admin
مدیر کل سایت
عضو شده در: 7 مهر 1385
پست: 10936
تشکر: 202
تشکر شده 329 بار در 224 پست
محل سکونت: همدان 
امتیاز: 61406
[وضعيت كاربر:آفلاین]
|
 تاریخ: چهارشنبه 21 مرداد 1388 - 21:29 عنوان: 13th World Conference on Earthquake Engineering |
 |
|
13th World Conference on Earthquake Engineering
Vancouver, B.C., Canada
August 1-6, 2004
Paper No. 2576
CYCLIC BEHAVIOR OF AN INNOVATIVE STEEL SHEAR WALL
SYSTEM
Qiuhong ZHAO1 and Abolhassan ASTANEH-ASL2
SUMMARY
Steel shear walls have been used more frequently in recent years as the lateral load-resisting system in the
design and retrofit of high-rise buildings. This paper concentrates on the experimental studies of an
innovative steel shear wall system used in buildings in USA and presents a summary of test results.
Steel plate shear wall system studied herein consists of steel plate shear walls placed inside a multi-bay
steel moment frame. In this system, steel walls are welded to the boundary steel moment frame. The steel
moment frame consists of very large concrete filled steel tubes (CFT) at the edges, internal wide flange
(WF) columns, and horizontal WF beams. Most of the gravity is resisted by the CFT columns, and lateral
loads are resisted by the dual system consisting of the moment frame and the steel shear wall.
Cyclic static tests were conducted on two half-scale specimens representing this system with different
span-to-height ratio for the steel wall panels. Both specimens showed highly ductile and stable inelastic
behavior, in the sense that they were able to tolerate more than 30 cycles of inelastic shear displacements
before reaching an inter-story drift more than 0.03. Inter-story drift herein is defined as lateral movement
of the floor divided by the story height. Throughout the test, the gravity load carrying CFT’s remained
essentially elastic while non-gravity carrying lateral load resisting elements underwent well-distributed
and desirable yielding. The experimental results and their implication in seismic design are summarized
and discussed.
INTRODUCTION
Reinforced concrete shear walls have been widely used as lateral load resisting system in concrete
buildings in the past, especially in high-rise buildings. In steel buildings, in most cases concrete shear
walls are used with a perimeter steel moment frame to resist seismic effects. However, tension cracks and
compression crushing failures in a concrete shear wall can result in spalling and splitting failure of the
wall, and lead to serious deterioration of stiffness and reduction of energy-dissipation capacity.
Furthermore, the casting and curing of concrete wall makes the construction not so efficient compared to
1
Doctoral Graduate Student, Dept. of Civil and Environmental Engineering, 721 Davis Hall, University of
California, Berkeley, CA 94720-1710, E-mail: qhzhao@berkeley.edu.
2
Professor, Dept. of Civil and Environmental Engineering, 781 Davis Hall, University of California,
Berkeley, CA 94720-1710, E-mail: astaneh@ce.berkeley.edu, Web: www.ce.berkeley.edu/~astaneh.
other systems such as braced frames or moment frames. In recent years, more attention has been paid to
steel shear walls that could be constructed economically and efficiently in high rise buildings. Seismic
behavior of this system and the design guidelines for them are therefore of high interest.
Steel plate shear walls have been used as the primary lateral load resisting system in several modern and
important structures in Japan and USA during the last 20 years. In Japan, stiffened steel plate shear walls
were used in new building construction since the 1970’s, and recently research has been conducted on
steel shear walls made of low strength steel and composite shear walls with “grooves” . (Astaneh-Asl [4])
In USA, stiffened shear walls were first used in seismic retrofit of existing hospitals in California.
Stiffened shear walls and shear walls with “grooves” are seldom studied and used in USA due to the high
labor cost. The focus of industry and this paper is on un-stiffened steel shear walls that were proven to be
more efficient in USA.
BACKGROUND
Steel Shear Wall Systems and an Innovative System
Three common types of steel plate shear wall systems are shown in Figure 1. In type I, the steel plate shear
wall is welded (or bolted) to the boundary elements in only one bay. The system is a “dual” system with
the moment frame and the steel shear wall in the same bay. In type II, two type I systems are connected by
the horizontal coupling beams and work together. Type III is an innovative steel shear wall system
developed and used by Magnusson Klemencic Associates. It is similar to type II, except that the edge
columns in the moment frame are very large concrete-filled steel tubes (CFT’s) instead of the usual wide
flange (WF) section. Due to the high axial stiffness of the CFT’s, most of the gravity load is carried by
them. These large CFT columns are connected to horizontal beams using special moment connections to
form special ductile moment frame.
Coupling CFT
Shear Wall
Beam
Beams and
Columns of
Simple
Seismic Special
Connection
Moment Frame
Moment
Connection
Type I Type II Type III
Figure 1. Three common types of steel shear wall systems. (Zhao [3])
The focus of steel shear wall project was the behavior of the innovative steel shear wall system Type III,
which was developed and used by Magnusson Klemencic Associates. The gravity load resisting system of
the building consisted of WF columns at perimeter and four CFT columns at the core. The lateral load
resisting system consisted of a steel shear wall system in one direction and steel braced frame system in
the other, as shown in Figure 2.
Coupling Beam
Non-Gravity Members
(for seismic loads only)
Bolted Splice
Steel Plate
Shear Wall
CFT columns
Figure 2. Lateral Load Resisting System of a Steel Building
Objectives of the Study
The main objectives of the research program summarized herein were to:
1. Collect information on actual behavior of steel shear walls and results of cyclic tests conducted on
the system;
2. Conduct analytical parametric studies on the innovative steel shear wall systems to identify key
parameters affecting seismic behavior;
3. Conduct cyclic tests on the system and establish its cyclic behavior regarding strength, stiffness,
energy dissipation and damageability characteristics;
4. Develop design guidelines and recommendations.
CYCLIC TEST ON STEEL SHEAR WALL SYSTEM
Specimens
The test program consisted of cyclic testing of two half-scale specimens. The specimens were constructed
as sub-assemblies of the prototype building over two floors (for Specimen One) and three floors (for
Specimen Two) with different span-to-height ratio for the walls, as shown in Figure 3. The test specimens
represented a dual lateral load-resisting system where steel shear wall is the “Primary” lateral load
resisting system welded to a “Back-up” system of special moment-resisting frame. Due to symmetry,
specimens only included half of the system it represented in the actual building and a roller was put at the
end of the coupling beam to simulate the boundary conditions, as shown in Figure 3.
Center Line Center Line Center Line
Specimen One Specimen Two Actual Building
Figure 3. Specimens as Sub-assemblies of Actual Building
The specimens were composed of a boundary CFT column, WF beams, interior WF column and steel plate
walls, as shown in Table 1 and Figure 4. The steel tube, WF column and beams were made of A572 Grade
50 with specified yield stress of 345 MPa (50 ksi), and the wall plate was made of A36 with specified
yield stress of 248 MPa (36 ksi). The concrete had a minimum specified f’c of 21 MPa (3 ksi). The WF
beams and columns were welded to each other and welded to the steel plate shear wall in the shop using
flux cored arc welds with E70T-7 electrodes and a specified strength of 483MPa (70 ksi). Then the steel
unit is welded to the CFT column in the field. The two specimens were fabricated according to the
specifications and details provided by the designers (Magnusson Klemencic Associates).
Table 1. Components of Test Specimens
CFT column
Specimen No. & Steel Wall Plate Beam Column
Designation Thickness Thickness Dia. Section* Section*
8 mm 610 mm
1 two-story 6 mm(1/4 inch) W18x86 W18x86
(5/16 inch) (24 inch)
8 mm 610 mm
2 three-story 10 mm( 3/8 inch) W18x86 W18x86
(5/16 inch) (24 inch)
* Properties of cross sections refer to the AISC Manuals (AISC [1]).
The WF beams were continuous through the WF column with their top and bottom flanges welded to the
column using full-penetration welds. The coupling beams were the extension of WF beams. Two sliding
load cells were added to the end of each coupling beam to simulate the roller condition. The ends of the
coupling beams were reinforced to support the load cells, as shown in Figure 4.
Each shear wall had an all-bolted splice at mid-height of story. The bolts in the steel plate splices were 16
mm (5/8 inch) diameter A490 slip-critical bolts tightened according to the AISC Specifications (AISC
[1]). The splice plates were 6 mm (¼ inch) thick A36 steel plates on both sides. The WF columns in
both specimens had also all-bolted field splice at mid-height of each story. All bolts were 22 mm (7/8
inch) diameter A490 Slip Critical bolts tightened as per AISC Specifications (AISC [1]). The material of
plates used in splices was all specified as A36 steel.
The CFT column was a 610 mm (24 inch) diameter steel tube with a thickness of 8 mm (5/16 inch) filled
with concrete. The concrete-filled tubes in the actual structures carry a large portion of the gravity load. To
represent the gravity compression, eight 38 mm (1-1/2 inch) diameter DYWIDAG W/FPU 1034 MPa (150
ksi) were placed inside the tube before casting concrete. Besides the DYWIDAG bars, there were also
shear studs welded inside the steel tube to ensure the composite action. Prior to fabrication of specimens,
the tube was cut in half, shear studs were welded and then the two halves of tubes were welded back to
each other using 5 mm (3/16 inch) partial penetration welds to make the full tube. The shear studs in
specimens were 19 mm (¾ inch) diameter by 76 mm (3 inch) length WHSS Nelson studs. The studs were
spaced in a staggered pattern with 165 mm (6.5 inch) spacing in circumferential direction and 254 mm
(10 inch) in longitudinal direction making the actual spacing of each stud from nearby studs almost 254
mm (10 inch).
The moment connections of WF beams to CFT column consisted of eight 19mm (¾ inch) diameter steel
deformed re-bars (four on each flange) which were embedded in the concrete-filled steel tube and fillet-
welded to the flanges of the WF beams, as shown in Figure 5.
Figure 4. Structural Details of Test Specimens
Figure 5. Details of Wide Flange Beam to CFT Column Connections
Test Set-up
The test set-up is shown in Figure 6. Main components of the test set-up are: Actuator, Top Loading
Beam, Bottom Reaction Beam, R/C Reaction Blocks, Bearing Support and Bracings.
The Actuator, Top Loading Beam, Bottom Reaction Beam and R/C Reaction Blocks were the same as the
components of the test set-up for the composite shear wall project (Zhao [5]). Cyclic shear forces were
applied by the only actuator to the top of the specimen, therefore all the stories in the specimen had the
same story shear, which represented the shear distribution in the prototype high-rise building.
In order to simulate bracing effects provided to the shear wall system by the story floors in actual
buildings, two types of bracings were used in the specimen. A point bracing was used on the top Loading
beam to brace the middle of this beam and prevent out of plane movement. A set of braces was also used
to brace the end of coupling beams and to prevent out of plane movements. The bearing support was
designed to support the ends of coupling beams with load cells and transfer the reactions to laboratory
floor.
OUT-OF-PLANE
BRACING
Specimen One Specimen Two
Figure 6. Test Set-up and Components
Loading History
After 1994 Northridge earthquake, the SAC Joint Venture developed a loading history and test
procedures. The loading history developed by SAC Joint Venture and included in the Protocol for
Fabrication, Inspection, Testing, and Documentation of Beam-Column Connection Tests and Other
Experimental Specimens (SAC [2]) was later included in the AISC Seismic Provisions (AISC [1]).
Due to differences between a beam-column connection and steel shear wall system, the SAC protocol
could not be used directly. However, in developing loading history an attempt was made to be consistent
with the SAC Protocol. The specimens were tested under cyclic displacement regime. Cyclic
displacements applied to both specimens were the same as shown in Figure 7. The overall drift is defined
as actuator displacement divided by the total height of the specimen 6.19m (20 ft - 4 inch).
0.03
0.02 Inelastic
0.01 Elastic
Overall 0
Drift
-0.01
0.006-0.007
-0.02
Specimens Yield 0.032
-0.03 End of Test
0 10 20 30 40 50 60 70 80
Cycles
Figure 7. Loading History Applied to Both Specimens
Instrumentation
A number of Linear Variable Displacement Transducers (LVDT’s) were installed on the specimens to
measure the global as well as local displacement of critical points such as intersection points of member
centerlines, the corner points of steel panels, the corner points of panel zones, etc. In this way movement
of the members were monitored and important data such as panel zone deformation, shear wall panel
deformation, slippage in the bolted splices and movement of the top and bottom flanges of a beam in a
moment connection were measured. For safety purposes, a series of displacement transducers were
mounted on the test set-up to detect any slippage of the actuator or the reaction blocks.
In order to measure strains at various critical locations, linear as well as “rosette” strain gauges were
mounted on the specimen. Critical locations included quarter points on a cross section of the steel tube,
flange and web of the column, middle point of the steel panel, quarter point of the steel panel, etc.
In order to measure reaction forces at the end of the coupling beams special load cells were designed and
installed. The load cells consisted of solid steel cylinders with semi-circular ends to act as rollers. The
cylinder portion was strain gauged and calibrated prior to tests.
CYCLIC BEHAVIOR OF STEEL SHEAR WALL SPECIMENS
Steel Shear Wall Specimen One with Two Stories
Specimen One behaved in a very ductile and desirable manner. Up to overall drift of about 0.006, the
specimen was almost elastic. At this drift level some yield lines appeared on the wall plate as well as WF
column (non-gravity column). Up to overall drift of about 0.022, the compression diagonal in the wall
panels was buckling and the diagonal tension field was yielding. At this level, the WF column developed
local buckling. The specimen could tolerate 79 cycles, out of which 35 cycles were inelastic, before
reaching an inter-story drift of 0.032 and maximum shear strength of about 4079 kN (917 kips). Notice
here the inter-story drift value equals to the lateral displacement of a floor divided by the story height,
which is different from the overall drift value used in loading history. At this drift level, the upper floor
coupling beam fractured at the face of the column (due to low-cycle fatigue) and the shear strength of the
specimen dropped to below 75% of the maximum capacity. The specimen was then considered failed.
Behavior of Specimen One during cyclic testing is summarized in Table 2. Figure 8 shows Specimen One
at various stages of cyclic loading.
Table 2. Key Test Observations on Behavior of Specimen One
Loading Actuator
Description
Groups Drift (Rad)
0 0.0002 Very small warm-up cycles were applied to check test set-up & Instruments.
1 0.00075 Actual test started. Specimen remained elastic.
2 0.001 Specimen remained elastic.
3 0.0015 Bolts on reaction blocks made noise. Specimen still appeared elastic.
4 0.002 Big bang noise from reaction beam. Specimen remained elastic.
5 0.0025 Continuous bang and squeaking noise. Specimen remained elastic.
6 0.003 Bang and squeaking noise from time to time. Specimen remained elastic.
7 0.0035 Bang and squeaking noise from time to time. Specimen remained elastic.
8 0.004 Some minor non-linearity was observed on the force-disp. curves.
9 0.0045 Bolt slipped on base plate connection. No sign of non-linearity on the curves.
10 0.005 Hysteretic curves were getting slightly fatter, sign of non-linearity.
11 0.0055 Middle panel buckled in a second mode shape. Tension field formed. No major
yielding. “Proportional Limit Point”.
12 0.006 Entire middle panel yielded. Buckling mode changed. Force-disp. curve was still
almost elastic. Slight yielding at column base. “Significant Yield Point”.
13 0.0065 Yielding in the middle panel continued. Buckling changed to first mode with
compression diagonal buckling from lower beam to upper beam.
14 0.007 Dropping of load at peak point indicated plate buckling. Column splice slipped
slightly. Yielding at top beam to CFT column connection.
15 0.008 Permanent out-of-plane deformation of middle panel.
16 0.009 Yielding was visible in areas around panel zone. Yielding of top flange of
bottom beam occurred between rebar to CFT column connections.
17 0.01 First and second buckling modes interacted in the middle panel. Two kinks
formed in the right half of the panel.
18 0.011 More yielding occurred on steel shear wall panels and ends of WF column.
19 0.012 Fracture and more kinks in the middle panel. Clear bending of column.
20 0.014 Second fracture in the middle panel. More yielding at the end of top coupling
beam and panel zones. Web of top beam yielded near the CFT column.
21 0.016 Local buckling at the base of WF column above the bottom beam. Epoxy
fracture was observed showing yielding of rebar inside the CFT column.
22 0.018 WF column had permanent deformation and yielded heavily inside middle
panel. Two more fractures and big out-of-plane bump formed in the middle
panel.
23 0.02 Very obvious yielding at the WF column base inside the middle panel. Slight
yielding of WF column around the base cover plate.
24 0.022 Plastic hinge formed and flange local buckled at the top coupling beam – WF
column connection. WF column flange yielded at the ends inside middle panel.
25 0.024 Rebar’s epoxy out at the connections of top and bottom beams to CFT column.
Twisting at the locally buckled part of WF column above the bottom panel zone.
26 0.026 WF column flange locally buckled below the top beam panel zone. Rebar
fractured at the bottom beam to CFT column connection. Corner of middle
panel fractured in diagonal direction.
27 0.028 Top coupling beam fractured at the face of WF column. WF column base
fractured at locally buckled area. “Point of Maximum Shear Strength”.
28 0.030 Flange of WF column fractured below the top beam. Fracture of the WF column
base inside the middle panel propagated into the already buckled web. The top
coupling beam was totally separated from the horizontal WF beam. Load
dropped to about 60% of maximum. “Point of Maximum Ductility”.
29 0.032 Rebar yielded at the connection of top beam to CFT column. Small fracture
occurred at “kink” locations of the top steel plate panel. Test stopped.
Specimen One Yielding
Specimen One at End of
Top Coupling Beam
Figure 8. Specimen One at Different Stages of Cyclic Test
Steel Shear Wall Specimen Two with Three Stories
Specimen Two also behaved in a ductile and desirable manner. Up to overall drift of about 0.007, the
specimen was almost elastic. At this drift level some yield lines appeared on the wall plate and the force-
displacement curve started to deviate from the straight elastic line. During later cycles a distinct X-shaped
yield line was visible on the steel plate shear walls. The specimen could tolerate 79 cycles, out of which
30 cycles were inelastic. The specimen reached maximum shear force of 5449 kN (1,225 kips) under an
overall drift of 0.022. In the next drift level, the top (fourth story) coupling beam fractured at the face of
the column (due to low-cycle fatigue). At an overall drift of 0.032, the CFT column fractured at the base
and the load dropped below 75% of maximum strength, then the specimen was considered failed and the
test stopped. Behavior of Specimen Two is summarized in Table 3. Figure 9 shows Specimen Two at end
of cyclic loading.
Table 3. Key Test Observations on Behavior of Specimen Two
Loading Actuator
Description
Groups Drift (Rad)
0 0.0002 Very small warm-up cycle was applied to check test set-up & Instruments
0 0.0004 Very small warm-up cycle was applied to check test set-up & Instruments
1 0.00075 Actual test started. Specimen was elastic.
2 0.001 Specimen remained elastic.
3 0.0015 Specimen remained elastic.
4 0.002 Specimen remained elastic.
5 0.0025 Noise of friction. Specimen remained elastic.
6 0.003 Specimen remained elastic.
7 0.0035 Specimen remained elastic.
8 0.004 Specimen was still elastic.
9 0.0045 Specimen was still elastic.
10 0.005 Loud squeaking noise from bracing and load cell. Specimen remained elastic.
11 0.0055 Big bang noise. Specimen remained elastic.
12 0.006 Very minor vertical yield lines in members.
13 0.0065 Top,middle coupling beam flange and web yielded. “Proportional Limit Point”.
14 0.007 The second story panel formed a tension field and diagonal yield lines were
observed across the splices. “Significant Yield Point”.
15 0.008 More yielding in the top and middle coupling beam, and third story wall panel.
16 0.009 Some tension field action shown in third story panel.
17 0.01 In the third story panel both shear yielding mode and buckling mode were
clearly observed. In the second story panel buckling mode was more obvious.
18 0.011 WF column in the second story showed much yielding.
19 0.012 Very obvious X shaped tension field was formed in the middle wall panels.
Some yielding in the middle horizontal beam web.
20 0.014 Very obvious buckling of two middle wall panels and distortion of wall splices.
Some yielding in the bottom horizontal beam.
21 0.016 Kinks formed in the lower half of the middle wall panels.Change of vibration
mode from first to second. WF column and middle panel zone heavily yielded.
22 0.018 Rebars at bottom and middle beam to CFT column connections had been
elongated and epoxy was pulled out.
23 0.02 X shaped valley was formed on the middle panels. Kinks formed at quarter
points of the valley. Flange and web of top coupling beam buckled slightly.
24 0.022 Obvious buckling in the first floor wall panel. Middle coupling beam had web and
flange buckling near panel zone. Very slight web buckling at the WF column
base in the second story. “Point of Maximum Shear Strength”.
25 0.024 Top coupling beam fractured at the face of WF column, extending from top
flange into web. Middle coupling beam has flange and web local buckling near
panel zone. WF Column of second story had flange local buckling at the ends.
26 0.026 Base of the column showed slight flange local buckling. Wall splice in the third
story slipped. More yielding of splice in the WF column on second story.
27 0.028 Top coupling beam totally separated. Kinks developed into fracture in the wall.
Very obvious web and flange local buckling at the WF column base on the
second and first story. Heavy yielding on the top horizontal beam.
28 0.030 CFT column fractured at the base and buckled.
29 0.032 First story WF column flange totally yielded and locally buckled. Load dropped
to about 75% of maximum. “Point of Maximum Ductility”. T |
|
