Very interesting articles. To answer Tom's question, I do feel that aero testing is quite relelvant, but the German experiment is what really fascinates me over the windtunnel. Running a rider in an inside velodrome at a constant speed while measuring wattage effort and heart rate strikes me as an almost bullet proof method of testing wheels, frames etc. If higher heart rate and the wattage required to maintain that speed on frame Y is more than frame X or with wheels, it strikes me that we can easily assume that one is more aero than another, or at least faster than another. Perhaps there could other factors, say wheel rolling resistance for example, but at any rate it demonstrates that one is faster than the other and as a consumer on my way to Tom's shop to make a wheel set purchase, then this is all I want to know. We may not be able to get exact drag co-efficient using this method, but do I or any other Joe Consumer AG tri-geek really care when all we really want to know is which is faster.

I'd love to see this kind of an experiment set up over here. I wonder what Andrew Coggan woud have to say about this approach.

Whatever happened to Andrew?

---

Tom Demerly
The Tri Shop.com

Dunno where Andrew went to but he would certainly be the guy to ask about this approach since he is an exercise physiology PHD working in academia. Let's hope he shows up.

Tom, Another thing I noticed the other day while thumbing through tri mags while on the throne: When one looks at photos head on or from frontal oblique angles, it really jumps out at you how little frontal area ANY bike really presents to the wind when compared with the huge hulking un-aero bear riding it! The more I look at it, the more that just really sticks out. Most of us aren't anywhere near our peaks as far as physio, so shaving seconds from a 40k by spending $1400 on a set of Zipps w/tires seems silly to me. Sure, if one has the $$ spend it, as we've said elsewhere. I won't look askance. But those Zipps are not the reason you improve your Oly time 30 seconds. Like the "magic" diet pills, we're only chasing a dream...

Make sure your bike fits, use aero bars, and train your ass off!

Just read through the first article and there are some interesting point made. .....

Frame drag and wheel drag seem to be the same - 8% of your power and the rolling resistance was 12% compared to the rider which was 60%. The area of improvement though was interesting. When the frame drag was reduced by 50% the percentage of power required dropped 3.&5, when the wheel drag was reduced the improvements were 1.8% for the front and only 0.6% for the rear (this would support the thread from a couple of days ago where we all said buy the front wheel first!) but the rolling resistance improvement was 9.7% and the rider position was 26% when each of these factors were improved by 50%.

..... this would indicate that working on your position and buying really good tires has a much bigger impact than a radical aero frame and wheel set (and it would be a lot less $$$$$). The other interesting factor they had was dropping the rider weight by 9.9lbs to save 1.8% so diets are probably better than Ti cassettes :)

...cool article!

Andy hangs out on the "wattage" forum at www.topica.com. He posts at least 4-5 times a day, and is one of the most prolific posters there. Its fairly elevated discussion about power and bicycling.

I too am very skeptical about the applications and/or answers being derived from controlled, wind-tunnel studies to the un-controlled outdoors world.

I have seen first hand how Cobb and others do wind-tunnel testing and it has little to resemble how you or I would ride our bikes on the road.

For one, the bike is pinned down. No bike movement at all. That usually never happens in the "real world"! Then, the wind comes predominantly from one angle. Again, not exactly what you experience on the road. Third of all, the rider and bike are always on flat ground. Again, not exactly the "real world" (unless you live in FL!!). Then ofcourse, you have to look at the fact that nobody else will ride in exactly the same position as the rider being tested, since few of us have identical body measurements or riding styles. Finally, I would be shocked if there were no errors in all this testing. Not sure you on earth checks all those calculations and data.

I'm sure there are even more reasons why to be skeptical about wind-tunnel testing. But hey, it helps to sell bikes and other equipment and makes for some great debates.

i think those are good points by bearis, and i am wondering some of the same things. particualrly that bit about the bike rocking and moving around - very insightful !! over at tno mr coggin has responded to a question of mine: what if i had a (cheap) steel bike made using the exact geometry of , say, a p3 made and rode it with the same position/parts/clothes in an IM? the reason for doing so we could say is purely cost savings. there are numerous places youcould get such a frame for well under $1k vs close to double that for the p3 itself. mr coggin says the difference between the two is 4 min in an IM. so then the cost is over $1K difference - and, thinking along the lines of bearis i kinda wonder about that 4 min. at MOO for example a fellow is turning and climbing and passing and getting out of the aero-position to eat, and going thru towns and barely eking up some of those hills in the woods with no wind around or a tailwind. it seems reasonable to calcualte down from the windtunnel 4 min in light of these things........let us say 2-2.5 min, then??? if that ??? it has been and is an interesting topic.

