Toshio Terasaki

Mechanical Behaviour of Joints in FSW: Residual Stress, Inherent Strain and Heat Input Generated by Friction Stir Welding

For ordinary arc welding processes such as MIG, welding deformation, residual stress and inherent strain are important factors to control the reliability of weldments. Friction stir welding (FSW) is a new welding method, especially for light metals. The distortion of FSW is reportedly small but reasons can not be determined in detail. This paper deals with thermal cycles and inherent strains which create both residual stresses and welding deformations, in order to make clear effective factors that lower apparent strains on FSW than on MIG. Inherent strains were measured by using the layer removal technique. Thermal cycles and maximum temperature rise were measured by type K thermocouples. It is concluded from experimental data that the main cause lowering apparent strain in FSW is the compressive load that makes tensile plastic strain in weldments. The second cause is that the heat input is smaller than in ordinary arc welding processes.

Development of advanced resistance spot welding process using control of electrode force and welding current during welding

In resistance spot welding of thin sheet–thick sheet–thick sheet joint, when the sheet thickness ratio is large (sheet thickness ratio = total thickness of sheet joint/thickness of the thin sheet positioned on the outside of the joint), how to stably secure the nugget between the thin sheet and the adjoining thick sheet is a key issue. If the sheet thickness ratio is so large, nugget formation between the thin sheet and thick sheet is extremely difficult. In order to control of the nugget (position of formation, shape, etc. of the nugget) during welding for three sheets joint with a high sheet thickness ratio, optimum welding process was investigated. The developed ‘two-step force, two-step current’ welding process was suitable for high sheet thickness ratio joint and relaxed the constraints on the sheet thickness ratio. In Step 1 (first part of welding period) of the welding process, a nugget is reliably formed between the thin sheet and thick sheet by applying conditions of low electrode force, short welding time, and high current. In the subsequent Step 2 (second part of welding period), a nugget is formed between the two thick sheets by applying high welding force and a long welding time. In the weld results of a three sheet joint (0.7+2.3+2.3 mm; sheet thickness ratio: 7.6) using mild steel GA (0.7 mm) as the thin sheet and 780 MPa high strength GA (2.3 mm) in the two thick sheets, ‘two-step force, two-step current’ spot welding process showed the wide available welding current range.

Characterization of blasted austenitic stainless steel and its corrosion resistance

It is known that the corrosion resistance of stainless steel is deteriorated by blasting, but the reason for this deterioration is not clear. A blasted austenitic stainless steel plate (JIS-SUS304) has been characterized with comparison to the scraped and non-blasted specimens. The surface roughness of the blasted specimen is larger than that of materials finished with #180 paper. A martensite phase is formed in the surface layer of both blasted and scraped specimens. Compressive residual stress is generated in the blasted specimen and the maximum residual stress is formed at 50–100 µm from the surface. The corrosion potentials of the blasted specimen and subsequently solution treated specimen are lower than that of the non-blasted specimen. The passivation current densities of the blasted specimens are higher those of the non-blasted specimen. The blasted specimen and the subsequently solution treated specimen exhibit rust in 5% sodium chloride (NaCl) solution, while the non-blasted specimen and ground specimen do not rust in the solution. It is concluded that the deterioration of corrosion resistance of austenitic stainless steel through blasting is caused by the roughed morphology of the surface.

New Concept of Equivalent Inherent Strain for Measuring Axisymmetric Residual Stresses

A new concept, the equivalem inherent strain g θ eq , is proposed for measuring axisymmetric residual stresses, based on the inherent strain theory. g θ eq is a parameter obtained by incorporating the radial inherent strain component into the tangential inherent strain component in computing the axially uniform axisymmetric residual stresses. The application of this concept can avoid the restriction of radial inherent strain component estimation and hence leads to a more simple and accurate method for measuring axisymmetric residual stresses. Experimental measurement and numerical simulation were carried out to demonstrate that this method has both the simplicity of the Sachs method and the high accuracy of the inherent strain method.

Influence of Dehydrogenation Heat Treatment on Hydrogen Distribution in Multi-Layer Welds Of Cr-Mo-V Steel

Cr-Mo-V steel normally undergoes a dehydrogenation heat treatment (DHT) of 350 °C × 4 h after welding, to minimize the susceptibility to cold cracking due to residual hydrogen in the weld. For low ambient temperatures or failure to sufficiently heat the weld during the DHT holding period, it is sometimes difficult to maintain and guarantee the 350 °C for 4 h. We therefore surmised that lower temperature DHT over a longer time period could be substituted for the standard DHT conditions and still achieve the same dehydrogenation effect. In this paper, using experimental and numerical methods to measure hydrogen diffusivity and the influence of DHT on Cr-Mo-V steel welds, we demonstrate that even at temperatures as low as 280 °C held over longer time periods, there is an equivalent dehydrogenation effect as in the existing conditions.

