Shake Table Testing of GRS Bridge Abutments Prof. John S. McCartney and Yewei Zheng University of California San Diego Department of Structural Engineering Presentation to: NHERI-UCSD Geotechnical Workshop May 31 st, 217
2 2 Project sponsors: Acknowledgements Caltrans Pooled fund members (WashDOT, UDOT, MDOT) Collaborators Yewei Zheng, Ph.D. Candidate Prof. Benson Shing, Chair of SE at UCSD Prof. Patrick Fox, Head of CEE at Penn State University
3 3 Presentation Overview Research Motivation Shaking Table Testing Program Material Properties Test Configuration Construction Instrumentation Input Motions Test Results and Analysis Longitudinal Tests Transverse Test Conclusions and Future Work
4 4 Geosynthetics in transportation applications: GRS Bridge Abutments Roadways Embankments Slopes Retaining walls Bridge abutments GRS retaining walls as bridge abutments with bridge loads applied directly to the reinforced soil mass Many advantages, including lower cost, easier and faster construction, and smoother approach roadway transition
5 5 GRS Bridge Abutments GRS bridge abutments have been widely used in US, but have not been adopted in California due to potential seismic issues: Geotechnical: backfill settlement and facing displacement Structural: bridge beam and seat movements, impact forces between bridge beam and seat, and interaction between bridge superstructure and GRS abutment Post-earthquake investigations for the 21 Maule earthquake, Chile (Yen et al. 211)
6 6 Shake Table Testing Program Shaking table tests have been used successfully to investigate seismic performance of GRS structures (El-Emam and Bathurst 24, 25, 27; Ling et al. 25, 212; Tatsuoka et al. 212; Helwany et al. 212) UCSD South Powell Structural Lab Shaking Table: Dimensions: 3 m x 5 m Shaking DOF: 1 D in N-S direction Maximum gravity load: 35 kn Dynamic stroke: ± 15 mm Dynamic capacity: 4 kn
7 7 Longitudinal Testing Upper wall Bridge beam Bridge seat GRS abutment Support wall Steel beams Powell lab shaking table Sliding platform
8 8 Transverse Testing Bridge beam Bridge seat GRS abutment Powell lab shaking table
9 9 Similitude Relationships Similitude for 1g shaking table tests (Iai 1989) Variable Theoretical scaling factor Scaling factor for λ = 2 Length λ 2 Material density 1 1 Strain 1 1 Mass λ 3 8 Acceleration 1 1 Velocity λ 1/2 1.414 Stress λ 2 Modulus λ 2 Stiffness λ 2 4 Force λ 3 8 Time λ 1/2 1.414 Frequency λ -1/2.77 Model geometry, reinforcement stiffness, soil modulus, bridge load, and frequencies of earthquake motions were scaled Goal: Similar response in model and prototype Stress-strain relationships for model and prototype (Rocha 1957; Roscoe 1968)
1 1 Model Design Model geometry scaling Prototype Model Wall height (m) 4.2 2.1 Bridge seat thickness (m).3.15 Clearance height (m) 4.5 2.25 Wall length (m) 4.7 2.35 Wall width (m) 4.2 2.1 Bridge width (m) 1.8.9 Prototype Model Product - Keystone Dimensions (L x W x H) Block scaling.6 m x.5 m x.3 m.3 m x.25 m x.15 m Prototype Model Product UX17 LH8 Stiffness (kn/m) Reinforcement scaling 15 38
