Flow field analysis of a turbulent boundary layer over a riblet surface

The near-wall flow structures of a turbulent boundary layer over a riblet surface with semi-circular grooves were investigated experimentally for the cases of drag decreasing (s ⁺=25.2) and drag increasing (s ⁺=40.6). One thousand instantaneous velocity fields over riblets were measured using the velocity field measurement technique and compared with those above a smooth flat plate. The field of view was 6.75 × 6.75 mm² in physical dimension, containing two grooves. Those instantaneous velocity fields were ensemble averaged to get turbulent statistics including turbulent intensities and turbulent kinetic energy. To see the global flow structure qualitatively, flow visualization was also carried out using the synchronized smoke-wire technique under the same experimental conditions. For the case of drag decreasing (s ⁺=25.2), most of the streamwise vortices stay above the riblets, interacting with the riblet tips frequently. The riblet tips impede the spanwise movement of the streamwise vortices and induce secondary vortices. The normalized rms velocity fluctuations and turbulent kinetic energy are small near the riblet surface, compared with those over a smooth flat plate. Inside the riblet valleys, these are sufficiently small that the increased wetted surface area of the riblets can be compensated. In addition, in the outer region (y ⁺ > 30), these values are almost equal to or slightly smaller than those for the smooth plate. For the case of drag increasing (s ⁺=40.6), however, most of the streamwise vortices stay inside the riblet valleys and contact directly with the riblet surface. The high-speed down-wash flow penetrating into the riblet valley interacts actively with the wetted riblet surface and increases the skin friction. The rms velocity fluctuations and turbulent kinetic energy have larger values compared with those over a smooth flat plate. Received: 24 March 1999/Accepted: 10 March 2000     

Drag measurements on V-grooved surfaces on a body of revolution in axial flow

In water flows with velocities of up to 9 m/s the friction drag of a body of revolution in axial flow was investigated for dependence on the body surface structure. This was done for different types of riblet film fixed on the surface with the riblet direction aligned with the flow. The lateral spacing between the triangular shaped riblets varied between 0.033 mm and 0.152 mm. In all cases the riblet spacing was equal to the riblet height. For comparison a smooth reference film was used.Depending on the Reynolds number and the non-dimensional riblet spacings ⁺, a turbulent drag reduction of up to 9% could be achieved with riblets in comparison with the flow over a smooth surface.In the region of transition to turbulent flow and with non-dimensional riblet spacings ofs ⁺10–15 drag reductions of up to 13% were obtained. It is therefore conjectured, that in addition to hampering the near wall momentum exchange, the riblets can delay the development of initial turbulent structures in time and space.     

Large-scale spanwise periodicity in a turbulent boundary layer induced by highly ordered and directional surface roughness

The effect of converging–diverging riblet-type surface roughness (riblets arranged in a ‘herringbone’ pattern) are investigated experimentally in a zero pressure gradient turbulent boundary layer. For this initial parametric investigation three different parameters of the surface roughness are analysed in detail; the converging–diverging riblet yaw angle α, the streamwise fetch or development length over the rough surface F_x and the viscous-scaled riblet height h⁺. It is observed that this highly directional surface roughness pattern induces a large-scale spanwise periodicity onto the boundary layer, resulting in a pronounced spanwise modification of the boundary layer thickness. Hot-wire measurements reveal that above the diverging region, the local mean velocity increases while the turbulent intensity decreases, resulting in a thinner overall boundary layer thickness in these locations. The opposite situation occurs over the converging region, where the local mean velocity is decreased and the turbulent intensity increases, producing a locally thicker boundary layer. Increasing the converging–diverging angle or the viscous-scaled riblet height results in stronger spanwise perturbations. For the strongest convergent–divergent angle, the spanwise variation of the boundary layer thickness between the diverging and converging region is almost a factor of two. Such a large variation is remarkable considering that the riblet height is only 1% of the unperturbed boundary layer thickness. Increasing the fetch seems to cause the perturbations to grow further from the surface, while the overall strength of the induced high and low speed regions remain relatively unaltered. Further analysis of the pre-multiplied energy spectra suggests that the surface roughness has modified or redistributed the largest scale energetic structures.     

