- Email: [email protected]
An experimental study on strengthening of masonry infilled RC frames using diagonal CFRP strips
Available online at www.sciencedirect.com
Composites: Part B 39 (2008) 680–693 www.elsevier.com/locate/compositesb
An experimental study on strengthening of masonry infilled RC frames using diagonal CFRP strips ¨ zgu¨r Anil *, M. Emin Kara, Mustafa Kaya Sinan Altin, O Civil Engineering Department, Gazi University, 06570 Maltepe, Ankara, Turkiye Received 27 September 2006; received in revised form 16 May 2007; accepted 2 June 2007 Available online 17 August 2007
Abstract The purpose of this study was to investigate experimentally the behavior of strengthened masonry infilled reinforced concrete (RC) frames using diagonal CFRP strips under cyclic loads. Ten test specimens were constructed and tested under cyclic lateral loading. Specimens were constructed as 1/3 scale, one-bay, one-storey perforated clay brick-infilled nonductile RC frames. The aspect ratio (lw/hw, where lw is the infill length and hw is the infill height) of masonry-infilled wall was 1.73. CFRP strips were applied with different widths and with three different arrangements such as on both sides (i.e. symmetrically) and on the interior side or the exterior side of the masonry walls. This experimental study investigated the effects of CFRP strips’ width and arrangement type on specimens’ behavior. Strength, stiffness and storey drifts of the test specimens were measured. Test results indicated that, CFRP strips significantly increased the lateral strength and stiffness of perforated clay brick infilled nonductile RC frames. Specimens receiving symmetrical strengthening showed higher lateral strength and stiffness. Specimens at which CFRP strips of the same width were applied to one of the interior or exterior surface of the infill wall showed similar lateral strength and stiffness. 2007 Elsevier Ltd. All rights reserved. Keywords: A. Carbon fibre; B. Debonding; B. Stress concentrations; Brick masonry infills
1. Introduction A great number of reinforced concrete (RC) structures were severely damaged or totally collapsed during earthquakes that occurred all around the world. For those structures constructed with insufficient strength and ductility, the required stiffness was not provided. Nowadays, a large number of structures in use have similar deficiencies such as flexible columns, soft stories, strong beam–weak column joints, nonseismic reinforcement details for earthquake loading, and low concrete strength. Because of the high risk of earthquake damage and the potential for great loss of life, a large number of RC buildings require extensive strengthening. In strengthening of those kinds of structures, different techniques have been developed and applied *
Corresponding author. Tel.: +90 312 231 74 00x2215; fax: +90 312 230 84 34. ¨ . Anil). E-mail address: [email protected] (O 1359-8368/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.compositesb.2007.06.001
in practice. Many researchers have investigated the strengthening techniques such as introduction of RC infill walls, precast panels, steel bracing, and concrete jacketing of frame, which can be used for strengthening of the seismically deficient structures. Among the available techniques, the addition of RC infill walls was found to be the most feasible strengthening technique for medium rise RC buildings in many countries and Turkey [1–4]. RC infills increased the lateral load capacity of the RC frame and reduced to lateral drift at ultimate load. While this upgrading technique is effective, it requires a great deal of preparation work, its construction may disturb the ongoing building functions, and the new structural elements may affect the architectural aesthetics of the building. Also these techniques add considerable mass leading to higher seismic loads. Hence, an alternative method of retrofitting is worth considering. During the last decade the use of fiber reinforced polymers (FRP) for retrofitting and strengthening became a valid alternative because of their small thickness,
S. Altin et al. / Composites: Part B 39 (2008) 680–693
681
Nomenclature dw diagonal length of infill ECFRP elastic modulus of CFRP Emasonry elastic modulus of masonry fc compression strength of concrete, mortar and plaster fmasonry compression strength of masonry unit fw infill height lw infill length n CFRP strip number Vcal calculated lateral strength capacities of specimens
and relative ease of application. FRP’s not only have the advantage of very high strength over conventional materials, but also are light weight and highly durable in many environments. The light weight of FRP makes rehabilitation techniques much easier in constricted spaces. In addition there is no need to have large equipment for FRP application. The strength and stiffness of a structure can be increased with very little increase in mass, distinctly advantageous from the seismic perspective. Carbon fiber reinforced polymers (CFRP) have been widely used to strengthen RC structures such as bridge girders, piers, beams, columns, slabs, beam–column joints as well as masonry structures. Numerous tests have been conducted around the world to examine the behavior of structural members strengthened by using CFRP sheets. In these studies, researchers investigated the effect of the type, amount, and pattern of FRP [5–23]. These tests showed that this technique of strengthening is an effective and convenient method to improve the member strength and/or stiffness. CFRP can be used to increase the flexural capacity, and also increase the shear strength and stiffness of RC beams [5–10]. Similarly it is shown that the RC columns strengthened with CFRP jackets have higher strength and stiffness [11–13]. Strengthening of RC beam–column joints with CFRP sheets showed that shear strength of the joints were improved significantly [14–16]. CFRP strips applied to tension regions of RC slabs increased the punching shear strength of the slabs significantly [17,18]. Literature contains different types of experimental studies including seismic strengthening or rehabilitation of entire RC systems using CFRP. In addition, studies were made on strengthening individual RC members such as beams, columns, etc. by using CFRP laminates or strips. One of these studies was about repairing a full scale RC building at which beam–column joints was wrapped by using CFRP laminates [19]. The repaired structure displacement capacity was increased significantly without loosing strength. The structure was dissipated significant amount of energy without showing any crucial damage at strengthened elements. Another experimental study investigated a two storey structure with single bay frames. Speci-
Vmasonry calculated equivalent diagonal strut capacities of masonry wall VCFRP calculated diagonal CFRP strip tension capacities of specimens tCFRP thickness of CFRP strip wCFRP width of CFRP strip eCFRP measured strain values of CFRP strip h main diagonal angle of masonry wall
mens were constructed with seismic deficiencies encountered in typical clay tile infilled frames, and were strengthened using different patterns of CFRP sheets. Test of specimens were carried out under cyclic lateral loading [20]. The authors stated that this technique was found to be an efficient, practical and economical solution for the strengthening of seismically deficient buildings. Experiments showed that when the CFRP strips were connected correctly with the frames as well as infill wall and new lateral load carrying system was generated, the lateral load carrying capacity was increased and storey drift ratio of original frame was reduced significantly. One other study [21] compares the strengthening techniques of RC frames with addition of RC infill wall or masonry infill wall strengthened with diagonal CFRP strips. In this study, two-storey three-bay undamaged frames were strengthened with two techniques and were tested under lateral cyclic loading. Test results showed that, both of the strengthening techniques provide approximately same lateral strength, and the capacity of the frame using CFRP strips depends on the amount and effectiveness of CFRP anchors. When the literature was surveyed, few studies were encountered about the behavior of strengthened masonry infilled RC frames using CFRP strips [22,23]. In addition, the question about the effective width of CFRP strips that were used for strengthening was not answered yet. The new Turkish seismic regulation advised the effective width of CFRP strip be equal to width of equivalent compression diagonal of infill wall [24]. An investigation of the effect on CFRP strip width to lateral strength and stiffness of specimen was necessary for safe and economical strengthening. The main objective of this study was to develop alternative strengthening techniques for reinforced concrete buildings, which could be applied with minimum disturbance to the occupants. One-third scale one storey, one bay perforated clay brick infilled test frames were designed such that they reflect the common deficiencies of the major building stock in many countries. The specimens were strengthened with diagonal CFRP strips and then tested under lateral cyclic loads. CFRP strips with three different widths were applied on the masonry infill walls with three different
682
S. Altin et al. / Composites: Part B 39 (2008) 680–693
arrangements: on both sides (i.e. symmetrically), and on the interior side or the exterior side of the masonry walls. The effect of CFRP strips’ width and application type on specimens’ lateral strength, stiffness and storey drift ratio were, investigated. 2. Experimental program 2.1. Test specimens Dimensions and reinforcement details of the test frames are given in Fig. 1. In all test specimens, the geometric dimensions and reinforcement patterns of the frames were identical. The test frame was a 1/3 scale, one-bay, one-storey nonductile RC frame. This test frame was detailed and constructed purposely with some deficiencies commonly observed in residential buildings in many countries, such as inadequate lateral stiffness, weak column–strong beam joints. Insufficient confinement of concrete was provided at column and beam ends and no confinement was provided at beam–column joints. The ties used in beams and columns of the test specimens had 90 hooks at their free ends. The columns were 100 · 150 mm and the beams were 150 · 150 mm. In the columns four 8 mm diameter plain bars and in the beams six 8 mm diameter plain bars were
used as longitudinal reinforcement. Plain bars with a diameter of 4 mm spaced at 100 mm were used as ties in both the beams and columns. All the frames were infilled with 65 · 95 · 95 mm 1/3 scaled perforated clay tiles, and infill walls were not constructed on the symmetry axis of the frame for simulating exterior walls of the building. The ratio of net area to the gross area of the clay tiles was about 0.56. Clay tiles were laid such that their holes were oriented horizontally. Then both sides of the wall were plastered with the same material used as the mortar. The thickness of the plaster that was used at two faces of brick walls was 7.5 mm. The total thickness of the clay tile with plaster was 80 mm. Finally, diagonally placed CFRP strips were used for strengthening the perforated clay tile infill wall. The aspect ratio (lw/hw, where lw is the infill length and hw is the infill height) of frame with masonry wall was 1.73. The specimens’ properties were given in Table 1. Specimen 1 was manufactured with masonry infill wall as reference member. The experimental parameters included the width and the application type of the CFRP strips. The CFRP configurations that were oriented diagonally were given in Fig. 2. Two-hundred mm wide CFRP strips were used for Specimens 2, 5 and 8 (0.13dw, where dw is the diagonal length of infill), 300 mm wide CFRP strips were used for Specimens 3, 6 and 9 (0.20dw), and finally 400 mm wide
Fig. 1. Dimensions and reinforcement details of test frames.
