The Transference of Axial Load From Reinforced Concrete Columns to Bases

Abstract

[Master of Philosophy thesis].  The research work described in this thesis is concerned with the transference of axial load from reinforced concrete column to concrete base. The load is transferred to the base by column longitudinal reinforcement, core concrete, and cover concrete. The first stage is primarily concerned with the literature review, design, preparation, and testing of specimens , which are designed to show the effect of the following base slab properties on the transference of the axial load where H-T square twisted steel bars are used for base and column reinforcement: (1) The overall depth (h) and therefore the bond length of the column bars; (2) Amount of tensile steel reinforcement (As ) and hence (p); and (3) The lateral dimensions (Ax B). The second stage is concerned with the results and calculations and shows how the above variations affected the transference of the axial load from the column steel to the base by anchorage bond and on the column strength as a whole. From these calculations it is found that the British code uses conservative ultimate anchorage bond stresses and a low allowable stress for steel in compression where the American code uses high or sometimes unsafe ultimate anchorage bond stresses and a very high all owable stress for steel in compression. It is found that part of the column length is acting as an extra anchorage length for the column longitudinal steel in addition to the embedment length in the base slab concrete. The third stage is concerned with the transference of axial load from column concrete to base and shows that this load is transferred from the core concrete at a higher stress than that from the cover concrete. A three part addition formula is suggested to calculate the ultimate axial load transferrable from a short reinforced column to a concrete base according to the above findlng. Finally the distribution of the anchorage bond stress along the anchorage length is predicted. This shows that a parabola plus a constant gives the best fit with the experimental results and that the results are best related to √fcu rather than fcu.