The seismic behavior of short circular reinforced concrete columns was studied by testing seven columns retrofitted with prestressing steel wire (PSW), subjected to combined constant axial compression and lateral cyclic load. The main test parameters were configuration index of PSW, prestressing level of PSW, and axial compression ratio. An analysis and discussion of the test results including failure mode, hysteresis curves, skeleton curves, ductility, and degradation of stiffness was done. The results show that the seismic performance of the retrofitted specimens could be effectively enhanced even if the axial compression ratio of columns reached 0.81. The ductility index and the energy absorption capacity of the retrofitted specimens increase with the prestressing level of PSW. The formulas for calculating shear capacity of RC short columns strengthened with PSW were proposed which may be useful for future engineering designs and researches.
Many of the catastrophic failures of bridges or frame structures that occurred during the past earthquakes were due to the failure of one or more of the RC columns. Shear damage sustained by short and stubby columns is often responsible for the collapse of the entire structure. The deficiencies in the seismic shear resistance may be attributed to the old design provisions, which resulted in lack of sufficient transverse reinforcement. This deficiency may still be found in some of the more recently designed columns. While lack of adequate transverse reinforcement may lead to diagonal tension failure in shear dominant columns [
A large number of existing bridge columns or frame columns are also susceptible to anchorage failure because of the location of longitudinal reinforcement splices, which often coincide with potential plastic hinges regions. Lack of flexural ductility problem is common in the bridge piers or short frame columns, which arises from two sources: (1) insufficient transverse reinforcement that results in lack of adequate confinement and (2) inadequate lap splice length in the plastic hinge zone. Such events motivated many researchers to do study on the reason behind RC bridge columns failure during earthquakes.
It has also been proven that the ductility capacity of columns is enhanced significantly when the core concrete is well confined. Passive confinement of concrete using external steel jackets or fiber reinforced polymers (FRPs) wrapping is the most common method used to improve the ductility capacity of the columns [
There have been a number of studies that attempted to explore the feasibility of using active confinement for seismic retrofit of concrete columns. Seismic behaviors and compressive strength of square cross section RC columns retrofitted with prestressed CFRP and AFRP strips was researched [
University of Ottawa [
The authors therefore developed a new type of prestressing anchoring system for retrofitting concrete column with high-strength steel wires [
The diameter of RC cylindrical columns is 300 mm and of height 430 mm. The column is prepared using C35 concrete. Vertical casting is applied, where first the ground beam is cast then three weeks later the column and column head are cast. The circular cross section column template is made using PVC pipe; ground beams and column head are cast in wooden templates.
The mixing ratio for the C35 concrete is cement 311 kg/m3, sand 854 kg/m3, gravel 1035 kg/m3, water 170 kg/m3, ordinary Portland cement PO.42.5, medium sand, and maximum size of gravel particle 25 mm. The mixing ratio for the epoxy polymer mortar is cement 711 kg/m3, sand 1434 kg/m3, water 340 kg/m3, polypropylene fiber 2.1 kg/m3, epoxy polymer 2.7 kg/m3, ordinary Portland cement PO.42.5, and medium sand.
Three cubic specimens and three prismatic specimens were prepared for concrete and polymer mortar under the same conditions with columns, to test the compressive strength of the concrete and mortar. The tested compressive strength of the cubic and prismatic concrete specimens at 40 days is 41.5 MPa and 33.0 MPa, respectively; the compressive strength of the polymer mortar is 53.60 MPa. The nominal diameter of the steel strands is 4.5 mm and cross section area is 9.62 mm2, the proportional and ultimate tensile strengths of strands are 1320 MPa and 1750 MPa, respectively. The stirrup grade is HPB235, of diameter 8 mm, yield tensile strength 407.3 MPa, and ultimate tensile strength 459.0 MPa. Longitudinal rebar is HRB335, of diameter 16 mm, yield tensile strength 385.3 MPa, and ultimate tensile strength 537.3 MPa. The curing age is 40 days and curing is done at room temperature for all columns, cubic and prismatic concrete specimens.
