Boundary layer transition in attached and separated flows at low Reynolds numbers

Abstract

The transition of a boundary layer from a laminar to a turbulent state is known to be affected by several flow and surface conditions, including flow Reynolds number, streamwise pressure gradient, free-stream turbulence, surface roughness, and periodic oscillations in free-stream velocity. Although the effects of these parameters on the transition process have been the subject of numerous investigations over the past decades, the interactions between these conditions, when present simultaneously, have yet to be fully documented. The current research program was undertaken to increase our understanding of the process of boundary layer transition, and to quantify the relative importance of these conditions. The majority of the present study consists of hot-wire anemometer measurements of a two-dimensional boundary layer developing over a flat surface. The ranges of Reynolds number, pressure distribution, free-stream turbulence, periodic¬ unsteadiness, and surface roughness were representative of the suction side of low pressure turbine blades, and residted in both attached-flow and separation-bubble transition processes. Based on the experimental results, an improved transition model to predict these effects is presented. This model is applicable to a broader range of flow and surface conditions than available alternatives. The experiments are complemented by numerical simulation of one of the test cases, by means of large-eddy simulation. The simulation provides an opportunity for detailed study of the unsteady flow phenomena and vortex shedding processes that accompany transition and reattachment of the separation bubble. Yaras, for his enthusiastic participation and continuous support during this research. 1 consider myself very fortunate that many thankless tasks, such as the majority of design and commissioning of the experimental test section, had already been completed before I joined this project. I am also very grateful for the cooperation of Profs. J.

The Multi-Objective Design of Flatback Wind Turbine Airfoils

“…83 m (6 ft) long test section, with a width of 0.762 m (30 in) and a height of 0.508 m (20 in). Centre-line velocities up to about 60 m/s, as measured with a pitot-static pressure tube connected to a U-tube manometer by the current author, are possible in the test section which has good flow angularity and turbulence intensities of ±0.15 deg and about 0.1-0.5%, respectively, as measured by El-Ramly and Roberts[117,118].These good flow characteristics are vital in the testing of airfoils where for e.g. high turbulence can lead to premature transition.…”

An algorithm for the aerodynamic analysis of rotating blades using a weak viscous-inviscid interaction method

“…This was a reasonable assumption, except in turbomachinery (p a rticu la rly low pressure turbines), where the length o f separation bubbles may be significant, up to 30% o f the suction surface o f a turbine blade (Roberts, 2005).…”

“…For transition in short separation bubbles (separation bubbles that have only a local effect on the pressure distribution and consist o f only one recirculation zone (Gaster, 1969)), studies including M ayle (1991), Roberts (1975), Davis et al (1987), Hatman & Wang (1999), Yaras (2002), and more recently Roberts & Yaras (2005b) have provided relationships between the location o f transition inception and the state o f the boundary layer at the separation location. Predicting the transition length in separation bubbles is more d iffic u lt because transition occurs very rapidly (Roberts, 2005). In itia lly , such as in the studies by Horton (1969) and Roberts (1980), point transition was assumed.…”

“…To the author' s knowledge, no correlation has been developed to predict the displacement thickness w ith in separation bubbles, and therefore, the only inform ation available in the literature is experimental data. Roberts (2005) published displacement thickness inform ation fo r separation bubbles in flow s w ith streamwise Reynolds numbers ranging from 3.5 x 105 to 4.7 xlO5, and freestream turbulence intensities between 0.4% and 4.5%, along w ith the corresponding separation, transition inception, and reattachment locations. This experimental data, summarized in Table 4.…”

“…Roberts (2005) explains, researchers have classified boundary-layer transition into three modes; natural, bypass, and separation-bubble transition. The fo llo w in g is a summary o f the discussion provided byRoberts & Yaras (2005b), andRoberts (2005) on the three transition modes. The prim ary mode fo r attached flow s, where the ffeestream turbulence is low or moderate and the surface is smooth, is natural transition.…”