Experiments with three-dimensional riblets as an idealized model of shark skin

The skin of fast sharks exhibits a rather intriguing three-dimensional rib pattern. Therefore, the question arises whether or not such three-dimensional riblet surfaces may produce an equivalent or even higher drag reduction than straight two-dimensional riblets. Previously, the latter have been shown to reduce turbulent wall shear stress by up to 10%. Hence, the drag reduction by three-dimensional riblet surfaces is investigated experimentally. Our idealized 3D-surface consists of sharp-edged fin-shaped elements arranged in an interlocking array. The turbulent wall shear stress on this surface is measured using direct force balances. In a first attempt, wind tunnel experiments with about 365,000 tiny fin elements per test surface have been carried out. Due to the complexity of the surface manufacturing process, a comprehensive parametric study was not possible. These initial wind tunnel data, however, hinted at an appreciable drag reduction. Subsequently, in order to have a better judgement on the potential of these 3D-surfaces, oil channel experiments are carried out. In our new oil channel, the geometrical dimensions of the fins can be magnified 10 times in size as compared to the initial wind tunnel experiments, i.e., from typically 0.5 mm to 5 mm. For these latter oil channel experiments, novel test plates with variable fin configuration have been manufactured, with 1,920–4,000 fins. This enhanced variability permits measurements with a comparatively large parameter range. As a result of our measurements, it can be concluded, that 3D-riblet surfaces do indeed produce an appreciable drag reduction. We found as much as 7.3% decreased turbulent shear stress, as compared to a smooth reference plate. However, in direct comparison with 2D riblets, the performance of 3D-riblets is still inferior by about 1.7%. On the other hand, it appears conceivable, with a careful design of the fin shape (possibly supported by theory), that this inferiority in performance might be reduced. Nevertheless, at present, it seems to be rather unlikely, that 3D-riblets can significantly outperform 2D-riblets. Finally, one interesting finding remains to be mentioned: The optimum drag reduction for short 3D-riblets occurs at a lower rib height than for longer 3D-riblets or for infinitely long 2D-riblets. The same observation had been made previously on shark scales of different species with differing rib lengths, but no explanation could be given. Received: 1 March 1999/Accepted: 16 July 1999     

Selective enhancement of oblique waves caused by finite amplitude second mode in supersonic boundary layer

Nonlinear interactions of the two-dimensional(2D) second mode with oblique modes are studied numerically in a Mach 6.0 flat-plate boundary layer, focusing on its selective enhancement effect on amplification of different oblique waves. Evolution of oblique modes with various frequencies and spanwise wavenumbers in the presence of 2D second mode is simulated successively, using a modified parabolized stability equation(PSE) method, which is able to simulate interaction of two modes with different frequencies efficiently. Numerical results show that oblique modes in a broad band of frequencies and spanwise wavenumbers can be enhanced by the finite amplitude 2D second mode instability wave. The enhancement effect is accomplished by interaction of the 2D second mode, the oblique mode, and a forced mode with difference frequency. Two types of oblique modes are found to be more amplified, i.e., oblique modes with frequency close to that of the 2D second mode and low-frequency first mode oblique waves. Each of them may correspond to one type of transition routes found in transition experiments. The spanwise wavenumber of the oblique wave preferred by the nonlinear interaction is also determined by numerical simulations.     

On near-wall turbulence-generating events in a turbulent boundary layer on a riblet surface

The boundary layer over a drag reducing riblet surface is investigated using hot-wire anemometry and flow visualisation. The concept of a riblet sublayer is introduced, and a definition is proposed in terms of a region of reduced turbulence energy production formed near the wall by the addition of riblets. The hot wire records are examined using a modified form of quadrant analysis, and results obtained over plain and riblet surfaces are compared. Close to the wall, the addition of riblets produces a marked reduction in the occurrence of ejection (2nd quadrant) events. A corresponding increase in the incidence of sweep (4th quadrant) events is accompanied by the development of a strong tendency toward a preferred event duration, and a preferred interval between events. These changes diminish rapidly with distance from the surface, becoming almost undetectable beyondy ⁺=40. They are discussed in the light of flow visualisation results, and interpreted in terms of mechanisms associated with the interaction between the riblets and the inner boundary layer flow structures. A conceptual model of the flow mechanisms in the riblet sublayer is proposed.     

The effect of riblets on laminar to turbulent transition

Experiments conducted on the effect of riblets on the laminar-to-turbulent transition of a flat plate in a water tunnel are reported. Transition was determined using a Laser Doppler Velocimeter (LDV). A smooth reference surface was compared to five riblet surfaces for a range of Reynolds numbers. Smooth surface transition Reynolds number was about 2.75 × 10⁶. All of the five tested riblet surfaces had lower transition Reynolds numbers. A critical roughness Reynolds number of about 6 was determined for one of the riblet surfaces. This is much lower than the generally accepted value of 25, considered safe for distributed roughness.     

