A new model for damping controlled fluidelastic instability in heat exchanger tube arrays

Abstract

Heat exchnager tube arrays are susceptible to damage due to flow-induced vibration, with fluidelastic instability the potnetially the most destructive mechanism. In this paper a simple wake model consisting of a convecting vortex sheet is proposed to represent the transient nature of fluidelastic forces present in a tube array. Using this model, the memory function proposed by Granger and Paidoussis has been obtained without the need to calibrate the model with experimental data. The resulting function is found to compare well with the first and second order empirical approximations. The memory function has been combined with experimental data for the static fluid force to produce a prediction of the critical velocity for a range of mass damping parameter. This stability threshold is in reasonable agreement with experimental data. Therefore it is concluded that vorticity transport may well form part of the underlying mechanism responsible for damping controlled fluidelastic instability, as previously proposed.

Development of a time delay formulation for fluidelastic instability model

“…Some theoretical or semi-empirical FEI models have been developed to predict the critical velocity such as the quasi-static (Blevins, 1974;Connors, 1970), the quasi-steady (Price & Paidoussis, 1984), the quasi-unsteady model (Granger & Paidoussis, 1996;Meskell, 2009), the semi-analytical channel flow model (Lever & Weaver, 1982) and the unsteady model (Chen, 1987;Tanaka & Takahara, 1981). The differences among these models lie in the definitions of the dynamic inviscid, and the pressure in the wake regions to be constant, the pressure distribution can be obtained analytically and then the drag force can be achieved by integration.…”

“…To support the assumption of the physics of the time delay proposed by Granger and Paidoussis (1996), Meskell (2009) proposed a simple wake model to clarify that the underlying process of vorticity transport accounts for the time delay. The memory function is obtained by numerical integration of the governing equations without calibrating the model with experimental data.…”

Surface pressure survey in a parallel triangular tube array

“…This was extended by Granger and Paidoussis (1996) to include an empirical memory function rather than a simple time lag to better represent the convective nature of the flow. Meskell (2009) proposed a theoretical approximation for this memory function, so that the only empirical input needed for the model of fluidelastic instability would be the force coefficients. For simplicity, the original quasi-steady model, with a constant empirical time delay will be considered here.…”

A CFD study on the fluctuating flow field across a parallel triangular array with one tube oscillating transversely

“…More specifically, Granger and Paidoussis (1996) describe it as a memory effect due to the diffusion of thin vorticity layers from the oscillating tube surface into its boundary layer, to be finally convected downstream. Posteriorly, Meskell (2009) proposed a simple wake model to estimate Granger and Paidoussis's memory function, showing that that process of vorticity transport can produce the time delay of the fluidelastic force.…”