S. Altin et al. / Composites: Part B 39 (2008) 680–693
683
Table 1 Test specimens Specimen no.
Frame fc (MPa)
Mortar plaster fc (MPa)
CFRP strip
1 2 3 4 5 6 7 8 9 10
17.3 18.0 16.8 17.1 18.2 17.8 17.7 17.0 17.5 17.6
4.0 3.8 4.2 3.7 3.6 4.1 3.9 3.8 3.5 3.7
Reference specimen with masonry wall 200 – 13 Both sides of masonry 300 – 20 wall 400 – 27 200 – 13 Interior side of the 300 – 20 masonry wall 400 – 27 200 – 13 Exterior side of the 300 – 20 masonry wall 400 – 27
a
Width (mm) – ratioa
Application
Ratio: (width of CFRP · 100)/diagonal length of masonry infill wall.
CFRP strips were used for Specimens 4, 7 and 10 (0.27dw). CFRP strips were applied symmetrically on both sides of the infill walls of Specimens 2, 3, and 4. Specimens 5, 6 and 7 received CFRP strips on the interior side only, whereas Specimens 8, 9 and 10 received CFRP strips on the exterior side only. Three different arrangements were used in total. Cross-marks in Fig. 2 indicates anchorage points of the CFRP strips in the infill walls. The surfaces where the CFRP strips were applied on were marked on the specimens. The lengths of the CFRP strips were determined from the specimens by measurement. CFRP strips were anchored in both the infill wall and the RC frame. Anchorage details are given in Fig. 3. Due to the fact that the infill wall was not constructed on the symmetry axis of the RC frame, anchorages were applied in two different types: one for the internal face of the infill wall, and one for the external. For the former case, first, 14 mm diameter holes were drilled on the frame columns as well as on the beam, and cleaned with compressed air. The ends of the CFRP strips were divided into 100 mm wide segments and wrapped over a 12 mm diameter bar. Then, these wrapped ends were inserted into the holes which were subsequently filled with epoxy resin. The preparation stages of the end section of the 200 mm wide CFRP strip are shown in Fig. 4 before anchoring it into the frame. Two, three and four anchorages were used for 200 mm, 300 mm and 400 mm wide CFRP strips, respectively. The CFRP anchor dowels which were used in the exterior face of the RC frame members were formed from 30 · 240 mm2 carbon fiber strips. These strips were rolled in the shorter direction, tied with ordinary fibers and folded. Steel wires were placed in the dowels to place them in the holes easily. The preparation stages of the exterior face CFRP strips anchors are shown in Fig. 5. 300 mm long CFRP sheet with the same width of the CFRP strips were bonded to both of the anchorage regions for preventing and minimizing stress concentration. In order to achieve a good connection between the infill wall and the CFRP strips, additional special CFRP anchor dowels were used [20]. The CFRP anchor dowels which were used in the perforated clay brick infills were made
Fig. 2. CFRP strip configuration used in specimens.
from 50 · 200 mm2 carbon fiber strips. These strips were rolled and tied with ordinary fibers. Infills were drilled with a 14 mm diameter drill and CFRP anchors were extended from one face of the infill to the other face. The locations of the masonry infill anchor dowels are given as cross marks in Fig. 2. The surfaces where CFRP strips were applied on were marked and brushed clean using a wire brush; dust was removed by wiping with clothes. Then 1.5 mm thick Sikadur 330 epoxy was applied on the cleaned surfaces. Before applied epoxy was hardened, the anchorages of the CFRP strips were inserted into holes that were filled with Sikadur 330 epoxy. Then Sikawrap 230-C fiber sheets were laid
684
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Fig. 4. RC frame interior face CFRP anchor photo.
Fig. 3. Details of CFRP anchorages.
according to the specifications given by the manufacturers. Finally, anchorages at the infill walls were applied. During these applications extensive care was taken for not to change fiber directions. Fig. 5. RC frame exterior face CFRP anchor photo.
2.2. Materials Specimen frames were constructed with a low compressive strength concrete to simulate the concrete strength of existent buildings. The concrete strength of the test frames was 17.5 MPa on average and the elastic modulus was 30,000 MPa on the day of testing. The yield strength values of 4, 8, 10 and 16 mm diameter plain bars are 326, 410, 475 and 450 MPa, respectively. The mix designs for the mortar used in the construction of the brick walls and for the plaster were identical. Mix
proportions are given in Table 2. Average compressive strength of the mortar used in the construction of the masonry walls of the specimens was found to be equal to 3.8 MPa. The average compressive strength of the masonry unit in the direction of its holes was calculated as fmasonry = 6.6 MPa considering the gross area of the brick. Modulus of elasticity of the perforated masonry clay unit was Emasonry = 3629 MPa. Properties of unidirectional Sikawrap 230-C CFRP sheets and Sikadur 330 epoxy used in this study are given in Table 3.
S. Altin et al. / Composites: Part B 39 (2008) 680–693 Table 2 Mixture design of mortar and plaster Material
Percentage by weight (%)
0–3 mm aggregate Cement Lime Water
61.0 10.5 10.5 18.0
Table 3 Properties of CFRP Sikawrap 230-C (unidirectional) and Resin Sikadur 330 Properties of CFRP
Remarks of CFRP
Construction
Areal weight (g/m2) Density (g/m3) Thickness (mm) Tensile strength (MPa) Elastic modulus (MPa) Ultimate tensile strain (%)
Warp: carbon fibers (99% of total areal weight); Weft: thermoplastic heat-set fibers (1% of total areal weight) 220 ± 10 1.78 · 106 0.12 4100 231,000 1.7
Properties of resin
Remarks of resin
Tensile strength (MPa) Elastic modulus (MPa)
30 3800
685
floor through high strength steel bolts. Specimens were tested under cyclic lateral loading. Lateral load was applied to specimens at the beam level, using a hydraulic jack and the applied load was measured with a load cell (the capacities being 600 kN in compression and 300 kN in tension). In the tests, in both push and pull half cycles, 10 kN lateral load was applied at the beginning, and each cycle was repeated two times. The lateral load was increased by 10 kN at each cycle up to the ultimate load level of specimen. After reaching specimen’s lateral load carrying capacity, tests were completed. During the test, the storey displacements and the lateral loads were monitored. Axial load was applied to the columns by prestressing tendons as shown in Fig. 6. The level of applied axial load in each test was about 10% of the axial load capacity of the frame columns. A rigid steel frame was constructed around the test specimen to prevent out of plane displacements. In the instrumentation of the specimens, linear variable differential transducers (LVDTs) were used for displacement measurements. In addition, the strains of CFRP strips were measured using strain gauges. Strain-gauges placement at the CFRP strips used in specimens is shown in Fig. 6. 3. Experimental results 3.1. Behavior of test specimens
2.3. Test setup and instrumentation Details of the test setup, the loading system, and the instrumentation are shown in Fig. 6. The foundation of the test specimen was anchored to the laboratory’s strong
Lateral load–displacement hysteretic curves for the test specimens are shown in Figs. 7–10. As indicated in these figures, diagonal CFRP strips increased the lateral strength and the stiffness significantly. Reference specimen reached
Fig. 6. Details of test setup, loading system, and instrumentation.
686
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Fig. 7. Load–displacement curve of specimen 1 (Reference).
