Art. XVI. — Experiments on Model Girders. 

 By W. C. Kernot, M.A., C.E. 



[Read 14th September, 1882.] 



It has been the practice for several years past, in connection 

 with the Engineering School of the University, to try 

 various experiments upon the strength of model girders and 

 frames, representing to scale existing or proposed structures. 

 These experiments have, until recently, been made on 

 wooden or cardboard models only. The desirability of test- 

 ing iron models riveted together in the same manner as the 

 actual structures, has all along been recognised, but practical 

 difficulties in the way of obtaining the proper sections of 

 metal in sufficiently small sizes, have hitherto prevented 

 anything being done in this direction. Recently, however, 

 a quantity of very small angle iron has been obtained, and 

 this has removed the obstacle. 



The first case chosen for experimental investigation was 

 that of a small bridge for pedestrian traffic, crossing the 

 lines of railway in the large goods station at Spencer-street. 

 This structure is as a rule very lightly loaded, it rarely 

 occurring that more than two or three persons are upon it 

 at once, but, at the same time, it is quite within the bounds 

 of possibility for a dense crowd to assemble upon it, testing- 

 its endurance to the utmost, and rendering any weakness a 

 matter of most serious importance. The reasons why this 

 particular bridge was selected were, first, that its construc- 

 tion being comparatively simple, the labour of making the 

 model was but moderate ; and, second, that calculation based 

 upon established dynamical laws led to the conclusion that 

 some parts were considerably deficient in strength, while 

 others were needlessly lavish, both in material and the amount 

 of riveting employed. The former criticism applies to the 

 compression diagonals, which consist merely of thin flat bars, 

 while the latter refers to the end vertical pieces, which are 

 of comparatively masssive and complex construction, con- 

 taining many times the quantity of iron theoretically 

 necessary. 



The model marked A represents accurately to a scale of 

 one-twelfth full size one of the girders of this structure. 

 That marked B is constructed according to an amended 



