Download (PDF)

10NCEE
Tenth U.S. National Conference on Earthquake Engineering
Frontiers of Earthquake Engineering
July 21-25, 2014
Anchorage, Alaska
SEISMIC PERFORMANCE OF
INTERMEDIATE MOMENT FRAMES WITH
RBS-B CONNECTIONS
S. W. Han1, K. H. Moon2 and B. Stojadinovic3
ABSTRACT
The connections of special moment frames (SMF) should have a drift capacity greater than 4%,
which is the minimum drift angle required by ANSI/AISC 358-05 for SMF connections. Since
some reduced beam section with a bolted web (RBS-B) connection specimens failed to achieve 4%
total rotation capacity, RBS-B connection is permitted for use only in Intermediate Moment
Frames (IMF) according to the ANSI/AISC 358-05. In previous studies, some RBS-B
connections could experience brittle connection fracture during earthquakes, which can also be
detrimental to the seismic performance of IMF systems with RBS-B connections. The purpose of
this study is to investigate whether IMFs with RBS-B connections provide a satisfactory seismic
performance. The seismic performance of IMFs with pre-qualified RBS-B connections is
evaluated based on the ATC-63 procedure. It is found that several IMFs with RBS-B connections
do not satisfy the acceptance criteria specified in ATC 63.
1
Professor, Dept. of Architectural Engineering, Hanyang University, Seoul 133-791, Korea
Post-Doc. Dept. of Architectural Engineering, Hanyang University, Seoul 133-791, Korea
3
Professor, Structural Dynamics and Earthquake Engineering, Swiss Federal Institute of Technology (ETH) Zurich,
Switzerland
2
10NCEE
Tenth U.S. National Conference on Earthquake Engineering
Frontiers of Earthquake Engineering
July 21-25, 2014
Anchorage, Alaska
Seismic Performance of Intermediate Moment Frames with RBS-B
Connections
S. W. Han1, K. H. Moon2 and B. Stojadinovic3
ABSTRACT
The connections of special moment frames (SMF) should have a drift capacity greater than 4%,
which is the minimum drift angle required by ANSI/AISC 358-05 for SMF connections. Since
some reduced beam section with a bolted web (RBS-B) connection specimens failed to achieve
4% total rotation capacity, RBS-B connection is permitted for use only in Intermediate Moment
Frames (IMF) according to the ANSI/AISC 358-05. In previous studies, some RBS-B connections
could experience brittle connection fracture during earthquakes, which can also be detrimental to
the seismic performance of IMF systems with RBS-B connections. The purpose of this study is to
investigate whether IMFs with RBS-B connections provide a satisfactory seismic performance.
The seismic performance of IMFs with pre-qualified RBS-B connections is evaluated based on the
ATC-63 procedure. It is found that several IMFs with RBS-B connections do not satisfy the
acceptance criteria specified in ATC 63.
Introduction
Reduced beam section welded web (RBS-W) connections is one of the prequalified connection
for Special Moment Frames specified in AISC 358 [1], which should have a total drift capacity
greater than 0.04 radian. Unlike RBS-W connections, reduced beam section bolted web (RBS-B)
connections have a high incidence of brittle beam flange weld fracture [2]. AISC 358 [1] permits
the RBS-B connection only as a prequalified moment connection for Intermediate Moment
Frames (IMF). Han et al. [3] investigated the cause of fracture in RBS-B connections designed
according to FEMA 350 [4] and found that the connection moment strength equation specified in
FEMA 350 [4] overestimates the actual moment strength of RBS-B connections, leading to
connection failure before the plastic moment capacity is reached at reduced beam sections. Han
et al. [5] reported that fractured RBS-B connections might provide only limited rotation
capacities.
When a moment frame experiences brittle connection fracture, the seismic response
demands on the remaining frame element increase substantially [6], increasing the probability of
undesirable seismic response. This study investigates whether IMF systems with RBS-B
connections provide satisfactory seismic performance using a seismic performance evaluation
according to ATC-63 [7]. Three 3-, 9-, and 20-story model buildings with RBS-B connections
are designed in accordance with current design code requirements [1, 8].
1
Professor, Dept. of Architectural Engineering, Hanyang University, Seoul 133-791, Korea
GPost-Doc., Dept. of Architectural Engineering, Hanyang University, Seoul 133-791, Korea
3
Professor, Structural Dynamics and Earthquake Engineering, Swiss Federal Institute of Technology (ETH) Zurich,
Switzerland
2
Prediction of Fracture in RBS-B Connections
RBS-B connections (Fig. 1) should be designed to ensure that the connection moment strength
(Mn) is larger than the probable maximum moment at the column face (Mf-pr) transferred from the
RBS beam when the RBS reaches its maximum probable moment [1]. The connection moment
strength is specified as follows [1]:
Mn = ΦbCprRyFyZb
(1)
where Φb is a resistance factor. For RBS, Φb is taken as 1.0. But some of test results of
RBS-B connections experiment showed unexpected early connections fractures before the
moment (Mf-pr) at connection reach the connection moment strength (Mn) proposed in
ANSI/AISC 358 [1]. Figure 1 shows the ratio for 18 fracture-resistant and 12 fracture-prone
RBS-B connection specimens [9]. The fracture-prone connection is defined as a connection
experiencing fracture near the beam flange groove welds whereas the fracture-resistant
connection fails in ways other than fracture near the beam flange groove welds. As shown in Fig.
1 (b) shows that only six of twelve specimens have a ratio greater than 1, which indicates Eq. (1)
may overestimate the moment strength (Mn) of some RBS-B connections. Therefore, for
fracture-prone RBS-B connections, Eq. (1) do not indicate the occurrence of fracture with
acceptable confidence. The overestimate of the moment strength (Mn) can be attributed to bolt
slip at the shear tab of the RBS-B connections.
M f − pr
b
Mn
M f − pr < M n
Lph = a + b 2
1.2
V
M pr
Lb 2
C
L
(a) fracture-resistant RBS-B specimens
1.3
1.2
(b) fracture-prone RBS-B specimens
Mn(Han et al.) 1.1
Mn(AISC-358)
1.1
Mf-pr /Mn
c
a
1.3
4c 2 + b 2
8c
Mf-pr /Mn
Radius =
1
0.9
1
0.9
0.8
0.8
0.7
0.7
0.6
0.65
0.6
0.7
Zf /Zb
0.75
0.8 0.65
Mn(Han et al.)
Mn(AISC-358)
0.7
Zf /Zb
0.75
0.8
Figure 1. RBS connection and Mf-pr / Mn values: Mn is computed using the equation specified in
AISC 358 [1], and using the equation proposed by Han et al. [3]
To more accurately predict the connection fracture, Han et al. [3] proposed an empirical equation
for computing the moment strength (Mn) for RBS-B connections. Detailed information about
computing connection moment strength is summarized in Han et al. [3]. Figure 1 also show Mfpr/Mn calculated by the moment strength (Mn) using Han et al. [3] equation for the fractureresistant and fracture-prone specimens, respectively. As shown in Fig. 1(a), Mf-pr/Mn is less than
1 for all 19 fracture-resistant connection specimens, which indicates that Mf-pr/Mn perfectly
predict no occurrence of connection fracture for the fracture-resistant specimens. As seen in Fig.
1(b), all fracture-prone specimens have Mf-pr/Mn values greater than 1 except for one specimen
(DBT-2B). Note that the ratio for specimen DBT-2B is very close to 1 (=0.99). In order to
predict the incidence of connection fracture, we use connection moment strength (Mn) as
proposed by Han et al. [3].
Analytic Model
An analytical model for RBS-B connections is proposed based on the M2 model [10] which can
reflect clear lengths for beams and columns with explicit modeling of the panel zone. Columns
are modeled using a fiber section model consisting of steel material with a strain hardening ratio
of 3%. To represent the hysteretic behavior of the panel zone, two spring elements are placed to
simulate the tri-linear force-deformation relationship including the contribution of post-yield
stiffness and strength of column flanges after the column web yields. The RBS is modeled with
an inelastic rotational spring element [11], placed at the center of the RBS to simulate the
strength and the stiffness deterioration using a tri-linear backbone curve. For the beam-column
connection, a fracture spring is installed to fail when the moment demand at the connection
reaches the connection moment strength [3]. After the connection fracture occurs, the residual
strength of the connection spring is assumed as zero, which represents the test results conducted
by Lee et al. [12].
