Lift and Damping Characteristics of Bare and Straked Cylinders at Riser Scale Reynolds Numbers

Abstract

ExxonMobil recently completed a series of Vortex-Induced Vibration (VIV) tests on rigid cylinders. The objectives were to provide insight into the physical phenomena that drive the VIV of riser sections at realistic, riser-scale Reynolds Numbers and to provide data that can be used to develop riser VIV lift and damping models for use in riser VIV analyses. These tests were performed using an innovative, resonant VIV Test Rig designed to provide credible VIV data for fullscale riser sections. We tested a smooth bare cylinder, bare cylinders with three levels of roughness, and a straked cylinder. The tests covered ranges of reduced velocity, Reynolds number, frequency of vibration, and vibration amplitude that are of interest for deepwater risers, at a realistic riser mass ratio. This VIV Test Rig allowed quantification of VIV lift and damping forces for cylinders and provided information on the complex phenomena that affect VIV amplitude. The VIV response of low mass ratio, bare cylinders shows pronounced sensitivity to Reynolds number and roughness when Reynolds numbers are in the critical region. Amplitudes of response can be as large as twice the cylinder diameter for smooth, bare cylinders in certain Reynolds number regions, larger than those reported previously for cylinders with properties similar to those of production risers. Measured motion and force time histories suggest that the strakes we tested completely disrupted the Karman vortex street, as was shown by the random, rather than periodic, nature of the cylinder vibrations. These findings are important for the design of risers subject to VIV. The need to enhance riser VIV prediction methods in light of these discoveries should be evaluated. Introduction As offshore oil and gas exploration and production moves into deep and ultra-deep water, high-current environments are more frequently encountered in the field. Vortex shedding due to high currents may excite high bending modes of risers, resulting in high rates of fatigue damage. VIV suppression is often required to reduce long-term fatigue damage rates to acceptable levels. The behavior of long, flexible risers in ocean currents is complex, with significant hydrodynamic and dynamic nonlinearities. Simplifications are required to make the problem tractable to analysis. In typical riser VIV analyses [1, 2], risers are modeled as tensioned beams and vortex-induced hydrodynamic loads are approximated using strip theory. For VIV, the global response amplitudes are determined by balancing vortex-induced excitations on the riser with dissipative forces. Thus, to predict riser fatigue with confidence, it is important that accurate strip-theory models are used to represent the flow physics of vortex-shedding and dissipative forces. Vortex-induced forces cannot yet be calculated using computational fluid dynamics (CFD) at Reynolds numbers consistent with riser-scale VIV [3], and so are typically based on model tests of rigid cylinders. In April 2003, ExxonMobil performed VIV testing on rigid cylinders to develop VIV lift and damping data at riserscale Reynolds numbers. A wide range of cylinder configurations and flow conditions was tested. The objective was to develop high-quality measurements of hydrodynamic forces and motions for incorporation into VIV response models as one element in the development of a high-confidence, validated design basis for riser VIV fatigue analysis.

Circular cylinder wakes and vortex-induced vibrations

“…Raghavan and Bernitsas (2011) have published experimental results showing a maximum amplitude of 1.9D at a Reynolds number of 8.7 Â 10 4 for a value of m n x around 0.2. This result is in agreement with measurements of Ding et al (2004) at slightly higher Reynolds numbers. These results fall well above an extrapolation of the straight line fit shown in Fig.…”

Vortex-induced vibrations of elastically mounted circular cylinder at Re of the O(105)

“…Ding et al (2004) reported drag and lift coefficients for smooth and rough cylinders. There is an indication of drag crisis in the case of smooth and small roughness cylinders from Re from about 0:3 Â 10 6 up to 1 Â 10 6 .…”

Numerical study of vortex-induced vibration of pivoted cylinders

“…In practical appliations, the Reynolds number can be one or two orders of magnitude higher. As the effect of Reynolds number on VIV has been studied in Bernitsas and Raghavan (2008), Resvanis et al (2012), Govardhan and Williamson (2006), Klamo et al (2005) and Ding et al (2004), the lift coefficient increases as the Reynolds number increases. Hence, we anticipate that the energy conversion gets more efficient.…”

“…Our numerical simulations are performed at Reynolds number 500, which is much smaller than that of actual operating conditions. Being aware of the limitations of this numerical study, we anticipate that the energy conversion efficiency increases in high Reynolds number flows (Bernitsas and Raghavan, 2008;Ding et al, 2004).…”