Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout

Triathlon Forum
21

Whenever this topic comes up I always think we should view the wheel + tyre as a system. Test each wheel with the very best tyre that the manufacturer can recommend, or which has been shown in testing to work best with that wheel.

I mean, the idea of the thing is to find out how to go as fast as possible so just test everything in its fastest configuration.

If the manufacturer cannot recommend the tyre that works best with their wheel, or no additional test data is available, then screw 'em they can have a 28 on there as punishment because if they are trying to sell something to us based on its aerodynamic merit then they should have learned this by now, and be able to provide an answer.

22

On the XXX4, they actually brake pretty well - light years better than the old Aeolus’, where you pretty much had to pray. Haven’t ridden them in the rain though.

For the “old Aeolus” do you mean the Aeolus Pro 5 and similar? This one:

https://www.trekbikes.com/us/en_US/equipment/cycling-components/bike-wheels/road-bike-wheels-wheelsets/bontrager-aeolus-pro-5-tlr-road-wheel/p/22641/?colorCode=black_grey

If so, that’s disappointing, because that price had been tempting me for a while.

23

I am not sure what can be learned from such incomplete data. Yaw angle is obviously a huge factor, which model 808 and 404? Which hubs? What fork were these on and at what blade width? What happened to Hed and Profile Design?
There are simply way too many unknowns to draw any conclusions.

It’s been a few years, but I’m pretty sure this was the exact same opening and closing line for my thesis defense.

24

… I think we need significantly more information about how this test was set up.
Yep, there’s a lot I’m not clear on including how they gathered the data to generate the transient template and whether or not this took airflow velocity gradient into account (i.e. did they measure airflow magnitude and direction at a single height above the ground or did they do a sweep from the ground up - which would be very difficult to do well, I suspect. This is a pretty fundamental factor.
Also, is the wheel rotation mapped directly against airspeed, or does it take account of the fact groundspeed and airspeed (however that’s measured) will not be the same except when travelling in a homogeneous body of still air - which is not what the test is about.

25

https://www.hambini.com/...which-one-is-fastest

My notes:

There isn’t a ton of information on the “protocol” i.e. what yaw distribution was used.There is some illusion made by the author to measuring “transient drag” instead of “steady state” drag.No tire information is givenThe power differences between 30 km/h and 50km/h look about rightOther than the occasional oddball, the ordinal rankings look about right based on depth.I’m guessing the “Bontrager 60mm” are the new Aeolus XXX 60mm rims as the D3 profile was not offered in a 60mm depth.I find it interesting that both the Bontrager and the (presumably) Aeolus XXX performed so well. They are both very similar in shape which is a somewhat “new” shape. I tend to believe the data that Trek Bontrager produces so that gives me some confidence in the ordinal results shown here.Flo wheels did not do well in this test which makes me wonder which version of their wheels were tested. The 45 is obviously the carbon model.
…I’ll go get my popcorn.

Bunch of horseshit test by Yoeleo , they finally figured out how to market/move their stock. They use to have YT channel with some chick talking/showing their frames and wheels… It was painful to watch, but with 500-600$ for set of wheels/frame people were watching.

They describe testing protocol without reviling testing protocol…relaxed hoods at 50km/h who was doing this test in UK ? Wiggo, Mr G or Froome?

You can’t go wrong with those prices if you are on the budget, I just don’t understand why people need validation from some BS test, like few watts would matter. You want to show off in front of friends: envy and zipp, no $, and want to lie to yourself you are faster get Yoyo.

If I offended anyone I’m sorry, I just call it how i see it.

26

Thus far, I’d purposely not mentioned that brand/wheel in hopes that nobody would ever consider buying a wheel made by a shady company that sounds suspiciously look “Yolo”. I hear you with regards to this being a marketing ploy. That said, the results… the Shimano wheels in particular… seem too realistic to have been faked by someone who doesn’t know that much about this stuff.

27

So I read this as a don’t buy FLO…

I will continue to buy Flo because they are transparent in their wind tunnel testing procedures.

28

On the XXX4, they actually brake pretty well - light years better than the old Aeolus’, where you pretty much had to pray. Haven’t ridden them in the rain though.

For the “old Aeolus” do you mean the Aeolus Pro 5 and similar? This one:

https://www.trekbikes.com/...colorCode=black_grey

If so, that’s disappointing, because that price had been tempting me for a while.
Edited/fixed: not sure that’s the same wheel as 2017 Aeolus 5 so don’t know.

29

On the XXX4, they actually brake pretty well - light years better than the old Aeolus’, where you pretty much had to pray. Haven’t ridden them in the rain though.