0.40% lateral drift at ultimate load. After the infill was crushed at the upper corners due to diagonal compression, the specimen lost its lateral load carrying capacity and thus, failed. Specimens 2, 3 and 4 that were strengthened with CFRP strips from both sides of the infill wall showed similar behavior and failure modes. CFRP strips at the interior side of the infill wall were ruptured due to tension (Fig. 11), and the anchorages of the CFRP strip at the exterior side of the infill wall were ruptured from the upper corner of the frame due to shearing. CFRP strips were ruptured at ultimate load level, and specimens lost their lateral load carrying capacities and stiffness abruptly. Storey drift ratio of Specimen 2 was 0.57% at ultimate load, and Specimens 3 and 4 had two times greater storey drift ratio than that of the reference specimen. Specimens that were strengthened non symmetrically (i.e. received CFRP strips on either the exterior or the interior sides of the infill wall) showed storey drift ratios between 0.50% and 0.61% at the ultimate load level and failed due to debonding of the CFRP strip from the upper corner of the infill wall (Fig. 12) at same storey drift ratios as can be seen in Figs. 9 and 10. CFRP strips were debonded at the ultimate load levels of the specimens. In addition, infill wall joints were crushed at the upper corners of the infill wall. After debonding occurred during next cycle, specimens failed due to rupture of the CFRP anchors at the upper corner of the frame (Fig. 13). The load level at which CFRP anchorages failed was lower than the ultimate load levels for all specimens. 4. Discussion of test results 4.1. Strength and stiffness The test results are summarized, and presented in Table 4. This table was prepared for illustrating the effect of applied strengthening technique on the ultimate strength, initial stiffness and lateral storey drift ratio.
Fig. 8. Load–displacement curves of Specimens 2, 3, and 4 (CFRP strips are applied on both faces of the wall).
The ratio of the ultimate lateral strength of Specimens 2, 3 and 4 that were strengthened with CFRP strips from both sides of infill wall to that of the reference specimen ranged between 2.18 and 2.61. The respective ratio values for the specimens that were strengthened with CFRP strips from one side of the infill wall varied between 1.57 and 1.85. Specimens receiving CFRP strips of the same width on one side of the infill wall had approximately the same ultimate lateral strength. On average, specimens receiving CFRP strips on one side of the infill wall had 29% less
S. Altin et al. / Composites: Part B 39 (2008) 680–693
687
Fig. 9. Load–displacement curves of Specimens 5, 6, and 7 (CFRP strips are applied on interior face of the wall).
Fig. 10. Load–displacement curves of Specimens 8, 9, and 10 (CFRP strips are applied on exterior face of the wall).
ultimate lateral strength than specimens receiving CFRP strips on both side of the infill wall. Response envelopes shown in Fig. 14 are plotted by connecting the peak points of the load–displacement hysteretic curves for each specimen. Response envelope curves can be used for evaluating the strength and stiffness characteristics of the specimens and also general behaviors. As can be seen from this figure, strength and stiffness of strengthened
frames were significantly higher than those of the reference specimen. All Seismic Codes provide similar limits to prevent extensive structural and non-structural damage and to minimize the second order effects. Turkish seismic code specifies the interstorey drift limit as 0.35% for the RC. framed systems used in this study, yet this value in Eurocode 8 regulations, for brittle nonsrructural infills in contact with the RC frame, is taken equal to 0.5%.
688
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Fig. 11. CFRP strip rupture photo (Specimen 3).
Fig. 12. CFRP strip debonding photo (Specimen 5).
The 0.35% limit drift ratio was marked on the response envelope curves shown in Fig. 14. There was no significant drop of the lateral load carrying capacity of the strengthened specimens up to this limit. Regarding the stiffness, this is true for specimens 2, 3, and 4, but not for the remaining ones. Furthermore, stiffness degradation (i.e. from the initial one to the one corresponding to a drift ratio of 0.35%) is more pronounced for specimens 8, 9, and 10 than for 5, 6, and 7. This can be attributed to the different anchorage detailing that the CFRP strips had depending on whether they were applied on the exterior or on the interior side
of the infill wall. Measured storey drift ratios for all strengthened specimens at ultimate load were greater than the limit drift ratio that was suggested by the regulations. The initial stiffnesses of the test specimens are given in Table 4. Initial stiffness was defined as the initial slope of the load–displacement curve in the first push half cycle. Diagonal CFRP strips increased initial stiffness of the specimens significantly. When the CFRP strips’ width was increased, initial stiffness also increased. Initial stiffnesses of specimens 2, 3, and 4 that were strengthened with CFRP strips from both sides of the infill wall were 4.0–6.4 times
S. Altin et al. / Composites: Part B 39 (2008) 680–693
689
Fig. 13. CFRP anchor rupture photo (Specimen 8).
Table 4 Summary of test results Specimen no.
1 (reference) 2 3 4 5 6 7 8 9 10 a b
Ultimate load (kN) Push
Pull
76.7 167.0 187.2 200.4 114.0 131.5 139.5 118.3 131.0 142.1
67.4 161.0 183.3 199.6 120.7 136.9 140.6 111.0 127.0 136.6
Ratioa
Drift ratio at ultimate load (%)
Initial stiffness (kN/mm)
Ratiob
1.00 2.18 2.44 2.61 1.57 1.78 1.83 1.54 1.71 1.85
0.40 0.57 0.80 0.82 0.55 0.50 0.61 0.50 0.52 0.58
50.00 200.00 300.00 320.00 190.48 250.00 265.00 196.22 266.67 285.00
1.00 4.00 6.00 6.40 3.81 5.00 5.30 3.92 5.33 5.70
Ratio of ultimate load of strengthened infilled frame to ultimate load of reference specimen. Ratio of initial stiffness of strengthened infilled frame to that of the reference specimen; initial stiffness was calculated as using first push half cycles.
greater than that of the reference specimen. The stiffnesses of specimens’ that were strengthened with CFRP strips from one side of the infill wall were from 3.81 to 5.70 times greater than that of the reference specimen. Should the initial stiffnesses of specimens with same CFRP strip width compared, the initial stiffness of the specimens receiving symmetrically strengthened were between 2% and 17% greater than that of specimens at which CFRP strips were applied on one side of the infill wall. Only there was no difference between specimens that were strengthened with diagonal CFRP strips from one side of infill wall. When the width of diagonal CFRP strips was increased from 0.13dw to 0.20dw, initial stiffness of the specimens receiving CFRP strips on both sides of infill wall was increased by 50%, and that of specimens receiving on one side of the infill wall were increased between 31% and
36%. It is important to note that, independent of the application method, for CFRP strips with 0.27dw width, initial stiffness of the specimens were only 7% greater than the specimens at which 0.20dw width CFRP strips were used. 4.2. Strains of CFRP strips Typical examples of strain measurements that were taken from different CFRP strips are given in Fig. 15 for Specimens 3, 5, 6 and 9. In this study, as an application technique CFRP strips bonded without removing the plaster and strain values were measured in this condition. As can be seen from Fig. 15, strain values that were measured from specimens at which CFRP strips were applied on both sides of the infill wall were greater than that of specimens at which CFRP strips were applied on one side of the infill
690
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Maximum tensile strain values that were measured from CFRP strips are shown in Fig. 16. CFRP strips that were bonded on the interior side of Specimens 2, 3 and 4 infill walls were ruptured at a strain level ranging between 0.0042 and 0.0054. Lateral displacement of the frame and cracks initiated at the masonry infill wall was caused direction changes in CFRP fibers. For this reason, CFRP strips that were affected by not only axial force but also transverse loads were failed at lower strain values. Specimens at which CFRP strips were applied from one side of infill wall had maximum tension strain values between 0.0028 and 0.0032. These values were corresponding to failure strains at which CFRP anchorages were ruptured. Average tensile strain values were marked on Fig. 16 for specimens that were strengthened with CFRP strips from both and one sides. Average tensile strain values that were obtained from specimens strengthened with CFRP strips form both sides were 1.6 times greater than that of specimens strengthened with CFRP strips from one sides. 4.3. Comparison of experimental and analytical results Comparisons of calculated and experimental ultimate lateral strength of specimens and ratios of them are presented in Table 5. Infill walls were taken as equivalent compression struts in analytical model. Lateral strength of strengthened specimens was calculated by superposition of the compression strength of the equivalent compression strut and the tension strength of the CFRP strip. The lateral strength (Vcal) of specimens was calculated using Eq. (1). V cal ¼ cos h ðV masonry þ V CFRP Þ
Fig. 14. Response envelopes of specimens.
wall. However the width of CFRP strips did not have significant effect on the measured strains. Diagonal CFRP strips of specimens strengthened with strips from one side of the infill wall were debonded from upper corner of infill wall under compression load at 0.002 average strains. Debonding occurred at each push and pull cycles at the upper corner of CFRP strip under compression.
ð1Þ
where Vcal is the calculated lateral strength capacities of specimens, Vmasonry the calculated equivalent diagonal strut capacities of masonry wall, VCFRP the calculated diagonal CFRP strip tension capacities of specimens and h is the main diagonal angle of masonry wall. The width of the equivalent diagonal compression strut was calculated according to FEMA356 regulation as 165 mm [25]. Material properties that were used in the analytical calculations were taken from the test results that were performed in the laboratory. Masonry unit’s compression strength and elastic modulus were determined according to FEMA356 regulation. Analytical calculations were performed using these values taking into account both clay brick, and mortar together. Tension strength of CFRP strips along fiber direction was calculated using Eq. (2). Strain values that were measured from experiments were used as strains in equations. V CFRP ¼ n ðeCFRP ECFRP W CFRP tCFRP Þ
ð2Þ
where n is the CFRP strip number, ECFRP the elastic modulus of CFRP, VCFRP the calculated diagonal CFRP strip tension capacities of specimens, tCFRP the thickness of CFRP strip, wCFRP the width of CFRP strip and eCFRP is the strain of CFRP strip.
