PSI - Issue 82

Abhijit Joshi et al. / Procedia Structural Integrity 82 (2026) 91–97 A. Joshi et al./ Structural Integrity Procedia 00 (2026) 000–000

96 6

The stresses and creep strains at the end of 100-hour creep hold time are shown in Figs. 6a and 6b, respectively. Some key observations can be made based on these results: • There is a significant stress relaxation - from 375.6 MPa at the start of the creep step to 283.3 MPa at the end of 100-hour creep hold. The stress contours in Figs. 5 and 6a highlights the extent of stress redistribution caused by creep near the peak-stress location. • The maximum predicted axial creep strain is 4.15%. Based on the tensile tests reported in (Joshi et al., 2025), this level of strain is too high for this material and it is expected to rupture at such strains. Moreover, the creep tests reported in (Joshi et al., 2025) did not show any signs of cracking or rupture in the 150 MPa and 500 °C creep tests. This result suggests that modelling of graphite particles as void potentially overpredicts the local peak strains, so this approach may not be appropriate for creep simulations. • Even though the approach of modelling graphite using voids may not be suitable for creep simulations for the cast irons, there are few conclusions from this study that can be used in creep analysis of any local stress-concentration features. Comparison of results for two studied models - with and without the circular void - is given in Table 2. The results show that with a Kt of 3.34 for stress, the maximum creep strain produced in the microstructural model is almost 56 times higher. These results highlight a considerable difference between the macroscopic strains, measured in experiments, and their micro-scale levels, with local stress increases resulting in significantly higher creep strains over a period of time, potentially leading to the failure initiations from these locations.

Fig. 6. Simulation results after100 hour creep: (a) axial stress; (b) axial creep strain.

Table 2. Comparison of creep simulation results for matrix-only model and model with graphite presented as circular void

Model with graphite modelled as circular void

Ratio of results for two models

Parameter

Matrix-only model

Maximum elastic axial stress S22 (MPa)

150

500.4

3.34

Maximum axial creep strain CE22 (mm/mm)

7.425E-4

4.151E-2

55.91

4. Discussion and conclusions The paper presents the detailed methodology of development of micromechanical models for evaluation of creep in cast irons using 2D RVE microstructural models. Different creep laws available in Abaqus, derivation of creep parameters, as well as loads and boundary conditions defined for the model are discussed. The obtained results are presented for the matrix-only model (equivalent to the macroscopic behaviour of the studied CGI) and the model with graphite particle modelled as circular void. The key conclusions based on the analysis reported in this paper are as follows: • The micromechanical models highlight the significant increase in local stresses around the graphite particles even for low macroscopic stresses. These local stresses can exceed the material’s yield strength resulting in local plasticity.