Two multi-kilowatts Nd: YAG laser beams are condensed separately and combined at surface of workpiece in order to increase penetration depth. In this system, one or two laser beams are inevitably inclined depending on size of condensing optics, thus, welding characteristics might be fairly complicated than that of a single laser beam welding. Welding characteristics are evaluated by penetration depth and penetration shape in bead-on-plate tests of stainless steel (304). This paper is intended as an investigation on the welding condition of the deepest weld penetration, especially on arrangement of condensing optics. The deepest weld penetration of 6.3 and 7.7 mm are obtained by two CW laser beams and two rectangularly modulated CW laser beams with total average power of four kilowatts at a welding speed of 5 mm/s, respectively. Deepened penetration depths of weld beads by laser beams in combination due to the increase in total laser power are obtained in comparison to penetration depths of 2.8 and 4.3 mm by a single CW laser beam and a single rectangular modulated CW laser beam with average powers of two kilowatts, respectively. Penetration depths of weld beads obtained with symmetric irradiation exceed those with asymmetric irradiation. The angle between two laser beams for the deepest penetration is 60 degrees for CW laser beams, while the angle is 30 degrees for modulated CW laser beams. The difference in the appropriate angles for the deepest penetration might be explained by different effect of keyhole bottom separation on penetration depths depending on waveform of laser beams. With asymmetric beam arrangement, a deeper penetration is obtained in backhand welding direction for the inclined laser beam than that in forehand direction both in welding using two CW laser beams in combination and two modulated CW beams in combination.
To investigate the brazing characteristics of three CsF-AlF3 non-corrosive fluxes for aluminum, fillet formation test has been performed using magnesium containing base metal and zinc containing filler metals. The brazeability was evaluated by the length of formed fillet on tee-type joint and penetration depth of flux into magnesium containing base metals. The increase in magnesium content of base metal deteriorated the brazeability in all fluxes. The CsF-AlF3 fluxes showed inferior brazeability to KAlF4-K2AlF6 flux, however, one of the CsF-AlF3 flux with mole fraction of CsF/AlF3=1.2 (CA flux) gave similar brazeability and narrower flux penetration depth for magnesium containing base metal. Brazeability and flux penetration depth were closely related to fluorine content per unit mole of flux, the increase of fluorine content enhanced fillet formation, however, it enhanced penetration simultaneously. The CA flux offered excellent brazeability for zinc containing filler metal, whereas KAlF4-K2AlF6 flux showed poor brazeability in filler metal more than 5%Zn. The use of CA flux is recommended for brazing of magnesium containing base metals and zinc containing filler metals.
Recently, short-pulse lasers have been applied to microprocessing in the field of various industries. It becomes more difficult to experimentally observe the microprocessing phenomena as pulse width becomes short. Numerical analysis with a continuum model has a limit in elucidation of such phenomena, therefore, computer simulations at the atomic or molecular level must be important. The authors have carried out the molecular dynamics simulation for the first time in the field of laser materials processing and have been making fusion and evaporation phenomena clear at the atomic level. In this paper, a modified molecular dynamics method for metal was developed, in which calculation of the molecular dynamics is carried out compensating the heat conduction by free electrons at each time step. Its validity was confirmed by calculating thermal conductivity and thermal diffusivity. Pulse width dependence of damage threshold was also investigated.
The effect of chemical compositions on anode area formation in tungsten inert gas arc welding was investigated to clarify the influence of alloying elements on the weldability of high alloyed steels in fabrication. Welding tests were carried out on 25%Cr- 22%Ni stainless steel plates with various amounts of surface active elements, manganese, silicon and aluminum. In these welding tests, the molten pool surface was observed using a high-speed video camera and spectrum analysis of arc plasma was done by using a spectroscope. Chemical composition of welding fumes was also analyzed. The results showed that the size of the anode area and the direction of the molten metal flow were influenced only by surface active elements. The slag containing manganese was observed in the quenched molten pool surface and the size of the distribution area of the slag corresponded closely to the size of the anode area. Based on the results of spectrum analysis of the arc plasma and chemical analysis of the welding fumes, the mechanism of anode area formation is considered to be as follows. Manganese is vaporized from slag carried by the molten metal flow and ionized in the arc column. Consequently, the anode area is formed on the slag gathered area that is a conduction path of arc current. Therefore, the size of anode area appears to be determined by surface active elements because they influence the direction of the molten metal flow.
The effect of chemical compositions on the penetration ratio and the uniformity of bead width in tungsten inert gas arc welding was investigated from the point of view of the effect of alloying elements on the behavior of the molten pool and the anode area. Welding tests were carried out on 25%Cr -22%Ni stainless steel plates with various amounts of sulfur, manganese and silicon. The penetration ratio and the uniformity of bead width were evaluated by measuring bead depth and width. Additionally, in order to evaluate the stability of the molten pool, it was observed using a high-speed video camera during welding. The penetration ratio was increased by the addition of sulfur and manganese. Silicon had no influence on penetration ratio. The uniformity of bead width was deteriorated by increasing the sulfur content but not influenced by increasing the others. This deterioration was caused by the instability of the molten pool, which increased in proportion to the shrinkage of the anode area caused by the increase of sulfur content. The increase of penetration is thought to be caused by : (1) the concentration of arc energy resulting from the anode area shrinkage, which is induced by the molten metal flow to the molten pool center driven by sulfur ; and (2) the increase of energy density resulting from vaporized manganese.
In the arc welding, spatters are frequently bonded to the surface of base metal. Therefore, to keep the quality of the weld products, the weld metal surfaces were usually brushed with bevel type or cup type wire brushes. Since the spatters cannot be removed completely under a certain brushing condition, it is very important how to finish the surface of base metal. This study tries to show the brushing conditions under which spatters can be removed completely. As the first step to find the proper condition, we made a brushing apparatus that can control the pushing force of the wire brush, and measured the force (we call it the brushing force) in the steel plate. The brushing force is determined by the two forces. One (Fy) is from the direction of brushing and the other (Fx) is perpendicular to the brush. In brushing the metal, the pushing force was set at 9.8 N, 19.6 N, or 29.4 N and the angle of the brush and workpiece was 15, 20, 25 or 30 degrees. The metals we used were two types of SS400. One had the scale and the other had no scale, which had been removed by the surface grinder. When the steel plate is brushed, various kinds of brushed traces appeared on its surface. The brushed traces were examined through a metallographical microscope and we showed the relations between the brushed traces and the brushing conditions. We also examined the brushing force when a tiny part of the surface (8.04 mm2 in size) was brushed (we call it the local brushing force), so that we could show the relations between the brushed traces and local brushing forces.
The setting operation of suitable welding condition is very important in welding procedure. In this study, paying attention to the volume of deposited metal for unit welding length (Vw), the setting method of the welding condition in GMA welding process with flux cored wire is proposed and its propriety is investigated. Influence of Vw and deviation of arc center on bead formation, too, are investigated. The unifying method of welding condition using Vw can select the suitable conditions easily and widely. In above mentioned welding process, it is advisable to select the welding condition with high current and high speed. It is effective to consider Vw in regard to the maximum welding speed without undercut. In the case of horizontal fillet weld as small Vw as 6 mm leg length, it has a broad allowable range to aim the arc center at 0.5 mm in front of root. In large Vw horizontal fillet weld, the process that arc center aims at 3-4 mm in front of root is effective to expand to the permitted limit of deviation of arc center.
In order to produce a vaporizing combustion chamber instead of brazing, laser welding was applied to the joint between SUS304 and brass, since it is now fabricated by brazing after many complicated processes. Materials used in the laser welding were 0.6 mm thick Cu-30%Zn alloy (brass) and 0.5 mm thick austenitic stainless steel SUS304. Stainless steel was lap-welded to Cu-Zn alloy. As a result of leak test under 0.4 MPa pressure, many leaks were recognized in the weld bead. The observation by means of EPMA and microscopy indicated that the leaks were caused by cracks which resulted from liquid-metal embrittlement of the brass or solidification cracking at grain boundaries of the stainless steel. The number of the cracks increased with increasing in the volume of melted Cu-Zn alloy, and when the volume was lower than 3 vol.% of the melted SUS304, few cracks occurred. Therefore, to eliminate the cracks by re-melting of only the surface of the stainless steel was conducted with the laser beam after the welding. By the elimination process, the cracks were eliminated completely ; consequently, no leakage was recognized except the welding conditions under which many cracks occurred in the weld beads.
