Document Sample
buell Powered By Docstoc
					                                          Buell set up a large drawing board
                                          in the “barn” so he could draw the
                                          RW 750 engine components.
                                          The building was not really
                                          a barn—the barn was across the
                                          drive—but rather a large farm
                                          outbuilding with three garage
                                          doors and a concrete floor. Heat
                                          was by an antique oil burner that
                                          had its own small oil tank. Every
                                          day someone filled the tank by
                                          carrying a pail to the 300 gallon
                                          outside tank, filling it and hauling
                                          it into the shop. When it was really
                                          cold out it would be an hour
                                          before the coats came off.

         An original drawing of the
       RW 750 crankcase produced
     by Erik Buell. This is part of the
      drawing that is on the drawing
         board in the picture above.

Erik Buell was born in 1950 in Gibsonia, Pennsylvania. His dad was a
top notch patent attorney and his mom taught English. (Don’t turn in
any document to Erik with grammatical errors; it will come back with
all the errors marked up!) He was exposed to a lot of mechanical stuff
on the farm; his dad collected interesting cars and by the 1960s his
motorcycle passion was in full bloom. His first motorcycle ride at age
12 had ignited the flame that continues to burn today. By the late ’60s
he’d begun working at bike shops and was racing motocross. He
worked for shops throughout the Pittsburgh area but perhaps the most
important was the one belonging to Walter “Zero” Finnegan. Zero was
a bit of a local legend, choosing to run his own small shop rather than
use his considerable mechanical skills to work for someone else. He
was not a good fit for many of these shops as he was a meticulous
craftsman and committed to giving customers the best service he
could. Erik took to heart Zero’s commitment to customer service and it
remains a guiding principle for Erik and Buell Motorcycles.
    Erik had tried his hand at being a rock star—didn’t we all at that
age—but it was tough to make any money. The baby boomer explosion
in the motorcycle marketplace was well underway and competent me-
chanics were in demand. Seriously good mechanics could make some
real money. Erik was one of the good ones. He essentially turned him-
self into an independent contractor, working for whichever shop
needed a top-notch wrench. He could work on any brand and usually
could beat the book rate while doing first class work. If the shop asked
him to do something he considered unethical, he would simply move           1979–1985 RW 750
to another shop. The fellow who wanted him to clean-up and quiet
down the little Hodaka comes to mind. This dealer planned to sell it to
some sucker as a 4-speed. It had come from the factory as a 5-speed but
the tranny was failing. Erik refused to fake it and walked. Some shops
were a little rough. At one, a customer showed up with a shotgun and
the owners faced him down with the pistols they carried; Erik knew it
was time to move on.
    By the early ’70s the racing bug had bitten really hard and he was
riding on the fire roads in the area for hours each day to train. In 1972
he was running his F81M Kawasaki hard on a fire road when disaster
struck. He was flat out and sideways when coming the other way was a
young kid on a little dirt bike. The kid froze. Erik was running so hard

     he could not avoid a collision and took out the entire front end of the
     bike with his leg. The front wheel of Erik’s bike was destroyed and his
     leg was crushed. The forks bent backwards so far that they broke some
     of the fins off the cylinders. With a shattered leg and bleeding badly
     Erik was lucky that a Jeep came along and got him to the ER before he
     bled to death. It was a close thing. When he was stabilized, one of the
     doctors wanted to amputate. Fortunately a different doc thought the
     leg could be saved. What pieces remained were screwed into a metal
     shaft and they casted Erik from the hip to the toes. That was the end of
     his motocross/flat track career.
        Instead, when he could finally get up and move around, he took the
     motor from his crushed bike and combined it with a 350cc cylinder
     and a H1R Kawasaki frame to make a road-racing bike for a friend.
     They made it to Daytona in 1973 but shortly afterwards the fellow de-
     cided he wanted out. Erik bought the half his friend had provided and
     went road racing. He had realized that even though he could not race
     motocross with a cast he could get on a road racer.
        He turned out to be pretty good, running at the front of a Novice
     class that included the likes of Dave Roper, Richard Schlacter, and Dale
     Singleton. He saw Gary Scott beat Kenny Roberts at Laconia on the
     Harley-Davidson RR250 GP bike and when they became available for
     purchase he just had to have one. He borrowed a lot of money to buy
     one and when it showed up for the 1974 season, it was horrible. The
     production version was not even close to the bike Scott rode; they were
     unreliable and to top it off you couldn’t get spares. Erik watched in
     frustration as the fellows he had diced with as a Novice continued to
     win races and qualify for Expert while his bike would break and he
     would slide further down the list.
        Perhaps the low point of the season was the Talladega round. He had
     crashed the RR again by pushing too hard in a valiant but fruitless effort
     to make a slow bike go fast, and had a white gel-coated Yamaha fairing
     mounted on the bike (you couldn’t buy a Harley-Davidson fairing). No
     paint, just a number. While driving all night to get to the race he stopped
     at a truck stop and saw a very large decal of a leaping bass. You know the
     ones; they decorate the rear windows of campers all across America. Erik
     decided it was just the thing to brighten up the boring white gel coat.
     This decision may have had something to do with lack of sleep, but for
     whatever reason a pair of these beauties ended up on the bike. When he
     rolled the bike into tech, the inspectors took one look and flunked him.
     They would not let him race until he took the decals off.

