V. 112, NO.

2
MARCH-APRIL 2015

ACI
STRUCTURAL J O U R N A L

A JOURNAL OF THE AMERICAN CONCRETE INSTITUTE

 CONTENTS
Board of Direction ACI Structural Journal
President
William E. Rushing Jr. March-April 2015, V. 112, No. 2
a journal of the american concrete institute
Vice Presidents an international technical society
Sharon L. Wood
Michael J. Schneider
123 Evaluation of Column Load for Generally Uniform Grid-Reinforced
Directors Pile Cap Failing in Punching, by Honglei Guo
Roger J. Becker
Dean A. Browning
Jeffrey W. Coleman 135 Design Implications of Large-Scale Shake-Table Test on Four-Story
Alejandro Durán-Herrera Reinforced Concrete Building, by T. Nagae, W. M. Ghannoum, J. Kwon,
Robert J. Frosch K. Tahara, K. Fukuyama, T. Matsumori, H. Shiohara, T. Kabeyasawa,
Augusto H. Holmberg
Cary S. Kopczynski S. Kono, M. Nishiyama, R. Sause, J. W. Wallace, and J. P. Moehle
Steven H. Kosmatka
Kevin A. MacDonald 147 Inverted-T Beams: Experiments and Strut-and-Tie Modeling, by
Fred Meyer
Michael M. Sprinkel N. L. Varney, E. Fernández-Gómez, D. B. Garber, W. M. Ghannoum, and
David M. Suchorski O. Bayrak

Past President Board Members 157 Energy-Based Hysteresis Model for Reinforced Concrete Beam-
Anne M. Ellis
James K. Wight Column Connections, by Tae-Sung Eom, Hyeon-Jong Hwang, and
Kenneth C. Hover Hong-Gun Park

Executive Vice President 167 Ductility Enhancement in Beam-Column Connections Using Hybrid
Ron Burg
Fiber-Reinforced Concrete, by Dhaval Kheni, Richard H. Scott,
Technical Activities Committee S. K. Deb, and Anjan Dutta
Ronald Janowiak, Chair
Daniel W. Falconer, Staff Liaison 179 Behavior and Simplified Modeling of Mechanical Reinforcing Bar
JoAnn P. Browning
Catherine E. French
Splices, by Zachary B. Haber, M. Saiid Saiidi, and David H. Sanders
Fred R. Goodwin
Trey Hamilton 189 Bond-Splitting Strength of Reinforced Strain-Hardening Cement
Neven Krstulovic-Opara Composite Elements with Small Bar Spacing, by Toshiyuki Kanakubo
Kimberly Kurtis
Kevin A. MacDonald and Hiroshi Hosoya
Jan Olek
Michael Stenko 199 Wide Beam Shear Behavior with Diverse Types of Reinforcement,
Pericles C. Stivaros
Andrew W. Taylor by S. E. Mohammadyan-Yasouj, A. K. Marsono, R. Abdullah, and
Eldon G. Tipping M. Moghadasi

Staff 209 Effect of Axial Compression on Shear Behavior of High-Strength

Executive Vice President
Ron Burg Reinforced Concrete Columns, by Yu-Chen Ou and Dimas P. Kurniawan

Engineering 221 Experimental Investigations on Prestressed Concrete Beams with

Managing Director Openings, by Martin Classen and Tobias Dressen
Daniel W. Falconer
Managing Editor 233 Discussion
Khaled Nahlawi
Staff Engineers 
Bond-Slip-Strain Relationship in Transfer Zone of Pretensioned Concrete
Matthew R. Senecal
Elements. Paper by Ho Park and Jae-Yeol Cho
Gregory M. Zeisler
Jerzy Z. Zemajtis
Contents cont. on next page
Publishing Services
Manager
Discussion is welcomed for all materials published in this issue and will appear ten months from
Barry M. Bergin this journal’s date if the discussion is received within four months of the paper’s print publication.
Discussion of material received after specified dates will be considered individually for publication or
Editors private response. ACI Standards published in ACI Journals for public comment have discussion due
Carl R. Bischof dates printed with the Standard.
Tiesha Elam Annual index published online at http://concrete.org/Publications/ACIStructuralJournal.
Kaitlyn Hinman ACI Structural Journal
Kelli R. Slayden Copyright © 2015 American Concrete Institute. Printed in the United States of America.
The ACI Structural Journal (ISSN 0889-3241) is published bimonthly by the American Concrete Institute. Publica-
Editorial Assistant tion office: 38800 Country Club Drive, Farmington Hills, MI 48331. Periodicals postage paid at Farmington, MI, and
Angela R. Matthews at additional mailing offices. Subscription rates: $166 per year (U.S. and possessions), $175 (elsewhere), payable in
advance. POSTMASTER: Send address changes to: ACI Structural Journal, 38800 Country Club Drive, Farmington
Hills, MI 48331.
Canadian GST: R 1226213149.
Direct correspondence to 38800 Country Club Drive, Farmington Hills, MI 48331. Telephone: +1.248.848.3700. Fac-
simile (FAX): +1.248.848.3701. Website: http://www.concrete.org.

ACI Structural Journal/March-April 2015 121

 
Fire Protection for Beams with Fiber-Reinforced Polymer Flexural Strength- Contributions to
ening Systems. Paper by Nabil Grace and Mena Bebawy ACI Structural Journal

Analysis and Prediction of Transfer Length in Pretensioned, Prestressed The ACI Structural Journal is an open
Concrete Members. Paper by Byung Hwan Oh, Si N. Lim, Myung K. Lee, and forum on concrete technology and papers
Sung W. Yoo related to this field are always welcome.
All material submitted for possible publi-

Flexural Testing of Reinforced Concrete Beams with Recycled Concrete cation must meet the requirements of
Aggregates. Paper by Thomas H.-K. Kang, Woosuk Kim, Yoon-Keun Kwak, and the “American Concrete Institute Publi-
Sung-Gul Hong cation Policy” and “Author Guidelines
and Submission Procedures.” Prospective
241 Reviewers in 2014 authors should request a copy of the Policy
and Guidelines from ACI or visit ACI’s
website at www.concrete.org prior to
MEETINGS submitting contributions.
Papers reporting research must include
a statement indicating the significance of
MARCH/APRIL 11-13—2015 International Concrete the research.
Sustainability Conference, Miami, FL, The Institute reserves the right to return,
30-2—Concrete Sawing & Drilling www.concretesustainabilityconference.org without review, contributions not meeting
Association Convention and Tech Fair, the requirements of the Publication Policy.
St. Petersburg, FL, www.csda.org/events/ 14-16—The American Institute of All materials conforming to the Policy
event_details.asp?id=444478&group Architects Convention, Atlanta, GA, requirements will be reviewed for editorial
http://convention.aia.org/event/homepage.aspx quality and technical content, and every
APRIL effort will be made to put all acceptable
14-17—The Masonry Society 2015 papers into the information channel.
13-15—BEST Conference Building Spring Meetings, Denver, CO, http:// However, potentially good papers may be
Enclosure Science & Technology™, www.masonrysociety.org/index. returned to authors when it is not possible
Kansas City, MO, www.nibs.org/?page=best cfm?showincenter=http%3A//www. to publish them in a reasonable time.
masonrysociety.org/html/calendar/index.htm
26-29—2015 Post-Tensioning Institute Discussion
Convention, Houston, TX, www.post- 17-20—12th North American Masonry All technical material appearing in the
tensioning.org/page/pti-convention Conference, Denver, CO, www. ACI Structural Journal may be discussed.
masonrysociety.org/NAMC/index.html If the deadline indicated on the contents
26-30—57th Annual IEEE-IAS/ page is observed, discussion can appear
PCA Cement Industry Technical 24-26 - Fifth International Symposium on in the designated issue. Discussion should
Conference, Toronto, ON, Canada, www. Nanotechnology in Construction, Chicago, be complete and ready for publication,
cementconference.org IL, www.nicom5.org including finished, reproducible illustra-
tions. Discussion must be confined to the
MAY JUNE scope of the paper and meet the ACI Publi-
cation Policy.
3-7—International Cement Microscopy 1-3—7th RILEM Conference on Follow the style of the current issue.
Association Annual Conference, Seattle, High Performance Fiber Reinforced Be brief—1800 words of double spaced,
WA, www.cemmicro.org Cement Composites, Stuttgart, typewritten copy, including illustrations
Germany, www.rilem.org/gene/main. and tables, is maximum. Count illustrations
4-7—2015 World of Coal Ash Conference, php?base=600040#next_614 and tables as 300 words each and submit
Nashville, TN, www.worldofcoalash.org them on individual sheets. As an approxi-
mation, 1 page of text is about 300 words.
Submit one original typescript on 8-1/2 x
11 plain white paper, use 1 in. margins,
THE ACI CONCRETE CONVENTION AND EXPOSITION: FUTURE DATES and include two good quality copies of the
entire discussion. References should be
2015—April 12-16, Marriott & Kansas City Convention Center, Kansas City, MO complete. Do not repeat references cited
2015—November 8-12, Sheraton Denver, Denver, CO in original paper; cite them by original
2016—April 17-21, Hyatt & Wisconsin Center, Milwaukee, WI number. Closures responding to a single
discussion should not exceed 1800-word
For additional information, contact: equivalents in length, and to multiple
Event Services, ACI discussions, approximately one half of
38800 Country Club Drive, Farmington Hills, MI 48331 the combined lengths of all discussions.
Telephone: +1.248.848.3795 Closures are published together with
the discussions.
e-mail: conventions@concrete.org
Discuss the paper, not some new or
outside work on the same subject. Use
references wherever possible instead of
ON COVER: 112-S12, p. 136, Fig. 2—Reinforced concrete (left) and prestressed concrete (right) specimens repeating available information.
on the E-Defense shake table.
Discussion offered for publication should
offer some benefit to the general reader.
Discussion which does not meet this
Permission is granted by the American Concrete Institute for libraries and other users registered with the Copyright requirement will be returned or referred to
Clearance Center (CCC) to photocopy any article contained herein for a fee of $3.00 per copy of the article. Payments
should be sent directly to the Copyright Clearance Center, 21 Congress Street, Salem, MA 01970. ISSN 0889-3241/98 the author for private reply.
$3.00. Copying done for other than personal or internal reference use without the express written permission of the
American Concrete Institute is prohibited. Requests for special permission or bulk copying should be addressed to the
Managing Editor, ACI Structural Journal, American Concrete Institute. Send manuscripts to:
The Institute is not responsible for statements or opinions expressed in its publications. Institute publications are not able http://mc.manuscriptcentral.com/aci
to, nor intend to, supplant individual training, responsibility, or judgment of the user, or the supplier, of the information
presented.
Send discussions to:
Papers appearing in the ACI Structural Journal are reviewed according to the Institute’s Publication Policy by individuals
expert in the subject area of the papers. Journals.Manuscripts@concrete.org

122 ACI Structural Journal/March-April 2015

 ACI STRUCTURAL JOURNAL
Title No. 112-S11
Evaluation of Column Load for Generally Uniform Grid-
Reinforced Pile Cap Failing in Punching
by Honglei Guo
Currently, the punching shear resistance of pile caps is frequently
evaluated empirically, and although the strut-and-tie model (STM)
may be used to calculate the issue, the two weaknesses of STM—
conservative nature and difficult configuration—hinder its rational
solution. To attempt to solve these issues, this paper presents a
generalized method of spatial STMs to evaluate punching shear
resistance of general pile caps with uniform grid reinforcement
(TPM). Based on results of the spatial strut-and-tie bearing mech-
anism of pile cap punching failure, three-dimensional (3-D) rather
than two-dimensional (2-D) strut strength is derived. During this
process, nonlinear finite element analysis in conjunction with the
derivation of a gradual least-square method for multiple variables
is adopted. TPM is verified by 98 specimens in the literature, whose
parameters (reinforcement ratio of tension tie, punching-span ratio,
concrete strength, pile number, and pile arrangement) vary, respec-
tively; the comparisons with the other four methods are made. It
is indicated that TPM is extensively applicable to the evaluation
of the punching shear resistance of general pile caps with uniform
grid reinforcement.
Keywords: building code; pile cap; punching shear resistance; strut-and-tie
model (STM).
INTRODUCTION
A pile cap is the load-transfer story between the super-
structure and pile, while the evaluation of its punching shear
resistance is an important basis for determining its thickness
and arrangement of reinforcement.
Generally speaking, the evaluation of punching shear
resistance of a pile cap can be classified into two types
according to the theory of plasticity:
Type 1—The collapse mechanism is assumed so that
the upper-bound solution to punching shear resistance is
obtained using the theory of plasticity, called the “upper-
bound method” for short. This method is adopted in the
critical section stress method of the ACI 318-08 code (ACI
CSM)1 and the Chinese JGJ94-94 code.2 (Although an
empirical method in appearance, ACI CSM is theoretically
an upper-bound method in essence).
Of the aforementioned, as shown in the Appendix* of the
paper, ACI CSM,1 (also, the details of JGJ94-94, ACI STM,
CRSI,3 and TPM at the back being given in the Appendix
of the paper) similar to the calculating method used for
punching shear resistance of slab in the ACI 318-08 code, is
divided into two steps:
1. For simplicity of evaluation, the critical sections
perpendicular to the plane of the pile cap are used instead
*
The Appendix is available at www.concrete.org/publications in PDF format,
appended to the online version of the published paper. It is also available in hard copy
from ACI headquarters for a fee equal to the cost of reproduction plus handling at the
time of the request.
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
of the oblique sections of the punching cone, and the perim-
eter of the critical sections is kept minimum but no closer to
the column edge than d/2 (the definition of d being given in
Eq. (1) and Fig. 4); and
2. Take the minimum of the three kinds of punching shear
resistance in these sections as the ultimate.
Whereas the method in JGJ94-94 code2 is divided into
three steps: 1) take the link line between the column side and
the nearest pile side to form the punching cone; 2) modify
the inclination of the punching cone to ensure it to vary from
45 to 78.7 degrees; and 3) in the end, use a punching coef-
ficient containing the punching-span ratio λ to correct the
punching shear resistance (the definition of λ being given
in Eq. (1)).
Type 2—The rational stress field is assumed according
to the load-transfer route so that the lower-bound solution
to punching shear resistance is obtained, called the “lower-
bound method” for short. As far as the practical evaluation
of the reinforced concrete is concerned, it has often been the
best choice for this method to have the structure likened to
a certain kind of structure or a combination of certain struc-
tures whose bearing mechanism is well known.
In technical codes, the text and Appendix A of
ACI 318-08,1 the CRSI handbook,3 CAN/CSA A23.3-04,4
BSEN 1992-1-1:2004,5 and AS 3600-20016 either adopt or
contain this method.
Of the aforementioned, when the center of any one pile is
at or within twice the distance between the top of the pile cap
and the top of the pile, Section 15.5 in ACI 318-081 states that
punching of the pile cap can be likened to an idealized truss,
and Appendix A of ACI 318-081 gives the basic components
of the truss: strut, tie, and nodal zone, and there is a series
of systematic provisions for the strength and dimensions
of these components. In fact, a general strut-and-tie design
procedure for all discontinuity (D)-regions was introduced.
As a supplement to the ACI 318-08 code, the CRSI hand-
book3 recommends another calculating method, separated
by three steps: 1) the applicable condition is the horizontal
distance between the column side and the nearest axis of the
pile is no larger than d/2; 2) the critical section is taken at the
perimeter of the column face; and 3) the additional contri-
bution of concrete to the punching strength resulting from
the small punching span is considered. This shows that the
CRSI handbook method effectively likens the evaluation of
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2010-415.R3, doi: 10.14359/51687420, received July 29, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
123
Fig. 1—Load-transfer mechanism of SSTM for punching
failure of pile caps.

Fig. 3—Failure form of the strut by nonlinear finite element

analysis. (Note: Model is one-fourth of four-pile cap of
symmetrical and determinant pile arrangement, and crack
surfaces are represented by circles.)
bottom; Zone II, the splitting zone in the midpart of the strut;
and Zone I, the shear-compression zone intersecting the pile
top. The forming process of the punching cone is as follows:
when the principal tensile stress in Zone II reaches the
splitting strength, the first crack is generated and, with the
Fig. 2—Damage mechanism of SSTM for punching failure column load increased, the oblique crack develops toward
of pile caps. the two ends of the strut. Soon after, the strut is split into
two (Struts A and B) connected at its two ends (Zones I and
the two-way shear of the pile caps to the superposition of
III), the column load being jointly borne by Struts A and B.
the one-way shear of two mutually orthogonal deep beams
Part of the column load is transferred to the longitudinal
whose width is equal to the length of the column edges.
reinforcement and the uncracked concrete of Zone I by
In the theoretical study, Wen7 modeled the punching of pile
Strut A, and the other part is transferred to the pile by Strut B.
caps as the coupling between two orthogonal deep beams,
When punching failure occurs, Strut A is punched out rela-
while Kinnunen and Nylander8 regarded it as a spatial shell.
tive to Strut B to have the punching cone formed. It can be
However, recent studies and practice have proved that it
considered that the column load at this moment is jointly
is more rational to liken the bearing mechanism of punching
borne by Zones I and III, together with the dowel action of
failure of pile caps to a spatial STM (SSTM).9-12
the bottom longitudinal reinforcement. The two parts are
Herein, as the basis for the derivation of the column load
correlated, and the loss of the punching shear resistance is a
of pile cap failing in punching to be conducted later, a brief
result of the damages in the aforementioned parts occurring
introduction to the author’s research conclusions is given
one after another so that, with no additional external load, the
as follows.11,12
oblique section suddenly collapses. Therefore, the punching
Load-transfer mechanism of punching failure—As shown
failure of pile caps is either the strut failure, which begins
in Fig. 1, the load-transfer system of the punching failure of
with the splitting in the midpart of the strut (Zone II) and
pile caps is analogous to the SSTM, where the compression
ends with shear-compression failure at the two ends of the
struts are used to model the zones of concrete with primarily
strut (Zones I and III) or the yield failure of the tension tie
unidirectional compressive stresses, while the reinforce-
resulting from insufficient tension tie reinforcement amount.
ments within the range of primarily unidirectional tensile
But the tension tie failure is also accompanied by the strut
stresses are approximated by tension ties.11,12
failure, so the strut failure is an indication of the loss of the
The pile load distribution during the punching failure of
pile cap punching shear resistance.12
pile caps can approximately adopt the value of the pile caps
The two basic factors influencing the strut strength are the
in the elastic stage.11
punching-span ratio and concrete strength.12 The strengths at
Damage mechanism of punching failure—As shown in
the two ends of the strut are not appreciably different; their
Fig. 2 and 3, the strut is represented as three zones: namely,
average can be taken as the strut strength.12
Zone III, the shear-compression zone intersecting the column

124 ACI Structural Journal/March-April 2015

Fig. 4—Effective depth and location of SSTM upper node.
Fig. 5—Location of SSTM lower node.
Dimensions of SSTM—Construct the true rather than
imaginary stress field to achieve the dimensions as follows:
1. During the elastic stage, the cross-sectional area at the
strut end for the pile near the column is larger than that far
from the column. But when the pile cap fails, because the
plastic internal force redistributes, the strut for the pile near
the column unloads (except for the strut between the column
and the pile beneath the column), and the strut for the pile far
from the column increases its load; therefore, at the end, as
shown in Fig. 2, all the cross-sectional areas at the strut end
basically stabilize at the same value—0.6 times that of the pile
(except for the strut between the column and the pile beneath Fig. 6—Effective range of tension tie, punching-span,
the column)—whatever the distance of pile to column.12 and As.
2. As shown in Fig. 4, the upper node of the SSTM is
although no larger than the experimental ones, are signifi-
located at 0.1 times the effective depth vertically downwards
cantly variable. Hence, the hidden safety risks; and 2) the
from the column center on the top surface of the pile caps.12
theory of STM applied to solve the punching of pile caps
3. As shown in Fig. 5, for simplicity, take a two-pile cap
is significantly conservative. This paper focuses on the
as an example to illustrate the location of the SSTM lower
derivation of the three-dimensional (3-D) rather than two-
node, which is obtained in accordance with the three steps:
dimensional (2-D) strut strength, from which the calculating
1) link upper node A to pile center B to obtain line segment
method for the punching shear resistance of the pile caps—
AB; 2) project AB onto the plane where the longitudinal
that is, the column load of pile cap failing in punching—is
reinforcement centroid is located to obtain line segment CD,
developed. Careful verification, comparison, and analysis
while obtaining the projection line L of the pile periphery
show that the results obtained in this paper should contribute
onto the same plane; 3) intercept CD with L to obtain line
to improving the aforementioned situation, and the informa-
segment ED, and the midpoint of ED is just lower node F of
tion presented in this paper should prove useful to organiza-
the SSTM.12
tions that publish design aids for pile caps.
4. As shown in Fig. 6, the effective range of the tension tie
is twice the pile diameter that is concentric with the lower
OVERALL CONSIDERATIONS FOR DERIVATION
node of the SSTM.11
OF EVALUATION
It should be noted that compared with the currently avail-
First, two variables are defined as follows:
able extensive literature on the bearing mechanism of the
1. Punching-span ratio λ
SSTM, investigation on the evaluation of the punching shear
resistance of pile caps with uniform grid reinforcement is as
λ = w/d (1)
of yet inadequate. So, based on the previously mentioned
research conclusions about the punching bearing mecha-
where, as shown in Fig. 4, the effective depth d is the depth
nism, further studies will be made along these lines.
to the centroid of the bottom longitudinal reinforcement. As
shown in Fig. 6, the punching span w is the distance GB1,
RESEARCH SIGNIFICANCE
where line segment AB1 is obtained by linking column
Many punching shear resistances of pile caps are eval-
center A to pile center B1, and point G is obtained through
uated by design aids with the rule-of-thumb procedures,
the interception of AB1 by the periphery of column. If not a
which have at least two drawbacks: 1) the theoretical calcu-
round column, convert its cross section to a circular one of
lation values either far exceed the experimental ones or,
equal perimeter.

