Encyclopedia of Earthquake Engineering

DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

Performance-Based Design Procedure for Structures with

Magneto-Rheological Dampers
Yunbyeong Chaea*, Baiping Dongb, James M. Riclesc and Richard Saused
a
Department of Civil and Environmental Engineering, Old Dominion University, Norfolk, VA, USA
b
Department of Civil and Environmental Engineering, Lehigh University, Bethlehem, PA, USA
c
Bruce G. Johnston Professor of Structural Engineering, Department of Civil and Environmental Engineering, Lehigh
University, Bethlehem, PA, USA
d
Joseph T. Stuart Professor of Structural Engineering, Department of Civil and Environmental Engineering, Lehigh University,
Bethlehem, PA, USA

Synonyms
Magneto-rheological damper; Performance-based seismic design; Simplified design procedure

Introduction
It is well-known that supplemental damping devices increase the energy dissipation capacity of structures,
reducing the seismic demand on the primary structure (Constantinou et al. 1998; Soong and Dargush
1997). A structural system with supplemental dampers is often represented by an equivalent linear system.
Kwan and Billington (2003) derived optimal equations for the equivalent period and damping ratio of
SDOF systems with various nonlinear hysteresis loops based on time-history analysis and regression
analysis. Symans and Constantinou (1998) studied the dynamic behavior of SDOF systems with linear or
nonlinear viscous fluid dampers and derived an equation for the equivalent damping ratio of the nonlinear
viscous fluid damper. Ramirez et al. (2002) proposed a simplified method to estimate displacement,
velocity, and acceleration for yielding structures with linear or nonlinear viscous dampers. Lin and Chopra
(2003) investigated the behavior of SDOF systems with a diagonal brace in series with a nonlinear viscous
damper by transforming the system to an equivalent linear Kelvin model.
Fan (1998) investigated the behavior of nonductile reinforced concrete frame buildings with visco-
elastic dampers. He derived an equivalent elastic-viscous model based on the complex stiffness and
energy dissipation of the viscoelastic system and proposed a simplified design procedure for a structure
with viscoelastic dampers. Lee et al. (2005, 2009) applied this method to structures with elastomeric
dampers and validated the simplified design procedure by comparing the design demand with the results
from nonlinear time-history analysis.
In this paper, a systematic procedure for the design of structures with MR dampers, referred to as the
Simplified Design Procedure (SDP), is developed. The procedure is similar to that developed by Lee
et al. (2005, 2009), but with modifications to account for the characteristics of the MR dampers.
A quasistatic MR damper model for determining the loss factor and the effective stiffness of an MR
damper is introduced and incorporated into the procedure to calculate the design demand for the structure
with MR dampers. The procedure is evaluated by comparing the predicted design demand to the seismic
response determined from nonlinear time-history analysis.

*Email: ychae@odu.edu

Page 1 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

Simplified Design Procedure (SDP)

In the SDP developed by Lee et al. (2005, 2009), the supplemental damper properties are represented by b,
which is the ratio of the damper stiffness per story in the global direction to the lateral load resisting frame
story stiffness, k0, without dampers and braces of the structural system. The structural system with
dampers is converted into a linear elastic system characterized by the initial stiffness of the structure, a
(the ratio of brace stiffness per story in the global direction to the lateral load resisting frame story stiffness
k0), b, and
. a, b, and
may vary among the stories of the structure. By conducting an elastic-static
analysis with the RSA method, the design demand for the structure is determined.
Since the loss factor of an MR damper depends on the displacement of the structure, as can be shown
below in Eq. 8, the SDP for an elastomeric damper developed by Lee et al. (2005, 2009) needs to be
modified for structures with MR dampers. The loss factor
is associated with the energy dissipation of the
damper over a cycle. For the purpose of determining the energy dissipation over a cycle of displacement,
the properties of the MR damper are assumed to remain constant. The linearization and energy dissipation
of an MR damper are discussed later.
Figure 1 summarizes the SDP for structures with MR dampers. In Step 1, the seismic performance
objectives and associated design criteria are established for the design of the structure. In Step 2, the

