betaFeedback

WorksPhysics and Astronomy

Longitudinal vortices in unsteady Taylor-Couette flow: solution to a 60-year-old mystery

Abstract

Applying a sufficiently rapid start-stop to the outer cylinder of the Couette-Taylor system, structures approximately aligned with the axis were recorded in the classic work of Coles (1965). These short-lived rolls are oriented perpendicular to the classic Taylor-vortex rolls. In this work we report numerical observation of this instability, guided by a more recent experimental observation. The instability is shown to be related to an inflection in the azimuthal velocity profile, a finding consistent with the experimental observations of its emergence during the deceleration phase. Despite the transient nature of start-stop experiments, we show that the instability can be linked to that of the oscillating boundary layer problem of Stokes. There are several reasons why the instability may have remained elusive, both for experimental observation and intrinsic to the idealized system. We look in more detail at dependence on the radius ratio for the Taylor-Couette system and find that, in the case where the size of the rolls scales with the gap width, for radius ratios any lower than that used by Coles, R_i/R_o=0.874, the instability is quickly overrun by axisymmetric rolls of Gortler type.

Introduction

The nature of the flow of fluid between two rotating cylinders has long played a special role in the study of fluid dynamics and dynamical systems. Taylor [18] demonstrated that linear stability analysis, using the no-slip condition, could accurately predict the onset of rolls. A large number of distinct flow regimes were soon identified, and Coles [7] (figure 2 therein) created a summary graphic in the plane defined by the rotation rates of the inner and outer cylinders. This plot included Taylor’s original stability line, above which greater rotation of the inner cylinder rendered the flow unstable to ‘singly periodic flow’ (Taylor-vortices), and catalogued regions of ‘doubly periodic flow’ (wavy Taylor-vortices), in addition to spiral and turbulent flows. Coles was particularly interested in the multiplicity of wavy Taylor vortex states, of various azimuthal wavenumber, and transitions between them. A celebrated version of this diagram, including a greater range of flows, was produced by Andereck et al. [3].

It is now a little over 100 years since Taylor’s landmark paper, 60 years since the classic work of Coles, and the rich dynamical landscape of the Taylor–Couette flow, for the steady rotating case alone, is well recognised.

Multiple flow states for the same rotation rates are not easily represented on a summary graphic, and figure 8 of Coles [7] is a prime example of this complexity. The addition of time-dependence likewise introduces an extra dimension of complexity, and expeditions in these directions are fewer but significant. Görtler rolls appear when flow encounters a curved surface, and like Taylor-vortex rolls, they are aligned with the direction of the flow and caused by a centrifugal instability. Görtler rolls have been observed to form and eventually evolve into Taylor-vortices for the case of an impulsive start of the inner cylinder, see e.g. Neitzel et al. [14]. Studies of an impulsive stop for the outer cylinder include Kohuth & Neitzel [10]; Verschoof et al. [19]; Ostilla-Mónico et al. [16]; Singh & Prigent [17], where rolls of many sizes appear and decay, leaving behind the largest scale.

Periodic modulations of the rotation rate of either cylinder have been studied. For example, Ern & Wesfreid [9] found regimes of both destabilisation and stabilisation, while Youd et al. [23] observed that Taylor-vortex rolls may or may-not reverse the direction of the rolls. Axial oscillations of the inner cylinder have been shown to suppress Taylor instability by Marques & Lopez [13]. In the context of planetary libration, an interesting set of experiments and calculations with modulated rotation rates have been performed, e.g. Noir et al. [15]; Lopez & Marques [11]. In all cases discussed above, the same instabilities are manifest, i.e. the appearance of modified Taylor-vortices or Görtler rolls. Other related studies have focused on the effect of surface modulation on a turbulent background state, e.g. Verschoof et al. [20] applied oscillations of the inner cylinder in high Reynolds number experiments and observed how the flow responded to different oscillation rates.

Experimental observations of longitudinal vortices

Coles [7] includes an image of a flow, figure 22(o) therein and reproduced in Figure 1(a) here, that appears unlike any of those mentioned above. While those flows are dominated by azimuthally aligned rolls that wrap around the cylinder, here the waves are aligned with the axis. In a footnote (p410), Coles describes his method for establishing wavy Taylor vortex flow with a high ‘tangential’ wavenumber, now more commonly referred to as the azimuthal wavenumber mm: “The usual experimental method for establishing a flow with a high tangential wavenumber was to approach the final operating condition from the singly-periodic side [from axisymmetric Taylor rolls] with two cylinders initially rotating in the same direction. A sufficiently fast stop for the outer cylinder then usually led to a flow with a large number of tangential waves, probably as a result of Tollmien instability in the unsteady viscous layer on the wall of the outer cylinder (cf. figure 22(o), plate 12).” The Tollmien–Schlichting instability is a viscous instability that can occur in a boundary layer flow, such as when a unidirectional flow encounters a flat plate. The rolls that arise within the boundary layer are aligned spanwise to the direction of the flow. In principle, it is possible the sudden-stop could indeed set up a boundary-layer type flow and hence Tollmien instability. However, Coles’ caption to figure 22(o) reads only “(o) Instability following start-stop motion of outer cylinder.” The sudden-stop and start-stop experiments are potentially quite different! Unfortunately, no parameters for the start-stop experiment were provided.

