page 1

page 2

Japan 1


new Cosmo Sport




from home

the Start of rotary power

the first Mazda rotary,
the Cosmo

Tsuneji Matsuda succeeded his father in 1951. Then in 1959 he first learned of a strange engine from aero engine manufacturer Curtiss-Wright.

Matsuda's interest was immediate and he tried to gain a manufacturing licence for the Wankel engine from NSU the following year.

But NSU were cool towards any Japanese, but undeterred Matsuda and engineers visited NSU and were amazed by the trick of coins balancing upright on running engines, (so thats where the party trick originated!)

Mazda was granted a licence to use supplied engines in Japanese market cars only, but bureautic delays meant that they finally got a license in 1961...after rival Yanmar Diesel, (who now make engines for speedboats).

However NSU were very slow to build the Wankel engine and in 1961 vice President Kohei Matsuda visited NSU to find out what was holding things up. They returned home with drawings, technical info and many promises for production of the NSU KKM400 Wankel engine, an engine was to be shipped to Hiroshima.

However the Mazda engineers were disappointed with the single rotor NSU engine, It had excessive vibration when idling, clouds of white smoke at all times, appalling oil and fuel consumption, then after 200 hours of running the thing suddenly lost power as its chattering rotor tip seals ate the electro plated sides of the combustion chamber.

NSU's twin rotor powered and uncannily Audi 80 shaped Ro80 continued to be offered from 1967 to 1977 gaining the Wankel engine a reputation for abysmal economy, and unbelievable reliability.

Every other company who had an NSU license gave up! and struggling NSU was swallowed up by Volkswagen in 1971.
But Mazda would carry on.

Company President Tsuneji Matsuda, appointed Kenichi Yamamoto to lead the Rotary project.

"Mr. Rotary" Kenichi Yamamoto

"I was shocked," Kenichi Yamamoto told Automotive news in 1993, speaking of his appointment to lead the development of the Rotary engine for Mazda 30 years earlier.

"I had a rather negative view of the engine Yamamoto continued, and Japan was being rapidly motorized, we were anxious to have any engine, so I thought: why should I take up the Rotary engine?"

For Yamamoto it was a rather dull prize, developing the new mid sized rotary Familia after a career that dated back to 1946.

The problem Matsuda said, was that Japans Ministry of International Trade and Industry wanted to merge Toyo Kogyo out of existence. As part of its general plan for the automotive industry.

Yamamoto recalled: "every other company saw the Wankel engine as a source of potential revenue, while Mr. Matsuda saw it as an engine that would give Mazda a unique character and as such would save us from a merger with other companies, Mr. Matsuda showed leadership, he championed the engine and I in turn felt I had to champion the engine and lead my staff."

A $750'000 Rotary engine development centre was built and staffed by 180 employees and equipped with the latest computers for engine analysis.

By August 1964 30 test benches were in operation and Yamamoto decided to appraise the new engine by the same standards as the piston engine, because that is what potential owners would do.

The Mazda Rotary was different to the NSU Wankel engine, Carbon Aluminium apex rotor seals replaced NSU's cast iron, while Mazda chose to use double spark plugs for each chamber instead of NSU's single plugs.

Mazda also switched to side ports for the intake with one on either side of the rotor, this significantly reduced the overlap of intake and exhaust port openings and though this sacrificed top end power it improved idle smoothness. To guarantee smoothness, Mazda decided to use 2 rotor designs and abandon Felix Wankels original single rotor design.

The special engine deserved a unique home, so work began on a special chassis for the "engine of tomorrow", not a Kei car but a full size sports coupe in shape like a much longer and racier R360.

The prototype was a Tokyo Motorshow debutante in late 1964, though Tsuneji Matsuda had actually driven one of two L8A 798cc powered prototypes to the 1963 show, and back home to Hiroshima, stopping to visit Mazda dealers along the way!

Named "Cosmo" the prototype had a painted nose and faired in headlights. At the back large rocket tail light clusters were bisected horizontally by a chromed bumper.

The shape was attractive if festooned with a few too many chromed features. The interior was superbe, a thin wooden steering wheel and classic British sports car dash with all the right dials for the enthusiast.

The suspension used double A arms at the front and a deDion axle at the back. The 64 prototype had a L10A 982cc engine.

That 64 prototype looked ready to sell, and it just about was. In January 1965 sixty Cosmo's were released to ordinary citizens and dealerships across the land as part of a real world test. Differing engine porting arrangements on the cars provided valuable data on the best arrangement. The cars totaled 375'000 test miles with no mechanical failures.

When production began in May 1967 the Cosmo was equipped with a dual side port engine, and a four barrel Hitachi carburetor. The twin rotor engine displaced 982cc and produced 108 to 110 bhp dependent on who you believed, from the second quote came the Cosmo's 110S export designation.

