This post was written to answer a question that I was asked. For the benefit of people without curve fitting software, I did some basic analysis of aerodynamic data* for rims and have a sort of opinion here. (Also, I made this for fun since I finally have MATLAB 2010 which supports W7 on my laptop! Yes!!!)
Click to enlarge
My statistics sucks... but maybe someone can make better work of this: the norm of residuals is 15.053 (higher polynomials are badly conditioned), and the standard deviation in y is 3.999... which should imply that the fit is 74%, which is a pretty strong correlation.
The other thing I want to suggest is that we can figure out the range of rim depths that a wheelset for overall usage should fall under. From the equation, we understand that there exists a region $37.5 \leq x \leq 53.1$ where a relatively large change in rim depth incurs only small aerodynamic benefit/loss of $\pm1.0W$ at 50km/h. Now, waiving all the assumptions that we have to clear, this sounds very reasonable indeed:
Hill climbing wheelsets: <37.5mm
General-use wheelsets: 37.5 to 53.1mm
TT wheelsets: >53.1mm
Now, my point was that the Zipp 303 has kind of disposed 404 as the better overall wheelset.
Now, my point was that the Zipp 303 has kind of disposed 404 as the better overall wheelset.
*Compiled from Roues Artisanales. I removed 2 outliers from the data: Ambrosio X-Carbo '05 (46, 31.5) and Zipp 808 (82, 16.7).
No comments:
Post a Comment