RoadBikeReview.com's Forum Archives - General


Archive Home >> General(1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 )


aerodynamics test(19 posts)

aerodynamics testDougSloan
Jun 17, 2002 6:23 AM
I wanted to do a test of a "normal" bike vs. a time trial bike, and I thought top speed down a hill might give some useful information.

There is a hill about 10 miles out of town here that is around 8-10%. I takes me about 5 minutes to climb it. It has some variation in grade, and has several turns that can be taken flat out, if you are familiar with the road and keep saying to yourself "don't brake, don't brake, it'll be ok".

The top speed I ever saw descending this hill in the most radical position I could achieve, that is, cheek down against the stem, hands narrow on the tops, knees in, was 55.0 mph (that may have been an exceptional day) a year ago. That was on the Colnago with Ksyrium wheels.

Last week I rode the hill on my Colnago with Nucleon wheels. Not exactly an aero set up, but the Nucleons do have low count, bladed spokes. With hands in drops and in a "normal" low aero descending postion, I hit 51.6 mph, pedaling a 53x11 gear in a sprint at the top.

On Saturday, I ride the Cervelo P3 with Zipp disc, Zipp 404, Tufo 215 tubulars, integrated aerobars, skinsuit, but with my regular helmet, one water bottle, and 56x11 gears. The latter variable might have nullified the experiment, though, as it introduced variable other than drag.

Both times recently, the computers were calibrated with rollout measurements to the nearest millimeter with the tire inflation pressures used on the runs. Each bike had it's own Cateye computer calibrated for that bike's wheel.

Sprinting at the top and descending in the normal aerobar position on the Cervelo, I hit 56.0 mph. This is interesting, as the position can be held all day long, as opposed to the racing bike aero position, which becomes uncomfortable in less than a minute, and it is extremely difficult, if not counter-productive, to pedal in that position.

The gearing variable migh be significant. I didn't notice the first run, but on the Cervelo I was able to pedal up to 51 mph before coasting. With a 56x11 gear, that's not very difficult.

I seem to be able to get much, much lower, actually, on a standard racing bike than on the time trial bike. The aerobars, with arms close togther and out front, won't allow the head and upper body to get nearly as low as a standard bike. The difference is that on the time trial bike, the position is easily held, actually restful, while the racing bike tuck is almost as fatiguing to hold.

I was able to at least quantify some difference in the two bikes down the hill with the two types of bikes. I'll retest the Cervelo in a 53x11 gear for closer comparison, but I doubt that it will be a significant difference. Sprinting power at the top might be different, also, on different days. I'm not sure how to sprint and get the same speed at a certain point in the course just before coasting to control for this. Maybe I'll just do this a number of times to help control for initial speed. To me, it was more important to get the top speed as high to make aerdynamics as much as a factor as possible. In any event, maybe the test shouldn't be looked upon as purely looking at aerodynamics, but as which bike is faster, period, as outfitted and normally ridden (considering it's ability to sprint, coast, etc.).

Any suggestions for further experimentation (assuming time to do this)? BTW, this was incidental to a regular workout, too, not blowing time just doing this hill.

Doug
correctionDougSloan
Jun 17, 2002 6:27 AM
...while the racing bike tuck is almost as fatiguing to hold *as pedaling hard.*
Don't pedal,TJeanloz
Jun 17, 2002 6:51 AM
When you pedal, you're introducing more than aerodynamics into the equation. A coast-down test would probably give you a better idea of what portion of the speed increase can be accounted for by aerodynamics. You probably should still turn the cranks- but not in a gear that provides you any acceleration.
That's what I was going to saycory
Jun 17, 2002 6:56 AM
On the basis of some roughly similar tests with cars, I think the pedaling introduces a variable you can't ignore--a difference in output of even 1 or 2 percent will queer the results. A sampling of dead-stick coastdowns, from a standing start, might be more accurate. I think you'd need several, though, to rule out minor differences in wind and pavement irregularities.
hmmDougSloan
Jun 17, 2002 7:26 AM
Thanks, guys.