Photographs of longitudinal vortices

Figure 1. (a) Figure 22(o) of Coles [7], captioned ‘Instability following start-stop motion of outer cylinder.’ (b) Similar observation from the experiments of Burin & Czarnocki (2012)

More recently, during an experimental campaign to characterize turbulence driven by outer cylinder motion (Burin & Czarnocki 2012), a brief set of start-stop observations were made, revealing what appears to be the same azimuthal instability as described above. See Figure 1(b). The instability in this case was observed to arise during the decelerating phase, and for only a narrow gap width (d=6.3d = 6.3 mm and radius ratio Ri/Ro=0.97R_i/R_o = 0.97), although its apparent absence for wider gap widths could be due to visualization shortcomings. A qualitative dependence of the instability wavelength with respect to forcing was observed, with initial acceleration provided by falling weights coupled to the drive system, while mechanical friction provided deceleration. As the acceleration increased, the wavelength increased by a factor of two: from about 1.3 to 2.5 cm, or (approximately) 2 to 4 gap widths. For smaller accelerations the instability was not seen, and for higher ones the turbulence emanating from the end-caps engulfed the central region before the instability could presumably appear. The maximum Reynolds number of the outer cylinder varied in these initial exploratory experiments, although the period and rate of both the acceleration and deceleration phases contain significant uncertainty due to the simple driving mechanism. Nonetheless, this second appearance of the instability in question did reveal that it occurs during the deceleration phase.

To our knowledge, there has been no systematic study of this transitory TC flow state. In this paper, using new simulations alongside these scarce laboratory observations, we aim to clarify when the longitudinal vortices should appear, with respect to the type of motion of the outer cylinder, and to identify the instability mechanism. We show how this particular unsteady flow may be understood within the wider context of unsteady boundary flows, along the way linking Taylor–Couette flow to a branch of fluid mechanics reaching back to Stokes’ oscillating boundary layer problem.

Methods

We consider the flow between concentric cylinders of radius RiR_i and RoR_o, where the subscripts indicate the inner and outer cylinders respectively, with rotation rates Ωi\Omega_i and Ωo\Omega_o. The radius ratio is denoted by η=Ri/Ro\eta= R_i/R_o and gap width d=Ro−Rid = R_o - R_i. Scales used in the non-dimensionalisation of the governing equations are the gap width dd and the viscous diffusion timescale d2/νd^2/\nu, where ν\nu is the kinematic viscosity, which combine to give a velocity scale ν/d\nu/d. Quantities in all figures are non-dimensionalised with these scales.

The dimensionless Navier–Stokes equations are

∂tu+u⋅∇u=−∇p+∇2u,∇⋅u=0.(1)\partial_t \mathbf{u}+\mathbf{u}\cdot\nabla\mathbf{u}=-\nabla p+\nabla^2\mathbf{u},\qquad\nabla\cdot\mathbf{u}=0. \tag*{(1)}

Note that the Reynolds numbers, Rei,o=Ri,oΩi,od/νRe_{i,o}=R_{i,o}\Omega_{i,o}d/\nu, do not appear in the governing equation but determine the speed of the boundaries: uθ=Reiu_\theta=Re_i and uθ=Reou_\theta=Re_o at dimensionless radii ri=Ri/d=η/(1−η)r_i=R_i/d=\eta/(1-\eta) and ro=Ro/d=ri+1r_o=R_o/d=r_i+1 respectively.

When Rei,oRe_{i,o} are constant, steady circular-Couette flow is given by

u=V0(r)θ^,V0(r)=11+η[(Reo−ηRei)r+η(1−η)2(Rei−ηReo)1r].(2)\mathbf{u}=V_0(r)\hat{\boldsymbol{\theta}},\qquad V_0(r)=\frac{1}{1+\eta}\left[(Re_o-\eta Re_i)r+\frac{\eta}{(1-\eta)^2}(Re_i-\eta Re_o)\frac{1}{r}\right]. \tag*{(2)}

To implement the case when time-dependent Ωi,o(t)\Omega_{i,o}(t) imply time-dependent Rei,o(t)Re_{i,o}(t), we consider the deviation u′\mathbf{u}' of the velocity field from (2) evaluated at each instant, V0(r,t)=V0(r;Rei(t),Reo(t))V_0(r,t)=V_0(r;Re_i(t),Re_o(t)), i.e.

u(x,t)=V0(r,t)θ^+u′(x,t).(3)\mathbf{u}(\mathbf{x},t)=V_0(r,t)\hat{\boldsymbol{\theta}}+\mathbf{u}'(\mathbf{x},t). \tag*{(3)}

The field V0(r,t)V_0(r,t) is used only in the computations and should be distinguished from the ‘mean profile’ V(r,t)V(r,t), which denotes the azimuthal component of the θ\theta- and zz-average of the total flow field u\mathbf{u}. The base component of the flow u=V0θ^\mathbf{u}=V_0\hat{\boldsymbol{\theta}} satisfies ∇2u=0\nabla^2\mathbf{u}=0. Substituting (3) into (1) and using this property, the governing equation for the deviation field u′\mathbf{u}' can be written

(∂t−∇2)u′=u×(∇×u)−∇p−(∂tV0)θ^,(4)(\partial_t-\nabla^2)\mathbf{u}'=\mathbf{u}\times(\nabla\times\mathbf{u})-\nabla p-(\partial_tV_0)\hat{\boldsymbol{\theta}}, \tag*{(4)}

subject to ∇⋅u′=0\nabla\cdot\mathbf{u}'=0 with boundary conditions u′=0\mathbf{u}'=0 at r=ri,or=r_{i,o}. A vector calculus identity has been applied to the advection term to give the rotational form of the governing equation, with the double-cross term on the right. This is slightly easier and less expensive to evaluate numerically.

