November Dave wrote:
Flo is a MUCH better A2 customer than any entity involved in yesterday's test, and for what it's worth we weren't the paying customer. A2 wouldn't have had any idea which wheel to influence yesterday if they were inclined to do so. But just to counterpoint skepticism, if A2 were inclined to fudge results, don't you think the Flo 30 would have done "better"? That result has the potential to piss off a customer who does a lot of work at A2.
As to your second point, yes, I agree wholeheartedly. With every hair on my head, and there are fortunately still tons of them.
A2 isn't in the business of fudging numbers. We've easily spent enough time there to determine that. A2 runs the tests that you ask them to run, and gives you the data. What does this mean? It means that the people designing/running the test are 100% responsible for the data/content they are producing. We've always chosen to be as transparent with our testing protocol as possible, for obvious reasons. I'm certainly not trying to take a stab at you, or suggest that your testing protocol was poorly designed. I'm just brining these points up because a lot of people don't realize how easily wind tunnel numbers can vary, even when the person designing/running the test has the best intentions. Here are a couple examples of some things we've learned and how easily numbers can be inadvertently be skewed.
1.
Using the EXACT same tire. Two tires that are identical in make/model/size can vary by up to 100 grams of drag in a test. Wind tunnel time is very expensive. As a result, testers will often use two or more brand new tires and rotate between them, not knowing there is up to a 100g difference between the tires.
2.
Setting Pressure. How you set your pressure matters. Our tests have shown that a difference in tire pressure of only 5psi can effect aero drag by more than 90g. Even when using the exact same tire and wheel. Why is this important? Our good friends at Silca tell us that the standard floor pump - used by many testers - has an accuracy of 5%. If you are inflating to 100psi for your test, your accuracy variance could be skewing the numbers by close to 100g of drag. This is why we use Silca's own testing rig "The Truth" for our tests. Here's a quote from Silca's page regarding "The Truth":
Quote:
We built a gauge setup we jokingly called 'The Truth' using a $500 Ashcroft 0.1% accuracy digital gauge, an old SILCA disc adapter and some precision industrial components to create a gauge with very high accuracy bleed. The Truth was capable of bleeding pressure at a rate which could yield repeatable 0.05psi readings. -
full article here 3.
Compound Effect. Take the "bad" tire from example one, and combine it with an inaccurate reading from a floor pump in example two, and you could have nearly a 200g difference in drag, even with the best intentions.
As I said above, I think the only way to really compare brands is to have everyone at the same tunnel, using the exact same protocol and taking their best shot. We personally haven't studied Zipp/HED/Bontrager/November/etc wheels to the nth degree, so we don't know what tire, at what pressure performs best on each of their wheels. Because we don't know these facts, we don't feel it is fair to test against them.
Chris Thornham
Co-Founder And Previous Owner Of FLO Cycling