Train loading on bridges since Stephenson's Rocket

Abstract

Railway locomotives increased in weight from less than 10 tons in 1825 to more than 100 tons by the 1930s. Originally loading on bridges was assumed as an average weight per unit length ignoring axle concentration and generous factors of safety were used to compensate for the absence of any dynamic allowance. After the Dee Bridge collapse of 1847 Board of Trade rules were imposed. The comparatively few bridge failures under train loading are summarised. After several failures of flawed cast-iron girders these were gradually replaced after 1891. There was no rigorous consideration of dynamic effects until the 1900s. Civil engineers had been rejecting heavier new locomotives without comparing hammer blow characteristics. Dynamic effects were eventually investigated for the Bridge Stress Committee Report of 1928. With the elimination of steam traction, loading as developed for Europe was introduced and is now adopted in the Eurocodes. Comparisons are made between early and modern loadings taking account of the effect of increasing speeds.

Historical load effects on fatigue of metallic railway bridges

“…A detailed history of locomotive development and the significant changes to freight wagons and passenger coaches over time is given by Hayward (2010Hayward ( , 2013. The 1960s saw the introduction of fuel tanker bogie wagons weighing 100 t loaded.…”

“…The light-traffic historical model, unlike the heavy-traffic model, was not created from the annual tonnage of trains as the weight of passenger coaches have not changed significantly through the periods (Hayward, 2010(Hayward, , 2013. The light-traffic model was created by considering the amount of coaches required per year to move the volume of passengers.…”

“…For medium-span bridges, truss girders or plate girder arches were preferred, with plate girder arches and large box section bridges being preferred for longer spans (Hayward, 2010(Hayward, , 2013. The majority of the bridges on the rail network have a span around or less than 10 m (Hayward, 2010(Hayward, , 2013Imam, 2006) and are of plate girder construction. For example, about 45% of the metallic bridges in Europe have a span of less than 10 m (Bell, 2007).…”

Evolution of load conditions in the Norwegian railway network and imprecision of historic railway load data

“…Hammer blow was studied in 1920s by the British railways and these studies concluded that hammer blow dominated the dynamic loads on bridges (Hayward, 2014). The American Railway Engineering Association (AREA) also investigated dynamic loads on steel bridges.…”

“…In the absence of data on historic loading conditions at a bridge site, the current train loadings and traffic intensity are commonly applied to the entire history of the structure in service life estimation. Hayward (2011Hayward ( , 2014 presents train loads on bridges in the British rail network throughout its history and shows that both axle weights and train speeds evolved tremendously from the first railways to today's modern railways. As the fatigue damage mechanism is highly sensitive to the magnitude of the stress range, assuming today's traffic over the history of the structure will yield grossly conservative results (Imam, Righiniotis, & Chryssanthopoulos, 2008).…”