Numerical Simulation of Tube Bundle Vibrations in Cross Flows

2002 ◽  
Author(s):  
E. Longatte ◽  
Z. Bendjeddou ◽  
M. Souli
Keyword(s):  
Numerical Simulation ◽  
Dynamic Behaviour ◽  
Tube Bundle ◽  
Numerical Estimate ◽  
Time History ◽  
Numerical Prediction ◽  
Critical Flow Velocity ◽  
Ale Formulation ◽  
Elastic Forces ◽  
Tube Bundles

In many industrial applications, mechanical structures like heat exchanger tube bundles are subjected to complex flows causing possible vibrations and damage. Part of fluid forces are coupled with tube motion and these so-called fluid-elastic forces can affect the structure dynamic behaviour generating possible instabilities and leading to short term failures through high amplitude vibrations. Most classical fluid force identification methods rely on structure response experimental measurements associated with convenient data processes. Owing to recent improvements in Computational Fluid Dynamics, numerical simulation of flow-induced vibrations is now practicable for industrial purposes. The present paper is devoted to the computation of fluid-elastic forces acting on tube bundles subjected to one-phase cross flows. What is the numerical process ? In the case where fluid-elastic effects are not significant and are restricted to added mass effects, there is no real coupling between structure and fluid motion. The structure displacement is not supposed to affect flow patterns. Thus it is possible to solve the fluid and the structure problems separately by using a fixed non-moving mesh for the fluid dynamic computation. Lift and drag forces acting on tube bundles can be computed numerically by using Large Eddy Simulation. Their spectrum and time history can be introduced as inlet conditions in the mechanical calculation providing the tube vibratory response. On the contrary when fluid-elastic effects can not be neglected, in presence tube bundles subjected to cross flows for example, a coupling between flow and structure computations is required. Such a calculation is performed in the present work. An improved numerical approach has been developed and applied to the fully numerical prediction of the dynamic behaviour of a flexible tube belonging to a fixed tube bundle subjected to cross flows. The purpose is to be able to provide a numerical estimate of the critical flow velocity for the threshold of fluidelastic instability of tube bundle without experimental investigation. The methodology consists in simulating in the same time thermohydraulics and mechanics problems by using an Arbitrary Euler Lagrange (ALE) formulation for the fluid computation. Numerical results turn out to be consistent with available experimental data obtained in the same configuration. This work is a first step in the numerical prediction of tube bundle vibrations in presence of cross flows.

Numerical Study of a Tube Bundle Vibrations in Cross-Flows: ALE and Transpiration Methods

2006 ◽  
Author(s):  
Marcus Vinicius G. de Morais ◽  
Rene-Jean Gibert ◽  
Franck Baj ◽  
Jean-Paul Magnaud
Keyword(s):  
Numerical Study ◽  
Tube Bundle ◽  
Numerical Estimate ◽  
Numerical Approach ◽  
Numerical Results ◽  
Small Vibration ◽  
Critical Flow Velocity ◽  
Simplified Approach ◽  
Transpiration Method ◽  
Elastic Forces

In this paper, we compare the performances of ALE and Transpiration methods. The ALE approach is a powerful tool to treat coupled problems. We can mention for ALE, more precisely, the approach in finite elements of Donea and Hughes. However, the ALE performance for determining fluid-elastic forces to small vibration amplitudes is still ignored. The Transpiration method is a simplified approach for calculating fluid-elastic forces to relatively small vibration amplitudes. Based on a first order development of velocity boundary conditions, this method allows the use of a fluid domain fixed in time during a dynamic computation, by avoiding the problems due to the mesh distortions. The purpose of this work is to provide a numerical estimate of the critical flow velocity for the threshold of fluid-elastic instability of tube bundle without experimental investigation. A staggered coupled numerical approach is suggested and applied to the numerical prediction of the vibration frequency of a flexible tube belonging to a fixed tube bundle in fluid flow. Numerical results turn out to be consistent with available experimental data obtained in the same configuration. This work presents our numerical results for a prediction of tube bundle vibrations induced by flows implemented in CAST3M, a numerical platform of French Nuclear Agency (CEA-Saclay).

