National Research Council Resident Research Associate, NASA Langley Research Center, Hampton, VA
Head Structures Technology Research & Development Group, Boeing, Philadelphia, PA
ICASE Staff Scientist, NASA Langley Research Center, Hampton, VA
Senior Research Scientist, NASA Langley Research Center, Hampton, VA
A methodology is presented for determining the fatigue life of composite structures based on fatigue characterization data and geometric nonlinear finite element analyses. To demonstrate the approach, predicted results were compared to fatigue tests performed on specimens which consisted of a tapered composite flange, representing a stringer or frame, bonded onto a composite skin. In a first step, quasi-static tension and fatigue tests were performed to evaluate the debonding mechanisms between the skin and the bonded stringer. Specimen edges were examined under the microscope to document the damage occurrence. In a second step, a two-dimensional finite element model was developed to analyze the tests. To predict matrix cracking onset, the relationship between the externally applied tension load and the maximum principal stresses transverse to the fiber direction was determined through geometrically nonlinear analysis. Transverse tension fatigue life data were used to generate an onset fatigue life P-N curve for matrix cracking. The resulting prediction was in good agreement with measured data from the fatigue tests. In a third step, a fracture mechanics approach based on geometrically nonlinear analysis was used to determine the relationship between the externally applied tension load and the critical energy release rate. Mixed mode energy release rate fatigue life data from DCB, 4ENF, and MMB tests were used to create a fatigue life onset G-N curve for delamination. The resulting prediction was in good agreement with data from the fatigue tests. Additionally, the prediction curve for cumulative life to failure was generated from the matrix onset and delamination onset fatigue life curves. The results were in good agreement with data from the fatigue tests, which demonstrated that the methodology offers a significant potential to predict cumulative fatigue life of composite structures.
Paper ID: CTR10561J