Erik Buell was your typical racer: always
struggling to get enough money for racing.
The really good parts were expensive, so
Buell, not content with just being a
consumer, decided to become the U.S.
distributor, under the name of Pittsburgh
Performance Products, for some of the
products he wanted to use on his bike. He
got on board with manufacturers of the
best equipment out there. He sold the very
lightweight Dymag magnesium wheels, AP
Racing brakes, and Interstate Leathers
(despite the name these were high quality
race leathers made in England). The
exchange rate really favored the dollar so
this was a very nice little business.
   Buell also developed and patented a
“sandwich” brake rotor. At the time brake
rotors were either cast iron, stainless steel,
or aluminum. Cast iron has excellent brake
feel, good heat transfer, and good life but is
prone to rust and is very heavy. Aluminum
is very light and has excellent heat transfer
properties but wears out quickly.
   Buell addressed these issues in his
patent application:
                                                                 Sales literature was produced for the
   “It has long been recognized that the
                                                                       products Buell sold through his
removal of even a few ounces of unsprung
weight was of tremendous advantage.”                                Pittsburgh Performance Products
   “I have discovered a brake disc which                               company to support his racing.
solves all these problems. It is light in
weight, has high thermal transfer                    1980s using just one set of brake rotors for an
characteristics, and is highly resistant to the      entire year.
wear and abrasion of the brake pucks.”                  One of the points that Buell stressed for the
   Buell’s solution was to sandwich an               Dymag wheels and the Buell Brakes was the
aluminum core between two sheets of stainless        reduction in rotational mass and unsprung
steel. They had one-third less weight than a steel   weight. You can see from the ads that Buell’s
rotor plus much better heat transfer and a very      current ZTL brake system is not just “marketing
long life. The Wilke Brothers’ race team won two     speak” but a further development of a long-held
USAC National Midget Championships in the            belief.                                          ■

                         Barton Motors
             Barton was a tiny company founded in the mid-1970s
             to produce an English Formula One motorcycle. Most
             of the Japanese Formula One bikes were run by teams
             located in England so there was a lot of expertise
             available to make parts. The engine was derived from a
             Suzuki. The first, and apparently only, complete bike
             they built was for the movie “Silver Dream Racer”
             which starred David Essex. The chassis followed
             common wisdom at the time that too much stiffness
             was not good, and that the right amount of flex in the
             chassis would result in a nice-handling motorcycle.
             This sometimes actually worked for the underpowered,
             skinny-tired bikes of the early two-stroke era but with
             the power the 750 Barton was making by 1981, a
             flexible frame was almost unridable.
                This brochure for the Barton company shows the
             Silver Dream Racer bike that was featured in the movie
             of the same name.                                      ■

