1 Seismically Isolated Structures
2 Presentation Objectives
• Present background material and basic concepts of seismic isolation
• Review seismic-code design requirements:
• Chapter 17 – ASCE Standard ASCE/SEI 7-05, Minimum Design Loads for Buildings and
Other Structures (referred to as the Standard)
• Illustrate typical application with a design example of seismically isolated structure
• Hypothetical three-story emergency operation center (EOC) located in a region of high seismicity
•
3 Background and Basic Concepts
Seismic Codes/Source Documents - Past
4 Background and Basic Concepts
Seismic Codes/Source Documents - Current
5 Background and Basic Concepts
Earthquake Response Modification
• De-couple structure above the isolation interface from potential damaging earthquake ground motions
• De-couple structure from earthquake ground motions by increasing period of the isolated structure to several times the period of the same structure on a fixed base
• Trade displacement (of the isolation system) for force (in the structure above the isolation system)
•
6 Background and Basic Concepts
Trade Displacement for Force
7 Example – Map of ASCE 7-10 Ground Motions
1-Second MCER Spectral Acceleration (Site Class D)
8 Background and Basic Concepts
Video of Earthquake Shaking
9 Background and Basic Concepts Seismic-Code Performance Objectives (Section 1.1, 2009 NEHRP Provisions)
• Intent of these Provisions is to provide reasonable assurance of seismic performance:
• Avoid serious injury and life loss
• Avoid loss of function in critical facilities
• Minimize nonstructural repair costs (where practical to do so)
• Objectives addressed by:
• Avoiding structural collapse in very rare, extreme ground shaking
• Limiting damage to structural and nonstructural systems that could lead to injury, economic loss or loss of functions for smaller more frequent ground motions
•
10 Background and Basic Concepts Seismic-Code Performance Objectives (Table C17.2-1 , 2009 NEHRP Provisions)
11 Background and Basic Concepts
Seismic-Code Performance Objectives
(Implicit for seismically isolated structures)
• Intent of these Provisions is to provide reasonable assurance of seismic performance:
• Avoid serious injury and life loss
• Avoid loss of function in critical all facilities
• Minimize structural, nonstructural and contents repair costs
• Objectives addressed by:
• Avoiding structural collapse in very rare, extreme ground shaking
• Avoiding Limiting damage to structural and nonstructural systems and contents that could lead to injury, economic loss or loss of functions for smaller more frequent ground motions by reducing earthquake demands on these systems
•
12 Background and Basic Concepts
Seismic Isolation Applications – New Buildings
• Motivating Factors
• Maintain functionality
• Protect contents
• Avoid economic loss
• Typical Applications
• Hospitals
• Emergency operations centers
• Other critical facilities (Risk Category IV)
• Research facilities (laboratories)
• Hi-tech manufacturing facilities
• Art museums
13 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
14 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
15 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
16 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
17 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
18 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
19 Background and Basic Concepts
Example Protection of Contents (and Function) New de Young Museum – San Francisco
20 Background and Basic Concepts
Isolation System Terminology
• Isolation System
“The collection of structural elements that includes all individual isolator units, all structural elements that transfer force between elements of the isolation system, and all connections to other structural elements. The isolation system also includes the wind-restraint system, energy-dissipation devices, and/or the displacement restraint system if such systems and devices are used to meet the design requirements of this chapter.”
•
21 Background and Basic Concepts
Isolation System Terminology
• Isolator Units
“A horizontally flexible and vertically stiff element of the isolation system that permits large lateral deformations under design seismic load. An isolator unit is permitted to be used either as part of, or in addition to, the weight-supporting system of the structure.”
• Isolation Interface
“The boundary between the upper portion of the structure, which is isolated, and the lower portion of the structure, which moves rigidly with the ground.”
•
22 Background and Basic Concepts
Isolation System Terminology
23 Background and Basic Concepts
Isolation Products Used in the United States
• Elastomeric (rubber) Isolators
• High-damping rubber (HDR) bearings
• Lead-rubber (LR) bearings
• Sliding Isolators
• Friction pendulum system (FPS)
• Single-concave sliding surface bearings
• Double-concave sliding surface bearings
• Triple-pendulum bearings
• Flat sliding bearings (used with rubber isolators)
• Supplementary Dampers
• Fluid-viscous dampers
24 Background and Basic Concepts
Acceptable Isolation Systems
• The Standard permits the use of any type of isolation system or product provided that the system/isolators:
• Remain stable for maximum earthquake displacements
• Provide increasing resistance with increasing displacement
• Have limited degradation under repeated cycles of earthquake load
• Have well-established and repeatable engineering properties (effective stiffness and damping)
• The Standard does not preclude, but does not fully address 3-D isolation systems that isolate the structure in the vertical, as well as the horizontal direction
25 Background and Basic Concepts
General Design Requirements – Isolation System
• The Standard (Section 17.2.4) prescribes general design requirements for the isolation system regarding:
• Environmental Conditions
• Wind Forces
• Fire Resistance
• Lateral Restoring Force
• Displacement Restraint
• Vertical-load Stability
• Overturning
• Inspection and Replacement
• Quality Control
26 Background and Basic Concepts
General Design Requirements – Structural System and Nonstructural Components
• The Standard (Section 17.2.5) prescribes general design requirements for the structural system regarding:
• Horizontal Distribution of Force
• Building Separations
• Nonbuilding Structures
• The Standard (Section 17.2.6) prescribes general design requirements for nonstructural components regarding:
• Components at or above the Isolation Interface
• Components Crossing the Isolation Interface
• Components below the Isolation Interface
27 Criteria Selection
Acceptable Methods of Analysis*
28 Background and Basic Concepts
Design Approach
• Design the structure above the isolation system for forces associated with design earthquake ground motions, reduced by only a fraction of the factor permitted for design of conventional, fixed-base buildings (RI = 3/8R ? 2.0)
• Design the isolation system and the structure below the isolation system (e.g., the foundation) for unreduced design earthquake forces (RI = 1.0)
• Design and prototype test isolator units for forces (including effects of overturning) and displacements associated with the maximum considered earthquake (MCER) ground motions
• Provide sufficient separation between the isolated structure and surrounding retaining walls and other fixed obstructions to allow unrestricted movement during MCER ground motions
•
29 Background and Basic Concepts
Design Approach
• Design the structure above the isolation system, the isolation system, and the structure below the isolation system (e.g., the foundation) for more critical of loads based on bounding values of isolation system force-deflection properties:
• Design the isolation system for displacements based on minimum effective stiffness of the isolation system
• Design the structure above for forces based on maximum effective stiffness of the isolation system
• Variations in Material Properties (Section 17.1.1):
“The analysis of seismically isolated structures, including the substructure, isolators, and superstructure, shall consider variations in seismic isolator material properties including changes due to aging, contamination, environmental exposure, loading rate, scragging and temperature.”
•
30 Background and Basic Concepts
Effective Stiffness and Damping
31 Equivalent Lateral Force Procedure
Isolation System Displacement (DD and DM)
32 Equivalent Lateral Force Procedure
Total Maximum Displacement (DTD and DTM)
33 Equivalent Lateral Force Procedure
Design Forces (Vb, Vs and Fx)
34 Equivalent Lateral Force Procedure
Response Modification Factor (RI)
35 Equivalent Lateral Force Procedure
Example Values of Design Parameters (Steel SCBF)
36 Equivalent Lateral Force Procedure
Example Values of Design Parameters (Steel OCBF)
37 Modeling and Analysis
Moments due to P-Delta Effects (and horizontal shear)
38 Modeling and Analysis
Moments due to P-Delta Effects (and horizontal shear)
39 Modeling and Analysis
Bilinear Idealization of Isolator Unit Behavior
40 Modeling and Analysis
Bilinear Idealization of Double-Concave FPS Bearing
41 Modeling and Analysis
Comparison of Modeled and Tested Hysteresis Loops
42 Modeling and Analysis
Force-Deflection Behavior of Double-Concave FPS Bearing
43 Modeling and Analysis
Effective Period of Double-Concave FPS Bearing
44 Modeling and Analysis
Effective Damping of Double-Concave FPS Bearing
45 Dynamic Lateral Response Procedures
RSA and RHA Procedures
• General – While the equivalent lateral force (ELF) procedure is useful for preliminary design, the Standard requires dynamic analysis for most isolated structures (and is
commonly used for design even when not required)
• Response Spectrum Analysis (RSA) Procedure – RSA is useful for design of the superstructure which remains essentially elastic for design earthquake ground motions
• Response History Analysis (RHA) Procedure – RHA procedure is useful for verification of maximum isolation system displacement, etc., for MCER ground motions
•
46 Dynamic Lateral Response Procedures
Minimum Design Criteria
• The Standard encourages the use of dynamic analysis but recognizes that along with the benefits of more complex model methods also comes an increased chance of error – to avoid possible under-design, the Standard establishes lower-bound limits of the results of RSA and RHA as a percentage of the ELF design parameter:
•
47 Dynamic Lateral Response Procedures
Modeling Requirements
• Configuration - Dynamic analysis models should account for:
• Spatial distribution of individual isolator units
• Effects of actual (and accidental) mass eccentricity
• Overturning forces and uplift of individual isolator units
• Variability of isolation system properties (i.e., upper-bound and lower-bound values of stiffness and damping)
• Nonlinear Properties of the Isolators – Model should incorporate nonlinear properties of isolators determined from testing of prototype units (e.g., consistent with effective stiffness and effective damping properties of the ELF procedure)
• Nonlinear Properties of the Superstructure – Model should incorporate nonlinear properties of the superstructure, if RSA is used to justify loads less than those permitted for ELF (not typical)
•
48 Dynamic Lateral Response Procedures
Response Spectrum Analysis (RSA)
• Amplitude-dependent values of isolator properties:
• Same effective stiffness and effective damping properties of isolators as those of the ELF procedure (including separate models/analyses of maximum and minimum values of effective stiffness)
• Modal Damping
• Effective damping of isolated modes limited to 30 percent of critical
• Higher modes typically assumed to have 2 to 5 percent damping
• 100%-30% Combination of Horizontal Earthquake Effects
• QE = Max (1.0QEX + 0.3QEY, 0.3QEX + 1.0QEY)
• Story Design Shear Force Limit
• Design story shear forces are limited to those of the ELF distribution (over height)
anchored to the RSA value of design base shear, Vs
49 Dynamic Lateral Response Procedures
Response History Analysis (RSA)
• Explicit modeling of nonlinear properties:
• Typical for modeling of Isolator units
• Not typical for other elements of the structure
• At least 3 earthquake records:
• Design based on the maximum response of the 3 records
• Design based on the average response if 7, or more, records
• Earthquake record selection and scaling:
• Records are selected with site properties (e.g., soil type), site-to-source distances, and source properties (i.e., fault type, magnitude, etc.) consistent with those that dominate seismic hazard at the site of interest
• Selected records are scaled to match the “target” spectrum of either design earthquake or MCER ground motions over the period range of interest (e.g., 0.5TM to 1.25TM).