The code used for time integration of (4) is a long-standing variant of the Openpipeflow solver [22], and modification for this case required only the addition of the forcing term, i.e. the known last term on the right-hand side of (4). Periodicity is assumed in the axial dimension over a length 2π/α2\pi/\alpha, and mm-fold periodicity can be imposed in the azimuthal dimension. Variables are discretised via a double-Fourier decomposition, and finite differences are used in the radial dimension:

a(rn,θ,z)=∑∣k∣<K∑∣m′∣<MAkm(rn)ei(αkz+mm′θ),n=1,…,N.(5)a(r_n,\theta,z)=\sum_{|k|<K}\sum_{|m'|<M}A_{km}(r_n)e^{i(\alpha kz+mm'\theta)},\qquad n=1,\ldots,N. \tag*{(5)}

Each Fourier mode AkmA_{km} is evaluated at the radial points rnr_n, which are located at the NN extrema the Tchebyshev polynomial TN−1(x)T^{N-1}(x) defined on x=(r−ri)∈[0,1]x=(r-r_i)\in[0,1]. These points are bunched towards the walls within the boundary layers. N=100N=100 for most calculations, while, for linear stability calculations, KK and MM are only as large as required to capture the first few modes. The Crank–Nicolson method is applied to the diffusion term, while an Euler-predictor and Crank–Nicolson corrector are applied the first and last terms on the right-hand side of (4). Via the pressure-Poisson equation, pp is used to project the velocity onto the space of divergence-free functions, using an influence-matrix technique to implement the no-slip and divergence-free condition at the wall simultaneously. The difference between the predictor and corrector may be used to control the timestep, but for the calculations here, the timestep is fixed at Δt=10−6\Delta t=10^{-6}. (Note that the dimensionless advective timescale 1/∣u∣∼1/Reo1/|\mathbf{u}|\sim1/Re_o.)

The mean profile (θ,z-average)(\theta,z\text{-average}) is the component V(r,t)=uθ,00′+V0(r,t)V(r,t)=u'_{\theta,00}+V_0(r,t), where the subscript 00 indicates the k=m′=0k=m'=0 mode in (5). In the absence of perturbations, the mean profile is quickly computed by simulating with K=M=1K=M=1. Eigenvalues, which give the growth rates of linear perturbations to the base flow, are calculated by combining time-integration with Arnoldi iteration. This method applies repeated re-orthogonalisation to the perturbation via the Gram–Schmidt process, which accelerates convergence to the leading mode, and enables capture of further eigenmodes and eigenvalues. For details, see e.g. Chu et al. [6] §2.2.

Results

Approximation of experiment

The recent experimental observations of longitudinal vortices for a start-stop of the outer cylinder provided the following parameter configuration, which is later rationalised in §3.2. An example video may be found in the supplementary material; 1 second in the video corresponds to approximately 0.024 viscous times d2/νd^{2}/\nu.

Instability was observed for the narrow gap, η=0.97\eta= 0.97, and systematic experiments used a weight-and-pully system to start the motion, implying approximately constant acceleration. The outer cylinder was also observed to come to an abrupt halt, suggesting use of a constant decelleration until no motion is reached, rather than say an exponential slow down. Therefore, we begin with a linear ramp up and down in the Reynolds number for the outer cylinder

Reo(t)={U1t/T1,0≤t<T1,U1(T2−t)/(T2−T1),T1≤t<T2,0,T2≤t,(6)Re_o(t) = \begin{cases} U_1 t/T_1, & 0 \le t < T_1,\\ U_1 (T_2-t)/(T_2-T_1), & T_1 \le t < T_2,\\ 0, & T_2 \le t, \end{cases} \tag*{(6)}

with experimentally-estimated dimensionless parameters U1=8000U_1 = 8000, T1=0.01T_1 = 0.01, T2=0.05T_2 = 0.05. The inner cylinder is fixed, i.e. Rei=0Re_i = 0. Instability was most frequently observed with an azimuthal wavelength λ≈3.5\lambda\approx3.5 (gap widths). Using λ≈2πro/m\lambda\approx2\pi r_o/m, where ro=1/(1−η)r_o = 1/(1-\eta) and mm is the rotational symmetry (see (5)), implied setting m=60m = 60.

Figure 2(a) shows calculations of the azimuthal mean flow profile V(r,t)V(r,t) during the ramp up and ramp down of the outer cylinder. The viscous time-scale d2/νd^2/\nu was used for the non-dimensionalisation. Similarly, a time scale for the ramp-up stage may be written δ2/ν\delta^2/\nu, where δ\delta measures how far the change in velocity of the outer boundary is expected to penetrate into the flow. Expressing this in the timescale used for the non-dimensionalisation, we have δ2/ν=t d2/ν\delta^2/\nu= t\,d^2/\nu, which rearranges to δ/d=t\delta/d = \sqrt{t}. This implies that by the end of the ramp-up stage, at time t=T1=0.01t = T_1 = 0.01, non-zero velocity is expected to penetrate approximately one tenth of the way across the gap, in good agreement with the curve for t=0.01t = 0.01 in Figure 2(a). From this time on, the ramp-down is seen in the velocity at the outer boundary, leaving behind a non-zero velocity profile in the interior.