11 11 Backfill Soil 1 26 Percent Finer (%) 8 6 4 2 C u = 6.1, C z = 1. Well-graded sand (SW) Dry Unit Weight (kn/m 3 ) 24 22 2 18 16 zero air void curve.1.1 1 1 Particle Size (mm) Gradation curve 14 2 4 6 8 1 12 14 16 Gravimetric Water Content (%) Compaction curve
12 12 Backfill Soil Properties Properties Value Specific gravity, G s 2.61 Coefficient of uniformity, C u 6.1 Coefficient of curvature, C z 1. Maximum void ratio, e max.853 Minimum void ratio, e min.371 Peak friction angle, φ ( ) 51.3 van Genuchten (198) SWRC model parameter, a vg (kpa -1 ).5 van Genuchten (198) SWRC model parameter, N vg 2.1 Drying curve volumetric water content at zero suction, θ d.32 Wetting curve volumetric water content at zero suction, θ w.2 Residual volumetric water content, θ r
13 13 Selection of Compaction Conditions Target relative density (D r ) for prototype structures = 85% (96% standard relative compaction) 6 e =.636, D r = 45%, σ 3 ' = 34 kpa e =.564, D r = 6%, σ 3 ' = 34 kpa e =.515, D r = 7%, σ 3 ' = 34 kpa e =.443, D r = 85%, σ 3 ' = 34 kpa e =.443, D r = 85%, σ 3 ' = 69 kpa 12 e =.636, D r = 45%, σ 3 ' = 34 kpa e =.564, D r = 6%, σ 3 ' = 34 kpa e =.515, D r = 7%, σ 3 ' = 34 kpa e =.443, D r = 85%, σ 3 ' = 34 kpa e =.443, D r = 85%, σ 3 ' = 69 kpa Deviatoric Stress, σ 1 '-σ 3 ' (kpa) 5 4 3 2 1 5 1 15 Axial Strain (%) Stress Ratio, σ 1 '/σ 3 ' 1 8 6 4 2 5 1 15 Axial Strain (%)
14 14 Selection of Compaction Conditions Target relative density (D r ) for prototype structures = 85% (96% standard relative compaction) Target relative density (D r ) for model specimens = 7% (92% standard relative compaction) Secant Modulus (kpa) 4 35 3 25 2 15 1 5 e =.636, D r = 45%, σ 3 ' = 34 kpa e =.564, D r = 6%, σ 3 ' = 34 kpa e =.515, D r = 7%, σ 3 ' = 34 kpa e =.443, D r = 85%, σ 3 ' = 34 kpa e =.443, D r = 85%, σ 3 ' = 69 kpa Theoretical for λ = 2, σ 3 ' = 34 kpa 1 2 3 4 5 Axial Strain (%) Dry Unit Weight (kn/m 3 ) 26 24 22 2 zero air void curve 18 Target compaction conditions: w 16 c = 5% γ d = 16.9 kn/m 3 D r = 7% 14 2 4 6 8 1 12 14 16 Gravimetric Water Content (%)
15 15 Backfill Soil φ' = 51.3 c = kpa Deviatoric Stress (kpa) 6 5 4 3 2 1 γ d = 16.9 kn/m 3 e =.515 D r = 7% σ 3 ' = 69 kpa σ 3 ' = 34 kpa σ 3 ' = 14 kpa Volumetric Strain (%) 8 6 4 2 σ 3 ' = 14 kpa σ 3 ' = 34 kpa σ 3 ' = 69 kpa -2 5 1 15 5 1 15 Axial Strain (%) Axial Strain (%) Triaxial test results for Dr = 7%
16 16 Geogrid Reinforcement Tensile Force (kn/m) Prototype: Tensar UX 17 Model: Tensar LH 8 Index stiffness = 38 kn/m Stiffness scaling factor = 4 5 4 3 2 1 Machine direction J 5% = 38 kn/m Cross-machine direction J 5% = 8 kn/m 5 1 15 2 Axial Strain (%) Tensile Force (kn/m) 6 5 4 3 2 1 5%/min 1%/min 1%/min 5%/min 1%/min Typical strain range in tests 1% 5 1 15 2 Axial Strain (%)
17 17 Shaking Table Testing Plan Test No. 1 2 3 4 5 6 Testing Purpose Bridge Stress (kpa) Reinforcement Spacing (m) Reinforcement Stiffness (kn/m) Shaking Direction Baseline Reduced Bridge Load Increased Reinforcement Spacing Reduced Reinforcement Stiffness Steel Welded Wire Mesh Baseline 66 43 66 66 66 66.15.15.3.15.15.15 38 38 38 19 48 38 Long. Long. Long. Long. Long. Trans.
