Finite elements analysis of the side grooved I-beam specimen for mode II fatigue crack growth rates determination

Purpose: Carefully investigate the stress-strain state of the side grooved I-beam specimen with edge crack and determine the effect of crack length and crack faces friction on stress intensity factor at transverse shear. Design/methodology/approach: The finite element method was used to estimate the stress-strain state of I-beam specimen at transverse shear. For this purpose, a fullscale, three-dimensional model of the specimen was created, which precisely reproduces its geometry and fatigue crack faces contact. For the correct reproduction of the stress singularity at the crack tip, a special sub-model was used, which has been tested earlier in solving similar problems of fracture mechanics. In order to improve the accuracy of the calculations, for crack plane and cross-section of the specimen on the crack extension modeling, an algorithm for changing the crack length without changing the total number of elements in the model was developed and applied. Young's modulus and Poisson's ratio of structural steels were specified for the model material. The static loading of the model was realized assuming small scale yielding condition. The stress intensity factor was found through the displacement of nodes in the prismatic elements adjacent to the plane and the front of the crack. Findings: Mathematical dependences, which show an increase of stress intensity factor in the I-beam specimen with an increase in the crack length, and its decrease with an increase of crack faces friction factor at transverse shear, were established. The results are compared with the partial cases known from the literature and their good convergence was shown. Research limitations/implications: By analyzing the obtained graphical dependences, it is established that for relative crack lengths less than 0.4 there is a significant influence of the initial notch on the stress-strain state of the specimen, and for the lengths greater than 0.9 an influence of constrained gripping part took place. For this reason, all subsequent calculations were carried out in the range of relative crack length from 0.4 to 0.9, which represents the applicability range of the final calculation formula. Increasing of the crack faces friction factor from 0 to 1 monotonically reduces the stress at the crack tip. For a short crack, this effect is 1.5 times greater than for a long one, which is reflected by the calculation formula. Practical implications: Using the proposed calculation formula, one can calculate