as shown in Fig. 1 (b). The geometry o specimens is shown in Table 2. All specimens were coated with epoxy resin, and cured or a week. GFRP tabs were

Transcription

1 コンクリート工学年次論文集,Vol.36,No.2,214 - Technical Paper - EXPERIMENTAL STUDY ON SHEAR BEHAVIOR OF RC BEAMS JACKETED BY FLAX FIBER SHEET Kosuke TAKASAKI *1, Tidarut JIRAWATTANASOMKUL *2, Dawei ZHANG *3 and Tamon UEDA *4 ABSTRACT Flax is planted in rather cold climate or clothing and ood, and can be made into nature iber. The cost o lax iber sheet is signiicantly less than conventional Fiber Reinorced Polymer (FRP) sheet. Flax abric contains lax ibers in two orthogonal directions (warp and wet directions). In this study, beam test or six rectangular beam specimens with FRP sheet and one control specimen are conducted. As results, it is veriied that lax iber strengthening is eective or enhancement o shear behavior. In addition, the result o strengthening beams with lax ibers is discussed by comparing it with the case o beams strengthened with PET ibers.. Keywords: lax iber, FRP jacketing, shear strengthening, coupon test, beam test 1. INTRODUCTION Several Reinorced concrete (RC) structure without properly designed or shear reinorcement usually collapse catastrophically in shear under severe loading. In recent years, Fiber Reinorced Polymer (FRP) sheets have become increasingly in use or shear strengthening RC structures owing to their high stiness, strength-to-weight ratios and design lexibility [1]. Conventional FRPs such as Carbon Fiber Reinorced Polymers (CFRP) and Aramid Fiber Reinorced Polymers (AFRP) are eective in enhancing shear strength, but their applications in structures is limited mainly by their high initial cost and relatively high environmental impact [2]. Instead o those synthetic FRPs, use o natural FRP (NFRP) such as lax FRP has gained popularity in engineering applications since they are more cost eective and environmental riendly while maintaining high mechanical properties comparable to those o Glass Fiber Reinorced Polymers (GFRP) used as reinorcement [3]. Although lax FRP can be applied in composite structures, its structural perormance or strengthening RC structures has not been clearly investigated to ensure its structural application and saety. Flax FRP sheets usually contain ibers in two orthogonal directions, termed as warp and wet directions as shown in Fig. 1. It was ound in this study that tensile strength, elastic modulus and racturing strain are dierent in the dierent iber direction. In the wet direction, their tensile strength and stiness tend to be stronger than that in the warp direction. On the other hand, in the warp direction the racturing strain is higher than that in the wet direction. This research aims or the irst time to study shear strengthening o RC beams jacketed by lax FRP. In this study, tensile test o lax FRP coupons was conducted to investigate their mechanical properties. To examine the structural perormance o RC beams jacketed by lax FRP, in total seven rectangular beam specimens were prepared including one control specimen without lax-frp jacketing. Three specimens were jacketed in which the wet direction was set normal to the member axis, while the other three specimens had the warp direction set normal to the member axis. The number o lax sheet layers is one, two and three or each group o three specimens. All the six specimens with jacketing showed similar shear ailure iediately ater the tensile racture o lax iber sheet along a main shear crack. The enhancement o shear strength could be observed in all jacketed specimens. This act proved that lax FRP can eectively increase shear strength. In addition, the experimental shear strength o RC beams was compared with that obtained rom design equation o JSCE code in order to investigate its applicability. 