1. Establish performance objectives

and design criteria

2. Design structure satisfying

strength requirement without
dampers

3. For each value of a and b, develop

preliminary design by conducting
elastic-static analysis with a simple
frictional damper model (h =4/p) for
each a and b Revise performance
objectives and/or
design criteria if design
is unsatisfactory
4. Select the MR approximate
damper capacity from the smallest
b value that meets the criteria of
Step 1

5. Perform elastic-static analysis with

MR damper properties from
selected MR damper utilizing the
Hershel-Bulkley quasi-static model
& simplified analysis procedure Adjust size of members or
MR dampers
(size, number of dampers,
location, etc)

Meet the criteria No

of Step 1 ?

Yes

End

Fig. 1 Generalized simplified design procedure (SDP) for structures with MR dampers

Page 2 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

structure is designed without MR dampers in accordance with the design code selected in Step 1 to satisfy
the strength requirement for the members in the structure. In Step 3, the MR dampers are incorporated into
the design of the structure to satisfy the specified performance objectives. The design demand for the
structure is estimated for a range of selected values for a, b, and a constant loss factor
¼ 4=p using a
simplified analysis procedure, discussed later. In the simplified analysis procedure, the MR damper
behavior is based on the simple frictional MR damper model (Chae 2011). The required MR damper
sizes are then selected in Step 4 based on the smallest b value that meets the design criteria and
performance objectives in Step 1. Since the simple frictional damper model does not account for the
velocity dependent behavior of an MR damper, a more accurate determination of the design demand is
determined in Step 5 using a more sophisticated MR damper model (i.e., Hershel-Bulkley model) in the
simplified analysis procedure. The design is then revised with final member sizes and the MR damper
sizes are selected (location, number, force capacity, etc.). If the performance objectives cannot be met in
an economical manner, then the performance objectives and/or structural system design need to be revised
as indicated in Fig. 1.

Equivalent Linear System for an MR Damper

The SDP requires that the structure with the nonlinear MR dampers be linearized. In order to linearize the
system for estimating the response of structures with MR dampers, the Hershel-Bulkley quasi-static MR
damper model is used. Figure 2 shows the force-velocity and force-displacement relationships for the
model where the damper force f is given as

f ¼ signðu_ Þff0 þ C ju_ jn g (1)

In Eq. 1 u is the damper displacement relative to the initial position of the damper and u_ is the damper
velocity. f0 is the frictional force. C and n are the coefficients of the nonlinear dashpot. Suppose that the
MR damper is subjected to a harmonic displacement motion

uðt Þ ¼ u0 sin ðot Þ (2)

where u0 is the amplitude of displacement and o is the excitation frequency of the damper. The energy
dissipated by the damper over one cycle of the harmonic motion is equal to

a f b f
fmax
fmax
f0
f0

. .
u0 u u0 u

Fig. 2 Hershel-Bulkley visco-plasticity MR damper model: (a) force-velocity relationship; (b) force-displacement
relationship

Page 3 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

a b y x
x
m
kbr
MR
damper MR damper
c
ko m

kbr
ko

Fig. 3 SDOF system: (a) schematic of equivalent SDOF system with MR damper and brace; (b) mechanical model

ð 2p
o
E M RD ¼ f ðt Þu_ðtÞdt ¼ 4f0 u0 þ 2nþ2 CgðnÞu1þn
0 o
n
(3)
0

where
 n
G 1þ2

gðnÞ ¼ 2 (4)
Gð2 þ nÞ

In Eq. 4, G() is the gamma function (Soong and Dargush 1997). In general, diagonal bracing is installed in
the building in series with the dampers. Therefore, the energy dissipation of an MR damper needs to be
studied considering the stiffness of the diagonal bracing. Figure 3 shows an SDOF system with an MR
damper and diagonal bracing. Under the harmonic motion x ¼ x0 sin ðot Þ , the maximum damper
displacement, ud0, and velocity, u_ d0 , of the MR damper occurs when the displacement x and velocity ẋ
are a maximum, respectively, where ud0 and u_ d0 can be calculated as (Chae 2011)

ud0 ¼ x0  f0 =k br (5a)

u_ d0 ¼ x_0 ¼ x0 o (5b)