Experimental Investigation of Fluid-Elastic Vibration of Square Array Tube Bundle in Two Phase Flow

2019 ◽  
Author(s):  
Ryoichi Kawakami ◽  
Seinosuke Azuma ◽  
Toshifumi Nariai ◽  
Kazuo Hirota ◽  
Hideyuki Morita ◽  
...  
Keyword(s):  
Two Phase Flow ◽  
Tube Bundle ◽  
Flow Direction ◽  
Phase Flow ◽  
Elastic Instability ◽  
Steam Generators ◽  
Two Phase ◽  
Critical Flow Velocity ◽  
Tube Bundles ◽  
Square Array

Abstract The in-plane (in-flow) fluid-elastic instability (in-plane FEI) of triangular tube arrays caused tube-to-tube wear indications as observed in the U-bend regions of tube bundles of the San Onofre Unit-3 steam generators[1]. Several researches revealed that the in-plane FEI is likely to occur in a tightly packed triangular tube array under high velocity and low friction conditions, while it is not likely to occur in a square array tube bundle. In order to confirm the potential of steam-wise fluid-elastic instability of square arrays, the critical flow velocity in two-phase flow, (sulfur hexafluoride-ethanol) which simulates steam-water flow, was investigated. Two types of test rigs were prepared to confirm the effect of the tube diameter and tube pitch ratio on the critical velocity. In both rigs, vibration amplitudes were measured in both in-flow and out-of-flow directions in various flow conditions. In any case, in-flow fluid elastic instability was not detected. Based on the results of the tests, it is concluded that the flow interaction force is small for concern to occur the fluid-elastic instability in the in-flow direction of the square tube bundles of steam generators.

Fluctuating Lift Forces of the Karman Vortex Streets on Single Circular Cylinders and in Tube Bundles: Part 3—Lift Forces in Tube Bundles

1972 ◽  
Vol 94 (2) ◽  
pp. 623-628 ◽  
Author(s):  
Y. N. Chen
Keyword(s):  
Second Harmonic ◽  
Tube Bundle ◽  
Lift Coefficient ◽  
Reynolds Numbers ◽  
Circular Cylinders ◽  
Critical Flow Velocity ◽  
Pressure Drag ◽  
Tube Bundles ◽  
Karman Vortex ◽  
Lift Forces

The trend of the fluctuating lift coefficient CL and the dimensionless shedding frequency S (Strouhal number) of the vortex in tube bundles at higher Reynolds numbers R will be predicted by the course of the steady pressure drag coefficient CD at the corresponding R ranges. Furthermore, some measurements of the vortex lift forces in tube bundles will be given. It reveals that the lift force for certain small transverse tube spacings possesses a strong second harmonic. The tubes and, therefore, the transverse gas column in the tube bundle channel can be excited to vibrate in resonance either at the critical flow velocity or at its half value. Finally, the coupled vibration between the vortex shedding and the transverse gas column will be covered with some experiments.

Fluid Structure Interactions for Tube Bundles: Physical Meaning of a Rayleigh Damping

2010 ◽  
Author(s):  
Daniel Broc ◽  
Jean Franc¸ois Sigrist
Keyword(s):  
Fluid Flow ◽  
Numerical Simulations ◽  
Physical Meaning ◽  
Dynamic Behaviour ◽  
Stokes Equations ◽  
Tube Bundle ◽  
Inertial Effects ◽  
Rayleigh Damping ◽  
Dissipative Effects ◽  
Tube Bundles

It is well known that a fluid may strongly influence the dynamic behaviour of a structure. Many different physical phenomena may take place, depending on the conditions: fluid flow, fluid at rest, small or high displacements of the structure. Inertial effects can take place, with lower vibration frequencies, dissipative effects also, with damping, instabilities due to the fluid flow (Fluid Induced Vibration). In this last case the structure is excited by the fluid. The paper deals with the vibration of tube bundles under a seismic excitation or an impact. In this case the structure moves under an external excitation, and the movement is influenced by the fluid. The main point in such system is that the geometry is complex, and could lead to very huge sizes for a numerical analysis. Important developments have been made in the last years to develop homogenization methods for the dynamic behaviour of tube bundles. The numerical size of the problem is reduced, and it is possible to make numerical simulations on large tube bundles with reasonable computer times. These methods consider that the fluid movement is governed by the Euler equations for the fluid. They are based on an analysis on an elementary cell, corresponding to one tube, and on an expression of the forces applied by the fluid to the structure. This force only depends on the fluid’s and tube’s acceleration. Only “inertial effects” will theoretically take place, with globally lower frequencies. A research program is under progress to take into account dissipative effects also, with a homogenization of the Navier-Stokes equations in the tube bundle. It is common, in numerical simulations, to add a damping for the structures by using a global Rayleigh damping. The paper deals with the physical meaning of this Rayleigh damping in the Euler homogenized equations. It can be demonstrated that this damping corresponds to a force applied by the fluid to the structure depending not only on the acceleration, but on the fluid and structure velocity also. This Rayleigh damping is a first step to take into account the dissipative effects for FSI in tube bundles.

