This document discusses assessing the collapse hazard of base isolated buildings considering pounding to moat walls using the FEMA P695 methodology. Key points: 1) Experimental shake table tests were conducted on a 1/4 scale base isolated model with different concrete wall thicknesses to examine the effect of wall stiffness on pounding behavior. 2) Numerical models were developed including a simplified moat wall model, isolator models, and 3D superstructure model. 3) Nonlinear static (pushover) and dynamic (response history) analyses were performed and collapse evaluation was conducted in accordance with FEMA P695, calculating parameters like period-based ductility and spectral shape factor.

1. Assessing the Collapse Hazard of Base Isolated
Buildings Considering Pounding to Moat Walls
Using the FEMA P695 Methodology
Armin Masroor
Advanced Technology & Research, Arup
Advisor: Professor Gilberto Mosqueda
Department of Civil, Structural and Environmental Engineering
University at Buffalo

2. Problem Description
• Seismic isolation offers a simple and
direct opportunity to control or even
eliminate damage to structures.
– Decoupling a building from its foundation.
– Resulting in elongation of fundamental
period.
• Shifting period results in decreasing
acceleration (force) but increasing
displacement demand at base level.
(Constantinou et al. 2007)

3. Problem Description
• A typical base isolated basement design requires a space in which the building is
free to move sideways without hitting the surrounding structure. This space is
commonly referred to as the "moat".
• Structural design codes such as ASCE 7-05
regulates the minimum moat wall
clearance distance.
DM 
gS M 1T M  12e 
DTM  D M 1  y 2 2 
4 2 B M  b d  Moat
Impact
• Minimum moat wall clearance distance
equals to the total maximum displacement
at the base of the structure under the
Maximum Considered Earthquake (MCE).
• Despite the cautious regulation for moat wall gap distance, pounding of base
isolated building to moat walls has been reported in previous earthquake.

4. 1994 Northridge Earthquake
• The base-isolated Fire Command and Control (FCC) building in Los Angeles
experienced strong motion during the 1994 Northridge earthquake.
• Post earthquake observations showed that the base isolated FCC building
performed well, except for impact, which increased structure shear, and drift
demands. The effectiveness of base isolation was reduced because of impact
(Nagarajaiah et al 2001).

5. 2011 Christchurch Earthquake
• The Christchurch Women’s Hospital, is the only base-isolated building in the South
Island of New Zealand.
• Pounding occurred during 2010 Darfield Earthquake and the 2011 Christchurch
Earthquake (Gavin et al. 2012).
• Rolling trolleys, items falling from shelves and walls, and sloshing of water from a
full birthing pool to a distance of 6 to 9 ft from the pool was reported from the
hospital staff during the ground motion.

6. Objectives and Scope of Research
• The goal of this study is to determine the effects of pounding on the global
response of base isolated buildings:
– Realistic experimental testing.
– Development of reliable analytical models for moat wall pounding.
– Compare response of base isolated structures with and without moat wall.
– Evaluate the efficacy of code specifications in accounting for pounding in base
isolated buildings.
Numerical
Prototype Experimental 3D Numerical Collapse
Simulation of
Buildings Study Simulation Evaluation
Impact

7. Prototype Buildings
 Code : IBC 2006, ASCE 7-05, and AISC Steel Manual
 Building Location: Los Angeles, CA
 Site Class: D (Vs=180 m/s to 360 m/s)
 Mapped spectral accelerations: Ss = 2.2 g, S1 = 0.74 g
 Lateral System R First Mode Period
Intermediate Moment Frame (IMRF) 1.67 1.4 sec
Ordinary Concentric Braced Frame (OCBF) 1.00 0.4 sec

8. Isolators Design Parameters
Isolator Properties DBE MCE
Effective Period (TD, TM) 2.77 s 3.07 s
Effective Damping (BD,BM) 24.2% 15.8%
Isolator Displacement (DD, DM) 12.7 in. 24.3 in.
Total isolator displacement (DTD, DTM) 15.3 in. 29.4 in.

9. Shake Table Test Setup
• The test specimen represents ¼ scale
single bay of an internal moment frame
in the prototype structure.
• The test setup consisted of :
– Structural frame (¼ scale 3-story IMRF)
– Gravity frame (one by one bay frame
with, pin-pin columns and braced out of
plane)
– Isolators (single friction pendulum R=30
in. and displacement capacity of 8 in.).
The effective period of the isolated
model at MCE displacement is 1.5 sec.
– Concrete blocks (designed to simulate
impact surfaces)
– Retaining walls (consist of concrete wall
with soil back fill and rigid steel wall)

10. Moat wall setup
• Different scaled concrete wall thicknesses of 2, 4, and 6 in were tested to examine
the effect of wall stiffness on the pounding behavior.
• A rigid steel wall was also used to cover a wider range of wall properties (With and
without weld reinforcement).

12. Experimental Results
50 7
40 4 in gap
30 5 6 in gap
Acceleration (g)
No impact
Velocity (in/s)
20 3
10 1
0
-1
-10
-20 -3
-30 -5
-40 -7
-50
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6
Displacement (in) Time (s)
• The sudden drop in base velocity at the instances of impact can be observed for both
the 4 and 6 in. gaps.
• This increased acceleration could consist of the effect of both rigid body motion and
also local waves in the steel plate where the accelerometers were installed.

13. Impact Force
• In a structural collision, contact between two objects consists first of a local
phase followed by a second global (vibration) response phase.
• Local behavior: The first phase of impact is indentation of two
objects at the point of the contact. The contact force generated in
this phase is generally a function of the shape and material
properties of colliding objects as well as impact velocity.
• Vibration aspect of impact: The contact force in this second phase
can be affected by external seismic forces, and dynamic properties of
the two objects including mass and stiffness.
75 75
15 cm thick concrete wall
65 Steel wall w/o weld 65
Contact Force (kips)
Contact Force (kips)
55 Steel wall with weld 55
45 45
35 35
25 25
15 15
5 5
-5 -5
0 0.1 0.2 0.3 0.4 -1 0 1 2 3
Time (s) Displacement (in)

14. Numerical Modeling
• Detailed numerical models were
developed for:
1- Moat wall model
2- Seismic isolators
3- Superstructures

15. Simplified Moat wall Model
• A continuous cantilever beam supported by an elastic foundation and
external distributed damping was assumed.
• A rotational spring was assumed at the base of the beam to capture the post-
elastic behavior due to the formation of a plastic hinge.
2   2v  v  2v
 EI  x  2   C  Kv  m  x  2  F (x , t )
x 2  x  t t
Boundary conditions:
 2v v
v  x 0  0; EI  K 
x  x 0
2
x  x 0
 2v  3v
 0; 0
x 2  x  L  x 3  x  L 

16. Simplified Moat wall Model
• Simulation of impact forces in structural analysis should consider the two
phases of impact to capture both the effects of local deformation at the
impact point and the vibration aspect of the colliding objects.
• Hertz damped model captures forces during the first phase of impact.
• The force obtained in the first phase
can be implemented in Single
degree of freedom (generalized
forced vibration equation) to find
lateral displacement of the wall and
also resisting force imposed on the
striker body.

17. Impact Force 7 7
Contact Force (kips)
Contact Force (kips)
Numerical Numerical
5 Experimental 5 Experimental
Simulation 3 3
1 1
60 -1 -1
Experimental 0 0.05 0.1 0.15 0.2 -0.1 0.4 0.9 1.4
Contact Force (kips)
Numerical Time (sec) Displacement (in)
40 9.5 9.5
Contact Force (kips)
Contact Force (kips)
Numerical Numerical
7.5 7.5
Experimental Experimental
20 5.5 5.5
3.5 3.5
0 1.5 1.5
-0.1 0 0.1 0.2 0.3 0.4
Displacement (in) -0.5 -0.5
0 0.05 0.1 0.15 0.2 -0.1 0.1 0.3 0.5 0.7 0.9 1
60 Time (sec) Displacement (in)
Experimental
Contact Force (kips)
Numerical 34 34
Contact Force (kips)
Contact Force (kips)
40 29 Numerical 29 Numerical
24 Experimental 24 Experimental
19 19
20
14 14
9 9
0 4 4
-1 -1
0 0.02 0.04 0.06 0.08 0.1 0 0.1 0.2 0.3 0.4 -0.1 0.4 0.9 1.4 1.9 2.4 2.9
Time (sec) Time (sec) Displacement (in)

18. 3D Moat Wall Model
• The proposed impact element was extended to 3D to capture the collision
between the base level of a 3D base isolated building and surrounding moat
wall.
• The proposed 3D impact element captures the nonlinearity in moat wall as well
as the soil back fill.
• A more generic moat wall model is
presented here to be used in 3D
numerical studies.

19. 3D Moat Wall Model
• Under unidirectional excitation with all points of the base level on one side in
contact with the moat wall springs.
• A corner point of the base level first touches the moat wall and pushes back
this point

20. Superstructure Modeling
Fiber Section
• All columns and moment-resisting beams were 2
x 10
4 Zero Length Section
modeled using force-based nonlinear elements with
Moment (Kips.in)
1
stress-strain relationships that were modified to
include strength and stiffness degradation. 0
• A fiber section for the column geometry was -1
generated with the Modified Ibarra Krawinkler -2
-0.2 -0.1 0 0.1 0.2
Deterioration Model (Lignos et al. 2011) model Chord Rotation (rad)
assigned to each fiber element as stress-strain
behavior.
• Multiple nonlinear beam–column elements were
strung together to physically simulate the inelastic
buckling behavior in braces (Uriz et al. 2008).
• An initial camber of 0.1% of the brace length was
applied at the brace midpoint to initiate buckling.

