Seismic Design of Friction-Damped Precast Concrete

Frame Structures
Brian G. Morgen1 and Yahya C. Kurama, M.ASCE2

Abstract: This paper presents the seismic design of unbonded posttensioned precast concrete frame structures that use friction dampers
for supplemental energy dissipation. A procedure is described to determine the friction damper slip forces and posttensioning steel areas
needed to satisfy prescribed design lateral strength and energy dissipation requirements for a trial frame with given geometry, beam and
column member dimensions, and selected damper distribution. The proposed design procedure assumes that the lateral strength require-
ments for the frame have been obtained from a linear elastic analysis of the structure under equivalent lateral forces. Nonlinear reversed-
cyclic analyses of friction-damped precast concrete beam-column subassemblies and multistory frame structures under lateral loads are
conducted to critically evaluate the design procedure and to identify areas where improvement may be needed. The analytical results show
that friction-damped precast frames can be designed to achieve significant energy dissipation levels while maintaining a large level of
self-centering capability due to the posttensioning force.
DOI: 10.1061/共ASCE兲0733-9445共2007兲133:11共1501兲
CE Database subject headings: Concrete, posttensioned; Concrete, precast; Concrete, prestressed; Concrete, reinforced; Damping;
Energy; Frames; Seismic analysis; Seismic design.

Introduction and Background
Since the 1990s, a significant amount of research has been con-
ducted on the seismic behavior and design of precast concrete
structures that do not emulate the behavior of monolithic cast-
in-place reinforced concrete construction. One of the “non-
emulative” precast frame systems that have successfully emerged
from these research efforts uses unbonded posttensioning between
the precast beam and column members to achieve the lateral load
resistance needed in seismic regions 共e.g., Cheok and Lew 1993;
Priestley and Tao 1993; Priestley and MacRae 1996; El-Sheikh
et al. 1999兲.
Unbonded posttensioned precast concrete frame structures
have desirable seismic characteristics such as a large self-
centering capability 共i.e., ability of the structure to return towards
its original undisplaced position upon unloading from a nonlinear
lateral displacement兲 and an ability to undergo large nonlinear
lateral displacements without significant damage. The postten-
sioning steel is placed inside the beams at the floor and roof levels
to provide lateral strength and stiffness to the frame as well as the
restoring force needed to develop the self-centering capability.
1 reversed-cyclic loading were recently conducted at the University
Project Engineer, Magnusson Klemencic Associates, 1301 Fifth Ave.,
Suite 3200, Seattle, WA 98101 共corresponding author兲. E-mail:
bmorgen@mka.com
2 precast beam specimen and Fig. 2共b兲 shows a prototype friction
Associate Professor, Dept. of Civil Engineering and Geological
Sciences, Univ. of Notre Dame, Notre Dame, IN 46556. E-mail:
ykurama@nd.edu
Note. Associate Editor: Michael D. Symans. Discussion open until
April 1, 2008. Separate discussions must be submitted for individual
papers. To extend the closing date by one month, a written request must
be filed with the ASCE Managing Editor. The manuscript for this paper
was submitted for review and possible publication on January 19, 2005;
approved on September 18, 2006. This paper is part of the Journal of
Structural Engineering, Vol. 133, No. 11, November 1, 2007. ©ASCE,
ISSN 0733-9445/2007/11-1501–1511/$25.00.
The greatest setback to the use of unbonded posttensioned
precast concrete frame structures in seismic regions is that, as a
result of small energy dissipation, their lateral displacement de-
mands during a severe earthquake may be larger than acceptable
共Priestley and Tao 1993兲. The research described in this paper
focuses on this issue, with the broad objective of increasing the
energy dissipation of these structures by using supplemental fric-
tion dampers.
As shown in Fig. 1, the friction dampers are placed at selected
beam ends in a frame. In the event of a large earthquake, gap
opening occurs at the joints between the precast concrete beam
and column members. In addition to providing a part of the lateral
strength of the frame, the friction dampers utilize the gap opening
displacements at the beam ends to develop slip displacements at
the friction interfaces and provide supplemental energy dissipa-
tion to the frame 共Morgen and Kurama 2004a,b兲. Furthermore, the
dampers can be designed not to slip under service loads due to
wind and small earthquakes, providing additional resistance
against gap opening at the beam-to-column interfaces, and thus,
additional lateral stiffness to the frame.
Large-scale experiments of friction-damped unbonded postten-
sioned precast concrete beam-column subassemblies under
of Notre Dame 共Morgen and Kurama 2004b兲. As an example, Fig.
2共a兲 shows the displaced shape at the end of a friction-damped
damper. Due to space limitations in the laboratory, the beam was
oriented in a vertical configuration with the column placed hori-
zontally. The gap opening behavior at the beam-to-column inter-
face and the resulting slip displacements in the dampers can be
seen in Fig. 2共a兲. Figs. 2共c and d兲 show the hysteretic beam end
moment 共M b兲 versus beam chord rotation 共␪b兲 results from a test
with friction dampers 共Test 44兲 and a baseline test with no damp-
ers 共Test 43兲, respectively.
The hysteresis loops in Fig. 2共d兲 show that the specimen with-

JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007 / 1501

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
 Fig. 1. Friction-damped precast concrete frame 共adapted from Morgen and Kurama 2004b, with permission兲

