Evaluation of Building Period Formulas

EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS
Earthquake Engng Struct. Dyn. 2010; 39:15691583

Published online 23 March 2010 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/eqe.998
Evaluation of building period formulas for seismic design

Oh-Sung Kwon, and Eung Soo Kim
Department of Civil, Architectural and Environmental Engineering, Missouri University
of Science and Technology, Rolla, MO 65409, U.S.A.
SUMMARY
Building period formulas in seismic design code are evaluated with over 800 apparent building periods
from 191 building stations and 67 earthquake events. The evaluation is carried out with the formulas in
ASCE 7-05 for steel and RC moment-resisting frames, shear wall buildings, braced frames, and other
structural types. Qualitative comparison of measured periods and periods calculated from the code formulas
shows that the formula for steel moment-resisting frames generally predicts well the lower bound of the
measured periods for all building heights. But the differences between the periods from code formula and
measured periods of low- to-medium rise buildings are relatively high. In addition, the periods of essential
buildings designed with the importance factor are about 40% shorter than the periods of non-essential
buildings. The code formula for RC moment-resisting frames describes well the lower bound of measured
periods. The formula for braced frames accurately predicts the lower bound periods of low-to-medium
rise buildings. The formula for shear wall buildings overestimates periods for all building heights. For
buildings that are classified as other structural types, the measured building periods can be much shorter
than the periods calculated with the code formula. Based on these observations, it is suggested to use
Cr factor of 0.015 for shear walls and other structural types. Copyright q 2010 John Wiley & Sons, Ltd.
Received 27 August 2009; Revised 7 January 2010; Accepted 8 January 2010
KEY WORDS:
building period formula; seismic design; system identification
1. INTRODUCTION
The fundamental period of a building is a key parameter for the seismic design of a building
structure using the equivalent lateral force procedure. As the building period cannot be analytically
calculated before the building is designed, periods from the empirical period formulas recommended
in seismic design codes or from finite element analysis with assumed mass and stiffness are used
during the preliminary design stage. In most building design projects, empirical building period
Correspondence
to: Oh-Sung Kwon, Department of Civil, Architectural and Environmental Engineering, Missouri
University of Science and Technology, Rolla, MO 65409, U.S.A.
E-mail: kwono@mst.edu
Contract/grant sponsor: Department of Civil, Architectural, and Environmental Engineering at the Missouri University
of Science and Technology
Copyright q
2010 John Wiley & Sons, Ltd.
1570
O. KWON AND E. KIM
formulas are used to initiate the design process. The period from the empirical period formula
also serves as a basis to limit the period from a finite element model by applying the upper bound
factor, Cu , suggested in the 2003 NEHRP Recommended Provisions for Seismic Regulations for
New Buildings ([1], referred to as NEHRP 2003 hereafter) and subsequently in ASCE 7-05 [2].
In the 1970s design codes, such as UBC-70 [3] and BOCA-75 [4], two formulas were used to
estimate building periods: one for moment-resisting frames (MRFs hereafter) and the other one
for all other structural types as summarized in Table I. These formulas remained in the code until
UBC-82 [5]. From the ATC 3-06 project [6], the period formulas for reinforced concrete and steel
moment-resisting frames (RC MRFs and steel MRFs hereafter) were calibrated based on identified
building periods from the 1971 San Fernando Earthquake. Seventeen steel MRFs and 14 RC MRFs
were used for this calibration. The form of the formulas for the RC and steel MRFs in ATC 3-06
[6] were developed based on the assumption that lateral forces are distributed linearly over the
height of a building and that the deflections of the building are controlled by drift limitation.
The calibrated building formulas in ATC 3-06 [6] were reflected in BOCA-87 [7] and UBC88 [8] with minor refinement. The same form of the formula is also applied to other structural
types in UBC-88 [8]. More recently, Goel and Chopra [911] calibrated the formula for MRFs
in the code and developed a new formula for shear wall buildings with measured (or apparent)
building periods from several earthquake events. In their study, 42 steel MRFs, 27 RC MRFs, and
9 shear wall buildings were used. In the study of shear walls [10], it was found that the building
period formula for shear walls should be a function of equivalent shear area and building height
rather than a function of only building height. Hence, rather than calibrating parameters in the
existing code formula, a new formula was suggested. The suggested period formula for shear walls
and calibrated parameters for MRFs in Goel and Chopra [911] were reflected in NEHRP 2000
[12] and 2003 which is the basis of the current minimum design load for buildings and other
structures [2]. Table I summarizes the revision history of approximate building period formulas
in design specifications (UBC, BOCA, NEHRP, ASCE 7, Eurocode, and ATC 3-06 [6]) since
1970s.
While the building period formulas have been calibrated and revised over the past 30 years,
the number of data points that were used for the previous calibrations was limited to take account
the wide variability of structural periods over building heights. Housner and Brady [24] noted
a wide spread in the available data set of periods for shear wall structures, mainly due to large
variations in building stiffness and to a lesser extent effective mass. In addition, well documented
calibration or evaluation results are not reported for several structural types. For example, the current
approximate period formula for eccentrically braced frames (EBF), shear walls, and other structural
types first appeared in UBC-88 [8]. But these formulas have not been calibrated. Furthermore, all
instrumented buildings are in strong seismic regions. To account for the effects of the design base
shear to structural periods, the upper limit coefficient, Cu , is proposed, which varies with design
spectral acceleration level, purely based on engineering judgment. Hence, the empirical building
period formulas derived from periods of buildings in strong seismic regions may or may not be
applicable to the buildings in a region with lower seismic activity.
The California Strong Motion Instrumentation Program (CSMIP) developed by the California
Geological Survey (CGS) has instrumented over 170 buildings in California since its establishment
in 1972. Among the instrumented buildings, around 40 buildings were instrumented after the
1994 Northridge Earthquake. These instrumented buildings have recorded many minor to moderate
seismic events in the US for the past several decades. The objective of this study is to evaluate
the current building period formulas in seismic design codes with measured (or apparent) building
Copyright q

