- Email: [email protected]
External reinforcement of high strength concrete columns
Composite Structures 65 (2004) 279–287 www.elsevier.com/locate/compstruct
External reinforcement of high strength concrete columns M.N.S. Hadi *, J. Li Faculty of Engineering, University of Wollongong, Wollongong NSW 2522, Australia Available online 11 December 2003
Abstract With the technology development on the compressive strength of concrete over the years, the use of high strength concrete has proved most popular in terms of economy, superior strength, stiffness and durability due to many advantages it could offer. However, strength and ductility are inversely proportional [J. Mater. Civil Eng. 11 (1999) 21]. High strength concrete is a brittle material causing failure to be quite sudden and Ôexplosive’ under loads. It is also known that structural concrete columns axially compressed rarely occur in practice. The stress concentrations caused by the eccentric loading further reduce the strength and ductility of high strength concrete. Therefore, studies for high strength concrete columns under eccentric loading are essential for the practical use. This paper experimentally investigates a number of high strength concrete columns that are externally reinforced with galvanised steel straps and fibre-reinforced polymers subjected to concentric and eccentric loading. The experimental results show that external reinforcement can enhance the properties of high strength concrete columns. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Eccentric loading; Columns; External reinforcement; Galvanised steel straps; Fibre reinforced polymers; High strength concrete
1. Introduction Since high strength concrete has become a familiar phase in concrete technology in the late 1980s, the application of high strength concrete in the construction industry has steadily increased over the past two decades. The wide application of high strength concrete has stimulated a number of research studies in many countries including Australia during the last few years. However, the studies are not enough to predict the behaviour of the material with reasonable accuracy. As a consequence, important issues related to design and construction of high strength concrete structures are not adequately addressed in building codes, therefore, structural designers are unable to take full advantage of the material because of the inadequate information. The increase in brittleness with the increase of strength of concrete is of major concern in using the high strength concrete [1]. The lack of ductility of high strength concrete results in sudden failure without warning, which is a serious drawback of high strength concrete. Extensive *
Corresponding author. Tel.: +61-2-4221-4762; fax: +61-2-42213238. E-mail address: [email protected] (M.N.S. Hadi). 0263-8223/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.compstruct.2003.11.003
previous studies have shown that addition of compressive reinforcement and confinement will increase the ductility as well as the strength of material effectively [2,3]. The higher the concrete strength, the more it becomes necessary to provide confinement [4]. Confining the concrete can reduce its brittleness. In the recent years, considerable attention has been focused on the external reinforcement, as one of the methods of confinement, which has been proved by previous studies as an effective method to enhance the structural properties of high strength concrete members [5,6]. Externally reinforcing high strength concrete enhances the properties of concrete columns, most importantly reducing the effect of its brittle behaviour, and allowing the column to attain maximum load carrying capacity. These higher strengths are achieved as a result of the lateral pressures, applied by the external reinforcement, to the extreme fibres of the concrete column. The confinement prevents the lateral expansion of the specimen under axial load, improving the column’s stiffness. As a result, the high strength concrete column is able to carry higher loads than if it were unreinforced. Among the various external reinforcements, steel straps and fibre reinforced polymers are being used popularly. Previous studies have shown that external
280
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
steel reinforcement increases a column’s strength and enables the steel straps to be smaller in size than internal steel reinforcement. As the corrosion for the steel straps resulted in bond deterioration, the steel is galvanised to resist against corrosion. In recent years, fibre-reinforced polymers wrapping in lieu of steel jacket has become an increasingly popular method for external reinforcement in which fibre-reinforced polymers offer improved corrosion and fatigue resistance compared to the steel reinforcement [7]. The high tensile strength and low weight make fibre-reinforced polymers ideal for use in the construction industry. Another attractive advantage of fibre reinforced polymers over steel straps as external reinforcement is its easy handling, thus minimal time and labour are required to implement them. Li and Hadi [5] conducted a series of experimental tests for concrete columns under eccentric loading. The non-prismatic column geometry with ends haunched on one side enabled the application of eccentric loading. Two types of confining materials were investigated, Eglass and Carbon fibre. The conclusion of this research was both the confining materials were effective in producing columns with high strength and high ductility. This study considers various types of external reinforcement and compares them with experimental results. The effect of the two types of external reinforcing material, galvanised steel straps and fibre reinforced polymers are evaluated. Two sets of tests in terms of eccentric loading and concentric loading are conducted. Then, the effectiveness of the external reinforcement as a confining material under different loading conditions are investigated.
2.1. Mechanics of confinement
2. Modelling of external confinement
fcc0 fr ¼ 1 þ k1 0 0 fco fco
Lateral confinement has been proven to significantly enhance the strength and ductility of concrete members due to the presence of triaxial compression [8]. Such a triaxial state of compression is achieved by providing ‘‘confinement’’ for the concrete. When concrete columns are subjected to axial loading, concrete will expand normal to the loading direction due to the Poisson’s effect. If the confinement and reinforcement are placed to confine this expansion, the desired stress state is achieved. In internally reinforced columns, the confinement can be achieved by providing transverse reinforcement in the form of stirrups or helix. For externally confined columns, this confinement is attained by applying the fibre-reinforced composites around the perimeter of concrete columns, which exerts a continuously increasing confining action. The performance of the member is dependent upon the interaction of the confined concrete and the confining material.
The confinement supplied to concrete using external confinement is of the passive type. As the concrete tries to expand under load, the jacket prevents this expansion, creating hoop stresses in the jacket. Fig. 1 [9] illustrates this effect. To calculate the confining stress fr applied to the concrete by confinement, the following relationship proposed by Samaan et al. [10] is used: fj tj ð1Þ D where fr is the confining stress; fj the hoop stress in jacket; tj the thickness of the jacket; D the diameter of the concrete column. While for the internally reinforced and steel strapped specimens, the following relationship from Pessiki and Peroni [11] is to be used:
fr ¼ 2
fr ¼ fl ke
ð2Þ 2fus As Dh s
0
s Þ2 ð12d 1qcc
where fl ¼ and ke ¼ where fl is the lateral confinement strength, fus the stress in the helix at maximum column load, As the area of the helical bar, and Dh and s are the outside diameter and pitch, respectively, of the helix, d is the diameter of concrete column, s0 is the clear spacing between straps and qcc is ratio of area of confined concrete to unconfined. 2.2. Confinement effectiveness Most empirical expressions for predicting the strength of FRP-confined concrete take the following form based on the suggestion of linear relationship between confinement effectiveness and the confinement ratio by Richart et al. [8]: ð3Þ
where fcc0 is the compressive strength of confined concrete; fco0 the compressive strength of unconfined concrete; fr the lateral confining pressure; k1 the effectiveness coefficient. Richart et al. [8] assumed a constant value for the coefficient k1 equal to 4.1. However, recent studies revealed that the existing models suitable for concrete confined with steel spirals or ties rather than concrete confined with fibre-reinforced polymers composites. A
Fig. 1. Confinement action for continuous jacket.
