I recently had the opportunity to test additional 23C Continental GP4000S tires, along with a retesting of my original sample after having been ridden ~200 miles as a rear wheel. I figured this would help give a good indication of both the repeatability of the roller testing and also an idea of the consistency across different tires of the same models. Here's how it went:
- 04/05/13 - New tire with ~20 miles of use - Crr = .00336
- 04/14/13 - Same tire after ~200 miles of use - Crr = .00343
- 04/14/13 - Tire used in Flo aero tests - Crr = .00344
- 04/17/13 - New tire, fresh out of box - Crr = .00334
So, across those 4 samples, we get an average of .00339 (I'd round to .0034, which happens to be the result and number of digits I report in the spreadsheet) and a standard deviation of .00005.
If I'm doing my stats right, then this means there's a 99% confidence range of .0033-.0035.
Granted, this is a fairly small sample set, but it matches pretty well with my "gut feel" that the measurements reported in my Crr spreadsheet should be considered to have a tolerance of around +/- .0001, and that tires listed within .0001 of each other are basically "tied".
I also acquired 20C Continental GP4000 in the black color. My intention there was to first confirm that the black color GP4000 20C tires have the "Black Chili" tread compound (They do...it says so right on the package), and additionally to see how well it rolls. The idea was that since it has a similar shape and tread markings as the 23C tire, then it possibly would work as well aerodynamically on narrow rims as the 23C tire appears to do on the wider rims.
The result? 20C Continental GP4000 (Black) - Crr = .0041
That's basically the same as what I found the old 19C Bontrager AeroWing TT tire to exhibit (.0043), in which case, I think I'd still prefer the 20C Continental SuperSonic (Crr = .0034) tire for narrow rims, especially for front wheel uses. As we learned in my last blog post, it would take a LOT of aerodynamic advantage to make up for that much of a Crr difference.