S. Altin et al. / Composites: Part B 39 (2008) 680–693
691
Fig. 15. Typical shear force–strain curve of Specimens 3, 5, 6 and 9.
Fig. 16. Measured maximum strains of CFRP strip.
Reference specimen experimental strength was calculated with analytical approach very closely. All of the ana-
lytically calculated lateral strengths of strengthened specimens were smaller than the experimental values. On
692
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Table 5 Comparison of experimental and calculated ultimate load capacities Specimen no.
Experimental ultimate load (kN)
Calculated ultimate load (kN)
Ratioa
1 (reference) 2 3 4 5 6 7 8 9 10
76.7 167.0 187.2 200.4 120.7 136.9 140.6 118.3 131.0 142.1
75.5 121.6 144.7 169.7 89.9 97.1 104.3 89.9 97.1 104.3
1.02 1.37 1.29 1.18 1.34 1.41 1.35 1.32 1.35 1.36
a
Ratio of experimental ultimate load values to calculated ultimate load.
average, experimental strengths of Specimens 2, 3 and 4 that were strengthened with CFRP strips from both sides of the infill wall were 28% greater than the analytically calculated values. Experimental strengths of specimens that were strengthened with CFRP strips from one sides of the infill wall were 35% greater than the analytically calculated values on average. 5. Conclusions In the study presented, 1/3 scale, one-bay, one-storey nonductile RC frames with masonry infill walls were strengthened by diagonal CFRP strips and experimentally investigated under cyclic lateral loading. The following conclusions can be drawn in the light of experimental study. (1) The diagonal CFRP strips that were used to retrofit brick masonry infilled reinforced concrete frames were effective in increasing lateral strength and lateral stiffness significantly. An important advantage of this technique is that it can be applied without evacuating the building during its application, thus causing minimum disturbance to the occupants. The success of this technique depends mainly on the successful application of anchorages between the CFRP strips and the infill wall or frame. (2) Lateral strengths of specimens receiving CFRP strips on both sides of the infill wall increased by 2.18 and 2.61 times compared to specimens receiving CFRP strips on one side only. The increases in lateral stiffness for the same specimens were 4.00 and 6.00 times. Ultimate lateral load increase for the specimens receiving CFRP strips on one side of infill were 1.57 and 1.85 times, and stiffness increases were from 3.81 and 5.70 times. Ultimate lateral strength and stiffness of the specimens that were strengthened with same CFRP strips width from one side of infill wall were approximately the same. (3) When the width of CFRP strip was increased, increase in strength and stiffness could be limited. If CFRP strip width was increased from 0.13dw to 0.20dw, 50% stiffness increase was obtained for the
specimens receiving CFRP strips from both sides of infill wall, and the stiffness increased between 31% and 36% compared to specimens strengthened with CFRP strips from one side of infill wall. (4) Storey drift ratio of all specimens was obtained to be greater than 0.35% limit value that was suggested by the new Turkish seismic regulations. There was no significant degradation in lateral load carrying capacity of all specimens up to this limit value. This is the case for the stiffnesses of specimens 2, 3, and 4, but not for the remaining ones. Furthermore, stiffness degradation is more pronounced for specimens 8, 9, and 10 than for 5, 6, and 7. Different anchorage detailing that the CFRP strips had depending on whether they were applied on the exterior or on the interior side of the infill wall caused this difference. (5) As an application technique CFRP strips bonded without removing the plaster, and strain values were measured in this condition. Strain values obtained from CFRP strips that were bonded to both sides of infill wall were 1.6 times greater than that of bonded to one side of infill wall. However width of CFRP strips did not have significant effect on measured strains. CFRP strips that were bonded to both sides of infill wall failed with rupture. Stress concentrations that were encountered at CFRP anchorages caused the premature failure of anchorages with rupture. Therefore, rectangular CFRP sheets that were used for preventing stress concentration at the region of anchorages showed limited success. (6) Analytically calculated lateral load carrying capacities were obtained to be 33% less than the experimental results. Authors believe that, the suggested analytical approach can give beneficial ideas to designers for the strengthening of masonry infilled RC frames by using CFRP strips, and the behavior of this complicated composite structure can be simulated quite accurately.
Acknowledgements This study was conducted at the Structural Mechanics Laboratory of Gazi University. The research is supported by the Scientific and Technical Research Council of Turkey (TUBITAK) through Research Grand 105M250 is great fully acknowledged. In addition, authors specially thanks to Mr. Kadir Basßogˇlu and Aslan Masonry Inc. References ¨ zsoy S. The behavior and strength of infilled frame. [1] Ersoy U, U TUBITAK MAG-205 Report, 1971, Ankara, Turkey. 95pp. [in Turkish]. [2] Altin S, Ersoy U, Tankut T. Hysteretic response of reinforced concrete infilled frames. ASCE, J Struct Eng 1992;118(8):2133–50.
S. Altin et al. / Composites: Part B 39 (2008) 680–693 ¨ zcebe G, Ersoy U. Rehabilitation of reinforced [3] Sonuvar MO, O concrete frames with reinforced concrete infills. ACI, Struct J 2004;101(4):494–500. ¨ zcebe G. Contribution of reinforced concrete [4] Canbay E, Ersoy U, O infills to seismic behavior of structural system. ACI, Struct J 2003;100(5):637–43. [5] Diagana C, Li A, Gedalia B, Dlemas Y. Shear strengthening effectiveness with CFF strips. Eng Struct 2003(25):507–16. [6] Taljsten B. Strengthening concrete beams for shear with CFRP sheets. Constr Build Mater 2003(17):15–26. [7] Khalifa A, Nanni A. Rehabilitation of rectangular simply supported RC beams with shear deficiencies using CFRP composites. Constr Build Mater 2002(16):135–46. [8] Taljsten B, Elfgren L. Strengthening concrete beams for shear using CFRP materials: evaluation of different application methods. Compos Part B 2000(33):87–96. [9] Kachlakev D, McCurry CC. Behavior of full scale reinforced concrete beams retrofitted for shear and flexural with FRP laminates. Compos Part B 2000(31):445–52. [10] Li A, Diagana C, Delmas Y. CFRP contribution to shear capacity of strengthened RC beams. Eng Struct 2001(23):1212–20. [11] Saadarmanesh H, Ehsani MR, Jin L. Strength and ductility of concrete columns externally reinforced with fiber composite straps. Struct J ACI 1994;91(4):434–47. [12] Chaallal O, Shahawy M, Hassan M. Performance of axially loaded short rectangular columns strengthened with carbon fiber-reinforced polymer wrapping. J Compos Constr 2003;7(3):200–8. [13] Ye L, Yue Q, Zhao S, Li Q. Shear strength of reinforced concrete columns strengthened with carbon-fiber reinforced plastic sheet. J Struct Eng ASCE 2002;128(12):1527–34. [14] Ghobarah A, Said A. Shear strengthening of beam–column joints. Eng Struct 2004;24:881–8.
693
[15] Prota A, Nanni A, Manfred G, Cosenza E. Selective upgrade of under designed RC beam–column joints using CFRP. Struct J ACI 2004;101(5):699–707. [16] Antonopulos CP, Triantafillou TC. Analysis of FRP-strengthening RC beam–column joints. J Compos Constr 2002;6(1):41–51. [17] Robertson IN, Johnson G. Repair of slab–column connections using epoxy and carbon fiber reinforced polymer. J Compos Constr 2004;8(5):376–83. [18] Limam O, Nguyen VY, Foret G. Numerical and experimental analysis of two way slabs strengthened with CFRP strips. Eng Struct 2005:841–5. [19] Balsamo A, Colombo A, Manfredi G, Negro P, Prota A. Seismic behavior of a full scale RC frame repaired using CFRP laminates. Eng Struct 2005(27):769–80. ¨ zcebe G, Ersoy U, Tankut T, Erduran E, Keskin RSO, Mertol HC. [20] O Strengthening of brick infilled RC frames with CFRP. SERU_ Structural Engineering Research Unit, Rep. No. 2003/1, TUBITAKMETU, Ankara, Turkey; 2003. _ Akyu¨z U, Ersoy U, O ¨ zcebe G. An experimental study on [21] Erdem I, two different strengthening techniques for RC frames. Eng Struct 2006:1843–51. [22] El Gawady MA, Lestuzzi P, Badoux M. Aseismic retrofitting of unreinforced masonry walls using FRP. Compos Part B: Eng 2006:148–62. [23] Cecchi A, Milani G, Trelli A. In plane loaded CFRP reinforced masonry walls mechanical characteristics by homogenization procures. Compos Sci Technol 2004(64):2097–112. [24] Turkish Seismic Code 4-25, Turkey; 2006. [25] FEMA 356, Prestandard and commentary for the seismic rehabilitation of buildings. ASCE, Federal Emergency Management Agency; November 2000.
Composites: Part B 39 (2008) 680–693 www.elsevier.com/locate/compositesb
An experimental study on strengthening of masonry infilled RC frames using diagonal CFRP strips ¨ zgu¨r Anil *, M. Emin Kara, Mustafa Kaya Sinan Altin, O Civil Engineering Department, Gazi University, 06570 Maltepe, Ankara, Turkiye Received 27 September 2006; received in revised form 16 May 2007; accepted 2 June 2007 Available online 17 August 2007
Abstract The purpose of this study was to investigate experimentally the behavior of strengthened masonry infilled reinforced concrete (RC) frames using diagonal CFRP strips under cyclic loads. Ten test specimens were constructed and tested under cyclic lateral loading. Specimens were constructed as 1/3 scale, one-bay, one-storey perforated clay brick-infilled nonductile RC frames. The aspect ratio (lw/hw, where lw is the infill length and hw is the infill height) of masonry-infilled wall was 1.73. CFRP strips were applied with different widths and with three different arrangements such as on both sides (i.e. symmetrically) and on the interior side or the exterior side of the masonry walls. This experimental study investigated the effects of CFRP strips’ width and arrangement type on specimens’ behavior. Strength, stiffness and storey drifts of the test specimens were measured. Test results indicated that, CFRP strips significantly increased the lateral strength and stiffness of perforated clay brick infilled nonductile RC frames. Specimens receiving symmetrical strengthening showed higher lateral strength and stiffness. Specimens at which CFRP strips of the same width were applied to one of the interior or exterior surface of the infill wall showed similar lateral strength and stiffness. 2007 Elsevier Ltd. All rights reserved. Keywords: A. Carbon fibre; B. Debonding; B. Stress concentrations; Brick masonry infills
1. Introduction A great number of reinforced concrete (RC) structures were severely damaged or totally collapsed during earthquakes that occurred all around the world. For those structures constructed with insufficient strength and ductility, the required stiffness was not provided. Nowadays, a large number of structures in use have similar deficiencies such as flexible columns, soft stories, strong beam–weak column joints, nonseismic reinforcement details for earthquake loading, and low concrete strength. Because of the high risk of earthquake damage and the potential for great loss of life, a large number of RC buildings require extensive strengthening. In strengthening of those kinds of structures, different techniques have been developed and applied *
Corresponding author. Tel.: +90 312 231 74 00x2215; fax: +90 312 230 84 34. ¨ . Anil). E-mail address: [email protected] (O 1359-8368/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.compositesb.2007.06.001
in practice. Many researchers have investigated the strengthening techniques such as introduction of RC infill walls, precast panels, steel bracing, and concrete jacketing of frame, which can be used for strengthening of the seismically deficient structures. Among the available techniques, the addition of RC infill walls was found to be the most feasible strengthening technique for medium rise RC buildings in many countries and Turkey [1–4]. RC infills increased the lateral load capacity of the RC frame and reduced to lateral drift at ultimate load. While this upgrading technique is effective, it requires a great deal of preparation work, its construction may disturb the ongoing building functions, and the new structural elements may affect the architectural aesthetics of the building. Also these techniques add considerable mass leading to higher seismic loads. Hence, an alternative method of retrofitting is worth considering. During the last decade the use of fiber reinforced polymers (FRP) for retrofitting and strengthening became a valid alternative because of their small thickness,
S. Altin et al. / Composites: Part B 39 (2008) 680–693
681
Nomenclature dw diagonal length of infill ECFRP elastic modulus of CFRP Emasonry elastic modulus of masonry fc compression strength of concrete, mortar and plaster fmasonry compression strength of masonry unit fw infill height lw infill length n CFRP strip number Vcal calculated lateral strength capacities of specimens
and relative ease of application. FRP’s not only have the advantage of very high strength over conventional materials, but also are light weight and highly durable in many environments. The light weight of FRP makes rehabilitation techniques much easier in constricted spaces. In addition there is no need to have large equipment for FRP application. The strength and stiffness of a structure can be increased with very little increase in mass, distinctly advantageous from the seismic perspective. Carbon fiber reinforced polymers (CFRP) have been widely used to strengthen RC structures such as bridge girders, piers, beams, columns, slabs, beam–column joints as well as masonry structures. Numerous tests have been conducted around the world to examine the behavior of structural members strengthened by using CFRP sheets. In these studies, researchers investigated the effect of the type, amount, and pattern of FRP [5–23]. These tests showed that this technique of strengthening is an effective and convenient method to improve the member strength and/or stiffness. CFRP can be used to increase the flexural capacity, and also increase the shear strength and stiffness of RC beams [5–10]. Similarly it is shown that the RC columns strengthened with CFRP jackets have higher strength and stiffness [11–13]. Strengthening of RC beam–column joints with CFRP sheets showed that shear strength of the joints were improved significantly [14–16]. CFRP strips applied to tension regions of RC slabs increased the punching shear strength of the slabs significantly [17,18]. Literature contains different types of experimental studies including seismic strengthening or rehabilitation of entire RC systems using CFRP. In addition, studies were made on strengthening individual RC members such as beams, columns, etc. by using CFRP laminates or strips. One of these studies was about repairing a full scale RC building at which beam–column joints was wrapped by using CFRP laminates [19]. The repaired structure displacement capacity was increased significantly without loosing strength. The structure was dissipated significant amount of energy without showing any crucial damage at strengthened elements. Another experimental study investigated a two storey structure with single bay frames. Speci-
Vmasonry calculated equivalent diagonal strut capacities of masonry wall VCFRP calculated diagonal CFRP strip tension capacities of specimens tCFRP thickness of CFRP strip wCFRP width of CFRP strip eCFRP measured strain values of CFRP strip h main diagonal angle of masonry wall
mens were constructed with seismic deficiencies encountered in typical clay tile infilled frames, and were strengthened using different patterns of CFRP sheets. Test of specimens were carried out under cyclic lateral loading [20]. The authors stated that this technique was found to be an efficient, practical and economical solution for the strengthening of seismically deficient buildings. Experiments showed that when the CFRP strips were connected correctly with the frames as well as infill wall and new lateral load carrying system was generated, the lateral load carrying capacity was increased and storey drift ratio of original frame was reduced significantly. One other study [21] compares the strengthening techniques of RC frames with addition of RC infill wall or masonry infill wall strengthened with diagonal CFRP strips. In this study, two-storey three-bay undamaged frames were strengthened with two techniques and were tested under lateral cyclic loading. Test results showed that, both of the strengthening techniques provide approximately same lateral strength, and the capacity of the frame using CFRP strips depends on the amount and effectiveness of CFRP anchors. When the literature was surveyed, few studies were encountered about the behavior of strengthened masonry infilled RC frames using CFRP strips [22,23]. In addition, the question about the effective width of CFRP strips that were used for strengthening was not answered yet. The new Turkish seismic regulation advised the effective width of CFRP strip be equal to width of equivalent compression diagonal of infill wall [24]. An investigation of the effect on CFRP strip width to lateral strength and stiffness of specimen was necessary for safe and economical strengthening. The main objective of this study was to develop alternative strengthening techniques for reinforced concrete buildings, which could be applied with minimum disturbance to the occupants. One-third scale one storey, one bay perforated clay brick infilled test frames were designed such that they reflect the common deficiencies of the major building stock in many countries. The specimens were strengthened with diagonal CFRP strips and then tested under lateral cyclic loads. CFRP strips with three different widths were applied on the masonry infill walls with three different
682
S. Altin et al. / Composites: Part B 39 (2008) 680–693
arrangements: on both sides (i.e. symmetrically), and on the interior side or the exterior side of the masonry walls. The effect of CFRP strips’ width and application type on specimens’ lateral strength, stiffness and storey drift ratio were, investigated. 2. Experimental program 2.1. Test specimens Dimensions and reinforcement details of the test frames are given in Fig. 1. In all test specimens, the geometric dimensions and reinforcement patterns of the frames were identical. The test frame was a 1/3 scale, one-bay, one-storey nonductile RC frame. This test frame was detailed and constructed purposely with some deficiencies commonly observed in residential buildings in many countries, such as inadequate lateral stiffness, weak column–strong beam joints. Insufficient confinement of concrete was provided at column and beam ends and no confinement was provided at beam–column joints. The ties used in beams and columns of the test specimens had 90 hooks at their free ends. The columns were 100 · 150 mm and the beams were 150 · 150 mm. In the columns four 8 mm diameter plain bars and in the beams six 8 mm diameter plain bars were
used as longitudinal reinforcement. Plain bars with a diameter of 4 mm spaced at 100 mm were used as ties in both the beams and columns. All the frames were infilled with 65 · 95 · 95 mm 1/3 scaled perforated clay tiles, and infill walls were not constructed on the symmetry axis of the frame for simulating exterior walls of the building. The ratio of net area to the gross area of the clay tiles was about 0.56. Clay tiles were laid such that their holes were oriented horizontally. Then both sides of the wall were plastered with the same material used as the mortar. The thickness of the plaster that was used at two faces of brick walls was 7.5 mm. The total thickness of the clay tile with plaster was 80 mm. Finally, diagonally placed CFRP strips were used for strengthening the perforated clay tile infill wall. The aspect ratio (lw/hw, where lw is the infill length and hw is the infill height) of frame with masonry wall was 1.73. The specimens’ properties were given in Table 1. Specimen 1 was manufactured with masonry infill wall as reference member. The experimental parameters included the width and the application type of the CFRP strips. The CFRP configurations that were oriented diagonally were given in Fig. 2. Two-hundred mm wide CFRP strips were used for Specimens 2, 5 and 8 (0.13dw, where dw is the diagonal length of infill), 300 mm wide CFRP strips were used for Specimens 3, 6 and 9 (0.20dw), and finally 400 mm wide
Fig. 1. Dimensions and reinforcement details of test frames.
S. Altin et al. / Composites: Part B 39 (2008) 680–693
683
Table 1 Test specimens Specimen no.
Frame fc (MPa)
Mortar plaster fc (MPa)
CFRP strip
1 2 3 4 5 6 7 8 9 10
17.3 18.0 16.8 17.1 18.2 17.8 17.7 17.0 17.5 17.6
4.0 3.8 4.2 3.7 3.6 4.1 3.9 3.8 3.5 3.7
Reference specimen with masonry wall 200 – 13 Both sides of masonry 300 – 20 wall 400 – 27 200 – 13 Interior side of the 300 – 20 masonry wall 400 – 27 200 – 13 Exterior side of the 300 – 20 masonry wall 400 – 27
a
Width (mm) – ratioa
Application
Ratio: (width of CFRP · 100)/diagonal length of masonry infill wall.
CFRP strips were used for Specimens 4, 7 and 10 (0.27dw). CFRP strips were applied symmetrically on both sides of the infill walls of Specimens 2, 3, and 4. Specimens 5, 6 and 7 received CFRP strips on the interior side only, whereas Specimens 8, 9 and 10 received CFRP strips on the exterior side only. Three different arrangements were used in total. Cross-marks in Fig. 2 indicates anchorage points of the CFRP strips in the infill walls. The surfaces where the CFRP strips were applied on were marked on the specimens. The lengths of the CFRP strips were determined from the specimens by measurement. CFRP strips were anchored in both the infill wall and the RC frame. Anchorage details are given in Fig. 3. Due to the fact that the infill wall was not constructed on the symmetry axis of the RC frame, anchorages were applied in two different types: one for the internal face of the infill wall, and one for the external. For the former case, first, 14 mm diameter holes were drilled on the frame columns as well as on the beam, and cleaned with compressed air. The ends of the CFRP strips were divided into 100 mm wide segments and wrapped over a 12 mm diameter bar. Then, these wrapped ends were inserted into the holes which were subsequently filled with epoxy resin. The preparation stages of the end section of the 200 mm wide CFRP strip are shown in Fig. 4 before anchoring it into the frame. Two, three and four anchorages were used for 200 mm, 300 mm and 400 mm wide CFRP strips, respectively. The CFRP anchor dowels which were used in the exterior face of the RC frame members were formed from 30 · 240 mm2 carbon fiber strips. These strips were rolled in the shorter direction, tied with ordinary fibers and folded. Steel wires were placed in the dowels to place them in the holes easily. The preparation stages of the exterior face CFRP strips anchors are shown in Fig. 5. 300 mm long CFRP sheet with the same width of the CFRP strips were bonded to both of the anchorage regions for preventing and minimizing stress concentration. In order to achieve a good connection between the infill wall and the CFRP strips, additional special CFRP anchor dowels were used [20]. The CFRP anchor dowels which were used in the perforated clay brick infills were made
Fig. 2. CFRP strip configuration used in specimens.
from 50 · 200 mm2 carbon fiber strips. These strips were rolled and tied with ordinary fibers. Infills were drilled with a 14 mm diameter drill and CFRP anchors were extended from one face of the infill to the other face. The locations of the masonry infill anchor dowels are given as cross marks in Fig. 2. The surfaces where CFRP strips were applied on were marked and brushed clean using a wire brush; dust was removed by wiping with clothes. Then 1.5 mm thick Sikadur 330 epoxy was applied on the cleaned surfaces. Before applied epoxy was hardened, the anchorages of the CFRP strips were inserted into holes that were filled with Sikadur 330 epoxy. Then Sikawrap 230-C fiber sheets were laid
684
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Fig. 4. RC frame interior face CFRP anchor photo.
Fig. 3. Details of CFRP anchorages.
according to the specifications given by the manufacturers. Finally, anchorages at the infill walls were applied. During these applications extensive care was taken for not to change fiber directions. Fig. 5. RC frame exterior face CFRP anchor photo.
2.2. Materials Specimen frames were constructed with a low compressive strength concrete to simulate the concrete strength of existent buildings. The concrete strength of the test frames was 17.5 MPa on average and the elastic modulus was 30,000 MPa on the day of testing. The yield strength values of 4, 8, 10 and 16 mm diameter plain bars are 326, 410, 475 and 450 MPa, respectively. The mix designs for the mortar used in the construction of the brick walls and for the plaster were identical. Mix
proportions are given in Table 2. Average compressive strength of the mortar used in the construction of the masonry walls of the specimens was found to be equal to 3.8 MPa. The average compressive strength of the masonry unit in the direction of its holes was calculated as fmasonry = 6.6 MPa considering the gross area of the brick. Modulus of elasticity of the perforated masonry clay unit was Emasonry = 3629 MPa. Properties of unidirectional Sikawrap 230-C CFRP sheets and Sikadur 330 epoxy used in this study are given in Table 3.
S. Altin et al. / Composites: Part B 39 (2008) 680–693 Table 2 Mixture design of mortar and plaster Material
Percentage by weight (%)
0–3 mm aggregate Cement Lime Water
61.0 10.5 10.5 18.0
Table 3 Properties of CFRP Sikawrap 230-C (unidirectional) and Resin Sikadur 330 Properties of CFRP
Remarks of CFRP
Construction
Areal weight (g/m2) Density (g/m3) Thickness (mm) Tensile strength (MPa) Elastic modulus (MPa) Ultimate tensile strain (%)
Warp: carbon fibers (99% of total areal weight); Weft: thermoplastic heat-set fibers (1% of total areal weight) 220 ± 10 1.78 · 106 0.12 4100 231,000 1.7
Properties of resin
Remarks of resin
Tensile strength (MPa) Elastic modulus (MPa)
30 3800
685
floor through high strength steel bolts. Specimens were tested under cyclic lateral loading. Lateral load was applied to specimens at the beam level, using a hydraulic jack and the applied load was measured with a load cell (the capacities being 600 kN in compression and 300 kN in tension). In the tests, in both push and pull half cycles, 10 kN lateral load was applied at the beginning, and each cycle was repeated two times. The lateral load was increased by 10 kN at each cycle up to the ultimate load level of specimen. After reaching specimen’s lateral load carrying capacity, tests were completed. During the test, the storey displacements and the lateral loads were monitored. Axial load was applied to the columns by prestressing tendons as shown in Fig. 6. The level of applied axial load in each test was about 10% of the axial load capacity of the frame columns. A rigid steel frame was constructed around the test specimen to prevent out of plane displacements. In the instrumentation of the specimens, linear variable differential transducers (LVDTs) were used for displacement measurements. In addition, the strains of CFRP strips were measured using strain gauges. Strain-gauges placement at the CFRP strips used in specimens is shown in Fig. 6. 3. Experimental results 3.1. Behavior of test specimens
2.3. Test setup and instrumentation Details of the test setup, the loading system, and the instrumentation are shown in Fig. 6. The foundation of the test specimen was anchored to the laboratory’s strong
Lateral load–displacement hysteretic curves for the test specimens are shown in Figs. 7–10. As indicated in these figures, diagonal CFRP strips increased the lateral strength and the stiffness significantly. Reference specimen reached
Fig. 6. Details of test setup, loading system, and instrumentation.
686
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Fig. 7. Load–displacement curve of specimen 1 (Reference).
0.40% lateral drift at ultimate load. After the infill was crushed at the upper corners due to diagonal compression, the specimen lost its lateral load carrying capacity and thus, failed. Specimens 2, 3 and 4 that were strengthened with CFRP strips from both sides of the infill wall showed similar behavior and failure modes. CFRP strips at the interior side of the infill wall were ruptured due to tension (Fig. 11), and the anchorages of the CFRP strip at the exterior side of the infill wall were ruptured from the upper corner of the frame due to shearing. CFRP strips were ruptured at ultimate load level, and specimens lost their lateral load carrying capacities and stiffness abruptly. Storey drift ratio of Specimen 2 was 0.57% at ultimate load, and Specimens 3 and 4 had two times greater storey drift ratio than that of the reference specimen. Specimens that were strengthened non symmetrically (i.e. received CFRP strips on either the exterior or the interior sides of the infill wall) showed storey drift ratios between 0.50% and 0.61% at the ultimate load level and failed due to debonding of the CFRP strip from the upper corner of the infill wall (Fig. 12) at same storey drift ratios as can be seen in Figs. 9 and 10. CFRP strips were debonded at the ultimate load levels of the specimens. In addition, infill wall joints were crushed at the upper corners of the infill wall. After debonding occurred during next cycle, specimens failed due to rupture of the CFRP anchors at the upper corner of the frame (Fig. 13). The load level at which CFRP anchorages failed was lower than the ultimate load levels for all specimens. 4. Discussion of test results 4.1. Strength and stiffness The test results are summarized, and presented in Table 4. This table was prepared for illustrating the effect of applied strengthening technique on the ultimate strength, initial stiffness and lateral storey drift ratio.
