I agree, interesting.

I would think things would be pretty subtle between a u shaped and kamm tail rim profile.

What i don’t get is aero when the inside of the wheel is the leading edge. I thought a worn tire (flatter front) reduces aero, so one would think a kamm tail would be no better.

Either way, no matter if they cooked data, or if this is revolutionary (no pun intended), i am not about to go out and buy new wheels. But good on them for fooling with the idea.

But... with all we know about kamm tails on frames... why have other companies not gone here before if it truly is faster?

Maybe the answer is that it is not faster (we know the shape is inferior to naca). Or that side forces have too much of an assymetrical effect on the steering.

I'm no aero expert, but that graph looks really funky to me. I don't care what brands were involved, or which brand did the testing. How is it possible that a wheel can go from roughly 10 grams of drag at 80 degrees, then go to 1130 grams of drag at about 82 degrees? Is testing at 80 degrees even applicable in the real world? Would this be the equivalent of flying down Hawi at 45 mph being super aero, then suddenly not just being blown off the road, but being blown straight into the ocean? I'm joking here...but how can results like that come out of a computer...and someone say...yeah...that looks right.

I have no idea what this means.

First, it's a CFD "simulation". Second, yes, 82 degrees is ridiculous. Honestly, anything beyond 30 degrees is pretty absurd. It's not that it can't happen but it's so incredibly rare that it's not a worthwhile input to a development or testing protocol (IMO).

With that said, at such an extreme yaw wild swings in drag wouldn't really surprise me so that alone shouldn't disqualify the data (IMO).

is this the reynolds ar58 or the previous generation reynolds "aero 58" wheels, also designed by Paul Lew (and also featuring an unusual design with a narrow inner edge)...?

I am curious how every other wind tunnel plot shows less drag as yaw increases, but this is the opposite.

Armchair aerodynamicist: if something like the texture/tread pattern on say Continental tyres compared to others seems to have a notable benefit, why doesn't anyone apply this on the texture of the rim, say when the inner edge is leading?

I don't know but my guess would be the prior generation since the AR 58 is relatively new.

Oh shit... cfd. And i just looked at the numbers.

Next...

I could type out a lengthy explanation but... I don't think I'll do it correctly. Here's a quick but not entirely correct explanation: it has to do with the axis and speed in which drag is computed and then normalized. For example, look at the vectors in a 50kph wind tunnel test with a (for demonstrative purposes) 45 degree yaw angle. What's the speed in the forward direction? What's the cross wind? How much side force do you have? How much drag do you have in the forward direction? Most people are used to looking at charts that show drag in the forward direction. This style chart shows total drag.

A common misconception is that you're actually more aero/faster in a mild crosswind versus no crosswind at all because practically every bike frame and wheel set shows lower drag at 10 degrees of yaw than 0 degrees. So let's work in reverse: you're going 20mph and you're consistently hit from the side by a 20mph crosswind. If you were to recreate this condition in the wind tunnel, how would you orient the wheel and at what speed would you run the tunnel? How would you attribute the resulting total drag to the respective vectors (crosswind force vs drag in the direction of travel)?

I'm really kind of hoping someone who's better at explaining this sort of thing chimes in with force vector diagrams to explain this because I'm struggling to do this in written form. For what it's worth, this isn't the first time data has been published in this form. It's rare but it's not the first time.

With regards to your idea regarding texturing the inside edge of the rim, it's a good idea but I suspect nobody has tried for fear of running afoul of Zipp's patent.

Not sure about CFD, but judging by my new feeds if it was CBD they might have something.

I'm no expert, but I thought some of the road bike and other frame kamm tails were to both abide by the UCI 3:1 ratio and also provide a tight spot to hide a UCI style water bottle? If you use a complete naca foil, you either can't meet the 3:1 sometimes OR your bottle is like a turd in a punch bowl after the downtube.

Like this?




I still don't know what their CFD chart means. Compare it to this one:

Interesting

Kamm tail indeed is an interesting profile, in some conditions.