The Mazda rotary with dual ignition had the unusual feature of two distributors. The transmission was a 4 speed with a direct drive 4th ratio and a final drive of 4.111:1.

Front suspension was unequal length A-arms with coil springs, an anti-roll bar, and tubular shock absorbers.

The upper arms were welded tubes, the lower arms were stamped. The rear suspension featured a 51mm deDion axle on semi-eliptic springs, with trailing links and tubular shock absorbers.

The differential was mounted to the body on a tee shaped member with noise vibration insulated by rubber. Axle half shafts used ballbearing sliding splines.

Stamped steel wheels measured 14x4.5 in with either 6.45x14in cross ply or 165HR-14in radial Bridgestone tyres.

Neither brakes or the rack and pinion steering needed power assistance, the fronts being discs and the rears were drums.

continued next column



who invented the rotary?

The rotary engine idea has been around for several centuries with steam powered versions employing up to 12 chambers suggested during the 18th century.

However the inventor of the now accepted rotary concept (a 3 chambered rotor within a trochoidal housing), and certainly the man who's name stuck with it was the self taught genius: Felix Wankel.

"Genius" Felix Wankel

He put his idea down to "a liverish induced state brought on by an excess of Christmas pudding," but in fact he had devoted his life to researching and developing the design.

Felix Wankels design was actually considerably more complex, (involving a rotating rotor and rotating housing), than the engine which was the first rotary to ever run. This modified version of the Wankel concept fired up on February 1st 1957 and was designed by Walter Froede.

Walter Froede working alongside Felix Wankel realised that something simpler was required, and secretly developed the classic rotary engine design (rotating 3 sided rotor within a fixed Trochoidal chamber), after Wankel had dismissed Froed's proposal. The modified concept however was proven by the first test engine.

The changes required to make practical that smoking, thirsty, though smooth running and powerful design into the engine that propels the Mazda Rx7 and Cosmo were many and protracted. At first Wankel was infuriated with Mazda's "meddling" but later expressed gratitude for Mazda's extreme perseverance.

Production styling was tidied by eliminating the targa band and addition of cabin-ventilation "ears". Otherwise little changed from the original show cars.

Over fourteen months, 343 of the L10A Cosmo Sports coupe were built.

In July 1968, production switched to the type L10B Cosmo. Although the designation refers to a new engine, the entire car was subtly altered. The new engine still displaced 982cc but revised port timing boosted power to 128bhp at 7000rpm.

A fifth gear (overdrive at 0.841:1) was added to the transmission, the brakes got vacuum assistance and air conditioning now became an option.

Fifteen inch wheels with 155HR-15 in radials became standard equipment.

The Cosmo also grew gaining 5.9 in between the axles. Overall length, however, decreased slightly.

From the front a larger mesh opening for the radiator could now be seen below the front bumper.

Performance was noticeably improved whereas Toyo Kogyo claimed 16.4 sec for the quarter mile and a top speed of 115mph for the L10A engine, the L10B tripped through the quarter mile in 15.8sec and topped out at 124 mph.

Unfortunately the price rose as well. The L10A was priced at 1.48 million yen, or about $4100 US at the 1967 exchange rate. That wasn't cheap, but as Mazda enthusiasts will point out it was less than two-thirds the price of a Toyota Celica 2000 GT.

The L10B Cosmo retailed at 1.48 million yen, which is $4390 US, still cheaper than the Toyota.

Production of the Cosmo crept along, quite literally. The assembly line was not automated, resembling (as much as possible for the Japanese) that of an Italian exotic manufacturer. Cars were moved along on chassis dollies, pushed by hand.

No wonder that only about twenty Cosmos were produced per month, about one per day.

Toyo Kogyo built the Cosmo until September 1972. The final count for the L10B Cosmo was 1,176. The five year production total of all Cosmos was 1,519, this seems to make the car less than a success.

If you key in development cost, you can be sure that Toyo Kogyo lost money on every single one. However this misses the point. The Cosmo was intended as a test bed for the rotary engine and as a high visibility vehicle for Toyo Kogyo, in this it succeeded admirably.

Toyo Kogyo brought the Cosmo 110S to market in amazingly short time, particularly for a car with a new type of engine.

Unfortunately, the Cosmo had no immediate successor. While Mazda enthusiastically focussed on fitting rotary engines to everything, several projects graced the drawing boards.

Project X605 and the next Cosmo would pick up the story.

The next chapter (Round and Round) deals with those vehicles adapted to the rotary, during which those 2 brothers were born:

The new rotary sports car, the Rx7.

The Cosmo becomes an elite super coupe.


click to continue
the Rx7 story in
Round and Round


click to continue
the Cosmo story

need money for that new RX ?

Are you Paid to Travel do you want Fun For Life

A Pre Paid Credit Card with Offshore Bank Account
+ Affiliate earnings
The art of Making Money