Publications

A new method for characterising bleed system loss is introduced, using research compressor test results as a demonstration case. A loss coefficient is defined for a control volume including only flow passing through the bleed system. The coefficient takes a measured value of 95% bleed system inlet dynamic head, and is shown to be a weak function of compressor operating point and bleed rate, varying by +/-2.2% over all tested conditions. This loss coefficient is the correct non-dimensional metric for quantifying and comparing bleed system performance.

This paper presents a novel method that can completely condition the flow into a turbomachinery experiment. A single, thick, 3D-printed gauze can be tailored to provide an exact stagnation pressure profile, flow angle distribution and turbulence intensity. The new method is superior to existing techniques as it provides accurate and cheap flow conditioning in just one component. It removes the requirement for separate endwall boundary layer generators, inlet guide vanes and turbulence grids. The paper is presented in two parts: first, the methods for designing complete flow conditioning gauzes are presented. In the second part, two gauzes are designed and manufactured for two compressor testing applications. Both applications demonstrate the fine control that can be achieved in an experiment using these gauzes.

The cost effectiveness of a tidal stream turbine can be improved by maximising the power extracted for a given rotor diameter. This paper presents a numerical and experimental study showing that winglets could be used to this end. The numerical simulations were conducted using Tornado, a vortex lattice code, which can model the interaction between different spanwise sections unlike Blade Element Momentum methods. Tornado was used to identify the important winglet design parameters such as dihedral angle. Tornado cannot capture viscous effects and so an experimental study was conducted on four designs. These were tested on a small-scale horizontal axis turbine in the Ifremer flume tank. The impact of winglets on the blade spanwise flow was found to have a significant effect on the amount of loss generated. The inviscid code used in this paper could complement existing quasi-3D design tools.

An unsteady five-hole probe has been developed for the measurement of turbulent flow in tidal channels. Such measurements are vital for accurate prediction of unsteady loads on tidal turbines. Existing field-based velocimeters are either unable to capture the required range of frequencies or are too expensive to profile the variation of turbulence across a typical tidal power site. This work adapts the traditional five-hole wind tunnel probe to achieve a low-cost device with sufficient frequency range for tidal turbine applications. The main issue in the marine environment is that the ambient hydrostatic pressure is much higher than the dynamic pressure. This has been overcome by using novel calibration coefficients and differential transducers. In flume tank tests against laser Doppler velocimeter measurements, the frequency response of the probe has been shown to be sufficient to capture all the frequencies necessary for tidal turbine design.

This paper introduces a pneumatic 9-hole probe which can measure flow angles, stagnation and static pressures, and spatial derivatives of stagnation pressure. It does this through direct measurement at a single location, rather than empirical corrections using measurements at multiple points. The new design resembles a 5-hole probe with 4 additional holes positioned around the side of the probe head. This arrangement enables the probe to distinguish between flows with stagnation pressure gradient and flows at an angle. Mapping between the inputs, the probe hole pressures, and outputs, the calibration reference measurements, is achieved with a trained neural network which takes the place of a conventional calibration map.