Paralympic tandem cycling and hand-cycling: Computational and wind tunnel analysis of aerodynamic performance
Mannion, Paul
Mannion, Paul
Loading...
Repository DOI
Publication Date
2019-03-27
Type
Thesis
Downloads
Citation
Abstract
There are several key resistive forces affecting the speed of cyclists; namely aerodynamic resistance, gravity induced by road gradients, rolling resistance, and drive train and wheel bearing resistances. Of these, in most scenarios, aerodynamic resistance is the leading challenge for cyclists to overcome. Aerodynamic improvements, particularly on flat to rolling terrain, offer the greatest potential for improvements in cycling speed. On flat roads, descents and inclines of up to about 5%, air resistance is the single biggest force a cyclist must overcome. For example, at speeds in excess of 50 km/h on flat terrain the aerodynamic resistance is up to 90% of the total resistance experienced by the cyclist. The power to overcome aerodynamic resistance acting on a body is proportional to the cube of speed, meaning small increases in speed require significant increases in power. Para-cycling (competitive cycling for people with physical disabilities) has not experienced the same level of aerodynamics research that has been invested in able-bodied cycling. However, Para-athletes compete at the highest level of sporting events in the Paralympics and world championships. There is a diverse range of cycling categories within competitive Para-cycling, including hand-cycling, tricycle, tandem and traditional cycling (with a standard bicycle). Within each category there are additional sub-categories for different disability types which minimise the impact of impairment on the result of the competition. Para-cyclists can reach the velocities typically achieved by able-bodied cyclists, and in the case of hand-cyclists during a hill descent, they can exceed the velocities capable by able-bodied cyclists. Thus, aerodynamics play an important role in para-cycling, and is a key area for performance optimisation. This research focused on the aerodynamics of tandem cycling and hand-cycling (classes H1-H4) within the competitive UCI Para-cycling categories and uses computational fluid dynamics simulations for the aerodynamics analysis, supported by wind tunnel tests on reduced-scale models. A tandem cycling setup is composed of a sighted (pilot) and a visually impaired (stoker) athlete competing on a tandem bicycle. Aerodynamic equipment for tandem cycling is largely derived from traditional able-bodied cycling; including helmets, skinsuits and wheels. Cycling aerodynamics is typically considered to be a bluff body aerodynamics problem, and tandem cycling presents a unique case in cycling with two bluff bodies in close proximity to each other. Parallels can be drawn to two solo cyclists drafting, but the aerodynamics of tandem cycling has not been explored in the literature prior to this research. Hand-cycling differs from traditional able-bodied cycling to a greater degree than tandem cycling. Within the H1-H4 competition classes, the athletes adopt a recumbent position on the hand-cycle and provide propulsion power with their arms. A key output of this research includes mapping the influence of the near-wall grid resolution, and Reynolds-Averaged Navier-Stokes turbulence modelling for tandem cycling and hand-cycling. It was found that there was no single turbulence model suitable for tandem cycling, with the SST k-ω turbulence model found to be most suitable for yaw angles of 0° and 5°, and the k-kl-ω turbulence model for 10°, 15° and 20° yaw. Tandem-cycling also exhibited a dependence on the near-wall grid resolution, with counter-intuitive and incorrect flow occurrences at y* values exceeding 30. The locations for flow separation were incorrectly predicted if parameters outside of the aforementioned grid requirements and turbulence model were used, inferred by the incorrect drag predictions where the stoker experienced a larger drag than the pilot. Road race and time-trial setups for competitive race events were investigated to determine aerodynamic benefits associated with unique tandem setups such as the so-called ‘frame-clench’ time-trial position. In addition, the aerodynamic interaction between the pilot and the stoker athletes was explored by considering the drag of each athlete with respect to a solo cyclist in equivalent positions. It was found that the pilot and stoker can experience up to 92.7% and 47.8%, respectively, of the drag of an equivalent solo cyclist for a standard road race setup. This new knowledge was further expanded upon with torso angle studies and crosswind investigations that broadened the knowledge of the aerodynamic interactions between the pilot and stoker athletes. Hand-cycling did not exhibit the same pronounced dependencies on grid resolution and turbulence modelling as tandem cycling. However, the SST k-ω turbulence model coupled with a low y* grid was found to provide accurate results at 0° yaw and in crosswind conditions. The wheel selection for hand-cycling proved to have the largest impact on the aerodynamic drag. The athlete and frame reside within the vertical profiles of the three wheels utilised by a hand-cyclist, and the three wheels exhibit strong aerodynamic interactions with each other, the athlete, and the hand-cycle frame geometry. 20-inch diameter spoked wheels on the rear axle were found to be more favourable aerodynamically than 26-inch wheels. In addition, a front disk wheel coupled with two rear spoked wheels was found to provide the best combination of low drag and low lateral forces in crosswind conditions; improving stability and aerodynamic performance. These results were leveraged to develop best practice guidelines for the modelling of hand-cycling and tandem aerodynamics using computational fluid dynamics, supported by reduced-scale wind tunnel experiments to validate the numerical simulations. This research provides knowledge to athletes, coaches, and researchers regarding tandem and hand-cycling aerodynamics, and provides directions to attain improved aerodynamic performances.
Publisher
NUI Galway
Publisher DOI
Rights
Attribution-NonCommercial-NoDerivs 3.0 Ireland