EVALUATING FINITE-ELEMENT MODELING TECHNIQUES FOR USE WITH THE VIRTUAL CRACK CLOSURE TECHNIQUE (VCCT)

摘要

The virtual crack closure technique (VCCT) has been used extensively over the past few decades to determine the components of the energy release rate, G. The ability to decompose the total strain energy release rate into its respective Mode I, Mode II, and Mode III components significantly increases the versatility of this method over other analytical techniques. The purpose of this effort was to investigate various parameters that could affect the results of the VCCT methodology when utilized in a very costeffective and time-efficient manner for industry applications. The parameters that were investigated include the element size ahead of the crack, constant and gradient mesh size across the width of the specimen, and ply-by-ply and "averaged property" element use. A double cantilever beam (DCB) specimen was modeled in ANSYS. to provide analytical results for this comparison. The DCB specimen is the standard specimen used to determine the Mode I fracture toughness of a composite material. Results of the models were compared to the theoretical Mode I Energy Release Rate for a DCB specimen. The traditional DCB specimen was modeled in ANSYS. to evaluate the effects of several parameters, such as element size and node spacing, as well as effect of element type, and modeling technique, i.e., ply-by-ply modeling versus layer-lumping. The effects of modeling techniques including ply-by-ply or "averaged property" modeling as well as the selection of element type can have a significant effect on the analytical results. This indicates that the selection of element type and modeling technique also affects the calculated energy release rate values. The work in this study demonstrates that using a layered SOLID46 element actually yields slightly better results than its "averaged property" counterpart, SOLID45. In addition, a previously proposed 0/45 deg DCB specimen was analyzed and very interesting results were found. It was seen that the G profile for the 0/45 deg specimen proposed by Martin and Rousseau [2] varies considerably from the edge of the specimen to its center, with the maximum value occurring at the center. The unidirectional DCB specimen; however, exhibits a more constant shape across the width. This G profile in the 0/45 deg DCB specimen clearly exhibits anticlastic curvature.