Bike Wheels and Ocean Creatures

On to part 2!!

Ground Boundary Layers

If I did not make it clear in the past post, I really enjoy new innovations in bicycling. However, there is a difference between some new kick-starter trinket that solves a problem in an innovative way, and a $4,000 wheel-set without any published data yet besides “It looks like humpback whales”.

Anyway, back to birds and whales.

The other important lesson to take from the Albatross doesn’t have to do with it’s wings, but HOW it actually fly’s 1000’s of miles without flapping:

Once over the open ocean, the albatross will perform a circular swooping flight to cheat the wind. Flying 30 or so feat above the ocean the albatross will dive down with the wind to the ocean surface gaining velocity from gravity and the wind. Once at the surface it’ll turn around right at the surface and head into the wind. Once angled towards the wind the bird then pulls up, using it’s high speed and incoming wind to climb back to it’s initial flight level. It’ll then repeat this process thousands of times to cross an ocean.  (or just watch the video below)

Albatross flight

The albatross takes advantage of the surface boundary layer to generate “free speed”. Wind speed is always lower the closer you are to the ground and this gradient is a VERY well known phenomenon. This is important because a large wind gradient close to the ground undermines the whole idea of minimizing crosswind drag. If you want a more exhaustive discussion of Atmospheric Boundary Layer check out This Post. In short , and in my humble opinion, the magical 15 degrees of cross wind that wheel and bike industry has adopted as the targeted drag angle to minimize drag for is WRONG!!!!!!!!!!!!!!!!!!!WindAngleIsTooDamnHigh

This high crosswind angle is Zipp’s ENTIRE marketing shtick. It’s fine to try to reduce drag at high angles of attack but every design choice you make when making a wheel, bike, plane, whatever represents a trade-off. Usually, high angle of attack drag reduction results in higher low angle drag, especially if you introduce drag creating turbulators, which is exactly what Zipp has done.

Now I will add, that between the testing that we did when designing our wheels we noticed that for near zero angles of attack, drag results were more insensitive to actual rim shape than at the magical 15-degree: 60mmcfd

These are results from a variety of the “best-case” 60mm rim shapes from our CFD runs along with some industry rim shapes (cough, zipp, cough). If you look closely the rims that perform best at high angles of cross wind tend to perform worse at near zero angles of attack (but results are all closer at the Zero point). I believe that in order to get a true 15-degree wind angle on your wheel you have to have near hurricane force winds and that most people would rather have a wheel that performs slightly better during the remaining 98% non-hurricane-force wind conditions. The cynic in me even would go so far as to say most bike industry has focused on 15-degrees simply because it’s the low point on the drag curve, the magical point where 90mm wheels get to the ‘negative drag’ point.

Scaling the Flow

When I posted the last post a friend brought up the important point of flow similarity between objects at different sizes and speeds. In Aerospace design its usually pretty difficult to fit a full sized aircraft or rocket into your wind tunnel. Fortunately flow is somewhat scale-able. Generally speaking if as long as Reynolds Number and Mach number of a flow is the same between say a full sized aircraft and a model in a wind tunnel.

So lets take a look at the various Re and Mach numbers for the 787, Albatross, Humpack Whale, and a Zipp 454:reynoldsnumberwheels
 What this tells us is that none of these are REALLY like any other. However it does tell us that most of the flows above are generally in the turbulent region (what this is is even highly dependent on the individual situation). If anything the bike wheelset is the one closest to the laminar region of flow.

Zipp modeled these fins off the ‘tubercles’ of a humpback whales and there’s some decent research on the topic. However, the question of the humpback whale brings up a few interesting questions and points. The paper does indeed show that flow separation on the flipper is delayed due to the tubercles. However, the tubercles are only found on the high Aspect Ratio (long) flippers of the Humpback. Faster whales like the Orca (ok litterally just learned Orca’s actually a dolphin) and dolphins have smooth contoured fins, additionally no birds that I’m familiar with have tubercles (although they do have more intricate flow control). So why do humpbacks have these features while other animals do not?figure-3-humpback-whale-megaptera-novaeangliae-flipper-leading-edge-tubercles-left

The other important question to ask about the tubercles is how it’s applied to the Zipp wheels. On the whales the tubercles are on the leading edge of the fin. With a wheelset, only a quarter of the rim is showing leading edge tubercles to the wind at any given cross-wind moment. The rest of the time, the flow that’s being turbulated by the Zipp fins is flowing off into the free-stream and not working to help flow remain attached. Also…um importantly…the humpback fins don’t rotate.

capture

So the tubercles have to provide a large benefit in that small area to offset the detriment they cause around the rest of wheel.

Most good engineering innovations have sort of an ‘ah-ha’ sense to them where the design just sort of makes sense before you even look at any data. For instance, once the Wright brothers first solved the problem of powered flight there were dozens of  other non-wright airplanes in the sky within a few years because they quickly recognized that control/balance was one of the primary problems that the wrights solved and had a bit of an ‘ah-ha’ moment.

I personally don’t get that “Oh that makes a lot of sense” feeling with the Zipp tubercles. I don’t think there’s some special magical design that’s going to cheat wind. Of course I might be very wrong since I don’t know the actual numbers, and I’m just some stupid blogger on the internet…..but I do know you don’t see airplanes that look like this:Stupid Looking Airplane

StumbleUponRedditShare

2 Comments

  1. Good question…. These tubercles are not on the humpbacks tails either, which i think they used more for their propulsion. The two fins, i think it’s just for stability and looked to fold back during fast dives.

  2. Chris – It’s always good to follow your engineering and cycling exploits. Regarding the humpback whale, there is a very good reason for tubercles. When they eat, they encircle schools of fish by swimming around them in circles with an ever decreasing radius of air bubbles that prevents the fish from escaping. They then rise up through the encircled pod of fish with their mouths open. Lunchtime!

    They don’t use their fins for propulsion. Rather, it allows them to bank sharply. Why? The tubercles delay separation due to stream wise vorticity that is induced as the flow passes over the tubercles, transporting high momentum fluid in the farfield down into the boundary layer. Therefore separation is delayed, and the whale can turn a sharper corner because the fin can reach a higher angle of attack before separation occurs. This makes them better hunters because they can corral fish better/eat better.

    There is a wealth of research on how tubercles delay stall, particularly in plunging/pitching airfoils. Interestingly, about 3 years ago I sat on a Ph.D. defense on just this topic and thought “that might work on a bike wheel as the trailing edge becomes the leading edge.” I convinced an undergrad honors student to work on it with me, and we ran some simple CFD simulations and it looked promising. A couple months after he graduated, and we were mulling next steps, Zipp announced their new wheel. We were beat to the proverbial punch.

    However, as you astutely noticed, and we saw this exact behavior in our studies, while the wheel performs better at higher yaw angles (where you spend little time riding) you pay a penalty at lower yaw angles. So the question is, “is it worth it?”. My opinion is, probably not. The only other interesting thing to consider, though, is that the tubercled wheels also produce much less severe stall characteristics (stall develops very slowly with increasing angle of attack, instead of very quickly as with most airfoil shapes). Consequently, there could be some handling benefits where wheel handling characteristics in a crosswind may also not change as suddenly (making it easier to control), but this would of course be a secondary consideration relative to the drag considerations.

Leave a Reply

Your email address will not be published.