out dampers behaves in an essentially elastic manner through analytical study on the seismic design of friction-damped un-
nonlinear displacements 共i.e., nonlinear-elastic兲, with very little bonded posttensioned precast concrete frame structures. The
energy dissipation but extremely good self-centering capability. focus of the paper is the determination of the damper slip forces
As shown in Fig. 2共c兲, the energy dissipation of the subassembly and posttensioning steel areas needed to achieve prescribed de-
can be significantly increased by using friction dampers, while sign lateral strength and energy dissipation requirements for a
preserving the desirable self-centering capability. As an additional frame. The proposed procedures consider in-plane loading only
benefit, the hysteresis loops of the friction-damped specimen and are intended for the design of structures that can be classified
show significantly reduced stiffness degradation under reversed- as “regular” based on current building code requirements 共e.g.,
cyclic loading. ICC 2006兲. For “irregular” structures and for structures with sig-
The dampers that were developed as part of this research use nificant torsional response or torsional loading, three-dimensional
friction forces between metallic surfaces. Knowledge gained from dynamic time-history analyses may need to be conducted.
past investigations of friction dampers in structural applications It is assumed that the design beam end moment demands have
共e.g., Grigorian et al. 1993; Way 1996兲 led to the use of two types been determined from a linear-elastic analysis of the structure
of friction interfaces: 共1兲 leaded-bronze against stainless steel; under equivalent lateral forces 共e.g., as described in ICC 2006兲.
and 共2兲 leaded-bronze against alloy steel. These configurations Recommendations on the energy dissipation requirements for
were previously shown to provide consistent and repeatable concrete moment frame structures can be found in ACI T1.1-01
damper slip force versus displacement characteristics. The leaded- Acceptance criteria for moment frames based on structural test-
bronze surfaces at the friction interfaces were created by sand- ing 共ACI Innovation Task Group 1 2001a兲. The design approach
wiching 12.7-mm-thick leaded-bronze disks 共CDA 932/SAE 660 developed in this paper provides a methodology to satisfy these
bearing bronze—a popular alloy for bearing applications under requirements by using friction dampers at selected beam ends of
moderate loads and moderate to high speeds兲 at the friction inter-
faces. According to Way 共1996兲, the use of a leaded-bronze alloy
at the friction interface is desirable because the material continu-
ally “self-lubricates” when rubbing against an adjacent metal sur-
face due to the development of a mixture of lead and its oxide
at the surface of the leaded-bronze alloy. By self-lubricating the
frictional surface, the leaded-bronze alloy helps reduce the phe-
nomenon of stick-slip and results in a consistent value for the
coefficient of friction.
Full-scale experiments of isolated friction dampers under dy-
namic loading 共Morgen and Kurama 2004a兲 have shown that
some of the benefits of using friction as a supplemental energy
dissipation mechanism for seismic protection include: 共1兲 repeat-
able and reliable hysteretic behavior that is relatively independent
of velocity and displacement amplitude 共assuming that the damp-
ers are used indoors under controlled environmental conditions兲;
共2兲 close-to-rectangular damper force versus displacement char-
acteristics providing a large amount of energy dissipation per
cycle; and 共3兲 large damper initial stiffness allowing slip to occur
early in the response, and thus, provide energy dissipation begin-
ning at small lateral displacements of the frame.

Research Significance and Scope Fig. 2. Research results adapted from Morgen and Kurama 共2004b,
with permission兲: 共a兲 friction-damped beam end displaced shape; 共b兲
Based on the recent experimental investigations summarized prototype damper; 共c兲 behavior of beam with friction dampers; and
above 共Morgen and Kurama 2004a,b兲, this paper describes an 共d兲 behavior of beam without friction dampers

1502 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
an unbonded posttensioned precast concrete frame.
Previous research on the use of supplemental energy dissipa-
tion in precast concrete construction is limited and has primarily
focused on shear wall and braced frame structures. In comparison,
the dampers described in this paper are used locally at the beam
ends, without the need for bracing members in the structure. The
application of similar “local-damper” configurations has been in-
vestigated for cast-in-place reinforced concrete frames 共Way
1996兲, steel frames 共Christopoulos et al. 2002; Garlock et al.
2005兲, and coupled walls 共Kurama et al. 2006; Weldon and
Kurama 2007兲. The use of mild 共e.g., Grade 60兲 steel reinforce-
ment through the precast beam-column joints, in addition to the
posttensioning steel, has also been investigated 共Stone et al. 1995;
Priestley et al. 1999; ACI Innovation Task Group 1 2001b兲 and
successfully applied in practice 共Englekirk 2002兲.
The proposed design approach is developed below in two
steps: 共1兲 design of friction-damped beam ends; and 共2兲 design of
multistory frames. Nonlinear reversed-cyclic analyses of friction-
damped beam-column subassemblies and multistory frame struc- Fig. 3. Equilibrium of forces at friction-damped beam end
tures under lateral loads are conducted to critically evaluate the
design procedures and identify areas where improvement may be
needed. The research is described in full detail in Morgen and ␮ = coefficient of friction for the damper slip interfaces 共Morgen
Kurama 共2007兲, including the mechanical design of the friction and Kurama 2004a,b兲; and Fdn = damper normal force 共i.e.,
dampers and nonlinear dynamic time-history analyses of frames the damper “clamping” force acting normal to the friction slip
with and without dampers under earthquake loading. interfaces兲.
In Eqs. 共2兲–共4兲, hb = beam depth; hd = distance from the damper
normal bolt 共Morgen and Kurama 2004a,b兲 to the extreme con-
Design of Friction-Damped Beam Ends crete fiber of the beam; and Cc = concrete compressive stress re-
sultant given by
This section describes the development and analytical verification
of a design procedure to satisfy prescribed flexural strength and Cc = Nb = A p f pi 共6兲
energy dissipation requirements for a friction-damped unbonded
posttensioned precast concrete beam end. where Nb = axial force in the beam; A p = total area of the postten-
sioning steel; and f pi=design initial stress in the posttensioning
steel after losses. Eq. 共6兲 for Cc assumes that all of the initial
Damper Slip Force and Posttensioning Steel Area 共after losses兲 posttensioning force is transferred into the beam and
In order to determine the damper slip force and posttensioning ignores any additional axial forces that develop in the beam due
steel area needed, the nominal moment strength of a friction- to the displacements of the frame.
damped precast concrete beam end is divided into two compo- The variable a used in Eqs. 共2兲 and 共3兲 is the depth of the
nents, M bs and M bp, representing the contributions of the friction assumed uniform concrete compression stress block given by
dampers and the beam posttensioning steel reinforcement, respec-
A p f pi
tively, to satisfy the total design beam end moment demand, M bd a= 共7兲
as 0.85f ⬘c bb

M bd = M bs + M bp 共1兲 where f ⬘c = design unconfined concrete compressive strength and

bb = beam width 共assuming that the cover concrete has not spal-
Using the equilibrium of the forces in Fig. 3 at the beam end, led兲. Additionally, c in Fig. 3 represents the neutral axis depth.
Eq. 共1兲 can be written as In order to determine the required damper slip force and post-
M bs = Fds共hb + 2hd兲 共2兲 tensioning steel area, a new parameter, referred to as the design
damper moment ratio, is defined as