DOI: 10.1002/eqe
EVALUATION OF BUILDING PERIOD FORMULAS FOR SEISMIC DESIGN
1571
Table I. Approximate fundamental period formulas.

RC MRF
Steel MRF
UBC-70, 82 [13, 14]

BOCA-75 [4]
Ta = 0.10N
ATC 3-06 [6]
Ta = Ct h n
RC/Masonry shear wall
Ct = 0.035

Ta = 0.05h n / D
3/4
Ta = Ct h n
Ct = 0.030
Other
Ta = 0.05h n / D
Ta = 0.05h n / D
3/4
Ct = 0.025
BOCA 87 [7]
EBF
See note
Ct = 0.035
UBC-88, 94, 97
Eurocode 8 [15]
3/4
Ta = Ct h n
Ct = 0.030
Ct = 0.035
Ct = 0.02 or,

Ct = 0.1/ AC
Ct = 0.020
Ct = 0.020
Ct = 0.020
Cr = 0.020
x = 0.75
Cr = 0.020
x = 0.75

or, Ta = 0.0019h n / Cw
Ct = 0.030
3/4
ASCE 7-98 [16]
Ta = Ct h n
BOCA-96 [17, 18]

Ct = 0.030
NEHRP 94, 97 [19, 20]
Ct = 0.035
or, Ta = 0.10N
Ct = 0.030
NEHRP 00 [12], 03
ASCE 7-02,05 [2, 21]
Ta = Cr h nx
Cr = 0.016
x = 0.9
Cr = 0.028
x = 0.8
or, Ta = 0.10N
Cr = 0.030
x = 0.75
Note: Rayleighs method is also suggested as a period formula for all structural types in BOCA-87 [7],
UBC-82-97 [14, 22]. As the equation is not a function of geometry and needs a structural model a priori, the
equation is not included in this table.
For shear walls or exterior concrete frames utilizing deep beams or wide piers, or both, D is the dimension
of the building in ft in a direction parallel to the applied force. For isolated shear walls not interconnected
by frames or for braced frames, D is the dimension of the shear wall or braced frame in a direction parallel
to the applied force.
Applicable to structures not exceeding 12 stories in height and having a minimum story height of not less
than 10 ft.
Refer to NEHRP 2003 for the definition of C .
w
Refer to UBC-94 [23] for the definition of A .
c
Eurocode 8 also suggests using T = 2 (d), where d is the lateral elastic displacement of top of the building
a
in m due to the gravity loads applied in the horizontal direction. Since the equation is not a function of
geometry and needs a structural model a priori, the equation is not included in this table.
BOCA-96 [17, 18] allows the use of C = 0.03 for both EBF systems and dual systems using EBF.
t
Copyright q

DOI: 10.1002/eqe
1572
O. KWON AND E. KIM
periods from the instrumented buildings. To achieve this objective, 141 buildings are selected
from the CGS stations and apparent periods of the buildings are identified utilizing the transfer
function method. In addition, periods of 50 buildings from previous studies [6, 9, 10] are compiled
to build a database of total 191 buildings. A total of 411 earthquake events recorded from these
buildings are used to identify more than 800 building periods including periods in the transverse and
longitudinal directions. These measured periods are used to evaluate the building period formulas
in [2]. The effects of Occupancy Category, building height, number of stories, and lateral loadresisting systems to the fundamental periods of the buildings are evaluated. From the evaluation,
several recommendations for future studies are proposed.
In the following section, selected building stations and adopted system identification method
are presented. In Section 3, the identified building periods are used to evaluate building period
formula in the current seismic design code. This is followed by the conclusions and suggestions
for future study in Section 4.
2. SELECTED BUILDINGS AND IDENTIFICATION OF PERIODS