18 18 Construction
19 19 Longitudinal Test Configuration
2 2 Longitudinal Model Geometry
21 21 Longitudinal Model Geometry
22 22 Transverse Test Configuration
23 23 Construction Elevation, z (m) 2.5 2.5 2. 1.5 1..5 Test Test 11 Test Test 22 Test Test 33 Test Test 44 Test Test 66 2 1 4 62 8 1 3 12 414 165 Gravimetric Apparent Water Cohesion Content (kpa) (%) Volumetric Water Content (%) Elevation, z (m) 5 2.5 4 2. 3 1.5 21. 1.5 Drying Test 11 Wetting Test 22 van Genuchten model (198) - Drying Test Test 33 van Genuchten model (198) - Wetting Test Test 44 Test 6 Test 6.1 12 1 4 1 26 8 1 12 14 Suction 13 (kpa) 14 1 5 Apparent Suction Cohesion (kpa) (kpa) SWRC is needed for to estimate the apparent cohesion, but otherwise the material properties for saturated/dry soil can be used Test No. 1 2 3 4 6 Target Average dry unit weight (kn/m 3 ) 16.6 17.1 17.1 16.7 16.6 16.9 Average relative density (%) 64 73 73 67 65 7 Average water content (%) 4.7 6.7 5.5 4.3 5. 5.
24 24 Sensors Strain gauges String potentiometers Linear potentiometers Pressure cells Load cells Accelerometers Dielectric sensors
25 25 Longitudinal Test Instrumentation Plan Longitudinal Section L2 Longitudinal Section L1 Transverse Section T1 Longitudinal Section L1 Transverse Longitudinal Section T1 Section 2
26 26 Input Motions Acceleration (g).4.3.2.1 White noise characterize system frequencies Shaking event Motion PGA (g) PGD (mm) Earthquake motions (scaled frequencies, same accelerations) 1 White Noise.1-1. 194 Imperial Valley earthquake (El Centro) PGA =.31 g/ PGD = 65 mm 2 194 Imperial Valley.31 65 2. 21 Maule earthquake (Concepcion) PGA =.4 g/ PGD = 18 mm 3 White Noise.1-3. 1994 Northridge earthquake (Newhall) PGA =.58 g/ PGD = 89 mm 4 21 Maule.4 18 Sinusoidal motions (.5, 1, 2, 5 Hz) 5 White Noise.1-6 1994 Northridge*.58 89 7 White Noise.1 - Original Scaled Displacement (mm) 8 Sin @.5 Hz 5.5 5. 9 Sin @ 1 Hz.1 25. -.1-5 -.2 -.3 1 Sin @ 2 Hz -1.2 12.5 -.4 11 Sin @ 5 Hz -15.25 2.5 5 1 15 2 25 3 35 4 5 1 15 2 25 3 Time (s) Time (s) 35 4 Imperial Valley - Acceleration 15 1 Imperial Valley - Displacement 12 White Noise.1 2.7 Original Scaled
27 27 Testing System Performance Pseudo Displacement Spectral Acceleration (mm) (g) 1 1.2 1. 5.8.6.4-5.2 Target input Shaking table Support wall Reaction wall -1.1 4 8 12 1 16 2 1 24 28 Frequency Time (s)(hz) Acceleration (g).4.2 -.2 -.4 Target input Shaking table Support wall Displacement Response time spectra history Peak =.42 g Acceleration time history Target input Shaking table 4 8 12 16 2 24 28 Time (s) The shaking table performed well in displacement-control mode for earthquake motions The steel connection beams and sliding platform successfully transmitted table motions to the base of the support wall The pseudo-spectral accelerations of the shaking table and target motion are in good agreement, which indicates that the shaking table adequately reproduced the salient characteristics of the target input motions
28 28 4 3 2 1-1 -2-3 -4 4 3 2 1-1 -2-3 -4 4 3 2 1-1 -2-3 -4 2. Elevation (m) Lateral Facing Displacement (mm) Facing Displacements (long.) z = 1.875 m 1.5 z =.975 m 1. Imperial Valley - Residual Imperial Valley - Max Maule - Residual z =.75 m Maule - Max Northridge - Residual Northridge - Max.5 4 2 8 5 4 12 1 6 16 8 15 2 1 (s) Lateral Facing Time Displacement (mm) 24 12 2 L1 T1 lateral Time histories fordisplacement the Imperial profiles Valley motion (mode scale) scale) Test Test 1 1 (model Seismic displacements at the top are larger than the bottom Longitudinal shaking results in displacements in transverse direction 28
