Cole C. Doupe

Date of Award


Document Type


Degree Name

Master of Science in Astronautical Engineering


Department of Aeronautics and Astronautics

First Advisor

Eric D. Swenson, PhD


Space launch vehicles produce tremendously harsh environments for their payloads. One of the worst contributors to this harsh environment is vibration. Launch vehicle contractors require accurate dynamic models in order to perform coupled loads analyses with each payload to mitigate risks. Accurate predictions of the dynamic response of the payload are not achieved easily. The Finite Element (FE) method has proven to be the best approach in creating accurate dynamic models of complex structures. To improve the agreement between an FE model and the structure it represents, a common practice is to `tune' or adjust parameters of the FE model to match experimentally measured data. In order to collect spatially dense and accurate dynamic responses from a satellite, a Polytec laser vibrometer is used which measures the Doppler shift to determine the frequency response from an excitation. To illustrate the benefits of employing this approach, a process is developed to measure dense modal data and tune an FE model of the US Air Force Academy's FalconSAT-5 Structural Engineering Model. The first step in the process developed in this research involves measuring and tuning models of the satellite structure panels individually. In tuning the structural panels, material stiffness is the major design variable. The tuned FE models of the panels are integrated into the full satellite model which is then tuned based on the spring constant of the connections between the panels. The first eight modes of each side panel, six modes of the top panel, and five modes of the base panel were tuned with eigenvalues matching measured natural frequencies within 2%. Next, the first five modes ranging through 154 Hz were tuned on the full satellite FE model. Predicted natural frequencies were within 3% of measured values for most cases and modes. Modal assurance criterion values comparing tuned FE model eigenvectors and measured mode shapes decreased with increasing numbers of modes tuned, but remained above 0.75 through tuning five modes.

AFIT Designator


DTIC Accession Number