This paper reports the result of low cycle fatigue test. The laod of this test was given by pure bending shown in Fig. 4, and the amplitude of moment was constant. The specimens consist of two kinds. One is butt-welded in centre part of the plate as shown in Fig. 2, and the other is base metal. Variations of steel classfication are four: SM41, SM50, SM50Y, and SM53. The results of this test are summarized in Table 3. It was ascertained by this test that fatigue life in low cycle fatigue range is influenced by tensile strength more than by yield stress of coupon test, and that fatigue strength of welded specimens is as high as that of base metal if welding is good. The trait of crack propagation was distinguished in two types. (a) propagation in weld metal zone (b) propagation in heat-affected zone It was observed in this test that (b) type was more preralent than (a) type.
Measurements of the reaction stress and the intensity of restraint were undertaken and the results were compared with the analytical results in Part 1. Experiments were made for two types of specimens, H-type and T-type (see Figs. 14 (a) (b) ), in which bending moment by reaction forces may or may not be neglected respectively. Some analyses showed good agreement between test results and calculations (see Figs. 17 (a) (b)). It was found that the occurrence of weld joints might be estimated by comparison of the Kcr-value with the K-value. The necessary value of uncompleted fillet-weld length 2Δ to prevent weld cracking may be obtained from equation (14) for given values of joint geometry, size of fillet and welding heat input.
It is generally acknowledged that the ductile joint of molybdenum is difficult to make by fusion welding because of its embrittleness due to recrystallization and grain growth. The purpose of this investigation is to obtain a ductile joint of pure molybdenum without recrystallization and extreme deformation by diffusion or pressure bonding processes with or without insert metals. The following resuls were obtained. 1. Although the joint of molybdenum without insert metals was accomplished at higher temperature than the recrystallization temperature with 5-10% deformation, the joint made at higher temperature than 1200Δ was brittle due to recrystallization. A transfacial grain growth across the initial bonding interface was observed to occur at 1250Δ for 5 min and such migration of grain boundary was easily obtained relatively at lower temperature compared with that for various metals. 2. It was found that the joint of molybdenum without recrystallization and extreme deformation was accomplished using insert metals such as iron, nickel, copper and silver by electroplating. The remelting temperature of these joints became too higher than the melting points of the insert metals. Although nikel formed an intermetallic compound with molybdenum, nickle insert metal made a stronger joint at low temperature than the others. Copper was also found to be applicable as an insert metal, but it did not alloy with molybdenum.
The present study was undertaken to determine the fatigue strength of weld joint of high tensile steel of 80 kg/mm2 (≈ 110 ksi) level. The endurance limit of this steel was 46 kg/mm2 obtained from repeated bending fatigue test. The endurance limit was remarkably decreased to only about 13 kg/mm2 due to welding. The metallurgical structural change caused by welding seems not to be a dominant factor for the decrease in the fatigue strength. Because the endurance limit increased to 40 kg/mm2 when the reinforcement was completely machind off after welding. When the reinforcement was almost completely dressed, leaving the micro-defects on the surface of the plate, the endurance limit was again 40 kg/mm2. Therefore, the micro-defects appearing at the toe of the reinforcement have little effect upon the endurance limit, although such defects were always the sites from which fatigue crack was initiated. Macroscopic stress concentration due to reinforcement may be a major factor for the reduction of fatigue strength of the joint. However, fatigue test results on the simulated weld specimens, machined from the base metal, showed that the fatigue strength reduction factor was never higher than 3.0 in the present investigation. It is concluded that the drastic reduction in fatigue strength of weld joint can not be completely explained by any one of factors alone but the strength reduction by the reinforcement must be intensified by the other factors such as micro-defects and metallurgical structural changes. To determine the endurance limit of joint, an apparent fatigue strength reduction factor, Kfw, is proposed as follows, Kfw=Kf⋅Cu⋅Cn⋅Cs⋅Cx where Kf is the fatigue reduction factor due to reinforcement, Cu, Cn and Cs correction factors for micro-defects, notch sensitivity of material and structural changes, respectively. And Cx is also a correction factor for other unknown effects. The value of each factor except Cx was determined from the result of present investigation and the calculation of endurance limit was made. A good agreement was obtained between the calculated and observed endurance limits of weld joint.
Nickel, copper, chromium and tin plated steel in 1 mm thickness are spot-welded and cross sections of the welds are subjected to microprobe analysis to make clear the behavior of the plated metal during spot welding. The observation of macrophotographs and microprobe analysis elucidate that in spot welding of the nickel, copper and chromium plated steel, plated metals mingle into the weld nugget under violent stirring motion of the fused metals, resulting in formation of homogeneous alloys in the short weld time (about 10 cycles) and that, on the other hand, in spot welding of the tin plated steel, tin is expelled from the weld interface because of its low melting point and it does not mingle into the weld nugget.
Deoxidation equilibriums in Arc melting were studied as the first step in the investigation of metallurgical reactions in arc welding. The iron specimens containing oxygen and some deoxidation alloying elementss were melted under arc using a tungsten electrode in an Argon atmosphere, and these deoxidation reactions were investigated. The results of the Si, Mn, Si-Mn, V, Ti, Al deoxidations and the oxygen solubility in pure iron are as follows. (1) Within 100 sec., all the reactions reach the apparent equilibriums. (2) Being compared with the thermodynamical equilibrium data, the results of the apparent equilibriums of the Si (<0.2%), Mn, V, Ti deoxidations and the oxygen solubility under the same melting condition are almost consistent with the equilibrium data at the same temperature (e.g., corresponding equilibrium temperature 16401670°C at Argon pressure 500 mm Hg, current 500 A). With increased Argon pressure or current, the corresponding equilibrium temperature is increased at the rate of 1015°C/100 mm Hg or 2030°C, /100 A in the case of the oxygen solubility. However, the presence of Mn in the Si-Mn deoxidation is not effective as expected and the Al deoxidation is less active than calculated from the data of equilibrium., It is probalbe that the fact of the latter is caused by the present experimental method. (3) The Si and Si-Mn deoxidations (Si>0.5%) are very active, and these results can not be interpreted by the SiO2 forming reaction. It is anticipated that the equilibriums will be controlled by the SiO forming reaction at the constant SiO pressure. This a characteristic phenomenon in arc melting.