As far as the athlete is concerned, it seems to me that field testing (especially since there seems to be a very sensitive tool if one takes enough care) would be far superior to wind tunnel testing for several reasons.

1. One is getting effort into the equation. It does no good to have the best aero position if one can only generate 50% of the power one can do at a slightly less aero position. Field testing is likely to discover this. Might be harder in a wind tunnel.

2. Bike handling skills are also put into the equation. An aero bike or position that cannot be ridden in a straight line may be slower overall also.

3. The only cost is the time cost, at least once one has the power meter.

4. There are probably other reasons I cannot think of right now.

However, it seems to me that field testing would be particularly sucky if one was trying to determine how aero the frame was, since so many variables that could interfere with the result cannot be well controlled.

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

Hey Tom,
The times seemed to be about 1 to 2-1/2min slower for most. Fastest time last year=46:?? for Thurlow, Fastest time this year 48:16. Thurlows timw this year 48:50?

Also just to fuel the fire, I went 2:38 seconds faster this year ON A P3C compared to last year, on an Abici TT.

Pretty much, and I do like your use of analyticcycling to "ballpark" the estimated savings. That said, those aero gains were also basically in line with what my test measurements said they would be in that they were ~2 to 2.5 s/km. That's pretty damned good agreement, if I must say so myself ;-)

Just so we're all "talking the same language", I'll run through a similar estimate and give all the inputs. The main assumptions are that it is a flat course (it's not, but that's OK in doing this comparison) and that the wind conditions were similar (they weren't, but I'd say the wind conditions in 2007 were more favorable...so this will be a conservative estimate of aero gains).

So, in 2007, using the following inputs:

Area = .225 m^2
Cd = 1 (this plus the Area were obviously chosen to make the CdA product the measured .225 m^2)
Air Density = 1.08 kg/m^3
Mass = 84 kg
Crr = .004
Slope = 0
Power = 230W

That results in a predicted speed of 11.64 m/s, for a total time over the 37.1 km course of 53:07

Now, in 2008, the temperature was 20F lower and the barometric pressure was higher, which resulted in the air density increasing to 1.124 kg/m^3. Another temperature effect is on the Crr. A commonly accepted compensation for temp is 0.6%/deg F. This means the Crr would increase to .0045 from .004.

So, if I'd run the same setup as 2007 and put out exactly the same power, just from atmospheric changes alone (with effects on air density and Crr) the calculator predicts a speed of 11.41 m/s, for a time of 54:03 over the 31.1 km.

That means the course was basically 1 minute slow this year just due to conditions. In reality, I think it was even slower than that due to the slightly different winds, but we'll disregard this since it just makes the potential savings more conservative of an estimate.

OK...I went 1:45 faster on a day that was 1 minute slower....that means my "total" time gain over last year was 2:45. Checking that with the calculator, if I hold the CdA the same and just increase the power, it comes up with a speed of 11.74 m/s, which would be a 37.1 km time of 52:40, or basically 1:23 faster.

So, of the 2:45 of time gain, basically exactly half (1:23) was due to the higher power (just as you found), which leaves a 1:22 gain due to aero changes, or 2.2 s/km. Well, the ROT says that should be equivalent to a CdA drop of .022 m^2, which would put my 2008 CdA at .203 m^2. Hmmm...does that number look familiar?? Pretty close to my measured .205 m^2, right?

If I plug a CdA of .205 m^2 into the calculator with my 2008 power it comes up with a speed of 12.08 m/s, which would be a predicted time of 51:11 vs. the 52:40 with a CdA of .225 m^2...or, about 1:30 faster. That's pretty darn close to the 1:22 difference calculated above, huh?

Now...I know the bike was different (Soloist vs. P2K), but measured CdA of the setup I ran last year was within .002 m^2 of what I measured for the P2K, so it's safe to say that I would've been the same speed on either bike. My position (as measured by the location of the "touch points" relative to the BB) was identical.

The evidence that the vast majority of the aero gain was due to the different frame is pretty damned compelling, if you ask me...

---

http://bikeblather.blogspot.com/

See my post to Frank above...my opinion that the conditions were at least a minute slower just from atmospheric effects (temp, density, etc.) and probably greater than that due to less favorable wind conditions.

From what I understand...that "fastest time" from last year may be off by a digit in the minutes column. If you think the "timing issues" were bad this year in the 45-49 group, you should've seen last year ;-)

BTW, good job out there!

---

http://bikeblather.blogspot.com/

Let us presume you are right. Your body position is optimum and all the aerodynamic improvements came from the frame. Since you didn't win the race, unless Cervelo has a few more miracles up their sleeve, what is says to me is, if you want to see any more improvements they are all going to come from the engine. The bike can do no more for you. Might want to consider PowerCranks. :-)

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

Hmmm...so you mean to say that if I put the "touch points" of the 2 bikes to within a millimeter of the same location on both bikes that I'll somehow be able to sit on the P3C in a manner that's faster than the way I sit on the P2K, which is a bike and position I've "optimized" over the last year and a half?

And those minor, if any, differences are going to DROP the measured CdA as large as the drop that was measured, or even 1/10 as large? All the while my measured drag on the P2K is consistent with past measurements (meaning I didn't "sandbag" the P2K)?

Puhleeeeeze...if I could make minor body "shifts" that could gain even 1/10 to 1/5 of the measured difference, I guarantee you I would've found it in the past 1.5 years of monkeying around testing my stuff. You're reaching...

---

http://bikeblather.blogspot.com/

Frank: what makes you think that Cervelo doesn't have a few more miracles (e.g., a P4C) up their sleeves?

Tom: good ride, and thanks for sharing all the data!

I don't. However, I feel quite confident that when they get the CdA of the bike down to zero that further improvements can only come from the engine.

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

That's hilarious! Whenever has this thread been about saying that a measurable reduction in frame drag means you're guaranteed to win a race? Besides, didn't we just go through the exercise of determining that ~half the apparent gains in my performance from 2007 to 2008 were just from the power increase?

If you want to go faster, you can address the demand side and/or the supply side. Didn't I do both?

I plan on still doing both in the future.

Well...it's a sure sign that you've "won" an argument with someone when they change the subject... ;-)

---

http://bikeblather.blogspot.com/

Let me ask you this. Presume next year you are still on the P3C. How is it you expect to further reduce your CdA? What are you going to do? If you think it possible to further reduce your CdA, how is it you are so sure you haven't done some of those things unconsciously when you moved to the P3C, making the frame seem better than it really is?

---

--------------
Frank,
An original Ironman and the Inventor of PowerCranks

Because none of those "things" I plan on exploring are related to my body position.

For example, on neither of the bikes are the upturns of the Vision base bars trimmed off. I've got some other equipment ideas as well. There's plenty of room for improvement left in that area IMHO. I always think of Andy telling me "Sometimes the lightest, most aerodynamic bike part is the one you leave on the bench" ;-)

Also...you forget, it's not my P3C. I borrowed it. That means I'll be in the market for something of that aerodynamic "class" or better. You'll be sure that whatever it is, it'll get fully tested out.

Please...give it up...I held my position the same in both runs...and in both of the races that were used as "backup" data. It's a "dead end" for you...

---

http://bikeblather.blogspot.com/

Uh, yeah. I didn't know there was an expiration date on wind tunnel data. Is there something you know about that data that invalidates it?



Well, for starters, the target audience for P3C marketing isn't exactly people riding round tube frames.




Maybe. But weren't you making that guess in a sarcastic mocking fashion?



Based on the wind tunnel data, I think I'd be about 2 to 4 minutes faster over 40km. Based on the data of you and Frank talking out of your a$$, my estimate would be revised downward. ;)


Just totally guessing? You don't say? Because your gut tells you so?



Give me your CdA assumptions for both scenarios (preferably backed up by testing) and we can discuss. Until then, you are just talking out of your a$$.