For the “old Aeolus” do you mean the Aeolus Pro 5 and similar? This one:

https://www.trekbikes.com/...colorCode=black_grey

If so, that’s disappointing, because that price had been tempting me for a while.
Yeah those are the ones. Inexcusably bad braking IMO. Wheels like these are what led us to disc brakes on road bikes.

The Pro 5 is a new wheel that came out at the same time as the XXX though. I wonder how it compares to the old non-Pro models

30

On the XXX4, they actually brake pretty well - light years better than the old Aeolus’, where you pretty much had to pray. Haven’t ridden them in the rain though.

For the “old Aeolus” do you mean the Aeolus Pro 5 and similar? This one:

https://www.trekbikes.com/...colorCode=black_grey

If so, that’s disappointing, because that price had been tempting me for a while.
Yeah those are the ones. Inexcusably bad braking IMO. Wheels like these are what led us to disc brakes on road bikes.

The Pro 5 is a new wheel that came out at the same time as the XXX though. I wonder how it compares to the old non-Pro models
Thanks for pointing that out! I’m going to delete my post because now I’m not sure it’s the same wheel as last year.

31

Thus far, I’d purposely not mentioned that brand/wheel in hopes that nobody would ever consider buying a wheel made by a shady company that sounds suspiciously look “Yolo”. I hear you with regards to this being a marketing ploy. That said, the results… the Shimano wheels in particular… seem too realistic to have been faked by someone who doesn’t know that much about this stuff.

That blog is totally not “realistic” there is nothing in it, numbers are “could be realistic” but the rest looks fake, except 2 random charts… anyone could come up with, just get the numbers 50-50% correct…
Having said that, those guys are doing carbon components for long time (they are not shady, since they are in business for long time), instead of trying to be the best, they should place them self just below (1-2W slower) Envy and Zip, with wheels that cost quarter of the price it would be no brainier choice for market their target. With current data posted they will just get laugh at. ENVY guys will get envy, zip guys will get zippy-zappy and everyone will call it BS. Otherwise none of Envy guys would even notice that blog, guys on the budget would still buy the product 1W slower and 70% less expensive.

32

I’ll go get my popcorn.

They went to an awful lot of trouble and expense to come up with that really poor presentation.

What frame and fork were used? This has a big effect on wheel performance.
Why “relaxed hoods position” for aero wheels… and at 50km/hr?
Exactly how did they perform the “transient yaw angle” test, and what yaw angles were tested?
For a full rider on bike test we really need to see the repeatability.

33

I’ll go get my popcorn.

They went to an awful lot of trouble and expense to come up with that really poor presentation.

What frame and fork were used? This has a big effect on wheel performance.
Why “relaxed hoods position” for aero wheels… and at 50km/hr?
Exactly how did they perform the “transient yaw angle” test, and what yaw angles were tested?
For a full rider on bike test we really need to see the repeatability.

I agree. The main reason I posted this was that I found it interesting that the Enve 7.8 and Bontrager Aeolus XXX 6 ended up so close to each other and that Flo, which has tested well elsewhere, tested so poorly. What I’m saying here is that maybe… maybe… something about this testing methodology, exposes strengths in some wheel designs and weaknesses in others that we have not previously seen.

34

Thus far, I’d purposely not mentioned that brand/wheel in hopes that nobody would ever consider buying a wheel made by a shady company that sounds suspiciously look “Yolo”. I hear you with regards to this being a marketing ploy. That said, the results… the Shimano wheels in particular… seem too realistic to have been faked by someone who doesn’t know that much about this stuff.

Wait, so the people who did this test have a stake in one of the wheels tested? That’s not clear to me from the site linked to.

35

Thus far, I’d purposely not mentioned that brand/wheel in hopes that nobody would ever consider buying a wheel made by a shady company that sounds suspiciously look “Yolo”. I hear you with regards to this being a marketing ploy. That said, the results… the Shimano wheels in particular… seem too realistic to have been faked by someone who doesn’t know that much about this stuff.

Wait, so the people who did this test have a stake in one of the wheels tested? That’s not clear to me from the site linked to.

I agree and I can’t find a direct link… though that doesn’t mean it shouldn’t be ruled out.

36

If you wanted to know the outright fastest complete wheel system, why can’t a list of wheels simply be tested in the purported ‘fastest configuration’?

Hey I know some guys who can do that haha

But what most people don’t realize is that those dudes probably spent 20-50 hours planning everything before they ever set foot in the tunnel.

