shear stress amplitude τa to the normal one σa is less than 0.63 [10], or when the ratio of the shear strain range ∆γ to the normal one ∆ε is less than 1.5 [3]. Such ratios depend on the kind of material, stress or strain level and temperature. Some theoretical predictions of the fatigue fracture plane position have been presented by Kanazawa et al. [11], Findley [12] and Stulen and Cummings [13]. Numerous models of fatigue crack initiation and propagation for multiaxial cyclic loading do not take into account the change of the principal strain and stress axes. That is a possible reason why they can only be applied to some particular cases. Consequently, averaged principal stress directions should be determined, and the averaging procedure could be carried out, for instance, by employing suitable weight functions [1417]. In the following, the position of the principal stress axes 123 at the generic time instant is described by the three “principal” Euler angles ϕ, θ, ψ. The aim of the present paper is to verify whether the weight function method can allow us to determine the expected fatigue fracture plane position. The algorithm presented in [15–17] is applied to experimental data related to one steel and two cast irons [18,19]. AVERAGING PROCEDURE The principal axes 123 system can be considered as a transformation of the XYZ system, fulfilling a condition of orthogonal conversion or specific rotation [20]. A transformation between two Cartesian systems can be described in different ways, but the direction cosines matrix is usually employed. On the other hand, such a matrix includes nine dependent components, and we cannot average all the matrix elements since the averaged matrix would not satisfy all the conditions of transformation. Moreover, we are not able to select which three independent elements of the direction cosines matrix should be averaged. Thus, the three independent Euler angles ϕ, θ, ψ are selected for description of the transformation from the XYZ system to the 123 system (Fig.1). Let us assume that the 1-axis and the 3-axis represent the directions of maximum and minimum principal stresses, respectively. Figure 1: Principal stress axes 123 described by the Euler angles ϕ, θ, ψ Under variable loading, the Euler angles are time-varying functions as well as the stress tensor components. Different sets of the Euler angles could describe the same position of the principal axes 123. Consequently, in order to average correctly the results determined for different time instants, the ranges of the Euler angles have to be reduced through a two-stage procedure, bringing them to 0 ≤ ϕ, θ ≤ π /2 and -π/2 ≤ ψ ≤ π/2. A detailed description of an algorithm to perform such a reduction can be found in Refs [15,16]. From a physical point of view, it seems logical to carry out the Euler angles averaging through suitable weight functions [14-17] in order to include various factors influencing the fatigue fracture behaviour (stress
RkJQdWJsaXNoZXIy MjM0NDE=