One of the important subject in the study of automatic arc welding is to establish control system of welding parameters for controlling bead shape, namely penetration depth, bead width and bead height, by detecting or predicting behavior of molten pool owing to change of groove shape. This report describes the effect of welding parameters on the bead shape and the study of simultaneous control system of penetration depth and bead height in the variation of root gap in gas metal arc welding. From the fundamental investigation of optimum penetration control method, following results were obtained. 1) By the method of controlling additional filler wire feeding rate in order to maintain constant torch height during automatic current control (ACC), the penetration depth and the bead height can be controlled constant at the variation of root gap width in Y-groove. 2) The penetration depth can be effectively controlled by the welding current control which is corelated to the welding speed control in order to keep constant bead height.
One of the important subject in the study of automatic arc welding is to establish control system of welding parameters for controlling bead shape, namely penetration depth, bead width and bead height, by detecting or predicting behavior of molten pool owing to change of groove shape. This report describes the study of simultaneous control system of penetration depth and bead height despite the variation of root gap in gas metal arc welding with X-groove. In this system, root gap width is detected by image processing with CCD (charge coupled device) camera, and welding current and welding speed are controlled in order to keep constant penetration depth and bead height. Furthermore in this system, wire feeding rate and terminal voltage are controlled by using equations proposed by Lesnewich and Halmφy so that both wire extension length and arc length can be kept constant during torch height control with ACC method. This control of both wire extension length and arc length is very efective to maintain stabilized welding arc. From the result of an experiment with X-groove of SUS304 stainless steel, both penetration depth and bead height can be simultaneously controlled to be almost constant despite the variation from 0 to 3 mm of root gap width.
Tensile test of base metal and electron beam welded joint of some aluminum alloys (A7N01, A5083, A6061) and a ferritic special steel (JFMS) was conducted at cryogenic temperature. These materials are candidate materials for nuclear fusion reactor in Japan. Testing temperatures were 293K, 77K and 4.2K. Main testing results obtained are as follows: 1. Most aluminum alloys and their welded joint exhibit serration in the tensile test at 4.2K. 2. Base metal and welded joint of A5083-0 alloy exhibit comparatively high strength and high ductility at cryogenic temperature. 3. Welded joint of A6061-T4 alloy and A7N01-T6 alloy exhibits relatively low ductility at cryogenic temperature compaired with base metal. 4. Base metal of JFMS exhibits brittleness at 77K. Welded joint of this steel exhibits high notch brittleness at the temperature between 293K and 77K.
Liquid sodium is used as the coolant for liquid metal fast breeder reactor. The 21/4Cr-1Mo steel is used as structural material for evaporator and SUS F321 steel is used for super heater of steam generator of "Monju", a prototype fast breeder reactor in Japan. The trial manufacture of small sodium loop made of SUS 316 steel and a study on the sodium corrosion of explosive plugged part are already reported by us. Study on the sodium corrosion of arc welded part of SUS 321 steel, SUS 304 steel and 21/4Cr-1Mo steel is reported in this paper. Measurement of weight loss and analysis by X-ray photoelectron spectrum are conducted for the corroded specimen. Effect of welding on the corrosion characteristic of these materials is not evidently observed.
Pure iron plates were welded using Fe-Si alloy electrode wires in a controlled arc atmosphere. The effects of silicon on the oxygen contents and non-metallic inclusions in the weld metals were investigated under various welding conditions in Ar-O2 and Ar-CO2 atmospheres. The oxygen contents of the Fe-Si weld metals decrease with an increasing welding current and increase with an increasing arc voltage at low oxidizing gas partial pressure. The traveling speed has not a marked effect on the oxygen contents of the Fe-Si weld metals at low oxidizing gas partial pressure. The oxygen contents of the Fe-Si weld metals increase with an increasing the partial pressure of oxidizing gases and decrease with an increasing silicon contents in the weld metals. The oxygen contents of the weld metals welded in Ar-CO2 are lower than those in Ar-O2. The nonmetallic inclusions in the Fe-Si weld metals increase with an increasing oxygen contents of the weld metals. The nonmetallic inclusions are silicon-dioxide accompanying with iron-oxide. Behavior of the oxygen absorption into the Fe-Si weld metals is discussed using thermodynamic data.
Pure iron plates were welded using Fe-Mn and Fe-Si-Mn alloy electrode wires in a controlled arc atmosphere. The effects of manganese and silicon-manganese on the oxygen contents and non-metallic inclusions in the weld metals were investigated under various welding conditions in Ar-O2 and Ar-CO2 welding atmospheres. The oxygen contents of the Fe-Mn and Fe-Si-Mn weld metals increase with an increasing the partial pressure of oxidizing gases and decrease with an increasing manganese and silicon-manganese contents in the weld metals. The oxygen contents of the weld metals welded in Ar-CO2 are lower than those in Ar-O2. The oxygen contents of the Fe-Si-Mn weld metals significantly decrease comparing with those of the Fe-Si and Fe-Mn weld metals. The nonmetallic inclusions in the Fe-Mn and Fe-Si-Mn weld metals increase with an increasing oxygen contents of the weld metals. The nonmetallic inclusions in the Fe-Mn weld metals are oxide of manganese-oxide accompanying with iron-oxide. The nonmetallic inclusions in the Fe-Si-Mn weld metals are oxide composed of silica, manganese-oxide and iron-oxide. Behavior of the oxygen absorption into the Fe-Mn and Fe-Si-Mn weld metals is discussed using thermodynamic data.
MIG welding under high pressure Ar atmosphere (0-6 MPa) has been studied by means of welding chamber. Effects of pressure on the arc voltage, welding current and weld metal shape are studied. MIG welding process by constant arc voltage condition and constant arc length condition are compared. The main results are summarized as follows; 1) Increasing pressure from 0 to 6 MPa, arc became to constrictive, plural cathode spots and unstable arc were observed. Brightness of arc column was increased. A lot of spattered droplets and fine particles were produced. 2) As welding s carried out using constant arc voltage condition, welding current was decreased above about 2 MPa. Decreasing of heat input under high pressure was observed. Crowned weld metals, narrow bead width, weld metals which was not deep penetration were shown. 3) As welding was carried out using constant arc length condition, arc voltage was raised remarkably with pressure then also, heat input was increased. Crowned welds were not obtained. Bead width, penetration depth were not changed remarkably in-creasing pressure from 0 MPa to 6 MPa. Area of HAZ was affected by increasing pressure. Tensile and bending test results were satisfactory under condition of this study. 4) It was estimated that constant arc length condition is good for MIG welding process under high pres-sure arc atmosphere.
Rectangular wave AC TIG arc welding phenomena were investigated with respect to the welding current waveform, using an experimental AC power source of the transistor controlled inverter type, which was designed and set up. Current level and duration of each polarity in a cycle were separately varied from 0 to 300 amp. in the frequency range up to 500 Hz. It is shown that the current level and the duration of EN polarity in a cycle are dominant parameters in the current waveform for the arc pressure and the bead formation, but those of EP polarity are responsible for the cleaning action in AC TIG welding of Al alloy.
It is well-known that CO2 welding of galvanized steel often makes a porosity, such as a blow hole or a worm hole, in weld metal. In this study, in order to investigate the mechanism of porosity formation process in weld metal of galvanized steel, zinc analysis on inner surface of blow hole, gas chromatography analysis for gas in porosity and gas evolution test of weld metal of galvanized steel were carried out. Moreover, several types of welded joint were fabricated by the best welding conditions expected for giving less porosity, and their weld results and static strength was evaluated. As the results, the following matters were concluded. (1) The formation process of porosity was considered as follows: Hydrogen which was expected to exist in Fe-Zn compound layer formed on mild steel surface as zinc was coated, was came out into molten pool during welding. As the thermal instability, such as large spatter formation, was occurred in molten pool, continuous gas evolution from melted metal was intercepted, and the porosity was formed in weld metal at that time. (2) It was clarified that the welding condition which gave less porosity was low welding voltage, low welding current and low welding speed for 1.6-2.3 mm thichness galvanized plate. And it was noticed that arc point must be hold the best position to make arc stable. (3) Appearance of obtained weld bead and X-ray results of welded joints were estimated good, and joint efficiency was about 1.0 in all types of welded joint except for specimens in which the large weld distortion or melt out hole were occurred.
