aerodynamics

Claud Butter said:

How can they be improved? A standard racer has drop handlbars, light weight metals, ethiopian tyres and sod all else (never see these bikes with disks, suspensions etc). Is it possible to get bikes with bicycle fairings? (they have these on sports motorbikes for reducing air resistance).

If I were to take my drink holder and bicycle pump off the frame of my bike, would that increase my speed by a mph or 2? What about wearing aerodynamic helmets?

Areodynamics are the one area where overall the biggest improvements can be made for a cyclist (apart from training obviously!). However, certain compromises need to be made. A cyclist needs good freedom of movment to take drinks and food whilst riding and to trim components on the fly. Fairings make this difficult to acheive and make it hard to ride in a pack with stuff sticking out all over the place.

Aerodynamic helmets are used widely in Time trials but not in road racing for the same reason (they are too cumbersome). There have been trials and modelling conducted into this area. Hence aerobars, different clothing materials, boot covers, aero wheels, tear drop profile tubes and seatposts have become commonplace.
However one thing you should know is that one drink bottle on the seat tube is faster than on the down tube or no bottles at all.

Check Analytical cycling www.analyticalcycling.com

http://www.analyticcycling.com/

Aero seat posts only make you bike faster when you're not on it. The pressure loss from having that much surface area parallel to your thighs outweighs the gain from the airfoil shape.

artmichalek said:

Aero seat posts only make you bike faster when you're not on it. The pressure loss from having that much surface area parallel to your thighs outweighs the gain from the airfoil shape.

Got any evidance to back that assertion?

I did aerodynamic design of jet engines for 3 years, and your explanation sounds extremly, well, wrong.

They don't have it up on their web site any more, but two seasons ago, the Saeco team ran a series of wind tunnel tests with their tt bikes. With everything else constant, the round seat posts always produced less drag than the aero ones.

artmichalek said:

They don't have it up on their web site any more, but two seasons ago, the Saeco team ran a series of wind tunnel tests with their tt bikes. With everything else constant, the round seat posts always produced less drag than the aero ones.

It is really hard to do good wind tunnel testing... Particularly to acribe the cause of the losses to a feature of the item tested. That almost always requires a computer model that has been validated against the results in question.

I suspect these results are not exactly diffinative, because if they were, certainly all of the other teams would have gone back to round seat posts since the tests results were made public. This is not the case, at least for phonak and postal, they both have oversized airfoil seat posts on their TT bikes. My guess is that other teams did their own testing and found different results, if this is the case, all results are suspect...

artmichalek said:

Aero seat posts only make you bike faster when you're not on it. The pressure loss from having that much surface area parallel to your thighs outweighs the gain from the airfoil shape.

I have often thought about that. Your thighs constantly move back and forth beside the seatpost and (to an extent) the downtube. The almost certainly disturb the airflow over the frame. A jet engine has a stationary exterior so aerodynamic design of that would be much easier.
Yes I would like to know if bike manufacturers actually simulate a rider riding on their frame to develop tube profiles. It seems ludicrous to me that the profiles should be as simple in shape as they are.

tamman2000 said:

It is really hard to do good wind tunnel testing... Particularly to acribe the cause of the losses to a feature of the item tested. That almost always requires a computer model that has been validated against the results in question.

I suspect these results are not exactly diffinative, because if they were, certainly all of the other teams would have gone back to round seat posts since the tests results were made public. This is not the case, at least for phonak and postal, they both have oversized airfoil seat posts on their TT bikes. My guess is that other teams did their own testing and found different results, if this is the case, all results are suspect...

A trial and error method using wind tunnel tests is probably the best method of checking the performance of an aerodynamic bike. With this method it would be easier to find which components or changes produced the best drag results. A computer simulatiuon using something like Fluent would be very cumbersome and more inaccurate as I dont think you can simulate accurately the many different types of body shape, clothing materials (and wrinkling of the clothing). Obviously the frame is the easy part.
Don't forget for accurate design you need a rider (whose legs are moving) and rotating wheels and cranks.

tafi said:

I have often thought about that. Your thighs constantly move back and forth beside the seatpost and (to an extent) the downtube. The almost certainly disturb the airflow over the frame. A jet engine has a stationary exterior so aerodynamic design of that would be much easier.

