A flow-excited Helmholtz resonator was investigated experimentally and theoretically. The analysis was focused on a simplified momentum balance integrated over the region of the orifice. The resulting expressions were used to guide an experimental programme designed to obtain measurements of the resonator pressure under flow excitation, as well as the dynamics of the shear layer in the orifice using particle image velocimetry (PIV). The pressure measurements indicated a number of distinctive features as the flow speed varied. The PIV results provided a detailed representation of the shear layer vorticity field, as well as the equivalent hydrodynamic forcing of the resonator. The forcing magnitude was found to be roughly constant over a range of flow speeds. A model was proposed that provides a prediction of the resonator pressure fluctuations based on the thickness of the approach boundary layer, the free stream speed and the acoustic properties of the resonator. The model was shown to provide an accurate representation of the resonating frequency as well as the magnitude of the resonance to within a few decibels.

In this experimental study both smoke visualization and three-component hot-wire measurements have been performed in order to characterize the streamwise evolution of longitudinal counter-rotating vortices in a turbulent boundary layer. The vortices were generated by means of vortex generators (VGs) in different configurations. Both single pairs and arrays in a natural setting as well as in yaw have been considered. Moreover three different vortex blade heights h, with the spacing d and the distance to the neighbouring vortex pair D for the array configuration, were studied keeping the same d/h and D/h ratios. It is shown that the vortex core paths scale with h in the streamwise direction and with D and h in the spanwise and wall-normal directions, respectively. A new peculiar ‘hooklike’ vortex core motion, seen in the cross-flow plane, has been identified in the far region, starting around 200h and 50h for the pair and the array configuration, respectively. This behaviour is explained in the paper. Furthermore the experimental data indicate that the vortex paths asymptote to a prescribed location in the cross-flow plane, which first was stated as a hypothesis and later verified. This observation goes against previously reported numerical results based on inviscid theory. An account for the important viscous effects is taken in a pseudo-viscous vortex model which is able to capture the streamwise core evolution throughout the measurement region down to 450h. Finally, the effect of yawing is reported, and it is shown that spanwise-averaged quantities such as the shape factor and the circulation are hardly perceptible. However, the evolution of the vortex cores are different both between the pair and the array configuration and in the natural setting versus the case with yaw. From a general point of view the present paper reports on fundamental results concerning the vortex evolution in a fully developed turbulent boundary layer.

In this paper, we investigate the linear stability of oscillating zonal flows on the equatorial β-plane in the presence of fully three-dimensional disturbances. To exclude inflection point effects, we focus on the simplest case of a linear meridional shear with time-mean and oscillating components. For purely oscillatory background flows we find that in addition to resonant excitation of ‘additive’ type that occurs in the zonally invariant case, resonant excitation of ‘difference’ type is also possible. For flows with an oscillatory shear superimposed on an unstable time-mean shear it is shown that while the oscillatory shear has a stabilizing influence on disturbances with a small zonal wave number k, at higher k the effect of the oscillating shear diminishes and can even be destabilizing. Overall, a small oscillatory shear tends to reduce the fastest growth rate in the system and pushes the dominant k to higher values. Calculation of dominant zonal and vertical modes shows that the zonally asymmetric modes dominate a large portion of the parameter space, especially at high time-mean background shear and low oscillatory shear. As a result, the dominant vertical mode can have a somewhat larger vertical scale than in the zonally invariant case. At intermediate values of the time-mean shear the growth rate is relatively flat with respect to the zonal mode number, with maximum growth rate occurring in bands of high and low k. We have uncovered a rich assortment of vertical and zonal modes which are likely to play a role in the nonlinear evolution of equatorial flows.