Mean profiles and perturbation energies

Figure 2. (a) Mean profile during the ramp up of the outer cylinder (from t=0t = 0 to T1=0.01T_1 = 0.01) and ramp down of the outer cylinder (from t=T1t = T_1 to T2=0.05T_2 = 0.05). (b) Normalised energy of the mean flow, Em=0(t)/Em=0(T1)E_{m=0}(t)/E_{m=0}(T_1) (black) and energy of non-axisymmetric (m≠0m \ne0) perturbations, Em≠0(t)/Em=0(T1)E_{m\ne0}(t)/E_{m=0}(T_1) (blue). Each blue line shows the development of a perturbation introduced at a different time, where the line starts. Vertical lines indicate the times T1T_1 and T2T_2 and the slope indicates the growth rate for the wavelength λ≈3.5\lambda\approx3.5 at t=0.04t = 0.04 of Figure 3.

Velocity perturbations are added to V(r,t)V(r,t) that are very small, so that the Fourier modes of the perturbation may be considered to be linearly independent. For the case of axially-aligned rolls, we consider only modes with k=0k = 0 in (5) so that the calculations are two-dimensional (axially-independent). Small perturbations are added for m′m' up to 7, but the m′=1m'=1 mode is always found to be dominant. Figure 2(b) shows time series of the energy summed over all azimuthally-dependent (m′≠0m' \ne0) perturbations. It was found that perturbations added at t=0t = 0 would not grow substantially, such as the bottom-most curve. Instead, perturbations added at later times during the development of the mean profile were necessary to trigger transient instability. (In the physical experiment, it is perhaps reasonable that some noise is always present, from end conditions at least.) Where a growing mode becomes established, the growth rates appear to be approximately constant for times 0.02-0.04, which is during the decelerating phase, consistent with the appearance of the instability in the experiment.

It is possible in our numerical calculations to freeze the mean profile V(r,t)V(r,t) at a particular time and to calculate the growth rates of infinitesimal disturbances. For this to produce physically reasonable growth rates, it assumes that there is time for the disturbances to grow substantially more rapidly than changes in the base profile. In Figure 2(b), it is seen that perturbations can grow many orders of magnitude in a short time. With this frozen profile assumption, Figure 3(a) shows the growth rates of perturbations to the profile as a function of azimuthal wavelength λ\lambda at several times. Peak growth is observed at t=0.04t = 0.04 for a wavelength of λ≈2.5\lambda\approx2.5, but observe that instability starts earlier at longer wavelengths, λ≈3\lambda\approx3 for t=0.02t = 0.02 (approximately when growth is first observed), which is a reasonable match to the wavelength observed in the experiment.

Growth rates as functions of azimuthal wavelength

Figure 3. (a) Growth rates σ\sigma as a function of azimuthal wavelength λ\lambda, assuming frozen mean profiles at the given times. U1=8000U_1 = 8000, T1=0.01T_1 = 0.01, T2=0.05T_2 = 0.05. (b) Growth rates at t=0.04t = 0.04 changing only U1U_1.

For the purely azimuthal base flow u=V(r,t)θ^\mathbf{u} = V(r,t)\hat{\boldsymbol{\theta}}, the nonlinear terms in (1) are zero, so that u\mathbf{u} can be multiplied through by an arbitrary constant and remains a solution. Therefore, if U1U_1 is changed without altering T1T_1 and T2T_2, the mean flow is multiplied by a factor but does not change structure. Looking at the time t=0.04t = 0.04 where growth was maximised, Figure 3(b) shows that a change in amplitude of the start-stop changes the growth rate of perturbations, but has no noticeable effect on the preferred wavelength of instability.

In the Introduction, it was noted that Coles [7] had postulated that the rolls ‘probably’ appeared as the result of Tollmien instability, which appears in the boundary layer problem. Acceleration of the boundary sets up a velocity profile of the boundary layer type, but the decelerating phase introduces an inflection point, and it is during this phase that the instability is observed. Figure 4(a) shows the second derivative of the mean profiles at several times, so that the zero of this function tracks the position of an inflection point. Using a wavelength λ≈2.6\lambda\approx2.6 (m=80m = 80), around the value of theoretical preferred wavelength from Figure 3, Figure 4(b) shows eigenfunctions at times 0.02 and 0.04 (as viewed from above, looking down the axis of the cylinders). Although the rolls are not centred precisely on the radial location of the inflection point, they are close and move further from the boundary as tt increases. Figure 5(a) shows the leading growth rates (of modes with strictly mm-periodic dependence) as a function of tt. At t=0t = 0, the mean flow is zero, so all growth rates are associated with free decay modes. Modes associated with Tollmien instability should arise during the ramp up from t=0t = 0 to t=T1=0.01t = T_1 = 0.01, when the mean profile is similar to that for the boundary layer flow. We only see slow adjustment of the growth rates of the free-decay modes, however, duing this period. From t=T1t = T_1, deceleration occurs, the inflection is introduced into the mean profile, and the mode associated with the observed longitudinal vortices quickly appears.

Figure 4. (a) Second derivative of the mean profile for the parameters of Figure 2 – roots indicate the location of an inflection point that appears strictly after T1=0.01T_1 = 0.01. (b) Eigenfunctions of the leading unstable mode with λ≈2.6\lambda\approx2.6 at the given times; contours of streamfunction ψ\psi such that eigenfunction u′=∇×(ψz^)\mathbf{u}^{\prime} = \nabla\times(\psi\hat{\mathbf{z}}).

