Wheel Measurements

Now that I have the 'results' section up and running, I'd like a bit of help. What wheelsets would you most like to see measured? Eventually I will be measuring everything I can get my hands on, but I'd like to start with those wheels that would make the biggest difference to people. I figure a set of Ksyriums and Zipps is obvious, but... Help?

ScienceIsCool said:

Hi everyone.

I've been working for a while on a technique to measure the performance of a bicycle wheel. It worked out much better than I expected, so I wrote a paper on the subject. The response has been very positive, so I've decided to create a website and release my findings as a resource for cyclists.

I expect that I will have a lot of data for individual wheelsets up within a week. More will be added on a daily and/or weekly basis.

Anyone who is interested is invited to come and check out www.bikephysics.com

ScienceIsCool said:

Now that I have the 'results' section up and running, I'd like a bit of help. What wheelsets would you most like to see measured? Eventually I will be measuring everything I can get my hands on, but I'd like to start with those wheels that would make the biggest difference to people. I figure a set of Ksyriums and Zipps is obvious, but... Help?

I'll be selfish and put my own up for assessment,

DT Swiss RR1450.

Others of interest would be

• American Classic 420
  
• Bontrager Race X Lite Aero
  
• Campy Eurus
  
• Shimano 560
  
• Shimano Dura Ace
  

Great piece of work. Look forward to watching it progress.

How about an overall score. Yes I know that there would be debate about the weightings, but its still usefull. Also, could you note if the wheels are tubular or clincher.

Are you using the same tyres and tubes for the test?

Walrus said:

I'll be selfish and put my own up for assessment,

DT Swiss RR1450.

Others of interest would be

• American Classic 420
  
• Bontrager Race X Lite Aero
  
• Campy Eurus
  
• Shimano 560
  
• Shimano Dura Ace
  

This list is a good starting point.
Could you also compare/contrast/relate to:
http://www.sheldonbrown.com/rinard/wheel/grignon.htm

Hmmm. This is really interesting, though their method of measurement isn't explained at all. Also, all those torques are a function of velocity... Hmmm. I'd love to compare and contrast, but there really isn't any information to go on. I'm willing to discuss and debate the topic though. Thanks for the link! I'll give it some more thought.

John Swanson

daveornee said:

This list is a good starting point.
Could you also compare/contrast/relate to:
sheldonbrown.comgrignon.htmOpen ↗

A ranking system sounds like a good idea. Suggestions anyone? Maybe we could base it on moment of inertia and power at 40 km/hr? Also, I didn't note in my paper (though should have) that all wheels were measured without any tire or tube, but with a rim strip. Without a rim strip, the spoke holes would have caused excessive drag.

John Swanson

mikesbytes said:

Great piece of work. Look forward to watching it progress.

How about an overall score. Yes I know that there would be debate about the weightings, but its still usefull. Also, could you note if the wheels are tubular or clincher.

Are you using the same tyres and tubes for the test?

ScienceIsCool said:

A ranking system sounds like a good idea. Suggestions anyone? Maybe we could base it on moment of inertia and power at 40 km/hr? Also, I didn't note in my paper (though should have) that all wheels were measured without any tire or tube, but with a rim strip. Without a rim strip, the spoke holes would have caused excessive drag.

John Swanson

Would suggest that you take all of the measurements and apply some sort of weighting.

How do you do the aerodynamics with no tyre on?

Thanks for the efforts. ( Of course it makes me depressed when I look at the power differences, compared to what the total power output required to maintain a bicycle at 50kmh on the flats with no wind 🙁 )

One thing that comes to mind is you are likely to get questions in regards to the "Custom" sample that you have listed as to whether the DA22 is 32 or 36 and 2X, 3x, radial, etc. (Of course then everyone will want you to have a comparison of the same wheel build 2x versus 3x.

On the technical side, would the relative variance between the different wheelsets of the "C" component to the power requirement change substantially in a real world case where you were encountering say a 15-20mph headwind versus calm (as I assume is the basis for the testing) or tailwind?

You're right, I definitely will add that the wheel was laced 3x. I've been missing small details like that. Thanks for pointing it out.

Without getting too technical, the value of 'c' should not change when the wheel is in a velocity field with small magnitude (i.e., you're going between 20 and 60 km/hr).

