanes.

Monty Python finds them really embarrassing.
The last mainstream production automobile with a rotary engine was the Mazda RX-7. The 'R' stood for 'Rotary.' These were excellent engines, light, well-behaved, reasonably powerful and reliable. The final model year of the RX-7 (1993?) went like a greased goose, possibly due to the twin turbos it sported; it fell victim, however, to the same forces that felled the MR2 Mk. II in the U.S. - a slowing economy, lack of a convertible model, and the success of its smaller sibling the Miata.

Apropos of nothing, I can tell you that the RX-7 mill made an excellent modern replacement for the Triumph 2500cc beast that powers the TR6. By a freak accident (I suppose) the tranny and engine mount almost perfectly in a TR6.

Note: Ashley Pomeroy tells me that the RX-7 is still in production in Japan; it simply became too much to try to pass the increasingly restrictive U.S. emissions standards without an engine redesign. I also note from various web sources that there is a new RX-7 in the works for the U.S. market; presumably it incorporates such a redesign.


Okay, perdedor, here goes. This might suck. Just warning you. :-)

The key thing to keep in mind is that the central rotation point around which the rotors spin is not always (in fact, may not be ever) at the center of the rotor. The rotor is offset, and this is why a combustion cycle can spin it. If it was centered, there would be no reason for combustion and the resultant expansion to exert anything other than a uniform force on the rotor's combustion surface. Got that? Okay, with this in mind, here we go.

Let's take the Mazda 13B engine for reference. This is one of the more popular and readily available Wankel engines around; it powered the RX-7 around 1990. It is a two-rotor engine.

Think of the engine as a longitudinal sandwich; that is, it has layers if you move from end to end. At the extreme ends are the side housings, which close off the flat outer sides of the rotor housings. Each rotor housing is a metal ring which contains a rotor within its circumference and allows it to spin (wobble, really) as if it were 'rolling around' the inside of the ring. Remember Spirograph? Same principle. Anyhow. In between the two rotor housings, and serving to seal the inner faces of the rotor chambers in the rotor housings, are the intermediate housings.

What you have, then, looks a little like this:




      i = intermediate housing
               |       |
        ______           ______
    |ss|| rr |ii       ii| rr ||ss|  ----- s = side housing (similar at other end)
 ---|ss||-rr-|ii-------ii|-rr-||ss|-----
    |ss|| rr |ii       ii| rr ||ss|
        ------           ------
          |                 |
          |                 |
     Rotor Housing       Rotor Housing

...whew. Okay. Now, the rotors spin around the crankshaft (called in the rotary engine the eccentric shaft which is represented in the diagram with dashes (---) through the middle. So the diagram is looking at the engine from the side.

Here comes the cool part. The rotors are nearly triangular, which means that when they are inside the rotor housings, there are three gaps between the sides of the triangle and the inner surface of the rotor housing. As the rotor spins, these three gaps move around the inner surface of the rotor housing. With me so far? Good, 'cuz I'm not. Anyhow. Think of each of these gaps as a potential combustion chamber! At one point in the rotor's spin, there are intake valves. Unlike those on a car, these are simply open ports. When an apex of the rotor passes these points, the chamber (or gap) behind that apex draws in fuel/air mixture.

It draws the mixture in because the chamber is growing in size as it passes the intake port. This is because the rotor, as I mentioned, is off-center; at this point, its center is moving away from the intake area, and thus the space in the chamber is increasing. Okay. So eventually, the next apex passes the intake port, and seals the chamber against the rotor housing. The rotor wobbles on, and while rotating, its side moves closer to the other side of the rotor housing, making the chamber full of fuel smaller. This results in compression! Eventually, when the chamber is almost gone, meaning the mixture is maximally compressed, spark plugs projecting into the chamber fire. The mixture begins to burn and expand just as the rotor, still moving, begins to wobble away from the rotor housing wall, allowing the expansion to drive the wobble and thus the rotation onward.

After a small time, the leading apex of the chamber passes another set of ports, this time for exhaust. As the chamber moves on, it begins to grow smaller in preparation for the intake cycle, and in the process forces the exhaust out the open port. Finally, it moves past the intake port (now small again) and the wobble continues.

I hope that made sense. In any case, to sum up, the offcenter rotor means that for half of the rotation of a particular chamber, it will be decreasing in volume (exhaust and compression) and the other half increasing (intake and combustion).

Here's the next key point: the rotor moves at one-third the speed of the eccentric shaft. The inner hole of the rotor (where the shaft passes through) and the shaft itself are toothed like a gear; the inner hole of the rotor is larger than the shaft and has three times the circumference. Since there are three chambers, this means that a rotor will fire once per revolution of the eccentric shaft, in one of its three chambers.

The Mazda 13B is a two-rotor engine. This means that for every rotation of the eccentric shaft, two of the six total combustion chambers (one per rotor) will fire. Ergo, each rotor will undergo one power stroke per revolution. Compare this to a conventional reciprocating engine, where half the cylinders fire per revolution (the other half the time, they are on an intake, not combustion stroke). This is one reason the rotary is more efficient than the cylinder-based engine. Another is that since there are no moving parts in the valves, you don't need a cam, rockers, valves, and the like absorbing energy.

One final note: the valves and spark plugs can be in the rotor housing, or they can be in intermediate or side housings - as long as they are placed properly around the circumference of the rotor's travel. It is better to have them in side and intermediate housings, as that way the rotor housing (which, along with the rotor, bears the force of the engine) can be a single casting and not have weak points.

Wow. That better, perdedor?

Neil: One problem I could see occurring with Diesel rotary engines is that the Diesel requires much higher compression ratios and pressure, since the ignition of the mixture is done completely by compression heating as opposed to an externally-supplied spark. The greater forces on the 'chamber seals' (i.e. the rotor edges) might negate any advantage seen from the lubricating nature of the fuel.

You may ask yourself, ``If the Wankel engine is so great, why isn't it still in use in consumer autos?'' There are two reasons.

The first is durability. For the proper operation of the combustion cycle, the chamber created between the rotor and the housing must be perfectly sealed. The problem is that, since the rotor barely brushes the housing, and does so only at three locations, there is little tolerance for error. Thus, normal wear and tear to the rotor and housing will relatively quickly lead to an incomplete seal, which is bad. Pistons last longer because they have a much better seal.

Second is fuel efficiency and emissions. Because the chamber has an odd triangular shape, fuel and air pulled into chamber in the intake stroke do not get mixed evenly within the chamber (i.e., there is bad swirl). Thus there is, compared to the reciprocating engine, a larger amount of fuel left incompletely burned. Thus, not only are you getting less energy out of each unit of fuel injected into the engine; but you are also emitting more unburned gunk, which is bad for the environment (and, incidentally, for emissions standards).

A friend of mine (Carl Arthur), who is a mechanical guru, had the idea of making a rotary diesel engine. This, he claims, would solve or at least ameliorate these problems. As far as the first problem goes, diesel fuel is thick enough to serve as a lubricant, thus reducing wear. As for the second problem, the injection valves atomize the fuel pretty thoroughly, giving you enough surface area to ensure a relatively clean burn. I do not know whether the peculiar operation of the Wankel engine is incompatible with the diesel cycle, or if there is another reason such engines are not made.


Stewacide informs me that a rotary engine may have problems providing the higher compression ratio that diesel engines require. I'll have to get back to Carl on this one.

Log in or register to write something here or to contact authors.