We use a dynamical systems approach to identify coherent structures from often chaotic motions of inertial particles in open flows. We show that particle Lagrangian coherent structures (pLCS) act as boundaries between regions in which particles have different kinematics. They provide direct geometric information about the motion of ensembles of inertial particles, which is helpful to understand their transport. As an application, we apply the methodology to a planktonic predator–prey system in which moon jellyfish Aurelia aurita uses its body motion to generate a flow that transports small plankton such as copepods to its vicinity for feeding. With the flow field generated by the jellyfish measured experimentally and the dynamics of plankton described by a modified Maxey–Riley equation, we use the pLCS to identify a capture region in which prey can be captured by the jellyfish. The properties of the pLCS and the capture region enable analysis of the effect of several physiological and mechanical parameters on the predator–prey interaction, such as prey size, escape force, predator perception, etc. The methods developed here are equally applicable to multiphase and granular flows, and can be generalized to any other particle equation of motion, e.g. equations governing the motion of reacting particles or charged particles.

The internal flow over a backward-facing step in the transitional regime (ReD = 6000) was studied based on direct numerical simulations. The predictions were carried out with the help of a finite-volume Navier–Stokes solver equipped with a co-visualization facility which allows one to investigate the flow dynamics at high temporal resolution. First, grid-induced oscillations were precluded by a careful grid design. Second, the strong influence of the velocity profile approaching the step was studied and outlined. The main objective, however, was to provide a comprehensive insight into the dynamic flow behaviour, especially oscillations of the reattachment length of the primary recirculation region. The origin of this well-known flapping behaviour of the reattachment line is not yet completely understood. In the present work, the mechanisms leading to the oscillations of the reattachment length were extensively investigated by analysing the time-dependent flow. Besides the oscillations of the primary recirculation region, oscillations of the separation and the reattachment line of the secondary recirculation bubble at the upper channel wall were also observed. The results clearly show that in the present flow case the flapping of the primary reattachment and the secondary separation line is due to vortical structures in the unstable shear layers between the main flow and the recirculation bubbles. Vortices emerging in the shear layers and sweeping downstream convectively induce small zones of backward-flowing fluid at the channel walls while passing the recirculation regions. In the case of the primary recirculation region, the rotational movement of the shear-layer vortices impinging on the lower channel wall was found to cause zones of negative fluid velocity at the end of the recirculation bubble and thus flapping of the reattachment line. In contrast, in the case of the secondary recirculation region, the shear-layer vortices moved away from the upper channel wall so that their rotational movement did not reach the boundary. In this case, the pressure gradients originating from local pressure minima located in the shear-layer vortices were identified as being responsible for the oscillations of the separation line at the upper channel wall. While moving downstream with the shear-layer vortices, the pressure gradients were found to influence the top boundary of the channel and create alternating zones of forward- and backward-flowing fluid along the wall. All of these unsteady processes can best be seen from animations which are provided on the Journal of Fluid Mechanics website: journals.cambridge.org/FLM.

In this study, we revisit two simplified models of hovering motion for fruit fly and dragonfly from the perspective of force decomposition. The unsteady aerodynamics are analysed by examining the lift force and its four constituent components, each of which is directly related to a physical effect. These force components include one from the vorticity within the flow, one from the surface vorticity and two contributions credited to the motion of the insect wing. According to the phase difference in the models, a hovering motion can be classified into one of three types: symmetric, advanced and delayed rotations. The relative importance of the force components under various flow conditions are carefully analysed. It is shown that the symmetric rotation has the maximum vorticity lift (from volume and surface vorticity), but the optimal average lift is attained for an advanced rotation, which, compared to the symmetric rotation, increases the force contribution due to the unsteady surface motion at the expense of sacrificing contribution from the vorticity. By identifying the variations of the vorticity lift with flow characteristics, we may further explore the detailed mechanisms associated with the unsteady aerodynamics at different phases of hovering motion. For the different types of rotation, the insect wing shares the same mechanism of gaining lift when in the phase of driving with a fuller speed but exhibits different mechanisms at turning from one phase of motion to another. Moreover, we also examine the effects of the Reynolds number in an appropriate range and evaluate the performance of different wing profiles from symmetric to largely cambered.

