Hi all

Has anyone done any CFD analysis of any sports? For instance, the spin on a golf ball, aerodynamics of a javelin, that sort of thing. I am a journalist researching an article on the use of technology in sports equipment, and I would like to include a reference to some real work.

Cheers

Dave

Hi there,

That's a difficult task. THe spin of a golf ball in the air includes more than just hydrodynamics. THis has to do with Classical Mechanics (a branch of Physics) that deals with the motion of rigid bodies. So that's a particular difficult task. You need not only to solve for the motion of the ball in the air (which includes simple mechanics and complicated hydrodynamics) but also the motion of the ball around its axis (like a spining top) and the forces on it (moments of forces etc..). I have not heard of javelin either.

However, I think that there has been some work done on things like Frisbee, maybe also on Boomerang (this one is a tough one!). But I have no reference on these.

Good luck, PG.

(1). My son sent me e-mail the other day about CFD pictures of tennis ball. I can't access my yahoo e-mail right now, till tonight. (2). So, check the site http://wings,ucdavis.edu/Tennis/CFD/Images/gallery.html This maybe is the same place where my son got the pictures. It is likely that you also have seen it.

Golf balls have dimples structure to reduce aerodynamic drag caused by friction. That is why a golf ball can be thrown to a longer distance than a regular ball of the same size and weight. Javelin has a sharp spear like structure at the tip to cut through the aerodynamic drag of air. Spinning of a golf ball is similar like spinning of a top or spinning of a spear except that one is of a conical shape (top); and the other one is of round shape (ball) or the latter one is of spear shape (javelin). You may go through some dynamics textbook if you are more interested to find out the dynamics equations involved in modeling a golf ball or a javelin.

Hi Dave,

I know they have used CFD in sports like ski jumping (to shape the skis for optimal lift) and I believe in speed skating too.

In motorsports like F1 CFD is of course in extensive use.

Kaare

Hi, As a journalist you must of heard of the America's Cup, a sailboat race between countries where sindicates spend millions of dollars. A great part of the money is spent on the boat, and each sindicate is restricted to building two boats, so tank testing and CFD analysis is crucial. If you look in the Internet for information about this race (it will be in New Zeland this year, and a very interesting boat is sponsored by Prada, in Italy) I am sure you will find lots of CFD used for sail, mast, hull, rudder and keel development. Hoping you have good luck, Astrid

Actually, dimples do not reduce the frictional drag but the increased roughness transitions the boundary layer to turbulent flow earlier which allows it to remain attached thereby reducing the form drag (the main source of drag on a golf ball at typical Reynold's numbers).

Yes, you are right. Dimples in a golf ball i.e. surface roughness tends to bring the transitional flows thereby bringing the turbulent flows earliar. Thus the form drag is reduced.

Several of the Americas Cup entrants use our software, Tecplot, to view the results of their numerical simulations. I could put you in touch with some people that might be willing to provide information about how they use CFD to improve their designs. They might even be able to supply a cool image or two! Call me at 800-676-7568 or email kelly@amtec.com if you are interested.

Good luck, Kelly

Kelly, Does Tecplot have any kind of visualization for 3-D flows? I am a doc. student and my research is in developing a FEM code for the flow arround a sailboat (the model may be an ILC25...) and I don't have tools for the 3-d visualization. If you have, are there any special packages for Universities? Thank you Astrid

Tecplot can plot 3-D volumes and 3-D surfaces, X-Y and 2-D data, vectors/streamtraces and more. There are tools within Tecplot that allow you to alter, slice and animate your data too. Tecplot's macro language allows you to automate your work and the Tecplot Add-On Developer's Kit (ADK) enables you to create custom extensions to Tecplot.

I suspect that the readers of this board would rather that I didn't go into a detailed commercial message here so I will email you directly and we can discuss whether Tecplot is a fit for your needs. In the meantime please feel free to visit our Web site (www.amtec.com) for a free demo version of Tecplot and more information about The CFD Analyzer, INCA and the rest of our products.

Kelly

PS Amtec will be releasing a new CFD-related product in the very near future. I can't say much about it right now other than it fills a big gap in the current commercial CFD software market. This product should be available in early July. Visit our Web site or contact our sales department then for more details. (Sorry folks, I couldn't resist the plug. It's not often that we launch a brand new product.)

Two classes of problems have been mentioned. One is as pointed out in the first message is wehre one wants to study the flight of objects in air and there are effects such as gravity, and applied external force (transient) that are also needed to be included in the simulation. The other class is where aerodyncamics is of primary interst such as car racing.