ACI Structural Journal/March-April 2015 125

Table 1—Relationship between g and l, fc' for l ≥ 0.95 It is known from the foregoing conclusion that the two
6.7 MPa (971.5 psi) ≤ fc′ ≤ 35 MPa (5075 psi) ≤ fc′ ≤
basic factors influencing the strut strength are the punching-
λ 35 MPa (5075 psi) 50 MPa (7252 psi) span ratio and concrete strength; thus, γ = γ(λ,fc′). To find
the specific expression for γ, the ADINA nonlinear finite
0.95 γ = 2.89255 – 0.31042√fc′* γ = 1.059
element (NFE) program, which has successfully evaluated
1.0 γ = 2.89255 – 0.31042√fc′ γ = 1.056 the punching shear resistance of pier deck13 (pier deck is
1.2 γ = 2.8618 – 0.30712√fc′ γ = 1.04775 similar to pile cap), is adopted. Analysis and derivation of
1.4 γ = 2.8782 – 0.30888√fc′ γ = 1.05225 the expression for γ are made by referring to mathematical
deduction of the gradual least-square method for multiple
1.6 γ = 2.8659 – 0.30756√fc′ γ = 1.04775
variables (GLSMV).
1.8 γ = 2.8618 – 0.30712√fc′ γ = 1.04925 In selecting the model for computerization, as the purpose
2.0 γ = 2.88025 – 0.3091√fc′ γ = 1.0545 of computerization is simply to derive the strut strength,
*
Unit of fc′ is MPa; γ is nondimensional. Similarly hereinafter. there is no need for consideration of the pile number or pile
arrangement other than the choice of the strut. Therefore, a
Table 2—Relationship between g and l, fc' for quarter of the four-pile cap of symmetrical and determinant
0.15 ≤ l ≤ 0.95 pile arrangement is selected, as shown in Fig. 3.
6.7 MPa (971.5 psi) ≤ fc′ ≤ 35 MPa (5075 psi) ≤ fc′ ≤
In developing the numerical model, the concrete of pile
λ 35 MPa (5075 psi) 50 MPa (7252 psi) caps is divided into four layers, the greater part of which
are 3-D isoparametric elements with eight nodes and three
γ = 2.05 × (2.341 – 0.9751λ)
0.75 ≤ λ
– 0.22 × (2.398 –1.057λ) γ = 0.75 × (2.32 – 0.96λ) degrees-of-freedom per node, a few triangular prism-shaped
≤ 0.95 degenerate elements being taken as transition ones. Where
× √fc′
0.35 ≤ λ γ = 2.05 × (1.972 – 0.521λ) –
the pile cap is near the column and pile, the 3 × 3 × 3 inte-
γ = 0.75 × (1.9789 – 0.493λ) gration order is adopted, while the 2 × 2 × 2 integration
≤ 0.75 0.22 × (1.9738 –0.488λ)×√fc′
order is used elsewhere. The column and pile are linked to
γ = 2.05 × (2.125 – 0.973λ)
0.15 ≤ λ
– 0.22 × (2.14258 –0.945λ) γ = 0.75 × (2.18292 – 1.085λ) the pile cap also in the form of 3-D isoparametric elements.
≤ 0.35 The concrete material model adopted is a nonlinear one
× √fc′
with compression crushing, tensile cutoff with strain soft-
2. Reinforcement ratio of tension tie ρ ening, and shear stress transferring across the cracks taken
into account.13 The reinforcing bars are represented by truss
As elements with two nodes, the constitutive law for which is an
ρ= (2) elastic-plastic material model.
2Dp d
As shown in Fig. 3, for the strut between the column
bottom and the pile top in question, take λ and fc′ as 0.15
where, as shown in Fig. 6, As is the sum of the cross- to 2.0 and 6.7 to 50 MPa (971.7 to 7252 psi), respectively.
sectional areas of the longitudinal reinforcements within the Thus, a total of 102 cases of combination is investigated.
effective range of the tension tie; and Dp is the pile diameter. In the process of analysis, as ρ of the pile caps with uniform
As pointed out earlier, the strut failure is an indication grid reinforcement is in general rather small, no larger than
of the loss of the pile cap punching shear resistance. So 1.2% at most and has little influence on the strut strength,14,15
the evaluation of punching shear resistance of pile caps is it can be maintained at 0.6% throughout.
exactly an evaluation of the strut bearing load, while strut The relationships between γ and λ, fc′ for λ ≥ 0.95 are
bearing load F is the cross-sectional area at the strut end S × shown in Table 1 as an example.
strut strength fce. It is known from the earlier statement that S The expressions in Table 1 are summed up as follows:
is 0.6 times the cross-sectional area of the pile, and fce is the 1. γ = a – b√fc′, for 6.7 MPa (971.5 psi) ≤ fc′≤ 35 MPa
average of the strengths at the two ends of the strut. So F can (5075 psi)
be expressed as follows 2. γ = c, for 35 MPa (5075 psi) ≤ fc′≤ 50 MPa (7252 psi)
Obviously, a, b, c are the functions of λ. Use the least-
f ce1 + f ce 2 square method once again to obtain
F = S × f ce = 0.6πR 2 × (3)
2
a = 2.90045 – 0.0170961λ ≈ 2.05 × 1.41485
where R is the radius of the pile; fce1 is the strength at one end
of the strut; and fce2 is that at the other end.
Thus, if only the specific expression of fce is found, F
b = 0.30668 – 0.00183469λ ≈ 0.22 × 1.394
will be obtained. Then, depending on static equilibrium at
the upper node of the SSTM, the column load of pile caps
failing in punching will be readily solved.
c = 1.05888 – 0.00459047λ ≈ 0.75 × 1.41184
DERIVATION FOR fce
In conclusion, for λ ≥ 0.95:
Define γ = fce/fc′, where fc′ is the cylinder compressive
strength of the strut concrete.

126 ACI Structural Journal/March-April 2015

 1) γ = 2.05 × 1.41485 – 0.22 × 1.394√fc′, for 6.7 MPa
(971.5 psi) ≤ fc′ ≤ 35 MPa (5075 psi); and
2) γ = 0.75 × 1.41184, for 35 MPa (5075 psi) ≤ fc′ ≤ 50
MPa (7252 psi).
Similarly, the relationships between γ and λ, fc′ for other
ranges of λ are obtained in Table 2. Observing the situation
of λ in each of its ranges shown in Tables 1 and 2 to know:
1. Whatever the range λ is in, for 6.7 MPa (971.5 psi) ≤ fc′ ≤
35 MPa (5075 psi)
γ = 2.05f1(λ) – 0.22f2(λ)√fc′ (4)
2. Whatever the range λ is in, for 35 MPa (5075 psi) ≤ fc′ ≤
50 MPa (7252 psi)
γ = 0.75f3(λ) (5)
3. In the same range of λ, the expressions for f1(λ), f2(λ),
and f3(λ) are almost identical, so a unified expression can
be taken
f1 ( λ ) + f 2 ( λ ) + f 3 ( λ )
β( λ ) =
3 ACI 318-08 code1 (ACI STM); 4) the American CRSI hand-
Take β(λ) out of Eq. (4) and (5), then
γ = α(fc′) × β(λ)
where, for α(fc′):
1) α(fc′) = 2.05 – 0.22√fc′, for 6.7 MPa (971.5 psi) ≤ fc′ ≤
35 MPa (5075 psi)
(6a)
2) α(fc′) = 0.75, for 35 MPa (5075 psi) ≤ fc′ ≤ 50 MPa
(7252 psi)
whereas, for β(λ):
1) β(λ) = 1.4, for λ ≥ 0.95
2) β(λ) = 2.35 – λ, for 0.75 ≤ λ ≤ 0.95
(6b)
3) β(λ) = 1.975 – 0.5λ, for 0.35 ≤ λ ≤ 0.75
4) β(λ) = 2.15 – λ, for 0.15 ≤ λ ≤ 0.35
Observation of Eq. (6b) shows, for 0.15 ≤ λ ≤ 0.95, that
the slopes of all the fold line segments making up β(λ) are
almost identical. So the straight line linked by point λ = 0.15
and point λ = 0.95 can be used to represent β(λ) in this range
(0.15 ≤ λ ≤ 0.95) in a unified manner; that is, ultimately
1) β(λ) = 2.1125 – 0.75λ, for 0.15 ≤ λ ≤ 0.95
(6c)
2) β(λ) = 1.4, for λ ≥ 0.95
Thus, the ultimate expression of the strut strength fce is
ACI Structural Journal/March-April 2015
f ce = γ × f c′ = α ( f c′)β(λ ) f c′ (7)
which, substituted back into Eq. (3), gives the ultimate
bearing load expression of the strut
F = 0.6πR 2 f ce = 1.885 R 2 α ( f c′) β ( λ ) f c′ (8)
where α(fc′) and β(λ) are found in Eq. (6a) and (6c), respec-
tively. As shown in Eq. (7), fce is a constantly increasing
function of fc′, whatever the range fc′ is in; for 0.15 ≤ λ ≤
0.95, fce is a decreasing function of λ, while for λ ≥ 0.95, fce
is a constant function of λ.
RESULTS AND DISCUSSION
Table 3 lists the published test data of 98 specimens on
the punching failure of the pile caps with uniform grid rein-
forcement in literature, whose pile number, pile arrange-
ment, punching-span ratio, concrete strength, and reinforce-
ment ratio of tension tie vary, respectively, while Table 4
gives the Pe/Pp (experimental column load/predicted column
load) of five theoretical methods, as compared with: 1)
the method proposed in this paper (TPM); 2) the critical
section stress method of the ACI 318-08 code (ACI CSM)1;
3) the strut-and-tie model method in Appendix A of the
book method3 (CRSI); and 5) the method of the Chinese
JGJ94-94 code2 (JGJ94-94). For illustrating the calculating
process of the five aforementioned methods, as an example,
in the Appendix of the paper, give the detailed calculations
of specimen TDS3-1 in Table 3.
It is necessary to point out that: 1) in Table 4, the punching
shear resistance is represented by the column load of pile cap
failing in punching; 2) the bending failure and the failure of
one-way shear are not included in Tables 3 and 4 because
their failure types are not consistent with the failure of the
two-way shear studied in this paper; and 3) as the bottom
reinforcement layout concentrated in the vicinity of the pile
top and the diagonal on the plane of the pile caps have a
larger punching shear resistance than the uniform grid rein-
forcement,9,11,12,16 they will be studied elsewhere.
Table 5 summarizes the statistical appraisal of the Pe/Pp
obtained by all the theoretical methods in Table 4.
Accuracy
It is known from Table 5 that, when all the calculable
specimens are taken, or after the asterisked specimens (the
asterisk implies that the specimens may fail in bending;
more details will be given later) in Table 4 are removed,
although TPM has the largest number of specimens, it has
the highest accuracy. As for evaluations with the remaining
four methods, despite their fewer specimens, they agree well
only for certain of them.
It is known from Table 4 after further analysis that, as far
as individual Pe/Pp calculated by TPM is concerned, except
for the two asterisked specimens, PC454 and T441, which
have rather large calculating deviation (Pe/Pp of T441* is
the minimum in all 98 specimens, while Pe/Pp of PC454*
is the maximum in all 98 specimens), the accuracy of the
remaining specimens is basically good, whereas for PC454,
127
Table 3—Summary of pile cap test results
Column size, mm Reinforcement layout, Test column load
Specimen (diameter or side length) d, mm fc′, MPa No. of bar × bar diameter, mm Bar yield stress fy, MPa at failure, kN
Sabnis and Gogate14 (No. of pile: 4; pile arrangement: determinant; pile diameter, 76.2 mm)
SS01 111.44 31.3 3 × 5.715 each way 499.4 250.4
SS02 111.62 31.3 3 × 3.429 + 4 × 2.68 each way :886.0; :410.1 244.6
SS03 110.87 31.3 7 × 3.429 each way 886 248.0
SS04 111.62 31.3 3 × 5.715 + 3 × 2.68 each way :499.4; :410.1 225.7
76.2 round
SS05 108.59 41.0 7 × 5.715 + 4 × 2.032 each way :499.4; :480.2 263.5
SS06 108.59 41.0 11 × 5.715 each way 499.4 280.2
SG02 117.48 17.9 3 × 9.525 each way 251.2 173.5
SG03 117.48 17.9 4 × 9.525 each way 251.2 176.8
Jimenez-Perez et al. (No. of pile: 4; pile arrangement: determinant; pile diameter, 76.2 mm)
15

MS01 114.30 28.7 275.5

MS02 114.30 28.7 275.5
MS03 114.30 28.7 306.6
MS04 120.65 28.7 291.1
MS05 120.65 31.5 231.1
MS06 107.95 28.7 261.1
MS07 107.95 28.7 287.7
MS15 117.48 31.5 300.0
MS16 76.2 round 117.50 31.5 ♂ ♂ 288.9
MS17 114.30 31.5 310.0
MS19 114.30 31.5 320.0
MS20 107.95 31.5 310.0
MS23 107.95 31.5 313.3
MS24 107.95 31.5 331.1
MS28 101.60 28.7 318.9
MS29 101.60 28.7 293.3
MS30 101.60 31.5 313.3
Taylor and Clarke (No. of pile: 4; pile arrangement: determinant; pile diameter, 200 mm)
16

A001 20.9 10 × 10 each way 410 1110

200 square 400
A009 26.8 10 × 10 each way 410 1450
Adebar et al. (No. of pile: 4; pile arrangement: diamond; pile diameter, 200 mm)
9

A 300 square 445 24.8 9 × 11.3 one way; 15 × 11.3 other way 479 1781
Shen (No. of pile: 4; pile arrangement: determinant; pile diameter, 50 mm)
17

T415 96 16.3 23 × 2.2 each way 233 73.5

T417 79 16.3 12 × 2.2 each way 233 67.6
T420 104 8.4 18 × 1.57 each way 285.5 51.0
T421 102 8.4 23 × 1.57 each way 285.5 52.5
T422 95 10.7 18 × 2.2 each way 249.9 50.7
60 square
T423 92 8.4 20 × 2.2 each way 249.9 57.1
T424 93 8.4 23 × 2.2 each way 249.9 55.9
T425 100 8.4 25 × 2.2 each way 249.9 61.3
T426 100 10.7 17 × 2.8 each way 276.9 59.3
T427 95 10.7 18 × 2.8 each way 276.9 60.5
Notes: ♂ is no reinforcement data provided in the literature; 1 mm = 0.0394 in; 1 MPa = 0.145 ksi; 1 kN = 0.225 kip.

128 ACI Structural Journal/March-April 2015

Table 3 (cont.)—Summary of pile cap test results
Column size, mm Reinforcement layout, Test column load
Specimen (diameter or side length) d, mm fc′, MPa No. of bar × bar diameter, mm Bar yield stress fy, MPa at failure, kN
T428 103 10.7 20 × 2.8 each way 276.9 66.6
T429 97 10.8 22 × 2.8 each way 276.9 78.9
T430 97 10.8 24 × 2.8 each way 276.9 69.1
T432 95 12.5 18 × 2.2 each way 249.9 65.2
T433 96 12.5 18 × 2.8 each way 276.9 78.6
60 square
T435 93 9.8 18 × 2.2 each way 249.9 55.9
T436 98 9.8 18 × 2.8 each way 276.9 52.9
T439 105 9.8 24 × 1.57 each way 285.5 50.0
T441* 104 10.7 24 × 2.2 each way 249.9 41.2
T442 92 12.5 24 × 2.8 each way 276.9 72.8
Shen (No. of pile: 4; pile arrangement: determinant; pile diameter, 100 mm)
17

T452 150 square 225 9.4 11 × 8 each way 276.5 364.6

Shen (No. of pile: 6; pile arrangement: determinant; pile diameter, 100 mm)
17

T601 225 13.5 12 × 10+20 × 6 each way :272.0;:276.3 460.6

150 square
T602 225 9.4 12 × 8+20 × 8 each way , :276.5 441.0
Zhuang18 (No. of pile: 4; pile arrangement: determinant; pile diameter, 100 mm)
PC453 215 12.2 7 × 12 each way 280.2 370
150 square
PC454* 185 17.2 6 × 12 each way 283.5 500
Guo et al.11 (No. of pile: 6; pile arrangement: determinant; pile diameter, 180 mm)
20 × 12 one way;
S1 220 square 259 15.4 318.6 1250
18 × 12 other way
Wu et al.19 (No. of pile: 3; pile arrangement: equilateral triangle; pile diameter, 110 mm)
PC1-1 200 square 400.0 25.4 3 × 10 each way 304.8 910
PC1-2 150 square 399.2 25.5 3 × 10 each way 304.8 790
PC1-3 150 square 399.2 26.9 3 × 8 each way 285.6 790
PC1-4 150 square 400.0 32.8 3 × 10 each way 304.8 880
PC2-1 200 square 330.9 30.1 3 × 10 each way 304.8 780
PC2-2 150 square 329.9 25.0 3 × 10 each way 304.8 720
PC2-3 150 square 329.9 28.7 3 × 8 each way 285.6 680
PC2-4 150 square 331.3 25.5 3 × 10 each way 304.8 650
PC3-1 180 square 260.0 27.1 3 × 10 each way 304.8 670
PC3-2 150 square 260.4 26.4 3 × 10 each way 304.8 620
PC3-3 150 square 260.4 29.1 3 × 8 each way 285.6 550
PC3-4 150 square 261.5 26.9 3 × 10 each way 304.8 630
PC4-1 180 square 179.9 24.1 3 × 10 each way 304.8 530
PC4-2 150 square 180.0 24.0 3 × 10 each way 304.8 490
PC4-3 150 square 180.0 25.0 3 × 8 each way 285.6 426
PC4-4 150 square 180.8 25.4 3 × 10 each way 304.8 610
Wu and Fang20 (No. of pile: 4; pile arrangement: determinant; pile side length, 100 mm)
C2-1 150 square 520 7.54 5 × 8 each way 289.3 559
C2-2 150 square 320 13.4 7 × 8 each way 289.3 630
Yang (No. of pile: 3; pile arrangement: equilateral triangle; pile diameter, 100 mm)
21

YZ1 100 square 210 13.2 3 × 12 each way 310 441

Notes: ♂ is no reinforcement data provided in the literature; 1 mm = 0.0394 in; 1 MPa = 0.145 ksi; 1 kN = 0.225 kip.