Here, a 55mm Kamm tail profile is compared to :
Zipp 404 Fircrest
Reynolds 58

... and the results looks good.

However, CFD is sometimes not that accurate. A real wind tunnel test on a real bike with a rider will be better.

And, not sure the high yaw (more than 45° ?) are of any interest ?

But for sure, the results at 25° / 30° are raising interest... for more info.

These results are very puzzling. The graph says drag, not total force magnitude, but I suspect it's the latter.... But even so it's pretty strange. I would want to know what these CFD simulations are. Someone in the comments says he only does 2d simulations, But again, even so these results seem inexplicable.

There's a reason no one has made a kamm-leading-edge airfoil before.

---

-------------
Ed O'Malley
www.VeloVetta.com
Founder of VeloVetta Cycling Shoes
Instagram • Facebook

Why is that a misconception? Axial drag is....axial drag. Off-axis forces might affect rolling resistance slightly in some way or cause the rider to expend some marginal energy to keep the bike aligned in the right direction. But there's a reason that drag is normally reported as axial drag.

In a wind tunnel, when increasing yaw (at same wind speed), axial force go down generally, logically... but maybe axial CdA goes up.

Because you need to compare axial force generated vs axial wind speed, not axial force generated vs total wind speed.

This is why sometimes graph results show drag in grams lowering (because of yaw).... and from this derivate the misconception the CdA is better at yaw, while, maybe not. Sometimes it is simply axial force going down, because less axial wind speed.

Here for exemple it is not clear if it is about axial drag or total drag. But OK, it moves up...

There aren't any wheels with 20g of drag at 0 yaw at 50kph. That's almost an order of magnitude too low.

Hah! I did not even look at the numbers!

Just doing small things wrong can make big differences in CFD. But I suspect this guy did big things wrong.

---

-------------
Ed O'Malley
www.VeloVetta.com
Founder of VeloVetta Cycling Shoes
Instagram • Facebook

Very possible.

Paul Lew is the gift that keeps on giving... Aerodynamically, I'm not sure he's ever made a wheel that's been tested independently to verify his often outlandish claims. What I find surprising is that companies keep hiring him. Reynolds hiring him was a surprise given that they have the technical chops to evaluate his claims, but when his wheels there were basically a flop, I thought he'd have disappeared. Don't know why Mercury would hire him. I guess he remains convincing to certain groups...

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

What's hanabi s opinion on this lol

[quote Rappstar Reynolds hiring him was a surprise given that they have the technical chops to evaluate his claims, but when his wheels there were basically a flop, I thought he'd have disappeared.[/quote]
Flop? They're not like a huge market success probably, but they're pretty well-regarded, speaking of both generations of the "Aero" line of wheels. And there hasn't been a ton of independent testing, but the below Tour Magazine (fwiw) results show the first generation doing great at low yaw.

I don't know if Lew's contribution had anything to do with it or not, but I don't think it's appropriate to throw a manufacturer under the bus without any real evidence one way or the other except a personal antipathy towards a single personality.

 

Reminds me of this Madfiber thread:

https://forum.slowtwitch.com/...3F_P3249950/?page=-1


and the SarahConner replies.


.

---

Once, I was fast. But I got over it.

This graph looks unlike anything I've ever seen in wheel aero.. but at least good to see that Paul, being old like me, still thinks in grams of drag.. I know that CdA is so much better for working the math, but those of us who grew up at Texas A&M, MIT and the early San Diego still find grams of drag to be sort of comforting!!

My guess would be that this model is based on some sort of x-section from the front centerline of the wheel.. but even that would/should have different stalling properties and would not explain the crazy high numbers out at ~90 degrees for the Zipp and not the others.

10 years ago CFD could get within 10-15% of tunnel and the work we did with Intelligent Light and Matt Godo really helped refine this down considerably so that we were typically better than 5% of tunnel values when I left Zipp 5 years ago.

https://altairuniversity.com/..._1431_082211_web.pdf

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc