Hydrogen concentration distribution in multipass weld metal has been studied using hydrogen diffusion theory under various welding conditions. It is shown that hydrogen concentration distributions are predicted by means of parameters given by welding conditions, namely, heat input, plate thickness, weld metal width and so on. Influence of welding conditions on hydrogen concentration distributions is described.
The existence of the unusual fracture path transition phenomenon in Charpy impact test specimen had been already found out on the electron beam (EB)-weld metals of constructional high strength steels in the previous report. Then in this report, the occurring possibility of such "fracture path transition phenomenon" and its mechanism were theoretically studied by means of the stress analysis in static load for the bending model of impact test specimen of EB-weld metal using finite element method (FEM). The stress analysis was performed on the bending model for HT50 and 80 weld metals of 12 and 40 KJ/cm weld heat inputs. The remarkable conclusions are as follows; (1) As compared the distribution of plastic zone in the model specimen for EB-weld metal between 12 and 40KJ/cm weld heat inputs, the remarkable large extensions of the plastic zone were observed in the part of base metal near the fusion boundary (Bond) in case of the former model specimen. From this results it is presupposed that the fracture path of this impact test specimen has a chance to deviate into base metal side in case of the weld metal of low weld heat input. (2) It was expected that the fracture path transition has a tendency to occur with a decrease of the bead width in the test specimen dB and with an increase of the hardness in the weld metal Hv, w as a result of the comparison for the variation of the plastic strain at the base metal near the fusion boundary (εp, B) and the center of weld metal (εp, w) from V-notch bottom in the model specimen between 12 and 40KJ/ cm EB-weld metals.
The feasibility of underwater welding by the submerged arc welding process is investigated. In this process, a special flux rrixed water glass with MF43 is used to stabilize arc and to protect the part being welded from the sorrounding water. Welding experiments are done by using steel for welded structure SM41 of 9mm in thickness and it is ascertained that this process may be put to practical use. Main results are summerized as follows: (1) Welding arc can be generated easily and kept in stable, if only proper welding conditions are selected. Especially, good welds can be obtained easily by using a special flux mixed 40wt% water glass with MF43. (2) The weld metal includes blowholes to some extent. (3) The hardness of the underwater weld seems to be relatively low. (4) Weld metal has the tensile strength of 82% of that of base metal. (5) Cooling rate at 500°C at bond is about 23°C/sec under water and 10°C/sec in air, the former being about only 2 times higher than the latter. The former seems to be lower value than those by any other underwater welding processes.
The characteristics of the image in the optical joint configuration sensing method reported in the previous paper are obtained in connection with the welding condition, such as arc voltage and arc current. The one or two dimensional parallel processing method, in which the obtained image is converted into binary data by the array of light comparators installed on the screen of the image receiver and the required values are extracted by the logic circuit, is described, in the next stage. The result of the feedfoward arc welding control with this sensing method is shown as demonstrating the usefulness of the method in the last stage. All the experiments were performed by CO2 arc welding on the square grooved joint.
Effects of finishes of weld toe and mean stresses on bending fatigue strength of transverse fillet welded joints were investigated. Variation of radius and position of weld toe and flank angle along the weld bead were statistically analyzed. The results are summarized as follows i) Fatigue strength of SM58Q (MIG⋅as-welded), SM58Q (TIG finished), HT80 (ground), HT80 (ground+peended) increased to 114%, 130%, 120% and 214% at Nf=106 under minimum stress σmin=0 compared with that of the corresponding low hydrogen electrode·as-welded specimens. However, the effect of peening vanished in the low cycle region under σmin=0 as well as in the entire region under high mean stress. ii) Statistical analysis of the values of curvature (1/ρ, ρ : radius of weld toe) along the weld bead were performed for normal, logarithmic normal, Weibull and Gumbel distribution. It was found that the curvature of weld toe in all cases of weld finishes obeys the Gumbel distribution. iii) Fatigue strength reduction factor Kf was calculated using equivalent stress concentration factor Kreq which was derived from averaging the scatter of stress concentration factor Kt (caused by the variation of ρ) and the calculated Kf values were in good agreement with the test results. iv) Fatigue strength under high mean stresses in the neighborhood of yield point ay decreased little compared with that under σmin=0 except for the peened specimen.
