Initial Welding Residual Stresses Research Articles

Re-Distribution of Welding Residual Stress in Fatigue Crack Propagation Considering Elastic–Plastic Behavior

The welding residual stress re-distribution behavior during fatigue crack propagation in butt-welded high-strength steel plates for ship construction is investigated based on experimental test results and numerical analyses. The specimens’ initial welding residual stresses are obtained from X-ray for middle tensile (MT) specimens cut from butt-welded high-strength steel plates. Then, fatigue crack propagation experiments on MT specimens are conducted, and a strain gauge is used to measure the residual stress re-distribution field around cracks. A practical fatigue crack propagation simulation procedure is developed with a dynamic update of in-situ welding residual stress, where the residual stress intensity factor Kres of the MT specimen is deduced. The stress ratio effect on Kres during fatigue crack propagation is analyzed and a good agreement between experimental and numerical results is achieved.

Approximate residual stress and plastic strain profiles for laser-peened alloy 600 surfaces

This paper presents approximate in-depth residual stress and plastic strain profiles for laser-peened alloy 600 surface via FE analysis. In approximations, effects of the initial welding residual stress and the number of shots are quantified. Based on FE analysis results, residual stress profiles are quantified by two variables; the maximum difference in stress before and after LSP, and the depth up to which the compressive residual stress exists. Plastic strain profiles are quantified by one variable, the maximum equivalent plastic strain at the surface. The proposed profiles are validated by comparing with published LSP experimental results for welded plates. Effects of the initial welding residual stress and the number of shots on these variables are discussed. The proposed profile can be directly applied to predict the mitigation effect of LSP on PWSCC and to efficiently perform structural integrity assessment of laser peened nuclear components.

A simplified method for fatigue life evaluation of welded joints in OSD considering welding residual stress relaxation

The welding residual stress at the welded joints of orthotropic steel decks (OSD) has a significant influence on the fatigue life, and welding residual stress relaxation (WRSR) occurs under vehicle loading. The existing fatigue evaluation method for steel bridges does not consider the WRSR on fatigue life and may cause a large deviation between the fatigue life prediction results and the fatigue cracking in the field. This paper proposes a simplified fatigue life evaluation method that can consider WRSR quantitatively. First, the initial welding residual stress (IWRS) distribution model is established based on residual stress tests and numerical simulation. Second, the simplified mechanical model of the coupled effect of IWRS and vehicle-induced structural stress as well as the law of WRSR are established via stress analysis at the weld toe section of DRWJs under external loading. Furthermore, the peak/valley value of the coupled effect of IWRS and vehicle-induced structural stress are determined, and a stress time-history simplification method considering WRSR is proposed. On this basis, a simplified fatigue life evaluation method considering the WRSR of DRWJs on OSDs is finally proposed and combined with the Q345 steel full-range SN curve model. The Jiangyin Yangtze River Highway Bridge is used as an example to demonstrate the fatigue life evaluation procedure of this proposed model in detail. Based on the multiscale finite element model (FEM) considering the coupling effect of the WRS and vehicle loading stress, the real stress time-history curves of DRWJs under vehicle loading cases are extracted and compared to those of the simplified method. Based on the full-range SN curve model considering IWRS and WRSR, the fatigue life of the welding joints of Jiangyin Bridge is finally calculated, and the results are more accurate and closer in value to the field observations of fatigue cracking when compared with the Eruocode3 standard. This provides a theoretical basis for fatigue analysis and lifetime assessment technology of large span highway steel box girder bridges.

Distribution and relaxation of welding residual stress in U-rib-to-deck joints considering actual material properties

Welding residual stress (WRS) adversely affects the fatigue cracking of U-rib-to-deck joints. The distribution and relaxation of the WRS of a U-rib-to-deck joint were calculated and analyzed by combining measurement and simulation and considering the inhomogeneity of the material. First, the initial WRS was simulated by using the actual high-temperature material properties of the base metal (BM) and weld metal (WM), and the simulation value was verified against measured data. Second, based on the simulated initial WRS, its relaxation was simulated using the Chaboche cyclic-plasticity constitutive model of BM and WM, and the influence of cycle number, stress amplitude, and stress ratio on WRS relaxation was investigated. The results show that the WRS relaxation changes its value but not its distribution trend. Most of the WRS relaxation occurs in the first cyclic loading, accounting for approximately 90% of the total relaxation. The relaxation increases with cycle number increasing until a stable state is reached after between seven and ten cycles. The WRS relaxation increases with stress amplitude or stress ratio increasing. Finally, a predicted equation for WRS relaxation was proposed based on the cycle number, stress amplitude, and stress ratio.

