*Tour Aero Bike Tests*

[C36]

If memory is correct, Tour averages across yaw equally.

In reality, bikes like S5 might be even faster as their low angle performance is historically extremely strong

I think this is incorrect and there is a ponderation for each angle. I can’t remember which one they use but they use one (at one point they used the Mavic one, I think it was the first one made public).

Where do you get that cervelos are good at low angles?

[Lina]

Historically the Cervelos have dominated low yaw by a considerable margin (in comparison to competition). Thus, they're penalized by Tour's weighting methodology, which nobody agrees with since it doesn't match real world weighting. Kona is most certainly an edge case, but even then it usually is a sub 12 degree yaw

Depends entirely on where you live. Global averages matter very little for most people. Someone lives in an almost still climate and is always riding practically into 0 yaw. Then on the other hand you have people who are regularly riding in +10 m/s winds. Optimally you'd have 0 yaw and high yaw results but it's a lot easier to just give 1 number.

[CAAD8FRED]

If memory is correct, Tour averages across yaw equally.

In reality, bikes like S5 might be even faster as their low angle performance is historically extremely strong

I think this is incorrect and there is a ponderation for each angle. I can’t remember which one they use but they use one (at one point they used the Mavic one, I think it was the first one made public).

Where do you get that cervelos are good at low angles?

I believe the Mavic distribution of yaw is what the magazine uses AND that distribution of wind yaw angles is pretty close to a standard normal distribution with a mean of 0

[C36]

Depends entirely on where you live. Global averages matter very little for most people. Someone lives in an almost still climate and is always riding practically into 0 yaw. Then on the other hand you have people who are regularly riding in +10 m/s winds. Optimally you'd have 0 yaw and high yaw results but it's a lot easier to just give 1 number.

Indeed it "depends" where you live, https://www.slowtwitch.com/Tech/Real_Wo ... _5844.html they compiled few of the publised yaw angle studdies and clearly the ones measured at Kona see higher angles than the others.

I recall the 2013 paper from Mavic at Kona, I also know they had a more std yaw angle law way before this. I also know some brands used the data published for building / wind farms... but they are usually measured at 10m hight so quite different from the ones seen by a bike.

[HBike]

If memory is correct, Tour averages across yaw equally.

In reality, bikes like S5 might be even faster as their low angle performance is historically extremely strong

Almost every bike is stronger at common lower yaw angles (with differences there not that huge). Nothing special. Whether the S5 was especially optimised for this set of yaw angles - I don't know. Main differentiation of the Tour results is based on differences at larger yaw angles. And here design goals come into play (see below). Similar with deeper wheels (look at the spread where strong differences occur north of 15 degrees yaw).

From an article at Cyclist:
Trek: ‘In the real world 2.5° to 12.5° are the most prevalent yaw angles riders encounter.’
Yu at Specialized adds, ‘For an average cyclist, unless you’re riding in extremely windy conditions, the typical angles are less than 10°.’
This slight difference in results is why one aero bike doesn’t look identical to another. Specialized designed the Venge ViAS based on its vision of the perfect range of yaw, while Trek designed the Madone to fit a different range.
So it seems that if you’re Peter Sagan, driving the peloton along at 50kmh, you want a bike optimised to deal with yaw angles of around 3°-7°, while the rest of us want a bike designed to tackle yaws of up to 10°-12°.
‘If you look at a WorldTour sprinter coming off a wheel in the last 200m of a race, the effective yaw is extraordinarily low – close to 0°. That’s because they’re going really fast, more than 60kmh, and finishing straights are typically well shielded by barriers and crowds, which serve to block any crosswinds.
‘On the other hand, if you go to the Kona Ironman World Championships, they ride up the Hawaiian coast, with the wind blowing in across the water, so for an age-grouper at Kona the effective yaw angles hit up to the 15° range if it’s gusting. Pros will be going a little bit faster, so they’ll see yaw angles of up to 10° or so – maybe low teens,’ says Yu.

Historically the Cervelos have dominated low yaw by a considerable margin (in comparison to competition). Thus, they're penalized by Tour's weighting methodology, which nobody agrees with since it doesn't match real world weighting. Kona is most certainly an edge case, but even then it usually is a sub 12 degree yaw

Without data proving this for actual Cevelos, this is just heresay. Tour's weighting is acceptable and transparent. I would wish for curves instead of single values when testing bikes, however.

[justkeepedaling]

If memory is correct, Tour averages across yaw equally.