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
number of strength models have thus been proposed specifically for FRP-confined concrete, which employ Eq. (3) with modified expressions for k1 [12]. In this study, the expression from Karbhari and Gao [13] was used for the theoretical calculation on the strength of FRP wrapped specimens: k1 ¼ 2:1
fr fco0
0:13 ð4Þ
281
3.1. Columns’ details 3.1.1. Concentrically loaded columns Five columns without internal reinforcement were designed for this testing. Each column had a diameter of 205 mm and a height of 910 mm. Four columns continually wrapped with FRP had the following configurations: one-layered and three-layered Carbon fibres, one-layered and three-layered Kevlar Fibres. The remaining plain column was used as control column. The configuration of this set of columns is summarised in Table 1.
3. Experimental programme In order to test the performance of concrete columns confined with various reinforcing materials, two sets of tests were designed: five cylindrical concrete columns of 205 mm diameter and 910 mm height were tested under concentric loading. Another set of six cylindrical concrete columns of 205 mm diameter and 920 mm height were tested under eccentric loading with an eccentricity of 50 mm. The configuration of external reinforcement varies for both sets of columns as well. The testing variables selected for this study are (1) the type of external reinforcement: galvanised steel straps and fibrereinforced polymers, (2) the number of layers for FRP, (3) the spacing of steel straps, (4) the types of FRP materials and (5) the loading pattern. In order to have a better insight about the contribution of FRP on confinement and to be able to conduct the theoretical analysis on the behaviour of the column specimens in the main testing program of this study, a preliminary testing on the all reinforcing materials used in this study was conducted as well.
3.1.2. Eccentrically loaded columns Li and Hadi [5] achieved eccentric loading by changing the geometry of the column to non-prismatic. In this study, the columns were kept prismatic but the load was applied at an offset from the centre thus producing eccentric loading. Six concrete columns were cast and tested. Three of the columns were wrapped with three layers of unidirectional fibre-reinforced polymers. Two were externally reinforced with galvanised steel straps, each steel strap was 20 mm wide and 0.5 mm thick. Two spacings for the steel straps were used: 10 and 20 mm. The final column was internally reinforced with steel helix and longitudinal reinforcement. All the columns were eccentrically loaded until failure with an eccentricity of 50 mm. The testing matrix is summarised in Table 2. 3.2. Eccentric loading Where eccentric loading differs from concentric loading is that it involves concentrating the load a
Table 1 Configuration of the concentrically loaded columns Column (1)
Diameter (mm) (2)
Height (mm) (3)
Reinforcing type (4)
Reinforcing material (5)
Loading pattern (6)
1 2 3 4 5
205 205 205 205 205
910 910 910 910 910
External External External External External
Single-layered Carbon Single-layered Kevlar Three-layered Carbon Three-layered Kevlar
Concentric Concentric Concentric Concentric Concentric
Table 2 Testing matrix of the eccentrically loaded columns Column (1)
Diameter (mm) (2)
Height (mm) (3)
Reinforcing type (4)
Reinforcing material (5)
Loading pattern (6)
1 2 3 4 5 6
205 205 205 205 205 205
920 920 920 920 920 920
External External External External External Internal
Three-layered Carbon Three-layered E-glass Three-layered Kevlar Galvanised steel straps at 10 mm spacing Galvanised steel straps at 20 mm spacing 6 N12 Bars and N10 helix
Eccentric Eccentric Eccentric Eccentric Eccentric Eccentric
282
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
The band-it method was employed to apply the galvanised straps on the two concrete columns in this study. The galvanised steel straps were placed along the length of column at 20 mm spacing for one column and 10 mm spacing for the second column. 3.4. Specimen testing
Fig. 2. Steel end plates for eccentric loading.
certain distance form the neutral axis of the cross section. As shown in Fig. 2, two plates were designed and manufactured in order to apply eccentric loading on the columns. These plates were used on either end of the columns during loading.
The testing program consisted of testing the five concrete columns under concentric loading and testing the six cylindrical concrete columns with different external confinement under the eccentric load. The hydraulically operated 5000 kN Denison compression testing machine, located in the Engineering Laboratory at the University of Wollongong was used to test all the columns in this study. All the columns were tested to failure.
4. Observed behaviour and test results 3.3. Specimen preparation Two batches of concrete were used to cast the concentrically and eccentrically loaded columns. The design compressive strength of both batches of concrete was 100 MPa. However, 73.62 MPa and 51 MPa were achieved for the concentrically loaded and eccentrically loaded columns, respectively. The three types of fibre-reinforced polymers used in this study were Carbon, Kevlar and E-glass. The epoxy system consisted of two parts, resin and slow hardener, were used to bond the FRP to the surface of the concrete columns. The process of applying the FRP is known as the wet lay up method and was used to wrap all the columns with external FRP confinement.
The failure of the columns in all cases was brittle and in the case of the plain specimen, a very explosive failure. In the case of the FRP confined columns, the snapping of the fibres could be heard throughout the loading as the concrete tried to expand. While for the two galvanised steel straps reinforced columns, failure was sudden and soundless. In each case the straps may have yielded but did not break. This suggested that the failure of the columns was a direct result of cracking of the concrete tensile flexure. This can be explained that this type of reinforcement may not be suitable for columns under eccentric load. Tables 3 and 4 present the testing results of the concentrically loaded and eccentrically loaded columns, respectively.
Table 3 Testing results of the concentrically loaded columns Column (1)
Configuration (2)
Ultimate load (kN) (3)
Axial stress (MPa) (4)
Axial deflection (mm) (5)
1 2 3 4 5
Plain Single-layered Carbon Single-layered Kevlar Three-layered Carbon Three-layered Kevlar
2351 2860 2490 2980 2490
71.23 86.65 75.44 90.29 75.44
5.048 4.404 4.234 6.514 5.574
Table 4 Testing results of the eccentrically loaded columns Column (1)
Configuration (2)
Ultimate load (kN) (3)
Axial stress (MPa) (4)
Axial deflection (mm) (5)
1 2 3 4 5 6
Three-layered Carbon Three-layered E-glass Three-layered Kevlar Galvanised steel straps at 10 mm spacing Galvanised steel straps at 20 mm spacing Internally reinforced
840.0 630.8 906.0 720.0 704.9 636.8
25.45 19.11 27.45 21.81 21.36 19.29
5.20 4.38 5.50 4.22 3.94 –
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
283
4.1. Concentrically loaded plain column This plain concrete column, as expected, presented a very brittle explosive failure. The column did not experience any excess deflection after reaching the maximum compressive load due to the lack of confinement, which led to the brittle failure. 4.2. Single-layered Carbon fibres column under concentric loading The external confinement provided to this column resulted in a higher ultimate load. However, the failure was still quite explosive and resulted in no increased deflection after reaching the maximum load as well. 4.3. Single-layered Kevlar column under concentric loading The single-layered Kevlar column achieved a slight increase in ultimate load over the plain specimen. And the most promising aspect about this column is that there was a small amount of excess deflection achieved after ultimate load. The failure was less explosive and the column was almost fully confined even after failure. 4.4. Three-layered Carbon specimen under concentric loading This specimen achieved significantly better results than the single layered specimen both in strength and deflection. It is of significance to note that the column still appeared to be fully intact after failure. Upon closer inspection, it could be seen that the jacket had a section where the jacket has frayed rather than actually fractured. This meant that even after failure the column still had the ability to withstand load and still maintain its integrity. 4.5. Three-layered Kevlar specimen under concentric loading The three-layered Kevlar specimen also outperformed the single-layered specimen, also achieving higher strength and ductility. However, there was not as much excess deflection achieved as the Carbon wrapped specimen. This specimen also remained intact after failure except for the presence of small fractures in the jacket.
