Democritus University of Thrace (DUTH)
Μηχανισμός μεταφοράς δυνάμεων ενισχυμένων δομικών στοιχειών από σκυρόδεμα
Abstract
dc:descriptionInternationally reinforced concrete (RC) structures constitute a large percentage of the existing buildings or infrastructures, with the most of them incompatible with the modern safety requirements. The intervention in a construction arises from several factors and leads either to repair or to repair and strengthening as a necessity for rehabilitation/upgrading of the safety level of reinforced concrete structures. Various repair and strengthening techniques have been developed in recent decades, with emphasis on jacketing of various materials. Concrete jacketing consists of a rehabilitation method with various advantages regarding the upgrading of the bearing capacity and ductility of the existing element or structure and at the same time being cost-effective solution with easy application. The main reasons that usually lead to intervention in RC buildings are: a) failures in critical structural components due to overloading (e.g., earthquake etc), b) initial construction imperfections (poor consolidation of concrete during casting), c) the need to upgrade the overall carrying capacity of the construction, particularly to constructions designed with older regulations that have no ductility requirements, which leads to increased seismic vulnerability, d) the need to upgrade the bearing capacity of the construction due to changes of usage and therefore change of applicable loads or due to storey super induction. Regardless the jacketing material, the key element of the strengthening design is proven to be the interface between old and new material. Through interface loads are transferred and the composite (strengthened) element is fully activated. The strengthened element is aimed to have the same behaviour with the corresponding monolithic one of the same section and characteristics, though the behaviour of the composite element is altered due to the mechanical behaviour of the interface of different materials or materials casted at different times. With numerous techniques (such as anchors, mechanical or chemical) the element’s behaviour is upgraded together with its monolithic connection of the interface. Interfaces are governed by shear transfer mechanisms such as aggregate interlock, cohesion and friction and dowel or clump action and the assessment of the interface behaviour is still a subject of research that the current thesis studies. The current study aims at the research of the transfer mechanisms of strengthened reinforced concrete elements and includes both experimental and analytical investigation. Seventy five column specimens were constructed, fifty of which were strengthened with reinforced concrete jackets. A large number of experimental result can be found internationally related to the interface behaviour of uni-side connection, of defaulted joints and created cracking of structural elements. In fact, many analytical models in the form of empirical relationships are proposed in order to define friction, cohesion, dowel and clump action and slip of core and jacket, while simplified equations of jacket design are incorporated in international guidelines. The mechanical behaviour of the interface of a composite element is related to the following parameters: a) the normal stress applied at the interface due to jacket’s confinement which limits the crack width and increases the element’s stiffness in shear, b) the roughness of the interface with various methods (resins, chipping, etc.) which affects favourably the element’s final shear capacity, c) the jacket’s confinement which causes the normal stress and limits the debonding of the contacted concretes and d) the connector bars (percentage, characteristics, diameter, etc.) crossing the interface of the two concretes (dowels) or connector of the longitudinal reinforcement bars (welded bend-down bars). Based on these parameters an extended experimental program was designed for the investigation of the mechanical behaviour of structural elements strengthened with reinforced concrete jackets. Seventy five specimens of square section (150x150mm, section scale 1:2) and total height 500mm were constructed, containing two different confinement ratios: low (ωcc=0.075) and moderate (ωcc=0.15), made of moderate concrete compressive strength (fc=24.37MPa) which was considered as moderate for structures with long lasting concrete curing. Steel reinforcement for the longitudinal bars had tensile characteristic strength of fy=500MPa (Β500c) while stirrups were of lower strength (fy=220MPa, S220). Fifty specimens were strengthened with stiff four-sided concrete jacket of 80mm width made of concrete with higher compressive strength (fcj=31MPa). All steel bars embedded were of the same quality with the cores’ reinforcement, all connectors were made of steel of Β500c. The reinforcement percentages of longitudinal steel and connectors were chosen according to international standards. The experimental setup favoured the study of the shear transfer mechanisms:A. direct loading of the interface. The imposed load acted at the core’s area while jacketing was seated, permitting the core element to slip (LPA). Nine specimens were tested with the following parameters: a) Interface roughness of old and new concrete, b) normal stress applied due to jacket’s confinement and c) steel bar concrete connectors (dowels) and reinforcement connector (bend-down bars) crossing the interface. Results show that dowel action is mainly shear in all values of slip while the axial component is negligible. With the dowel bar increase the rate of load reduction after peak is higher, accompanied by local crushing of concrete around the dowel bar. The normal stress applied by the activation of the jacket’s confinement affects favourably the maximum shear stress in smooth interfaces while doubling the confinement ratio doesn’t cause proportional increase of the shear stress. Roughening of the interface increases the maximum load and the corresponding slip. Finally, reinforcement connectors (bend-down bars) cause buckling of the