To evaluate the approximate period formulas in the current seismic design code, a database of
191 building periods is developed. There are about 170 building stations instrumented through
California Strong Motion Instrumentation Program. After reviewing the plans of the buildings,
buildings with large irregularity, base isolation system, or energy dissipation system are excluded
from the selection. Total 141 CGS stations are selected out of the 170 stations. The 50 non-CGS
stations are from ATC 3-06 [6] (26 buildings), National Oceanic and Atmospheric Administration
(NOAA, 14 buildings), United States Geological Survey (USGS, 4 buildings), and unknown sources
(6 buildings). All buildings investigated in this study are located in California. The data from the
selected buildings consist of a total of 67 earthquake events beginning with the 1970 Lytle Creek
Earthquake (M 5.2) to the 2008 Yucaipa Earthquake (M 4.1).
The selected buildings have various lateral load-resisting systems including 125 steel MRFs, 58
RC MRFs, 56 RC shear walls, 34 concentrically braced frames (CBFs), 8 EBFs, 23 reinforced
or unreinforced masonry shear walls (RM or URM shear walls), 12 precast concrete tilt-up shear
walls (PC shear walls), and 66 other structural types which cannot be classified as one of the
above lateral load-resisting systems. Some of the unclassified buildings consist of dual or multiple
systems. Since transverse and longitudinal lateral load-resisting systems are classified as separate systems, the total number of lateral load-resisting systems evaluated in this study is 382.
The number of data points for RC MRFs (58) is not notably larger than the number of data points
(50) that were used in the previous calibration by Goel and Chopra [9]. But other structural types
provide a considerably large number of new data points that were not available in the previous
studies for the systematic evaluation of period formulas.
The selected buildings are classified into two occupancy categories, namely essential facilities
(Occupancy Category IV) and non-essential facilities. Among the 141 CGS stations, 34 buildings
belong to Occupancy Category III, such as hospitals and buildings for post-earthquake emergency
responses and communications. Since buildings in Occupancy Category IV are designed with
higher design base shear than the buildings in other occupancy categories, the investigation on the
periods of these buildings can provide insight into the effects of seismic design level on structural
periods, from which the applicability of the current empirical formula to buildings in low seismic
areas can be indirectly evaluated.
Copyright q

DOI: 10.1002/eqe
1573
Plan
Transfer function
Ch. Avg = 0.51 sec
14
Roof
5th
4th
3rd
2nd
12
10
8
6
4
2
0
0
0.2
0.4
0.6
0.8
Period, sec
Elevation
Figure 1. Transfer function in NS direction (C13214 station, 2008 Chinohills Earthquake).
The fundamental period in this study refers to the apparent first mode period identified from
recorded accelerations. It is termed apparent as the true fundamental period of a building is hard
to be identified due to several factors including the variation of the periods from inelastic response
of soil and structure, soilstructure-interaction, inadequate distribution or insufficient number of
accelerometers, and noise in the measurement system [2527]. To minimize the effects of nonlinear
responses on the apparent periods, the periods from low-intensity seismic events are used in this
study. As there is no clear transition between linear elastic response and nonlinear response in
most structures, the cut-off PGA of 0.15g is selected, which was used for the calibration of the
period formulas by Goel and Chopra [9]. By using the cut-off period consistent with the previous
studies, the periods identified from this study can be systematically compared with those in the
previous studies and with the periods calculated from approximate period formulas in the seismic
design code.
Transfer functions of each building for each direction (transverse and longitudinal direction)
are identified with one input channel (typically base acceleration) and several output channels
along the height of the building. As the base acceleration is used as an input channel, the effect
of soil properties or soilstructure-interaction on the apparent building periods is not considered
for evaluation. Input and output channels of all 141 CGS stations are carefully selected after
inspection of the sensor locations. Output channels at a specific location of the buildings, which
may independently vibrate or which may not capture the global vibration of the building, such
as at a penthouse or at one of the peculiar wings of the building, are not included in the system
identification. The average of periods from multiple channels is defined as the apparent period of
the building. For instance, the transfer functions of CGS station C13214 in Figure 1 show slightly
different periods depending on output channels. The average of these periods is used as the period
of the building in the considered direction.
3. EVALUATION OF BUILDING PERIOD FORMULAS

The following sections compare identified building periods with the approximate code formulas.
To minimize the period variations due to inelastic deformation of structures, periods from earthquake events with peak input channel acceleration less than 0.15g are selected for comparisons
Copyright q

