174 NOTES ON CHAIN CABLES. 



In reference to the fender chain for the canal locks at the Isthmus, it was recommended 

 that it be annealed if perchance it was called upon to stop a ship and thereby subjected 

 to severe straining — to anneal after each such occasion. 



In order to assure ourselves that good welds have been made in the chain, proving 

 stresses are necessary, and to be efficient they should be somewhat above the working 

 stresses. Furthermore, the proof stresses should be sustained by the chain a reasonable 

 length of time, long enough to exhaust about all the stretch of the links which the proof 

 stress is capable of causing. In the subsequent use of the cable the links will not assume 

 the same positions as in the proof tests, but will act upon each other differently, in conse- 

 quence of which there may be an increased stretch, even if the approach to the proving 

 stress were not very close. But the only assurance we have of good welded links is found 

 in the results of the proof test. 



In the making of welds, if we could get the material into close contact at a suitable 

 temperature and some mechanical work given the iron at that time, an excellent weld would 

 result, and possibly the junction of the welded parts might not be detected even microscop- 

 ically. Such a weld was in fact made in some experimental work with a mild grade of 

 steel. The welded surfaces were brought together with a rubbing movement, one side 

 against the other, and so well were the surfaces united that well-defined crystals spanned 

 the junction, as found upon subsequent examination of a polished and etched surface. 



A bodily sliding or rubbing movement seems to facilitate welding, as many short and 

 slightly oblique butt welds have shown. With long scarfed ends, overlapped and ham- 

 mered down, there is danger of enclosing cinder or in closing down upon each other oxi- 

 dized surfaces which will not weld. 



Chief Constructor Richard M. Watt, Vice-President: — I simply want to add to 

 this paper the fact that it represents several years' exhaustive study on this subject, and that 

 in his capacity as shop superintendent of the Hull Division of the Navy Yard at Boston, Mr. 

 Otterson has markedly improved the output of chain at that yard, so that his efforts have 

 resulted in the saving of many thousands of dollars to the government. I believe that that 

 attention, which has so increased the hand output, will in time develop an entirely satis- 

 factory method or process of power manufacture of chain. 



Mr. Francis T. Bowles, Past President: — No doubt we are under many obligations to 

 the author of this paper for the information he has given us, and I do not want to appear as 

 too discriminating when I suggest that notes on the subject of chain cable are absolutely in- 

 complete without any reference to the subject of swivels and shackles. My own interest in 

 the subject of chain cable is entirely connected with operating troubles with shackles and 

 swivels which cannot be properly described in polite language. 



Mr. Robert Haig, Member (Communicated) : — With regard to the statement made on 

 page 158 of Mr. Otterson's paper it is hardly correct to state that "Lloyd's will not accept 

 chain made by this process;" the real facts are that as test links must be cut out of every IS 

 fathoms the makers have not yet been able to put in a joining link with the machine as it is 

 found difficult to bunch the two ends of the chain on to the one link, but while this no doubt 

 will yet be accomplished, at present joining links have to be put in by hand, and as this 

 brings the strength of the chain down to that of iron hand-made chain the value of the me- 

 chanical process is somewhat neutralized. 



