Very cool. I've had some ideas about that but of course I've never had a chance (or the equipment, or really, the time) to test them. Very, very nice.

From a development standpoint, it would be useful to isolate rotational aero drag from bearing drag. From an end-user's standpoint, it doesn't matter. Rotational drag is rotational drag, and if one wheel has less drag than another, who cares what the relative contributions are of the bearings or the spoke/rim/hub aero behavior.

---

"They're made of latex, not nitroglycerin"

Thanks again for that connection. I should have archived his page and images back then. He seemed like a nice guy when I exchanged email with him. I hope he's well, wherever he is.

Yes, I have...not up on blog yet though.


Maybe...but only at yaw angles above 10d, if so.


Ummm...not so much. It didn't roll near as well for me as it did for VN when they tested it at Wheel Energy, despite the 26C being VERY wide (>28mm) when mounted on a rim with an ~21mm internal width.


Not so much...you'd be far faster with a TC w/latex tube (and sealant in that if you're worried about goat heads, and such).

Then again, you are saying that you're a road/crit racer who's switching to discs...so, I'm not sure how much you value overall speed/watts vs other properties :-/

---

http://bikeblather.blogspot.com/

But those aren't Specialized tires...that exercise is left up to you ;-)

---

http://bikeblather.blogspot.com/

Right...I guess I should have said that Damon confirmed it :-)

As long as we're mentioning old-timey internet contributors, I should mention that all of Damon's data on wheel stiffness is still hosted on Sheldon Brown's (RIP) most excellent website.

---

http://bikeblather.blogspot.com/

Hmm. I thought the watts to spin between different wheels would be a pretty strong signal. The Moore and Bloomfield study put the range of wattage values between a disc wheel and a non-aero, regularly spoked wheel at about 10 watts. The difference between a disc wheel (~8 watts) and TriSpoke (11 watts) was about 3 watts. Could you comment on whether the general wattage values in this study are similar to what you find (I.e., the rotational drag on a disc is about 10ish watts)?

Thanks, Tom. I'm fully aware of the weight & aero penalties associated with discs. I hope that the other advantages, especially in rainy crits & on technical descents (both of which I encounter on the regular) are worth the penalty.

I'll take your advice & go Turbo Cotton with latex. I love that tire & actually use it as a daily driver in the summer. Maybe I'll try the Schwalbe tubeless sometime, which I understand also roll quits nicely.

Best,

Mike

Awesome, and thank you. Re: 26 TC vs. 24, it depends (mostly on the road surface/realized Crr difference and your average effective wind vs. ground speed...e.g. if you're in the pack vs. off the front). For some quick and dirty estimates, you can use a total CdA delta of 0.001 m^2 and, say, the Velonews or Tom A. published Crr data on your favorite cycling calculator to see what the tradeoff would be at various speeds (keep in mind: if you're in the pack, the wind speed is significantly lower than your ground speed).

We've found (as has Tom via his blog) that you have to be moving pretty fast on the front or alone for a super skinny tire's CdA advantage to overcome the total combination of something like the Turbo Cotton. Typically >60 km/h

Chris

We've tested a fair number of other tires, but won't claim to know about all of them. But for the ones we have complete data on (like the GP4000s in various sizes), the statement re: Turbo Cotton being the fastest total combo stands.

Chris

These relations of numbers are really interesting and the Chung method has a nice spin to dig deeper, thanks for your reply ..

---

*
___/\___/\___/\___ the s u r f b o a r d of the K u r p f a l z is the r o a d b i k e .. oSo >>

Yes, the total magnitudes you're referencing are in the ballpark for race speeds. However, for R&D and isolating the resulting change of single variables at a time, the deltas are very small. So my fault earlier for misspeaking: the signal isn't small in an absolute sense, the resolution we needed was very high.

Hi everyone,
My name is Ingmar, and I joined Specialized’s aero R&D Department last March. I’m a long time Slowtwitch reader. Before I moved over from Germany, I was an avid triathlete. Right now I am getting settled in the US, enjoying the lunch rides at Specialized and will see if I get into the US tri scene later this year.
I wanted to add a few notes regarding the way we calculate the rotational CdA and the combination with rotational drag. After we established a way of measuring this drag component, coming up with a way to communicate this metric in an accurate and easy to use way was a big part of the project.
We looked at a wide variety of variables to see what impacts rotational drag and therefore how to create an accurate model for future simulations and testing. Testing variables included tire size, tire direction, fork/no fork, air density, bearing drag, relative humidity, airspeed, and wheel speed. From this data we created a model that covers the vast majority of conditions a rider would experience outside:
P = ½*rho*CdA_rotational*v_wheel^2*v_air
If you compare this equation with the translational drag equation, you notice it is almost the same with one difference: Here wheel speed is the squared speed component, while in the other equation air speed is the squared equation. When the wind speed and the ground speed are the same the translational and rotational CdA can just be added But when those speeds are different (due to antural wind), then the model diverges somewhat. In the most extreme wind conditions there are instances where this simplification introduces an error of less than 10% rotational drag (= less than 3% total wheel drag = less than 0.2% total system drag). We feel that with every model of reality you have to balance accuracy with complexity. Directly adding up rotational and translational CdA into total CdA allows to keep the complexity of our model at the minimum and makes the metric easy to understand and use. Our goal is to be transparent and diligent in how we do our research here at Specialized!
We learned a lot through this project, and, as always, answering these questions led to more interesting questions that we would like to find answers to in the future. If there is anything you want me to clarify, I’ll try to check in here from time to time over the next days, so let me know.

Best,

Ingmar

Welcome, Ingmar.

Very cool. I had wondered about a model like this but I was never sure whether the difference would have been empirically measurable. Congratulations on getting it to work.