Congratulations - you just won the Frank Day Award for a Complete Lack of Understanding of Physics!

Rik

So it sounds to me like you're saying "yes."

But that would also mean that you don't believe in wind tunnel testing for body position since that's not blinded, either. That's fine -- I was just trying to understand the basis, and consequences, of your critique.

Field testing has some advantages but one large disadvantage: it's hard to do controlled yaw testing. As we're seeing from Tom's tests, there can be pretty sizable differences between straight-on drag and low-angle drag. Which is not to say that yaw testing is impossible -- it's that controlled yaw testing is very nearly impossible.

Taking into account that the times were off, and listed fast by :30s last year, my 49:54 would actually be 50:24.

This year I went 50:28, but it took me an addl 10w to do it.

Wind gusts were an issue for me, I came out of the bars on corners 3 and 4. The 1080 is fast, but the 808 may have been a better choice with the gusts.

Yep...which is why at Greg Steele's constant urging, I'm seriously considering purchasing a data-logging portable weather station :-)

---

http://bikeblather.blogspot.com/

:P

I once asked kestrel to send me a dozen so I could set them up every .1km along a ~1km course to confirm my assumptions that the measurements at the s/f area are ~= to measurements along the course over time...

still only have one :(

goes to alex's new thread... mine doesn't look as pretty as his yet, but similar idea...

g

---

greg
www.wattagetraining.com

What an iBike Aero?

That also could be an option...I just need a ANT+ Sport compatible power meter to go along with it (ouch!) :-)

The nice thing about the weather station is that you can set it up while warming up and gather some pre-race data. That might help in some equipment and/or pacing decisions for the particular race.

---

http://bikeblather.blogspot.com/

Well, ANT+ would certainly be easier and slicker but with a portable weather station you'd still need to do post-ride analysis anyway.

Of course, aero positioning might screw up the iBike's wind pressure port.

I'm not saying "yes", I'm saying that research results are always subject to uncertainty and bias, and that large potential sources of bias are ignored by guys who do wind tunnel testing and the sort of field testing you describe. Yet these sources of bias are acknowledged to be potentially large in other fields of research, such as medicine and psychology. Why are they ignored by bike testers? Probably because they're inconvenient to deal with, and because dealing with them makes the data less malleable by people with a vested interest.

other than literally riding blindforded - which is clearly inane - how do you propose to get around this problem?

So it sounds to me like you're saying "yes."

But that would also mean that you don't believe in wind tunnel testing for body position since that's not blinded, either. That's fine -- I was just trying to understand the basis, and consequences, of your critique.[/reply]I'm not saying "yes", I'm saying that research results are always subject to uncertainty and bias, and that large potential sources of bias are ignored by guys who do wind tunnel testing and the sort of field testing you describe. Yet these sources of bias are acknowledged to be potentially large in other fields of research, such as medicine and psychology. Why are they ignored by bike testers? Probably because they're inconvenient to deal with, and because dealing with them makes the data less malleable by people with a vested interest.[/reply]
So you're not saying "yes, you don't believe in wind tunnel testing for body position" you're saying:

"Because the rider wasn't blinded to the bike he was riding, there is a possibility that he would hold his head a touch lower, hunch his shoulders a little more, or otherwise make subtle improvements to position while on the "faster" frame. He may deny that he did it, or he may not even be aware that he did, but he may have done it and you have no way to control for it."

"The subtle changes wouldn't have to be that large to completely invalidate any estimate of time savings over 40k."

"While the lack of blinding may not explain all of the effect, it may explain some. If it does, then any estimate of the magnitude of difference [..] has to be taken with a pinch of salt."

"So how much of a pinch of salt do we need to take it with? Oh that's right, we can't accurately adjust for threats to validity - that's what makes them threats to validity. So we don't actually know with a huge degree of certainty that there was any difference"

"large potential sources of bias are ignored by guys who do wind tunnel testing"

So how is this different from saying, "yes, I don't believe in wind tunnel testing for body position?"