Local and global energy densities associated with welding residual stresses of cracked structures

The residual stresses associated with fusion welding processes have recently become a popular issue considering the reliable design of structural components. In a welding problem, it is quite common that fracture is observed to initiate either directly from the welds or from fatigue cracks that develop in the weld’s heat affected zone (HAZ). A recent study has investigated the effects of subsequent dynamic loading with stationary crack fronts, which assumes load being applied to the structure following the welding process. To analyze these types of problems, a computational procedure based on finite element methodology is developed to integrate the welding residual stresses with subsequent dynamic fracture behavior. The welding analysis is performed with a welding simulation program (SYSWELD). The inclusion of residual stresses and the determination of classical fracture parameters are handled with a specialized finite element program, called FRAC3D. The calculation of energy densities that are recently used in the analysis of cracked structures was shown to be an effective method for the solution of fracture problems. In this paper, a new methodology that is developed to determine local and global energy densities for the simulation of dynamic fracture behavior with initial welding residual stresses is presented.

Redistribution of Welding Residual Stresses of Crack Tip Opening Displacement Specimen by Local Compression.

The demand of crack tip opening displacement (CTOD) test which evaluates fracture toughness of a cracked material is very important to ensure the stability of structure under severe service environment. The validity of the CTOD test result is judged using several criterions of the specification standards. One of them is the artificially generated fatigue pre-crack length inside the specimen. For acceptable CTOD test results, fatigue pre-crack must have a reasonable sharp crack front. The propagation of fatigue crack started from the tip of the machined notch, which might have propagated irregularly due to residual stress field. To overcome this problem, test codes suggest local compression method, reversed bending method and stepwise high-R ratio method to reduce the disparity of residual stress distribution inside the specimen. In this paper, the relation between the degree of local compression and distribution of welding residual stress has been analyzed by finite element analyses in order to determine the amount of effective local compression of the test piece. Analysis results show that initial welding residual stress is dramatically varied three-dimensionally while cutting, notch machining and local compressing due to the change of internal restraint force. From the simulation result, the authors find that there is an optimum amount of local compression to modify regularly for generating fatigue pre-crack propagation. In the case of 0.5% compressions of the model width is the most effective for uniforming residual stress distribution.

Redistribution of welding residual stress following high plastic deformation in seamless pipes

This paper summarises the findings of a Joint Industry Project (JIP) led by DNV GL on the treatment of residual stress on pipes undergoing high plastic deformation. The paper presents measured residual stresses and some of the full-scale test results supported by Finite Element Analysis (FEA) for the seamless pipes before and after high plastic deformation on selected girth welds. Several FEA models of the reeling simulation have been developed including mapping an initial axial residual stress (transverse to the weld) profile onto a seamless girth-welded pipe. The initial welding residual stress distribution used for mapping was based on a number of through thickness measurement along the circumference of the girth welds. The predicted residual stresses after reeling simulation was subsequently compared with experimental measurements. It is evident that FEA models can predict welding residual stress re-distribution during the reeling process, but some differences with measurements are present through the thickness.

