An asymmetric Abel inversion processing system was developed to establish a method for diagnosing arc plasma that has an elliptical cross section and is a non-axisymmetric in the direction perpendicular to the observation direction during an arc welding process. This method was applied to an arc plasma in a tilted tungsten inert gas (TIG) welding process. As a result, when this method was applied to the non-axisymmetric intensity distribution due to integration, it was confirmed that the relative error near the central axis and the edge was large regardless of the difference in the cross-sectional shape. In addition, after applying this method to the tilted TIG arc plasma, it was shown that an error of about 350 K occurred in the temperature distribution when emission intensity distribution was converted to the temperature distribution by the Fowler-Milne method. Furthermore, the temperature distribution of the arc plasma obtained in this study was slightly wider than the previous study, while the high-temperature area where the temperature was higher than 14000 K showed a good agreement with the previous study. Therefore, it was concluded that, by applying this method to the tilted TIG arc plasma, the equivalent result to the three-dimensional emission spectroscopy could be obtained by photographing a tilted arc plasma from two directions.

抄録全体を表示

This study tested a two-color temperature measurement method using a multi-spectral camera designed for biomedical use. We selected the images of 735 nm and 930 nm as wavelengths. Two calibrations were performed before using them for temperature conversion. One was sensitivity calibration between wavelengths, which was conducted by heating the steel surface coated with blackbody spray from the opposite surface and comparing it with the temperature measured by thermography. Another calibration was a pixel shift that took into account the array of elements. This is a correction peculiar to a multi-spectral camera, and it was possible to eliminate a steep temperature error at the end of the temperature discontinuity. In this report, the electrode temperature of the gas tungsten arc was reported as an application example.

抄録全体を表示

The light emitting elements around a tungsten electrode during tungsten inert gas welding were identified using an optical emission spectroscopy. A white and blue umbrella-shaped light emitting bilayer region was observed above the tip of the tungsten electrode containing the lanthanum oxide during the welding on a water-cooled copper plate. On the other hand, the umbrella-shaped light emitting region was not clearly observed when the pure tungsten electrode was used. The optical emission spectroscopy revealed that the umbrella-shaped light emitting region was the emission from lanthanum vapor. The white region in the upper layer mainly consisted of the emission from lanthanum atoms, and the blue region in the lower layer mainly consisted of the emission from lanthanum ions. The lanthanum atoms emitted strong light at 5800 K, and the lanthanum ions emitted strong light at 10800 K. Therefore, it was considered that the bilayer regions appeared like the umbrella depending on the temperature field and the shape of the arc plasma around the electrodes. The temperature range in which the tungsten vapor emitted strong light was almost the same as that of the lanthanum vapor. Nevertheless, the spectrum of tungsten vapor was not measured. This was because the electrode surface during this welding was covered with the molten lanthanum, and the amount of tungsten vapor was smaller than that of lanthanum vapor. The temperature range in which the cerium and thorium vapors emitted strong light was almost the same as that of the lanthanum vapor. It was suggested that the umbrella-shaped light emitting region appeared around the electrode during welding when the tungsten electrode containing the cerium oxide or the thorium oxide was used as well.

抄録全体を表示

The droplet ejection from an electrode during an alternative current tungsten inert gas (AC TIG) welding process was observed using a visualization system composed by a high-speed camera with a band-pass filter to clarify dominant factors of the droplet ejection. Different welding currents, electrode diameters and electrode positive (EP) ratios were set. The numbers of droplets ejected from the electrode tip were measured for the different conditions. The timings of droplet ejections from the electrode were also determined in one AC cycle. The results indicated that droplets were likely to be ejected when the welding current was high, when the electrode diameter was small, when the EP ratio was large, and in the latter half of the EP term. Because the electrode temperature under these welding conditions was higher, the high electrode temperature was considered to be a dominant factor for the droplet ejection. However, immediately after the start of the electrode negative (EN) term, the number of droplets decreased even though the electrode temperature was the highest in one AC cycle. Therefore, it was suggested that other factors affected the electrode ejection. Moreover, raised portions were formed on the surface of the molten electrode right before droplet ejections. It was considered that the formation of the raised part might be suppressed by the collision of positive ions in the arc plasma with the electrode at EN term. Estimated pressure due to the ion collision acting on the ridge was larger than that due to the surface tension and the electromagnetic force, respectively. Therefore, it was suggested that the collision of positive ions at EN term suppressed the droplet ejection.