Like Jim said you’d need to devout some serious time tim finding the fastest wheel /tire combo. Then I suspect you’ll need to redo that for every bike since fork blade are going to very in widths and blade widths won’t be uniform either

Wind tunnel testing is waaaaaaaaaaaaaaay harder than most people think, even when we make it look easy.

37

[quote desert dude Then I suspect you’ll need to redo that for every bike since fork blade are going to very in widths and blade widths won’t be uniform either

This. It’s a fun to look at these tests, but the wheel that is fastest by itself isn’t necessarily going to be the fastest one in the frame. And the fastest wheel in once frame isn’t going to be the fastest in another. Especially if you are looking at the difference between a beam bike which doesn’t cover the back wheel at all versus a regular bike. Or between a tririg omni or ventum which there isn’t a downtube and a cervelo p3 whose downtube is right behind the wheel.

38

Whenever this topic comes up I always think we should view the wheel + tyre as a system. Test each wheel with the very best tyre that the manufacturer can recommend, or which has been shown in testing to work best with that wheel.

I mean, the idea of the thing is to find out how to go as fast as possible so just test everything in its fastest configuration.

If the manufacturer cannot recommend the tyre that works best with their wheel, or no additional test data is available, then screw 'em they can have a 28 on there as punishment because if they are trying to sell something to us based on its aerodynamic merit then they should have learned this by now, and be able to provide an answer.

i’ll 2nd that!
of course then you have frame/fork interactions but we’ll ignore that for now…

39

Thus far, I’d purposely not mentioned that brand/wheel in hopes that nobody would ever consider buying a wheel made by a shady company that sounds suspiciously look “Yolo”. I hear you with regards to this being a marketing ploy. That said, the results… the Shimano wheels in particular… seem too realistic to have been faked by someone who doesn’t know that much about this stuff.

Wait, so the people who did this test have a stake in one of the wheels tested? That’s not clear to me from the site linked to.

I agree and I can’t find a direct link… though that doesn’t mean it shouldn’t be ruled out.

my immediate reaction on seeing some no-name brand near the top of the results was that there is probably some influence there…

40

Tire choice and pressure is absolutely crucial to any wheel testing and, for that reason, it’s practically impossible to conduct a test like this properly. If you try to make the testing “fair” by using the same tire and pressure for each wheel, you’ve likely produced near meaningless data. There are simply too many variables; it would take weeks to test all the different combinations. Then, of course, you’ll need to repeat those results to give them value. Fun!

I couldn’t agree more, and it’s the exact reason we’ve never done a FLO vs. Them study.

I have to say that I find it funny how the manufacturers are always made out to be the villains in these studies. As manufacturers, we certainly do not know everything, but I can tell you as a collective group, we have learned a lot about testing products. Honestly, most of us work together to help ensure the data we are producing is as good as possible. Some of the best advice we have ever received has come from our biggest competitors.

I’m not saying that all brands are 100% transparent and honest with their marketing, but I think most of our customers are smart enough to spot the fakers when they see them.

I think it’s important for people to understand how the smallest changes in a wind tunnel can result in massive changes in drag. In most cases, these changes are so small or non-obvious that the people conducting the tests have no idea they are introducing error. Your protocol can render tens of thousands of dollars worth of wind tunnel time completely useless.

Independent and crowdsourced testing is great, but people need to pay close attention to their protocols and heaven forbid, ask the manufacturers for advice. Take the following as an example. There have been several crowd-sourced studies on this forum. One, in particular, added hundreds of grams of drag worth of error to their results in how they set up their tires. Yes, hundreds of grams in just the tires. Conclusion, the results are useless.

We own a $2,000 pressure sensor that we use for all of our studies. It’s accurate to +/- 0.1 psi or better. Why? Because tire pressure is critically important. Had the people conducting the above study called and asked for advice, we would have gladly loaned them the pressure sensor, and in a single conversation eliminated hundreds of grams of error. Honestly, the more good data, the better.

A while back we contacted all of the brightest minds in our industry to create a Collective Testing Protocol (CTP) that all manufacturers could adhere to. The idea was to develop a standard protocol—available to the public—that eliminated as much error as possible. Wind tunnels could then become certified to conduct CTP studies, and your results would be stamped with a CTP logo by an independent 3rd party. We had 100% buy-in when we presented the idea, but when the actual work had to be done, most of the people backed out. I still have the wheel protocol somewhere.

Anyway, my moment on the soapbox is over. We’ll be discussing all of this stuff in future podcasts simply because we think it’s important to share what we’ve learned with our audience.