Global aerodynamic instability of twin cylinders in cross flow

This paper comprises an in-depth physical discussion of the flow-induced vibration of two circular cylinders in view of the time-mean lift force on stationary cylinders and interaction mechanisms. The gap-spacing ratio T/D is varied from 0.1 to 5 and the attack angle α from 0° to 180° where T is the gap width between the cylinders and D is the diameter of a cylinder. Mechanisms of interaction between two cylinders are discussed based on time-mean lift, fluctuating lift, flow structures and flow-induced responses. The whole regime is classified into seven interaction regimes, i.e., no interaction regime; boundary layer and cylinder interaction regime; shear-layer/wake and cylinder interaction regime; shear-layer and shear-layer interaction regime; vortex and cylinder interaction regime; vortex and shear-layer interaction regime; and vortex and vortex interaction regime. Though a single non-interfering circular cylinder does not correspond to a galloping following quasi-steady galloping theory, two circular cylinders experience violent galloping vibration due to shear-layer/wake and cylinder interaction as well as boundary layer and cylinder interaction. A larger magnitude of fluctuating lift communicates to a larger amplitude vortex excitation.     

Features of flow-induced forces deduced from wavelet analysis

In the present study, the effect of Reynolds number (Re) on flow interference between two side-by-side stationary cylinders and the associated flow-induced forces are investigated using finite element method and wavelet analysis. The pitch ratio chosen is T/D=1.7, where T is the separation distance measured between cylinder centers and D is the diameter, and Re, based on the free-stream velocity and the diameter of the cylinder, is varied within the laminar flow regime, i.e., 60<Re<200. The method of continuous wavelet transform is used to analyze time-variant features of flow-induced forces in the time–frequency domain. Flow patterns in the form of vorticity plots are presented to demonstrate the underlying physics. It is found that flow interference initially occurs in the inner vortices shed from the two cylinders, and extends to the outer vortices with increasing Re. The flow behind two cylinders undergoes three regimes: Regime I—unbiased gap flow, Regime II—stable biased gap flow, and Regime III—unstable gap flow. Flow-induced forces show significant variations when the flow transits from one regime to another. In particular, during the transition from Regimes II to III, the forces not only increase by amplitude, but also change their nature from deterministic to random, and show some nonstationary features. This is shown to be caused by the amalgamation of inner and outer vortices behind the two cylinders when the flow interference extends from inner vortices to outer vortices. Whenever possible, the present results are compared with experimental measurements and theoretical predictions. The numerical simulations are consistent with these other results.     

The aerodynamics of a cylinder submerged in the wake of another

Flow-induced fluctuating lift (C_Lf) and drag (C_Df) forces and Strouhal numbers (St) of a cylinder submerged in the wake of another cylinder are investigated experimentally for Reynolds number (Re)=9.7×10³–6.5×10⁴. The spacing ratio L^⁎ (=L/D) between the cylinders is varied from 1.1 to 4.5, where L is the spacing between the cylinders and D is the cylinder diameter. The results show that C_Lf, C_Df and St are highly sensitive to Re due to change in the inherent nature of the flow structure. How the flow structure is dependent on Re and L^⁎ is presented in a flow structure map. Zdravkovich and Pridden (1977) observed a ‘kink’ in time-mean drag distribution at L^⁎≈2.5 for Re>3.1×10⁴, but not for Re≤3.1×10⁴. The physics is provided here behind the presence and absence of the ‘kink’ that was left unexplained since then.     