In reality, bikes like S5 might be even faster as their low angle performance is historically extremely strong

Almost every bike is stronger at common lower yaw angles (with differences there not that huge). Nothing special. Whether the S5 was especially optimised for this set of yaw angles - I don't know. Main differentiation of the Tour results is based on differences at larger yaw angles. And here design goals come into play (see below). Similar with deeper wheels (look at the spread where strong differences occur north of 15 degrees yaw).

From an article at Cyclist:
Trek: ‘In the real world 2.5° to 12.5° are the most prevalent yaw angles riders encounter.’
Yu at Specialized adds, ‘For an average cyclist, unless you’re riding in extremely windy conditions, the typical angles are less than 10°.’
This slight difference in results is why one aero bike doesn’t look identical to another. Specialized designed the Venge ViAS based on its vision of the perfect range of yaw, while Trek designed the Madone to fit a different range.
So it seems that if you’re Peter Sagan, driving the peloton along at 50kmh, you want a bike optimised to deal with yaw angles of around 3°-7°, while the rest of us want a bike designed to tackle yaws of up to 10°-12°.
‘If you look at a WorldTour sprinter coming off a wheel in the last 200m of a race, the effective yaw is extraordinarily low – close to 0°. That’s because they’re going really fast, more than 60kmh, and finishing straights are typically well shielded by barriers and crowds, which serve to block any crosswinds.
‘On the other hand, if you go to the Kona Ironman World Championships, they ride up the Hawaiian coast, with the wind blowing in across the water, so for an age-grouper at Kona the effective yaw angles hit up to the 15° range if it’s gusting. Pros will be going a little bit faster, so they’ll see yaw angles of up to 10° or so – maybe low teens,’ says Yu.

Historically the Cervelos have dominated low yaw by a considerable margin (in comparison to competition). Thus, they're penalized by Tour's weighting methodology, which nobody agrees with since it doesn't match real world weighting. Kona is most certainly an edge case, but even then it usually is a sub 12 degree yaw

Without data proving this for actual Cevelos, this is just heresay. Tour's weighting is acceptable and transparent. I would wish for curves instead of single values when testing bikes, however.

As I said, it's historically true. Since the S3 first came out. The reason being they have one of the narrowest headtubes out there. Even in competitor data, they always were/are the lowest drag at low yaw. Those include Trek, Felt, Giant, QR, and a couple others that I've since forgotten. It also shows up in independent testing

https://aerogeeks.com/2013/09/24/2014-f ... -analysis/

https://aerogeeks.wordpress.com/2013/12 ... irst-ride/

https://cyclingmagazine.ca/sections/new ... alifornia/

Since we're now talking Kona, Cervelo TT/Tri has dominated low yaw for over a decade

https://aerogeeks.com/2014/10/29/felt-i ... -analysis/

https://www.slowtwitch.com/Products/Tri ... _6756.html

https://www.slowtwitch.com/Tech/Triathl ... _6429.html

https://www.speedtheory.co.nz/blog/post ... l-testing/

Ironically, Tour mag is the only test I've ever seen with a poor Cervelo low yaw result. Partially the reason why I've taken their results with a grain of salt over the years

http://djconnel.blogspot.com/2011/02/to ... rames.html

[HBike]

Almost every bike is stronger at common lower yaw angles (with differences there not that huge). Nothing special. Whether the S5 was especially optimised for this set of yaw angles - I don't know. Main differentiation of the Tour results is based on differences at larger yaw angles. And here design goals come into play (see below). Similar with deeper wheels (look at the spread where strong differences occur north of 15 degrees yaw).

From an article at Cyclist:
Trek: ‘In the real world 2.5° to 12.5° are the most prevalent yaw angles riders encounter.’
Yu at Specialized adds, ‘For an average cyclist, unless you’re riding in extremely windy conditions, the typical angles are less than 10°.’
This slight difference in results is why one aero bike doesn’t look identical to another. Specialized designed the Venge ViAS based on its vision of the perfect range of yaw, while Trek designed the Madone to fit a different range.
So it seems that if you’re Peter Sagan, driving the peloton along at 50kmh, you want a bike optimised to deal with yaw angles of around 3°-7°, while the rest of us want a bike designed to tackle yaws of up to 10°-12°.
‘If you look at a WorldTour sprinter coming off a wheel in the last 200m of a race, the effective yaw is extraordinarily low – close to 0°. That’s because they’re going really fast, more than 60kmh, and finishing straights are typically well shielded by barriers and crowds, which serve to block any crosswinds.
‘On the other hand, if you go to the Kona Ironman World Championships, they ride up the Hawaiian coast, with the wind blowing in across the water, so for an age-grouper at Kona the effective yaw angles hit up to the 15° range if it’s gusting. Pros will be going a little bit faster, so they’ll see yaw angles of up to 10° or so – maybe low teens,’ says Yu.

