An Unsteady Theory For Fluidelastic Instability in an Array of Flexible Tubes in Cross-Flow. Part I: Theory

1993 ◽  
Vol 7 (7) ◽  
pp. 751-766 ◽  
Author(s):  
M. Yetisir ◽  
D.S. Weaver
2015 ◽  
Vol 285 ◽  
pp. 58-64 ◽  
Author(s):  
Nai-bin Jiang ◽  
Bin Chen ◽  
Feng-gang Zang ◽  
Yi-xiong Zhang

Author(s):  
Stephen Olala ◽  
Njuki Mureithi

In-plane fluidelastic instability is a dynamic phenomenon requiring fluid coupling of at least two degrees-of-freedom, in this case, at least two flexible tubes. Due to the nature of the mechanism causing streamwise fluidelastic instability, a purely experimental or an unsteady determination would require intensive experimental effort. As a compromise between experimental effort and prediction accuracy, the quasi-steady model is used in the current study. In the present work, previously measured quasi-steady and unsteady forces are used to estimate the time delay first between the displacement of an oscillating tube and the forces generated on itself then the time delay between the displacement of a central oscillating tube and the forces induced on the adjacent tubes. The estimated time delays are then used together with drag coefficients and derivatives to predict the in-plane fluidelastic instability in a rotated triangular tube array of P/D = 1.5 subjected to two-phase flow. The results closely replicate dynamic test results and confirm the predominance of the stiffness controlled mechanism and the potential of the quasi-steady model in accurately predicting streamwise fluidelastic instability in arrays subjected to two-phase flows.


Author(s):  
T. F. Joly ◽  
N. W. Mureithi ◽  
M. J. Pettigrew

Tests were done to study the effect of angle of attack on the fluidelastic instability of a fully flexible tube bundle subjected to two-phase (Air-Water) cross-flow. A test array having nineteen flexible tubes in a rotated triangular configuration with a pitch-to-diameter ratio of 1.5 was tested. Four different angles of attack ranging for 0 degree (inline flexibility) through 30 and 60 degrees to 90 degrees (transverse flexibility) were studied. For each angle of attack several homogeneous void fractions have been tested (70%, 80%, 90%, and 95%). Stability test results show that the angle of attack strongly affect the tube bundle dynamic behavior. The different mechanisms underlying the fluidelastic instability are highlighted and the results compared to existing data on fluidelastic instability.


2005 ◽  
Vol 128 (1) ◽  
pp. 148-159 ◽  
Author(s):  
R. Violette ◽  
M. J. Pettigrew ◽  
N. W. Mureithi

Almost all the available data about fluidelastic instability of heat exchanger tube bundles concerns tubes that are axisymmetrically flexible. In those cases, the instability is found to be mostly in the direction transverse to the flow. Thus, the direction parallel to the flow has raised less concern in terms of bundle stability. However, the flat bar supports used in steam generator for preventing U-tube vibration may not be as effective in the in-plane direction than in the out-of-plane direction. The possibility that fluidelastic instability can develop in the flow direction must then be assessed. In the present work, tests were done to study the fluidelastic instability of a cluster of seven tubes much more flexible in the flow direction than in the lift direction. The array configuration is rotated triangular with a pitch to diameter ratio of 1.5. The array was subjected to two-phase (air-water) cross flow. Fluidelastic instability was observed when the flexible tubes were located at the center of the test section and also when the seven flexible tubes were placed over two adjacent columns. No instability was found for a single flexible tube in a rigid array, nor for the case where the seven flexible tubes were placed in a single column. Tests were also done with tubes that are axisymmetrically flexible for comparison purposes. It was found that fluidelastic instability occurs at higher velocities when the tubes are flexible only in the flow direction. These results and additional wind tunnel results are compared to existing data on fluidelastic instability. Two-phase flow damping results are also presented in this paper.


2002 ◽  
Vol 16 (2) ◽  
pp. 153-173 ◽  
Author(s):  
K. HIROTA ◽  
T. NAKAMURA ◽  
J. KASAHARA ◽  
N.W. MUREITHI ◽  
T. KUSAKABE ◽  
...  

2010 ◽  
Vol 132 (4) ◽  
Author(s):  
Marwan Hassan ◽  
Achraf Hossen

This paper presents simulations of a loosely supported cantilever tube subjected to turbulence and fluidelastic instability forces. Several time domain fluid force models are presented to simulate the damping-controlled fluidelastic instability mechanism in tube arrays. These models include a negative damping model based on the Connors equation, fluid force coefficient-based models (Chen, 1983, “Instability Mechanisms and Stability Criteria of a Group of Cylinders Subjected to Cross-Flow. Part 1: Theory,” Trans. ASME, J. Vib., Acoust., Stress, Reliab. Des., 105, pp. 51–58; Tanaka and Takahara, 1981, “Fluid Elastic Vibration of Tube Array in Cross Flow,” J. Sound Vib., 77, pp. 19–37), and two semi-analytical models (Price and Païdoussis, 1984, “An Improved Mathematical Model for the Stability of Cylinder Rows Subjected to Cross-Flow,” J. Sound Vib., 97(4), pp. 615–640; Lever and Weaver, 1982, “A Theoretical Model for the Fluidelastic Instability in Heat Exchanger Tube Bundles,” ASME J. Pressure Vessel Technol., 104, pp. 104–147). Time domain modeling and implementation challenges for each of these theories were discussed. For each model, the flow velocity and the support clearance were varied. Special attention was paid to the tube/support interaction parameters that affect wear, such as impact forces and normal work rate. As the prediction of the linear threshold varies depending on the model utilized, the nonlinear response also differs. The investigated models exhibit similar response characteristics for the lift response. The greatest differences were seen in the prediction of the drag response, the impact force level, and the normal work rate. Simulation results show that the Connors-based model consistently underestimates the response and the tube/support interaction parameters for the loose support case.


Sign in / Sign up

Export Citation Format

Share Document