Leading growth rates for four cases

Figure 5. Leading growth rates σ\sigma as a function of tt for T1=0.01T_1 = 0.01, T2=0.05T_2 = 0.05, λ≈2.6\lambda\approx2.6 (m=80m = 80) with (a) U1=8000U_1 = 8000, (b) U1=24000U_1 = 24000, and (c) for lower U1=1000U_1 = 1000 shown over a longer period. Vertical lines indicate the times T1T_1 and T2T_2. Only one longitudinal mode goes unstable; all other modes for t≲0.03t \lesssim0.03 connect to the free-decay modes of t=0t = 0 when u=0\mathbf{u} = 0. (d) Growth rates for the ramp-down of a developed flow: Re(t)=U1=8000Re(t) = U_1 = 8000 for −∞<t<T1-\infty< t < T_1, then Re(t)Re(t) as before for t>T1t > T_1.

Using the viscous length scale δ\delta introduced in relation to Figure 2(a), the accelerating phase may be compared with Stokes’ problem for a flat plate impulsively accelerated to velocity U0U_0. A critical value for ReI=U0δ/νRe_I = U_0\delta/\nu close to ReI=1485Re_I = 1485 has been calculated by Luchini & Bottaro [12] for the appearance of Tollmien–Schlichting waves. Using the relationship δ/d=t\delta/d = \sqrt{t} and equating U0U_0 with RoΩoR_o\Omega_o gives Reo=ReI/tRe_o = Re_I/\sqrt{t}. Given that our boundary velocity is only ramped up to U1U_1, rather than impulsively started, this provides only a lower bound, Reo(t)≥1485/tRe_o(t) \ge1485/\sqrt{t}, where max⁡(Reo(t))=U1\max(Re_o(t)) = U_1. Within the ramp-up phase, the bound is lowest at t=T1=0.01t = T_1 = 0.01, implying an ReoRe_o greater than 14 850 would be required for Tollmien instability. Calculations with U1U_1 three times larger at U1=24 000U_1 = 24\,000 showed growth rates similar in form (Figure 5b), still with no sign of instability before T1T_1 and instability arising immediately after T1T_1. Although Tollmien instability could in principle appear later tt, it is unlikely since the velocity profile decays rapidly. Moreover, the appearance of instability at the same time adds extra weight to distinguish it from Tollmien instability, as when different Re0(t)Re_0(t) are achieved, the bound implies that such instability is expected to arise at a different time tt.

As the instability arises during the decelerating phase, it is reasonable to ask whether the accelerating phase is necessary. Figure 5(d) shows the growth rate for the case where the mean flow is developed with Re(t)=U1=8000Re(t) = U_1 = 8000 for all tt up until T1=0.01T_1 = 0.01, then is followed by the same ramp down to Re(t)=0Re(t) = 0 at t=T2=0.05t = T_2 = 0.05. All eigenvalues are negative, indicating that both the ramp up and ramp down phases are required for instability.

Comparison with Stokes’ oscillating boundary problem

Stokes found an analytic solution for the flow induced in a semi-infinite body of fluid by a sinusoidal oscillation of an infinite plate. For a plate moving at (dimensional) speed U=U0cos⁡(ωt)U = U_0 \cos(\omega t), the viscous length scale for the time-varying boundary velocity is often defined δ=2ν/ω\delta= \sqrt{2\nu/\omega}. Scaling with U0U_0 and δ\delta gives a Reynolds number Rδ=U0δ/νR^\delta= U_0\delta/\nu. Von Kerczek & Davis [21] considered the linear stability of the base flow at each instant for a channel of width dd, where the second wall is stationary. For most calculations, the ratio of scales was β=d/δ=8\beta= d/\delta= 8, for which it was shown that the influence of channel width was negligible. The most unstable velocity profile was found to be at time t=π/2t = \pi/2 and that rolls of wavenumber

α=0.5\alpha= 0.5 first go unstable at Rδ=86R^\delta= 86. Blondeaux & Vittori [4] have more recently revisited this analysis for β→∞\beta\to\infty, confirming the results of Von Kerczek & Davis [21]. They also examined for which RδR^\delta there is instability in the accelerating phase or over the whole cycle.

To produce the counterpart instability in the Taylor–Couette apparatus, we suppose that the gap for the channel, d=βδd = \beta\delta, coincides with the Taylor–Couette gap width dd, and expect that correspondence will be best for the narrow gap, where curvature terms are small relative to the viscous terms. For the Taylor–Couette system we put Reo=U1sin⁡(2πt/T)Re_o = U_1 \sin(2\pi t/T).

Converting scales from the channel to Taylor–Couette system, we have δ=2ν/ω=2ν(T(d2/ν)/(2π))=dT/π\delta= \sqrt{2\nu/\omega} = \sqrt{2\nu(T(d^2/\nu)/(2\pi))} = d\sqrt{T/\pi}, then using β=d/δ\beta= d/\delta gives T=π/β2T = \pi/\beta^2. Maintaining that β=8\beta= 8 for the Taylor–Couette system determines a period for the oscillation T=π/82≈0.05T = \pi/8^2 \approx0.05. For the velocity amplitude, U1(ν/d)=U0U_1(\nu/d) = U_0 implies U1=U0d/ν=(d/δ)(U0δ/ν)=βRδU_1 = U_0d/\nu= (d/\delta)(U_0\delta/\nu) = \beta R^\delta. The critical Rδ=86R^\delta= 86 predicts that the amplitude is critical for U1=688U_1 = 688. The most unstable wavenumber α=0.5\alpha= 0.5 implies m=2πRo/λ=2π(1/(1−η))d/((2π/α)δ)m = 2\pi R_o/\lambda= 2\pi(1/(1-\eta))d/((2\pi/\alpha)\delta), which for radius ratio η=0.97\eta= 0.97 and β=8\beta= 8 give m≈133m \approx133. Using these parameters for the Taylor–Couette setup, Figure 6(a) shows the real part of the three leading growth rates for perturbations to frozen mean profiles. The range in tt shown corresponds to half a period, as the mean flow is reversed in the second half and has the same growth rates. (Nine oscillations are simulated first to eliminate any transient originating from the initial condition u′=0\boldsymbol{u}' = 0 at t=0t = 0.) The leading growth rate just touches criticality at the expected time in the cycle, indicating that the prediction derived from the channel calculation of Von Kerczek & Davis [21] is very good.

