PERFORMANCE OF PRE-CRACKED RC BEAMS SHEAR STRENGTHENED WITH NSM CFRP LAMINATES

Transcription

1 PERORMANCE O PRE-CRACKED RC BEAMS SHEAR STRENGTHENED WITH NSM CRP LAMINATES Salvador DIAS Assistant Proessor ISISE - University o Minho Azurém, Guimarães, Portugal sdias@civil.uminho.pt Joaquim BARROS ull Proessor ISISE - University o Minho Azurém, Guimarães, Portugal barros@civil.uminho.pt Abstract The strengthening o an existing Reinorced Concrete (RC) structure oten involves concrete in a cracked state. To evaluate the eect o existing cracks on the behavior o RC beams shear strengthened according to the Near Surace Mounted (NSM) technique with Carbon iber Reinorced Polymer (CRP) laminates, an experimental program was carried. NSM CRP laminates were applied in RC beams with or without cracks prior the application o the shear strengthening intervention. The main results o this experimental research are presented and analyzed in terms o the structural behavior o the beams, ailure modes and eectiveness o the NSM technique with CRP laminates. The principal dierence o the behavior o NSM CRP beams with and without pre-cracks can be resumed to an expected loss o initial stiness in the pre-cracked specimens. However, the pre-cracking did not aect the eicacy o the NSM shear strengthening technique in terms o load carrying capacity and ultimate delection. Keywords: CRP laminates, NSM, Pre-cracking, RC beams, Shear strengthening. 1. Introduction A RC beam needs to be shear strengthened when is deiciently reinorced in shear or when its shear capacity alls below its lexural capacity ater lexural strengthening. The shear ailure mode o a RC beam should be avoided since it is brittle and unpredictable. Advanced composite materials like CRP laminates applied according to NSM technique can be used in order to increase the shear resistance o RC beams [1-3]. This technique is based on the introduction o laminates into slits made on the concrete cover o the lateral aces o the beams to be strengthened. The strengthening intervention oten involves concrete elements already cracked. To evaluate the inluence, on the strengthening eectiveness, o already existing cracks when a RC beam is shear strengthened with NSM CRP laminates an experimental program was carried out. The relevant results are presented and analyzed. 2. Experimental program ig. 1 presents the T cross section o the RC beams adopted in the experimental program, the lateral geometry o the type o beam and the steel reinorcement common to all tested beams (nine). The reinorcement systems were designed in order that all beams ail in shear. The dierences between the tested beams are restricted to the shear reinorcement systems applied Page 1 o 8

2 in the L i beam span and the presence, or not, o the pre-cracks in the concrete beore the application o the CRP. The experimental program was made up o our beams with steel stirrups φ6@300mm (ρ sw = 0.10%) and ive beams with steel stirrups φ6@mm (ρ sw = 0.16%). According to Table 1, one NSM CRP shear strengthening coniguration (ive laminates at 45º) was applied in three beams with ρ sw = 0.10% (beams 3S-5LI45, 3S-5LI451 and 3S-5LI452) and in two beams with ρ sw = 0.16% (beams 5S-5LI45 and 5S-5LI45). In the beams 5S-5LI60 and 5S-5LI60, both with ρ sw = 0.16%, it was applied ive NSM CRP laminates at 60º. The 3S-5LI451, 3S-5LI452, 5S-5LI45 and 5S-5LI60 beams were precracked beore have been strengthened. As schematically represented in ig. 1, the laminates were distributed along the AB line, where A represents the beam s support at its test side and B is obtained assuming load degradation at 45º. φ6//65 (horizontal stirrups) Steel plate φ10//50 (vertical stirrups) A = 1 L i 45º B 1 15x80 (stirrups φ8) = 1 L r φ10//50 (vertical stirrups) φ6//65 (horizontal stirrups) Steel plate 6φ12 φ8//80 in L r φ6//150 in Li and φ6//75 in L r φ32+2φ (lateral concrete cover = 22 mm) ig. 1 - Geometry o the type o beam, steel reinorcements common to all beams, support and load conditions (dimensions in mm). d = 360 Table 1. CRP shear strengthening conigurations o the tested beams. Beams ρ sw (%) a Pre-cracking Laminates CRP angle [θ ] CRP spacing [s ] CRP percentage [ρ ] (º) b (mm) (%) c 3S-5LI No 3S-5LI Yes 3S-5LI Yes S-5LI No 2 5 5S-5LI Yes 5S-5LI No 5S-5LI Yes a 3S-R is the erence beam without CRP or the beams with ρ sw = 0.10% (ig. 2) and 5S-R is the erence beam without CRP or the beams with ρ sw = ρ = 2a b b s sinθ where 0.16% (ig. 2); b Angle between the CRP iber direction and the beam axis; c The CRP percentage was obtained rom ( ) ( ) a = 1.4 mm and b = 9.5 mm are the dimensions o the laminate cross section, and b w = 180 mm is the beam web width. The three point beam bending tests (ig. 1) were carried out using a servo closed-loop control equipment, taking the signal read in the displacement transducer (LVDT), placed at the loaded section, to control the test at a delection rate o 0.01 mm/second. To prevent brittle spalling o the concrete cover at the supports, the beam ends were strengthened by conining the concrete with a two-directional cage o φ6@65mm horizontal stirrups and φ10@50mm vertical stirrups (ig. 1). To overcome the diiculties to bend φ32 mm longitudinal tensile bars, their ends were welded to steel plates. With the purpose o obtaining the strain variation along the three laminates (CRP A, CRP B and CRP C) that have the highest probability o providing the largest contribution or the shear strengthening o the RC beam, strain gauges (SG_L) were bonded in these laminates according to the arrangements represented in Table 2. The localization o the SG_L in the precracked beams was governed by the crack pattern ormed at the end o the pre-cracking test. Theore, in order to obtain the imum strain values, strain gauges were positioned in the interceptions with the shear cracks. In each o the tested beam, one steel stirrup was monitored with three strain gauges (SG_S). The location o the monitored laminates and stirrups in the tested beams is represented in ig. 2. w Page 2 o 8

3 Reerence beams NSM beams with ρ sw = 0.16% Beam 3S-R Beam 5S-5LI45 Beam 5S-5LI Beam 5S-R 4x300 (stirrups φ6) 15x80 (stirrups φ8) Beam 5S-5LI60 Beam 5S-5LI60 6x (stirrups φ6) 15x80 (stirrups φ8) Beam 3S-5LI45 Beam 3S-5LI451 Beam 3S-5LI452 CRP C 1 6x (stirrups φ6) CRP B 1 NSM beams with ρ sw = 0.10% monitored stirrup CRP A 15x80 (stirrups φ8) x (stirrups φ6) 15x80 (stirrups φ8) Legend (see exemple in 3S-5LI45 beam): monitored stirrup or monitored laminate x300 (stirrups φ6) 225 monitored laminates 1 15x80 (stirrups φ8) Monitored laminates: CRP A: nearest the loaded section; CRP B between the CRP A and CRP C ig. 2 - Localization o the steel stirrups (continuous line) and CRP laminates (dashed line) in the tested beams (dimensions in mm). Table 2 - Position o the strain gauges in the monitored CRP laminates. Position o the strain gauges in the monitored laminates a Beams CRP A CRP B CRP C SG_L4 SG_L5 3S-5LI S-5LI S-5LI S-5LI S-5LI S-5LI S-5LI S-5LI451 3S-5LI452 CRP A CRP B CRP C SG_L4 SG_L5 CRP A CRP B CRP C SG_L4 SG_L5 5S-5LI45 CRP A CRP B CRP C SG_L4 SG_L5 5S-5LI60 CRP A CRP B CRP C SG_L4 a or each monitored laminate the value in this table ers to the length (in centimeters) o the laminate that is counted rom the top the web beam to the position o the SG. or the beams without pre-cracks the spacing between SG s is: L/3 (CRP A and B) and L/5 (CRP B) where L is length (in centimeters) o the laminate. Page 3 o 8

