Different aspects concerning the resistance of steel girders subjected to concentrated forces have interested a large amount of researchers all over the world. The subject most dealt with among researchers is what is called "patch loading", a concentrated force applied at one flange. Generally, the resistance to concentrated forces is considered as being a problem of very complex nature. Therefore all studies aiming at predicting the ultimate resistance of steel girders to concentrated forces gives more or less empirical solutions. This thesis focus on three different applications of concentrated forces. Concentrated forces applied at one flange, opposite concentrated forces applied at two flanges and concentrated forces applied at an unstiffened girder end. The first application is recognized as patch loading, the second and third are herein defined as opposite patch loading and end patch loading, respectively. The report is divided in two parts, the main text and the appendices. The appendices are used to give results form tests performed herein and to present data for tests taken from the literature, but also to give results from a comparison with a design model suggested herein and the tests. The main text starts with an introduction giving the basis in the subject and a review of earlier work and ends up in a suggestion for a design model for the ultimate resistance for the three different applications of concentrated forces. The design model is developed with the purpose to be consistent for the three load applications, simple enough to be fit for use and also harmonized with those used for other buckling problems. The design model is based on a von Karman approach and includes three parts, an expression for the yield resistance, the elastic buckling load and a resistance function. The expression for the yield resistance is based on observations and conclusions from an experimental investigation, including a total of 48 tests on welded girders made from the high strength, quenched and tempered steel Weldox 700 and 12 tests on rolled beams. All three load applications were tested, with the majority of the tests performed as end patch load test. Approximate solutions for the elastic buckling load for patch loading, opposite patch loading and end patch loading are formulated on the basis of results from FEM. The buckling coefficients includes a contribution from the flange stiffness and the approximate solutions gives a reasonable accuracy compared to the result from FEM. The resistance function is empirically determined from about 250 patch load tests, since this is the only load application for which one can find tests on girders with slender webs in the literature. The resistance function is also proven suitable for opposite patch loading and end patch loading. Compared to results from a total of about 540 tests, the design model suggested herein gives a better accuracy in the prediction of the ultimate resistance for all three load applications than other leading design models from the literature. What can be considered as new and original in this thesis is that it gives a consistent model for prediction of the resistance of steel girders subjected to concentrated forces which is harmonized with the models for other instability problems a design model that extends the applicability to high strength, quenched and tempered steels additional solutions for the critical buckling loads for various boundary conditions / Godkänd; 1995; 20070426 (ysko)
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:ltu-26700 |
| Date | January 1995 |
| Creators | Lagerqvist, Ove |
| Publisher | Luleå tekniska universitet, Byggkonstruktion och -produktion, Luleå |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Doctoral thesis, comprehensive summary, info:eu-repo/semantics/doctoralThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | Doktorsavhandling / Högskolan i Luleå, 0348-8373 ; 159 |
Page generated in 0.0103 seconds