ACI Structural Journal/March-April 2015 129

Table 3 (cont.)—Summary of pile cap test results
Column size, mm Reinforcement layout, Test column load
Specimen (diameter or side length) d, mm fc′, MPa No. of bar × bar diameter, mm Bar yield stress fy, MPa at failure, kN
Ma22 (No. of pile: 3; pile arrangement: isosceles triangle; pile diameter, 90 mm)
P5 100(one side) × 222.5
180 20.1 4 × 6 each way 340
P6 140(other side) rectangle 226.4
Suzuki et al. (No. of pile: 4; pile arrangement: determinant; pile diameter, 150 mm)
23

TDS3-1 300 28.0 11 × 9.53 each way 356 1299

250 square
TDM3-1 250 27.0 10 × 12.72 each way 370 1245
Suzuki et al. (No. of pile: 4; pile arrangement: determinant; pile diameter, 150 mm)
24

BDA-30-
200 square 250 24.6 6 × 9.53 each way 549
20-70-2
358
BDA-40-
250 square 350 25.9 8 × 9.53 each way 1019
25-70-1
Suzuki and Otsuki25 (No. of pile: 4; pile arrangement: determinant; pile diameter, 150 mm)
BPB-35-
200 square 290 20.4 9 × 9.53 each way 353 755
20-1
Chan et al.26 (No. of pile: 4; pile arrangement: determinant; pile side length, 150 mm)
C(Chan) 200 square 200 30.74 12 × 10 each way 480.7 870
Ahmad et al. (No. of pile: 4; pile arrangement: determinant; pile diameter, 150 mm)
27

A(Saeed) 150 round 230 20.68 10 × 12.8 + 6 × 6.5 each way 480
, :413
F(Saeed) 150 round 230 27.6 12 × 12.8 + 6 × 6.5 each way 560
Blévot and Frémy28 (No. of pile: 4; pile arrangement: determinant; pile side length, mm: except that 9A3 is 140, others are 350)
4N1 500 square 670 37.3 8 × 32 + 7 × 16 each way :276.2; :279.3 7000
4N1b 500 square 680 40.8 8 × 25 + 7 × 12 each way :440.3; :516.7 6700
:250.6; :281.2;
4N3 500 square 920 34.15 4 × 32 + 4 × 25 + 8 × 12 each way 6500
:293.1
:484.5; :446;
4N3b 500 square 920 49.3 4 × 25 + 4 × 20 + 8 × 10 each way 9000
:429.5
9A3 150 square 470 34.4 16 × 12 450.25 1700
Blévot and Frémy (No. of pile: 3; pile arrangement: equilateral triangle; pile side length, 350 mm)
28

3N2 450 square 462.5 37.7 3 × 32 each way 255 3800

3N2b 450 square 480 43.7 4 × 25 each way 442 4500
3NH 450 square 715 32.65 3 × 32 + 1 × 25 each way :261; :333 5200
3NHb 450 square 730 42.45 4 × 25 439 7200
Miguel et al. (No. of pile: 3; pile arrangement: equilateral triangle; pile diameter, mm: except that B30A4 is 300, others are 200)
29

B20A1/1 350 square 500 27.4 3 × 12.5 each way 591 1512
B20A1/2 350 square 500 33.0 3 × 12.5 each way 591 1648
B20A3 350 square 500 37.9 3 × 12.5 each way 591 1945
B20A4 350 square 500 35.6 3 × 12.5 each way 591 2375
B30A4 350 square 500 24.6 3 × 12.5 each way 591 2283
Chao and Bo (No. of pile: 9; pile arrangement: determinant; pile diameter, 150 mm)
30

CTA 300 square 314 24.88 6 × 16 + 5 × 14 each way :374; :369 1900

Notes: ♂ is no reinforcement data provided in the literature; 1 mm = 0.0394 in.; 1 MPa = 0.145 ksi; 1 kN = 0.225 kip.

there is a statement in Reference 18 that says, “As there’s
no law about the crack distribution of PC454 in the limit
state, it’s hard to say whether the pile cap failure is caused
by bending or punching from the final crack shape.” So this
calculating deviation is probably due to a bending failure.
With respect to T441, no description of the test phenomenon
is provided in literature. But, be it TPM or ACI CSM, CRSI
and JGJ94-94, or the evaluation in Reference 31, whose
author and test conductor of T441 are in the same project
group,31 Pe/Pp values all tend to be small. Furthermore,

130 ACI Structural Journal/March-April 2015

Table 4—Pe/Pp of five theoretical methods
Specimen TPM ACI CSM ACI STM CRSI JGJ94-94 Specimen TPM ACI CSM ACI STM CRSI JGJ94-94
SS01 1.1205 2.0694 3.1300 — 1.5493 T439 0.6900 0.2450 2.3697 0.5952 0.8139
SS02 1.0934 2.0049 2.4757 — 1.5087 T441 *
0.5349 0.2013 1.5907 0.4791 0.6873
SS03 1.1094 2.0667 2.5101 — 1.5471 T442 0.9434 0.5352 2.3560 1.0866 1.5390
SS04 1.0093 1.8500 2.2844 — 1.3916 T452 1.3847 0.4005 2.7189 0.9421 0.7880
SS05 0.9955 1.9812 2.0332 — 1.4482 T601 1.0280 0.9343 3.3474 0.9948 0.6984
SS06 1.0580 2.1068 2.1620 — 1.5431 T602 1.2746 1.0730 2.6018 1.1395 0.8579
SG02 1.0580 1.7350 3.0654 — 1.2674 PC453 1.2084 0.4344 1.8974 0.9439 1.1550
SG03 1.0580 1.7680 3.1237 — 1.2929 PC454* 1.5010 0.8463 2.0610 — 1.3163
MS01 1.1140 2.2769 ♂ — 1.7386 S1 0.9789 1.8629 2.8090 — 1.6480
MS02 1.1140 2.2769 ♂ — 1.7386 PC1-1 1.2248 ↑ 2.6157 0.2124 0.9141
MS03 1.2400 2.5339 ♂ — 1.9347 PC1-2 1.0957 ↑ 2.2526 0.2458 0.7544
MS04 1.1120 2.2053 ♂ — 1.6814 PC1-3 1.0719 ↑ 2.2443 0.2393 0.7279
MS05 0.8593 1.6746 ♂ — 1.2528 PC1-4 1.0602 ↑ 1.9336 0.2412 0.7533
MS06 1.1080 2.3736 ♂ — 1.8099 PC2-1 1.0414 ↑ 1.9593 0.2395 0.8462
MS07 1.2220 2.6155 ♂ — 1.9948 PC2-2 1.0651 ↑ 2.1635 0.3734 0.8422
MS15 1.1470 2.2727 ♂ — 1.6986 PC2-3 0.9564 ↑ 1.9501 0.3296 0.7267
MS16 1.1042 2.1722 ♂ — 1.6357 PC2-4 0.8808 ↑ 1.8705 0.2436 0.7949
MS17 1.2199 2.4409 ♂ — 1.8354 PC3-1 1.0276 ↑ 1.9911 0.4880 0.9706
MS19 1.2590 2.5197 ♂ — 1.8952 PC3-2 0.9888 0.1786 1.8937 0.5871 0.8863
MS20 1.2807 2.6724 ♂ — 2.0164 PC3-3 0.8475 0.1510 1.9978 0.4964 0.7370
MS23 1.2940 2.7009 ♂ — 2.0385 PC3-4 0.8936 ↑ 1.7954 0.4062 0.9418
MS24 1.3680 2.8543 ♂ — 2.1543 PC4-1 1.0474 0.5268 2.1388 1.0454 1.2011
MS28 1.4070 3.1890 ♂ — 2.4443 PC4-2 1.0145 0.7891 1.9862 1.2694 1.2316
MS29 1.2940 2.9330 ♂ — 2.2487 PC4-3 0.8659 0.6709 2.2386 1.0785 1.0406
MS30 1.3450 2.9838 ♂ — 2.2544 PC4-4 1.0357 0.4404 2.1085 1.0535 1.3710
A001 0.6257 0.3833 2.5850 0.7613 0.7056 C2-1 1.1768 0.4381 3.8819 0.2455 0.8534
A009 0.7302 0.4421 3.3768 0.8783 0.7809 C2-2 0.9722 0.8005 2.3684 0.3652 1.0096
A 1.0730 0.7713 5.9605 —  YZ1 1.2250 0.4244 3.9305 1.0023 1.3823
T415 0.7691 0.4047 2.3786 0.8547 0.8762 P5 1.0183 0.7986 2.0488 — 1.1502
T417 0.8524 0.7042 3.3137 — 1.099 P6 1.0362 0.8130 2.0847 — 1.1705
T420 0.7971 0.2818 3.0539 0.6711 0.9508 TDS3-1 1.0384 0.1860 3.5395 0.7466 0.8620
T421 0.8346 0.3165 2.7202 0.7292 1.0100 TDM3-1 1.4596 0.7188 2.5305 1.4426 1.0854
T422 0.7110 0.3587 2.0199 0.7456 0.9741 BDA-30-20-70-2 0.6175 0.3690 1.7825 0.7409 0.6046
T423 0.9918 0.5134 2.9133 1.0382 1.2900 BDA-40-25-70-1 0.7998 ↑ 2.3643 0.4101 0.6105
T424 0.9612 0.4834 2.8376 0.9982 1.2410 BPB-35-20-1 0.9107 0.4704 2.4754 — 0.8224
T425 0.9894 0.4012 3.0049 0.9015 1.2140 C(Chan) 1.1545 1.4711 2.5285 — 1.3894
T426 0.7955 0.3433 2.2548 0.7701 1.052 A(Saeed) 0.9878 2.5236 1.2171 — 2.8267
T427 0.8484 0.4283 2.2000 0.8897 1.1630 F(Saeed) 1.0145 2.7488 1.6577 — 2.9343
T428 0.8721 0.3401 2.4667 0.7929 1.1300 4N1 0.9685 0.5856 1.9958 — 0.7903
T429 1.0787 0.5135 2.8901 1.0958 1.4250 4N1b 0.8333 0.5122 2.2440 — 0.6955
T430 0.9447 0.4497 2.5311 0.9597 1.2480 4N3 0.7217 0.1617 1.1856 0.4876 0.4911
T432 0.8167 0.4245 2.4791 0.8932 1.3100 4N3b 0.7054 0.1864 1.4860 0.5619 0.5324
T433 0.9760 0.4944 2.4952 1.0480 1.5540 9A3 0.9433 ↑ 1.1479 0.4301 0.7856
T435 0.8548 0.4477 2.4735 0.9164 1.0990 3N2 0.9521 1.1006 2.3088 — \
Notes: Pe/Pp is experimental column load/predicted column load; ↑ is infinite bearing load because piles are totally within critical section; ♂ is evaluation cannot be conducted because no
reinforcement data provided; — is calculating condition not applicable;  is not easy to evaluate; and \ is evaluation cannot be conducted because no free end dimensions of pile cap provided.

ACI Structural Journal/March-April 2015 131

Table 4 (cont.)—Pe/Pp of five theoretical methods
Specimen TPM ACI CSM ACI STM CRSI JGJ94-94 Specimen TPM ACI CSM ACI STM CRSI JGJ94-94
T436 0.7733 0.3483 2.1331 0.7557 0.9592 3N2b 0.9436 1.1437 1.8692 — \
3NH 0.9345 0.6902 1.5854 — \ B20A3 1.1254 0.4720 2.1248 — \
3NHb 1.0314 0.2116 1.9773 — \ B20A4 1.4631 0.5947 2.5946 — \
B20A1/1 1.0100 0.4315 1.9508 — \ B30A4 0.6881 0.1063 2.4941 — \
B20A1/2 1.0448 0.4286 1.8004 — \ CTA 1.0270 1.3172 1.7544 — 0.9895
Notes: Pe/Pp is experimental column load/predicted column load; ↑ is infinite bearing load because piles are totally within critical section; ♂ is evaluation cannot be conducted because
no reinforcement data provided; — is calculating condition not applicable;  is not easy to evaluate; and \ is evaluation cannot be conducted because no free end dimensions of pile
cap provided.

Table 5—Statistical appraisal of Pe/Pp obtained by all theoretical formulas in Table 4

Predicting method Total number of specimens Average Standard deviation Coefficient of variation Minimum Maximum
All calculable specimens 98 1.0179 0.1940 0.1906 0.5349 (T441*) 1.501 (PC454*)
TPM Asterisked specimens in
96 1.0179 0.1832 0.1800 0.6175 1.4631
Table 3 removed
All calculable specimens 86 1.1177 0.9079 0.8123 0.1063 3.1890
ACI
CSM Asterisked specimens in
84 1.1318 0.9126 0.8063 0.1063 3.1890
Table 3 removed
All calculable specimens 81 2.3789 0.6796 0.2857 1.1479 5.9605
ACI
STM Asterisked specimens in
79 2.3929 0.6813 0.2847 1.1479 5.9605
Table 3 removed
All calculable specimens 51 0.7228 0.3126 0.4325 0.2124 1.4426
CRSI Asterisked specimens in
50 0.7277 0.3138 0.4313 0.2124 1.4426
Table 3 removed
All calculable specimens 88 1.2526 0.5066 0.4044 0.4911 2.9343
JGJ
94-94 Asterisked specimens in
86 1.2585 0.5087 0.4042 0.4911 2.9343
Table 3 removed

although Pe/Pp of T441 calculated by ACI STM has reached Applicability

1.5907, it is the relatively small value of all the calculable
specimens by ACI STM. Therefore, it can be inferred just
as well that the rather large calculating deviation of T441 is
also attributable to it being probably a bending failure.
In other words, of the five methods, TPM is always capable
of maintaining good accuracy whatever the situation.
Variability
It is known from Table 5 that, when all the calculable
specimens are selected, or after the asterisked specimens in
Table 4 are removed, the variation coefficient of Pe/Pp with
TPM is always smaller than the other four methods. Hence,
TPM is best in terms of calculating stability.
After further analysis, it is known from Tables 4 and 5 that:
1) the average of Pe/Pp with TPM is only slightly larger than
1.0. With the theoretical essence of lower-bound solution of
the SSTM method taken into account, this value should be
rational, while that with the other four methods makes some
deviation from 1.0. In addition, the degree of variability of
the other four methods is also larger, and there is a tendency
that the smaller the punching span is, the larger the column
load calculated will be; and 2) when the asterisked speci-
mens in Table 4 are removed, the minimum or maximum of
Pe/Pp value with TPM is still comparatively rational.
In a word, TPM is safe and reliable, and the potential of
bearing load is also appropriate.
TPM is capable of evaluating all the specimens in Table 3,
so its calculating mode is comparatively unified and not
restricted by the number of piles and the form of pile arrange-
ment. ACI CSM is incapable of evaluation when all the
piles are within the critical section, while ACI STM can not
perform evaluation unless it meets certain restrictions on the
punching span and all the specimen parameters, including
reinforcement, have to be provided at the same time. Like-
wise, CRSI is not applicable unless it is confined to a certain
small punching-span condition. Constrained by the form of
pile arrangement, such as the diamond pile arrangement, as
shown in Table 4, it is not easy to perform evaluation using
JGJ94-94; furthermore, regarding triangle pile arrangement,
JGJ94-94 cannot carry out evaluation unless the free-end
dimensions of the pile cap are provided. Thus, the applica-
bility of TPM is recommendable.
Further analysis is as follows:
1. As mentioned earlier, it has been anticipated during
computerization that the ρ of pile caps with uniform grid
reinforcement has little influence on its punching shear resis-
tance, which is confirmed by test verification in Tables  3
through 5. It should be noted that none of most of the codes
in the world has considered the impact of the longitudinal
reinforcement ratio on the punching shear resistance of
pile caps with uniform grid reinforcement. For instance, it
is not considered in ACI CSM,1 JGJ94-94,2 CRSI3 and the

