Finite element analysis of three common tensile tests with induced flaws.

摘要

Three common tests to evaluate the tensile strength of dental composites are the three point bending, four point bending, and biaxial flexural test. The protocol for these tests are laid out in the ASTM standards D790 and ISO 6872. The purpose of this study is to investigate these tests for mathematical validity when compared to finite element analysis solutions as well as explore the effect of induced flaws on the outcome of these tests. The tests were modeled using the finite element analysis program Cosmos/M 2.5 following the ASDM and ISO standards. Flaws were introduced into the test by randomly assigning two, five, and ten percent of the samples elements with the material properties of air, which would represent voids, in the sample. Twenty tests were conducted for each flaw group. The FEA of the three and four point bending tests was found to be in agreement with the analog equations (analog S = 750, FEA 3 point bending = 759, FEA 4 point bending = 754) while the biaxial flexure test stress was found to be lower then the analog equation (analog S = 76, FEA S = 57.5). The randomness of the flaws produced a variation in the data points that appeared to fit the Weibull distribution. An increasing the flaw percentage, which may lead to larger flaw size, produced little change in the characteristic strength but significant decreases in the Weibull modulus (m). The differences between 2, 5, and 10 percent flaw distribution in the Weibull modulus are significant at the 80% confidence level. This confirms Weibull's assertion that flaws influence the Weibull modulus more then the characteristic strength.