PSI - Issue 57

Elena Sidorov et al. / Procedia Structural Integrity 57 (2024) 316–326 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

322

7

The software package ANSYS is used to generate the FE model. In vicinity of the wheel load introduction, a regular mesh of 8-node solid prism elements (SOLID185) is implemented. The remaining parts of the beam are freely meshed with 10-node tetrahedral solid elements (SOLID187). Contact elements (CONTAC174) are placed at the bottom surface of the rail. The corresponding target elements (TARGE170) are located at the top surface of the girder flange. Augmented Lagrange contact formulation with a friction coefficient of 0.5 is used. Ideal elastic behavior of steel is assumed with a modulus of elasticity E = 210 000 N/mm² and a Poisson’s ratio  = 0.33. The rail welds are modelled as symmetrical fillet welds. The two different gap types, that are shown in Figures 5b and c, can be considered by the FE model. Kuhlmann et al. (2022) performed fatigue tests on crane runway beams with hot-rolled section HE 280 A and flat material 50  30 mm as crane rail. The rail was fastened to the top flange by chain intermittent fillet welds with weld size a = 5 mm that measured h = 50 mm in length with a spacing of g = 250 mm. This configuration is chosen as reference case for the following parametric study. The span of the crane runway beams and the number of rail welds are specified in Figure 6a. Figure 7a shows the pressure p of the rail welds measured in [N/mm] for different wheel load levels if a uniform gap is modelled underneath the rail with gap size g con . The upper thick continuous curve represents the case where the gap size g con is so great that no contact between rail and flange can be achieved independently of the wheel load level. The lower broken curve stands for the case with full contact. All other curves show the weld pressure in case of partial rail-flange contact. Obviously, higher wheel load levels result in an increased rail-flange contact. In case of the wedge shaped gap modelling, the partial contact is achieved at lower wheel load levels as shown in Figure 7b. 4.2. Reference case

Fig. 7. Dependency of rail weld pressure on gap modelling: (a) uniform gap, (b) wedge-shaped gap

The principal shape of the contact area between rail and flange depends on the gap modelling. In case of a uniform gap according to Figure 5b, partial contact develops at the middle of the rail in the loading section as visible in Figure 8a. In case of a wedge-shaped gap according to Figure 5c, the contact area primarily develops along the rail edges in longitudinal direction of the rail. Under loading, the contact area enlarges in transverse direction as shown in Figure 8b. With respect of the observations presented in Section 3, the wedge-shaped gap modelling according to Figure 5c is regarded as more realistic.