132 ACI Structural Journal/March-April 2015

critical section stress method of the British code,5 nor is it
mentioned in the formulae of punching shear resistance of
the German32 and Japanese codes.33 This, of course, may
involve considerations of the strength reserve, but there
is also a factor that should not be ruled out—namely, the
punching shear resistance of pile caps with uniform grid
reinforcement is not sensitive to its longitudinal reinforce-
ment ratio, as demonstrated in References 14 and 15. The
aforementioned discussion shows that if longitudinal rein-
forcement is arranged according to a uniform grid, it is not
necessary to impose restrictions on the reinforcement ratio
of tension tie ρ for TPM.
2. As previously mentioned, during derivation of fce, under
the prerequisite for punching failure, the punching-span
ratio λ is given a large range of variation—namely, 0.15 to
2.0—and concrete strength fc′ basically contains the whole
range of ordinary concrete strength as well—namely, 6.7 to
50 MPa (971.5 to 7252 psi). Likewise, a large range of vari-
ation of λ and fc′ is also embodied in Table 4. But it can
be seen that for all ascertained punching failure specimens,
their theoretical values calculated by TPM just agree well
with the test values. Therefore, it can be asserted that, if only
what has happened is a punching failure, on the one hand,
there is no need to restrict the punching-span ratio for TPM;
on the other hand, TPM is also applicable to all the pile caps
with ordinary concrete strength.
3. As previously mentioned, the model in computerization
is not exclusively developed for a certain pile number or a
certain form of pile arrangement. What it selects is the strut
model, so, as a result, the obtained results should be generally
applicable to an arbitrary pile number and an arbitrary form
of pile arrangement as can be seen from Tables 3 through 5.
It is necessary to point out that, for pile caps with the pile
beneath the column, as frequently seen in engineering prac-
tice, it can be imagined that the bearing mechanism of the
SSTM is still tenable. But as the punching failure of the pile
caps is a result of extension and development of diagonally
splitting crack, it is unlikely for punching failure to occur in
the strut located between the column and the pile beneath the
column. Consequently, the column load of punching failure
of this kind of pile caps should be the sum of the following
two parts: the first part, the column load of punching failure
with no pile beneath the column; and the second part, the
actual load borne by the pile beneath the column. Of the
two parts, the evaluation of the first part can be carried out
with TPM, while that of the second part, as can be seen from
the previously mentioned load-transfer mechanism, can be
performed reversely with pile load distribution at the elastic
stage, thus bypassing quite a lot of inconvenience in the
evaluation of the statically indeterminate spatial truss at the
plastic stage. Therefore, TPM has extensive applicability.
CONCLUSIONS
In this paper, through the NFE analysis and the derivation
of GLSMV, a new method, TPM, for evaluating punching
shear resistance of pile cap with uniform grid reinforcement
is presented. In view of the good agreement between TPM
and experimental data, with ρ, λ, fc′, pile number, and pile
ACI Structural Journal/March-April 2015
arrangement form variable, and the definite advantages in
terms of accuracy, variability, and applicability as compared
with the other four methods, TPM can be widely applicable
to the evaluation of the punching shear resistance of the
general pile cap with uniform grid reinforcement.
AUTHOR BIOS
Honglei Guo is a Professor in the Department of Civil Engineering at
Wuhan Polytechnic University, Wuhan, China. He received his BS and
MS from Wuhan University in 1988 and 1993, respectively, and his PhD
from Southeast University, Nanjing, China, in 1997. His research interests
include shear strength and optimal design of reinforced concrete structures.
REFERENCES
1. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-08) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2008, 473 pp.
2. China Academy of Building Research, Technical Code for Building
Pile Foundations, China Architecture and Building Press, China, 1995,
pp. 64-66. (in Chinese)
3. Concrete Reinforcing Steel Institute, CRSI Handbook, seventh edition,
Schaumburg, IL, 1992, 840 pp.
4. CAN/CSA A23.3-04(R2010), “Design of Concrete Structures,” Cana-
dian Standards Association, Toronto, ON, Canada, 2010, pp. 63-65.
5. BSEN 1992-1-1:2004, “Eurocode 2: Design of Concrete Struc-
tures-Part 1-1: General Rules and Rules for Buildings,” British Standards
Institution, London, UK, 2004, pp. 107-110.
6. AS3600-2001, “Concrete Structures,” Council of Standards Australia,
Sydney, Australia, 2001, pp. 124-125.
7. Wen, B. S., “Strut-and-Tie Model for Shear Behavior in Deep Beams
and Pile Caps Failing in Diagonal Splitting,” ACI Structural Journal, V. 90,
No. 4, July-Aug. 1993, pp. 356-363.
8. Kinnunen, S., and Nylander, H., “Punching of Concrete Slabs without
Shear Reinforcement,” Transactions No. 158, Royal Institute of Tech-
nology, Stockholm, Sweden, 1960, 112 pp.
9. Adebar, P.; Kuchma, D.; and Collins, M. P., “Strut-and-Tie Models for
the Design of Pile Caps: An Experimental Study,” ACI Structural Journal,
V. 87, No. 1, Jan.-Feb. 1990, pp. 81-92.
10. Breña, S. F., and Morrison, M. C., “Factors Affecting Strength of
Elements Designed Using Strut-and-Tie Models,” ACI Structural Journal,
V. 104, No. 3, May-June 2007, pp. 267-277.
11. Guo, H. L.; Ding, D. J.; and Jiang, Y. S., “Study for Load Transfer
Mechanism of Space Truss Model Simulating Thick Pile Caps (1),” Indus-
trial Construction, China, V. 27, No. 8, Aug. 1997, pp. 30-35. (in Chinese)
12. Guo, H. L.; Ding, D. J.; and Jiang, Y. S., “Study for Load Transfer
Mechanism of Space Truss Model Simulating Thick Pile Caps (2),” Indus-
trial Construction, China, V. 27, No. 9, Sept. 1997, pp. 36-40. (in Chinese)
13. Malvar, L. J., “Punching Shear Failure of a Reinforced Concrete
Pier Deck Model,” ACI Structural Journal, V. 89, No. 5, Sept.-Oct. 1992,
pp. 569-576.
14. Sabnis, G. M., and Gogate, A. B., “Investigation of Thick Slab (Pile
Cap) Behavior,” ACI Journal Proceedings, V. 81, No. 1, Jan.-Feb. 1984,
pp. 35-39.
15. Jimenez-Perez, R.; Sabnis, G. M.; and Gogate, A. B., Experimental
Behavior of Thick Pile Caps Design of Concrete Structures—The Use of
Model Analysis, Elsevier Applied Science Publishers, 1985, pp. 221-229.
16. Taylor, H. P. J., and Clarke, J. L., “Some Detailing Problems in
Concrete Frame Structures,” The Structural Engineer, V. 54, No. 1,
Jan. 1976, pp. 19-29.
17. Shen, J. H., “Elastic and Plastic Analysis of Pile Caps,” MASc thesis,
Tong Ji University, Shanghai, China, 1985, pp. 102-103. (in Chinese)
18. Zhuang, G. M., “Analysis of Three-Dimensional Nonlinear Finite
Element of Pile Caps,” MASc thesis, Tong Ji University, Shanghai, China,
1988, pp. 75-82. (in Chinese)
19. Wu, R. P.; You, H. M.; and Ji, J., “Study on the Bearing Capacity of
Thick Pile Cap with Three Piles,” Journal of Building Structures, China,
V. 14, No. 1, Jan.-Feb. 1993, pp. 63-71. (in Chinese)
20. Wu, R. P., and Fang, X. D., “Bearing Capacity Study and Test of Four-
Pile Deep Pile Cap,” Proceedings of High-Rise Buildings and Bridge Foun-
dation Engineering Academic Conference, Guang Zhou, Rock Mechanics
and Engineering Institute of Guang Dong Province, 1989, pp. 32-51.
(in Chinese)
21. Yang, Z., “Analysis of Elastic Stress and Internal Force Atlas of Pile
Caps,” MASc thesis, Tong Ji University, Shanghai, China, 1986, pp. 9-74.
(in Chinese)
133
 22. Ma, X. Q., “Analysis of Ultimate Strength of Pile Cap with Three
Piles,” MASc thesis, Tong Ji University, Shanghai, China, 1989, pp. 55-59.
(in Chinese)
23. Suzuki, K.; Otsuki, K.; and Tsubata, T., “Experimental Study on
Four-Pile Caps with Taper,” Transactions of the Japan Concrete Institute,
V. 21, 1999, pp. 327-334.
24. Suzuki, K.; Otsuki, K.; and Tsuhiya, T., “Influence of Edge Distance
on Failure Mechanism of Pile Caps,” Transactions of the Japan Concrete
Institute, V. 22, 2000, pp. 361-368.
25. Suzuki, K., and Otsuki, K., “Experimental Study on Corner Shear
Failure of Pile Caps,” Transactions of the Japan Concrete Institute, V. 23, 2002.
26. Chan, T. K., and Poh, C. K., “Behaviour of Precast Reinforced
Concrete Pile Caps,” Construction and Building Materials, V. 14, No. 2,
2000, pp. 73-78. doi: 10.1016/S0950-0618(00)00006-4
27. Ahmad, S.; Shah, A.; and Zaman, S., “Evaluation of the Shear
Strength of Four Pile Cap Using Strut and Tie Model (STM),” Journal of
the Chinese Institute of Engineers, V. 32, No. 2, 2009, pp. 243-249. doi:
10.1080/02533839.2009.9671501
134
28. Blévot, J. L., and Frémy, R., “Semelles sur Pieux,” Institute Technique
du Bâtiment et des Travaux Publics, V. 20, No. 230, 1967, pp. 223-295.
29. Miguel, M. G.; Takeya, T.; and Giongo, J. S., “Structural Behaviour of
Three-Pile Caps Subjected to Axial Compressive Loading,” Materials and
Structures, V. 41, No. 1, 2007, pp. 85-98. doi: 10.1617/s11527-007-9221-5
30. Guo, C., and Lu, B., “Experimental Study on the Load-Carrying
Properties of Nine-Pile Thick Caps under a Column,” China Civil Engi-
neering Journal, V. 43, No. 1, 2010, pp. 95-102. (in Chinese)
31. Zhou, K. R., “Study of the Whole Process, Mechanism and Bearing
Capacity of Punching,” PhD thesis, Tong Ji University, Shanghai, China,
1990, 74 pp. (in Chinese)
32. “Tragwerke aus Beton, Stahlbeton und Spannbeton Teil 1:
Bemessung und Konstruktion,” Normenausschuss Bauwesen (NABau) im
DIN Deutsches Institut für Normung e. V., Berlin, Germany, 2001.
33. Japan Road Association, “Specifications for Highway Bridges IV;
Substructures,” Tokyo, Japan, 2002.
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S12
Design Implications of Large-Scale Shake-Table Test on
Four-Story Reinforced Concrete Building
by T. Nagae, W. M. Ghannoum, J. Kwon, K. Tahara, K. Fukuyama, T. Matsumori, H. Shiohara,
T. Kabeyasawa, S. Kono, M. Nishiyama, R. Sause, J. W. Wallace, and J. P. Moehle
A full-scale, four-story, reinforced concrete building designed in
accordance with the current Japanese seismic design code was
tested under multi-directional shaking on the E-Defense shake
table. A two-bay moment frame system was adopted in the longer
plan direction and a pair of multi-story walls was incorporated
in the exterior frames in the shorter plan direction. Minor adjust-
ments to the designs were made to bring the final structure closer to
U.S. practice and thereby benefit a broader audience. The resulting
details of the test building reflected most current U.S. seismic
design provisions. The structure remained stable throughout the
series of severe shaking tests, even though lateral story drift ratios
exceeded 0.04. The structure did, however, sustain severe damage
in the walls and beam-column joints. Beams and columns showed
limited damage and maintained core integrity throughout the series
of tests. Implications of test results for the seismic design provi-
sions of ACI 318-11 are discussed.
Keywords: collapse; damage; design; full-scale; moment frame; multi-
story; shake table; shear wall.
INTRODUCTION
Code requirements for reinforced concrete have evolved
significantly around the world in the past decades. In the
United States, the 1971 San Fernando, CA, earthquake was
a watershed event leading to the introduction of require-
ments for ductile reinforced concrete buildings, which
have evolved incrementally since that time based on field
and laboratory experiences. In Japan, following a history of
several damaging earthquakes and many laboratory tests, the
Japanese seismic design code was substantially revised in
1981. In the 1995 Hyogoken-Nanbu earthquake, many rein-
forced concrete buildings designed before 1981 experienced
major failures, especially in the first-story columns and walls.
Although newer reinforced concrete buildings designed in
accordance with the revised 1981 code showed improved
resistance against collapse, several sustained severe damage
due to their large deformations. Such damage made it diffi-
cult to continue using them after the earthquake and resulted
in high repair costs. This experience demonstrates that
further improvements in seismic design of concrete build-
ings might be desirable for the future.
It was in light of the aforementioned experiences that a
large-scale shake-table testing program was conducted in
2010. Within the program, a full-scale, four-story, rein-
forced concrete building designed in accordance with the
present Japanese seismic design code was tested by using
the E-Defense shake table. The main objectives of the study
related to the concrete building were: 1) to verify methods
for assessing performance such as strength, deformation
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
capacity, and failure mode; 2) to identify suitable compu-
tational methods to reproduce the seismic responses of the
building; and 3) to develop a practical method for assessing
damage states regarding reparability.
Design and instrumentation of the test structure were
performed with input from U.S. co-authors. Wherever
possible, minor adjustments to the designs were made to
bring the final structure closer to U.S. practice and thereby
benefit a broader audience. The resulting details of the test
building reflected the most current U.S. seismic design
provisions (Nagae et al. 2011b).
Summaries of the global behavior of the test building
and key local damage and deformation observations are
presented. A comparison between the details of the test
structure and U.S. seismic design practices is also provided.
Implications of test results for the seismic design provi-
sions of ASCE 7-10 (ASCE/SEI Committee 7 2010) and
ACI 318-11 (ACI Committee 318 2011) are discussed. In a
related publication (Nagae et al. 2011a), the seismic design
provisions of the Architectural Institute of Japan (AIJ 1999)
were evaluated in light of test results.
RESEARCH SIGNIFICANCE
Current Japanese and U.S. seismic design provisions are
based on pseudo-dynamic component tests, sub-assembly
tests, and limited dynamic tests of partial structural systems.
The test presented is a first-of-its-kind, multi-directional,
dynamic test of a complete, full-scale reinforced concrete
building system to near collapse damage states. The test
provides unique data on component and system performance
that are used to evaluate current seismic design provisions
and highlight potential code changes.
SPECIMEN DETAILS
Figure 1 shows the plans and framing elevations of the
reinforced concrete test building. Figure 2 shows a photo-
graph of the test building on the E-Defense shake table. The
height of each story is 3 m (118.1 in.). The building footprint
measures 14.4 m (47 ft 3 in.) in the longer (X) direction, and
7.2 m (23 ft 7.5 in.) in the shorter (Y) direction. A two-bay
moment frame system was adopted in the longer (X) plan
direction and a pair of multi-story walls were incorporated
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-022.R2, doi: 10.14359/51687421, received May 21, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
135
Fig. 1—Framing and reinforcing details. (Note: Dimensions are in mm; 1 mm = 0.039 in.)
Fig. 2—Reinforced concrete (left) and prestressed concrete
(right) specimens on the E-Defense shake table.
in the exterior frames in the shorter (Y) plan direction. The
thickness of the top slab was 130 mm (5.1 in.). Rigid steel
frames were set within the open stories of the test specimen
for collapse prevention and measurement of story defor-
mations. Representative building mechanical equipment
136
was incorporated to assess potential damage during strong
seismic motions. Table 1 lists the various weights of the test
specimen. The weight was estimated based on the reinforced
concrete members, the fixed steel frames, and the equip-
ment. Figure 1 shows dimensions and reinforcement details
of typical members. The test building was designed in accor-
dance with current Japanese seismic design practice.
When constructing the test building, columns, walls,
beams, and the floor slab were cast monolithically. The
longitudinal reinforcement of columns, beams, and the wall
boundaries were connected by gas pressure welding. Lap
splices were used for the reinforcement of other parts of
the walls and the floor slabs. The frames in the test building
were nominally identical in design and detailing. The shear
walls at axes A and C contained the same amount of longitu-
dinal reinforcement but differed in the spacing of transverse
reinforcement (Fig. 1). A complete set of drawings and spec-
imen details can be found in Nagae et al. (2011b). Additional
test data can be found on the NEEShub website (NEEShub
2011) and in Tuna (2012).
SPECIMEN DESIGN
The extent to which the test structure satisfies the seismic
design provisions of ASCE 7-10 and ACI 318-11 is explored
ACI Structural Journal/March-April 2015
Table 1—Weight and design forces Table 2—Material properties of concrete
Fourth Third Second Fc, σB, Ec,
(A) Structural elements, kN Roof floor floor floor N/mm2 N/mm2 N/mm2
Column 53 106 106 106 Cast of fourth story and roof floor slab 27 41.0 30.5
Beam 240 240 240 240 Cast of third story and fourth floor slab 27 30.2 30.3
RC
Wall 40 79 79 79 Cast of second story and third floor slab 27 39.2 32.8
Slab 484 428 424 420 Cast of first story and second floor slab 27 39.6 32.9
Sum 816 853 849 845 Notes: Fc is specified concrete compressive strength; σB is measured concrete compres-
sive strength; and Ec is measured secant modulus of concrete; 1 N/mm2 = 0.145 ksi.
(B) Non-structural Fourth Third Second
elements, kN Roof floor floor floor Table 3—Material properties of steel
Stair and handrail 6 6 6 6
Steel Grade Anominal, mm2 σy, N/mm2 σt, N/mm2 Es, kN/mm2
Measurement frame 0 3 17 17
D22 SD345 387 370 555 209
Equipment 112 5 0 0
D19 SD345 287 380 563 195
Sum 118 14 23 23
D13 SD295 127 372 522 199
Total of (A) and (B), kN 934 867 872 867
D10 SD295 71 388 513 191
Fourth Third Second First
D10 SD295 71 448 545 188
story story story story
D10 KSS785 71 952 1055 203
ΣWi, kN 934 1801 2673 3541
Notes: Anominal is nominal area of reinforcing bars; σy is measured yield strength of
Ci = 0.2 × Ai 0.29 0.25 0.22 0.20 steel reinforcement; σt is measured ultimate strength of steel reinforcement; and Es
Qi, kN 273 450 593 708 is measured elastic modulus of steel reinforcement; 1 mm2 = 0.0016 in.2; 1 N/mm2 =
0.145 ksi.
Notes: Wi is weight of floor i; Ai is shape factor for vertical distribution of lateral forces
for floor i; Ci is lateral force at floor i as a fraction of ƩWi; and Qi is shear at story i; 1 relationship. The secant stiffness corresponding to the flex-
kN = 0.225 kip.
ural strength was calculated in accordance with provisions
in this section. The building specimen was designed to with- of the MLIT standard (2007). Beyond flexural yielding, the
stand the seismic lateral forces presented in Table 1 (MLIT stiffness was reduced to 0.01 times the initial effective stiff-
2007) without members exceeding their elastic limits. These ness. The pushover analysis indicates that the ultimate base-
forces, which sum to 20% of the weight of the structure, are shear strength of the building specimen is approximately
higher than those that would be specified by ASCE 7-10 0.42W (1500 kN [337 kip]) in the frame direction and 0.51W
(Section 12.8.1.3), which caps seismic lateral forces for (1800 kN [405 kip]) in the wall direction.
a low-rise building to 1/R times the structure weight for a Figure 4 shows the column-beam moment strength ratios.
design basis earthquake, where R is the response modifica- Reinforcement of the top slab was reflected in the moment
tion coefficient (8 for special reinforced concrete moment strength of beams in negative bending (top in tension). Effec-
frames and 6 for special reinforced concrete shear walls). tive flange widths of beams were adopted in accordance
The vertical distribution of the design forces, given by the with the recommendations of the 2007 MLIT Standard or
parameter Ai in Table 1, is similar to the ASCE 7-10 specifi- ACI 318-11, which produced roughly similar flange widths.
cation (approximate inverted triangular distribution). Variations of column axial forces due to lateral forces were
Results of material tests are given in Tables 2 and 3. In estimated from pushover analysis in the Japanese calcu-
subsequent evaluations, the moment and shear strengths lations. In the U.S. calculations, a plastic mechanism was
of each member were calculated adopting the compressive assumed in which hinging of the columns occurs at the foun-
strength of concrete and the yield strength of steel reinforce- dation and just below the roof, and beam hinging occurs at
ment obtained by averaging material test results. column faces at intermediate floors in the frame direction. In
To aid in the design of the test specimen, pushover the wall direction, the assumed plastic mechanism considered
(nonlinear static) analyses were conducted on line-element hinging of the columns and walls at the foundation, and beam
models of the structure. Figure 3 presents pushover results hinging at column and wall faces. Discrepancies in column-
for the final test specimen details. The analytical model used beam moment strength ratios evaluated using ACI and MLIT
for pushover analyses was built following work by Kabeya- procedures (Fig. 4) can mostly be attributed to differences in
sawa et al. (1984). The effective flange width of a top slab the estimates of axial forces on columns. From the second to
was adopted in accordance with the recommendations of fourth floors, the column-beam moment strength ratios were
the 2007 MLIT Standard. A vertical distribution defined by slightly below 1.0 for interior columns, while those of exte-
the parameter Ai (Table 1) was adopted for the lateral force rior columns ranged from approximately 1.0 to 1.87.
distribution. In the analytical model, inelastic deformations
of beam elements were represented by rotational springs Assessment of specimen design in accordance
at the ends of elements. The first and second break points with U.S. seismic design practice
corresponding to member cracking strength and flexural The structure was assessed in both the x- and y-direc-
strength were assigned in the tri-linear moment-rotation tions using ACI 318-11 and ASCE 7-10 provisions. The

ACI Structural Journal/March-April 2015 137

Fig. 3—Pushover analysis results. (Note: 1 kN = 0.225 kip.)
Fig. 4—Moment strength ratios of columns to beams.
goal was to determine how well the structure compares
with U.S. seismic design practices. Rather than presume
that the building was to be constructed at a particular site
with corresponding site seismic hazard, the assessments of
seismic design requirements are based on a seismic hazard
represented by the linear response spectrum for the 100%
JMA-Kobe ground motion to which the test structure was
subjected.
Shear wall direction (y-direction)—The approximate
natural period in the shear wall direction is 0.31 seconds
based on Eq. 12.8-7 in ASCE 7-10. The spectral acceleration
corresponding to this period is approximately 2.5g for the
100% JMA-Kobe ground motion imparted to the structure
(Fig. 5, y-direction). Elastic analysis was performed using
equivalent (static) lateral forces corresponding to the spec-
tral acceleration divided by an R factor of 6, as specified in
ASCE 7-10 for a building frame system with special rein-
138
forced concrete shear walls. Equivalent lateral forces were
distributed over the height of the structure in accordance with
provisions of ASCE 7-10. An effective moment of inertia
equal to 50% of the gross moment of inertia was used over
the full wall height: an intermediate value between the effec-
tive moments of inertia provided in ACI 318-11 for cracked
and uncracked walls. Selected wall effective moments of
inertia are also consistent with values recommended by
ASCE 41-06 (ASCE/SEI Committee 41 2007a) for cracked
walls. An effective moment of inertia equal to 30% of the
gross moment of inertia was used for beams and columns as
per ASCE 41-06 – supplement 1 (ASCE/SEI Committee 41
2007b) provisions for beams and columns with low axial
loads. Beams were considered T-beams with an effective
flange width evaluated in accordance with provisions of
ACI 318-11. Joints were taken as rigid. Elastic analysis of
the walls decoupled from frames at Axes A and C indicates
ACI Structural Journal/March-April 2015