29 29 Facing Displacements (long.) Test No. 1 2 3 4 Increased Reduced Reduced Testing Purpose Baseline Reinforcement Reinforcement Bridge Load Spacing Stiffness 2. L1 end of construction 2. L1 the Maule motion Elevation (m) 1.5 1. Test 1.5 Test 2 Test 3 Test 4 1 2 3 4 5 6 7 8 Lateral Facing Displacement (mm) Elevation (m) 1.5 1. Test 1.5 Test 2 Test 3 Test 4 1 2 3 4 5 6 7 8 Lateral Facing Displacement (mm) Reinforcement spacing and stiffness have most significant effects Greater bridge load results in larger displacements for static loading, but smaller displacements for seismic loading
3 3 Bridge Seat Settlements for Test 1 1 NE SE SW NW Settlement (mm) 8 6 4 2 Average Settlement NW NE SW SE -2 2 4 6 8 1 Time (s) Bridge seat instrumentation Bridge seat settlements for the Maule motion (model scale) Test 1 Average incremental bridge seat settlements (model scale) Test 1 Earthquake motion Max Settlement (mm) Min Settlement (mm) Residual Settlement (mm) 194 Imperial Valley 3.1 -.1 1.4 21 Maule 7. -.2 1.4 1994 Northridge 7. -.7 2.2 Vertical strain.1%
31 31 Settlement (mm) 1 8 6 4 2 Bridge Seat Settlements (long.) Test No. 1 2 3 4 Increased Reduced Reduced Testing Purpose Baseline Reinforcement Reinforcement Bridge Load Spacing Stiffness Test 1 Test 2 Test 3 Test 4 Incremental Settlement Settlement (mm) 25 2 15 1 5 Test 1 Test 2 Test 3 Test 4 Total Settlement EOC Imperial Valley Maule Northridge Testing Stage EOC Imperial Valley Maule Northridge Testing Stage Reinforcement spacing and stiffness have the most significant effects Greater bridge load results in larger settlements for static loading, but smaller settlements for seismic loading
32 32 Acceleration Amplification (long.) Elevation, z (m) 2. 1.5 1..5 Root-mean-square (RMS) acceleration Wall facing Reinforced zone Retained zone 1. 1.1 1.2 1.3 1.4 1.5 1.6 RMS Acceleration Ratio Imperial Valley motion in Test 1 Horizontal Acceleration (g).8.6.4.2 -.2 -.4 -.6 -.8 Peak =.63g Peak =.53g Bridge seat Bridge beam 4 8 12 16 2 24 28 Time (s) Amplification ratio = 1.6 for bridge seat Amplification ratio = 1.8 for bridge beam Acceleration amplification increases with elevation in the GRS bridge abutment Amplification ratios increase from retained zone to reinforced zone to wall facing Amplification ratios for bridge beam are larger than bridge seat
33 33 Reinforcement Strain (%) (%).1 Reinforcement Strains (long.) Initial Maximum bridge load Minimum Residual Transverse Longitudinal section T1 section L1.3 Initial Minimum.2 z = 1.95 m Maximum Residual layer 13.1 bridge load.3 z = 1.95 1.5 m.2 layer 13 1.1.3 z = 1.5 m.2 layer 7.1.3.6.2 z =.15 m layer 14.1.3.1.2.3.4.5.6.7.8 Distance from West Side Wall Facing,.2 z y = w (m).15 m layer 1.2.4.6.8 1. 1.2 1.4 1.6 Distance from Front Wall Facing, x (m) The location of the seismic maximum reinforcement strain was observed to be under the bridge seat in the upper reinforcement layers, but was near the facing block connections in the lower layers. Residual strains under the bridge seat for the upper layers increased significantly Shaking in the longitudinal direction also caused increases in reinforcement strain in the transverse direction
