Structural Engineering and Earthquake Simulation Laboratory SG-1: Lateral Spreading – SG-1: Lateral Spreading – Observations & Analysis Observations & Analysis Raghudeep B. & Thevanayagam S. 20 Aug 2007: 9-9:30 AM, NEESR- Workshop PI: R. Dobry, co-PI’s: A. Elgamal, S. Thevanayagam, T. Abdoun, M. Zeghal UB-NEES Lab: A. Reinhorn, M. Pitman, J. Hanley, SEESL-Staff Tulane: Usama El Shamy Students & Staff: UB (N. Ecemis, Raghudeep B.) and RPI (J. Ubilla, M. Gonzalez, V. Bennett, C. Medina, Hassan, Inthuorn)
47
Embed
Structural Engineering and Earthquake Simulation Laboratory SG-1: Lateral Spreading – Observations & Analysis Raghudeep B. & Thevanayagam S. 20 Aug 2007:
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Structural Engineering and Earthquake Simulation Laboratory
PI: R. Dobry, co-PI’s: A. Elgamal, S. Thevanayagam, T. Abdoun, M. ZeghalUB-NEES Lab: A. Reinhorn, M. Pitman, J. Hanley, SEESL-StaffTulane: Usama El ShamyStudents & Staff: UB (N. Ecemis, Raghudeep B.) and RPI (J. Ubilla, M. Gonzalez, V. Bennett, C. Medina, Hassan, Inthuorn)
Structural Engineering and Earthquake Simulation Laboratory2
OutlineOutline
Review of Test SG-1 Lateral Spreading Observation
Comparisons of LG-0 and SG-1 Highlights – Similarities & Differences (flat versus sloping
ground) Reanalysis of Lateral Spreading
Initiation of spreading – hypothesis Newmark analysis - Sliding Some thoughts
Thoughts on lateral spreading
Structural Engineering and Earthquake Simulation Laboratory3
Review of SG-1 TestReview of SG-1 Test
Structural Engineering and Earthquake Simulation Laboratory4
Review of Test SG-1Review of Test SG-1
Inclined Box (2o) Hydraulic Fill (Dr = 45~55%) 5.58m [18 ft] Deep Saturated Sand Dense Instrumentation Design Base Motion (5s/10s/10s/10s) Uninterrupted Base Motion (5s ~0.01g/3s ~0.05g) Soil Liquefied Large lateral spreading observed
Structural Engineering and Earthquake Simulation Laboratory5
SG-1 Test ConfigurationSG-1 Test Configuration
Top View
Side View
Structural Engineering and Earthquake Simulation Laboratory6
Input Base MotionInput Base Motion
14 16 18 20 22 24 26 28-0.5
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
Time [s]
Hor
izon
tal D
ispl
acem
ent [
cm]
POL1X 5.58m - Tied to the Base Shaker
1st Stage Motion
Damped Motion
Actuator Cut-Off
Data Analyzed in this Range
2nd Stage Motion2 Hz
Structural Engineering and Earthquake Simulation Laboratory7
Base AccelerationsBase Accelerations
14 16 18 20 22-0.2
-0.1
0
0.1
0.2X-Motion of the Base
b1x
14 16 18 20 22-0.2
-0.1
0
0.1
0.2
Acc
ele
ratio
n [g
] b2x
14 16 18 20 22-0.2
-0.1
0
0.1
0.2
Time [s]
b3x
14 16 18 20 22-0.2
-0.1
0
0.1
0.2Y-Motion of the Base
b1y
14 16 18 20 22-0.2
-0.1
0
0.1
0.2b2y
14 16 18 20 22-0.2
-0.1
0
0.1
0.2
Time [s]
b3y
Base Input Motion
Structural Engineering and Earthquake Simulation Laboratory8
Structural Engineering and Earthquake Simulation Laboratory13
Shear StressesShear Stresses
2.5
5
2.5
5
2.5
5
[k
Pa
]
2.5
5
14 15 16 17 18 19 20 21 220
2.5
5
14 15 16 17 18 19 20 21 22
at 0.122m
at 0.381m
at 0.777m
at 1.295m
at 1.936m
at 2.713m
at 3.368m
at 4m
at 4.648m
at 5.166m
Time [s]
Top Rings
Bottom Rings
Structural Engineering and Earthquake Simulation Laboratory14
Factors Conducive for Lateral Spreading: Factors Conducive for Lateral Spreading: LG-0 versus SG-1LG-0 versus SG-1
SG-1
Shear Strain
0
Cyclic
Flow Failure
Strain Accumulation
Monotonic Strength Envelope
Sh
ear
Str
ess
Shear Strain
LG-0
Monotonic Strength Envelope
Small Strain Accumulation
• No Static Shear• Very little Strain Accumulation• No Flow observed• Failure termed as Liquefaction
• Non-zero Static Shear• Strain Accumulation until curve hits the strength envelope• Large Flow thereafter and curve follows the envelope• Failure termed as Flow Failure
Structural Engineering and Earthquake Simulation Laboratory15
Triaxial Test Data Triaxial Test Data (no initial shear, Theva 2003)
• e=0.779 (Moist tamping) • e = 0.778 (MT)
• e=0.804 (MT)
Is this what is seen in SG-1?Is hydraulic fill creating meta-stable structure more prone to collapse?Is collapse potential higher if static shear is present (i.e occurs at higher densities)?