21. Collapse Evaluation
• Collapse capacity of base isolated structures considering pounding to a moat wall:
The Methodology (FEMA P695).
1. Develop model: detailed finite element model suitable for nonlinear time history
analysis.
2. Analyze model: nonlinear static (pushover) and nonlinear dynamic (response
history) analysis for a set of pre-defined ground motions.
3. Evaluate performance: Collapse margin ratio is adjusted by a Spectral Shape Factor
(SSF).
• It is suggested that the probability of collapse due to Maximum Considered
Earthquake (MCE) ground motions be limited to 10%.
1.0
0.9
Adjusted
Probabilty of Collapse
0.8 Fragility Fragility
0.7 Curve Curve
0.6
0.5
0.4
0.3
Accepted
0.2 probability
0.1
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Spectral Acceleration

22. Nonlinear Static Analysis
• Period-based ductility, is an important parameter to calculate adjusting parameters
in the Methodology.
ult
T 
 y ,eff
0.25 0.5
Vmax 2nd Floor Buckle
Base Shear Coefficient
Base Shear Coefficient
0.2 0.4
0.8Vmax
0.15 0.3
3rd Floor Buckle
0.1 0.2

 ult
0.05 y,eff 0.1
0 0
0 20 40 60 80 100 0 20 40 60 80 100 120
Roof Displacement(in) Roof Displacement(in)
(a) (b)
• The period-base ductility was calculated as 4.0 and 1.15 for base-isolated IMRF
and OCBF models, respectively. SSF of 1.12 and 1.47 was calculated for base
isolated OCBF and IMRF model, respectively.

23. 4 4
IDA Analysis
Intensity Scale Factor
Intensity Scale Factor
3 3
2 2
• Nonlinear time history analysis for
1 1
different moat wall gap distances.
0 0
0 1 2 3 4 0 1 2 3 4
• The Far-Field ground motion set was Maximum Interstory Drift Ratio (%)
(a) Without Moat Wall
Maximum Interstory Drift Ratio (%)
(b) Moat Wall at 35 in Gap Distance
scaled to match the median spectral 4 4
Intensity Scale Factor
Intensity Scale Factor
acceleration to various acceleration 3 3
intensities.
2 2
• IDA was conducted twice for each 1 1
ground motion. 0 0
0 1 2 3 4 0 1 2 3 4
Maximum Interstory Drift Ratio (%) Maximum Interstory Drift Ratio (%)
• Adding moat wall and decreasing 4
(c) Moat Wall at 30 in Gap Distance (d) Moat Wall at 25 in Gap Distance
the moat wall gap distance leads to Intensity Scale Factor
3
flattening the IDA curve indicating
that larger drift ratios are obtained 2
at lower intensity scale factors. 1
0
0 1 2 3 4
Maximum Interstory Drift Ratio (%)
(e) Moat Wall at 20 in Gap Distance

24. Collapse Probability
• OCBF model without moat walls shows more conservative margins in comparison
to IMRF model which is slightly less than 10%.
• Pounding to moat wall at required gap distance by ASCE7-05 results in acceptable
probability of collapse (less than 10%) for flexible and ductile IMRF model.
• The OCBF frame shows a notable increase in collapse probability because of
pounding to moat wall at 30 in. gap distance and requires increasing gap distance
to 35 in. to have acceptable collapse probability.
1.0 1.0
0.9 0.9
MCE intensity
MCE intensity
Probabilty of Collapse
Probabilty of Collapse
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 0.4 Without Wall
Accepted Accepted
0.3 Without Wall 0.3 probability 35 in Gap
probability 30 in Gap 30 in Gap
0.2 0.2
25 in Gap 25 in Gap
0.1 0.1
20 in Gap 20 in Gap
0 0
0 0.5 1 1.5 2 2.5 3 3.5 4 0 0.5 1 1.5 2 2.5 3 3.5 4
Intensity Scale Factor Intensity Scale Factor
(a) (b)

25. Conclusions
• For the first time, a series of shake table experiments were conducted on base
isolated buildings pounding against moat walls of various stiffness and set at various
distance in order to investigate the effects on superstructure response.
• An impact element considering moat wall flexibility was proposed based on impact
theory and observations during experimental simulations.
• The proposed 2D impact element was extended to a 3D simulation to capture the
collision between the base level of a 3D base isolated building to a moat wall.
• The OCBF model without moat wall shows more conservative collapse probability
(2%) in comparison to the IMRF model (8.1%). Installing moat wall at 30 in. gap distance
(DTM ,minimum required gap distance by ASCE7-05) results in increasing the probability
of collapse for OCBF frame to 12.9% but it is still slightly less than the accepted value
for the IMRF model (9.7%).
• Pounding to moat wall at required gap distance by ASCE7-05 result in acceptable
probability of collapse for flexible and ductile IMRF model. However, the stiff and brittle
OCBF frame requires a gap distance of 35 in. to have acceptable collapse probability.

26. ACKNOWLEDGEMENTS
• This project was supported by the National Science Foundation (NSF) NEES
Program under Grants No. CMMI-0724208 and CMMI-1113275.
• Prof. Keri Ryan (PI) and Prof. Stephen Mahin, (Co-PI) provided valuable
input to the project
• Earthquake Engineering Research Institute (EERI).

27. Thank you!