A comparison of triaxially braided composites using the generalized method of cells

Two distinct triaxial braid architectures are compared in this numerical study: (1) the "raditional" triaxial braid in which the axial tows are merely laid in between the woven biased (±θ°) tows and (2) the "true" triaxial braid in which the axial tows are interleaved through the biased tows. The microstructure of the triaxial braids is constrained as a function of volume fraction, braid angle, and tow geometries. A multiscale modeling methodology is developed to use the Generalized Method of Cells (GMC) micromechanics model recursively over multiple length scales in a three step homogenization process to compare the effective elastic properties of the two different types of triaxial braids. This methodology has previously been used to accurately predict experimental results for the "traditional" triaxial braid architecture; however, this paper simply compares the two types of braid architectures numerically. Preliminary results show that the "true" triaxial braid follows most of the trends of the "aditional" triaxial braid in effective stiffness properties as a function of braid angle, except for the axial modulus at higher braid angles. At braid angles in excess of 45°, the "true" triaxial braid shows an increase in axial stiffness. The axial modulus of the "true" triaxial braid exceeds that of the "traditional" triaxial braid for braid angles greater than 50°. Therefore, the "true" triaxial braid does present a viable alternate braid that may offer advantages over the "traditional" triaxial braid for certain applications. An experimental study is being prepared to compare the two triaxial braids experimentally and to validate the numerical model used in the present study.