Bond Tests of Short NSM-FRP and Steel Bar Anchorages D. G. Novidis 1 and S. J. Pantazopoulou, M.ASCE 2 Abstract: Performance of near-surface mounted NSM bars as additional reinforcement in strengthening of existing reinforced-concrete construction largely depends on the development capacity of the bar inside the groove. This is controlled primarily by the surface characteristics of the bar and its interaction with the surrounding groove ﬁller and the cover concrete. In this paper, the bond strength of short NSM-bar anchorages is explored experimentally, using a modiﬁed eccentric pullout test specimen designed to alleviate some of the deﬁciencies associated with bond testing. A total of 45 tests were conducted to study the inﬂuence of the most important technological parameters of this upgrading method, namely the groove dimensions, the embedment length, and the surface pattern of the bars. NSM bars were 12 mm diameter rods: CFRP bars used had a sandblasted surface and winding helical lengthwise indentations. Steel bars were either standard deformed bars or smooth reinforcement. The variety of bar type considered was intended to highlight and to quantify through the tests the bond strength that may be mobilized by postinstalled reinforcement according to surface proﬁle and stiffness. Test results are used to establish a limit-state bond-slip model for near-surface mounted bars, so as to enable implementation of this emerging technology in practical design. DOI: 10.1061/ASCE1090-0268200812:3323 CE Database subject headings: Steel; Bars; Pull-out resistance; Finite element method; Bonding; Anchorages. Introduction Over the past decade, the advent of ﬁber-reinforced polymer FRP reinforcing products has provided opportunities for inno- vative applications in the ﬁeld of upgrading/strengthening of existing structures. These reinforcing materials offer several ad- vantages such as high strength and stiffness, corrosion resistance, low weight, and fast placement without excessive scaffolding. An effective method for this purpose is by postinstalling rebars in surface grooves cut on the tension side or the web and is known as near-surface mounting NSM technology, De Lorenzis and Nanni 2001. This is a relatively old technology for postinstalling reinforcement in structures with deﬁcient ﬂexural strength that received renewed attention with the extensive availability of new FRP reinforcing products such as bars and laminates. The NSM method comprises three basic steps: a preparation of a surface groove in the concrete cover of the member that requires strengthening; b detailing and postinstalling of rebars; and c groove ﬁlling and surface leveling with a pertinent ﬁller material. The ﬁller material and added-bar combination are re- ferred to as the strengthening system De Lorenzis and Nanni 2001. Performance of the bar largely depends on its development capacity, which is controlled by its surface characteristics, its ex- tensional stiffness, and by the fracture properties of the ﬁller. Once placed, NSM bars are subjected to similar displacement conditions as the existing reinforcement. If force transfer can be supported through bond, then the bar may effectively function as the existing reinforcement, developing axial strain and a commen- surate tensile force, thereby contributing to the member resis- tance. With this strengthening method, apart from securing composite action, bond of NSM reinforcement controls strength, spacing and width of cracks, and the eventual failure mode of the upgraded member Teng et al. 2006. Stress transfer and interaction occurs along two interfaces, namely, the cylindrical contact surface between the NSM bar and the surrounding ﬁller, and along the rectangular boundary of the groove, between groove ﬁller and the surrounding concrete. In both cases, force transfer is achieved through chemical adhesion at the early stages and through friction for higher levels of slip. Clearly, the roughness of the groove surface is an important vari- able, although it cannot be easily regulated. Mechanical interlock- ing between bar and ﬁller may occur if the bar used has surface deformations. Therefore, when using NSM reinforcement, the dependable bond strength is controlled entirely by the surface characteristics and rod stiffness, the groove dimensions and its surface rough- ness, and the shear behavior of the ﬁller material. Usually failure occurs at one of the two contact interfaces; pullout tests in con- trolled laboratory conditions usually lead to failures at the concrete-epoxy paste surface, although failures by splitting have also been reported if a cementitious mortar is used instead of epoxy paste as ﬁller material Teng et al. 2006; Novidis et al. 2007. The ductility of the failure mode depends entirely on the properties of the system. If the ﬁller material is epoxy paste, i.e., an elastic brittle material with a high tensile resistance, failure by precipitous debonding is expected to occur in either of the two contact surfaces without stress redistribution. In this paper, the bond properties of FRP and steel bars used as NSM reinforcement are explored through testing of 45 short- anchorage eccentric pullout concrete specimens, using a novel 1 Ph.D. Candidate, Civil Engineering, Dept. Civil Engineering, Demokritus Univ. of Thrace, Vas. Soﬁas #12A, Xanthi 67100, Greece. E-mail: dnovidis@civil.duth.gr 2 Professor, Dept. of Civil Engineering, Demokritus Univ. of Thrace, Vas. Soﬁas #12A, Xanthi 67100, Greece. E-mail: pantaz@civil.duth.gr Note. Discussion open until November 1, 2008. Separate discussions must be submitted for individual papers. To extend the closing date by one month, a written request must be ﬁled with the ASCE Managing Editor. The manuscript for this paper was submitted for review and pos- sible publication on March 5, 2007; approved on June 18, 2007. This paper is part of the Journal of Composites for Construction, Vol. 12, No. 3, June 1, 2008. ©ASCE, ISSN 1090-0268/2008/3-323–333/$25.00. JOURNAL OF COMPOSITES FOR CONSTRUCTION © ASCE / MAY/JUNE 2008 / 323