Flow over convergent and divergent wall riblets

Fast swimming sharks have small riblets on their skin, which are assumed to improve the swimming performance of the fish. Fluid dynamic experiments in water as well as in air confirm this assumption. With riblet surfaces as compared to smooth surfaces, drag reductions up to about 10% were measured. The overall riblet pattern on sharks shows parallel riblets directed from head to tail, but besides this overall pattern fast swimming sharks have also small areas with converging riblets and others with diverging riblets. In the present study the velocity field over convergent and divergent riblet patterns is investigated by hot-wire measurements in turbulent pipe flow. Significant changes in the near wall velocity field were found.     

Coherent structures and riblets

This work deals with the effect of the riblets on the coherent structures near the wall. The emphasis is put on the genesis of the quasi-streamwise vortices in the presence of the riblets. The quasi-streamwise vortices regenerate by the tilting of wall normal vorticity induced by prevailing structures. This requires a mechanism which leads to a temporal streamwise dependence near the elongated flow structures and to a subsequent formation of new wall normal vorticity. It is suggested here that the action of existing quasi-streamwise vortices on the sidewalls of wall normal vorticity may create a local, streamwise dependent spanwise velocity and therefore, a secondary wall normal vorticity field. A preliminary analysis of the set-up and the time and space development of this secondary three-dimensional flow associated with the regeneration mechanism, is given. An attempt is made, in order to explain the drag reduction performed by the riblets through an intermittent model, based on the protrusion height. Logical estimates of the amount of drag reduction are obtained. The differences between the mechanism suggested here and those based on forced control experiments are also discussed.     

Riblet drag reduction in mild adverse pressure gradients: A numerical investigation

Riblet films are a passive method of turbulent boundary layer control that can reduce viscous drag. They have been studied with great detail for over 30 years. Although common riblet applications include flows with Adverse Pressure Gradients (APG), nearly all research thus far has been performed in channel flows. Recent research has provided motivation to study riblets in more complicated turbulent flows with claims that riblet drag reduction can double in mild APG common to airfoils at moderate angles of attack. Therefore, in this study, we compare drag reduction by scalloped riblet films between riblets in a zero pressure gradient and those in a mild APG using high-resolution large eddy simulations. In order to gain a fundamental understanding of the relationship between drag reduction and pressure gradient, we simulated several different riblet sizes that encompassed a broad range of s⁺ (riblet width in wall units), similarly to many previously published experimental studies. We found that there was only a slight improvement in drag reduction for riblets in the mild APG. We also observed that peak values of streamwise turbulence intensity, turbulent kinetic energy, and streamwise vorticity scale with riblet width. Primary Reynolds shear stresses and turbulence kinetic energy production however scale with the ability of the riblet to reduce skin-friction.     

Modelling the time-dependent flow over riblets in the viscous wall region

The flow over drag reducing riblets is examined computationally using a time-dependent model of the viscous wall region. The flow at the upper bound of the computational domain (y ⁺?40) is described using a streamwise eddy model consisting of two scales. A control-volume finite-element method utilizing triangular meshes is used to exactly fit the riblet cross-sectional geometry. Observations of the transient flow conditions suggest that the riblets limit the lateral spread of fluid inrushes towards the wall and retain low momentum fluid in the riblet valleys effectively isolating much of the wall from such inrushes. The generation of intermittent secondary vortices within the riblet valleys also occurs; however, these appear to be quite weak and fairly short-lived.     

Drag characteristics of longitudinal and transverse riblets at low dimensionless spacings

A surface grooved with microscopic riblets aligned parallel to the flow is an effective means to reduce the turbulent skin friction up to 10% compared to a smooth surface. The maximum drag reduction is found for a dimensionless rib spacing s ⁺ in the range of 15–17. For s ⁺ < 10, a linear behaviour of the drag reduction curve is predicted by viscous theory. This linear slope of the drag reduction curve is in contradiction to Schlichting’s postulation of a hydraulically smooth behaviour of small-scale roughness in a turbulent flow. This regime of evanescent dimensionless rib spacings is investigated experimentally by direct wall shear stress measurements in a fully developed channel flow. Additionally, a numerical calculation of the viscous flow over riblets was carried out to predict the drag reducing behaviour. The experimental results show a linear drag reducing behaviour down to s ⁺ = 0.3, which is in good agreement with the numerical results of the viscous simulation. The postulation of Schlichting’s hydraulically smooth regime of a rough surface was not confirmed, neither for a riblet surface nor for a surface geometry with grooves oriented perpendicular to the flow. In the latter case, the drag increases as a quadratic function of the roughness height.     