M bp = Cc 冉 冊
hb − a
2
共3兲 ␬d =
M bs
M bp
共8兲

冉 冊
M bd = Fds共hb + 2hd兲 + Cc
hb − a
2
共4兲
Substituting ␬d into Eq. 共1兲 yields

␬d
M bs = M bd 共9兲
where Fds = slip force of the friction dampers 共assumed to be equal 1 + ␬d
for the two dampers兲 given by
Fds = n␮Fdn 共5兲 M bd
M bp = 共10兲
1 + ␬d
It is assumed that Fds acts in a direction parallel to the beam
共as shown in Morgen and Kurama 2004b兲. In Eq. 共5兲, Then, by substituting Eqs. 共9兲 and 共10兲 into Eqs. 共2兲 and 共3兲,
n = number of friction slip interfaces in each damper; respectively, the required damper slip force, Fds, and posttension-

JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007 / 1503

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright

Fig. 4. Beam end moment-rotation behavior: 共a兲 idealized behavior;
共b兲 friction damper contribution; and 共c兲 posttensioning contribution
ing steel area, A p, for a prescribed design beam end moment, M bd
and selected value of the design damper moment ratio, ␬d can be
estimated as
␬dM bd
Fds =
共1 + ␬d兲共hb + 2hd兲
M bd
冉 冊
Ap =
hb − a
f pi共1 + ␬d兲
2 Dh M bs
It is assumed that bb, hb, hd, f ⬘c , and f pi are known for the trial
structure. Note that the determination of the posttensioning steel
area in Eq. 共12兲 requires an iterative process using the rectangular
stress block depth a from Eq. 共7兲.
Relative Energy Dissipation Ratio
Recommendations on the energy dissipation requirements for
concrete moment frame structures can be found in ACI T1.1-01
共ACI Innovation Task Group 1 2001a兲. According to ACI T1.1-
01, the relative energy dissipation ratio, ␤ of a structure should be
larger than or equal to 0.125. The relative energy dissipation ratio
is defined for a beam end moment versus chord rotation cycle as
the ratio of the area Dh enclosed by the hysteresis loop for that
cycle 关e.g., shaded area enclosed by the hysteresis curve in Fig.
4共a兲兴 to the area Ah of the circumscribing parallelogram 关dashed
lines in Fig. 4共a兲兴. The circumscribing parallelogram area Ah is
defined by the initial positive and negative stiffnesses during the
first linear-elastic cycle of loading and the maximum positive and
negative beam end moment resistances during the cycle for which
the relative energy dissipation ratio is calculated.
It can be shown that the relative energy dissipation ratio, ␤, is
a measure of the amount of viscous damping in an equivalent
linear-elastic system that would result in a similar amount of en-
ergy dissipation as the nonlinear system. ACI T1.1-01 recom-
mends that if ␤ is smaller than 0.125, there may be inadequate
damping for the frame as a whole, and the oscillations of the
structure may continue for a considerable time after an earth-
quake, possibly producing low-cycle fatigue effects and excessive
displacements.
Fig. 4共a兲 depicts the idealized hysteretic moment versus rota-
tion 共M b-␪b兲 behavior of a friction-damped beam end satisfying
the design beam end moment demand, M bd. The basis for this
idealized behavior can be seen from Fig. 2共c兲 and from other
experimentally obtained results presented in Morgen and Kurama
共2004a,b兲. Figs. 4共b and c兲 show the idealized contributions from
the friction dampers, M bs and the posttensioning steel, M bp, re-
spectively, to the total moment resistance, M bd in Fig. 4共a兲, where
M bs, M bp, and M bd are assumed to be reached at the same beam
chord rotation. It is also assumed that the beam end moment
1504 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007
Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
contribution from the friction dampers possesses an elastic-
perfectly plastic 共i.e., elastoplastic兲 hysteretic behavior and that
the entire energy dissipation in Fig. 4共a兲 is provided by the fric-
tion dampers. Thus, the beam end moment contribution from the
posttensioning steel is represented using an elastic though nonlin-
ear 共bilinear兲 moment-rotation behavior as shown in Fig. 4共c兲
关i.e., bilinear-elastic behavior, similar to Fig. 2共d兲兴. This bilinear-
elastic component provides self-centering capability to the struc-
ture, while allowing softening 共i.e., stiffness reduction兲 and period
elongation to occur. Note that the nonlinear behavior in Fig. 4共c兲
is governed by the opening of a gap at the beam end 关see Fig.
2共a兲兴, and thus occurs without significant damage to the structure.
For design purposes, a relationship can be developed between
the damper moment ratio ␬d used in Eqs. 共9兲–共12兲 and the relative
energy dissipation ratio, ␤, defined in ACI T1.1-01. Two types of
relative energy dissipation ratios are used in this paper: 共1兲
␤b⫽relative energy dissipation ratio at a beam end; and 共2兲
共11兲
␤⫽relative energy dissipation ratio for an entire multistory frame.
Ignoring the “postyield” stiffness of the idealized moment-
rotation relationship in Fig. 4共a兲 共i.e., assuming ␣ = 0兲, the relative
共12兲 energy dissipation ratio, ␤b at a beam end can be written as
␤b = = 共13兲
Ah M bd
Then, combining Eqs. 共9兲 and 共13兲, the design damper moment
ratio, ␬d can be related to ␤b as
␤b
␬d = 共14兲
1 − ␤b
The proposed design procedure requires that a value for the
beam end relative energy dissipation ratio ␤b be selected 共note
that ␤b 艌 0.125 is required by ACI T1.1-01兲. It is recommended
that ␬d values smaller than 1.0, and thus, ␤b values smaller than
0.5, be used in design to ensure a sufficient amount of self-
centering capability in the structure. Then, the damper slip force
and posttensioning steel area needed at the beam end can be de-
termined by substituting Eq. 共14兲 into Eqs. 共11兲 and 共12兲.
Analytical Verification of Beam Design Equations
The beam design procedure described above is verified based on
nonlinear analyses of a series of 12 cruciform-shaped friction-
damped precast concrete beam-column subassembly models 共ex-
tending between midstory heights of column and midspan lengths
of beams兲 as shown in Fig. 5. It is assumed that four identical
friction dampers are used in each subassembly, with two dampers
at each beam end.
The main parameters varied in the study are the design beam
end relative energy dissipation ratio, ␤b and the design beam end
moment demand, M bd. As shown in Table 1, four relative energy
dissipation ratios 共␤b = 1 / 8, 1 / 4, 5 / 16, and 5 / 12兲 and three beam
end moment demands 共M bd = 678, 1,356, and 2,034 kN m兲 are
used. The increasing M bd demands, which were determined based
on the design demands of a prototype unbonded posttensioned
precast concrete frame structure described in El-Sheikh et al.
共1999兲, are assumed to correspond to increasing beam depths of
hb = 610, 813, and 1,016 mm, respectively.
The required damper slip forces, Fds and posttensioning steel
areas, A p for the 12 subassemblies, as determined using Eqs. 共11兲,
共12兲, 共14兲, and 共7兲, are given in Table 1. It is assumed that the
critical sections of the structures are located at the beam ends