DOI: 10.1002/eqe
1574
O. KWON AND E. KIM
except for CBFs. For the CBFs, a threshold of 0.20g is used due to limited number of data points.
The effects of importance factors are investigated only for steel MRFs.
3.1. Steel MRFs
There are two approximate period formulas for steel MRFs in the current seismic design provisions.
The first period formula, Equation (1), is based on the study by Goel and Chopra [9] where periods
from 42 steel buildings (a total of 81 lateral and transverse steel MRF systems) were used for
the calibration. The second formula, Equation (2), has been in the code since the 1970s and is
applicable to structures not exceeding 12 stories and with a minimum story height greater than 10ft.
Ta = Cr h nx
(1)
Ta = 0.10N
(2)
where Cr and x are parameters whose values vary depending on the lateral load-resisting system
of the building, h n is height in ft, and N is number of stories. For the steel MRFs, Cr is 0.028
and x is 0.8. The database in this study includes 65 steel MRF buildings (total 125 lateral and
transverse steel MRF systems). The newly added building periods in this study include periods
from several low-to-medium rise buildings lower than 100 ft in height. Figure 2 compares the
periods of steel MRFs with the approximate period formula in ASCE 7-05 [2]. The lower bound
period is calculated using Equation (1). The upper bound period is calculated for a site with
S D1 0.4g. Figure 2(a) shows that the current code formula conservatively predicts the lower bound
of structural periods for all building heights. However, Figure 2(b) indicates that the difference
between the periods from code formula and apparent periods is relatively large, especially for the
buildings with heights less than 100 ft, which corresponds to 68 story buildings. Based on the
study by the US Department of Energy with randomly chosen commercial buildings in the United
States, 98% of commercial buildings, including office buildings, have less than 100 000 ft2 of floor
space which is equivalent to 38 story office buildings [28]. Considering that the majority of
buildings are low-to-medium rise buildings and that Figure 2(b) shows large uncertainty in the code
formula for these buildings, the code formula may need further refinement depending on building
heights.
Figure 2(c) compares apparent building periods in terms of the number of stories with periods
using Equation (2) which is applicable to buildings with a minimum story height not less than 10 ft
and with the number of stories not exceeding 12. For the purpose of evaluation of the formula, all
buildings are plotted in Figure 2(c) including buildings with minimum story heights of less than
10 ft or with more than 12 stories. The database of the selected buildings shows that the buildings
that have a minimum story height less than 10 ft are mostly residential buildings and hotels. From
Figure 2(c), it can be seen that the formula that has been used for over 30 years predicts the lower
bound of building periods, especially for buildings with less than 5 stories. For buildings with
more than 12 stories, the code formula largely underestimates structural periods of all buildings
including buildings with minimum story height less than 10 ft. As the code formula provides
conservative period (short periods), Equation (2) may not need to be restricted to buildings with
number of stories less than 12 and story heights less than 10 ft unless the economic design is a
concern.
Seismic design codes mandate the use of higher design base shear for essential facilities which
are required for post-earthquake recovery or which contain substantial quantities of hazardous
Copyright q

DOI: 10.1002/eqe
1575
8
Ta=0.028h0.8, Cu=1.4
Period, sec
7
6
5
4
Ta=0.028h0.8
3
2
New data points

From literature
1
0
200
400
Height, ft
(a)
600
800
3.0
Period, sec
2.5
Ta=0.028h0.8, Cu=1.4
2.0
1.5
1.0
Ta=0.028h0.8
New data points
From literature
0.5
0.0
50
100
150
Height, ft
(b)
Period, sec
4.0
3.0
Ta=0.1N, Cu=1.4
2.0
Ta =0.1N
1.0
Code limit
MinStoryHeight <10ft
MinStoryHeight >=10ft
0.0
0
(c)
10
15
Number of Stories
20
25
Figure 2. Comparison of periods of steel MRFs from records: (a) comparison of Equation (1) with periods
from all buildings; (b) comparison of Equation (1) with periods from low-to-medium rise buildings; and
(c) comparison of Equation (2) with periods from steel MRFs.
substances. For the design of the essential facilities, which belong to Occupancy Category IV in
the ASCE 7-05 [2], higher occupancy importance factors are applied. The occupancy importance
factors were not specified in the design codes in the 1970s. The importance factors, I, from
1.25 to 1.5 for essential facilities have been used since 1980s; UBC-82 [14] (I = 1.5), BOCA-87
[7] (I = 1.5), UBC-88, 94 [8, 23] (I = 1.25), and ASCE 7-02, 05 [21, 2] (I = 1.5). Hence, it is
expected that essential facilities designed and constructed after 1980 were designed with 2550%
Copyright q

DOI: 10.1002/eqe
1576
O. KWON AND E. KIM
2.0
Essential facilities
Non-essential facilities
Period, sec
1.6
1.2
T = 0.034h0.8
0.8
0.4
T = 0.02h0.8
0.0
0
20
40
Height, ft
60
80
100
Figure 3. Effects of occupancy categories to steel MRFs.
higher design base shear than buildings in other Occupancy Category assuming that other design
parameters are similar.
To investigate the effects of higher design load on the fundamental periods of steel MRFs,
Figure 2 is re-plotted with two categories of periods for essential facilities, such as hospitals and
emergency response agencies, and non-essential facilities. There are total 8 steel MRF buildings
that belong to Occupancy Category IV which were designed after 1980. The heights of these
buildings are less than 100 ft. The periods of the essential facilities and non-essential facilities in
this height range are compared in Figure 3 and regression analyses are conducted with the two
sets of data points for qualitative comparison. For the purpose of comparison, Equation (1) with
x = 0.8 is used for the regression analysis. Figure 3 shows that the periods of the low-to-medium
rise essential facilities are about 40% shorter than that of the non-essential buildings in the similar
height range. It needs more data to generalize this observation to high-rise buildings. Based on the
observation though, it is suggested to use different building period formulas for low-to-medium
rise essential buildings and non-essential buildings in the same height range. As an alternative,
a factor of 0.6 can be applied to the existing code formula for low-to-medium rise steel MRF
essential buildings.
The decrease in period with the use of the importance factor is consistent with the findings of
Tremblay [29] where it was found that CBF structures may reduce periods by as much as 42%
due to application of the importance factor. Nakashima et al. [30] reported similar variations of
building period with seismic hazard levels for high-rise steel frames built in Japan between 1968
and 1988. These findings have two implications: (1) Use of the importance factor tends to shorten
the period which in general leads to a larger seismic demand depending on the period. Hence,
buildings designed with importance factors may not provide the margin of safety required to ensure
the operational performance of facilities after an earthquake event. (2) The building period depends
on the level of design base shear. Buildings designed for lower seismic regions are expected to have
longer periods than buildings in built in higher seismic regions. In the current code, the variation
of building period is indirectly considered by allowing higher Cu factor (up to Cu = 1.7) in regions
with low seismic base shear demand. With a lack of period data in low seismic regions in the
US, however, further data collection is required to calibrate empirical building period formulas for
buildings in low seismic regions.
Copyright q

DOI: 10.1002/eqe
1577
Figure 4. Periods of concentrically braced frames.
3.2. Braced steel frames

In the current seismic design provisions, empirical period formula for EBF follows the format of
Equation (1) with Cr = 0.030 and x = 0.75. This equation has not been calibrated or evaluated since
late 1980s when the equation was first recommended for EBFs in UBC-88 [8]. The period formula
for CBFs has the same format except Cr = 0.020. Hence the EBFs could be designed for longer
periods than CBFs in order to account for the flexibility in the link beams. Recently, Tremblay
[29] evaluated the period formula for braced steel frames by reviewing periods of 220 braced
frame buildings. Among the buildings, however, only two data points from a two-story building
were measured periods while all others are periods from analytical models. Through theoretical
and parametric studies, Tremblay [29] proposed an equation for CBFs as Ta = 0.0076h n , where
h n is building height in feet.
Among the selected CGS stations, there are 34 CBFs and 8 EBFs excluding dual systems in
which both braced frames and steel MRFs are used as lateral load-resisting systems. And among
the 34 CBFs, 17 frames were installed with base isolation or seismic dampers. It is speculated
that the CBFs were used to concentrate deformation demand to the isolation bearings in the base
isolated buildings. The buildings with isolators are not included in the period comparison. Hence,
the measured periods of 17 CBFs and 8 EBFs are compared with the formulas in ASCE 7-05 [2]
and by Tremblay [29].
Figure 4 compares the measured periods of CBFs with the periods from ASCE 7-05 [2] (Ta =
0.02h 0.75
n ) and Tremblay (2005) (Ta = 0.0076h n ). Station C24643 has moment frames in the longitudinal direction and CBFs in the transverse direction. Since the main lateral load-resisting system
in the transverse direction is CBFs, the period in this direction is included in the comparison. It can
be clearly seen from Figure 4 that the equation for CBF in ASCE 7-05 [2] greatly underestimates
the period of a high rise building (station C24643), which may lead to the uneconomical design.
For low-to-medium rise buildings, the formula follows the lower bound of the measured periods.
For all building heights, the formula developed by Tremblay [29] follows the apparent building
periods better than the code formula.
Figure 5 compares the measured periods of EBFs with period from ASCE 7-05 [2] (Ta =
0.03Hn0.75 ). Owing to the limited number of data points, it is difficult to properly evaluate the
current code formula. But it can be noted that for the given number of measured data points, the
equation in ASCE 7-05 [2] describes the relationship between building height and the apparent
Copyright q