Fig. 8. Load–displacement curves of Specimens 2, 3, and 4 (CFRP strips are applied on both faces of the wall).
The ratio of the ultimate lateral strength of Specimens 2, 3 and 4 that were strengthened with CFRP strips from both sides of infill wall to that of the reference specimen ranged between 2.18 and 2.61. The respective ratio values for the specimens that were strengthened with CFRP strips from one side of the infill wall varied between 1.57 and 1.85. Specimens receiving CFRP strips of the same width on one side of the infill wall had approximately the same ultimate lateral strength. On average, specimens receiving CFRP strips on one side of the infill wall had 29% less
S. Altin et al. / Composites: Part B 39 (2008) 680–693
687
Fig. 9. Load–displacement curves of Specimens 5, 6, and 7 (CFRP strips are applied on interior face of the wall).
Fig. 10. Load–displacement curves of Specimens 8, 9, and 10 (CFRP strips are applied on exterior face of the wall).
ultimate lateral strength than specimens receiving CFRP strips on both side of the infill wall. Response envelopes shown in Fig. 14 are plotted by connecting the peak points of the load–displacement hysteretic curves for each specimen. Response envelope curves can be used for evaluating the strength and stiffness characteristics of the specimens and also general behaviors. As can be seen from this figure, strength and stiffness of strengthened
frames were significantly higher than those of the reference specimen. All Seismic Codes provide similar limits to prevent extensive structural and non-structural damage and to minimize the second order effects. Turkish seismic code specifies the interstorey drift limit as 0.35% for the RC. framed systems used in this study, yet this value in Eurocode 8 regulations, for brittle nonsrructural infills in contact with the RC frame, is taken equal to 0.5%.
688
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Fig. 11. CFRP strip rupture photo (Specimen 3).
Fig. 12. CFRP strip debonding photo (Specimen 5).
The 0.35% limit drift ratio was marked on the response envelope curves shown in Fig. 14. There was no significant drop of the lateral load carrying capacity of the strengthened specimens up to this limit. Regarding the stiffness, this is true for specimens 2, 3, and 4, but not for the remaining ones. Furthermore, stiffness degradation (i.e. from the initial one to the one corresponding to a drift ratio of 0.35%) is more pronounced for specimens 8, 9, and 10 than for 5, 6, and 7. This can be attributed to the different anchorage detailing that the CFRP strips had depending on whether they were applied on the exterior or on the interior side
of the infill wall. Measured storey drift ratios for all strengthened specimens at ultimate load were greater than the limit drift ratio that was suggested by the regulations. The initial stiffnesses of the test specimens are given in Table 4. Initial stiffness was defined as the initial slope of the load–displacement curve in the first push half cycle. Diagonal CFRP strips increased initial stiffness of the specimens significantly. When the CFRP strips’ width was increased, initial stiffness also increased. Initial stiffnesses of specimens 2, 3, and 4 that were strengthened with CFRP strips from both sides of the infill wall were 4.0–6.4 times
S. Altin et al. / Composites: Part B 39 (2008) 680–693
689
Fig. 13. CFRP anchor rupture photo (Specimen 8).
Table 4 Summary of test results Specimen no.
1 (reference) 2 3 4 5 6 7 8 9 10 a b
Ultimate load (kN) Push
Pull
76.7 167.0 187.2 200.4 114.0 131.5 139.5 118.3 131.0 142.1
67.4 161.0 183.3 199.6 120.7 136.9 140.6 111.0 127.0 136.6
Ratioa
Drift ratio at ultimate load (%)
Initial stiffness (kN/mm)
Ratiob
1.00 2.18 2.44 2.61 1.57 1.78 1.83 1.54 1.71 1.85
0.40 0.57 0.80 0.82 0.55 0.50 0.61 0.50 0.52 0.58
50.00 200.00 300.00 320.00 190.48 250.00 265.00 196.22 266.67 285.00
1.00 4.00 6.00 6.40 3.81 5.00 5.30 3.92 5.33 5.70
Ratio of ultimate load of strengthened infilled frame to ultimate load of reference specimen. Ratio of initial stiffness of strengthened infilled frame to that of the reference specimen; initial stiffness was calculated as using first push half cycles.
greater than that of the reference specimen. The stiffnesses of specimens’ that were strengthened with CFRP strips from one side of the infill wall were from 3.81 to 5.70 times greater than that of the reference specimen. Should the initial stiffnesses of specimens with same CFRP strip width compared, the initial stiffness of the specimens receiving symmetrically strengthened were between 2% and 17% greater than that of specimens at which CFRP strips were applied on one side of the infill wall. Only there was no difference between specimens that were strengthened with diagonal CFRP strips from one side of infill wall. When the width of diagonal CFRP strips was increased from 0.13dw to 0.20dw, initial stiffness of the specimens receiving CFRP strips on both sides of infill wall was increased by 50%, and that of specimens receiving on one side of the infill wall were increased between 31% and
36%. It is important to note that, independent of the application method, for CFRP strips with 0.27dw width, initial stiffness of the specimens were only 7% greater than the specimens at which 0.20dw width CFRP strips were used. 4.2. Strains of CFRP strips Typical examples of strain measurements that were taken from different CFRP strips are given in Fig. 15 for Specimens 3, 5, 6 and 9. In this study, as an application technique CFRP strips bonded without removing the plaster and strain values were measured in this condition. As can be seen from Fig. 15, strain values that were measured from specimens at which CFRP strips were applied on both sides of the infill wall were greater than that of specimens at which CFRP strips were applied on one side of the infill
690
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Maximum tensile strain values that were measured from CFRP strips are shown in Fig. 16. CFRP strips that were bonded on the interior side of Specimens 2, 3 and 4 infill walls were ruptured at a strain level ranging between 0.0042 and 0.0054. Lateral displacement of the frame and cracks initiated at the masonry infill wall was caused direction changes in CFRP fibers. For this reason, CFRP strips that were affected by not only axial force but also transverse loads were failed at lower strain values. Specimens at which CFRP strips were applied from one side of infill wall had maximum tension strain values between 0.0028 and 0.0032. These values were corresponding to failure strains at which CFRP anchorages were ruptured. Average tensile strain values were marked on Fig. 16 for specimens that were strengthened with CFRP strips from both and one sides. Average tensile strain values that were obtained from specimens strengthened with CFRP strips form both sides were 1.6 times greater than that of specimens strengthened with CFRP strips from one sides. 4.3. Comparison of experimental and analytical results Comparisons of calculated and experimental ultimate lateral strength of specimens and ratios of them are presented in Table 5. Infill walls were taken as equivalent compression struts in analytical model. Lateral strength of strengthened specimens was calculated by superposition of the compression strength of the equivalent compression strut and the tension strength of the CFRP strip. The lateral strength (Vcal) of specimens was calculated using Eq. (1). V cal ¼ cos h ðV masonry þ V CFRP Þ
Fig. 14. Response envelopes of specimens.
wall. However the width of CFRP strips did not have significant effect on the measured strains. Diagonal CFRP strips of specimens strengthened with strips from one side of the infill wall were debonded from upper corner of infill wall under compression load at 0.002 average strains. Debonding occurred at each push and pull cycles at the upper corner of CFRP strip under compression.
ð1Þ
where Vcal is the calculated lateral strength capacities of specimens, Vmasonry the calculated equivalent diagonal strut capacities of masonry wall, VCFRP the calculated diagonal CFRP strip tension capacities of specimens and h is the main diagonal angle of masonry wall. The width of the equivalent diagonal compression strut was calculated according to FEMA356 regulation as 165 mm [25]. Material properties that were used in the analytical calculations were taken from the test results that were performed in the laboratory. Masonry unit’s compression strength and elastic modulus were determined according to FEMA356 regulation. Analytical calculations were performed using these values taking into account both clay brick, and mortar together. Tension strength of CFRP strips along fiber direction was calculated using Eq. (2). Strain values that were measured from experiments were used as strains in equations. V CFRP ¼ n ðeCFRP ECFRP W CFRP tCFRP Þ
ð2Þ
where n is the CFRP strip number, ECFRP the elastic modulus of CFRP, VCFRP the calculated diagonal CFRP strip tension capacities of specimens, tCFRP the thickness of CFRP strip, wCFRP the width of CFRP strip and eCFRP is the strain of CFRP strip.