50 Emergency Operations Center Design Example
Overview
• Design example illustrates the following:
• Determination of seismic design parameters
• Preliminary design using ELF procedures
• Final design (design verification using dynamic analysis)
• Specification of isolation system testing criteria
• Hypothetical emergency operations center (EOC)
• Essential Facility - Risk Category IV
• High Seismic Site – 6 km from an active fault (SDC F)
• Configuration – approx. 50,000 sf, 3-stories plus mechanical penthouse with helipad
• Structure - Steel special concentric braced frames
• Isolators – Double-concave FPS sliding bearings (35 isolators)
•
•
51 Emergency Operations Center Design Example
Structural Design Criteria – Special SCBF
• Height limit (Table 12.2-1, SDC F) h < 100 ft
• Response modification factor (R and RI):
• Fixed-base (Table 12.2-1): R = 6
• Isolated (Sec. 17.5.4.2): RI = 2 (Cd = 2)
• Importance factor, Ie (Risk Category IV):
• Fixed-base (Sec. 11.5.1/Table 1.5-2): Ie = 1.5
• Isolated (Sec. 17.2.1): Ie = 1.0
• Plan irregularity of superstructure (Table 12.3-1): None
• Vertical irregularity of superstructure (Table 12.3-2): None
• Lateral response procedure (Sec. 17.4.1, S1 > 0.6g): Dynamic Analysis
• Redundancy factor, r:
• Fixed-base (Table 12.3.4): r > 1.0
• Isolated (inferred): r = 1.0
52 Emergency Operations Center Design Example
3-D ETABS Model of the Structure
53 Emergency Operations Center Design Example
Typical Floor Framing Plan
54 Emergency Operations Center Design Example
Penthouse Roof Framing Plan
55 Emergency Operations Center Design Example
Longitudinal Bracing Elevation
56 Emergency Operations Center Design Example
Transverse Bracing Elevations
57 Emergency Operations Center Design Example
Basic Design Requirements
• Seismic Codes and Standards
• General: ASCE 7-05 (Standard)
• Seismic: 2009 NEHRP Recommended Provisions
• Other Loads (load combinations): 2006 IBC
• Materials
• Concrete: floor slabs fc’ = 3 ksi foundations fc’ = 5 ksi normal weight 150 psf
• Steel: columns Fy = 50 ksi primary girders (1st-floor) Fy = 50 ksi other girders and beams Fy = 36 ksi braces Fy = 46 ksi
• Steel Deck 3-inch deep, 20-gauge deck
58 Emergency Operations Center Design Example
Gravity Loads (by elevation)
59 Emergency Operations Center Design Example
Maximum Gravity (Dead/Live Load) Forces on Isolators
60 Emergency Operations Center Design Example
Seismic Design Parameters (USGS)
• Design Parameters at USGS website: http://geohazards.usgs.gov/designmaps/
• User enters design data:
• Code: 2009 NEHRP Provisions
• Site Classification: Site Class C or D
• Risk Category: Risk Category IV
• Site Lat. 37.800 Site Long. - 122.250
• Summary report provides:
• Echo print of design data
• Map showing site location
• MCER and design ground motions:
• SMS = 1.861 g; SDS = 1.241 g
• SM1 (D) = 1.121 g; SD1 = 0.747 g
• SM1 (C) = 0.972 g; SD1 = 0.648 g
• Plots of MCER and Design Spectra
• Supporting Data (long report)
61 Emergency Operations Center Design Example
Design and MCER Response Spectra
62 Equivalent Lateral Force Procedure
Example Values of Design Parameters (Steel SCBF)
63 Emergency Operations Center Design Example
Preliminary Design – Isolation System
• Isolation system (isolator bearing) selection criteria:
• Large maximum displacement capacity, DTM ? 30 inches to accommodate very high seismic demands
• Effective period (design level), TD ? 2.5 sec., to reduce forces on superstructure and overturning loads on bearings
• Effective damping (MCER level), bM ? 10%, to limit MCER displacement
• High-damping rubber (HDR) bearings, lead-rubber (LR) bearings and sliding (FPS)
bearings are all possible choices
• Double-concave FPS bearing (FPT8844/12-12/8-6) selected:
• Maximum displacement capacity of about 33 inches
• Effective period, TD ? 3.5 sec. at displacement, D > 16 inches
• Effective damping, bM ? 12.5% at displacement, D = 30 inches
• Load capacity: > 500 kips (long term), > 1,000 kips (short term)
•
64 Modeling and Analysis
Double-Concave FPS Bearing
65 Emergency Operations Center Design Example
Seismic Force Analysis – ETABS Model
• A linear, 3-D (ETABS) model of the EOC structure was used to expedite calculation of the following loads and load combinations:
• Gravity loads, including maximum long-term loads on isolators:
• 1.2D + 1.6L
• Superstructure design forces for combined gravity and reduced design earthquake load effects ignoring potential uplift of isolators (pushover using ELF lateral forces):
• 1.2D + 0.5L + E = (1.2 + 0.2SDS)D + 0.5L + QDE/2
• 0.9D – E = (0.9 – 0.2SDS)D - QDE/2
• Isolation system and foundation design forces for combined gravity and unreduced
design earthquake loads and permitting local uplift of individual isolator units (pushover using ELF lateral forces):
• 1.2D + 0.5L + E = (1.2 + 0.2SDS)D + 0.5L + QDE
• 0.9D – E = (0.9 – 0.2SDS)D - QDE
66 Emergency Operations Center Design Example
Seismic Force Analysis – ETABS Model
• A linear, 3-D (ETABS) model of the EOC structure was used to expedite calculation of the following loads and load combinations:
• Maximum short-term (downward) and minimum short-term (downward) forces on individual isolators for combined gravity and unreduced design earthquake loads (pushover using ELF lateral forces and permitting local uplift of individual isolators)
• 1.2D + 1.0L + E = (1.2 + 0.2SMS)D + 1.0L + QDE
• 0.9D – E = (0.9 – 0.2SMS)D - QDE
• Maximum short-term (downward) and minimum short-term (maximum uplift
displacement) forces on individual isolators for combined gravity and unreduced MCER
loads (pushover using ELF lateral forces and permitting local uplift of individual isolators)
• 1.2D + 1.0L + E = (1.2 + 0.2SMS)D + 1.0L + QMCE
• 0.9D – E = (0.9 – 0.2SMS)D - QMCE
•
67 Emergency Operations Center Design Example
Preliminary Design – ELF Displacement
• Design Displacement, DD:
68 Emergency Operations Center Design Example
Preliminary Design – ELF Displacement
• Maximum Displacement, DM:
69 Emergency Operations Center Design Example
Preliminary Design – ELF Displacement
• Total Design and Maximum Displacements, DTD and DTM, (e = 0.05d):
•
•
•
• FPS bearings mitigate the effects of mass eccentricity, but additional displacement due to actual plus accidental torsion cannot be taken as less than 1.1 times translation-only displacement which corresponds to e = 0.02d for the geometry of the EOC building:
•
70 Emergency Operations Center Design Example
Preliminary Design – ELF Effective Stiffness
• Minimum and Maximum Effective Design Stiffness:
71 Emergency Operations Center Design Example
Preliminary Design – ELF Effective Stiffness
• Minimum and Maximum Effective MCER Stiffness:
72 Emergency Operations Center Design Example
Preliminary Design – ELF Lateral Design Force
73 Emergency Operations Center Design Example
Preliminary Design - Hysteresis Loops Used for ELF Design
74 Emergency Operations Center Design Example
Preliminary Design – ELF Distribution of Lateral Design Force
75 Emergency Operations Center Design Example
Preliminary Design – ELF Distribution of Lateral Design Force
76 Emergency Operations Center Design Example
Framing on Lines 2 and 6 – Preliminary Design
77 Emergency Operations Center Design Example
Framing on Line 4 – Preliminary Design
78 Emergency Operations Center Design Example
Framing on Lines B and D – Preliminary Design
79 Emergency Operations Center Design Example
First-Floor Framing – Preliminary Design
80 Emergency Operations Center Design Example
Typical Detail of Isolation System - Preliminary Design
81 Emergency Operations Center Design Example
Typical Gravity (Dead/Live Load) Weight on Isolators
82 Emergency Operations Center Design Example
Maximum Downward Design Forces on Isolators
83 Emergency Operations Center Design Example
Minimum Downward Design Forces on Isolators
84 Emergency Operations Center Design Example
Maximum (Downward) MCER Forces on Isolators
85 Emergency Operations Center Design Example
Maximum MCER Uplift Displacement of Isolators
86 Emergency Operations Center Design Example
RHA Final Design (Design Verification)
• Dynamic analysis (RSA or RHA) is required for design of the EOC building since S1 ? 0.60g and TM > 3.0s
• RHA is not required for design of the EOC building since site conditions are not “soft” and the isolation system meets the criteria of Section 17.4.1.7, but is used in this example to:
• Verify lateral ELF forces used for preliminary design of the structure above the isolation system
• Calculate maximum displacements used for final design of the isolation system (and testing of individual isolator units)
• Verify maximum forces used for preliminary design of the isolation system and foundations
• Verify uplift displacements of individual isolator units
87 Emergency Operations Center Design Example
RHA Design Verification – Target Response Spectra
• Target design and MCER response spectra of this example use 100 percent of “Code” spectra in lieu of site-specific spectra required for design isolated structures located at sites with S1 ? 0.6 g (Section 11.4.7)
88 Emergency Operations Center Design Example
RHA Design Verification - Earthquake Record Selection
• Select earthquake ground motion records to match seismic source and site conditions of the EOC facility
• Seismic source (dominant fault) information available from site hazard de-aggregation
(https://geohazards.usgs.gov/deaggint/2008/)
• Site conditions may be assumed (e.g., Site Class D) or determined by geotechnical study
(i.e., vs,30)
• EOC site seismic hazard dominated by the Hayward fault:
• Fault type - Strike-slip
• Characteristic magnitude – M7+
• Fault Proximity – Within 6 km (near source)
• EOC site conditions:
• Site Class – Site Class C/D (vs,30 = 450 meters/sec.)
89 Emergency Operations Center Design Example
Selection of Earthquake Ground Motion Records
• Seven strike-slip records selected from the near-field (NF) and far-field (FF) record sets of
FEMA P695 with mean properties:
• Magnitude = M7.37
• Distance to source = 5.2 km (JB)
• Shear wave velocity, vs,30 = 446 mps
•
90 Emergency Operations Center Design Example
Scaling of Earthquake Ground Motion Records
• Earthquake ground motion records oriented to have a common axis of stronger shaking response (at long periods) and scaled:
• Average spectrum of SRSS combination of scaled records envelops MCER spectrum from
1.75 seconds (0.5 TD) to 4.9 seconds (1.25 TM)
• Average spectrum of the stronger components is comparable to MCER spectrum at response periods of interest (e.g., 3.9 seconds for MCER analysis)
•
91 Emergency Operations Center Design Example
Comparison of Average Spectra of Scaled Records and Target MCER Spectrum
92 Emergency Operations Center Design Example
RHA Design Verification – Modeling
• Isolated Structure Modeling Requirements:
• Linear elastic model of “essentially elastic” superstructure
• Explicit nonlinear modeling of isolator units
• Isolation System Modeling Requirements:
• Properties developed and verified by prototype test (same as ELF)
• Account for spatial distribution of isolators
• Consider translation in both horizontal direction (3-dimensional)
• Access overturning/uplift forces on individual isolator units
• Account for the effects of vertical load, etc., on isolators
• ETABS Model
• Same model as that used for pushover (with ELF lateral forces)
• Isolators modeled as bi-linear elements (representing upper-bound and lower-bound properties of bearing stiffness)
93 Emergency Operations Center Design Example
Comparison of Modeled and Tested Hysteresis Loops - RHA
94 Emergency Operations Center Design Example
Design Verification - RHA
95 Emergency Operations Center Design Example
Design Verification - RHA
96 Emergency Operations Center Design Example
Design Verification - RHA
97 Emergency Operations Center Design Example
Comparison of ELF and RHA Methods – Individual Records
98 Emergency Operations Center Design Example
Comparison of ELF and RHA Methods – Individual Records
99 Emergency Operations Center Design Example
Prototype Testing – Number and Type of Test Specimens
• Two of Each Isolator Type and Size. Prototype tests shall be performed separately on two full-sized specimens (or sets of specimens, as appropriate) of each predominant type and
size of isolator unit of the isolation system
• Wind Restraint System. Test specimens shall include the wind-restraint system as well as individual isolator units is such systems are used in the design
• Prototype Test Specimens Not Permitted for Construction. Test specimens shall not be used for construction unless accepted by the registered design professional
• (Make) Use of Prior Prototype Testing. Prototype testing may be based on prior prototype testing of the same type and size of isolator unit for comparable test loads
100
101
102
103
104
105
106
107
•
•
Emergency Operations Center Design Example
Prototype Testing – Sequence and Cycles
Emergency Operations Center Design Example
Prototype Testing – Effective Properties of Isolator Units
• Effective stiffness, keff, and effective damping, beff, of each prototype isolator unit is calculated for each cycle of test loading:
Emergency Operations Center Design Example
Prototype Testing – Maximum and Minimum Effective Properties of the Isolation
System at the Design Displacement
Modeling and Analysis
Comparison of Modeled and Tested Hysteresis Loops
Emergency Operations Center Design Example
Prototype Testing – Acceptance Criteria of Test Specimens
• Cyclic-load tests to establish effective stiffness and damping:
• Force-deflection plots have positive incremental restoring force capacity
• For each increment of test displacement and vertical load:
• For each test specimen, the effective stiffness at each of the 3 cycles of test loading is within 15 percent of the average stiffness over the 3 cycles of test load
• For each of two test specimens (of common type and size), the effective stiffness of one specimen is within 15 percent of the effective stiffness of the other (at each of the 3 cycles of test loading, and on average)
• Cyclic-load tests to check durability – for each test specimen:
• There is no more than 20 percent change in effective stiffness
• There is no more than a 20 percent reduction in effective damping
• Static-load tests to verify isolator unit stability
• All test specimens remain stable (for maximum MCER loads)
Emergency Operations Center Design Example
Prototype Testing of Double-Concave FPS Bearing (FPT8844/12-12/8-6)
Emergency Operations Center Design Example
Post-Test Inspection of Double-Concave FPS Bearing (FPT8844/12-12/8-6) Questions
This
set
of
instructional
slides
presents
the
design
example
and
background
material
for
“seismically
isolated
structures,”
Chapter
12,
of
FEMA
P-751,
2009
NEHRP
Recommended
Seismic
Provisions:
Design
Examples.
Seismically
Isolated
Structures -1
The
three
primary
objectives
of
this
instructional
presentation
of
Chapter
12
material
are:
(1)
to
provide
background
on
seismic
isolation,
basic
concepts
and
analysis
methods,
(2)
to
review
the
seismic-code
design
requirements
for
seismically
isolated
structures
(Chapter
17
of
ASCE
7-05,
referred
to
as
the
Standard),
and
(3)
to
illustrate
a
typical
application
seismic-code
requirements
with
an
example
design
of
a
hypothetical
seismically-isolated
3-story
emergency
operation
center
located
in
a
region
of
high
seismicity.
In
the
example,
seismic
loads
are
based
on
the
new
“risk-targeted”
seismic
design
values
of
the
2009
NEHRP
Provisions.
Note.
The
most
current
version
of
ASCE
7
is
ASCE
7-10
which
adopted
the
new
“risk-targeted”
ground
motions
of
2009
NEHRP
Provisions
with
only
slight
editorial
changes,
and
did
not
incorporate
substantive
changes
to
Chapter
17
of
ASCE
7-05.
Thus,
the
material
presented
in
these
slides
applies
equally
well
to
the
design
requirements
of
Chapter
17
of
ASCE
710
for
seismically
isolated
structures.
Seismically
Isolated
Structures -2
Initial
development
of
design
requirements
for
base-isolated
buildings
began
with
ad
hoc
groups
of
the
Structural
Engineers
Association
of
California
(SEAOC).
These
requirements
were
used
by
the
California
Office
of
Statewide
Planning
and
Development
(OSHPD)
for
regulation
of
first
base-isolated
hospital
in
California
(University
of
Southern
California
Teaching
Hospital)
and
subsequently
adopted
by
the
1990
SEAOC
Blue
Book
and
as
a
non-mandatory
appendix
of
the
1991
Uniform
Building
Code
(1991
UBC).
At
that
time,
the
SEAOC
Blue
Book
served
as
the
role
model
for
the
UBC,
the
model
building
code
with
the
most
up-to-date
and
widely
respected
seismic
design
requirements.
In
the
1990’s,
the
design
requirements
for
seismically-isolated
structures
were
adopted
as
a
mandatory
section
of
the
UBC
and
as
a
new
chapter
of
the
NEHRP
Provisions.
(Click).
Around
the
year
2000,
the
three
major
model
building
codes
(SBC,
BOCA
and
UBC)
merged
to
form
the
new
International
Building
Code
(IBC)
with
seismic
design
requirements
taken
from
the
NEHRP
Provisions
(which
were
also
adopted
and
incorporated
into
the
American
Society
of
Civil
Engineers
Standard,
ASCE
7.
Seismically
Isolated
Structures -3
Today,
model
building
codes
(e.g.,
national
model
codes
such
as
the
IBC,
or
regional
derivatives,
such
as
the
California
Building
Code,
CBC)
adopt
by
reference
the
seismic
design
requirements
of
ASCE
7
which
are
based
on
the
NEHRP
Provisions.
As
such,
the
design
requirements
for
seismically
isolated
structures
described
in
Chapter
12
of
FEMA
P-751
are
taken
from
ASCE
7
(referred
to
simply
as
the
Standard).
The
most
current
version
of
the
Standard,
ASCE
7-10,
has
been
adopted
by
the
2012
IBC
(by
reference).
Most
jurisdictions
and
regulatory
authorities
in
the
United
States
have
(or
will)
adopt
the
2012
IBC
(or
regional
derivatives
thereof).
Seismically
Isolated
Structures -4
The
basic
concept
of
seismic
isolation
(also
referred
to
as
base
isolation)
is
to
“de-couple”
the
“superstructure”
(structure
above
the
isolation
interface)
from
potentially
damaging
ground
motions
by
adding
“seismic
isolators”
which
support
the
structure
above
while
permitting
large
relative
horizontal
displacement
of
the
structure
in
an
earthquake.
Note.
Earthquake
ground
motions
shake
buildings
in
the
vertical
as
well
as
the
horizontal
direction,
but
typically
cause
the
most
damage
due
to
building
response
in
the
horizontal
direction.
The
seismic
isolators
de-couple
response
by
essentially
making
the
fundamental-mode
period
mode
of
the
isolated
structure
several
time
longer
than
the
period
of
the
structure
above
the
isolation
system
–that
is,
several
times
longer
than
the
period
of
the
same
structure
on
a
fixed
base.
In
this
manner,
displacement
of
the
isolation
system
is
“traded”
for
the
force
in
the
structure
above
the
isolation
system.
Seismically
Isolated
Structures -5
This
figure
shows
5%-damped
and
20%-damped
acceleration-
displacement
response
spectra
(ADRS)
typical
of
a
high
seismic
region.
The
ADRS
is
a
plot
of
response
spectral
acceleration
on
the
vertical
axis
and
response
spectral
displacement
on
the
horizontal
axis.