Prediction of static fracture strength of laser-welded lap joints by numerical analysis

In a previously reported experimental study of the static fracture strength of laser-welded lap joints, the authors have determined that static fracture strength is susceptible to the effects of bead configuration parameters. The bending moment generated when the axial forces of the two sheets constituting the lap joint are parallel and opposite affects the static fracture strength of joints. The static fracture strength of lap joints is mostly investigated experimentally, with numerical treatments being little studied. In a previous study addressing fundamental aspects of the ductile fracture strength of round bars, however, Otsuka et al have shown the ductile fracture strength to be amenable to numerical treatment through consideration of the equivalent plastic strain. The static fracture strength of lap joints as a previously reported experimental parameter also offers scope for numerical treatment through consideration of the equivalent plastic strain. This article initially describes an investigation of a numerical calculation method which explains the experimental values of load–displacement curves obtained in tensile shear tests. The proposed numerical calculation method is then used to investigate whether the experimentally determined static fracture strength of lap joints can be predicted by the equivalent plastic strain.

Welding deformation produced by two-pass welding

Multilayer welding is considered to be an accumulation of two-pass welding, and it was clarified that the basic factors of welding deformation were produced by two-pass welding. From comparison between the experimental results and numerical analysis, it was found that the numerical analysis method used in this research was very accurate to predict the welding deformation. Also, the influences of the welding heat input and the distance between two-passes on the welding deformation were examined by using the numerical analyses. As a result, it was found that the longitudinal shrinkage made from each pass was decided by the inherent strain distribution parallel to the weld line. The inherent strain distribution parallel to the weld line after two-pass welding was a larger value at one- or two-pass welding. On the other hand, it was revealed that transverse shrinkage made from each pass was decided by the integration of inherent strain distribution perpendicular to the weld line. In the case of the inherent strain distribution perpendicular to the weld line after two-pass welding, it was clarified that the inherent strain was almost same as the sum of inherent strain distribution perpendicular to the weld line made from each pass.

Source of welding residual stress and deformation

Figure 1 shows the expansion of a 10-yen coin heated steadily from room temperature (assumed to be 208C) to 1008C. In this case, the coin is subjected to a steady temperature rise of 808C from 20 to 1008C and consistent thermal strain occurs without any thermal stress. Since the linear coefficient of expansion a (8C) corresponds to the strain per 18C of temperature change, at a temperature change of DT (8C), the thermal strain is aDT. In the example shown in Figure 1, DT 1⁄4 100 2 20 1⁄4 808C and the thermal strain is 80a. If the 10-yen coin was to be composed of aluminium, since the linear coefficient of expansion of aluminium is 2.4 £ 10 (8C), the thermal strain is 80 £ 2.4 £ 10 1⁄4 0.00192 and, if the diameter of the 10-yen coin is D (mm), then the thermal expansion is 0.00192D (mm) (expansion is deformation). If the 10-yen coin was to be composed of iron, since the linear coefficient of expansion of iron is 1.2 £ 10(8C), the thermal strain is 80 £ 1.2 £ 10 1⁄4 0.00096 and the thermal expansion is 0.00096D (mm). It is thus clear that the difference in the thermal expansion of aluminium and iron is manifested only in a difference in their linear coefficients of expansion. It is important to note that thermal expansion strain aDT is caused by temperature change DT. Thus, no thermal strain occurs when the temperature change is zero. 2.2 Source in deposited metal

Numerical Analysis Method for Temperature Distribution in Cylindrical Steel during Quenching Process

The technics for predicting thermal cycle of a quenching by using the finite element method have been studied by comparing theoretical values with numerical analysis results. The basic equation of a minimum element length for cylinder thermal cycle was proposed, which was able to calculate from informations such as an initial temperature, a radius of cylinder, a heat transfer coefficient and the aimed temperature change at the cylinder surface. The minimum time increment for guaranteeing accuracy was given corresponding to the minimum element length. It was shown that the minimum element length for pipe thermal cycle was replaced by that of cylinder. Temperature distribution in cylinder during quenching process can be sufficiently predicted by the finite element method with the above, mentioned minimum element length and the minimum time increment.

Prevention of HAZ cracking in corner joints of high heat input SAW box columns

Summary One-pass submerged-arc welding (SAW) with high heat input is extensively used for the corner welds of box columns, though internal cracking of lamellar tearing type in some cases occurs along the centreline of the flange plate thickness. This paper describes an investigation of weld microstructures, hydrogen concentrations, and residual stresses at the cracking position to clarify the cracking mechanism. Appropriate measures to prevent cracking in terms of material- and fabrication-related factors are also studied on the basis of laboratory-scale one-pass SAW experiments and numerical analysis of hydrogen diffusion. The generation and propagation of cracking are closely related to the transition of the hydrogen concentration at the cracking position, whereas the effect of the residual stress on cracking is small. The material-related factors controlling this cracking include elongated manganese sulphide (MnS) in the centre segregation band and martensite-austenite constituent (M-A) around the MnS ...