Wakes of elliptical cylinders at low Reynolds number

The wakes of elliptical cylinders are numerically investigated at a Reynolds number Re_D = 150. ANSYS-Fluent, based on the finite volume method, is used to simulate two-dimensional Newtonian fluid flow. The cylinder cross-sectional aspect ratio (AR) is varied from 0.25 to 1.0 (circular cylinder), and the angle of attack (α) of the cylinder is changed as α = 0° – 90°. With the changes in AR and α, three distinct wake patterns (patterns I, II, III) are observed, associated with different characteristics of fluid forces. Steady wake (pattern I) is characterised by two steady bubbles forming behind the cylinder, occurring at AR < 0.37 and α < 2.5°. Time-mean drag and fluctuating lift coefficients are small. Pattern II refers to Karman wake followed by steady wake (AR ≥ 0.37 – 0.67, depending on α) with the Karman street transitioning to two steady shear layers downstream. An inflection angle α_i is identified where the time-mean drag of the elliptical cylinder is identical to that of a circular cylinder. Pattern III is the Karman wake followed by secondary wake (AR ≤ 0.67, α > 52°), where the Karman street forming behind the cylinder is modified to a secondary vortex street with a low frequency. The Time-mean drag coefficient is maximum for this pattern.     

Dependence of flow classification on the Reynolds number for a two-cylinder wake

This work aims to investigate the dependence of flow classification on the Reynolds number (Re) for the wake of two staggered cylinders. The Re examined ranges from 1.5×10³ to 2.0×10⁴. The pitch ratio, P^⁎=P/d examined is 1.2–6.0 (d is the cylinder diameter), and angle (α) is 0–90°, where P is the center-to-center spacing between two cylinders and α is the angle between the incident flow and the line through the cylinder centers. Two single hotwires were used to measure simultaneously the fluctuating streamwise velocities (u) in the vortex streets generated by the two cylinders. The power spectral density functions and the Strouhal numbers were then obtained from the u signals, based on which the flow structure pattern or mode could be determined. Over two hundred configurations of two staggered cylinders have been examined for each Re. It is found that Re has an appreciable effect on the dependence of the flow mode on P^⁎ and α. The observation is connected to the Re effect on the generic features of a two-cylinder wake such as flow separation, boundary layer thickness, gap flow deflection and vortex formation length.     

Numerical simulation of flow-induced bi-directional oscillations

Flow-induced vibration (FIV) by vortex shedding behind a submerged cylinder can lead to damage of nuclear components. With respect to such a serious scenario, various experiments and numerical simulations have been conducted to predict the vibration phenomena. Especially in simulation, the immersed finite element method (IFEM) is a promising approach to solve fluid-structure interaction problems because it needs less computational resources. In this paper, two-dimensional motions of cylinders are simulated by using IFEM to obtain their vibration characteristics. Three benchmark tests such as flow past a fixed circular cylinder, in-line oscillation of a circular cylinder and flow-induced vibration with uni-directional motion are performed to verify the proposed numerical method. Furthermore, bi-directional motions of two horizontally and vertically arranged cylinders as well as that of a single cylinder in fluid flow are analyzed, and then key findings are fully discussed.     

Flow-induced vibrations of two circular cylinders in tandem with shear flow at low Reynolds number

The flow-induced vibrations of two elastically mounted circular cylinders subjected to the planar shear flow in tandem arrangement are studied numerically at Re=160. A four-step semi-implicit Characteristic-based split (4-SICBS) finite element method is developed under the framework of the fractional step method to cope with the vortex-induced vibration (VIV) problem. For the computational code verification, two benchmark problems are examined in the laminar region: flow-induced vibration of an elastically mounted cylinder having two degrees of freedom and past two stationary ones in tandem arrangement. Regarding the two-cylinder VIVs in shear flow, the computation is conducted with the cylinder reduced mass M_r=2.5π and the structural damping ratio ξ=0.0. The effects of some key parameters, such as shear rate (k=0.0, 0.05, 0.1), reduced velocity (U_r=3.0–18.0) and spacing ratio (L_x/D=2.5, 3.5, 4.5, 8.0), are demonstrated. It is observed that the shear rate and reduced velocity play an important role in the VIVs of both cylinders at various center-to-center distances. Additionally, in comparison with the single cylinder case, a further study indicated that the gap flow has a significant impact on such a dynamic system, leading it to be more complex. The results show that, the performances of ‘dual-resonant’ are discovered in the shear flow. A valley is formed in transverse oscillation amplitude of DC for each spacing ratio when U_r is about 6.0. For the X–Y trajectories of the circular cylinders, figure-eight, figure-O and oval shape are obtained. Finally, the interactions between cylinders are revealed, together with the wake-induced vibration (WIV) mechanism underlying the oscillation characteristics of both cylinders exposed to shear flow. Besides, the “T+P” wake pattern is discovered herein.     