In Eq. 5a kbr is the stiffness of the diagonal bracing. Substitution of Eq. 5a into Eq. 3 results in the
expression for the energy dissipation of the MR damper for a SDOF system

E M RD ¼ 4f0 ud0 þ 2nþ2 CgðnÞunþ1

d0 o
n
(6)

The strain energy of the MR damper, ES, is calculated from the equivalent stiffness of the damper and the
maximum damper displacement:

1
E S ¼ k eq u2d0 (7)
2
where keq is the equivalent stiffness of the MR damper, defined as k eq ¼ f0 =ud0 and based on the secant
stiffness from the force-displacement relationship of the Hershel-Bulkley model. The loss factor
of the
MR damper by definition is

1 EM RD 4ff0 þ CgðnÞð2ud0 oÞn g

¼ ¼ (8)
2p E S pk eq ud0

Page 4 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

Simplified Analysis Procedure

The simplified analysis procedure provides an elastic-static method for calculation of the design demand
of an MDOF system with MR dampers. The simplified analysis procedure utilizing the response spectrum
analysis (RSA) method is summarized in Figs. 4 and 5.
In the simplified analysis procedure, the maximum structural displacements are determined by the well-
known equal displacement rule. In the equal displacement rule, the maximum displacement of the
nonlinear structure, whose lateral stiffness is based on its initial tangent stiffness, is assumed to be
equal to that of a linear structure. The equal displacement rule is only applicable to structures that lie in
the low-frequency and medium-frequency spectral regions (Newmark and Hall 1973).
In order to obtain the equivalent period for an MDOF structure with MR dampers, the combined
stiffness of the MR dampers and diagonal bracing needs to be added to the stiffness of the structure. Thus,
the global effective stiffness of the MDOF system is given as

Given:
∑ MR damper properties: , , ( : index for MR damper location)
∑ Structural properties: , , , (inherent damping ratio of the -th mode)
Step 1. Assume and set = (fundamental frequency of structure without MR dampers)

Step 2. Determine maximum damper displacements

= − /
th
: maximum deformation of damper and bracing of the MR damper

Step 3. Calculate equivalent stiffness of each MR damper

= /

Step 4. Determine K = for each MR damper and update

Step 5. Update modal frequency and modal vector ( = 1,…, )

, Á eig( , )
where is the mass matrix of the structure

Step 6. Calculate loss factor of each MR damper

4 + ( ) 2
=

where is the fundamental modal frequency

Step 7. Perform modal analysis described in Figure 5

Step 8. Apply the modal combination rule (e.g., SRSS or CQC) to get the final displacement and velocity of each MR
.
damper
= function of ( ,…, ),
. . .
= function of ( ,…, )

Step 9. Repeat Step 2 ~ Step 8 until convergence is achieved.

Step 10. Calculate maximum damper force for each MR damper

.
= +

Fig. 4 Simplified analysis procedure used to design MDOF structures with passive MR dampers utilizing response spectrum
analysis (RSA) method

Page 5 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

th
Step 7. For =1 to mode

Substep 1. Assume modal displacement vector

Substep 2. Determine maximum displacement for each damper
= − /
th
: maximum deformation of damper and bracing at the MR damper in mode
Substep 3. Calculate equivalent modal damping ratio

Σ ( )
= +

Substep 4. Find maximum modal displacement ( ) from response spectrum

= ( , ) where = 2 /

Substep 5. Update modal displacement vector

=
: mode vector; Γ = / M ; 1: unit vector; M : modal mass (= )

Substep 6. Repeat Substep 2 ~ 5 until convergence in is achieved

th
Substep 7. Calculate maximum velocity for each MR damper, where for the MR damper
=

Fig. 5 Modal analysis method for the simplified analysis procedure utilizing response spectrum analysis (RSA) method