I don't want rolling resistance, internal friction, or weight to affect the results, either. I understand that those things dominate at low speeds. Maybe it would be best to pedal up to, say, 20 mph and then coast? The Zipp rear hub does seem to have quite a lot of friction when coasting; the ratcheting mechanism just seems "tighter" than a Campy or Mavic. That might prevent it from gaining speed early on.

On this hill, the terminal speed occurs very near the bottom, so initial speed probably does matter.

Maybe it would be more significant to ride each bike in the "normal" riding position, too. I think then the results will be vastly different between a racing bike in the drops and the time trial bike.

The racing bike had tires at 120 psi, and the time trial bike at 140 psi. Sure, that's a variable, but that's how they normally are ridden (by me).

I'll try various ways and see that happens over the next few months. Thanks.

Doug
coasting vs spinningcyclopathic
Jun 17, 2002 9:34 AM
Doug,

I found that at speed 27mph+ on 7-8% hill you're actually faster coasting (cranks parallel, knees hug TT). I'd suggest sprinting up to 30mph then let it go. Second with respect to freehub the data I've seen suggest the diff in 3-8wt range, try to feed it in analyticcycling.

btw roadbike with aerobars is comfortable just takes time to get used to.
coasting vs spinning ?filtersweep
Jun 17, 2002 7:27 PM
"It has some variation in grade, and has several turns that can be taken flat out, if you are familiar with the road and keep saying to yourself "don't brake, don't brake, it'll be ok"."

I have a friend who does TTs who SWEARS that coasting down a windy hill is dangerous because it is more difficult to control the bike if the rear wheel is not under power. My own experience on a motorcycle seems to support this, you simply do NOT want to coast through a curve or turn.

His buddy broke his collarbone on a bad crash coasting through a turn... apparently he lost control and crashed in the ditch.

While I realize speed is the ultimate goal in any sort of race situation (as well as minor things such as finishing the race), does anyone have a thought on the relative merits of pedaling through descents under the guise of maintaining better control of the bike?
Removing variables.Len J
Jun 17, 2002 9:50 AM
If you truly want to see the effect of the aerodynamic frame & rider position, put the same wheels on both & coast.

If you want the Aero effect of just the aero wheels, put them on both bikes & coast.

If you want the aero efficiency of the TT bike with Zipps compared to the Racing with K's. Coast each of them. The friction you talk about in the hub is part of your TT bike, so it should be part of the comparison.

My .02

Len
actuallyDougSloan
Jun 17, 2002 12:55 PM
I'm wanting to measure the complete package of both, as typically ridden.

What I should do is assume a "normal" riding position on the racing bike, as positioning is the dominate factor in aerodyamic drag, it seems. I would bet that the difference will be enormous (my hypothesis).

The hub friction when coasting (in the freewheel mechanism) really should be eliminated as a variable, if possible, as I don't think it's a significant variable in a race. In a time trial, if you are coasting, you are probably on the brakes, too.

Thanks.

Doug
actuallyLen J
Jun 17, 2002 1:01 PM
You say:
"The hub friction when coasting (in the freewheel mechanism) really should be eliminated as a variable, if possible, as I don't think it's a significant variable in a race. In a time trial, if you are coasting, you are probably on the brakes, too. "

Don't you have the hub friction weather your peddling or not? Therefor the only way to eliminate it is to truly eliminate the friction, coasting or peddling is irrelevant. What am I missing?

Len
ratchetingDougSloan
Jun 17, 2002 1:38 PM
I'm just referring to the ratcheting mechanism, not the bearings. When pedaling, the mechanism engages instead of ratcheting.

Doug
Um, do what you want, butdjg
Jun 17, 2002 2:53 PM
if you're trying to sell us on the idea that pedaling is an effective experimental control in your attempt to measure...whatever...then I'm not sure what to say. So maybe I'll just stop.
re: aerodynamics testSTEELYeyed
Jun 17, 2002 6:55 AM
Thats an interesting test,although I think each bike tested should be ridden in a normal riding position for the test to get a true evaluation,maybe a clip on aero bar on the road bike? Check out the link http://popularmechanicsdoors/bicycles/1998/8/Wind_Tunnel_Tuneup/
Great test....rwbadley
Jun 17, 2002 7:15 AM
Great Idea Doug,
I have wondered about this same thing. The closest I have gotten to an answer, is by going down my favorite long hill and experimenting with (rider) position. Lower being better of coarse, but head angle knee/pedal position etc. all make some difference.
I will stick my neck out with a guess that a true "coast down" would be a more true test than with pedaling.

