.225 +/-.4 depending on how tall they are

.04 you mean? haha

CDA at what airspeed? CDA isn’t a constant.

This.

Also, given how few people actually have tunnel tested or own aerometers I would take a lot of the CdA's with a grain of salt.

But, otherwise, for "FOP"........0.225 doesn't sound unreasonable. If someone said "MOP and slower" I would put it closer to 0.245.

Not to mention how many people simply can't stay glued down in their extensions for the entirety of an event. Their "effective" CdA winds up being a lot worse.

Tangent, really fun game I'd like to try. I'd love to look at the front/side profile of a bunch of triathletes and guess the CdA, then get scored on my accuracy.

I bet I'd be way off. I've been shocked at few few of the "eye tests" I've guessed at that have ended up the opposite of what I thought.

---

Too old to go pro but doing it anyway
http://instagram.com/tgarvey4

Yes, looks like I missed a zer0...oh, there it is.

Depends on if they’re FOP because they’re strong or if because they’re aero.

I was the latter and tested/optimized to a .222 @ 40km/h. Best splits were 2:13 half on 230w and 4:42 full on 206w. Both flat courses.

Cd is a coefficient. A is area. Neither is dependent on speed.

Drag is dependent on speed.

---

blog

Independence of Cd from speed is only a first approximation. Most of time it is a good approximation for a cyclist, but there are important exceptions.

You are confusing how to calculate Cd or CdA for a cyclist vs Cd/CdA dependence on the variability of speed. A cyclist riding at 20 mph vs 30 mph with the same exact setup and position, will have the same Cd and CdA.

---

blog

Just search for drag crisis to get a clue what can happen with Cd.

I would say you are correct, because the total drag is based upon the equation which includes CdA where Cd is a coefficient that is dependent on the characteristics of the object under study. The total drag has a number of factors and drag is speed dependent. One interesting aside, on a recent podcast Rohan Denis mentioned that of all the bits that would concern him about drag, the one that was foremost is the clothing. As he put it, your body is the largest drag so the surface of that object is important, as is the way it is presented to the wind.

So back to the question, CdA is dependent on the coefficient caused by the surface of the body in question and A is dependent upon how that person presents themselves (ie their position on the bike). This is not speed dependent unless they change their position due to speed, or their clothing surface changes due to speed. I think I have seen numbers for the total CdA of the system (bike, rider, kit etc) of between 0.18 (I wonder if this is something a person can hold for any amount of time?)and 0.25 so that is quite a range. If @marcag or jim@aero sees and responds or one of the other aero guru's on ST they may have a more definitive reply to your question.

Yup, as others noted, that is an over simplification. The Reynolds number at which a round object transitions from laminar to turbulent flow happen to coincide with things like a rider's head going from 10 mph to 30 mph. That will change the Cd over those speeds.

Drag crisis is a (big) bit of a stretch…

Reynolds number is just length*velocity/nu.

To suggest that velocity changes 10x (length and nu are constant) doesn’t make sense.

If the suggestion is that some tests show trends in CdA that imply transition affecting attachment as functions of velocity, I’d suspect something else (seems like “Crr” varies significantly with speed and load actually). I doubt anyone is testing this carefully. It’s much more likely that some strap/garment or posture change, or numerical issue are to blame.

That chart is misleading, as a stand-alone. I’m not even sure it’s correct at all. The y-axis for Re and Cd is normally shown as logarithmic. See attached graphic from Princeton.edu

https://www.princeton.edu/~maelabs/hpt/mechanics/mecha_54.htm

This is a critical difference.

Since in cycling one is concerned about changes in Cd of 5%, 10%, 20%, then Cd absolutely changes in ways that matter over typical tri speeds (20-30 mph).

I just came across this paper that includes a chart of Cd over speed for an actual cyclist:

https://mdpi-res.com/...f?version=1594566693

https://www.researchgate.net/publication/340042252_Cyclist_Reynolds_number_effects_and_drag_crisis_distribution

This is focusing on the arms and legs and the left leg flow investigated at angles where the bluff body Reynolds number effects are likely to be most detectable. It is also on a polished plastic 3d printed mannequin with no cloths so no surface roughness effects from skin, hair or clothing.


Bob

@Hugi introduced the 'Cd changes' which it does and pointed to the drag crisis phenomenon. But the exam question includes "typical" and defines a cohort.
The conclusion from the useful paper quoted says:
"The Cd variations of a bicycle–cyclist system differ from different object forms. This may suggest
that the cyclists’ different adopted positions might have different Cd variations and the drag crisis
phenomenon may vary between positions and body forms. In conclusion, a drag crisis phenomenon
is noted in a bicycle–cyclist system in the aero position. The crisis happens at speeds from 3 m/s to
9 m/s and Re between 3.21 × 105 and 9.63 × 105. The drag crisis is mostly due to a crisis in pressure Cd."

