SOIL STRUCTURE INTERACTION OF BURIED CULVERTS

Abstract

Arching is a soil-structure interaction concept that plays a key role in most of the culvert installations. The term soil arching has generally been used to describe the phenomenon of stress transfer in a soil mass above a culvert as a result of mobilization and redistribution of shear stresses in the soil mass. Depending on the stiffness of the culvert relative to the surrounding backfill, soil arching above culverts can be positive or negative, i.e. the culvert may be subjected to stresses less than or greater than the overburden. Terzaghi (1936) stated that the amount of arching can only be obtained by direct measurement under field conditions. This thesis describes two instrumented full scale tests for monitoring the arching effects during construction and on a long-term basis. An extensive and critical literature study is carried out to review the present design methods for buried culverts. The limitations of the existing design methods are pointed out, and a plastic design approach based on stress field considerations is outlined for the design of buried culverts. The proposed design approach may be used both for a ultimate failure condition and in the working range. The method allows for mobilized shear stresses above buried culverts, to be handled in a consistent manner. The proposed design method is compared with other methods, and good agreement is found. Model tests and full scale test from the literature are presented, and the need for monitoring long-term behaviour is pointed out. The first full scale test is a long-span flexible steel culvert. The experimental study shows that the culvert undergo changes in structural response as time progresses after installation. The thrust force in the steel is increasing and shows negative arching for the structure as a whole, although the measured earth pressure above the crown shows positive arching. The second full scale test is a rigid concrete pipe under high fill, with a compressible layer of expanded polystyrene in the soil above in order to promote positive arching. The experimental study shows that the soft layer compresses during construction and mobilizes positive arching. The average vertical earth pressure on top of the pipe is 43 % of the overburden pressure, and lang term observations indicate that the positive arching effect is permanent. The experimental data is compared with the proposed design method and results obtained by the finite element method, and good agreement is found. The finite element method using a hyperbolic stress strain relationship for the soil clearly shows the mobilization of shear stresses and redistribution of soil stress due to the presence of the compressible layer. The finite element method is used to optimize the location and size of the compressible layer. The experimental and theoretical study shows that this method is reliable and can be used to construct more economic rigid culverts below high embankments.