160 CORRESPONDENCE. 



Later on Mr. Haycroft states that he has developed "the true 

 maxima stresses in the several parts of the truss, and finds them 

 differ considerably from those given in the paper." This appears 

 at first sight as if Mr. Haycroft considered the stresses in the 

 paper were incorrect, but must be taken in conjunction with the 

 above " I think " and his statement later on that — " In Y 3 and Y s 

 the stress is given as 21 tons, although I think it will be found no 

 stress can exist in these members from the class of loading under 

 consideration. These braces, however, should not, I think, be 

 omitted, as they preserve uniformity in appearance, and, if 

 properly utilised, are a source of strength to braces Y 11 and Y 6 ." 

 It appears to me strange for an engineer to come forward at a 

 meeting like the present and make vague charges of error in a 

 calculation which he does not prove ; and to recommend the put- 

 ting in of struts which he states carry no stress so that they may 

 be ' ; a source of strength to braces Y xx and Y 6 ." The calculation 

 of stresses in redundant structures is so well understood that it 

 is needless for me to say much further than has already been 

 given in the paper, but I would point out that the live and dead 

 loads combined at apices 1 and 2 produce a certain reaction on 

 the right abutment, that this reaction is taken by the chord hori- 

 zontally and by an imaginary diagonal from right abutment to 

 the intersection of Z z and V 2 . The amount of the stress on 

 this diagonal can then be resolved along Z \ and ]T 3 , and if this 

 simple principle of a triangle of forces be adopted, it will be seen 

 at once that there is a stress of 21*19 tons on bar Y z . In a 

 similar manner the dead loads on apices 3, 4, 5, 6 and 7 produce 

 a stress of 24'38 tons on Y 11 . The two systems of diagonals 

 in this case are entirely independent, and the strut Y 3 can in no 

 way relieve the stress on 7 n . Mr. Haycroft's surmise in refer- 

 ence to the concentrated rolling load is, to say the least, strange ; 

 however, as he may mean it seriously, I will explain how this 

 load has been taken. The traction engine is assumed to weigh 

 in steam 16 tons, of which 9^ tons are on the driving wheels, 

 and 6^ tons on the trailing wheels. The wheels are 1' 4" wide, 

 and 5' 0" centres, the distance between centres of driving and 

 trailing wheels being 10' 4". The maximum stress in one truss 

 occurs when the traction engine is close up to the kerb, and the 

 driving wheels are over a cross girder. The total load at this apex 

 is then 9.86 tons due to both driving and trailing wheels. (The 

 detail calculation being so simple is omitted). Mr. Haycroft 

 states that he does not consider it good practice to design all the 

 suspension bolts the same size. Of course this is a mere state- 

 ment of opinion on his part, and as the additional cost for surplus 

 material is only <£15 per truss, the advantage of uniformity more 

 than compensates for this waste. 



