I don't have any direct experience in shell in tube heat exchanger vibration measurements but my general impression is that external shell measurements would not provide reliable data on internal tube vibrations because of the many non-tube component responses likely to be present in accelerometers mounted on the shell. I think you would have to get acceleration measurements directly on the tubes or at least the tubesheets supporting the tube ends. A recent text that treats heat exchanger vibration analysis extensively,is "Au-Yang,M.K. (2001), "Flow Induced Vibration of Power and Process Plant Components - A Practical Workbook", ASME Press, ISBN: 0-7918-1066-7. It cites experimental work but doesn't address test instrumentation. In Sect. 12.5 on "Heat Exchanger Acoustics", Au-Yang shows and describes resonance maps developed by Zaida etal (1989)in the Journal of Fluids and Structures Vol. 3 pp. 293-314 and 315-324. The resonance parameter G uses fluid and acoustic Reynolds numbers with staggered and in-line tube array dimensional parameters to define resonant and non-resonant conditions. Au Yang notes that two-phase flow and inhomogeneous single phase flow tend to disorganize and randomize tube vortex shedding and structural lock-in effects. Another source addressing the same issues of heat exchanger tube vibrations is Lucas,M.J. etal(1997), "Handbook of Acoustic Characteristics of Turbomachinery Cavities", ASME Press, ISBN 0-7918-0054-7 where Chapter 9 is on "Vortex Shedding from Bluff Bodies"