         The primary function of racing tech inspection is to ensure the
     safety of the participants; secondary functions involve meeting the ba-
     sic rules of the class you are running. They are not trying to catch
     cheating—if you have used the valve springs from a 1936 Super Duper
     while only 1935 valves are allowed, the post-race teardown is supposed
     to catch that. They also enforce standards involving appropriate decals
     and the like but safety is the first and foremost task of the inspectors.
     Imagine his surprise when Erik rolled his bike back to the pits to ready
     it for the first race only to find that the frame was cracked in three
     places. Seems it was okay if his bike folded in half at 150 mph but it was
     not gonna have a big old bass on the side when it crashed!
         In the mid ’70s, to help raise money to support his racing and also to
     cut costs for parts, Erik set himself up under the name of Pittsburgh
     Performance Products as a distributor for some of the parts he needed.
     This enterprise would prove to be a lifeline at key points in the future.
         While Erik was chasing his road racing dreams and wrenching at
     bike shops he was also pursuing a mechanical engineering degree at the
     University of Pittsburgh. As he worked his way through school he also
     worked his way up the ranks of AMA road racing. He made Light-
     weight Expert in 1977 and Formula One Expert in 1978. He graduated
     from Pitt with a Mechanical Engineering degree in 1979.

Erik had done well in school and his look of steely determination surely
did nothing to discourage potential employers, so he had his choice of
a few jobs including a very nice one in aerospace. Instead, he pursued
Harley-Davidson. They were the only American motorcycle company
and Erik had it very bad for motorcycles. Harley-Davidson would only
recruit locals and were not all that big on college grads. Erik had to get
himself to Milwaukee and talk his way in.
   That he did, and once there his talent and drive quickly made an im-
pact. He started out doing testing and analysis; his first job was track-
ing the oil usage on Shovelhead test bikes. This consumed a lot of time
as you might imagine. Shovels were never known for being oil-tight.
He quickly rose through the ranks to lead the chassis development
team for the FXR project.
   His personal racing was also in progress. He owned and raced a
Ducati and a TZ750 Yamaha, the latter running in the old Formula
One (F1) class. His racing program was moving along fairly well for a
couple of years but things were about to change.
   The career at Harley-Davidson was going well. He was leading the
chassis team on the FXRT project and as a result learning an enormous
amount about real world motorcycle chassis dynamics. The team
hooked up test bikes with recording gear and then flogged them around
the Talladega test track, recording what was happening. The result was
one of the sweetest handling Harleys of all time (these bikes became one
of the favorite rides of the Hell’s Angels because of this), and a head full
of knowledge that Erik would find useful for years to come.
   Erik’s career was going well but his racing was not. He was fast but
the Yamaha was old and wearing out. He was also concerned about
racing a Yamaha while Harley-Davidson was locked in a life-and-
death struggle with Japanese manufacturers; it was becoming increas-
ingly obvious that racing a Yamaha while working for Harley was not
appreciated by people in the corner offices.
   Erik considered two solutions to these problems. The first involved
an obscure British company called Barton Motors, which had built a
750cc square four two-stroke race bike. The second involved convinc-
ing Harley-Davidson that they should take the Sportster motor and
build themselves a real sportbike.

         This dyno chart is from an RW 750 run that was made to tune the exhaust
         pipes. By this time there was more Buell than Barton in the engine. It gains
     over 23 hp in less than 600 rpm when you hit 7500 rpm; ultimately gaining 90
      hp in 3000 rpm to reach 163.5 hp at 10500 rpm. This is manageable in a race
     car with nice fat slicks but trying to feed in throttle while leaned over on a bike
     was treacherous. Even with modern slicks this would be a handful; with 1980s
           tires it was a recipe for high sides. Buell was able to smooth it out a bit by
       tuning the pipes and carbs but it would never be an easy bike to ride. Before
                         Buell designed a nice stiff chassis, it bordered on the suicidal.