Three leading growth rates and eigenfunction of the neutral mode

Figure 6. (a) Three leading growth rates for a sinusoidal oscillation of period TT as a function of t′=t−9.25Tt' = t - 9.25T in the Taylor–Couette configuration. Parameters for critical U1U_1 and wavenumber mm predicted from Stokes oscillation in a channel. Calculations cover half of the period T=0.05T = 0.05, as the base flow reverses in the second half and eigenvalues repeat. Reo(t)Re_o(t) is maximum at t′=0t' = 0 and is zero at t′=T/4=0.0125t' = T/4 = 0.0125. (b) Eigenfunction of the neutral mode.

Qualitatively, the structure of the growth rates in Figure 6(a) is not dissimilar to those of Figure 5(c) for the start-stop experiment of § 3.1, likewise the structure of the eigenfunction of Figure 6(b) is similar to that of Figure 4(b) for the start-stop experiment at the latter time, when it grows most rapidly. To make an approximately quantitative link, and to a posteriori rationalise the choice of parameters for the start-stop experiment, we make a rough approximation that the linear ramp up and ramp down over a time T2T_2 corresponds to one half of a sinusoidal oscillation of period T=2T2T = 2T_2. Then, T=π/β2T = \pi/\beta^2 for our experiment with T2=0.05T_2 = 0.05 implies β=π/0.1≈5.6\beta= \sqrt{\pi/0.1} \approx5.6. This value is not so far from the β=8\beta= 8 of

Von Kerczek & Davis [21], suggesting that, if the instability is of the same nature, the result of § 3.1 should not have been substantially influenced by the inner boundary. It should then be possible to predict the most unstable wavelength of the instability from that of the Stokes problem. The wavenumber α=0.5\alpha= 0.5 based on units δ\delta for the channel corresponds to an azimuthal wavelength of λ=(2π/α)/β≈2.2\lambda= (2\pi/\alpha)/\beta\approx2.2 in gap widths dd. In Figure 3 it was found that the peak growth rate occurred for λ≈2.5\lambda\approx2.5 in the start-stop experiment, a difference of around 10-20%.

Dependence on radius ratio

In this section we return to the start-stop problem and consider the case where RoR_o is fixed and RiR_i decreases, so that the gap dd increases and η=Ri/Ro\eta= R_i/R_o decreases.

If Ωo(t)\Omega_o(t) is also unchanged, then the penetration depth δ\delta does not change, and β=d/δ\beta= d/\delta increases. It has been confirmed in calculations that the flow near the outer cylinder is independent of the inner cylinder, and there is no change to the instability (therefore not shown).

If dd is increased and Ωo(t)\Omega_o(t) scaled such that δ\delta increases proportionately, i.e. β=d/δ\beta= d/\delta is held fixed, then wavelength of the instability will continue to scale with dd and the wavenumber mm decreases. In this case we expect more influence from the curvature. Conveniently, the dimensionless velocity at the boundary, V(ro,t)V(r_o,t), and the time-dependent Reynolds number Reo(t)Re_o(t) are unchanged, as our non-dimensionalisation is already based on the viscous time scale and on dd. Only the wavenumber mm changes with η\eta: for a change d→d~d \to\tilde{d} and η→η~\eta\to\tilde{\eta}, we have m~=m d/d~=m (1−η)/(1−η~)\tilde{m} = m\,d/\tilde{d} = m\,(1-\eta)/(1-\tilde{\eta}).

Figure 7 compares the growth of different Fourier modes for η=0.97\eta= 0.97, 0.8740.874, 0.80.8. The longitudinal vortices are axially independent, α=0\alpha= 0, and the non-zero mm corresponds to λ≈2.6\lambda\approx2.6 for all radius ratios η\eta. The growth of this longitudinal mode is similar for all cases (thick purple line; note the scale on the yy-axis). Supposing that the penetration depth δ\delta approximately gives the distance from the outer wall to the centre of a vortex, then it is estimated that the wavelength over a pair of azimuthally independent (m=0m = 0) Görtler rolls might be 4δ/d=4/β4\delta/d = 4/\beta, in units of gap width. For β=5.6\beta= 5.6, the axial wavenumber is then α=2.8π≈8.8\alpha= 2.8\pi\approx8.8. In Figure 7 it is seen that such Görtler modes do not grow for the narrow gap at η=0.97\eta= 0.97, but for η=0.8\eta= 0.8 they are expected to swamp any trace of the longitudinal vortices in a flash. Interestingly, Coles’ choice of η=0.874\eta= 0.874 appears to have been a serendipitous one, close to the borderline where both modes might be observed.

Energy plots and Stokes streamfunction contours

Figure 7. Energy of longitudinal modes (m≠0m \ne0, α=0\alpha= 0) and Görtler modes (m=0m = 0, α≠0\alpha\ne0) during a start-stop with parameters U1=8000U_1 = 8000, T1=0.01T_1 = 0.01, T2=0.05T_2 = 0.05 at radius ratios (a) η=0.97\eta= 0.97, (b) η=0.874\eta= 0.874, (c) η=0.8\eta= 0.8. For all η\eta, the m≠0m \ne0 for longitudinal modes correspond to azimuthal wavelength λ≈2.6\lambda\approx2.6.

