Debuted in the 1964 Tokyo motor show and went into production in 1967, Cosmo was the first Wankel-engined car by Mazda and only the second in the world - just after NSU Spider. However, I would say it was far more significant than the NSU in terms of influence to the fate of rotary engines.
The Japanese did not invent rotary engines, of course. Mazda purchased license from NSU in 1961, which was quite a gamble to a small manufacturer located at Hiroshima. The original German design was full of problems – excessive wear at the tip seals caused high fuel consumption, oil leak and early breakdown – so bad that it actually led to the bankruptcy of NSU. To make it practical for road use, Mazda spent a lot of time and efforts to perfect it. With some luck, it found that an engine housing made of aluminum / carbon compound would deliver a hard yet low friction surface for the rotors to spin against, and rubber tip seals would reduce leakage yet withstand the required temperature. As a result, the world's first reasonably durable Wankel engine was born.
It was a twin-rotor unit. Each combustion chamber displaced 492 cc thus the effective capacity was 1968 cc. Like most other Wankel engines following it, it was compact and lightweight (only 102kg), high-revving (with 7000 rpm redline) and creamy smooth. The early engine produced a respectable 110 horsepower. Cars built from 1968, or Series 2, raised output further to 130hp. However, also in typical Wankel engine way, it was short of torque, so it needed to be squeezed to get performance out of it.
To the surprise of many, this engine was put into a stylish 2-seater coupe. Designed in-house by Heiji Kobayashi, the Cosmo had traces of contemporary Jaguar E-type, Ferrari and Ford T-bird. This made Mazda's first rotary attempt all the more memorable.
||Cosmo Series 2
|Year of production
||1,519 units (all)
||1,519 units (all)|
||2 x 491 cc
||2 x 491 cc|
||115 mph (c) / 104 mph*
||124 mph (c)
||7.5 sec (est)
In theory, the rotary engine has a lot of fascinating advantages. Firstly, the rotors spin in one direction, unlike piston which changes direction reciprocately, thus it should be smoother and waste less energy. Secondly, it uses less components - a twin-rotor Wankel engine equals to a conventional 6-cylinder engine, and it had no valve gears, crankshaft or connecting rods, thus it could be smaller, lighter and simpler than piston engines. Thirdly, lack of valve gears and crankshaft leads to less vibration and noise.
NSU's rotary engines
Dr. Wankel experimented his first rotary engine in NSU, which eliminated the imperfect eccentric movement of rotor by a revolving housing which was held by another housing. It was perfectly smooth, but too complicated to implement in economical way. Therefore at last a more compromised design, with an eccentric rotor and a single fixed housing (that is the one Mazda still using today) was adopted.
The first production rotary car was the 1963 NSU Spyder, a small car like Fiat 850 Spyder. It had single rotor with capacity of 498 cc. That was just a low volume experimental project.
The really ambitious application came in 1967 when NSU launched the famous Ro80 mid-size sedan, which won European Car Of The Year that year. Two rotors produced 115 smooth horsepower. Unfortunately, it was quickly found to be unreliable. Tip sealing worn out quickly, increasing fuel consumption, decreasing power and even leaking oil. This cost NSU a lot for repair and established a poor reputation for Wankel engines. To a certain extent, NSU’s bankruptcy and absorption by the Volkswagen group was due to the Wankel engine. Since then it had never produced rotary engine again.
However, before that happened Wankel engines had already attracted much attention globally. To raise development budget, NSU sold the license for United States to an aircraft engine maker called Curtiss-Wright, which then sub-licensed to General Motors, Ford and other car makers.
General Motors was the most eager player besides NSU and Mazda. It obtained worldwide license from NSU in 1970, then displayed two prototypes in 1973, Chevrolet Corvette 2-Rotor and Corvette 4-Rotor, both were mid-engined sports cars. Having invested a lot of money on rotary projects, GM planned for 80% of its 1980 production to be rotary cars.