Shear span ratio of column is
In the experiment, 8 circular cross section short columns were designed, of which 7 were retrofitted and one control specimen. Specimen shear span ratio is 1.93, volume ratio of stirrup reinforcement is 0.25%, and longitudinal reinforcement ratio is 2.28%. Specimen geometry and reinforcement details are shown in Figure
Details of the specimens.
The variables include 4 different prestressing levels, 3 different strand spacing measurements (30 mm, 60 mm, and 90 mm) and 3 different design axial compression ratios (0.6, 0.81, and 0.9). Design axial compression ratios
The retrofitting setup and strand tensioning anchoring are shown in Figure
Test parameters.
Specimen |
Design axial compression ratio |
Strands spacing (mm) |
|
|
---|---|---|---|---|
PZ1 | 0.81 | — | — | — |
PZ2 | 0.81 | 30 | 0.237 | 0 |
PZ3 | 0.81 | 30 | 0.237 | 0.30 |
PZ4 | 0.81 | 30 | 0.237 | 0.40 |
PZ5 | 0.81 | 30 | 0.237 | 0.50 |
PZ6 | 0.60 | 60 | 0.119 | 0.40 |
PZ7 | 0.81 | 60 | 0.119 | 0.50 |
PZ8 | 0.90 | 90 | 0.079 | 0.65 |
Circular columns prestressed with steel wires.
Strands configuration value
Loading setup is shown in Figure
Test setup.
Force or displacement control method is applied. Horizontal loading is by load or displacement control method: before yielding of the longitudinal reinforcement, load control is applied, where the loading level increased after each loading cycle. After yielding, the displacement control method is adopted, where the displacement value just as the longitudinal reinforcement yields is taken as the displacement control value for each loading stage, where loading is twice for each cycle. When the horizontal load drops to 85% of the ultimate load, the specimen has attained failure load and deformation; therefore the test is terminated.
The value of axial load and horizontal cyclic load is measured by the force sensor. A dial indicator is fixed on the foundation beam to measure the foundation horizontal displacement, two rod displacement meters attached the top of the column side, take the difference between the mean displacement at the top and bottom of the beam as the final column displacement under horizontal loading.
Paste strain gages on the column longitudinal reinforcement and stirrups to measure the strain. For retrofitted column, apart from longitudinal reinforcement and stirrups strain gages, others are also attached on the steel strands to measure the strand strain under horizontal load. Horizontal load and displacement as well as all the data are recorded by computer-controlled data acquisition system.
The main parameters of each specimen like design axial compression ratio, strand stirrup configuration index, and prestressed levels vary, hence cracking and failure modes are not the same. Typical specimen failure pattern is shown in Figure
Failure patterns of specimens.
PZ1
PZ2
PZ3
PZ4
PZ5
PZ6
PZ7
PZ8
Specimens PZ4 (
The specimen PZ6 (
The failure process of specimen PZ8 (
Throughout the experiment the steel wires work normally during loading, since the anchorage system is highly efficient and reliable. For the retrofitted specimens, the lateral expansion of the concrete is contained by the steel strands and stirrups such that the deformation capacity of the specimen is significantly improved. Concrete confining effect can be reflected by the respective specimen’s strands tensile strain level. The specimens stirrups and strands strain are compared by recording the second stirrup strain (180 mm from the column top surface) and the strand corresponding to stirrup’s position. Figure
Comparison of strains in stirrups and steel wires.
From Figure
With other parameters remaining equal, the specimens with lower prestressed level, stirrups strain increases rapidly; these specimens stirrups strain value is significantly higher than specimens with high prestress level, as seen in Figure
The measured force-drift typical hysteresis curve is shown in Figure
Typical force-drift hysteretic curves.