For the first class of problems an approach such as using overset (chimera) grids with a 6 DOF model would be ideal to use. We have implemented this approch in our CFD-FASTRAN code and it has been used for such problems as store separation and pilot ejection.

Those interested in learning more about CFD-FASTRAN are welcome to visit CFD Reserch Corporation website at http://www.cfdrc.com. You will find some very neat examples on the applications of CFD-FASTRAN.

David,

Dr. Keith Koenig at Mississippi State University has done significant research on the aerodynamics and structural mechanics of sports equipment (mostly using baseball and aluminum bats.) You might want to contact him and see if he would like to share some information.

If you are interested, email me directly and I'll give you his number.

-John

Please go to the file of one of the CFD gurus (CFD Resources Online, Misc). His name is Dr. Parviz Moin. Dr. Moin described by diagram how the drag of a golf ball is reduced due to its dimple shapes.

Hello, David,

We have done a number of sports-related CFD simulations using FLUENT over the years. The most recent was the flow past a soccer ball. You can see some of the details at: http://www.fluent.com/news/fall98/around.htm

Some researchers at Dartmouth College have been involved with yacht designs for the America's Cup. A very general overview of this work is available at: http://www.fluent.com/news/spring97/americup.htm

Finally, several years ago we performed a simulation of the flow around a ski jumper in various configurations.

If you'd like more details on these examples, please let me know.

Regards,

Eric Grald, Fluent Inc.

Drag on a golf ball comes mainly from air-pressure forces. This drag arises when the pressure in front of the ball is significantly higher than that behind the ball. The only practical way of reducing this differential is to design the ball so that the main stream of air flowing by it is as close to the surface as possible. This situation is achieved by a golf ball's dimples, which augment the turbulence very close to the surface, bringing the high-speed airstream closer. Therefore, a golf ball's dimples increase the pressure behind the ball while significantly decreasing the pressure in front of the ball. For the Reynolds number achievable by hitting the ball with a club, the co-efficient of drag is much lower for the dimpled ball.

(1). I don't play golf. I think it is interesting that people would hit the ball and try to land it in the hole, over and over again. It would be easier to develope a remote guided ball such that it will land in the hole each time one hit it. But, somehow it coincides with my theory that the only way to succeed is to fail first, to fail many times first. So, each time one hit the ball, one fails. And he does this over and over again until the ball get in the hole. (2). Back to the basic. Flow over a ball will create a separation bubble behind the ball. This is because the flow is viscous. ( If you solve inviscid flow over a ball, you are not going to get the separation bubble behind the ball. And the ball also is not going to experience the drag. No drag! So, you have to solve the viscous equations to get the drag.) (3). For flow over a smooth ball, this separation bubble starts at the location slightly less than 90 degree from the nose. And the separation bubble is as large as the ball itself in terms of the diameter. Since the pressure in the separation bubble is lower than the pressure at the nose, the net force is the drag on the ball. (4). When the ball has a rough surface, it will turn the laminar flow into the turbulent flow on the ball surface. For turbulent flow over a ball, the separation bubble will become smaller because the point of separation is now moved further downstream pass the 90 degree location. As the size of the low pressure region in this separation bubble is reduced, the drag force is reduced. ( In the inviscid flow, the separation bubble disappears completely, no drag force is created because the pressures at the nose and the rear end are equal. In addition, there is no surface frictional force.) (5). I think, it is the separation bubble size reduction which is responsible for the drag reduction. (6). I have not seen the 3-D calculation of flow over a golf ball, so it must be the result of testing, not the CFD analysis.

Nice quote. Too bad you didn't mention the source of your quote and the fact that there are some accompanying pictures coming with the story. The quoted text comes from Scientific American and can be found at:

http://www.sciam.com/0197issue/0197moinbox3.html

Kind regards, Niels

(1). It is not very scientific at all. (2). There is no mentioning of the source of the plot for drag vs Reynolds number. Was it simulated by DNS or LES or turbulence models, or wind tunnel testing? (3). How was the picture derived? Was it derived from the lab testing? Or was it computed results? What is in that separated wake region? (4). Can someone just include the complete geometry of a golf ball in the 3-D CFD calculation? I mean, is it possible to use CFD to determine the drag reduction for a real golf ball? This should be a good real world test case for large commercial CFD codes.

The text about the golfball is part of an online article written by Moin and Kim. In the article the following is said:

"Moin is now Franklin and Caroline Johnson Professor of Engineering at Stanford University and director of the Center for Turbulence Research there."

I'd say, ask Moin. He seems to be the person who came up with the fancy pictures and can give an answer wether those are LES, DNS or whatsoever.