Fig. 5. Beam-column subassembly analytical model 共adapted from
Morgen and Kurama 2004b, with permission兲
where gap opening occurs. Additional structural properties
assumed for the design of the subassemblies are: bb = 610 mm,
hd = 292 mm, f ⬘c = 41 MPa, and f pi = 0.65f pu, where f pu = 1,862
MPa⫽design ultimate strength of the posttensioning steel.
Nonlinear reversed-cyclic lateral load analyses of the subas-
semblies were conducted using the model in Fig. 5 and the
DRAIN-2DX program 共Prakash et al. 1993兲. This analytical
model includes fiber beam-column elements to represent the
precast beam and column members, bilinear truss elements to
represent the unbonded posttensioning steel, and elastic-perfectly
plastic truss elements to represent the friction dampers. More
detailed information on the analytical model, as well as model
verification, can be found in Morgen and Kurama 共2004b兲. It is
assumed that the subassemblies are properly designed such that
the nonlinear behavior is governed by axial-flexural stresses and
gap opening, and all premature failure modes 共e.g., shear failure兲
are prevented.
As examples of representative behavior, Figs. 6共a–d兲 plot the
beam end moment, M b, versus beam chord rotation, ␪b behaviors
corresponding to ␤b = 1 / 8, 1 / 4, 5 / 16, and 5 / 12, respectively, for
the subassembly with hb = 813 mm and M bd = 1,356 kN m. The
four different subassembly designs in Fig. 6 have similar beam
end moment strengths satisfying the design moment demand,
M bd = 1,356 kN m with increasing amounts of energy dissipation,
while maintaining a large self-centering capability due to the
posttensioning force.
The relative energy dissipation ratios, ␤b of the 12 beam-
column subassemblies, as determined from the analysis results
according to the ACI T1.1-01 definition, are listed in Table 2.
Each calculated ␤b ratio in Table 2 represents the average value
from all of the rotation cycles conducted during the analysis of
each structure 共i.e., ␪b = ± 0.5, ±1.0, ±1.5, and ±2.0%兲. It is ob-
served that the calculated ␤b ratios are reasonably close to the
Table 1. Required Subassembly Fds and A p
␤b = 1 / 8
M bd hb Fds Ap
共kN m兲 共mm兲 共kN兲 共mm2兲
2,034 1,016 151 3,626
1,356 813 116 3,065
678 610 67 1,968
Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
prescribed values used in design, with the largest difference be-
tween the calculated and prescribed ␤b ratios being equal to 0.04.
Thus, it is concluded that the proposed design formulation based
on Eqs. 共11兲, 共12兲, 共14兲, and 共7兲 results in a predictable beam end
moment versus beam chord rotation hysteretic behavior, including
the moment strength and energy dissipation characteristics.
Figs. 7共a and b兲 plot the calculated damper slip forces, Fds and
posttensioning steel areas, A p from Table 1 against the calculated
average beam end relative energy dissipation ratios, ␤b from
Table 2. As can be expected from Eqs. 共11兲, 共12兲, and 共14兲, the
relationships between Fds and ␤b and between A p and ␤b are close
to linear, with Fds increasing and A p decreasing as ␤b is increased
for each prescribed beam end moment demand, M bd.
Design of Multistory Frames
In order to extrapolate the above design methodology for a
friction-damped beam end to multistory frame structures, this sec-
tion describes a design approach in which the number of friction
dampers in a frame is flexible and is a choice in the design pro-
cedure. The goal is to determine the required posttensioning steel
areas and damper slip forces for a trial structure with given ge-
ometry, beam and column member dimensions, and selected
damper distribution. It is assumed that the damper distribution is
reasonably uniform within the structure 共i.e., concentration of
dampers in certain regions of the structure is not allowed兲.
As an example, Fig. 8 depicts two floor levels in a four-bay
friction-damped precast concrete moment frame. The lower floor
共level i兲 has friction dampers at every beam end, whereas the
upper floor 共level i + 1兲 has friction dampers at four beam ends.
One method would be to design each beam end in the structure
using Eqs. 共11兲, 共12兲, 共14兲, and 共7兲. The difficulty with this ap-
proach is that different amounts of posttensioning steel would
result for each beam end, requiring the posttensioning area to be
varied from section to section and resulting in an impractical
design. Therefore, the goal of the frame design formulation below
is to provide a constant A p at each floor/roof level and to provide
the required damper slip force at the friction-damped beam ends
to satisfy the prescribed lateral strength and energy dissipation
requirements for the frame.
To develop design equations for use in multistory frame struc-
tures, the following two new variables are introduced: 共1兲
njt⫽total number of beam ends at a floor/roof level 共e.g., njt = 8 for
levels i and i + 1 in Fig. 8兲; and 共2兲 njs⫽total number of friction-
damped beam ends at the floor/roof level 共e.g., njs = 8 for level i
and njs = 4 for level i + 1 in Fig. 8兲. It is assumed that the damper
contribution to the beam end moment resistance, M bs is constant
at all of the friction-damped beam ends at a floor/roof level. Simi-
larly, the posttensioning contribution to the beam end moment
resistance, M bp is assumed to be constant at all of the beam ends
␤b = 1 / 4 ␤b = 5 / 16 ␤b = 5 / 12
Fds Ap Fds Ap Fds Ap
共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲
302 2,974 378 2,671 503 2,200
227 2,503 285 2,245 383 1,845
133 1,619 165 1,465 222 1,206
JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007 / 1505
 Fig. 6. Subassembly M b-␪b behaviors: 共a兲 ␤b = 1 / 8; 共b兲 ␤b = 1 / 4; 共c兲 ␤b = 5 / 16; and 共d兲 ␤b = 5 / 12