Keff ¼ K0 þ Kbrsystem (9)

where K0 is the stiffness of the structure without diagonal braces and MR dampers, and Kbrsystem is the
stiffness associated with the braces and MR dampers. The structure is assumed to have N DOF, thus the
dimension of Keff is N  N. The combined stiffness Kibrsystem of the diagonal bracing and MR damper at
the i-th MR damper location is

k ibr k ieq
Kibrsystem ¼ (10)
k ibr þ k ieq

where kibr and kieq are the horizontal stiffness of the diagonal bracing and the MR damper associated with
the ith MR damper, respectively. kieq can be calculated utilizing the secant stiffness method as noted above.
The individual combined stiffnesses based on Eq. 10 are appropriately assembled to form Kbrsystem. The
effective periods and mode shapes of the structure are then obtained by performing an eigenvalue analysis
of the structure, considering the seismic mass of the structure.
The equivalent damping ratio xeq of an MDOF system is determined using the lateral force energy
method proposed by Sause et al. (1994), where
XL
1
F i ui
i¼1 i d d
xeq ¼ þ xin (11)
2 FT x0

In Eq. 11
i and uid are the loss factor and maximum damper displacement of the ith MR damper,
respectively, and L is the number of MR dampers. Since the damper displacement is unique for each MR
damper, the loss factor of each MR damper, which is a function of damper displacement, is also unique for
each damper. For each damper
i is obtained from Eq. 8. xin in Eq. 11 is the inherent damping ratio and x0

Page 6 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

is the vector of the displacements of the structure that develop under the lateral force F. The individual
damper force Fid and the lateral force vector F are defined as

F id ¼ k ieq uid , F ¼ Keff x0 (12)

In the simplified analysis procedure using the RSA, the relationships in Eq. 12 are substituted into Eq. 11,
and the inherent damping xin and x0 from each mode are considered, as indicated in Substep 3 in Fig. 5.

Performance-Based Design of a Three-Story Building with MR Dampers

Prototype Building Structure
Based on the proposed SDP, a three-story building with MR dampers is designed. The floor plan and
elevation of the prototype structure is shown in Figs. 6 and 7, respectively. It consists of a three-story,
six-bay building and represents a typical office building located in Southern California. Lateral loads are
resisted by a four perimeter moment resisting frames (MRFs) and four damped braced frames (DBFs) in
the two orthogonal principle directions of the building’s floor plan. MR dampers are installed in the DBFs

MRF MRF

DBF
6@25⬘=150⬘

6@25⬘=150⬘

Fig. 6 Floor plan of prototype building

MRF MRF
3
4@12.5⬘=50.0⬘

2
Gravity
frame
1
Ground

Base

6@25⬘=150⬘

Fig. 7 Elevation of prototype building

Page 7 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

to control the drift of the building, adding supplemental damping to the structure. The DBFs have
continuous columns, with pin connections at the beam-to-column connections and at the ends of the
diagonal bracing. A rigid diaphragm system is assumed to exist at each floor level and the roof of the
building to transfer the floor inertia loads to the MRFs and DBFs. The building has a basement where a
point of inflection is assigned at one third of the height of the column from the column base in the analysis
model.

Performance Objectives
In this design, three different performance objectives for the building are considered:

1. Limit the story drift to 1.5 % under the DBE ground motion
2. Limit the story drift to 3.0 % under the MCE ground motion
3. Design strength of members in the DBF shall not be exceeded by the demand imposed by the DBE
ground motion

The MCE ground motion is represented by a response spectra that has a 2 % probability of exceedance
in 50 years, and the DBE ground motion is two third the intensity of the MCE ground motion (FEMA
2000a). The performance objectives of 1.5 % story drift satisfies the life safety performance level under
the DBE, and the 3 % story drift satisfies the collapse prevention level under the MCE. The performance
levels are defined in FEMA-356 (2000b). To minimize the damage and repair cost to the DBF structure,
the third performance objective is adopted to have the DBF structure remain elastic under the DBE.

Performance-Based Design
The prototype building structure is intended to provide the basis for an MRF and DBF with MR dampers
which can be constructed in the laboratory for future tests. Due to laboratory constraints, the prototype
building structure and resulting MRF and DBF were designed at 0.6-scale. The MRFs are designed to
satisfy the strength requirement of the current building seismic code of ICC (2006); the member design
criteria is based on the AISC steel design provisions (2005b). The design response spectrum is based on a
site in Southern California where the spectral acceleration for the short period, SS, and for a 1 s period, S1,
are equal to 1.5 and 0.6, respectively. The strength contribution from the DBFs and MR dampers is not
considered when the MRFs are designed since, as noted above, the DBFs and MR dampers are intended
only to control the story drift of the building system. More detailed information on the design of the MRFs
and gravity frames can be found in Chae (2011).
Once the MRFs and gravity frames are designed for strength, the required capacity of the MR dampers
to control the drift is determined. The DBF members are then designed by imposing the displacement and
damper force demands on the DBF, which are obtained from the simplified analysis procedure and the
required MR damper capacity. The maximum displacements and the maximum MR damper forces are
assumed to occur concurrently in the SDP. The design of the three-story building is revised until the
performance objectives and strength requirements are satisfied.
Large-scale MR dampers were used for the study which can generate a 200 kN damper force at a
velocity of 0.1 m/s (Chae et al. 2010). The parameters for the Hershel-Bulkley model associated with the
large-scale MR damper are: f0 ¼ 138:5 kN, C ¼ 161:8 kNsec/m, and n ¼ 0:46.
The optimal damper location which satisfies the performance objectives is determined by using the
simplified analysis procedure, resulting in one large-scale MR damper in the second and third stories,
respectively, with a ¼ 10 and b ¼ 0:3 (Chae 2011). Tables 1 and 2 summarize the member sizes for the
MRFs, gravity frames, and DBFs. Table 3 summarizes the calculated design demand associated with
maximum story drift and maximum damper forces. As can be observed, the design demand for the story

Page 8 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

Table 1 Member sizes for MRFs and gravity frames

Story (or floor level) MRFs Gravity frames
Column Beam Column Beam
1 W8X67 W18X46 W8X48 W10X30
2 W8X67 W14X38 W8X48 W10X30
3 W8X67 W10X17 W8X48 W10X30

Table 2 Member sizes for DBFs

Story (or floor level) Column Beam Diagonal bracing
1 W10X33 W10X30 
2 W10X33 W10X30 W6X20
3 W10X33 W10X30 W6X20

Table 3 Calculated design demand associated with maximum story drift and maximum damper force from SDP
Maximum story drift (%) Maximum damper force (kN)
Story DBE MCE DBE MCE
1 1.18 1.91  
2 1.35 2.32 222.9 244.4
3 1.41 2.57 233.6 261.6

Table 4 Demand-to-capacity ratio for DBF frames, DBE

Column Beam Brace
Story (or floor level) (W10X33) (W10X30) (W6X20)
1 0.955 0.521 
2 0.303 0.576 0.270
3 0.079 0.354 0.283

drift under the DBE and MCE are less than 1.5 % and 3.0 %, respectively, in order to satisfy the
performance objectives. Table 4 shows the DBE demand-to-capacity ratios for the DBF members. The
demand-to-capacity ratio for each member is less than 1.0, which means the members are designed to
remain elastic under the DBE. The design of the braces was controlled by stiffness, a ¼ 10, and not
strength, hence, the demand-to-capacity ratios for the braces are small in Table 4.

Assessment of Simplified Design Procedure

The SDP is assessed by comparing the design demand from the SDP with results from a series of
nonlinear time-history analyses (NTHA) of the three-story building using the nonlinear finite element
program OpenSees (2009).

OpenSees Model
Symmetry in the floor plan and ground motions along only one principal axis of the building were
considered in the analysis. Hence, only one-quarter of the building was modeled consisting of one MRF,
one DBF, and the gravity frames that are within the tributary area of the MRF and DBF. The OpenSees