Vibration of Tube Bundles in Two-Phase Cross-Flow: Part 2—Fluid-Elastic Instability

1989 ◽  
Vol 111 (4) ◽  
pp. 478-487 ◽  
Author(s):  
M. J. Pettigrew ◽  
J. H. Tromp ◽  
C. E. Taylor ◽  
B. S. Kim
Keyword(s):  
Tube Bundle ◽  
Cross Flow ◽  
Design Guidelines ◽  
Experimental Program ◽  
Elastic Instability ◽  
Two Phase ◽  
Critical Flow Velocity ◽  
Flexible Tube ◽  
Elastic Instabilities ◽  
Tube Bundles

An extensive experimental program was carried out to study the vibration behavior of tube bundles subjected to two-phase cross-flow. Fluid-elastic instability is discussed in Part 2 of this series of three papers. Four tube bundle configurations were subjected to increasing flow up to the onset of fluid-elastic instability. The tests were done on bundles with all-flexible tubes and on bundles with one flexible tube surrounded by rigid tubes. Fluid-elastic instabilities have been observed for all tube bundles and all flow conditions. The critical flow velocity for fluid-elastic instability is significantly lower for the all-flexible tube bundles. The fluid-elastic instability behavior is different for intermittent flows than for continuous flow regimes such as bubbly or froth flows. For continuous flows, the observed instabilities satisfy the relationship V/fd = K(2πζm/ρd2)0.5 in which the minimum instability factor K was found to be around 4 for bundles of p/d = 1.47 and significantly less for p/d = 1.32. Design guidelines are recommended to avoid fluid-elastic instabilities in two-phase cross-flows.

Steam Generator Tube Vibrations: Experimental Determination Versus ALE Computation of Fluidelastic Forces

2003 ◽  
Author(s):  
Z. Bendjeddou ◽  
E. Longatte ◽  
A. Adobes ◽  
M. Souli
Keyword(s):  
Experimental Determination ◽  
Moving Boundary ◽  
Reynolds Numbers ◽  
High Amplitude ◽  
Industrial Applications ◽  
Critical Flow Velocity ◽  
Tube Arrays ◽  
Elastic Forces ◽  
Tube Bundles ◽  
High Reynolds

In heat exchanger tube bundles like in many others industrial applications, fluid structure interaction is a crucial problem to overcome. Flow-induced tube vibration in tube bundles is due to two main kinds of physical effects: (1) fluid-elastic forces caused by structure motion; (2) turbulent forces due to vortex generation at high Reynolds numbers. The second component, turbulent excitation, is independent on structure motion and may generate wear and fatigue damage while the first component may lead to fluid-elastic instability inducing high amplitude displacement and possible tube short term failure. In this context many studies are carried out in order to develop methods for the identification of critical flow velocity in tube arrays. In the present work two methods are presented: (1) the first one relies on experimental measurements, it is fitted with analytical modeling and provides fluid-elastic coefficients; (2) the second one relies on numerical simulation using Computational Fluid Dynamics Codes (CFD) involving moving boundary techniques; it provides fluid force estimates and in some cases it makes it possible to simulate tube vibrations. The first part is devoted to experimental determination of fluid-elastic forces. A numerical method for prediction of fluid-elastic effects in fluid at rest is presented in the second section. Results of both methods are compared in the third part.

A Global Model for Flow-Induced Vibration of Tube Bundles in Cross-Flow

1991 ◽  
Vol 113 (3) ◽  
pp. 446-458 ◽  
Author(s):  
S. Granger
Keyword(s):  
Tube Bundle ◽  
Damping Ratio ◽  
Cross Flow ◽  
Global Model ◽  
Stable Region ◽  
Predictive Analysis ◽  
System Response ◽  
Critical Flow Velocity ◽  
Global System ◽  
Tube Bundles

This paper presents an approximate model which can be used for predictive analysis of industrial tube bundles subjected to cross-flow. A tube bundle in cross-flow is locally approximated, in a global sense, by a single-degree-of-freedom system, called the global system. The critical flow velocity can be predicted by computing the velocity at which the damping ratio of the global system becomes zero. In the stable region, tube response amplitude can be approximated by the amplitude of the global system response. In this approach, the knowledge of four fluid force coefficients is required to solve the problem. They are determined experimentally by dynamic response measurements. The global model so defined is compared with the simplified method usually used for industrial predictive analysis purposes. It is shown that the conventional method can be considered as a particular simplified case of the present model. Practical examples concerning square-in-line tube bundles are given. They show that the first results obtained with the global model are promising.

