Comparison of experimental and computational procedures for the evaluation of polymer matrix composites

Abstract

It has been found that composite materials offer desirables mechanical properties for the naval, structural, aerospace, automotive (among others) industries. The study of those materials has advanced significatively the fields of the physical and mechanical characterization using analytic formulations, and also in the field of the computational simulations particularly in the use of the Finite Elements Method (FEM). This method has been found useful in the prediction of the mechanical properties of the composites in cases in which the geometries or the load state is too complex for the use of analytical formulations or too expensive for its experimental evaluation. In this study the goal is to stablish a methodology for the definition of FEM models with predictive capabilities based in a limited series of inexpensive experimental tests and stablish the accuracy of those models. A series of laminates are produced by the specifications of a local firm in which the fiber orientation varies using the hand lay-up technique and those are characterized using traction, ignition and defects experimental methods. FEM models of the traction test for each laminated were defined using ANSYS ®and the results obtained were compared with the experimental results stablishing a mechanical response pattern under different configurations of fiber orientation, type and layer numbers. The results of the comparison between experimental and FEM approximations for the Mat (random oriented) laminates agree with an error lower than 6% while the error for the woven fabric in a 90° orientation reach the 34.2% which can be attributed to variations in the manufacturing process.