Experimental validation of finite element codes for welding deformations

doi:10.1016/j.jmatprotec.2010.05.014

Journal of Materials Processing Technology

Volume 210, Issue 13, 1 October 2010, Pages 1681-1689

https://doi.org/10.1016/j.jmatprotec.2010.05.014 Get rights and content

Abstract

A single pass metal inert gas welding on an austenitic steel plate has been presented for the purpose of providing controlled experimental data against which numerical codes quantifying welding stresses can be validated. It includes a moving heat source with material deposit, and completes thus existing validation data. The experiment has been addressed by a numerical code, WeldSimS, reproducing qualitatively the distortion during welding quite well. Quantitative differences between the numerical and experimental results, however, indicate the need for more accurate modelling tools than those presently available, which are all based on commonly accepted modelling principles and input data.

Introduction

During welding of e.g. steel, the heat source leads to rapid heating and melting of the material and the formation of a weld pool that subsequently cools and solidifies. Welding stresses arise due to the inhomogeneous cooling, and in steel these stresses are caused by the effects of non-uniformly distributed cooling contractions and, in ferritic steels, solid state phase transformations. Modelling of the welding stresses and associated deformations requires, therefore, a numerical code able to predict both the thermometallurgical and the mechanical development during the process. Ideally, a numerical code should account for a series of complex phenomena such as the material deposition, weld arc heat input, melting and solidification, solid-state phase transformations, work hardening, strain rate sensitivity, and the flow stress dependency on the specific mixture of phases appearing at the different temperatures. Welding in general includes several mechanical phenomena such as transverse, longitudinal, and angular shrinkage, as well as bending distortion (Pilipenko, 2001, chap. 2.2).

However, due to this complexity, as well as the need for minimising the computation time, simplified models are often applied (Taljat et al., 1998, Ferro et al., 2006, Brown et al., 2006, Dean and Hidekazu, 2006). This means that the codes should be carefully validated against controlled experiments before being used to predict welding stresses in real industrial situations. Such validation experiments have been carried out in the present study. The study also addresses the presented validation experiment numerically, investigating the validity of the numerical code WeldSimS (Fjær et al., 2006) implemented in FORTRAN90 by the Institute for Energy Technology.

Validation cases have previously been published, e.g. by Volden (1998) presenting a Satoh, 1972b, Satoh, 1972a experiment, Taljat et al. (1998) who presented welding of a disc, Roux and Billardon (2006) who presented a “multipass” welding of a disc, and Vincent et al., 1999, Vincent et al., 2005, Bergheau et al., 2004 and Depradeux and Jullien, 2004a, Depradeux and Jullien, 2004b at INSA-Lyon who also presented a Satoh experiment and a disc experiment in addition to a single pass tungsten inert gas (TIG) welding. These validation cases include one-, two- and three-dimensional cases represented by the Satoh test, a disc case, and a plate experiment, respectively. They do, however, not include material deposit.

The experiment of the present study is a single pass metal inert gas (MIG) welding on an austenitic steel plate, carried out at the CEMEF¹ research centre. By involving material deposit and a moving heat source, our experiment complements the validation cases referred to above. The objectives of the present paper is firstly to present a validation experiment that completes existing work. Secondly, the present study addresses the lack of accuracy in commonly used modelling equations used in welding simulations.

Section 2 of the paper presents the welding case and the experimental details along with the experimental results. The numerical code including the modelling equations and implementation is summarized in Section 3. The modelling results are presented and compared to the experimental results in Section 4.

Section snippets

The welding case

Fig. 1 shows the experimental setup including the torch and the sample positioned as when ready to be welded. The thick plate used in the experiment is furthermore shown in Fig. 2. It is 136 mm wide, 10.5 mm thick, and 250 mm long in the welding direction. The geometry was chosen small enough to preform 3D modelling with limited calculation time, but still large enough in the welding direction to obtain a thermal quasi-stationary situation. The weld line is marked in Fig. 2 with dashed line, and

Governing equations

Neglecting mechanical dissipation and assuming thermal equilibrium at the microscopic scale, the transient heat transfer is governed by (Haug and Langtangen, 1997): $ρ c_{p} \dot{T} + ρ {\dot{g}}_{liq} Δ H = \nabla \cdot (λ \nabla T)$ where $ρ$ , $c_{p}$ , $λ$ , $T$ , $Δ H$ , and $g_{liq}$ denote the density, specific heat capacity, thermal conductivity, temperature, latent heat of fusion, and the liquid fraction, respectively. Thermal boundary conditions Mortensen, 1999, Carslaw and Jaeger, 1959 include both convection from Fourier’s law and radiation from

Modelling results and comparison to the experiment

The temperature development obtained by WeldSimS and the experimental results from Test 3 are shown in Fig. 16. The figure shows that the heating rate, heating time, and cooling rate obtained by the numerical codes coincide with the experimental results. However, it is clear that the models overshoot the experimentally measured peak temperature. We believe one of the reasons for this discrepancy is the fact that the temperature was not measured at the very tip of the applied 1 mm insulated

Conclusions

A single pass metal inert gas welding on an austenitic steel plate, which includes material deposit and a moving heat source, has been presented for the purpose of providing controlled experimental data against which numerical codes quantifying welding stresses can be validated. Some experimental problems, especially the need for more accurate temperature measurements, have been discussed. Improved temperature measurements could in future experiments be addressed by use of thermocouples without

Acknowledgments

The authors wish to express their thanks to Professor Mohammed M’Hamdi, SINTEF and Faculty of Engineering Science and Technology, Norwegian University of Science and Technology, Norway for contribution of relevant input and suggestions. The funding from the research Council of Norway through the “STORFORSK” Project No. 167397/V30 is greatly acknowledged.

