International Journal of Civil, Structural, Environmental and Infrastructure Engineering Research and Development (IJCSEIERD) ISSN 2249-6866 Vol.

2 Issue 4 Dec - 2012 87-100 TJPRC Pvt. Ltd.,

ENERGY DISSIPATION IN LOW STRENGTH CONCRETE BRIDGE COLUMNS

ALI M. SYED, BASHIR ALAM & MOHAMMAD JAVED Faculty member, Department of Civil Engineering, University of Engineering & Technology Peshawar, Peshawar 25120, Pakistan

ABSTRACT
Low strength concrete is seen in many bridges in developing countries like Pakistan. The high seismic demand warrants investigation to see how such bridges would dissipate energy in extreme loading that may occur due to earthquakes. Two scaled bridge columns were fabricated with target strength of 2,400 psi. The columns were tested by subjecting them to reverse cyclic quasi-static loading. Load and displacement data was recorded which was then used to plot the hysteretic curves. The amount of energy that dissipated in every cycle and cumulative energy dissipation for each column was calculated. This experimental study provided high value results regarding the energy dissipation and the ultimate displacement capacity. Further, number of cyclic loading also affects the amount of energy dissipated and maximum displacement of column.

KEY WORDS: Energy, Dissipation, Hysteresis, Bridge Column, Quasi-Static Loading INTRODUCTION
The earthquakes cause the ground to move. The ground excitation imparts inertial forces to bridge structure. The causative ground motion results in significant lateral loading (AASHTO LRFD, 2007) of the bridges and these forces are to be resisted by the bridge columns. Single column bridges have only one load path and are most vulnerable due to lack of redundancy (Syed, 2009). Seismic performance of reinforced concrete (RC) bridge columns depends on amount of energy it can dissipate (Syed, 2009). When lateral loads act on the bridge, plastic hinge is formed in RC column due to forces that exceed the elastic moment capacity (Syed, 2009). The formation of plastic hinge results in columns that are flexural dominant (Syed, 2009). Bridge designs generally revolve around energy dissipation in a seismic event, which, if avoided, requires uneconomical sections that may seem to be impracticable Syed (2009), Preistley et. al (1996). However, relying on energy dissipation requires reasonable ductility to be present in plastic hinge zone without which the concept of energy dissipation through the plastic hinge cannot be used (Preistley et. al, 1996). The lateral motion of bridges due to ground shaking that causes the cracking and yielding in the plastic hinge zone of columns, results in energy dissipation through hysteresis Kawashima (2006), Poljansek et al. (2009). The reverse cyclic nature of inelastic deformation is the main cause of hysteresis in RC columns (Ming-Liang & Surendra, 1987). In single column bridges that have a flexural dominant design, plastic hinge usually forms near the column base (Ming-Liang & Surendra, 1987). This paper presents laboratory testing of two scaled RC bridge columns that have solid circular section. A comprehensive field survey was undertaken and it was revealed that many bridges in northern part of Pakistan had strength below 2,500 pounds per square inch (psi) and some even had strength as low as 2,000 psi, Syed (2009), Ming-Liang & Surendra, (1987). Pakistan lies in highly seismic active area with potential of large seismic events around Mw8 or above (Bilham and Wallace, 2006), therefore a study of bridge column made of low strength concrete reinforced with mild steel

88

Ali M. Syed, Bashir Alam & Mohammad Javed

bars was undertaken to evaluate their capacity. It was decided to fabricate two columns of 2,400 psi target strength (Syed, 2009). An experimental program was devised to perform the reverse quasi-static cyclic testing in the laboratories of Earthquake Engineering Center. The experimental investigations provided quantifiable results of the hysteretic energy that the columns could dissipate.

DESCRIPTION OF TEST COLUMNS

Two test columns with target strength of 2,400 psi were prepared in the laboratory. They are named QSCT-2-003 and QSCT-3-004. These columns had solid circular cross-section with dead mass supported on top, and these represented typical hammer-head bridge columns generally found in study areas in northern part of Pakistan. However, the authors believe that similar columns do exist in other parts of the country and in other countries as well. Both the test columns were similar in geometry and size, as shown in Figure 1 with similar materials as indicated in Table 1. The features of test columns were deduced from the extensive field survey Syed (2009), Syed M. Ali et al, (2011), Naeem et al. (2005), EERI (2006), Dellow et al. (2006) and Syed and Shakal (2007). The summary of field survey of bridges is presented in Table 2. Scaling After similitude analysis, scaling factors were finalized as presented in Table 3. During selection of scale factor, the limitations that were considered were capacity of lab facilities (hydraulic actuator capacity, size of testing floors and walls, capacity of sensors), cost and ease of construction. The side view of test column is presented in Figure 2 in which geometric details can be seen. The top view of the test column can be seen in Figure 3. Materials As seen from Table 3 that scale factor for material properties

