Investigation of bearing capacity of barrette piles with laboratory tests

Foundations are structural elements that transfer loads from structures to the soil. There are many parameters that need to be taken into consideration when designing the foundation, such as soil type and conditions, bearing capacity, settlement, liquefaction. Soil type and conditions are the most important of these parameters and they are used directly in determining the type of foundation. If there is no risk in terms of bearing capacity, settlement and liquefaction in the soil environment, shallow foundations are used. However, the structural loads are at high levels, the shallow foundation thickness is not economical, the shallow foundation is insufficient from geotechnical points etc. deep foundation systems are used. In many countries, as in our country, piles with circular cross sections are widely used in deep foundation applications. However, due to its advantages such as providing economy and ease of application, it is seen that the applications of barrette piles may be an alternative to circular section piles. In this study, model piles have been constructed to compare circular section piles and rectangular barrette piles in a laboratory environment. Rectangular shaped model barrette piles are produced in sizes that correspond to 1/10, 1/20, 1/30 and 1/40 of the real barrette pile dimensions. Circular shaped model piles are produced in approximately the same cross-sectional area as model barrette piles. Then, compressive loading tests were performed on model piles in case of loose soil conditions and bearing capacities of piles were investigated. As a result of loading tests, the effect of pile geometry, pile cross-sectional area and pile length on bearing capacity was analyzed. Scale effect was examined in rectangular barrette piles. Results obtained with laboratory model tests were subjected to regression analysis and models were formed. Linear, interaction, quadratic and pure-quadratic regression types were used in the created models and some statistical correlations were derived.