Strengthening of Concrete Structures using Steel Wire Reinforced Polymer

Transcription

1 SP Strengthening of Concrete Structures using Stee Wire Reinforced Poymer by W. Figeys, L. Schueremans, K. Brosens, and D. Van Gemert Synopsis: This paper deas with a new materia for externa reinforcement: Stee Wire Reinforced Poymer (SWRP). It consists of thin high-strength stee fibres embedded in a poymer aminate. This innovative materia combines the advantages of stee pates and CFRP, which are aready used today. The materia cost of SWRP is reativey ow, and the aminate is quite fexibe. In the feasibiity part, the practica use of SWRP is studied. Further, the avaiabe design mode for externay bonded reinforcement for concrete eements is confronted with the resuts of an experimenta program, carried out at the Reyntjens Laboratory of KULeuven. The mode is adapted accordingy. Keywords: design mode; externa reinforcement; stee wire reinforced poymer (SWRP) 743

2 744 Figeys et a. Wine Figeys, ir., obtained Master of Science in Architectura Engineering in 004. She is PhD researcher at the Buiding Materias Division at the KULeuven. Her research topic concerns structura restoration techniques, focusing on epoxy bonded externa reinforcement. She has a research grant of the Institute for the Promotion of Innovation through Science and Technoogy in Fanders (IWT-Vaanderen). Luc Schueremans, dr. ir., graduated as a civi engineer at the KULeuven in He defended his Ph. D. thesis at the end of 001. Since 00, he is a post-doctora researcher at the Buiding Materias Division of KULeuven, Begium. His research focuses on structura behavior of masonry, materia modeing, and reiabiity based design and safety assessment of existing structures. Kris Brosens, dr. ir., obtained his PhD at the Department of Civi Engineering, K.U.Leuven, Begium in 001. His research concerned the strengthening of concrete structures with externay bonded stee pates and fibre reinforced materias. Nowadays he works as a project engineer for Triconsut N.V., a spin-off company of the K.U.Leuven, speciaized in the structura restoration of historica monuments. Dionys Van Gemert, dr. ir., is professor of buiding materias science and renovation of constructions at the Department of Civi Engineering of KULeuven, Begium. He is head of the Reyntjens Laboratory for Materias Testing. His research concerns repairing and strengthening of constructions, deterioration and protection of buiding materias, and concrete poymer composites. 1. INTRODUCTION The capacity of a structure can be enhanced using the technique of externay bonded reinforcement. By adding extra reinforcement the fexura capacity, the fexura stiffness and the shear capacity are infuenced. Today, mosty stee pates and carbon fibre reinforced poymer (CFRP) sheets and aminates are used. Since both materias have their own properties, they aso have their preferabe appications [1; ]. Stee pates have a ow materia cost and can easiy be appied in arger sections. Therefore, deformation probems are often tacked with stee pates. As stee pates have no fibrous structure, bots can be used to reduce the anchorage ength. Disadvantages are the high density of stee which hampers the appication, and stee pates need a specia treatment against corrosion. CFRP is a ightweight fexibe composite. This makes it easier to appy. It has an E-moduus comparabe to stee. The tensie strength is 5 to 10 times higher than standard stee. CFRP is used for strengthening of concrete pates, because their strength can be better expoited. As the fexibe sheets can easiy be wrapped, CFRP is aso often appied as externa shear reinforcement. But, as carbon can not take shear stresses, beams have to be rounded with a radius of 3 cm. Disadvantages are the high materia cost and its britteness. Therefore, arge safety factors are required.