Interference effect of an upstream larger cylinder on the lock-in vibration of a flexibly mounted circular cylinder

Experiments have been carried out to investigate the flow-induced vibration response of a flexibly mounted circular cylinder located in the vicinity of a larger cylinder and subjected to cross-flow. The interfering larger cylinder was placed upstream and had a diameter twice that of the vibrating cylinder. Complex interaction was observed between the flow over the two cylinders. The vibration responses of the flexible cylinder were classified into different regimes according to the relative positions of the two cylinders. In the-side-by-side arrangement and the tandem or near-tandem arrangement, flow-induced vibrations of the flexible cylinder were greatly suppressed. In the staggered arrangement which covered a large portion of the relative cylinder positions being investigated, vibrations of the smaller cylinder were greatly amplified. The vibration response curves were also largely modified with a broadening of the lock-in resonance range. A shift of the peak reduced velocity for maximum vibration response was also found. Flow visualizations and wake velocity measurements suggested that the modifications of the vibration responses were related to the presence or absence of constant or intermittent flow through the gap region between the two cylinders. The proposed mechanisms of flow interactions and the resulting vibration response characteristics could explain previous observations on flow-induced vibrations of two equal-sized circular cylinders reported in the literature.     

Experimental investigation of the wave-induced flow around a surface-touching cylinder

The wave-induced flow around a circular cylinder near both a rigid wall and an erodible bed is experimentally investigated using Particle Tracking Velocimetry (PTV). The aim of this study is to gain quantitative information on the local mean flow, the vorticity dynamics and the evolution of the erodible bed. The flow is characterized in terms of the Keulegan–Carpenter (KC), Reynolds (Re) and Ursell (U_r) numbers. The effects of changing these parameters over the ranges 1<KC<31, 3×10³<Re<2.6×10⁴ and 1.5<U_r<152 are investigated. For KC<1.1 the flow does not separate. When KC increases, the flow becomes unstable and large-scale vortical structures develop. The dimensionless intensity (|Γ^⁎|) depends non-monotonically on KC, with a local maximum at KC=17, and the dimensionless area of the same macrovortex (A^⁎) follows a somewhat similar law. Although the dimensionless boundary layer thickness (δ^⁎) exhibits some discontinuities between KC regimes, it decreases with KC at x/D=0.5, as x/D=1 weakly depends on KC and can be regarded as constant (δ^⁎=0.7) and then, increases with KC when moving away from the cylinder. These findings are used to interpret the physics governing the flow around a cylinder touching a wall and are compared with available results from the literature (Sumer et al., 1991). The evolution of the scour mechanism occurring over an erodible sandy bed is also investigated. The validity of some empirical formulas in the literature is also tested on the basis of the available dataset. The empirical relationships of Cevik and Yuksel (1999) and Sumer and Fredsøe (1990) for the dimensionless scour depth (S/D) agree well with our results. The dimensionless scour width (W_s/D) is predicted well by Sumer and Fredsøe's (2002) empirical equation for KC<23, whereas Catano-Lopera and Garcia's (2007) formula is more accurate for higher values of KC.     

Simulation of cross-flow-induced vibration of cylinder arrays by surface vorticity method

The surface vorticity method (SVM), which is a fast and practical grid-free two-dimensional (2-D) method, and a fluid–structure interaction model incorporating the effects of cylinder motions and displacements is used to simulate the vortex-induced vibration of cylinder arrays at sub-critical Reynolds number Re=2.67×10⁴. The SVM is found to be most suitable for simulating a 2-D cylinder row with large-amplitude vibrations where the vorticity field and the fluid forces of the cylinder row change drastically, and the effect of the stream on the transverse direction vibration is very significant. The fluidelastic instability of a flexible cylinder row at small pitch ratio is also investigated, and the critical reduced velocity of the cylinder row at a reduced damping parameter S_G=1.29 is calculated, which is in good agreement with experimental and analytical results of the unsteady model. Vortex-induced vibration of a staggered cylinder array is simulated using different structural parameters. When the cylinders are relatively more flexible, the flow pattern changes dramatically and the fluid–structure interaction has a dominant impact on the flow field. Compared with grid-based methods, the grid-free SVM is a fast and practical method for the simulation of the FIV of cylinder arrays due to vortex shedding at sub-critical Reynolds numbers.     