E of concrete and reinforcing steel was taken

unity; this indicates that no scaling was done for the material properties of the model concrete, also called micro-concrete. This is generally recommended for inelastic testing (Reinhorn, 1992). The two columns QSCT-2-003 and QSCT-3-004 had target strength of 2400 psi (16.5 MPa). The mix design for micro-concrete is presented in Table 1 and the mechanical properties of reinforcing steel used in test columns are provided in Table 4. The reinforcement detail of test column is shown in Figure 4, whereas Figure 5 shows the details of rebar and confinement steel in cross sectional view. Dead Mass The required dead mass in test column was 19.24 tons shown in Figure 1. This mass would produce the same level of stresses as found in prototype bridge column.

Fig. 1. Scaled RC Bridge Columns under Reverse Quasi-Static Cyclic Testing

Energy Dissipation In Low Strength Concrete Bridge Columns

89

Fig.2: Side View (Elevation) of Components and Geometric Details of the Model Bridge Column
2438

2133

914

2438

Column base

305mm 762mm square top pedestal

Fig.3: To View of the Test Column with Modular Mass on Top

Fig.4: Reinforcement Details of Bridge Column and Base of Column in Elevation

90

Ali M. Syed, Bashir Alam & Mohammad Javed

26-7.37mm rebar

5-1mm spiral @ 37.5mm pitch

=305mm
Fig.5: Cross Section of Test Column Showing Reinforcement Details

EXPERIMENTAL PROGRAM
The reverse quasi-static cyclic testing to study the hysteretic energy dissipation was conducted in laboratories of Earthquake Engineering Center of University of Engineering and Technology Peshawar. Hydraulic actuator of 50 ton force capacity was used for lateral loading in longitudinal (North-South) direction. The corresponding displacements were measured with the displacement transducers attached to the midpoint of pedestal on column top as seen in Figure 2. This is centerline of lateral load applied by the hydraulic actuator. Testing Rig The test setup comprised of strong floor and strong wall used for anchoring the test specimen and hydraulic actuator respectively as shown in Figure 1. The base of the column was anchored down with strong floor using four 44 millimeter diameter mild-steel bolts. Instrumentation and Data Acquisition Data was recorded for the lateral force applied and displacement of the end of hydraulic actuator. The data of the two channels was recorded using UCAM-70 data acquisition system. In this study, the data was sampled at frequency of around 1.5 Hz. It is important to recall here that the frequency of cyclic testing was around 0.0067 Hz, which shows that the sampling frequency was well above the Nyquist-Shannon sampling frequency (Dally, Riley, & McConnell, 2004). The data acquisition system with other equipment is shown in Figure 6.

Fig.6: Data Acquisition System with Allied Accessories

Energy Dissipation In Low Strength Concrete Bridge Columns

91

Testing Protocol The tooth-saw loading waveform in reverse quasi-static testing was used with a frequency of 0.0067 Hz (150 seconds). Two testing schemes were used, Scheme-1 for first column QSCT-2-003 shown in Figure 7. In this scheme three cycles per drift were applied until failure occurred at 4% drift and total of 37 cycles were applied. For all the cases the testing was to be continued at least till 20% reduction in strength was observed, which means that during the test the maximum force was monitored for each cycle until a cycle experiences a maximum force that is 80% of the maximum force in the preceding cycles Kawashima (2006) and Poljansek et al. (2009). This point corresponds to a failure state or the state beyond which performance is not acceptable.

Fig 7: Scheme-1, used for Testing Only First Column QSCT-2-003 Scheme-2 was used for the column QSCT-3-004 and is as shown in Figure 7. In this scheme 2 cycles per drift were used with frequency of a cycle of 0.0067 Hz. In this scheme total of 13 cycles were applied.

Fig.8: Scheme-2, used for Testing the Remaining Three Columns Sign Convention A positive force is used to describe push of actuator which is acting in north direction on the column whereas negative force is used for pull acting in south direction. In similar way, positive displacement is used for movement of column in north direction and negative displacement is used to describe displacement in south direction.

92

Ali M. Syed, Bashir Alam & Mohammad Javed

EXPERIMENTAL RESULTS
The two test columns are discussed here one-by-one. First Column QSCT-2-003 This column had strength of 2,420 psi cylinder strength when tested as per ASTM C39 (ASTM C39/ C39M, 2003) as described in Table 1. The testing began with one cycle of 0.1% drift which was followed by three cycles each of 0.2%, 0.4%, 0.5% and 0.75% drift. Minor hair line cracks were seen at around 0.5% drift level, which were so narrow that they were hardly visible. At 1.0% drift, visible cracks appeared that are shown in Figure 9 for the north and south face of the column.

Fig.9: Minor Cracks At 1% Drift of First Column QSCT-2-003

Following abbreviations are used in the discussion that follows in rest of the paper, Refer to Figure 10 for elaboration. Pc = Force at initial cracking; Uc = Displacement at initial cracking; Pyo = Force at initial yield; Uyo = Displacement at initial yield; Py = Force at yielding; Uy = Displacement at yielding; and Uu = Ultimate displacement at 80% of maximum restoring force The maximum restoring force in this test was (+) 4.96 kips in north direction and (-) 6.76 kips in south direction. From the analysis of hysteresis curves, it is seen that this point is the initial yield point, which means that at 1% drift (1.91 mm) the steel was at onset of yielding. Further from the analysis of data plotted for the displacement hysteresis curves it is observed that initial cracking of concrete occurred at around 0.48% drift, initial yield at 1.0% and yield at 1.13% for the force applied in north direction and for the force applied in south direction the cracking occurred at 0.50%, initial yield at 0.98% and yield at 1.10%, which is evident from the plot of backbone curve shown in Figure 10. The values for cracking, initial yield and yield are provided in Table 5.

Energy Dissipation In Low Strength Concrete Bridge Columns

93

Fig.10: Minor Cracks at 1% Drift of First Column QSCT-2-003 From the load-deformation data the hysteresis curves were plotted and energy dissipated in each cycle was calculated. Hysteresis curves for various stages of cyclic testing corresponding to 0.50%, 1.0%, 2.0%, 3.0%, 3.50% and 4.0% are shown in Figure 11.

Fig.11: Hysteretic Energy Dissipation Curves of First Column QSCT-2-003 at Various Drift Levels

94

Ali M. Syed, Bashir Alam & Mohammad Javed

It is noticed that energy dissipated per cycle increased with the increase in drift. The maximum energy dissipated was in first cycle of 4.0% drift. It is further noticed that energy dissipation per cycle is more in first cycle than second or third cycle and difference is less among the second and third cycle. The energy dissipated per cycle is plotted in Figure 12, from this figure it is clear that energy dissipation starts at around 1.0% drift which is onset of yielding. Here it is worth mention that energy dissipation happens when the system yields, prior to that, in elastic system there is no hysteretic energy dissipation.

Fig.12: Hysteretic Energy Dissipation for Each Cycle of First Column QSCT-3-004 The cumulative energy dissipated in this column QSCT-2-003 is 174.2 kip-in and the numbers of cycles are 34 before failure at around 4.0% is reached. Second Column QSCT-3-004 The main features of the testing were 2 cycles per drift until failure was observed. The loading scheme is shown in Figure 8. This column had strength of 2,300 psi cylinder strength when tested as per ASTM C39 (ASTM C39/ C39M, 2003) as described in Table 1. The testing began with one cycle of 0.1% drift which was followed by two cycles each of 0.25%, 0.50%, 1.0%, 2.0, 3.0% and 4.0% drift. Very thin hair line cracks appeared before 0.5% drift, which were so fine that they were not clearly visible. However, at 1.0% drift, visible cracks appeared on north and south face of the column. From the hysteresis curves, it is seen that 1% drift was the point of initial yield. Further from the analysis of data plotted for the displacement hysteresis curves it is observed that for north direction force initial cracking of concrete occurred at around 0.35% drift, initial yield at 1.0% and yield at 1.10% and for the force applied in south direction the cracking occurred at 0.50%, initial yield at 1.0% and yield at 1.10%. The values for cracking, initial yield and yield are provided in Table 5. From the load-deformation data the hysteresis curves were plotted and energy dissipated in each cycle was calculated. It is noticed that energy dissipated per cycle increased with the increase in drift. The maximum energy dissipated was in first cycle of 4.0% drift. It is further noticed that energy dissipation per cycle is more in first cycle than second cycle.

Energy Dissipation In Low Strength Concrete Bridge Columns

95

The energy dissipated per cycle is plotted in Figure 13, from this it is clear that energy dissipation starts at around 1.0% drift. The cumulative energy dissipated in all cycles is 110.39 k-in. Here in this column it is important to note that the numbers of cycles are 13 before failure which is observed around 4.0% drift.

Fig.13: Hysteretic Energy Dissipation for Each Cycle of Second Column QSCT-3-004.