Structural Engineering and Earthquake Simulation Laboratory16
Lateral spreading begins when cyclic shear stress meets monotonic failure envelope.Soil is not necessarily at liquefied state when lateral spread begins.
Ultimate
int
Monotonic Strength Envelope
Flow Begins !!!
Structural Engineering and Earthquake Simulation Laboratory17
Small DeformationsLarge Deformations, primarily initiated
by gravitational static shear
Structural Engineering and Earthquake Simulation Laboratory19
Comments on LG-0 Vs SG-1Comments on LG-0 Vs SG-1 Soil degraded faster in SG-1 compared to LG-0 Mostly Cyclic Strains in LG-0; Monotonic strains dominate in
Initial Static shear has a significant influence in initiating large strains.
Cyclic shear in SG-1 degrades the soil sufficiently to a point where the cyclic shear stress meets undrained strength envelope, lateral spreading begins?
Identify lateral spread initiation points from stress-strain curves deduced from acceleration data (next)
Then perform Newmark analysis modified to account for strength degradation
Structural Engineering and Earthquake Simulation Laboratory20
Coupled with Strength Coupled with Strength Degradation and variable yield Degradation and variable yield
accelerationacceleration
Structural Engineering and Earthquake Simulation Laboratory36
Model DescriptionModel Description
Original Laminar Box
Rings
Soil
a1(t)
a2(t)
ai(t)
an(t)
an-1(t)
Rigid Blockaavg(t)
• Weight of Rings, including the unfilled incorporated• Weight of each ring 11% of the weight of saturated soil filled in one ring• Horizontal Ground surface considered
Acceleration of the rigid block = Average of accelerometers present above the sliding surface
Yield Acceleration
Structural Engineering and Earthquake Simulation Laboratory37
Other AssumptionsOther Assumptions
Strength & Yield Acceleration varying with strain (& time) Yield Acceleration obtained from shear strength during
sliding
ayield = (fA-Wsin)/M f() = (1-ru)’v0tan
f() = shear strength during sliding = Friction angle varying with Strain (& time) A = Area of Laminar Box (12.75 m2), M = mass of the rigid
block
tatata yieldavgrel
t
rel ddatd0
1
0
22
1
Structural Engineering and Earthquake Simulation Laboratory38
Structural Engineering and Earthquake Simulation Laboratory44
Some Variations in Initiation PointSome Variations in Initiation Point(More to come later)(More to come later)
Shear Strain
Sh
ea
r S
tre
ss
Shear Strain
Sh
ea
r S
tre
ss
Higher Amplitude
Shear Strain
Sh
ea
r S
tre
ss
Steady State Strength ≠ Initial Static Shear
Shear Strain
Sh
ea
r S
tre
ss
Peak Strength ≈ Initial Static Shear
Structural Engineering and Earthquake Simulation Laboratory45
Factors Conducive for Lateral Factors Conducive for Lateral SpreadingSpreading
SG-1
Shear Strain
0
Cyclic
Flow Failure
Strain Accumulation
Monotonic Strength Envelope
Sh
ear
Str
ess
Shear Strain
LG-0
Monotonic Strength Envelope
Small Strain Accumulation
• No Static Shear• Very little Strain Accumulation• No Flow observed• Failure termed as Liquefaction
• Non-zero Static Shear• Strain Accumulation until curve hits the strength envelope• Large Flow thereafter and curve follows the envelope• Failure termed as Flow Failure
Structural Engineering and Earthquake Simulation Laboratory46
Triaxial Test Data Triaxial Test Data (no initial shear, Theva 2003)
• e=0.779 (Moist tamping) • e = 0.778 (MT)
• e=0.804 (MT)
Is this what is seen in SG-1?Is hydraulic fill creating meta-stable structure more prone to collapse?Is collapse potential higher if static shear is present (i.e occurs at higher densities)?
Structural Engineering and Earthquake Simulation Laboratory47
ConclusionsConclusions Soil does not have to fully liquefy for lateral spreading to begin. But soil must be degraded to a ‘threshold’ strength for lateral spreading to
begin. Lateral spread likely begins when the soil is sufficiently degraded and the
cyclic curve hits monotonic strength envelope Threshold spreading point depends on initial static shear, cyclic shear amplitude,
pore pressure generation and associated degradation of soil strength. Once lateral spreading begins, little or no cyclic component exist. ‘Effective’ soil
friction angle during spreading increases. Modified Newmark Analysis, coupled with strength degradation, traces the
measured lateral displacements well. Undrained strength ratio at threshold lateral spreading falls in a narrow range
of about 0.08 Does hydraulic fill method create soil structure prone to collapse?