Vortex induced vibration and vortex shedding characteristics of two side-by-side circular cylinders of different diameters in close proximity in steady flow

This paper presents results of a numerical study of vortex-induced vibrations of two side-by-side circular cylinders of different diameters in steady incompressible flow. The two-dimensional Reynolds-averaged Navier–Stokes equations with a SST k–ω turbulence model are solved using the Petrov–Galerkin finite element method and the Arbitrary-Lagrangian–Eulerian scheme. The diameter ratio of the two cylinders is fixed at 0.1 and the mass ratio of both cylinders is 5.0. Both cylinders are constrained to oscillate in the transverse direction only. The Reynolds number based on the large cylinder diameter and free stream velocity is fixed at 5000. The effects of the reduced velocities of the cylinders on the vibration amplitude and vortex shedding regimes are investigated. It is found that for the range of parameters considered, collision between the two cylinders is dependent on the difference of the reduced velocities of the cylinders. Presence of the small cylinder in the proximity of the large one appears to have significant effects on the vortex shedding regime and vibration amplitude of the large cylinder.     

Vortex shedding from two finite circular cylinders in a staggered configuration

Wind tunnel experiments were conducted to measure the vortex shedding frequencies for two circular cylinders of finite height arranged in a staggered configuration. The cylinders were mounted normal to a ground plane and were partially immersed in a flat-plate turbulent boundary layer. The Reynolds number based on the cylinder diameter was Re_D=2.4×10⁴, the cylinder aspect ratio was AR=9, the boundary layer thickness relative to the cylinder height was δ/H=0.4, the centre-to-centre pitch ratio was varied from P/D=1.125 to 5, and the incidence angle was incremented in small steps from α=0° to 90°. The Strouhal numbers were obtained behind the upstream and downstream cylinders using hot-wire anemometry. From the behaviour of the Strouhal number data obtained at the mid-height position, the staggered configuration could be broadly classified by the pitch ratio as closely spaced (P/D<1.5), moderately spaced (1.5?P/D?3), or widely spaced (P/D>3). The closely spaced staggered finite cylinders were characterized by the same Strouhal number measured behind both cylinders, an indication of single bluff-body behaviour. Moderately spaced staggered finite cylinders were characterized by two Strouhal numbers at most incidence angles. Widely spaced staggered cylinders were characterized by a single Strouhal number for both cylinders, indicative of synchronized vortex shedding from both cylinders at all incidence angles. For selected staggered configurations representative of closely spaced, moderately spaced, or widely spaced behaviour, Strouhal number measurements were also made along the vertical lengths of the cylinders, from the ground plane to the free end. The power spectra showed that for certain cylinder arrangements, because of the influences of the cylinder–wall junction and free-end flow fields, the Strouhal numbers and flow patterns change along the cylinder.     

Flow-induced oscillation of two circular cylinders in tandem arrangement at low Re