Electromagnetic welding is classified as a high energy rate welding, and hence, the feasibility of this welding depends, in principle, on the velocity of flying body at the time of collision, as experienced well in the case of explosive welding. The authors pointed out in this paper that its acceleration could be an important factor in this welding, as well as its velocity. The deformation test under a high velocity on the aluminum tube and the welding test on the joint of tube and core brought the following conclusions. (1) Under the high strain rate condition of electromagnetic welding, the increment of flow stress caused by the strain rate hardening should be accounted, other than that by the strain hardening, because the former increased very significantly with increasing the strain rate. (2) Both the velocity and acceleration of flying tube controlled the feasibility of welding; the larger the acceleration was, the smaller became the velocity required for obtaining the welding joint. (3) The result of (2) suggested that the welding was performed when the sum of the kinematic energy and strain energy given by the pressure at collision reached a critical value.
Detection of groove information such as gap center-weld line and gap width of groove by projecting vibrating laser slit-like light is shown based on the analyeis of intensity distribution of laser slit-like light reflected or scattered from the surface of two kinds of steel sheets, each involved gap width (0.1-1.0 mm) with I shape groove. Intensity distributions of laser spot light and laser slit-like reflected light vary in space or time, but the intensity variation decrease nearly 50% by vibrating laser slit-like light. In addition, the intensity variation of laser light is influenced by the surface property of the object. Detection. of gap center-weld line and gap width of I shape groove of specimen is based on sharp reduction of intensity of reflected light at root gap. Furthermore, an algorithm for detection of weld line and gap width of groove has been developed and it's effectiveness has been proven experimentally.
In this report, mechanism of grain boundary liquation was investigated by high temperature microscope. The results indicated that (1) temperature of grain boundary liquation by increasing heating rate was lower than solidus temperature (2) grain boundary liquation was closely related to melting of inclusion.
22% Cr-3% Mo duplex stainless steels in which Ni and N contents were varied independently were prepared and base metal and TIG weld metal were subjected to pitting corrosion tests in 6% FeCl3. The critical pitting temperature at which pitting corrosion occurred was evaluated and the pitting site was observed. The pitting test results were examined metallurgically from the viewpoint of the precipitation of Cr carbide and nitride and the distribution of elements between ferrite phase and austenite phase. Since the pitting corrosion resistance of austenite phase is mainly dominant in base metal, the decrease of Ni and the increase of N cause the increase of N in austenite phase and improve the total corrosion resistance. As the decrease of Ni and N reduces the generation of austenite phase and results in the precipitation of Cr carbide and nitride in weld metal, the decrease of Ni and N deteriorates the pitting corrosion resistance of weld metal.
The condition for occurrence of weld solidification cracking is analyzed in relation to strain rate by direct observation technique MISO for plain carbon steels, stainless steels and an Inconel alloy. All the materials used show a dependence of critical strain required for crack initiation on strain rate. Namely, the critical strain increases with an increase in strain rate. The dependence becomes bigger with a decrease in crack susceptibility. Therefore, the critical strain in a high strain rate is more likely to differentiate the crack susceptibilities among materials than in a low strain rate. On the other hand, the strain rate in Houldcroft-type cracking test and in case of some welding fabrications is shown to be so low that the difference of the critical strain among materials is little. Under such low strain rate, the critical strain rate below which the crack can not occur is useful for the comparison of crack susceptibilities.
The reason why the critical strain required for the initiation of solidification crack increases with an increase in strain rate is studied. Weld solidification cracking tests and simulated hot ductility test are carried out for stainless steels and Inconel alloy to observe grain boundary sliding. Under the low strain rate as critical strain rate below which the crack can not occur, grain boundary sliding occurs within brittleness temperature range during weld solidification. On the contrary, it hardly occurs under a high strain rate. The contribution ratio of the deformation due to grain boundary sliding to total deformation in weld metal is about 0.2 with simulated hot ductility test under the condition of low strain rate as critical strain rate. Therefore, it is considered that the grain boundary sliding has an important role to the solidification crack initiation in welding. Then, the reason why the critical strain increases with an increase in strain rate is explained from the view-point of contribution of grain boundary sliding to the total strain of weld metal.
In the present paper, Elasto-Plastic Fracture Mechanics (EPFM) was applied to an assessment of initial defect in fatigue design. Considering fatigue design methods of welded steel structures, assessment methods of iditial defect in notched components were classified into three cases. In the Case 1, fatigue design is based on joint classification, where allowable size of initial defect is specified according to joints shape. Safety factor to fatigue life varied with shape of components. In the Case 2A, fatigue design is based on joint classification, where allowable size of initial defect is, in this case, assessed by ΔJ and experimental method. The size depends upon stress level and materials (SS41 or HT80). And in the Case 2B, fatigue desing is based on apriori specified initial defect size and EPFM. By this design method, to increase sectional area is better than to decrease tensile residual stress when fatigue strength is critical in low cycle region, and vice versa.