Fig. 5—Acceleration response spectra of input waves. (Note:
Damping ratio = 0.05; 1 m/s2 = 39.37 in./s2.)
that the walls would develop their design moment strength
(0.9 × nominal moment strength) at approximately 0.37/R
of the JMA-Kobe 100% motion. If wall-frame interaction is
taken into account, however, the wall-frame system would
develop its design moment strength at approximately 0.55/R
of the 100% JMA-Kobe motion. Thus, the building in the
wall direction has only 55% of the strength that would be
required for the JMA-Kobe motion if that motion is consid-
ered as the design earthquake shaking level. In subsequent
discussion, wall-frame interaction is taken into account.
When applying the equivalent lateral-force distribution in
accordance with ASCE 7-10, wall flexural yielding occurs
at a lower load than that generating the wall’s factored shear
strength. Distributed vertical and horizontal steel satisfied all
shear reinforcement requirements of ACI 318-11.
The wall-foundation interface was not intentionally
roughened prior to casting the walls. Given the amount
of longitudinal steel crossing the interface, the axial force
on the walls, and a friction coefficient of 0.6, nominal
shear-friction strength in accordance with ACI 318-11 of
both wall bases was approximately 2140 kN (482 kip). That
shear-friction strength exceeded estimated shear demands by
approximately 55% based on the 100% JMA-Kobe ground
motion. Nominal shear-friction strength was, however, only
20% higher than maximum base shear demand estimated
from pushover analysis (approximately 1800 kN [405 kip]),
which accounts to some extent for member over-strength.
ACI 318-11 allows the use of two methods to determine
if boundary elements are required in walls. If the drift-based
method is considered (ACI 318-11, Section 21.9.6.2), no
boundary elements are required in the walls for the 100%
JMA-Kobe motion, whether drift estimates are obtained
considering wall-frame interaction or not. If the stress-
based method is considered (ACI 318-11, Section 21.9.6.3),
however, boundary elements are required in the walls up to
a height of 7550 mm (297 in.) from the base of the wall if
walls are considered decoupled from the frames, and a height
of 5060 mm (199 in.) if wall-frame interaction is accounted.
If one considers that boundary elements are not required in
the walls, minimum boundary detailing in both walls satis-
fies ACI 318-11 provisions. If one considers that boundary
elements are required, however, the provided spacing
of hoops in the boundary elements of the wall at Axis C
(100 mm [3.94 in.]) marginally exceeds the required spacing
ACI Structural Journal/March-April 2015
(83 mm [3.26 in.]). In the wall at Axis A, hoops were spaced
at 80 mm (3.15 in.) in the first story and this spacing satis-
fies all ACI 318 hoop spacing requirements for the boundary
element. In the upper stories of the wall at Axis A, hoops
in the boundary regions were spaced at 100 mm (3.93 in.)
and therefore did not satisfy the ACI 318-required spacing
of 83 mm (3.26 in.).
If wall-frame interaction was considered, beams spanning
between shear walls and corner columns were found to have
sufficient moment strength to resist moments from elastic
analysis based on the 100% JMA-Kobe motion hazard level.
Shear strengths of the beams were sufficient to develop
beam probable moment strengths.
Because demands on corner columns in the shear wall
direction were significantly lower than demands on the same
columns in the frame direction, capacity and detailing of
corner columns will be described in the section discussing
the frame direction (x-direction).
Frame direction (x-direction)—The approximate natural
period in the moment frame direction is 0.44 secomds based
on ASCE 7-10 Eq. 12.8-7. The spectral acceleration corre-
sponding to this period is approximately 1.45g for the 100%
JMA-Kobe ground motion imparted to the structure (Fig. 5,
x-direction). Elastic analysis was performed using equivalent
(static) lateral forces corresponding to the spectral accelera-
tion divided by an R factor of 8, as specified in ASCE 7-10
for special reinforced concrete moment frames. Equivalent
lateral forces were distributed over the height of the structure
in accordance with ASCE 7-10. Elastic analysis of the frames
indicates that the first-story corner columns reach design
flexural strength at a shaking level corresponding to approx-
imately 1.4/R of the JMA-Kobe 100% motion. All frame
member strengths therefore exceeded the required design
strength corresponding to a 100% JMA-Kobe hazard level.
Factored shear strengths of all beams were not sufficient
to develop probable moment strengths due to the require-
ment that concrete shear contribution be taken as zero
(ACI 318-11, Section 21.5.4.2). Maximum beam shear
stresses corresponding to the development of probable
moment strengths ranged from 2.0 to 2.7 times the square
root of the concrete compressive strength in psi (0.17 to
0.22 MPa). The spacing of beam transverse reinforcement
was 200 mm (7.87 in.) in the critical plastic hinge regions,
which exceeds the maximum allowable spacing of 120 mm
(4.72 in.) as required by ACI 318-11.
Factored shear strengths of the third- and fourth-story
columns were not sufficient to develop probable moment
strengths. Column shear stresses corresponding to the devel-
opment of column probable moment strengths ranged from
1.4 to 3.8 times the square root of the concrete compressive
strength in psi (0.114 to 0.315 MPa). Column-end trans-
verse reinforcement met spacing and layout requirements of
ACI 318-11 in the first two stories but not the top two stories.
No columns met the requirement for minimum volumetric
reinforcement ratio in the critical end regions; columns had
20 to 50% of the hoop volumes required by ACI 318-11
in the critical end regions. Transverse reinforcement ratios
varied substantially between columns in different stories due
to differences in numbers of crossties.
139
Table 4—Key response values at roof
Test No. Maximum roof acceleration
Input wave x-direction, m/s 2
y-direction, m/s 2
1 JMA-Kobe 25% 3.12
2 JMA-Kobe 50% 7.03
3 JMA-Kobe 100% 9.65
4 JR-Takatori 40% 6.46
5 JR-Takatori 60% 8.09
*
Maximum roof drifts do not include residual drifts accrued from previous tests.
Notes: 1 m2/s = 39.37 in./s2; 1 mm = 0.039 in.
Fig. 6—Maximum interstory drift distribution.
Joint shear demands for both interior and exterior joints
were calculated considering force equilibrium on a hori-
zontal plane at the midheight of the joints, in accordance
with ACI 318-11. Joint shear demands calculated including
the contribution of slab flexural tension reinforcement within
the ACI 318 effective flange width were found to be approx-
imately 20 to 40% higher than demands computed ignoring
the slab contribution. Note that ACI 318 does not require
consideration of the slab reinforcement in calculations of
joint shear demand. Regardless of whether slab contribution
was taken into account, all joint design shear strengths, based
on ACI 318-11, exceeded joint shear demands. Because
joints were only confined by hoops without crossties, the
maximum center-to-center horizontal spacing between hoop
or crosstie legs was larger than the ACI 318-11 limit of
350 mm (14 in.). The provided hoop spacing in the joints
of 140 mm (5.5 in.) was larger than the maximum spacing
allowed by ACI 318-11 of approximately 25 mm (1 in.) for
the provided arrangement of hoops without crossties (limited
by minimum volumetric reinforcement ratio requirements).
Other joint detailing satisfied ACI 318-11 requirements,
including those for longitudinal bar anchorage.
Figure 4 shows column-beam nominal moment strength
ratios. Below the roof, all strength ratios for exterior columns
140
Maximum roof drift* Residual roof drift
x-direction, mm y-direction, mm x-direction, mm y-direction, mm
6.37 16.9 24.2 0.5 0.4
11.01 122.4 106.9 1.1 5.4
14.01 242.7 323.9 6.2 22.5
8.13 240.4 240.8 1.3 7.9
9.99 278.1 414.0 8.0 11.6
satisfied the 6/5 minimum requirement of ACI 318-11. That
requirement was not satisfied at interior joints.
E-DEFENSE SHAKE-TABLE FACILITY AND
TEST CONDITIONS
The E-Defense shake-table facility has been operated
by the National Research Institute for Earth Science and
Disaster Prevention of Japan since 2005. The table is 20 x
15 m (65 ft 7 in. x 49 ft 3 in.) in plan dimension and can
produce a velocity of 2.0 m/s (78.7 in./s) and a displace-
ment of 1.0 m (39.4 in.) in two horizontal directions simul-
taneously. It can accommodate a specimen weighing up to
1200 tonnes (1323 tons). In this series of tests, the consid-
ered reinforced concrete building was tested side-by-side
with a prestressed concrete building having almost the same
configuration and overall dimensions (Fig. 2). More detail
about the test structure, including detailed drawings, can be
found in Nagae et al. (2011b).
LOADING PROGRAM
Ground motions designated as JMA-Kobe and JR-Taka-
tori, recorded in the 1995 Hyogoken-Nanbu earthquake,
were adopted as the input base motions. The North-South-
direction wave, East-West-direction wave, and vertical-
direction wave were input to the y-direction, x-direction, and
vertical direction of the specimen, respectively. The inten-
sity of input motions was gradually increased to observe
damage progression. The adopted amplitude scaling factors
for JMA-Kobe were 10, 25, 50, and 100%. Following the
JMA-Kobe motions, the JR-Takatori motion scaled to
40 and 60% was applied to impart large cyclic deformations.
Figure 5 presents the acceleration response spectra for the
input motions. JMA-Kobe 100% has a strong intensity in the
short-period range corresponding to the natural period of the
specimen, as can be seen in Fig. 5. The JR-Takatori 60% has
a strong intensity in the longer-period ranges corresponding
to estimated damaged specimen periods.
TEST RESULTS
Maximum recorded story drift and global behavior
White-noise inputs were applied prior to each main test.
From these, the initial natural periods of the test building
were found to be 0.43 seconds in the frame direction and
0.31 seconds in the wall direction, which compare favor-
ably with periods estimated using ASCE 7-10 Eq. 12.8-7
(0.44 seconds in the frame direction and 0.31 seconds in the
ACI Structural Journal/March-April 2015
Fig. 7—Damage state of moment frame with cracks highlighted.
wall direction). Figure 6 shows the distribution of maximum
story drift over the height of the specimen for the shaking
tests. In the frame direction, the story drift is larger in the
first and second stories than in the third and fourth stories.
In the wall direction, the story drifts are relatively uniform,
although the drifts become larger in the first story than drifts
of other stories in the JMA-Kobe 100% test and JR-Takatori
tests. The structure remained stable through all the severe
dynamic tests and thus satisfied the minimum collapse-
prevention performance objective. Table 4 lists the maximum
recorded roof level accelerations, drifts, and residual drifts
for all earthquake simulation tests. Residual drifts were
relatively low, with a maximum recorded value of 22.5 mm
(0.88 in.) in the wall direction at the end of the JMA-Kobe
100% motion.
Damage states of members
Figures 7 through 9 show images of damage in the lower
parts of the specimen. After the JMA-Kobe 50% test, the
interior beam-column joints of the second floor and the
column and wall bases of the first story showed minor
cracking. In the interior beam-column joints, the maximum
measured inclined crack width of 0.5 mm (0.02 in.) after the
JMA-Kobe 50% test increased to 2.5 mm (0.1 in.) after the
JMA-Kobe 100% test. Eventually, inclined cracks in the inte-
rior beam-column joints at the second floor reached 5.3 mm
(0.21 in.) after the JR-Takatori 60% test. Maximum inclined
crack widths at beam ends and exterior beam-column joints
were limited to approximately 1.5 mm (0.06 in.), even after
the JR-Takatori 60% test. Compressive failure of concrete
apparently due to large flexural deformations was observed
in column and wall bases. The cover concrete of column
bases partially spalled to a height of 250 mm (9.8 in.) in the
ACI Structural Journal/March-April 2015
JMA-Kobe 100% test, and completely spalled to a height
of 200 to 400 mm (7.9 to 15.8 in.) in the JR-Takatori 60%
test. The core concrete of column bases remained adequately
confined by transverse reinforcement even after the JR-Taka-
tori 60% test.
The corner portion of both wall bases suffered compressive
failure to a height of 300 mm (11.8 in.) and length of 600 mm
(23.6 in.) in the JMA-Kobe 100% test. The longitudinal rein-
forcement in that region had lateral offset due to inelastic
buckling. Wall sliding at both wall bases was observed in the
JMA-Kobe 100% and subsequent tests. Significant sliding
was primarily observed following crushing of the wall
boundary zones (Wallace 2012), which may have weakened
the wall-foundation interface shear friction resistance. The
sliding mechanism affected the maximum drift and deforma-
tion demands in the test structure and may have accentuated
the damage observed in the wall boundary regions. Sliding
of the walls at their base reached approximately 100 mm
(3.93 in.) during the JMA-Kobe 100% test and accounted for
up to 10% of the roof drifts during that motion.
Local deformations
The shear deformations of the second-floor interior
joints are highlighted first because these joints sustained
severe damage and degradation. Shear deformations of the
second-floor interior beam-column joints were measured
in the frame direction, as shown in Fig. 8(a). Figure 8(b)
shows the history of the shear deformation angles as well
as the average story drift angles of the upper and lower
stories during the JMA-Kobe 100% test. Peaks a to e in the
response history (Fig. 8(b)) are identified for later reference.
Assuming that the shear deformation angle of the beam-
column joint contributes to the average story drift angle, as
141
Fig. 8—Deformation of interior beam-column joint in JMA-Kobe 100%.
shown in Fig. 8(c), the deformation ratio is defined as the
ratio of the shear deformation angle to the average story drift
angle. Figure 8(d) shows the deformation ratio from Peaks a
to e. The deformation ratio was 0.35 at Peak a (when the
average story drift ratio reached 0.009) and reached more
than 0.6 at Peak d. Figure 8(e) shows the development of
inclined cracks in the joint at Peaks b, c, and d.
The rotation and lateral slip deformations of the wall base
were measured in the y-direction using instrumentation
shown schematically in Fig. 9(a). The histories of the base
rotation angle, lateral slip, and first-story drift and drift angle
during the JMA-Kobe 100% test are shown in Fig. 9(b) and
(f). Peaks of story drift are denoted a to g for cross reference
with other figures. Figure 9(c) shows an overall photograph
of the wall at Peak c. A local compressive failure is seen at
the base corner of the A-side, and several tension cracks are
seen at the lower part of the B-side. Figure 9(d) shows the
deformation ratio at the peak story drifts in the JMA-Kobe
100% test. The deformation ratio is defined as the ratio of
drift due to base rotation and lateral sliding to story drift. At
Peak c, the story drift was mostly derived from the rotation
and lateral sliding of the wall base. Because the maximum
lateral sliding displacement becomes approximately constant
after the maximum deformation of Peak c, the deformation
ratio of lateral sliding increased at Peaks e and g. Figure 9(e)
shows the damage of a wall base after the test. From video
observations, lateral sliding became significant at Peak c and
142
the local buckling of bars occurred at the base of B-side in
the cycle when the story drift approached Peak d.
Global hysteretic behavior and strength
The global drift angle is defined as the relative horizontal
displacement of the fourth floor level (Fig. 1) divided by its
height above the base. The base shear force was calculated
based on the horizontal inertia forces given by the estimated
weight of each floor and the corresponding floor accelerations.
In shear force calculations, the weights of vertical elements
were lumped with floor weights as presented in Table 1.
Figure 10 shows the relationship between the base shear
force and global drift angle. In the relationships, the hyster-
etic loops show inelastic behavior, while the stiffness is
observed to decrease with an increase in the drift angle, as
evinced by the decrease in reloading stiffness with increasing
drift angles. The history of story shear force (Fig. 10) indi-
cates that the elongation of the first-mode period is more
significant in the frame direction than in the wall direction in
the JMA-Kobe 50% test, while the period in the wall direc-
tion elongated noticeably in the JMA-Kobe 100% test due to
the damage incurred by the shear walls. The apparent lowest
periods of the structure estimated by the white-noise input
were 0.99 seconds in the frame direction and 0.88 seconds
in the wall direction after the JMA-Kobe 100% test. It is
useful to note that measured base shear forces reached a
maximum of approximately 85% of the building weight
ACI Structural Journal/March-April 2015
Fig. 9—Deformation of wall base in JMA-Kobe 100%. (Note: 1 mm = 0.039 in.)
in the wall direction and 55% in the frame direction. Thus,
actual strength was well in excess of the design lateral force
level of 0.2W in each direction.
Figure 11 shows the distribution of the story shear coef-
ficient over the height of the structure. The story shear
coefficient is defined as the story shear force divided by
the weight of the floors above that story, normalized by the
value of the coefficient at the first story. The figure presents
values of the coefficient evaluated using the maximum story
shear forces recorded during a given motion (“Max” in the
figure), and values of the coefficient evaluated using story
shear forces occurring at the same time instant when the
base shear reaches its maximum (“Base Peak” in the figure).
Also presented in the figure are the design shear coefficients
prescribed in Japanese design practice (given by the factor
Ai in the 2007 MLIT standard). Equivalent story shear coef-
ACI Structural Journal/March-April 2015
ficients estimated using the ASCE 7-10 equivalent later-
al-force procedures are also shown in the figure. It is useful
to note that the distribution of the story shear coefficients
corresponds to a similar distribution of applied floor inertia
forces; for example, an inverted triangular distribution of
story shear force coefficients implies an inverted triangular
distribution of floor inertia forces. Figure 11 indicates that
floor inertia forces at peak base shear had a relatively uniform
distribution over the height of the building, as opposed to
an inverted triangular distribution often assumed in design,
especially in the JMA-Kobe 100% and the JR-Takatori 60%
tests. Such uniform vertical seismic force distributions have
been observed in previous shake-table tests (for example,
Kabeyasawa et al. 1984). Higher mode contributions and
localization of damage may have influenced the observed
vertical distribution of lateral forces. Such observation
143