Results are presented for flow-induced vibrations of a pair of equal-sized circular cylinders of low nondimensional mass (m^*=10) in a tandem arrangement. The cylinders are free to oscillate both in streamwise and transverse directions. The Reynolds number, based on the free-stream speed and the diameter of the cylinders, D is 100 and the centre-to-centre distance between the cylinders is 5.5D. The computations are carried out for reduced velocities in the range 2≤U^*≤15. The structural damping is set to zero for enabling maximum amplitudes of oscillation. A stabilized finite element method is utilized to carry out the computations in two dimensions. Even though the response of the upstream cylinder is found to be qualitatively similar to that of an isolated cylinder, the presence of a downstream cylinder is found to have significant effect on the behaviour of the upstream cylinder. The downstream cylinder undergoes very large amplitude of oscillations in both transverse and streamwise directions. The maximum amplitude of transverse response of the downstream cylinder is quite similar to that of a single cylinder at higher Re beyond the laminar regime. Lock-in and hysteresis are observed for both upstream and downstream cylinders. The downstream cylinder undergoes large amplitude oscillations even beyond the lock-in state. The phase between transverse oscillations and lift force suffers a 180 jump for both the cylinders almost in the middle of the synchronization regime. The phase between the transverse response of the two cylinders is also studied. Complex flow patterns are observed in the wake of the freely vibrating cylinders. Based on the phase difference and the flow patterns, the entire flow range is divided into five sub-regions.     

On vortex shedding from low aspect ratio dual step cylinders

A dual-step cylinder is comprised of two cylinders of different diameters. A large diameter cylinder (D) with low aspect ratio (L/D) is attached to the mid-span of a small diameter cylinder (d). The present study investigates the effect of Reynolds number (Re_D) and L/D on dual step cylinder wake development for D/d=2, 0.2≤L/D≤3, and two Reynolds numbers, Re_D=1050 and 2100. Experiments have been performed in a water flume facility utilizing flow visualization, Laser Doppler Velocimetry (LDV), and Particle Image Velocimetry (PIV). The results show that vortex shedding occurs from both the large and small diameter cylinders for 1≤L/D≤3 at Re_D=2100 and 2≤L/D≤3 at Re_D=1050. At these conditions, large cylinder vortices predominantly form vortex loops in the wake and small cylinder vortices form half-loop vortex connections. At lower aspect ratios, vortex shedding from the large cylinder ceases, with the dominant frequency in the large cylinder wake attributed to the passage of vortex filaments connecting small cylinder vortices. At these lower aspect ratios, the presence of the large cylinder induces periodic vortex dislocations. Increasing L/D increases the frequency of occurrence of vortex dislocations and decreases the dominant frequency in the large cylinder wake. The identified changes in wake topology are related to substantial variations in the location of boundary layer separation on the large cylinder, and, consequently, changes in the size of the vortex formation region. The results also show that the Reynolds number has a substantial effect on wake vortex shedding frequency, which is more profound than that expected for a uniform cylinder.     

Flow patterns and turbulence effects in large cylinder arrays

Studies on flow-induced vibrations in large tube bundles are usually focused solely on frequency analysis, without considering the flow patterns which are responsible for the fluid forces. Furthermore, investigations which involve variations in the spacing ratios do not separate transversal and longitudinal proximity effects. The purpose of this article is to separately analyze the influence of the transversal (T/D) and longitudinal (L/D) spacing ratios of a confined in-line cylinder array with five rows on the flow characteristics and to identify flow patterns. The laser Doppler anemometry technique was employed to acquire the mean velocity and its fluctuations in the transversal and longitudinal directions between the cylinder rows. Strouhal numbers and regimes reported in the literature were identified in the experiments. The same regime did not always persist along all cylinder rows for a given spacing ratio, as a result of the combined longitudinal and transversal proximity effects and also of the generation of turbulence by the array. For the smallest T/D ratio, a quasi-steady behavior associated with the biased flow pattern was noted in the experimental set-up and flip-flopping was observed in one case. Additionally, the flow characteristics in these arrays diverged from tube bundle classifications described in the literature. The behavior of the fluid forces and susceptibility to vibrations in the array were predicted based on the turbulence intensity of the incident flow of the cylinders. The results reinforced the need to extend flow pattern investigations to arrays with more cylinder rows and to consider both transversal and longitudinal proximity effects, when studying flow-induced vibrations.     

