A set of experiments has been conducted, in water, on long circular cylinders at sub-critical and critical Reynolds numbers typical of production risers in the Gulf of Mexico (ranging from about 1.5x105 to 3.4x105) experiencing currents. These experiments were conducted in the Rotating Arm Facility at the Naval Surface Warfare Center in Carderock, Maryland. The cylinders were mounted under, and parallel to, the arm that provided a (approximately) linear sheared flow relative to the cylinders. Drag, acceleration, and tension were all measured to determine the cylinders' vortex-induced vibration response. Potentially excited transverse bending modes ranged from 1 to 15.
Vortex-induced vibration (VIV) of cylindrical structures is a fairly well known phenomenon that can cause rapid fatigue failure and/or excessive drag on a structure. In the ocean, many cylindrical structures are potentially subject to VIV from ocean currents. These structures include risers, tendons, mooring lines and the hull of a spar-type structure.
One of the most important parameters affecting VIV is known as the Reynolds number, defined as Re=V*D/?, where: V is the relative velocity of the flowing fluid experienced by the cylinder; D is the cylinder outside diameter (including coatings); and ??is the kinematic viscosity of the fluid. Since the Reynolds number is proportional to both diameter and velocity, performing tests at Reynolds numbers that correspond to those of offshore risers and tendons experiencing high ocean currents (for which the velocity and diameter are often large) is quite difficult. Note that this problem is especially important for Reynolds numbers typical of offshore production risers, since the boundary layers typically become turbulent for this Reynolds number range (larger than 100,000 or so).
Another important parameter affecting VIV fatigue damage estimates is the mode number of vibration. While shallowwater riser spans often experience only first mode vibration, deepwater risers can experience very high modes of vibration and can also experience more than one mode of vibration at a time. Thus, it is important that relevant experiments for VIV of deepwater risers have fairly high potentially excited mode numbers, in addition to appropriate Reynolds numbers. Note that, unless otherwise designated, in this paper "mode" and "transverse bending mode" are used interchangeably. This is because most VIV modeling concentrates on the larger transverse (normal to flow) motions. The authors recognize that the in-line vibrations are equally important, and bothdirections were measured during these experiments.
While several researchers have performed tests on fixed cylinders at high Reynolds numbers (e.g. Jones et. Al1 and Shih2), and while the authors and others have performed VIV tests on flexible cylinders at low Reynolds numbers (e.g. Allen3–7), controlled VIV experiments on flexible cylinders at high Reynolds numbers have mostly been limited to those by the authors (Allen and Henning8 and numerous unpublished experiments by the authors). An exception was the Stride Joint Industry Program experiments (Willis and Thethi9) thatconsisted of a long pipe towed offshore by tugs using long cables.