Search

With utilization of penetration mechanism in electron-beam welding, an equaption which estimates the penetration depth in bead by calculation is induced from thermal conduction theory in this report.Moreover, to examine the valiuity of the equation, each penetration depth which is estimated by the equation is compared and discussed with the actual depth in numerous beads under the same welding conditions.As a result of this discussion, it becomes apparent that the induced equation for calculation of the penetration depth gives successfully the penetration depth in actual bead.Therefore, if the welding condition and physical constants for the used material are known, the penetration depth in actual bead can be easily estimated by simple calculation using the above equation.

Defining new dimensionless parameters β, in, jn, k and functions Fon, F∞n, the temperature distribution of welded plates under unstationary states, which was obtained in the authors' preceding paper is more simply shown in equations (2) to (5).1)2)In equations (6) to (9) the cooling rate of welded plates is shown.The cooling rate passing through a constant temperature, which is very large at the start of the welding, rapidly decreases during welding, and increases again near the finishing point of the welding (crater).The rapid cooling at the both ends of weld bead is due to the unstationarity of the temperature distribution.The effect of the welding conditions on the unstationarity of the cooling rate is discussed. Figs 6 to 9.

Occurring and developing mechanism of longitudinal hot cracking due to rotational deformation of specimen of thin aluminum plate is studied experimentally based on the results in previous report.Fursthermore, a new method of assessment of hot cracking susceptibility is suggested basedon the above experimental results and modification of Houldcroft hot cracking test.It is observed that in welding a rectangular specimen starting rom edge, longitudinal cracking occurs in weld bead and when the width of specimen is increased, the cracking ratio changes abruptly from 100% to 0% at a certain width of specimen-critical width-in high speed welding, but at low welding speed, the cracking ratio decreases gradually from 100% and drops to 0%. In these cases, it is observed that the cracking occurs in the brittle area behind the molten pool in the start part of weld bead and it develops from A to B in Fig. 2 with the molten pool.The above mentioned results may be understood from the existence of minimum deformation -l for thecracking (which is described in 1st report) and the actual deformation. The latter is affected bythe rigidity of the plate and the temperature distribution in the width direction.From above results, it is concluded that the factor which stops the cracking in Houldcroft type cracking test is uniformity of temperature distribution in the width direction due to the decrease of effective width by a slit inserted. On the other hand, there is another tendency of the development of cracking being accelerated by the decrease of rigidity due to the slit. Accordingly, it is understood that Houldcroft type cracking test is an obscure one with an inconsistent factor in stopping mechanism of cracking. From above results, a new improved hot cracking test is suggested.

As the most fundamental case of shrinkage distortion in welded joints, we have studied on the transverse shrinkage distortion of bead-on plates. Characteristics of transverse. shrinkage distortion are shown in Fig. 4. Both angular distortion and shrinkage produce near the weld bead, and the distorted portion correspondes to the range heated above 150-200°Cwhen welding (see Fig. 5). Comparing the results of Fig 1, 2 and 3, we know the shrinkage distortion appears approximately uniform except for the ends of weld.To analyse the deformation of bead-on plate we assumed that shrinkage forces and shrinkage moments act as shown in Fig. 6, and inherent strain, the cause of distortion, distribute as shown in Fig. 7. Transverse shrinkage and angular change are then represented by eqs. (1) and (2). Fig. 8 shows the characteristics of these equations.Simple formulas of transverse shrinkage and angular change are obtained from Fig. 8 and the results of our former study (see Reference 1), and they are eqs. (6) and (7). Experimental results are shown in Fig. 9 and 10.

It is generally known that the majority of fillet weld cracks are toe cracks, underbead cracks and so on. But we have found an unusual crack in the.fillet welds of 50 kg/mm2 grade high strength steels. The crack occurred at the heel-opposite side of the toe-of fillet weld. Therefore we refer to them by the name of "Heel-Crack" in this paper. We have carried out various experiments to make clear the characteristics of the crack and the causes of its formation, and established a practical preventive method for the crack based on these test results. The test results are summarized as follows:(1) Heel-cracks are located at the heel of fillt welds and they are hardly detectable by inspection methods such as visual observation and magnetic particle inspection.(2) It has been difficult to reproduce the heel-cracks by well-known weld cracking tests such as C.T.S. test and Reeve cracking test. We have developed a Non-restraint T type Cracking Test which can reproduce heel-cracks successfully.(3) Heel-crack is a cold crack which occurs at temperatures below 100°C.(4) Main factors which affect the heel-crack formation are the cooling process of the weld, rigidity and hydrogen in the weld.(5) Many 50 kg/mm2 grade high strength steels are susceptible to the formation of heel-cracks. But semikilled steels seems somewhat less susceptible to heel-cracks than killed steels.(6) Crack sensitivity is increased by increasing the carbon equivalent and/or the maximum hardness at underbead.(7) Heel-crack could be prevented by preheating or controlling the bead length. To prevent the heelcrack, higher preheating temperature and longer weld bead are desirable.