at each floor/roof level. The relationship between M bs and M bp is 共3兲. The depth of the assumed uniform concrete compression
given by Eq. 共8兲. stress block, a in Eq. 共20兲 is given by Eq. 共7兲.
As shown in Eqs. 共15兲 and 共16兲, the sum of the prescribed Note that Fds and A p are assumed to remain constant within the
design beam end moment demands at a floor/roof level, ⌺ M bd floor/roof level being designed. This simple approach allows fric-
共from a linear-elastic analysis of the frame under equivalent lat- tion dampers to be used selectively at the beam ends at a floor/
eral forces兲 is assumed to be equal to the sum of the damper roof level, while keeping the posttensioning steel area constant
contribution, M bs times the number of friction-damped beam across the spans. It is assumed that the structure is designed to
ends, njs and the posttensioning contribution, M bp times the total ensure a full yield mechanism involving all floor/roof levels 共i.e.,
number of beam ends, njt story mechanisms are prevented兲 and all spans in the frame.
Based on this assumption, the location of the friction-damped
⌺M bd = n jsM bs + n jtM bp 共15兲
beam ends at a floor/roof level does not have a significant effect
on the behavior of the frame. This consideration forms the basis
⌺M bd = n js␬dM bp + n jtM bp 共16兲
of the design formulation above.
Then, Eqs. 共17兲 and 共20兲 can be developed to determine the It can be seen that when njs = 0 共i.e., floor/roof level with no
required damper slip force Fds and posttensioning steel area A p for friction dampers兲, then, M bp = ⌺M bd / njt from Eq. 共18兲, indicating
a floor/roof level with a selected number of friction-damped beam as expected that all of the beam end moment resistances at the
ends floor/roof level are provided by the posttensioning steel. Note also
the similarities between the floor/roof design Eqs. 共19兲 and 共20兲
⌺M bd and the beam end design Eqs. 共11兲 and 共12兲 when njs = njt 共i.e.,
M bs = ␬d 共17兲
n jt + ␬dn js dampers placed at all beam ends兲.
Similar to the beam end design equations, the proposed frame
⌺M bd design procedure requires that a value for the frame relative en-
M bp = 共18兲
n jt + ␬dn js ergy dissipation ratio ␤ be selected 共note that ␤ 艌 0.125 is re-
quired by ACI T1.1-01兲. It is recommended that ␤ ratios smaller
⌺M bd
Fds = ␬d 共19兲
共n jt + ␬dn js兲共hb + 2hd兲
Table 2. Calculated Subassembly Relative Energy Dissipation Ratios
⌺M bd hb

冉 冊
Ap = 共20兲
hb − a 共mm兲 ␤b = 1 / 8 ␤b = 1 / 4 ␤b = 5 / 16 ␤b = 5 / 12
f pi共n jt + ␬dn js兲
2 1,016 0.137 0.228 0.298 0.385
813 0.143 0.260 0.317 0.406
where, Eq. 共19兲 is obtained by substituting Eq. 共17兲 into Eq. 共2兲
610 0.163 0.280 0.334 0.411
and Eq. 共20兲 is obtained by substituting Eqs. 共6兲 and 共18兲 into Eq.

1506 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
 Fig. 7. ␤b-Fds and ␤b-A p relationships: 共a兲 ␤b-Fds; 共b兲 ␤b-A p

than 0.5 be used in design to ensure a sufficient amount of self-
centering capability in the structure.
The beam end relative energy dissipation ratio, ␤b is related to
the frame relative energy dissipation ratio ␤ as follows. If damp-
ers are used at every beam end in the structure, then, ␤b = ␤. If
dampers are used at selected beam ends, then ␤b ⬎ ␤ and is de-
termined based on the number of dampers in the structure as
described in the next section. Note that ␤b is assumed to be the
same for all of the friction-damped beam ends in the structure.
Once the ␤b ratio corresponding to the selected ␤ ratio is
found, the required damper slip force and posttensioning steel
area at each floor and roof level in the structure can be determined
by substituting Eq. 共14兲 into Eqs. 共19兲 and 共20兲.
The only difference within each of the six-story/four-bay, nine-
story/four-bay, and six-story/six-bay frame groups is the number
of friction-damped beam ends. Fig. 9共a兲 depicts the baseline con-
figuration of the six-story/four-bay frame with friction dampers
located at every beam end, except at the roof level 共resulting in a
total number of friction-damped beam ends in the frame, nd = 40兲.
This baseline frame was designed using a constant design beam
end relative energy dissipation ratio of ␤b = 1 / 8 at each damper
location in the structure.
Additional six-story/four-bay frame configurations targeting
␤ = 1 / 8 were considered with nd = 20, 16, and 12 as shown in Figs.
9共b–d兲. The design beam end relative energy dissipation ratios

Analytical Verification of Frame Design Equations

Eight different frame and damper configurations targeting the ACI

T1.1–01 minimum of ␤ = 1 / 8 were designed based on the multi-
story frame design procedure described above. As shown in Fig.
9, the structures consist of four six-story/four-bay frames, two
nine-story/four-bay frames, and two six-story/six-bay frames. The
center-to-center bay spacing of the frames is 7.62 m and the
center-to-center story height is 4.88 m for the first story and
3.96 m for the upper stories. The columns are 711 by 965 mm in
cross section. The beam width is 610 mm. Additional structural
properties used in design are: f ⬘c = 41 MPa and f pi = 0.65f pu, where
f pu = 1,862 MPa⫽design ultimate strength of the posttensioning
steel.
It was assumed that M bd = 2,034 kN m and hb = 1,016 mm at
floor levels 1–3, M bd = 1,356 kN m and hb = 813 mm at floor lev-
els 4 and 5, and M bd = 678 kN m and hb = 610 mm at levels 6–9.
Note that these design beam end moment demands and beam
dimensions are the same as the values used in the beam-column
subassembly investigation described previously. The M bd de-
mands were assumed to be the same at every beam end at a
floor/roof level.

