Boundary layers and drag

I will try not to get too complicated, (as I will probably lose myself) but I was wondering:

I think most people know that with bike aerodynamics achieving laminar air flow over an object is the ideal, followed next by turbulent flow which has slightly higher drag and finally separated flow incurring by far the worst drag figures, awful aerodynamically.

Many companies have however caught onto the idea that laminar flow is very hard to achieve with separated flow often occurring. They have realised that if you put a turbulator (a dimple for example) on the surface of an object, for example golf balls, Zipp discs, Lazer helmets and the Nike Swift Spin body suits, you can artificially turbulate airflow, which although slightly increases drag it greatly helps to prevent separation and thus decreases overall drag. Anybody disagree so far?

(I know there are arguments about bike frames not being bluff bodies, but in conjunction with the rider, the components and the associated interference drag can laminar flow realistically be achieved? I would suggest that the bike and rider be treated as one bluff body?)

My question to the forum is; If bike frames and components had turbulators would they not also reduce overall drag? It needen’t be a totally dimpled bike something more simple like putting sand (it doesn’t have to be as low tech as sand!), in the final lacquer coat would certainly turbulate the boundary layer

Your thoughts please

AndyA

The phenomenon you are referring to is caused by a lower pressure (vacuum) pocket behind the object after shedding the streamlines. Dimpling causes the streamlines to cling to the object longer thereby reducing the area of low pressure (vacuum) that is pulling the object backward.

The book Boundry Layer Theory by Schlichting explains the theory.

One can argue this theory all day long, but in my opinion the only way to get a definitive answer is to put said items in the windtunnel and obtain empirical data. I understand that the esteemed Mr. Cobb has indeed done this and that dimpled helmets and wheels showed no benefit. I’m not sure why, but I would guess it is related to speed of the object and overall shape of the object. Most triathletes don’t look like golf balls.

Agreed, but why then have Nike spent vast amounts of money on their Swift Spin Body Suits if they don’t work, and again, if they do why not apply the same science to a bike? Quote from Nike “The effect of the differently textured fabrics on the body is similar to the one that dimples have on a golf ball during flight”
I would ask Dave Cobb myself but for the life of me I can’t poston his forum

The full Nikes press release can be read below:

        ![http://www.nike.com/nikebiz/img/s.gif](http://www.nike.com/nikebiz/img/s.gif)    ![http://www.nike.com/nikebiz/img/s.gif](http://www.nike.com/nikebiz/img/s.gif)  ![http://www.nike.com/nikebiz/img/s.gif](http://www.nike.com/nikebiz/img/s.gif)      

Nike Design Exclusive, Highly Aerodynamic “Swift Spin” Body Suit For Lance Armstrong And His 2002 Tour De France U.S. Postal Service Team Speed Suit Represents Cycling’s Most Innovative Competition Apparel Ever; Debuts At The 2002 Tour de France Time Trials

BEAVERTON, ORE (8 July, 2002) - After more than four years of research, testing and design, Project Swift, Nike’s elite Apparel Innovation Team, has unveiled the Swift Spin. The highly aerodynamic time-trial cycling skin is making it’s exclusive debut with the United States Postal Service Team at the 2002 Tour de France. The Swift Spin follows in the wake of Project Swift’s international success with track and field’s “Swift Suit,” debuted in 2000 at Sydney, and speedskating’s “Swift Skin,” showcased in 2002 at Salt Lake City. The suit, one of many used by the USPS riders during the Tour, is designed for the time trials (stages in which the cyclist is racing the clock, not other riders).

With this latest innovation, the aerodynamic speed suits created under the Project Swift banner now stand as the most innovative pieces of competitive sports apparel to date. The Swift Spin was designed in U.S. and Canadian wind tunnels to deliver a unique performance benefit to three-time Tour de France Champion Lance Armstrong and his United State Postal Service Team.