A new method for in-process monitoring of welding based on non-contacting ultrasonic technique is proposed. When a laser pulse is irradiated onto a surface of base metal, an ultrasonic pulse is generated due to ablation of material from the surface. The ultrasonic waves, which travel through both liquid and solid metal, are detected with a laser interferometer as micro displacement on the surface and then propagation time of the ultrasonic pulse is measured. Since sound velocity depends on the temperature distribution along the propagation path, it is capable to estimate the location of fusion point on the sound path from the propagation time with a temperature distribution model. Moreover, a configuration of the molten pool is reconstructed by scanning the laser beams on the surface which is being observed. Feasibility tests are performed with a pulsed Nd : YAG laser source (wavelength : 1064 nm, pulse energy : <450 mJ/pulse) for generating ultrasonic waves and Michelson interferometer for detecting. Molten pools are formed by static arc current using GTAW (welding current : 70-150 A) on type 304 stainless steel having 32 mm thickness. The estimated configurations molten pool are compared with lateral cross section views. Consequently, by applying the present method, it is found that ultrasonic signals are detected with good signal-to-noise ratio despite under welding operations and the molten pool configurations are estimated as a similar geometry within 1 mm or less accuracy. Accuracy of the estimated depth tends to be inferior as an increase of welding current ; therefore advanced temperature distribution model and simultaneous detection system of the propagation time over the observing area are required in order to determine the more accurate configurations.
Automatic control system with a laser slit sensor was developed and applied to full automatic multi-layer GTA welding for pipe joints. The laser slit sensor was able to measure dimensions of groove of the welding section. It was also able to detect the off-positioning and the groove area which required for multi-layer welding control, and the defects on the welding surface that might be in each welding pass. The dimensions of groove with hindering blocks were able to be measured. In addition, a tracking control method that was automatically corrected by following the value off-positioning of the welding torch was established. It was shown that torch position control was able to be appropriately performed at intermittently tack welding for avoid blocks, back welding of first layer and to finish welding of multi-layer. A controlling method for amount of weld deposit was useful. Using the sensor and system, It was clarified that the sound welding joint was obtained fully automatically.
As reheating temperature of coarse grain HAZ is around 1623 K, TiN and austenite compositions are reported to reach equilibrium. This means that size and a volume fraction of the TiN precipitate could be predicted by thermodynamic data. As austenite grain size has close relation with average size and a volume fraction of TiN, prediction of the austenite grain size is also possible. In order to establish a prediction way of sizes of TiN and austenite grain of coarse grain HAZ, the relationship between size of the TiN and composition of the TiN that was calculated by thermodynamic data was investigated. The relation between the TiN precipitate morphology and size of the austenite was also examined. The effect of microstructure and the TiN size on coarse grain HAZ toughness of Ti containing steels was also investigated. The size of the TiN showed close relationship with chemistry of the TiN and the austenite that equilibrate each other. The austenite grain size showed to follow the Zener's relation with the TiN size and a volume fraction. These results show that the prediction of the austenite grain size of coarse grain HAZ is possible by calculating equilibrium chemistry of TiN and austenite. HAZ toughness of Ti containing steels reveals close relation with the grain size and the TiN size.
Microstructural features of LASER welded duplex stainless steels have been studied from viewpoints of nitrogen distribution and its effects on pitting corrosion resistance. The experiment was conducted by means of 25 kW CO2 LASER welding process without filler metals. Nitrogen supersaturation in ferrite phase due to rapid solidification of LASER weld metal was observed in the as-welded conditions. Pitting resistance was extremely improved in case of high nitrogen steels welded by LASER. On the other hand, local nitrogen depletion near the interface in the ferrite phase, where pitting occurred, was also observed. It was due to the migration of nitrogen from ferrite to austenite. Ferrite-austenite ratio and distributions of elements were restored as same as base metal after short time solution heat treatment. Pitting resistance was restored when the cooling rate was high by preventing nitrogen evolution. These nitrogen supersaturation, local depletion and evolution phenomena could be explained by the calculation based on nitrogen diffusion.