S. Altin et al. / Composites: Part B 39 (2008) 680–693
691
Fig. 15. Typical shear force–strain curve of Specimens 3, 5, 6 and 9.
Fig. 16. Measured maximum strains of CFRP strip.
Reference specimen experimental strength was calculated with analytical approach very closely. All of the ana-
lytically calculated lateral strengths of strengthened specimens were smaller than the experimental values. On
692
S. Altin et al. / Composites: Part B 39 (2008) 680–693
Table 5 Comparison of experimental and calculated ultimate load capacities Specimen no.
Experimental ultimate load (kN)
Calculated ultimate load (kN)
Ratioa
1 (reference) 2 3 4 5 6 7 8 9 10
76.7 167.0 187.2 200.4 120.7 136.9 140.6 118.3 131.0 142.1
75.5 121.6 144.7 169.7 89.9 97.1 104.3 89.9 97.1 104.3
1.02 1.37 1.29 1.18 1.34 1.41 1.35 1.32 1.35 1.36
a
Ratio of experimental ultimate load values to calculated ultimate load.
average, experimental strengths of Specimens 2, 3 and 4 that were strengthened with CFRP strips from both sides of the infill wall were 28% greater than the analytically calculated values. Experimental strengths of specimens that were strengthened with CFRP strips from one sides of the infill wall were 35% greater than the analytically calculated values on average. 5. Conclusions In the study presented, 1/3 scale, one-bay, one-storey nonductile RC frames with masonry infill walls were strengthened by diagonal CFRP strips and experimentally investigated under cyclic lateral loading. The following conclusions can be drawn in the light of experimental study. (1) The diagonal CFRP strips that were used to retrofit brick masonry infilled reinforced concrete frames were effective in increasing lateral strength and lateral stiffness significantly. An important advantage of this technique is that it can be applied without evacuating the building during its application, thus causing minimum disturbance to the occupants. The success of this technique depends mainly on the successful application of anchorages between the CFRP strips and the infill wall or frame. (2) Lateral strengths of specimens receiving CFRP strips on both sides of the infill wall increased by 2.18 and 2.61 times compared to specimens receiving CFRP strips on one side only. The increases in lateral stiffness for the same specimens were 4.00 and 6.00 times. Ultimate lateral load increase for the specimens receiving CFRP strips on one side of infill were 1.57 and 1.85 times, and stiffness increases were from 3.81 and 5.70 times. Ultimate lateral strength and stiffness of the specimens that were strengthened with same CFRP strips width from one side of infill wall were approximately the same. (3) When the width of CFRP strip was increased, increase in strength and stiffness could be limited. If CFRP strip width was increased from 0.13dw to 0.20dw, 50% stiffness increase was obtained for the
specimens receiving CFRP strips from both sides of infill wall, and the stiffness increased between 31% and 36% compared to specimens strengthened with CFRP strips from one side of infill wall. (4) Storey drift ratio of all specimens was obtained to be greater than 0.35% limit value that was suggested by the new Turkish seismic regulations. There was no significant degradation in lateral load carrying capacity of all specimens up to this limit value. This is the case for the stiffnesses of specimens 2, 3, and 4, but not for the remaining ones. Furthermore, stiffness degradation is more pronounced for specimens 8, 9, and 10 than for 5, 6, and 7. Different anchorage detailing that the CFRP strips had depending on whether they were applied on the exterior or on the interior side of the infill wall caused this difference. (5) As an application technique CFRP strips bonded without removing the plaster, and strain values were measured in this condition. Strain values obtained from CFRP strips that were bonded to both sides of infill wall were 1.6 times greater than that of bonded to one side of infill wall. However width of CFRP strips did not have significant effect on measured strains. CFRP strips that were bonded to both sides of infill wall failed with rupture. Stress concentrations that were encountered at CFRP anchorages caused the premature failure of anchorages with rupture. Therefore, rectangular CFRP sheets that were used for preventing stress concentration at the region of anchorages showed limited success. (6) Analytically calculated lateral load carrying capacities were obtained to be 33% less than the experimental results. Authors believe that, the suggested analytical approach can give beneficial ideas to designers for the strengthening of masonry infilled RC frames by using CFRP strips, and the behavior of this complicated composite structure can be simulated quite accurately.
Acknowledgements This study was conducted at the Structural Mechanics Laboratory of Gazi University. The research is supported by the Scientific and Technical Research Council of Turkey (TUBITAK) through Research Grand 105M250 is great fully acknowledged. In addition, authors specially thanks to Mr. Kadir Basßogˇlu and Aslan Masonry Inc. References ¨ zsoy S. The behavior and strength of infilled frame. [1] Ersoy U, U TUBITAK MAG-205 Report, 1971, Ankara, Turkey. 95pp. [in Turkish]. [2] Altin S, Ersoy U, Tankut T. Hysteretic response of reinforced concrete infilled frames. ASCE, J Struct Eng 1992;118(8):2133–50.
S. Altin et al. / Composites: Part B 39 (2008) 680–693 ¨ zcebe G, Ersoy U. Rehabilitation of reinforced [3] Sonuvar MO, O concrete frames with reinforced concrete infills. ACI, Struct J 2004;101(4):494–500. ¨ zcebe G. Contribution of reinforced concrete [4] Canbay E, Ersoy U, O infills to seismic behavior of structural system. ACI, Struct J 2003;100(5):637–43. [5] Diagana C, Li A, Gedalia B, Dlemas Y. Shear strengthening effectiveness with CFF strips. Eng Struct 2003(25):507–16. [6] Taljsten B. Strengthening concrete beams for shear with CFRP sheets. Constr Build Mater 2003(17):15–26. [7] Khalifa A, Nanni A. Rehabilitation of rectangular simply supported RC beams with shear deficiencies using CFRP composites. Constr Build Mater 2002(16):135–46. [8] Taljsten B, Elfgren L. Strengthening concrete beams for shear using CFRP materials: evaluation of different application methods. Compos Part B 2000(33):87–96. [9] Kachlakev D, McCurry CC. Behavior of full scale reinforced concrete beams retrofitted for shear and flexural with FRP laminates. Compos Part B 2000(31):445–52. [10] Li A, Diagana C, Delmas Y. CFRP contribution to shear capacity of strengthened RC beams. Eng Struct 2001(23):1212–20. [11] Saadarmanesh H, Ehsani MR, Jin L. Strength and ductility of concrete columns externally reinforced with fiber composite straps. Struct J ACI 1994;91(4):434–47. [12] Chaallal O, Shahawy M, Hassan M. Performance of axially loaded short rectangular columns strengthened with carbon fiber-reinforced polymer wrapping. J Compos Constr 2003;7(3):200–8. [13] Ye L, Yue Q, Zhao S, Li Q. Shear strength of reinforced concrete columns strengthened with carbon-fiber reinforced plastic sheet. J Struct Eng ASCE 2002;128(12):1527–34. [14] Ghobarah A, Said A. Shear strengthening of beam–column joints. Eng Struct 2004;24:881–8.
693
[15] Prota A, Nanni A, Manfred G, Cosenza E. Selective upgrade of under designed RC beam–column joints using CFRP. Struct J ACI 2004;101(5):699–707. [16] Antonopulos CP, Triantafillou TC. Analysis of FRP-strengthening RC beam–column joints. J Compos Constr 2002;6(1):41–51. [17] Robertson IN, Johnson G. Repair of slab–column connections using epoxy and carbon fiber reinforced polymer. J Compos Constr 2004;8(5):376–83. [18] Limam O, Nguyen VY, Foret G. Numerical and experimental analysis of two way slabs strengthened with CFRP strips. Eng Struct 2005:841–5. [19] Balsamo A, Colombo A, Manfredi G, Negro P, Prota A. Seismic behavior of a full scale RC frame repaired using CFRP laminates. Eng Struct 2005(27):769–80. ¨ zcebe G, Ersoy U, Tankut T, Erduran E, Keskin RSO, Mertol HC. [20] O Strengthening of brick infilled RC frames with CFRP. SERU_ Structural Engineering Research Unit, Rep. No. 2003/1, TUBITAKMETU, Ankara, Turkey; 2003. _ Akyu¨z U, Ersoy U, O ¨ zcebe G. An experimental study on [21] Erdem I, two different strengthening techniques for RC frames. Eng Struct 2006:1843–51. [22] El Gawady MA, Lestuzzi P, Badoux M. Aseismic retrofitting of unreinforced masonry walls using FRP. Compos Part B: Eng 2006:148–62. [23] Cecchi A, Milani G, Trelli A. In plane loaded CFRP reinforced masonry walls mechanical characteristics by homogenization procures. Compos Sci Technol 2004(64):2097–112. [24] Turkish Seismic Code 4-25, Turkey; 2006. [25] FEMA 356, Prestandard and commentary for the seismic rehabilitation of buildings. ASCE, Federal Emergency Management Agency; November 2000.