All samples described below were tested at the Oran W. Nicks low speed wind tunnel at Texas A&M University using dedicated cycling fixtures in January of 2003. Data is presented as axial force (which is equivalent to drag at 0 degree beta/yaw) at 30 mph. Relative ordering will be the same at lower speeds, but the magnitudes will be decreased.
Chester Kyle has previously indicated that the tire interface with the rim is an important detail to be managed when trying to minimize axial force. Details about this tire effect and its magnitude are shady. In an attempt to quantify whether or not tire size even matters, I executed a test schedule that would shed some light on the issue.
Two distinctly different aero wheel constructions (the DuPont Three-spoke clincher, and the Zipp 404 clincher) were tested. The same wheels were tested with different size tires (700x20c and 700x25c), but of the same tire brand and model. The tire used was a Contintental Ultra 2000 kevlar beaded tire inflated to 110 psi. The wheels in the test are shown below: