A new method and its principle for sensing droplet transfer in GMAW of steel and aluminum alloy have been researched in this paper. A practical arc light sensing and controlling system has been developed. The reliability of the arc light characteristic signal that indicates droplet detachment and the control accuracy of the system have been verified using high-speed photography. Based on the mathematical model set up in the paper, the mechanism of the signal and relative phenomena were analyzed. The parameters which influence arc light radiation were given and the limitation of this sensing method was discussed.
The evaluation of an interface migration for the diffusion bonding process in which the temperature continuously changed was examined based on the kinetics and the Ni diffusion at the bonding interface of dissimilar materials of both carbon-steel/ Ni and ferritic stainless steel/austenitic stainless steel. On the bonding interface of SUS444/316L the γ phase was transformed into the a phase by Ni diffusion from the γ phase to the α phase side. The distance of interface migration was obtained analytically from the balance of Ni atoms that enter and exit at the interface. An interface migration in the case of multistep and continuous cooling was confirmed to be predicted from the calculation of an interface migration in the state of constant temperature.
A spatter reduction process in pulsed CO2 gas shielded arc welding has been studied by focusing on the rectangular pulsed wave form control and the chemical compositions of welding wire. The spatter in pulsed CO2 arc welding is mainly caused by the reason that a molten droplet on the welding wire is blown off by arc repulsion force at the peak current duration. The droplet blow off depends on a up-slope of peak current, and the rapid-up-slope induces the increase of a large spatter. Furthermore, the droplet blow off is influenced by the deoxidizing elements. The increase of deoxidizing elements (Al, Ti, Si) in the chemical composition of welding wire greatly reduces the spatter from molten pool and droplet necking. However, surplus content of deoxidizing elements spoils the stability of 1 pulse 1 drop transfer, and a molten droplet is blown off at the peak current duration. With the optimization in wave form control and content of deoxidizing elements in the wire, the amount of spatter decreases to 1/6 in comparizon of that the conventional CO2 arc welding.
The static and fatigue strength of laser welded lap joints with different zinc coating weight of steel sheets were quantitatively investigated. The welding phenomenon of zinc coated steel sheet was observed by the laser strobe. From the observation results, the mechanism of the weld defect formation was examined. The following results have been confirmed: (1) The quantity of spatters from the weld pool increased with the increase of zinc coating weight of steel sheets. (2) In the case of steel sheets with zinc coating weight of more than 40 g/m2, 30-40% of the weight of molten metal scattered as spatters. (3) Blow holes were formed by sputters. (4) The static and fatigue strength of laser welded lap joints decreased with the increase of zinc coating weight of steel sheets. (5) The gap width between the sheets suppressed spatters and blow holes. (6) With regard to the mechanism of spatter formation, the driving force of sputter generation is laser energy, not zinc vaporization, which disturbs the molten metal flow around the keyhole.
In order to develop a fast surface treatment method or a fast welding method using an electric arc with a laser beam, stabilizing effects of a CO2 laser on a TIG arc were investigated. A TIG arc makes discontinuous molten pools even when the travelling speed of a base metal is high and the distance between a base metal and a tungsten electrode is large. In such a case, the behavior of an arc usually becomes unstable and the arc voltage changes in the range of 5-8 V. Whereas, the arc behavior became stable by using a CO2 laser beam simultaneously, and good continuous weld beads were made on conditions that the TIG arc current, the laser output, the distance between a SUS304 base metal and a tungsten electrode, and the travelling speed of a base metal were 100 A (DCEN), 2 kW, 15 mm and 200 mm/s, respectively. It was found that the anode spot was fixed on the molten pool formed by the laser beam, and, the molten pool, the laser plume and the metallic vapor affected the stabilization of the path of a TIG arc and its voltage. Therefore, it is effective to controle a long arc by a laser that the beam irradiates near the tip of the arc where the arc stiffness becomes weak.
The formation of porosity in CO2 laser welding of brazed Al alloy plate was investigated. In order to evaluate the influence of brazing layer on the porosity formation, two types of specimens were welded. One was the brazed plate in which two plates were bonded, and the other was the Al plate without brazing layer. The obtained results are as follows. (1) The number of porosity was less in the weld fusion zone of Al alloy plate than that of the brazed plate. It was made clear that melting the brazing layer caused the porosity formation. (2) The joint shape, in which the brazing layer in weld groove was removed, was significant for reducing the porosity. (3) High speed welding was effective for reducing the porosity in the brazed plate. It was suggested that high speed welding narrowed the melted zone of brazing layer, leading to the reduction in porosity.
