This thesis presents the work undertaken to assess potential improvements in high performance bicycles. There are several wheel options available for elite riders to use in competition and this research has investigated the aerodynamic properties of different wheel type. The research has also developed CFD and FEA models of carbon fibre bicycle wheels to assist in the wheel improvements process.

An accurate and repeatable experimental test rig was developed to measure the aerodynamic properties of bicycle wheels in the wind tunnel, namely translational drag, rotational drag and side force. Both disk wheels and spoked wheels were tested.

It was found that disk wheels of different hub widths have different aerodynamic properties with the 53mm wide Zen disk wheel requiring the lowest total power of the wheels tested. There was little difference between the translational power requirements of the wheels but there was greater variation in the rotational power requirements.

Compression spoked wheels of 3 and 5 spokes were found to require less power than wire spoked wheels. There was little difference between the total power requirements of the compression spoked wheels tested, with the differences at 50km/hr being less than the experimental uncertainty.

The Zipp 808 wheel demonstrated considerably lower axial force than all other wheels at 10° yaw angle, confirming Zipp design intention to have optimum wheel performance between 0-20°. The Zen 3-spoke wheel showed the lowest axial drag and side force at yaw of the compression spoked wheels tested and had similar side force results to the Zipp 808.

CFD models of the disk and 3-spoke wheel achieved good agreement with the experimental results in terms of translational drag. Rotational drag did not agree so well, most likely due to the turbulence model being designed for higher Reynolds number flows.

A FE model of the disk wheel was validated with experimental testing. In order to simplify modelling, the FE model of the 3-spoke wheel did not include the hub, which led to a large discrepancy with experimental results for the particular loading scenario.

The experimental rig and CFD models were used to develop aerodynamic improvements to the wheel and the FE models were used to identify the implication of geometric changes to the wheel structural integrity. These improvements are not reported in this thesis due to the results being commercially sensitive.