longitudinal bars that are welded on. Β. Quasi real loading of core and jacket. The repeating loading is applied at the core’s section with the entire element seated (LPB) in order to simulate the real loading conditions in which the existing element is already deformed due to shrinkage that creates joints. With the same parameters as before ( a) roughness- smooth- teflon, b) normal stress on the interface due to jacket’s confinement and c) kind of connectors: concrete (dowels) or reinforcement connectors (bend-down bars), twenty eight prismatic specimens were tested. From the experimental study it is extracted that the failure mode depends on the kind of connector crossing the interface. Concrete connectors (dowel bars) act in shear in all deformation values and cause plastic regions of the weakest concrete around the bar, while the reinforcement connectors (bend-down bars) provoke buckling of the longitudinal reinforcement steel bar in the region they are welded, buckling which is decreased with the increase of the confinement ratio. What is more, the confinement ratio increase leads to lower rate of load reduction after the peak point. Finally, roughening of the interface contributes to the mechanical behaviour up to peak load and the following softening branch. In all aforementioned cases, the strengthened member presents an upgraded behaviour (v>1). C. quasi monolithic loading in which the composite element is seated and loaded in the entire strengthened section (LPC). Another parameter is incorporated at the experimental program, except for the normal stress applied at the interface due to jacket’s confinement and the kind of connectors, the initial construction imperfections due to incomplete or inadequate concrete consolidation or due to non-desired granulometry. Such imperfections are quantified through three geometrical indexes in order to predict the bearing capacity in axial loads. Specimens were repaired with the use of high strength thixotropic mortar with admixtures that eliminate shrinkage with polyacrylic fibres. Repaired specimens presented a steep softening branch after the maximum load provided that imperfections were extended at the confined area. The quantification indexes are used in a proposed novel and simple prediction model of the bearing capacity that comes in excellent agreement with experimental data. Specimens, then, were strengthened with reinforced concrete jacket and were retested. The experimental results show that jacket’s confinement compensate the bearing capacity of core upgrading the element. Finite element models were used in order to simulate and define the mechanical behaviour of selected strengthened specimens of the various loading patterns (LPA, LPB) and of elements with initial construction imperfections. Analytical results are compared with the corresponding experimental ones. For the analysis the following were selected: a) suitable stress-strain constitutive concrete law under triaxial stress and a two parameter failure criteria was applied (modified Drucker-Prager failure criteria), b) suitable 3D finite element with the option of simulating the aforementioned constitutive law, c) suitable algorithm of solving the equilibrium equations that permits the response if the member after the maximum capacity, d) suitable surface finite elements for the simulation of the maximum shear stress and the coefficient of elastic friction response and finally e) suitable multipliers simulating the interface debonding. Comparing experimental and analytical results leads to the conclusion that suitable simulation predicts the mechanical behaviour of the interface and the divergence up to peak is negligible since a softening branch cannot be simulated. Finite element analysis of specimens show that the jacket’s confinement works favourably at the bearing capacity of core concrete and relieves the connector bars (dowels). The finite element analysis’s results of specimens with initial construction imperfections are tolerably differenced from the corresponding experimental ones and confirm the failure mode of interface between repair mortar and core concrete (debonding) and of the failure mode in case of extended imperfections. To conclude, taking into consideration the codes’ provisions word widely it is extracted that the shear stress is conservatively calculated along interfaces for safety design matters. The degree of monolithic contact and interface capacity in transferring loads was quantified through a proposed interfacial monolithic coefficient. In fact, it is extracted that with the assumption of a quasi-monolithic loading and connection of old and new concrete the capacity and response of the strengthened element is overestimated. The initial construction imperfections effect needs further investigation. Specifically, it is suggested the repair of such imperfections with the use of other kind of repair mortars according to international standards, strengthening with fibre reinforced polymers (FRP’s) and the development of appropriate analytical model of predicting the capacity and response of the repaired element and of the interface of the strengthened and composite member. It is also necessary, to develop a generalised contact law with application in every kind of material and connector based on plasticity models and failure criteria. The reliability of such a simulator can be certifies with the application in complete software of finite element analysis. What is more, the investigation of the interface response in cyclic loading with the aforementioned parameters and in elements with predefined joints is necessary, together with an extended research in the applications of modern materials of high performativity for the repair of initial construction imperfections and damages due to overloading (ultra-high strength concrete, carbon nanotubes cement composites- CNTs).
Degree
thesis:*- Grantor dc:publisher
- Democritus University of Thrace (DUTH)
- Year dc:date
- 2014
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Achillopoulou, Dimitra
- Αχιλλοπούλου, Δήμητρα
Subjects
dc:subject × 8Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/34868