I think there is too a point, but after that it gets into serious voodoo. Cobb's article (trinewbies) is a great prover (is that a word) that someone who is TOTALLY out of the ballpark can see pretty significant gains, but once,you hit that point, it starts to become very negligible - diminishing returns. I think (esp. in the IM world) other things have to factor in like comfort, weight, type of course, and so forth to get a fully accurate picture. Aero is a good chunk of the equation, but by far not the only one.

---

---

"To give less than your best is to sacrifice the gift." - Pre

MattMizenko.com

I think it's quite valid even though I like the above mentioned German experiment better if I had a choice as a way to get info regarding bicycle frames, wheels, etc. I'm not an engineer but from what I have read, am under the impression that the wind tunnel/math modeling approach is fairly valid despite some unique challenges testing bicycles in a wind tunnel. The Williams Formula One team just invested around 25 million for a new windtunnel in their attempt to catch up to Ferrari aerodynamics, so there must be something legit to this.

The problem is with bicycles that there is no independent group doing aero work making this info readily available for us consumers. This is being done by only a few people who have commercial afiliations. For example if company X hires John Cobb to test their new gollygeewhiz wheel or frame that they are marketing as superduper aero. Cobb's info remains the property of the company and if it turns out that John finds the new wheel or frame isn't really that aero do you think the company will let us know or change their advertising. I don't think soooo!

This is why I respect Hed so much. Their aero data is to be found on their website. None of the other wheel makers are doing this. Gerard from Cervelo is supposed to be doing aero testing at the MIT windtunnel this winter. I'm hoping he'll make his info available.

Relavence is an interesting word to use...

If I understand what you mean I think that the question should be... Is Aero Testing predictive?

Having said that I dcided to put my 2 cents in.

When you are operating in the "real world" of actual performance there are additional factors that come in to play. These include things like psychology (how does your bike / wheel / aerobar make you feel) There is something to be said about feeling fast, or confident that your equipment is not going to hold you back...
There is also the factor of Fit Fit FIt... Not all of us are 300 pound clydesdales... Personally I am a small person that tips in between 135 adn 140. If you are talking abotu relavence threre is also the factor of handling... kinda hard to go fast when you have been blown off the road and your bike is broken into 5 pieces.

Relavence... has the implication of being personal. Is having the most aero thing out there "relvent to you" Considering my budget and my goals in tri, I would say no... someone else may say "hell Yeah"

Which brings me to my non-conclusion and the end to my rambling... Aero testing is like Lactate threshold testing or muscle content testing... it may give you an idea about your maximum potential but they mean absolutely anything in the real world. In scientific measurement there is no such thing as an absolute value for anything that you measure. This is why you owuld see values like X + or - a... when you go into the real world the value of a changes incredibly due to other factors (rider, clothes, conditions, etc. ) The fluctuation of a would be much graeter than any measured difference between bike frames or wheels.

actually le me ramble on for a little longer...
The mentality with competitors and aeroness reminds me of a arms race mentality. Everyone else is riding this or that... gotta keep up or else I will get left in the dust...

food for thought... (if there was a complete thought in there somewhere)

As for what Andrew Coggan thinks about the German study, just go back to earlier threads and find out. He participated in aero testing by measuring power and heartrate and has written about it numerous times, and I don't think his opinion has changed.

The basic problem with this discussion is that it's about the concept that real life measurements are not necessarily the best way to measure real life. It's why some medicine experiments are better done with patients in a controlled environment, even though patients will eventually use those medicines at home. Testing is all about reducing the noise, isolating the proper measurement. Also in an older thread Andrew explains this quite clearly, much better than I could.

If two equal cyclists race each other and one has slightly better equipment, will he always win? No, not if he encounters a few strong gusts of wind that his opponent didn't? Does that mean that his equipment wasn't really better? No it means that there are several factors and equipment is only one. But suggesting that to test in gusty condition would improve the accuracy of the measurements is twisted logic.

