In order to make clear the effect of stress relief heat treatment on the reduction of residual stress in base metal, high temperature tensile test, creep test and tensile relaxation test, both anisothermal and isothermal, were carried out using two kinds of HT80 steels. Relaxation property was discussed on the basis of creep phenomena and a method of forecasting stress drop during S.R. treatment was studied. The findings are summarized as follows: 1. In the anisothermal stress relaxation test, stress drop during heating was larger than that predicted from high temperature tensile properties. This fact points to the occurrence of creep strain during heating. this creep strain became larger with increasing initial stress and with temperature rise. 2. In the isothermal stress relaxation test, stress drop occurred mainly in the initial 1 hour from beginning. The isothermal stress relaxation curve could be calculated from creep test data on the application of strain hardening concept. 3. A higher heating rate during anisothermal stress relaxation test led to an increase in residual stress immediatelly after heating, but this effect became almost negligible as holding continued more than 1 hour and then the stress decreased to a certain value regardless of heating rate.
A study has been made on the reaction for dissolution of solid iron in liquid copper-iron solutions of various concentrations and on the migration of molten copper into solid iron. Experiments were carried out; cylindrical samples of base metal iron were immersed in liquid copper-iron alloys with various iron concentrations and then rotated at given revolutions in argon atmosphere in a resistance furnace, and the cylinders of the solid were padded with a small quantity of copper and this copper was melted under vacuum in an induction furnace. The rate of dissolution of solid iron into liquid copper-iron alloy is dependent on the rotational speed, the temperature and the iron concentration. The reaction process was found to be rate-controlled by the diffusion of solute from the iron-copper interface to the liquid and the obtained value of activation energy for dissolution of solid iron in liquid copper-iron alloy was about 14.3 K calf mole. At higher velocities and higher temperatures, however, the transition from pure diffusion control to mixed control is likely to occur with an increasing solute concentration. On the other hand, small amounts of the molten copper migrate through the copper-iron interface into the solid iron and a thin laver of the base metal adjacent to the interface appears to be saturated with the molten copper. The principal mechanism of the transfer of molten copper into the solid iron was shown to be by volume diffusion and by grain boundary diffusion, without liquid metal grain-boundary penetration. The mechanism of the migration was considered variable with the interfacial energy, the grain-size and the temperature. The structure of the solidified iron-copper interface was investigated by means of metallography.
In the previous works, the authors pointed out that the change of cathode mode of a graphite electrode could be explained rationally by considering the energy balance and ionization process in the cathode fall region. There are still left, however, some difficulties, in a carbon arc, to clarify the cathode mechanism, since there is evaporation of carbon out of electrodes and, consequently, the composition of a plasma at each condition is not consistent. Therefore, systematic researches on low pressure arcs were made with nonconsumable electrodes, such as a thoriated tungsten cathode.and a water-cooled copper anode, to eliminate the influence of vapours from electrodes. Experimental conditions were: the current range of 20-100 A, and the pressure of 20-100 Torr in a pure hydrogen atmosphere. A tungsten cathode did not show the cold cathode mode which was recognized with a graphite cathode, but it took the modes of the plasma cathode and the hot one under this experimental conditions. In an arc with a tungsten cathode, however, the plasma cathode mode clearly appeared in wide ranges of arc currents and pressures, and the mode transition was more exactly observed than in a carbon arc. The anode mode, which was not described in the previous papers, was systematically studied. Under the conditions of low pressure and small current, a glow-like discharge with light violet colour emerged on a water cooled copper anode, and this anode glow gave eminent effects on electrical and thermal characteristics of an arc. Thus, two kinds of anode modes, i.e. the normal and the glow modes, were recognized in low pressure arcs with a copper anode. The anode drop VA in the range of normal anode mode gradually increased, particularly it showed remarkable increase when the current or the pressure came near to the critical conditions of the mode transition. When the glow anode mode emerged, anode energy VA*, as well as arc voltage VA, increased abruptly by 10 volts or so, and this change in VA* and VA was due to the increase of the anode fall VA. Governing factors of the anode modes and their correlations were clarified. They are (a) arc current I, (b) ambient gas pressure p, (3) kinds of atmosphere and (d) anode material. These facts were very similar to the governing factors of the cathode modes and the correlations among factors.
The effect of deposit metals at the root of a collar on the fatigue strength was studied in order to improve the life of a repaired crankshaft. A round fatigue test specimen with a collar was used and its root was deposited by TIG welding with different filler metals. 0.45% carbon steel was used as the base metal. 20Cr-lONi and 25Cr-2ONi stainless steels, nickel, monel metal and brass were applied as the filler wires. The following experimental results were obtained: (1) The fatigue strength is increased by the deposit of 20Cr-lONi steel, but is not increased by the deposit of the other filler metals. (2) The effect of the deposit of 20Cr-lONi steel wire is more remarkable, when the fillet radius becomes smaller. (3) The thickness of deposit layer has little effect on the fatigue strength.