In designing an elevated temperature vessel with no small deformation of creep, it is required to predict its accumulation throughout the service life. For welded joints to build such a vessel, however, any practical way for prediction. has not been definitive yet. In the present study, some measurements and analyses by a finite element method (FEM) were made of the creep behavior of a 50 mm thick 304 stainless steel butt-welded joint at 550°C. Narrow gap submerged arc welding was used for preparing it. Considering metallurgical inhomogeneity in the joint, creep tests were made of the base plate and weld metal in advance. For the heat-affected zone, a "synthetic heat-affected zone" technique was attempted to prepare test specimens. The obtained data were supplied for the FEM computations. The computations were excuted by elastic analyses with incremental time, while strain contours on the cross-section of the joint were measured by moire photography. The computations agreed with measurements fairly for creep curves and approximately for strain contours.
Brazeability of high purity alumina (99.9 wt% Al2O3) to commercially pure titanium (CP Ti) and Ti-6A1-4V alloy with Ti-Zr base filler metals has been investigated by microscopy, EDX microanalysis, X-ray diffraction, and shear test of the joints. Filler metals used were two types of amorphous foils, developed for trial, of 25Ti-25Zr-5OCu (Type 5000) and 37.5Ti-37.5Zr-l5Cu-lONi (Type 1510), whose melting points were about 815°C and were apploximately 100°C lower than those of conventional Ti-base fillers. Shear strength of joints brazed with Type 5000 filler metal had a tendency to decrease with increase in both brazing temperature and time. However, in case of brazing at comparatively low temperature for a short time using Type 5000 filler metal, it was possible to get sufficient joints having shear strength of above 150 MPa. Metallurgical test results showed that in the joint of alumina to CP Ti with Type 5000 filler metal, (Ti, Zr)2Cu was formed at joint region nearby CP Ti base metal at the beginning of brazing, and that the composition of the region varied with increase in brazing time. A thin titanium and/or zirconium oxide layer was also detected at the interface between alumina and the joint region, and the growth of this oxide layer was related to the decrease in joint strength and to the occurrence of cracking at the joint. As for using Type 1510 filler metal, the joints were flaked during cooling after brazing, because of rapid formation of brittle Ti and Zr rich layer at the joint interface, which indicates that the brazeability of Type 1510 filler metal is fairly inferior to that of Type 5000 filler metal.
In a new Chromium (Cr) metallization process, effects of thickness of the Cr metallized layer and reaction products in that layer on the bond strength and Helium leak rate of the brazed layer were investigated. The results of this study are as follows, (1) The bond strength and Helium leak rate of the brazed layer were influenced considerably by the thickness of the Cr metallized layer and the reaction products in the metallized lyer (2) A high bond strength and very small Helium leak rate were obtained when the thickness of Cr metallized layer was between 0.6 μm and 2μm. (3) The reaction products were guessed as a complex compound (Cr.Si.C) and Cr silicide (Cr3Si2) from the SiC side when the thickness of Cr metallized layer was 1.4μm. The reaction products were the same as that when the thickness of metallized layer was 6μm. But the grain sizes of reaction products ranged from 0.07 μm to 0.18μm when thickness of Cr metallized layer was 1.4μm. Those reaction products were about 1/10 in grain size as compared with that when the thickness of Cr metallized layer was 6μm.
Bonding of Si3N4 to metal using liquid insert metals of Cu-base binary alloys containing Cr, Nb, V, Ti or Zr was conducted in vacuum of about 6 mPa. Joint strength was evaluated by tensile-shear tests. The heighest strength of Si3N4 to W joints bonded using Cu-5%Cr, Cu-1 %Nb, Cu-3%V, Cu-5%Ti and Cu-10%Zr insert metals were about 150 MPa, 140 MPa, 140 MPa, 180 MPa and 180 MPa, respectively. Effect of bonding conditions on area fraction of void in Si3N4 to W joints was investigated. Area fraction of void increased with increasing bonding temperature and holding time. It was considered that formation of void was mainly caused by outflow or vaporization of insert metal.