   Erik was watching the ambitious Barton motorcycle, which co-
starred with David Essex in the B movie “Silver Dream Racer,” when
the TZ’s chassis broke. The Barton looked to be the answer to his prob-
lems, so he bought the Barton bike in late 1980. Actually the bike was
just the beginning of his problems.
   The second solution got about as far as you might imagine. It would
reappear after the first solution collapsed but we will get to that later.
   The Barton bikes had shown potential with 500cc engines but the
larger 750cc displacement did nothing to help their already question-
able reliability. In addition, the chassis was decidedly marginal. Not to
disparage Barton—after all it was a couple guys in an ancient aban-
doned church attempting the impossible—but their lack of resources
resulted in a motorcycle that was not going to take Erik into the upper
reaches of the AMA pro ranks. However, it was fast. Erik felt he could
fix the problems. He was, after all, very knowledgeable about motor-
cycle chassis design.                                                        Carmine Vara and Erik produced
                                                                             these packets in an effort to attract
   Did we say the Barton was fast? It was very fast. Erik would eventu-
                                                                             sponsorship for what would
ally get it to 178 mph with a peak horsepower of 163 at 10,500 rpm.          eventually become the RW 750.
This was more than enough to compete with the ubiquitous TZ750 in            They got no takers.
AMA F1 racing. There was of course one small issue. Its very flexible
frame would wind and unwind itself under hard cornering. The result
was unpredictable handling. If the motor had given you a nice broad
power curve, you could ride around the handling. Trouble was the
power band was about as wide as a matchbook and when the power hit,
all the matches lit at once. Plus, it liked to seize—often and without
warning. The thing was a wicked, mean, and nasty SOB but when it
was right it had the power to kick the TZ750’s butt; a very seductive
package for a determined young engineer/racer.
   With Carmine Vara, his friend and partner in the enterprise they
called Ermine Racing, Erik struggled with the Barton through 1981 but
the beast was constantly breaking. Erik was constantly calling England
for parts. They persevered, intrigued by its potential.
   Then fate intervened. In early 1982, Erik learned that Barry Hart
(the mad Welshman behind Barton) had gotten a job offer from British
bike maker Armstrong that he just could not refuse. The idea of an ac-
tual steady income was too much to pass up so he took it. Barton parts
would become unavailable. Erik, as he likes to do, jumped in with both
feet, buying the whole Barton deal, lock, stock, and porous crankcase
castings. He really had little choice. If parts became unavailable, Erik
would not have a bike to race in F1. Going back to race a production

                                               Bill Meyer in his Buell powered Ocelot at Road Atlanta for the 1985
                                           SCCA National Championship races. Meyer was the last of the sports car
                                            guys to give the engine a try. Meyer found the motor “reliable and fast.”
                                           He won the SCCA round at Summit Point in August, but at the National
                                           Championship races the ignition system failed. That was the last gasp for
                                                                                           the Buell-Barton engine.

                                                    Not on Two Wheels
                                         Dave Ammen and Greg Rutherford purchased Barton engines to
                                         use in a SCCA D-sports racer. This class allowed 850cc engines of
                                         any configuration. The Barton had a real power advantage over
                                         other available engines but proved to be very fragile. “We were
                                         consistently faster than everyone else, but we could not get the
                                         engine to last,” Ammen said. “It needed more work and financial
                                         backing than we were able to give it.” Ammen had to make his
                                         own set of gears. “The gears were just not able to handle the
                                         weight of the car,” said Ammen. Fortunately he was in the
                                         business of making gears. He managed a couple of wins and
                                         qualified for the National Championships a few times but gave
                                         up after not being able to overcome the reliability issues.
                                            The 750cc engine also saw use in Nigel Rollason’s sidecar rig
                                         at the Isle of Man TT and it is there that Barton had its best
                                         results. Rollason managed a second in 1979 and a first and third
     This advertisement for the sports
                                         in 1986. He was consistently fast from 1979 until 1987. By the
          car motor ran in Sports Car
                                         time of his victory Buell owned Barton and some of the pieces in
                           magazine.     the winning chair were more Buell than Barton.                   ■

bike after running in the premier class was not an acceptable option so
Erik really had to buy Barton. He was convinced he could make the
business work and sell some motorcycles. There was a market for a de-
cently-priced replacement for the TZ750 in AMA racing. It took until
very late in 1982 to arrange all the details and then in early 1983 Erik
wrote the check. He was in the motorcycle business. Unfortunately, by
the time the pieces all arrived, the 1983 race season was half over and
valuable development time was lost.

Development of the RW 750
Erik labored on continuing to develop what was now being called the
RW 750 motorcycle. Then in the spring of ’84 he convinced Harley-
                                                                           It usually looked like this in the shop
Davidson to witness a test at Talladega (a track where Harley-             at the farm, where you can see one
Davidson rents countless hours of testing time) hoping to get them to      partially assembled RW 750 and
take the project under their wing and get him the resources to compete     one bare frame.

