Image based measurement techniques for aircraft propeller flow diagnostics

Abstract

The aim of the thesis is to measure the propeller slipstream properties (velocity and vorticity) and to assess the unsteady and instantaneous behavior of the propeller flow field at high disk loadings, zero thrust and thrust reverse using the image based measurement techniques. Along with its implementation of the techniques in industrial facilities particular propeller phenomena are addressed, such as the propeller slipstream behavior at varying angles of attack, and its interaction with the wing for thrust cases and no thrust. In addition the flowfield behavior at propeller thrust reverse is investigated. The application of PIV in industrial facilities requires special attention. Accurate and timely calibrations are performed with a tachymeter. Laser light sheet reflections are reduced using Rhodamine acrylic paint on wind tunnel models and black adhesive foils at wind tunnel walls. Very fruitful is the application of a traversing system, which enables additional time-efficient calibration procedures while only disparity corrections are necessary as well as a timely acquisition of data in multiple planes. Using these methods industrial measurements are performed with the focus on the determination of unsteady propeller phenomena. Initial measurements are performed for the investigation of a single sting propeller. Particle Image Velocimetry (PIV) is applied in order to study the vortex pattern behind the propeller. Now, several planes are measured with the intent to determine the forces on the propeller using a simplified momentum method approach. It is shown that this method depends severely on the inflow boundary conditions, making such a method impracticable for engineering estimations. Nevertheless is the investigation of the vorticity distribution considerably useful in the quantification of high loading events. Additional BOS measurements show that the propeller slipstream contraction can be determined which is useful for a validation of less advanced CFD codes. PIV is then applied to propeller flow features at high lift conditions. The slipstream of a propeller on a half-model is investigated for increasing angles of attack. The root vortices from the propeller blade are shown to introduce secondary vortices in the flow field distribution over the nacelle and the wing. A numerical calculation is available which allows addressing several aspects of the level of agreement between unsteady Reynolds-averaged Navier-Stokes calculations and phase-locked Particle Image Velocimetry measurements. Although in general the flow field distributions are comparable there are distinct differences in the vortical regions. In particular the secondary vortices caused by the root vortices are not visible in the boundary layer in the numerical computations. A comparison of the circulation strength of the tip vortices indicates that the numerical calculations suffer from significant numerical dissipation. During the analysis of the final PIV data is was found that the symmetric phase-only filter has a detrimental effect on the peak-locking. A more detailed analysis based on synthetic images shows that the phase-only filter performs badly in regions with varying particle diameter and fluctuation free flows. An artificial introduced fluctuation field shows that it overshadows the peak-locking. The symmetric phase-only was originally introduced for PIV measurements with significant reflections. It is suggested to use this filter only if it can be shown that no peak-locking is present in the data (e.g. on a sample of the free stream flow). The final PIV measurement is performed for the investigation at propeller thrust reverse. Thrust reverse is investigated for two cases. One setting has identical blade pitch angles and the second setting has alternating blade pitch angles. It is shown that the phase-locked flow fields at both cases differ significantly, although both settings deliver the same amount on negative thrust. The lesser amount of fluctuations in the second case is supported by hot-film measurements. The alternating blade pitch angles are thus a viable option for producing negative thrust on propellers. The outlook presents PIV measurements and CFD calculations on an ‘open rotor’ which indicate that there is future need for PIV measurements, in order to get an understanding of the flow field behavior as well as for the validation of numerical codes.