Spokes
from
the
origin
show
lines
of
constant
period
ranging
from
0.5
seconds
to
6
seconds.
(Click).
For
shorter,
stiffer
conventional
(fixed-base)
structures
with
fundamental-mode
elastic
periods
of
less
than
about
1.0
second,
5%damped
spectral
acceleration
ranges
from
about
0.9g
to
1.8g.
(Click).
By
adding
seismic
isolators,
the
fundamental-mode
period
is
typically
increased
to
about
2
to
4
seconds
and
damping
is
increased
to
at
least
10%,
reducing
spectral
response
to
about
0.2
to
0.4
g
–
corresponding
to
about
a
factor
of
4
reduction
in
peak
lateral
force.
However,
to
provide
this
reduction
in
lateral
force,
the
seismic
isolators
must
be
able
to
accommodate
about
15
to
30
inches
of
peak
lateral
displacement.
Even
larger
displacement
capacity
would
be
required
for
isolation
systems
with
longer
fundamental-mode
periods.
Seismically
Isolated
Structures -6
Ground
motion
intensity
varies
greatly
across
the
United
States
as
shown
by
this
map
of
1-second
maximum
considered
earthquake
(MCER)
spectral
acceleration
for
assumed
Site
Class
D
site/soil
conditions.
Typically,
seismic
isolation
has
been
used
in
regions
of
high
seismicity
such
as
the
coastal
areas
of
California,
Wasatch
fault
zone
(e.g.,
Salt
Lake
City,
Utah),
the
New
Madrid
seismic
zone
(e.g.,
Memphis,
Tennessee)
and
the
Charleston,
South
Carolina,
seismic
zone.
Regions
of
high
seismicity
provide
the
greatest
opportunity
for
realizing
the
benefits
of
isolation,
but
also
the
greatest
challenges
to
the
design
of
the
isolation
system
to
accommodate
large
earthquake
displacements.
Seismically
Isolated
Structures -7
This
slide
begins
with
a
photo
of
a
commercial
building
severely
damaged
by
the
1995
M6.8
Kobe
earthquake.
The
building
which
housed
television
equipment
and
related
media
operations
for
Nippon
Hoso
Kyokai
(NHK)
prior
to
the
earthquake
was
subsequently
demolished.
A
Japanese
video
will
show
a
sequence
of
three
clips
(Click).
The
first
clip
is
from
a
surveillance
video
camera
inside
the
NHK
during
the
Kobe
earthquake.
The
next
two
clips
are
from
shake-table
tests
of
simulated
Kobe
earthquake
response
of
a
typical
office
and
contents
–
first
clip
shows
office
contents
response
of
conventional,
fixed-base
building,
the
second
clip
shows
office
contents
response
of
an
isolated
building.
Seismically
Isolated
Structures -8
Section
1.1
of
the
Provisions
describes
seismic-code
performance
objectives
applicable
in
concept
to
all
building
types,
although
intended
primarily
for
conventional,
fixed-base,
buildings.
Life-safety
(avoiding
serious
injury
and
life
loss)
is
explicitly
addressed
by
design
for
MCER
ground
motions
that
are
intended
to
avoid
collapse
in
vey
rare,
extreme
ground
shaking
(i.e.,
less
than
10
percent
probability
of
collapse
for
MCER
ground
motions).
It
should
be
noted
that
the
NHK
building
met
these
criteria
in
the
1995
Kobe
earthquake
(i.e.,
the
building
although
severely
damage
did
not
collapse).
Functional
and
economic
performance
objectives
are
not
explicitly
addressed
by
seismic-code
design
requirements.
Rather,
it
is
hoped
that
additional
design
strength
(i.e.,
Ie
=
1.5)
will
adequate
protect
the
structure
from
damage
that
could
close
a
critical
facility
(i.e.,
hospital)
and
that
somehow,
nonstructural
systems
(and
contents)
can
survive
the
shaking
without
loss
of
function
or
significant
economic
loss.
Note.
It
is
difficult,
if
not
impossible,
to
design
nonstructural
systems
(and
anchor
contents)
of
a
conventional
fixed-base
building
such
that
significant
economic
and
functional
losses
would
not
occur
for
the
violent
shaking
shown
in
the
video
of
NHK
building
response
during
the
1995
Kobe
earthquake.
Seismically
Isolated
Structures -9
Seismically
isolated
structures
are
expected
to
perform
much
better
than
conventional,
fixed-base,
structures
during
moderate
and
major
earthquake
ground
motions
as
shown
in
commentary
Table
C17.2-1
of
the
Provisions
which
compares
expected
performance
for
fixed-base
structures
(designated
with
an
“F”)
and
isolated
structures
(designated
with
an
“I”)
Seismically
Isolated
Structures -10
Hypothetically,
if
Section
1.1
of
the
Provisions
was
revised
to
specifically
address
seismically
isolated
structure
performance,
then
it
might
read
as
shown
in
this
slide.
The
life-safety
performance
would
be
the
same
for
fixed-base
and
isolated
structures
–that
is,
avoid
structural
collapse
for
very
rare,
extreme
(MCER)
ground
motions.
However,
avoiding
loss
of
function
would
not
be
limited
to
critical
facilities,
but
apply
to
all
isolated
structures,
since
the
same
conservative
criteria
are
required
for
design
of
the
structure
above
the
isolation
system
regardless
of
Risk
Category
(i.e.,
RI
= 2.0 and Ie
=
1.0).
Similarly,
by
reducing
earthquake
shaking,
isolation
would
provide
a
practical
basis
for
avoiding
damage
to
all
structural
and
nonstructural
systems
and
contents
above
the
isolation
interface.
Seismically
Isolated
Structures -11
The
Standard
applies
to
new
construction.
Use
of
isolation
for
seismic
retrofit
has
additional
motivating
factors
including
protection
of
historical
architecture,
minimizing
construction
cost
and
impact
on
facility
operation.
Typical
applications
of
isolation
to
new
structures
include
primarily
essential
(Risk
Category
IV)
facilities
and
other
facilities
whose
operation
immediately
after
an
earthquake
is
considered
to
be
of
particular
importance
to
the
owner
(hi-tech
manufacturing),
or
which
house
valuable
contents
susceptible
to
earthquake
damage
(art
museums).
Seismically
Isolated
Structures -12
An
example
application
of
isolation
to
protect
contents
(as
well
as
function)
is
the
“new”
de
Young
Museum
in
the
Golden
Gate
Park
of
San
Francisco,
California.
The
new
de
Young
Museum
replaced
an
existing
structure
that
was
programmatically
inadequate
and
seismically
deficient,
having
suffered
significant
structural
damage
during
the
1989
Loma
Prieta
earthquake.
The
museum
is
located
less
than
8
km
from
the
San
Andreas
Fault,
and
the
need
to
protect
the
eclectic
collections
from
earthquake
damage
prompted
the
owner
to
opt
for
base
isolation
of
the
low-rise
building
housing
the
galleries.
The
museum
building
is
seismically
isolated
with
a
combination
of
76
high-damping
elastomeric
(rubber)
bearings,
76
flat
sliding
bearings
(sliders)
and
24
fluid
viscous
dampers.
Bearings
and
dampers
are
located
in
the
crawl
space
below
the
first
floor
and
interior
courtyards,
and
do
not
affect
museum
architecture
or
function.
Unless
informed,
visitors
to
the
museum
are
not
aware
that
the
building
is
base
isolated.
The
isolation
system
selected
for
the
new
de
Young
museum
was
one
of
20
different
systems
considered
for
the
museum
and
found
by
engineering
evaluation
to
be
the
alternative
that
had
the
lowest
base
shear
(best
system
for
superstructure
design),
the
lowest
floor
acceleration
(best
system
for
collection
protection)
and
the
lowest
cost
of
the
alternatives
Seismically
Isolated
Structures -13
considered.
Seismically
Isolated
Structures -13
With
isolation,
the
curators
of
the
new
de
Young
Museum
can
brace
or
anchor
artifacts
in
a
conventional
manner
and
have
more
freedom
with
temporary
exhibitions.
In
most
cases,
bracing
can
be
avoided
which
would
be
problematic
for
many
exhibits
such
as
the
glass
sculpture
shown
in
this
photo.
Base
isolation
also
made
design
of
the
highly
irregular
superstructure
easier,
complying
with
the
owner’s
directive
to
have
as
many
open
spaces
as
possible.
Seismically
Isolated
Structures -14
This
photo
taken
during
construction
of
the
new
de
Young
Museum
shows
grade
beams
and
foundations
at
individual
isolator
locations
Seismically
Isolated
Structures -15
This
photo
taken
during
construction
of
the
new
de
Young
Museum
shows
isolators
(e.g.,
sliding
bearing
at
an
interior
location
and
a
rubber
bearing
at
a
perimeter
location)
and
1st-floor
steel
framing
Seismically
Isolated
Structures -16
This
photo
taken
during
construction
of
the
new
de
Young
Museum
shows
steel
concentric-braces
and
upper-floor
framing
Seismically
Isolated
Structures -17
This
photo
taken
in
crawl
space
of
the
new
de
Young
Museum
before
addition
of
fireproofing
shows
rubber
bearings
on
reinforced-concrete
pedestals.
A
total
of
76
high-damping
rubber
bearings
provide
restoring
force
and
tend
to
be
located
near
the
perimeter
of
the
building
to
resist
torsion
(i.e.,
rotation
of
the
building
during
an
earthquake).
Seismically
Isolated
Structures -18
This
photo
taken
in
crawl
space
of
the
new
de
Young
Museum
after
addition
of
fireproofing
shows
a
typical
stub
column
on
top
of
a
sliding
bearing
and
a
fluid
viscous
damper
connected
at
one
end
to
a
1st-floor
girder
above
and
at
the
other
end
to
a
reinforced-concrete
pedestal
and
foundation
below.
A
total
of
76
flat
sliders
provide
support,
add
damping
(due
to
friction)
without
adding
stiffness
to
the
isolation
system.
Due
to
the
relatively
close
proximity
to
the
fault
and
the
potential
for
large
ground
motion
“pulses,”
the
isolation
system
incorporates
a
total
of
24
fluid
viscous
dampers,
providing
additional
displacement
control.
The
covered
“moat”
around
the
perimeter
of
the
building
accommodates
36
inches
of
isolated
structure
displacement
in
any
direction.
This
clearance
includes
substantial
cushion
on
the
calculated
maximum
earthquake
displacement
of
26
inches.
Seismically
Isolated
Structures -19
The
next
three
slides
describe
Standard
terminology
for
the
isolation
system
and
elements
thereof.
The
isolation
system
includes
the
individual
isolator
units
and
other
structural
elements
(e.g.,
connections)
that
transfer
force
from
the
isolator
units
to
the
structure
and
foundation
below
and
to
structure
above
the
isolation
system.
The
isolation
system
also
includes
energy-
dissipation
devices
(e.g.,
viscous
dampers),
wind-restraint
system
and
displacement
restraint
system,
if
such
systems
and
devices
are
used
to
meet
Standard
requirements.
In
most
applications,
wind
restraint
is
an
inherent
feature
of
the
isolator
unit
–that
is,
the
initial
stiffness
of
rubber
bearings
(or
the
static
friction
level
of
sliding
bearings)
is
typically
large
enough
to
resist
wind
design
loads
without
significant
displacement.
Although
uncommon,
a
displacement
restraint
system
(e.g.,
moat
bumpers,
etc.)
could
be
used
to
limit
maximum
considered
earthquake
displacement
of
the
isolation
system,
provided
it
is
shown
that
such
restraint
would
not
adversely
affect
the
stability
of
structure
above
the
isolation
system.
Seismically
Isolated
Structures -20
The
Standard
defines
isolator
units
as
horizontally
flexible
and
vertically
stiff
elements,
assuming
that
isolation
system
provides
only
horizontal
isolation.
While
earthquake
damage
to
buildings
and
their
contents
is
due
primarily
to
horizontal
ground
motions,
vertical
ground
motions
can
adversely
affect
certain
vibration-sensitive
equipment
and
systems.
Protection
against
damage
due
to
vertical
(as
well
as
horizontal)
ground
motions
would
require
a
3-dimensional
isolation
system,
which
is
not
practical
for
most
applications
(exceptions
include
special
military
facilities
and
possibly
nuclear
power
plants).
The
Standard
defines
the
isolation
interface
as
an
imaginary
boundary
between
the
upper
portion
of
the
structure
which
is
isolated
and
the
lower
portion
of
the
structure
which
moves
rigidly
with
the
ground.
Seismically
Isolated
Structures -21
This
figure
illustrates
the
isolation
interface,
the
imaginary
boundary
between
the
isolated
and
non-isolated
portions
of
the
structure.
The
isolation
interface
is
often
referred
to
as
the
“plane
of
isolation,”
although
the
isolation
interface
need
not
be
located
at
single
horizontal
plane.
This
figure
also
illustrates
the
boundaries
between
(1)
structure
above
the
isolation
system,
(2)
the
isolation
system,
and
(3)
the
structure
below
the
isolation
system.
Typically,
there
is
a
heavy
girder
or
slab
just
above
isolator
units
to
resist
large
P-Delta
moments
that
occur
at
peak
earthquake
displacements
of
isolator
units
supporting
vertical
loads.
In
this
capacity,
the
heavy
girder
or
slab
is
considered
an
element
of
the
isolation
system,
since
it
is
required
for
stability
of
isolators.
Seismically
Isolated
Structures -22
The
first
seismic
isolation
systems
in
buildings
in
the
United
States
were
composed
of
either
high-damping
rubber
(HDR)
or
lead-rubber
(LR)
elastomeric
bearings
(low-damping
rubber
bearing
with
a
lead
core
that
adds
damping).
Other
types
of
isolation
systems
in
the
United
States
include
sliding
systems
,such
as
the
friction
pendulum
system,
or
some
combination
of
elastomeric
and
sliding
isolators.
The
FPS
may
be
composed
of
single-concave
sliding
surface
bearings
(original
“dish”
concept),
double-concave
sliding
surface
bearings
(each
with
two
dishes,
one
facing
up
and
one
facing
down),
or
triple-pendulum
bearings
(a
more
sophisticated
version
of
the
double-concave
bearings).
In
each
case,
gravity
is
used
as
the
restoring
force
of
the
FPS.
Some
applications
at
sites
of
very
high
seismicity
(such
as
that
of
the
de
Young
Museum
in
San
Francisco)
use
supplementary
fluid-viscous
dampers
in
parallel
with
either
sliding
or
elastomeric
bearings.
Seismically
Isolated
Structures -23
The
Standard
permits
a
broad
range
of
isolator
products
that
meet
certain
basic
requirements
for
stability,
strength,
degradation
and
reliability.
The
Standard
recognizes
that
the
engineering
properties
of
an
isolation
system,
such
as
effective
stiffness
and
damping,
can
change
during
repeated
cycles
of
earthquake
response
(or
otherwise
have
a
range
of
values).
Such
changes
or
variability
of
design
parameters
are
acceptable
provided
that
the
design
is
based
on
analyses
that
conservatively
bound
the
range
of
possible
values
of
design
parameters.