抄録全体を表示

In this study, to identify dominant factors of droplet temperature in a gas metal arc welding, the effects polarity on droplet temperature were investigated. In particular, the droplet temperature, the wire heat input and the wire melting rate with electrode positive (EP) polarity and electrode negative (EN) polarity using 100% Ar gas and 100% CO2 gas were measured. As a result, the droplet temperature with EP polarity was higher than EN polarity’s one using 100% Ar gas within the range of 130A to 230A. This result showed the reverse tendency compared with the case of 100% CO2. Moreover, the wire heat input and the wire melting rate with EN polarity were larger than EP polarity’s ones regardless of the shielding gas type. The simplified calculation suggested that this was because the ion current ratio was about 80% or more. Besides, especially with EP polarity, the wire melting rate using 100% CO2 was larger than 100% Ar’s one. It was suggested that this was because the wire melting rate was made a difference by the wire preheating effect depend on the droplet frequency determined by the specific heat of the shielding gas. These results identified the dominant factors of droplet temperature as the welding current, the ion current ratio and the specific heat of the shielding gas.

抄録全体を表示

Numerical simulation based on an incompressible smoothed particle hydrodynamics method was performed to clarify dominant factors of slag transfer and deposition processes in a metal active gas welding with a computational model which considered the effects of a shielding gas flow on a slag floating on a weld pool surface. As a result, the slags generated on the weld pool surface near the center of a heat source were transferred to the edge of the pool, which was similar to an experiment result. The slags stayed at the trailing region of the pool and then deposited on a weld bead. These simulated behaviors supported the validity of the present computational result. In order to identify the dominant factors of the slag behavior on the weld pool, numerical experiments were conducted with individually acting the forces due to the Marangoni effect, the shearing force, the Lorentz force, and the drag by the shielding gas flow. From the computational results, it was suggested that the shearing force and the drag due to the shielding gas were dominant near the center of the heat source, and the slag was transferred from the vicinity of the heat source to the end of the weld pool. On the other hand, it was clarified that the force due to the Marangoni effect was dominant in the trailing region of the weld pool, and consequently the slag was transferred from there to the front region in the welding direction.

抄録全体を表示

Arc welding process is indispensable technology in various fields of industry. In order to develop a high quality and high productivity welding process, it is important to control the welding process. In flux cored arc welding (FCAW), a flux cored wire (FCW) is used. The FCW is composed of a metal sheath and a central flux core. There are various types of the flux such as a deoxidizing agent, a stabilizer of the arc plasma, a slag-forming agent and so on. When the TiO2-based FCW is used, the slag covers the weld metal surface. The appearance of weld metal after the welding process is beautiful and it can be applied to all-position welding, but the phenomena during the welding process is not clear completely. In this study, to make clear the welding phenomena, a numerical simulation model of weld pool phenomena in welding process using TiO2-based FCW is constructed. The model calculates fluid flow in the weld pool and the distributions of the metal and the slag are calculated by VOF method. The model is applied to the vertical upward welding. In case of the solid wire, the molten metal flows downward strongly and forms humping bead. On the other hand, in case of the TiO2-based FCW is used, the slag covers the molten metal surface and the strong downward flow is prevented, and sound bead is formed. The molten metal flow at the surface is very important to determine the bead formation phenomena. In addition, the interfacial tension and the melting temperature of the slag are important to determine the distribution of the slag at the molten pool surface and bead formation phenomena.

抄録全体を表示

Gas metal arc welding applying inert gas typically required a little oxidizing gas for arc discharge stabilization; thus, oxide precipitates potentially deteriorating fracture toughness of the weld metals are generated. Although several methods in order to reduce oxide gas in the MIG welding have been proposed for example applying a special nozzle which can add inert and active gas separately, it has not been applied for actual production yet, since weld flaws may occur due to arc destabilization especially in groove welding. Therefore, the aim of this research is to achieve the low oxidation defect free weld in narrow groove welding by applying a special nozzle we developed. The effect of the nozzle, influence of arc phenomenon such as pulsed arc and continuous current arc, and frequency of adding active gas were investigated by some experiences and simulations. As the results the nozzle is contributed arc stability and pulsed arc affects arc length and time for arcing and the active gas adding frequency of 30Hz provide maintaining higher CO2 molar fraction in arc. Extremely low oxygen in weld metals were achieved by those combined effects.

抄録全体を表示

While automatic welding process has been introduced at the manufacturing site today to improve the welding efficiency and weldment quality, there are still some joint which is difficult to be automatically welded. Horizontal penetration bead welding in Shipyard, for instance, where weld pool shape varies easily and tracing technique for its variation is required, is manually welded by skilled welder. In order to automate such skillful welding, our research team works on development of weld pool recognition technique with visual sensor and control robot system. In this research, feature points of weld pool are recognized by using CNNs based learning model in real time during CO2 welding on V-groove joint with gap variation. The chemical composition of the flux cored wire is specially designed for bridge performance and back bead quality. It is adopted the straight stepped weaving to adapt a weld pool shape with gap variation. In order to reduce work processes of ceramic backing attachment, with and without ceramic backing welding has been studied in this research. From the images by a CMOS camera, it is confirmed that the pool lead length and width (PLL, PLW) which are calculated by feature points are recognized with high accuracy by CNNs learning model. On the other hand, it is also found that a large corpus of labeled images is required to obtain the high performance of learning model. In order to reduce costly expert annotation, we propose a self-training method which uses unlabeled images. As a result, it is confirmed that the PLL and PLW are recognized accurately by the self-training method proposed. Finally, results of demonstration of automatic welding with real time image recognition and robot control are described. These results show that horizontal penetration bead welding with and without ceramic backing is possible to be automated by robot system proposed.