PZ1
PZ2
PZ3
PZ4
PZ5
PZ6
PZ7
PZ8
From Figure Before yielding, the area under the force-drift hysteresis curves of all specimens is small, and after unloading the residual deformation is very small. After yielding, increase in loading cycles and the horizontal displacement leads to significantly reduced force-drift curve slope. After unloading residual deformation gradually increases, which indicates that the stiffness of the specimen is deteriorating. Once reaching displacement control loading stage, the unretrofitted column PZ1 undergoes shear failure, hysteresis loop at elastic-plastic phase takes an arched shape, and pinching of the curve is observed, which displays shear deformation characteristics and poor seismic performance. By applying prestressed steel strand retrofitting, hysteresis loop is generally plumped; ductility and energy dissipation capacity are good. Comparing the force-drift curves of PZ1, PZ2, PZ3, and PZ4 specimens, it can be seen under the same Under the same axial compression ratio and strand prestressing level (for specimens PZ5 and PZ7), increasing As the axial compression ratio increases, the hysteresis curve becomes relatively narrow. After yielding of the specimen there is significant strength and stiffness degradation. The greater the displacement, the more the strength and stiffness degradation. At later loading stages, the force-drift curve becomes more unstable.
The specimen’s
Skeleton curves of specimens.
The characteristics of each specimen (load point and displacement) are listed in Table
Test results of key points in
Specimen number | Load (kN) | Displacement (mm) | ductility | Failure mode | ||||
---|---|---|---|---|---|---|---|---|
|
|
|
|
|
|
Increase (%) | ||
PZ1 | 337.1 | 368.0 | 312.8 | 6.07 | 20.4 | 3.36 | — | Shear |
PZ2 | 359.7 | 405.0 | 344.3 | 8.09 | 39.9 | 4.93 | 33.6 | Bending shear |
PZ3 | 376.7 | 419.3 | 356.4 | 7.02 | 41.14 | 5.86 | 74.4 | Bending |
PZ4 | 382.5 | 431.1 | 366.4 | 5.15 | 32.20 | 6.25 | 86.0 | Bending |
PZ5 | 389.0 | 419.8 | 356.8 | 7.52 | 48.46 | 6.44 | 91.7 | Bending |
PZ6 | 367.0 | 381.4 | 324.2 | 6.74 | 37.56 | 5.57 | 65.6 | Bending |
PZ7 | 385.0 | 408.5 | 347.2 | 7.33 | 37.09 | 5.06 | 50.6 | Bending shear |
PZ8 | 397.9 | 451.2 | 383.5 | 6.75 | 25.60 | 3.71 | 10.4 | Bending shear |
Note: due to equipment failure during the loading process of specimen PZ5, resulting in larger axial pressure fluctuations, the measured horizontal load is small.
From Figure
After yielding prestressing steel strands reinforced columns exhibit good carrying capacity; after peak load, the skeleton curve enters a horizontal stage and a gentle descending stage, and seismic performance is effectively improved. Compared with the control column, nonprestressed steel strand retrofitted column PZ2 displays improved deformation, but at late loading stage skeleton curve decreases rapidly and the descending segment is shorter. For lower strand configuration index specimen PZ8, though strand prestressed level is high, due to high axial compression ratio, descending stage exhibits sharp decline and bending-shear failure. When steel strand configuration index is too low, the prestressing effect on improving the seismic performance of short columns with high axial compression ratio is not significant. Under high axial compression ratio (
Ductility is an important parameter to determine the deformation capacity. In this paper ductility coefficient
From Table when the axial compression ratio and the strand configuration index are fixed, increasing the level of strand prestressing can improve the deformation capacity of the specimens. Compared with nonprestressed retrofitted specimen PZ2, the ductility coefficient of prestressed reinforced short columns PZ3 ( Specimen PZ5 compared with PZ7, strand configuration index is doubled and ductility factor is increased by 27.3%. This is because other conditions are fixed, and confinement effect of core concrete increases with increase in steel strands configuration index. As axial compression ratio increases, the ultimate load increases, ultimate displacement decreases, the descending segment of the skeleton curve becomes steeper, bearing capacity of concrete members after ultimate load decreases rapidly, and deformation capacity reduces. Comparing PZ6 and PZ7; with equal strand configuration index and the prestressed level increased from Comparing the nonprestressed specimen PZ2 (with high strand configuration index, The strand configuration index of highly prestressed level specimen PZ8 is only 0.079. Compared with the control specimen, the ductility index of PZ8 is not significantly improved, indicating that in order to guarantee the retrofit effectiveness of high axial compression ratio, on the one hand strand configuration index should not be too small, and on the other hand strand prestressing level should be increased.