34 34 Bridge Seat and Beam Interaction Horizontal Relative Displacement Contact Force (mm) (kn) Seismic Joint Size (mm) 3 2-2 1-4 -6-1 -2-8 -1-3 4 8 12 12 16 16 2 2 24 24 28 28 Time (s) Relative Impact displacement force time time history history for the for Northridge the Northridge motion motion 6 5 4 3 2 1 Impact force 1 kn closed Seismic Joint Bridge beam moving away from bridge seat Bridge beam moving toward bridge seat East West 4 8 12 16 2 24 28 Time (s) open
35 35 Transverse Test Instrumentation Plan Transverse Section T2 Transverse Section T1 Longitudinal Section L1 Transverse Section T1 Transverse Section T2 Longitudinal Section L1
36 36 Facing Displacements (trans.) 2. 2. Northridge motion Elevation (m) 1.5 1. T1 - South North Imperial Valley - Residual Imperial Valley - Max.5 Maule - Residual Maule - Max Northridge - Residual Northridge - Max -5 5 1 5 15 1 2 25 15 3 235 25 4 Lateral Facing Displacement (mm) Elevation (m) 1.5 1..5 T1 - South T1 - North L1 - Residual L1 - Max T1-South - Residual T1-South - Max T1-North - Residual T1-North - Max 1 2 3 4 Lateral Facing Displacement (mm) T1-South had outward residual displacements, whereas T1-North had inward residual displacements for the Northridge motion Transverse shaking results in displacements in longitudinal direction
37 37 Bridge Seat Settlements (trans.) 1 8 Maule motion Longitudinal shaking Transverse shaking Settlement (mm) 6 4 2-2 2 4 6 8 1 Time (s) Average incremental bridge seat settlement time history (model scale) Average incremental residual bridge seat settlement (model scale) Shaking Direction Imperial Valley (mm) Maule (mm) Northridge (mm) Longitudinal 1.4 1.4 2.2 Transverse 2.5 4.8 4.7 Transverse shaking > Longitudinal shaking
38 38 Conclusions Incremental bridge seat settlements under seismic loading are relatively small for all tests (ranging from 1.5 mm to 7. mm), which would not be expected to cause significant damage to bridge structures Reducing reinforcement spacing and increasing reinforcement stiffness are the most effective means to reduce facing displacements and bridge seat settlements under seismic loading Greater bridge load resulted in larger deformations for static loading, but smaller deformations for seismic loading, which is attributed to the larger soil stiffness under greater bridge load Incremental bridge seat settlements due to transverse shaking are larger than for the longitudinal shaking Overall, the MSE bridge abutments show good seismic performance in terms of facing displacements and bridge seat settlements
39 39 Ongoing Work FLAC 2D/3D numerical model validation for static and dynamic conditions Detailed investigations on the seismic design of GRS bridge abutments using validated numerical models The testing program performed in this study was limited by the size and payload capacity of the shaking table in the Powell Structural Lab, so full-scale testing on the NHERI shaking table will help to alleviate these effects Impact of unsaturated soil conditions on the seismic compression of backfill soils requires further investigations