Tubular Tires, Wheels and Aerodynamics

Everything about building wheels, glueing tubs, etc.
Jannekallio
Posts: 4
Joined: Tue Jan 15, 2019 2:44 am

by Jannekallio

Based on this:

https://docs.google.com/spreadsheets/d/ ... /htmlview#
One should use zipp tangente


https://www.bicyclerollingresistance.co ... =7&minpr=7
And based on this vittoria corsa speed

Any other suggestions?

RocketRacing
Posts: 878
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

RyanH wrote:
Sat Jan 12, 2019 1:39 am
@mep

Check the link I put in my OP. In part 2 they added the aero drag to the crr penalty to come up with a total system cost.

Unless Hambini suggests otherwise, it seems there is a concensus that tread pattern matters. Conti Gp4k seem to be the optimal tread pattern.

Maybe I'm misinterpreting, but I think Hambini's test suggest that the rule of 105% might not be an actual rule. On many of the wheels in his chart, the 23mm Conti GP4K (they're possibly 25mm/26mm actual width at the tested 119 psi) measures wider than the rim yet they still perform well. In one of his comments, he suggested to not get bogged down by a millimeter here or there on tire size.
I think that was my comment. Note the error of his test is 2.5%. That brings a mid scoring wheel of 593w to 578w to 608w. That covers the entire range of the test, minus the top two or three, and bottom handfull of wheels.

The data from zipp on the silca blog adresses tire pressure, tire width, and aero losses. But the rule of 105 may only apply to u shaped wheels. The zipp data on the silca blog was for 404 firecrest wheels. Other shapes may have different tolerances to wider tires. That to me is the most interesting question from hambini’s data. If this is the case, what design feature makes it so? Is it the tire/brake track interface?

Basically, narrower tires, even on a wider rim is faster. 23c beats 25. Silca/zipp data showed that with the 404, Different pressures of 23c tires (different widths) had no significant effect on aero... because they always remained in the acceptable width for the rim (narrower). 25c, when inflated, could grow beyond a critical width (rule of 105), at which point aero losses became more substantial. It becomes harder for air to reattach. Deeper rims increase the chance of reattachment, as do wider rims. As such, one can imagine that effective yaw angle will also play into the picture, as wider tires will cause the air to detatch at a lower angle of yaw.

Tires that measure as wide, or narrower than the brake track will be faster (look at the new superwide knott rima on the system six with the narrow 23c tires). If you must, go wider, but don’t exceed the rule of 105 (for the total rim width), or aero losses will start to stack up more significantly.

Ignoring a 2.5% margin of error, lets assume that some rims handle a wider tire better. Iagree that at 119psi, 23c gp4000s2’s measure closer to 26mm. But be mindful, at my riding pressure of 70psi, my 25c gp4000s2’s measure at 26mm!!!

Who runs tires at 120psi these days? More silca data shows us the errors in our ways (impedance losses). 90psi is probably closer to realistic. At that psi, the 23c tire will be closer to 25mm wide. Only 1mm, but for rims edging close to the rule of 105... if could be more than 2.5% difference. If memory serves, josh of zipp/silca estimated a 1-9w loss.

Now throw in rim internal widths effect on measured tire width.

So, ignoring the 2.5% error of his tests (again most of his wheels have overlapping error bars), what he is testing is the aero performance of wheels with tires that likely measure between 25-26mm, depending on the rim. One could argue, he is testing how each rim copes with this tire width.

To help us analyze the data, i would love to see measured tire width, brake track width, and max rim width for each wheel.

To know what rim is fastest, i think we need to use an optimal tire width for each rim. Why do any less? That being said, it would be cool to know what rims can tolerate a wider tire.

by Weenie


RocketRacing
Posts: 878
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

Mep wrote:
Fri Jan 11, 2019 9:32 pm
Has anyone seen research indicating that the aero effect from tires outweigh their rolling resistance?

We all know about the 105% rule and that's there's an aerodynamic penalty from running too-wide tires. But dissecting the problem one level deeper, is the tire's tread pattern and how it interfaces with the rim actually important?
Aerocoach has a recient blog post on the gp5000’s where they estimate crr, and aero losses, and then factor the two together for a final total score. The compare gt4000, gt5000, and the conti tt’s. The tt’s are great for crr, but suck for aero, presumably due to the lack of tread pattern to disturb (energise) the surface layer of air (helping to maintain attachment to the rim).