Historically the Cervelos have dominated low yaw by a considerable margin (in comparison to competition). Thus, they're penalized by Tour's weighting methodology, which nobody agrees with since it doesn't match real world weighting. Kona is most certainly an edge case, but even then it usually is a sub 12 degree yaw

Without data proving this for actual Cevelos, this is just heresay. Tour's weighting is acceptable and transparent. I would wish for curves instead of single values when testing bikes, however.

As I said, it's historically true. Since the S3 first came out. The reason being they have one of the narrowest headtubes out there. Even in competitor data, they always were/are the lowest drag at low yaw. Those include Trek, Felt, Giant, QR, and a couple others that I've since forgotten. It also shows up in independent testing

https://aerogeeks.com/2013/09/24/2014-f ... -analysis/

https://aerogeeks.wordpress.com/2013/12 ... irst-ride/

https://cyclingmagazine.ca/sections/new ... alifornia/

Since we're now talking Kona, Cervelo TT/Tri has dominated low yaw for over a decade

https://aerogeeks.com/2014/10/29/felt-i ... -analysis/

https://www.slowtwitch.com/Products/Tri ... _6756.html

https://www.slowtwitch.com/Tech/Triathl ... _6429.html

https://www.speedtheory.co.nz/blog/post ... l-testing/

Ironically, Tour mag is the only test I've ever seen with a poor Cervelo low yaw result. Partially the reason why I've taken their results with a grain of salt over the years

http://djconnel.blogspot.com/2011/02/to ... rames.html

Thanks. You provided some nice data points. We don't know whether this translates to the most recent model (S5), too, but I am inclinded to belive you, here.

[SixThirteen]

How far back does the data go?

I have a 2010 Cervelo S3. Only the daddy of aero, the original Cervelo Soloist, is older. Curious how that era stacks up to more modern frames.

Using Cervelo's figures, the first gen S5 was 9 watts faster than the S3. Again using Cervélo's figures, the 2015 S5 (which tested at 205 watts @ 45kmh according to tour mag) was 21 watts faster than the 2012 Gen 1 S5. So 30 watts difference between the S3 and the 2016 S5? (The new disc S5s are basically no faster than the 2015 rim S5). Seems optimistic. But if those tests were done with bottles, it could make sense. I had an S2 and it looked so obviously unaero optimized with bottles on the bike. If you put that crazy fat 2015 S5 bar on an S3, that would cut the delt down by something like 5 watts IIRC. 25 watts for the rest of the frame design (wheel cut out, bottle optimized DT, dropped stays, smoothed fork transition to frame and deeper sections)... I dunno. Maybe?

I had an S3 which broke. Tour had a test where they took top of the range and bottom of the range aero bikes and switched wheels, so basically they tested S2 (2017 I think, same frame as S3) with the same wheels as same gen S5, and it was about 5 or 10W if memory serves. I've got the data somewhere. That difference narrows if you put aero bars on the S3, and is within the limit for tweaks to fit and position.

[SixThirteen]

It’s 2.5 watts per tube/tire at 18mph. And I didn’t say specifically the absolute speed difference would be detectable. What would be detectable is the difference in RPE while sitting in a group going 25mph. Do you agree?

I'm not convinced it would be noticeable to be honest, but there's a measureable difference, so that's enough to make it worthwhile. However, your choice of clothing that morning might make even more of a difference.

https://www.bicyclerollingresistance.com/ is your friend. Jano specifically tested Gatorskin (approx 20W per tyre) and GP5000 with butyl tube (approx 11W per tyre).

That 18W saving works in the bunch of course

Also latex tubes vs tubless https://www.bicyclerollingresistance.co ... s-vs-tubes

Spoiler alert, the latex tubes are worth about 1.5W per tyre over the butyl, and the tubeless a further approx 2W per tyre

[HBike]

It’s 2.5 watts per tube/tire at 18mph. And I didn’t say specifically the absolute speed difference would be detectable. What would be detectable is the difference in RPE while sitting in a group going 25mph. Do you agree?