4 The concrete compressive strength at the date o beam testing was evaluated rom uniaxial compression tests with cylinders (150 mm diameter and 300 mm height) according to EN [4], and the average value obtained was 59.4 MPa. The properties o the steel bars (Table 3) were obtained rom uniaxial tensile tests, carried out according to EN [5]. or the laminates (S&P Laminates CK 150/0), uniaxial tensile tests were carried out according to the ISO recommendations [6], rom which the ollowing average values were obtained: imum tensile strength = MPa, Young s modulus = MPa, ultimate tensile strain = 1.63%. The MBrace Resin 220 was used to bond the laminates to the concrete. This type o adhesive was tested by Bonaldo et al. [7] and the average values obtained in terms o imum tensile strength, Young s modulus and ultimate tensile strain were 33 MPa, 7470 MPa and 4.83%, respectively. Table 3 - Properties o the steel bars (average values). Property φ6 φ8 φ12 φ16 (type 1) φ16 (type 2) φ32 Yield stress (MPa) Tensile strength (MPa) Experimental program 3.1 Pre-cracking test As already mentioned, prior to the application o the NSM CRP laminates, our RC beams (3S-5LI451, 3S-5LI452, 5S-5LI45 and 5S-5LI60) were loaded up to a shear crack pattern was ormed. or this purpose, and taking into account the behavior o the 3S-R and 5S-R erence beams (these beams were previously tested up to ailure), a test stop criterion o a delection o 3 mm at the loaded section was adopted. This value is about 30% o the delection corresponding to l/250, which is the imum allowed delection or serviceability limit states according to the Eurocode [8], where l is the beam span length. Considering the above mentioned stop criterion, the imum load applied in the precracking test was kn, kn, kn and kn or the 3S-5LI451, 3S- 5LI452, 5S-5LI45 and 5S-5LI60 beams, respectively. a) Ater the pre-cracking test b) Ater marking the position o the slits/laminates c) Ater opening the slites d) Ater the aplication o the CRP laminates ig. 3 - Strengthening activities ater the pre-cracking test. Page 4 o 8

5 Ater the pre-cracking test, the strengthening activities were executed with the beams in the unloaded state (ig. 3). In ig. 4 are shown the cracking pattern o the beams ater the precracking test, and the adopted NSM CRP shear strengthened conigurations. The dierences between 3S-5LI451 and 3S-5LI452 beams are restricted to the number o laminates crossing the shear crack ormed during the pre-cracking test (three in 3S-5LI451 beam and two in 3S-5LI452 beam). 3S-5LI451 3S-5LI452 5S-5LI45 5S-5LI60 ig. 4 - Pre-cracked RC beams shear strengthened with NSM CRP laminates. 3.2 Test up to ailure The orce-displacement diagrams (-u) in the loaded section obtained or the tested beams are reported in ig. 5. Assuming that Δ =, being and the imum orce o the erence beam (3S-R or 5S-R) and o the shear strengthened beam, respectively, the Δ ratio was evaluated. The values or, Δ, and the delection at loaded section corresponding to ( u ) are included in Table (a) 700 (b) orce (kn) S-R orce (kn) S-R 5S-5LI S-5LI45 3S-5LI S-5LI60 5S-5LI45 0 3S-5LI Delection at loaded-section (mm) 0 5S-5LI Delection at loaded-section (mm) ig. 5 - orce vs delection at the loaded-section or the tested beams with the lower (a) and higher (b) percentage o steel stirrups. Page 5 o 8

6 Table 4 - Relevant results. Beams (kn) Δ (%) u (mm) 3S-R ε CRP (%) 3S-5LI S-5LI S-5LI S-R S-5LI S-5LI S-5LI S-5LI ig. 5 shows that the adopted NSM CRP shear strengthening conigurations provided an increase in terms o stiness and in terms o imum load (between 35% and 49%). urthermore, the NSM CRP shear strengthening conigurations provided an increase in terms o delection at the loaded section in correspondence to ( u ) that ranged between 42% and 78%. The main dierence between the behavior o strengthened beams with or without pre-cracks resides in an expected loss o initial stiness in the pre-cracked specimens (up to the imum load applied in the pre-cracking test). This dierence was more evident in the beams with ρ sw = 0.10%. Above the load corresponding to the end o the pre-cracking test, the structural perormance o 3S-5LI451 and 5S-5LI45 beams was slightly higher than the respective beams without a pre-cracking test (3S-5LI45 and 5S-5LI45, respectively). The imum load ( ) o 3S-5LI452 and 5S-5LI60 beams had similar values to the respective uncracked strengthened beams (3S-5LI45 and 5S-5LI60, respectively). The better perormance o the 3S-5LI451 beam when is compared with that o the 3S-5LI452 beam can be justiied by the number o laminates crossing the shear ailure crack (the same that was ormed during the pre-cracking test). The obtained results showed that the eicacy o the NSM shear strengthening technique with CRP laminates is not negatively aected by the presence o a crack pattern that may exists when a strengthening intervention is needed. Table 4 also compares the values o the imum strain recorded in the monitored laminates up to imum load ( ε CRP ) in the beams with NSM CRP laminates. The values o CRP has ranged between 1.20% (3S-5LI45 beam) and 1.54% (5S-5LI45 beam). These values correspond to 74% and 94% o the CRP ultimate strain obtained in the uniaxial tensile tests o the laminates (ε u = 1.63%) and conirm the high level o mobilization o the CRP in the tested beams. A very important aspect o the eectiveness o the NSM shear strengthening technique with CRP laminates, regarding the analyzed beams, is the capacity o this technique to mobilize the yield stress o the stirrups beore the imum load o the strengthened beams has been attained. As expected, all tested beams ailed in shear in the L i shear span. or the erence beams (3S-R and 5S-R), the imum load was attained when one stirrup crossing the shear ailure crack has ruptured. Debond through the laminate-adhesive interace (laminate sliding) was the ε Page 6 o 8