RESULTS & DISCUSSIONS

In this paper reverse quasi-static cyclic testing of four low strength concrete columns is discussed. The first column QSCT-2-003 with strength of 2,420 psi is tested using 3 cycles per drift level. This scheme has effect of subjecting relatively large number of displacement cycles as compared to second scheme of 2 cycles per drift. Due to smaller increments of drift it resulted in more number of cycles until the failure of column. In this column failure was seen in second cycle of 4% drift which is the 37th cycle. The energy dissipated in this column is the greatest among the entire four test columns amounting to 174.2 k-in, which is due to the fact that the column has undergone greater number of cycles at various increments of increasing drift beyond elastic range. The ultimate displacement in this column was 3.05% which is least in all the four columns and this is also due to the fact that this column went under around three times more cyclic demand. The second column QSCT-3-004 had concrete strength of 2,300 psi and was tested using second scheme of loading which comprised of 2 cycles per drift. The column had undergone 13 cycles of loading until failure at second cycle 4% drift. The total energy dissipated was 110.39 k-in. The cycle of maximum energy dissipation was first cycle of 4% drift with 28.25 k-in energy dissipation. The overall energy dissipation was less than first column, around 37% less and it can be attributed to less number of cycles to which this column was subjected in post elastic state. The ultimate displacement in this column was 3.73% which is more than first column but is the highest among all the four columns.

CONCLUSIONS
This paper discusses the hysteretic energy dissipation in low strength concrete columns. From this study the following conclusions are derived: 1. 2. The hysteretic energy dissipation starts at around 1% drift in all columns which is onset of yielding. The energy dissipation also depends on the displacement demand and number of cycles of reversals.

96

Ali M. Syed, Bashir Alam & Mohammad Javed

3. 4.

More energy is dissipated in first cycle of any drift level than subsequent cycles. The failure in first columns with more number of reversals (that are 37) occurs earlier at 3.05% drift, whereas in second column with less number of reversals (that are 13) occurs at higher drift level of 3.75%. This concludes that displacement ductility is reduced due to more number of stress reversals.

5.

The energy dissipation per cycle in first column with more reversals is less than the second column with lesser number of reversals.

ACKNOWLEDGMENTS
The first author wishes to thank the Higher Education Commission (HEC) Islamabad for their PhD funding to undertake this research. The author also thanks the UET administration for their facilitation for carrying out this research.

REFERENCES
1. AASHTO LRFD Standard, 2007, Specifications AASHTO LRFD, Bridge Design Specification, 4th edition. Washington D.C. (USA): The American Association of State Highway Officials 2. Syed A. M. 2009, Study of Energy Dissipation Capacity of RC Bridge Columns under Seismic Demand Ph.D. dissertation. NWFP University of Engineering and Technology Peshawar, Pakistan 3. 4. M.J.N Priestley, F. Seible, G.M. Calvi, 1996, Seismic Design & Retrofit of Bridges, 1st Edition. Kawashima, K. (2006). Seismic Design, Isolation and Retrofit of Bridge. Tokyo: Department of Civil Engineering Tokyo Institute of Technology, Japan. 5. Poljansek, K., Perus, I., & Fajfar, P, 2009, Hysteretic energy dissipation capacity and the cyclic to monotonic drift ratio for rectangular RC columns in flexure, Earthquake Engineering and Structural Dynamics, 38, 907-928. 6. Ming-Liang Wang1, Surendra P. Shah, 1987, Reinforced Concrete Hysteresis Model Based on the Damage Concept, Earthquake Engineering & Structural Dynamics Volume 15, Issue 8, pages 9931003. 7. Bilham R, Wallace K. Future Mw>8 earthquakes in the Himalaya: implications from the 26 Dec 2004 Mw=9.0 earthquake on India's eastern plate margin. CIRES and Geological Sciences, University of Colorado Boulder 2006. (http://cires.colorado.edu/~bilham/HimalayanEarthquakes/KangraCentenaryFinal.htm) 8. Dellow GD, Ali Q, Syed AM, Hussain S, Khazai B, Nisar A., 2006 Preliminary Reconnaissance Report For The Kashmir Earthquake Of 8 October 2005. In: NZSEE Napier Conf., paper 31. 9. EERI 2006, The Kashmir Earthquake of October 8, 2005: Impacts in Pakistan Learning from Earthquakes, Earthquake Engineering Research Institute, Oakland, California, 2005. (http://www.eeri.org/lfe/pdf/kashmir_eeri_2nd_report.pdf) 10. Naeem A, Scawthorn C, Syed AM, Ali Q, Javed M, Ahmed I, et al., 2005 First Report on the Kashmir Earthquake of October 8, 2005 Learning from Earthquakes, Earthquake Engineering Research Institute, Oakland, California, 2005. (http://www.eeri.org/lfe/pdf/kashmir_eeri_1st_report.pdf) 11. Reinhorn, A. M. (2008). Lecture 2 - Modeling of Structures and Similitude retrieved Dec 23, 2008, from CIE616 - EXPERIMENTAL METHODS IN STRUCTURAL ENG.:

Energy Dissipation In Low Strength Concrete Bridge Columns

97

(http://civil.eng.buffalo.edu/CIE616/LECTURES/Lecture%202%20%20Modeling%20and%20Scaling/Slides%202%20%20Modeling%20and%20Scaling.pdf) 12. Syed A. M., Shakal, A. Response to the Pakistan Earthquake of October 8, 2005 The National Academies 2007 (http//www7.nationalacademies.org/dsc/Quake_Report_2007.pdf) 13. Syed A. M., Khan N. A., Rahman S., Reinhorn A., M. 2011, A Survey of Damages to Bridges in Pakistan after a Major Earthquake of October 8, 2005. Earthquake Spectra, 27: 947-970. 14. Dally, J. W., Riley, w. F., & McConnell, K. G, 2004, Instrumentation for Engineering Measurements, Second edition, Singapore: John Wiley & Sons. 15. ASTM C 39/C 39M 03, Standard Test Method for Compressive Strength of Cylindrical Concrete Specimens, ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States. Table 1: Identification of Columns, Target Strength and Mix Design of Concrete Used for Test Columns

Item

QSCT-2003

QSCT-3-004 2,400

Target concrete strength (psi) Proportion by weight Cement : Sand : Coarse Agg. Water/Cement ratio Average cylinder strength achieved (psi) Modulus of Rupture (psi) Modulus of Elasticity (ksi*) * ksi = kilo pounds per square inch 1:1.5:3 0.62 2,420 595 2,798

1:1.5:3 0.58 2,300 595 2,737

Table 2: Summary of Field Survey of Concrete Bridges in Pakistan S.# 1 2 3 4 5 6 7 8 9 10 11 Span Number of lanes Column height Column diameter Rebar size Rebar ratio Spiral bar size for confinement Pitch of spiral Rebar Grade Concrete cylinder strength Load on column Parameter Average Value * 85 feet 2 23.3 feet 4.27 feet 1 in 1.37% 0.39 in 6 in 60 ksi 2,400 psi 683.43 kips

* in = inch; kips = kilo pounds; psi = pounds per square inch

98

Ali M. Syed, Bashir Alam & Mohammad Javed

Table 3: Summary of Scale Factors Used for Test Columns Item Length, l Area, Scale Factor Required 4.0 16.0 Provided 4.0 16.0 64.0 4.0 1.0 1.0 1.0 1.0 1.0 1.0 16.0 16.0 64.0 16.0 1.0 64.0

A
I D E

Moment of inertia,

64.0 4.0 1.0 1.0 1.0

Linear displacement,

Angular displacement, Modulus of elasticity, Stress, Specific mass for column only-static case, Poissons Ratio, Strain, Concentrated load, Shear force, V Moment,

0.25 1.0 1.0

16.0 16.0 64.0 16.0

Mass on column top, m Gravitational acceleration, Energy, e

1.0 64.0

Table 4: Mechanical Properties of Rebar Used in Test Columns Parameter Type Diameter Number Spiral Pitch Yield strength Ultimate strength % elongation Value Rebar Deformed 0.28 in 26 52.94 ksi 70.34 ksi 20.10% 1.5 in 89.05 ksi Value Confinement Plain 5 plain wires, each having 0.039 in diameter

Energy Dissipation In Low Strength Concrete Bridge Columns

99

Table 5: Values for Cracking, Initial Yield and Yield for Column QSCT-2-003 and QSCT-3-004 QSCT-2-003 Value Item Pc Uc Pyo Uyo Py Uy Uu North / South Direction 3.70 kips / -4.45 kips 0.48% (0.36 in) / -0.50% (-0.75 in) 5.00 kips/ -6.80 kips 1.0% (0.75 in) / -0.98% (-0.74 in) 6.49 kips / -6.76 kips 1.13% (0.85 in) / -1.10% (-0.83 in) 3.05% QSCT-3-004 Value North / South Direction 3.6 kips / -5.0 kips 0.35% (0.26 in) / -0.50% (-0.37) 4.85 kips / -7.0 kips 1.0% (0.75 in) / -1.0% (-0.75 in) 6.34 kips /-6.61 kips 1.10% (0.83 in)/ -1.10% (-0.83 in) 3.73%