CAST-IRON BEAMS. 179 



should be made with the pattern placed in the mould with 

 the greatest depth vertical, and the bar in a horizontal 

 position. After the casting has been taken out of the sand 

 the separate pieces can be cut off, thus forming a beam for 

 a bending test, a tension specimen, and a compression 

 .specimen. 



When a bar is tested under a cross-breaking stress the 

 central load is to be gradually applied until fracture takes 

 place by the tension side failing, and the beam breaking 

 into two parts. At the same time as the load is being 

 increased the deflections must be measured at frequent 

 intervals, the readings being taken more frequently as the 

 point of fracture is approached. These deflections may be 

 most suitably measured directly from a light frame attached 

 to the beam itself, as has been described. 



From the fact that it is unusual to get a number of 

 castings of exactly the same dimensions, and still less 

 frequently of the nominal dimensions, it is the common 

 practice to ascertain the dimensions of the beam, and 

 then to calculate the load which it would have taken to 

 fracture a bar of the nominal dimensions. For instance, if 

 the nominal dimensions of the beam are 2 in. deep and 1 in. 

 wide, and the actual dimensions are found to be 2*04 in. deep 

 and 0'98 in. wide, the result must be reduced to what 

 it would have been if the dimensions had been 2 in. by 

 1 in. This can be done by taking it that the strength 

 varies directly as the width, and directly as the square of 

 the depth. The reason for doing this is obviously in order 

 to preserve uniformity and to help comparison. It is 

 analogous to reducing the breaking loads on tension bars 

 to so many pounds or tops per square inch. An example 

 of a test of this kind is given so as to show the method 

 of working out the results. 



TEST OF A CAST-IRON BEAM. 

 Original Dimensions. Depth, 2'03 in. ; width T04 in. 



LOADS. DEFLECTIONS. 



Lb. In. 



500 0-046 



1000 0-095 



1500 d-150 



2000 0-200 



2500 0-260 



3000 340 



3125 0-300 breaking load 



