Flow-Induced Vibration of Cross-Shaped Tube Bundle: The Effect of Tube Bundle Arrangement

2003 ◽  
Author(s):  
Fumio Inada ◽  
Takashi Nishihara ◽  
Akira Yasuo ◽  
Ryo Morita ◽  
Akihiro Sakashita ◽  
...  

A cross-shaped tube bundle is proposed for the lower plenum structure in the next-generation LWR. The effect of tube bundle arrangement on the flow-induced vibration characteristics of the cross-shaped tube bundle in cross flow was considered experimentally. Regarding random vibration, the power spectral density of the fluid force of the staggered arrangement as well as the correlation length was measured and those of the normal arrangement were compared with those of the staggered arrangement. Regarding self-excited vibration, vibration response was compared. The trend of the power spectral densities, correlation length, and the critical velocity of the normal arrangement were similar to those of the staggered arrangement.

Author(s):  
Fumio Inada ◽  
Takashi Nishihara ◽  
Akira Yasuo ◽  
Ryo Morita ◽  
Akihiro Sakashita ◽  
...  

Cross-shaped tube bundle is proposed for a lower plenum structure in a next generation LWR. Vibration response of cross-shaped tube bundle in cross flow has been measured in water tunnel tests. First, small-scale test was conducted. Tests were conducted with 3×3 flexible tubes as well as single flexible tube in rigid tube bundle. The flexible tubes could vibrate in lift, drag, and torsional direction. The effect of arrangements of tube bundle and the natural frequency ratio of bending and torsional vibrations were considered. Second, a large-scale test was conducted for only one case to check the effect of Reynolds number, in which Reynolds number was 10 times larger than that of small-scale test. In all the cases, large amplitude vibration could appear when the flow velocity became larger than a critical value, and a self-excited vibration was found to occur. The nondimensional critical gap velocity of the large-scale test agreed well with that of the small-scale test, which suggested that the effect of Reynolds number was not so large. A design guideline to prevent self-excited vibration was proposed for cross-shaped tube bundle.


2003 ◽  
Vol 18 (5) ◽  
pp. 651-661 ◽  
Author(s):  
F. Inada ◽  
T. Nishihara ◽  
A. Yasuo ◽  
R. Morita ◽  
A. Sakashita ◽  
...  

Author(s):  
Fumio Inada ◽  
Takashi Nishihara ◽  
Jun Mizutani

A cross-shaped control rod guide tube bundle is proposed for the lower plenum structure in the next-generation LWR, ABWR-II. In our previous studies, we measured the local fluid excitation forces acting on a cross-shaped tube bundle as well as the self-excited vibration characteristics in pure cross flow in water tunnel tests. In the reactor conditions, the flow field around the tube bundles contains mixed cross and parallel flow components. In this study, water tunnel tests under mixed cross and parallel flow conditions were preformed to understand the influence of the balance of parallel and cross flow components on vibration response. The distributions of the flow direction and flow velocity in the gap between the adjacent tubes were measured with circular Pilot tubes in detail. It was found that the critical flow velocity of self-excited vibration was not influenced by the parallel flow component, but depended only on the cross flow component.


2006 ◽  
Vol 129 (1) ◽  
pp. 162-168
Author(s):  
Fumio Inada ◽  
Kimitoshi Yoneda ◽  
Akira Yasuo ◽  
Takashi Nishihara

The local fluid excitation force acting on a rotated square tube bundle having transverse pitch-to-diameter ratio of T∕D=3.1, in a single-phase cross-flow was measured, and the normalized power spectral density (NPSD) and correlation length in the axial direction of a tube were examined. The fluid excitation force acting on the interior tube was from three to ten times larger than that acting on the upstream tube. The fluid force was almost fully developed after the third row. NPSD of the fluid excitation force could be almost plotted on a single universal curve. Regarding the lift direction, there was a peak in NPSD at fD∕u∼0.3 caused by vortex shedding. Regarding the drag direction, there could be another peak in NPSD around twice the vortex shedding frequency. In the region of fD∕u>0.5, where the effect of the vortex shedding was assumed to be small in the lift direction, the correlation length of the lift direction was ∼1.1D. NPSD was a little larger than previous results for tube bundles of relatively small pitch to diameter ratios summarized by Axisa, Antunes, and Villard (1990, J. Fluid Struct., 4, pp. 321–341).


Author(s):  
Takashi Nishihara ◽  
Fumio Inada ◽  
Akira Yasuo ◽  
Ryo Morita ◽  
Akihiro Sakashita ◽  
...  

A cross-shaped tube bundle with dense arrangement may be designed for a lower plenum structure in a next generation LWR, though the characteristics of flow-induced vibration of this type of tube bundle remain virtually unknown. In this study, turbulence-induced fluid dynamic forces acting on a cross-shaped tube bundle with a dense arrangement subject to cross flow were measured by water tunnel tests with two types of scale models. One is a small-scale model to measure local fluid dynamic forces and their correlation length in the lift and drag direction. The other is a large-scale model to investigate the effect of the Reynolds number on fluid dynamic forces in the lift, drag and torsional directions. Free oscillation tests with another small-scale model were also conducted to measure vibration amplitude by random excitation force. In conclusion, the following results were obtained. Vortex-induced vibration cannot arise in the cross-shaped tube bundle, since a typical peak corresponding to periodic vortex shedding was not observed in power spectral density for fluid excitation force. Power spectral densities of fluid dynamic forces in the drag, lift and torsional directions have mutually similar properties and they are hardly dependent on the Reynolds number. The experimental results were compiled into dimensionless correlation equations composed of the power spectral density for the local fluid excitation force and its correlation length. They are useful for evaluating the random vibration amplitude. The estimated amplitudes of turbulence-induced vibration by the correlation equation coincide with those of the experimental results obtained by the free-oscillation tests.