Three-dimensional computation for flow-induced vibrations of an upstream circular cylinder in two tandem circular cylinders

It is well known from a lot of experimental data that fluid forces acting on two tandem circular cylinders are quite different from those acting on a single circular cylinder. Therefore, we first present numerical results for fluid forces acting on two tandem circular cylinders, which are mounted at various spacings in a smooth flow, and second we present numerical results for flow-induced vibrations of the upstream circular cylinder in the tandem arrangement. The two circular cylinders are arranged at close spacing in a flow field. The upstream circular cylinder is elastically placed by damper-spring systems and moves in both the in-line and cross-flow directions. In such models, each circular cylinder is assumed as a rigid body. On the other hand, we do not introduce a turbulent model such as the Large Eddy Simulation (LES) or Reynolds Averaged Navier-Stokes (RANS) models into the numerical scheme to compute the fluid flow. Our numerical procedure to capture the flow-induced vibration phenomena of the upstream circular cylinder is treated as a fluid-structure interaction problem in which the ideas of weak coupling is taken into consideration.     

Flow-induced vibration of a flexibly mounted circular cylinder in the proximity of a larger cylinder downstream

A flexibly mounted circular cylinder is placed upstream of a stationary cylinder twice as large. Flow-induced vibration response of the small cylinder is measured with the interfering cylinder placed at 57 relative locations. In most situations, reduced-amplitude vibration or even no vibration is observed. Lock-in resonance remains the dominant vibration behavior, but the reduced velocity of peak lock-in is found to shift to a value higher or lower than the isolated cylinder value, depending on the lateral separation between the two cylinders. When the flexible cylinder is located just in front of the large cylinder, galloping-type vibration of very large amplitude occurs at reduced velocities above 12. Mechanisms of flow-induced vibration are discussed with the aid of flow visualizations. The present study supplements a previous paper reporting amplified vibration of the flexible cylinder with the interfering cylinder placed in various upstream locations.     

Flow-induced vibration of a circular cylinder subjected to wake interference at low Reynolds number

Two- and three-dimensional numerical simulations of the flow around two circular cylinders in tandem arrangements are performed. The upstream cylinder is fixed and the downstream cylinder is free to oscillate in the transverse direction, in response to the fluid loads. The Reynolds number is kept constant at 150 for the two-dimensional simulations and at 300 for the three-dimensional simulations, and the reduced velocity is varied by changing the structural stiffness. The in-line centre-to-centre distance is varied from 1.5 to 8.0 diameters, and the results are compared to that of a single isolated flexible cylinder with the same structural characteristics, m^?=2.0 and ζ=0.007. The calculations show that significant changes occur in the dynamic behaviour of the cylinders, when comparing the flow around the tandem arrangements to that around an isolated cylinder: for the tandem arrangements, the lock-in boundaries are wider, the maximum displacement amplitudes are greater and the amplitudes of vibration for high reduced velocities, outside the lock-in, are very significant. The main responsible for these changes appears to be the oscillatory flow in the gap between the cylinders.     

VIV suppression of a two-degree-of-freedom circular cylinder and drag reduction of a fixed circular cylinder by the use of helical grooves

Experimental investigations have been carried out to examine the effects of triple-starting helical grooves on the drag of fixed circular cylinders and the vortex-induced vibration of elastically supported cylinders. For the elastically supported cylinder, the Reynolds number varied from 1.3×10⁴ to 4.6×10⁴, whilst for the fixed cylinder from 3.1×10⁴ to 3.75×10⁵. A comparative approach which allows direct comparisons of the results was adopted where two cylinders of identical dimensions and physical properties with or without helical surface grooves were tested in exactly same experimental set-ups. In the elastically supported cylinder tests, the cylinders were attached to a vertically cantilevered supporting rod and towed in a towing tank. Both the in-line and cross-flow vibrations were permitted. In the fixed cylinder tests, the cylinders were supported on rigid vertical struts and towed horizontally in the same towing tank. It is found that for the case investigated the helical grooves were effective in suppressing the vortex-induced cross-flow vibration amplitudes with the peak amplitude reduced by 64%. Drag reductions of up to 25% were also achieved in the sub-critical Reynolds number range tested in the study for the fixed cylinders.