The purpose of this report is to review critically the available information on structural failures in welded ships.

Although buckling failures in a few naval vessels and several transversely framed European tankers have been reported, this report will review failures from the brittle fracture point of view.

Welded ship fractures were of the brittle-cleavage type and usually propagated at high velocity. There was no evidence of fatigue although high, local cyclic stresses in some cases undoubtedly contributed to crack initiation. The loud noise accompanying extensive fracture indicated the instantaneous release of a large amount of energy.

As far as the engineer is concerned the basis of any tensile strength criterion of steel in a structure must be its ability to resist brittle-cleavage type fractures. If there can be no brittle failure, the structure will not fail in tension under service loading. Specifically, the engineer wants to know:

1. The conditions under which a brittle-cleavage crack will start.
2. The conditions under which a brittle-cleavage crack will propagate.

In order to find realistic answers to the above questions, considerable weight must be given to the actual service performance of structures, and all research must be related to it.

Although the present ship structure research was prompted by welded ship failures, it should be remembered that riveted ships are not immune to fractures. Since 1900, over a dozen riveted merchant ships have broken in two during heavy weather or are listed as missing. It is significant that most of these vessels were of the tanker type, the same type that has given the most trouble as far as welded ships are concerned. Several of the riveted tankers which broke in two were said to have been heavily loaded amidships. This was also the most prevalent loading condition when serious failures occurred in welded tankers. In most cases, failure occurred in riveted ships when the ships were less than 10 years old: in a few cases, however, the ships were over 20 years old.

Sizable cracks developed in several large passenger liners. Both the Leviathan and Majestic experienced cracks in upper strength decks, the cracks starting at square uptake openings and extending to the side shell. Some of the breaks extended down the shell. In at least one case a loud report was heard when the structure gave way indicating that the fracture was probably of the brittle-cleavage type. The Europa had deck cracks starting from square uptake openings and sheer strake cracks starting at airports.

Minor fractures have occurred in riveted ships, and frequent mention of cracks at hatch corners, bulwarks. etc., has appeared in the technical literature. It is probable that many of these cracks were of the brittle-cleavage type rather than the fatigue type as was generally suspected in earlier years. It is not unlikely that some of the serious riveted ship failures may have been associated with plate buckling.