Riblet flow calculation with a low Reynolds number κ - ε model

A low Reynolds number κ - ε model has been used to calculate the turbulent boundary layer over riblets. Calculated mean velocity, Reynolds shear stress and kinetic energy distributions are generally in good agreement with available experimental data. The comparison between these distributions and those in a corner flow points to a significant difference between the two flows and the unlikelihood of counter-rotating vortices within the riblet grooves. One shortcoming of the present κ - ε model is the relatively slow return to a two-dimensional turbulence state as the distance from the riblet surface increases.     

一维与二维爆轰传播的时空关联特性数值研究

一维爆轰传播的理论完备、计算准确, 二维斜爆轰传播由于壁面与黏性效应, 大尺度、高精度预测还有一定难度. 利用Euler方程和H₂-Air基元反应模型, 对二维有限长楔面诱导的斜爆轰和活塞驱动一维非定常正爆轰进行计算比较研究, 从时空两个维度方面, 分析了两者在起爆过程、稀疏波传播、爆轰波面演化中的关联特性. 研究结果表明: 在过驱动度相同的条件下, 经过时空变换的活塞驱动一维爆轰传播与二维驻定斜爆轰在起爆区波系结构、波面演化特征和主要参数分布规律方面无论定性或者定量对比均符合较好, 所以, 一维非定常爆轰和二维驻定斜爆轰具有时空相关性. 两者的差异主要体现在过驱动斜爆轰受稀疏波影响过渡到近Chapman-Jouguet (C-J)爆轰状态所需的弛豫时间不同, 原因可能是起源于活塞和壁面稀疏波强度的差异. 本文提出的一维与二维爆轰传播的时空关联方法不仅有助于认知斜爆轰起爆、过驱爆轰产生、胞格爆轰演化的三阶段规律, 还可以对比揭示壁面、边界层和黏性效应的影响, 应用在斜爆轰发动机燃烧室设计中能够有效节约计算时间和成本, 并降低复杂度.      

Modelling the time dependent flow over riblets in the viscous wall region

The flow over riblets is examined computationally using a time dependent model of the viscous wall region. This 2 1/2 D model, developed by Hatziavramidis and Hanratty (1979) and modified by Nikolaides (1984) and Chapman and Kuhn (1981, 1986) assumes homogeneity in the streamwise direction so that the flow is solved only in the cross-sectional plane. The flow at the upper boundary of the computational domain (y ⁺ 40) is described using a streamwise eddy model consisting of two scales, one of the streak spacing (⁺ 100), which dominates vertical momentum transport, and a larger scale that accounts for the influence of large outer flow eddies.The protrusion height concept (Bechert and Bartenwerfer, 1989) is used to define ay ⁺=0 location for surfaces with riblets. A control volume finite element method utilizing triangular meshes is used to exactly fit the riblet cross-sectional geometry. Results obtained using fairly large riblets compare well with the limited experimental evidence available. Observations of the transient flow suggest that the riblets interact with the near-wall streamwise vortices, weakening them by the generation of intermittent secondary vortices within the riblet valleys. The riblets also appear to limit the lateral spread of inrushes towards the wall and retain low momentum fluid in the riblet valleys effectively isolating much of the wall from such inrushes.     

FLOW STRUCTURE IN THE TURBULENT BOUNDARY LAYER OVER DIRECTIONAL RIBLETS SURFACES

本文采用时间解析的二维粒子图像测速技术,对零压力梯度光滑以及汇聚和发散沟槽表面平板湍流边界层统计特性和流动结构进行了研究.结果表明在垂直于汇聚和发散沟槽表面的对称平面内,相对于光滑壁面,发散沟槽壁面使当地边界层厚度、壁面摩擦阻力、湍流脉动、雷诺应力等明显减小;而汇聚沟槽壁面对湍流边界层特性和流动结构的影响正好相反,汇聚沟槽使壁面流体有远离壁面向上运动的趋势,因而导致边界层厚度增加了约43%;同时,在汇聚沟槽表面情况下流向大尺度相干结构更容易形成,这对减阻是不利的.此外,顺向涡数量在湍流边界层的对数区均存在一个极大值,发散沟槽表面所对应的极大值位置更靠近沟槽壁面,而在汇聚沟槽表面则有远离壁面的趋势,由顺向涡诱导产生的较强的喷射和扫掠运动会在湍流边界层中产生较强的剪切作用,顺向涡数量的减少是发散沟槽壁面当地摩擦阻力降低的主要原因.     