Project Swift is led by two members of Nike Apparel’s Advanced Innovation Team (AIT): Innovation Director Rick MacDonald, based in Oregon, and Senior Designer Eddy Harber, based in London. Through their work in creating aerodynamic suits for speedskating gold medal winners like Derek Parra of the U.S. and Jochem Uytdehaage of the Netherlands, along with Australian 400 meter gold medalist Cathy Freeman and America’s five-time track and field medal-winner Marion Jones, the Nike designers were able to bring extensive knowledge and experience from other highly wind-sensitive sports to the design of the Swift Spin. “We’ve taken everything that we learned from the Sydney Swift Suit and the Salt Lake Swift Skin,” says MacDonald, “and applied it to cycling. Each time we deal with a new sport, we have to relearn that sport. Our team looked at cycling in a way that no one ever had. Then we combined that unique insight with the technology we already had in-hand from track and field, and speedskating. With cycling, the body is in a different position, the speeds are different, different parts of the body are moving at different speeds. The environment can be totally different, as well. Nike took all of those learnings and reconfigured the suit to provide a unique benefit to the USPS riders. We worked on the Swift Spin for a long time with a great deal of input from world-reknowned cyclists and internationally lauded aerodynamicists. There is nothing out there on the Tour like it.”

The Swift Spin most closely resembles speedskating’s Swift Skin. Though the body position and speeds of the two sports are similar, says MacDonald, the airflow over a cyclist is different. “Our adaptations of the Swift Spin are based on the airflow over the cyclists we positioned in the wind tunnels,” he says.

The bottom line is that the countless hours of wind testing by Nike, says MacDonald, “show a significant benefit to wearing the Swift Spin, over any other suit we tested - and we tested a lot of other suits.”

In the beginning, Project Swift investigated more than 50 textiles to meet the unique performance attributes intended for the Swift Spin. These fabrics were pulled from around the globe and tested for several key qualities including wind resistance, elasticity, and breathability. Next they employed an advanced form of “body mapping” known as Nike Zoned Aerodynamic Technology (patent pending). Through this process they scientifically placed the six selected fabrics on certain body locations to work strategically and harmoniously with the athlete’s unique motion in relation to airflow. This maximizes the performance output against the negative effect of air friction, as well other physiological and environmental factors.

Of particular concern for cyclists was heat, both of the body and environment. “The conditions are relatively cooler in speedskating. And track sprinters are in their suits for relatively little time when compared to a cyclist. We used lighter-weight fabrics on the Swift Spin in most places, although the suits do share a lot of similar textiles. We had to rearrange those fabrics in a new pattern due to thermoregulation demands on the bike.”

In the final stages of production the Swift Spin was meticulously crafted. When possible, seams were aligned to correspond with the airflow direction or placed completely out of its way to further reduce drag. Where appropriate, the Swift Skin was articulated to minimize creasing, which could “trap” air and slow a cyclist. Simply stated, this full body suit reduces drag like no other. The effect of the differently textured fabrics on the body is similar to the one that dimples have on a golf ball during flight.

From a distance, the suit resembles a conventional cycling time-trial suit. And while it’s available in both short- and long-sleeved versions, it’s actually quite different. The leg extensions are actually a bit longer than conventional suits. “That extension is longer and made of a different fabric than other parts of the suit for specific aerodynamic reasons,” says Harber. “The difference between this and a conventional suit really lies in the aerodynamic characteristics of the fabrics and where they’re placed.” Also critical to the Swift Spin are the direction and placement of seams in the suit. Seams in all Swift products are placed to move in harmony with the airflow and not against it."

The fit of suit has also been trimmed down. “We changed the fit to be more streamlined and close to the body when the rider is in the time-trial, aerodynamic position,” says MacDonald. “This suit is tight. Any kind of streamlining that you do on a suit like this reduces drag. It not a comfortable suit for walking around or sitting up on the bike, but when you’re in that special aerodynamic position, it fits like a glove.”

With the Swift Spin, Nike also has delivered unique aerodynamic shoe covers and cycling gloves, similar to the ones designed for the Salt Lake speedskaters. Armstrong’s long-sleeved Swift Skin suit has gloves incorporated into the design. Both covers and gloves are made of the silver aerodynamic textile previously employed in both the track-and-field and speedskating suits. “It’s just as effective on the bike,” MacDonald says. "It’s a very fast fabric for the small, fast-moving parts like the hands and feet.