Characteristics of the heat-affected zone cracking of cast alloy 718 has been investigated using the bead on plate test and microscopic observation for isothermally processed specimens. The results obtained in this study are summarized as follows : (1) It was clarified that cracking in HAZ of cast alloy 718 was intergranular cracking which was caused by liquation of grain boundary phases. (2) According to the microscopic observation of specimens heated at various temperatures, incipient melting of the grain boundaries was found to occur at the temperature of 1383 K and above. (3) The Laves cluster was observed to liquate at the ternary eutectic temperature of 1383 K, which was lower than the NbC liquation temperature of 1413 K. (4) The liquation cracking in HAZ of cast alloy 718 was considered to be caused by constitutional liquation of Laves cluster at the grain boundaries.
Low-ductility creep-fracture (LDCF), which arised occasionally in heat affected zone (HAZ) of heat-resisting Cr-Mo steels, was reproduced in laboratory. Synthetic HAZ specimens (HAZ specimens) were prepared by giving weld-thermal-cycle to round bars of 1 1/4Cr-1/2Mo and 2 1/4Cr-lMo steels, and creep-rupture tests were made on those specimens at the testing temperatures of 775 and 825 K. HAZ specimens showed very small ductility (reduction in area smaller than 10%) at each time to fracture. The specimens produced intergranular fracture along the grain boundaries of prior-austenite. The heat treatments of stress-relieving (SR) was given to the HAZ specimens ; heating temperatures of 975 K and 925 K, respectively, were selected for 2 1/4Cr-lMo and 1 1/4Cr-1/2Mo steels on the basis of practical SR conditions of those steels. The ductility of HAZ specimens of 2 1/4Cr-lMo steel was improved by SR for 20 h at 975 K. That of 1 1/4Cr-1/2Mo steel was improved as well by SR for 200 h at 925 K, but this treatment reduced significantly its fracture stress. Therefore, it is practically difficult for 1 1/4Cr-1/2Mo steel to remove LDCF by SR treatment. The influence of SR treatment on LDCF was discussed from the metallurgical view point in some details.
We evaluated not only total hydrogen content in duplex stainless steel and its weld metal but also hydrogen content in austenite by using internal friction measurement. The internal friction measurement was conducted over the temperature range from liquid nitrogen temperature to room temperature at a frequency of about 1.5 Hz. We investigated relationship between peak height and volume fraction of austenite and, also cleared influence of microstructure on internal friction peak. (1) SUS304 austenitic stainless steel was cathodically charged with hydrogen. After charging, internal friction peak due to hydrogen was found at 260 K. This hydrogen induced peak height increased with increase of hydrogen content. Consequently it is apparent that this peak is attributed to hydrogen in austenite. (2) The internal friction peak was observed at 260 K in SUS329J1 duplex stainless steel subjected to hydrogen charging. However this peak was not found for SUS444 ferritic stainless steel. Therefore it is clear that this peak is associated with hydrogen in austenite in SUS329J1 duplex stainless steel. (3) The volume fraction of austenite for SUS329J1 duplex stainless steel and SMAW 329J4L and SMAW 329J3L weld metal was approximately 40%. The base metal has the microstructure with austenite elongated toward rolling direction in rod shape in ferrite matrix, and the weld metal has the microstructure with widmanstatten austenite in ferrite matrix. Hydrogen content and peak height value was plotted on same straight line, therefore their relationship is independent of microstructure morphology. The relationship between peak height and hydrogen content in austenite in SUS329J1 base metal is applicable to that of weld metal. Internal friction peak height due to hydrogen is proportional to volume fraction of austenite in case hydrogen content in austenite is same.
Microstructural analyses and tensile testing of directly diffusion-bonded joints of A6061 to SUS316 were conducted in order to evaluate the joinability of aluminum alloy to stainless steel. Diffusion bonding of A6061/SUS316 was carried out at 723-848 K for 0.3-9.0 ks applying 9.8 MPa in vacuum. SEM observation revealed that the reaction layer was formed at the A6061/ SUS316 interface. The intermetallic compounds such as Fe2Al5 and FeAl3 were identified in the reaction layer. It was elucidated that the reaction layer growth could be expressed by parabolic growth law, and that the apparent activation energy for the reaction layer growth was 187 kJ/mol. The tensile strength of A6061/SUS316 joint was risen up to about 200 MPa by maintaining the reaction layer thickness at around 1-2 μm. The fracture modes shifted from the interfacial fracture to the brittle fracture in reaction layer with increase in the reaction layer thickness.