     When the first “production” RW 750
       was finished, Erik Buell and Dave
     Gess set up a photo shoot. Dave had
        access to a portrait studio at that
      time; it was decidedly marginal for
      what he had in mind but when you
     have no money, you have to make it
            work. (Making it work was a
            recurring Buell theme!) Dave
          borrowed yards of black velvet
     material from a photographer friend
        and draped the studio in it. Two
      exposures were taken, one with the
        bodywork off and one with it on.
       The result is this ghost image. Erik
     would later credit this picture as the
           inspiration for the translucent
             bodywork on the City-Cross
                            model of 2005.

in the Daytona 200. Doug Brauneck would ride the bike, Erik would
wrench, and Harley-Davidson would watch. They offered no money to
help the test along so Erik arrived on his usual shoestring.
   Brauneck ran a dozen laps and was very fast indeed. A speed of 178
mph was recorded on the radar gun and the handling was great. His
speed was such that Harley-Davidson began to believe that running
competitively at Daytona was not a fantasy. They wanted to see just
how the bike would do over 200 miles.
   This seemingly delightful result of the test was the RW’s downfall.
Because of the time constraints Harley needed to run the test that
weekend. The only tires available that would survive the banking for
any length of time lowered the bikes gearing. Erik, being broke, did not
have any gears that would allow the bike to run fast and stay below red-
line on the Talladega banking. Brauneck ran some very, very fast laps.
The engine seized. Erik fixed it. It seized again. Turned out one of the
rotary valves was incorrectly machined and that cylinder was leaning
out at maximum rpm. Harley was not happy; Erik was still on his own.
   Now a regular guy like you or I might say at this point “Damn, that
didn’t work. Glad I have a day job to pay the bills.” Erik is not a regular
guy. He decided that if he had more time he could get this thing work-
ing right and have bikes to sell by the fall. To get the extra time he quit a
very nice job at Harley-Davidson. (Actually, he tried to quit, but Jeff
Bleustein made him take an extended leave of absence so he would

                The RW 750 engine was a 66.4 x
                54.0 mm bore and stroke, rotary
               valve square four. It featured two
               twin-cylinder crankshafts geared
                together. There are four separate
                     cylinders. Many of the parts
                     were designed or redesigned
                  by Buell, including the pistons
                    and the flywheel. The engine
                  started life as a 500; there was
                 also an 850 version in addition
                      to the 750 that Buell raced.

                           Doug Brauneck on the RW 750 prototype during practice at Road America in 1984.

                                          Road America
     By the summer of 1984, Buell had fixed most
     of the RW’s problems and was running the
     bike in as many AMA Nationals as he could
     afford. The goal was to sell bikes. At Road
     America that year Doug Brauneck, one of the
     top privateers, was entered to ride the bike. He
     did a lot of practice laps and was impressed,
     but he was concerned about reliability and, at
     the very last minute, decided to stick with the
     slower but reliable Yamaha TZ750. Erik
     literally jumped into his leathers and ran to
     the start line. He only completed one lap before
     the water pump drive sheared. Dave Gess
     managed to get one shot of him aboard the
     bike. (See page 6)
         The next race was at Pocono. On the
     Monday before the race, Buell joined a few
     friends who had rented Blackhawk Farms race
     track outside of Beloit, Wisconsin, and ran a
     few laps to make sure he had fixed the water
     pump problem. The bike ran great until it
     seized in turn one and highsided Buell to the
                                                                       Erik Buell and John Hasty discuss the RW.
     ground. He broke three ribs. The bike did not
     run at Pocono.                                 ■