抄録全体を表示

Robotic welding has been introduced in GMA welding to save labor. Estimation of the penetration depth of the molten pool is important to obtain good welding results in this welding. The authors used the image of the weld pool as input for deep learning to identify the welding state and estimate the melting state of the root of the groove. In this study, a single-sided downward-facing weld with a ceramic backing material is used. In order to apply deep learning, it is necessary to construct training data. A large amount of time is required to construct the training data. The relationship between the melting state of the root of the groove and that of the ceramic backing material was determined by basic experiments. The ceramic backing material melts when the brightness of the state in front of the weld pool increases. Using this feature, the state of the tip of the weld pool in the molten pool image was classified into three types. This facilitated the image classification. In addition, since only the tip ofthe molten part was targeted, the image was less susceptible to changes in the bevel shape. This was applied to the case where the gap varied from 4 mm to 6 mm. The effectiveness of the estimation of the melting state of the root of the groove was confirmed.

抄録全体を表示

For the development of in-process Blow Hole(BH) detection and prediction technology, welding system with infrared camera and welding robot was integrated and the authors verified the detection of welding phenomena directly associated with the occurrence of BH. In addition, an image analysis index for welding defect detection and prediction was considered. It was clarified that it is possible to observe the bubbling phenomena just under the arc due to the blow hole generation by the welding system. In addition, it was considered possible to detect and predict the occurrence of BH with high accuracy by analyzing the observation frequency of the bubble generation phenomenon directly under the arc with an appropriate index.

抄録全体を表示

In ESW process, welding phenomena and mechanism have not been understood enough. In this study, phenomena in the slag bath during ESW were observed by using X-ray transmission imaging system. By changing welding conditions, various droplet transfer mode similar to GMAW were observed in the slag bath. According to the temperature measurement result in the slag bath, upper and middle part was around 1700-1900 K that was near the melting point of iron, and bottom part was more than 2300 K. Therefore, it is presumed that most of welding current is concentrated at the bottom of the slag bath and spreads radially from the wire tip, and then the electromagnetic pinch force acts effective in the droplet detachment. In addition, in the bottom of slag bath, high-speed rotating convection was observed. It is estimated that penetration depth is governed by this convective heat transport. These observations indicate that wire tip position and temperature distribution in the slag bath are determined by the setting of welding conditions, then the welding current path in the slag bath changes by these acting mutually. It is suggested that droplet transfer mode and convection phenomenon are also determined by the current path in the slag bath. Then the effects of wire extension length and slag bath depth on wire melting rate were also investigated. In the case of the same wire extension length and the same welding current, the wire melting rate of ESW was lower than that of EGW. Furthermore, it was clarified that Joule heat at the wire extension and wire melting rate varied with the slag bath depth also, despite the same wire extension length. Therefore, the slag bath depth control method should be developed to expand the application of ESW process in the future.

抄録全体を表示

A penetration process in electroslag welding is simulated using a two-dimensional smoothed particle hydrodynamics method to elucidate the heat transfer mechanism. The base metals are melted and penetration forms in the orthogonal direction of the weld line. From the base metal, molten metal flows downward, and then a molten metal pool forms below a slag bath. The distributions of temperature increasing rates due to heat conduction and Joule heating are also visualized to clarify the dominant factors contributing to the temperature increases of the molten slag and the base metal. Joule heating occurs mainly at the side and bottom of the wire in the slag bath. The temperature of the base metal is increased by the heat conduction from the high temperature slag. The factors of energy transfer in the system are evaluated quantitatively as well. Most of the input electric energy is converted into the thermal energy by Joule heating in the slag bath. Approximately 80% of the generated thermal energy is transported to the base metal, while the other 20% is transported to the molten pool. Compared with the heat transfer in a gas metal arc welding, the thermal energies in the electroslag welding are larger into the base metal and smaller than those in a gas metal arc welding.

抄録全体を表示

A steel/aluminum alloy dissimilar resistance spot welding process was simulated by a three-dimensional smoothed particle hydrodynamics method. Furthermore, the time dependent increase of the intermetallic compound thickness on the joining interface was estimated using the numerical data of the temperature history obtained by the simulation. As a result, the steel sheet started to melt from the center of the sheet in the thickness direction and formed a nugget, while the aluminum alloy sheet started to melt from the joining interface and formed a nugget. The convection in the molten aluminum alloy caused by the electromagnetic force promoted the heat transfer at the solid-liquid interface because the temperature gradient become steeper due to the conduction, whereas the temperature near the nugget center decreased. Moreover, the numerical estimation indicated that the intermetallic compound layer was thicker near the center of the joining interface and thinner toward its edge. The maximum thickness was estimated to be approximately 1 μm, which was the same order of magnitude as the experimentally obtained value. These results support the validity of the computational model developed in this study for simulating the nugget formation process during dissimilar resistance spot welding and estimating the intermetallic compound thickness.

抄録全体を表示

Friction stir spot welding (FSSW) is one of the solid state welding method and dissimilar material joining by FSSW have been investigated. However, a removable coating is required to form a clean surface on the lower plate in conventional FSSW method. Therefore, we have developed a new dissimilar material joining method—scrubbing refill FSSW (Sc-RFSSW)—, and Sc-RFSSW provides higher joint strength than conventional FSSW methods in joining the aluminum alloy and non-coated mild steel. In this study, the mechanical properties and metallurgical joining interface of an aluminum alloy and a hot-dip galvanized (GI) coated steel joint by Sc-RFSSW were investigated. In addition, a new tool tip shape was examined to improve the interfacial cleaning effect.

抄録全体を表示

Since welding hot cracking is one of the most harmful defects which may largely reduce the strength of welded joints and structures, it is very important to study the prediction and prevention of its generation. Hot cracking is caused by both mechanical and metallurgical factors. In this research, authors proposed a new computational method for evaluating the occurrence of hot cracking through considering both mechanical and metallurgical behavior. As the metallurgical behavior, strain due to solidification shrinkage and mechanical resistance of molten metal were modelled using the solidification ratio as a parameter. The mechanical behavior, an index for generation of hot cracking using the increment of plastic strain is proposed. In addition, authors proposed a simplified prediction method using the temperature gradient for the direction of columnar crystal growth, which affects the location of hot cracking, and is used to evaluate the associate angle of the columnar crystals. The developed method was applied to butt welding. As a result, it was found that the predicted direction of columnar crystal growth by the proposed method is in good agreement with the experimental results. In addition, the generation of hot cracking in the same butt welding was predicted using the proposed method and the obtained cracking positions agreed well with those observed by experiment. The Influence of the strain of solidification shrinkage on hot cracking was investigated in relation to the surrounding constraints, the temperature gradient and solidification mode, which are presumed to be highly related to the occurrence of hot cracking. From the above results, it can be concluded that the proposed method can analyze the effects of mechanical and metallurgical factors on hot cracking.

抄録全体を表示

For the automation of butt welding, it has been considered to introduce tack welds instead of out-of-plane restraint jigs to perform welding with large heat input. However, the occurrence of hot cracking has been a significant issue in large heat input welding. In the previous report, the authors have developed a prediction method of hot cracking that can investigate the influence of not only the high temperature strain generated in the weld metal but also the phenomena related to solidification such as solidification shrinkage, latent heat of solidification and crystals growth direction. In this study, the conditions for prevention of hot cracking were investigated by numerical analysis and experiments through controlling the temperature gradient at solidification temperature using a trailing torch in tandem welding. As a result, it was confirmed that the large strain existing position predicted using proposed hot cracking analysis has a good agreement with the cracking position observed experimentally in tandem butt weld. It was also found that with increasing the electrode distance, both the columnar crystals associate angle became larger and the local strain became smaller. It can be concluded that the hot cracking can be successfully prevented in the practical experiments using the proposed large electrode distance.

抄録全体を表示

Fatigue crack propagation stage is considered to be the dominant in fatigue fracture of welded joints and some solutions regarding stress issue have been developed. It is known that fatigue crack forms slip band at crack tip and propagates by accumulation of irreversible deformation against cyclic loading. Then if angle difference between (110) plane which is the main slip plane of body centered cubic lattice and the plane of which angle is 45° to the crack plane with respect to crack propagation direction is maximum, suppression of fatigue crack growth is expected. And the delay of fatigue crack growth is also expected, when the compressive stress is developed in the steel plate. In the present study, fatigue crack path and fatigue crack growth rate in the test steel with highly oriented texture and internal stress distribution were investigated. The angular difference with (110) plane and crack plane got largest by accumulating the (110) plane perpendicular to the plane thickness direction and propagating the crack in the thickness direction. Fatigue crack growth rate of steel with highly oriented texture was lower than that of normalized steel, and the crack growth was remarkably suppressed when compressive stress was superimposed inside the steel plate. It is considered that crack closure effect by fracture surface contact with increase of the fracture surface ruggedness appeared, because fatigue crack branching and bending in steel with highly oriented texture were observed.

抄録全体を表示