Felix Wankel, a German engineer, worked for NSU (NSU Motorenwerke AG) and was the given the task of developing his “brainchild” rotary engine, an internal combustion powerplant that uses an eccentrically-turning rotor instead of reciprocating pistons. Although he had started development of a rotary engine as far back as 1924, it was in 1957 that he finally managed to get one running (the KKM 57) at the NSU research department. The engine showed great promise in that it was smooth, small, quiet and relatively simple, at least in concept. A number of companies – among them Rolls Royce, Curtiss-Wright, Norton, Suzuki, John Deere and GM - around the world attempted their own designs over the next few years, but none continued development.

No one, that is, except Toyo Kogyo, the little automotive company in Japan that would brand its cars Mazda. The company bought the patent rights from NSU in 1961, along with a prototype engine. That engine failed on the test stand within a few tens of hours, launching Toyo Kogyo on an intensive development program to solve the technical problems inherent in the rotary engine. Millions of dollars were spent on the metallurgical and machining aspects of the design, much of which was devoted to computer time to solve the “K Factor.”

And what is the K Factor? We’ll get to that in a moment, but first some theory.
Rotary engines use a triangular-shaped, rounded tip steel rotor that spins (eccentrically) inside a shaped housing. The shape scribed by the tips of the rotor form a geometric figure known as an epitrochoid. As the rotor moves inside the housing it comes very close to the walls of the epitrochoid, thus creating areas of compression. Valves and spark plugs can be situated in those areas, thereby creating combustion chambers. This powers the rotor, which was Felix Wankel’s great idea.

However, Toyo Kogyo found that there were two major problems to be overcome. The first was the form and materials needed for the seals on the sides and tips of the rotor. Without seals there would be no compression, of course, but the seals themselves need to travel with the rotor without causing wear in the housing, called “chatter marks.” Great amounts of time and engineering efforts were made to solve the sealing problem. To eliminate this phenomenon, a cross-hollow seal was developed by drilling small holes inside the metallic apex seal. This greatly improved the durability of the prototype engine, enabling it to complete 300 hours of continuous high-speed operation. Also in the initial stage of rotary engine development, engineers were faced with the further problem of engine oil leaking into the combustion chamber, causing excessive oil consumption. The team identified the oil seal as the cause of the problem, and developed an innovative Mazda-unique oil seal in conjunction with Nippon Piston Ring Co. and Nippon Oil Seal Co.

The second problem was the efficiency of the rotor itself. It was found that, as it turned, the radius of its rotation (from the center of the crankshaft attached to center of the rotor) relative to the eccentricity (the distance from the center of the rotor to the tip of the rotor) was critical to engine life and power output. Vast amounts of computer time were devoted to calculating the optimum radius-to-eccentricity ratio, or K Factor.

[Show as slideshow]

Toyo Kogyo risked everything on the outcome, but it was worth it. On May 30 of 1967 they unveiled the first production rotary-engine Mazda Cosmo 110S, a little sports car that captured the hearts of enthusiasts with its 110-horsepower Type 10A engine. The Cosmo Sport recorded more than 3 million kilometers of road tests over six years, winning a number of races in the process.

Mazda was finally ready for the US market and, in 1970, introduced the R100. It was quickly followed by the RX-2, RX-3, RX-4 and RX-5 Cosmo along with its little Rotary Pickup truck. Owners loved the cars, but the oil crisis of 1973 and shifting needs of the buying public reflected the relatively poor fuel mileage of rotary engines. Solving the fuel economy and emissions problems of rotary engines took Mazda the remainder of the decade, but in 1979 the sports car world was completely shaken up with the arrival of the legendary RX-7.

The car was so successful that several subsequent generations followed in 1986 and 1993, culminating in the first turbocharged rotary engine RX-7, a highly sought-after collectible. Mazda’s rotary engine cars achieved over 100 professional sports car racing wins during these years, culminating in an overall win at the 1991 24-Hours of Le Mans and going into the history books as the first – and still the only – Japanese car company to ever win the famous race.

Mazda continued heavily into piston-engine products in the 1990s and early 2000s, but the rotary engine was far from forgotten. In 2003 they launched the RX-8, the first-of-its-kind 4-door pure sports car powered by its newest Renesis engine. Although it only displaced 1.3 liters and is the size of a toaster over, the 9,000 rpm engine developed 238 horsepower and 159 lb-ft of torque. Its sales remain brisk as the current model year marks 40 years of rotary engines for the company.

Where Mazda might take the rotary engine in the future is anyone’s guess, but it’s sure to be with us for many years to come. No other company ever succeeded with rotary engine design and engineering and no other company ever did so much with it. Rotary-powered Mazda cars are certain to hold a very significant place in automotive history.

Torque vs Horsepower: Which Is More Important?

Any gear head or car guy loves power, and the more the better. Few experiences are as satisfying as the neck-snapping, back-pressed-into-the-seat feeling of acceleration that can be had at the expense of some gasoline and tire tread. Engine power has to be adequate for such acceleration, however, and that’s the subject of this article.

Engine output has for decades been quantified by “horsepower,” although in recent years manufacturers have been listing torque values as well. The reason is that, frankly, torque is more important. Let’s look at what these measurements are and then put them into practical use.

First, let’s not confuse power with work. “Work” is the force used to lift or push or pull over a distance. When you lift the coffeepot and hold it there you are doing work. Power, by definition, is the act of producing work over some specific time. The familiar “horsepower” is a term that stems from the 1700s and was created by James Watt to sell his steam-powered water pumps. He calculated (optimistically, as things turned out) that one horsepower is the ability to lift 550 pounds one foot in one second, presumably the power an average horse can exert without killing itself.