You have completly (and understandably) misunderstood the job of a jet aerodynamicist. A jet engine is a machine designed to move huge amounts of air, all of the aerodynamics of any consequence is on the inside, where there are fans, compressors, combustors and turbines that all have to interact together through their aerodynamics, and on top of that each on is aerodynamicly complicated by itself. If you ever get the chance to look at the main gas path (the heart of the engine) of a jet, you will see what looks like hundreds of small wings that all have to work together, that is what I worked on.

This is not a hard and fast rule (there aren't many of them in aero), but if the streamwise (in the direction of flow) dimension of the smaller of 2 interacting bodies is less than the distance between them their interactive effects will be negligable compaired to their primary aerodynamics.

In other words, unless your thigh is within 1 diameter (or chord if you have an aero seat post) of your seat post, your thigh ain't doin' nothin' to it. I don't know about you, but my legs are much farther apart than that...

tafi said:

Yes I would like to know if bike manufacturers actually simulate a rider riding on their frame to develop tube profiles. It seems ludicrous to me that the profiles should be as simple in shape as they are.

They do test with riders on, the problem is that all riders are different and they need to come up with something that will work for everyone.

tafi said:

A trial and error method using wind tunnel tests is probably the best method of checking the performance of an aerodynamic bike. With this method it would be easier to find which components or changes produced the best drag results. A computer simulatiuon using something like Fluent would be very cumbersome and more inaccurate as I dont think you can simulate accurately the many different types of body shape, clothing materials (and wrinkling of the clothing). Obviously the frame is the easy part.
Don't forget for accurate design you need a rider (whose legs are moving) and rotating wheels and cranks.

Trial and error in a wind tunnel is the best way to come up with the best combo, but it is awful for determining the cause of losses. A computational fluid dynamics (CFD) code like Fluent lets you see more information at more places (this was my specialty, although we wrote our own CFD codes instead of using a commercial code like fluent), this is nessisary for understanding the loss mechanism in most cases. They are two separate issues (best set up, and why are there losses), that was what I was trying to say ealier, and there are different tools that are better for each purpose.

btw, how is it that someone who knows about fluent, could not know what is entailed in jet engine aero? Fluent is a pretty specialized piece of software...

tamman2000 said:

You have completly (and understandably) misunderstood the job of a jet aerodynamicist. A jet engine is a machine designed to move huge amounts of air, all of the aerodynamics of any consequence is on the inside, where there are fans, compressors, combustors and turbines that all have to interact together through their aerodynamics, and on top of that each on is aerodynamicly complicated by itself. If you ever get the chance to look at the main gas path (the heart of the engine) of a jet, you will see what looks like hundreds of small wings that all have to work together, that is what I worked on.

This is not a hard and fast rule (there aren't many of them in aero), but if the streamwise (in the direction of flow) dimension of the smaller of 2 interacting bodies is less than the distance between them their interactive effects will be negligable compaired to their primary aerodynamics.

In other words, unless your thigh is within 1 diameter (or chord if you have an aero seat post) of your seat post, your thigh ain't doin' nothin' to it. I don't know about you, but my legs are much farther apart than that...

They do test with riders on, the problem is that all riders are different and they need to come up with something that will work for everyone.

Trial and error in a wind tunnel is the best way to come up with the best combo, but it is awful for determining the cause of losses. A computational fluid dynamics (CFD) code like Fluent lets you see more information at more places (this was my specialty, although we wrote our own CFD codes instead of using a commercial code like fluent), this is nessisary for understanding the loss mechanism in most cases. They are two separate issues (best set up, and why are there losses), that was what I was trying to say ealier, and there are different tools that are better for each purpose.

btw, how is it that someone who knows about fluent, could not know what is entailed in jet engine aero? Fluent is a pretty specialized piece of software...

I do know a fair bit about the internals of a jet engine but since we don't often deal with complexities such as compressible flow in riding a bike I had assumed you may have been talking about cowlings which are stationary. My apologies for this misunderstanding.

Does the rule of thumb about interacting bodies work as well if one of the bodies (ie: the thigh) is moving back and forth?
This point is interesting when looing at the current trend of oversized downtubes though (particularly around the bottom bracket area).

I agree that when dealing with a systme like this which is essentially chaotic, small changes to one aspect of a frame can have large implications for the rest of the aerodynamics. It would take a very long time for someone to come up with anything aproaching an optimal frame design with respect to aerodynamics.

tafi said:

Does the rule of thumb about interacting bodies work as well if one of the bodies (ie: the thigh) is moving back and forth?
This point is interesting when looing at the current trend of oversized downtubes though (particularly around the bottom bracket area).

