![[PukiWiki]](image/pukiwiki.png "[PukiWiki]")
*Steel Railway Bridge in Norway [#l658d09f] **1.Field of application [#q32f651d] >The "Iron Ore" railway line between Lulea, Sweden, and Narvik, Norway. **2.Circumstances of repair [#be166874] >Completed in 1902, this bridge was renewed at the beginning of the 1960s and several of the old riveted bridges were replaced by welded plate girder bridges. Fatigue cracks were found in the webs of some plate girders at the ends of the vertical stiffeners after a few years of service. >A total five bridges of the same structural type exhibited 100-200 mm long fatigue cracks at the ends of the vertical stiffeners. **3.Types of structure [#q5cc1116] >Welded plate girder bridge. >Five bridge of same structural type were constructed of two parallel, simply supported welded steel I-beams, with the sleepers set directly on the top flange (Fig. 1) One bridge is over the Maunojokk River, built in 1961 and has a span length of 11.6 m. The track lies in a horizontal curve with a radius of 600 m. The two main girders, 1900mm apart, are straight, 1000 mm in depth and connected by cross-framing approximately every 2.0m. The web plate of both girders is strengthened on the inside by a vertical stiffener at every cross frame. These stiffeners are not directly attached to the top and bottom flanges. Instead, they are in contact with the flanges through a steel plate fitted at their ends. This steel plate is welded to the stiffener only, and not to the flange. The bridge is stabilized by a horizontal truss at level of the top flange. >Five bridge of same structural type were constructed of two parallel, simply supported welded steel I-beams, with the sleepers set directly on the top flange ([[Fig. 1>#fig1]]) One bridge is over the Maunojokk River, built in 1961 and has a span length of 11.6 m. The track lies in a horizontal curve with a radius of 600 m. The two main girders, 1900mm apart, are straight, 1000 mm in depth and connected by cross-framing approximately every 2.0m. The web plate of both girders is strengthened on the inside by a vertical stiffener at every cross frame. These stiffeners are not directly attached to the top and bottom flanges. Instead, they are in contact with the flanges through a steel plate fitted at their ends. This steel plate is welded to the stiffener only, and not to the flange. The bridge is stabilized by a horizontal truss at level of the top flange. &br; CENTER:&aname(fig3);&attachref(King_h37.jpg);&br; CENTER:&aname(fig1);&attachref(Norway1.gif);&br; &br; CENTER:Fig.3 Cross section. CENTER:Fig.1 Bridge type where fatigue cracks occurred. &br; **4.Details of loading [#yeab6531] >Fluctuating loads due to vehicles live load. >Fluctuating loads due to railway live load. >One span collapsed under the load of s semi-trailer vehicle with a total weight of 47tons; a load within the permissible limits for the bridge. **5.Description of damage [#x57ecd95] >The collapsed span was made up of four 100 ft long suspended girders, and each girder fractured at a point 16 ft from the northern end of the span. The structure sagged 18 in. Inch-wide cracks spread acrossed the deck. >Fatigue cracks occurred in the webs of some plate girders at the ends of the vertical stiffeners. These cracks had propagated first horizontally and then gradually towards the neutral axis ([[Fig.2>#fig2]]). The orientation of the fatigue cracks were almost horizontal and parallel to the main girder stresses. >The cause of the failure was given as brittle fracture, at a point on each girder where a change in flange section occurred. The night temperature contributed to the collapse. The flanges of the girders were thickened locally by welded-on cover plates ([[Fig.4>#fig4]]), and each fracture was at the point of termination of one of these cover plates ([[Fig.5>#fig5]]). The cover plates were welded manually. >Cause of fatigue cracks is as follow. -Out-of-plane displacement of the gap between the flange and the end of the weld of the vertical stiffener due to train pass ([[Fig.3>#fig3]]). >Under bead cracking was observed in welds at ends of tension-flange cover plates. It was suggested that the cause of the cracks might have been due to a lack of preheating at these points or to the presence of excess hydrogen or inadequate precautions for the difficulties of welding. &br; CENTER:&aname(fig4);&attachref(King_h38.jpg);&br; CENTER:&aname(fig2);&attachref(Norway2.gif);&br; &br; CENTER:Fig.4 Flange of a girder with cover plate. CENTER:Fig.2 Typical fatigue cracks; propagation arrested by holes drilled at crack tips. &br; &br; CENTER:&aname(fig5);&attachref(King_h39.jpg);&br; CENTER:&aname(fig3);&attachref(Norway3.gif);&br; &br; CENTER:Fig.5 Clack and fracture location. CENTER:Fig.3 The effect of displacement on the transverse bending stresses in the web plate. &br; **6.Repair method applied [#qec07795] >The recommended repair was to pre-stress the girders with steel cables or other devices within the depth of the present structure so that flanges liable to cracking under tension would be subjected to compressive stresses. >The cracks were arrested by drilling holes at each crack tip. No cracks so treated continued to propagate. >To improve structural behavior, authors suggested some alternative bracing systems due to finite element analysis. The design changes which had the beneficial effects is as follows. Alternative 1 : The horizontal bracing of the as-built bridge consists of six horizontal diagonal truss bars. Alternative 2 : Removing all the vertical cross-framing diagonal bars between the supports. Alternative 3 : All vertical web stiffeners are attached to the bottom flange. To minimize any reduction in fatigue strength of the tension flange, the connection could use high strength friction grip bolts. **Reference [#ze343660] >'''Engineering News Record,'''Sept., 20, 1962. >'''Engineering,''' Sept., 21, 1962. >Nishimura T. and Miki C., Fracture of Steel Bridges Caused by Tensile Stress, '''J. Japanese SCE,''' Nov., 1975. (In Japanese) >Akesson B., Edlund B., Shen D., Fatigue Cracking in a steel Railway Bridge, '''Structural Engineering International,''' Vol. 7, No. 2, 1997. &br; &br;
