Simple Experiments to Validate Modal Substructure Models

While significant strides have been made in recent years, experimental/analytical substructuring methods can be quite sensitive to seemingly small measurement errors, to modal truncation (for modal methods), small residual terms (for frequency based methods), etc... As a result, one tends to have less confidence in a substructuring prediction than, for example, a finite element model, even though both may have similar accuracy in some situations. This work explores ways of estimating the uncertainty in modal substructure models, seeking to provide the experimentalist with an approach that could be used to evaluate the fidelity of a substructure model. This would allow one to detect cases where the substructuring problem is very sensitive to uncertainty, so a remedy can be sought, and perhaps even provide a measure of the expected scatter in the predictions. Simple experiments are proposed, for example obtaining the natural frequencies of the subcomponent after attaching a well characterized subcomponent at a point, in order to verify the subcomponent model and to estimate the sensitivity of the substructuring predictions to uncertainties. Special attention is paid to the adequacy of the modal basis of the substructure.