7 predominant ailure mode o the tested beams. The ailure o the 5S-5LI45 beam occurred with the rupture o the intermediate laminate. 4. Conclusions The adopted NSM CRP shear strengthening conigurations (ρ = %), applied in RC beams with an average concrete compressive strength o about 60 MPa, provided an average increase o the imum load ( ) and o the delection at the loaded section in correspondence to ( u ) equal to 40% and 53%, respectively. Due to the relatively high-strength concrete used, the resistance to the concrete racture propagation during the debond process o the laminates crossing the critical diagonal crack has contributed to signiicantly mobilize the tensile capacity o the CRP laminates. In act, the imum strain recorded in the laminates up to the imum load has ranged between 74% and 94% o the CRP ultimate strain obtained in the uniaxial tensile tests o the laminates. The main dierence o the behavior o NSM CRP beams with and without pre-cracks resides in an expected loss o initial stiness in the pre-cracked beams. In these beams the mobilization o the CRP laminates started just ater the opening process o the pre-cracks, while the mobilization o the CRP laminates in the non pre-cracked beams only occurred when the shear crack has ormed. However, the pre-cracking did not aect the eicacy o the NSM shear strengthening technique in terms o load carrying capacity and ultimate delection. 5. Acknowledgements The authors wish to acknowledge the support provided by the Empreiteiros Casais, Degussa, S&P and Secil (Unibetão, Braga). The study reported in this paper is part o the research project PTDC/ECM/114511/9, supported by the Portuguese oundation or Science and Technology (CT). 6. Reerences [1] BARROS, J., DIAS, S., Near surace mounted CRP laminates or shear strengthening o concrete beams, Journal Cement & Concrete Composites, Vol. 28, No. 3, March 6, pp [2] KOTYNIA, R., Shear strengthening o RC beams without NSM CRP laminates, 8 th International Symposium on iber Reinorced Polymer Reinorcement or Concrete Structures (RPRCS-8), July 7, 10 pp. (CD-ROM). [3] EL-HACHA, R., WAGNER, M., Shear Strengthening o Reinorced Concrete Beams using Near-Surace Mounted CRP Strips, 9 th International Symposium on iber Reinorced Polymers Reinorcement or Concrete Structures (RPRCS-9), July 9, 4 pp. (CD-ROM). [4] EN 206-1, Concrete - Part 1: Speciication, perormance, production and conormity, European standard - CEN, 0, 69 pp. [5] EN , Metallic materials - Tensile testing. Part 1: Method o test (at ambient temperature), European standard - CEN, 1990, 35 pp. [6] ISO 527-5, Plastics - Determination o tensile properties - Part 5: Test conditions or unidirectional ibre-reinorced plastic composites, International Organization or Standardization (ISO), 1997, 7 pp. Page 7 o 8

8 [7] BONALDO, E., BARROS, J., LOURENÇO, P., Steel ibre reinorced concrete and CRP laminate strips or high eective lexural strengthening o RC slabs, Technical report 05-DEC/E-14 Department o Civil Engineering, University o Minho, October 5, 103 pp. [8] EN , Eurocode 2: Design o concrete structures Part 1-1: General rules and rules or buildings, European standard - CEN, 4, 225 pp. Page 8 o 8