Fig. 9. Friction-damped frame damper layouts targeting ␤ = 1 / 8: 共a兲

␤b = 1 / 8, nd = 40; 共b兲 ␤b = 1 / 4, nd = 20; 共c兲 ␤b = 5 / 16, nd = 16; 共d兲
␤b = 5 / 12, nd = 12; 共e兲 ␤b = 1 / 8, nd = 64; 共f兲 ␤b = 1 / 4, nd = 32; 共g兲
Fig. 8. Friction-damped floor elevation ␤b = 1 / 8, nd = 60; and 共h兲 ␤b = 1 / 4, nd = 30

JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007 / 1507

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Table 3. Required Frame Fds and A p Targeting ␤ = 1 / 8
Six-story/four-bay frame Nine-story/four-bay frame Six-story/six-bay frame
␤b = 1 / 8 ␤b = 1 / 4 ␤b = 5 / 16 ␤b = 5 / 12 ␤b = 1 / 8 ␤b = 1 / 4 ␤b = 1 / 8 ␤b = 1 / 4
nd = 40 nd = 20 nd = 16 nd = 12 nd = 64 nd = 32 nd = 60 nd = 30
Floor/roof Fds Ap Fds Ap Fds Ap Fds Ap Fds Ap Fds Ap Fds Ap Fds Ap
level 共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲 共kN兲 共mm2兲
9 — — — — — — — — 0 2,348 0 2,348 — — — —
8 — — — — — — — — 71 1,968 160 1,916 — — — —
7 — — — — — — — — 61 1,968 160 1,916 — — — —
6 0 2,348 0 2,348 0 2,348 0 2,348 71 1,968 160 1,916 0 2,348 0 2,348
5 120 3,065 276 2,981 360 2,787 512 2,445 120 3,065 276 2,981 120 3,065 276 2,981
4 120 3,065 276 2,981 0 3,716 0 3,716 120 3,065 276 2,981 120 3,065 276 2,981
3 160 3,626 365 3,529 431 2,955 667 2,910 160 3,626 365 3,529 160 3,626 365 3,529
2 160 3,626 365 3,529 0 4,368 0 4,368 160 3,626 365 3,529 160 3,626 365 3,529
1 160 3,626 365 3,529 431 2,955 667 2,910 160 3,626 365 3,529 160 3,626 365 3,529

共prescribed only at the friction-damped beam ends兲 were scaled
proportionally based on the number of dampers in the structure
such that ␤b = 1 / 4, 5/16, and 5/12 were used for the frames with
nd = 20, 16, and 12, respectively. The analyses investigated, for
example, if decreasing the total number of dampers by a factor of
two and increasing the beam end relative energy dissipation ratio,
␤b by a factor of two results in roughly the same relative energy
dissipation ratio, ␤ for the entire frame. Note that according to
ACI T1.2 共ACI 2001b兲, which defines procedures for the seismic
design of hybrid precast concrete moment frames, the minimum ␤
ratio of 1/8 has to be satisfied at every beam-column joint in the
frame. In comparison, the frame configurations in Figs. 9共b–d兲
aim to satisfy the prescribed frame energy dissipation requirement
at the “system” level rather than at the “joint” level based on the
placement of dampers at a reduced number of beam ends in the
structure.
Figs. 9共e and f兲 show the two nine-story/four-bay frame con-
figurations and Figs. 9共g and h兲 show the two six-story/six-bay
frame configurations targeting the ACI T1.1–01 minimum of
␤ = 1 / 8. The baseline configurations in Figs. 9共e and g兲 have fric-
tion dampers located at every beam end except at the roof level
共resulting in total number of friction-damped beam ends of
nd = 64 and 60 in the two frames, respectively, with ␤b = 1 / 8 at
each damper location兲. The number of friction-damped beam ends
is reduced by half in the configurations shown in Figs. 9共f and h兲,

Fig. 10. Six-story/four-bay frame base shear-roof drift behaviors: 共a兲 ␤b = 1 / 8, nd = 40; 共b兲 ␤b = 1 / 4, nd = 20; 共c兲 ␤b = 5 / 16, nd = 16; and 共d兲
␤b = 5 / 12, nd = 12

1508 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
 Fig. 11. Nine-story/four-bay frame base shear-roof drift behaviors: 共a兲 ␤b = 1 / 8, nd = 64; 共b兲 ␤b = 1 / 4, nd = 32

Fig. 12. Six-story/six-bay frame base shear-roof drift behaviors: 共a兲 ␤b = 1 / 8, nd = 60; 共b兲 ␤b = 1 / 4, nd = 30

Table 4. Calculated Frame Relative Energy Dissipation Ratios Targeting ␤ = 1 / 8

Six-story/four-bay Nine-story/four-bay Six-story/six-bay
frame frame frame

Roof ␤b = 1 / 8 ␤b = 1 / 4 ␤b = 5 / 16 ␤b = 5 / 12 ␤b = 1 / 8 ␤b = 1 / 4 ␤b = 1 / 8 ␤b = 1 / 4
drift, ⌬ nd = 40 nd = 20 nd = 16 nd = 12 nd = 64 nd = 32 nd = 60 nd = 30
共%兲 ␤ ␤ ␤ ␤ ␤ ␤ ␤ ␤
±0.5 0.147 0.141 0.140 0.120 0.140 0.131 0.116 0.111
±1.0 0.147 0.151 0.148 0.147 0.140 0.143 0.123 0.124
±1.5 0.161 0.167 0.164 0.167 0.126 0.132 0.138 0.138
±2.0 0.171 0.178 0.174 0.179 0.118 0.125 0.147 0.150
Average 0.156 0.159 0.156 0.153 0.131 0.133 0.131 0.131

