AND THOSE MEASURED WHILE RECEIVING CARGO. 123 



The measured strains represent changes in length of the plates, and it was 

 obvious at the time of their development which they were, whether they were in 

 a tensile or a compressive direction as the gauge successively opened and closed, 

 but pre-existing strains in a finished structure at repose defy definition. In order 

 to ascertain what strains exist in different parts of the ship, of tension and com- 

 pression or zero strains at neutral zones, it is necessary to establish reference 

 lengths initially on the component members and remeasure them at desired stages. 

 It involves considerable work to acquire reliable information upon the distribu- 

 tion of strains in the principal members of a ship when launched, with the ma- 

 chinery installed, with a cargo aboard, and finally the live loads at sea, but it is en- 

 tirely feasible to acquire such data and in fairly complete detail. The method of 

 measured strains devised by the writer, which has been used in the investigation of 

 internal strains in steel forgings since 1887, and more recently in the investigation 

 of strains in engineering structures, admits of furnishing this information. This 

 method of investigating strains was employed on the present occasion in measur- 

 ing the deck plates during the time the ship was receiving cargo. 



The purpose of making the strain gauge measurements at the time of taking 

 cargo aboard was primarily to ascertain what strains resulted directly from loading 

 the ship. Notwithstanding the intent of the observations, the results evidently 

 showed effects which were chiefly due to changes in the temperature of the plates, 

 with obscure effects, if any, attributable to the cargo. With stevedore work going 

 on at the several hatches, few places on the deck were accessible at all times or 

 available even in the early hours of the morning, when most of the measurements 

 were made. The places selected were just aft the bulkhead, as shown on Plate 69. 

 Reference lengths were established, of 20 inches each, on six courses of plating, 

 three on each side of the ship, and one on each bulwark rail. The latter were on 

 the angles directly over the double-riveted butt-strap joints of the bulwarks. All 

 reference lengths were fore and aft in their direction. 



Measurements were inaugurated soon after the ship came out of dry-dock, and 

 were continued until the day of sailing. The diagram, Plate 70, has entered upon it 

 the results of these measurements, also the temperature of the deck at each of the 

 places measured, the hour of the day, and the draught of the ship forward and aft. 



On June 13 the ship was at dock in Brooklyn, with bow heading east, on June 

 15 and 16 at Communipaw with bow toward the west, and on succeeding days at 

 Pier No. 52, North River, with the bow heading east. These changes in position in- 

 troduced differences in thermal conditions which doubtless influenced certain of 

 the measurements. 



On the morning of June 20, sailing day, the temperature of the deck was nearly 

 uniform, the range being from 64° to 67° F. For convenience, the stresses were 

 regarded as having zero value at that time, and the state of strain or strains on 

 other days is given with reference to those on the day of sailing. 



The manner in which the strain gauge readings were taken and results arrived 

 at were as follows : — A reading is taken with the strain gauge on the reference bar, 