Hi Ingmar,

It would seem to me that the fork would have the same effect as changing the air density for only the top part of the wheel. I am not an engineer, but I don't see how your model would help you understand or communicate the effect of wheel/fork/frame interactions.

I know Tom A. hates the use of watts when discussing drag, but isn't the power required to turn the wheel really what you are interested in, so it might make more sense to express the total energy to move the wheel in watts at a few common standardized speeds, like 20, 25, and 30 mph?

Hi Robert,

Thanks for the welcome!
Your work was a big influence on my path into bicycling aerodynamics and to Specialized.
It is measurable (at 0.05W, repeatability a lot becomes measurable), but v_air^2*v_wheel is close enough that you do not introduce much of an error. If you would base the equation purely on a curve fitting approach, the reality is a combination of the two (more biased to v_wheel^2). We ended up not going down that path since it added a non necessary degree of complexity and increases the risk to overfit the data. Also it is more in line with our theoretical model.

Halo Ingmar,
Sounds like a very pragmatic, engineering approach. Good stuff.

---

http://bikeblather.blogspot.com/

Hi Mike,

You are right, for studying interaction effects this approach does not add much. But we needed to study interaction effects in order to validate the approach.
If the interaction between different forks would have been so substantial, that you could not compare wheel data from one frame to another, than having a model that allows you to add one CdA to another would not have been necessary and could have been misleading.

Its true, as the rider you want to know if the watt savings from your awesome new wheels helps you beat your buddy who is X-watts stronger than you. But just like your friends power depends on his daily form, the watt savings you get from your wheel depend on changing variables (like the weather, altitude, etc.). If you want to know for sure, you have to do some calculations, and those are done easiest with CdA numbers. If you just need an estimate, we at Specialized use 0.001m^2 CdA (one division in all our charts) [/font]≈ 1 Watt saved at 40 km/h ≈ 1 second saved per kilometer.
If you want to compare wheels from different manufacturers power number can be very misleading. People often get lazy to write down the exact testing conditions and this crucial info gets lost. For example, Germans always report at 40km/h (track riders 60km/h), Americans like 30mph. Some people assume air density is 1.225 kg/m^3, some assume 1.2 kg/m^3 (i've even seen 1.12754kg/m^3) and some seem to use whatever the air density was the day they tested. This makes it really hard to compare manufacturer claims.

I hope this helps

I think you mean 1W~= 0.1s/km ;-)

Yes, your last statement is why I prefer drag reported in CdA...it doesn't require "qualification" to be useful :-)

---

http://bikeblather.blogspot.com/

Wow. I'm sure you're exaggerating but it's nice to hear anyway.


Yeah, the real world issue is whether any estimation procedure is robust for the amount of effort people are willing to put into it. Glad it's working for you.

So would your reply suggest that there isn't much difference between aero forks and that the wide versus narrow fork spacing doesn't have a big impact on rotational drag. Or do I have to wait and see what the new Shiv looks like to find out? ;-)

Hello grumpier.mike and All,

Anyone seen more numbers for just the Mavic CX01 blade portion concept of the Mavic CXR80 tire+rim+blade? .... (tire+rim+blade plot below) .... while designed for tubulars ..... is it there a worthwhile gain for clinchers by smoothing the tire/rim interface?

I don't see the CXR80 listed in the present Mavic wheel lineup .... looks like it was dropped.

https://cycletechreview.com/...s-wheels-disallowed/

Excerpt:

"The UCI, naturally, doesn’t like it and has said no, leaving triathletes and British time triallists the only two-wheeled competitors able to ride the wheels with their blades fitted."









The Bontrager 'wings' tire looks like a good way to get the benefits of a smoother tire/rim interface ...

From Bontrager advertisement:

"Bontrager's R4 Aero Tire is all about making you faster and more efficient. Its 220tpi casing creates less drag and provides a supple road feel you'll love, while the tread minimizes rolling resistance and grips great in the corners. The R4 Aero also features integrated flat-tire protection. Plus, this amazing tire has Bontrager's unique aero wing design that saves you energy mile after mile. In fact, it has been tested to save over three minutes in the bike leg of the Ironman Triathlon! No wonder Keith Bontrager himself calls the R4, "the World's Fastest Tire."

....... but what is the best practice method for smoothing other brand tires tire/rim interface?



One of the home brew methods that have been around for many years is to smooth the tire/rim interface crack with silicone sealant.

Other methods?

Or is the aero gain worth the effort?

---

Cheers, Neal

+1 mph Faster

Wheels look nice. Not sure why a 21 spoke rear, that just screams we want to be proprietary, which screams run the other way.


Can the rims be purchased separate?
Any component weights? (rims /. hubs?)

Weight on the only review so far shows 1416, which is still light, but not sub 1400g. What is your +/- spec on that?

At least for the CLX 64s, the front weight like 695 grams IIRC. I can't remember what the rear was but I think around 850 or so.

---

---

Save: $50 on Speed Hound Recovery Boots | $20 on Air Relax| $100 on Normatec| 15% on Most Absorbable Magnesium

Blogs: Best CHEAP Zwift / Bike Trainer Desk | Theragun G3 vs $140 Bivi Percussive Massager | Normatec Pulse 2.0 vs Normatec Pulse | Speed Hound vs Normatec | Air Relax vs Normatec | Q1 2018 Blood Test Results | | Why HED JET+ Is The BEST value wheelset

Although I'll likely race on TCs with latex, I'm looking forward to trying tubeless with these, especially for off-season work. From a weight and performance perspective, would the Silca valve & tape be better than the stock valve & spoke-hold plugs? I've no experience with either, obviously, but I'm a huge fan of what Josh is doing & would happily support him if there's a performance advantage. Thanks.