Boundary Layer Measurements on a Rotating Disk

Author(s):  
Vrishank Raghav ◽  
Narayanan Komerath

Dynamic stall occurs on helicopter rotor blades, wind turbine blades, and even insect and bird wings. Although most studies on dynamic stall are conducted assuming two-dimensional behavior, it has been shown that this phenomenon is highly three-dimensional. Recent studies of dynamic stall on a rotating blade of helicopter in forward flight and wind turbines in yaw have shown that the nature of the radial flow near the surface has first-order significance in the stalled flow field. Past literature suggests that the boundary layer over a rotating disk has been correlated to that over a swept wing. Drawing a similar parallel it is hypothesized that fundamental insights into dynamic stall may be derived from studying the boundary layer over a rotating blade in dynamic stall is similar to that over a rotating disk. Particle image velocimetry (PIV) is used to investigate the boundary layer development on a rotating disk. The first set of measurements was conducted using PIV with a micro lens attachment to the camera. Next, a microscope was used to conduct μ-PIV measurements. The measured radial velocity profiles show substantially higher radial jet peak velocities than the analytical solution for a mirror-smooth disk. This difference is narrowed down to the effect of surface roughness of the painted disk, representative of reality on rotor blades used in PIV. This results in a much higher effective viscosity in the near-surface layer, contributing to additional radial flow, as seen from the centrifugal pump literature. However, the non dimensional radial velocity profile exhibits the expected self similar behavior at various radial locations.

2009 ◽  
Vol 23 (03) ◽  
pp. 505-508 ◽  
Author(s):  
RUI YANG ◽  
REN-NIAN LI ◽  
WEI HAN ◽  
DE-SHUN LI

The flow field past the rotating blade of a horizontal axial wind turbine has been modeled with a full 3–D steady–RANS approach. Flow computations have been performed using the commercial finite–volume solver Fluent. The NREL phase VI wind turbine blade sections from the 3–D rotating geometry were chosen and the corresponding 2–D flow computations have been carried out for comparison with different angles of attack and in stalled conditions. The simulation results are analyzed. The main features of the boundary layer flow are described, for both the rotating blade and the corresponding 2–D profiles. Computed pressure distributions and aerodynamic coefficients show evidence of less lift losses after separation in the 3–D rotating case, mostly for the inward sections of the blade and the highest angles of attack, which is in agreement with the literature.


2017 ◽  
Vol 139 (3) ◽  
Author(s):  
Hamidreza Abedi ◽  
Lars Davidson ◽  
Spyros Voutsinas

The aerodynamics of a wind turbine is governed by the flow around the rotor, where the prediction of air loads on rotor blades in different operational conditions and its relation to rotor structural dynamics is one of the most important challenges in wind turbine rotor blade design. Because of the unsteady flow field around wind turbine blades, prediction of aerodynamic loads with high level of accuracy is difficult and increases the uncertainty of load calculations. An in-house vortex lattice free wake (VLFW) code, based on the inviscid, incompressible, and irrotational flow (potential flow), was developed to study the aerodynamic loads. Since it is based on the potential flow, it cannot be used to predict viscous phenomena such as drag and boundary layer separation. Therefore, it must be coupled to tabulated airfoil data to take the viscosity effects into account. Additionally, a dynamic approach must be introduced to modify the aerodynamic coefficients for unsteady operating conditions. This approach, which is called dynamic stall, adjusts the lift, the drag, and the moment coefficients for each blade element on the basis of the two-dimensional (2D) static airfoil data together with the correction for separated flow. Two different turbines, NREL and MEXICO, are used in the simulations. Predicted normal and tangential forces using the VLFW method are compared with the blade element momentum (BEM) method, the GENUVP code, and the MEXICO wind tunnel measurements. The results show that coupling to the 2D static airfoil data improves the load and power predictions while employing the dynamic stall model to take the time-varying operating conditions into consideration is crucial.


Author(s):  
A. Brown ◽  
B. W. Martin

This paper reviews the methods for predicting boundary-layer behavior on flat and curved surfaces under conditions experienced in gas turbine engines and the resultant heat transfer to the turbine rotor blades. Particular attention is given to the effects of streamwise pressure gradient and the intensity of mainstream turbulence on transition phenomena. The time-mean heat transfer across a boundary-layer under unidirectional oscillatory mainstream flow, such as might be initiated in a combustion chamber, is considered. The relevance of flat plate predictions and correlations to rotating turbine blades is also discussed.


2019 ◽  
Vol XVI (2) ◽  
pp. 13-22
Author(s):  
Muhammad Ehtisham Siddiqui

Three-dimensional boundary-layer flow is well known for its abrupt and sharp transition from laminar to turbulent regime. The presented study is a first attempt to achieve the target of delaying the natural transition to turbulence. The behaviour of two different shaped and sized stationary disturbances (in the laboratory frame) on the rotating-disk boundary layer flow is investigated. These disturbances are placed at dimensionless radial location (Rf = 340) which lies within the convectively unstable zone over a rotating-disk. Mean velocity profiles were measured using constant-temperature hot-wire anemometry. By careful analysis of experimental data, the instability of these disturbance wakes and its estimated orientation within the boundary-layer were investigated.


1985 ◽  
Vol 40 (8) ◽  
pp. 789-799 ◽  
Author(s):  
A. F. Borghesani

The Navier-Stokes equations for the fluid motion induced by a disk rotating inside a cylindrical cavity have been integrated for several values of the boundary layer thickness d. The equivalence of such a device to a rotating disk immersed in an infinite medium has been shown in the limit as d → 0. From that solution and taking into account edge effect corrections an equation for the viscous torque acting on the disk has been derived, which depends only on d. Moreover, these results justify the use of a rotating disk to perform accurate viscosity measurements.


Author(s):  
Lena Pfister ◽  
Karl Lapo ◽  
Larry Mahrt ◽  
Christoph K. Thomas

AbstractIn the stable boundary layer, thermal submesofronts (TSFs) are detected during the Shallow Cold Pool experiment in the Colorado plains, Colorado, USA in 2012. The topography induces TSFs by forming two different air layers converging on the valley-side wall while being stacked vertically above the valley bottom. The warm-air layer is mechanically generated by lee turbulence that consistently elevates near-surface temperatures, while the cold-air layer is thermodynamically driven by radiative cooling and the corresponding cold-air drainage decreases near-surface temperatures. The semi-stationary TSFs can only be detected, tracked, and investigated in detail when using fibre-optic distributed sensing (FODS), as point observations miss TSFs most of the time. Neither the occurrence of TSFs nor the characteristics of each air layer are connected to a specific wind or thermal regime. However, each air layer is characterized by a specific relationship between the wind speed and the friction velocity. Accordingly, a single threshold separating different flow regimes within the boundary layer is an oversimplification, especially during the occurrence of TSFs. No local forcings or their combination could predict the occurrence of TSFs except that they are less likely to occur during stronger near-surface or synoptic-scale flow. While classical conceptualizations and techniques of the boundary layer fail in describing the formation of TSFs, the use of spatially continuous data obtained from FODS provide new insights. Future studies need to incorporate spatially continuous data in the horizontal and vertical planes, in addition to classic sensor networks of sonic anemometry and thermohygrometers to fully characterize and describe boundary-layer phenomena.


Sign in / Sign up

Export Citation Format

Share Document