It is well-known that the lower the speed of rotation is, the higher the rate of burn off becomes in heating phase under steady state in continuous drive friction welding. This report deals with the effect of welding parameters on the rate of burn off in heating phase. In order to analyze the effect, the axial compressive strain rate of the lead disc, selected as a model of the zone of workpieces being deformed, is investigated. Furthermore, the improvement in the quality of welds is discussed on the basis of the fact that the amount of burn off is increased with the friction torque in the braking time. The results of experiments may be summarized as follows: 1) When a higher heating pressure and/or a lower rotational speed is used, the rate of burn off becomes higher. 2) In the above-mentioned case, the friction torque becomes higher and the zone being deformed becomes thinner. 3) Under the same heating pressure, the rate of burn off per unit thickness of zone being deformed vh increases straightly with an increase of the friction torque. The gradient of the straight line showing the relation between them becomes steeper with an increaes of the heating pressure. 4) From the results of the compression-torsion test of the lead disc, it becomes evident that the axial compressive strain rate of the disc becomes higher as the axial load and/ or the torsional moment increases and the ratio of the diameter to the thickness of the disc D/l decreases. 5) Based on the results of the test by the lead disc, the factors, which increase the rate of burn off Vh a to lower rotational speed, are considered as follows. The friction torque contributes largely to the increase of Vh, whereas the decrease of the thickness of the zone being deformed and the decrease of the axial compressive strain rate dependent on D/l do not so much. 6) The quality of welds is improved by the increment of the amount of burn off, which depends on vh, raised by the increased friction torque in braking time. It is estimated that the increment of the amount of burn off has the same effect as upsetting at the end of heating phase.
This thesis deals with the bending fatigue test results of welded built-up beams of SS 41, SM 50, SM 58, and HT 80, respectively. The fractures of all these beams initiated at the flange-to-web fillet weld under bending, and it took about 10, 000 or 36, 000 cycles of loading from a visible crack to separation of the flange plate. The test results are compared with those of the welded built-up beams and the longitudinal fillet welded specimens obtained by other investigators. The longitudinal fillet welded specimens indicate the same as or a somewhat greater fatigue strength, than the beams in bending.
Recently two main types of high temperature weld cracking have been recognized. One is liquation cracking which is associated with microsegregation leading to intergranular films, the other ductility-clip cracking which occurs at newly migrated grain-boundaries free from films. An investigation has been carried out using a high temperature microscope and an electron probe X-ray microanalyser to obtain a basic understanding of the grain-boundary liquation cracking in the weld heat-affected zone of steels. In this paper, Type 310 and Type 304 stainless steels were used. The main results obtained are as follows. 1) For Type 310 steel which seems to be comparatively susceptible to hot cracking, low-melting point liquation of grain boundary was observed at about 1355°C-1360°C in rapid heating (heating rate about 140°C/sec). This temperature was extremely lower than the solidus temperature (about 1420°C) in rapid heating. 2) It was demonstrated how the phenomenon presented above was caused by segregation of low-melting point compositions and how such grain-boundary segregation, was induced only when both grain-boundary migration accompanied with heating and solute-rich zone were in simultaneous exsistence. 3) It was presumed that the most important factor to cause hot liquation cracking of Type 310 steel was the low-melting point eutectic reaction beween Cr and Ni which had been enriched to grain-boundary simultaneously. 4) The reasons why Type 304 steel is less susceptible to hot cracking in HAZ than Type 310 steel are as follows. a) As a great deal of Cr element dissolves into δ-ferrite, Cr is separated from Ni element, the low-melting eutectic reaction between them is suppressed. b) The grain-boundary migration on heating is apt to be interrupted by δ-ferrite, so it becomes difficult for solute atoms to be swept up to the grain-boundary.
Recently, some brazed parts of mechanical products are used under a conditiom receiving fatigue stress in the mechanical industry. The purpose of this study is to clarify the effect of surface roughness of base metal on the fatigue strength of the brazed joint. The materials used in this ezxperiment are mild steel (SS41) and BAg-8 (Ag 72%, Cu 28%). In this experiment, each brazing surface of base metal was polished with various grade emery papers. (No 60, 120, 240, 400, and 600 mesh) And also, the brazing surface of each specimen was machined by the following processes : (1) Maching by plaim milling cutter (2) Grinding by 60 mesh grinding wheel (3) Grinding by 80 mesh grinding wheel (4) Electrolytic polishing (5) Super fimishing by 240 mesh grinding stone In these specimens, the roughness of base metal machined by milling cutter was very high, its value being 14μ Rmax, and the others were lower, their values being 1.6-0.6μ Rmax. As carried out hithrto, these specimens were brazed with BAg-8 filler metal in the fixture as shown in Photo 1 using the 35 KW resistance furnace in hyrogen atomsphere. Then, these brazed spcimens were machined and ground according to design of Fig I. Fatigue test of these specimens was carried out by a rotary bending machine, and it gave results as follows : (1) In the specimens polished with emery papers, the fatigue strength of specimen polished with 120 mesh emery paper was very high, its value being 17.5 Kg/mm2 and that of one polished with 600 mesh emery paper was lowest its value being 10.6 Kg/mm2. (2) In the case of the machined specimens, the fatigue strength of specimens machined by milling cutter was high, its value being 17.4 Kg/mm2, and those of other specimens machined by various processes were lower, their values being 12.0 Kg/mm2-10. Kg/mm2. (3) From these results, it is clear that the fatigue strength of coarse surface specimens is higher than that of smooth surface specimens. This comes from the difference in the actual brazed surface area of each specimn and other physical properties. And such a tendency of fatigue strength in these specimen; is much the same as tendency of tensile strength in mild steel joints brazed with BAg-8.