Obviously the cross over from where crr is trumped by aero losses varies, but if you ride over 30km/hr... focus on aero, as aero resistance is exponential as speed goes up.

For your second question, swissside found similar data to aerocoach, and they found the gp4000 to be the most aero tire (ignoring fitting the right width to your rim). Continental claimed it was not intentional.

I also think the interface between tire and brake track is key. I think hambinis post with diagrams covers that, as does the proof of concept with the mavic strips to fill this gap. Silicon or tape can also serve a similar purpose.
Last edited by RocketRacing on Fri Jan 18, 2019 4:32 am, edited 1 time in total.

RocketRacing
Posts: 878
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

Hexsense wrote:
Sat Jan 12, 2019 6:10 am
I agree that it is just an observation from some Zipp/HED/Bontrager wheelsets. They were traditional U-shape. Widest far below brake track. Somehow, their brake track are roughly 5-7% narrower than max width of the rim too. Coincident or not, that actually imply same brake track width to tire width. Who knows if they make observation on a rim that is widest at the brake track itself, would there be a 100% rule rather than a 105? Or different rim shape doesn't follow the same aerodynamic implication at all? Who knows.
Back to the zipp data on the silca blog. That 105 rule i believe was for the u shaped 404. But i think it makes sense for most if not all designs... just realize that some shapes may better tolerate different widths of tires to different aero losses.

Big width differences (too wide tire for wheel) will not be good, and narrower tires will always be faster on the same wheel. The rule of 105 is a rough guide. The real devil is in the details... and thise details are hard to measure. But we can visualize the ideal tire shape for a rim. Keep it narrow.

User avatar
C36
Posts: 644
Joined: Fri Mar 03, 2017 3:24 am

by C36

RocketRacing wrote:
I think that was my comment. Note the error of his test is 2.5%. That brings a mid scoring wheel of 593w to 578w to 608w. That covers the entire range of the test, minus the top two or three, and bottom handfull of wheels.
.
You are mixing accuracy and repeatability.
Here the “absolute real value” could be within 2.5% of the measured value. It doesn’t mean that 2 exact same wheels could be 5% appart.
Somewhere on the original post the repeatability was mentioned a lot better



Envoyé de mon iPhone en utilisant Tapatalk

RocketRacing
Posts: 878
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

C36 wrote:
Fri Jan 18, 2019 5:17 am
RocketRacing wrote:
I think that was my comment. Note the error of his test is 2.5%. That brings a mid scoring wheel of 593w to 578w to 608w. That covers the entire range of the test, minus the top two or three, and bottom handfull of wheels.
.
You are mixing accuracy and repeatability.
Here the “absolute real value” could be within 2.5% of the measured value. It doesn’t mean that 2 exact same wheels could be 5% appart.
Somewhere on the original post the repeatability was mentioned a lot better

Envoyé de mon iPhone en utilisant Tapatalk
I am not mixing them up at all. Error was stated to be 2.5%. The same wheel could be tested twice, and could be 5% apart. But remember, the 2.5% was an estimate. But i may have neglected semantics, as i forget the term he used in his blog. Repeatability... or consistency, or the lack of, is one of the major critiques with this data set.

The rider is the main source of drag. 1mm in a wheel/tire can have a big effect. How can a live rider be expected to remain in a consistent position for all of the hours of testing for all these wheels (with data collected over a series of months). Positioned by lasers yes... but same clothing? Same body weight? Fatiguing of a rider over the course of an entire day of testing... can cause small position shifts, which may have far more effect on data than subtle differences in wheelset. Manaquins have their own issues, but consistency of position due to a lack of fatigue is one advantage.

Throw in the tire widths into the equation... and there are a lot of confounders in the data.

A test of repeatability would be to retest a wheel or three again, blinded. Heck, test the same wheel on three different weeks. See where they land.

My takehome from his data was what hambini originally stated: it is less about profile, and more about depth. That is a general trend we see when we unfocus our eyes and take a step back.

Mep
Posts: 498
Joined: Fri May 28, 2004 4:11 pm

by Mep

You're probably mixing them up, or at least definitely not using the right semantics because repeatability was described to be high. Ultimately depth is the primary determinant for a wheel's aerodynamics, which would certainly make sense. That said, this isn't the thread for Hambini's wheel test. Back on track..
RocketRacing wrote:
Mep wrote:
Fri Jan 11, 2019 9:32 pm
Has anyone seen research indicating that the aero effect from tires outweigh their rolling resistance?