References (32)

T. Belytschko et al.
Hourglass control in linear and nonlinear problems
Computer Methods in Applied Mechanics and Engineering
(1984)
T. Brown et al.
Prediction and measurements of residual stress in repair welds in plates
International Journal of Pressure Vessels and Piping
(2006)
B. Taljat et al.
Numerical analysis of gta welding process with emphasis on post-solidification phase transformation effects on residual stresses
Materials Science and Engineering A
(1998)
Y. Vincent et al.
Thermo-mechanical consequences of phase transformations in the heat-affected zone using a cyclic uniaxial test
International Journal of Solids and Structures
(2005)
T. Belytschko et al.
Nonlinear Finite Elements for Continua and Structures
(2000)
J. Bergheau et al.
Viscoplastic behavior of steels during welding
Science and Technology of Welding and Joining
(2004)
F. Bonollo et al.
Model for CO₂ laser welding of stainless steel, titanium, and nickel: parametric study
Materials Science and Technology
(1993)
H.S. Carslaw et al.
Conduction of Heat in Solids
(1959)
D. Dean et al.
Prediction of welding residual stress in multi-pass butt-welded modified 9Cr–1Mo steel pipe considering phase transformation effects
Computational Materials Science
(2006)
L. Depradeux et al.
2d and 3d numerical simulation of tig welding of a 316L steel sheet
Numerical Simulation of Welding
(2004)

L. Depradeux et al.

Experimental and numerical simulation of thermomechanical phenomena during a tig welding process

Journal De Physique IV

(2004)

P. Ferro et al.

The influence of phase transformations on residual stress induced by the welding process—3D and 2D numerical models

Modelling and Simulation in Materials Science and Engineering

(2006)

H. Fjær et al.

Alspen—a mathematical model for thermal stresses in direct chill casting of aluminium billets

Metallurgical and Materials Transactions B

(1990)

H.G. Fjær et al.

On the use of residual stress from welding simulations in failure assessment analyses for steel structures

J. Goldak et al.

Computational Welding Mechanics

(2005)

J. Goldak et al.

Computer modeling of heat flow in welds

Metallurgical Transactions B

(1986)

Cited by (33)