250
600
150
Beam tip load (kN)
Beam tip load (kN)
200
800
(a) TRS2A specimen
100
50
0
-50
-100
-150
Analysis
Experiment
-200
-250
-0.08 -0.06 -0.04 -0.02
0
0.02 0.04 0.06 0.08
(b) DB700-SB specimen
400
200
0
-200
-400
-600
-800
-0.08 -0.06 -0.04 -0.02
0
0.02 0.04 0.06 0.08
Total rotaion angle [rad]
Total rotaion angle (rad)
Figure 2. Verification for (a) fracture-resistant and (b) fracture-prone RBS-B connection
specimens
In order to verify the accuracy of the proposed analytical model, fracture-resistant
specimen [13] and fracture–prone specimen [12] are examined. The hysteretic curves recorded
during the tests of those specimens are plotted in Fig. 2 (a) and (b). To simulate those hysteretic
curves, nonlinear static analyses are conducted using Opensees software [14] with the proposed
analytical model. Figure 2 shows actual and simulated hysteretic curves for the fracture-resistant
and fracture-prone connection specimens. It is evident that the proposed analytical connection
model satisfactorily predicts the actual hysteretic curves of the RBS-B connections including
their plastic rotation capacity and the occurrence of connection fracture. The intermediate
moment frame is modeled as 2-dimensional frames using the nonlinear analysis software
OpenSees [14]. The P- effect is included.
Seismic Performance Evaluation
ATC-63 [7] provided a methodology for quantifying system performance and response
parameters for use in seismic design. The two performance objectives used in ATC-63 [7] are:
(1) The probability of collapse for the maximum considered earthquake (MCE) ground motions
is recommended to be 10%, or less on average across a performance group that contains model
frames having a specific seismic force resisting system with different configurations, and (2) For
an individual model frame, the probability of collapse is 20%, or less. To achieve those
objectives, the adjusted collapse margin ratio (ACMR) is estimated, and compared with limiting
values of ACMR. A step-by-step procedure to determine the probability of collapse and the
adjusted collapse margin ratio for the performance group and individual model frame is
summarized in ATC-63 [7].
This study adopts the seismic performance evaluation procedure prescribed by ATC-63
(2009). Four performance groups comprising model frames with varying number of stories, bay
widths, and different seismic design categories (SDC) are considered. The seismic force resisting
system for the model frame is the IMF with RBS-B connections. Frames with 3-, 9- and 20-story
are considered. The bay widths considered are 6.0 m (20 ft) and 9.1 m (30 ft). Two seismic
design categories are considered, the lower and upper bounds of Seismic Design Category C
(SDC Cmin and SDC Cmax). Note that the SDC C building height is not limited by ASCE/SEI 7
[15] for IMF systems
All model buildings have a symmetric floor plan and the perimeter frames are the seismic
load bearing IMFs. The bay of 3-story, 9-story and 20-story are 4x6, 5x5, 5x6, respectively. All
story heights for model building are 4m except first story height of 9-story and 20-story. The first
story height of 9-story and 20-story is 5.5m. The buildings are assumed to be standard office
buildings (occupancy category II) located on sites classified as site class D. The gravity dead and
live loads for design are 4.12 kPa and 0.96 kPa, respectively. The basic wind speeds for wind
loads is assumed as 51 m/s, which can be applied for the buildings (occupancy category II)
located on most regions in the U.S. The IMFs are designed according to ASCE/SEI 7 [15] and
ANSI/AISC-341 [8], and drift requirements [15]. RBS connections are designed according to the
procedures specified in ANSI/AISC 358 [1].
To conduct the incremental dynamic analysis, 44 far-field ground motions, obtained by
ATC-63 [7] from the PEER NGA database considering earthquake magnitudes ranging from 6.5
to 7.6 and site classes C and D are used. Seismic performance evaluation is conducted for the 12
model frames shown in Table 1 using the procedure prescribed in ATC-63 [7]. Six of the 12
model frames and two of the four performance groups fail to meet the ATC-63 acceptance
criteria. The failed six model frames are 9CMIN-6.0, 20CMIN-6.0 and 20CMIN-9.1 designed
for SDC Cmin and 9CMAX-6.0, 20CMAX-6.0 and 20CMAX-9.1 designed for SDC Cmax. The
failed performance groups are Group 1 and Group 2 that contain model frames with a bay length
of 6 m designed for SDC Cmin and Cmax.
In order to investigate the effects of the design variables, namely, the number of stories,
the bay size, and the SDC, on the probability of collapse subjected to maximum considered
earthquake, the probability of collapse [P(collapse|SMT) = P(SCT < SMT)] for each model frame.
Figure 3 shows the probability of collapse with respect to different design variables. The
probability of collapse becomes higher with an increase in the number of stories. The median
collapse probability for 20-story model frames is 0.249, whereas that for 3-story model frames is
0.019. The probability of collapse for all 3-story model frames is less than 0.200, which is the
acceptance probability for individual model frames specified by ATC-63 [7]. No 20-story model
frames satisfy this criterion. The effect of bay size on the seismic performance of the model
frames is also significant. The median collapse probability for model frames with a bay width of
6.0 m is 0.267, six times higher than that for the model frames with a bay width of 9.1 m (0.046).
As shown in Fig. 3 (b), the median collapse probability decreases with an increase of bay size.
Regarding the SDC, the median collapse probability for the model frames designed for SDC
Cmax (SM1=0.2g) is 0.121, whereas the median collapse probability for the model frames
designed for SDC Cmin (SM1=0.1g) is 0.173. Thus, median collapse probability increases with a
higher SDC as shown in Fig. 3 (c). However, this is only valid for the systems and SDC values
considered in this study and cannot be generalized without further investigation.
Table 1. Summary of collapse margin parameters and acceptance check.
Perfor
mance
group
Group
1
Computed overstrength and collapse margin
parameters
ACMR
SMT
Ω μT SCT CMR SSF -a
Arch.ID
4.60
2.08
0.788
3.38
1.09
3.69
0.68
1.77
Pass
9CMINF-6.0
0.094
7.6
1.37
0.124
1.32
1.10
1.45
0.65
1.73
Fail
20CMINF-6.0
0.050
9.5
1.58
0.070
1.40
1.12
1.56
0.66
1.74
Fail
7.23
1.68
2.23
0.66
2.35
Fail
0.381
4.2
1.32
0.844
2.21
1.05
2.33
0.65
1.73
Pass
9CMAXF-6.0
0.153
4.6
1.31
0.193
1.26
1.09
1.37
0.65
1.73
Fail
20CMAXF-6.0
0.082
7.7
1.47
0.086
1.04
1.11
1.15
0.65
1.73
Fail
5.50
1.37
1.62
0.65
2.30
Fail
1.50
3CMINF-9.1
0.233
4.9
2.48
1.359
5.83
1.11
6.45
0.70
1.80
Pass
9CMINF-9.1
0.094
9.5
3.06
0.308
3.29
1.21
3.98
0.73
1.84
Pass
20CMINF-9.1
0.050
10.7
1.74
0.075
1.50
1.13
1.70
0.67
1.75
Fail
8.37
2.43
4.04
0.70
2.47
Pass
3.54
3CMAXF-9.1
0.381
5.0
3.60
1.802
4.73
1.13
5.33
0.73
1.84
Pass
9CMAXF-9.1
0.153
6.3
1.95
0.373
2.44
1.15
2.80
0.67
1.76
Pass
20CMAXF-9.1
0.082
7.7
2.06
0.113
1.37
1.15
1.59
0.68
1.77
Fail
6.33
2.54
3.24
0.69
2.42
Pass
Average
2.85
0.45
Collapse Probability of each Architype
0.4
Median
P(Collapse|SMT)
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0.45
0.4
0.4
0.35
0.35
P(Collapse|SMT)
0.45
P(Collapse|SMT)
2.03
3CMAXF-6.0
Average
Group
4
0.3
0.25
0.2
0.15
0.1
0.05
0
0
ACMR
P/F
-b
0.233
Average
Group
3
βTOT
3CMINF-6.0
Average
Group
2
Acceptance check
5
10
15
Number of story
20
0
25 5
0.3
0.25
0.2
0.15
0.1
0.05
6
7
8
Bay size (m)
9
0
10
Cmin
Cmax
Seismic Design Category (SDC)
Figure 3. Probability of collapse for MCE earthquake with respect to (a) number of story, (b) bay
length, and (c) seismic design category
Conclusions
This study evaluated the seismic performance of intermediate moment frames with RBS-B
connections designed according to current seismic design provisions. The procedure for seismic
performance evaluation prescribed in ATC-63 [7] was used. An analytical model for RBS-B
connections was developed to emulate the hysteretic behaviors of fracture-resistant and fractureprone RBS-B connection specimens. 12 model frames and 4 performance groups with different
number of stories, bay size and SDC values were considered. 6 of 12 model frames and 2 of 4
performance groups failed to meet acceptable performance levels against collapse according to
ATC-63 [7]. It is observed that the probability of collapse for the model frames increases with an
increase of the number of stories and SDC level, and a decrease in bay size.
Acknowledgments
Authors acknowledge the financial supports provided by the National Research Foundation of
Korea No.2012R1A2A2A06045129 and No. 2011-0028552.
References
1.
AISC. Prequalified connections for special and intermediate steel moment frames for seismic applications,
ANSI/AISC 358-10, Chicago, IL, 2010.
2.
Engelhardt MD, Fry GT, Jones SL, Venti MJ, Holliday SD. Behavior and design of radius-cut, reduced beam
section connections, SAC/BD-00/17, Sacramento, CA. 2000
3.
Han SW, Moon KH, Stojadinovic B. Design equations for moment strength of RBS-B connections, Journal of
Constructional Steel Research 2009; 65: 1087-1095.
4.
FEMA. Recommended seismic design criteria for new steel moment frame buildings, FEMA 350; Washington,
D.C., 2000
5.
Han SW, Moon KH, Hwang SH, Stojadinovic B. Rotation capacities of reduced beam section with bolted web
(RBS-B) connections, Journal of Constructional Steel Research 2012; 70: 256-263.
6.
Luco N, Cornell CA. Effects of Connection Fractures on SMRF Seismic Drift Demands, ASCE Journal of
Structural Engineering 2000; 126: 127-136
7.
ATC. Quantification of building seismic performance factors, ATC-63; Redwood City, CA., 2009
8.
AISC. Seismic provisions for structural steel buildings, ANSI/AISC 341-10; Chicago, IL., 2010
9.
SAC. Connection database, www.sacsteel.org/connections, SAC Joint Venture; Richmond, Calif., 2003
10. Gupta A, Krawinkler H. Seismic demands for performance evaluation of steel moment resisting frame
structures, John A. Blume Earthquake Engineering Center Rep. No. 132, Dept. of Civil and Environmental
Engineering, Stanford University, CA. 1999
11. Lignos D. Sidesway collapse of deteriorating structural systems under seismic excitations, PhD thesis, Stanford
University, CA. , 2008
12. Lee CH, Jeon SW, Kim JH, Uang CM. Effects of panel zone strength and beam web connection method on
seismic performance of reduced beam section steel moment connections, Journal of Structural Engineering
2005; 131: 1854-1865.
13. Tremblay R, Tchebotarev N, Filiatrault A. Seismic performance of RBS connections for steel moment resisting
frames: influence of loading rate and floor slab, Behavior of Steel Structures in Seismic Areas, Proceedings of
the Second International Conference, STESSA‘97, Kyoto, Japan, 664–671., 1997
14. Mazzoni S, McKenna F, Fenves GL. Opensees command language manual, Pacific Earthquake Engineering
Research Center (PEER), University of California, Berkeley, Available at http://opensees.berkeley.edu/., 2005
15. ASCE. Minimum design loads for buildings and other structures, ASCE/SEI 7-10, Reston, VA., 2010

Download Report