However, it would seem to me that terminal velocity would be bumped up against in either case.

If you throw a rock down the crevasse it will hit the same top speed as the rock that is merely dropped, agreed? (given distance for speed to be reached would be altered)

Since the grade you descended was definitely long enough for your top speed to be reached in either case, it would appear the aero bike gave significant advantage.

It would be interesting to quantify other variables, such as tire pressure difference of say 15-20 lbs. Or 36 spoke std vs 24 spoke bladed. I know these could be quantified mathematically, but I have always been better influenced by 'real world' enactment.

Thanks Doug,
RW
An Aerodynamicist responds. . . (very long)yfoiler
Jun 17, 2002 11:58 AM
The following was in response to a question I posed to a NASA aero specialist about why my TT's were quicker on my Trek "y"foil bike than on my disk wheeled, slope tubed, aero front wheeled, TT bike.

I found the response most interesting and hope you all do too.

************* Text Follows *************

Date: Thu, 4 Apr 2002 10:56:00 -0800 (PST)

From: "Doug Isaacson" Subject: Re: The Y-foil aero?

To: "yfoiler"

hey marty,

too many variables to make an educated guess as to why you are faster on your y than your tt bike...

so i'll make an ignorant guess ;) my guess would be fit, comfort, and 'muscle memory'... not sure if that is the right term, but your muscles have to train the exact movements over and over and over again to become proficient... you may tire more quickly on your tt bike if it is not the exact same geometry you train in. to illustrate, i've started competitive swimming when i was 6 yrs old, swam through high school and college, and started triathlons after college. never used a wetsuit until doing some cold water tris... bought a nice wetsuit with plenty of flexibility, but couldn't swim as fast in it (while i theoreticllay should have been faster)... turns out, i was still using my standard kick, but without the wetsuit, the only thing i use my kick for is to level my body position... with the wetsuit, the only way i replicated my training (non-wetsuit) position was to stop kicking entirely... then i gained about 5% speed at the same heartrate... same thing applies on bikes (or any sport for that matter). i think there was a book written about this very topic as it applied to michael jordan's attempt (failed) at baseball... it was titled "Why Michael Can't Hit".

now to the real question... is the y more aero??? my answer is... it is not less aero than other bikes. lack of a pocket??? i doubt that has much to do with the y; probably just good positioning... the body is still the VAST majority of the drag (something like 85%), so even taking away the bike entirely would have roughly the same pocket.

that sidewind thing had been debated so much on newsgroups with disc wheels that i wrote a webpage to state my viewpoint ;)

http://www.geocities.com/doug_isaacson/discthrust.html

there are some links there where i derive the basic aero theory and show results of a frequently quoted study.

the same thing applies (to some extent) to the y's frame (as well as any frame, but with longer 'chord' the y gets more benefit).

it is amazing to me how much marketing there is on the aerodynamics of bikes, when the fact of the matter is that NO company has invested enough to even come close to what the ideal aerodynamic bike would look like. cervelo is really pushing their 'true aero' tubing... they extoll the virtues of using the industry's only truly aerodynamic tubing; a NACA airfoil. having done my share of computational CFD, as well as wind tunnel tests, and with a NASA library at my disposal :) i set out to investigate if theirs was a wise choice. first thing was to look at an industry standard book from college "theory of wing sections" to check out the airfoil(s) they chose... in my estimation a symmetrical NACA0012 airfoil or something VERY similar (doesn't really matter). the first thing i was reminded of was that all of these airfoils were designed for WWII bomber vertical tail fins... ie high speed, big wing. the performance of these airfoils deteriorated rapidly at off-design points. airfoils are designed for a given 'reynolds number'... the NACA airfoils were designed for a reynolds number of roughly 6 MILLION. an 'aero' bicycle downtube has a reynolds number in the 50-80 THOUSAND range. result? guaranteed flow separation near the leading edge (just like a round tube). so i looked into low reynolds number aerodynamics, and this is where i realized why no bike company has produced a
FINISH THE POST!!MisJG
Jun 17, 2002 12:08 PM
just as it was getting good. . . I wanna know why no company has produced the ideal aero bike! I gotta know what the tubes would look like!
Sorry it got chopped off HERE IT IS !!!!!!yfoiler
Jun 19, 2002 10:55 AM
hey marty,

too many variables to make an educated guess as to why you are faster on your y than your tt bike...