Thanks AndyA

“why then have Nike spent vast amounts of money on their Swift Spin Body Suits if they don’t work,”

To sum it up in a single word - marketing.

What causes the streamlines to “cling to the object longer” is the fact that because of the dimpling the laminar boundary layer transition to turbulent occurs sooner. Because turbulent boundary layer separation occurs later than laminar separation, the separated region is smaller, thus reducing pressure-related drag. Skin-friction drag will rise a bit, but because pressure-related drag is responsible for the most part of the total drag, an effective drag reduction is achieved.

Dimples are appropriate for the Reynolds numbers typical of the flight of a golf ball, but are not appropriate for the flows around bikes and/or bike components. Any experienced engineer with an aerodynamics background will tell you that, you don’t have to go test it on a wind tunnel to know it. So all those helmets and disc wheels with dimples on them are just products that play with the buyers imagination.

The same can be said for Nike’s super-fast fabric. The efect of reducing the skin-friction contributes very little to the overal drag. More cleverly, and that has been done by Cobb I think, it’s to have a suit with the seams in the right places, in order to trip the laminar bounday layer in the right locations, thus reducing the separated region and with it the pressure-related drag. Such a suit made of normal lycra can be faster than a seamless suit made with the fastest fabric.

As for the bikes with devices to trip the boundary layer, you would be working on an area that would bring very small gains in drag reduction. But I can see it being a good selling point, just like that ridiculous nose on the headtube of the Litespeed Blade of some years ago :slight_smile:

Paulo

Thanks

Interesting so dimples don’t work but trip strips do (but not much!). Don’t dimples act as trips?

See below for a really good article on the subject.

What do Skinsuits Have To Do With Boundary Layers?
An aerodynamics primer by Kraig Willet

http://www.bike.com/template.asp?date=7%2F16%2F2002&page=2&lsectionnumber=6&lsectionname=Tech+Smart&lsectiondirectory=techno

Cheers but I am still not sure what to think?

AndyA

(Oh and as far as I am aware the suits are not for sale?)

Good article and a lot easier to digest than Schlichting’s book :slight_smile:

First of all, strips that trip the boundary layer are effective only when placed in the right places. In order to know where to place them, you need to know exactly how is the flow around the body. While you can do that with a cyclists’ body, especially his upper body, it’s a lot more difficult with the legs, because they’re moving, and even more difficult with the flow around the bike.

Secondly, the gains in drag reduction on the bike itself would be very small.

That said, if you REALLY want to try trip strips, why don’t you place a couple of them on your head tube, around 45deg? Or streamline the cables? OR ride an extra 15mins on every bike workout? Being a mechanical engineer and a coach I know what works best… ;-)))

Paulo

Um… What I had seen John Cobb post was that dimpled wheels work, helmets don’t.

The reason is that for dimples to work they have to be subject to very fast speeds. (Above fifty-miles and hour.) Zipp put dimples on their disc because it does travel considerably faster than the bike. (Basic physics, an object spinning is traveling faster than an object not.) The dimples not only help the airflow to put it self back together sooner coming off the rear wheel, it also seems to suck air away from the bike frame.

While not completely bland, I don’t believe Zipp is all about marketing hype; if they were they’d still be selling the 3-spoke wheel.

the dimpled Zipp, while intended to be faster than a standard flat disc (Corima, Zipp '2002, etc.) they are faster ON the bike and slower OFF, but minimally I guess Cobb comments]. Unfortunately, the idea was nice, but it did not really make much difference.

The marketing of the wheel has been great however. I believe that it is a good wheel, but the Corima has a very nice hub and is time proven. I have heard of faults with the new Zipp hubs http://topica.com/lists/timetrial/read/message.html?mid=1714321993&sort=d&start=0] but I’m sure they have corrected them by now.

