This paper describes an experimental study of the three-dimensional flow within a high-speed linear cascade of low-pressure turbine blades. Data were obtained using pneumatic probes and a surface flow visualization technique. It is found that in general, the flow may be described using concepts derived from previous studies of high-pressure turbines. In detail, however, there are differences. These include the existence of a significant trailing shed vortex and the interaction of the endwall fluid with the suction surface flow. At an aspect ratio of 1.8, the primary and secondary losses are of equal magnitude.

Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details. The manuscript was received on 10 July 2006 and was accepted after revision for publication on 14 September 2006. DOI: 10.1243/09576509JPE340Abstract: Darrieus-type vertical axis wind turbines have a number of potential advantages for small-scale and domestic applications. For such applications, the issues of cost and reliability are paramount and hence simplicity of design of the structure, the generator, and any control system is vital. A particular concern relating to Darrieus turbines is their potential to selfstart. If, as has been suggested by several authors, they require external assistance to start then much of their advantage is lost. The purpose of the study described here is, therefore, to investigate their starting performance through the development and validation of computational simulation and to determine the parameters that govern the capability to self-start. A case study is presented based upon the use of the widely used and well documented, symmetrical NACA 0012 turbine blade profile. It is shown that a lightly loaded, three-bladed rotor always has the potential to self start under steady wind conditions, whereas the starting of a two-bladed device is dependent upon its initial starting orientation.

As the demand for renewable energy grows, the use of small wind turbines becomes increasingly attractive. Turbines using vertical-axis geometries are particularly suited to the urban environment by virtue of their operation being independent of wind direction. However, such designs have received much less attention than the more common ‘propeller type’ designs and the understanding of some aspects of their operation remains weak. This is particularly true of their starting characteristics. Indeed, some authors maintain that they cannot start without external assistance. In this investigation a numerical model is used to simulate the starting of an H-rotor Darrieus turbine under steady wind conditions. Experimental wind-tunnel data for a small prototype is presented, demonstrating unaided start-up of a three-bladed Darrieus in a steady wind. Discrepancy between the modelled and experimental results demonstrate that modelling remains constrained by the quality of data on aerofoil characteristics.

The effects of Reynolds number, Mach number, and turbulence on the calibrations of commonly used types of five-hole probe are discussed. The majority of the probes were calibrated at the exit from a transonic nozzle over a range of Reynolds numbers (7 × 103 < Re < 80 × 103 based on probe tip diameter) at subsonic and transonic Mach numbers. Additional information relating to the flow structure were obtained from a large-scale, low-speed wind tunnel. The results confirmed the existence of two distinct Reynolds number effects. Flow separation around the probe head affects the calibrations at relatively low Reynolds numbers while changes in the detailed structure of the flow around the sensing holes affects the calibrations even when the probe is nulled. Compressibility is shown to have little influence upon the general behavior of these probes in terms of Reynolds number sensitivity but turbulence can affect the reliability of probe calibrations at typical test Reynolds numbers.