High-Mode-Number Vortex-Induced-Vibration Field Experiments

Abstract

This article, written by Technology Editor Dennis Denney, contains highlights of paper OTC 17383, "High-Mode-Number Vortex-Induced-Vibration Field Experiments," by J.K. Vandiver, Massachusetts Inst. Of Technology; H. Marcollo, AMOG Consulting; and S. Swithenbank and V. Jhingran, Massachusetts Inst. of Technology, prepared for the 2005 Offshore Technology Conference, Houston, 2-5 May. This paper presents the initial results from vortex-induced-vibration (VIV) field testing of a long, flexible, model riser at a high mode number. The experiments were designed to gain understanding of the dynamic behavior of a long riser in uniform flow responding at mode numbers ranging from 10 to 25 in crossflow vibration. The observed reduced velocity and root-mean-square (RMS) displacement-response levels were reported. Mean values of the drag coefficient, Cd, and hydrodynamic damping derived from measured data were compared with calculated values from formulas commonly used in engineering design of offshore systems. Introduction Experiments at Lake Seneca, New York, took place in the summer of 2004. They were part of a larger testing program developed by Deepstar (a joint-industry technology-development project) for improving the ability to model and mitigate VIV. While these particular tests were focused on investigating uniform-flow conditions, a second set of tests designed for shear-flow conditions was performed subsequently in November 2004 and will be reported separately. The initial motivation for this field experiment was to improve the under-standing of VIV at high mode numbers. Most model testing has been conducted at low mode numbers (<10). Drilling and production in 1000- to 3000-m depths requires understanding VIV behavior at higher mode numbers but still short of infinite-length behavior. Experiment Description The Lake Seneca test facility is a fully equipped field-test station moored in calm, deep water. It was ideal for conducting a controlled test on a long, circular pipe in uniform flow. As Fig. 1 shows, the tests were accomplished by towing a vertical, composite pipe with a suspended bottom weight to produce the desired tension. The length, diameter, and tension of the pipe were chosen to permit crossflow excitation of up to the 25th mode. The maxi-mum speed possible with the system was limited by the maximum allowable deflection angle of the pipe. Typical towing speeds were approximately 0.3 to 1.1 m/s. The pipe was constructed in 30.5-m-long segments, which were joined together. Total pipe length was limited to 137 m by the depth of Lake Seneca. During the experiment, total lengths of 61.26 and 122.23 m were tested. Each 30.5-m-long section of pipe contained six evenly spaced triaxial accelerometers. The sampling rate was 60 Hz per accelerometer. The same serial network was used to sample the tension- and tilt-measuring devices very close to the top U-joint, connection. Two mechanical current meters measured towing speed; one was suspended underneath the towed weight, and the other hung over the side of the towing vessel. A load cell and tiltmeter were attached at the top of the pipe. This instrumentation allowed measuring the tension in the pipe and the top angle of inclination.