Paper was based on studies using a light male of average height (ie aligned with cohort but weight below average).
This cohort averages roughly 40kph (11m/s) for the bike: in any case >9m/s (Re @ >9.63 × 10^5 in the bicycle–cyclist system). So the drag crisis occurs at speeds lower than this cohort cycles (and can therefore be discarded in providing an answer). A "typical" Cd can be read off the first graph in the paper (0.65). [ CdA will depend on the size and position of the athlete/bike, obv.]
The Drag Crisis Phenomenon on an Elite Road Cyclist—A Preliminary Numerical Simulations Analysis in the Aero Position at Different Speeds
https://mdpi-res.com/...f?version=1594566693

Great analysis, I had gone through the paper for an answer but missed the points you make here! Very helpful for me personally.

On the less slick side I'd say ~ .220 and on the most aero side .205.

On avg I'd say that the avg person coming to me in the tunnel is in that .225-.240 range and typically I can knock off .015-.020.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Way worse than that. MOP and slower is gonna be .27+ish.

I got to 70.3 worlds/KQd on a position and setup that I honestly didn't think was that bad-- visually it was ehh (not great) and performance wise--just fine. certainly didn't stand out from the median ironman setup you see looking around among mid pack dudes. I go low 2:20s in my 70.3s, I just had 5th fastest split in 30-34m at Choo last year... so i thought I would be going for some marginal gains when i connected with desert dude and went to the A2 tunnel.

I was .27 to start and he called my original position the 'human parachute'-- so MOP is gonna be that or worse since I honestly wasn't THAT bad visually.

I went with a group of FOP type-guys to the tunnel in march--the positions that looked good to start probably are more like .24 (most guys who were fairly dialed in our tunnel trip started around here) and the super slippery ones got to like .22 by the end.

Grain of salt and all that but recent out/back TT Notio testing had me at .195 to .198 combined with a guess of crr at 0.400 on a “country” road surface. 28mph test speed.

If BRR lists my tires at 0.300 at my pressure then a country road is slower for sure.

I am trying push the 10w per mph threshold as FAR as possible so I have a gnat’s ass chance at 30mph 10mi TT….someday.

That isn’t a quality study. I don’t think anyone serious would use such a small domain. Mesh is pretty coarse, no mention of prism layers or y+ or if wall functions used or not. Turbulence intensity at 1e-6%?? Massively overconfident in scan precision. If you’re going to study transition, you can’t use a smooth/uniform boundary condition for skin and fabrics… and 42M cells is nowhere near enough to assume these kinds of details are in the mesh.

RANS (and in particular k-epsilon) is bad at predicting transition anyway, but too many other issues here to take this paper seriously.

codygo, you seem to really understand this stuff. For those of us less knowledgeable, how does Cd change with speed? My weak understanding is Cd is a dimensionless coefficient, and if so why would that coefficient change with speed? I can certainly understand that there is some aspects of fluid properties that may be speed related, but the Cd change thing has me baffled. I guess in truth I am curious what in the drag equation is speed dependent since speed is already accounted for in the basic equation. My math skills are less than stellar, but if 2 variables in an equation change with one input change that is one weird curve. Any insights would be appreciated.

Cody knows his stuff, hopefully he can dumb it down so I can understand it :-)

Someone kind of derailed the thread with the "at what speed" to which I would answer "at the speed of a front of the pack triathlete". I am guessing, but the OP means draft illegal triathlon so the answer is probably 40ish km/h / 25mph.

Measured Cda will vary with speed and there is a whole conversation of the difference in how fabrics and head tube behave differently and if you are testing between 36 and 48km/h you probably have MUCH bigger things to worry about when trying to get accurate numbers.

RChung said it best with :
'I think the important question isn't "are CdA and Crr speed-dependent?" We think they are. The important question is "how speed-dependent are they over the range of speeds we experience as cyclists?"'

The answer : not much

I would add another dimension to Burn, DD and others input.

A "aero conscious" MOP may be .24ish. These are the kind of people DD sees. They go see him because they are aero conscious.

But the real MOPer who's level of consciousness is "I need Zipp wheels to go fast" is probably .27ish.

At the pointy edge more people are aero conscious athletes. If I go to sportstats, pick Oceanside, take bib numbers, look at finishpix, I would say the FOP guys are .23ish-.24ish. Not in the tunnel where they put their head in a perfect position but where they are on race day.

I have seen a ton of guys north of .25 but a lot of them are aero unconscious.

Aerometers don't sell much because there is a whole lot of aero unconscousness/incompetence out there (along with a bunch of other problems)