Fig. 3. Failure of the internally reinforced concrete column.
even after the concrete cover had spalled away, the confined core continued to carry an increasing load. Fig. 3 shows the column after failure. 4.7. Carbon fibre jacketed concrete columns under eccentric loading Failure of the column was marked by brittle rupture of the hardened fibres at the bottom of the column, which can be seen in Fig. 4. Failure was sudden and quite explosive. During various stages of loading, the snapping sounds could be heard, which were attributed to the cracking of the concrete and the stretching of the hardened fibres. The test results show that this column could withstand much higher ultimate load than the internally reinforced column. This reveals that Carbon confinement could provide significantly greater confining pressure to the high strength concrete column. 4.8. E-glass wrapped specimen under eccentric loading Failure of the E-glass wrapped column specimen was marked by fibre rupture at the top of the column. Although it was sudden, the failure could be predicated by the appearance of white patches at the top of the column as the result of the fibre stretching. From Fig. 5, it can be seen the layers of E-glass were torn as a result of the eccentric load applied to the column. As the external Eglass confinement tried to prevent the concrete from expansion under loading, it was ruptured when the tensile stress, applied by the concrete lateral expansion, became too large.
4.6. Internally reinforced specimen under eccentric loading This specimen exhibited a brittle failure under the eccentric loading. The concrete cover started to fall away due to lateral dilation under the loading. However,
Fig. 4. Failure of the Carbon fibres column.
284
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
Fig. 7. Failure of the steel strapped column with 10 mm spacing.
Fig. 5. Failure of the E-glass wrapped column.
The results show that the load carrying capacity of this column wrapped with E-glass was slightly lower than the internally reinforced column. 4.9. Kevlar wrapped concrete column under eccentric loading The material used to wrap this column is one sheet of Kevlar in 920 mm wide rather than the roll of tape. The failure mode of this column was similar to that of Eglass specimen: fibres were ruptured at the top end of the column. This can be seen in Fig. 6. Cracking of Kevlar fibre could be heard throughout the testing with the failure of the column signified by a loud snap of the Kevlar jacket. The largest load carrying capacity was achieved by this column compared to other eccentrically loaded columns. This is contributed to the external confinement in terms of continuous sheet. Also, the failure was sudden and loud. 4.10. Galvanised steel strapped column at 10 mm spacing under eccentric loading It was found this column failed in the tensile bending region under eccentric load. Fig. 7 shows the failure of this column occurred in the space of two straps. This can
Fig. 8. Failure of the steel strapped column with 20 mm spacing.
be explained as a result of there being no reinforcement in this region. However, the crack was much smaller and failure occurred closer to the bottom of the column when compared to the column with steel straps in 20 mm spacing. The failure was brittle and soundless. 4.11. Galvanised steel straps at 20 mm spacing under eccentric loading Failure of this column was similar to that of the other galvanised steel straps wrapped column, in that the cracking of the concrete on the tension side marked the failure. Also evident in Fig. 8, is that failure, again occurred in between the galvanised steel straps and the straps themselves again, did not show any sign of failing. As the increased spacing of straps resulted in a larger area of the column being un-reinforced, the column failed in a substantial crack in the concrete. The results shown in Table 4 confirm that the larger the spacing between the straps, results in a lower load carrying capacity.
5. Comparisons and analysis
Fig. 6. Failure of the Kevlar wrapped column.
From the two sets of experiments conducted in this study, it can be noted that the Carbon wrapped columns outperformed the other types of reinforced columns except the Kevlar sheet wrapped column, which was
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
proved by the testing results of the eccentrically loaded columns. The testing results indicated that Carbon fibres wrapping is more effective for the external confinement compared to the galvanised steel straps and E-glass. However, this is not the case in the Kelvar sheet wrapping column, which presented the largest loading capacity as the continuous sheet was used as the external confinement. Again, this proved the layout of fibres has a significant influence on the behaviour of the eccentrically loaded columns. The comparison among the eccentrically loaded columns shows that all the externally reinforced columns out-performed the internally reinforced column excluding the E-glass specimen, which almost achieved the same strength. The E-glass was confirmed to be the weakest reinforcing material, which presented an ultimate load 10% lower than that of the two Band-It columns and a 44% decrease in compressive load compared to the Kevlar fibre sheet confined column. Another comparison made between the two galvanised steel straps wrapped columns shows that the larger the spacing between the straps results in a lower load carrying capacity. However, the column with galvanised steel straps in 20 mm spacing exhibited only 2.2% decrease in the loading carrying capacity. Nonetheless, there was a 26% and 16% decrease in ultimate load over the Kevlar fibre sheet and Carbon fibre confined columns respectively. And both columns achieved slightly higher ultimate load compared to the internally reinforced column, which proved the external confinement with galvanised steel straps is also more effective than the internal reinforcement. But the failure of the columns with this type of reinforcement is sudden, which indicates that the galvanised steel straps have very little effect on improving the ductility of the columns. The comparison among the concentrically loading columns confirmed that the confinement significantly enhances the strength, stiffness and ductility of high strength concrete, in particular when applied in multiple layers.
285
6. Theoretical calculations The analytical model outlined above was used to calculate the confining stresses in the columns during testing. It is to be noted that these models have been established for concentrically loaded specimens. For eccentrically loaded specimens, two types of stresses are encountered, axial stresses and bending stresses. The first step in calculating the confining stress is to determine the value of the hoop stress. This can be determined by using the experimental values obtained from the tensile tests conducted on the FRP coupons and steel straps. Table 5 presents the values of tensile strengths that were obtained for the different materials, which include two lots of materials used for the concentrically loaded and eccentrically loaded columns, respectively. The tensile strength of the material was used to predict the confining stress in the columns. This was achieved by using Eqs. (1) and (2). Table 6 shows the calculated values of confining stresses, which indicates that the Carbon fibres to be superior. The confining stresses of the helically reinforced columns were calculated using Eq. (2). These confining stresses are shown in Table 7. As shown in Table 8, the confinement effectiveness was obtained for the different specimens using Eq. (4). Table 8 compares the theoretical calculating results of all the externally reinforced columns under concentric loading with their experimental results. The theoretical ultimate load was calculated based on the calculated compressive strength of confined concrete using Eq. (3) and the concrete column cross sectional area. It shows that the theoretically determined confining stresses generally overestimates the experimental results. The most matched results were obtained by the one-layered FRP wrapped specimens. This may be due to the tensile testing on the material was restricted to a single-layered coupon. The strength of three-layered FRP was assumed to be three times of the strength of single-layered coupon.
Table 5 Tensile strength of reinforcing materials Specimen (1)
Thickness (mm) (2)
Tensile strength (MPa) (3)
Concentrically loaded columns
One-layered Carbon One-layered Kevlar Three-layered Carbon Three-layered Kevlar
1.360 1.635 3.060 3.700
305.15 104.59 406.90 138.65
Eccentrically loaded columns
Three-layered Carbon Three-layered E-glass Three-layered Kevlar Galvanised steel straps Helix
0.900 0.500 1.000 0.500 10.000a
692.27 196.00 162.00 627.50 662.10
a
Diameter of steel bars.
286
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
Table 6 Confining pressure of FRP confinement Specimen (1)
Confining stress fr (MPa) (2)
Concentrically loaded columns
One-layered Carbon One-layered Kevlar Three-layered Carbon Three-layered Kevlar
4.05 1.67 12.15 5.00
Eccentrically loaded columns
Three-layered Carbon Three-layered E-glass Three-layered Kevlar
6.08 0.96 1.58
Table 7 Confining pressure of steel reinforcement
(1)
Helix stress fus (MPa) (2)
Steel area Ast (mm2 ) (3)
Diameter Dh (mm) (4)
Pitch s (mm) (5)
Confining pressure fr (MPa) (6)
Internal helix Straps at 10 mm spacing Straps at 20 mm spacing
662.10 627.50 627.50
78.5 10.0 10.0
165 205 205
60 30 40
1.04 1.74 1.23
Material
Table 8 Comparison between theoretical and experimental results for the concentrically loaded columns Specimen (1)
fr (MPa) (2)
k1 (3)
fcc0 (MPa) (4)
Theoretical ult. load (kN) (5)
Experimental ult. load (kN) (6)
One-layered Carbon One-layered Kevlar Three-layered Carbon Three-layered Kevlar
4.05 1.67 12.15 5.00
3.06 3.44 2.65 2.98
86.01 79.36 105.82 88.52
2837 2618 3491 2920
2860 2490 2980 2490
7. Conclusions The work carried out in this study involved two sets of testing: five columns under concentric loading and six columns under eccentric loading, which are mainly set to evaluate the effectiveness of various types of the external reinforcement. The results from both sets of tests allow the following conclusions to be drawn: 1. The methods of external reinforcement can be used as an alternative method of reinforcement to enhance the properties of high strength concrete. It has been shown that the confinement of the concrete prevents the concrete from expanding and therefore allows the concrete to absorb higher stresses, resulting in a higher load carrying capacity. 2. The tests proved that the benefits of confinement could be enhanced by applying multiple layers, which can be seen from the results of testing the concentric loading columns. 3. The test results also indicated that the Carbon fibres provides the greatest amount of confinement, and
4.
5.
6.
7.
had significantly better results, if the external confinement was achieved by the application of FRP in roll of tape. The highest load carrying capacity achieved by Kevlar sheet wrapped column confirms that the wider rolls of the fibre reinforcement can provide a greater confining stress. This also can be concluded that the fibre layout has significant influence on the behaviour of concrete structural members. The external confinement with galvanised steel straps improves the strength of the column to a certain extent. The brittle, sudden, soundless failure of the galvanised steel straps wrapped columns shows that the galvanised steel straps had very little effect on improving the ductility of the columns. The E-glass proved to be the weakest reinforcing material in this study. The ultimate load achieved by the E-glass wrapped specimen even lower than the internally reinforced column. The comparisons made between theoretical calculations and experimental results show that the models prepared by Samaan et al. [10] and Pessiki and Peroni
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
[11] and can be used to estimate the ultimate strengths of externally reinforced columns under concentric loading to a certain degree accuracy.
Acknowledgements The authors acknowledge the financial assistances of the University of Wollongong through its Grant scheme, which enabled conducting this research. The technical assistance of Messers Ian Bridge, Alan Grant and Bob Rowlan of the University of Wollongong are appreciated. References [1] Mansur MA, Chin MS, Wee TH. Stress–strain relationship of high-strength fiber concrete in compression. J Civil Eng 1999;11(February):21–9. [2] Razvi SR, Saatcioglu M. Strength and deformability of confined high-strength concrete columns. ACI Struct J 1994;91(November– December):678–87. [3] Hadi MNS, Schmidt LC. Use of helices in reinforced concrete beams. ACI Struct J 2002;99(2):191–8.
287
[4] Attard MM, Mendis PA. Ductility of high strength columns. Australian Civil Eng Trans 1993;35(4):295–306. [5] Li J, Hadi MNS. Behaviour of externally confined high strength concrete columns under eccentric loading. J Compos Struct 2003;62(2):145–53. [6] Pessiki S, Harries KA, Kestner JT, Sause R, Ricles MJ. Axial behaviour of reinforced concrete columns confined with FRP jackets. J Compos Constr 2001;5(November):237–45. [7] Pantazopoulou SJ, Bonacci JF, Thomas MDA, Hearn N. Repair of corrosion-damaged columns with FRP wraps. J Compos Constr 2001;5(February):3–11. [8] Richart FE, Brandtzaeg A, Prown RL. A study of the failure of concrete under combined compressive stress. Bulletin No. 190, 1928, Engineering Experiment Station, University of Illinois, Urbana, Apr. p. 73. [9] Xiao Y, Wu H. Compressive behaviour of concrete confined by carbon fibre composite jackets. J Mate Civil Eng 2000;12(2):139– 46. [10] Samaan M, Mirmiran A, Shahawy M. Model of concrete confined by fibre composite. J Struct Eng 1998;124:1025–31. [11] Pessiki S, Pieroni A. Axial load behaviour of large-scale spirally reinforced high-strength concrete columns. ACI Struct J 1997;94:305–15. [12] Lam L, Teng JG. Strength models for fiber-reinforced plasticconfined concrete. J Struct Eng 2002;128(5):612–23. [13] Karbhari VM, Gao Y. Composite jacketed concrete under uniaxial compression-verification of simple design equations. J Thermoplast Compos 1997;6:92–107.
External reinforcement of high strength concrete columns M.N.S. Hadi *, J. Li Faculty of Engineering, University of Wollongong, Wollongong NSW 2522, Australia Available online 11 December 2003
Abstract With the technology development on the compressive strength of concrete over the years, the use of high strength concrete has proved most popular in terms of economy, superior strength, stiffness and durability due to many advantages it could offer. However, strength and ductility are inversely proportional [J. Mater. Civil Eng. 11 (1999) 21]. High strength concrete is a brittle material causing failure to be quite sudden and Ôexplosive’ under loads. It is also known that structural concrete columns axially compressed rarely occur in practice. The stress concentrations caused by the eccentric loading further reduce the strength and ductility of high strength concrete. Therefore, studies for high strength concrete columns under eccentric loading are essential for the practical use. This paper experimentally investigates a number of high strength concrete columns that are externally reinforced with galvanised steel straps and fibre-reinforced polymers subjected to concentric and eccentric loading. The experimental results show that external reinforcement can enhance the properties of high strength concrete columns. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: Eccentric loading; Columns; External reinforcement; Galvanised steel straps; Fibre reinforced polymers; High strength concrete
1. Introduction Since high strength concrete has become a familiar phase in concrete technology in the late 1980s, the application of high strength concrete in the construction industry has steadily increased over the past two decades. The wide application of high strength concrete has stimulated a number of research studies in many countries including Australia during the last few years. However, the studies are not enough to predict the behaviour of the material with reasonable accuracy. As a consequence, important issues related to design and construction of high strength concrete structures are not adequately addressed in building codes, therefore, structural designers are unable to take full advantage of the material because of the inadequate information. The increase in brittleness with the increase of strength of concrete is of major concern in using the high strength concrete [1]. The lack of ductility of high strength concrete results in sudden failure without warning, which is a serious drawback of high strength concrete. Extensive *
Corresponding author. Tel.: +61-2-4221-4762; fax: +61-2-42213238. E-mail address: [email protected] (M.N.S. Hadi). 0263-8223/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.compstruct.2003.11.003
previous studies have shown that addition of compressive reinforcement and confinement will increase the ductility as well as the strength of material effectively [2,3]. The higher the concrete strength, the more it becomes necessary to provide confinement [4]. Confining the concrete can reduce its brittleness. In the recent years, considerable attention has been focused on the external reinforcement, as one of the methods of confinement, which has been proved by previous studies as an effective method to enhance the structural properties of high strength concrete members [5,6]. Externally reinforcing high strength concrete enhances the properties of concrete columns, most importantly reducing the effect of its brittle behaviour, and allowing the column to attain maximum load carrying capacity. These higher strengths are achieved as a result of the lateral pressures, applied by the external reinforcement, to the extreme fibres of the concrete column. The confinement prevents the lateral expansion of the specimen under axial load, improving the column’s stiffness. As a result, the high strength concrete column is able to carry higher loads than if it were unreinforced. Among the various external reinforcements, steel straps and fibre reinforced polymers are being used popularly. Previous studies have shown that external
280
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
steel reinforcement increases a column’s strength and enables the steel straps to be smaller in size than internal steel reinforcement. As the corrosion for the steel straps resulted in bond deterioration, the steel is galvanised to resist against corrosion. In recent years, fibre-reinforced polymers wrapping in lieu of steel jacket has become an increasingly popular method for external reinforcement in which fibre-reinforced polymers offer improved corrosion and fatigue resistance compared to the steel reinforcement [7]. The high tensile strength and low weight make fibre-reinforced polymers ideal for use in the construction industry. Another attractive advantage of fibre reinforced polymers over steel straps as external reinforcement is its easy handling, thus minimal time and labour are required to implement them. Li and Hadi [5] conducted a series of experimental tests for concrete columns under eccentric loading. The non-prismatic column geometry with ends haunched on one side enabled the application of eccentric loading. Two types of confining materials were investigated, Eglass and Carbon fibre. The conclusion of this research was both the confining materials were effective in producing columns with high strength and high ductility. This study considers various types of external reinforcement and compares them with experimental results. The effect of the two types of external reinforcing material, galvanised steel straps and fibre reinforced polymers are evaluated. Two sets of tests in terms of eccentric loading and concentric loading are conducted. Then, the effectiveness of the external reinforcement as a confining material under different loading conditions are investigated.
2.1. Mechanics of confinement
2. Modelling of external confinement
fcc0 fr ¼ 1 þ k1 0 0 fco fco
Lateral confinement has been proven to significantly enhance the strength and ductility of concrete members due to the presence of triaxial compression [8]. Such a triaxial state of compression is achieved by providing ‘‘confinement’’ for the concrete. When concrete columns are subjected to axial loading, concrete will expand normal to the loading direction due to the Poisson’s effect. If the confinement and reinforcement are placed to confine this expansion, the desired stress state is achieved. In internally reinforced columns, the confinement can be achieved by providing transverse reinforcement in the form of stirrups or helix. For externally confined columns, this confinement is attained by applying the fibre-reinforced composites around the perimeter of concrete columns, which exerts a continuously increasing confining action. The performance of the member is dependent upon the interaction of the confined concrete and the confining material.
The confinement supplied to concrete using external confinement is of the passive type. As the concrete tries to expand under load, the jacket prevents this expansion, creating hoop stresses in the jacket. Fig. 1 [9] illustrates this effect. To calculate the confining stress fr applied to the concrete by confinement, the following relationship proposed by Samaan et al. [10] is used: fj tj ð1Þ D where fr is the confining stress; fj the hoop stress in jacket; tj the thickness of the jacket; D the diameter of the concrete column. While for the internally reinforced and steel strapped specimens, the following relationship from Pessiki and Peroni [11] is to be used:
fr ¼ 2
fr ¼ fl ke
ð2Þ 2fus As Dh s
0
s Þ2 ð12d 1qcc
where fl ¼ and ke ¼ where fl is the lateral confinement strength, fus the stress in the helix at maximum column load, As the area of the helical bar, and Dh and s are the outside diameter and pitch, respectively, of the helix, d is the diameter of concrete column, s0 is the clear spacing between straps and qcc is ratio of area of confined concrete to unconfined. 2.2. Confinement effectiveness Most empirical expressions for predicting the strength of FRP-confined concrete take the following form based on the suggestion of linear relationship between confinement effectiveness and the confinement ratio by Richart et al. [8]: ð3Þ
where fcc0 is the compressive strength of confined concrete; fco0 the compressive strength of unconfined concrete; fr the lateral confining pressure; k1 the effectiveness coefficient. Richart et al. [8] assumed a constant value for the coefficient k1 equal to 4.1. However, recent studies revealed that the existing models suitable for concrete confined with steel spirals or ties rather than concrete confined with fibre-reinforced polymers composites. A
Fig. 1. Confinement action for continuous jacket.
