224 KENNEDY'S AUTOGRAPHIC APPARATUS. 



specimen, and the elastic stretch of spring piece being pro- 

 portional to the load upon it, it follows that this stretch is 

 a direct measure of the load upon the specimen. This fact 

 is taken advantage of in the working of the apparatus. 



The spring piece is made of such dimensions and 

 material that the maximum loads on the specimens which 

 are tested never approach its elastic limit. The pointer 

 G, which is used to draw the diagram, is carried by a small 

 roller forming its centre of rotation. This roller lies 

 between and is pressed upon by two plates, one of which 

 is attached to each end of the spring piece. As 

 the spring piece stretches under the load the two 

 plates move with respect to one another, with the result 

 that the roller is rotated, and with it the pointer. 

 The rotation of the index, and therefore the circum- 

 ferential movement of its point, as it traces out the diagram 

 on the glass, is a measure of the stretch of the spring piece. 

 This is a measure of the load upon the spring piece, and 

 therefore of the load upon the specimen also. A large 

 magnification of the stretch of the spring piece is obtained 

 by the large ratio which the arm of the pointer bears to 

 the radius of its roller. By carefully selecting the spring 

 piece, the load can be made exactly proportional to the 

 circumferential movement of the pointer, and by direct 

 calibration with actual loads in the testing machine, the 

 movement of the pointer which corresponds to given load 

 can be ascertained. 



It is only necessary, then, after adjusting the apparatus, 

 to go on increasing the load upon the specimen until 

 fracture takes place, and the whole history of the process, 

 as regards relation of stretch to load, will be told upon the 

 smoked glass by the specimen itself. When the diagram 

 is completed it is varnished and used as a negative, from 

 which photographic prints may be taVen. 



This instrument is truly automatic, and is quite inde- 

 pendent of any outside manipulation, but at the same 

 time it possesses several disadvantages which militate 

 against its general use. One of these is that the load 

 movement of the pointer is in a circumferential direction 

 instead of being in a straight line at right angles to the 

 direction of the stretch movement. For many purposes 

 this matters little, but where the diagrams have to be com- 

 pared with the diagrams plotted from rectangular co- 

 ordinates a somewhat tedious process of reduction is 

 necessary. 



