CHAPTER VII. 



Q/Oni i!iJlJ~eJ5 iOiljIy ,T>>J 0~J .'.'3 '.'.J.' [ CO.Z d 



.,"oi;aJi, ad.j^oiJvA : uno/ Boiiiiig^ii * i;i v.; 8 j.jJ7.i, 

 TESTING OPERATIONS (Continued.) 



APPLIANCES FOR DRAWING AUTOGRAPHIC 

 DIAGRAMS. 



123. .In the examples of tests which have been quoted 

 that is to say, the tests in tension, compression, bending, and 

 torsion it has been shown that the results of these tests 

 may be recorded or, expressed graphically as well as by 

 means of written figures. This is done in most cases by 

 making use of rectangular co-ordinates, and plotting a 

 curve whose abscissae represent loads, and whose corres- 

 ponding ordinates rejfresent to some scale the strains of 

 the test specimen produced by the loads in question. A 

 curve of this kind has been called a load-strain diagram, 

 and forms a very ready means of recording a test in a form 

 which conveys to the mind an idea of the test much more 

 rapidly than by inspecting columns of figures. 



In these examples the stresses and their corresponding 

 strains were measured at frequent intervals, and plotted 

 afterwards. Attempts have been made, with more or less 

 success, to provide means whereby the diagram can be 

 drawn, as the test proceeds, by an appliance attached to the 

 specimen itself. A diagram drawn in this way is called an 

 "autographic" or "automatic" diagram. 



It is exceedingly difficult to design a really satisfactory 

 and accurate autographic apparatus. Let us see what 

 are the principal conditions which have to be satisfied 

 in doing this. First, take a simple tension test. It is 

 required to be able to draw a curve whose abscissae repre- 

 sent, say, the loads on the specimen, and whose ordinates 

 represent the extensions produced by the corresponding 

 stresses. It is the usual, but not invariable, custom to 

 arrange the appliance so that the diagram is drawn upon 

 a sheet of paper. In all cases the pencil which is to trace 

 out the curve must have two independent movements with 

 respect to the paper, one controlled by the load and the 

 other by the strain of the specimen. Generally, these 

 two movements are at right angles to one another. 