I'm not convinced it would be noticeable to be honest, but there's a measureable difference, so that's enough to make it worthwhile. However, your choice of clothing that morning might make even more of a difference.

https://www.bicyclerollingresistance.com/ is your friend. Jano specifically tested Gatorskin (approx 20W per tyre) and GP5000 with butyl tube (approx 11W per tyre).

That 18W saving works in the bunch of course

Also latex tubes vs tubless https://www.bicyclerollingresistance.co ... s-vs-tubes

Spoiler alert, the latex tubes are worth about 1.5W per tyre over the butyl, and the tubeless a further approx 2W per tyre

Difference latex to sealant is about 0.2W-0.4W depending on pressure. And, it is with only 20 ml of sealant. With 40ml you add another 0.5W according to aero-coach. So real-life differences including margin of error are very small.

[wheelsONfire]

I was told S5 ang Dogma F have delivery times at + 6 months.

[Steve Curtis]

How far back does the data go?

I have a 2010 Cervelo S3. Only the daddy of aero, the original Cervelo Soloist, is older. Curious how that era stacks up to more modern frames.

Using Cervelo's figures, the first gen S5 was 9 watts faster than the S3. Again using Cervélo's figures, the 2015 S5 (which tested at 205 watts @ 45kmh according to tour mag) was 21 watts faster than the 2012 Gen 1 S5. So 30 watts difference between the S3 and the 2016 S5? (The new disc S5s are basically no faster than the 2015 rim S5). Seems optimistic. But if those tests were done with bottles, it could make sense. I had an S2 and it looked so obviously unaero optimized with bottles on the bike. If you put that crazy fat 2015 S5 bar on an S3, that would cut the delt down by something like 5 watts IIRC. 25 watts for the rest of the frame design (wheel cut out, bottle optimized DT, dropped stays, smoothed fork transition to frame and deeper sections)... I dunno. Maybe?

I had an S3 which broke. Tour had a test where they took top of the range and bottom of the range aero bikes and switched wheels, so basically they tested S2 (2017 I think, same frame as S3) with the same wheels as same gen S5, and it was about 5 or 10W if memory serves. I've got the data somewhere. That difference narrows if you put aero bars on the S3, and is within the limit for tweaks to fit and position.

The s2 and s3 frames are the same but different forks.
I read on the tour site that when they tested the s2 back in 2016, they have never seen a bikes drag drop more than that bike with aero wheels. I think they quoted 20watts from the factory mavics to some 404s.

Another fact was they didn't consider the bottle when Cervelo tested that frame so bottles really effect the numbers.
No bottle or an aero bottle if you want speed

[SixThirteen]

How far back does the data go?

I have a 2010 Cervelo S3. Only the daddy of aero, the original Cervelo Soloist, is older. Curious how that era stacks up to more modern frames.

Using Cervelo's figures, the first gen S5 was 9 watts faster than the S3. Again using Cervélo's figures, the 2015 S5 (which tested at 205 watts @ 45kmh according to tour mag) was 21 watts faster than the 2012 Gen 1 S5. So 30 watts difference between the S3 and the 2016 S5? (The new disc S5s are basically no faster than the 2015 rim S5). Seems optimistic. But if those tests were done with bottles, it could make sense. I had an S2 and it looked so obviously unaero optimized with bottles on the bike. If you put that crazy fat 2015 S5 bar on an S3, that would cut the delt down by something like 5 watts IIRC. 25 watts for the rest of the frame design (wheel cut out, bottle optimized DT, dropped stays, smoothed fork transition to frame and deeper sections)... I dunno. Maybe?

I had an S3 which broke. Tour had a test where they took top of the range and bottom of the range aero bikes and switched wheels, so basically they tested S2 (2017 I think, same frame as S3) with the same wheels as same gen S5, and it was about 5 or 10W if memory serves. I've got the data somewhere. That difference narrows if you put aero bars on the S3, and is within the limit for tweaks to fit and position.

The s2 and s3 frames are the same but different forks.
I read on the tour site that when they tested the s2 back in 2016, they have never seen a bikes drag drop more than that bike with aero wheels. I think they quoted 20watts from the factory mavics to some 404s.

Another fact was they didn't consider the bottle when Cervelo tested that frame so bottles really effect the numbers.
No bottle or an aero bottle if you want speed

Here you go

[spartan]

update tour mag wrote a article about how fast 2023 bikes are up tdf mtn stage.




mmm both the cervelo s5 and canyon are at 202 watts . the new madone slr is 207. surprised .
https://www.tour-magazin.de/profi-radsp ... -5-etappe/

[spdntrxi]

^ that tells me +/- 30watts does not matter much at all