Secant stiffness The relative stiffness degradation curves for all specimens generally have three stages: sharp decline of stiffness stage at initial loading, gradual stiffness degradation, and stiffness stabilizing stage. At initial loading ( At descending stages, the stiffness degradation rate increases with increase in specimen axial compression ratio but decreases with increase in the strand configuration index and prestressing level, and when
Relative stiffness degradation curves of specimens.
When the existing columns with insufficient shear resistance are subjected to seismic loading, the stirrups yield prematurely, causing the columns to undergo shear failure. Appearance of crossed cracks can be observed when the columns undergo stress reversals, and this will lead to damage of concrete. The external prestressed retrofitting by the strands plays the role of inhibiting the development of diagonal cracks, controlling of crack width, and also significantly increases the shear resistance, compression strength, and durability of the columns. It is therefore very clear that retrofitting by transverse prestressing strands significantly improves the bearing capacity of concrete columns and also promotes shear resistance.
From the experiment results and analysis, a formula to calculate the shear capacity of prestressing strand retrofitted cylindrical concrete columns is proposed:
In formula (
From the experiment it is found, when prestressed steel strand retrofitted specimen horizontal displacement is 20 mm (the ultimate displacement angle 1/29), specimen cracks are many and dense, wide through cracks are not formed, and the aggregates interlocking role is existent. When stirrups strain has reached yield value, the strand strain increment is greater than 2500
In Table
Comparison of calculation and test results.
Specimen number |
|
|
|
|
|
---|---|---|---|---|---|
PZ2 | 188.53 | 56.75 | 405.03 | 245.28 | 1.65 |
PZ3 | 188.53 | 192.95 | 419.30 | 381.48 | 1.10 |
PZ4 | 188.53 | 238.34 | 431.06 | 426.87 | 1.02 |
PZ5 | 188.53 | 218.35 | 419.80 | 406.88 | 1.03 |
PZ6 | 188.53 | 118.35 | 381.36 | 306.88 | 1.24 |
PZ7 | 188.53 | 141.73 | 408.94 | 330.26 | 1.24 |
PZ8 | 188.53 | 117.28 | 451.19 | 305.81 | 1.48 |
Prestressing strand retrofitting technology can significantly improve the seismic performance of short columns with large axial compression ratio and makes short columns failure pattern to change from shear failure to flexural-shear failure or flexural failure. Under high axial compression ratio from 0.60 to 0.90, the specimens still exhibit good ductility, indicating that the plastic deformation capacity of RC column is greatly improved by prestressed steel strands retrofitting. With the increase in horizontal displacement, the confinement function of stirrups and shear capacity is gradually shared by the steel strands. The higher the prestressed level and configuration index get, the earlier the steel strands get subjected to forces, and thus the greater the shear force is shared. The ductility of the specimen increases nonlinearly with increase in strand configuration index. When the strand configuration index increases to a certain value, there is slower increment of the ductility of the specimen with increase in configuration index; for RC column specimens with axial compression ratio greater than 0.80, a configuration index value not less than 0.237 is suggested. The retrofitted specimen magnitude of stiffness degradation increases with increase in axial compression ratio and decreases with increase in strand configuration index and prestressed level; when A formula for shear capacity of prestressed steel strand retrofitted RC cylindrical columns is established, the calculated and experimental values agree well, and safety is guaranteed; it can be used for practical engineering designs.
The authors declare that there is no conflict of interests regarding the publication of this paper.