2. TEST PROGRAMS 2.1 Material test (1) FRP coupon Tensile properties o lax ibers are imperative in order to consider them as reinorcement. To that purpose, coupon test o lax FRP was conducted considering iber alignment in both warp and wet directions with nominal iber mass o 128 and 12 g/m 2, respectively. The properties o the epoxy resin used in the test to uniormly gather the ibers are shown in Table 1. From the lax iber sheet (see Fig. 1 (a)), three coupon specimens with warp and wet iber directions were prepared based on JSCE standard [4], *1 Graduate School o Engineering, Hokkaido University, JCI Member *2 Ph.D., Graduate School o Engineering, Hokkaido University, JCI Member *3 Associate Pro., Graduate School o Engineering, Zhejiang University, JCI Member *4 Pro., Graduate School o Engineering, Hokkaido University, JCI Member

2 as shown in Fig. 1 (b). The geometry o specimens is shown in Table 2. All specimens were coated with epoxy resin, and cured or a week. GFRP tabs were attached at the both edges o specimens to prevent rom slipping between specimen and the wedge grip o the tensile machine. Strain gauges were attached at the mid-height o each specimen. The test setup or a coupon test is shown in Fig. 1 (c). Table 1 Properties o epoxy resin Name No. Bending strength Tensile strength Elastic modulus Dine FF D ,35 WEFT WARP Stress-strain curve o the six coupon specimens is shown in Fig. 2. While WARP specimens showed higher racture strain than WEFT specimens, the latter showed higher strength than the ormer. Thereore, the dierence in iber direction aects the load-carrying capacity and ductility o strengthening o RC members ater strengthening by lax iber jacketing. Stress () Strain (µm) Fig. 2 Stress-strain curve o coupon specimens GFRP tab a) Flax iber Strain gauge 25 (2) Concrete and steel reinorcement The compressive strength was 33 at 14 days using high early strength cement with coarse aggregate having maximum size o 2. Slump value was set to be less than 15. Casting o the beams was made with ready mixed concrete in sti steel molds placed horizontally. Longitudinal reinorcement used in the beam test had diameter o 25 and yielding strength o 539, whereas stirrup used had diameter o 6 and yielding strength o 35. The properties o steel reinorcement are shown in Table 3. Fig. 1 Flax iber, coupon test setup and specimen From the coupon test, tensile strength, elastic modulus and rupture strain were obtained and the suary o results o tensile test or six coupon specimens are shown in Table 2. Coupon specimens or WARP and WEFT were denoted as to and to, respectively. Coupon b) Coupon specimen Table 2 Suary o result o tensile test w t A 2 c) Test setup tu ɛ u µɛ E ,864 13, ,61 15, ,47 15, ,739 18, ,481 2, ,296 2,8 * w = width, t = nominal thickness, A = area, tu = tensile strength, ɛ u = racture strain, and E = elastic modulus o coupon specimens. Type Table 3 Properties o steel reinorcement Diameter Area 2 y E s ɛ y µɛ Longitudinal D , 2,1 Stirrup D , 2, * E s = elastic modulus, y = yielding strength and ɛ y = yielding strain. 2.2 Beam test Ater observing the tensile testing results o lax-frp coupon specimens, loading test or beam specimens with and without FRP jacketing were conducted. All seven beam specimens were loaded statically to their ultimate state by a hydraulic jack. Each specimen had a cross section (b h) o 25 27, whose corners were chamered with a radius o 11 to prevent stress concentration in the lax sheet, and the shear span a was 6, resulting in a shear-span to eective-depth ratio o 2.5. The spacing o stirrups was 15 in all specimens. The longitudinal reinorcement and stirrup ratios were 3.4% and.17%, respectively, in all seven specimens, whereas the volumetric ratio (i.e., calculated using the nominal thickness o the FRP sheets) o the wrapped -119-

3 lax-frp composites varied rom.7 % to.21%. Continuous lax-frp sheets with the main ibers oriented in the transverse direction were ully wrapped around the RC beam. The main direction o lax FRP was in either WARP or WEFT directions while the number o lax-frp layers was the experimental parameters. The details o seven beam specimens are shown in Table 4. Specimen Table 4 Details o beam specimens Layer No. ρ % b d h a Strains o longitudinal reinorcement, stirrups and lax-frp sheet were measured using strain gauges. The strain gauges were located in the region where shear cracks were expected to occur. Displacement at mid-span and supports were measured using Linear Variable Dierential Transducers (LVDT). The characteristics o beam specimens, test setup, and location o strain gauges are shown in Fig. 3. specimen showed concrete crushing along main shear crack with loud noise o crack opening when the peak load was reached. In case o all six specimens with lax-frp jacketing, they all ailed in shear with FRP rupture along shear cracks. According to the test results, the longitudinal reinorcement in the control specimen yielded at the near support. It is evident that shear-crack opening signiicantly aects to the yielding o such longitudinal reinorcement. This phenomenon is coonly known as tension shit eect. Fig. 4 Failure modes beore removing iber sheets 15 Stirrup Longitudinal reinorcement Fig. 5 Failure modes ater removing iber sheets Strain gauge Strain gauges used in V s and V calculation 2D25 D6 4D25 25 Flax FRP Fig. 3 Characteristics o beam specimens, test setup and location o strain gauges 3. EXPERIMENTAL RESULTS Failure modes Failure modes beore and ater removing lax-frp sheets are shown in Figs. 4 and Chamering corner 3.2 Shear-orce and vertical-displacement relationships Relationships between shear orce and vertical displacement at mid-span are shown in Figs. 6, 7 and 8. Figs. 6 and 7 show such relationships o RC beams jacketed by WARP and WEFT FRP sheets, respectively. Fig. 8 shows the shear orce-displacement relationships o all beam specimens. Shear strengths o both WARP and WEFT specimens are enhanced obviously when the number o layers o lax FRP applied increases. Displacements at the peak load also increase gradually when increasing the number o layers o lax FRP. Thereore, wrapping lax iber sheets on RC beams has shear strengthening eect. Comparing WARP with WEFT specimens, maximum shear strength and corresponding displacement o WEFT specimens are higher than those o WARP specimens, as shown in Fig. 8. The enhanced shear strength increases with the

4 amount o lax FRP. The enhancements o shear strength are rom 45% to 72% or WEFT specimens and rom 22% to 67% or WARP specimens. Despite the dierence in the tensile strength and racture strain, both WEFT and WARP specimens show similar eiciency in shear strengthening. Jacketing with lax-frp sheet in WEFT direction would improve more shear strengthening eect or the RC beams than jacketing in WARP direction. ; Shear strength (kn) Displacement () Fig. 6 Relation between shear orce and vertical displacement o WARP specimens ; Shear strength (kn) Fig. 7 Relation between shear orce and vertical displacement o WEFT specimens ; Shear strength (kn) Fig. 8 Relation between shear orce and vertical displacement o RC beams 3.3 Shear orce components Total shear orce (V tot ) consists o shear orces carried by concrete (V c ), stirrup (V s ) and NFRP sheet (V ), as shown in Eq. (1). Shear orces carried by stirrup and FRP sheet are calculated using strains o stirrup and FRP sheet measured along shear crack, as shown in Fig. 3 (see strain gauges in circles). Shear orce carried by concrete (V c ) can be obtained by subtracting the total shear orce with the shear orce carried by stirrup and FRP (V tot -V s -V ). The equation or the total shear orce is show as below. V = V + V + V total c s WARP 1 WARP 2 WARP 3 WEFT 1 WEFT 2 WEFT Displacement () WARP 1 WARP 2 WARP 3 WEFT 1 WEFT 2 WEFT Displacement () (1) Shear orce carried by concrete, stirrup and lax-frp sheets are shown in Fig. 9., (kn),, V (kn),, V (kn),, V (kn) d) WARP 2 e) WEFT 2 24 V V ) WARP 3 g) WEFT 3 8 a) b) WARP 1 c) V 16 V 8 8 V,, V (kn),, V (kn),, V (kn) Fig. 9 Contributions o each shear orce component to total shear orce In all beam specimens, stirrup and FRP sheet started carrying shear orce when irst shear crack exhibited with the load over 1 kn. When the number o layer o FRP sheets increases, shear orce carried by FRP (V ) increases in both o WARP and WEFT specimens. In WEFT specimens, V at the peak load contributes approximately twice as large as that in the WARP specimens. This is because the tensile strength o WEFT specimen is larger than WARP specimen. 