TEST OF A CAST-IRON BEAM. 187" 



In order to compare the modulus, as obtained from the 

 cross-breaking tests, with that obtained from direct tension, 

 one of the broken halves of the beam may be put in the 

 machine as a tension specimen, and treated in the usual 

 way. Without giving all the figures in detail, it will le 

 sufficient to quote the results obtained for such a test- 

 applied to the case in question. 



Here, 



area of the section = 2'00 x 1'03 

 = 2'06 sq. in. 



Mean extension ) AATO 



, v = 0'0073 in. 

 per ton ot load j 



The elastic modulus 



x 



2240 10 



2-06 1 * 0-0073 

 = 14,900,000 Ib. per sq. in. 

 = 6,600 tons per sq. in. 



This result agrees fairly well with the modulus as obtained 

 from the bending experiment. 



Care should be taken to discriminate clearly between 

 the maximum stress in a bar of cast iron, as given by a test in 

 direct tension, and the maximum stress as calculated from 

 the breaking load in a cross-breaking test by the ordinary 

 beam formula. For a beam of a perfectly elastic substance 

 the beam formula does truly express the relation existing 

 between the load, the dimensions, and the stress in those 

 fibres furthest away from the neutral surface. But this is 

 true only so long as the beam does retain its elasticity, and 

 as soon as the limit has been passed the formula no longer 

 holds. In the case of cast-iron beams, the elastic state ceases 

 to exist before actual rupture occurs, and the maxi- 

 mum stress, as calculated from the breaking load by 

 the beam formula, does not correspond to the maximum 

 stress, as deduced from a direct-tension test. In fact, these 

 two bear to one another no constant ratio, except that 

 the bending stress is always greater than the tension stress, 

 and less than the compressive stress for the same metal. 

 The precise relation between the two depends very largely 

 upon the form of the cross-section. It is therefore wrong 