We investigate the mixing of a stratified fluid of finite volume by a turbulent buoyant plume. We develop a model to describe the mixing and apply this to both the cases of a two-layer stratification and a continuous stratification. With a two-layer stratification, the plume intrudes at the interface where it supplies an intermediate layer of fluid. This new layer gradually deepens, primarily mixing the original near-source layer of fluid through entrainment. Eventually, this intermediate layer becomes sufficiently buoyant that the plume penetrates into the more distal layer, leaving a partially mixed region between the original layers of fluid. Analysis of new experiments shows that the growth of the intermediate layer depends primarily on the ratio λ of (i) the filling box time, during which the plume entrains a volume of fluid equal to that in the near-source layer, and (ii) the time for the buoyancy of the near-source layer to increase to that of the more distal layer. For small values of λ, the near-source layer becomes approximately well mixed, and the penetration time of the plume scales with the buoyancy evolution time of the near-source layer. In the limit λ ~ O(1), however, the plume penetrates through into the distal layer long before the near-source layer becomes well mixed; instead, at the time of penetration, the plume leaves an intermediate partially mixed zone between the two original layers. We develop a new phenomenological model to account for the mixing in this intermediate layer based on the effective turbulent diffusion associated with the kinetic energy in the plume and compare this with the model for penetrative entrainment proposed by Kumagai (J. Fluid Mech., vol. 147, 1984, p. 105). In comparison with the experimental data, the models provide a reasonably accurate prediction of the plume penetration time, while the diffusive mixing model provides a somewhat more accurate description of the evolution of the density profile for a range 0 < λ < 1. The diffusive mixing model also leads to predictions which are consistent with some new experimental data for the case in which a plume mixes a continuously stratified layer. In particular, the model is able to predict the initial transient mixing of the region between the source and the height at which the plume intrudes laterally in the ambient fluid, thereby providing an advance on the late-time mixing model of Cardoso and Woods (J. Fluid Mech., vol. 250, 1993, p. 277). We consider the implications of these results on the turbulent penetrative entrainment associated with buoyant plumes.

The interaction between free-stream disturbances and the boundary layer on a body with a rounded leading edge is considered in this paper. A method which incorporates calculations using the parabolized stability equation in the Orr–Sommerfeld region, along with an upstream boundary condition derived from asymptotic theory in the vicinity of the leading edge, is generalized to bodies with an inviscid slip velocity which tends to a constant far downstream. We present results for the position of the lower branch neutral stability point and the magnitude of the unstable Tollmien–Schlichting (T-S) mode at this point for both a parabolic body and the Rankine body. For the Rankine body, which has an adverse pressure gradient along its surface far from the nose, we find a double maximum in the T-S wave amplitude for sufficiently large Reynolds numbers.

Three-dimensional flows over impulsively translated low-aspect-ratio flat plates are investigated for Reynolds numbers of 300 and 500, with a focus on the unsteady vortex dynamics at post-stall angles of attack. Numerical simulations, validated by an oil tow-tank experiment, are performed to study the influence of aspect ratio, angle of attack and planform geometry on the wake vortices and the resulting forces on the plate. Immediately following the impulsive start, the separated flows create wake vortices that share the same topology for all aspect ratios. At large time, the tip vortices significantly influence the vortex dynamics and the corresponding forces on the wings. Depending on the aspect ratio, angle of attack and Reynolds number, the flow at large time reaches a stable steady state, a periodic cycle or aperiodic shedding. For cases of high angles of attack, an asymmetric wake develops in the spanwise direction at large time. The present results are compared to higher Reynolds number flows. Some non-rectangular planforms are also considered to examine the difference in the wakes and forces. After the impulsive start, the time at which maximum lift occurs is fairly constant for a wide range of flow conditions during the initial transient. Due to the influence of the tip vortices, the three-dimensional dynamics of the wake vortices are found to be quite different from the two-dimensional von Kármán vortex street in terms of stability and shedding frequency.