(d) Görtler mode at time indicated by the point in (c); contours of the Stokes streamfunction ψ\psi, such that ur=−(1/r)∂zψu_r = -(1/r)\partial_z\psi.

Conclusions

The figure 22(o) of Coles [7], captioned “(o) instability following start-stop motion of the outer cylinder,” with its longitudinal vortices, has been a bit of a longstanding mystery. When Coles described his procedure for introducing different azimuthal wavenumbers mm for wavy Taylor–vortex rolls, he mentioned a sudden-stop that induced a large number of waves, “probably as a result of Tollmien instability,” but referred to the image captioned for a start-stop experiment.

In this work, we have simulated the appearance of longitudinal rolls in the Taylor–Couette system, guided by its unexpected appearance in a more recent experimental campaign (Burin & Czarnocki 2012). We have shown that instability is linked to an inflection point in the mean profile (rather than the Tollmien instability) — despite the transient nature of the start-stop flow, it is possible to link the instability to that of Stokes oscillating boundary problem, which has a Rayleigh instability for sufficiently large Reynolds numbers (Cowley [8]).

There are a number of reasons why the instability may have remained elusive in Taylor–Couette flow. One, of course, is the rich array of other instabilities exhibited by the system, which create other interesting and more manifest diversions. In the recent experiment, longitudinal vortices were only observed for the case η=0.97\eta= 0.97. Guided by this experiment, our parameters suggested a ratio for the viscous scale δ\delta and the gap dd of β=d/δ≈5.6\beta= d/\delta\approx5.6. From the radial extent of the rolls, seen in Figure 4(b), it is expected if β\beta is significantly decreased, then without sufficient physical space for the rolls, the instability will be suppressed.

In our calculations where β\beta is fixed while decreasing the radius ratio, i.e. the rolls scale with the gap width, the radius ratio η=0.874\eta= 0.874 of Coles [7] appears to be around the lower limit at which the instability might be observable. At lower η\eta, the magnitude of curvature terms raises relative to viscous terms, and the instability is almost immediately swamped by centrifugal instability and the formation of Görtler rolls.

It is less obvious why the instability has not been observed for larger β\beta, where the gap dd widens but δ\delta is kept the same. This occurs when RoR_o and Ωo(t)\Omega_o(t) are unchanged and only RiR_i is reduced. Although the instability should be observable in principle, there are practical challenges, and few, if any, of the experiments we have referenced have been looking in the right regime, combining a start-stop or sinusoidal oscillation with a large Reynolds number. For the traditional case of a stationary outer cylinder and steadily rotating inner cylinder, for η=0.97\eta= 0.97, 0.8740.874, 0.80.8 the critical Reynolds numbers are respectively Rei≈239Re_i \approx239, 118118, 9595. Here, for a change d→d~d \to\tilde{d}, matching the peak boundary velocity gives U~1(ν/d~)=U1(ν/d)\tilde{U}_1(\nu/\tilde{d}) = U_1(\nu/d), which implies U~1=U1d~/d=U1(1−η~)/(1−η)\tilde{U}_1 = U_1\tilde{d}/d = U_1(1-\tilde{\eta})/(1-\eta). For the U1=8000U_1 = 8000, inferred from the experiment with η=0.97\eta= 0.97 and used in most calculations, this gives U~1≈3.4×104\widetilde{U}_1 \approx3.4 \times10^4 at η=0.874\eta= 0.874 and U~1≈5.3×104\widetilde{U}_1 \approx5.3 \times10^4 at η=0.8\eta= 0.8. These are the peak values for Reo(t)Re_o(t) for the experiment, and while they are achievable, it is perhaps not usual to be examining linear stability at such large Reynolds numbers.

Another practical difficulty for the case of large β\beta is that, even if the rolls were present, they only occupy a small fraction of the depth that scales like 1/β1/\beta. This would lead to only a small contrast for traditional visualisation techniques using crystalline platelets.

A final question is the source of perturbations necessary to excite the instability. Endplates for the cylinders naturally break axial independence of the flow. They may lead to Ekmann vortices, and can aid the formation of azimuthally aligned Taylor-vortex rolls. In both Cole’s and our experiment pictured in Figure 1, turbulence is present at the ends, but often split end plates and other systems are employed to minimise end-effects. If the instability can only be excited from very low level disturbances, in combination with the start-stop being only a transient flow, it is quite possible that in many situations it would not reach sufficient amplitude to be observed.

For the future, with laser-based techniques such as particle image velocimetry (PIV) now more prevalent, such diagnostics may aid further experiments, and possibly lead to more frequent unexpected observations, of this instability over a wider range of radius ratios. Artificial perturbations, such as axially aligned grooves may also enhance the instability, possibly pushing it into regimes where Taylor or Görtler rolls would otherwise take over.

It would also be interesting to revisit whether a signature of the instability may be observed in the presence of turbulence –Verschoof et al. [20] examined turbulent flow with rapid oscillations of the inner cylinder, parameterised by the Womersley number Wo=dω/νWo = d\sqrt{\omega/\nu}, that correspond to values of β=Wo/2\beta= Wo/\sqrt{2} in the range 10.5 to 80.8, commensurate with the instability discussed above. They observed that for shorter modulation periods, the flow responds with a phase delay. Numerical work at large Reynolds numbers has been performed for planar Couette flow by Akhtar & Ostilla-Mónico [1] with and without rotation, and Akhtar & Ostilla-Mónico [2] for larger amplitudes and more varied waveforms of the oscillation. They found that modified versions of Stokes’ laminar solution persist subject to modified turbulent viscosity, except in the case of cyclonic rotation. It would be interesting to determine whether turbulent mean profiles become unstable to a form of longitudinal instability at very large Reynolds numbers.

