PSI - Issue 57

Kashif Kamran Toor et al. / Procedia Structural Integrity 57 (2024) 772–784 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

780

9

• Weld model using oblique shell elements. • Weld model as solid elements including the notch.

Boundary conditions and solution procedure was as described in section 2.4. For the oblique shell element model, INFs were generated based on stress extrapolation method type ‘b’ where read out point distance s were measured from the weld toe. The solid element model was based on the effective notch stress method, hence, INFs were generated by reading the maximum principal stress at the 1 mm notch radius. 3.3. Weld model - Shell elements The weld model using the oblique shell elements is considered to be sufficiently accurate to include the weld stiffness into the FE model. General engineering practice recommends a minimum weld throat thickness of 1/3 of the main plate which in the FE model is rounded to 15 mm. The weld was modelled using the oblique shell element with a thickness of 15 mm and oriented at 45 deg. representing the weld of 20 mm leg length. The edge weld was also modelled assuming the same weld leg and shell thickness equal to the main plate thickness. Two shell models were analysed based on the weld toe location measured from the shell intersection. In model ‘A’ a leg length of 20 mm from the shell intersection was used whereas model ‘B’ was based on measurement shown in Fig. 8c.

a

b

Fig. 9. (a) Weld geometry shell type A; (b) Weld geometry shell type B.

The stress read out path is from the weld toe at the edge side. The weld modelling and stress path is shown in Fig. 9. For details of this type of modelling technique reference is made to, Aygul et al. (2011) and Eriksson et al. (2003). 3.4. Weld model – Solid elements The weld modelling using the solid element technique is assessed as the most accurate stiffness representation of the weld in a FE model. A sub-modelling technique was used such that a part of the joint was modelled as solid elements including the weld profile and the notch at the toe. The weld toe location for the solid models is measured for a leg dimension of 10 mm (Model A) and 20 mm (Model B) for comparison of results with the two corresponding shell models A and B. The local solid model was created based on effective notch stress methodology using a 1 mm notch radius at the weld toe. The sub-model extent in the global model and weld profile at the hot spot is shown in Fig. 10.

a

b