The effect is examined on infinitesimal standing waves over a plane beach when restricted by the arbitrary placing of a finite rigid (or permeable) lid of length ℓ on the undisturbed surface. A uniformly bounded solution for the potential function is obtained by a Green's function method. The Green's function is derived and manipulated, for subsequent computational expedience, from a previously known solution for the problem of an oscillating line source placed at an arbitrary location in the sector. Applications are made to both the case of plate anchored at the origin and the case of plate anchored some distance at sea (drifted plate problem). In both cases water column potentials and equipotentials are constructed from the numerical solution of a Fredholm equation of the second kind by finite difference discretization. Solutions are further extended to include the logarithmically singular standing wave, combination with which allows the construction of progressing waves. Computation of initially incoming progressing wave envelopes demonstrates the emergence of a partially standing wave pattern shoreward of the plate. There is no difficulty, in principle, to extend the theory to any number of plates, and this is verified by computation for the case of two plates. A new shoreline radiation condition is constructed to allow formulation, in the usual way, of the reflection/transmission problem for the plate, and results are in good qualitative agreement with a similar model on a horizontal plane bed. It is argued that the Green's function constructed here could be used in a number of diverse problems, of this linear nature, where all, or part, of the submerged boundary is that of a plane incline.

The absolute and convective instability properties of plane mixing layers are investigated for linearized inviscid disturbances. It is shown that confinement by plates parallel to the flow can enhance the absolute instability so much that even a co-flow plane mixing layer becomes absolutely unstable when the ratio of distances of the plates from the mixing layer lies in a certain range. Even when the plates are placed equidistantly from the mixing layer, a co-flow mixing layer can become absolutely unstable if the velocity profile has an asymmetry about its mid-plane. ‘Semiconfinement’, where a plate is only added to one side of the mixing layer, is also investigated. A substantial destabilization is possible when the plate is added on the side of the faster stream. Previous investigations seem only to have found absolute instability when the streams flow in opposite directions.

The constraints necessary for equilibrium solutions of the boundary layer equations are explored for turbulent boundary layers subject to lateral convergence and divergence and with longitudinal pressure gradients. It is shown that in addition to the well-known equilibrium solutions for two-dimensional boundary layers there are additional possible equilibrium states for boundary layers with these extra rates-of-strain acting. The necessary constraints for equilibrium are derived and discussed.

A low-dimensional Galerkin model is proposed for the flow around a high-lift configuration, describing natural vortex shedding, the high-frequency actuated flow with increased lift and transients between both states. The form of the dynamical system has been derived from a generalized mean-field consideration. Steady state and transient URANS (unsteady Reynolds-averaged Navier–Stokes) simulation data are employed to derive the expansion modes and to calibrate the system parameters. The model identifies the mean field as the mediator between the high-frequency actuation and the low-frequency natural shedding instability.

We present analytical solutions for the initial rise height zm of two-dimensional turbulent fountains issuing from a horizontal linear source of width 2b0 into a quiescent environment of uniform density. Using the initial rise height prediction as a measure we classify line fountains into three types depending on their source conditions. For source Froude numbers Fr0 ≫ 1, the near-source flow of the ‘forced’ fountain is dominated by source momentum flux and behaves like a jet; the asymptotic solution to the fountain equations yields, in agreement with previous studies, zm/b0 ~ Fr04/3. For Fr0 = O(1) the fountain is ‘weak’ and fluid is projected vertically to a height that is consistent with an energy-conserving flow – the sensitivity of the rise height with Fr0 increases as zm/b0 ~ Fr02. For Fr0 ≪ 1, the fountain is ‘very weak’ and we find that zm/b0 ~ Fr02/3. As the local value of the Froude number decreases with height, all three forms of fountain behaviour identified are expected above a highly forced source and we provide scalings for the three lengths that contribute to the total rise height. Comparisons between our predicted rise heights and the previous experimental results show good agreement across a wide range of Fr0. The collated data highlights that experiments have focused in the majority on fountains above sources with intermediate Fr0. Notably there is a lack of measurements on very weak line fountains and of independent experimental confirmation of the initial rise heights across the range of Fr0.

