Self-Centering Steel Frame
Systems
NEESR-SG: Self-Centering Damage-Free SeismicResistant Steel Frame Systems
Project Team
Richard Sause, James Ricles, David Roke,
Choung-Yeol Seo, Michael Wolski, Geoff Madrazo
ATLSS Center, Lehigh University
Maria Garlock, Erik VanMarcke, Li-Shiuan Peh
Princeton University
Judy Liu
Purdue University
Keh-Chyuan Tsai
NCREE, National Taiwan University
Current NEESR Project
NEESR-SG: Self-Centering
Damage-Free Seismic-Resistant
Steel Frame Systems
This material is based on work supported by the
National Science Foundation, Award No. CMS0420974, in the George E. Brown, Jr. Network for
Earthquake Engineering Simulation Research
(NEESR) program, and Award No. CMS-0402490
NEES Consortium Operation.
Motivation: Expected Damage
for Conventional Steel Frames
(a)
(b)
Conventional
Moment
Resisting
Frame
System
Reduced beam section (RBS) beam-column test
specimen with slab: (a) at 3% drift, (b) at 4% drift.
Self Centering (SC) SeismicResistant System Concepts
 Discrete structural members are
post-tensioned to pre-compress
joints.
 Gap opening at joints at selected
earthquake load levels provides
softening of lateral force-drift
behavior without damage to
members.
 PT forces close joints and
permanent lateral drift is avoided.
M
Previous Work on SC Steel Moment
Resisting (MRF) Connections
W24x62 (Fy 248 MPa),
or W36x150 (Fy 350 MPa)
W14x311, W14x398 (Fy 350 MPa)
or CFT406x406x13 (Fy 345 MPa)
PT Strands
3660 to
4120 mm
MRF Subassembly
with PT Connections
6100 to 8540 mm
Initial Stiffness Is Similar to
Stiffness of Conventional Systems
PT Steel MRF
500

400
Lateral Load - H (KN)
H
Stiffness with
welded connection
FR
300
200
100
0
MRF subassembly
with post-tensioned
connections
-100
-200
-300
-400
-500
-150
-100
-50
0
50
Lateral Displacement -  (mm)
100
150
Lateral Force-Drift Behavior
Softens Due to Gap Opening
Steel MRF subassembly with
post-tensioned connections
and angles at 3% drift
Lateral Force-Drift Behavior Softens
Without Significant Damage
• SC steel MRF
softens by gap
opening and
reduced contact
area at joints
Welded Connection
600

H
400
Lateral Load, H (kips)
• Conventional steel
MRFs soften by
inelastic
deformation, which
damages main
structural members
and results in
residual drift
200
0
-200
PostTensioned
Connection
-400
-600
-8
-6
-4
-2
0
2
Displacement,  (in)
4
6
8
Energy Dissipation from Energy
Dissipation (ED) Elements
Steel MRF

400
H
Lateral Load, H [KN]
Steel MRF
subassemblies
with posttensioned
connections with
different size ED
elements.
300
Specimen PC2
L6x6x5/16, g/t = 4
200
100
0
-100
Specimen PC4
L8x8x5/8, g/t = 4
-200
-300
-400
-150
-100
-50
0
50
100
Lateral Displacement,  [mm]
150
Limited, Repairable Damage
Angle fracture
Before testing
@ 4% Drift
After testing
Summary of SC Seismic-Resistant
Structural System Behavior
• Initial lateral stiffness is similar to that of
conventional seismic-resistant systems.
• Lateral force-drift behavior softens due to gap
opening at selected joints and without significant
damage to main structural members.
• Lateral force-drift behavior softening due to gap
opening controls force demands.
• Energy dissipation provided by energy
dissipation (ED) elements, not from damage to
main structural members.
NEESR-SG: Self-Centering
Damage-Free SeismicResistant Steel Frame Systems
•
•
•
•
Project Scope.
Project Goals.
Status of Selected Research Tasks.
Summary.
NEESR-SG: SC Steel Frame
Systems Project Scope
• Develop two SC steel frame systems:
– Moment-resisting frames (SC-MRFs).
– Concentrically-braced frames (SC-CBFs).
• Conduct large-scale experiments utilizing:
– NEES ES (RTMD facility) at Lehigh.
– non-NEES laboratory (Purdue).
– international collaborating laboratory (NCREE)
• Conduct analytical and design studies of
prototype buildings.
• Develop design criteria and design procedures.
NEESR-SG: SC Steel Frame
Systems Project Goals
• Overall: self-centering steel systems that are
constructible, economical, and structurally
damage-free under design earthquake.
• Specific:
– Fundamental knowledge of seismic behavior of SCMRF systems and SC-CBF systems.
– Integrated design, analysis, and experimental
research using NEES facilities.
– Performance-based, reliability-based seismic design
procedures.
NEESR-SG: Self-Centering
Damage-Free SeismicResistant Steel Frame Systems
•
•
•
•
Project Scope.
Project Goals.
Status of Selected Research Tasks.
Summary.
NEESR-SG: SC Steel Frame
Systems Project Research Tasks
1.
2.
3.
4.
5.
6.
7.
8.
9.
Develop reliability-based seismic design and assessment procedures.
Develop SC-CBF systems.
Further develop SC-MRF systems.
Develop energy dissipation elements for SC-MRFs and SC-CBFs.
Develop sensor networks for damage monitoring and integrity
assessment.
Design prototype buildings.
Perform nonlinear analyses of prototype buildings.
Conduct large-scale laboratory tests of SC-MRFs and SC-CBFs.
Collaborate on 3-D large-scale laboratory tests on SC-MRF and SCCBF systems.
Task 2. Develop SC-CBF Systems:
SC-CBF System Concept
Rocking
behavior
of simple
SC-CBF
system.
More Complex SC-CBF
Configurations Being Considered
P01+P
g
V
P02 P02+P
roof
P01
g
g
g
g
g
g
g
g
g
g
g
base
col
V
base
SC-CBF Design Criteria
Lateral Force
MCE
DBE
Failure of
Frame
Members
PT Yielding
Δgap
Column
Decompression
PT steelyields
yields
Member
IO
LS
Significant
Yielding of
Frame
Members
CP
Roof Drift
Current Work on SC-CBF Systems
• Evaluate frame configurations.
• Evaluate effect of energy dissipation
(ED) elements.
• Develop and evaluate performancebased design approach.
SC-CBF Configurations Studied
PT
PT
PT
ED
ED
Frame A
Frame B12
Frame B12ED
Dynamic Analysis Results (DBE)
– Effect of
frame
configuration.
Drift (%)
• Roof drift:
3.0
2.0
1.0
0.0
-1.0
-2.0
-3.0
2.7%
0
5
10
15
20
A
B12
25
30
35
40
Time (s)
1.4%
2.7%
2.0
Drift (%)
– Effect of ED
elements.
3.0
1.0
0.0
-1.0
B12
-2.0
2.4%
-3.0
0
5
10
15
20
Time (s)
B12ED
25
30
35
40
Pushover Analysis Results
PT Yield
Base Shear (V/Vdes)
2.5
V/Vdes
2.0
1.5
1.0
0.5
A
B24
B12
B12ED
Decompression
0.0
0.0
0.5
1.0
1.5
2.0
2.5
Drift (% )
3.0
Roof Drift (%)
3.5
4.0
4.5
5.0
Preliminary Results for SC-CBF
• Dynamic analysis results indicate selfcentering behavior is achieved under DBE.
• Frame A has lower drift capacity before PT
yielding than Frame B:
– PT steel is at column lines rather than mid-bay.
• Frame A also has lower drift demand.
• Energy dissipation helps to reduce drift
demand and improve response.
Task 3. Further Develop SC-MRF
Systems: Current Work
• Study of interaction between SC-MRFs
and floor diaphragms by Princeton and
Purdue.
• SC column base connections for SCMRFs being studied by Purdue.
Interaction of SC-MRFs and
Floor Diaphragms (Princeton)
Approach 1.
Transmit
inertial forces 2gap
from floor
diaphragm
without
excessive
restraint of
connection
regions using
flexible
collectors.
gap
r
gap
Collector Beams
2gap
Interaction of SC-MRFs and
Floor Diaphragms (Purdue)
Approach 2.
Transmit inertial
forces from floor
diaphragm within
one (composite)
bay for each
frame.
SC Column Base Connections
for SC-MRFs (Purdue)
Post-Tensioned Bars
Reinforcing Plate
Energy Dissipation Plate
Slotted Keeper Angle
Beam at Grade
Moment-Rotation Response at
Column Base
Identifying appropriate level of column base moment capacity
and connection details, leading to laboratory experiments.
Task 4. Develop Energy
Dissipation Elements for SC-MRFs
• SC systems have no significant energy
dissipation from main structural elements:
– Behavior of energy dissipation elements
determined SC system energy dissipation.
• Energy dissipation elements may be
damaged during earthquake and replaced.
• For SC-MRFs, energy dissipation elements
are located at beam-column connections.
Quantifying Energy Dissipation
• Define relative hysteretic ED ratio bE
bE : Relative ED capacity
bE =
Area of yellow
Area of blue
x 100(%)
For SC systems: 0 ≤ bE ≤ 50%
Target value: bE = 25%
Hysteresis Loop
ED Element Assessment
• Consider several ED elements:
– Metallic yielding, friction, viscoelastic,
elastomeric, and viscous fluid.
