AN ANALYSIS OF TESTS OF WATER-TIGHT BULKHEADS. 91 



has been assumed about 8 inches in case of the 12 -inch I stiff eners with 

 face plate, while, for stiffeners with only one flange and one line of rivets 

 on the bulkhead, as in tests 12, 13 and 14, the width is assumed 5 inches. 

 (See Plates 30 and 31.) 



We might expect that single stiffeners would behave differently, 

 according as the pressure was applied on the same side of the bulkhead 

 as the stiffeners or on the opposite side. With the limited number of tests 

 that would bring out this point, and which have been at the disposal of 

 the author, it has not, however, been possible to arrive at any conclusion 

 on this question, especially as some of the results are in this respect 

 contradictory. 



In accordance with the considerations concerning the effect of brackets, 

 which have been given above, the stresses in the stiffeners have been calcu- 

 lated corresponding to the maximum bending moment existing in the free 

 part of the stiff ener between the brackets. This maximum was found, in 

 all the cases here analyzed, near the middle of the stiff ener, being here 

 greater than at the apex of the brackets. 



C. RESULTS OF THE ANALYSIS WITH DISCUSSION. 



Comparison between double and single stiffeners. — As will be seen by 

 an inspection and comparison of the diagrams for double stiffeners given 

 on Plates 28 and 29 and those for single stiffeners given on Plates 30 and 31 

 a very marked difference was found in the behavior of the two kinds of 

 stiffeners. In practically all cases the observed curve of deflection of 

 double stiffeners, tests Nos. i to 6, corresponded fairly with the curve calcu- 

 lated under the assumption that the stiffeners were freely supported at the 

 apices of the brackets, and this in spite of the fact that the brackets, fitted 

 both at the top and the bottom of the stiffeners, were very substantial. 

 In most cases the brackets of the double stiffeners were indeed small, only 

 of 18 inches effective height or ^ to rsh and this might be thought to account 

 for the great deflection, but even in cases where, as in test No. 6, the 

 brackets were replaced by a rigid engine foundation, the deflection was 

 still very considerable and approached that of a freely supported stifi'ener. 



The single stiffeners, where bracketed efficiently, behaved on the other 

 hand as if fixed at the ends, showing, relatively, a much smaller deflection 

 than the double stiffeners. Considering first the tests Nos. 7 to 11, all of 

 which took place in vessels of the United States Navy, the stift'eners are 

 provided with large, very efficient 3-foot brackets, but are otherwise of much 

 lighter construction than the double stiffeners. The weight per foot run of 



