carpediemracing 

I'm not a physicist but something doesn't seem quite right with the experiment, although I think it's great that you did it. When I find a physicist or someone who actually knows vectors and stuff I'll have to report his/her findings or explain the flaw in the experiment.

Perhaps it is a percentage thing (5W over 7W = 29% decrease in required power)? Maybe acceleration related to time? The speeds used in your experiment? Not sure.

It seems illogical esppecially when looking at cars. The effective savings of horsepower at the wheels of a 14 pound lighter flywheel (14 lbs vs 28 lbs stock) in a car can gain something like 100 hp depending on rpm/gear (or as little as 2-3 HP). In other words, the force required to accelerate a wheel (or a car's drivetrain) increases radically (exponentially?) with rpm. The reported "performance increase", done by alleged independent shops, finds that a light flywheel will save a lot of HP (into three digits) in first gear but by fourth or fifth gear, there is barely any difference. This is because at the higher speeds the engine cannot accelerate as briskly and so the time required to accelerate is much longer. Therefore lighter weight doesn't matter.

Likewise but in an opposite direction, a car magazine, looking to build a low-buck140 hp or so car (handling + acceleration) did a month-to-month article of upgrades. They modded the suspenesion, engine (minor mods - exhaust, intake, etc), trying to stay within some budget or some rules. At the end of the engine "stuff" they found a measurable but small HP increase (at the wheels it went from 105 to something like 112 HP, I think about 7-8 HP increase). The next issue, in order to put big brakes on, they put 17" wheels (from 15"). Their measured HP dropped back to, or maybe below, the stock power reading. This is because the HP is measured at the tires and the larger (and heavier) wheel/tire combo ended up soaking up a lot of HP. They tried to justify this as being acceptable for the gains in cornering and braking traction but in the real world, if you look at autocrossers and drivers who regularly do laps on the track, they use smaller diameter wheels with the same width tires - the wheels are much lighter and therefore there is a corresponding increase in performance. So they'll put the 15" on for the race or laps, then the 17" for the drive home.

Having said that, regarding the original post, wheel weight/aero requirements depend on the course. If you are racing a cross shaped hot dog course (i.e. imagine tracing a route around a plus sign), the four 180's plus four 90 degree turns will probably favor a lighter wheel, aero won't be as much a factor. If you're racing on a kidney shaped course with no turns, then aero will be more important than acceleration.

If you're really good, you can do what Coors Light did in the early 90s - go to the front with your front and rear TriSpokes, go 35 mph for a while, and eventually blow the field apart until 4 or 5 Coors guys are together with maybe an odd rider for company. Ride away from field and let designated leader win. With an aero wheelset, no matter what the weight, if you can stay at the very front of the field, you can go through most turns at full speed. The lack of deceleration and acceleration will negate any advantage a light wheel (might) have. This happened at the second last all-road Tour of Michigan, almost all eight stages, and they did that even on a 1/2 mile course, 8 turn, narrow road, rainy and windy day.

cdr