"Fluid Motion," Feature Article, September 2006

fluid motion

Engineers are playing with the hulls to generate
more speed for Olympic rowers.

by Nicolas Warzecha and Andreas Spille- Kohoff

Competition among world-class athletes has become so intense that tiny improvements in performance mark the difference between gold medal winners and also-rans. So it should come as no surprise that national teams are concentrating not only on training and conditioning Olympic athletes, but also on improving their equipment to obtain a competitive edge.

It is the role of the Institute for Research and Development of Sports Equipment in Berlin to work with sports associations and other research establishments to develop equipment for the top athletes of the German national teams. The institute, known by its German acronym, FES, developed the innovative racing bikes ridden by the East German team at the Seoul Olympics in 1988. These bikes featured small front wheels that forced the athletes to lean far forward to decrease cross-sectional area and aerodynamic drag. The East German cyclists left the other national teams far behind.

Researchers prepare to test a skiff's behavior in the water at the Potsdam Model Basin. Results of this and other lab tests are being combined with CFD simulations to optimize boats for Olympic competition.

Less publicized are the bobsleds that the German team used in this year's Winter Olympics at Turin. The German team won 11 gold medals at Turin, and three were in bobsled events. The sleds were optimized by FES engineers using computational fluid dynamics software.

The bicycles for the East German team were developed by building prototypes and using wind tunnels to measure their drag. This approach is effective, but limiting because it can take weeks or even months to build a prototype. The time and expense severely limit the number of designs that can be evaluated.

Evaluating More Designs

FES made significant, though less visible, improvements in the air resistance of the bobsled design for Turin. But the process of designing equipment has changed since 1988. For the first time, FES engineers simulated many different alternative bobsleds using computational fluid dynamics software to calculate the wind resistance on the computer before a prototype was built. This made it possible for the engineers to evaluate more than 10 times the number of designs that would have been possible using build-and-test methods. The engineers also used CFD to fine-tune the new sled's braking and steering characteristics. Then they spent two years welding and bolting together the prototype.

The German team won a total of 29 medals at the 2006 Winter Games. It edged out the United States, which won 25 medals, including nine gold.

The institute's engineers, guided by simulations using Ansys CFX software from CFX Berlin Software GmbH, are engaged in the much more challenging task of optimizing the performance of some of the 14 different classes of boats used in Olympic rowing and canoeing. FES researchers say that they have already succeeded in developing a preliminary design that promises to reduce drag by 3 percent compared to the design used during the last Olympics.

A boat moves in a complex manner in the water as it is propelled by the rower. As a result, its water resistance changes continually. Simulating the water resistance requires tracking the trim of the boat as it bobs up and down in response to the rower's efforts and the water's effects.

Physical and virtual: A racing canoe makes its way through real water at the model basin (above) and through computerized fluid dynamics modeling (below). The model was run in parallel on 24 Linux processors.

FES engineers used the fluid dynamics software to simulate the performance of various boat designs. The software includes a feature called Expression Language, which allows users to create physical models quickly from within the user interface and add new variables. It also allows Fortran applications to be coupled to Ansys CFX. This capability made it possible for FES engineers to accurately model the heaving and pitching of the boat as it moves through the water, which was critical to achieving accurate results.

The engineers began by simulating experiments in which boats were towed, to eliminate the effect of a rower's strokes on the boat's motion. They created a block structured grid model with three million elements and used the software's free-surface multiphase model to analyze both the motion of the water and the air trapped by the movement of the boat and the water. This simulation showed a very good correlation with the drag measured in towing experiments. All cases were simulated in parallel on a cluster of 24 64-bit Linux processors, whose installation was supported by CFX Berlin, an independent distributor of Ansys software in Berlin.

The wetted surface of the boat hull in this model did not match the experiments, since the simplified model didn't account for the forces acting on the boat. So FES engineers simulated the boat at its position with the solver to calculate these events and estimated the new position of the boatÑthat is, its sink and trim values. They continued with a series of manual steps that slowly converged to a final position showing good agreement with experiments. Once they had determined that the approach provides realistic results, this approach was automated. Using a combination of Expression Language and Fortran code, the software calculated a boat's sink and trim position on the fly during solution.

More Than Drag

A key advantage of CFD analysis is that it provides far more information than physical testing does. Physical testing typically provides a single number, the amount of drag at a certain speed. This number, it should be noted, does not take into account the forces exerted by the rowers. On the other hand, the analysis performed by FES records calculates drag as well as information on the movement of the water around the boat, the position of the boat, and the forces acting on the boat.

In particular, CFD makes it possible to precisely measure the bow wave, aft wave, and wake of the boat. These three phenomena are generated by the energy expended in drag. By understanding them, engineers get a good indication of what is causing drag in a design and what aspects of the design they should change to improve it.

CFD made it possible to evaluate in a few months a number of designs that would have taken far more than an Olympic cycle to build and test. Instead, FES will build and test only the most promising designs. A prototype boat is under construction, and physical testing will be used to verify simulation results. Testing is particularly important because CFD is not yet able to predict stability, which is another critical factor in a boat's performance. A boat in the Athens Olympics in 2004 capsized when it was struck by a small wave.

Simulation by itself will not lock up the rowing competition for a team in the Olympics. The performance of athletes is still by far the most important factor. And, it is important to note that many other countries are also using simulation to design sports equipment in a behind-the-scenes competition that will help to determine the course of the 2008 games.

In the meantime, FES engineers are working to expand the scope of CFD analysis to analyze other factors that are difficult to measure, such as the effect of initiating rowing forces at different times and variations in the velocity of the boat. Including these effects in the analysis may make it possible to send the performance of the boats and athletes to even higher levels.

Nicolas Warzecha is responsible for CFD simulations for water sports equipment at the Institute for Research and Development of Sports Equipment in Berlin, Germany; Andreas Spille-Kohoff is responsible for research and development at CFX Berlin Software GmbH.