New aero test: 12 aero frames vs 12 "unaero" light frames

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

User avatar
neeb
Posts: 1101
Joined: Tue Jan 27, 2009 8:19 pm

by neeb

ichobi wrote:I think the direction of Cervelo R5 / Madone / Scott Addict that try to combine weight saving and comfort and aerodynamic is the way to go. Save a few watts here and there, while enjoying a stable and smooth ride is a nice package, all while ridding myself the assumption that I am riding an aerodynamically inferior bike compared to those with aero specific frame. Save myself a few thousands dollars.
Sort of agree, except that for me, the Foil is comfortable and nicely stiff, so I suspect that I wouldn't benefit from a slightly more "comfortable" but slightly less stiff, slightly lighter bike such the Addict.

Seems to me that except for the outliers, the aero thing is a bit of a red herring and what matters is the trade-off between stiffness/stability on the one hand, and comfort/light weight on the other. But there is no absolute best place to be on that spectrum, it depends on personal preference and riding style.

Illuminate
Posts: 404
Joined: Mon May 24, 2010 4:18 am
Location: Australia

by Illuminate

100km/4hr20min = about 23kph. Who here has raced with an ave speed of 23 (aside from audax riders)?

The thing to remember is that power needed increases as the cube of the velocity so whilst the result doesn't look impressive here, how would it look if you bumped the average up to a more realistic speed (say 35 or 40km/h) or perhaps even higher on a flatter course?

How about sprinting at the end (65km/h) - I wonder what the differences would be starting from 40 and going up to 65 over 300m in a side by side drag (everything else being equal)?

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



User avatar
SolidSnake03
Posts: 556
Joined: Wed Jan 18, 2012 7:09 pm

by SolidSnake03

dynaserve wrote:For those arguing that the measurement error is too large to make the test worthwhile - If you were to guess the fastest aero & light bikes what would they have been? Also, based on what has been published previously, if you were to predict the order of the aero bikes it would probably be pretty close (with a few exceptions)?

To me the test looks credible.


Honestly? I wouldn't be able to "guess" which is fastest because I believe that most of that is borderline non-sense. Guessing on which is fastest to me would require just going off of looks or the company name because my aerodynamic knowledge is a bit limited.

Additionally, I'm not saying the test isn't credible in any way, it looks to be a well set up test that attempts to incorporate a variety of different bikes from other brands. If you read what I posted more closely you would see that I'm not arguing about the test credibility but that the results show that the whole "aero" thing is basically a wash. The differences observed are so small that to me they show it doesn't really matte much which frame you were/are on. That was my point, a test that shows a lack of significant difference doesn't mean the test isn't credible, it might simply be that the items in question ACTUALLY AREN'T THAT DIFFERENT.
Looks like I made a new 90 Proof friend

exFictitiouZ
Posts: 55
Joined: Thu Feb 20, 2014 2:20 pm

by exFictitiouZ

The TOUR mag generated those fictitious times on a fictitious course using wind tunnel data + weight to put into picture how aero frames fare in real world, not in a final 200m-sprint. Remember that aero frames are designed primarily with flat courses and solo efforts in mind. In their parcour of 100km that gains an epic elevation of 2,000m, TOUR mag goes on to show that aero frames are no slower than, if not faster (taking into consideration the margin of error), their lighter counterparts. Now imagine if they blew up the numbers by instead assuming a totally flat, blazing fast course that averaged 40-45km/h; that would show even bigger differences.

What I am trying to say here is that the differences are there, demonstrable by scientific means. But whether those differences matter or not, depends on you and your viewpoint.

PS. The test protocols are very credible to my eyes.
Last edited by exFictitiouZ on Thu Feb 20, 2014 3:23 pm, edited 2 times in total.

HillRPete
Posts: 2284
Joined: Tue Apr 12, 2011 8:08 am
Location: Pedal Square

by HillRPete

exFictitiouZ wrote:What I am trying to say here is that the differences are there, demonstrable by scientific means.

But this is exactly the point of contention, there is no absolute accuracy, and we have been wondering whether a, say, 30s difference could be down to measurement errors just as well, with the given methodology.

Also it is dangerous to assume that the fastest frame at a given speed and yaw will be the fastest at any given speed and yaw. There are too many nonlinearities involved.

I am not challenging the claim that gear aerodynamics might decide a sprint finish, but I have to agree with SolidSnake03 that what this test shows is, at best, that the difference between frames is smaller than the manufactorers want to make you believe.

aaric
Posts: 430
Joined: Thu Jun 09, 2011 4:10 pm

by aaric

Well, the article simulates a 200w constant power (23kph? assuming Illuminate's math is right), which is relatively slow. My recollection is that the bike manufacturers generally cite much much higher numbers - 40 kph-ish, if I recall - somebody else can do the math for the watts for that.

I'm not sure what the average cyclists' wattage output is, but I'd say 200w is not competitive from my limited racing experience - so they did a great job of proving that for your average mamil, they wouldn't see great gains with an aero frame.

The reality though is that even at these low wattage/speed numbers, you'll see a bigger difference in aero gains than you will with cutting out 400g of weight. Whether or not those numbers make up for the sacrifices of comfort / fit are an individual choice.

efeballi
Posts: 492
Joined: Tue Dec 17, 2013 2:50 pm
Location: Istanbul, hopefully not for long

by efeballi

Illuminate wrote:100km/4hr20min = about 23kph. Who here has raced with an ave speed of 23 (aside from audax riders)?

The thing to remember is that power needed increases as the cube of the velocity so whilst the result doesn't look impressive here, how would it look if you bumped the average up to a more realistic speed (say 35 or 40km/h) or perhaps even higher on a flatter course?