DOI: 10.1002/eqe
1578
O. KWON AND E. KIM
3.5
Period, sec
3.0
2.5
T = 0.03h0.75
2.0
1.5
1.0
0.5
0.0
0
50
100
150
Height, ft
200
250
300
Figure 5. Periods of eccentrically braced frames.
building period. More data points are required to properly evaluate the formula for high-rise
buildings.
3.3. RM, URM, and PC shear walls
In ASCE 7-05 [2], two equations, Equations (3) and (4), can be used to calculate the approximate
periods of shear wall structures. The Equation (3) has the same form as Equation (1) except that
Cr is 0.02 and x is 0.75. In fact, Equation (3) is applicable to all structural systems as it is the most
conservative equation in the current design code. The equation has been in the code since UBC-88
[8] and has not been recalibrated for shear walls. Goel and Chopra [10] evaluated Equation (3)
with 16 buildings and found that for majority of buildings the code formula gives periods longer
than the measured values, which are non-conservative. The second formula in ASCE 7-05 [2],
Equation (4), proposed by Goel and Chopra [10], is based on measured periods of nine shear wall
buildings. The format of Equation (4) was theoretically derived and calibrated with the measured
building periods. In order to apply Equation (4) in a design process, however, a designer should
have the dimensions of the shear walls a priori.
Ta = Cr h nx
where Cr = 0.02,

Ta = 0.0019h n / Cw
x = 0.75
(3)
(4)
where,
Cw =
2
n
100
hn

A B i=1 h i
Ai
2
hi
1+0.83
Di
A B = base area of the structure in ft2

Ai = the area of shear wall i in ft2
Di = the length of shear wall i in ft
h i = the height of shear wall i in ft
n = the number of shear walls in the building effective in resisting lateral forces
in the direction under consideration
Copyright q

DOI: 10.1002/eqe
1579
3
C58483
Period, sec
C24541
T = 0.02h 0.75 , Cu =1.4
1
T = 0.02h0.75
T = 0.015h 0.75
0
0
50
100
150
Height, ft
200
RC SW
PC SW
RM & URM SW
250
300
Figure 6. Periods of shear wall structures.
Owing to a lack of detailed design information for the buildings in the database, Equation
(4) could not be properly evaluated. Therefore, only Equation (3) is evaluated in this study.
The measured periods of 56 RC shear walls, 23 RM and URM shear walls, and 12 PC shear walls
are compared with the code formula as shown in Figure 6. The figure shows that the data points
of periods of shear walls form a trend except the two outlier data points, denoted as C24541 or
C58483, which have longer periods than other structures. It can be seen from Figure 6 that for
a large number of RC shear wall buildings, the code formula overestimates the lower bound of
structural period which may lead to unconservative seismic design. This observation is consistent
with the findings of Goel and Chopra [10]. If the simple format of Equation (3) is to be used
for the design, it is suggested to use Cr of 0.015 which predicts the lower bound of the apparent
building periods better than Equation (3) as shown in Figure 6. Another observation that can be
made is that the periods of PC, RM, and URM shear wall buildings are similar to the periods of
the reinforced concrete shear wall. Hence the suggested Cr factor of 0.015 can be applied to other
shear wall building types. As only two shear wall buildings are in Occupancy Category IV, the
effects of the importance factor on the fundamental periods are not investigated for the shear wall
buildings.
3.4. RC MRF
The period formula for RC MRFs has the same format as Equation (1) except Cr = 0.016 and
x = 0.9. Equation (2) can be also applied to calculate the period of RC MRFs. The data points
in Figure 7(a) show that the current period formula follows the lower bound of the measured
periods. Figure 7(b) compares the measured periods with Equation (2), which is applicable to a
structure with a minimum story height of not less than 10 ft and a maximum number of stories
not exceeding 12. Equation (2) also follows the lower bound of periods of buildings even up to
25 story buildings and even for buildings with a minimum story height less than 10 ft. However,
the measured periods are more dispersed when the period is a function of the number of stories
than when the period is a function of building height.
Copyright q

DOI: 10.1002/eqe
1580
O. KWON AND E. KIM
Period, sec
3.0
New data points

From literature
2.5
2.0
Ta=0.016h0.9, Cu=1.4
1.5
Ta=0.016h0.9
1.0
0.5
0.0
0
50
100
(a)
Period, sec
2.5
150
Height, ft
250
200
300
Ta=0.1N, Cu=1.4
2.0
1.5
Ta=0.1N
1.0
0.5
Code limit
0.0
0
(b)
Min Story Height < 10ft

