Beschreibung
In this study, the impact of heat input model parameters (arc efficiency and heat source shape parameters) using Goldaks double-ellipsoid heat source on the calculated welding temperature field, residual stress and deformation have been systematically investigated. The main conclusions are illuminated as below: The predicted size of FZ varies with all the heat input model parameters. The reason is that the size of FZ is mainly associated with both the distribution and the magnitude of heat flux in numerical welding simulations. The calculated position of HAZ boundary and cooling rate t8/5 time can be affected by arc efficiency (obviously) and heat source width as well as depth (slightly) overall. This is because the predicted HAZ boundary location and t8/5 time are mainly related to the magnitude of net heat input and the distribution of heat flux in the transverse crosssection. Arc efficiency is crucial to the computed welding temperature field, in particular, the position of HAZ boundary and t8/5 time. This is due to that arc efficiency can significantly affect the magnitudes of both heat flux and net heat input. The suggested adjustment procedure of heat input model is: (i) to acquire a reasonable value or even a small value range of arc efficiency in light of the measured position of HAZ boundary at first, while the values of shape parameters of heat source are equal to the measured size of FZ; (ii) then to adjust heat source shape parameters in terms of the measured size of FZ until the predicted weld dimension matches the measurements well; (iii) to compare the predicted t8/5 time and phase fractions (if SSPT is considered) with the experimental results finally. It is expected that the calculated t8/5 time and phase percentages could also fit the measurements well if the used material properties are reliable. Heat input model parameters almost have no influence on the computed distribution shape and the magnitude of longitudinal residual stress overall. Among the heat input model parameters, the calculated width of longitudinal tensile residual stress distribution is largely determined by arc efficiency. The simulated distribution shape of transverse residual stress nearly does not change with heat input model parameters. Transverse residual stress largely depends on the temperature gradient in the thickness direction, which can mainly be affected by arc efficiency, heat source width and depth in the thickplate beadon (butt) welds, and heat source depth as well as arc efficiency in the highrestrained filletwelded joints. However, the calculated magnitude of transverse residual stress almost does not change with heat input model parameters in the thinplate welds. In numerical simulations, arc efficiency is critical to both the mode and the magnitude of outofplane deflection (buckling distortion) in the thinplate welds. For the thickplate beadon (butt) welds or the highrestrained welds like T joint, arc efficiency plays a decisive role in the calculated magnitude of angular distortion (or transverse bending) which provides the main outofplane deformation in these welds.