While a velodrome is more controlled than the normal roads, there are still many changing factors even on a velodrome, which frustrate attempts to measure accurately. Of course with enough time it can be done, just like plain outside testing is accurate as long as you do a couple thousand repetitions.

The real question how ever is, if you test the same set-up in real life a few thousand times and get to a "final result", what would give you the best estimate of that result, a windtunnel test or a real-life test? Here on slowtwitch Andrew indicated he would choose the windtunnel without hesitation, and based on my experience with both methods I would have to agree. Not to mention that in a velodrome you may get yaw angles of 10 degrees, whereas in the windtunnel one can easily simulate real life yaw angles of say 30 degrees.

---

Gerard Vroomen
3T.bike
OPEN cycle

"Maybe I am just being too cynical ;-) "

Just maybe, as if your average Joe tri-geek is going to be able to take fifteen different wheelsets or frames along to test. That's why I like Hed for making their info available to us.

Would be interested though in your disc vs wheel cover findings. I believe that Cobb did this and felt that air entered around the circumference of the wheel cover nulifying an aero benefit. There is however, a $.98 solution called black electrical tape. Just touches the rim and forms a seal with the cover. I've tried it and it and it holds the duration of a tri easily. Not every solution has to be rocket science.

"As for what Andrew Coggan thinks "

I hadn't seen the previous posts and was unaware that Coggan had done work in this area. That was what I wanted to aske him about.

BTW, how is the Cervelo wind tunnel testing at MIT coming along? Have you started it yet?

i am still wondering about part of this point:

".... suggesting that to test in gusty condition would improve the accuracy of the measurements is twisted logic. "

respectfully, i do not see why this would be. gusts are part of riding outdoors. the issue is not that one rider hits a gust and another does not. the issue is whether the things that make a bike fast in the tunnel still do so in different conditions than the tunnel tested. to me it seems a sort of blind leap of faith to say that a bike that is fast in smooth winds at measurable yaw angles is also faster in gusts of winds hitting it from every which way. i don't have a solution but it doesn't seem that twisted to suggest that maybe under gusts from various angles and for various durations certain things might shine that otherwise would not, and vice versa. you can test your f1 motorcycle in the tunnel but if it makes it harder to steer you go back to square one. not that aero frames are hard to steer - the point being perhaps under such (nonlaminar winds) conditions some of the mathematical extrapolations taken in the tunnel concerning the bikes slipperiness lose something, or even reverse themselves !

maybe not, i really don't know. i am asking things that are on my mind. if gerard or mr coggin have covered this in their reseach or papers i apologize. still, there seems to be SOME confusion on the issue or contradictory views held by several key players mentioned previously (c-dale engineers, east european framebuilders.....) so i'm still left wondering, like i say.

oh, and i thought the post on bikes rocking side to side ( and, presumably also steering the front wheel about its axis) raised awhole passel of interesting ramifications. what of it? how aero is a frame while rocking to and fro, with the front wheel moving back and forth across the down tube?? wouldn't a test of a bike held still miss this entire issue? and yet, intutively it seems to me that if you take a front whell and rotate it in front of an wing shaped tube like you actually do when riding it is gonna do SOMETHING to the airflow over said tube ! ditto the rocking thing. lots to wonder about.

[reply]oh, and i thought the post on bikes rocking side to side ( and, presumably also steering the front wheel about its axis) raised awhole passel of interesting ramifications. what of it? how aero is a frame while rocking to and fro, with the front wheel moving back and forth across the down tube?? [/reply]

Of course that has an effect, and it can be measured, but again much easier in a windtunnel than out on the road. After we had developed the Baracchi, we tested exactly this. Since the frame was designed to pick up the airflow coming off the front wheel, it was very important to test it at the front wheel angles that one encounters in the real world.

The same goes for the gusts argument, if you want to test the performance of equipment in gusts, you need to create carefully defined gusts in the tunnel and expose all the models you're testing to the same gusts. That will give you the answer, testing outside in gusty conditions won't.

---

Gerard Vroomen
3T.bike
OPEN cycle

a hearty thanx to gerard and mr willet for their time and patience. very interesting lesson, guys.

Good references. Just what I was looking for. Thanks.