Analytical modelling of shot-peening residual stress on welding carbon steel surface layer

The residual stress distribution was studied by an analytical model, due to shot peening on the welding carbon steel surface layer. The initial welding residual stresses before shot peening were taken into consideration in this analytical model. The Hertzian elastic contact theory was used to get the elastic compression stress state after impact on the surface layer. The initial welding stress field and the shot peening stress field would superpose and the welding surface layer would yield based on the elastic-plastic evaluation, then the residual stress after shot peening can be achieved. The influence of initial welding residual stress on the stress distribution after shot peening was analyzed and discussed. A series of experiments were carried out and the residual stress on the welding surface was determined by X-ray diffractometer before and after shot peening. The calculation results of the analytical model are consistent with the experimental results. The critical shot velocities when welding surface layer yielded and reverse yielded were calculated. While the welded joint surface material reversely yielded, the maximum compressive residual stress would not obviously increase with the increase of shot velocity, the thickness of the compressive stress layer would be increased. Welding residual tensile stress can enlarge the thickness of the compressive stress layer at the same shot velocity when reverse yield appeared.

Welding Residual Stress Behavior in Tubular Steel Joints under Multiaxial Loading

AbstractThe lack of insight into the initial welding residual stress field and its behavior during the different phases of the fatigue damage namely crack initiation and propagation has led to conservative assumptions in fatigue design codes of welds. Since a significant amount of fatigue failures in welded joints are caused by torsion or combined tension-torsion in machinery components, estimating the potential threat of the inevitable residual stresses to structural integrity seems to be mandatory for the design of the future tubular lightweight welded components and structures. In this paper the residual stresses in cylindrical specimens with bead on tube welds out of S355J2H were determined experimentally by means of X-ray and neutron diffraction. Welding residual stresses as high as almost the yield strength of the base metal were observed in the weld bead. At the weld toes which are potentially fatigue crack initiation sites, however compressive residual stresses were present. After discussing briefly the sources and origins of the residual stress field, its behavior under pure torsional and combined tension-torsion loading will be presented.

Welding Residual Stresses in Tubular Joints

In spite of an increased awareness of welding residual stress threat to structural integrity, the extent of its influence on fatigue especially under multiaxial loading is still unclear and is a matter of debate. One important reason for this lack of clarities is that the determination of the initial welding residual stress field in welded structures even at the fatigue crack initiation sites is difficult and requires complementary instruments. Since the fatigue crack initiation in sound welds almost always occurs on the surface, the determination of surface residual stresses could increase the awareness of the extent of their threat to the structural safety. In this paper the development of residual stresses in different TIG-welded tubular specimens out of S355J2H and S690QL steel is studied and compared. The mechanisms of the development of residual stresses based on heat input and cooling rate are discussed. The welding parameters and thus heat inputs are varied and the mechanisms leading to different residual stress states are investigated. X-ray method was used for residual stress state characterization.

Welding residual stress behavior under mechanical loading

For an accurate estimation of the fatigue performance of welded joints, not only the initial residual stress field but also its variation under load is decisive. The portion of residual stress which stays stable can shift the load stress range much the same way as the mean stresses do in fatigue loading. The relaxation of residual stresses during fatigue loading however reduces this hazard to the structural health. That is, before considering the influence of residual stresses in fatigue, the effect of fatigue on residual stresses should be understood. In this work, the influence of the quasi-static and cyclic loading on the relaxation of welding residual stresses in small specimens and large components out of different steels namely S235JRG2, S355J2G3, P460NL, S690QL, S960QL, and S1100QL has been investigated experimentally. X-ray diffraction analysis as a reliable method has been used for the determination of residual stress profiles in surface layers. For residual stress analysis in deeper layers, synchrotron and neutron diffraction techniques were applied as complementary methods. A clear recognition of the difference between initial welding residual stresses in small- and large-scale specimens was observed. Tensile residual stresses as high as the yield strength could appear in large-scale welded specimens. That was not the case for small-scale welds due to the low grade of restraint. Nevertheless, in both types of samples, the highest residual stresses were obtained in the weld centerline. At the weld toe which is critical regarding fatigue crack initiation, the magnitude of the residual stresses is lower. Beside of that, it was observed that in the case of relaxation, the first load cycles especially in components out of low-strength steels are decisive. The influence of loading conditions and local mechanical properties on the relaxation of welding residual stresses was investigated based on principles of solid mechanics. The von Mises failure criterion was able to describe the relaxation behavior.

