AN INTERACTIVE DESIGN OF CAPPING BEAMS SUPPORTED OVER ROWS OF SHORT BORED PILES

Abstract

Quite often, masonry structures are supported on rows of short bored piles, like under-reamed piles. Pile-capping beams are used to tie the heads of the piles,- which under normal circumstances are supported on the soil, to carry the masonry walls of the structure. However, while designing such beams very little consideration,is given to the load transferred by the capping beams directly to the soil and usually they are designed as if they are isolated from the soil. Some empirical rules based on measurements are also used. The piles, thus, have to be designed for higher loads. Another beneficial factor often ignored are the stiffening effects of masonry .walls.on the capping beams. Thus the design often turns out to be quite distant from reality. One of the reasons for ignoring the interaction of various elements of structure and soil is that such analyses are, often, quite involved. Even with the availability of versatile methods of solution, like Finite Element Method, and the easier accessibility of fast computers, the analysis remains quite costly due to substantial cost of computation and input-output data processing and interpretation. This is more so because the problem is essentially a 3-dimensional one because of 3-dimensional nature of the soil, though other elements may be adequately idealised as 2-dimensional members. Thus, if some improvisation could be incorporated in the programme by which one could treat the problem as a 2-D one, at the same time the effect of the 3rd dimension of soil V could . be included, it would reduce both computation, time and data processing effort, and, thus, would economise such an analysis. With this in view a Q3D finite element-method of analysis was fteveloped (called Quasi-3D Finite Element Method (Q3D) hereafter) and Q3D analysis capability was incorporated in an available computer programme for solu- tion of 'plane stress', 'plane strain' and 'axisymmetric' problems. The programme was used for the analysis of a beam resting on elastic medium and was found to be quite in agreement w~,th results obtained by earlier workers. The problem under consideration was also analysed by this method varying various parameters, like spacing of pile, width of capping beams, elastic properties of soil and uniform and 2-layer media. The piles., wall and beam were Idealised as 'plane stresst elements and the soil, as 'Q3D' element. Soil was taken as isotropic material while at pile sections soil/pile/soil was idealised as an trans- versely layered material and an equivalent elastic property was used for this. Various (masonry/beam, pile/soil,, soil/ beam) interfaces were assumed to be perfectly adhering. Soil was ''assumed to-have deformation only in the vertical (v) and along the length of beam (u) while the deformation transversely horizontal to the beam axis was considered negligible. The analysis showed, among other things, that: vi i) a considerable share of the total load (20-70 °., depending on various parameters) is directly trans-mitted to the soil by 'footing action' of the capping beam; ii) this share of load is directly proportional to spacing of piles and, also, to the modulus of elasticity of the supporting soil medium; iii) but, this is only marginally affected by the variation of the width of the capping beam or that in the relative stiffness of the stiffer underlayer in a 2-layer soil system; iv) the masonry wall, supported on the beam, acting in composition of the beam, transfers a substantial por-- tion of the total load to the pile by its 'arching action' ; v) when the beam continues over two or more spans, the pile loads (i.e. the support reactions tend to become equal, due to differential settlements of the piles and, consequently, little negative bending moment develops in the beam over the piles; vi) considerable friction develops in the beam/soil interface, which reduces the bending moment and the axial force in the beam; and vii) the bending moment and axial tension in the beam is dependant on the total pile load and the spacing of piles. vii Based on these observations, a step-by-step design methci has been suggester'. for determination of pile loads and forces - bending moment and axial tension/compression in the capping beams. Thus, it may be seen that the study of interaction of soil-structure not only ensures that a structure will per-form satisfactorily, but also, may, often, lead to sub-stantial economy.