The influence of flow unsteadiness on erosive cavity dynamics

This paper presents investigations of unsteady flow conditions influence over cavitation dynamics. Measurements of wall pressure fluctuations in a model turbine are first introduced. It is shown that the wake of the spiral casing tongue modulates cavity dynamics near the trailing edge of runner blades. A new device, which permits a 2D hydrofoil to oscillate around its mean incidence angle, is used in order to recreate flow conditions similar to those of blade row wakes experienced in industrial flows. This device is mounted in the LMH High-Speed Cavitation Tunnel test section. Flow visualizations and surface pressure measurements are performed concurrently. Perturbations strongly affect leading edge cavitation at some flow conditions. Complete disappearance of leading edge cavitation can be imposed at a particular operating point. It is proposed that this behavior is associated with turbulent boundary layer transition. Pressure time signals obtained with the oscillating hydrofoil and the model turbine share common features that are different from those obtained with the hydrofoil set at fixed incidences.

Etude de l'influence des instationnarités des écoulements sur le développement de la cavitation

Jean-François Caron

Flows in hydraulic machinery show interactions between guiding and rotating components. The wakes of the fixed components will be part responsible for the pressure fluctuations sensed on the blades. Those fluctuations become more important when the distance between the fixed and rotating components decreases. Global instabilities induced by the rope in the draft tube will also be sensed all over the installation. Cavitation on the blades will react in some ways to those fluctuations. This work is devoted to the influence of flow unsteadiness on the cavitation phenomenon. Two different flow configurations are tested, both subject to macroscopic fluctuations of the flow variables. The first part of this work involves the use of a fixed and oscillating NACA hydrofoil in the test section of the High Speed Cavitation Tunnel. Different leading edge configurations corresponding to hydrodynamically smooth and rough surfaces are tested. Observations of cavitation figures are done in parallel with pressure measurements on the hydrofoil suction side. It is observed that the pressure field leads the hydrofoil motion. This positive phase lag increases with the oscillation reduced frequency. For a particular reduced frequency, the phase lag between the movement and the pressure signals increases from the leading to the trailing edge of the hydrofoil. The transition location of the boundary layer on the suction side is modified by the movement. This transition follows the hydrofoil movement if compared with the fixed incidence case. Leading edge cavitation on the hydrofoil suction side interacts with the fluctuating pressure field. Cavitation appearance becomes intermittent when the leading edge is smooth. It disappears over all oscillation cycles if the oscillation frequency is further increased. This phenomenon is not observed when the leading edge is rough. Cavitation is also observed on the smooth parts of the leading edge when a rough patch is put at mid span. Intermittence is therefore not related to the modification of the pressure field, but to the state of the laminar boundary layer in the low pressure region. The positive phase lag of the pressure field also act to lower cavitation aggressiveness when the leading edge is rough on the oscillating hydrofoil. Those observations are validated by the measured acceleration. The second part of this work is devoted to the study of cavitating flow field in Francis turbines. A model turbine is used for pressure measurements on the blades without cavitation. The operating conditions correspond to the best efficiency point, highest power output at highest head, part load at the lowest head and part load at the highest head. It is observed that pressure fluctuations measured on the blades have the highest amplitude at part load. Those pressure fluctuations are related to the swirl of the flow in the draft tube. Other pressure fluctuations sources are the distribution of velocity at runner inlet and the wakes of the guide vanes. Experiments with cavitation on the blades show that the distribution of the flow at the runner inlet has a strong influence on the pressure pulses measured. Those pulses are detected in the pressure signals before cavitation becomes visible on the blades. An increase in the Reynolds number of the flow acts to higher the incipient cavitation parameter. This observation suggests that cavitation tests should be done at the highest Reynolds number reachable with the model turbine.

EPFL2002

Analyse des erreurs dues à l'écoulement dans la mesure de débit par la méthode acoustique

The precise flow-rate measurement is a basic problem in the management and the exploitation of a hydraulic power-plant. The efficiency characterisation of an hydraulic machinery or of a water distribution net is linked to the accurate knowledge of the flow-rate. As well as the usual measurement methods, acoustic discharge measurement has become of great importance during the last ten years, specially due to the last development in the field of electronic and signal processing. The acoustic flow measurement uncertainties show that geometry, integration method, influence of the flow field and the interaction between the acoustic transducer and the pipe have an important influence on the measurement accuracy. In this work, the influence of the flow field on the accuracy is of great importance. To estimate the uncertainties of the acoustic discharge measurement for disturbed flow fields, laboratory tests are carried out at the hydraulic machinery laboratory LMH in order to qualify the measurement accuracy depending on the upstream distortion induced by different typical pipe configurations. Furthermore, numerical flow computations with software packages are performed to simulate the turbulent flow profiles at the location of the acoustic flow measurement. These simulations are performed for laboratory and also industrial flow configurations. Comparative discharge measurements for typical disturbed flow fields are performed on a special test rig of LMH between a reference electromagnetic flow meter and an 8 paths in two crossed planes acoustic flow-meter. The accuracy analysis, based upon all laboratory measurements shows that globally the measurement quality is excellent even in the case of a heavily disturbed flow field in the meter section. A specific analysis has to be performed at the flow-rate measurement installation in order to take into account the pipe configuration influence (for example an elbow or a gate), the developing length between the meter section and the disturbance and finally the use of crossed planes. Such analysis allows the reduction of the acoustic measurement uncertainties. The protrusion effect can also introduce an important measurement error in the acoustic measurement. Laboratory measurements in the high speed cavitation tunnel of LMH show clearly the influence of the protrusion effect. In such small dimensions in comparison with the industrial pipes, these effects are amplified and they introduce important errors. The second part of this work is based on the simulation of the velocity profiles at the location of the metering section by numerical flow computations. Comparisons between numerical results and Laser Doppler flow measurements show that the simulation and the flow physics results are in agreement with each other. Based on these convincing results, numerical flow computations are performed for an industrial power-plant in order to obtain an approximation of the flow profile in the metering section. Based on these results, an optimisation of the acoustic path installation is proposed in order to limit the uncertainties due to the flow-field in the pipe. This study is applied to the laboratory configurations and allows one to find the optimum measured orientations of the acoustic paths. Finally, this study also applies to an industrial installation.

EPFL1997

Cavitation Erosion Prediction on Francis Turbines Part 3 : Methodologies of Prediction

François Avellan, Philippe Dupont

In the frame of a joint research program between EDF, Hydro Québec and IMHEF, different methods are investigated to predict cavitation erosion on Francis turbines from model. They are based on measurement of pitting, pressure fluctuations and acceleration. The measurement techniques have been detailed in Part I and Part 2. The present article describes essentially the theoretical and practical aspects of the methods and discusses the results obtained until now from the mode! and prototype tests. The first analysis shows that the methods proposed are suitable to measure cavitation aggressiveness on mode! and on prototype, and that the level on the mode! is several orders of magnitude smaller than on the prototype. To adjust transposition laws, a more complete set of data is needed.

Kluwer Academic Publishers1996

Analyse des erreurs dues à l'écoulement dans la mesure de débit par la méthode acoustique

EPFL1997

Etude de l'influence des instationnarités des écoulements sur le développement de la cavitation

Jean-François Caron

EPFL2002