PSI - Issue 5

Rui F. Martins et al. / Procedia Structural Integrity 5 (2017) 633–639 Diogo F. Almeida et al. / Structural Integrity Procedia 00 (2017) 000 – 000

637

5

(45mm/min). A convection coefficient value of hc = 10W.m -2 .ºK -1 was set for the free surfaces of the plate. In addition, as the plates to be welded were in contact with a fixed structure, the temperature of the lower face plate was set equal to room temperature (293ºK). During thermal analyses an emissivity value of 0.75 was also considered for the material.

1 st weld pass

cooling 2 nd weld pass

cooling

time [s]

Fig. 4. Finite element mesh used in 3D FEA. Schematic representation of the passages of the welding process as a function of time.

ǡ ǡ ǡ ൌ ͸ ͵ − ͵

0.0025 m 0.0075 m 0.0025 m 0,0015 m

a f a r

ʹ ʹ − ͵ ʹ ʹ − ͵ ൅ ሺ − ሻ ʹ ʹ

b

c

Fig. 5. Parameters used in the design of the heat source. Goldak’s double ellipsoid model.

3.2. Structural analyses

The structural numerical simulation aimed to determine the residual stresses and distortions generated by the heat cycle caused by TIG welding process. The finite element mesh used was the same defined during thermal analysis; however, the element type was changed using the "ETCHG" command provided in ANSYS®, for transforming the thermal elements in structural elements. The corresponding structural element to the heating element used in thermal analysis (SOLID70) was the SOLID185 finite element. For temperatures above melting temperature of AISI316L SS, a very small value for the Young's Modulus was defined. In addition, a multilinear isotropic hardening based on the criteria of von Mises plasticity was assumed throughout the analyses and the mechanical properties of AISI 316L SS used in this work, namely thermal coefficient of expansion, stress vs. strain, Young's Modulus, varying with temperature, were obtained from the works of Depradeux (2004), Bezerra (2006) and Pozo-Morejón et al. (2011). Although there is a variation of the Poisson ’ s coefficient with temperature, in this work it was considered constant and equal to 0.3. Finally, for the structural analysis it is quite important to model correctly the boundary conditions, as these will have a major influence on the distribution of residual stresses and distortions in the welded plates. In the case of this work, one of the plates was welded along the outer edges (spots) to a fixed structure, so as to minimize its deformation (Fig. 1), and this boundary condition was defined in the FEA. Moreover, in the numerical simulations the weld metal used during the TIG welding process was not considered.