Critical Velocity of a Non-Linearly Supported Multi-Span Tube Bundle

2006 ◽  
Author(s):  
M. K. Au-Yang ◽  
J. A. Burgess
Keyword(s):  
Flow Velocity ◽  
Critical Velocity ◽  
Tube Bundle ◽  
Normal Modes ◽  
Cross Flow ◽  
Time History ◽  
Tube Bundles ◽  
Cross Flow Velocity ◽  
The Cross ◽  
The Time Domain

The phenomenon of fluid-elastic instability and the velocity at which a heat exchanger tube bundle becomes unstable, known as the critical velocity, was discovered and empirically determined based upon single-span, linearly supported tube bundles. In this idealized configuration, the normal modes are well separated in frequency with negligible cross-modal contribution to the critical velocity. As a result, a critical velocity can be defined and determined for each mode. In an industrial heat exchanger or steam generator, not only do the tube bundles have multiple spans, they are also supported in over-sized holes. The normal modes of a multi-span tube bundle are closely spaced in frequency, and the non-linear effect of the tube-support plate interaction further promotes cross-modal contribution to the tube responses. The net effect of cross-modal participation in the tube vibration is to delay the instability threshold. Tube bundles in industrial exchangers often have critical velocities far above what were determined in the laboratory based upon single-span, linearly supported tube bundles. In this paper, the authors attempt to solve this non-linear problem in the time domain, using a time history modal superposition method. Time history forcing functions are first obtained by inverse Fourier transform of the power spectral density function used in classical turbulence-induced vibration analyses. The fluid-structure coupling force, which is dependent on the cross-flow velocity, is linearly superimposed onto the turbulence forcing function. The tube responses are then computed by direct integration in the time domain. By gradually increasing the cross-flow velocity, a threshold value is obtained at which the tube response just starts to diverge. The value of the cross-flow velocity at which the tube response starts to diverge is defined as the critical velocity of this non-linearly supported, multi-span tube bundle.

scholarly journals Numerical Simulation of Tower Rotor Interaction for Downwind Wind Turbine

2010 ◽  
Vol 2010 ◽  
pp. 1-11 ◽  
Author(s):  
Isam Janajreh ◽  
Ilham Talab ◽  
Jill Macpherson
Keyword(s):  
Numerical Simulation ◽  
Cross Section ◽  
Cross Sections ◽  
High Speed ◽  
Stokes Equations ◽  
Circular Cross Section ◽  
Time History ◽  
Navier Stokes ◽  
Cross Sectional ◽  
Ale Formulation

Downwind wind turbines have lower upwind rotor misalignment, and thus lower turning moment and self-steered advantage over the upwind configuration. In this paper, numerical simulation to the downwind turbine is conducted to investigate the interaction between the tower and the blade during the intrinsic passage of the rotor in the wake of the tower. The moving rotor has been accounted for via ALE formulation of the incompressible, unsteady, turbulent Navier-Stokes equations. The localizedCP,CL, andCDare computed and compared to undisturbed flow evaluated by Panel method. The time history of theCP, aerodynamic forces (CLandCD), as well as moments were evaluated for three cross-sectional tower; asymmetrical airfoil (NACA0012) having four times the rotor's chord length, and two circular cross-sections having four and two chords lengths of the rotor's chord. 5%, 17%, and 57% reductions of the aerodynamic lift forces during the blade passage in the wake of the symmetrical airfoil tower, small circular cross-section tower and large circular cross-section tower were observed, respectively. The pronounced reduction, however, is confined to a short time/distance of three rotor chords. A net forward impulsive force is also observed on the tower due to the high speed rotor motion.

scholarly journals The dynamic behaviour of Archimede’s Bridges: Numerical simulation and design implications

2010 ◽  
Vol 4 ◽  
pp. 91-98 ◽  
Author(s):  
Federico Perotti ◽  
Gianluca Barbella ◽  
Mariagrazia Di Pilato
Keyword(s):  
Numerical Simulation ◽  
Dynamic Behaviour
Sign in / Sign up

Export Citation Format

Share Document