with the ␤b ratio increased proportionally to 1 / 4 at each damper
location.
Table 3 lists the calculated required damper slip forces and
posttensioning steel areas from Eqs. 共19兲 and 共20兲 for the eight
frame configurations in Fig. 9. Note that the damper slip forces
remain within practical values for all cases.
Nonlinear reversed-cyclic lateral load analyses of the eight
frames were conducted using the DRAIN-2DX program. The
frame models were constructed by combining the subassembly
models in Fig. 5 at the beam midspan and column midstory loca-
tions. The columns were allowed to yield at the bases but not over
the height of the structures, in accordance with a capacity design
approach.
The resulting base shear force, V versus roof drift, ⌬ relation-
ships for the six-story/four-bay, nine-story/four-bay, and six-story/
six-bay structures are plotted in Figs. 10–12, respectively. The
base shear force is equal to the sum of the lateral forces applied at
the floor and roof levels and the roof drift is equal to the lateral
displacement at the roof divided by the height of the roof from the
base. As expected 共since the frames targeted the same M bd and ␤
values兲, the different damper configurations 共with different post-
tensioning steel areas, damper slip forces, and damper locations兲
in each of the frame sets result in nearly identical base shear
resistances and overall shape of the hysteresis loops, regardless of
the number of dampers used.
The frame relative energy dissipation ratios, ␤ calculated from
the individual roof drift cycles in Figs. 10–12 共i.e., ⌬ = ± 0.5, ±1.0,
±1.5, and ±2.0%兲, as well as the average frame ␤ ratios, are
presented in Table 4. The calculated average ␤ ratios are, conser-
vatively, somewhat higher than the target value of ␤ = 1 / 8. This

JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007 / 1509

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Table 5. Required Frame Fds and A p Targeting ␤ = 5 / 16 and 5 / 12 Table 6. Calculated Frame Relative Energy Dissipation Ratios Targeting
␤ = 5 / 16 and 5 / 12
Six-story/four-bay frame
Six-story/four-bay frame
␤b = 5 / 16 ␤b = 5 / 12
nd = 40 nd = 40 Roof ␤b = 5 / 16 ␤b = 5 / 12
drift, ⌬ nd = 40 nd = 40
Floor/roof Fds Ap Fds Ap
共%兲 ␤ ␤
level 共kN兲 共mm2兲 共kN兲 共mm2兲
±0.5 0.277 0.342
6 0 2,348 0 2,348
±1.0 0.270 0.339
5 302 2,245 405 1,845
±1.5 0.275 0.344
4 302 2,245 405 1,845
±2.0 0.281 0.348
3 396 2,671 529 2,200
Average 0.276 0.343
2 396 2,671 529 2,200
1 396 2,671 529 2,200

smaller than the target values of ␤ = 5 / 16 and 5 / 12, respectively,

difference may be, in part, due to the contribution of yielding at
the column bases to the overall frame energy dissipation, which is
not included in the proposed design formulation. Note that the
frame configurations with friction dampers placed at a reduced
number of beam ends to satisfy the prescribed energy dissipation
requirement at the system level rather than at the joint level may
result in a more economical solution.
In addition to the frame investigations targeting the ACI
T1.1–01 minimum of ␤ = 1 / 8, configurations targeting larger ␤
ratios 共␤ = 5 / 16 and 5 / 12兲 were also considered using the six-
story/four-bay frame structure. The total number of friction-
damped beam ends was kept constant at nd = 40 and the prescribed
beam end relative energy dissipation ratios were increased to
␤b = 5 / 16 and 5 / 12. Similar to the frames in Fig. 9, Eqs. 共19兲 and
共20兲 were utilized to determine the required Fds and A p values,
respectively, for these additional frame configurations, as pro-
vided in Table 5. Note that the damper slip forces remain within
practical values.
Based on the analysis results, Fig. 13 shows the base shear
versus roof drift behaviors for the three six-story/four-bay frame
configurations with nd = 40 and ␤b = 1 / 8, 5 / 16, and 5 / 12 关note
that Fig. 13共a兲 is the same as Fig. 10共a兲兴. The hysteresis loops for
␤b = 5 / 16 and 5 / 12 show that friction-damped precast concrete
frames can be designed to have energy dissipation levels signifi-
cantly higher than the ACI T1.1–01 minimum, while maintaining
a large self-centering capability due to the posttensioning force.
The combination of friction dampers for energy dissipation with
posttensioning steel for self-centering provides designers with
flexibility in achieving the desired hysteretic characteristics for
the structure.
The resulting calculated relative energy dissipation ratios for
the frames in Figs. 13共b and c兲 are shown in Table 6, with average
values of ␤ = 0.276 and 0.343, respectively. These ␤ ratios are
indicating that the design energy dissipation requirement is not
satisfied 共contrary to the frames in Figs. 10–12 targeting ␤ = 1 / 8兲.
Investigating the results in Table 6, the unconservative discrep-
ancy between the target and calculated frame relative energy
dissipation ratios increases for increasing values of ␤, with the
largest discrepancy occurring for ␤ = 5 / 12. It is concluded that
improvement in the proposed frame design equations may be
needed for these large ␤ configurations. Note that a similar trend
can also be observed in the beam-column subassembly results in
Table 2.
Summary and Conclusions
A seismic design approach for friction-damped precast concrete
frame structures is presented. The structures use unbonded post-
tensioning steel in the beams to provide a part of the lateral re-
sistance at the floor and roof levels. The focus of the paper is to
determine the damper slip forces and posttensioning steel areas
needed to satisfy prescribed design lateral strength and energy
dissipation requirements for a trial frame with given geometry,
beam and column member dimensions, and selected damper
distribution.
For design purposes, the moment resistance at the end of a
friction-damped beam is decomposed into two components: 共1兲
resistance due to the friction dampers; and 共2兲 resistance due to
the posttensioning steel. A series of prototype friction-damped
beam-column subassemblies and multistory frame structures are
designed following the proposed procedures. ACI T1.1-01 共ACI
Innovation Task Group 1 2001a兲 is used to prescribe the amount
of energy dissipation in the structures. The parameters studied in
the investigation are: 共1兲 beam depth; 共2兲 number of stories; 共3兲
number of bays; 共4兲 number of friction-damped beam ends in the

Fig. 13. Six-story/four-bay frames with nd = 40: 共a兲 ␤b = 1 / 8 关same as Fig. 10共a兲兴; 共b兲 ␤b = 5 / 16; and 共c兲 ␤b = 5 / 12

