382 SECTIONAL TRANSACTIONS.— G. 



amount of stress per unit load and distribution of stress through structure 



Suggested testing to destruction of full-size bridges. 



Rivets. — Need for information as to (a) ultimate strength of rivet, (6) dis- 

 tribution of stress in riveted girders. Bridge Problems arising out of railway 

 amalgamations. 



{d) Mr. J. S. Wilson and Prof. B. P. Haigh.— T/ie Influence of 

 Rivet-holes on the Strength and Endurance of Steel Structures. 



The numerous rivet-holes that pierce the plates and rolled sections used in 

 bridges and other steel structures reduce the strength and endurance of tension 

 members, particularly under varying loads. The arbitrary allowances which 

 are made for this reduction are discussed in relation to the results of a number 

 of ' fatigue ' tests carried out by the authors. 



Small plates with different spacings of rivet-holes, tested in a Haigh fatigue- 

 testing machine adjusted to apply tensile stresses varying from a minimum 

 to a maximum 2,000 times per minute, are shown to crack through the holes in 

 directions perpendicular to the axis of the applied pull— not necessarily across 

 the shortest alternative path between the holes. The results are compared with 

 theoretical deductions from establislied general principles. 



The fatigue tests are supplemented with steady-load extensometer tests on 

 plates with different rivet-hole spacings. 



In the afternoon the Canister Works of Messrs. Eeckitt & Sons, 

 Ltd., were visited. 



Friday, September 8. 



2, Mr. G. V. Maxted. — The Equipment of a modern Cement Works, 



with special reference to the work of the Humber Portland 

 Cement Co. 

 The works were visited in the afternoon. 



3. Prof. F. C. Lea, O.B.E., and Mr. E. E. Steadling.— T/zc 



Resistance to Fire of Concrete and Reinforced Concrete. 



The paper describes a series of experiments carried out in the Department 

 of Civil Engineering of the University of Birmingham, with the object of 

 investigating the effect of high temperature on Portland cement concrete, plain 

 and reinforced. 



It is shown that concretes as ordinarily used in practice to-day, containing 

 quartz sand as a fine aggregate, lose about 20 per cent, of their strength when 

 heated up to a temperature of 550° C. Above this temperature the loss in 

 strength is very much greater, and at about 700° C. the loss is of the order of 

 70 per cent, to 80 per cent. It is suggested that this is probably due to the 

 expansion of the quartz at 575° C, when the a-(3 transformation takes place. 



Much more fire-resistant concretes can be made by using a fine aggregate 

 made of brick or natural rock (such as basalts or dolerites), which do not 

 contain free quartz in .-my large quantities. The loss with such a material is 

 only of the order of 30-40 per cent, at 700° C. and even higher (1000° C). 



It is pointed out, however, that although these special concretes may carry 

 their load during a fire, yet the after-effects may be suflScient to cause failure, 

 and the reasons for tliis are discussed. 



Experiments are also described to obtain data on the linear expansion of 

 Portland cement when exposed to temperatures up to 800° C. The data obtained 

 indicate the probable cause of the spalling off of concrete surrounding steel 

 reinforcement. 



The suggestion is made that the concrete as normally used in reinforced 

 concrete is not an effective fire-resisting material. It will spall off and expose 

 the steel during the fire, and if the building does not then fail, the moisture 

 from the atmosphere finds access to the dehydrated lime, causing cracking and 

 disintegration of the concrete that has been heated above a certain temperature. 