Page 9 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

15⬘ 15⬘
rigid floor diaphragm

W3
fiber element

element
fiber
MNS MR damper
model
W2
3@7.5⬘=22.5⬘

t
en

element
element
panel zone

elastic
em

fiber
element el
ss
tru
W1
truss element

element
fiber
truss element
5⬘

MRF DBF
Lean-on
Column

Fig. 8 OpenSees model for 0.6-scale building

model is shown in Fig. 8. The beams and columns of the MRF structure are modeled with a nonlinear
distributed plasticity force-based beam-column element with five fiber sections along the element length.
The cross section of the element is discretized into 18 fibers, including 12 fibers for the web and 3 fibers
each for the top and bottom flanges. Each fiber is modeled with a bilinear stress–strain relationship with a
post-yielding stiffness that is 0.01 times the elastic stiffness. The beam-column joints in the MRF are
modeled using a panel zone element, where shear and symmetric column bending deformations are
considered (Seo et al. 2009). The doubler plates in the panel zones of the MRF are included in the model.
The nonlinear force-based fiber element is also used to model the columns of the DBF. The beams and
braces of the DBF are modeled using linear elastic truss elements. The gravity frames are idealized using
the concept of a lean-on column, where an elastic beam-column element with geometric stiffness is used
to model the lean-on column. The section properties of the lean-on column are obtained by taking the sum
of the section properties of each gravity column within the tributary area (i.e., one quarter of the floor plan)
of the MRF and the DBF. The MR dampers are modeled using the MNS MR damper model implemented
into OpenSees by Chae et al. (2010). The MR damper is assumed to be located between the top of the
brace and the adjacent beam-column joint, as shown in Fig. 8. The results reported in this paper are for MR
dampers that are passive controlled with a constant current input of 2.5 A. Studies with the MR dampers in
semi-active control mode are presented in Chae (2011).
The gravity loads from the tributary gravity frames are applied to the lean-on column to account for the
P-D effect of the building. To model the effect of the rigid floor diaphragm, the top node of the panel zone
element in the MRF and the beam-column joint in the DBF are horizontally constrained to the node of the
lean-on column at each floor level, while the vertical and rotational degrees of freedom of these nodes are
unconstrained.
Rayleigh damping is used to model the inherent damping of the building with a 5 % damping ratio for
the first and second modes.

Page 10 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

Comparison of Response
An ensemble of 44 ground motions listed in FEMA P695 (ATC 2009) is scaled to the DBE and MCE
levels using the procedure by Somerville et al. (1997) for the NTHA.
A summary of the median and standard deviation of maximum story drift and residual story drift from
the NTHA is given in Table 5. Figures 9 and 10 compare the calculated design demand for drift from the
SDP with the median values for maximum story drift from the NTHA under the DBE and MCE ground
motions. The story drift design demand calculated by the SDP shows good agreement in Fig. 9 with the
median maximum story drifts from the NTHA. The calculated design demand for story drift from the SDP
also shows good agreement with the median values from the NTHA under the MCE, see Fig. 10. Figures 9
and 10 show that the median values of the maximum story drift from the NTHA satisfies the performance
objectives of 1.5 % and 3.0 % under the DBE and MCE levels, respectively. The residual story drift of the
building after the DBE has a maximum median value and standard deviation of 0.22 % and 0.27 %,
respectively, which occurred in the third story as summarized in Table 5. The residual drift is small.
Table 6 compares the design demand for maximum MR damper forces calculated by the SDP with the
median maximum MR damper forces from the NTHA. The MR damper force design demands from the
SDP are slightly smaller than the median NTHA results for the DBE. However, the differences between
the SDP and the NTHA are only 3.9 % and 3.3 % for the MR dampers in the second and third stories,
respectively. For the MCE, the differences between the median NTHA results and the SDP for the MR
damper forces in the second and third stories are 1.6 % and 0.5 %, respectively. The design demand

Table 5 Median and standard deviation of maximum and residual story drift from nonlinear time history analysis
DBE level MCE level
Story Max story drift (%) Residual drift (%) Max story drift (%) Residual drift (%)
1 1.18 (0.35)a 0.11 (0.21) 1.86 (0.85) 0.42 (0.62)
2 1.35 (0.36) 0.17 (0.26) 2.10 (0.85) 0.57 (0.66)
3 1.46 (0.33) 0.22 (0.27) 2.32 (0.84) 0.63 (0.69)
a
Value in () indicates standard deviation response

SDP
3rd
story
NTHA (median)

NTHA (median + stdev)

SDP
2nd
story NTHA (median)

Performance
objective
(1.5%)
SDP
1st
story NTHA (median)

0 0.5 1 1.5 2

Story drift (%)