Isolation
systems
typically
provide
only
horizontal
isolation
and
are
rigid
or
semi-rigid
in
the
vertical
direction.
While
the
basic
concepts
of
the
Standard
can
be
extended
to
full
(3-dieminsional)
isolation
systems,
the
requirements
are
only
intended
for
design
of
horizontal
isolation
systems.
The
design
of
a
full
isolation
system
would
require
special
analyses
that
explicitly
include
vertical
ground
motions
and
the
potential
for
rocking
response
of
the
structure
above
the
isolation
interface.
Seismically
Isolated
Structures -24
The
Standard
prescribes
general
design
requirements
for
design
of
the
isolation
system,
including
environmental
conditions
(e.g.,
aging
effects,
creep,
fatigue,
operating
temperature
and
exposure
to
harmful
substances)
The
isolation
system
is
required
to
have
a
wind-restraint
system,
unless
shown
(by
testing
of
isolator
units)
to
not
displace
more
than
the
amount
permitted
for
fixed-base
structure
for
design
wind
loads.
The
isolation
system
is
required
to
have
the
same
fire
resistance
as
that
of
comparable
structural
elements
of
a
fixed-base
structure
(i.e.,
columns
in
the
basement
of
a
fixed-base
building).
The
isolation
system
is
required
to
have
a
minimum
amount
of
restoring
force
at
large
displacements
(i.e.,
positive
post-yield
slope)
to
ensure
that
isolation
system
does
not
accumulate
residual
displacement
in
a
given
direction
during
repeated
cycles
of
earthquake
response.
The
isolation
system
is
required
to
not
restrain
displacement
up
to
the
maximum
considered
earthquake
displacement
unless
the
isolated
structure
is
explicitly
designed
for
the
effects
thereof.
At
maximum
considered
earthquake
displacement,
isolators
must
be
stable
for
“worst-case”
vertical
loads
and
the
isolated
structure
must
be
safe
against
global
overturning
(although
individual
isolators
are
permitted
to
uplift
during
earthquake
response,
if
such
does
not
affect
their
stability).
Access
must
be
provided
for
inspection
and
replacement
of
isolators,
Seismically
Isolated
Structures -25
including
pre-occupancy
inspection
of
structural
separation
areas
(moat
clearance)
and
components
that
cross
the
isolation
interface.
A
quality
control
testing
program
is
required
for
isolator
units
(i.e.,
in
addition
to
testing
of
isolator
unit
prototypes).
Seismically
Isolated
Structures -25
The
Standard
prescribes
general
design
requirements
for
design
of
the
structural
system
and
for
nonstructural
components.
The
isolated
structure
be
must
separated
from
surrounding
retaining
walls,
etc.
(moat
clearance)
by
at
least
the
maximum
considered
earthquake
displacement
Isolated
non-building
structures
(e.g.,
seismically
isolated
tank,
etc.)
must
be
designed
and
constructed
in
accordance
with
Chapter
15
of
the
Standard
using
displacements
and
forces
of
Sections
17.5
or
17.6.
Nonstructural
components
above
the
isolation
interface
(isolated
components)
must
be
anchored/braced
for
force
corresponding
to
maximum
dynamic
response
of
the
isolated
structure,
or
by
exception
may
be
anchored/braced
for
fixed-base
building
design
requirements
(which
would
be
conservative,
but
would
also
avoid
calculation
of
peak
dynamic
response
of
the
isolated
structure)
Nonstructural
components
that
cross
isolation
interface
(e.g.,
water
and
fire
piping,
electrical
conduit,
HVAC
ductwork,
etc.)
must
be
designed
to
accommodate
maximum
earthquake
displacement
Nonstructural
components
below
the
isolation
interface
must
be
anchored/braced
for
fixed-base
building
design
requirements.
Seismically
Isolated
Structures -26
The
equivalent
lateral
force
(ELF)
procedure
is
intended
primarily
to
prescribe
minimum
design
criteria
and
may
be
used
for
design
of
a
very
limited
class
of
isolated
structures
(without
confirmatory
dynamic
analyses).
The
simple
equations
of
the
ELF
procedure
are
useful
tools
for
preliminary
design
and
provide
a
means
of
expeditious
review
and
checking
of
more
complex
calculations.
Modal
(Response
Spectrum)
analysis
is
permitted
if
the
site
is
relative
stiff
(not
Site
Class
E
of
F)
and
the
superstructure
is
essentially
elastic
(does
not
require
explicit
modeling
of
nonlinear
elements)
and
the
isolation
system
is
not
“highly
nonlinear.”
The
last
criterion
is
often
assumed
to
not
be
met
(even
when
it
is)
and
most
isolated
building
designs
are
validated
using
seismic
response
history
(time
history)
analysis.
Seismically
Isolated
Structures -27
The
design
approach
of
the
Standard
for
isolated
structures
is
to
(1)
protect
the
isolated
structure
from
significant
earthquake
damage
for
design
earthquake
ground
motions,
and
(2)
to
protect
the
isolation
system
form
failure
for
maximum
considered
earthquake
ground
motions
(e.g.,
collapse
performance
comparable
to
that
of
fixed-base
structures).
Seismically
Isolated
Structures -28
The
design
approach
of
the
Standard
for
isolated
structures
is
to
explicitly
incorporate
uncertainty
in
the
properties
of
the
isolation
system
due
to
(1)
variation
in
effective
stiffness
and
damping
properties
determined
by
prototype
testing,
(2)
variation
in
material
properties
of
isolators
due
to
aging,
etc.,
and
(3)
other
potential
sources
of
uncertainty
such
as
those
due
to
manufacturing
(e.g.,
isolator
fabrication
tolerances).
Explicit
incorporation
of
uncertainty
in
the
design
properties
of
the
isolation
system
is
fundamentally
different
and
more
conservative
than
the
design
approach
used
for
other
(fixed-base)
structures.
Seismically
Isolated
Structures -29
This
figure
illustrates
the
calculation
of
the
effective
stiffness
and
the
effective
damping
for
an
isolator
unit
with
either
purely
hysteretic
or
purely
viscous
damping
behavior.
In
the
case
of
viscous
damping,
the
area
of
the
hysteresis
loop
(Eloop)
corresponds
to
dynamic
cyclic
response
at
the
period
of
the
isolated
structure
(i.e.,
at
velocities
representing
earthquake
response).
In
general
(and
for
all
hysteretic
systems),
effective
properties
are
amplitude-dependent.
This
slide
concludes
the
first
part
of
presentation
that
has
addressed
background
and
basic
concepts
of
seismic
isolation
and
Code
design
requirements.
The
next
part
of
the
presentation
will
focus
on
equivalent
lateral
force
(ELF)
design
methods,
modeling
and
analysis
of
the
isolation
system,
and
dynamic
lateral
response
analysis
procedures.
Seismically
Isolated
Structures -30
This
figure
illustrates
definitions
and
formulas
of
the
amplitude-
dependent
values
of
effective
period
(TD)
and
effective
damping
(.D)
used
to
calculate
the
design
displacement
(DD).
The
maximum
displacement
(DM)
is
calculated
in
the
same
manner
(only
for
MCER
ground
motions
which
are
50
percent
stronger)
using
amplitude-dependent
values
of
effective
period
(TM)
and
effective
damping
(.M).
Due
to
the
inherent
nonlinear
nature
of
isolation
system
stiffness,
the
effective
period
at
maximum
displacement
tends
to
be
a
somewhat
larger
than
that
at
the
design
displacement,
and
effective
damping
at
maximum
displacement
tends
to
be
somewhat
less
than
that
at
the
design
displacement
for
the
same
system.
The
value
of
1-second
MCER
5%-damped
spectral
acceleration
(SM1)
is
given
in
Section
11.4.3
of
the
Standard
and
is
the
product
of
the
site
factor
(Fv)
and
the
1-second
MCER
spectral
acceleration
(S1)
provided
by
USGS
maps
of
ground
motion
values.
The
value
of
1-second
design
5%damped
spectral
acceleration
(SD1),
defined
as
2/3
of
SM1
in
Section
11.4.4
of
the
Standard,
is
the
same
1-second
spectral
acceleration
as
that
used
for
design
of
conventional
fixed-base
structures
(albeit
with
a
very
different
response
modification
factor).
Seismically
Isolated
Structures -31
This
figure
illustrates
definitions
and
formulas
for
“total”
displacement
at
corners
of
the
isolated
structure
that
includes
potential
rotation,
as
well
as,
translation
of
the
isolation
system.
Total
displacement
is
calculated
as
a
factor
(not
less
than
1.1)
times
translation-only
displacement.
This
factor
is
based
on
the
buildings
center
of
rigidity,
plan
dimensions
and
actual
plus
accidental
mass
eccentricity.
The
key
assumption
underlying
these
equations
is
that
the
distribution
of
isolator
effective
stiffness
in
plan
is
proportional
to
the
distribution
of
mass
(supported
weight)
of
the
structure
above.
5%
percent
eccentricity
increases
corner
displacement
by
about
15%
for
buildings
square
in
plan,
and
by
abut
30%
for
buildings
that
longest
dimensions
many
times
greater
than
the
other.
Systems
with
proportional
stiffer
isolators
near
the
perimeter
of
structure
(e.g.,
de
Young
Museum)
provide
greater
resistance
to
torsion
and
the
effects
of
actual
plus
accidental
mass
eccentricity.
Total
maximum
displacement
(DTM)
is
used
to
verify
stability
of
isolators
for
vertical
loads
and
to
establish
minimum
moat
clearance.
Seismically
Isolated
Structures -32
The
unreduced
design
base
shear
force
(Vb),
the
product
of
the
upper-
bound
value
of
effective
stiffness
(kDmax)
and
the
design
displacement
(DD),
is
required
for
design
of
the
isolation
system,
foundations,
and
other
structure
below
the
isolation
system.
The
base
shear
required
for
design
of
the
structure
above
the
isolation
system
(Vs),
is
reduced
by
the
response
modification
factor
(RI)
which
has
values
ranging
between
1.0
and
2.0.
Design
forces
are
distributed
over
the
height
of
the
building
above
the
isolation
level
assuming
an
inverted-triangular
distribution
which
is,
in
general,
a
conservative
distribution
for
isolated
structures,
particularly
for
isolated
structures
with
a
heavy
first
floor.
Seismically
Isolated
Structures -33
The
response
modification
factor
(RI)
is
defined
as
3/8
of
the
R
factor
(from
Table
12.2-1)
of
the
seismic
force
resisting
system
of
the
structure
above
the
isolation
system,
but
not
greater
the
2.0.
Systems
not
permitted
for
use
in
conventional,
fixed-base,
buildings
(e.g.,
in
high
seismic
regions)
are
also
not
permitted
by
the
Standard
for
isolated
structures.
The
2006
IBC
modified
this
concept
to
allow
use
of
OCBFs
and
OMFs
in
high-seismic
regions
(i.e.,
SDC
D,
E
and
F
structures)
if
designed
for
RI
=
1.0
and
AISC
341.
Seismically
Isolated
Structures -34
This
figure
illustrates
the
trade-off
between
isolated
structure
design
forces
(normalized
by
building
weight)
and
isolation
system
displacement
as
a
function
of
effective
period
for
a
steel
special
concentric
braced
frame
(SCBF)
system.
Design
shear
forces
decrease
(subject
to
certain
limits)
and
design
displacements
increase
as
the
isolated
period
increases.
The
ground
motion
design
values
(i.e.,
corresponding
to
a
region
of
high
seismicity)
and
force-deflection
properties
of
the
isolation
system
used
to
develop
the
design
parameters
shown
in
this
figure
are
the
same
as
those
used
for
the
emergency
operation
center
(EOC)
design
example
covered
later
in
this
presentation.
Note.
The
minimum
value
of
design
base
shear
(Vs,min)
is
based
on
the
requirement
of
Standard
Section
17.5.4.3
that
the
structure
above
the
isolation
system
be
designed
for
not
less
than
the
base
shear
required
by
Section
12.8
for
a
fixed-base
structure
of
the
same
effective
seismic
weight
and
period
(TD)
as
that
of
the
isolated
structure.
As
shown
above,
minimum
value
of
base
shear
is
about
0.10W
applies
to
isolated
periods
of
about
2.5
seconds,
and
greater.
Seismically
Isolated
Structures -35
This
figure
illustrates
the
trade-off
between
isolated
structure
design
forces
(normalized
by
building
weight)
and
isolation
system
displacement
as
a
function
of
effective
period
for
a
steel
ordinary
concentric
braced
frame
(OCBF)
system.
The
ground
motion
design
values
(i.e.,
corresponding
to
a
region
of
high
seismicity)
and
force-deflection
properties
of
the
isolation
system
used
to
develop
the
design
parameters
shown
in
this
figure
are
the
same
as
those
of
the
previous
slide
(of
design
parameters
for
a
steel
SCBF
system.
The
only
differences
are
in
the
values
of
the
R
factor
(i.e.,
6
for
fixed-base
steel
SCBFs
and
3¼
for
steel
fixed-base
steel
OCBFs)
and
the
RI
factor
(2
for
isolated
steel
SCBFs
and
1.0
for
fixed-base
steel
OCBFs).
Seismically
Isolated
Structures -36
The
next
two
slides
illustrate
moments
in
the
structure
above
and
structure
(foundation)
below
due
to
P-Delta
effects
and
horizontal
shear
for
different
types
of
seismic
isolators.
Moments
due
to
P-Delta
effects
are
typically
quite
large
and
require
special
consideration
in
the
modeling
and
analysis
of
isolated
structures.
In
this
slide,
moments
are
shown
on
the
left
for
a
elastomeric
isolator
(i.e..,
either
high-damping
rubber
or
lead-rubber
bearing)
and
on
the
right
for
a
flat
sliding
isolator
with
the
sliding
surface
at
the
base
(face
up).
In
the
case
of
the
elastomeric
isolator
P-Delta
moment
is
shared
approximately
equally
between
the
structure
above
and
foundation
below.
In
the
case
of
the
flat
slider,
the
P-Delta
moment
is
resisted
entirely
by
the
foundation
below.
Note.
If
the
sliding
surface
was
at
the
top
(face
down),
then
the
P-Delta
moment
would
be
resisted
entirely
by
the
structure
above.
Seismically
Isolated
Structures -37
In
this
slide,
moments
are
shown
on
the
left
for
a
double
dish
(or
double
concave)
sliding
isolator
and
on
the
right
for
a
single
dish
sliding
isolator
with
the
sliding
surface
at
the
base
(face
up).
In
the
case
of
the
double
dish
sliding
isolator,
P-Delta
moment
is
shared
approximately
equally
between
the
structure
above
and
foundation
below
similar
to
an
elastomeric
isolator.
In
the
case
of
the
single
dish
sliding
isolator,
P-Delta
moment
is
resisted
entirely
by
the
foundation
below.
Note.
If
the
sliding
surface
was
at
the
top
(dish
facing
down),
then
the
P-Delta
moment
would
be
resisted
entirely
by
the
structure
above.