so i'll make an ignorant guess ;) my guess would be fit, comfort, and 'muscle memory'... not sure if that is the right term, but your muscles have to train the exact movements over and over and over again to become proficient... you may tire more quickly on your tt bike if it is not the exact same geometry you train in. to illustrate, i've started competitive swimming when i was 6 yrs old, swam through high school and college, and started triathlons after college. never used a wetsuit until doing some cold water tris... bought a nice wetsuit with plenty of flexibility, but couldn't swim as fast in it (while i theoreticllay should have been faster)... turns out, i was still using my standard kick, but without the wetsuit, the only thing i use my kick for is to level my body position... with the wetsuit, the only way i replicated my training (non-wetsuit) position was to stop kicking entirely... then i gained about 5% speed at the same heartrate... same thing applies on bikes (or any sport for that matter). i think there was a book written about this very topic as it applied to michael jordan's attempt (failed) at baseball... it was titled "Why Michael Can't Hit".

now to the real question... is the y more aero??? my answer is... it is not less aero than other bikes. lack of a pocket??? i doubt that has much to do with the y; probably just good positioning... the body is still the VAST majority of the drag (something like 85%), so even taking away the bike entirely would have roughly the same pocket.

that sidewind thing had been debated so much on newsgroups with disc wheels that i wrote a webpage to state my viewpoint ;)

http://www.geocities.com/doug_isaacson/discthrust.html

there are some links there where i derive the basic aero theory and show results of a frequently quoted study.

the same thing applies (to some extent) to the y's frame (as well as any frame, but with longer 'chord' the y gets more benefit).

it is amazing to me how much marketing there is on the aerodynamics of bikes, when the fact of the matter is that NO company has invested enough to even come close to what the ideal aerodynamic bike would look like. cervelo is really pushing their 'true aero' tubing... they extoll the virtues of using the industry's only truly aerodynamic tubing; a NACA airfoil. having done my share of computational CFD, as well as wind tunnel tests, and with a NASA library at my disposal :) i set out to investigate if theirs was a wise choice. first thing was to look at an industry standard book from college "theory of wing sections" to check out the airfoil(s) they chose... in my estimation a symmetrical NACA0012 airfoil or something VERY similar (doesn't really matter). the first thing i was reminded of was that all of these airfoils were designed for WWII bomber vertical tail fins... ie high speed, big wing. the performance of these airfoils deteriorated rapidly at off-design points. airfoils are designed for a given 'reynolds number'... the NACA airfoils were designed for a reynolds number of roughly 6 MILLION. an 'aero' bicycle downtube has a reynolds number in the 50-80 THOUSAND range. result? guaranteed flow separation near the leading edge (just like a round tube). so i looked into low reynolds number aerodynamics, and this is where i realized why no bike company has produced a truly aerodynamic bike. most wind-tunnels are incapable of testing at low reynolds numbers... even all the wind tunnel testing done for lance armstrong at Texas A&M is pushing the limits of that tunnel to the point i question the results (and that is the whole bike; a higher reynolds number since it increases proportionally with speed and length). the only choice a bike company has is to build a never before built wind tunnel (yeah right) or use CFD (and pay a crew of CFD engineers to build
d'oh! still there's more missing....(nm)JS Haiku Shop
Jun 19, 2002 11:22 AM
I GIVE UP. WHEN I POST, IT'S ALL THERE ?????yfoiler
Jun 19, 2002 3:06 PM
I don't know why it's getting chopped off.
When I post it the entire text of the message is there.
Then it gets cut off.
WTF??!! I've seen way longer posts.

I give up.
If anybody is that interested e-mail me directly and I will send you the entire thing as an e-mial

Marty
yfoiler@yahoo.com