Fig. 10—Hysteretic behavior and history of base shear
force. (Note: 1 kN = 0.225 kip.)
can partly explain the higher than estimated base shear
forces seen in Fig. 10. This is particularly the case in the
wall direction where observed base shear forces during the
JMA-Kobe 100% motions were more than 50% larger than
those estimated from pushover analysis; which was based on
an approximate inverted triangular lateral load distribution
(Fig. 3).
144
Overturning moment at the base of the first story is mostly
produced by the first-mode response of a structure and is
relatively insensitive to the distribution of the lateral forces
(Kabeyasawa et al. 1984). Roof drift is also relatively
insensitive to higher modes. Thus, the relation between
roof drift angle and base moment is a convenient measure
for comparing calculated and laboratory test strengths.
Figure 12 shows the measured relationships between roof
drift angle and overturning moment. Calculated overturning
moments, obtained by pushover analyses at maximum story
drift ratio of 0.02 (Fig. 4), are also shown in Fig. 12. In the
y-direction, the measured maximum overturning moment is
1.3 times the calculated value, while in the x-direction, the
measured maximum overturning moment is 1.5 times the
calculated value. Several factors may have contributed to
the measured overstrength, including underestimation of the
slab contribution to member strengths, other three-dimen-
sional effects, and strain-rate effects.
IMPLICATIONS OF TEST RESULTS TO ACI 318-11
Although columns had 20 to 50% of the hoop volumes
required by ACI 318-11 in the critical end regions, they
performed adequately, maintaining core integrity through
the full series of severe dynamic tests. It is noted, however,
that column axial forces were relatively low, varying
from an estimated tensile force on corner columns due to
uplift, to a maximum compressive axial force of approx-
imately 0.1Ag fc′ at the first story (where Ag is the column
gross-section area and fc′ is the measured concrete compres-
sive strength). This observation suggests that the volume
of transverse reinforcement required by ACI 318-11 may
be reduced in the axial force ranges of the tested columns.
Several design codes (including the Japanese MLIT Stan-
dard [2007], CSA A23.3-04 [2004], and NZS 3101 2006
[2006a,b]) account for the effects of axial force on confine-
ment requirements of concrete columns. While these codes
treat the effects of axial forces in different ways, they
generally require less confinement reinforcement for lower
axial forces.
Similarly, the volume ratios of hoops in the critical
regions of the beams were 60% of the ratios required. Beams
performed adequately and suffered relatively minor damage
while maintaining core integrity throughout the dynamic
tests. It is important to note that the beams were under rela-
tively low shear stresses. Such observations indicate that
beams under low shear stresses and conforming to the prin-
ciples of ACI 318-11 but with somewhat lighter transverse
reinforcement can meet life-safety performance objectives.
Both shear walls sustained notable damage, including
cover spalling and bar buckling, during the first high-
intensity ground motion (JMA-Kobe 100%). It is noteworthy
that confined boundary elements were not even required
by the ACI 318 provisions (using the displacement-based
approach). One of the reasons for the inconsistency here is
that the measured lateral displacements were approximately
twice the design values. Considering the measured displace-
ments, ACI 318 provisions would have required confined
boundary elements.
ACI Structural Journal/March-April 2015
Fig. 11—Distribution of floor lateral force coefficient.
Fig. 12—Hysteretic behavior based on overturning moment. (Note: 1 kN-m = 0.737 k-ft.)
Although confinement was not required by the ACI 318
provisions, the wall boundaries nonetheless contained
confinement reinforcement satisfying the ACI 318 special
boundary element requirements at Axis A and nearly satis-
fying them at Axis C. The observed concrete spalling and
longitudinal reinforcement buckling exceeded expectations
of some of the authors, and may suggest a need for improved
detailing requirements.
The nominal shear-friction strength at the wall-foundation
interface, calculated in accordance with ACI 318-11, was
2140 kN (482 kip) for both walls combined. Shear demands
on the first story were estimated to be 1400 kN (315 kip)
based on the JMA-Kobe 100% ground motion being the
design motion, 1800 kN (405 kip) based on pushover anal-
ysis, and 3000 kN (675 kip) based on recorded data. Measured
base shear demands were 40% larger than the calculated
shear-friction capacity of the wall-foundation interface. Test
data therefore indicate that improvements on methods for
estimating peak shear demands on wall systems should be
sought. Notably, the effects of higher modes and localized
damage on the vertical distribution of lateral loads should be
considered when estimating peak story-shear demands.
The interior beam-column joints sustained significant
damage during the earthquake simulation tests. Impli-
cations for ACI 318 are not readily extracted, however,
because the beam-column joint designs did not satisfy the
ACI 318 requirements. Deficiencies included deficient ratios
of column-beam flexural strength ratios and deficient volu-
metric ratio of joint transverse reinforcement.
ACI Structural Journal/March-April 2015
SUMMARY AND CONCLUSIONS
A full-scale, four-story, reinforced concrete building struc-
ture was tested on the E-Defense shake table. The structure
was designed in accordance with the present Japanese seismic
design code. Minor adjustments to the design were made to
bring the final structure closer to U.S. practice and thereby
benefit a broader audience. The structure was subjected to a
series of multi-directional seismic base motions including
three high-intensity motions. The following key observa-
tions were made:
1. The structure remained stable throughout the tests, even
though lateral drift ratios exceeded 0.04. Thus, the structure
satisfied a collapse-prevention performance objective. The
structure did, however, sustain severe damage in the walls
and beam-column joints.
2. At times of maximum base shear, the distribution of
lateral inertia forces was approximately uniform over height,
unlike the inverted triangular distribution used to design
the structure. The nearly uniform lateral force distribution,
along with other factors, resulted in a significant increase in
the maximum base shear during the tests. Test data therefore
indicate that improvements on methods for estimating peak
shear demands on wall systems should be sought.
3. Both walls suffered significant damage in their
boundary regions, including wall boundary crushing, longi-
tudinal reinforcement buckling, and lateral instability. Walls
had tightly spaced hoop reinforcement at the boundaries that
satisfied all ACI confinement requirements at Axis A and
nearly satisfied them at Axis C. ACI 318-11 provisions for
the transverse reinforcement of special structural walls may
need to be adjusted if more limited damage is desired, partic-
ularly for thin walls with relatively large cover.
145
 4. Significant sliding at the wall-foundation construction
joint was observed at the base of both walls. The sliding
mechanism affected the maximum drift and deformation
demands in the test structure and may have accentuated
the damage observed in the wall boundary regions. Three
factors may have contributed to the observed sliding. First,
although the construction joint between the walls and the
foundation were cleaned, they were not intentionally rough-
ened as required by ACI 318-11. Second, although design
shear demands were less than the sliding shear strength
calculated in accordance with ACI 318-11, the actual test
shears were much higher than the design values. Third,
damage to the wall-boundary regions may have reduced the
shear-friction strength at the wall-foundation joints. These
observations suggest two issues that may not be adequately
treated in current codes. First, that higher-mode contribu-
tions and effects of localized damage should be accounted
for when estimating shear force demands on shear walls, and
second, that integrity and stability of the wall boundary zone
is an important component of wall sliding shear resistance.
5. Columns performed adequately and maintained core
integrity throughout the series of severe tests even though
they did not satisfy the confinement volumetric reinforce-
ment ratio requirements of ACI 318-11. Column axial force
ratios were relatively low and did not exceed 10% of the
column gross-section axial capacity. Test results therefore
indicate that it might be possible to reduce the ACI 318-11
minimum volumes of confining reinforcement for columns
with low axial force ratios.
6. Beams also performed adequately and maintained core
integrity even though they did not satisfy the confinement
volumetric reinforcement ratio requirements of ACI 318-11.
Beam shear stresses were, however, relatively low and did
not exceed 2.7 times the square root of concrete compressive
strength in psi (0.22 in MPa).
7. Joints performed poorly, exhibiting wide inclined cracks
and deformations that accounted for up to 60% of floor drifts
at the end of the test series. Interior joints performed worse
than exterior joints. It is noted that the joint designs satisfied
Japanese code requirements but did not satisfy ACI 318-11
code requirements.
AUTHOR BIOS
T. Nagae is a Senior Researcher at the National Research Institute for
Earth Science and Disaster Prevention, Tsukuba, Japan.
W. M. Ghannoum is an Assistant Professor at the University of Texas at
Austin, Austin, TX.
J. Kwon is a PhD Candidate at the University of Texas at Austin.
K. Tahara is a Researcher at the National Research Institute for Earth
Science and Disaster Prevention.
K. Fukuyama is a Visiting Researcher at the National Research Institute
for Earth Science and Disaster Prevention.
T. Matsumori is a Senior Researcher at the National Research Institute for
Earth Science and Disaster Prevention.
H. Shiohara is a Professor at the University of Tokyo, Tokyo, Japan.
T. Kabeyasawa is a Professor at the University of Tokyo.
S. Kono is a Professor at the Tokyo Institute of Technology, Tokyo, Japan.
146
M. Nishiyama is a Professor at Kyoto University, Kyoto, Japan.
R. Sause is a Professor at Lehigh University, Bethlehem, PA.
J. W. Wallace is a Professor at the University of California, Los Angeles,
Los Angeles, CA.
J. P. Moehle is the T.Y. and Margaret Lin Professor at the University of
California, Berkeley, Berkeley, CA.
ACKNOWLEDGMENTS
The authors acknowledge the generous support of the Ministry of Educa-
tion, Culture, Sports, Science & Technology (MEXT) and of the National
Research Institute for Earth Science and Disaster Prevention of Japan for
carrying out the tests presented in this paper. Participation by the U.S.
co-authors was supported by the Pacific Earthquake Engineering Center
and by the Network for Earthquake Engineering Simulation of the National
Science Foundation under award CMMI-1000268. Additional instrumenta-
tion of the test structure using NEES@UCLA sensors was provided under
Award CMMI-1110860, while analysis of the data was partly funded by
the National Science Foundation under Award No. 1201168. Any opinions,
findings, and conclusions or recommendations expressed in this material are
those of the authors and do not necessarily reflect the views of the sponsors.
REFERENCES
ACI Committee 318, 2011, “Building Code Requirements for Structural
Concrete (ACI 318-11) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 503 pp.
AIJ, 1999, “Design Guidelines for Earthquake Resistant Reinforced
Concrete Buildings Based on the Inelastic Displacement Concept,” Archi-
tectural Institute of Japan, Tokyo, Japan, 440 pp. (in Japanese)
AIJ, 2010, “Standard for Structural Calculation of Reinforced Concrete
Structures,” Architectural Institute of Japan, Tokyo, Japan, 526 pp. (in Japanese)
ASCE/SEI Committee 41, 2007a, “Seismic Rehabilitation of Existing
Structures (ASCE/SEI 41-06),” American Society of Civil Engineers,
Reston, VA, 428 pp.
ASCE/SEI Committee 41, 2007b, “Supplement to Seismic Rehabilita-
tion of Existing Buildings (ASCE/SEI 41-06),” American Society of Civil
Engineers, Reston, VA, 428 pp.
ASCE/SEI Committee 7, 2010, “Minimum Design Loads for Buildings
and Other Structures (ASCE/SEI 7-10),” American Society of Civil Engi-
neers, Reston, VA, 636 pp.
CSA A23.3-04, 2004, “Design of Concrete Structures,” Canadian Stan-
dards Association, Mississauga, ON, Canada, 258 pp.
Kabeyasawa, T.; Shiohara, H.; and Otani, S., 1984, “U.S.-Japan Coop-
erative Research on R/C Full-Scale Building Test, Part 5: Discussion of
Dynamic Response System,” Proceedings of the 8th World Conference on
Earthquake Engineering, San Francisco, CA, pp. 627-634.
MLIT, 2007, “Technological Standard Related to Structures of Buildings,”
Ministry of Land, Infrastructure, Transport, and Tourism, Tokyo, Japan.
Nagae, T.; Tahara, K.; Fukuyama, K.; Matsumori, T.; Shiohara, H.;
Kabeyasawa, T.; Kono, S.; Nishiyama, M.; and Nishiyama, I., 2011a,
“Large-Scale Shaking Table Tests on A Four-Story RC Building,” Journal
of Structural and Construction Engineering, V. 76, No. 669, pp. 1961-1970.
doi: (Transactions of AIJ)10.3130/aijs.76.1961
Nagae, T.; Tahara, K.; Taiso, M.; Shiohara, H.; Kabeyasawa, T.; Kono,
S.; Nishiyama, M.; Wallace, J. W.; Ghannoum, W. M.; Moehle, J. P.; Sause,
R.; Keller, W.; and Tuna, Z., 2011b, “Design and Instrumentation of the
2010 E-Defense Four-Story Reinforced Concrete and Post-Tensioned
Concrete Buildings,” PEER Report 2011/104, Pacific Earthquake Engi-
neering Research Center (PEER), Berkeley, CA, 261 pp.
NEEShub Project 2011-1005, “U.S. Instrumentation and Data Processing
for the Four-Story Reinforced Concrete and Post-Tensioned E-Defense
Building Tests,” The George E. Brown, Jr. Network for Earthquake Engi-
neering Simulation, https://nees.org/warehouse/report/project/1005. (last
accessed Feb. 2014)
NZS3101, Part 1:2006, 2006a, “Concrete Structures Standard: Part 1—
The Design of Concrete Structures,” Standards Association of New
Zealand, Wellington, New Zealand, 309 pp.
NZS3101, Part 2:2006, 2006b, “Concrete Structures Standard—Commen-
tary,” Standards Association of New Zealand, Wellington, New Zealand, 397 pp.
Tuna, Z., 2012, “Seismic Performance, Modeling, and Failure Assess-
ment of Reinforced Concrete Shear Wall Buildings,” PhD dissertation,
University of California, Los Angeles, Los Angeles, CA, 298 pp.
Wallace, J. W., 2012, “Behavior, Design, and Modeling of Structural
Walls and Coupling Beams—Lessons from Recent Laboratory Tests and
Earthquakes,” International Journal of Concrete Structures and Materials,
V. 6, No. 1, pp. 3-18. doi: 10.1007/s40069-012-0001-4
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S13
Inverted-T Beams: Experiments and Strut-and-Tie Modeling
by N. L. Varney, E. Fernández-Gómez, D. B. Garber, W. M. Ghannoum, and O. Bayrak
Contrary to rectangular deep beams, inverted-T beams are loaded
on a ledge at the bottom chord of the beam. This loading configura-
tion induces a tension field into the web and the resulting complex
strain distribution renders sectional design provisions inadequate.
The applicability of strut-and-tie modeling (STM), developed for
rectangular deep beams and simpler, two-dimensional designs, was
evaluated. An experimental study was conducted in which 33 tests
were performed on 22 large-scale reinforced concrete inverted-T
beams and the effects of the following variables were investigated:
ledge geometry, quantity of web reinforcement, number of point
loads, member depth, and shear span-depth ratio. It was concluded
that strut-and-tie modeling, although developed for much simpler
structural components, offers a simple and accurate design method
for the more complex strain distributions in inverted-T beams.
The STM provisions developed for rectangular beams accurately
captured both failure mode and ultimate capacity and are recom-
mended for use in inverted-T beam design, as a major conclusion
of this research.
Keywords: D-region; inverted-T beam; laboratory testing; large-scale;
nonlinear design; reinforced concrete; shear; shear span; strut-and-tie.
INTRODUCTION
Inverted-T bent caps are often used in construction to
reduce the overall elevation of bridges and/or to improve
available clearance beneath the beams, as shown in Fig. 1.
The bent caps are beams that support bridge girders on
ledges near the bottom of the beam, effectively loading the
cap along its tension chord. Within a given cross section
(transverse direction), the loads are transferred from the
ledges to the bottom of the web and then “hung” vertically
to the compression chord, generating tension fields in the
web at the loading points. The loads are then transferred
in the longitudinal direction to the supports, as in a typical
compression-chord-loaded beam. This three-dimensional
flow of forces, in addition to the deep beam loading condi-
tions commonly encountered in bent caps, generate regions
of stress discontinuities that are traditionally designed using
empirical equations and rules of thumb.
Significant web shear cracking of recently built inverted-T
straddle bent caps has been reported in Texas, according to
the Texas Department of Transportation (TxDOT), trig-
gering concerns about the current design procedures. Most
inverted-T bent caps in Texas are designed using sectional
provisions for the web and a series of checks for the ledges
that closely follow the procedures found in the AASHTO
LRFD Bridge Design Specifications.1 Due to the load and
geometric discontinuities in inverted-T beams, this beam
theory is not valid; thus, sectional design provisions cannot
be used to properly design such structures.
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
In the past two decades, many structural design codes
have adopted strut-and-tie modeling (STM) as a more trans-
parent option for the design of deep beams and other struc-
tures with discontinuities. The current STM provisions were
developed for rectangular deep beams and simple structures
with two-dimensional strain distributions, but have not been
experimentally investigated for more complex structural
elements such as inverted-T beams.
Due to scarcity of experimental research on inverted-T
beams, a comprehensive large-scale experimental program
was undertaken to examine the behavior of such structural
elements and assess the validity of implementing STM
design. Thirty-three specimens were tested as part of the
research program. Unlike those found in the literature,
the test specimens in this program were considered more
representative of inverted-T beams designed in practice in
terms of their size, geometric and loading properties, and
reinforcement details. This paper presents the STM design
provisions as applied to inverted-T beams, the laboratory
test results, and the corresponding design recommendations.
RESEARCH SIGNIFICANCE
Significant diagonal web shear cracking of inverted-T
bent caps may represent a risk both in terms of strength
and serviceability. Due to the nonlinear distributions of
strains in inverted-T beams, STM offers a safe, lower-bound
design alternative to examine forces and predict the failure
mode in an element. Current strut-and-tie provisions were
developed for rectangular deep beams and have not been
investigated for the three-dimensional state of stress present
in these structures. The research presents an extensive large-
scale experimental program aimed at assessing the accuracy
and conservatism of strut-and-tie modeling for the design of
inverted-T bent caps. The unique experimental data presented
in this paper and the assessment of STM design provisions
is considered to be significant contributions to the literature.
BACKGROUND AND STRUT-AND-TIE MODELING
Typically for reinforced concrete beams, a designer makes
the assumption that plane sections remain plane, referred
to as the Bernoulli hypothesis or beam theory. Within this
theory, the strains in the beam are presumed to vary linearly
through the depth of a section; thus, the beam is said to be
dominated by sectional behavior. As shown in Fig. 2(a),
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-064.R1, doi: 10.14359/51687403, received June 10, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
147
Fig. 1—Typical inverted-T bent cap.
these regions of linear stress (or strain) are referred to as
B-regions (with the “B” for beam or Bernoulli).
A D-region (with “D” standing for discontinuity or
disturbed) can typically be found on either side of a B-region,
as shown in Fig. 2. These regions cannot be designed
using the sectional procedures as the strain distribution is
nonlinear and thus the assumptions used to derive the beam
theory are no longer valid. These disturbances are caused
either by abrupt changes in geometry or loading. Frame
corners, dapped ends, openings, and ledges are examples of
geometric discontinuities. Point loads such as girder bear-
ings or support reactions are examples of load discontinu-
ities. According to St. Venant’s Principle, an elastic stress
analysis would indicate that the effects of a disturbance
dissipate at approximately one member depth away from the
discontinuity.2 In other words, a D-region extends approx-
imately one effective member depth d from the load or
geometric change.
A deep beam is one in which the entire span involves
predominant nonlinear strain distributions through the depth
of the section. For this strain condition to exist, the shear
span a must be less than approximately 2 to 2.5 times the
effective member depth d. The right shear span in Fig. 2(a) is
entirely composed of D-regions and is thus considered a deep
beam. A beam with a greater shear span-depth ratio (a/d),
as shown in the left span, is assumed to behave according
to beam theory and can be designed using sectional proce-
dures. The design of deep beams requires the use of STM or
other non-linear procedures outside the scope of this paper.
STM offers an approach for obtaining lower-bound solu-
tions for the strength design based on simple truss theory.
The resultants of complex states of stress are idealized as
a system of uniaxial force elements, or a truss, within the
member as shown in Fig. 2(b). This system will yield a
conservative design if the resulting truss model is in equi-
librium with the external forces and the concrete has enough
deformation capacity to accommodate the assumed distri-
bution of forces.3 Proper anchorage of the reinforcement is
crucial. The factored forces also must not exceed the factored
strengths as shown in the following equation.
ϕFn ≥ Fu (1)
148
Fig. 2—(a) B- and D-regions in a rectangular beam; and
(b) corresponding STM.
Strut-and-tie models consist of three components: struts,
ties, and nodes. These are assembled together to represent
the flow of forces through a structure, as shown in Fig. 2(b).
After calculating the external reactions and defining the
geometry of the STM, the individual member forces of the
truss are determined through statics.
Struts are compression elements that vary in shape
depending on their location. Represented by dashed lines,
struts can be bottle-shaped if allowed to spread along their
length or prismatic in regions of uniform stresses, such as
the compression zone of a beam’s B-region. It is important
to provide transverse reinforcement to control the tensile
stresses caused by the spreading of compressive forces in
bottle-shaped struts.4
Tension elements, or ties, are generally made up of rein-
forcing steel and denoted by solid lines, as seen in Fig. 2.
Enough reinforcement must be provided to carry the tensile
demand of the tie and should be distributed so that the
centroid coincides with the tie location. Details such as bar
spacing and anchorage are essential for proper STM.
Due to the concentration of forces from intersecting truss
members, the nodes are the most highly stressed regions of a
structural member. Three types of nodes can exist within an
STM and are denoted by the intersecting elements. Within
the nodal designations, “C” refers to a compression element,
such as a strut, an externally applied load, or a support reac-
tion, and “T” stands for tension or tie. A CCC node is one
in which only struts intersect, a CCT node has tie(s) inter-
secting in only one direction, and a CTT node has ties inter-
secting in two different directions. The type of node governs
its behavior and thus its strength.
The versatility of STM allows it to be used to design any
structure and accommodate various load transfer mechanisms.
In theory, as long as the principles required to achieve a lower
bound solution are met, any model can be considered a safe
design. A model should, to the furthest extent possible, follow
the actual stress field as determined by an elastic stress anal-
ysis. If the model varies substantially from the stress field, the
structure will undergo substantial deformations leading to an
increased chance of cracking. Schlaich et al.3 provides addi-
tional discussion on using STMs to design structural concrete
members. The applicability of STMs has been validated with
experimental testing on rectangular deep beams.5 However,
ACI Structural Journal/March-April 2015
experimental testing on more complicated structures such as Table 1—ACI 318-116 and Birrcher et al.4 STM
inverted-T beams is limited. concrete efficiency factors
ACI 318-11
Recent advances in STM provisions
Strut, fce = Node, fce =
ACI 318-116 STM examines the strength of struts and
0.85βsfc′ βs 0.85βnfc′ βn
nodes separately. For the strut-node interface, the smaller
of the nominal compressive strength of the end of the strut, Prismatic 1.00 CCC 1.00
Fns, and of the face of the nodal zone at which the strut acts, Bottle-shaped* 0.75 CCT 0.80
Fnn, is used. These nominal compressive strengths are calcu- Tension flange 0.40 CTT 0.60
lated using their respective cross sectional areas, Acs and Anz,
Birrcher et al.
as shown in following equations. The concrete efficiency
factors, βs and βn, are summarized in Table 1. Node, fcu = Strut-node
mνfc′ Bearing face, ν Back face, ν interface†, ν

Fns = 0.85βs fc′Acs (2) CCC 0.85 f c′

0.45 ≤ 0.85 − ≤ 0.65
CCT 0.70 20
Fnn = 0.85βs fc′Acs (3) *
Without reinforcement satisfying ACI 318 Section A3.3.3, βs = 0.60λ.
†
Without reinforcement satisfying AASHTO Article 5.6.3.5, ν = 0.45.
Birrcher et al. conducted a thorough investigation of rect-
4

angular deep beam behavior to improve upon the current
STM provisions and recommend modifications to both
the ACI 318-116 and AASHTO LRFD1 codes. Potentially,
the most significant modifications that affect the design
of inverted T-beams focused on the node strength. This
procedure simplifies the design of struts by considering
the strut-node interfaces, which implicitly accounts for the
strut capacity and eliminates redundant stress checks at the
same location.
In Birrcher et al.,4 the design strength of the node, Fn, is deter-
mined by the limiting compressive stress at the faces of the node
by the concrete efficiency factor ν in the following equation.
Fs = 0.85νfc′Acn (4)
The appropriate concrete efficiency factor, ν, is used to
reduce the compressive strength of the concrete in the node
depending on the type of node (CCC, CCT, or CTT) and
face (bearing face, back face, strut-node interface) under
consideration. The factors developed by Birrcher et al.4 are
summarized in Table 1 along with the existing factors in
Appendix A of ACI 318-11.6 It should be noted that Birrcher
et al.4 recommended removing reference to CTT nodal
regions as they are typically smeared nodes and emphasis
for deep beam design should be placed on the more critical
CCC and CCT nodal regions. The cross-sectional area of the
node, Acn, such as the strut-node interface shown in Fig. 3, is
limited in the perpendicular direction by either the width of
the bearing plate or web width bw.
For bearing areas smaller than the width of the struc-
tural member, the concrete strength for all the faces in that
node was increased to account for triaxial confinement. The
triaxial confinement factor, m, is in Article 5.7.5 of AASHTO
LRFD1 and Section 10.14.1 of ACI 318-11.6
It can be noted from Table 1 that the efficiency factor for
a strut-node interface is given as the same for both CCC and
CCT nodes according to the recommendations put forth by
Birrcher et al.4 These provisions do not reduce the nodal
strength due to the presence of a tension field in an inverted-T
beam as the node below the applied load is a CCT node,
rather than a CCC node. It was of interest to observe how
well these provisions captured the strength and behavior of
the inverted-T beams with their additional tension field.
The effectiveness of the Birrcher et al.4 STM design provi-
sions were demonstrated by using an extensive database of
rectangular deep beams. Improvements were made in the
overall conservatism and accuracy as well as simplicity of
STM for deep beams as compared to current ACI 318-116 and
AASHTO1 procedures. However, the application of STMs
was not investigated experimentally for more complicated
three-dimensional structures such as inverted-T beams. Due
to their unique geometry, certain assumptions not addressed
in the current STM procedures had to be made in the design
of inverted-T beams.
Strut-and-tie modeling of inverted-T beam specimens
Inverted-T beams transfer load in three dimensions: from
the ledges to the web, from the tension to the compression
chord, and from the loading points to the supports. To capture
this behavior, it is necessary to consider a three-dimensional
strut-and-tie model. To simplify the analysis, the model is
divided into two compatible two-dimensional models.
The STM design of an inverted-T beam is often iterative,
as many factors are interdependent. First, the overall geom-
etry was determined based on the experimental variables
under consideration, and the preliminary loads and corre-
sponding reactions were calculated. The STMs were then
detailed to carry the required loads. Diagonal shear cracking
was a primary concern in this study, thus the beams were
designed to ensure a shear failure.
Defining geometry of longitudinal strut-and-tie model—
An example of a simple longitudinal STM for an inverted-T
beam with two shear spans is shown in Fig. 4. Each tie was
aligned with the centroid of the reinforcing bars. Vertical
hanger bars were placed at each load point with the tie corre-
sponding to the center of the bearing pad. A 45-degree spread
on the ledge under the loading plates defined the width of
hanger ties.7 For cut-off ledges, load spread was limited on
one side, as shown in the STM.
The horizontal tie along the bottom of the beam was
aligned with the centroid of the flexural reinforcement. The
width of the tie was assumed to be twice the distance from