Seismically
Isolated
Structures -38
This
figure
illustrates
the
bilinear
idealization
of
an
isolator
unit
and
simple
formulas
for
the
effective
period,
Teff,
and
effective
damping,
.eff,
based
on
the
yield
point
(Dy,
Fy)
and
the
point
of
peak
response
(D,
Y).
Although
more
sophisticated
idealizations
could
be
used,
the
relatively
simple
bilinear
idealization
of
isolator
behavior
provides
acceptably
accurate
estimates
of
effective
period
and
effective
damping
for
most
elastomeric
and
friction-pendulum
(dish)
sliding
bearings
at
large
displacements
(i.e.,
D
>>
Dy).
The
bilinear
idealization
reflects
the
inherent
amplitude-dependent
behavior
of
the
effective
period
which
tends
to
increase
with
increasing
displacement
and
effective
damping
which
tends
to
decrease
with
increasing
displacement.
The
bilinear
idealization
does
not
(can
not)
capture
behavior
of
isolators
at
extreme
displacements
which
include
stiffening
of
elastomeric
bearings
at
high
strains
in
the
rubber
(e.g.,
300%
strain),
engagement
of
the
sliding
element
and
the
lip
of
the
dish
of
the
friction-pendulum
sliding
bearing
(for
dish
isolators
that
have
lips),
or
the
complex
behavior
of
the
FPS
triple-pendulum
bearing.
Seismically
Isolated
Structures -39
This
slide
shows
a
section
view
and
dimensions
of
a
double-concave
friction
pendulum
system
(FPS)
bearing
(FPT8844/12-12/8-6,
manufactured
by
Earthquake
Protection
Systems).
FPS
bearings
can
be
fabricated
to
have
different
amounts
of
dynamic
friction
(i.e.,
friction
at
interfaces
between
the
articulated
slider
element
and
the
top
and
bottom
concave
plates).
For
this
isolator,
the
nominal
value
of
dynamic
friction
is
0.06
for
both
top
and
bottom
plate
surfaces,
and
the
lower-bound
and
upper-bound
values
of
dynamic
friction
are
assumed
to
range
from
0.04
to
0.08
considering
all
possible
sources
of
variability
(i.e.,
aging
and
environmental
effects,
manufacturing
tolerances
and
prototype
testing)
as
illustrated
by
the
hysteresis
loops
shown
in
the
next
slide.
Seismically
Isolated
Structures -40
This
figure
shows
modeled
and
tested
hysteresis
loops
for
the
double-
concave
FPS
bearing
at
peak
displacements
of
plus/minus
27
inches.
The
lower-bound
and
upper-bound
loop
properties
are
based
on
values
of
dynamic
friction
of
0.04
and
0.08,
respectively,
and
are
intended
to
bound
variation
in
properties
due
to
aging
and
environmental
effects,
manufacturing
tolerances
as
well
test
loop
variation.
Note.
The
effective
stiffness
(used
to
calculate
effective
period)
and
effective
damping
are
amplitude
dependent.
Therefore,
the
values
of
effective
period
and
effective
damping
used
for
ELF
design
are
a
function
of
ground
motion
intensity
(i.e.,
seismic
design
values
for
the
site
of
interest),
as
well
as
the
properties
of
the
bearing.
The
next
three
slides
illustrate
theoretical
values
of
normalized
force,
effective
period
and
effective
damping
as
a
function
of
bearing
displacement
of
the
double-
concave
FPS
bearing.
In
each
slide,
values
of
the
parameter
of
interest
are
shown
for
nominal
(0.06),
upper-bound
(0.08)
and
lower-bound
(0.04)
values
of
dynamic
friction.
These
curves
are
useful
aids
during
preliminary
ELF
design
of
the
isolation
system.
Seismically
Isolated
Structures -41
This
figure
shows
normalized
bearing
force
(i.e.,
lateral
restoring
force)
of
the
double-concave
bearing.
Lateral
restoring
force
is
a
function
of
the
dynamic
friction
coefficient
and
effective
radius
of
double-concave
configuration
(i.e.,
approximately
185
inches,
the
sum
of
the
two
dish
radii
less
the
height
sliding
elements).
Note.
When
friction
is
nil,
the
restoring
force
is
a
linear
function
of
displacement
divided
by
the
effective
radius.
Seismically
Isolated
Structures -42
This
figure
shows
the
effective
period
of
the
double-concave
FPS
bearing
as
a
function
of
the
dynamic
friction
coefficient
and
effective
radius
of
double-concave
configuration
(185
inches).
Note.
When
friction
is
nil,
the
effective
period
formula
is
same
as
that
of
a
pendulum,
185
inches
in
length.
This
figure
illustrates
the
increase
in
the
effective
period
with
increasing
displacement.
Seismically
Isolated
Structures -43
This
figure
shows
the
approximate
effective
damping
of
the
double-
concave
FPS
bearing
as
a
function
of
the
dynamic
friction
coefficient
and
effective
radius
of
double-concave
configuration
(185
inches).
The
curves
shown
are
intentionally
conservative
(at
small
displacements)
and
the
approximate
formula
only
applies
to
relatively
large
displacements
(D
>
12
inch),
which
are
of
primary
interest.
At
large
displacements,
the
figure
illustrates
the
decrease
in
the
effective
damping
with
increasing
displacement
and
the
importance
of
the
amount
of
dynamic
friction
on
the
value
of
effective
damping.
Seismically
Isolated
Structures -44
Standard
Section
17.4
requires
dynamic
analysis
for
isolated
structures
that:
1.
Are
potentially
near
an
active
fault
(S1
= 0.6g)
2. Are on a “soft” soil site (Site Class E or F)
3. Are “tall” (over 4 stories in height)
4. Have a very long isolated period (TM > 3.0 seconds)
5. Have a relatively flexible superstructure (3T > TD)
6. Have an irregular superstructure, or
7. Have an isolation system with excessive damping, inadequate
restoring force, or displacement restraint.
These criteria effectively require dynamic analysis for most isolated
structures, although ironically dynamic analysis is not required for
more dynamically complex fixed-base structures that do not meet
these criteria (when applicable)
Seismically
Isolated
Structures -45
The
Standard
establishes
a
“safety-net”
of
minimum
displacement
and
force
requirements
based
on
a
percentage
of
ELF
design
values.
The
primary
concern
is
that
response
history
analysis
could
be
misused.
The
ELF
formulas
provide
an
easy
means
of
checking
for
gross
errors.
Seismically
Isolated
Structures -46
The
Standard
requires
dynamic
analysis
models
to
accurately
represent
building
geometry
and
behavior,
including
the
capability
of
evaluating
uplift
of
individual
isolator
units.
Like
ELF
methods,
dynamic
analysis
models
must
evaluate
response
for
upper-bound
and
lower-bound
values
of
isolation
system
properties,
where
bounding
values
are
based
on
prototype
testing
of
isolator
units
and
incorporate
aging
and
environmental
effects
and
other
sources
of
variability.
Typically,
the
superstructure
is
modeled
with
linear
elastic
elements
and
only
the
isolators
(and
dampers,
if
used)
are
modeled
as
nonlinear
elements.
Seismically
Isolated
Structures -47
Response
spectrum
analysis
(RSA)
requires
equivalent
linear
properties
of
the
isolation
system
which
(like
ELF
methods)
are
amplitude
dependent.
Thus,
at
least
four
models
are
required
for
RSA
which
are
the
combinations
of
upper-bound
and
lower-bound
properties
of
the
isolation
system
at
design
earthquake
intensity
and
upper-bound
and
lower-bound
properties
at
MCER
intensity.
The
Standard
requires
100%-30%
of
horizontal
responses
for
RSA
which
is
conservative
for
peak
response
of
isolated
modes
using
maximum
direction
ground
motions.
The
RSA
story
design
shear
force
limit
is
required
to
avoid
underestimation
of
higher-mode
response
when
isolators
are
modeled
with
effective
rather
than
actual
properties.
The
RSA
story
design
shear
force
limit
is
based
on
the
values
of
Vs
calculated
by
RSA.
Although
not
explicitly
required
by
the
Standard,
the
value
of
Vs
used
for
RSA
design
should
not
be
taken
as
less
than
any
of
ELF
limits
on
base
shear
(e.g.,
1.5
times
shear
force
required
to
activate
the
isolation
system).
Seismically
Isolated
Structures -48
While
RHA
is
most
commonly
used
for
dynamic
analysis
of
isolated
structures,
it
is
problematic
for
design
since
the
number
of
analyses
required
to
address
bounding
values
of
properties,
multiple
locations
of
accidental
mass
eccentricity,
etc.,
produce
an
overwhelming
number
of
data.
While
the
Standard
permits
as
few
as
3
earthquake
records,
7
earthquake
records
are
typically
used
RHA,
so
that
the
design
may
be
based
on
the
average
value
of
the
response
parameter
of
interest.
There
is
no
unique
set
of
earthquake
records,
and
earthquake
records
are
typically
developed
on
a
project-specific
basis.
The
Standard
is
vague
on
the
details
for
scaling
earthquake
records
to
match
target
spectra,
and
on
how
the
two
horizontal
components
of
these
records
should
be
oriented
and
applied
to
the
model
(e.g.,
to
address
torsion
due
to
accidental
mas
eccentricity,
etc.).
This
slide
concludes
the
part
of
the
presentation
that
has
addressed
equivalent
lateral
force
(ELF)
design
methods,
modeling
and
analysis
of
the
isolation
system
and
dynamic
lateral
response
analysis
procedures.
The
next
part
of
the
presentation
will
apply
these
methods
and
procedures
to
an
example
design.
Seismically
Isolated
Structures -49
The
design
example
is
a
hypothetical
3-story
(plus
mechanical
penthouse)
emergency
operations
center
(EOC),
assumed
to
be
located
in
Oakland,
California,
approximately
6
kilometers
from
the
Hayward
fault
(i.e.,
high
seismic
location).
Seismic
isolation
is
an
appropriate
design
strategy
for
EOCs
and
other
essential
facilities
where
the
goal
is
to
limit
earthquake
damage
and
protect
facility
function.
Steel
special
concentric
braced
frames
are
used
for
the
seismic
force
resisting
system.
Steel
braced
frames
are
commonly
used
for
structure
isolated
buildings,
although
other
systems
could
have
been
used
in
this
example.
The
isolation
system
incorporates
double-concave
friction
pendulum
sliding
bearings,
although
other
types
of
isolators
could
have
been
used
in
this
example.
Seismically
Isolated
Structures -50
This
slide
summarizes
pertinent
structural
design
criteria
for
the
example
EOC.
Of
note,
Standard
Section
17.5.4.2
specifies
Ie
=
1.0
for
design
of
an
isolated
structure,
regardless
of
the
risk
category.
If
the
EOC
was
not
isolated
(fixed-base
design),
then
the
value
of
the
importance
factor
would
have
been
of
Ie
=
1.5,
the
value
required
by
Table
1.5-2
of
the
Standard
for
design
of
an
“essential”
(Risk
Category
IV)
structures.
Due
to
the
proximity
of
the
site
to
the
Hayward
fault
(i.e.,
S1
=
0.6g)
dynamic
analysis
is
required.
In
this
example,
however,
the
design
is
first
developed
using
ELF
methods
and
then
verified
using
RHA.
The
redundancy
factor
is
taken
as
.
=
1.0,
although
a
larger
value
would
be
required
by
the
Standard
if
the
structure
was
not
isolated.
Seismically
Isolated
Structures -51
This
figure
shows
the
3-dimensional
ETABS
model
of
structure
of
the
EOC
building.
The
ETABS
computer
program
(Computer
Structures
Inc.)
was
selected
for
performing
static
and
dynamic
analyses
of
the
isolated
structure
since
this
software
package
has
a
number
of
isolation-friendly
features,
although
other
commercially
available
structural
analysis
programs
could
have
been
used.
The
ETABS
model
was
used
to
perform
the
following
analyses:
1.
Gravity
Load
Evaluation
–
Distribution
of
building
weight
on
isolators
2.
ELF
–Gravity
and
earthquake
load
design
of
the
superstructure
3.
Pushover
analysis
(ELF
loads)
–
Distribution
of
gravity
and
earthquake
(design
earthquake
and
MCER)
loads
on
isolators
(and
foundations)
4.
RHA
–Peak
design
earthquake
and
MCER
displacement
of
the
isolation
system
and
peak
design
earthquake
and
MCER
story
shear
Seismically
Isolated
Structures -52
This
figure
is
a
plan
view
showing
typical
floor
framing
-4
bays
x
6
bays,
columns
at
25
feet,
on
center.
Seismically
Isolated
Structures -53
This
figure
is
a
plan
view
showing
penthouse
roof
framing.
Seismically
Isolated
Structures -54
This
figure
an
elevation
view
showing
longitudinal
bracing
on
Lines
B
and
D
Seismically
Isolated
Structures -55
This
figure
shows
transverse
bracing
on
Lines
2,
4
and
6
Seismically
Isolated
Structures -56
This
slide
summarizes
basic
design
requirements,
governing
codes
and
material
properties.
Note.
Example
EOC
design
was
developed
in
accordance
with
the
2006
IBC
(and
by
reference
ASCE
7-05)
except
for
seismic
requirements
which
were
based
on
new
ground
motions
and
other
earthquake
provisions
of
the
2009
NEHRP
Provisions
(which
are
essentially
the
same
as
those
of
ASCE
7-10)
Seismically
Isolated
Structures -57
The
weight
of
each
floor
and
distribution
of
the
total
weight
on
isolators
must
be
determined
as
a
necessary
first
step
in
the
design
of
the
isolation
system.
This
figure
shows
the
dead
load
(seismic)
weight
of
gravity
loads
by
floor
level.
The
total
dead
load
(seismic)
weight
on
isolators
is
9,100
kips
with
an
additional
reduced
live
load
weight
on
isolators
is
2,241
kips.
Seismic
weight
is
used
for
lateral
force
design
and
analysis.
Additional
reduced
live
loads
must
be
included
for
gravity
load
design
of
isolators.
Although
isolators
must
be
design
for
additional
requirements
(e.g.,
MCER
loads),
the
same
basic
load
combinations
(e.g.,
long-term
gravity
design
loads)
that
apply
to
columns
of
building
also
apply
to
the
design
of
isolators.
Seismically
Isolated
Structures -58
This
figure
shows
a
plan
view
of
one
quadrant
of
the
EOC
building
at
the
1st-floor.
This
plan
view
is
used
in
this
slide
and
in
subsequent
slides
to
show
the
values
of
loads
on
individual
isolators.
Due
to
building
symmetry,
loads
on
isolators
in
other
quadrants
are
very
similar
(click).
Values
of
dead
load
(D)
and
reduced
live
load
(L)
are
shown
at
each
isolator
location
(click).
Long-term
gravity
design
loads
on
isolators
are
defined
by
1.2D
+
1.6L
load
combination
of
the
2006
IBC.
Long-term
design
loads
on
isolators
range
from
about
220
kips
at
corner
isolators
(A1)
to
about
600
kips
at
an
interior
isolator
(C3).
The
maximum
value
of
long-term
design
load
is
an
important
parameter
for
preliminary
design
of
bearing
sizes.
Seismically
Isolated
Structures -59
The
seismic
design
parameters
must
be
determined
for
the
site
of
interest.