We all know about the 105% rule and that's there's an aerodynamic penalty from running too-wide tires. But dissecting the problem one level deeper, is the tire's tread pattern and how it interfaces with the rim actually important?


Obviously the cross over from where crr is trumped by aero losses varies, but if you ride over 30km/hr... focus on aero, as aero resistance is exponential as speed goes up.
Where are you getting 30km/h? The study linked in the first post used wind velocity of 48.3kph (30mph) and using that data suggests aerodynamics is a smaller factor than rolling resistance for a tire.

RocketRacing
Posts: 878
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

How do we know repeatability is high? For reasons i mentoned, the rider will make repeatability low... especially when data is collected months apart.

The 30km/hr number comes from hours of objective data combined with supercomputer analysis.

Or maybe i pulled it out of a hat. Seriously, it depends on many factors. Aero wheels? Wheel widths in question, rear wheel in question (aero matter less, how much less?), how much lower is the crr of the tires were are comparing (cause if we are comparing corsa speeds to oem rubber... no competition), how much aero loss are we looking at between the two tires in question? Did anyone get wind tunnel data on that to quantify the aero gains/losses? How heavy are you? How fast do you ride? Do you really ride that fast all the time? What road surfaces do you ride on? Are they as good as you think they are?

The aerocoach data gives a hint of the aero vs crr loses you get with different sizes, and where a crossover may be. My rough math was 30-35km/hr comparing the same tires in 23c vs 25c.

To simplify, i think 23c/25c vs 25c/25c is the debate for most of us. Lets simplify, and say gp5000, on 404’s, so we are comparing apples to apples. The choice for 23 vs 25 will depend on how your chosen tire measures on the rim. 23c will always be more aero (narrower) at any reasonable pressure. Looking at zipp/silca data, and the aerocoach data, 1 mm too wide up front can cause 1-9w loss or so. 25c has a crr that may better, but at an aero cost that will depend on your chosen pressures, which depend on roads and rider weight. Add to that, ride speed. Too many variables.

I am light enough to get away with 23c, even on kind of crappy roads, but kind of like the idea of extra comfort of 25c. But my rim width is in a very tight inbetween for the optimal for 23c vs 25c gp5000’s. With 26mm at the brake tracks, at a psi of 70, gp5000’s should measure no more than 26mm. Should be perfect, but i know 23c will be a bit faster still. I think i will go 25c...

Are you slow? Get the easy gains with wider tires. Are you fast? Keep it aero. Ideally, do both... but sadly wider is slower for aero.

aeroisnteverything
Posts: 193
Joined: Fri Aug 24, 2018 4:43 pm

by aeroisnteverything

Re Mavic CXR wheels and their edge strips - the concept seems sound, so the question is: why are they not making this more widely available on their other wheel lines - Cosmic Carbon SLs, Ultimates, etc.? And why is no one else making a similar product as an add-on option?

User avatar
C36
Posts: 644
Joined: Fri Mar 03, 2017 3:24 am

by C36

aeroisnteverything wrote:Re Mavic CXR wheels and their edge strips - the concept seems sound, so the question is: why are they not making this more widely available on their other wheel lines - Cosmic Carbon SLs, Ultimates, etc.? And why is no one else making a similar product as an add-on option?
UCI banned


Envoyé de mon iPhone en utilisant Tapatalk

jeanjacques
Posts: 193
Joined: Thu Apr 14, 2016 11:01 am
Location: France

by jeanjacques

Rocketracing, you talk a lot but I don't get your input here.

aeroisnteverything
Posts: 193
Joined: Fri Aug 24, 2018 4:43 pm

by aeroisnteverything

C36 wrote:
Fri Jan 18, 2019 10:44 pm
UCI banned


Envoyé de mon iPhone en utilisant Tapatalk
Yeah, but there are tons of products that are not UCI legal. There is a whole sport built around bikes that are not UCI legal. I am not saying make this default option, but sell as an add on - why not?

RocketRacing
Posts: 878
Joined: Thu May 10, 2018 2:43 am

by RocketRacing

jeanjacques wrote:
Fri Jan 18, 2019 11:00 pm
Rocketracing, you talk a lot but I don't get your input here.
There is a lot to it.

Post Reply
  • Similar Topics
    Replies
    Views
    Last post