Influence of wavy surfaces on coherent structures in a turbulent flow

We describe how outer flow turbulence phenomena depend on the interaction with the wall. We investigate coherent structures in turbulent flows over different wavy surfaces and specify the influence of the different surface geometries on the coherent structures. The most important contribution to the turbulent momentum transport is attributed to these structures, therefore this flow configuration is of large engineering interest. In order to achieve a homogeneous and inhomogeneous reference flow situation two different types of surface geometries are considered: (1) three sinusoidal bottom wall profiles with different amplitude-to-wavelength ratios of α = 2a/Λ = 0.2 (Λ = 30 mm), α = 0.2 (Λ = 15 mm), and α = 0.1 (Λ = 30 mm); and (2) a profile consisting of two superimposed sinusoidal waves with α = 0.1 (Λ = 30 mm). Measurements are carried out in a wide water channel facility (aspect ratio 12:1). Digital particle image velocimetry (PIV) is performed to examine the spatial variation of the streamwise, spanwise and wall-normal velocity components in three measurement planes. Measurements are performed at a Reynolds number of 11,200, defined with the half channel height h and the bulk velocity U _B. We apply the method of snapshots and perform a proper orthogonal decomposition (POD) of the streamwise, spanwise, and wall-normal velocity components to extract the most dominant flow structures. The structure of the most dominant eigenmode is related to counter-rotating, streamwise-oriented vortices. A qualitative comparison of the eigenfunctions for different sinusoidal wall profiles shows similar structures and comparable characteristic spanwise scales Λ _z = 1.5 H in the spanwise direction for each mode. The scale is observed to be slightly smaller for α = 0.2 (Λ = 15 mm) and slightly larger for α = 0.2 (Λ = 30 mm). This scaling for the flow over the basic wave geometries indicates that the size of the largest structures is neither directly linked to the solid wave amplitude, nor to the wavelength. The characteristic spanwise scale of the dominant eigenmode for the developed flow over the surface consisting of two superimposed waves reduces to 0.85 H. However, a scale in the order of 1.3 H is identified for the second mode. The eigenvalue spectra for the superimposed waves is much broader, more modes contribute to the energy-containing range. The turbulent flow with increased complexity of the bottom surface is characterized by an increased number of dominant large-scale structures with different spanwise scales.     

Numerical and experimental investigation of the laminar boundary layer over riblets

One of the main aims of this work is to show to what extent drag reduction in a turbulent boundary layer can be ascribed to a purely viscous effect. A numerical and experimental study is performed in a laminar boundary layer over triangular riblets. The 2-D parabolic equations of motion are integrated using an x marching method and the discretised system is solved with the MSI algorithm. The influence of the riblet geometrical parameters and of the number of grid points is studied. Measurements are carried out in a water tunnel with forward scatter and backscatter laser-Doppler velocimetry extending within the riblets. The longitudinal velocity component measurements and computations are practically identical. Numerical results presented herein show that a slight drag reduction is obtained for s/h=1.2. It appears that, as far as friction is concerned, the wetted area is not the surface to be considered. Thus, the boundary layer over riblets would behave like a boundary layer on an equivalent smooth plate located beneath the crest plane. The numerical study in terms of the riblet height h shows best results are for h tending to zero, with the ratio s/h being equal to 1.2.     

Active cancellation of artificially introduced Tollmien–Schlichting waves using plasma actuators

In the present work artificially excited Tollmien–Schlichting (TS) waves were cancelled using plasma actuators operated in pulsed mode. In order to achieve this a vibrating surface driven by an electromagnetic turbulator was flush mounted in a flat plate to excite the TS waves. These were amplified by an adverse pressure gradient induced by an insert on the upper wall of the test section. A control plasma actuator positioned downstream of the excitation actuator attenuates the waves by imparting an unsteady force into the boundary layer to counteract the oscillation. As a result the amplitude of the velocity fluctuations at the excitation frequency is reduced significantly depending on the distance from the wall. A parameter study was performed to identify the influence of several operation parameters of the control actuator.     

Riblet modelling using a second-moment closure

A low Reynolds number second-moment closure has been used to calculate a turbulent boundary layer which develops over a riblet surface with zero pressure gradient. The calculated mean velocity distributions compare favourably with measurements. Calculated Reynolds stresses away from the riblet surface region are also in agreement with measurements. In the vicinity of the riblets, the model reflects the increased anisotropy of the Reynolds stress tensor inadequately. Possible reasons for this shortcoming are discussed and suggestions for improving the model are made.