Author(s):  
H. Senez ◽  
N. W. Mureithi ◽  
M. J. Pettigrew

Two-phase cross flow exists in many shell-and-tube heat exchangers. Flow-induced vibration excitation forces can cause tube motion that will result in long-term fretting wear or fatigue. Detailed flow and vibration excitation force measurements in tube bundles subjected to two-phase cross flow are required to understand the underlying vibration excitation mechanisms. Studies on this subject have already been done, providing results on flow regimes, fluidelastic instabilities, and turbulence-induced vibration. The spectrum of turbulence-induced forces has usually been expected to be similar to that in single-phase flow. However, a recent study, using tubes with a diameter larger than that in a real steam generator, showed the existence of significant quasi-periodic forces in two-phase flow. An experimental program was undertaken with a rotated-triangular array of cylinders subjected to air-water cross-flow, to simulate two-phase mixtures. The tube bundle here has the same geometry as that of a real steam generator. The quasi-periodic forces have now also been observed in this tube bundle. The present work aims to understand turbulence-induced forces acting on the tube bundle, providing results on drag and lift force spectra and their behaviour according to flow parameters, and describing their correlations. Detailed experimental test results are presented in this paper. Comparison is also made with previous measurements with larger diameter tubes. The present results suggest that quasi-periodic fluid forces are not uncommon in tube arrays subjected to two-phase cross-flow.


Author(s):  
Shahab Khushnood ◽  
Zaffar M. Khan ◽  
M. Afzaal Malik ◽  
Zafarullah Koreshi ◽  
Mahmood Anwar Khan

Flow-induced vibration in steam generator and heat exchanger tube bundles has been a source of major concern in nuclear and process industry. Tubes in a bundle are the most flexible components of the assembly. Flow induced vibration mechanisms, like fluid-elastic instability, vortex shedding, turbulence induced excitation and acoustic resonance results in failure due to mechanical wear, fretting and fatigue cracking. The general trend in heat exchanger design is towards larger exchangers with increased shell side velocities. Costly plant shutdowns have been the motivation for research in the area of cross-flow induced vibration in steam generators and process exchangers. The current paper focuses on the development of a computer code (FIVPAK) for the design (natural frequencies, variable geometry, tube pitch & pattern, mass damping parameter, reduced velocity, strouhal and damage numbers, added mass, wear work rates, void fraction for two-phase, turbulence and acoustic considerations etc.) of tube bundles with respect to cross flow-induced vibration. The code has been validated against Tubular Exchanger Manufacturers (TEMA), Flow-Induced Vibration code (FIV), and results on an actual variable geometry exchanger, specially manufactured to simulate real systems. The proposed code is expected to prove a useful tool in designing a tube bundle and to evaluate the performance of an existing system.


Author(s):  
Paul Feenstra ◽  
Teguewinde Sawadogo ◽  
Bruce Smith ◽  
Victor Janzen ◽  
Helen Cothron

The tubes in the U-bend region of a recirculating type of nuclear steam generator are subjected to cross-flow of a two-phase mixture of steam and water. There is a concern that these tubes may experience flow-induced vibration, including the damaging effects of fluidelastic instability. This paper presents an update and results from a series of flow-induced vibration experiments performed by Canadian Nuclear Laboratories for the Electric Power Research Institute (EPRI) using the Multi-Span U-Bend test rig. In the present experiments, the main focus was to investigate fluidelastic instability of the U-tubes subjected to a cross-flow of air. The tube bundle is made of 22 U-tubes of 0.5 in (12.7 mm) diameter, arranged in a rotated triangular configuration with a pitch-over-diameter ratio of 1.5. The test rig could be equipped with variable clearance flat bar supports at two different locations to investigate a variety of tube and support configurations. The primary purpose of the overall project is to study the effect of flat bar supports on ‘in plane’ (‘streamwise’) instability in a U-tube bundle with realistic tube-to-support clearances or preloads, and eventually in two-phase flow conditions. Initially, the test rig was designed for tests in air-flow using an industrial air blower. Tests with two-phase Freon refrigerant (R-134a) will follow. This paper describes the test rig, experimental setup, and the challenges presented by simulating an accurate representation of current steam generator designs. Results from the first series of tests in air flow are described.


2001 ◽  
Vol 44 (2) ◽  
pp. 19-22
Author(s):  
Jason Tolomeo ◽  
Y. Lee

Designing an appropriate shaker random vibration test to accurately simulate a high-intensity acoustic environment is complicated by the fundamental differences in the nature of acoustic excitation and baseshake input, as well as test fixture and boundary condition differences. A coupon random vibration test of a spacecraft panel with an imbedded heatpipe component was performed to simulate the expected acoustic environment at liftoff and flight. Analytical predictions of the acoustic response of the full panel were made and then compared with predictions of the baseshake response of the coupon configuration at specific locations. An appropriate shaping profile was then designed for the random vibration test. The resulting responses produced by the random vibration test were found to be a good representation of the acoustic responses in both overall response levels and power spectral densities.


Sign in / Sign up

Export Citation Format

Share Document