MinStoryHeight >= 10ft
10
15
Number of Stories
20
25
Figure 7. Periods of RC MRFs with input ground motions less than 0.15g: (a) periods as a function of
building height and (b) periods as a function of number of stories.
3.5. Other structural types

Equation (1) can also be applied to other structural configurations. Among the 142 selected CGS
stations, 65 stations could not be classified as one of the structural systems evaluated in the
previous sections. These stations are in the Other Structures category and have lateral loadresisting systems such as plywood shear walls, dual systems (steel brace and MRFs, RC shear wall
and MRFs), and frames with composite elements. The measured periods of these buildings are
compared with Equation (1) with Cr = 0.02 and x = 0.75 in Figure 8 where it can be clearly seen
that the approximate period in the code largely overestimates the periods of these buildings. One
reason for the overestimation is the fact that a large number of buildings having dual or multiple
lateral load-resisting systems exhibit higher stiffness and strength than buildings with a single
lateral load-resisting system. Some of the outlier data points are denoted in Figure 8 with their
building code. The buildings with shorter measured periods than the periods predicted with the
code formula include C58718 (distributed concrete shear walls and braced steel frames), C24514
(concrete shear walls and perimeter steel shear walls), C03743 (concrete walls, concrete moment
frames, and steel braces), and C58257 (composite steel plate shear walls and steel MRFs). The
buildings with longer measured than predicted periods include C24581 (composite steel-concrete
frame with infill URM walls on the perimeter), C14654 (chevron-type braced frames and steel
MRF), and C57318 (steel MRF and RC shear wall). As there are not many data points for these
building types, the structural periods of buildings in the Other Structures category may need to
Copyright q

DOI: 10.1002/eqe
Period, sec
1581
C24581
C14654
C57318
2
T = 0.02h0.75, Cu=1.4
T = 0.02h0.75
1
C24514
C58718
0
0
50
100
C03743
C58257
T = 0.015h 0.75
150
Height, ft
200
250
300
Figure 8. Periods of other types of structures.
be estimated with a sound mechanical basis such as using the finite element method. To initiate
design process, however, it is suggested to use Cr factor of 0.015 which predicts the lower bound
building period better than the current code formula as shown in Figure 8.
4. CONCLUSIONS AND SUGGESTED FUTURE STUDIES

This study evaluates the period formulas in seismic design provisions through comparison of
apparent periods of instrumented buildings. Periods from 141 building stations in the CGS and
periods from 50 buildings available in the literature are used for this evaluation. The measured
periods of steel MRFs, braced frames, shear wall buildings, RC MRFs, and other types of structures
are compared with the code formulas. For steel MRFs, the effects of importance factors are also
investigated. The following is a summary of the findings from this study:
For steel MRFs, the current code formula, Equation (1), describes the general trend of the
lower bound of the measured periods. For low-to-medium rise buildings, the uncertainty in the
code formula is high. The code formula may need further refinement depending on building
heights.
The essential steel MRF buildings designed with an importance factor tend to have periods
about 40% shorter than the buildings designed without an importance factor. Hence it is
suggested to apply a factor of 0.6 to the approximate building period for the essential steel
MRFs. The intention of the importance factor in essential buildings is to increase the safety
margin (or to decrease probability of failure). Since the decreased period may increase seismic
demand depending on periods, the effect of importance factors to the seismic fragility of
essential buildings needs careful reevaluation.
As was shown in the comparisons involving essential buildings, the level of design base shear
affects the structural period. Since the building period formulas have been derived primarily
with data obtained from buildings in strong seismic regions with large design base shear, the
formulas can be conservatively applied to buildings in low seismic activity areas. But once
data are available, the actual periods of buildings in low seismic region need to be evaluated
to further calibrate code formulas and to better estimate seismic fragility of those buildings.
Copyright q

