A computational method of predicting creep lifetime is presented for welded joints. In the present method, time-incremental computations of stress by a finite element method and simultaneous cumulations of a variable as a measure of creep damage are reiterated up to the critical time for the cumulative value to exceed a specified limit. This critical time is regarded as the creep lifetime. A butt-welded joint of 50 mm thick Type 304 stainless steel was prepared by SAW with depositing a 308 wire. Specimens cut from the joint were subjected to creep tests at 823 K under applied stresses of 200 to 300 MPa. The obtained data was compared with the computations. The comparison showed that the creep lifetime of the joint predicted from the computations is within the factor of two (one half to two times) of that measured.
Recently, the serious problems facing heavy industries such as the increase in the average age of employees, shortage of skilled workers and environmental problems in the workplace are being addressed. The bridge panel fabrication system we have developed is such an approach to solve these problems. It consists of two parallel box girder panel sub-assembly lines, one exclusively for web panels and the other for flange. Each line is made up of stages for fitting, welding, re-forming, drilling and finishing. All of the line equipment is controlled by the line computer in the central control room. The box girder panel is reinforced by the stiffeners and the fillet welding is the objective of each welding stage. In total, 14 articulated-type arc welding robots are applied in the welding stages. In the web panel welding stage, 8 robots are hung from stationaly transverse girders with 2 m span in longitudinal direction and can be used for operations up to 4 m wide by 16 m long. In the flange panel welding stage, 2 welding robots are hung from each of the 3 transverse sliding units that are arranged under the mobile gantry. In total, 6 welding robots cover the 5 m wide by 16 m long work area. By adopting the High Speed Rotating Arc welding process, we have doubled welding efficiency as compared with conventional processes. Furthermore, this process is more effective to reduce panel deformation because of the lower welding heat input. The welding robot system is stabilized with precise accuracy by newly developed joint end and bead end sensing techniques as well as a seam tracking arc sensor system based on the High Speed Rotating Arc. All of the welding robot is linked with LAN to the newly developed, teachingless CAD/CAM system that is based on the computerized design fabrication system for bridge fabrication. This system is especially effective for bridge panel fabrication where almost all the panels are of different shapes. Thus, a step towards computer-integrated manufacturing (CIM) in steel structure fabrication has been achieved.
In the brazed joint of natural diamond to Fe-42 Ni alloy made with Ag-Cu eutectic filler including 4.5 wt% Ti, the behavior of titanium segregation changes according to the brazing thermal cycles, which causes the change in joint strength and the fracture mode. The chemical reaction products of diamond and the filler were identified by ESCA and EDS analyses on the fractured surfaces at the brazed interface, where the analyzed surfaces were made using a developed tensile shear test method, in order to propose the model of formation mechanism of the joint structure. The analytical results revealed that such a structure as consists of TiC nucleation surrounded by Ag and/or Cu is preferred as the joint structure. It was also indicated that the segregation of Ti at the joint interface reduces the bonding strength, where the fracture is liable to occur at the interfaces of diamond-Ti and/or TiC-Ti. The relation between the joint structure and the strength was also discussed using the disregistry values calculated in the previous report. The structure formation models of the interface between diamond and brazing filler metals were proposed to explain the change in the structure and strength of the joint made by different cooling rates.
Micrometer sized melting process is available to micro-welding for manufacturing microme-chanical systems. This report describes an in-situ observation of the micrometer sized melting phenomena by micro-discharge with a thin tungsten (W) electrode in a SEM vacuum environment. By the cathode electrode a melting spot can be formed on the surface of the specimen. Diameters of melting spots depend on the amount of discharge energy. Consequently, in the range of 1 μm to 10 μm, the diameter of a melting spot can be controlled by selecting the pulse condition. Additionally, a weld bead like melted-zone can be formed by applying repeated pulsed voltage between an electrode and a specimen. The width of a melted-zone depends on pulse voltage, pulse width, repetition frequency, and traveling velocity of an electrode.
The investigation results on the resistance spot weld quality of Al-Mg alloy sheet are summarized as follows; (1) Comparison of continuous spot weldability with mild steel sheet Compared with mild steel, the Al-Mg alloy exhibits degradation and variation of tensile shear strength associated with continuous resistance spot welding, and causes excessively serious wear in electrodes. (2) A study for variation of tensile shear strength in continuous resistance spot welding Variation of tensile shear strength in continuous resistace spot welding is due to nonuniform corona bond formation and nonuniform nugget size and profile. (3) Effects of oxide film and heat condition, (1) The oxide film causes localized wear in electrodes, and in addition, even prevents the formation of corona bond. (2) If the heat input after saturation of nugget growth is regaded as extra heat, it promotes wear in electrodes and causes corona bond formation. (3) Variation of tensile shear strength is improved mainly by removing oxide film. In addition by reducing the heat input, it causes degradation of tensile shear strength and makes the failure mode sparse, the formation of corona bond is suppressed and then tensile shear strength become stable and electrode life is extended.
We attemted to solder aluminum with the aid of ultrasonic vibration using solder of Zn, Ag, Cu, Mg, Si and Sn as a filler metal, and examined the tensile strength of the soldered joints. From the observation of the interface reaction, the mechanism by which the aluminum oxide film is broken and removed out of the interface was discussed. The mechanism of removing the oxide film was suggested as follows : First, the oxide film on the convex of faying surface is broken by the friction due to the ultrasonic vibration. Secondly, the interdiffusion through the fresh surface between the aluminum and the filler metal occurs, resulting in the formation of liquid by the eutectic reaction at the interface. The eutectic liquid, of which the amount is dependent on the erosion of the aluminum, aggresively promotes to remove the oxide film at the concave of faying surface. The metals other than zinc are also available as a filler metal, if they form low melting point liquid by the eutectic reaction with the aluminum, and several filler metals prove to provide the soldered joints with sufficient strength.
Fundamental characteristics of high power CO2 laser welding of 87%Al2O3 ceramics were investigated with an intention to develop a joining technique for thick ceramics. It was found that a penetration depth of 20 mm is possible at a welding speed of 6.5 mm/sec and a laser power of 10 kW. The porosity ratio was lower at lower input powers, and there was a tendency that the porosity ratio decreases with decreasing the welding speed at the same input power. The bending strength decreased as the porosity ratio increased. Porosities, in particular large porosities at bead center, were found to greatly influence the bending strength of welded joints of alumina ceramics. The root bending strength of butt weld of 4 mm thick plate had the same value as the base material.
This paper describes the consumption mechanism of Ru-Y2O3 electrode for air plasma cutting torch during arc discharge. To clarify the consumption mechanism, the direct observation by stereoscope and CCD camera was carried out during arcing, and the analyses by EMPA and micro X-ray diffractmeter of cathode surface after arcing were carried out. In case of Ar shielding gas, it was cleared during arcing that the fusion area of 80Ru-20Y2O3 cathode surface was sparated into the two regions of Ru and Y2O3 components and sputtering of the molten particles mostly occurred from Ru part. On the other hand, in case of Ar+20%O2 and air, the top of cathode surface was covered with the molten layer of Y2O3 and the plasma was stably sustained during arcing. The reason why the difference arises is considered due to possibility of oxygen supply during arcing.
This paper describes on the new developed Iridium(Ir) -Yttrium oxide (Y2O3) electrode for plasma cutting torch using compressed air in operating gas. From the two type consumption tests, the continuous operation and the periodic operation at 25 A in air, Ir-Y2O3 sytem showed extremely low consumption in the range from 10 to 25 wt% in Y2O3 content for 3.6 ks at 25 A in arcing. And Ir-Y2O3 system showed the same erosion characteristic of dependency on Y2O3 content in metal as Ru-Y2O3 system. Especially, Ir including 15 wt% Y2O3 electrode showed the same durability as 85Ru-15Y2O3 electrode, and showed a good erosion resistance compared with Hf electrode.
This paper describes the affection of O2 and N2 gases in operating gas on the electrode consumption for air plasma cutting, and discusses on the consumption mechanism of the developed new electrodes. From the consumption test under various shielding gases, it is concluded: Ru-Y2O3 and Ir-Y2O3 electrodes showed a good erosion resistance in oxidizing atmosphere (O2, Air), however these electrodes showed extremely increased weight loss in non-oxidizing atmosphere (At, N2). The characteristic of Ru-Y2O3 and Ir-Y2O3 electrodes is discussed from the qualitative model that the formation and evaporation of Y2O3 layer covered the cathode surface occurs.
Takahashi performed experiments on the reflection of the transverse waves by an inclined step on a free surface, and he also calculated the same cases using the sound wave theory proposed by Werneyer, and found large gap between the experimental results and calculated values. He also performed experiments on the reflection of the transverse waves by an inclined slit on a free surface having 1 mm height. In order to verify the elastic wave theory, numerical experiments of the same cases of reflections for the transverse waves have been executed. The incident angle of the transverse wave is given by 45 degree, and the inclination angles of steps or cracks are taken from 0 degree to 50 degree for every 10 degree. The results are found to agree with the experimental results within 2 or 3 dB error. The visualization of reflected waves has been realized using vectors diagrams. As the consequence, Physical mechanisms of new phenomena of reflections, such as head waves or corner waves, have been clarified in details.
The effects of M-A constituent on initiation and propagation of cracks has been investigated by the observation of fracture morphology and crack around M-A constituent. Energies for the crack initiation and propagation have been estimated by using an instrumented-Charpy impact testing machine. Furthermore, concentration and distribution of tensile plastic-distortion introduced in a soft phase around hard phase have been qualitatively analyzes with FEM in order to explain the effect of the M-A constituent on the crack initiation and propagation. As the result, remarkable decrease in the crack initiation energy and propagation energy were observed, as the area fraction of the massive M A constituent was increased. The FEM analysis have suggested that the greater hard phase decreases the energy for crack initiation in the surrounding soft phase and this can explained the observed decrease in the crack initiation energy by the increase in the area fraction of the massive M-A constituent. It has also been suggested that the decrease in the crack propagation energy is caused by the increase in the interfacial area between the massive M-A constituent and matrix, since the crack was observed to propagate preferentially along the interface between the M-A constituent and matrix.
The ductility characteristic of solidifying weld metal between liquidus and solidus temperatures during welding has been measured by externally-restraint solidification cracking tests, namely Trans-Varestraint test and Slow Bending type Trans-Varestraint test. Minimum augmented strain to cause cracking (εmin), Brittleness temperature range (BTR) and Critical strain rate for temperature drop (CST) were selected from ductility characteristics as criteria to evaluate weld solidification crack susceptibility. In addition, solidification crack susceptibility of 16 kinds of commercial aluminum alloys from 1000 to 7000 series has been evaluated qualitatively by self-restraint type solidification cracking tests, namely Ring-casting test, GTAW crater cracking test, Houldcroft type test and Fan-shaped test. Moreover, as a metallurgical factor, gain size of weld metal and dihedral angle of eutectic products in grain boundary and the amount of eutctic products were also measured. Correlation between the test results obtained by self-restraint tests, criteria showing ductility characteristics and metallurgical factors were discussed. CST is the most suitable criterion to evaluate weld solidification crack susceptibility of Al alloys, because crack susceptibility was monotonically decreased as the increase of CST. εmin and BTR are also important criteria to show the threshold at which crack susceptibility begin to increase steeply. At more than about 0.22% in εmin or less than 43°C in BTR, Al alloy showed very low susceptibility to cracking. Among metallurgical factors, dihedral angle and grain size showed close relation to crack susceptibility, but the amount of eutectic products didn't.