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
number of strength models have thus been proposed specifically for FRP-confined concrete, which employ Eq. (3) with modified expressions for k1 [12]. In this study, the expression from Karbhari and Gao [13] was used for the theoretical calculation on the strength of FRP wrapped specimens: k1 ¼ 2:1
fr fco0
0:13 ð4Þ
281
3.1. Columns’ details 3.1.1. Concentrically loaded columns Five columns without internal reinforcement were designed for this testing. Each column had a diameter of 205 mm and a height of 910 mm. Four columns continually wrapped with FRP had the following configurations: one-layered and three-layered Carbon fibres, one-layered and three-layered Kevlar Fibres. The remaining plain column was used as control column. The configuration of this set of columns is summarised in Table 1.
3. Experimental programme In order to test the performance of concrete columns confined with various reinforcing materials, two sets of tests were designed: five cylindrical concrete columns of 205 mm diameter and 910 mm height were tested under concentric loading. Another set of six cylindrical concrete columns of 205 mm diameter and 920 mm height were tested under eccentric loading with an eccentricity of 50 mm. The configuration of external reinforcement varies for both sets of columns as well. The testing variables selected for this study are (1) the type of external reinforcement: galvanised steel straps and fibrereinforced polymers, (2) the number of layers for FRP, (3) the spacing of steel straps, (4) the types of FRP materials and (5) the loading pattern. In order to have a better insight about the contribution of FRP on confinement and to be able to conduct the theoretical analysis on the behaviour of the column specimens in the main testing program of this study, a preliminary testing on the all reinforcing materials used in this study was conducted as well.
3.1.2. Eccentrically loaded columns Li and Hadi [5] achieved eccentric loading by changing the geometry of the column to non-prismatic. In this study, the columns were kept prismatic but the load was applied at an offset from the centre thus producing eccentric loading. Six concrete columns were cast and tested. Three of the columns were wrapped with three layers of unidirectional fibre-reinforced polymers. Two were externally reinforced with galvanised steel straps, each steel strap was 20 mm wide and 0.5 mm thick. Two spacings for the steel straps were used: 10 and 20 mm. The final column was internally reinforced with steel helix and longitudinal reinforcement. All the columns were eccentrically loaded until failure with an eccentricity of 50 mm. The testing matrix is summarised in Table 2. 3.2. Eccentric loading Where eccentric loading differs from concentric loading is that it involves concentrating the load a
Table 1 Configuration of the concentrically loaded columns Column (1)
Diameter (mm) (2)
Height (mm) (3)
Reinforcing type (4)
Reinforcing material (5)
Loading pattern (6)
1 2 3 4 5
205 205 205 205 205
910 910 910 910 910
External External External External External
Single-layered Carbon Single-layered Kevlar Three-layered Carbon Three-layered Kevlar
Concentric Concentric Concentric Concentric Concentric
Table 2 Testing matrix of the eccentrically loaded columns Column (1)
Diameter (mm) (2)
Height (mm) (3)
Reinforcing type (4)
Reinforcing material (5)
Loading pattern (6)
1 2 3 4 5 6
205 205 205 205 205 205
920 920 920 920 920 920
External External External External External Internal
Three-layered Carbon Three-layered E-glass Three-layered Kevlar Galvanised steel straps at 10 mm spacing Galvanised steel straps at 20 mm spacing 6 N12 Bars and N10 helix
Eccentric Eccentric Eccentric Eccentric Eccentric Eccentric
282
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
The band-it method was employed to apply the galvanised straps on the two concrete columns in this study. The galvanised steel straps were placed along the length of column at 20 mm spacing for one column and 10 mm spacing for the second column. 3.4. Specimen testing
Fig. 2. Steel end plates for eccentric loading.
certain distance form the neutral axis of the cross section. As shown in Fig. 2, two plates were designed and manufactured in order to apply eccentric loading on the columns. These plates were used on either end of the columns during loading.
The testing program consisted of testing the five concrete columns under concentric loading and testing the six cylindrical concrete columns with different external confinement under the eccentric load. The hydraulically operated 5000 kN Denison compression testing machine, located in the Engineering Laboratory at the University of Wollongong was used to test all the columns in this study. All the columns were tested to failure.
4. Observed behaviour and test results 3.3. Specimen preparation Two batches of concrete were used to cast the concentrically and eccentrically loaded columns. The design compressive strength of both batches of concrete was 100 MPa. However, 73.62 MPa and 51 MPa were achieved for the concentrically loaded and eccentrically loaded columns, respectively. The three types of fibre-reinforced polymers used in this study were Carbon, Kevlar and E-glass. The epoxy system consisted of two parts, resin and slow hardener, were used to bond the FRP to the surface of the concrete columns. The process of applying the FRP is known as the wet lay up method and was used to wrap all the columns with external FRP confinement.
The failure of the columns in all cases was brittle and in the case of the plain specimen, a very explosive failure. In the case of the FRP confined columns, the snapping of the fibres could be heard throughout the loading as the concrete tried to expand. While for the two galvanised steel straps reinforced columns, failure was sudden and soundless. In each case the straps may have yielded but did not break. This suggested that the failure of the columns was a direct result of cracking of the concrete tensile flexure. This can be explained that this type of reinforcement may not be suitable for columns under eccentric load. Tables 3 and 4 present the testing results of the concentrically loaded and eccentrically loaded columns, respectively.
Table 3 Testing results of the concentrically loaded columns Column (1)
Configuration (2)
Ultimate load (kN) (3)
Axial stress (MPa) (4)
Axial deflection (mm) (5)
1 2 3 4 5
Plain Single-layered Carbon Single-layered Kevlar Three-layered Carbon Three-layered Kevlar
2351 2860 2490 2980 2490
71.23 86.65 75.44 90.29 75.44
5.048 4.404 4.234 6.514 5.574
Table 4 Testing results of the eccentrically loaded columns Column (1)
Configuration (2)
Ultimate load (kN) (3)
Axial stress (MPa) (4)
Axial deflection (mm) (5)
1 2 3 4 5 6
Three-layered Carbon Three-layered E-glass Three-layered Kevlar Galvanised steel straps at 10 mm spacing Galvanised steel straps at 20 mm spacing Internally reinforced
840.0 630.8 906.0 720.0 704.9 636.8
25.45 19.11 27.45 21.81 21.36 19.29
5.20 4.38 5.50 4.22 3.94 –
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
283
4.1. Concentrically loaded plain column This plain concrete column, as expected, presented a very brittle explosive failure. The column did not experience any excess deflection after reaching the maximum compressive load due to the lack of confinement, which led to the brittle failure. 4.2. Single-layered Carbon fibres column under concentric loading The external confinement provided to this column resulted in a higher ultimate load. However, the failure was still quite explosive and resulted in no increased deflection after reaching the maximum load as well. 4.3. Single-layered Kevlar column under concentric loading The single-layered Kevlar column achieved a slight increase in ultimate load over the plain specimen. And the most promising aspect about this column is that there was a small amount of excess deflection achieved after ultimate load. The failure was less explosive and the column was almost fully confined even after failure. 4.4. Three-layered Carbon specimen under concentric loading This specimen achieved significantly better results than the single layered specimen both in strength and deflection. It is of significance to note that the column still appeared to be fully intact after failure. Upon closer inspection, it could be seen that the jacket had a section where the jacket has frayed rather than actually fractured. This meant that even after failure the column still had the ability to withstand load and still maintain its integrity. 4.5. Three-layered Kevlar specimen under concentric loading The three-layered Kevlar specimen also outperformed the single-layered specimen, also achieving higher strength and ductility. However, there was not as much excess deflection achieved as the Carbon wrapped specimen. This specimen also remained intact after failure except for the presence of small fractures in the jacket.