Fortunately,
the
USGS
has
developed
a
website
that
provides
values
of
these
parameters
for
various
editions
of
Seismic
Codes,
including
the
2009
NEHRP
Provisions
(and
ASCE
7-10).
Users
of
the
USGS
website
enter
the
name
of
the
governing
Seismic
Code,
the
site
classification
(e.g.,
from
a
geotechnical
study
of
the
site),
the
risk
category
of
the
building
and
latitude
and
longitude
of
the
building
site
(e.g.,
the
example
EOC
is
assumed
to
located
at
Lat.
37.800
and
Long.
122.250.
The
website
returns
a
report
(summary
report
is
shown
in
this
slide)
including
values
of
the
seismic
design
parameters
for
the
site
of
interest.
In
this
example,
the
site
is
classified
as
“CD”
and
seismic
design
parameters
are
taken
as
the
average
of
the
USGS
values
for
Site
Class
C
and
Site
Class
D.
Note.
The
design
parameters
at
short-periods
given
in
Chapter
12
FEMA
P-751
erroneously
used
two-thirds
of
the
values
shown
in
this
slide
(which
does
not
effect
the
design
example,
other
than
the
value
of
the
vertical
component
ground
motions
in
load
combinations).
Seismically
Isolated
Structures -60
This
figure
shows
design
earthquake
and
MCER
response
spectra
for
the
EOC
site.
These
spectra
are
constructed
in
accordance
with
the
procedure
of
Section
11.4
of
the
Provisions
(and
the
generic
spectrum
shape
of
Figure
11.4-1)
and
the
seismic
design
parameters
obtained
for
the
USGS
web
site.
As
note
previously,
site-specific
ground
motions
(site-specific
spectra)
are
required
for
design
of
isolated
structures
when
S1
=
0.6
which
is
the
case
for
example
EOC.
This
was
not
done
for
this
example,
rather
the
generic
spectra
shown
in
this
figure
were
used
in
lieu
of
site-specific
spectra.
Subject
to
other
limitations,
site-specific
spectra
can
be
taken
as
less
than
100
percent,
but
not
less
than
80
percent
of
generic
spectra
shown
in
the
figure.
For
this
example,
site-specific
spectra
were
conservatively
taken
as
100
percent
of
generic
spectra
shown
in
this
figure.
Seismically
Isolated
Structures -61
Preliminary
design
of
the
isolation
system
requires
selection
of
one
or
more
candidate
isolator
bearing
types
that
have
sufficient
load
and
displacement
capacity
to
meet
project
requirements.
The
formulas
of
the
ELF
procedure
may
be
used
with
site
seismic
design
parameters
to
develop
appropriate
selection
criteria,
as
illustrated
in
this
slide
(copy
of
Slide
35).
While
there
is
no
precise
“right”
set
of
criteria,
this
slide
suggests
that
isolators
will
require
large
(e.g.,
>
30
inch)
displacement
capacity
to
accommodate
MCER
displacements
(DTM),
if
unreduced
lateral
force
(Vb)
on
the
isolated
structure
is
limited
to
0.15W
–0.2W.
The
figure
also
shows
that
the
design
base
shear
for
the
superstructure
(Vs)
is
the
same
for
design
periods
(TD)
greater
than
about
2.5
seconds
(i.e.,
better
performance,
but
no
structure
design
economy
design
periods
longer
then
2.45
seconds).
Seismically
Isolated
Structures -62
This
figure
summarizes
isolation
system
selection
criteria
based
on
the
curves
shown
in
the
preceding
slide.
High-damping
rubber
(HDR),
lead-rubber
(LR)
and
sliding
(FPS)
isolator
bearings
could
all
be
configured
to
meet
these
criteria.
For
this
example,
the
double-concave
FPS
bearing
(FPT884412-12/8-6)
was
selected
for
use
at
each
of
the
35
isolator
locations.
This
bearing
has
a
relatively
long
period
(TD
>
3.5
seconds)
which
helps
to
limit
lateral
forces
and
related
overturning
loads,
and
potential
uplift
of
individual
isolators
below
braced
frames.
One
isolator
bearing
size
is
convenient,
but
typically
(and
especially
for
elastomeric
bearings),
isolated
structures
use
more
than
one
type
of
isolator
bearing
based
on
the
amount
of
vertical
load
supported
by
the
bearing.
Seismically
Isolated
Structures -63
This
figure
is
a
section
view
of
the
double-concave
FPS
bearing
used
in
this
example.
This
bearing
utilizes
an
articulated
slider
between
the
top
and
bottom
concave
plates
(dishes).
For
this
example
double-concave
application,
the
nominal
value
of
dynamic
friction
is
the
same
for
top
and
bottom
concave
plates
such
that
total
displacement
is
“symmetric”
and
shared
approximately
equally
between
the
top
and
the
bottom
concave
plates.
Other
configurations
of
this
bearing
(i.e.,
triple-pendulum
bearings)
utilize
different
nominal
values
of
dynamic
friction
for
the
top
and
bottom
plates
which
affect
a
more
complex
“asymmetric”
pattern
of
total
displacement.
Seismically
Isolated
Structures -64
This
slide
summarizes
ETABS
gravity
and
seismic
force
analyses
and
related
load
combinations
required
for
design
of
the
example
EOC.
Reduced
design
earthquake
loads
are
used
in
load
combinations
for
design
of
the
superstructure.
Unreduced
design
earthquake
loads
are
used
in
load
combinations
for
design
of
the
isolation
system,
and
foundation
elements
below.
Seismically
Isolated
Structures -65
This
slide
summarizes
ETABS
pushover
analyses
and
related
load
combinations
used
to
determine
maximum
and
minimum
short-term
forces
on
individual
isolators
due
to
the
design
earthquake
loads
and
maximum
short-term
(downward
forces)
and
minimum
short-term
(maximum
uplift
displacements)
on
individual
isolators
due
to
MCER
loads.
These
forces
(and
displacements)
are
used
for
design
of
isolators
and
for
establishing
load
criteria
for
prototype
testing
of
isolator
units.
Seismically
Isolated
Structures -66
The
next
sequence
of
eight
slides
illustrates
preliminary
design
of
EOC
using
the
formulas
of
the
ELF
procedure
and
site
seismic
parameters
and
the
effective
properties
of
the
double-concave
FPS
bearing.
In
this
figure,
the
design
displacement
(DD)
is
calculated
as
16.0
inches,
based
on
an
effective
period
of
TD
=
3.5
seconds
and
an
effective
damping
of
ßD
=
20%
(BD
=
1.5).
The
effective
period
and
damping
parameters
are
amplitude
dependent
and
the
process
to
determine
their
values
is
necessarily
iterative.
Figures
of
effective
period
and
damping
as
a
function
of
double-concave
FPS
bearing
displacement
shown
in
this
slide
(i.e.,
figures
shown
previous
as
Slides
42
and
43)
are
used
to
expedite
the
process.
As
shown
by
the
red
lines,
an
effective
period
of
TD
=
3.5
seconds
and
an
effective
damping
of
ßD
=
20%
correspond
to
about
16
inches
of
displacement
of
the
double
concave
FPS
bearing.
Seismically
Isolated
Structures -67
In
this
figure,
the
maximum
displacement
(DM)
is
calculated
as
about
31
inches,
based
on
an
effective
period
of
TD
=
3.9
seconds
and
an
effective
damping
of
ßD
=
13%
(BD
=
1.3).
The
effective
period
and
damping
parameters
are
amplitude
dependent
and
the
process
to
determine
their
values
is
necessarily
iterative.
Figures
of
effective
period
and
damping
as
a
function
of
double-concave
FPS
bearing
displacement
shown
in
this
slide
(i.e.,
figures
shown
previous
as
Slides
42
and
43)
are
used
to
expedite
the
process.
As
shown
by
the
red
lines,
an
effective
period
of
TD
=
3.9
seconds
and
an
effective
damping
of
ßD
=
13%
correspond
to
about
30
inches
of
displacement
of
the
double
concave
FPS
bearing.
Seismically
Isolated
Structures -68
This
slide
illustrates
the
ELF
formulas
used
to
calculate
“total”
design
and
maximum
isolation
system
displacement
where
“total”
means
displacement
(at
corners)
due
to
both
translation
and
rotation
due
to
actual
plus
accidental
mass
eccentricity.
Amplification
of
translation-only
displacement
is
based
on
plan
geometry
and
the
assumption
that
isolator
stiffness
is
distributed
in
plan
proportional
to
supported
weight.
For
the
example
EOC,
the
formulas
suggest
a
25
percent
increase
in
displacement
due
to
rotation.
The
Standard
permits
using
a
smaller
amount
displacement
amplification,
but
not
less
than
10
percent,
provided
such
can
be
justified.
In
the
case
of
sliding
isolators,
the
“stiffness”
of
the
bearing
is
approximately
proportional
to
the
weight
supported,
effectively
reducing
the
potential
for
rotation
due
to
accidental
mass
eccentricity.
On
this
basis,
the
example
EOC
is
designed
for
the
minimum
10
percent
increase
in
displacement
and
the
total
maximum
displacement
(DTM)
of
a
corner
of
the
example
EOC
is
34
inches
(approximately
the
displacement
capacity
of
the
double-concave
bearing).
Seismically
Isolated
Structures -69
In
this
figure,
the
minimum
value
of
effective
design
stiffness
is
calculated
for
TD
=
3.5
seconds
and
W
=
9,100
kips,
and
the
maximum
value
of
effective
stiffness
is
estimated
as
1.2
times
the
minimum
value
(by
comparing
normalized
bearing
force
at
16
inches,
as
shown
in
the
figure).
ELF
formulas
use
maximum
effective
stiffness
to
calculate
lateral
design
force.
This
approach,
originally
developed
for
elastomeric
bearings
also
works
for
sliding
bearings,
recognizing
that
the
stiffness
is
related
to
weight
on
the
bearing
(since
the
friction
force
is
proportional
to
the
weight
supported).
Seismically
Isolated
Structures -70
In
this
figure,
the
minimum
value
of
effective
stiffness
at
MCER
displacement
is
calculated
for
TM
=
3.9
seconds
and
W
=
9,100
kips,
and
the
maximum
value
of
effective
stiffness
at
MCER
displacement
is
estimated
as
1.15
times
the
minimum
value
(by
comparing
normalized
bearing
force
at
30
inches,
as
shown
in
the
figure).
Maximum
effective
stiffness
at
MCER
displacement
is
used
to
calculate
MCER
forces
on
the
isolation
system.
Seismically
Isolated
Structures -71
This
figure
illustrates
the
calculation
of
the
unreduced
shear
force
(Vb)
required
for
design
of
the
isolation
system
and
the
foundation,
the
calculation
of
the
unreduced
shear
force
at
MCER
displacement
used
to
check
the
stability
of
the
isolation
system,
and
calculation
of
the
reduced
shear
force
(Vs)
required
for
design
of
the
superstructure.
In
all
cases,
shear
force
is
expressed
as
a
fraction
of
the
seismic
weight
(W).
The
shear
force
required
for
design
of
the
superstructure
is
reduced
by
a
factor
of
2,
subject
to
other
limits
of
design
base
shear.
Without
these
limits,
the
design
shear
force
would
be
Vb
=
0.16W
reduced
by
RI
=
2,
or
Vs
=
0.08W.
However,
this
value
of
shear
force
is
less
than
both
the
minimum
force
required
for
design
of
fixed-base
building
of
period,
TD
=
3.5
seconds
(i.e.,
Vs
=
0.094W)
and
the
shear
force
required
to
activate
the
isolation
system
based
on
maximum
value
of
dynamic
friction,
µP, m a x =
0.08
(i.e.,
Vs
=
0.12W).
Hence,
the
superstructure
of
the
example
EOC
building
is
designed
for
Vs
=
0.12W
or
1,092
kips,
the
isolation
system
and
foundation
are
designed
for
Vb
=
0.16W,
or
1,456
kips,
and
the
stability
of
the
isolation
system
is
checked
for
VMCE
=
0.24W,
or
2,184
kips
of
peak
lateral
force.
Seismically
Isolated
Structures -72
This
figure
illustrates
two
of
the
hysteresis
loops
implicitly
used
for
ELF
design.
The
hysteresis
loop
with
plus/minus
31-inch
peak
displacement
(solid
line)
is
based
on
the
minimum
value
of
dynamic
friction
(0.04),
since
this
results
in
the
largest
value
of
displacement
(DM
=
30.9
inches).
The
hysteresis
loop
with
plus/minus
16-inch
peak
displacement
(dashed
line)
is
based
on
the
maximum
value
of
dynamic
friction
(0.08),
since
this
results
in
the
largest
value
of
design
force
(Vs
=
0.16W).
Note.
The
design
displacement,
DD
=
16
inches,
was
calculated
using
a
value
of
effective
damping,
ßD
=
20,
that
was
based
on
minimum
dynamic
friction
(0.04).
Thus,
ELF
forces
are
conservatively
based
on
maximum
effective
stiffness,
at
displacements
based
on
minimum
effective
stiffness.
This
is
an
intentional
conservatism
of
the
Standard
for
ELF-based
design.
Seismically
Isolated
Structures -73
This
figure
illustrates
the
ELF
calculation
of
story
forces
for
the
example
EOC
building
using
Standard
Eq.
17.5-9.
Height
is
measured
form
the
isolation
interface
such
that
the
first
floor
is
4
feet
above
the
base
and
the
Penthouse
(PH)
roof
is
54
feet
above
the
base
of
the
isolated
building.
Eq.
17.5-9
is
based
on
an
inverted
triangular
distribution
of
lateral
response
with
height
and
the
resulting
values
of
cumulative
force
normalized
by
cumulative
weight
at
each
story
may
be
seen
to
increase
from
12
percent
(Vs
=
0.12W)
at
base
(isolation
interface)
to
25
percent
at
the
penthouse
level.
In
general,
Eq.
17.5-9
is
conservative
for
isolated
structures
that
have
an
isolated
period
that
is
at
least
3
times
the
period
of
the
structure
above
the
isolation
on
a
fixed
base.
Seismically
Isolated
Structures -74
This
slide
is
a
figure
of
the
ELF
distribution
of
lateral
forces
corresponding
to:
(1)
design
of
the
superstructure
(reduced
design
earthquake
forces),
(2)
design
of
the
isolation
system
and
foundation
(unreduced
design
earthquake
forces)
and
(3)
checking
isolation
system
stability
(unreduced
MCER
forces).
Seismically
Isolated
Structures -75
The
next
four
slides
show
the
design
of
seismic
bracing
and
typical
framing
based
on
ELF
forces
(reduced
design
earthquake
forces).
This
first
figure
shows
seismic
bracing
and
typical
framing
on
Lines
2
and
6.
Seismic
braces
are
10-inch
square
tubes
(HSS
10
x
10
x
5/8)
at
all
locations.
Seismically
Isolated
Structures -76
This
figure
shows
seismic
bracing
and
typical
framing
on
Line
4.
Seismic
braces
are
10-inch
square
tubes
(HSS
10
x
10
x
1/2)
at
all
locations.
Seismically
Isolated
Structures -77
This
figure
shows
seismic
bracing
and
typical
framing
on
Lines
B
and
D.