Just before the turn of the 20th Century carriage makers turned to internal combustion engines – rather than the trusty, if not smelly and high-maintenance horse – to propel their new carriages, but they used the tried-and-true measurements of horsepower to describe their new engines’ capabilities. Everyone understood it, because they could relate to it. In today’s world few of us have any regular interaction with horses, so the term “horsepower” is somewhat esoteric. Besides, it really doesn’t explain how a car accelerates.

That takes us to Jan and Dean.

What? Hang on, readers, because this will all make sense. Jan and Dean wrote and performed what is, arguably, the greatest car song of the 1960s, Dead Man’s Curve. In it, a race between an XKE and a Corvette ends in disaster. The gist of the story is that a mindless idiot driving a Corvette is challenged to a drag race by another mindless idiot driving an XKE. They end up going too fast and, after dramatic sound effects of crashing cars, the singers end the song thusly: “Well, the last thing I remember, Doc, I started to swerve, and then I saw the Jag slide into the curve. I’ll never forget that horrible sight, and I found out that everyone’s right…Won’t come back from Dead Man’s Curve.”

The important point to the song is that it is technically accurate. That is, the Corvette out-accelerated the XKE, although both cars were virtually equal in power-to-weight ratios, at about 10.5 pounds to the horsepower (3,200 lbs/300 hp for the Corvette, 2,800 lbs/265 hp for the XKE). Both cars had nearly identical gear and rear axle ratios.

So how could the Corvette have been so much faster than the Jag? In a word: Torque. The Corvette had 100 lb-ft more torque than the Jag, making it over one second faster from 0-60 mph.

Arithmetic Time

We need to discuss the way engines actually accelerate cars. Obviously, a certain amount of power is required to keep a car rolling in the first place. In the case of a that Corvette, let’s say a force of 400 pounds is required to push it along at 60 mph on level ground. To translate that into horsepower (all we have is a force at the moment), we need to add the time element.

Since 60 mph is 88 feet-per-second, to calculate how much horsepower is needed we just multiply the 400 pounds of force times 88 ft/sec and we get 35,200 pounds-feet-second. Since we know that 550 pounds-feet-second equals one horsepower, we just divide 35,200 by 550 to get 64 pounds-feet-second, or 64 horsepower.

But that 64 horsepower isn’t accelerating the car, just moving it along. We need torque to accelerate. Torque, by definition, is a moment of force that produces rotation – or torsion – and is the product of tangential force multiplied by the radius of the part rotated.

Confused? Look at it this way: If you picture a horizontal arm one foot long and hang a one-pound weight on the end of it, you will have a torque of one pound-foot acting on whatever the other end of the arm is attached to. If we want to express this work as power, we must add the time dimension to it. Thus, we would have one pound-foot-second.

Back to the Engine

The engine above, while producing horsepower, is also producing torque. That’s because it is rotating, and we should be able to calculate its torque. In the case above, let’s assume the Corvette’s engine is spinning at 3000 rpm. By dividing the 3000 rpm by 60, we get 50 revolutions per second, right?

Okay, since we know from above that the engine produces 35,200 pounds-foot-second, all we need to do is eliminate the time factor to get the torque number. We do so by dividing 35,200 pounds-ft-sec by 50 revs-sec and we get 704 pounds-ft-revs.

So what the hell does that number mean? Not to worry. We’re talking about circular motion, right? All we need to do is remove the circular component from the number above and we get what we want. To do so is simply to introduce the concept of the “radian.” A radian is the length of the radius of a circle laid onto the circumference. Without going into painful – and boring – geometry lessons, just take our word for it that there are always 6.2832 radians in any circumference because of the relationship of Pi, or 3.14. Therefore, if one revolution has 6.2832 radians, we divide the 704 pounds-foot-revs by that number to get 112.04 pounds-foot-radian. Since radians have no actual value, we drop the word and end up with 112 pounds-feet of torque.

Okay, So Why Is The Corvette Faster?

It should be apparent that any engine produces torque because it is turning. Horsepower is produced by the exploding fuel in the cylinders, which turn the crankshaft, etc., creating the twisting force (torque) that you have at the flywheel. It follows that the greater the torque at the low end of the rev range, the less resistance the car’s weight has against the engine’s tendency to turn. The less resistance, the faster the car moves.

In the case of the Jag vs Corvette, the Vette’s engine has 100 pounds-feet more torque than the Jag’s. On top of that, the V8 configuration of the Corvette’s engine produces far greater torque at the low end of the rev range, hence greater acceleration. Of course, eight cylinders produce two more power strokes per revolution than the Jag’s six cylinders. It all adds up.

The real measure of how fast a car will accelerate is its torque, and at which rev range the maximum is developed. Horsepower is of secondary importance actually, but it sounds better than “torque.” Today’s cars with small displacement engines are fast and agile because they produce torque in another way: gearing. Five and six-speed transmissions use lower gears to multiply engine torque, creating terrific acceleration and good fuel mileage, although the “super cars” still use big engines.

Pounds-Feet or Foot-Pounds?

Technically, it’s “pound-feet,” because work is defined as force over distance. However, decades of usage of the term “foot-pound” by nearly everyone – including engineers – makes either term acceptable.

Anyway, who cares how torque is expressed, as long as there’s plenty of it!