The thermal cycle and microstructure in each portion of the weld heat-affected zone of a steel are examined. The CCT diagrams in the case of the max. heating temp. 1050°C for two steels, which one contain 0.27%C, 1.05%Mn, 1.3%Si, 0.3%Cr, 0.27%Ni, 0.20%Mo and the other 0.31%C, 0.63%Mn, 0.25%Si, 2.0%Cr, 0.5%Ni, 0.3%Mo, are plotted. The microstructure and hardness in the weld heat-affected zone are in a good accordance with those of the specimen subjected to the thermal cycle of the same max. heating temp. as above (1050°C). It seems to be reasonable to use CCT diagrams for the development of the welding technology of the low alloy and alloy steels and to be effective for the choice of the adequate welding conditions.
The two-dimensional temperature distribution in the longitudinal middle section of welding bead under quasi-stationary state during welding can be approximately represented by the following equation referring to the rectangular coordinate (x, y), the origin of which is moving with the heat source towards the positive direction of x-axis, u=u*+(u0-u*)eh2a2/υ0b xcosμ0(1-y/b), where u0 is the temperature of welding heat source ; u*, an imaginary temperature at which the temperature of bead is considered to be balanced with that of base metal while the bead is cooling, and this is to be determined experimentally ; b is thickness of bead, that is, distance from the deepest bottom to the top of the bead surface ; ν0, welding velocity ; h2, thermal diffusivity ; α2, coefficient of heat transfer from bead to base metal ; and μ0, minimum positive root of the equation cotμ=μ/α2b, By this equation the isotherm of the bead may be found as follows : x=υ0/2h2(b-y)2+ξ ; ξ=-υ0b/h2a2log(u0-u*/u-u*). Further, the formula which represents the shape of columnar crystal found in the longitudinal middle section of bead is obtained as follows by calculating the orthogonal trajectory of the melting-point isotherm shifting with time : x=-h2/υ0log(1-y/b). This describes the shape of practical columnar crystal with near precision and clarifies the following main properties. The higher the welding speed, the larger the thickness of bead and the smaller the thermal diffusivity of weld metal, the nearer the inclination of columnar crystal will be to perpendicular position.
The fish-bone type welding crack test which had been introduced by Houldcroft was applied to study the weldability of casting aluminum alloys of Japanese specifications (JIS) (Table 1). The results are shown in Fig. 4 & 5. Some subsidiary experiments were made in order to investigate the effect of welding condition and the mechanism of crack in addition to the micrographic observation. The conclusions are as follows : 1) The crack length in this test largely depends on the dimensions of the specimen. 2) The slots of this specimen are not always effective to stop the propagation of crack but are sometimes effective to promote it. 3) Not only hot cracking but also sub-solidus cracking seems to be caused. 4) The crack length is not much affected by the welding current in applicable range, but a little increases with the increase of welding speed. 5) Almost always the Si content above 3% in weld mdtal prevents the welding crack except in 7B (11% Mg), in which a bead edge crack happend. The Cu content 3-4% is not so harmful to crack sensitivity. In Al-Mg alloys the crack tendency is predominant but is comparatively small in the range of 4-8% Mg.
In the previous report, the author proposed the value of the deoxidation tendency (R1+1/2R2-A) calculated from the chemical composition of fluxe and proved the relation between its value and the oxygen contents in weld metal. In this report, further, the relation was investigated about several types of electrodes produced experimentally. The relation between (R1+1/2R2-A) and [O] is different from each other depending on the type of coating, as the basicity of slag is different. In electrodes used Fe-Si as deoxidizer, the relation between (R1+1/2R2-A) and [O] is different from the relation in electrodes used Fe-Mn in the cases of acidic types. The cause of which is presumed as follows: the reaction velocity of Si-deoxidation is slower than that of Mn-deoxidation, and weld metals are solidified before when the reaction rea-ches to equilibrium state. But that relation is not different in the case of basic type, because the reactions of Si-deoxidation and Mn-deoxidation are fast in the case of basic type. In spite of this experiment the consideration about the kind of deoxidizer is not necessary, because Fe-Mn is, only, used in acidic types electrode on the market. Therefore, it is able to presume the oxygen contents in weld metals from the chemical composition of fluxes. These considerations are based on the experimental values of Km, and Ksi, which shows the distribution of Mn and Si between slag and weld metal, obtained in this experiment. KMn and Ksi in basic type (low hydrogen type) electrodes are different from these in acidic types (the other types) electrodes. In acidic types, Si contents in weld metals depend on the basicity of slags (SiO2 contents in slags), so oxygen contents are lower in the cases of strong acidic slags than nearly neutral slags. Next, KSi in electrodes used Fe-Si are different from the one in electrodes used Fe-Mn in the case of acidic types, but no difference in the case of basic type. After these experiments, it is proved that the slag-metal interactions affect extremely to the deoxidation reactions during welding process.
Mechanical and corrosion properties were investigated of welded joints of pure zirconium plates, 1 and 2 mm thick, which were welded in a controlled atmosphere welding chamber filled with either pure argon or impure argon mixed with an impurity gas of air, nitrogen, oxygen or hydrogen. Effects of surface cleaning and impurity in argon atmosphere on mechanical properties and cor-rosion resistance against acids at room temperature and high temperature high pressure water were studied and the following conclusions were obtained : (1) It is necessary to pickle (HF 5%+HNO3 45%+H2O 50%) the zirconium specimen prior to welding. (2) Microstructure of weld metal shows α' (acicular α) and air, N2 and H2 in argon atmosphere increase the amount of α'. (3) Tensile fracture of welded joint occurs at the most softened zone in base metal. (4) The amount of absorbed nitrogen or oxygen in weld metal is proportional to its initial par-tial pressure in argon atmosphere. (5) The effects of impurity gases in argon atmosphere on mechanical and chemical properties of weld metal are as follows : (a) More than 200 ppm (vol.) of air or N2 decreases greatly the corrosion resistance against high temperature high pressure water. (b) More than 700-1, 000 ppm of each gas (air, N2 or O2) embrittles the mechanical properties, and hardens the weld metal. However, H2 does not change the hardness and tensile strength although it embrittles the weld metal. (c) More than 105 ppm of each gas decreases the corrosion resistance against HCl+Iron solution. (d) Up to 104 ppm of each gas does not decrease the corrosion resistance against 81% H2SO4. (6) It is necessary to pickle the weld zone after welding in order to improve high temperature high pressure corrosion resistance. (7) As far as the controlled atmosphere welding is concerned, it seems to be necessary for reactor purposes to use an initial vacuum less than 10-2 mmHg and filling with high pure grade (99.99%) argon to one atm. For chemical industry purposes the chamber should be initially evacuated less than 0.5mmHg and filled with same high pure grade (99.99%) argon.