Feature Article the strength of the bond between the GRP and the steel plate will be significantly weaker than that of the GRP-to-resin bond represented by the baseline CZM of Table 2. A parametric study using a FE model of the non-perforated joint shows that a factor of 0.30 needs to be applied to the baseline CZM in order to replicate the average failure load from the physical testing, see Figure 10. As previously mentioned the FE model is a half model, therefore the load of 2.2 kN from the analysis matches the 4.4 kN average failure load from the physical tests. Figure 10 shows the tension load versus time plot for the analysis of the first set, the state of cohesive zone damage at the bondline and the distribution of axial stress in the joint. The damage and stress images are taken just prior to failure of the joint as indicated by the vertical red bar on the load versus time plot. Further work with this FE model a. Cohesive zone damage b. Axial stress c. Load versus time plot Figure 10. Abaqus results for non-perforated joint. is ongoing; for example, mesh size sensitivity has yet to be studied. A similar study has been undertaken using a FE model of the perforated joint. In this case the 0.3 x baseline CZM is applied to all of the GRP-to-steel interfaces at the bond line and the baseline CZM is initially applied to the GRP-to-resin interface, i.e. where the resin that penetrates the steel perforations is bonded to the inside face of the GRP lap. A similar iteration methodology to that described above was used with the perforated joint. Although in this case the CZM for the GRPto-steel interface was not modified; only the CZM for the GRP-to-resin interface was modified. Using this method, it was found that applying a factor of 2.25 to the baseline CZM gives good correlation of failure loads comparing the FE model and physical test results. Unique Polyamide * * * * High Modulus and Strength Easy Molding Water Resistance Good Surface Appearance MITSUBISHI GAS CHEMICAL AMERICA www.mgc-a.com composite@mgc-a.com 12 SAMPE Journal, Volume 53, No. 1, January/February 2017http://www.mgc-a.com http://www.wichitech.com