• Evaluation criteria:
– Behavior, force capacity versus size,
constructability, and life-cycle maintenance.
• Friction ED elements selected for further
study.
Bottom Flange Friction Device
(a)
Anchorage
BFFD
PT
strands
Friction PL
Column
Beam
Friction bolts
Column angle
Col. angle bolts
(b)
A
Slotted keeper angle
Reinforcing plate
PT strands
Slotted plate
W36x300
Shim Plates
Friction bolts
with Belleville
washers
Reinforcing plate
Built-up angles
Slotted plate welded to
beam bottom flange
Brass friction plate at A
slotted plate/angle interface
Section A-A
Built-up
angles
bolted to
column
BFFD Moment Contribution
• BFFD contribution to connection moment capacity
r
COR+
r+
r
MFf+
MFf-
COR-
rFf
MFf = Ff∙r
Ff
MFf+ = Ff ∙r+
MFf- = Ff ∙r -
|MFf+ | > |MFf- |
Test Setup
W21x111
Beam
Test specimen subassembly
V
Clevis plates
Hydraulic actuator
12’-2 ½”
PT strands
Reaction wall
W21x111 Beam
PT Strands
– Denotes lateral bracing
BFFD
Strong floor
Column
N
S
3/5 Scale
Test setup elevation
Column
BFFD
Test setup
Test Matrix
Test Loading r,max
No. Protocol (rads)
1
2
3
4
5
6
7
CS
CS
CS
EQ
CS
CS
CS
0.035
0.030
0.030
0.025
0.065
0.035
0.065
CS: Cyclic Symmetric
Experimental
Parameter
Reduced Friction Force
Design Friction Force
Fillet Weld Repair
Response to EQ Loading
Effect of Bolt Bearing
Assess Column L Flex., CJP
Effect of Bolt Bearing, CJP
EQ: Chi-Chi MCE Level Earthquake Response
Test 2: Design Friction Force
Beginning of Test 2
r = +0.03 rads
r = -0.03 rads
Test 2: Response
Normalized Moment, M/Mp,n
1.0
r-
M
-
M+
r+
0.5
imminent gap opening
stiffness
reduction
theoretical decompression
0.0
-0.5
stiffness
reduction
-1.0
-0.04
theoretical decompression
imminent gap opening
-0.03
-0.02
-0.01
0.00
0.01
Rotation, r (rads)
0.02
0.03
0.04
Test 2: Comparison with
Simplified Model
Configuration 2, Test C
(b) Test 2
Normalized Moment, M/Mp,n
1.0
r-
M-
r+
Axial stiffness of PT
strands & beam
M+
1
0.5
2M+Ff
Md + M+Ff
model
0.0
experiment
-0.5
2M-Ff
Md + M-Ff
-1.0
-0.04
-0.02
0.00
Rotation, r (rads)
0.02
0.04
Results for ED Elements for
SC-MRFs
• Friction ED element:
– Reliable with repeatable and predictable behavior.
– Large force capacity in modest size.
• BFFD:
– Provides needed energy dissipation for SC-MRF
connections.
– When anticipated connection rotation demand is
exceeded, friction bolts can be designed to fail in
shear without damage to other components.
Task 8. Conduct Large-Scale
Laboratory Tests
• Two specimens, one SC-MRF and one SC-CBF,
tested at Lehigh NEES ES (RTMD facility).
• 2/3-scale 4 story frame.
• Utilize hybrid test method (pseudo dynamic with
analytical and laboratory substructures).
• Utilize real-time hybrid test method, if energy
dissipation elements are rate-sensitive.
9. Collaborate on 3-D LargeScale Laboratory Tests
• Large-scale 3-D SC steel frame system tests at
NCREE in Taiwan under direction of Dr. K.C. Tsai.
• Interaction of SC frame systems with floor
diaphragms and gravity frames will be studied.
• 3-D tests are part of Taiwan program on SC
systems.
• Project team is collaborating with Taiwan
researchers:
– US-Taiwan Workshop on Self-Centering Structural
Systems, June 6-7, 2005, at NCREE.
– 2nd workshop planned for October 2006 at NCREE.
Summary
• Two types of SC steel frame systems are being
developed:
– Moment-resisting frames (SC-MRFs).
– Concentrically-braced frames (SC-CBFs).
• Research plan includes 9 major tasks:
– Significant work completed on 7 tasks.
– Numerous conference publications available from
current project.
• Large-scale experiments utilizing NEES ES at
Lehigh are being conducted.
• Ongoing collaboration with NCREE in Taiwan.
Self-Centering Steel Frame
Systems
Thank you.