Relation between microstructure and stress corrosion cracking in high temperature water for duplex stainless steel was investigated using transmission electron microscopy and fractography. The samples were solution-treated at various temperatures and then sensitized at 923 K for 72 ks. The slow strain rate testing (SSRT) was carried out at 562 K in water with 8 ppm dissolved oxygen under 8 MPa pressure at a strain rate of 4.17×10-6 s-1. The volume fraction of α phase and Cr2N in the α phase increasedd with increasing the solution treatment temperature. For the sensitized sample, M23C6 precipitated at α/γ interface, the bowing area of secondary austenite, namely γ* phase occurred, and σ phase precipitated at the γ*/α interface. They increased with decreasing solution treatment temperature and with decreasing the volume fraction of α phase. The stress corrosion cracking in high temperature water hardly occurred for the solution-treated samples. The reduction of area of the sensitized sample was much lower than that of the solutiontreated samples and decreased with decreasing solution treatment temperature, because the γ* phases near the M23C6 and the σ phase were predominantly corroded and resolved during the test.
An insert metal for liquid diffusion bonding of HP40 was newly developed by using a mathematical programming method. The main procedure of this method is to obtain the optimum composition to maximize some index of an performance of the insert metal by a grid search. The procedure also involves data estimation by means of the interpolation method which was originally proposed by Schagen. The contour maps of objective function Z which were introduced as the index of a performance of the insert metal were obtained through the interpolation method. The curved surface being once obtained, its general view would give suggestions for further searches and the optimum composition could be efficiently searched. As the results of the search, a composition of Fe-24Cr-33Ni-2.7B-1.5C-4Si which minimized the objective function was determined. Scanning electron microscopic analysis revealed that the joint bonded at 1473 K with the newly developed insert metal (Fe-24Cr-33Ni-2.7B-1.5C-4Si) had a sound microstructure without forming brittle phases in the bonding interlayer. This joint also exhibited good tensile properties, that is, its strength and ductility were comparable to those of the base alloy.
Diffusion bonding of aluminum bronze to austenitic stainless steel was performed with a phosphor bronze foil as an insert metal. The joints were fabricated at temperatures lower than the solidus temperature and also they were fabricated at temperatures between the solidus and liquidus temperatures of the phosphor bronze (solid-liquid bonding). Influences of bonding time, bonding temperature, bonding pressure and roughness of bonding surface on the joint tensile strength and the fatigue strength have been investigated. Main results obtained are as follows; (1) In the solid-liquid bonding, Sn-riched liquid arose in the phosphor bronze and then it was discharged to the outside of the joint with increasing the bonding time at higher bonding pressures. At lower bonding pressures, however, the liquid stayed in the phosphor bronze and the fused structure in which voids were forms in the cooling process of the bonding operation was observed in the bond zone. (2) When the volume fraction of the fused structure was around 1%, the joint tensile strength took the maximum value. The strength was decreased with the increase in the amount of the fused structure. (3) The maximum joint strengths for the constant bonding load of 1 MPa and restrained bonding were 415 MPa and 465 MPa, respectively. (4) The joint strength increased with the decrease of roughness of the bonding surface, and the maximum strength obtained was 515 MPa at Rz 0.04μm. (5) The endurance ratio of the joints bonded at temperatures lower than the solidus temperature of the phosphor bronze was 36 to 46%, but it was higher than 50% for the solid-liquid bonding.
Surface coatings have come into a wider use to increase an ability of machine elements such as a paper machine roll. These surface-coated rolls have been designed empirically, because no evaluation method of strength for coating films has been developed. In this study, fatigue tests of chromium plated rolls were carried out under a contact loading condition and a fracture mode was investigated in detail to establish the evaluation method for coating film strength. As a result, we found out that a coating film began to break in the axial direction, then number of cracks initiated, and after that, the surface film became to be in flakes. Also, we confirmed that a number of stress cycles to flake off depended on a thickness of the coating film, but a number of stress cycles to initiate a crack did not depend on it. Then, to evaluate the dependence of the crack initiation on the coating film thickness, a stress analysis by FEM was performed. As a result, a tangential stress that must cause the cracking did not change remarkably when the coating film thickness was changed. From the above mentioned facts, we considered that the cracking was caused by the fatigue of the compressive tangential stress.