Fig. 9 Comparison of story drift between SDP and NTHA, DBE ground motions

Page 11 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

SDP
3rd
story
NTHA (median)

NTHA (median + stdev)

SDP
2nd
story NTHA (median)

Performance
objective
(3.0%)
SDP
1st
story NTHA (median)

0 0.5 1 1.5 2 2.5 3 3.5

Story drift (%)

Fig. 10 Comparison of story drift between SDP and NTHA, MCE ground motions

Table 6 Comparison of maximum damper forces

DBE level MCE level
Story SDP NTHA SDP NTHA
1    
2 222.9 231.9 (6.4)a 244.4 248.4 (7.3)
3 233.6 241.5 (8.3) 261.6 260.2 (9.4)
a
Value in () indicates standard deviation response

Table 7 Median and standard deviation of DBF maximum magnitude of column plastic rotation from nonlinear time-history
analysis and DBE ground motions
Story Location along column Max plastic rotation (rad %) Residual plastic rotation (rad %)
1 Bottom 0.00 (0.05)a 0.00 (0.04)
Top 0.00 (0.00) 0.00 (0.00)
2 Bottom 0.00 (0.00) 0.00 (0.00)
Top 0.00 (0.00) 0.00 (0.00)
3 Bottom 0.00 (0.00) 0.00 (0.00)
Top 0.00 (0.00) 0.00 (0.00)
a
Value in () indicates standard deviation response

calculated by the SDP shows reasonably good agreement with the median results from the NTHA for the
maximum MR damper forces.
The linear elastic behavior of the DBF columns under the DBE is confirmed by checking the plastic
rotation developed in the columns. Summarized in Table 7 are the median and standard deviation of the
DBF maximum magnitude of column plastic rotation from the NTHA for the DBE ground motions. In the
first story, some plastic rotation developed at the base of the column under the DBE. However, the median
is zero and the standard deviation is 0.0005 rad for the maximum plastic rotation. The median and
standard deviation of the maximum plastic rotation in the second and third stories columns are zero under
the DBE ground motion, as given in Table 7, which indicates linear elastic behavior of the columns at

Page 12 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

these stories. The median of the residual plastic rotation at the base of the column is zero and the standard
deviation for the residual plastic rotation is 0.0004 rad.
The building’s response under the DBE and MCE determined by the NTHA appears to have met the
performance objectives for the structure. Moreover, the performance under the DBE meets the immediate
occupancy level (FEMA 2000b), where the maximum residual drift is practically equal to the allowable
construction tolerance of 0.2 % for steel structures (AISC 2005a), with minimal damage occurring in the
building (the median of the maximum plastic rotations in the beams and columns of the MRF are 0.37 %
and 0.07 % radians, respectively, with the MRF having the same residual drift statistics as the DBF due to
the rigid diaphragm (Chae 2011)).

Summary
A simplified design procedure was developed to enable the performance-based design of structures with
MR dampers. The design procedure utilizes a systematic approach to calculate the design demand of
structures with MR dampers. The simplified analysis procedure enables the design demand to be
determined without performing a nonlinear time-history analysis by linearizing the structure and utilizing
the response spectrum analysis method.
To linear the nonlinear MR dampers, the simplified analysis procedure is based on an equivalent linear
MR damper model using the Hershel-Bulkley quasi-static MR damper model. The energy dissipated by
the MR damper over one cycle of a harmonic motion is calculated and the equivalent stiffness determined
based on the secant stiffness method from the damper force-displacement relationship. The loss factor of
the MR damper is obtained from the energy dissipated by the damper and the strain energy calculated
from the equivalent stiffness of the damper. Both the equivalent stiffness and loss factor of the MR damper
are dependent on the maximum displacement of the damper. A three-story building was designed using
the SDP, where three performance objectives associated with two seismic hazard levels were selected. The
SDP was assessed by comparing the design demand calculated by the SDP with the response determined
from nonlinear time-history analyses. The MNS MR damper model was implemented into the OpenSees
computer program and statistics for the response to DBE, and MCE ground motions were obtained from a
series of nonlinear time-history analyses using 44 different ground motions. The performance of the
building from the nonlinear time-history analysis indicated that building design satisfies the three
performance objectives. The design demand associated with story drift and maximum MR damper forces
from the SDP showed good agreement with the median values from the nonlinear time-history analyses,
confirming the robustness of the SDP.