Funding. This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

Declaration of interests. The authors report no conflict of interest.

References

  1. [1]AKHTAR, M WASY & OSTILLA-MÓNICO, RODOLFO 2022 The effect of modulated driving on non-rotating and rotating turbulent plane couette flow. Journal of Fluid Mechanics 943, A40.arxiv.org/abs/2111.01046
  2. [2]AKHTAR, M WASY & OSTILLA-MÓNICO, RODOLFO 2024 Superposition of system response in modulated turbulent plane couette flow. Physical review fluids 9, 114606–1.arxiv.org/abs/2407.20639
  3. [3]ANDERECK, C DAVID, LIU, SS & SWINNEY, HARRY L 1986 Flow regimes in a circular couette system with independently rotating cylinders. Journal of Fluid Mechanics 164, 155–183.DOI
  4. [4]BLONDEAUX, PAOLO & VITTORI, GIOVANNA 2021 Revisiting the momentary stability analysis of the stokes boundary layer. Journal of Fluid Mechanics 919, A36.DOI
  5. [6]CHU, SHIJUN, WILLIS, ASHLEY P & MARENSI, ELENA 2024 The minimal seed for transition to convective turbulence in heated pipe flow. Journal of Fluid Mechanics 997, A46.arxiv.org/abs/2403.08985
  6. [7]Coles, Donald 1965 Transition in circular couette flow. Journal of Fluid Mechanics 21 (3), 385–425.DOI
  7. [8]Cowley, Stephen J 1987 High frequency rayleigh instability of stokes layers. In Stability of Time Dependent and Spatially Varying Flows: Proceedings of the Symposium on the Stability of Time Dependent and Spatially Varying Flows Held August 19–23, 1985, at NASA Langley Research Center, Hampton, Virginia, pp. 261–275. Springer.DOI
  8. [9]Ern, Patricia & Wesfreid, Jose Eduardo 1999 Flow between time-periodically co-rotating cylinders. Journal of Fluid Mechanics 397, 73–98.DOI
  9. [10]Kohuth, Kerry Randall & Neitzel, GP 1988 Experiments on the stability of an impulsively-initiated circular couette flow. Experiments in fluids 6 (3), 199–208.DOI
  10. [11]Lopez, Juan M & Marques, Francisco 2011 Instabilities and inertial waves generated in a librating cylinder. Journal of Fluid Mechanics 687, 171–193.DOI
  11. [12]Lucchini, Paolo & Bottaro, Alessandro 2001 Linear stability and receptivity analyses of the stokes layer produced by an impulsively started plate. Physics of Fluids 13 (6), 1668–1678.
  12. [13]Marques, Francisco & Lopez, Juan M 1997 Taylor–couette flow with axial oscillations of the inner cylinder: Floquet analysis of the basic flow. Journal of Fluid Mechanics 348, 153–175.DOI
  13. [14]Neitzel, GP, Kirkconnell, CS & Little, LJ 1995 Transient, nonaxisymmetric modes in the instability of unsteady circular couette flow. laboratory and numerical experiments. Physics of Fluids 7 (2), 324–334.DOI
  14. [15]Noir, J, Calkins, MA, Lasbleis, M, Cantwell, J & Aurnou, JM 2010 Experimental study of libration-driven zonal flows in a straight cylinder. Physics of the Earth and Planetary Interiors 182 (1-2), 98–106.DOI
  15. [16]Ostilla-Mónico, Rodolfo, Zhu, Xiaojue, Spandan, Vamsi, Verzicco, Roberto & Lohse, Detlef 2017 Life stages of wall-bounded decay of taylor-couette turbulence. Physical Review Fluids 2 (11), 114601.
  16. [17]Singh, Harminder & Prigent, Arnaud 2021 Turbulence generation and decay in the taylor–couette system due to an abrupt stoppage. Journal of Fluid Mechanics 918, A21.DOI
  17. [18]Taylor, Geoffrey Ingram 1923 Viii. stability of a viscous liquid contained between two rotating cylinders. Philosophical Transactions of the Royal Society of London. Series A, Containing Papers of a Mathematical or Physical Character 223 (605–615), 289–343.
  18. [19]Verschoof, Ruben A, Huisman, Sander G, van der Veen, Roeland CA, Sun, Chao & Lohse, Detlef 2016 Self-similar decay of high reynolds number taylor-couette turbulence. Physical review fluids 1 (6), 062402.DOI
  19. [20]Verschoof, Ruben A, te Nijenhuis, Arne K, Huisman, Sander G, Sun, Chao & Lohse, Detlef 2018 Periodically driven taylor–couette turbulence. Journal of Fluid Mechanics 846, 834–845.DOI
  20. [21]Von Kerczek, Christian & Davis, Stephen H 1974 Linear stability theory of oscillatory stokes layers. Journal of Fluid Mechanics 62 (4), 753–773.DOI
  21. [22]Willis, Ashley P 2017 The openpipeflow navier–stokes solver. SoftwareX 6, 124–127.
  22. [23]Youd, Anthony J, Willis, Ashley P & Barenghi, Carlo F 2003 Reversing and non-reversing modulated taylor–couette flow. Journal of Fluid Mechanics 487, 367–376.DOI

Paper details

vlnc:W_yqbmvr2z
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents