124 STRAINS IN THE HULL OF A SHIP AT SEA, 



followed by a reading- with the gauge on the deck plate. The distance or gauged 

 length laid off on the deck is thus determined with reference to that laid off on the 

 reference bar, the strain gauge being used as a transfer instrument to compare 

 one gauged length with another. The reference bar is assumed to have a length of 

 exactly 20 inches. The temperature of the reference bar and that of the deck is 

 noted. If they are not the same the reading for the deck length is corrected to cor- 

 respond with that of the reference bar. The latter is made of structural steel or 

 steel of that grade and assumed to have the same coefficient of expansion as the 

 deck plates. 



Assuming the strains to have zero value on the deck plates on some given day 

 for comparison, on this occasion taking the lengths as they were measured on the 

 20th of June, the individual lengths are said to be in tension or compression on 

 other occasions according to their corrected lengths, that is, according to whether 

 they were longer or shorter than on the day selected for comparison. Referring to 

 examples on diagram, Plate 70, the corrected gauged length of stringer plate 

 No. 10, port side, on June 13 was .0019 inch shorter than it was on the 20th. 

 The force required to shorten the plate this amount is 2,850 pounds per square inch 

 compression. On June 16 the same gauged length, corrected, measured the same 

 as on the day selected for comparison. It therefore had zero stress upon it with 

 reference to its state on June 20. Again the gauged length was .0008 inch longer on 

 the 1 8th than on the day selected for comparison, which strain corresponds to a 

 stress of 1,200 pounds per square inch tension. 



The results exhibited on diagram, Plate 70, were deduced in the manner de- 

 scribed. Where stresses of tension are shown on the diagram the corrected gauged 

 length on the deck was longer than it was found on the day selected for compari- 

 son, namely, June 20. Where stresses of compression are shown the gauged length 

 was correspondingly shorter. 



There was no change in the cargo between the time of measurements of June 

 15 and 16, but a wide difference in temperature, accompanied by a wide difference 

 in the stresses. The starboard stringer plate passed through a range from 2,550 

 pounds per square inch compression to 1,050 pounds per square inch tension, a total 

 of 3,600 pounds per square inch. This and other changes which occurred between 

 those dates were clearly temperature effects. 



Significance is attached to the similarity between the shaded parts of the dia- 

 gram representing the stresses in the stringer plates and bulwark rails on June 13 

 and June 15, comparing the port side on the first day with the starboard side on 

 the second day. The ship was moved between those dates from Brooklyn to Com- 

 munipaw, and turned half around. Both days were sunny, and it would seem that 

 the effect of the sun on the port side when at Brooklyn was closely reproduced on 

 the starboard side when at Communipaw. 



The state of the strains on the mornings of June 16, 19 and 20 was nearly the 

 same. In the meantime, cargo was received which increased the draught from 12 

 feet 6 inches to 26 feet 6 inches forward, and 19 feet 6 inches to 27 feet 6 inches aft. 