have the option of coming back after six months. He was not coming           The one-car garage in Milwaukee,
back.) His parting was on exceptionally good terms—something that            where Erik started to develop the
would prove important in the future.                                         RW 750, became increasingly
                                                                             difficult to work in as the RW 750
   To develop a race bike Erik needed cash, so he sold his duplex and
                                                                             project grew. This farm near
found a run-down farm to rent in Mukwonago, Wisconsin. The farm              Mukwonago, Wisconsin, was a
was a long way from any neighbors and had a nice garage building as          perfect location for the project. The
well as a barn. It was an ideal place to build and test a race bike. He      barn itself, on the left, was used for
would generate some income by stepping up his Pittsburgh Perfor-             storage of leathers, wheels, and
mance Products business.                                                     general parts. The small building in
                                                                             front of the barn was the office. It
   The RW 750 really was, as the hot rodders say, “sanitary.” Lots of at-
                                                                             may have started life as a milking
tention to detail is evident in the construction of the chassis. It had
                                                                             shed and is attached to the barn.
mostly straight tubes for maximum stiffness. The engine was mounted          The out-building on the right was
well forward to keep weight on the front end. The front suspension was       the actual “factory” where the
Marzocchi with the patented Harley-Davidson electro-pneumatic                RW 750 was made and early
anti-dive system (which Erik invented). The rear suspension was a sin-       RR 1000s were built.
gle Works Performance unit. The axles were thicker than usual 1980s
practice; the front was 20 mm versus the usual 17 and the rear was 25
mm versus 20. The bearings were out against the fork tubes. None of
this was common 1980s practice but resulted in a much stiffer axle. In-
creasing stiffness is a theme continued across the entire history of Buell
   The bodywork not only looked good but shows the beginnings of
Buell’s interest in aerodynamics. It started with the basic shape of the
1969 Harley-Davidson Cal Rayburn streamliner developed in Cal

      In the winter of 1984-85 Kevin     Tech’s wind tunnel. Careful attention given not only to airflow over the
Cameron authored a feature story for     bodywork but also through the radiators resulted in a very efficient,
   Cycle magazine. This portrait was
                                         aerodynamic package. At the time, most fairings were simply covers,
  taken for that story. You can really
    see the pride radiating from Erik    with little attention paid to how they actually worked moving through
  Buell as he poses with the RW 750.     the air. Some period race bikes were faster without the fairings.
                                            By 1984, the new bike was ready and media kits went out. Cycle
                                         magazine sent Kevin Cameron out to write a major feature article that
                                         appeared in the July 1985 issue. One bike was sold and a second sale
                                         was pending. It looked like the gamble was going to work. America had
                                         a new motorbike company. Then, in spring of 1985, the AMA pulled
                                         the plug on the class. Formula One would go away and the growing
                                         Superbikes would become the premier class. Overnight Erik’s market
                                         was gone and along with it his business.

                                  Alan Ladd aboard the Machinist Union RW 750 prepares to go out
                                     for practice at Daytona. Ken Winpisinger is standing at the left.

                             Machinist Union Bike
One RW 750 was sold. On January 14, 1985,              amount of damage done to the rest of the engine
this was delivered to Kenneth Winpisinger of the       when it seized. We kept having issues at the
Machinist’s Union. “Alan Ladd and I had been           next several races and it finally dawned on me
racing locally and winning a lot,” said                that they all could be traced back to that
Winpisinger. “My dad was president of the              seizure. There was a lotta stuff we should have
Union and they sponsored an Indy Car team.             done right away that we didn’t think about. It
I’m thinking to myself ‘they should sponsor us’        was a learning curve for all of us, that’s for
and then I saw a magazine story on the Buell           sure,” Winpisinger said.
and showed it to my dad over dinner. The Union            The bike did not qualify at Daytona but they
was always looking to promote American-made            were told to be on the grid to replace any non-
products. I jokingly told Dad, ‘I should get one       starters. A few guys didn’t make the race but the
of these and you guys sponsor us.’ He said ‘Go         AMA official at the line straddled the front tire
get it.’ I just about fell over. I called Erik and     and grabbed the faring not letting them go out.
bought one.”                                           He actually wrenched the fairing so hard he
   The plan was to run the 1985 AMA season             cracked it. They never ran a lap. To add insult to
which kicks off at Daytona. “It would be better to     injury they were listed as 80th, last place in the
start a new race program somewhere other than          official results posted the next day. “I was
Daytona,” said Winpisinger, “if you are not 100        furious, they wouldn’t let us start and then they
percent ready to go, that place will hurt you.”        said we finished last,” said Winpisinger.
   It certainly did hurt them. The bike seized in         The best finish all year was second in a club
practice out on the banking between NASCAR             race at Summit Point at the end of year. “I felt
turn three and four. Top gear was about 160-           we had a pretty good package by the end of the
170 mph. Ladd managed not to crash and they            season and was confident that we would have
got the motor back up and running by race day.         done well if the class had survived,” Winpisinger
“What I didn’t realize at the time was the             said. “It was very frustrating.”                  ■