Creep analysis of steel fiber reinforced concrete based on beam tests and fracture mechanics concepts

Author(s)

Karla Peitl Miller

• Document type: Master's Dissertation
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Escola de Engenharia de São Carlos (EESC/SBD)
• Defense date: 2008-07-28

Examining board members

João Bento de Hanai; Luiz Eduardo Teixeira Ferreira; Walter Libardi

• Advisor: João Bento de Hanai

Abstract

Despite of the well known advantages of steel fiber addition to concrete (SFRC), especially the toughness improvement, only a few number of studies has been developed about creep on these composites. The main purpose of this research is to investigate the feasibility and inherent difficulties related to a particular creep evaluation method. This method is based on beam test results and their analysis by fracture mechanics theory. It is intended to become an alternative method instead of the usual creep analysis of axial compression test results. At the same time, looking at the development of hybrid composites - made of distinct kind of fibers to obtain the best responses for micro and macrocracking - an experimental program was performed. Specimens molded with plain concrete, ordinary SFRC and hybrid SFRC were tested in flexure, the last one made of an association of short and large steel fibers. Characterization tests were performed to obtain the main mechanical properties of these materials at several ages. The mixture proportions were based in previous studies, where good performance characteristics were observed in hybrid composites. Nevertheless, in this particular test series, the addition of shorter steel fibers resulted in high air contents, what probably caused the decrease of the composite\'s performance in some aspects. The test results displayed low influence of the fiber addition on mechanical properties such compression strength, modulus of elasticity and tensile strength. Creep performance showed to be worse in the SFRC and hybrid composites than in plain concrete matrix. However, the reinforcement with steel fibers improved the shrinkage restrain. The analysis of the long-term beam deflections was made by finding the corresponding strains in the sections. Afterwards, specific creep functions were obtained by regression methods. The experimental creep functions were compared to the existing ones in literature and design codes. Despite of some differences, such as higher initial creep rate, higher creep coefficients and faster stabilization, it may be concluded that these functions represented quite well the phenomenon. Also experimental functions for plain concrete showed good results when compared to creep prediction model given by design codes, such as the Brazilian NBR 6118:2003 and ACI 209:1982. Comparison with numerical modeling results also gave satisfactory results. Creep in flexure was also evaluated by means of notched beam tests, where the sustained load was performed only by the beam self-weight. The test results were analyzed by numerical modeling and application of fracture mechanic concepts. The overall results showed the feasibility of creep assessing by the beam test method, which can be, after further detailed test series, a good alternative method instead of axial compression tests. Also dynamic free vibration tests were performed, according to ASTM C-215:1991 recommendations, to investigate the beam stiffness loss due to long term loading effects. These tests showed that modal analysis can be a helpful method in the tests, since it does not introduce damages in the test specimens. (AU)