Fig. 3. Failure of the internally reinforced concrete column.
even after the concrete cover had spalled away, the confined core continued to carry an increasing load. Fig. 3 shows the column after failure. 4.7. Carbon fibre jacketed concrete columns under eccentric loading Failure of the column was marked by brittle rupture of the hardened fibres at the bottom of the column, which can be seen in Fig. 4. Failure was sudden and quite explosive. During various stages of loading, the snapping sounds could be heard, which were attributed to the cracking of the concrete and the stretching of the hardened fibres. The test results show that this column could withstand much higher ultimate load than the internally reinforced column. This reveals that Carbon confinement could provide significantly greater confining pressure to the high strength concrete column. 4.8. E-glass wrapped specimen under eccentric loading Failure of the E-glass wrapped column specimen was marked by fibre rupture at the top of the column. Although it was sudden, the failure could be predicated by the appearance of white patches at the top of the column as the result of the fibre stretching. From Fig. 5, it can be seen the layers of E-glass were torn as a result of the eccentric load applied to the column. As the external Eglass confinement tried to prevent the concrete from expansion under loading, it was ruptured when the tensile stress, applied by the concrete lateral expansion, became too large.
4.6. Internally reinforced specimen under eccentric loading This specimen exhibited a brittle failure under the eccentric loading. The concrete cover started to fall away due to lateral dilation under the loading. However,
Fig. 4. Failure of the Carbon fibres column.
284
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
Fig. 7. Failure of the steel strapped column with 10 mm spacing.
Fig. 5. Failure of the E-glass wrapped column.
The results show that the load carrying capacity of this column wrapped with E-glass was slightly lower than the internally reinforced column. 4.9. Kevlar wrapped concrete column under eccentric loading The material used to wrap this column is one sheet of Kevlar in 920 mm wide rather than the roll of tape. The failure mode of this column was similar to that of Eglass specimen: fibres were ruptured at the top end of the column. This can be seen in Fig. 6. Cracking of Kevlar fibre could be heard throughout the testing with the failure of the column signified by a loud snap of the Kevlar jacket. The largest load carrying capacity was achieved by this column compared to other eccentrically loaded columns. This is contributed to the external confinement in terms of continuous sheet. Also, the failure was sudden and loud. 4.10. Galvanised steel strapped column at 10 mm spacing under eccentric loading It was found this column failed in the tensile bending region under eccentric load. Fig. 7 shows the failure of this column occurred in the space of two straps. This can
Fig. 8. Failure of the steel strapped column with 20 mm spacing.
be explained as a result of there being no reinforcement in this region. However, the crack was much smaller and failure occurred closer to the bottom of the column when compared to the column with steel straps in 20 mm spacing. The failure was brittle and soundless. 4.11. Galvanised steel straps at 20 mm spacing under eccentric loading Failure of this column was similar to that of the other galvanised steel straps wrapped column, in that the cracking of the concrete on the tension side marked the failure. Also evident in Fig. 8, is that failure, again occurred in between the galvanised steel straps and the straps themselves again, did not show any sign of failing. As the increased spacing of straps resulted in a larger area of the column being un-reinforced, the column failed in a substantial crack in the concrete. The results shown in Table 4 confirm that the larger the spacing between the straps, results in a lower load carrying capacity.
5. Comparisons and analysis
Fig. 6. Failure of the Kevlar wrapped column.
From the two sets of experiments conducted in this study, it can be noted that the Carbon wrapped columns outperformed the other types of reinforced columns except the Kevlar sheet wrapped column, which was
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
proved by the testing results of the eccentrically loaded columns. The testing results indicated that Carbon fibres wrapping is more effective for the external confinement compared to the galvanised steel straps and E-glass. However, this is not the case in the Kelvar sheet wrapping column, which presented the largest loading capacity as the continuous sheet was used as the external confinement. Again, this proved the layout of fibres has a significant influence on the behaviour of the eccentrically loaded columns. The comparison among the eccentrically loaded columns shows that all the externally reinforced columns out-performed the internally reinforced column excluding the E-glass specimen, which almost achieved the same strength. The E-glass was confirmed to be the weakest reinforcing material, which presented an ultimate load 10% lower than that of the two Band-It columns and a 44% decrease in compressive load compared to the Kevlar fibre sheet confined column. Another comparison made between the two galvanised steel straps wrapped columns shows that the larger the spacing between the straps results in a lower load carrying capacity. However, the column with galvanised steel straps in 20 mm spacing exhibited only 2.2% decrease in the loading carrying capacity. Nonetheless, there was a 26% and 16% decrease in ultimate load over the Kevlar fibre sheet and Carbon fibre confined columns respectively. And both columns achieved slightly higher ultimate load compared to the internally reinforced column, which proved the external confinement with galvanised steel straps is also more effective than the internal reinforcement. But the failure of the columns with this type of reinforcement is sudden, which indicates that the galvanised steel straps have very little effect on improving the ductility of the columns. The comparison among the concentrically loading columns confirmed that the confinement significantly enhances the strength, stiffness and ductility of high strength concrete, in particular when applied in multiple layers.
285
6. Theoretical calculations The analytical model outlined above was used to calculate the confining stresses in the columns during testing. It is to be noted that these models have been established for concentrically loaded specimens. For eccentrically loaded specimens, two types of stresses are encountered, axial stresses and bending stresses. The first step in calculating the confining stress is to determine the value of the hoop stress. This can be determined by using the experimental values obtained from the tensile tests conducted on the FRP coupons and steel straps. Table 5 presents the values of tensile strengths that were obtained for the different materials, which include two lots of materials used for the concentrically loaded and eccentrically loaded columns, respectively. The tensile strength of the material was used to predict the confining stress in the columns. This was achieved by using Eqs. (1) and (2). Table 6 shows the calculated values of confining stresses, which indicates that the Carbon fibres to be superior. The confining stresses of the helically reinforced columns were calculated using Eq. (2). These confining stresses are shown in Table 7. As shown in Table 8, the confinement effectiveness was obtained for the different specimens using Eq. (4). Table 8 compares the theoretical calculating results of all the externally reinforced columns under concentric loading with their experimental results. The theoretical ultimate load was calculated based on the calculated compressive strength of confined concrete using Eq. (3) and the concrete column cross sectional area. It shows that the theoretically determined confining stresses generally overestimates the experimental results. The most matched results were obtained by the one-layered FRP wrapped specimens. This may be due to the tensile testing on the material was restricted to a single-layered coupon. The strength of three-layered FRP was assumed to be three times of the strength of single-layered coupon.
