Two main mechanical tests are commonly used to measure the critical shear stress of an adhesive, the Single Lap Shear (SLS) test and the Thick Adherend Shear Test (TAST). In these tests, the assessment is performed with mode mixity. In this work, a two-dimensional Finite Element Model of an asymmetric TAST is set-up and a few mechanical tests are performed on titanium/epoxy specimens. The model takes into account a shift of the adhesive joint on the loading axis. The impact of a shifted loading is estimated on two specimen geometries though a sensitivity analysis. Then, a method is suggested to assess the state of stress that caused the specimen failure. The results show that the failure location and the mode mixity can be tuned by the shift. Furthermore, it is shown that a standard TAST geometry is remarkably sensitive to a shifted loading. As a consequence, this parameter should be set with a submillimeter accuracy or a modified version of classical shear tests should be used. The reported results are consistent with one another and the method used provides a clear link between the critical stresses of the model and the mechanical response of the specimen until failure.

两种主要的机械测试通常用于测量粘合剂的临界剪切应力，即单层剪切 (SLS) 测试和厚粘附剪切测试 (TAST)。在这些测试中，评估是使用模式混合进行的。在这项工作中，建立了非对称 TAST 的二维有限元模型，并对钛/环氧树脂样品进行了一些机械测试。该模型考虑了粘合接头在加载轴上的移动。通过敏感性分析，估计位移载荷对两个几何形状的试样的影响。然后，提出了一种评估导致试样失效的应力状态的方法。结果表明，故障位置和模式混合可以通过移位进行调整。此外，还表明标准 TAST 几何结构对移位加载非常敏感。因此，该参数应设置为亚毫米精度或应使用经典剪切测试的修改版本。报告的结果彼此一致，并且所使用的方法在模型的临界应力与试样的机械响应之间提供了明确的联系，直至失效。