A comprehensive comparison of modeling strategies and simulation techniques applied in powder-based metallic additive manufacturing processes
2024, Journal of Manufacturing Processes
Additive manufacturing processes have been attracting extensive attention and developing greatly in recent years. These processes have been widely studied by industrial and academic. The manufactured parts are printed layer upon layer according to a computer-aided design model, hence very complex geometries can be created. Moreover, a wide range of materials (plastic, ceramic, metallic, etc.) can be used. Additive manufacturing processes require continuous analyses to investigate their performances and produce high-quality parts with minimal defects. The experimental approach is often used to assess these thermomechanical processes. However, it remains limited because of the exorbitant cost due to the numerous parameters. On the other side, numerical simulation can be an alternative solution to facilitate the optimization of the process parameters and predict the final characteristics of the manufactured parts with a reasonable cost. In the literature, numerous numerical methods and techniques have been developed and applied to simulate Additive manufacturing processes, such as Finite Element Method, Finite Volume Method, Lattice Boltzmann Method, Smooth Particle Hydrodynamics, Discrete Phase Method, etc. Depending on the objectives, different simulation scales have been developed (part scale, melt pool scale, powder scale, etc.). In this context, the proposed paper intends to present an overview of these numerical methods in order to provide a comprehensive comparison and a deep analysis of advantages and drawbacks among the available methods. Some promising techniques and future tendencies are also presented.
Filler metal distribution and processing stability in laser-arc hybrid welding of thick HSLA steel
2020, Journal of Manufacturing Processes
Welds made by high power laser beam have deep and narrow geometry. Addition of filler wire by the arc source, forming the laser-arc hybrid welding (LAHW) process, is very important to obtain required mechanical properties. Distribution of molten wire throughout the entire weld depth is of concern since it tends to have low transportation ability to the root. Accurate identification of filler metal distribution is very challenging. Metal-cored wires can provide high density of non-metallic inclusions (NMIs) which are important for acicular ferrite nucleation. Accurate filler distribution can be recognized based on statistical characterization of NMIs in the weld. In the present study, it was found that the amount of filler metal decreased linearly towards the root. The filler metal tends to accumulate in the upper part of the weld and has a steep decrease at 45–55 % depth which also has wavy pattern based on longitudinal cuts. Substantial hardness variation in longitudinal direction was observed, where in the root values can reach > 300 HV. Excessive porosity was generated at 75 % depth due to unstable and turbulent melt flow based on morphology of prior austenite grains. The delicate balance of process parameters is important factor for both process stability and filler metal distribution.
Residual Stress and Distortion Modeling of Electron Beam Direct Manufacturing Ti-6Al-4V
2018, Thermo-Mechanical Modeling of Additive Manufacturing
In this work, a finite element model is developed for predicting the thermo-mechanical response of Ti-6Al-4V during electron beam deposition. A three-dimensional thermo-elasto-plastic analysis is performed to model distortion and residual stress in the workpiece and experimental in situ temperature and distortion measurements are performed during the deposition of a single bead wide, 16 layer high wall build for model validation. Post-process Blind Hole Drilling residual stress measurements are also performed. Both the in situ distortion and post-process residual stress measurements suggest that stress relaxation occurs during the deposition of Ti-6Al-4V. A method of accounting for such stress relaxation in thermo-elasto-plastic simulations is proposed where both stress and plastic strain is reset to zero, when the temperature exceeds a prescribed stress relaxation temperature. Inverse simulation is used to determine the values of the absorption efficiency and the emissivity of electron beam deposited wire-fed Ti-6Al-4V, as well as the appropriate stress relaxation temperature.
Residual Stress and Distortion Modeling of Electron Beam Direct Manufacturing Ti-6Al-4V
2017, Thermo-Mechanical Modeling of Additive Manufacturing
In this work, a finite element model is developed for predicting the thermo-mechanical response of Ti-6Al-4V during electron beam deposition. A three-dimensional thermo-elasto-plastic analysis is performed to model distortion and residual stress in the workpiece and experimental in situ temperature and distortion measurements are performed during the deposition of a single bead wide, 16 layer high wall build for model validation. Post-process Blind Hole Drilling residual stress measurements are also performed. Both the in situ distortion and post-process residual stress measurements suggest that stress relaxation occurs during the deposition of Ti-6Al-4V. A method of accounting for such stress relaxation in thermo-elasto-plastic simulations is proposed where both stress and plastic strain is reset to zero, when the temperature exceeds a prescribed stress relaxation temperature. Inverse simulation is used to determine the values of the absorption efficiency and the emissivity of electron beam deposited wire-fed Ti-6Al-4V, as well as the appropriate stress relaxation temperature.
A simplified elliptic paraboloid heat source model for autogenous GTAW process
2016, International Journal of Heat and Mass Transfer
Citation Excerpt :
On the other hand, in order to simulate deep penetration welding processes like Laser Beam Welding (LBW) and Electron Beam Welding (EBW) the combination of double ellipsoidal and conical heat source models has been used [44]. The conical heat source has a Gaussian distribution of the power in the radial direction [21–45]. The conical source geometry represents an alternative to model the keyhole effect in the weldpool of the LBW and the characteristic “reversed bugle” configuration of the weld cross-section (Fig. 1a).
The applicability of a welding process depends on the rate of heat input, which determines the residual stresses, the heat affected zone and the microstructural changes in the base material. An adequate approach of heat input through a heat source represents a crucial step in the welding thermal field simulation affecting the accuracy of mechanical and microstructural studies. This research work proposes a volumetric-moving heat source for the Gas Tungsten Arc Welding process (GTAW) based on an elliptic paraboloid geometry capable of representing shallow and deep, wide and narrow fusion zones, considering as shape parameters of the heat source, the fusion width and depth penetration. The interaction of the melting material flow in the weldpool and the heat transfer process were analyzed taking into account the effect of convective heat transfer in the heat input distribution in the fusion zone, and the weldpool shape variations during its displacement. The mathematical model for the GTAW thermal field was solved numerically by means of Finite Volume Method (FVM). The elliptic paraboloid model provided a comparable heat input to the classic double ellipsoid model. The estimated temperatures and the predicted geometry of cross-section weld bead by the proposed model are in a good agreement with experimental results.
Determination of welding heat source parameters for fem simulation based on temperature history and real bead shape
2023, Materials Research Proceedings

View all citing articles on Scopus

View full text

Experimental validation of finite element codes for welding deformations

Abstract

Introduction

Section snippets

The welding case

Governing equations

Modelling results and comparison to the experiment

Conclusions

Acknowledgments

Computer Methods in Applied Mechanics and Engineering

International Journal of Pressure Vessels and Piping

Materials Science and Engineering A

International Journal of Solids and Structures

Nonlinear Finite Elements for Continua and Structures

Viscoplastic behavior of steels during welding

Science and Technology of Welding and Joining

Model for CO2 laser welding of stainless steel, titanium, and nickel: parametric study

Materials Science and Technology

Conduction of Heat in Solids

Prediction of welding residual stress in multi-pass butt-welded modified 9Cr–1Mo steel pipe considering phase transformation effects

Computational Materials Science

2d and 3d numerical simulation of tig welding of a 316L steel sheet

Numerical Simulation of Welding