3.4 Concrete shear orce As shown in Eq.(1), the concrete shear orce can be calculated rom the total member shear orce once the shear contributions o stirrup and FRP sheets are known rom the analyses on strain readings. The V

5 concrete shear orce in RC members wrapped with lax-frp sheets was ound to have reached its peak value beore the ull development o the member shear strength, as shown in Fig. 1. Concrete shear orce o beam specimens with lax-frp jacketing increases comparing to that o control beam. It can be seen that concrete shear orce increases when increasing the number o FRP layers. This indicates that lax-frp conines and prevents the shear crack opening o concrete, leading to higher concrete contribution to shear resistance. ; Concrete shear orce (kn) Displacement () WARP 1 WARP 2 WARP 3 WEFT 1 WEFT 2 WEFT 3 Fig. 1 Relationship between concrete shear orce and vertical displacement nonlinearly. Moreover, the rate o strain increase in lax-frp sheets increased ater yielding due to the stiness degradation o the stirrups. In addition, the large strain values observed in lax-frp sheets at the peak loads (i.e., 9, 13, µɛ or WEFT specimens and 5, 9, µɛ or WARP specimens) demonstrate the signiicance o using dierence in iber direction o lax-frp sheets and the number o layer applied. The more layers o lax-frp are applied, the higher strain development or maintaining the integrity and ductility o RC members can be observed. Strain o iber (µ) Yielding o stirrup Displacement () Fig. 12 Strain distributions o lax-frp sheets 3.5 Strain development in stirrups and FRP sheets Fig. 11 shows the development o the average strains in stirrups with the vertical displacement at mid-span until the peak load. The average strains o stirrups are the average values o all strain readings on each stirrup across the shear crack (see strain gauges with circles in Fig. 3). In the control specimen, increment o strain development in stirrups is higher than the other specimens jacketed by lax FRP. This might be because the shear crack propagation is aster in control specimen, leading to greater increase in strain. In all the strengthened beams, the stirrups have yielded beore the peak load was reached. Ater yielding point, the increment o strain becomes aster until peak load has been reached. Strain o stirrup (µ) Fig. 12 shows the development o the average strains in lax-frp sheet with the vertical displacement at mid-span until the peak load. An approximately linear increase o the average strains with the vertical displacement was seen beore yielding o stirrups. Ater yielding, the strain increase in lax-frp sheets behaved 4. DISCUSSIONS 4.1 Comparison between lax FRP and PET FRP Since the lax iber sheet has a rather low elastic modulus, the result o RC beam with the same dimension (b h = ), shear span (6 ) and stirrup ratio (.17%) as those lax FRP specimens, jacketed by PET sheet whose elastic modulus is also low [5], was compared with the beams with the lax FRP. The tension reinorcement ratio was slightly higher (4.22 % compared with 3.4 % o specimen with lax FRP). Fig. 13 presents a comparison o three RC beams jacketed by dierent types o FRP jackets, WARP, WEFT and PET FRP sheets, or the same amount o FRP ratio (i.e., ρ = %). It can be seen that with the same amount o FRP ratio, RC members jacketed by lax-frp sheet show the similar shear strength enhancement, namely nonlinear increase until peak-load, although the sheet stiness and tension reinorcement were less. 6 3 Peak load -ρ 4 WARP 3 =.21% Yielding o stirrup 25 -ρ 2 3 =.21% 2 PET-ρ SP4 =.22% Displacement () 1 Fig. 11 Strain increments o stirrups 5 ; Shear strength (kn) Displacement () Fig. 13 Comparison between RC beams jacketed by lax FRP and PET FRP sheets

