COMPARISON OF PROCESSING PARAMETERS FOR PULTRUDED GRAPHITE-EPOXY AND FIBERGLASS EPOXY - A HEAT-TRANSFER AND CURING MODEL

Mathematical modeling of any manufacturing process enhances the unders tanding of the mechanisms governing the process and helps to establish guidelines for optimizing the process. In the present study, a therma l model utilizing Patankar's method has been developed to simulate/opt imize the pultrusion process. The numerical model solves the steady-st ate continuity, momentum equations and the transient energy and specie s reaction equations in cylindrical coordinates for a two-dimensional product. This model is comprehensive in the sense that it includes a l iquid resin layer near the wall and involves solving for the velocity field. Most other researchers have assumed either flat or linear veloc ity profiles for this liquid region. The axial conduction term in the energy equation, neglected by most other researchers, is also included in the present model. The composite materials studied in this analysi s are graphite/epoxy and fiberglass/epoxy. Since the thermal propertie s of graphite and fiberglass are different, the bulk properties of the composites are different, and the thermal response of these two compo sites during pultrusion can be expected to be different. The present a rticle focuses on the comparison of the model's processing predictions for graphite/epoxy and fiberglass/epoxy and compares the predictions with experimental results. The key pultrusion process parameters exami ned were die wall temperature profile, fiber volume-fraction, and pull speed. Comparisons with experimental results were obtained using a 3/ 8'' diameter, 36'' long die for both graphite/epoxy and fiberglass/epo xy. The comparison of experimental data with numerical predictions are excellent and provide an ideal means by which to optimize the pultrus ion process for a given product geometry or to determine a start-up pr ocess window for a new product.