Seismic
braces
are
10-inch
square
tubes
(HSS
10
x
10
x
5/8)
at
all
locations.
Seismically
Isolated
Structures -78
This
figure
shows
first-floor
framing.
Blue
shading
shows
heavier
W24
x
146
girders
on
column
(isolator)
lines
where
framing
is
required
to
resist
moments
due
to
P-D
loads
and
horizontal
shear
in
isolators.
Seismically
Isolated
Structures -79
This
slide
shows
a
typical
detail
of
the
isolation
system
at
an
isolator
bearing.
The
isolator
bearing
is
installed
with
anchors
and
grout
directly
above
the
top
of
the
reinforced-concrete
foundation
(rebar
not
shown).
The
strength
of
the
grout
and
the
design
of
reinforced-concrete
foundation
is
governed
by
loads
from
bottom
plate
of
the
double-concave
bearing
considering
all
possible
displacements
of
the
articulated
slider.
Foundation
anchors
utilize
threaded
rods
and
couplers
to
permit
bearing
removal.
The
top
plate
to
the
bearing
bears
on
a
(milled)
heavy
steel
plate
attached
to
the
base
of
the
column.
Steel
sections
and
stiffeners
running
from
the
underside
of
the
girder
to
the
top
of
the
heavy
plate
are
designed
to
provide
stability
to
the
top
concave
plate
of
the
bearing
(again
considering
all
possible
displacements
of
the
articulated
slider).
Seismically
Isolated
Structures -80
The
figure
shows
typical
gravity
(dead
load
and
reduced
live
load)
weight
on
isolator
bearings
(same
loads
shown
previously
in
Slide
58).
Based
on
the
load
combination,
1.0D
+
0.5L,
weight
on
individual
bearings
varies
from
about
150
kips
to
about
400
kips
and
the
typical,
or
average
weight,
supported
by
all
bearing
is
about
290
kips
(i.e.,
35
bearings
x
290
kips/bearing
=
10,150
kips)
and
the.
The
typical
weight
on
isolators
(i.e.,
without
load
factors)
is
used
for
prototype
testing
of
isolator
units.
Seismically
Isolated
Structures -81
The
next
four
slides
summarize
forces
on
(or
uplift
displacements
of)
individual
isolator
bearings
due
to
dead,
live
and
seismic
(overturning)
loads
for
different
load
combinations.
Seismic
(overturning)
loads
on
individual
bearings
were
calculated
by
pushover
analysis
using
ELF
lateral
seismic
loads.
Forces
(or
uplift
displacements)
are
shown
for
both
the
X
and
Y
direction
of
earthquake
response
(i.e.,
direction
of
the
pushover).
The
first
figure
shows
maximum
downward
gravity
and
design
earthquake
forces
on
isolators,
based
on
the
load
combination,
(1.2D
+
0.2SDS)D
+
0.5L
+
QDE,
where
QDE
is
the
vertical
load
on
the
isolator
bearing
due
to
the
design
earthquake.
The
typical
value
of
the
maximum
downward
design
force,
about
500
kips,
is
used
to
establish
upper-bound
vertical
load
for
prototype
testing
of
isolator
units.
Seismically
Isolated
Structures -82
This
figure
shows
minimum
downward
gravity
an
earthquake
design
forces
on
isolators,
based
on
the
load
combination,
(0.9D -0.2SDS
)D -QDE,
where
QDE
is
the
vertical
load
on
the
isolator
bearing
due
to
the
design
earthquake.
The
typical
value
of
the
minimum
downward
design
force,
about
150
kips,
is
used
to
establish
lower-bound
vertical
load
for
prototype
testing
of
isolator
units.
Seismically
Isolated
Structures -83
This
figure
shows
maximum
downward
MCER
forces
on
isolators,
based
on
the
load
combination,
(1.2D
+
0.2SMS
)D
+
1.0L
+
QMCE,
where
QMCE
is
the
vertical
load
on
the
isolator
bearing
due
to
the
maximum
considered
earthquake.
The
maximum
downward
design
force,
about
1,000
kips,
is
used
for
prototype
testing
of
isolator
units
to
check
stability
at
maximum
MCER
displacement.
Seismically
Isolated
Structures -84
This
figure
shows
maximum
MCER
uplift
displacement
of
isolators,
based
on
the
load
combination,
(0.9D -0.2SMS
)D -QMCE,
where
QMCE
is
the
vertical
load
on
the
isolator
bearing
due
to
the
maximum
considered
earthquake.
The
maximum
uplift
displacement,
about
1/100th
of
an
inch,
is
used
for
prototype
testing
of
isolator
units
to
check
stability
at
maximum
MCER
displacement
(i.e.,
would
a
small
amount
of
uplift
cause
the
bearing
to
malfunction?).
This
slide
concludes
the
presentation
of
preliminary
design
using
ELF
methods,
the
next
series
of
slides
addresses
dynamic
analysis
requirements
and
verification
of
the
design
of
the
example
EOC
using
response
history
analysis
(RHA)
methods
Seismically
Isolated
Structures -85
The
Standard
requires
dynamic
analysis
for
design
of
example
EOC
since
the
site
is
potentially
located
near
an
active
source
(based
on
the
value
S1
=
0.6g) and because the isolated structure has period, TM < 3.0
seconds. These triggers for required use dynamic analysis date to
the original development of isolated structure provisions (more than
20 years ago) and reflect concerns about earthquake ground
motions, rather than the method of analysis. Today, ground motion
data available from the USGS is much more reliable at long periods,
and better incorporate near-field affects.
The requirement for dynamic analysis can be satisfied using
response spectrum analysis (RSA) when a site-specific ground
motion study is also required (i.e., S1 =
0.6g). Since RSA models
are linear elastic and use essentially the same effective stiffness
and damping properties as the ELF procedure, little is gained with
RSA other than a better and more convenient calculation of
responses in individual elements of the isolated structure.
The primary concern for isolated structures located at near-field
sites is the potential for ground motions to contain “pulses” that
could displace the isolated structure more than predicted by the ELF
formulas (or RSA methods). Response history analysis (RHA)
using ground motions recorded near fault rupture (and presumably
containing “pulses”) addresses this concern.
RHA is used in this example to verify ELF design forces (and uplift
Seismically
Isolated
Structures -86
displacements) and to calculate maximum displacements for design of the isolation
system (i.e., to slightly reduce maximum earthquake displacement required for design
from that calculated using the ELF procedure).
Seismically
Isolated
Structures -86
In
this
example,
site-specific
design
earthquake
MCER
spectra
were
taken
as
equal
to
100
percent
of
their
respective
“Code”
spectra
shown
in
this
figure,
rather
than
calculating
actual
site-specific
spectra
in
accordance
with
Standard
Section
21.1.
While
this
approach
would
not
be
permitted
for
design
of
real
building,
use
of
Code
spectra
was
convenient
for
the
example
EOC
and
also
provides
for
an
“apples-to-apples”
comparison
of
RHA
results
and
ELF
design
parameters.
Seismically
Isolated
Structures -87
Standard
Section
17.3.2
defers
to
Chapter
16
(“Seismic
Response
history
Procedure”)
for
ground
motion
record
selection
and
scaling
requirements,
with
the
exception
that
selected
ground
motion
records
need
only
envelop
the
target
spectrum
(e.g.,
Code
spectrum
in
this
example)
at
periods
from
0.5TD
to
1.25TM.(i.e.,
Chapter
16
requires
enveloping
the
target
spectrum
over
a
broader
range
of
fixed-base
building
periods).
Most
of
the
selection
and
scaling
requirements
of
Chapter
16
were
initially
developed
for
isolated
structures
and
may
be
found
in
the
isolation
provisions
of
older
editions
of
seismic
codes.
Ideally,
ground
motions
are
selected
from
recorded
events
whose
earthquake
magnitude,
fault
type,
distance
to
the
plane
of
fault
rupture
and
site
conditions
are
the
same
or
comparable
to
those
of
the
site
of
interest
and
the
fault
that
governs
seismic
hazard
at
the
site
of
interest.
The
fault
that
governs
seismic
hazard
at
a
given
site
may
obtained
from
a
USGS
web
site
that
provides
hazard
de-aggregation
data.
For
this
example,
the
Hayward
fault
was
found
to
govern
site
hazard.
The
Hayward
fault
is
a
strike-slip
system
that
has
the
capability
of
producing
M7+
events,
and
is
located
about
6
km
from
the
assumed
location
of
the
example
EOC.
Site
conditions
should
be
determined
by
a
geotechnical
study,
however,
for
this
example
the
site
conditions
were
assumed
to
be
Site
Class
C/D
(i.e.,
shear
wave
velocity,
vs,30
=
450
meters/sec.)
,
Seismically
Isolated
Structures -88
This
slide
lists
the
seven
earthquake
records
(two
components
each)
selected
for
RHA
of
the
example
EOC.
The
seven
records
were
selected
from
the
22
Far-Field
(FF)
records
and
28
Near-Field
(NF)
records
of
FEMA
P-695.
FEMA
P-695
is
a
convenient
source
of
the
strongest
ground
motion
records
recorded
to
date
and
available
form
the
PEER
NGA
database
(which
has
thousand
ground
motion
records,
but
only
a
limited
number
strong
ground
motion
records
from
large
magnitude
events
recorded
relatively
close
to
fault
rupture).
The
records
selected
for
RHA
of
example
EOC
were
the
seven
FEMA
P695
records
deemed
to
best
match
Hayward
fault
and
example
EOC
site
characteristics.
As
shown
in
the
table,
all
records
are
from
strike-slip
earthquakes
whose
magnitude
is
M7.37,
on
average,
whose
distance
to
fault
rupture
is
5.3
km,
on
average
(using
the
Joyner-Boore
definition
of
fault
distance)
and
whose
site
conditions
are
Site
Class
C
or
Site
Class
D
(i.e.,
vs,30
=-446
meters/sec.,
on
average).
It
may
be
noted
that
the
characteristics
of
these
seven
records
are,
on
average,
essentially
the
same
as
those
of
the
example
EOC
site,
and
the
nearby
Hayward
fault
which
governs
ground
motion
hazard
at
the
site.
Seismically
Isolated
Structures -89
This
slide
summarizes
scaling
factors
each
of
the
seven
records.
Chapter
16
of
the
Standard
is
clear
on
the
objective,
but
not
details
of
record
scaling.
For
this
example,
records
are
first
oriented
to
have
a
common
axis
of
stronger
shaking
at
long
periods
of
interest.
That
is,
the
stronger
component
of
each
record
is
grouped
together
(e.g.,
X
direction)
for
subsequent
application
to
the
RHA
model
of
example
EOC.
Records
are
scaled
by
a
two
step
process.
First,
records
are
“normalized”
in
terms
of
peak
ground
velocity
(PGV).
That
is,
each
record
is
scaled
up
or
down
to
have
the
same
value
of
PGV
(i.e.,
58.3
cm/s).
PGV
normalization
is
based
on
record
scaling
methods
of
FEMA
P695.
Second,
each
of
the
seven
records
is
scaled
by
the
same
factor,
as
required
to
envelop
the
target
spectrum
over
the
period
range
of
interest.
As
per
the
scaling
requirements
of
Standard
Chapter
16,
enveloping
criteria
require
the
square-root-sum-of-the-squares
(SRSS)
combination
of
the
spectra
of
two
horizontal
components
to
equal
or
exceed
the
target
spectrum
(i.e.,
100%
of
the
design
earthquake
spectrum
or
100
%
of
the
MCER
spectrum)
at
each
period
of
interest.
The
table
shows
the
individual
and
median
scale
factors
for
the
seven
records.
It
may
be
noted,
that
the
median
scale
factor
for
enveloping
the
design
earthquake
(DE)
spectrum
is
1.04
which
implies
that
the
records
are,
on
average,
representative
of
median
deterministic
ground
motions
that
would
be
expected
at
the
example
EOC
site
for
a
large
magnitude
(M7+)
event
on
Hayward
fault.
Seismically
Isolated
Structures -90
This
figure
compares
the
MCER
“target”
spectrum
(heavy
black
line)
with
the
average
spectrum
of
the
SRSS
combination
of
the
seven
scaled
records
(solid
black
line)
showing
that
average
spectrum
of
the
SRSS
combination
(of
the
two
horizontal
components)
envelops
the
target
spectrum
from
periods
of
0.5TD
(1.75
seconds)
to
1.25
TM
(4.9
seconds),
as
required
by
Standard
Section
17.3.2.
Also
shown
in
this
figure,
the
average
spectrum
of
the
seven
stronger
components
and
the
average
spectrum
of
the
seven
weaker
components
(i.e.,
stronger/weaker
at
long
periods
of
interest.
It
may
be
noted
that
the
average
spectrum
of
the
stronger
components
is
approximately
equal
to
the
target
spectrum
at
about
4
seconds.
Note.
The
MCER
spectrum
shown
in
Figure
12.5-7
of
FEMA
P751
is
not
shown
correctly
(i.e.,
short-period
spectral
accelerations
should
as
shown
in
this
slide)
–no
affect
on
the
design
of
the
example
EOC.
Seismically
Isolated
Structures -91
This
slide
summarizes
modeling
requirements
for
RHA
as
applied
to
the
ETABS
model
of
example
EOC
structure.
The
superstructure
of
an
isolated
structure
is
permitted
to
be
modeled
as
linear
elastic
provided
it
remains
“essentially
elastic.”
Such
is
the
case
for
the
example
EOC
which
remained
essentially
elastic
even
for
MCER
demands
(i.e.,
since
the
braces
are
conservatively
sized
using
ELF
forces).
The
isolation
system
of
the
EOC
incorporated
two
sources
of
nonlinearity
(1)
individual
isolators
were
modeled
as
essentially
bi-linear
elements
using
the
ETABS
“Isolator2”
element,
and
(2)
gap
elements
are
used
to
permit
uplift
at
individual
isolators.
The
ETABS
“Isolator2”
element
was
developed
to
represent
bi-linear,
hysteretic
behavior
of
FPS
bearings
and
accounts
for
changes
in
friction
properties
with
velocity
during
dynamic
response.
Two
primary
ETABS
models
were
developed
representing
upper-bound
(0.08)
and
lower-bound
(0.04)
values
of
dynamic
friction
(as
illustrated
in
the
next
slide).
Seismically
Isolated
Structures -92
This
figure
(copy
of
Slide
41)
shows
modeled
and
tested
hysteresis
loops
for
the
double-concave
FPS
bearing
at
peak
displacements
of
plus/minus
27
inches.
The
lower-bound
and
upper-bound
loop
properties
are
based
on
values
of
dynamic
friction
of
0.04
and
0.08,
respectively,
and
are
intended
to
bound
variation
in
properties
due
to
aging
and
environmental
effects,
manufacturing
tolerances,
as
well
test
loop
variation.
An
ETABS
model
of
the
isolated
structure
of
the
example
EOC
with
bearings
modeled
with
lower-bound
bi-linear
properties
was
analyzed
using
RHA
to
determine
peak
isolation
system
displacements.
