Creating the fastest motorcycle: breaking the world speed record for streamliners is mostly a labor of love--plus a dollop of computer analysis.

The BUB "Seven" official posted the new world speed record of 350.884 mph last September at the Bonneville Salt Flats. Not bad for what Denis Manning, founder and president of BUB Enterprises, calls the "sure-fire, money-losing proposition," according to Joe Harralson, who designed the engine and did the finite element analysis (FEA) on this record-breaking streamliner.

What's a streamliner?

There are essentially three categories of motorcycles when it comes to world records. At one extreme is the open motorcycle. It doesn't have any fairings. There's the partially streamlined motorcycle. It has fairings, mud guards, and other features to help air slip around the motorcycle. At the other extreme is the streamliner. This is a fully encased motorcycle that looks like a rocket on two wheels. People don't throw a leg over a streamliner as they do when hunkering down on a conventional motorcycle. Instead, the driver is strapped inside a cockpit, and instead of using "body English" to help steer, the driver's hands and feet steer and control the bike. Streamliners need about 11 miles of course to break a record: five miles to accelerate, five miles to decelerate, and a timed measured mile in between. In speed trials, two runs are made, in opposite directions, within a 2-hour window. The official time is the average of these two runs.

The record broken last September had held for 16 years. In 1990, Dave Campos went 322 mph. Incidentally, the world speed record for a conventional motorcycle is 255 mph, give or take, set two years ago.

Not the first time

The Seven is Manning's seventh streamliner. In 1970, he built the world record-breaking Harley-Davidson streamliner driven by the late Cal Rayborn at 265 mph. Manning wants to retain the record. He also wants to have a bike that is faster than the fastest piston-driven car. That speed record, depending on sanctioning body, is somewhere around 410 and 417 mph.

[ILLUSTRATION OMITTED]

A little about Manning. In 1978 he founded BUB Enterprises, a manufacturer of exhaust systems for motorcycles and ATVs that operated out of Manning's two-car garage. Now, BUB operates from a 15,000-f[t.sup.2] building in Grass Valley, CA (www.bub.com). In 1989, Manning spoke at an SAE meeting in northern California. He talked about racing, the streamliner, and world records. He also mentioned he was going to build his own engine for the next streamliner. This piqued Harralson's interest. "Up to that time, all motorcycle speed record attempts were either factory-based efforts or used modified production motorcycle engines," says Harralson.

Now a little about Harralson. He has worked for Mercury Marine (yes, outboard engine) and McCulloch Corporation (think chain saws) as a design engineer. Engines were his specialty. Eventually, he joined the California State University Sacramento and became a tenured professor of mechanical engineering. He's now retired. He still teaches. And he still designs engines.

Harralson has been using finite element analysis (FEA) since the early 1970s, back in the days of punched cards and mainframes, running Nastran when nobody but aerospace companies had that software. When he got to the university, Harralson got interested in FEA on PCs. He wrote a program. Then he saw the FEA software from Algor, Inc. (Pittsburgh, PA; www.algor.com). It had a better interface than what he had written, so he bought a license. This was the early 1980s. He's had an Algor software license ever since.

[ILLUSTRATION OMITTED]

When Harralson heard Manning, he thought, "Wouldn't that be a wonderful project to get involved in? A blank sheet of paper. The world's fastest motorcycle. It wasn't something a sensible person would have agreed to work on." Harralson scratched out some rough layouts of an engine, called Manning, and that was the beginning of a long friendship.

FEA on the streamliner

Harralson used Algor FEA software on the streamliner "quite a bit." Back then, the software was DOS-based and, he adds, "You had to draw the FE mesh yourself line by line." Harralson analyzed the stiffness of the crankshaft, the engine block, and heads. "The crankshaft was a particularly critical component because the engine is a 90-degree V4 and has an uneven firing order, which produces strong torsional loads," explains Harralson. The result of his analysis was the 4-cylinder, 16-valve, liquid-cooled, double overhead-cam, 3000-cc (2997-cc, actual) engine that powered the Seven. Fueled by methanol alcohol, the turbocharged engine produces 500 hp of power. All the FEA has paid off: This engine has never failed, according to Harralson.

At one point in the streamliner's life, the team built a conventional bike with a tubular steel frame. It was quite heavy and it wasn't fast. By this time, Harralson was modeling major parts of the streamliner in the CAD program from SolidWorks Corp. (Concord, MA; www.solidworks.com). Algor's FEMPRO interface used these CAD models directly for analysis. Harralson was able to show why the bike wasn't going fast. So the team basically built a whole new motorcycle on a carbon fiber composite frame, but kept the engine and the gearbox. That was the bike that set the record.

[ILLUSTRATION OMITTED]

FEA solved another problem. The team figured the original tube-chassis frame was adequate structurally, so if the new carbon-fiber frame was stiffer and stronger, it should be okay. Says Harralson, "I'm an old FEA hand. I know you've got to back up FEA with real tests. So we built a test fixture, loaded up both chassis, twisted and bent them, and measured the deflections." The measurements were within 5% of what the Algor software said they would be.

That wasn't the case with the rear suspension swing arm. The deflections were about half of what the software said. In the field, bolts were coming loose; one broke. It turns out a bolted joint was not modeled correctly in the FEA model. The team added more bolts, dowel pins, and such to the joint, and "suddenly we were close to the computer model's predictions on stiffness."

"FEA is not there to replace physical tests," says Harralson. "It's to work hand-in-hand with the tests to get better results." World record breaking results.

[e-mail]

by Lawrence S. Gould CONTRIBUTING EDITOR