IIW History 1948-1958

- 228 -

culate the maximum allowable load. The delegates were made responsible for calculating this load, either by following absolutely the rules for calculation in use in their respective countries, or by applying these rules and also taking account of the allowable stress laid down in advance for the weld seams ; it was thought that this arrangement would make it possible to separate the influence, on the one hand of the allowable stress and, on the other, of the method of cal– culation. Although, in the second case, the differences observed were less considerable, the results obtained showed remarkable differences as a whole (r ). An attempt to find common rules was therefore considered to be well justified and work to this ernl was immediately undertaken. From the beginning, the t wo following aspects of the problem were consid– ered : - Verification of the adequacy of a weld seam in relation to the stresses to which it is subjected. Calculation of these stresses. Under its second aspect, the problem may in cert ain cases be fairly simply resolved by applying almost directly the usual strength Gf materials methods; in many other cases, it is necessary to use more or less conventional hypotheses on the division of the stresses, either in certain seams or simultaneously in the seams and between certain of them. The first aspect of the problem was considered first because of a request for assistance which was received in this connexion from ISO/TC 44/SCz (Cal– culation of welded assemblies) in 1954. Important contributions to the study of this problem were received from the Dutch delegation which made known the experiments and studies carried out by VAN DER EB, KIST and VREEDEN– BURGH on the strength of weld seams subjected to loading in any direction, as well as the method of calculation proposed as a result of these studies by the « Centrum voor Lastechniek ». The Belgian delegation reported another method of calculation based on the principle of HUBER - voN M:rsES - HENCKEY (z). Finally, the Italian delegation presented a critical report on the two methods mentioned above, together with a proposal for a new method which modified and harmonised the two preceding methods (3) . In this latter document, the piroidal shaped polar diagram, which had been proposed by vAN DER EB, was used as the boundary for the ends of the vectors representing the resistance to fracture of the weld seam subjected to external loading directed according to these vectors ; it was proposed to replace this piroidal shape by an ellipsoidal shape which allowed of the establishment of a formula which was fairly simple and genoral and was reasonably close to the experimental data. The final formula of calculation, which is based on the consideratiom menti– oned above as well as on the H UBER - VoN MISES - H ENCKEY principle, is as follows :

In this formula, cr J_ and cr 11

are the stresses normal to the throat section and

to the cross section respectively of a weld seam, while -. 1- and -. 11 are the tangential stresses in the throat section, in the transverse and longitudinal directions res– pectively.

(1 ) Document X V-7-53 . (2) Document XV-24-54. (3) Document XV-13-54 .