4.          CONCLUDING REMARKS

 

The test-analysis model (TAM) is a very powerful tool because it provides a direct mathematical link and basis for comparison between finite element model predictions and modal survey test results using orthogonality. The major objective in selecting accelerometer locations should be to create the best TAM possible, meaning the most accurate and robust reduced mass matrix. However, there are two other considerations that should be taken into account in the selection of accelerometer sites. First, care must be used to make sure that selected accelerometer locations are accessible—i.e., that an accelerometer can be physically mounted at the desired location. Second, additional instrumentation may be appropriate at structurally important locations even if that loca­tion does not contribute highly to the mode shape observability or kinetic energy. Regions that will carry significant load (such as the interface between major compo­nents) or regions that are historically difficult to model (such as joints) should be instru­mented. These locations are usually designed to be stiff in anticipation of the loads to be transferred. Therefore, they often show little relative motion in the mode shapes and are rarely picked as accelerometer locations. However, if the objective of a model survey and subsequent correlation is to create a validated model for subsequent response analysis, instrumenting the locations that will most likely influence the analysis will safeguard against surprises in the test.

The selection of an accelerometer set and the creation of a TAM are not an exact science. Limitations posed by hardware accessibility, cost and schedule, and unanticipated behavior of structures make the selection of instrumentation an art as well as a science. The tools described in this manual can greatly aid the engineer in the design of a modal survey. Engineering judgment and experience, however, ultimately have the largest influ­ence on a successful test.