DOI: 10.1002/eqe
1582
O. KWON AND E. KIM
For CBFs and EBFs, there are not many available data points due to the limited number of
instrumented buildings and due to the fact that large portion of the instrumented CBFs have
base isolation systems. But based on the available data points, the code formula, in the form
of Equation (1), tends to underestimate the lower bound of the periods for buildings taller
than 200 ft. For low-to-medium rise buildings, the code formula predicts well the lower bound
periods.
For RM, URM, and RC shear walls, the code formula in the form of Equation (1) overestimates
the lower bound of measured periods, which can be unconservative. This finding is consistent
with the research by Goel and Chopra [10]. RM, URM, and RC shear walls have similar
trends so a single equation may be applicable to all shear wall types. If Equation (1) is used
for design, a revision of Cr factor from 0.02 to 0.015 is suggested.
For RC MRFs, the code formula predicts the lower bound of apparent periods.
For other structural types, a revision of Cr factor from 0.02 to 0.015 is suggested.
The approximate period formula in the current seismic provision ASCE 7-05 [2], has evolved for
over three decades. With the new information from the instrumented buildings, the current code
formulas are evaluated in this study. Code formulas for clearly defined lateral load-resisting systems
tend to predict the lower bound of the apparent periods. For other structural types, the equations
need more refinement and calibration. In addition, there are about 30 or more building stations
in low-to-medium seismic region in the National Strong-Motion Network (NSMN) operated by
USGS. Recorded data from these stations may provide a good basis to extrapolate building period
formulas developed and calibrated with buildings in strong seismic region to the buildings in low
seismic region.
ACKNOWLEDGEMENTS
This investigation was carried out with the support from the Department of Civil, Architectural, and
Environmental Engineering at the Missouri University of Science and Technology. The authors are grateful
for this support. The authors also appreciate valuable comments by Mr. Lim in ABS Consulting, Inc. on
seismic design practice in California.
REFERENCES
1. Building Seismic Safety Council (BSSC). NEHRP recommended provisions for the development of seismic
regulations for new buildings. FEMA 450, Washington, DC, 2003.
2. American Society of Civil Engineers. Minimum design loads for buildings and other structures. ASCE 7-05,
Reston, Virginia, 2005.
3. International Conference of Building Officials (ICBO). Uniform Building Code, Whittier, CA, 1970.
4. Building Officials and Code Administrators International (BOCA). The BOCA Basic Building Code/1975 (6th
edn). Building Officials and Code Administrators International: Chicago, IL, 1975.
5. International Conference of Building Officials (ICBO). Uniform Building Code, Whittier, CA, 1982.
6. Applied Technology Council. Tentative provisions for the development of seismic regulations for buildings.
ATC3-06, Applied Technology Council, Palo Alto, CA, 1978.
7. Building Officials and Code Administrators International (BOCA). The BOCA National Building Code/1987
(10th edn). Building Officials and Code Administrators International: Country Club Hills, IL, 1987.
8. International Conference of Building Officials. Uniform Building Code, International Conference of Building
Officials, Pasadena, CA, 1988.
9. Goel RK, Chopra AK. Period formulas for moment-resisting frame buildings. Journal of Structural Engineering
1997; 123(11):14541461.
Copyright q

DOI: 10.1002/eqe
1583
10. Goel RK, Chopra AK. Period formulas for concrete shear wall buildings. Journal of Structural Engineering
1998; 124(4):426433.
11. Goel RK, Chopra AK. Vibration properties of buildings determined from recorded earthquake motions. EERC97/14, Earthquake Engineering Research Center, University of California, Berkeley, Richmond, CA, 1997.
15. Comite Europeen de Normalisation (CEN). Eurocode 8: Design of Structures for Earthquake ResistancePart
1: General Rules, Seismic Actions and Rules for Buildings, 2004.
Reston, VA, 2000.
regulations for new buildings and other structures. FEMA 302, Washington, DC, 1997.
Reston, Virginia, 2002.
24. Housner GW, Brady AG. Natural periods of vibrations of buildings. Journal of the Engineering Mechanics
Division 1963; 89(4):3165.
25. Udwadia FE, Trifunac MD. Time and amplitude dependent response of structures. Earthquake Engineering and
Structural Dynamics 1974; 2(4):359378.
26. Trifunac MD, Ivanovic SS, Todorovska MI. Apparent periods of a building. I: Fourier analysis. Journal of
Structural Engineering 2001; 127:517526.
27. Trifunac MD, Ivanovic SS, Todorovska MI. Apparent periods of a building. II: time-frequency analysis. Journal
of Structural Engineering 2001; 127:527537.
28. U.S. Department of Energy. A look at commercial buildings in 1995: characteristics, energy consumption, and
energy expenditures, DOE/EIA-0625(95), Washington, DC, 1995.
29. Tremblay R. Fundamental periods of vibration of braced steel frames for seismic design. Earthquake Spectra
2005; 21(3):833.
30. Nakashima M, Yanagi H, Hosotsuji J. Simple expressions for prediction fundamental natural periods of highrise
buildings. Proceedings of the First International Conference on Constructional Steel Design, Acapulco, Mexico,
69 December 1992; 385394.
Copyright q

DOI: 10.1002/eqe

Evaluation of Building Period Formulas

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Evaluation of Building Period Formulas

Загружено:

Авторское право:

Доступные форматы

EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS

Earthquake Engng Struct. Dyn. 2010; 39:15691583