ACI Structural Journal/March-April 2015 149

Fig. 3—Geometry of CCT Support Node A.
Fig. 4—Longitudinal strut-and-tie model.
the extreme tension fiber to the centroid of the steel as shown
in the detail of Node A in Fig. 3.
All ties must be properly anchored to achieve the assumed
stress distribution. ACI requires that the yield strength of the
tie be developed at the critical point where the centroid of
the tie meets the end of the extended nodal zone or edge of
the diagonal strut as shown in Fig. 3.
The location of the intermediate (stirrup) Tie BC for the
two-panel mode in Fig. 4 was determined using the technique
proposed by Wight and Parra-Montesinos8 that any stirrup that
intersects an adjacent strut at an angle greater than 25 degrees
can be engaged as part of the vertical tie in the CCT node.
A line projected at a 25 degree angle from the edge of the
support plate at Node A to the top of the beam defined the
left limit of the tie. The right limit was defined by the edge of
the 45-degree load spread for hanger Tie DE and all stirrups
that fell within the rectangular shaded region were considered
part of the vertical Tie BC, as shown in Fig. 4. The 45-degree
load spread was an initial assumption validated with data from
strain gauges applied to the hanger reinforcement.
The horizontal strut along the top of the beam was
assumed to be prismatic with a depth equal to the depth
of the equivalent rectangular compression stress block as
defined from a typical flexural analysis. Although not tech-
nically valid in a D-region due to the nonlinear distribution
of strains, defining the depth of the strut using a flexural
analysis is considered conservative and the assumption is well-
established in practice.9
Diagonal bottle-shaped struts, represented by dashed lines,
complete the flow of forces in the longitudinal strut-and-tie
model. The angles between struts and ties were checked to
150
ensure they were greater than or equal to 25 degrees but
less than 65 degrees. This limit was enforced to prevent
an incompatibility of strains.4 A smaller angle would result
in the tension tie overlapping more of the diagonal strut,
decreasing its effectiveness. The forces in each element were
calculated using statics and the element size and location
adjusted as needed.
Defining geometry of cross-sectional strut-and-tie
model—A cross-section STM was required to design the
ledge of the inverted-T beam. The external loads were
applied equally to each ledge as shown in Fig. 5. The hanger
ties in the longitudinal model corresponded to the vertical
reinforcement. The top of the ledge reinforcement corre-
sponded to the horizontal ledge tie. The centroid of the hori-
zontal strut shown was located at the depth of the flexural
reinforcement from the longitudinal model. A diagonal strut
transferred the load from the bearing plate to the bottom of
the hanger bars.
Shear spans under investigation—Once the forces in the
truss members were calculated and the nodes checked using
the Birrcher et al.4 recommendations, the required steel
area was determined to satisfy the tie tensile forces. Proper
anchorage of the ties was provided within the extended
nodal regions.
Shear spans a equal to 2.50d and 1.85d were examined, as
shown in Fig. 4, with the two-panel model on the left end of
the beam and the single diagonal strut on the right. The a/d
is defined within the context of this paper as the ratio of the
distance from the center of the support to the center of the
nearest loading point a, with respect to the effective depth d.
The specimens were designed to be shear critical with two
ACI Structural Journal/March-April 2015
web shear failure modes, stirrup tie yielding for a/d of 2.50
and diagonal strut crushing for a/d of 1.85. The a/d of 2.50
was chosen to evaluate the limit of deep beam behavior and
compare with the Birrcher et al.4 studies. For this longer a/d,
the intermediate (stirrup) tie was designed to govern, thus
the capacity was determined by the quantity of stirrups in
the tie region. For specimens tested at an a/d of 1.85, the
strut-node interfaces of the diagonal strut were designed
to govern. Thus the capacity of the specimens tested at the
shorter a/d were determined by the size of the node (Fig. 3)
and the compressive strength of the concrete.
Resistance and load factors are required for STM design
but were taken as 1.0 for the purpose of this investigation as
nominal strengths were computed and compared with exper-
imental strengths. In general, no serviceability checks were
made before testing the specimens. Rather, the cracking
data obtained from loading the beams were used to validate
current serviceability provisions and/or make recommen-
dations for application to inverted-T beams.7 For step-by-
step STM design examples for inverted-T bent caps, refer to
Williams et al.9
DATABASE OF EXPERIMENTS CONDUCTED ON
INVERTED-T BEAMS
A thorough literature review was conducted prior to
establishing the experimental investigation. A total of 97
tension chord-loaded specimens reported within 13 unique
sources10-22 were ultimately compiled in a collection data-
base.7 Two sets of filters were used to develop the final
inverted-T database to meet the purposes of this project.
The first filter focused on data required to develop STMs.
Specimens not loaded to failure10,11; with complicated
support conditions, geometry, or reinforcement details12-21;
and with lack of information essential for the construction of
STMs12-18 were eliminated.
The majority of the specimens found in the literature
were unrepresentative of the bent caps in service in Texas,
10-12,16,19,21,22
requiring additional filters. A scaled comparison
of the cross sections of the specimens from the literature, the
inverted-T beams tested in the current project, and distressed
in-service bent caps within Texas is presented in Fig. 6.
The specimens are identified by their reference number.
A notable difference in size exists between the in-service
bent caps (hatched) and the majority of specimens found in
the literature (solid). A complete discussion of the filtering
process is provided in Larson et al.7
In summary, all of the 97 specimens from the 13 sources
were filtered out due to the reasons stated above, reinforcing
the need for a large number of specimens to evaluate the
behavior of inverted-T beams and investigate the applica-
bility of STM design provisions.
EXPERIMENTAL INVESTIGATION
Experimental variables
The five variables investigated are as follows: the length
of the ledge beyond the bearing of the exterior stringer, the
depth of the ledge, the amount of web reinforcement (trans-
verse and longitudinal), the number of point loads (girders)
on the ledge, and the height of the member, as shown in
ACI Structural Journal/March-April 2015
Fig. 5—Cross-sectional strut-and-tie model.
Fig. 7. Each beam was tested at an a/d of 1.85 or 2.5 to
observe the two web shear failure modes: diagonal strut
crushing or stirrup tie yielding.
Ledge length—The varying ledge lengths of inverted-T
bent caps were simplified to three types. A “cut-off” ledge
is one in which the ledge was interrupted just past the edge
of the bearing pad of the outermost girder. If the ledge ran
continuously to the support, it was considered a “long”
ledge. In a bent cap with a “short” ledge, the ledge continued
a distance approximately equal to the depth of the ledge past
the outermost girder, as shown in Fig. 7.
As previously mentioned, inverted-T bent caps are
tension chord-loaded structures in which the bridge girders
supported on the ledges induce a tension field in the web.
The size of this tension field is determined by the ledge
length and, as in the case of short and long ledges, the
tension field can engage all the “hanger” bars within the
45-degree load spread. For the cut-off ledges, the force can
only spread on one side of the bearing plate, concentrating
the load in a smaller area and increasing the tensile stresses.
Furthermore, by extending the ledge to the entire length of
a beam, the capacity of the support node can be increased.
The additional cross-sectional area in a longer ledge length
can provide confinement in the nodal region and increase
the bearing width at the support as compared to beams with
short and cut-off ledges.
Ledge depth—To fully capture the effect of ledge geom-
etry, two ledge depths were investigated as shown in Fig. 7.
Shallow ledge specimens had depth equal to one-third and
deep ledge specimens were one-half the total height of the
beam. The ledge depths were chosen to give an adequate
range of those seen in practice. As with the ledge length, the
ledge depth also has an effect on the width over which this
tension field spreads, to a lesser extent. Deeper ledges allow
the forces to spread over a wider area, decreasing the tensile
stress in the web.
Reinforcement ratio—Two amounts of orthogonal web
reinforcement were chosen with areas of steel equal to 0.3%
and 0.6% of the effective web area, as shown in Fig. 7. In most
tests, the amount of vertical and horizontal web reinforce-
ment, ρv and ρh, was equal. Two specimens were designed
with 0.3% in the horizontal direction (skin reinforcement) and
0.6% in the vertical direction (shear stirrups). The reinforce-
151
ment ratio of 0.003 (0.3%) corresponds to No. 4 (No. 13) bars
at 6.5 in. (165 mm) on center at each face of the beam. Like-
wise, a 0.006 (0.6%) ratio corresponds to No. 5 (No. 16) bars
at 5 in. (127 mm) on center. The lower limit of 0.3% is the
AASHTO LRFD1 minimum required skin reinforcement for
deep beams. The upper limit of 0.006 (0.6%) was selected to
encompass the maximum reinforcement ratio (0.57%) found
in the in-service distressed bents. The size and spacing of the
bars provides typical crack control.
Number of point loads—The beams in this investiga-
tion were loaded at either one or three points along their
length, as shown in Fig. 7. The load at each point was
equally divided and applied to the ledge on both sides of
the web using a U-shaped frame. Specimens with multiple
point loads allowed for shallower ledges by distributing the
Fig. 6—Scaled cross sections of literature specimens with corre-
sponding reference number, current specimens, and in-service
bent caps. (Note: Dimensions in inches; 1 in. = 25.4 mm.)
Fig. 7—Experimental variables.
152
total force to multiple locations and preventing local ledge
failure. The spacing of the three point loads was representa-
tive of in-service girders. They were also used to help quan-
tify the effect of multiple girders on bridge bent caps. Due
to limitations, laboratory testing is typically performed with
one loading point, but bent caps in service support multiple
girders on each side. By comparing beams tested at one and
three points, the validity of beams tested at a single load
point could be assessed.
Web depth—A review of the literature revealed a signifi-
cant difference in the size of the in-service bent caps when
compared to the specimens used to calibrate the shear provi-
sions in the current design code.7 Full-scale specimens with
web depths of 42 and 75 in. (1067 and 1905 mm) were
constructed and tested for the experimental program to fill
in this gap and validate the STM design provisions for use in
larger inverted-T beams.
Specimen description
A large testing program was required to fully evaluate all
of the experimental variables. The width of the beams was
proportioned to maximize the cross-sectional area of the
specimen, while keeping it narrow enough to ease instal-
lation and removal from the test setup. Typical dimensions
and reinforcement layouts are shown in Fig. 8 and bearing
plate sizes are given in Table 2. Flexural reinforcement
was comprised of 12 No. 11 (No. 36) bars for the 42 in.
(1067 mm) specimens and 22 for the 75 in. (1905 mm)
specimens. Hanger reinforcement was comprised of No. 6
(No. 19) bars and was detailed based on the estimated
ultimate load. Ledge reinforcement was either No. 5 or 6
(No. 16 or 19) bars, depending on the depth of the ledge and
the resulting demand. The web width was kept constant at
21 in. (533 mm) for each beam in the experimental program,
including the 75 in. (1905 mm) beams. The width of the
ledge was also the same, 10.5 in. (267 mm), on each side. All
other dimensions varied. To distinguish between the speci-
mens in Table 2 and their respective variables, the following
nomenclature was developed
Sample specimen designation: DC1-42-1.85-03
ACI Structural Journal/March-April 2015
Fig. 8—Typical reinforcement layout and dimensions. (Note: Dimensions in inches; 1 in. = 25.4 mm.)
where the first term refers to the ledge depth, either deep (D)
or shallow (S). The second term refers to the ledge length—
cut-off (C), short (S), or long (L). The third term refers to
the number of point loads, either one (1) or three (3). The
next number is the web depth in terms of inches, 42 or 75
(1067 or 1905 mm, respectively). Next is the a/d, either 1.85
or 2.5. The final term is the web reinforcement ratio, either
0.3% (03) or 0.6% (06).
The specimens were constructed using conventional mate-
rials and methods. Steel formwork was used to expedite the
fabrication process and to ensure dimensional accuracy.
Beams were tested approximately 28 days after concrete
placement. Domestic Grade 60 deformed bars satisfying the
requirements of ASTM A61523 were used for all steel rein-
forcement. Cross-sectional dimensions of the bars complied
with the nominal sizes given in ASTM A615.23 The tensile
strength of the coupons was measured in accordance with
ASTM A370.24 Material properties, including reinforcement
and concrete strength, are provided in Table 2.
Testing procedure
The specimens were tested at the University of Texas at
Austin’s Phil M. Ferguson Structural Engineering Labora-
tory. The upside-down simply supported beam test setup
used for testing is shown in Fig. 9. The load was applied
via a 5 million pound (22,200 kN) capacity, double-acting
hydraulic ram for single point load tests, and three 2 million
pound (9000 kN) capacity rams for multiple point load spec-
imens. U-shaped frames applied load evenly to the ledges of
the test specimens. At each support, six 3 in. (76 mm) diam-
eter threaded rods reacted against a 7000 lb (31 kN) transfer
girder to resist the applied load. 500 kip (2200 kN) capacity
load cells were placed between the transfer beam and the
reaction nut at each of the 12 rods to measure the applied
shear at each throughout the loading history.
Test specimens were monotonically loaded in 50 kip
(222 kN) increments to the appearance of the first diagonal
ACI Structural Journal/March-April 2015
crack, then in 100 kip (445 kN) increments to failure. After
each load increment, cracks were marked and diagonal crack
widths were measured and recorded as part of the service-
ability considerations of the experimental program.7
Each beam was designed with two test regions. Speci-
mens with a single point load were loaded a distance from
one support corresponding to the desired a/d. After a shear
failure was achieved, the load was removed and post-
tensioning clamps were installed to strengthen the first test
region. Then the hydraulic ram and U-frame were moved
and the test procedure was repeated. Specimens with three
loading points were designed such that both ends were
tested simultaneously and monitored until a shear failure
was achieved at one end of the beam. The load was then
removed, post-tensioned clamps were installed in the failed
test region, and the load was reapplied at the same location
until the opposite end of the beam failed in shear, as shown
in Fig. 9. Vtest, the maximum shear carried in the critical
section of the test region, including the self-weight of the
beam and test setup, is provided in Table 2.
COMPARISON OF STRUT-AND-TIE MODELS AND
EXPERIMENTAL RESULTS
A summary of the experimental versus calculated shear
strengths (Vtest/Vcalc) is provided in Table 2, where Vcalc is the
shear capacity calculated using the measured material proper-
ties and the Birrcher et al.4 STMs as implemented for inverted-
T beams. As shown in the table, all values of Vtest/Vcalc
are greater than 1.0, indicating that the STM provisions
as implemented for inverted-T beams are conservative for
all specimens tested. With the large number of specimens,
direct comparisons investigated each variable independently
while keeping all others constant as discussed extensively by
Larson et al.7 In these direct comparisons, the STMs showed
no bias to ledge depth, number of point loads, beam depth,
or chord loading; that is, the effects of these variables were
adequately captured as no trends were observed. The STMs
153
Table 2—Summary of beam details
Reinforcement fy, ksi
Support plate, Load
Specimen in. plate, in. No. 11 No. 6 No. 5 No. 4 fc′, ksi Vcalc, kip Vtest, kip Vtest/Vcalc
DS1-42-1.85-03 16 x 20 26 x 9 69.24 63.38 64.69 63.14 5.27 463 712 1.54
DS1-42-2.50-03 16 x 20 26 x 9 69.24 63.38 64.69 63.14 5.39 202 406 2.01
DS1-42-1.85-06 16 x 20 26 x 9 64.13 63.38 60.68 N/A 5.02 479 621 1.30
DS1-42-2.50-06 16 x 20 26 x 9 64.13 63.38 60.68 N/A 5.09 339 503 1.49
DL1-42-1.85-06 16 x 20 26 x 9 67.90 63.38 64.69 N/A 4.83 464 741 1.60
DL1-42-2.50-06 16 x 20 26 x 9 67.90 63.38 64.69 N/A 4.99 353 622 1.76
SS3-42-1.85-03 16 x 20 18 x 9 68.60 64.68 62.75 67.25 5.89 456 523 1.15
SS3-42-2.50-03 16 x 20 18 x 9 68.60 64.68 62.75 67.25 5.89 215 447 2.08
SS3-42-2.50-06 16 x 20 18 x 9 69.50 61.83 60.90 N/A 6.26 415 516 1.24
SC3-42-2.50-03 16 x 20 18 x 9 66.20 63.50 60.25 64.27 5.87 257 329 1.28
SC3-42-1.85-03 16 x 20 18 x 9 66.20 63.50 60.25 64.27 5.87 427 483 1.13
DS3-42-2.50-03 16 x 20 18 x 9 63.60 62.63 60.22 64.58 5.69 236 430 1.82
DL1-42-1.85-03 16 x 20 26 x 9 71.01 61.90 64.29 64.43 4.93 468 626 1.34
DL1-42-2.50-03 16 x 20 26 x 9 71.01 61.90 64.29 64.43 4.93 235 510 2.17
SL3-42-1.85-03 16 x 20 18 x 9 75.18 60.62 63.58 65.57 5.04 409 571 1.39
SL3-42-1.85-06 16 x 20 18 x 9 70.38 63.26 64.80 62.62 5.25 424 744 1.76
DC1-42-1.85-06 30 x 20 30 x 10 73.30 63.98 60.81 N/A 3.73 428 519 1.21
SS1-75-1.85-03 16 x 20 30 x 10 66.10 61.97 64.69 65.08 3.13 389 745 1.92
DC3-42-1.85-03 16 x 20 18 x 9 63.63 66.00 63.09 63.16 4.57 370 395 1.07
DS3-42-1.85-03 16 x 20 18 x 9 63.63 66.00 63.09 63.16 4.57 370 454 1.23
SS1-42-2.50-03 16 x 20 26 x 9 65.44 69.57 77.76 66.58 5.70 205 398 1.94
SS1-42-1.85-03 16 x 20 26 x 9 65.44 69.57 77.76 66.58 5.72 501 583 1.16
DC1-42-2.50-03 16 x 20 18 x 9 70.06 64.13 69.77 62.44 4.04 259 365 1.46
DL3-42-1.85-03 16 x 20 18 x 9 70.06 64.13 69.77 62.44 4.20 359 629 1.75
SL1-42-2.50-03 16 x 20 26 x 9 68.70 71.41 N/A 64.47 4.28 261 498 1.91
SC1-42-2.50-03 16 x 20 26 x 9 68.70 71.41 N/A 64.47 4.28 259 319 1.24
DS1-42-1.85-06/03 16 x 20 26 x 9 65.80 70.92 64.94 65.18 4.17 416 739 1.78
DS1-42-2.50-06/03 16 x 20 26 x 9 65.80 70.92 64.94 65.18 4.17 362 539 1.49
SC1-42-1.85-03 30 x 20 26 x 9 66.36 64.04 N/A 67.28 4.33 443 451 1.05
DC1-42-1.85-03 30 x 20 26 x 9 66.36 64.04 N/A 67.28 4.30 474 517 1.09
SC1-42-1.85-03 *
30 x 20 30 x 10 70.47 63.12 N/A 68.56 3.01 362 456 1.26
DC1-42-1.85-03* 30 x 20 30 x 10 70.47 63.12 N/A 68.56 3.00 362 424 1.17
SS1-75-2.50-03 16 x 20 26 x 9 65.22 63.85 63.62 63.76 5.16 357 649 1.82
*
Ledge length set equal to load plate length.
Notes: Shaded values indicate failure modes other than web shear; 1 in. = 25.4 mm; 1 ksi = 6.89 MPa; N/A is not available.

did show limited bias to ledge length and reinforcement
ratio, but produced conservative results in all cases with
reasonable safety margins.
Overall, the STM procedures offer a more transparent
approach to designing inverted-T deep beams than sectional
design, as they inherently consider all failure modes for the
ledges, web, and bearing points. The web shear failure mode
predicted by the STMs, either crushing of the strut-node
interface or yielding of the intermediate tie, was observed in
all specimens except for the five shaded in Table 2, in which
flexure (crushing of the compression stress block), shear
friction, ledge tie failure, and punching shear occurred. In
these few cases, the failure mode was related to the second
weakest element in the STM, which changed depending
of the experimental variables. Nonetheless, each of the 33
specimens carried loads well above the calculated web shear
capacity and thus, the strength estimates were conservative.
The statistical results for the strength ratios of the 33 test
specimens in the experimental program are summarized in
Table 3. As shown in the table, the design method yielded
conservative and accurate estimates of strength with a