3 FRPRCS Stee wire reinforced poymer (SWRP) is a new materia that can be used as externa reinforcement. It consists of thin high-strength stee fibres which are bunded into cords (see Fig. 1). These cords are woven into unidirectiona sheets with a synthetic textie. SWRP combines the advantages of stee and CFRP: the composite has the same strength as CFRP but is ductie. The materia cost is ow and the SWRP-aminate remains quite fexibe. The new composite enabes the same appications as stee pates and CFRP sheets and aminates, but aso new appication chaenges can be tacked: shear strengthening of compex shapes, wrapping of rectanguar beams, and improved uses of pre-stressing. This paper deas with the use of SWRP as externa reinforcement. It consists of three parts. In the first part a feasibiity study is made. Probems for the practica use of SWRP as externa reinforcement are recognized. In this part, the impregnation and the fexura stiffness of SWRP are studied. In the second part, the behaviour of the bonded connection with SWRP is examinated. If this behaviour can be predicted, anchorage ength and transferabe oad can be cacuated. Therefore, non-inear fracture mechanics is appied to mode and describe the shear-sip behaviour of the bonded connection. Some mode parameters, specific for the new materia, are determined by means of direct shear tests. In the ast section, the combination of bending and shear is studied. When a beam is strengthened, some extra faiure modes are introduced. One of them, so caed deamination, is studied in this section. The used mode is tested by means of four point bending tests on beams strengthened with SWRP.. MATERIAL PROPERTIES The SWRP used in the experiments of this paper is a prototype, produced by Bekaert Inc., Fig.. One sheet (width: 95 mm) consists of 65 stee cords. 19 fiaments are twisted in a cord. The fiament in the midde has a diameter of 0.5 mm, the other 18 have a diameter of 0. mm. By means of five tensie tests, the tensie strength and E-moduus of the SWRP are determined. The average tensie strength is 775 N/mm², the average E- moduus is N/mm². To ensure a good impregnation of the cords, SWRP can be pre-impregnated. This kind of SWRP can be compared with the pre-cured CFRP aminates where impregnation is aso more compete. In this research, some SWRP is pre-impregnated with an epoxy resin through vacuum impregnation foowed by autocaving at 15 C and 3 bar. Experiments are done on doube ayer SWRP with the epoxy resin F533 from Hexe Composites [3] with E-moduus 900 N/mm². 3. FEASIBILITY In practica appications, it is desirabe that SWRP is gued as easiy as CFRP. Such sheets can be easiy wrapped because of the ow fexura stiffness of the fibres. The stiffness of SWRP is investigated first, as we as the impregnation of SWRP with the adhesive.

4 746 Figeys et a. 3.1 Fexura stiffness of SWRP The new composite enabes the same appications as stee pates and CFRP but aso new appication domains are targeted: shear strengthening of compex forms, wrapping of rectanguar beams. Since the sheets are fexibe, CFRP can be easiy appied, which is an important advantage in practice. When wrapping CFRP, the reinforcement sticks to the beam or coumn without extra auxiiary actions. Aso SWRP seeks an easy appication. For the proposed appications, it is important that SWRP can be wrapped and kept in pace without specia arrangements. The stiffness of SWRP can be checked by cacuating the cross sectiona moment of inertia. Without taking into account the torsion of the fiaments, the moment of inertia equas 4.05 mm 4, Equation (1), for the type of SWRP studied in this paper. I cord = ( I fiament + Ai yi ) (1) i with A i section of fiament i [mm²] y i distance of fiament i to the neutra axis [mm] When the shape of a CFRP sheet is assumed to be a rectanguar pate (0.0167mm x 95 mm), the moment of inertia is mm 4. This means that the studied type of SWRP is more than 100 times stiffer than CFRP. A decrease in stiffness can be reached if ess fiaments beong to a cord, or if the fiaments are used singe, or if other types of SWRP are used. 3. Impregnation It is important that the externa reinforcement can be gued, without causing a weak ink in the connection. As the reinforcement is often appied above the head, viscous adhesive is used. When using this kind of adhesive, it has to be checked that a stee fibres are surrounded by the adhesive. If not, the gue is the weakest ink in the connection and can cause premature faiure. The impregnation of SWRP is tested by means of six pu-off tests. Severa pieces of SWRP are gued on concrete with Epico U [4], a viscous adhesive. It is necessary that the externa reinforcement is pushed into the gue, which is not convenient in practice. Therefore, new adhesives have to be deveoped. Pu-off test were carried out on the SWRP-aminates. On the SWRP, a cyindrica eement is gued. After hardening of the adhesive, a cyindrica saw cut is made to define the faiure area. Afterwards, the cyindrica eement is pued off. In Fig. 3, test sampes are presented after the pu-off test. The impregnation of the cords is sufficient. A test sampes have faied in the concrete, not in the adhesive or in the connection between cyinder and SWRP or between SWRP and concrete. But in the aboratory, the work is more accuratey done than in practice. Therefore, it seems that enhanced impregnation is necessary. This is possibe by deveoping new adhesives or with pre-impregnated aminates. These

5 FRPRCS aminates have a very good impregnation which does not depend on the circumstances of the site. 4. SHEAR BEHAVIOUR When using SWRP as externa reinforcement, the required anchorage ength and transferabe force must be known. Therefore, non inear fracture mechanics are appied to a mode which describes the shear-sip behaviour of the bonded connection. At the Reyntjens aboratory of the KULeuven, mode parameters are determined in a test program by means of direct shear tests. 4.1 Pure shear mode Using the equiibrium of forces, Vokersen [5] derived the foowing differentia equation which describes the shear stresses as function of the sip of the externa reinforcement: d s ( x) 1+ mγ dx E h E in which m = E c τ ( x) = 0 A γ = Ac (4) and s (x) sip of the externa reinforcement at x [mm] τ (x) shear stresses in the adhesive at x [N/mm²] E E-moduus of the externa reinforcement [N/mm²] E c E-moduus of the concrete [N/mm²] h thickness of the externa reinforcement [mm] A section of the externa reinforcement [mm²] A c concrete section [mm²] The best resuts are given by the assumption of a biineair shear-sip reationship, Fig. 4, as shown in [6; 7]. Initiay, the shear stress increases, unti the maximum shear stress τ m is reached. Afterwards, concrete cracks appear in the concrete and consequenty the shear stress decreases. When the utimate sip s 0 is reached, no forces are transferred anymore and the connection fais. With this assumption, the soution of Equation (3) becomes: () (3) with: s ω ( x) = Asinh( ω x) + B cosh( ωx) (5) τ = s m m 1+ m γ E h (6) Three mode parameters τ m, s m, s 0 are introduced in the biinear shear-sip reationship. The shear peak stress τ m depends ony on the strength properties of the concrete because faiure wi occur in the concrete. Appying a inear Mohr-Couomb faiure criterion [6],

6 748 Figeys et a. Equation (7) can be derived from the Mohr s circe for pure shear and a tangentia intrinsic curve [1]. with: f ctm f cm τ m = (7) f + f ctm cm f cm compressive strength of concrete [N/mm²] f ctm tensie strength of concrete [N/mm²] However, additiona parameters must be introduced to account for the reationship between the ab test and the reaity, Equation (8). fctm fcm τ m = kb1kbk c (8) f + f ctm cm k c is the concrete infuence factor, varying between 0,65 and 1. k c is 1 in case of good workmanship, reducing to 0.65 [1] in case of bad workmanship. k b1 describes the size effect for britte materias and is given by Equation (9) [7]. The mechanica strength increases when the test sampe becomes smaer. k b1 = k b ( k 1) (9) 1+ h ref where b is the width of the externa reinforcement and h ref is an empirica factor. h ref is the depth of the concrete that is infuenced by the shear stresses in the adhesive. Hozenkämpfer [7] proposed a vaue of h ref of.5 3 times the size of the biggest grave stones. k is an empirica factor which takes into consideration the muti-axia stress situation. This parameter has to be determined from experiments. k b introduces a second width effect: it accounts the spreading out of the forces in the concrete, Fig. 5. b kb = (10) bc in which b c is the width of the concrete. The second mode parameter s m is the vaue of the sip at peak shear stress. The sip is determined as the sum of the sip in the different ayers: the concrete (height h ref ), the adhesive and the externa reinforcement: h h = ref h i g h s = + + ( + ) m τ i τ m,4,5 1 ν (11) G i i Ec Eg E with: h thickness of the externa reinforcement [mm] h g thickness of the adhesive ayer [mm]

7 G i = E i (1 + ν ) shear moduus of the i th ayer FRPRCS [N/mm²] ν Poisson s ratio; [-] concrete: ν c = 0,; adhesive : ν g = 0,5; stee wire : ν = 0,3 As s m depends on τ m, it aso depends on the parameters k b1, k b, k c. The ast mode parameter is the sip s 0 at which no more forces can be transferred. Therefore, a new parameter is introduced: the fracture energy G f. This fracture energy is the energy per unit area, needed to bring a connection into compete faiure. It is given by the area under the shear-sip curve, Fig. 4. In Equation (1) gives the fracture energy for a biinear shear-sip reationship. G f τ = τ ( s ) ds = 0 s m 0 (1) If the fracture energy is known, the sip s 0 can easiy be cacuated. To determine the fracture energy, foowing expression is prosposed by Hozenkämpfer [6]: G f b1 b c = k k k C f (13) f ctm G f depends on the tensie strength of the concrete. Again, the parameters k b1, k b, k c are incuded. A new empirica factor is introduced. C f is a parameter fits the experiments to the mode. When the three mode parameters are known, the differentia equation of Vokersen can be soved. The shear stresses can be cacuated at every ocation as a function of the externa oading. Aso the anchorage ength and the corresponding maximum transferabe oad can be derived [1]. 4. Experiments Eight direct shear tests were executed at the Reyntjens aboratory. On four of them singe SWRP is tested. Pre-impregnated doube ayer SWRP (marked with I) are used in the four others. The test set-up is shown in Fig. 6. Two concrete prisms are bonded together with SWRP on two opposite sides which are grit basted. Between the two prisms, there is a gap of 18 mm. Bonding ength is 150 mm or 00 mm. On the other sides, stee pates are gued. They transfer the oad from the testing machine to the test sampes. Commerciay avaiabe adhesive Epico U [4] is used. This adhesive is a gue paste, its E-moduus equas 7000 MPa. After hardening, a saw-cut is made aongside the SWRP, to prohibit spreading out of forces. Therefore, k b equas 1. Since the test specimens are carefuy prepared under aboratory conditions, k c is aso 1.

8 750 Figeys et a. The test is deformation controed. The change of the gap between the two concrete bocks is monitored by means of two Linear Votage Differentia Transformers (LVDT) at two opposite sides of the bock, Fig. 6b. The gap increases at a constant mean rate of 0,001 mm/s. By comparing the two individua signas, one can check whether or not the tensie force acts centricay. Test 150a was obique and is not taken in consideration. In a test specimens, faiure was due to faiure of concrete. This means that the concrete is the weakest ink in the connection. Test specimen I150a after faiure is shown in Fig. 7. Brosens [1] determined empiricay the vaues of the parameters for CFRP: k = 1,47 mm and C f = 0,40. With these parameters, a first estimation of the transferabe oad can be made, tabe 1. The mode parameters are cacuated with the data from tabe 1. In Fig. 8, the measured tensie force and sip are compared to the mode using the CFRP mode parameters. In the test with singe SWRP a sma deviation is observed. At the end of the test, the measured curve diverges more from the mode and faiure took pace ater than predicted. The deviation on the force is about 1% and the test resuts are aways underestimated. The SWRP seems to behave stronger and stiffer compared to eements strengthened with CFRP. The mode parameters for CFRP predict the resuts reativey good, but are a rather conservative approach. The test resuts on pre-impregnated doube ayer SWRP are presented in Fig. 8 b. Smaer sip is measured, compared to the eements strengthened with singe SWRP because of the higher amount of reinforcement. The measured curves diverge more from the CFRP-mode (average deviation for the transferred forces is +0%). In [8] a sensitivity anaysis on the transferabe forces is performed. Based on this anaysis, it is concuded that a deviation on the materia characteristics can not expain a deviation of 1% and 0% from the mode. Aso the sensitivity of the modeparameters h ref, k, C f is examined. Changes on h ref cause ony sma changes in the transferabe forces. Parameter C f is a factor, needed to cacuate the fracture energy. It has amost no effect at sma bonding engths. The infuence of C f is ony observed for ong bonding ength, especiay onger than the theoretica anchorage ength. Anchorage ength can be cacuated as expained in [1] : with the CFRP parameters it becomes 85 mm for the singe SWRP and 399 mm for pre-impregnated SWRP with ayers. Finay, the parameter k takes into consideration the muti-axia stress situation. When changing this parameter, it has an infuence on a bonding engths. A test resuts were systematicay underestimated. The predictions of the test resuts on singe SWRP are rather we, but in a conservative manner. The pre-impregnated SWRP aminates have a bigger deviation, so that adapted mode parameters had to be proposed. The tests were made on test specimens with a sma bonding ength. As concuded in the sensitivity anaysis, ony the parameter k can be varied. Figeys proposed to decrease parameter k [8] : for singe SWRP, k can be taken 1. mm. For pre-impregnated doube SWRP, cacuations with k = 1.0 mm give improved resuts. In tabe, the transferabe force, cacuated with the CFRP-mode (F CFRP ) and with the adapted mode parameters

