A Conversation About Motorcycle Engines With Fujio Yoshimura

Over the past 40-odd years, whenever I've come back from a trip to Daytona or other centers of motorcycle activity, I've had far more conversations and observations than any editor could accept for publication. That extra material has rested on the pages of my notebooks.

In 1983, the big noise was Honda's Interceptor, a motorcycle intended as a Superbike homologation special—a rumored 1,200 in the initial build—and expected to be ignored by the riding public. But it didn't turn out as Honda planners had imagined: People were delighted with the quick, go-where-I-say handling of that bike, and it had the effect of giving the market a new direction. That, combined with Kawasaki's Ninjas, Suzuki's soon-to-arrive GSX-R750, and Yamaha's XJ600, defined a new category, the sportbike.

In that same notebook is a conversation with Fujio Yoshimura, son of Hideo “Pops” Yoshimura, who created the specialist tuning and aftermarket business that survives to this day. “We had an old Suzuki GS750ES with factory rods and bearing shells,” he told me. “It had a little problem: It ate the bearings.”

Yoshimura sent an oil sample and the bearing shells to Chevron Research for analysis: “The test oil was for roller-crank motors [previous GS engines had roller cranks].” He was at the time expecting production engines from Japan, especially cylinder heads. “It’s hard to get cylinder heads!

“First we try to make the power,” he continued. “Find the potential first, then reliability is only an engineering problem. Sometimes there are parts in the ceiling [of the dyno room].”

Yoshimura senior worked closely with Suzuki in Japan, but, “We make parts here to save time,” Fujio said. “And money is necessary to do that. We can get just about anything made in Southern California.

“For Superbike, we need 115 hp at the transmission shaft from 11,500 to 12,000 rpm, beginning from 8,000.”

That year, ’83, was the first for of AMA’s “small” Superbike, down to a 750cc limit from the 1,000cc of the early 1975–’82 heavyweights. Rob Muzzy’s best Kawasaki liter engine had made 152 hp at 10,250 rpm, so the new 750s would look for power at higher revs.

“We get to 105–110 hp,” Yoshimura said, “then begin track testing at Willow or Riverside [which closed in 1989].” Track rental then was $1,000 a day—expensive. The team also ran on track practice days and at some West Coast club races.

“Wes Cooley says it’s like riding a 250,” Yoshimura said. “Go in fast, no relying on gobs of power. I think a lot of riders will seem more able [on the new lighter and better-handling 750s]. Reliability is always a battle never solved in advance. Putting engines together is easy: Read the [service] book! Taking it apart is what’s hard; you must see how every part was working.”

I asked if he often uses the advice of specialist firms, such as Chevron Research. “I like that,” he said of failure analysis and advice from specialists. Yoshimura must discover faults himself, for Suzuki tells him nothing. “I like working in America,” he said. “It is an open society. Facts are available here.”

Yoshimura’s dad began tuning work after the Pacific War and became the Kyushu dealer for English and German motorcycles—Horex, BMW. Then he built a Super Rocket conversion for BSA twins. He thought Japanese machines of the time were terrible with their pressed-steel frames and leading-link front suspension. The Honda Super Hawk 305cc parallel twin was the first Japanese design he could admire.

“Four-valve engines are so much easier on springs,” Fujio said. “Every single piece must have some reason to be a certain way. A piston, a valve involves so much thought and technology. The dialectic!”

Yoshimura is attending classes to expand his knowledge. The goal: engines! “No rush, though,” he said. “Thirty or 40 years to live!”

At the 1978 Suzuka 8 Hours race won by four laps by Cooley and Mike Baldwin, a clutch problem was fixed between qualifying on Saturday and race day.

After winning the ’78 Daytona Superbike race on a Yoshimura GS1000, rider Steve McLaughlin told the crew, “The motor’s making a weird noise.” When crew members pulled the clutch cover, the clutch basket fell out. As a first step, the 9mm retaining rivets were made 10mm. They still broke.

Next, the factory sent a new-type clutch basket with no spring drive and six solid rivets. The team reasoned that, with nothing flexible between the crank and the grip of a sticky tire, there’d be trouble, so they didn’t use it. Cooley led the next AMA Superbike event, but the hammering from the engine broke the drive sprocket off his seven-spoke Morris magnesium wheel.

Yoshimura decided the solid sprocket mounting was killing the clutch. In June of that year, Suzuki made a new clutch assembly with a cush drive. In anticipation of Suzuka on July 31, the team tested at Ryuyo in Japan and had no trouble with the new cush-hub clutch. But with too few laps run and other riders on the track, they couldn’t be sure.

Thirty minutes of testing at Suzuka produced another failure. They could see that the hub springs—three heavy, three weak—were being bottomed and hammered by engine torque pulsing. So they tried six heavy springs and were able to prevent the metal-to-metal impacts that had caused failures elsewhere.

Suzuki had tried the solid drive because the two-stroke RG500 GP engines worked okay that way and so reasoned it could work on a four-stroke. But firing half as often, the four-stroke’s torque pulses were twice as powerful.

RELATED: Basic Motorcycle Engine Architecture—Transverse Inline-Four

Other problems to be overcome in Superbike racing included the usual cam-chain tensioner action (cam drives have to endure their own kind of bump and thump). On the GS1000, Yoshimura at first ran steel cages to guide the connecting-rod big-end needle rollers, then aluminum ones that were one-third the weight with silver plating.

Slippage at pressed-together crankshaft joints required welding. “Crank life depends on case flex, for the old two-valve engines cases lasted two races,” Yoshimura said. “The Katana case was stronger; we got half a season. There was so much [cylinder base-gasket] leakage. We were getting 3 to 4mm of gasket extrusion!”

Muzzy had the same experience with the Z1-based Kawasakis: Crankshaft motion caused case flex that scrubbed base gaskets to failure. “The turbo Suzuki [the XN85, a 673cc four-cylinder sold only in 1983 in the US] uses a metal base gasket,” Yoshimura said. Muzzy had to make his own out of solid sheet copper.

About the switch from rollers or balls to plain crankshaft bearings? “It is inevitable but has longer break-in,” Yoshimura explained. “We will just get over the strangeness and have a better setup in the end.”

Yoshimura was using a desktop computer for cam-profile calculations. It printed a menu and you selected a program, of which he had a long list. It queries the data, you enter it in steps, then a solution.

Auto racers in Japan had tried titanium valves in the late 1960s, but the hard facing fell off the sealing surface and the wear-resistant chrome plating on valve stems chipped. Cast pistons had been okay until engines began making a great deal of power.

Yoshimura spoke of taking intake-flow direction into account in designing cams for the new four-valve-per-cylinder engines. Today, BMW in at least one design opens one intake valve early, producing axial charge swirl that boosts midrange torque.

He also spoke of problems that arise when two separate organizations try to cooperate. Suzuki preferred to do everything by remote control, and, in Japan, committee decisions were so pervasive that motorcycles and their sales promotion lacked character. I thought of the two-stroke, three-cylinder GT750: too heavy to be a sportbike, too oddball to be accepted by touring riders. And the RE5 rotary, new technology that sent no other message. Every maker has produced such models that have been pushed aside by the marketplace.

“Decisions taken here are reversed by Japan,” Yoshimura said. “ ‘Don’t use four valve’ was the order, even though the two valve was no longer nearly competitive.”

Despite such differences, the many successes achieved by Yoshimura and Suzuki since 1983 suggest that agile minds applied themselves as much to organizational as to technical problems.

Sign up here to receive our newsletters. Get the latest in motorcycle reviews, tests, and industry news, subscribe here for our YouTube channel.