Where's the Like button?
Great post.

I think all we're lacking now is a rapid-oscillation wind tunnel of hambini's type, with a rolling treadmill that you actually ride on, not have the bike bolted to the floor. That way you can blow the rider around and watch the effect on their pedalling power as they try to maintain stability. Muscle strain gauges to monitor power wasted in the upper body countersteering/balancing. Would also account for rolling resistance and power-to-rotate.

It would put a lot more emphasis on the rider's ability to hold a position while being blown about, but then again, surely that's closer to real life?

I'd pay money for this testing...
As I'm sat on rollers most of the time through winter vs. a static trainer, I see a lot of wander and wonder how much of this might contribute to increases in yaw and drag - the micro corrections in side to side wheel, bike, and body.
My gut feel is that there is less of this on the road than on the rollers, but am not sure - rollers might exaggerate.

Work commitments aren't always a curse, sometimes they can be a blessing.

For what it's worth Sebi has tested on rollers in a wind tunnel (which tunnel I'm not sure) numerous times.

Question: how do you see yaw and side force affecting rolling resistance? In my mind, more side force would equal higher rolling resistance thus placing additional importance on the performance of wheels (and frames) at yaw.

In hopes of getting this sh!t-storm of a thread (which I regret starting at this point) somewhat on topic: what are your views on the effect of side force on rolling resistance (same question I posited to Josh above)? In my mind, increased side force would increase rolling resistance and thus should color our view of wheel/frame performance at yaw.

Well...at a minimum, it's another argument for always running low Crr tires ;-)

---

http://bikeblather.blogspot.com/

...now you've got me wondering how you got 30-60Hz data out of an aerostick? As I understand it, mine only transmits at 4 or 8 Hz, but the recording setup I have for it (WASP Util running on an iOS device) only records 1Hz values :-(

How did you do it?

Secondly, maybe those short "transitional yaw angles" don't matter as much in the grand scheme of things? (Going back to the utility of current models to get things pretty darned close to reality).

---

http://bikeblather.blogspot.com/

Thanks Josh, you are another one of my favorite posters!

It makes sense that a device with inertia would have a bit of lag and also possibly overshoot. A think you are saying that the Aerostick showed rapid and substantial changes in wind direction and intensity? Did you also notice a meaningful difference in the yaw frequency distribution between the two measuring devices?

I don't think Hambini has shown what he used to measure airspeed and yaw and the conditions when it was measured (I could have I've missed it) but it's interesting that his "typical" yaw values are so much higher than others have reported. If it was along roads with numerous cars, you'd expect a lot of disturbance in the flow and higher yaw.

At any rate I think it's an intriguing topic. It would be pretty cool if someone could verify the degree of fluctuation in the wind and the and the distribution, and then replicate this in the wind tunnel. Is the ability to disturb the flow in the tunnel not a common feature?

Chris @Flo are you watching? If you're inclined, it would be helpful to know more details regarding your study on real world yaw.

You would absolutely have some scrub and steering effects. I suspect it's substantial on a windy day, but no one has measured it AFAIK.

I'm not so sure that current models and measurement practices are treating wind properly. Particularly the large reduction in drag with yaw that is measured in the tunnel, which if "real" would mean that riders go substantially faster when there is a crosswind vs no wind... which I haven't experienced. The forward pressure is a bit higher with a crosswind so that would account for part of the difference but not all of it. Tire scrub and steering corrections are going to be something, and I would expect rapid fluctuations in real wind to increase drag as well.

doesn't AlphaMantis pick up an increase in system drag in the turns on the velodrome? I think I remember someone saying that once.

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

The drag area coefficient in the direction of travel (i.e. "CxA") drops with apparent wind yaw angle on good aero setups, but that doesn't mean you'll go faster than in still air. The apparent wind speed is significantly higher than ground speed in order to create those yaw angles. That higher apparent wind speed results in a much higher drag force to overcome (as compared to the ground speed)...make sense?

So...it's not so much that you'll go faster than you do in still air conditions under high yaw...just that you'll go faster than if your drag was best modeled by a cylindrical bluff body ;-)

That said, some of my lowest "apparent drag" values from races have occurred under light side-wind conditions...and I recall one incident of actually accelerating while coasting when a strong side gust of wind hit me...so there's that :-)

---

http://bikeblather.blogspot.com/

That's mostly due to higher tire loading from the banked turns. Rolling resistance is directly proportional to the tire load.

---

http://bikeblather.blogspot.com/

Yes, I believe that's true. Someone has probably modeled and measured this on the track (just guessing), but I'm not aware of anyone modeling the effect of wind on the road.

Yes, I stated this accounted for part of it. But not the ~15% drag reduction seen here @10deg yaw. The forward pressure increase with a 90 deg crosswind yielding 10 deg of yaw is only ~1.5%.

wow, DZ sails really well. Most people only see 5 or 10% reductions.