Table 5 Tensile strength of reinforcing materials Specimen (1)
Thickness (mm) (2)
Tensile strength (MPa) (3)
Concentrically loaded columns
One-layered Carbon One-layered Kevlar Three-layered Carbon Three-layered Kevlar
1.360 1.635 3.060 3.700
305.15 104.59 406.90 138.65
Eccentrically loaded columns
Three-layered Carbon Three-layered E-glass Three-layered Kevlar Galvanised steel straps Helix
0.900 0.500 1.000 0.500 10.000a
692.27 196.00 162.00 627.50 662.10
a
Diameter of steel bars.
286
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
Table 6 Confining pressure of FRP confinement Specimen (1)
Confining stress fr (MPa) (2)
Concentrically loaded columns
One-layered Carbon One-layered Kevlar Three-layered Carbon Three-layered Kevlar
4.05 1.67 12.15 5.00
Eccentrically loaded columns
Three-layered Carbon Three-layered E-glass Three-layered Kevlar
6.08 0.96 1.58
Table 7 Confining pressure of steel reinforcement
(1)
Helix stress fus (MPa) (2)
Steel area Ast (mm2 ) (3)
Diameter Dh (mm) (4)
Pitch s (mm) (5)
Confining pressure fr (MPa) (6)
Internal helix Straps at 10 mm spacing Straps at 20 mm spacing
662.10 627.50 627.50
78.5 10.0 10.0
165 205 205
60 30 40
1.04 1.74 1.23
Material
Table 8 Comparison between theoretical and experimental results for the concentrically loaded columns Specimen (1)
fr (MPa) (2)
k1 (3)
fcc0 (MPa) (4)
Theoretical ult. load (kN) (5)
Experimental ult. load (kN) (6)
One-layered Carbon One-layered Kevlar Three-layered Carbon Three-layered Kevlar
4.05 1.67 12.15 5.00
3.06 3.44 2.65 2.98
86.01 79.36 105.82 88.52
2837 2618 3491 2920
2860 2490 2980 2490
7. Conclusions The work carried out in this study involved two sets of testing: five columns under concentric loading and six columns under eccentric loading, which are mainly set to evaluate the effectiveness of various types of the external reinforcement. The results from both sets of tests allow the following conclusions to be drawn: 1. The methods of external reinforcement can be used as an alternative method of reinforcement to enhance the properties of high strength concrete. It has been shown that the confinement of the concrete prevents the concrete from expanding and therefore allows the concrete to absorb higher stresses, resulting in a higher load carrying capacity. 2. The tests proved that the benefits of confinement could be enhanced by applying multiple layers, which can be seen from the results of testing the concentric loading columns. 3. The test results also indicated that the Carbon fibres provides the greatest amount of confinement, and
4.
5.
6.
7.
had significantly better results, if the external confinement was achieved by the application of FRP in roll of tape. The highest load carrying capacity achieved by Kevlar sheet wrapped column confirms that the wider rolls of the fibre reinforcement can provide a greater confining stress. This also can be concluded that the fibre layout has significant influence on the behaviour of concrete structural members. The external confinement with galvanised steel straps improves the strength of the column to a certain extent. The brittle, sudden, soundless failure of the galvanised steel straps wrapped columns shows that the galvanised steel straps had very little effect on improving the ductility of the columns. The E-glass proved to be the weakest reinforcing material in this study. The ultimate load achieved by the E-glass wrapped specimen even lower than the internally reinforced column. The comparisons made between theoretical calculations and experimental results show that the models prepared by Samaan et al. [10] and Pessiki and Peroni
M.N.S. Hadi, J. Li / Composite Structures 65 (2004) 279–287
[11] and can be used to estimate the ultimate strengths of externally reinforced columns under concentric loading to a certain degree accuracy.
Acknowledgements The authors acknowledge the financial assistances of the University of Wollongong through its Grant scheme, which enabled conducting this research. The technical assistance of Messers Ian Bridge, Alan Grant and Bob Rowlan of the University of Wollongong are appreciated. References [1] Mansur MA, Chin MS, Wee TH. Stress–strain relationship of high-strength fiber concrete in compression. J Civil Eng 1999;11(February):21–9. [2] Razvi SR, Saatcioglu M. Strength and deformability of confined high-strength concrete columns. ACI Struct J 1994;91(November– December):678–87. [3] Hadi MNS, Schmidt LC. Use of helices in reinforced concrete beams. ACI Struct J 2002;99(2):191–8.
287
[4] Attard MM, Mendis PA. Ductility of high strength columns. Australian Civil Eng Trans 1993;35(4):295–306. [5] Li J, Hadi MNS. Behaviour of externally confined high strength concrete columns under eccentric loading. J Compos Struct 2003;62(2):145–53. [6] Pessiki S, Harries KA, Kestner JT, Sause R, Ricles MJ. Axial behaviour of reinforced concrete columns confined with FRP jackets. J Compos Constr 2001;5(November):237–45. [7] Pantazopoulou SJ, Bonacci JF, Thomas MDA, Hearn N. Repair of corrosion-damaged columns with FRP wraps. J Compos Constr 2001;5(February):3–11. [8] Richart FE, Brandtzaeg A, Prown RL. A study of the failure of concrete under combined compressive stress. Bulletin No. 190, 1928, Engineering Experiment Station, University of Illinois, Urbana, Apr. p. 73. [9] Xiao Y, Wu H. Compressive behaviour of concrete confined by carbon fibre composite jackets. J Mate Civil Eng 2000;12(2):139– 46. [10] Samaan M, Mirmiran A, Shahawy M. Model of concrete confined by fibre composite. J Struct Eng 1998;124:1025–31. [11] Pessiki S, Pieroni A. Axial load behaviour of large-scale spirally reinforced high-strength concrete columns. ACI Struct J 1997;94:305–15. [12] Lam L, Teng JG. Strength models for fiber-reinforced plasticconfined concrete. J Struct Eng 2002;128(5):612–23. [13] Karbhari VM, Gao Y. Composite jacketed concrete under uniaxial compression-verification of simple design equations. J Thermoplast Compos 1997;6:92–107.