Wedled structures generally consist of the following basic welded joints; the bead-on-plate welded joint, the edge welded joint and the groove welded joint. It is significant to clarify the relationship between the welding conditions and the residual stress distributions in above the basic welded joints in order to estimate the residual stress distributions in welded structures. This report deals with how to estimate the residual stress distributions in the basic welded joints. The main results obtained are as follows: 1) The residual stress distributions in the bead-on-plate and the edge welded joints are determined by the heat input parameter Tav calculated from the welding conditions the same as the residual stress distributions in groove welded joint. 2) The residual stress distributions in the fillet welded joint are approximately estimated from the residual stress distributions in the basic welded joints in case the web is greater.
In previous report, a new method was introduced for obtaining residual stress distributions in welds qualitatively by measuring surface relief induced due to cutting and annealing. This paper presents the micro-distributions of residual stress in welds obtained by this new method in four kinds of steel having a different transformation characteristics. Based on these test results, some considerations are performed on the micro-mechanism of welding residual stress, and it is clarified that a cause of reheat cracking by a subsequent welding pass occured in multi-pass weld of high strength steels.
Relations between ductility of the weld heat-affected zones (HAZ) and postheating conditions were investigated by means of bead bend test concerning the specimens of 13Cr, 13Cr-Ni and 13Cr-Ni-Mo cast steels. The results of these tests are summerized as follows. 1) In order to improve enough the ductility of the HAZ, it is necessary that the temperature of postheating is made above 600°C (below Ac1 point) and that the holding temperature before commence-ment of postheating is set lower than near the Mf point. 2) The ductility of the HAZ displays a strong tendency to depend upon the holding temperature. When the holding temperature is made higher than the vicinity of the Ms point, the ductility will not improve even with postheating at temperatures above 600°C. 3) In the range of suitable holding temperature, amount of martensite in the HAZ is approximately more than 80 percent.
The diffusion equation by which the local accumulation or segregation of mass can be analyzed is of non-Fickian type. Its equation form is not so simple as that of Fickian equation and it is, thus, difficult to solve mass segregation problems, especially in analytic manner. If the activity of diffusing mass instead of the concentration is taken as a variable in the equation the diffusion equation can be expressed by a very simple form as : ∂a/∂t=γ∇(D/γ∇a) In this equation the activity coefficient γ is a product of γo and γp which represent the chemical potential term and kinetic potential term, respectively. The form of this diffusion equation is as same as that of heat conduction, γ being equivalent to a product of density and specific heat and D/γ equivalent to heat conductivity. It follows that the solution for the mass diffusion involving the local accumulation can be easily given by looking into the analytical solutions of heat conduction. More complicated problems such as hydrogen accumulation in the weldments can benumerically solved by making use of heat transfer computer programs.
The plastic zone and the plastic strain behaviours near the propagating fatigue crack for SM41 and HT80 steels were observed continuously on various specimen thicknesses, using optical interference method. The relationship between the shape or size of the plastic zone and stress intensity factor range (ΔK) was clarified. In addition, the effects of plastic zone size, and the slope of plastic strain distribution on crack propagating rate were investigated on each test steel. Main results obtained are as follows. 1) The necking zone appears periodically ahead of a propagating carck in SM41 steel of which the lower yielding point is clear. The size of necking zone increases with a growth of crack. However, this periodic necking zone is not shown in HT80 steel. 2) The relationship between the length of plastic zone (Rx) and the ratio of ΔK/σY, appeared as follows; Rx=C1 (ΔK/σY)2 for SM41 steel Rx=C2 (ΔK/σY)1.5 for HT 80 steel Here, C1 is affected by specimen thickness, while C2 is almost constant in both specimen thicknesses. 3) The slope of plastic strain distribution ahead of a crack is unvariable during the propagating of the crack.