1510 / JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007

Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
frame; 共5兲 prescribed energy dissipation requirements; and 共6兲
prescribed lateral strength requirements.
Nonlinear reversed-cyclic analyses of the prototype structures
under lateral loads indicate that the prescribed design strength and
ACI T1.1–01 minimum energy dissipation requirements can be
satisfied using the proposed design procedures. The use of friction
dampers at selected beam ends to satisfy the prescribed frame
energy dissipation requirement at the “system” level rather than at
the “joint” level may lead to an economical solution. The results
also show that friction-damped precast concrete frames can
achieve energy dissipation levels significantly higher than the
ACI T1.1–01 minimum while maintaining a high level of self-
centering capability due to the posttensioning force; however, the
design approach may need to be improved for these cases. The
combination of friction dampers for energy dissipation with post-
tensioning steel for self-centering provides designers with flex-
ibility in achieving the desired hysteretic characteristics for a
structure.
Acknowledgments
This research was funded by the National Science Foundation
共NSF兲 under Grant No. CMS 98-74872 as a part of the
CAREER Program. The support of the NSF Program Directors
Dr. S. C. Liu and Dr. S. L. McCabe is gratefully acknowledged. In
addition, the writers recognize the technical and financial support
provided by industry partnerships with: R. W. Monroe and D. R.
Poweleit of the Steel Founders’ Society of America; R. Reddy of
Southwest Steel Casting Company of Longview, Texas; C. E.
Hilgeman and M. A. Fusani of Concrete Technology, Inc. of
Springboro, Ohio; and K. B. Allen and D. Martin of Dywidag-
Systems International of Bolingbrook, Ill. The opinions, findings,
and conclusions expressed in this paper are those of the writers
and do not necessarily reflect the views of the NSF or the indi-
viduals and organizations acknowledged above.
References
ACI Innovation Task Group 1. 共2001a兲. Acceptance criteria for moment
frames based on structural testing (T1.1–01) and commentary (T1.1R-
01), American Concrete Institute, Farmington Hills, Mich.
ACI Innovation Task Group 1. 共2001b兲. “Special hybrid moment frames
composed of discretely jointed precast and posttensioned concrete
members 共ACI T1.2-XX兲 and commentary 共T1.2R-XX兲.” ACI Struct.
J., 98共5兲, 771–784.
Cheok, G., and Lew, H. 共1993兲. “Model precast concrete beam-to-column
joints subject to cyclic loading.” PCI J., 38共4兲, 80–92.
JOURNAL OF STRUCTURAL ENGINEERING © ASCE / NOVEMBER 2007 / 1511
Downloaded 03 Feb 2009 to 129.74.250.206. Redistribution subject to ASCE license or copyright; see http://pubs.asce.org/copyright
Christopoulos, C., Filiatrault, A., Uang, C.-M., and Folz, B. 共2002兲. “Post
tensioned energy dissipating connections for moment-resisting steel
frames.” J. Struct. Eng., 128共9兲, 1111–1120.
El-Sheikh, M., Sause, R., Pessiki, S., and Lu, L. W. 共1999兲. “Seismic
behavior and design of unbonded posttensioned precast concrete
frames.” PCI J., 44共3兲, 54–71.
Englekirk, R. 共2002兲. “Design-construction of the Paramount—A 39-
story precast prestressed concrete apartment building.” PCI J., 47共4兲,
56–71.
Garlock, M. M., Ricles, J. M., and Sause, R. 共2005兲. “Experimental stud-
ies of full-scale posttensioned steel connections.” J. Struct. Eng.,
131共3兲, 438–448.
Grigorian, C., Yang, T., and Popov, E. 共1993兲. “Slotted bolted connection
energy dissipators.” Earthquake Spectra, 9共3兲, 491–504.
International Code Council 共ICC兲. 共2006兲. International building code
2006, Falls Church, Va.
Kurama, Y. C., Weldon, B. O., and Shen, Q. 共2006兲. “Experimental evalu-
ation of posttensioned hybrid coupled wall subassemblages.” J.
Struct. Eng., 132共7兲, 1017–1029.
Morgen, B., and Kurama, Y. 共2004a兲. “A friction damper for postten-
sioned precast concrete beam-to-column joints.” Proc., 13th World
Conf. on Earthquake Engineering 共CD-ROM兲, Vancouver, BC,
Canada.
Morgen, B., and Kurama, Y. 共2004b兲. “A friction damper for postten-
sioned precast concrete moment frames.” PCI J., 49共4兲, 112–133.
Morgen, B., and Kurama, Y. 共2007兲. “Friction-damped unbonded postten-
sioned precast concrete moment frame structures for seismic regions.”
Structural Engineering Research Rep. NDSE-07-01, Dept. of Civil
Engineering and Geological Sciences, Univ. of Notre Dame, Notre
Dame, Ind.
Prakash, V., Powell, G., and Campbell, S. 共1993兲. “DRAIN-2DX base
program description and user guide; Version 1.10.” Rep. No. UCB/
SEMM-93/17, Dept. of Civil and Environmental Engineering, Univ. of
California, Berkeley, Calif.
Priestley, M., and MacRae, G. 共1996兲. “Seismic tests of precast beam-to-
column joint subassemblages with unbonded tendons.” PCI J., 41共1兲,
64–81.
Priestley, M., Sritharan, S., Conley, J., and Pampanin, S. 共1999兲. “Pre-
liminary results and conclusions from the PRESSS five-story precast
concrete test building,” PCI J., 44共6兲, 42–67.
Priestley, M., and Tao, J. 共1993兲. “Seismic response of precast prestressed
concrete frames with partially debonded tendons.” PCI J., 38共1兲, 58–
67.
Stone, W., Cheok, G., and Stanton, J. 共1995兲. “Performance of hybrid
moment-resisting precast beam-column concrete connections sub-
jected to cyclic loading.” ACI Struct. J., 92共2兲, 229–249.
Way, D. 共1996兲. “Friction-damped moment resisting frames.” Earthquake
Spectra, 12共3兲, 623–633.
Weldon, B., and Kurama, Y. 共2007兲. “Unbonded posttensioned precast
concrete coupling beams: An experimental evaluation.” Proc., ASCE
Structures Congress, Long Beach, Calif.