Critical velocities for fluidelastic instabilities for tubes exposed to crossflow are typically predicted based on the normal vibratory mode shapes and frequencies. If the analysis is based on vibratory response shapes rather than mode shapes, significant differences can be shown to exist in the prediction for some cases. These differences tend to be more pronounced in cases of partial exposure of tubes to crossflow or non-uniform spacing of tube supports and at higher damping. A brief assessment and evaluation with two sample problems is given.

This content is only available via PDF.
You do not currently have access to this content.