In the previous report, to establish an evaluation method of a rolling contact fatigue strength for a coating film, an endurance limit of coated roll for contact loading was investigated and a crack initiation was found to determine the endurance strength dominantly. And we proposed that a mechanism of cracking was a fatigue process caused by a compressive tangential stress under contact loading. So, in this study, static and fatigue properties of Cr-plating film under a bending, tension and compression were obtained. And these properties were compared with rolling contact fatigue strength of Cr-plating film on a fatigue endurance diagram. As a result, we found that the rolling contact fatigue strength agrees well with the value estimated from the static and fatigue properties. And the rolling contact fatigue strength of Cr-plating film would be estimated from the static and fatigue properties. Also, it was found that a residual stress of Cr-plating had a large effect on the rolling contact fatigue strength.
To investigate the effect of surface cleanness upon ultrasonic bondability, ultrasonic micro-bonding apparatus was developed, which consisted of three vacuum chambers for bonding, surface analysis and gas analysis. The bonding chamber has an ultrasonic ball bonding system with ultrasonic power source oscillating at the range from 10 kHz to 300 kHz, a surface cleaning system of ion bombardment and laser irradiation, a heating system, and a pure-gas-introduction system. The surface analysis system consisted of an Auger electron spectroscopy analyzer, an ion gun with 500 um in ion beam diameter, and a mass spectroscopy analyzer. Gas analysis was done by a quadrupole mass spectrometer. For this apparatus, mechanical vibration characteristics of the bonding system was measured. Then, using this apparatus, ultrasonic bondability of Au wire to Au thin film deposited by physical vapor deposition was investigated. Bonding procedure was that at first Au thin film was cleaned in vacuum, next pure N2 gas was introducted about 1×105 Pa and Au ball was formed at the tip of a bonding wire by arc discharge, and then Au ball was ultrasonic-bonded to the cleaned surface with a ceramic capillary tool in pure N2 gas atmosphere. An initial surface, a cleaned surface, and a exposed surface in pure N2 gas were analyzed by AES. Main results were as follows; (1) Au surface, which was coated with an organic resist film for preventing plating and cleaned by acetone and water, is covered with two layers of a surface carbon-rich layer and an under layer including more water. (2) An organic material layer of two or three nanometers in thickness including water lubricates between Au film and Au ball in ultrasonic bonding, and so it deteriolates ultrasonic bondability. (3) Ultrasonic ball-bonding of Au wire 30 μm in diameter to Au surface cleaned by ion bombardment makes failure strength of bond more than wire strength under conditions of 0.38 um : vibration amplitude, 6 mW:ultrasonic power, 10 ms:bonding time, 0.5N: bonding load and 20°C: bonding temperature in pure N2 atmosphere.
Power-control semiconductors are widely used in industrial applications. The development of power semiconductor such as insulated-gate bipolar transister enables much faster switching, higher current and higher voltage rating. While high-performance of the semiconductor is accompanied by an evolution of heat. A substrate system, therefore, is very important part to dissipate heat from the semiconductor. Design and assembling process substrates system for the semiconductor have been researched and developed. In this report, especially, a conceptual design and performance of the substrate is discussed through a calculation of thermal stress by finite element method and an analysis of its function. It was clarified that the design accepted region of initial performance of the substrate was bounded by a dielectric breakdown voltage, a current capacity and a thermal conductivity. The substrate is consisted of conductors and an insulator. The dielectric breakdown voltage is determined by thickness of ceramics insulator. The current capacity is determined by thickness of copper conductor. The thermal conductivity is determined by materials and thickness between conductor and insulator. Furthermore, it was found that maximum principal stress is generated at a corner of ceramics and copper conductor at the lowest temperature, and the stress occurence crack at the ceramics. A reliability of the substrate was dominated by thermal stress generated in ceramics under temperature cycles.