Foreword By way of recapitulation (see these Proceedings, vol. 38 (1962)r p. 520), we express the critical stress state of breaking section in material by the distance pN of a tangent line of the stress circle from the origin of the diagram reading as -(a l-a3) T -(c1+ ~'3)C=pN, 2 2 where a1, a3=the principal stresses, maximum and minimum, C = cos cp with cp denoting the angle made by pH to the axis of ar pN = ao -aR + bM, ao representing the initial strength, and R, M the greatest

more »

... ences, vibratory and static r respectively. As applied to static strengths, aR vanishes so that pN = ao + b(a1--a3), bM being solely responsible to affect ao proper to the given materiall in one or the other way according to the kinds of stress. For uni-axial tension we put a1= aB and a3 = 0, aB being the breaking tensile stress acting on the fractured surface, which can be specified by C=1, and for torsion a1= -a3=zb, this being defined as it is. Since C drops out in the second case, the original formula becomes in these cases ao=aB (1-b)=zb (1-2b'), b' being written for b in the last expression to denote a negative number, because zb falls below o by yielding in contrast to aB rising up. As for ao we have ao = a1= a2 = a3 in the critical state under uniform tension. But this test appears less feasible, while ao can be estimated conveniently from the vibratory strengths by making a certain assumption (loc, cit., p. 524). It is possible in this way to compute ao and determine the relations of aB and zb to ao by adjusting b and b'. As a result of such calculation it is made known that b is some 0.33 or simply 1/3 in steels undergoing ductile fracture in. tension. To approach b' without knowing zb exactly we may resort to the ratio zb : aB, which is thought not less than 1/2, since we have then -2b'1-2b, <_ i.e. -b'1/6 _< for b=1/3. At any rate b and b' as such result from plastic strain in yielding. In case of no yielding b = b' = 0 corresponding to that ao = aB = zb.