There are several difficulties in the implementation of the SRS. One difficulty is in determining the actual time history with which to excite the structure under evaluation. The SRS does not uniquely identify the shock experienced by the part. This means shocks with different time histories, and parameters in spaces other than the SRS, could produce the same SRS without recreating the damage-causing potential of the structure’s service environment. Furthermore, the SRS accounts for neither the damaging potential caused by the interaction between masses on a system nor the damage potential caused by the coupling of translational and rotational movements [4]. This lack of correlation between test inputs and the resulting severity of damage of a shock environment leads to results that have an unknown level of conservatism. The inability to consistently and accurately replicate the severity of a mechanical shock environment warrants the development of new methods. Development of such a method requires the consideration and identification of parameters that are able to be implemented in a laboratory setting. The study discussed is a simulation based study in which measured mechanical shocks are applied to a finite-element model of an arbitrary test structure. In the context of this study, a simulation based approach offers several advantages over an experimental study. A simulation study allowed for a high volume of simulations which were used to evaluate the distributions of parameters and responses. Collecting this amount of data in an experimental based study would be time consuming and potentially expensive. Furthermore, the mechanical shocks were applied directly to the structure in the finite-element model simulations, and thus hardware constraints were not a factor. Data was collected at many locations on the structure for each finite-element model simulation, while doing so in an experimentally based study would not have been possible due to the amount of instrumentation that would be required. The objectives of this study are to identify possible parameters that correlate to the damage causing potential of a mechanical shock, identify a set of those parameters that can make a specification that uniquely specifies an acceleration time history, and compare the ability of the identified parameters to reproduce the damage causing potential of a shock to the ability of parameters that are currently being used. The process used to identify parameters of a mechanical shock started with using real transportation shock data to define parameters that could potentially have an effect on the severity of the shock. To determine how each input parameter correlated to the observed measures of severity shock realizations were developed and applied to the structure through finite-element model simulations. Similarly, the transportation shocks were applied directly to the finite-element model in order to observe how these unadulterated shocks, and their associated parameters, correlate to the observed measures of severity. The finite element model was initially evaluated by comparing the results from the model to the results of an experimental modal analysis. This was done to ensure that the finite element model was producing realistic results. The layout for the report is as follows. The test structure and model development are discussed in Sect. 5.2. Sections 5.3.1 and 5.3.2 describe the process used to identify the parameters that define the measure of severity associated with a mechanical shock. The results of the analytical study and identified parameters are shown and discussed in Sects. 5.3.3 and 5.3.4. Finally, the conclusions drawn from the analytical study are addressed in Sect. 5.4. 5.2 Model Development For this study the finite element-model was based on an existing structure (Fig. 5.1) as this allowed for confirmation that the finite-element model behaved realistically. The structure is made of approximately 6.5 mm thick aluminum and is composed of multiple components. The base of the structure has a diameter of 165 mm and the stand (the thinnest part) has a diameter of approximately 25 mm. The total height of the structure is approximately 305 mm tall. Composition of these different parts provides discontinuity throughout the structure producing varying measures of severity, which was desired for this experiment. A modal analysis was performed to compare the modes generated from the finite-element model and the modes collected from experimental data. The modal analysis was completed using a roving hammer technique with 3 uniaxial accelerometers and 17 impact points. Each accelerometer was placed on a plane that aligned with the axis of a Cartesian coordinate system in which the “z” axis is parallel to the vertical component of the structure. The impact points were spread around the structure such that multiple response directions on each face would be tested. For the modal analysis, free boundary conditions were simulated by placing the structure on foam. MEScope was used to extract the mode shapes based on the data at the impact and measurement locations. Finite-element modeling is critical to this study. The finite-element model was developed using Abaqus and consisted of four solid elements: the top and bottom plates, the main cylinder, and the stand. The top and bottom plates are thin cylinders that are connected to the main cylinder. The main cylinder is the largest part, the stand is connected to the top plate. Once the test structure was properly modeled both a modal analysis and the mechanical shocks were simulated. 44 M. Baker et al.
RkJQdWJsaXNoZXIy MTMzNzEzMQ==