References
AISC (2005a) Code of standard practice for steel buildings and bridges. American Institute of Steel
Construction, Chicago
AISC (2005b) Seismic provisions for structural steel buildings. American Institute of Steel Construction,
Chicago
Applied Technology Council (2009) Quantification of building seismic performance factors. ATC-63
project report (FEMA P695), Redwood City
Chae Y (2011) Seismic Hazard mitigation of building structures using magneto-rheological dampers,
PhD dissertation, Lehigh University

Page 13 of 14
Encyclopedia of Earthquake Engineering
DOI 10.1007/978-3-642-36197-5_310-1
# Springer-Verlag Berlin Heidelberg 2021

Chae Y, Ricles JM, Sause R (2010) Development of a large-scale MR Damper model for seismic hazard
mitigation assessment of structures. In: 9th US National and 10th Canadian conference on earthquake
engineering, Toronto
Constantinou MC, Soong TT, Dargush GF (1998) Passive energy dissipation systems for structural design
and retrofit. Monograph series, MCEER. State University of New York at Buffalo, Buffalo
Fan CP (1998) Seismic analysis, behavior, and retrofit of non-ductile reinforced concrete frame buildings
with viscoelastic dampers. PhD dissertation, Lehigh University, Bethlehem
Federal Emergency Management Agency (2000a) Recommended seismic design criteria for new steel
moment-frame buildings. Report no. FEMA-350, Washington, DC
Federal Emergency Management Agency (2000b) Prestandard and commentary for the seismic rehabil-
itation of buildings. Report no. FEMA-356, Washington, DC
International Code Council (2006) International building code. International Code Council, Falls Church
Kwan WP, Billington SL (2003) Influence of hysteretic behavior on equivalent period and damping of
structural systems. ASCE J Struct Eng 129(5):576–585
Lee K-S, Fan C-P, Sause R, Ricles J (2005) Simplified design procedure for frame buildings with
viscoelastic and elastomeric structural dampers. Earthq Eng Struct Dyn 34:1271–1284
Lee K-S, Ricles J, Sause R (2009) Performance-based seismic design of steel MRFs with elastomeric
dampers. J Struct Eng 135(5):489–498
Lin WH, Chopra AK (2003) Earthquake response of elastic Single-degree-of-freedom systems with
nonlinear viscoelastic dampers. ASCE J Eng Mech 129(6):597–606
Newmark NM, Hall WJ (1973) Seismic design criteria for nuclear reactor facilities. Report
no. 46, Building Practices for Disaster Mitigation, National Bureau of Standards, U.S. Department of
Commerce, pp 209–236
OpenSees (2009) Open system for earthquake engineering simulation. Pacific Earthquake Engineering
Research Center, University of California, Berkeley
Ramirez OM, Constantinou MC, Gomez JD, Whittaker AS (2002) Evaluation of simplified methods of
analysis of yielding structures with damping systems. Earthq Spectra 18(3):501–530
Sause R, Hemingway GJ, Kasai K (1994) Simplified seismic response analysis of viscoelastic-damped
frame structures. In: 5th U.S. National conference on earthquake engineering, Vol. I, pp 839–848
Seo CY, Lin YC, Sause R, Ricles JM (2009) Development of analytical models for 0.6 scale self-centering
MRF with beam web friction devices. In: 6th International conference for steel structures in seismic
area (STESSA), Philadelphia
Somerville P, Smith N, Punyamurthula S, Sun J (1997) Development of ground motion time histories for
Phase 2 of the FEMA/SAC steel project. Report no. SAC/BD-97/04, SAC Joint Venture, Sacramento
Soong TT, Dargush GF (1997) Passive energy dissipation systems in structural engineering. Wiley, West
Sussex
Symans MD, Constantinou MC (1998) Passive and fluid viscous damping systems for seismic energy
dissipation. ISET J Earthq Technol 35(4):185–206

Page 14 of 14