Finite Element Modeling of Dynamic Circuit Board Strain Response under High-G Shock Impact
Authors: Sergey M. Kaplan, M.S., Robert B. Greendyke, Ph.D. Company: Air Force Institute of Technology, Applied Research Date Published: 12/1/2015
Abstract: In an effort to predict, and ultimately improve the reliability of electrical circuits embedded in technology exposed to high-g environments an order of magnitude higher than those encountered in commercial applications, this study uses solid mechanics software to model dynamics of a Printed Circuit Board (PCB) undergoing drop tests in a laboratory. An accelerometer is used to obtain the impact acceleration profile of a steel fixture mounted on the droptable and a strain gage is used to collect uniaxial strain response of a rectangular PCB. Maximum measured acceleration in the laboratory was approximately 15,750 g with the total velocity change of 14.95 m/s, and maximum measured strain was approximately 6300 microstrains. The impact acceleration profile collected during experiments and a half-sine wave acceleration profile that matched the total change in kinetic energy were later applied as boundary conditions to the computational model for comparison. Varying Young’s modulus and density in the linear elastic material model, computational peak amplitude and dominant frequency responses of strain were compared to the experimentally obtained uniaxial strain. Both the experimentally obtained acceleration and the half-sine input boundary conditions applied to the fixture in the computational model produced a dominant frequency response, which closely matched experimental results. Using experimentally obtained accelerometer boundary conditions, the peak amplitude obtained from the computational model was approximately half of the measured strain. However, using the half-sine input acceleration profile, based on the total change in energy during impact, produced a close match between experimental and computational peak strain amplitudes. Variations in elastic modulus and density resulted in linear changes to the board strain, with density having the greatest percentage-wise effect due to the direct change in board mass.