9 FRPRCS for SWRP (F SWRP ), are compared. To evauate the parameter C f, additiona experiments are needed for confirmation, especiay tests with a onger bonding ength. 5. BENDING When a beam is strengthened, the required section is cacuated from the equiibrium of interna forces. A sufficient bonding ength can be provided to ensure that the forces can be introduced. However, the beam can fai in a different faiure mode [9]: debonding of the externa reinforcement, deamination, pate end shear faiure, peeing-off of aminates at intermediate ocations In this section the deamination is investigated for appications with SWRP. 5.1 Deamination This faiure occurs when the beam end remains unstrengthened, e. g. when the support rests on a coumn. At the beginning of the reinforcement, stress concentrations cause the externa reinforcement to pee off. Three stresses act at the pate end: τ,s, s n, see Fig. 9. Additiona interfacia shear stresses are deveoped on top of the Jourawski shear stresses. Aso, the pate experiences a norma stress, s n, which pees the aminate away from the beam. Maek, Sadaatmanesh and Ehsani [10] derived the Equations (14), (15) and (16) to describe these stresses. These are derived from the equiibrium of forces, Fig. 9, assuming inear eastic and isotropic materias, perfect bond between pate and concrete, a inear strain distribution through the fu depth of the section, and no interaction between shear strains and norma strains. dσ ( x) τ ( x) = h dx (14) d σ ( ) G 1 G x g g 1 E σ ( x) ( x) = σ c dx hg h E hg h E Ec (15) 4 d σ n ( x) 4 K n + 4β σ ( x) 4 n = dx Ec I c q( x) (16) in which σ (x) norma stress in the externa reinforcement [N/mm²] σ c (x) norma stress at the bottom of the concrete section [N/mm²] τ (x) shear stresses in the externa reinforcement [N/mm²] σ n (x) norma stresses acting between the concrete and pate [N/mm²] E E-moduus of the externa reinforcement [N/mm²] E c E-moduus of the concrete [N/mm²] E g E-moduus of the adhesive [N/mm²] I inertia moment of the externa reinforcement [mm 4 ]

10 75 Figeys et a. I c inertia moment of the concrete [mm 4 ] h thickness of the externa reinforcement [mm] h g thickness of the adhesive [mm] G g shear moduus of the adhesive [N/mm²] q(x) distributed oad on the concrete beam [N/mm] β = K nb E I [1/mm] (17) 4 4 E g K n = [N/mm³] (18) h g These equations are soved at the pate end [10]. The stresses are given by eq. (1), () and (3), given an interna bending moment M (Equation (19)) and a shear force V (Equation (0)), Fig. 10: 0 ) 1 x 0 M ( x = a + a x + a (19) 0 0 ) = dx dm ( x ) V ( x = a x + a (0) The stresses are: σ ( x) = b sinh( ωx) cosh( ωx + b x + b x b (1) in which [ ] ) + [ sinh( ωx) cosh( x) ] + h [ b x ] τ ( b () σ x) = b3hω ω 1 + n = ω K ( x) = β g n Gg 1 h y M E E h c 1 = m a 1 I tr b b b c 0 I c e βx [ cos( βx) sin( βx) ] q( x) E I + b E I c c (3) (4) (5) yc = m ( a10 + ) (6) I a tr 3 y b1 = m (7) ω c ( a10 + a0 + a3 ) + I tr 0 = a10 + a0 a3 M + 0 unpated ength, Fig. 10 y c distance between the centre of the transformed section and the bottom of the beam [mm] I tr inertia moment of the beam [mm 4 ]