The real production car, Chevrolet Monza, was to be launched in 1974, but the plan was hit by the newly announced smog regulations for 1977, which was so strict that GM feared the rotary engine might not pass. Eventually, Monza appeared with conventional engines. The prospect of rotary programme became uncertain, and virtually dead when its supporter, Ed Cole, president of GM, retired in September 1974.
Citroen, Mercedes and Nissan
Citroen collaborated with NSU and created 2 rotary cars, Ami 6 M35 and GS Birotor. Both were powered by NSU engines. However, they were short-lived and limited in number.
Mercedes was another giant European car maker to experiment Wankel engines. Mid-engined exotic concept car C111 stunned the world in the 1969 Frankfurt motor show. 3 rotors and 4 rotors version produced 320 hp and 405 hp respectively. But it did not come true.
Nissan took a Wankel license in 1970 and created a prototype in 1972. It planned to produce a small sports car at the rate of 120,000 cars per year, but eventually cancelled after the 1973 oil crisis.
Mazda - the only survivor
Mazda, called Toyo Kogyo then, obtained the license from NSU in 1960. Originally it wanted to put NSU's engine directly to its vehicle, but after experienced delivery delay and disappointing vibration and fuel consumption, it designed its own version started from drawing board.
Why was Mazda so eager to develop this technology ? Because in the 1960s, the Japanese government wanted to merge various car makers in order to enhance their global competitiveness. To retain independence, Mazda believed developing a unique technology would help. Therefore the decision to develop Wankel engine was not because of profit.
Its first rotary car, Cosmo, went into low volume production in 1967. That was a 2-seater GT like the original Ford Thunderbird. Twin-rotor each displaced 491 cc, capable of pumping out 110 hp. It was a true 200 km/h GT. 5 years of production totalled 1,519 units.
Having experimented, Mazda became confident to put Wankel engines into mass production. In 1968, it launched a 4-seat coupe called R100, which was based on the mass-produced Familia sedan. Powered by 10A twin-rotor engine with 100 hp output. It was shipped to the USA in 1970 and became a hit. Mazda created a rotary boom there with more rotary cars followed - R130, RX-2 and RX-3, all were based on the R100. In 1971, Mazda produced the 200,000th rotary cars.
The prospect was bright, and the company planned to produce 100% rotary cars by 1975. But the oil crisis changed that in one night. Since rotary cars were well known for thirsty, sales in the US was down by half in 1974 and did not recover to the 1973 level until RX-7 was launched. Wankel engine for econo cars became out of the plan, thus limited to only RX-7 in the following 20 years. The only exception was the 1992 Cosmo, which was a modernised version of the original Cosmo, with 280 hp twin-turbo 3-rotor Wankel, though it sales was bounded in Japan.
RX-8 and RENESIS
For 30 years, countless of car makers tried to develop Wankel engine but they eventually gave up. The only survivor, Mazda, made the most progress but still failed to catch up with the progress of piston engines. The biggest weaknesses of Wankel engines have always been heavy fuel consumption and poorer emission, the latter eventually killed RX-7 in 2002.
Just when everybody saw Wankel engine falling, Mazda worked quietly on a new generation Wankel engine called RENESIS (means Rotary Engine Genesis), which eventually applied to the RX-8 coupe. RENESIS differed from all the previous Wankel engines in exhaust ports configuration - the old ones used side intake ports and peripheral exhaust ports, but the new engine had side ports for both intake and exhaust. This eliminated overlap between intake and exhaust period, lengthened expansion stroke and enhanced thermal efficiency. It also improved breathing efficiency and reduced pollution, enabling the RX-8 to comply with tighter emission regulations.
However, the RENESIS was still too thirsty compared with conventional piston engines, especially when the development trend headed towards greenness in the late 2000s. Can Mazda's Wankel engines make another breakthrough ? It will be interesting to see.