In this paper, energy balance of an argon-helium mixed gas tungsten arc was systematically discussed as corresponding to the demixing effect proposed in the previous reports. Arc voltage increased as introducing helium gas into argon gas shielded arc. This tendency was caused by the increase of space potential near the cathode region in 75% below helium contents and the increase of potential gradient of arc plasma and space potential near the anode in almost pure helium content. Anode heat input and thermal efficiency increased as introducing helium gas into argon gas shielded arc. It was found that the increase of anode heat input was independent on the energy transported by electrons, but was depend on the heat conduction and convection. Furthermore, it was shown that the heat losses of the arcs reached to the minimum in 75% helium content. The results were caused by two phenomena that the radiative power decreased as increasing in the helium content, and the losses of heat conduction and convection gradually increased as increasing in the helium content and then steeply increased in almost pure helium content.
Diffusion bonding of copper at low bonding temperature and pressure was studied with the purpose of obtaining no deformation bonding. OFHC (JIS C1020P) was used for copper plate (80 mm diameter×8.7 mm thickness) and its surface was machined finely to roughness Ry 20 nm, flatness 0.3 μm/80 mm diameter by ultra precision lathe. Main bonding parameters were varied 573-973K (bonding temperature), 0.005-0.15 MPa (bonding pressure). Another parameters were applied constant value, 3.6 ks (bonding time), 4×10-4 Pa (atmosphere). Optimum bonding parameters attaining high joint strength and extreme small deformation was selected from test results. Finally, trial production of accelerating cell (X band cell) on Japan Linear Collider Development Plan was done by diffusion bonding method obtained with this study. Main results obtained are as follows : (1) Excellent joint properties which joint strength was equivalent to base metal and joint deformation was extremely small (limit of measurement) were obtained with following bonding parameters. Temperature : 773-973K Pressure : 0.01-0.15 MPa (Temperature : 773 K) 0.005-0.05 MPa (Temperature : 973 K) Time : 3.6 ks Atmosphere : 4×10-4 Pa (2) Trial production of accelerating cell (above mentioned copper plate was lapped 30 pieces) was done with diffusion bonding (temperature : 973 K, pressure : upper joint 0.02 MPa-bottom joint 0.04 MPa, time : 3.6 ks, atmosphere : 4×10-4 Pa). Excellent leak tightness (<3.6×10-12 Pa·m3/s) and extreme small deformation (direction of diameter and axis) were obtained. These results indicated practical use.
As for the blowholes to be generated in high speed fillet welding of hot-dipping galvanized and annealed steel sheet lap joints, the types of materials to cause them as well as their formation mechanism were examined. It seems to us that the formation of blowholes is not caused by the organic substances usually stuck to the surface of base materials but by the vaporization of zinc. As the result of chemical analysis of the gases sampled from the blowholes by breaking them, it was found that the main contents were hydrocarbon including methane gas. Hydrogen gas was also detected at the same time but it was only about a tenth of methane gas so that the possibility of hydrogen gas causing blowholes is low. The results of chemical analysis of zinc at both blowhole wall faces and root parts seem to indicate that zinc is supplied from heat affected zones to the origin of blowholes wherefrom it pushes up the top side of blowholes for their growth. From the above, the blowholes can be presumed to be caused by the vaporization of zinc. Intending to oxidize zinc as a means to control its vaporization, the effect of adding oxygen gas to the mixed gas of Ar-CO2 was examined. As the result, it was found that the effective addition rate of oxygen gas for the reduction of blowholes was about 5% and especially effective in the case of joint gaps of 0.2 to 0.4 mm. However, too much addition of oxygen gas spoils this effect. Therefore, using the consumable electrodes of different melting points, the vaporization characteristics of zinc oxide were examined. By selecting the type of filler metal having the melting point of about 1, 400°C and using the mixed gas including oxygen gas, the effectiveness of reducing blowholes was clarified.
In case of arc welding of hot-dip zinc-coated carbon steel sheets, the generation of spatters, blowholes and pitting causes some problems due to the vaporization of zinc while welding. To solve these problems, the method to positively vaporize zinc both in front of a molten pool and in joint gaps by taking full advantage of the property of zinc of easy vaporization has been examined in an ordinary single electrode process of high practical application. As the results, it has been found that for the reduction of generation of spatters, to do welding at high speed using comparatively high current and also to avoid the short-circuiting by refining the droplets from a wire using the pulse current wave shape of high peak current and a short time in a pulse arc are effective. Practically the effect of both current and welding speed on the generated amount of blowholes has hardly been seen. It is the basic guidelines for the reduction of generation of blowholes that the application of a pulse MAG welding process of high arc time ratio is more advantageous than a CO2 welding process accompanied by short-circuiting and to avoid the welding positions wherein molten metal is apt to cover zinc layers. As for pitting, even in the conditions wherein many blowholes are generated, it is little generated in either a flat or a horizontal welding position but significantly generated in a vertical downward welding position. It is advantageous for the reduction of generation of pitting in a vertical downward welding position, to decrease welding speed, to add oxygen to shielding gas and to tilt a welding torch in push angle.
