Slowman wrote:
if you look at BRR's tests - which we all have - one thing that sticks out obviously is that higher pressure always wins. even on larger tires. this speaks to a flaw in the protocol. drum testing - or at least BRR's drum testing - can't tweeze out "break point" to use jargon coined by a person participating in this very thread. when i look at BRR's data i stipulate to its reliability when testing compounds. but not pressures and since it's not reliable for pressures i'm highly suspicious of its reliability on tire widths.
Ok...I have to address this, since it's a common refrain from some quarters ("Quarterly's"? <LOL>) that because breakpoint pressure doesn't show up in Crr vs pressure testing on rollers, then ALL of the data from them is "suspect".
Of course, that conveniently ignores the fact that the only reason the concept of breakpoint pressure was identified was because there was a divergence in the correlation between roller testing and field testing above a certain pressure (for a given road condition, setup, speed, etc.). In other words, you can't accept the concept of a breakpoint without acknowledging that below that pressure roller tests and field tests are highly correlated.
Couple that with the fact that for most reasonable surfaces (especially for TT or Tri events) the breakpoint pressures end up being at a point quite a bit higher than most would suspect, then it's logical to utilize roller testing to discern Crr differences for reasonable pressures...especially if one understands that due to the highly asymmetric loss function of running to high of a pressure, that it's prudent to err on the side of too low of pressure than too high.
So...look at the results in the lower pressure ranges and be confident the results reflect "on the road" performance. It's that simple, really.
That said, the "rolling road" testing is really the same as roller testing, just with less sensitivity to differences in Crr than rollers (due to the lack of the "amplification effect" of the rollers). My questions about the 28 vs 30 results further up in the thread still stand...they go against previous results on tire width (edit to add: and logic).
http://bikeblather.blogspot.com/