OF SHIP CONSTRUCTION. 65 



if heavily loaded, it encounters a seaway. Immediate failure would perhaps not re- 

 sult, but the riveting would gradually give away, and troublesome leaks would 

 develop. 



Failures due to the three principal transverse stresses are indicated in Figs, i 

 to 5, inclusive (Plates 36 and 37).* 



DIAGONAL DISTORTION. 



A distortion of the hull in the manner and because of the stresses indicated in 

 Figs. I and 2 (Plate 36) would evidently mean a failure of the transverse framing 

 in the vicinity of the bilge, at or near the connection of the second deck to the side 

 of the ship, and possibly, but not necessarily, at or near the connection of the upper 

 deck to the side of the ship. Considering these in the order named, first for the 

 transverse vessel and then for the Isherwood vessel, we have : — 



The Bilge Connection. — From an inspection of the midship section we can con- 

 clude that the weakest point in the bilge connection is either through the riveting 

 connecting the side frame to the bilge bracket, or through the frame itself, in line 

 with two of the connecting rivets. The failure of the framing would in either case 

 be accompanied by a failure of the shell plating, so that we should include, as con- 

 tributing to the strength of the framing, a strip of the shell plating equal in width 

 (i. e., in the longitudinal direction) to two and one-half times the width of the shell 

 flange of the frame bar, reduced by one-half the diameter of the shell rivet in order 

 to obtain a result that will represent a mean between tension and compression 

 conditions.! ' 



To determine the strength of the riveted connection the polar moment of in- 

 ertia about the center of strength of the section must be found. This is most con- 

 veniently done by first finding the center of gravity of the connection and then cal- 

 culating the moments of inertia about vertical and horizontal axes through the cen- 

 ter of gravity. The sum of these two moments of inertia equals the desired value 

 of the polar moment of inertia. 



Taking the shearing strength of the rivets as equal to the tensile and compres- 

 sive strength of the steel, the moment of inertia about the horizontal axis is found 

 to be 577.5 inches*, and about the vertical axis 157.5 inches^ from which the polar 

 moment of inertia equals 735 inches*. If we regard the distance of the center of 

 the farthest rivet from the center of gravity of the connection (or, more correctly. 



*These sketches are borrowed from Holmes' Practical Shipbuilding. 



fThe width of the strip of shell plating that may be regarded as contributing, to the full extent of the 

 tensile strength of the material, to the strength of the frame bar or stiflPener is variously given as a function 

 of the thickness of the plating and as a function of the width of the connecting flange. The conservative values 

 here used, namely, two and one-half times the flange width for single bars and one and one-half times the sum 

 of the flange widths for double bars, do an injustice to the Isherwood ship, inasmuch as the width of the 

 shell strip included for the transversely framed vessel is 8.2S inches for each frame space, or 31.7 per cent of 

 the total shell; while for the double angle connection of the Isherwood vessel the width of the shell strip 

 becomes 16.2S inches for each transverse, or about 11.3 per cent of the total shell. It is difficult to conceive, 

 however, of a mode of reasoning that would warrant considering equal proportions of the shell as contributing 

 to the strength of the framing. 