It is important to develop a mathematical method for determining the optimum process parameters in welding engineering. An algorithm for optimizing process parameters to achieve the required temperature field on the specimens has been developed. The optimization of process parameters refers to the determination of optimum heat inputs, such as arc current to make the calculated temperature fields correspond with required ones in the present work. That is, the objective function is defined as the optimization problem of minimizing the difference between calculated and required temperatures. The nonlinear programming methods, such as the convex programming and the steepest descent method, are applied to solve the optimization problem in this study. It has been shown that the algorithms based on the optimizing methods are reliable enough to determine the optimum heat inputs in GTA (Gas Tungsten Arc) welding on both the plate and pipe specimens.
The temper embrittlement of the high-temperature type arises in the heat affected zones (HAZs) when the welded constructions of Cr-Mo steels were subjected to the stress-relief (SR) heat treatment in the temperature range above 900 K. In the temperature range below 900 K, there arose four types (the first to the fourth types) of temper emgrittlement as shown by the authors in the previous papers. The temper embrittlement of the high-temperature type (the fifth type) arising in 21/4Cr-1Mo and 11/4Cr-1/2Mo steels was examined in this research in connection with those four types. Charpy impact test was made on the synthetic HAZ specimens tempered at 900 K to 1000 K. Experimental results informed that 21/4Cr-1Mo steel in possiblly embrittled by SR treatment, because the SR temperature of this steel (about 975 K) meets that at which the fifth type arises. Shorter holding time is recommended for avoiding the embrittlement. 11/4Cr-1/2Mo steel is more possiblly suffered from the third type rather than the fifth type, because the SR temperature of this steel (about 925 K) meets that in which the third type arises. The role of carbide particles of inducing the fifth type of embrittlement was discussed in some detail.
Homogenization behavior in TLP-bonded joints of Ni-base single crystal superalloys, CMSX-2 and CMSX-4 were investigated using MBF-80 and F-24 insert metals. CMSX-2 and CMSX-4 were bonded at 1523-1548 K for 1.8-2.4 ks in vacuum. The (100) orientation of bonded specimen was aligned perpendicular to the joint interface. The homogenizing in the bonded interlayer of TLP-bonded joint was evaluated in terms of microstructures, line profile of elements analyzed by EPMA and microhardness distribution. Homogenization of the bonded region was attained quickly when the bonding temperature was increased. The microstructures, alloying elements and hardness distributions in the bonded interlayer following homogen-ization at 1523 K for 18 ks became uniform with those in the base metal. From the analyses of the bonded region following the solution treatment at 1589 K for 7.2 ks without any homogenization treatments, joint homogenization took place during the solution treatment. The homogenization treatment could be substituted by the sequential post-bonded heat treatments as solution treatment.
Mechanical properties of TLP-bonded joints of Ni-base single crystal superalloy, CMSX-2 were investigated using MBF -80 and F-24 insert metals. CMSX-2 was bonded at 1523-1548 K for 1.5-1.8 ks in vacuum. The (100) orientation of bonded specimen was aligned perpendicular to the joint interface. The tensile strength of the joints at elevated temperatures was equal to or greater than that of base metal in the range of testing temperature between 923 K and 1173 K. The creep rupture strength and rupture lives of the joints were the almost identical to ones of the base metal. The low cycle fatigue properties of the joints were also the same level as those of base metal. SEM observation of the fracture surfaces of joints after tensile test revealed that the fracture surfaces indicated the similar morphologies each other, and that the fracture of joints occurred in the base metal in any cases.
It is possible to reduce residual stresses of welding by the low temperature stress relieving. The technique is not applied in any field today because the optimum conditions are hard to define for every weldment. In this regard, the authors have already reported that the technique can be simulated by the finite element method based on thermo-elasto-plastic theory. In this paper, the calculated results of the technique for patch welding of disc are reported and the optimum conditions are shown in discussions. The conditions can be defined using the calculated results here, but it is shown that they are affected by boundary conditions. As a conclusion, the optimum conditions of the technique can be analized by the theory, but the examinations of the conditions are necessary for better treatments.