11 FRPRCS One coud derive that a peak shear stress (τ max ) and a peak norma peeing stress (σ nmax ) occur at the aminate end. Stresses are higher when the unpated ength 0 increases. If these two stresses make a critica combination, deamination wi take pace. A possibe faiure criterion is the Mohr-Couomb criterion. The Mohr Couomb ine is given as Equation (8)[1]. From this vaue, the maximum oad of the beam at deamination can be derived. 1 f cm τ = f cm f ctm 1 + ( 1) f ctm σ f cm (8) σ norma stress in an infinitesima part of the beam [N/mm²] τ shear stress in an infinitesima part of the beam [N/mm²] f cm mean compressive strength of concrete [N/mm²] f ctm mean tensie strength of concrete [N/mm²] 5. Experiments Five beams are examined in a four point bending test. The test setup is given in Fig. 11. The span of the beam is 1500mm. The cross section of the beam is 5mm by 15mm. The beam has a ength of 1700mm. Interna reinforcement (BE 500) is provided: 3 ø8mm as tensie reinforcement, ø6mm as compressive reinforcement and stirrups of ø6mm every 100mm. Concrete cover is 0mm. One beam remains unstrengthened and acts as a reference. Three beams are strengthened with one ayer of SWRP (A = 47,67mm²), the ast beam is strengthened with two ayers SWRP. The beams have an unpated ength 0 of 100mm or 50mm. The test specimens are isted in tabe 3. The beams were tested in a oad controed testing device. The oad increases with steps of 5 kn (rate: 5 kn/minute). After each step the mid span defection is measured. At both sides of the beam demec strain gauges are appied, Fig. 1, so that strain can be foowed during the test. Aso, the crack pattern is recorded. The reference beam faied by yieding of the interna stee. This occurs at a oad of 53 kn. A strengthened beams fai through deamination, e.g. beam 1L100b in Fig. 13. A test resuts are given in tabe Mode versus experiments A beam can fai through concrete crushing or yieding of the interna reinforcement. The oad at which this kind of faiure occurs can be easiy cacuated from the equiibrium of forces. In a cases, the interna tensie reinforcement yieds first. This happens at a oad of 57 kn for the unstrengtened beam, 87 kn for beams strengthened with one ayer SWRP and 119 kn for the beam strengthened with two ayers, tabe 4. These oads have to be reduced because of deamination. E.g. a beam strengthened with one ayer SWRP and an unstrengtened ength of 100mm wi fai at a oad of 77 kn. The deamination oads are isted in tabe 4. Remarkabe is the sma oad, 34 kn, when the unpated ength becomes arge, 50 mm. This means that the beam wi act as an unstrengthened beam for oads higher than 34 kn. The beam shoud fai through yieding of the interna reinforcement at a oad of 57 kn.

12 754 Figeys et a. The maximum oads of the beams 1L100a, L100 and the reference beam, are predicted we, with an error of respectivey 4%, 4% and 7%. Beam 1L100b has a arger deviation (17%) but this remains within acceptabe imits. Beam 1L50 fais in a different way than predicted. Theoreticay, the beam shoud fai as an unstrengtened beam because of yieding of the interna reinforcement at a oad of 57kN. However, the experimenta maximum oad is 70kN at deamination. An unpated ength of 50 mm is rather arge with a shear span of 500 mm. This means that the externa reinforcement starts at the midde of the span between the support and the point of appication of the oad. In practice, this is unusua. Possiby, this is caused by the interaction between shear, bending and spreading out of the oad under the appication point. It can be concuded that the described mode predicts the maximum oad quite we, for reduced unpated engths. 6. CONCLUSION SWRP is a new materia that can combine the advantages of stee pates and CFRP. It combines a reativey ow materia cost with a high strength and a fexibe shape. For some appications (wrapping), it is necessary that SWRP has a ow fexura stiffness. Therefore, it is necessary to reduce the stiffness of SWRP. Hence, new types of SWRP are required. Using SWRP as externa reinforcement, the appication with viscous adhesive is difficut. The impregnation is sufficient if accuratey appied. The deveopment of a new adhesive is necessary to improve impregnation. Aso, preimpregnated aminates are an interesting option. In this paper non-inear fracture mechanics is appied to mode and describe the shear-sip behaviour of the bonded connection. Experiments are presented which demonstrate that the new materia behaves stronger and stiffer than eements strengthened with CFRP. Adapted materia dependent parameters for SWRP for design purposes are proposed. Further experiments wi be performed to confirm the genera appicabiity of these mode parameters. When a beam is strengthened, additiona faiure modes have to be considered. One of them is deamination. By means of four point bending tests, it is verified that the mode needs no adaptation when using SWRP. Experiments show that the mode we predicts the maximum oad at reduced unpated engths. 7. ACKNOWLEDGMENT The authors woud ike to thank the Femish Institute for Promotion of Scientific and Technoogica Research in the Industry (IWT Vaams Instituut voor de Bevordering van Wetenschappeijk-Technoogisch Onderzoek in de Industrie) for their financia support and to Bekaert N.V. for the materias and the support.