I need to do some modelling and more thinking, but I'm sure that 'c' will change only a bit when encountering a cross-wind of moderate angle, say up to 30 degrees. I will try to verify that in the next while. Hmmm. That was an awesome question. Now I'm going to be thinking about that all day and not get any work done...

John Swanson

supergrill said:

Thanks for the efforts. ( Of course it makes me depressed when I look at the power differences, compared to what the total power output required to maintain a bicycle at 50kmh on the flats with no wind 🙁 )

One thing that comes to mind is you are likely to get questions in regards to the "Custom" sample that you have listed as to whether the DA22 is 32 or 36 and 2X, 3x, radial, etc. (Of course then everyone will want you to have a comparison of the same wheel build 2x versus 3x.

On the technical side, would the relative variance between the different wheelsets of the "C" component to the power requirement change substantially in a real world case where you were encountering say a 15-20mph headwind versus calm (as I assume is the basis for the testing) or tailwind?

I definitely need to add this to the paper. Thanks for pointing it out! The wheels were tested without tires and tubes for a few reasons. First is that it would completely screw up the measurement of moment of inertia. Second is that it's not necessary for rotational aerodynamics. Unlike translational aerodynamics where the tire forms the leading edge of an airfoil (i.e., the rim). Thirs is that a tire could make the rotational aerodynamics better or worse and would confuse the data. The reason is that each tire has some skin friction associated with it. That is, the air will stick to it in a small boundary layer. As the tire whips around the wheel, it carries that air with it, causing additional drag. Of course, not all tires are created equal and some will be better than others.

This is, of course, the topic of a paper I'd like to write in a month or two... 🙂 I believe there are real and substantial differences in the rotational aerodynamic properties of tires.

John Swanson

mikesbytes said:

Would suggest that you take all of the measurements and apply some sort of weighting.

How do you do the aerodynamics with no tyre on?

ScienceIsCool said:

I definitely need to add this to the paper. Thanks for pointing it out! The wheels were tested without tires and tubes for a few reasons. First is that it would completely screw up the measurement of moment of inertia. Second is that it's not necessary for rotational aerodynamics. Unlike translational aerodynamics where the tire forms the leading edge of an airfoil (i.e., the rim). Thirs is that a tire could make the rotational aerodynamics better or worse and would confuse the data. The reason is that each tire has some skin friction associated with it. That is, the air will stick to it in a small boundary layer. As the tire whips around the wheel, it carries that air with it, causing additional drag. Of course, not all tires are created equal and some will be better than others.

This is, of course, the topic of a paper I'd like to write in a month or two... 🙂 I believe there are real and substantial differences in the rotational aerodynamic properties of tires.

John Swanson

In the real world there would be a tyre and horizontal wind. I was thinking of a control tyre, perhaps picking the most popular clincher and the most popular tubular. Having said that, you probably don't have access to a wind tunnel. Whatever you do, the result is going to be really useful to a lot of people.

Actually... A few people are encouraging me to make a wind tunnel and are willing to offer their expertise. I have some ideas, but first I need to get a whole whack of wheel data up onto the site. I expect my first attempt at a design will be late August or early September.

John Swanson

mikesbytes said:

In the real world there would be a tyre and horizontal wind. I was thinking of a control tyre, perhaps picking the most popular clincher and the most popular tubular. Having said that, you probably don't have access to a wind tunnel. Whatever you do, the result is going to be really useful to a lot of people.

ScienceIsCool said:

Actually... A few people are encouraging me to make a wind tunnel and are willing to offer their expertise. I have some ideas, but first I need to get a whole whack of wheel data up onto the site. I expect my first attempt at a design will be late August or early September.

John Swanson

John, your project is fantastic.

Thanks! I appreciate the compliment. BTW, I have a new script to analyze data much more quickly. I've just used it to measure a Neuvation M28 Aero which is also posted to the website. Expect data to be added on a regular basis now. I'm pretty sure the next set will be a set of Mavic Elites. Followed by many, many more.

mikesbytes said:

John, your project is fantastic.

John, I'm looking at your wheel results and could you explain the wattage.

Using front wheels at P40 (I assume that 40kmh) I would require 2.14 watts of energy to maintain the front wheel of the Mavic Ksyrium ES compared with 1.15 watts of energy to maintain the Kult ProMotion, there for that front wheel uses 0.99 less watts to maintain 40kmh ?