6 4.2 Applicability o existing design equation To predict shear strength o beam strengthened by lax FRP, shear strength obtained rom experiment and JSCE code [6] are compared. Total shear strength can be calculated rom Eq. (1). The concrete and stirrup contributions to shear strength can be calculated as ollows: V c = " 1/ d 1ρ ( bd ) 3 c = Aw wy(sin αs + cos αs ) z / s where ʹc is the compressive strength o concrete; b is the width o member, d is the eective depth o member; ρ w is the ratio o transverse steel reinorcement; A w is the cross-sectional area o transverse steel reinorcement; wy is the yielding strength o transverse reinorcement; α s is the angle o transverse steel reinorcement to the member's axis; and z is d/1.15; s is the spacing o shear reinorcement. The shear contribution provided by FRP sheet is its tensile capacity and is computed based on the shear reinorcing eiciency o the FRP sheet (K) as shown in Eq. (4). V = KA (sin α + cos α ) z / s u where K= R in which.4 K.8 and ( ρ ) ( u / ) ( 1/ c" ) 1/4 2/3 1/3 R= E E in which.5 R 2. ; A is the cross-sectional area o the FRP sheet; u is the design tensile strength o the FRP sheet (N/ 2 ); s is the spacing o the FRP sheet; E is the modulus o elasticity o the FRP sheet (kn/ 2 ); ρ is the volumetric ratio o the FRP sheet; and α is the angle o the FRP sheet to the member axis. Comparison o shear strength between JSCE code and experiment is shown in Fig. 14. V tot o JSCE code is quite conservative as shown in the igure. This is partly because the equation o JSCE does not consider the coninement eect by jacketing sheet [7]. JSCE code (kn) control WARP WEFT exp. (kn) Fig. 14 Comparison between shear strength rom experiment and JSCE code w Conservativ e (2) (3) (4) 5. CONCLUSIONS (1) Flax iber is an eective material or strengthening o RC beams. WEFT direction is the better way to wrap lax iber on RC structures or shear strengthening. (2) Increasing the number o lax-frp sheet layer can increase the total shear strength and concrete shear strength. The enhancements o shear strength are rom 45% to 72% or WEFT specimens and rom 22% to 67% or WARP specimens. (3) The design equation in JSCE code underestimates the shear strength o beams with lax-frp sheet (4) Comparing with RC members jacketed with PET-FRP sheets, lax-frp sheets, whose stiness is smaller, can enhance shear strength as much as PET-FRP sheet. ACKNOWLEDGEMENT The authors would like to acknowledge Proessor XIAN Guijun at Harbin Institute o Technology, or providing invaluable advice and the lax iber sheet. The authors are also grateul to the Maeda Kosen Co. Ltd., Japan or providing or providing epoxy resins or this study. REFERENCES [1] Vistasp, M. K. and Lei, Z., Use o composites or 21st century civil inrastructure, J. o Comput. Methods Appl. Mech. Engrg., Vol. 185, May, 2, pp [2] A. K. Mohanty, M. Misra and G. Hinrichsen, Bioibres, biodegradable polymers and biocomposites: an overview, J. o Macromol. Matel. Engrg., Vol , Mar., 2, pp [3] Libo Y., Nawawi C. and Krishnan J., Flax ibre and its composites a review, J. o Composites Part B, Vol. 56, Aug., 213, pp [4] JSCE, Test method or tensile properties o continuous iber sheets, JSCE Concrete Library, JSCE-E [5] Jirawattansomkul, T., Dai, J.G., Zhang, D.W., Senda, M. and Ueda, T. (213), Experimental Study on Shear Behavior o Reinorced Concrete Members Fully Wrapped with Large Rupture Strain FRP Composites, J. o Composites or Construction, ASCE, CC [6] JSCE, Recoendations or Upgrading o Concrete Structures with Use o Continuous Fiber Sheets, JSCE Concrete Library, 22, pp [7] Jirawattansomkul, T., Zhang, D. and Ueda, T., Prediction o the post-peak behavior o reinorced concrete columns with and without FRP-jacketing, J o Engineering Structures, Vol.56, May, 213, pp