How about sprinting at the end (65km/h) - I wonder what the differences would be starting from 40 and going up to 65 over 300m in a side by side drag (everything else being equal)?

This.
Being alone or in a breakaway, sprinting and time trialling all increase the advantages of aero road bikes.
I'd be more than interested in the same test done with TT bikes and see how close TT bikes fare against standard road bikes.


Sent from my iPhone using Tapatalk
SHUT UP LEGS
2015 Giant Propel Advanced
2015 Cannondale Supersix Evo
2013 KTM Strada mod. (totaled)
2011 Pinarello Dogma 60.1(loaner)
2011 Scott SUB 45(sold)

Politecnico di Milano Ingegneria Meccanica

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

exFictitiouZ wrote:Remember that aero frames are designed primarily with flat courses and solo efforts in mind. In their parcour of 100km that gains an epic elevation of 2,000m, TOUR mag goes on to show that aero frames are no slower than, if not faster (taking into consideration the margin of error), their lighter counterparts.


My issue with it is that the scenario they chose is unrealistic. Solo riding, 2% grade, 100km. Anyone doing a loop at 200W will average greater than 23km/hr unless the course is extremely hilly. 2% grade doesn't seem that much, but when you completely remove all downhill and flat, it is.

Since it is computer generated based on WT data, why not run some cases that make sense? For instance for racers... a solo break on a flat road, a finishing sprint, or a climb. For a casual rider... try a hilly loop or a flat one, riding solo.
formerly rruff...

eric
Posts: 2196
Joined: Tue Nov 11, 2003 9:47 pm
Location: Santa Cruz, California, USA
Contact:

by eric

6500 feet (2000 meters) elevation gain in 62 miles (100km) is over the 100'/mile metric that many people would condsider extremely hilly. That explains the speed. 200w average for the whole 100km is reasonable for a less than top level amateur racer or strong recreational rider. Depending on the rider size in their model- I am assuming its not 180 lbs.

Simulating it at 2% for the entire 100km is totally unrealistic however. If that's what they actually did- I have not seen the article.
A course that's 1/3 climbing, 1/3 descent down that climb and 1/3 flat (by distance) would be more realistic of a hilly road race. The aero bike would have a greater advantage on the descent and flat and more of a disadvantage on the climb.

User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

Yep, scenario makes no sense. Doesn't even apply to a solo rider on a very hilly loop very well. It's easy enough to model something that makes sense, so why not?

Like I said, for racers the important part of the race will be at key points... sprint, breakaway, or climb... and there is no reason to worry about anything else.
formerly rruff...

User avatar
fa63
Posts: 2533
Joined: Mon Feb 19, 2007 7:26 am
Location: Atlanta, GA, US

by fa63

I think they have simulated a fairly heavy rider; in fact, the simulation results are very close to the numbers (in terms of distance, elevation gain and average speed) I see for myself on a few of my solo 100 km rides, where I am not riding slow but also not trying very hard either.

Regardless, it would appear to me that they created their model by creating a hypothetical course, discretizing the distance/elevation along the route, then estimating the velocity at any discrete point along the course for a given power, weight of rider & bike, frontal area, coefficient of wind drag, air density and rolling resistance. Obviously, lots of assumptions have to go into such a model, but I am sure they kept everything else constant and only changed the coefficient of wind drag based on the values they measured in the tunnel so the difference in the results would only be affected by that change. Whether or not the model differences translates into real world differences is of course debatable, but I think the results should not be dismissed nonetheless. It would be interesting to see what happens if the power was upper to the 250W range, and maybe for a lighter rider + bike combo.

hiro11
Posts: 41
Joined: Sat Jul 24, 2010 2:10 am

by hiro11

Even if these results are real (surely they're within the test's margin of error), the benefits of aero frames seem entirely meaningless. 0.4% savings in lab conditions? Who cares.

User avatar
DMF
Posts: 1062
Joined: Sat Dec 06, 2008 10:14 am
Location: Sweden

by DMF

Well, in an aero frames defence. What if you could have no net loss versus a conventional frame over the entire race (seen in total), but you could have some serious benefits when trying to solo break your way out of one group to close the gap to the group 300m in front of you... I'm fairly certain that the gains aren't as marginal at 65kph as they are at 23kph...

It's not about being overall better, that is just something you spit out when your intention all along is about undermining the benefits of aero frames (or wheels, helmets, clothes, whatever, always the same stupid argument), but what it's really about is having an ace up your sleeve at those key moments.

Aero won't take you to a two-man-sprint-showdown at the finish line, but if you get there yourself, it might tip the scales in your favour.

Take aero for what it is instead of pointing out everything it isn't.

User avatar
fa63
Posts: 2533
Joined: Mon Feb 19, 2007 7:26 am
Location: Atlanta, GA, US

by fa63

I think you guys are confusing the difference in drag between frames and the difference in the overall simulation times. I am sure the difference in drag between the frames is quantifiable and outside the margin of error in the measurements. However, frame drag is just one small component of the speed equation, so when that one variable is changed in the simulation the percentage difference in overall times is not as great as the percentage difference between the drag of the frames (for example, a 10 percent more aero frame is not 10 percent faster overall).

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



User avatar
WMW
in the industry
Posts: 893
Joined: Sun Jul 29, 2012 2:59 pm
Location: Ruidoso, NM

by WMW

fa63 wrote:Regardless, it would appear to me that they created their model by creating a hypothetical course, discretizing the distance/elevation along the route, then estimating the velocity at any discrete point along the course for a given power, weight of rider & bike, frontal area, coefficient of wind drag, air density and rolling resistance.


I don't have the article, but it looks like ~160lb rider... 200W and a constant 2% grade to get that speed.
formerly rruff...

Post Reply