The aero skinsuit is another nice idea, but it has to be custom sized to each rider, so I’ve read.

What I thought was hilarious was when they discussed how much research went into placing the seam at the base of the spine. You then got treated to images of Lance Armstrong in their skinsuit with a big old number 1 slapped right on that sweet spot.

Ken Lehner

$1100 for a dimpled disc to save what? At best probably about 7 seconds on a 40k. You got to be kidding me! I’m going to run more and save 7 sec per mile. Off to run again today.

Thanks guys

I take on board everything you are saying and I am not mechnical engineer but their is some convincing marketing out there for example take a look at the hype about the Alpha Q aero fork

I quote “The revolutionary Truespeed coating … drag is reduced by precisely controlling surface roughness…manipulates the seperation point of the boundary layer… technology banned by the US golf Assoc because it provides an unfair advantage… the fastet UCI-legal fork on the planet”

http://www.truetemper.com/performance_tubing/aero.html/alphaq.html

Can anybody see a disadvantage to it?

Cheers AndyA

“At best probably about 7 seconds on a 40k”

i’d argue zero seconds. the wheels are GREAT, as all zipp wheels are, but the dimples probably don’t do squat. stop and calculate REALLY how much faster any point on the disk is moving versus a point on the frame – relative to the wind. throw a rubber ball down your street. as it bounces down the road, that’s the trajectory taken by any point on your wheel. the farther from the hub, the higher and longer the hops. but the speed difference (esp relative to the wind) between that and your belly button, as you run down the road to fetch your ball, is not that great.

I am in the zero seconds camp, myself. I admit to not being in the know when it comes to all things aerodynamic, but I honestly think that a golf ball travels faster than a bicycle.

That being said, what do I think of the dimpled disk? Zipp makes a good disk, but in the end, I don’t think the dimples did more than increase tooling costs for the Indy-based company.

Just a quick thought here…

ZIPP makes a great disc. I’ve owned 3 and time trial with one currently. But I don’t buy the “dimple” concept for a rotating disc/wheel. A golf ball, yes. A disc, no. A golf ball is traveling very fast and it’s shape is a consistent all the way around. If you ovalized a golf ball (with dimples) it would not travel straight. That tells you right there how wind is effected by shape. But dimples DO work for the golf ball. Take the dimples off a golf ball and you can see for yourself how much slower and unstraight the ball will travel. But this doesn’t transfer over to the cycling world. Think for a second about the wind direction you and your bike are traveling through at speed in a race. How often is it zero degrees yaw (straight ahead)? Hardly ever. That would be best case scenario. So now, real-life situation, wind from all directions doing bad things to the airflow passing around your rig. I just don’t see how the dimples could help in this situation. Maybe in a straight TT with a headwind it MAY help, but I’ve never raced one of those, have you? What about this question as far as dimples on a disc – can the dimples slow you down?

Back to ZIPP. This is a company committed to improvements and refining. Their dimpled disc is a logical marketing step for them. And some people will buy it. And some will be faster on it. But I don’t think it’s due to the dimples.

.02 cents worth.

“the top of the wheel travels twice as fast through the air as the rest of the bicycle”

for an exceedingly small moment in time, the rest of which it’s spent going the same speed as the bike, or somewhat faster, or somewhat slower, and at one point it stops altogether.

so, as i said, it’s not rolling down the road traveling multiples of times the speed of the frame. but your point is taken. at one specific moment it is traveling at a multiple of 2.

It’s also worth noting that the back wheel is encountering turbulent air rather than the still air that the front wheel will see. Don’t the dimples require still, laminar air flow to work? It seems to me that dimpled disk would make more sense on the front than the rear.

Guys…, all i have to say about this topic is that MAYBE dimples do work. However since some argue that it works only above a certain speed (someone mentioned 30 mph), then why don’t you find any dimpled cars…? huh…?

Paul.

My neighbor has a dimpled car. Both sides on the doors and one side on the back quarter panel, the back bumper, and the roof. 16 yrs he has had that car and he collects dimples on cars.