---

Eric Reid AeroFit | Instagram Portfolio
Aerodynamic Retul Bike Fitting

“You are experiencing the criminal coverup of a foreign backed fascist hostile takeover of a mafia shakedown of an authoritarian religious slow motion coup. Persuade people to vote for Democracy.”

Hi Eric,

Yep.

Also, I believe that test (LSWT 1126) includes TT bikes with disc wheels & deep fronts, so sailing could be attributed to those, maybe more than to DZ.
That said, DZ happens to have posed for the scanner with his knees level. Other researchers have since measured the lowest drag with legs in that position. So bikes tested with foam Dave seem to have lower drag for that reason as well.

Cheers,
Damon

---

Damon Rinard
Engineering Manager,
CSG Road Engineering Department
Cannondale & GT Bicycles
(ex-Cervelo, ex-Trek, ex-Velomax, ex-Kestrel)

Aaah...don't forget drag force varies with the SQUARE of the apparent wind velocity. I'm getting ~3.2% increase in drag force for that example. (I used a ground speed of 25mph, which requires an apparent wind speed of 25.4mph at 90 deg to get a 10 deg apparent wind angle)

BUT...you picked the easiest crosswind angle to get that resultant apparent wind angle ;-)

On the opposite end of the spectrum, if traveling at 25mph, but against a crosswind that's 20 degrees off of straight ahead...well, to result in a 10 deg apparent wind angle that would require a 25mph crosswind speed, and results in an apparent wind speed of just over 49mph. The drag force in the direction of travel is nearly 4X the drag force traveling 25mph in still air.

Actually, I calculate that just a 5mph crosswind at 70deg from straight ahead would nearly totally negate a 15% drop in CxA (results in a 26.4mph apparent wind angle -> (26.4^2)/(25^2) = 1.12...close enough)

---

http://bikeblather.blogspot.com/

Those models aren't even precise enough to care which wheels you use, correct? That is, BBS doesn't ask if you use a 404FC, 404NSW, or old-school Al brake track 404 (or even a Jet 6+)? They didn't the last time I used it. That means that either 1) wheel shape doesn't matter other than depth (in which case what are we even doing putting new wheels into the tunnel) or 2) the BBS model isn't precise enough to care about the differences between wheels of a given depth, and these differences could reside in that imprecision. This does also put an upper bound on the possible differences he could be discerning, but still non-zero.

Also, other than a couple of outliers, none of the results were very shocking; it was mostly deeper is faster and wider brake tracks do better with wider tires. Much was made out of a minor reshuffling of the typical results.

And your ban over on WW is bullshit. I think he was making that up; I can't find a reference to it. There is a different reference to one, but I can't find the one he referred to. If it's that obscure, hanging you out to dry for it is... well I just said bullshit but that's what it is.

---

The point is, ladies and gentleman, that speed, for lack of a better word, is good. Speed is right, Speed works. Speed clarifies, cuts through, and captures the essence of the evolutionary spirit.

I'll show my work this time. Tan(10) = .1763. (1^2+.1763^2)^.5 = 1.0154. This is the relative wind velocity on the rider, acting at 10 deg yaw. Proportional pressure is 1.0154^2= 1.031. Proportional forward pressure (in line with rider) is 1.031* cos(10)= 1.0154 .... or 1.54% higher than the rider experiences with no wind.

Where did I goof?


That's because it's the only one that is the same in both directions. Else you have a headwind/tailwind situation.


But you have a headwind component in that example, and then you need to turn around 180 degrees. So a 5mph wind at 70 deg from straight behind... which is no longer 10 deg yaw. Headwinds and tailwinds are another matter.

As I understand it, you need to multiply the CdA by the total apparent wind velocity^2 (X the air density as well, of course), not the component in the axial direction.

Remember, the wind velocity is constant during the yaw sweeps (i.e. it's the apparent wind).

Then again, I'm on pain meds for a knee injury, so I may have totally stuffed this ;-)

---

http://bikeblather.blogspot.com/

I absolutely agree, but that's what I thought I did. The pressure is going to be acting on the same vector as the yaw, and we then need to compute the component in the X direction.

Heal quick, and I hope the meds are good!

Oh, no...the plots ARE the drag component in the direction of travel (why some prefer to label them CxA instead of CdA).

---

http://bikeblather.blogspot.com/

Yes, I know that's what the WT plots are... which is why we need to put the calculated pressure due to yaw in the direction of travel as well.

Well, it is what it is. Lame, yes...but not surprising for that place. It's more of a "timeout" though (just a one week ban).

How about Hambini immediately circumventing his 1 week ban by creating a new account and posting a reply anyway...and the resultant penalty is just another 1 week ban for his new account? That's an odd moderation technique :-/

---

http://bikeblather.blogspot.com/