Yes, provided that the velocity of the stream is large compaired to the velocity of the moving body. In the case of a thigh on a bike, your leg is moving a hell of a lot less than ~15kph (relative to the bike), and the aero losses on a bike don't pick up until you are going greater than ~20 kph, and of coarse the faster you are going, the more certain the rule becomes...

tamman2000 said:

It is really hard to do good wind tunnel testing... Particularly to acribe the cause of the losses to a feature of the item tested. That almost always requires a computer model that has been validated against the results in question.

I suspect these results are not exactly diffinative, because if they were, certainly all of the other teams would have gone back to round seat posts since the tests results were made public. This is not the case, at least for phonak and postal, they both have oversized airfoil seat posts on their TT bikes. My guess is that other teams did their own testing and found different results, if this is the case, all results are suspect...

Even if you've done wind tunnel testing you still need to use your common sense. Look at the seat post relative to your thighs when riding. Yes my thighs are within one diameter of the seat post - but for that matter, the seat post is easily within one diameter of your thigh (which is moving at the same speed). And do you really think there is clean air flow over it when you look at what it has to pass before it gets there? If there are gains to be made from aero seat posts I would suggest they are in the minds of the riders who use them. Real gains are made from exposed, fast moving areas of the bike and rider - helmets, aero wheels and aero bars.

If a cycling team has gone to the trouble of wind tunnel tests and you have nothing else to go on - its hard to argue.

tamman2000 said:

Got any evidance to back that assertion?

I did aerodynamic design of jet engines for 3 years, and your explanation sounds extremly, well, wrong.

lol....

+1 rep for you!

mitosis said:

Even if you've done wind tunnel testing you still need to use your common sense. Look at the seat post relative to your thighs when riding. Yes my thighs are within one diameter of the seat post - but for that matter, the seat post is easily within one diameter of your thigh (which is moving at the same speed). And do you really think there is clean air flow over it when you look at what it has to pass before it gets there?

The flow at the seat post will never be clean, but this is mostly because of the head tube, stem, and bars. And about the seat post being within the diameter of your thigh, that is pretty insignifigant. The variation in the flow in the direction normal to your thigh is small enough that you can treat it as constant over the width of your post, unless your are within the characteristic dimension of your post.

mitosis said:

If there are gains to be made from aero seat posts I would suggest they are in the minds of the riders who use them. Real gains are made from exposed, fast moving areas of the bike and rider - helmets, aero wheels and aero bars.

You will get no argument from me there, but if all of the teams have made these real gains, the small gains become matters of competition. What I am really suprised I haven't seen (maybe they would be illegal) is a back side fairing mounted to the seat. There is a huge separation behind the rider, if you could keep that flow attached, there would be big gains.

mitosis said:

If a cycling team has gone to the trouble of wind tunnel tests and you have nothing else to go on - its hard to argue.

1. All of the major teams do wind tunnel work. Postal and Phonak both use oversized airfoil seat posts on their TT bikes. If they are aware of the Saeco results, and still use what was indicated to be a suboptimal solution, they must have evidance to the contrary.

2. Blunt trailing edges are high loss features, regardless of upstream conditions. Now the scale of the post compaired to the scale of the bike leads me to believe that any differences measured from a change of post would be within the error of the experiment, and they quite possibly could have measured an improvement with the cylinder, but it would be meaningless. And without access to their data, or at least a detailed report on it, I can't make that assesment.

Just becasue a wind tunnel tells you something, doesn't mean you can take it to the bank. Do you have any idea how many times I have seen stuff that was "proven in a lab" fail? Experiments aren't perfect, and when you see extremely counterintuitive results, it calls for further investigation. You can be sure that other teams have done this test since Saeco did, and the evidance suggest they found different results.

If I had ever changed a design based on one experiment, with no confirming analysis or independant experiment, the FAA would have had my balls for lunch.

Do you really want to argue aerodynamics with a man who helped make the F-22 go mach (classified)?

Maybe what you are looking to find is information along the lines of:

timetrial.orgaerodynamics.htmOpen ↗

the reasons why being

sciencemuseum.org.ukaerodi.aspOpen ↗

all the information is gathered from wind tunnel testing.