PSI - Issue 7

Marton Groza et al. / Procedia Structural Integrity 7 (2017) 438–445 M. Groza et al. / Structural Integrity Procedia 00 (2017) 000–000

442

5

model space, except for the Y rotation axis, modelling a brake lever-like fixation, whereas the force was applied on the surface of the third bore. Fig. 5b. shows the von Mises equivalent stresses on the deformed shape with enlarged scale. Additional spherical bodies were created around the surface defects in order to get improved mesh properties, and to simplify the submodel creation process. The submodels with a diameter of max 2 ( ) A area ⋅ + for D1-3 contain 2,2, 1,7 and 3,4 million nodes to ensure mesh convergence, which has been checked and achieved. The surface defects were also meshed which made it possible to calculate directly comparable results with and without the defect through the change of the elastic properties of the meshed void. The point-by-point comparison of the with and without defect stress state near the defect was used to estimate the local stress gradient in an automated and efficient manner. The underlying theory and assumptions are explained in Chapter 4.

5.2. Modelling of cyclic plasticity

In the elastic-plastic submodels the Chaboche nonlinear kinematic material model from Chaboche (1991) was applied to describe the cyclic material behaviour. The Chaboche model is capable to adequately describe the cyclic evolution of the mean stress under force controlled loading. The parameters of the model for NCI material are summarized in Table 2.

Table 2. Parameters of the Chaboche nonlinear kinematic hardening model for NCI material.

1 ( ) γ −

( y MPa σ

1 ( ) C MPa

)

280

234 710

1076

6. Results and discussion

Calculation results are summarized in Table 3. and Fig. 5. All stress related quantities are evaluated at the hot-spot, which is defined as the location with the maximum Crossland equivalent stress in the elastic-plastic FEA computation. The sign of the hydrostatic stress at the hot-spot ,max h σ is in clear connection with the location of the defect on the lever component under bending load, D2 and D3 are on the tension (positive ,max h σ ), meanwhile D1 is on the compression (negative ,max h σ ) side. The local R values describe the effect of the acting pulsating force. The results of the elastic plastic computations are shown in Fig. 5a., where the relevant tensor stress components are plotted at the hot-spot, in a coordinate system oriented by the maximum principal stress. It is clearly visible, that the stress field at D1 is dominated by compressive, whilst at D2 and D3 by tensile stresses. In all three cases, after an initial local hardening

Fig. 2. Definition of the different quantities related to the local stresses.