Numerical Simulation of Welding Residual Stress and its Creep Strain for Cr5Mo Steel Welded Joint

By the finite element analysis codes ABAQUS and the function of RESTART, a sequential coupling calculating program for creep based on welding residual stress has been developed by establishing the model of temperature field, residual field and creep analysis. The welding residual stress and creep of welded joint for Cr5Mo furnace tubes was simulated by the program. This method established the basis for calculating the effects of welding residual stress on the creep of welded joint. A creep comparison between the welded joint tubes with welding residual stress and with internal pressure was made in the paper. The results show that although the residual stress is greatly decreased due to the creep relaxation behavior at high temperature, the initial welding residual stress has a great influence on the tube creep. The research results provide a reference for strength design of high temperature components.

Finite Element Analysis of the Effect of Residual Stress on Mechanical Behavior of Welded Plates

Residual stresses play an important role in the mechanical behavior of steels and welded structures. To examine the effect of residual stresses on tensile behavior and fatigue, residual stresses in the specimens were generated by welding. Experimental stress-strain curves of the specimens with/without residual stresses were obtained and compared to simulated curves obtained by the finite element analysis. The two results are in a good agreement. Finally, to study the relaxation of the residual stresses during fatigue crack propagation, we carried out fatigue crack propagation analysis by a 3-D cohesive zone model. Initial welding residual stresses decrease as the number of cycles increases.

용접잔류응력의 이완과 재분포 해석 및 실험적 검증

For the precise assessment of the effect of welding residual stresses on structural strength and fatigue crack growth behavior, new FE analysis algorithms for the estimation of residual stress relaxation due to external load and redistribution due to fatigue crack propagation were proposed in this paper. Initial welding residual stress field was obtained by thermal elasto-plastic analysis considering temperature dependent material properties, and the amount of residual stress relaxation and redistribution were assessed by subsequent elasto-plastic analysis In the analysis of fatigue crack propagation, the applied SIF(Stress Intensity Factor) range was evaluated by <TEX>$\frac{1}{4}$</TEX>-point displacement extrapolation method, and the effect of welding residual stresses on crack propagation was considered by introducing the effective SIF concept. The test results of crack propagations were compared with the predicted data obtained by the analysis.

CYCLIC CRACK RESISTANCE OF WELDED JOINTS WITH NON‐UNIFORM FIELDS OF RESIDUAL WELDING STRESSES

Abstract— The use of linear elastic fracture mechanics to describe the kinetics of fatigue fracture of welded joints with high welding residual stresses (WRS) is experimentally evaluated in this paper. A correction analysis is used to show that the crack propagation rate of cracks in joints, as a function of the applied stress intensity factor, is linear on a log‐log scale in the Paris regime when non‐uniform fields of WRS are superimposed on the applied cyclic loading. It is shown that crack growth rates in joints with high WRS do not depend on the characteristics of the loading cycle. The parameters of the Paris exponential equation are determined by the initial WRS distribution, by the range of cyclic stresses and by the load ratio. A method for calculating the cyclic crack resistance of joints is proposed which explicitly allows for a non‐uniform field of WRS that influences the fatigue crack growth rate.

溶接残留応力場での疲労き裂成長速度の評価

The present paper is one of researches which are carrying out to clarify the effect of local constraint in the weld zone on fatigue crack growth rates in welded joints. This local constraint may be produced by welding residual stresses, the microstructure and the mechanical properties. In particular, the effect of welding residual stresses on the behavior of which the fatigue crack propagates perpendicularly through the weld bead has quantitatively been investigated by means of fracture mechanics and fractography in association with the crack length to residual stresses field and the stress ratio.The fatigue crack growth rates in welded joints were markedly influenced by welding residual stresses. Namely, the growth rates of which the crack propagated perpendicularly through the weld bead decreased drastically by the compressive residual stresses at some stress intensity level and increased again with increasing stress intensity beyond the minimum value of crack growth rate. This minimum value depended on the initial welding residual stresses, the crack length to residual stresses field and the stress ratio. Therefore, it became clear that the fatigue crack propagation behavior into welding residual stresses fields is well explained by the effective stress intensity factor to define from both the welding residual stresses- and the plastic-induced crack closure.