# PIV-based load determination in aircraft propellers

PIV-based load determination in aircraft propellers

Author Contributor Faculty Department Date2012-02-10

AbstractThe thesis describes the application of particle image velocimetry (PIV) to study the aerodynamic loads of airfoils and aircraft propellers. The experimental work focuses on the development of a measurement procedure to infer the pressure of the flow field from the velocity distribution obtained by PIV velocimetry. The technique offers important advantages in aircraft propellers, since the loads can be locally inspected without the need to install pressure sensors and momentum balances in rotating devices. The applicability of the approach in problems relevant to the aerospace industry requires theoretically defining and experimentally verifying the principles of this method. Modern aircraft propellers operate in the compressible regime, in conditions where the blade profiles are usually experiencing transonic effects, especially at the blade tip. Therefore, the PIV based load determination is firstly applied in airfoils immersed in transonic flows (Mach 0.6 to 0.8) and further compared to conventional pressure measurement techniques to measure the surface pressure distribution, airfoil lift and drag. Three different fields of view have been used at different spatial resolutions, to obtain 2D velocity vector fields from which the surface pressure, as well as the lift and drag coefficients were computed. In the lift computation, the contour integration method is applied, whereas the drag is derived from a wake traverse approach. The surface pressure showed excellent agreement with the data recorded by the pressure orifices in the absence of shocks, although to correctly capture the pressure on the nose region of the airfoil an extrapolation of the PIV data towards the actual surface is needed, in view of the large pressure gradient normal to the surface and to the limited spatial resolution. The lift obtained by PIV can be reliably compared to conventional load determination approaches, showing uncertainties of the order of few percent. On the other side, the drag needs a proper experimental optimization to reach the same uncertainty values of the lift. In the drag coefficient computation, the wake based formulation is crucial to obtain accurate results. Main parameters in the drag determination from PIV data are the effect of the free stream pixel shift on the accuracy of the velocity derivative computation and the influence of the algorithm employed for the pressure reconstruction. An experimental investigation revealed that an increase in the pixel shift (increase in pulse separation) reduces the uncertainties in the outer part of the wake. The drag coefficient computation showed how the use of a nondirectional high order global integration algorithm improves the reliability of the drag computation by a factor of 2.5 in comparison to the one of a planar spatial marching algorithm. When using PIV on scaled down models operating in the transonic regime, the particle inertia cannot be considered negligible. Consequently, the flow tracer motion departs from the one of the surrounding fluid, compromising the accuracy of the measurement technique. The phenomenon is investigated examining several types of particle tracers (liquid droplets and solid particles) with the “shock wave test”. The main results of this study indicate that the temporal response of DEHS droplets of diameter in the one micron range is within 2 microseconds, overcome by properly treated (dehydrated) solid particles of sub micron diameter (SiO2 or TiO2 agglomerates) yielding a relaxation time as short as half of a microsecond. The second application focuses on aircraft propellers, where the PIV based load determination technique is compared to computational fluid dynamics (CFD), due to the technical limitations in instrumenting small blades operating at high rotational speed. The first study deals with the determination of the cross sectional aerodynamic loads on a scaled down aircraft propeller model operating at zero angle of incidence. The stereoscopic PIV technique is employed to perform phase locked measurements of the flow around the blade at 75% blade radius. The aerodynamic force is evaluated considering a control volume, for momentum integration, that moves at the same speed as the blade section. This formulation has the important advantage that the unsteady terms in the momentum equation are eliminated, simplifying the data acquisition and processing. The study of three dimensional effects occurring at the blade tip is conducted by dedicated experiments, whereby the three velocity components are measured at several planes by stereoscopic PIV with a traversing system along the blade radius. The consistent treatment of the three dimensional flow requires the velocity derivatives to be evaluated along the span wise direction. This is achieved by measurements with small separation along the radial direction, similarly to the dual plane stereoscopic PIV technique. Further comparison with a CFD simulation shows that the aerodynamic loads can be estimated in the three dimensional regime with similar accuracy as in the two dimensional one. The results demonstrated that phase locked measurements allow resolving the periodical flow of the blade revolution. A quantitative analysis of the measured pressure fields demonstrated how the propeller blade becomes less tractive as the measurement planes move to the tip, and shows the corresponding decrease in the blade resistance due to the presence of the trailing vortex. The numerical simulation confirmed the experimental pressure analysis, providing comparable results with maximum differences of the order of 10% in the numerical pressure coefficient, which may partially be ascribed to uncertainties in the representation of the real blade shape in the numerical model. The sectional PIV derived thrust proved the most in agreement with the one from simulation data, showing the expected decrease in traction up to the tip in both the CFD and PIV results. The experimental sectional drag, due to the limited resolution in capturing the thin blade wake, compares favorably to the numerical data mainly at the inboard part of the measurement domain, while a consistent deviation between experiment and simulation was observed at the immediate tip region. An unprecedented result of the multi plane analysis is the evaluation of the pressure over the entire surface of the blade, which shows the full potential of this technique for propeller blades load diagnostics. The techniques developed in the Aerodynamics Laboratories of TU Delft are finally employed at a larger scale in the industrial wind tunnels of the German Dutch Wind tunnels consortium (DNW), where the thrust and torque forces derived from PIV on a Hamilton Standard aircraft propeller model are meant to be compared to the forces obtained from a shaft balance and a multi component external one.

To reference this document use:http://resolver.tudelft.nl/uuid:5fac8b3d-f771-4842-910a-ba6118bfb5cf

ISBN9789461911940

Part of collectionInstitutional Repository

Document typedoctoral thesis

Rights(c) 2012 Ragni, D.