We report the results of physical experiments that demonstrate the strong influence of the thermal conductivity of the substrate on the evaporation of a pinned droplet. We show that this behaviour can be captured by a mathematical model including the variation of the saturation concentration with temperature, and hence coupling the problems for the vapour concentration in the atmosphere and the temperature in the liquid and the substrate. Furthermore, we show that including two ad hoc improvements to the model, namely a Newton's law of cooling on the unwetted surface of the substrate and the buoyancy of water vapour in the atmosphere, give excellent quantitative agreement for all of the combinations of liquid and substrate considered.

Direct numerical simulations (DNS) of two inviscid flows, the Taylor–Green flow and two orthogonal interacting Lamb dipoles, together with the DNS of forced isotropic turbulence, were performed to generate data for a comparative study. The isotropic turbulent field was considered after the transient and, in particular, when the velocity derivative skewness oscillates around −0.5. At this time, Rλ ≈ 257 and a one decade wide k−5/3 range was present in the energy spectrum. For the inviscid flows the fields were considered when a wide k−3 range was achieved. This power law spectral decay corresponds to infinite enstrophy and is considered one of the requirements to demonstrate that the Euler equations lead to a finite time singularity (FTS). Flow visualizations and statistics of the strain rate tensor and vorticity components in the principal axes of the strain rate tensor (λ, λ) were used to classify structures. The key role of the intermediate component 2 is demonstrated by its good correlation with enstrophy production. Filtering of the fields shows that the slope of the power law is directly connected to self-similar structures, whose radius of curvature is smaller the steeper the spectrum.

We investigate the instability and nonlinear saturation of temperature-stratified Taylor–Couette flows in a finite height cylindrical gap and calculate angular momentum transport in the nonlinear regime. The model is based on an incompressible fluid in Boussinesq approximation with a positive axial temperature gradient applied. While both ingredients, the differential rotation as well as the stratification due to the temperature gradient, are stable themselves, together the system becomes subject of the stratorotational instability and a non-axisymmetric flow pattern evolves. This flow configuration transports angular momentum outwards and will therefore be relevant for astrophysical applications. The belonging coefficient of β viscosity is of the order of unity if the results are adapted to the size of an accretion disk. The strength of the stratification, the fluid's Prandtl number and the boundary conditions applied in the simulations are well suited too for a laboratory experiment using water and a small temperature gradient around 5 K. With such a set-up the stratorotational instability and its angular momentum transport could be measured in an experiment.

We evaluate in this work the hydrodynamic transport coefficients of a granular binary mixture in d dimensions. In order to eliminate the observed disagreement (for strong dissipation) between computer simulations and previously calculated theoretical transport coefficients for a monocomponent gas, we obtain explicit expressions of the seven Navier–Stokes transport coefficients by the use of a new Sonine approach in the Chapman–Enskog (CE) theory. This new approach consists of replacing, where appropriate in the CE procedure, the Maxwell–Boltzmann distribution weight function (used in the standard first Sonine approximation) by the homogeneous cooling state distribution for each species. The rationale for doing this lies in the well-known fact that the non-Maxwellian contributions to the distribution function of the granular mixture are more important in the range of strong dissipation we are interested in. The form of the transport coefficients is quite common in both standard and modified Sonine approximations, the distinction appearing in the explicit form of the different collision frequencies associated with the transport coefficients. Additionally, we numerically solve by the direct simulation Monte Carlo method the inelastic Boltzmann equation to get the diffusion and the shear viscosity coefficients for two and three dimensions. As in the case of a monocomponent gas, the modified Sonine approximation improves the estimates of the standard one, showing again the reliability of this method at strong values of dissipation.