An
ETABS
model
of
the
isolated
structure
of
the
example
EOC
with
bearings
modeled
with
upper-bound
bi-linear
properties
was
analyzed
using
RHA
to
determine
peak
forces
in
the
isolation
system
and
superstructure.
Seismically
Isolated
Structures -93
This
slide
summarizes
the
average
values
of
peak
story
shear
force
in
the
X-axis
direction
and
in
the
Y-axis
direction
of
the
isolated
structure
calculated
using
RHA
and
compares
these
forces
with
the
corresponding
value
peak
story
shear
force
calculated
using
the
ELF
procedure.
Results
shown
for
RHA
results
represent
the
maximum
(worst
case)
of
four
orientations
the
larger
components
of
the
seven
records
which
were
applied
to
the
ETABS
model
in
separate
sets
of
analyses
in
the
(1)
positive
X-axis
direction,
(2)
negative
X-axis
direction,
(3)
positive
Y-axis
direction
and
(4)
negative
Y-axis
direction.
As
shown
in
the
table,
the
peak
story
shear
forces
in
the
X-axis
direction
and
Y-axis
direction
of
the
isolated
are
essentially
the
same
(i.e.,
both
governed
by
the
larger
components
oriented
in
the
direction
of
interest),
and
are
remarkably
similar
to
the
value
of
shear
force
at
the
isolation
level.
There
is,
however,
a
significant
difference
in
the
ELF
and
RHA
story
shear
force
results
at
levels
above
the
isolation
interface,
as
illustrated
by
the
figure
in
the
next
slide.
Seismically
Isolated
Structures -94
This
figure
shows
that
story
shear
forces
calculated
using
the
ELF
formula
is
generally
conservative
for
the
example
EOC,
as
compared
to
the
average
results
of
RHA.
This
result
indicates
that
higher
modes
of
the
isolated
structure
do
not
contribute
significantly
to
the
response
of
the
upper
floors
which
is,
in
part,
due
to
the
relatively
large
separation
in
the
period
of
superstructure
(on
a
fixed
base)
and
the
period
of
the
isolated
structure.
The
ratio
of
isolation
system
design
period
(TD
=
3.5
seconds)
is
over
six
time
the
period
of
the
superstructure
on
a
fixed-base.
The
results
of
the
RHA
verify
that
forces
calculated
using
the
ELF
procedure
and
used
for
preliminary
design
are
conservative.
Preliminary
sizes
of
superstructure
elements
could
be
refined
(e.g.,
wall
thickness
of
HSS
sections
could
be
reduced),
but
cost
savings
would
be
modest.
Seismically
Isolated
Structures -95
This
slide
summarizes
the
average
values
of
peak
isolation
system
displacement
calculated
using
RHA
and
compare
each
of
these
results
with
the
corresponding
value
calculated
using
the
ELF
procedure.
Note.
Multiple
record
orientations
of
the
seven
records
used
to
determine
“worst
case”
forces,
are
not
required
to
determine
the
peak
displacement
in
the
horizontal
plan
(i.e.,
same
for
all
orientations).
However,
multiple
orientations
of
records
were
used
to
check
for
potential
uplift
of
individual
isolators.
As
shown
in
the
table,
the
peak
displacements
calculated
using
ELF
formulas
and
“maximum
direction”
ground
motions
compare
well
with
the
average
displacement
in
the
X-Y
plane
(maximum
direction)
calculated
by
RHA.
The
slightly
smaller
values
of
peak
displacement
calculated
using
RHA
are
used
for
“final”
design
of
the
isolations
and
for
testing
of
isolator
prototypes.
Seismically
Isolated
Structures -96
This
slide
summarizes
peak
displacement
results
for
each
of
the
seven
records
as
well
for
the
average
of
the
set
of
seven.
RHA
results
are
based
on
analyses
with
the
larger
components
oriented
in
the
X-axis
direction.
Peak
displacement
results
are
reported
in
the
X-axis
direction,
in
the
Y-
axis
direction
and
the
in
X-Y
direction
(i.e.,
maximum
displacement
in
the
horizontal
plane)
for
records
scaled
to
the
design
earthquake
spectrum
(upper
set
of
brown
cells)
and
for
records
scaled
to
MCER
spectrum
(lower
set
of
brown
cells).
The
Standard
does
not
require
the
isolation
system
(or
the
superstructure)
to
be
designed
for
the
worst-case
response
of
the
seven
records
(i.e.
just
the
average
response).
However,
it
is
important
to
recognize
that
there
is
always
the
possibility
even,
if
very
remote,
that
design-basis
response
could
be
exceeded.
In
the
case
of
the
example
EOC,
none
of
the
seven
records
exceed
bearing
displacement
capacity
(i.e.,
33
inches)
for
design
earthquake
ground
motions,
but
two
of
the
seven
records
(FF-10,
NF-25)
significantly
exceed
isolator
displacement
capacity
for
MCER
ground
motions.
The
result
of
the
isolation
system
trying
to
respond
beyond
bearing
displacement
capacity
would
be
damage
to
the
bearings
(and
moat
wall),
higher
forces
in
the
superstructure
and
likely
damage
to
braces
(but
not
structural
failure,
since
the
steel
SCBFs
are
designed
to
yield
in
a
ductile
manner).
This
slide
also
summarizes
ELF
estimates
of
peak
displacement
of
individual
records
based
on
the
value
of
spectral
acceleration
at
the
isolated
period
and
the
next
slide
compares
these
displacements.
Seismically
Isolated
Structures -97
This
slide
compares
RHA
and
ELF
displacements
from
the
table
of
the
previous
slide.
The
very
close
agreement
between
displacements
calculated
using
RHA
and
ELF
methods
(applied
to
response
spectra
of
individual
records)
illustrates
the
usefulness
of
the
ELF
procedure,
and
the
related
concepts
of
effective
period
and
effective
damping,
to
provide
a
“sanity
check”
on
RHA
results.
This
slide
concludes
the
part
of
the
presentation
that
has
illustrated
the
example
design
of
a
hypothetical
seismically-isolated
emergency
operations
center
(EOC).
The
next
and
final
part
of
the
presentation
will
address
required
testing
of
prototype
isolator
units
using
material
from
the
EOC
design
example
Seismically
Isolated
Structures -98
Detailed
design
of
the
isolator
units
typically
is
the
responsibility
of
the
manufacturer
subject
to
design
and
testing
(performance)
criteria
included
in
the
construction
documents
(drawings
and/or
specifications).
Performance
criteria
typically
include
a
basic
description
and
size(s)
of
isolator
units,
design
criteria
(e.g.,
loads,
displacement
capacity,
force-
deflection
properties,
etc.),
quality
assurance
and
quality
control
requirements
(including
QC
testing
of
production
units)
and
prototype
testing
requirements.
Section
17.8
of
the
Standard
specifies
a
series
prototype
tests
for
establishing
and
validating
design
properties.
Note.
While
these
test
are
typically
performed
on
project
specific
basis,
they
need
only
be
performed
once
in
comprehensive
manner
to
establish
design
properties
for
“standardized”
isolator
products.
This
slide
summarizes
the
number
and
type
of
test
specimens
required
for
prototype
testing.
Seismically
Isolated
Structures -99
This
slide
illustrates
the
sequence
and
cycles
of
prototype
testing
required
by
Section
17.8.2.2
of
the
Standard
using
the
design
loads
and
design
displacements
of
the
example
EOC.
There
are
three
distinct
elements
of
prototype
testing.
First,
cyclic
load
tests
are
at
incremental
displacements
are
required
to
establish
the
force-
deflection
properties
of
isolators
(e.g.,
effective
stiffness
and
effective
damping)
for
typical,
upper-bound,
and
lower-bound
vertical
loads.
In
the
case
of
the
example
EOC,
incremental
test
displacements
are
4,
8,
16
and
30
inches,
and
vertical
loads
are
290
kips
(typical
vertical
load),
150
(lower-bound
vertical
load)
and
500
kips
(upper-bound
vertical
load).
Second,
11
cycles
of
load
at
the
design
displacement
(DTD
=
17.5
inches)
are
required
to
assess
isolator
prototype
“durability”
for
typical
vertical
load
(290
kips).
The
number
of
cycles
is
based
on
the
formula,
30
SD1/SDSBD
= 10 cycles
of
load
which
is
a
conservative
estimate
of
the
effective
number
of
cycles
for
two
maximum
considered
earthquake
events
(i.e.,
main
shock
plus
after
shock
as
large
as
the
main
shock)
Third,
a
static
load
test
is
required
to
check
isolator
stability
at
maximum
displacement
(DTM
=
32.5
inches)
for
both
maximum
downward
load
(1,000
kips)
and
minimum
downward
load
which
includes
uplift
if
the
minimum
downward
load
is
nil
and
structure
above
the
bearing
moves
upward
(e.g.,
0.1
inch
of
uplift
is
specified
for
example
EOC).
Note.
The
test
uplift
displacement
of
0.1
inch
(although
larger
than
the
0.01
inch
value
measured
by
dynamic
(RHA)
analysis)
is
too
small
to
cause
bearing
Seismically
Isolated
Structures -100
malfunction
(and
could
be
ignored).
Seismically
Isolated
Structures -100
This
figure
illustrates
typical
results
of
cyclic
load
tests
required
for
determining
the
force-deflection
properties
of
the
isolator.
In
this
illustration,
the
isolator
is
loaded
with
an
average
vertical
load
of
383
kips
and
cycled
at
low
velocities
(peak
velocity
is
less
than
5
in/sec.)
to
plus
and
minus
27
inches
of
displacement.
While
the
loads
of
this
illustration
are
not
quite
the
same
as
those
of
the
example
EOC,
they
are
actual
test
results
of
the
model
of
double-concave
FPS
bearing
as
that
used
in
the
design
example.
Effective
stiffness
and
damping
are
calculated
at
each
cycle
of
test,
and
in
this
illustration
are
very
similar
–the
dynamic
friction
coefficient
is
about
0.06
and
the
effective
damping
is
a
little
over
19%
(at
27
inches),
on
average.
As
shown
on
the
slide,
vertical
load
varies
during
cyclic
loading
(from
304
kips
to
467
kips).
Cyclic
load
testing
of
FPS
bearings
at
large
displacements
while
maintaining
a
constant
vertical
load
is
challenging,
since
the
height
of
the
bearing
increases
with
lateral
displacement.
Based
on
the
average
vertical
load,
the
normalized
stiffness
is
about
0.00769
kip/in./kip
which
corresponds
to
force
of
about
21
percent
of
the
supported
weight
27
inches
of
lateral
displacement
(i.e.,
0.21
˜
0.00769
kip/in//kip
x
27
inches).
Seismically
Isolated
Structures -101
This
slide
shows
formulas
for
calculating
maximum
and
minimum
effective
stiffness
and
effective
damping
of
the
isolation
system
at
the
design
displacement
(D).
Conceptually,
effective
stiffness
is
based
on
forces
at
the
design
displacement,
measured
by
prototype
testing,
summed
over
all
isolator
units,
and
effective
damping
is
based
on
the
hysteretic
loop
area
at
the
design
displacement,
measured
by
prototype
testing,
summed
over
all
isolators
(and
including
dampers,
if
such
are
used
as
part
of
the
isolation
system).
The
Standard
intentional
requires
conservative
values
of
forces
and
loop
areas
(e.g.,
maximum
of
3
cycles
at
a
given
load
level),
although
the
average
value
of
force
or
loop
area
is
typically
used
in
practice.
Formulas
for
maximum
and
minimum
effective
stiffness
(and
effective
damping)
only
address
variability
of
isolator
properties
measured
during
prototype
testing
and
should
be
modified
to
also
include
the
effects
of
aging
and
contamination,
etc.,
and
manufacturing
tolerances,
such
that
values
of
maximum
and
minimum
stiffness
(and
effective
damping)
used
for
design
reflect
the
full
range
of
possible
isolator
properties.
Thus,
in
the
design
of
the
example
EOC,
the
value
of
dynamic
friction
was
assumed
to
have
range
from
0.04
to
0.08,
although
the
cyclic
load
testing
showed
much
less
variability
around
the
nominal
value
of
dynamic
friction
(0.06),
as
shown
in
the
next
slide
(repeat
of
Slide
41).
Seismically
Isolated
Structures -102
This
figure
shows
modeled
and
tested
hysteresis
loops
for
the
double-
concave
FPS
bearing
at
peak
displacements
of
plus/minus
27
inches.
The
lower-bound
and
upper-bound
loop
properties
are
based
on
values
of
dynamic
friction
of
0.04
and
0.08,
respectively,
and
are
intentionally
conservative
with
respect
to
the
0.06
nominal
value
of
dynamic
friction
to
bound
potential
variation
in
properties
due
to
aging
and
environmental
effects,
manufacturing
tolerances
as
well
test
loop
variation.
Seismically
Isolated
Structures -103
The
Standard
provides
acceptance
criteria
that
ensures
that
isolators,
and
hence
the
isolation
system:
(1)
has
positive
incremental
restoring
force
capacity
–the
test
specimens
should
have
increasing
resistance
with
displacement
to
verify
that
isolation
system
will
not
accumulate
residual
displacement
in
a
given
direction
(2)
has
reliable
force-deflection
properties
–the
two
test
specimens
should
have
the
same
effective
properties
and
have
limited
variation
in
effective
stiffness
for
repeated
cycles
of
load
at
given
displacement
(3)
is
durable
–test
specimens
should
have
limited
degradation
of
their
effective
properties
for
repeated
cycles
of
load
such
that
the
isolation
system
would
still
be
functional
during
aftershocks
(4)
remains
stable
for
maximum
earthquake
loads
–the
tests
specimens
must
be
shown
capable
of
supporting
maximum
(and
minimum)
vertical
load
at
maximum
(MCER)
earthquake
displacement.
Seismically
Isolated
Structures -104
This
photograph
shows
prototype
testing
of
a
double-concave
FPS
bearing
(FPT8844/12-12/8-6)
in
large
test
machine
located
at
factory
of
Earthquake
Protection
Systems
(manufacture).
Top
concave
plate
is
displaced
approximately
two
feet
relative
to
the
bottom
concave
plate.
Articulated
slider
is
tilted
to
accommodate
the
curvatures
of
the
two
concave
plates.
Seismically
Isolated
Structures -105
This
photograph
shows
the
bottom
concave
plate
and
articulated
slider
of
the
double-concave
PFS
bearing
(FPT8844/12-12/8-6)
after
prototype
testing.
The
bearing
and
the
articulated
slider
have
been
disassembled
for
inspection
of
internal
surfaces
and
parts.
Top
and
bottom
concave
plates
and
articulated
slider
parts
are
cast
iron
with
materials
added
to
sliding
surfaces.
The
polished
surface
inside
the
bottom
concave
plate
is
a
stainless
steel
liner.
The
sliding
surfaces
of
the
articulated
slider
(black
surfaces
facing
up)
are
made
of
a
proprietary
Teflon-like
material
that
bears
on
the
stainless
steel
liners
of
the
top
and
bottom
concave
plates.
The
core
element
of
the
articulated
slider
is
the
object
in
the
center
of
the
bottom
concave
plate.
Seismically
Isolated
Structures -106
Slide
to
initiate
questions
from
the
participants.
Seismically
Isolated
Structures -107