Cheers Michael.

Yup, that's exactly right. Surprising, isn't it? It takes a whole Watt less to keep a Kult wheel spun at 40 km/hr. The reason is that the Ksyrium ES spokes are very large even if they are bladed. They are also attached very far on the edges of the hub which exposes them to more air.

What my measurements don't include is the forward or 'translational' aerodynamics, which is a function of rim shape. For that you need a wind tunnel. Coincidentally, I'm going to attempt building a wheel sized wind tunnel some time soon.

John Swanson
www.bikephysics.com

mikesbytes said:

John, I'm looking at your wheel results and could you explain the wattage.

Using front wheels at P40 (I assume that 40kmh) I would require 2.14 watts of energy to maintain the front wheel of the Mavic Ksyrium ES compared with 1.15 watts of energy to maintain the Kult ProMotion, there for that front wheel uses 0.99 less watts to maintain 40kmh ?

Cheers Michael.

I wish I had read thru your results yesterday. I bike has a broken rear wheel and I'm out of $ at the moment, so its going to be something cheap but I'm going thru decision pain, as I really just wanted to replace the rim, but you can't buy the origional rim. There was a WH-540 rear on ebay and I let it go. I'm not so impressed with the current 550/560's on sale at the moment. Velocity is a good choice in Australia as they are Australian made and the rims can be easily purchased for replacement.

I get the impression from your data that at the cheaper end, less spokes, ie better aerodynamics is better than lighter rims, hence the good result for the 540. Is this how you see it?

John,

I am challenging the relavance of the tests. The wheel is static, just rotating! About all these results tell me is which wheel to use on the Compu Trainer. 🙄

When I ride my bike, the bottom of the wheel is static, the top is doing twice the road speed. The major drag is from the air carried around the tyre meeting the air flow above the tyre. Deep V rims and smooth tread tyres use far less watts than a light wheel and a treaded tyre.

Ger the air flowing over the whole wheel at road speed, then publish the results. 🙂

Excellent points! You are correct that the tire/rim profile have a large effect on the overall drag on a wheelset. Added onto that is the fact that the tire likes to act like a viscous pump, dragging air along with it.

And you make an excellent point that the bottom of the wheel is always at a speed of 0 km/hr while the top of the wheel is going exactly twice what your cycling computer says. This means that while riding, the bottom of your wheel receives exactly zero drag due to forward wind resistance... Neat. But at the top it receives about 4 times what the middle does. Weird. To be technicaly correct, while riding, the wheel sits in a non-linear velocity field of air. That even negates wind tunnel data! Wind tunnels hold the wheel stationary and generate a constant velocity field. That is, the speed of the air is constant everywhere.

The best we can do, right now, is make approximations. Really, what we're after is the answer to a specific question. What is the best wheel? That is, what wheel takes the least amount of energy to accelerate and keep at speed.

I thin that the measurements I take can answer most of that question even if the exact values are not correct. Wheel A may take 1.372 W to keep spinning at 40 km/hr and not 0.968 W that I measure. However, I can tell you that wheel A is better than wheel B. Especially if they have similar rim cross sections (i.e., forward aerodynamics is also similar).

I can also tell you about the moment of inertia, which is valuable in figuring out how much power you'll need to sprint out of a corner. That one is an absolute number and not an approximation.

So overall, I think there is some validity to my testing. But I recognize and agree wholeheartedly that it is not perfect. Then again, I've shown that even a multi-million dollar wind tunnel is not perfect... 🙂

Let's collectively put our heads together and figure out how to improve the measurements. Until then, I'll stand by my methods and results. I'm sure you've got a good rebuttal or two, so let fly with both barrels.

John Swanson
www.bikephysics.com

gclark8 said:

John,

I am challenging the relavance of the tests. The wheel is static, just rotating! About all these results tell me is which wheel to use on the Compu Trainer. 🙄

When I ride my bike, the bottom of the wheel is static, the top is doing twice the road speed. The major drag is from the air carried around the tyre meeting the air flow above the tyre. Deep V rims and smooth tread tyres use far less watts than a light wheel and a treaded tyre.

Ger the air flowing over the whole wheel at road speed, then publish the results. 🙂