13 8. REFERENCES FRPRCS [1] Brosens, K., Anchorage of externay bonded stee pates and CFRP aminates for strengthening of concrete eements, doctora thesis, Kathoieke Universiteit Leuven, 001. [] Matthijs, S., Structura Behaviour and Design of Concrete Members Strengthened with Externay bonded FRP Reinforcement,doctora thesis, Universiteit Gent, 000. [3] Hexce Composites, technica data sheet F185, 000. [4] Resipast, Product cataogus betonhersteing, 003, Wommegem. [5] Vokersen, O., 1938, Die Nietkraftverteiung in zugbeanspruchten Nietverbindungen mit konstanten Laschenquerschnitten, Luftfahrtforschung, 15, pg [6] Ranisch, E.H., Zur Trägfahigkeit von Verkebungen zwischen Baustah und Beton- Gekebte Bewehrung, Heft 54, T.U. Braunschweig, 198. [7] Hozenkämpfer, P., Ingenieurmodee des Verbunds gekebter Bewehrung für Betonbauteie, doctoraatsverhandeing, Heft 108, IBMB, Braunschweig, pg 5-86, [8] Figeys, W., Strengthening of reinforced concrete structures with bandweave (in Dutch: Versterking van gewapend beton met bandweefse), Master of Science thesis, Kathoieke Universiteit Leuven, 004. [9] Teng, J.G.; Chen, J.F., FRP strengthened RC structures, John Wiey & Sons, Weinheim, 001. [10] Maek, A., Sadaatmanesh; H., Ehsani, R., Prediction of Faiure Load of R/C Beams strengthened with FRP Pate Due to Stress Concentration at the Pate End, ACI structura Journa, vo. 95, nr, 1998, pg

14 756 Figeys et a.

15 FRPRCS Figure 1 SWRP, top view. Figure SWRP, type studied in this paper, 65x [1x 0.5mm + 18x 0.mm].

16 758 Figeys et a. Figure 3 SWRP, section after pu-off test. Figure 4 Biinear t - s reationship. Figure 5 Spreading out of forces in the concrete.

17 FRPRCS Figure 6 a. Scheme of shear test [1]; b. photograph of test set sampe. Figure 7 Test I150a after faiure.

18 760 Figeys et a. Figure 8 a Comparison between experiment and mode. Parameters singe SWRP: h ref = 40 mm, k = 1.47 mm, C f = 0.40, τ m =.06 N/mm², s m = mm, s 0 = mm.

19 FRPRCS Figure 8 b Comparison between experiment and mode. Parameters pre-impregnated doube SWRP: h ref = 40 mm, k = 1.47 mm, C f = 0.40; τ m =.06 N/mm², s m = mm, s 0 = mm. Figure 9 Stress acting on an externa aminate. Figure 10 Strengthened concrete beam.

20 76 Figeys et a. Figure 11 Test set up four point bending test. Figure 1 Position of Demec strain gauges. Figure 13 Beam 1L100b, after faiure.