In failure analysis of a fractured component, estimation of residual stress distribution in the component before fracture is required. In this study, we propose a method to estimate the axisymmetric residual stresses in the original cylindrical component from its broken pieces. The method, which we refer to as broken piece method, employs an arc piece specimen and an axial piece specimen cut from the fractured component. The residual stresses are estimated from measured strains of the two pieces based on the equivalent inherent strain concept. The applicability of this method and appropriate dimensions of the piece specimens were theoretically discussed and experimentally verified by measuring residual stresses in a clad pipe. This method also applies to large cylindrical component case, where the residual stress measurement is quite difficult by using traditional measuring method with large cylindrical body specimen.
The dissimilar materials welded joints of high manganese non-magnetic steel/carbon steel (hereafter refered to as DMW joints), in which weld defects such as hot crack or blowhole are not found, were the good quality. Tensile strength of DMW joints was 10% higher than that of the base metal of carbon steel. In the bend tests, the DMW joints showed the good ductility without crack. Charpy absorved energy at 0(°C) of the DMW joints was over 120 (J) in the bond where it seems to be the lowest. Large hardening or softening was not detected in the heat affected zone. Fatigue strength of the DMW joints is almost the same with that of the welded joints of carbon steel/carbon steel. As the fatigue strength of the DMW joints exceeds the fatigue design standard curve of JSSC for carbon steel welded joints, the DMW joints can be treated the same as the welded joints of carbon steel/carbon steel of which strength is lower than that of high manganese non-magnetic steel, from the viewpoint of the fatigue design.
Tensile deformation characteristics of Sn-Pb eutectic system solder alloys have been investigated using a strain rate changing tensile test to obtain strain rate sensitivity index in order to correlate the behavior to thermal fatigue properties. Mechanical properties such as tensile stress and elongation were also obtained. These properties were compared Sn-Pb eutectic with Sn-Pb alloy containing small amount of Ag and Sb. The addition of small amount of Ag and Sb to Sn-Pb raised tensile strength and slightly decreased elongation. The alloys with elements showed finer microstructure and higher resistance to thermal fatigue than plain eutectic. The solution hardening and dispersion of Ag, Sn would be responsible to the enhanced tensile strength and resistance to thermal fatigue. Strain rate sensitivity index m was repeatedly odtained during tensile test. The plots between m and the strain where m was measured showed straight relation, therefore, the extrapolated value of m to strain zero (m0), and the gradient of the line (k) were used as a parameter for the estimation of thermal fatigue resistance. The m0 decreased with the coarsening of microstructure due to aging at elevated temperature, the value could be used as a measure of coarsening of microstructure. The m0 and k might be a good measure to estimate the thermal fatigue properties of solder alloys.
Effects of silver oxide additions on the bonding and superconducting properties of YBCO superconductors were studied. The YBCO superconductor ceramics were diffusion bonded with the parameters of 1223 K for 1 hour under the pressure of 2. 4 kPa in air. Thermal analysis and SEM/EDX analysis showed that during the bonding process, Ag2O additive was thermo-chemically decomposed and diffused into YBCO ceramics. The Ag did not react with the YBCO during the bonding processes and existed as the pure Ag in the YBCO matrix. With the increase of Ag2O content until 10 mass%, the strength of joints and critical current density Jc were increasing, but over addition of Ag2O made Jc decrease, although the critical transition temperature was nearly constant with the increase of Ag2O content. It was believed that during the bonding processes, thermochemically decomposed Ag diffused into the YBCO and filled into the voids in bonded surfaces, which could make a good contact between the bonded surfaces during the bonding processes, so the strength got increased. However, over addition of Ag2O would cause the decreasing in fraction of superconducting constitution, then the superconductivity became worse. It was feasible to bond the YBCO superconductors when the Ag2O was properly added. The diffusion bonding thermal cycle would not affect the crystal structure of the YBCO.
A new brazing method was developed and investigated in order to improve the bonding strength of insulated copper wires using fusion technology. The process was composed of fusing and press-soldering, which was a direct joining of copper terminal and insulated copper wire. Insulated copper wires wrapped with a tin plated copper terminal were heated simultaneously by resistance heating to be bonded. The tin plating layer was used for press-soldering. The room temperature tensile strength increased with resistance heating current until above 3.0 kA and the fracture in base metal occurred in the copper wires, for bonding above a current of 4.2 kA. In the elevated temperature tensile testings, rupture in base metal were observed at 150°C and 250°C. The results showed excellent thermal resistance of the joints.