Accuracy of indirect fatigue crack length measurement by potential drop method or by compliance technique may be affected at low load ratio due to fracture surface contact, crack closure or mixed mode fracture. As an alternative, the maximum value of crack opening displacement, COD max , from a clip gauge was utilized. Middle crack tension M(T) specimens were used to obtain conservative data at a low load ratio (R = 0.1). Thin sheet specimens (B = 3.2 mm) with different widths (100 mm ≤ W ≤ 400 mm) of AA6056-T4 were investigated in the mid-regime (Paris regime), which is of interest for damage tolerance analysis. The use of COD max is found to provide crack lengths equivalent to those measured optically. Hence, the method is very suitable for indirect crack length measurement. Furthermore, small width specimens provided data equivalent to large width specimens. Insofar, the size effect is found to be absent, and fatigue crack propagation data can be acquired on small width specimens when material availability is limited.