154 ACI Structural Journal/March-April 2015

Table 3—Summary of experimental/calculated
shear capacity
Rectangular deep beams,4
Inverted T-beams, 33 tests 179 tests
Birrcher Birrcher
Vtest/Vcalc et al.4 ACI 3186 et al.4
Minimum 1.05 1.04 0.73
Maximum 2.17 2.17 4.14
Mean 1.50 1.57 1.54
Unconservative 0.0% 0.0% 0.6%
Coefficient of
0.22 0.20 0.28
variation*
*
Coefficient of variation is standard deviation divided by mean.
minimum Vtest/Vcalc value of 1.05, a maximum of 2.17, and
an average of 1.51 for the inverted-T beams.
A comparison of STM procedures by Birrcher et al.4 and
ACI 318-116 revealed similar levels of accuracy for the
inverted-T beams as shown in Fig. 10. The average Vtest/Vcalc
values in Table 3 were equal to 1.50 and 1.57, respectively,
suggesting that ACI 318-116 is slightly more conservative.
The significant difference between the two STM procedures
is the treatment of the struts and nodes, thus most of the spec-
imens designed to fail due to yielding of the stirrup tie had the
same calculated capacity. However, ACI 318-116 predicted a
diagonal strut failure occurring in the cross-sectional model
before web shear in seven of the 33 specimens. This was due
to the low efficiency factor βs for struts in tension flanges of
beams, as shown in Table 1, which resulted in an increased
conservatism for several of the specimens shown in Fig. 10.
The Birrcher et al.4 STMs do not account for this out-of-
plane tension. If only web shear failure modes were consid-
ered using ACI 318-11,6 the mean Vtest/Vcalc would decrease
to 1.50.
In comparison with the rectangular deep beams, it can be
concluded that the Birrcher et al.4 STM provisions provided
equal, if not slightly better, predictions of shear strength.
The overall mean was similar with 1.50 for the inverted
T-beams and 1.54 for the rectangular bent caps evaluated
with the Birrcher et al.4 STMs. The scatter in the results was
also decreased for the inverted-T beams when the minimum
and maximum Vtest/Vcalc values were compared. The stan-
dard deviation and coefficient of variation also decreased.
Furthermore, no unconservative prediction of strength was
noted for the inverted-T beams, while a small number of
rectangular deep beams had Vcalc values greater than Vtest.
The comparison of the rectangular deep beams as eval-
uated using ACI 318-116 STMs is also provided in Table 3
to demonstrate its effectiveness. A significant improvement
was observed for minimum, maximum, and average Vtest/Vcalc
values. It was for this reason that Birrcher et al.4 STMs were
investigated for inverted-T beams.
SUMMARY AND CONCLUSIONS
In this investigation, the behavior of inverted-T beams
was studied through a comprehensive experimental program
composed of 33 tests on 22 large-scale beams. Most of the
beams were designed with two test regions, one on each end.
ACI Structural Journal/March-April 2015
ACI 3186
0.87
9.80
1.80
1.7%
0.58
Fig. 9—Test setup: specimen at failure of second test region.
Fig. 10—Comparison of Vtest/Vcalc for Birrcher et al.4 and
ACI 318-116 STMs.
The following variables were investigated to encompass the
full behavior of inverted-T beams: the length and depth of
the ledge, the quantity of web reinforcement, the number of
point loads, member depth, and the shear span-depth ratio.
With these results, previously proposed strut-and-tie provi-
sions were assessed for their applicability to inverted-T
beams. Both ACI 318-116 STM and Birrcher et al.4 design
provisions yield accurate and reasonably conservative results
for tension chord-loaded beams. The following conclusions
summarize the views of the authors:
• Use of STM is recommended for the design of
inverted-T beams. A comparison between the ulti-
mate shear capacity obtained from the test results and
the nominal shear capacity from the STM calculations4,6
revealed conservative strength estimates for every spec-
imen. Furthermore, the Birrcher et al.4 STMs accurately
predicted the web shear failure mode for 28 of the 33 spec-
imens. For the five that did not fail in shear, the calculated
shear capacity was exceeded and the actual failure mode
was the second weakest element in the model. Within
these provisions, a minimum web reinforcement ratio is
given as 0.3% in each orthogonal direction and is also
recommended for inverted-T beam design.7
• Valid assumptions were made in implementing the
STM provisions for inverted-T beams. The geom-
etry of inverted-T beams requires the use of a three-
dimensional STM model or two equivalent and compat-
ible two-dimensional models. Recommendations were
155
 given to aid in developing these models. A 45-degree load
spread at each load point satisfactorily models the hanger
and ledge reinforcement that engaged during loading.
The Birrcher et al.4 STM provisions provided accurate predic-
tions of failure mode and capacity for the inverted-T beams
but further investigation is recommended for struts in tension
members and tension flanges in other structural members.
AUTHOR BIOS
ACI member Nancy Larson Varney is a Staff II Structural Engineer with
Simpson Gumpertz & Heger, Inc. She received her BS from Lehigh Univer-
sity, Bethlehem, PA, in 2008, and her MS and PhD from the University of
Texas at Austin, Austin, TX, in 2010 and 2013, respectively. Her research
interests include strut-and-tie modeling of reinforced concrete.
ACI member Eulalio Fernández-Gómez is a structural engineer at
Osseous Structural
Engineering, Ciudad Juárez, Mexico. He received his
BS from Universidad Autónoma de Chihuahua, Chihuahua, Mexico, in
2004, and his MS and PhD from the University of Texas at Austin in 2009
and 2012, respectively.
ACI member David B. Garber is an Assistant Professor at Florida Inter-
national University, Miami, FL. He received his BS from Johns Hopkins
University, Baltimore, MD, in 2009, and his MS and PhD from the Univer-
sity of Texas at Austin in 2011 and 2014, respectively. His research inter-
ests include plasticity in structural concrete and behavior of prestressed
concrete members.
Wassim M. Ghannoum is an Assistant Professor in the Department of
Civil, Environmental, and Architectural Engineering at the University of
Texas at Austin. He is Chair of ACI Committee 369, Seismic Repair and
Rehabilitation, and a member of ACI Subcommittee 318-R, High-Strength
Reinforcement (Structural Concrete Building Code), and Joint ACI-ASCE
Committees 441, Reinforced Concrete Columns, and 447, Finite Element
Analysis of Reinforced Concrete Structures.
Oguzhan Bayrak, FACI, is a Professor in the Department of Civil, Envi-
ronmental, and Architectural Engineering and holds the Charles Elmer
Rowe Fellowship in Engineering at the University of Texas at Austin, where
he serves as Director of the Phil M. Ferguson Structural Engineering
Laboratory. He is a member of ACI Committees 341, Earthquake-Resistant
Concrete Bridges, and S803, Faculty Network Coordinating Committee;
and Joint ACI-ASCE Committees 441, Reinforced Concrete Columns, and
445, Shear and Torsion.
ACKNOWLEDGMENTS
The authors wish to thank the Texas Department of Transportation for
providing the financial support for this investigation, and the contribu-
tions of Project Director J. Farris and TxDOT Project Advisors including
C. Holle, D. Van Landuyt, G. Yowell, M. Stroope, N. Nemec, and R. Lopez.
The contribution of the students and the staff at the Ferguson Structural
Engineering Laboratory is also greatly appreciated. Opinions, findings,
conclusions, and recommendations in this paper are those of the authors.
REFERENCES
1. AASHTO LRFD, Bridge Design Specifications, American Associa-
tion of State Highway and Transportation Officials, Washington, DC, 2012,
1960 pp.
2. Joint ACI-ASCE Committee 445, “Recent Approaches to Shear
Design of Structural Concrete (ACI 445R-99),” American Concrete Insti-
tute, Farmington Hills, MI, 1999, 56 pp.
3. Schlaich, J.; Schäfer, K.; and Jennewein, M., “Toward a Consis-
tent Design of Structural Concrete,” PCI Journal, V. 32, No. 3, 1987,
pp. 74-150. doi: 10.15554/pcij.05011987.74.150
156
4. Birrcher, D.; Tuchscherer, R.; Huizinga, M.; Bayrak, O.; Wood, S.;
and Jirsa, J., “Strength and Serviceability Design of Reinforced Concrete
Deep Beams,” Report No. 0-5253-1, Center for Transportation Research,
the University of Texas at Austin, Austin, TX, 2009, 400 pp.
5. Tuchscherer, R.; Birrcher, D.; and Bayrak, O., “Experimental Exam-
ination of ACI 318 Strut and Tie Modeling Provisions,” Symposium
Honoring James O. Jirsa’s Contributions in Structural Concrete: A Time to
Reflect, SP-296, J. A. Pincheira and S. M. Alcocer, eds., American Concrete
Institute, Farmington Hills, MI, 2014, 20 pp.
6. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-11) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2011, 503 pp.
7. Larson, N.; Fernández-Gómez, E.; Garber, D.; Bayrak, O.; and Ghan-
noum, W., “Strength and Serviceability Design of Reinforced Concrete
Inverted T-Beams,” Report No. 0-6416, Center for Transportation Research,
University of Texas at Austin, Austin, TX, 2012, 234 pp.
8. Wight, J. K., and Parra-Montesinos, G., “Use of Strut-and-Tie Model
for Deep Beam Design as per ACI 318 Code,” Concrete International,
V. 25, No. 5, May 2003, pp. 63-70.
9. Williams, C.; Deschenes, D.; and Bayrak, O., “Strut-and-Tie Model
Design Examples for Bridges,” Report No. 5-5253-01-1, Center for Trans-
portation Research, University of Texas at Austin, Austin, TX, 2012, 276
pp.
10. Furlong, R. W.; Ferguson, P. M.; and Ma, J. S., “Shear and Anchorage
Study of Reinforcement in Inverted T-Beam Bent Cap Girders,” Report No.
113-4, Center for Highway Research, University of Texas at Austin, Austin,
TX, 1971, 81 pp.
11. Cussens, A. R., and Besser, I. I., “Shear Strength of Reinforced
Concrete Wall-Beams under Combined Top and Bottom Loads,” The Struc-
tural Engineer, V. 63, No. 15, Sept. 1985, pp. 50-56.
12. Graf, O.; Brenner, E.; and Bay, H., “Versuche mit einem wandartigen
Trager aus Stahlbeton,” Deutscher Ausschuss fur Stahlbeton, V. 99, 1943,
pp. 41-54.
13. Ferguson, P. M., “Some Implications of Recent Diagonal Tension
Tests,” ACI Journal Proceedings, V. 53, No. 8, Aug. 1956, pp. 157-172.
14. Schütt, H., “Über das Tragvermögen wandartiger Stahlbetonträger,”
Beton und Stahlbetonbau, V. 10, Oct. 1956, pp. 220-224.
15. Taylor, R., “Some Shear Tests on Reinforced Concrete Beams
without Shear Reinforcement,” Magazine of Concrete Research, V. 12, No.
36, 1960, pp. 145-154. doi: 10.1680/macr.1960.12.36.145
16. Furlong, R. W., and Mirza, S. A., “Strength and Serviceability of
Inverted T-Beam Bent Caps Subject to Combined Flexure, Shear, and
Torsion,” Report No. 153-1F, Center for Highway Research, University of
Texas at Austin, Austin, TX, 1974, 156 pp.
17. Smith, K. N., and Fereig, S. M., “Effect of Loading and Supporting
Condidtions on the Shear Strength of Deep Beams,” Shear in Reinforced
Concrete, SP-42, American Concrete Institute, Farmington Hills, MI, 1974,
pp. 441-460.
18. Fereig, S. M., and Smith, K. N., “Indirect Loading on Beams with
Short Shear Spans,” ACI Journal Proceedings, V. 74, No. 5, May 1977,
pp. 220-222.
19. Leonhardt, F., and Walther, R., “Wandartige Träger,” Deutscher
Ausschuss für Stahlbeton, V. 178, 1966.
20. Galal, K., and Sekar, M., “Rehabilitation of RC Inverted-T Girders
Using Anchored CFRP Sheets,” Composites. Part B, Engineering, V. 39,
No. 4, 2008, pp. 604-617. doi: 10.1016/j.compositesb.2007.09.001
21. Zhu, R. R.-H.; Dhonde, H.; and Hsu, T. T., “Crack Control for Ledges
in Inverted ‘T’ Bent Caps,” TxDOT Project 0-1854, University of Houston,
Houston, TX, 2003, 4 pp.
22. Tan, K. H.; Kong, F. K.; and Weng, L. W., “High Strength Concrete
Deep Beams Subjected to Combined Top- and Bottom-Loading,” The
Structural Engineer, V. 75, No. 11, 1997, pp. 191-197.
23. ASTM A615/A615M-08, “Standard Specification for Deformed and
Plain Carbon-Steel Bars for Concrete Reinforcement,” ASTM Interna-
tional, West Conshohocken, PA, 2008, 5 pp.
24. ASTM A370-08a, “Standard Test Methods for Mechanical Testing of
Steel Products,” ASTM International, West Conshohocken, PA, 2008, 47 pp.
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S14
Energy-Based Hysteresis Model for Reinforced Concrete
Beam-Column Connections
by Tae-Sung Eom, Hyeon-Jong Hwang, and Hong-Gun Park
The cyclic response of reinforced concrete beam-column connec-
tions is significantly affected by the bond slip of beam flexural
bars and joint shear deformations that occur at the joint panel. In
this study, using existing test results of 69 interior and 63 exterior
connections, the variation of energy dissipation (per load cycle)
according to the bond-slip and joint shear strength was statisti-
cally investigated. The results showed that the energy dissipation
correlated with the parameters of the bar bond slip better than with
the joint shear strength. On the basis of the result, the energy dissi-
pation of beam-column connections was defined as the function of
the bond parameters. By using the energy function and the existing
backbone curve of ASCE/SEI 41-06, an energy-based hysteresis
model was developed such that the area enclosed by the cyclic
curve is the same as the predicted energy dissipation. The proposed
model was applied to existing test specimens. The predictions were
compared with the test results and showed good agreement.
Keywords: beam-column connection; cyclic loading; energy dissipation;
hysteresis model; reinforced concrete; seismic design.
INTRODUCTION
In reinforced concrete moment-resisting frames subjected
to cyclic loading, the response, including stiffness degrada-
tion, strength degradation, and energy dissipation, is signifi-
cantly affected by the behavior of beam-column joints as well
as individual members.1-4 Figure 1 shows the cyclic response
(Fig. 1(a)) and joint load-transfer mechanism (Fig. 1(b) and
(c)) of beam-column connections that are affected by bar
bond slip and diagonal shear cracking. Under cyclic loading,
X-shaped diagonal cracks increase the shear deformation in
the joint. Furthermore, due to the plastic strains of the beam
flexural bars, the bond resistance of the joint is significantly
degraded. In the case of interior connections, the bar bond
demand is increased by the compressive force, as well as
the tensile forces (bar bond demand = T1 + C2 or T2 + C1
in Fig. 1(b)). Thus, the beam-column joints are susceptible
to bar bond slip. Once the bond slip of beam bars and the
shear deformation occur in the joint, the unloading/reloading
stiffness and energy dissipation are significantly degraded,
which appears as pinching in the cyclic response of Fig. 1(a).
To mitigate bond and shear strength degradations in the
joint, current earthquake design codes specify the minimum
requirement of column depth-bar diameter ratio (hc/db):
ACI 318-115 and ACI 352R-026 require hc/db > 20 and
hc/db > 20fy/420, respectively. However, previous test results
have shown that even when the minimum requirement
was satisfied, significant bond slip and shear deformation
occurred at the beam-column joints.3,7-9 Thus, to secure
the structural performance of beam-column joints, greater
development lengths are required for the beam flexural bars
as specified in NZS 3101:200610 and Eurocode 8.11
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
To address the effects of the bond slip and joint shear
deformation, various elaborate component models have been
developed.12-17 Lowes and Altoontash,12 Elmorsi et al.,13 and
Fleury et al.14 used continuum-type elements combined with
spring elements, maintaining compatibility with beam and
column line elements. Altoontash and Deierlein15 and Mitra
and Lowes16 proposed the models that consist of a shear
panel element and rotational spring elements. Uma and
Prasad17 proposed joint shear strength-deformation relation-
ship for nonlinear dynamic analysis. These models consist of
a shear-panel element for the joint, and vertical, horizontal,
and rotational spring elements. Although addressing all
components affecting the connection behavior, these models
require great time and effort in modeling and computations,
particularly when numerical analysis of the entire moment
frame structures is required.
More conveniently for the numerical analysis of the
moment frame structures, lumped plasticity spring elements
representing the overall cyclic response of a beam-column
connection can be used. El-Metwally and Chen18 and
Alath and Kunnath19 used zero-length rotational spring
elements between the joint and beams/columns, to decouple
the inelastic response of the beams, columns, and joints.
Kunnath20 used joint spring elements at the intersection
of beams and columns. Ghobarah and Biddah21 developed
a stress-strain relationship for beam-column joints with
transverse reinforcement, and Anderson et al.22 expanded
the stress-strain relationship to joints without transverse
reinforcement. Birely et al.23 used dual hinge elements at
joint interfaces. Magliulo and Ramasco24 used a lumped
plasticity model to perform three-dimensional nonlinear
dynamic analysis.
In the lumped plasticity models, the hysteresis constitutive
model of the spring elements should be able to address the
degradations of unloading/reloading stiffness, strength, and
energy dissipation under cyclic loading, which are signifi-
cantly affected by the bond slip and shear cracking at the
joint. To describe the strength and stiffness degradations,
various hysteresis models were developed by Clough,25
Otani,26 Saatcioglu,27 Takeda et al.,28 Song and Pincheira,29
and Sivaselvan and Reinhorn.30 The majority of the existing
models are stiffness-based models in which the degrada-
tion of unloading/reloading stiffness and strength under
cyclic loading was defined on the basis of existing test
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-192.R4, doi: 10.14359/51687404, received June 9, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
157
Fig. 1—Cyclic response and joint load-transfer mechanism of beam-column connections affected by bar bond-slip and diag-
onal shear cracking.
results. However, it is very difficult to accurately define the
unloading/reloading stiffness considering the complicated
joint behavior, such as the bar bond slip and diagonal shear
cracking. More importantly, in actual design of new struc-
tures without test results, it may not be feasible to accurately
define the model parameters.
As an alternative, energy-based models for beams and
columns were studied by Eom et al.,31 Eom and Park,32
Sucuoğlu and Acun,33 Sucuoğlu and Erberik,34 and Kwak
and Kim.35 Ibarra et al.36 proposed the energy-based model
for beam-column connection. Particularly, in the energy-
based model proposed by Eom et al.31 and Eom and Park,32
the load-displacement relationship and the stiffness were
defined such that the area enclosed by the cyclic curve is the
same as the predicted energy dissipation. Thus, if the energy
dissipation is accurately predicted considering the bar bond
slip and diagonal shear cracking, the load-displacement
relationship with pinching can be reversely created from the
energy dissipation.
In the present study, the concept of the energy-based
model was applied to beam-column connections to define
the load-displacement relationship under cyclic loading. For
this purpose, first, the energy dissipation of beam-column
connections was estimated from existing test results, consid-
ering the design parameters. Then, the energy-based hyster-
esis model was defined using the predicted energy dissipa-
tion; the unloading/reloading stiffness under cyclic loading
was determined to satisfy the predicted energy dissipation.
In the proposed model, the existing backbone curve of
ASCE/SEI 41-0637 was used, and the Pinching 4 model of
OpenSees38 was modified to implement the predicted energy
dissipation. For verification, the proposed model was applied
to existing test specimens, and the results were compared
with the test results. Limitations on the application of the
proposed method were also discussed.
RESEARCH SIGNIFICANCE
The present study focused on developing a beam-column
connection model which can be conveniently used for prac-
tical design/analysis of reinforced concrete moment frames.
A design equation was developed to accurately predict the
energy dissipation capacity using bar bond-slip parameters,
which are used in current design codes. Using the constraint
condition of energy dissipation, the proposed hysteresis
model can directly and accurately define the cyclic load-
158
displacement relationship of beam-column connections.
Because the proposed model is defined as the function of the
energy dissipation capacity, it can be conveniently used for
the performance-based design/analysis of moment frames.
Evaluation of energy dissipation capacity
To quantitatively evaluate the energy dissipation capacity
of the beam-column connections, existing cyclic test results
of 69 cruciform and 63 T-shaped beam-column connections
were investigated.9,A1-A28* The material and geometric prop-
erties of the specimens are presented in Tables A1 and A2.*
The test specimens had conventional reinforcement details
at the joints, such as transverse hoops, and no lap splices of
beam flexural reinforcing bars. The concrete strengths were
fc′ = 23.9 to 88.2 MPa (3.46 to 12.8 ksi). The yield strength
and diameter of the beam bars were fy = 276 to 710  MPa
(40.0 to 103 ksi) and db = 9.5 to 35.8 mm (0.37 to 1.41 in.),
respectively. The specimens exhibited various failure modes
from the joint failure to the beam failure, depending on
the design parameters, such as the beam moment-column
moment ratio, the joint shear capacity-demand ratio, and the
bar bond parameters. The limitations of the design parame-
ters and the proposed method were given in the “Applica-
tions” section.
For parametric study of the existing test results, the energy
dissipation ratio κ specified in ACI 374.1-0539 was used. As
shown in Fig. 2, κ is defined as the ratio of the actual energy
dissipation EII per load cycle to the idealized elastic-per-
fectly plastic energy dissipation Eep: κ = EII/Eep. Generally,
the κ value increases with the deformation.40 However,
energy dissipation capacity is important when large inelastic
deformations occur, and at small deformations, the energy
dissipation does not significantly affect the shape of the load-
displacement relationship. Thus, in the present study,
according to ACI 374.1-05,39 κ was defined at the third load
cycle of a relatively large story drift ratio δ = 3.5%. However,
when the existing test conditions did not satisfy the require-
ment of ACI 374.1-05,39 the κ was defined differently: when
the number of load cycles at δ = 3.5% was less than three, κ
was calculated for the second load cycle. When a specimen
failed before δ = 3.5%, or when the strength of the second
*
The Appendix is available at www.concrete.org/publications in PDF format,
appended to the online version of the published paper. It is also available in hard copy
from ACI headquarters for a fee equal to the cost of reproduction plus handling at the
time of the request.
ACI Structural Journal/March-April 2015

Fig. 2—Definition of energy dissipation ratio (ACI 374.1-05).
or third load cycle was less than 80% of that of the first load
cycle (this case can be regarded as the failure of the spec-
imen), κ was evaluated at a moderate drift ratio of less than
δ = 3.5%. In the calculation of Eep, the initial stiffness ki
was defined from the envelope curve (refer to Fig. 2).39 The
κ values of the interior and exterior connection specimens
are presented in Tables A1 and A2, respectively. It should
be noted that for a specific beam-column connection, the
present study used a single value of the energy dissipation
ratio, evaluated at the third load cycle of 3.5% drift ratio. It is
implicitly assumed that the energy dissipation ratio does not
significantly vary according to the drift ratios. The evidence
for the assumption is given in Fig. B1.
According to the previous studies,32,41-44 the energy dissi-
pation capacity of beam-column connections is affected by
various design parameters, such as the geometry and rein-
forcement details of the beams and columns. However, as
discussed in the Introduction and Fig. 1, the energy dissi-
pation capacity of beam-column connections is degraded
primarily by the bar bond-slip and diagonal cracking at the
joint. Thus, the joint shear strength and the bond resistance
of the beam flexural bars were considered as the primary
design parameters for the evaluation of energy dissipation.
In ACI 318-11,5 ACI 352R-02,6 and NZS 3101:2006,10 the
requirement for the bond resistance of the beam flexural bars
is defined as follows (refer to Fig. 1(b) and (c)).
For interior connections
 parameters and the energy dissipation ratios κ for the inte-
 rior and exterior connections, respectively. In the figures, the
20 for ACI 318-11
hc  fy
≥ 20 ≥ 20 for ACI 352R-02 (1)
db  420
 1.25 f y
 for NZS 3101:2006
 3.3α f α d γ f c′
For exterior connections
 fy
 for ACI 318-11
 5.4 f c′
 On the other hand, in Fig. 3(c) and 4(c), the parameters,
ldh  αf y
≥ for ACI 352R-02 (2)
db  6.2 f c′
 0.24α α f
 1 2 y
for NZS 3101:2006
 f c′
ACI Structural Journal/March-April 2015
where hc is column depth (or joint depth); db is the greatest
bar diameter of the beam flexural bars; and ldh is develop-
ment length of the beam flexural bars anchored inside the
joint in exterior connections. In Eq. (1), αf and αd are coef-
ficients addressing the direction of the beam flexural bars
and the ductility of beam plastic hinges; and γ is the coef-
ficient addressing the story drift ratio demand δ of the joint
(=1.53 – 0.29δ, δ in %). In Eq. (2), α is the overstrength factor
of steel reinforcing bars addressing the strain-hardening
behavior (=1.25); and α1 and α2 are coefficients addressing
the details of hook anchorage and the joint confinement by
transverse hoops.
The requirement for the joint shear strength is defined as
follows (refer to Fig. 1(b) and (c)).
V jn ≥ φV ju (3)
where Vjn = γj√fc′Aj (nominal shear strength of the joint);
Vju = T1 + C2 – Vc for interior connections; Vju = C1 – Vc for
exterior connections (shear demand of the joint); ϕ is strength
reduction factor for shear; γj is coefficients addressing the
confinement provided by the beams framing into the joint;
Aj is effective joint shear area; T1 is resultant tension force
at the beam critical section in the negative moment; C2 is
resultant compression force at the beam critical section in
the positive moment; and Vc is shear demand of the column.
For parametric study related to bond slip, from Eq. (1),
hc/db, (hc/db)(√fc′/fy), and (hc/db)(γ√fc′/fy) were chosen as
the bond parameters of the beam flexural bars for interior
connections, and from Eq. (2), (ldh/db), (ldh/db)(√fc′/fy), and
(ldh/db)(√fc′/α1α2fy) were chosen for exterior connections. In
the majority of the existing specimens investigated in this
study, the number of the beam flexural bars placed at the top
was greater than at the bottom. In this case, the bond slip of
the bottom bars is greater than that of the top bars because
the inelastic deformation of the bottom bars is greater than
that of the top bars due to the force-equilibrium in the cross
section. Therefore, the bond parameters of the test speci-
mens summarized in Tables A1 and A2 were defined using
the yield strength and maximum diameter of the bottom bars.
Figures 3 and 4 show the relationships between the bond
vertical and horizontal axes indicate the κ values and bond
parameters, respectively. The trend lines and correlation
coefficients R2 are presented in Fig. 3 and 4 (R2 close to 1.0
indicates a strong correlation).
For both interior and exterior connections, the κ values
correlated better with the bond parameters (hc/db)(√fc′/fy) and
(ldh/db)(√fc′/fy). In Fig. 3(b) and 4(b), the correlation coeffi-
cients R2 = 0.926 for the interior connections and 0.880 for
the exterior connections were relatively high, which means
good correlations between the energy dissipation capacity
and the bond parameters (hc/db)(√fc′/fy) and (ldh/db)(√fc′/fy).
which include the effects of the story drift ratio (that is, γ)
and the details of hook anchorage and transverse reinforce-
ment (that is, α1α2), did not show good correlations with the
κ values.
159
Fig. 3—Variation of energy dissipation ratio according to bond parameters: interior connections.9,A1-A19

Fig. 4—Variation of energy dissipation ratio according to bond parameters: exterior connections.A20-A28

Fig. 5—Variation of energy dissipation ratio according to joint shear parameters.

For parametric study for the joint shear strength, from ACI 352R-02,6 the cyclic responses were dominated by the
Eq. (3), Vjn/Vju was chosen as the parameter. The shear joint deformations rather than by the beams or columns.
parameters of the interior and exterior connection specimens On the basis of the results shown in Fig. 3(b) and 4(b),
are presented in Tables A1 and A2. Figures 5(a) and (b) show the energy dissipation ratios κ of the interior and exterior
the relationships between the energy dissipation ratios κ and connections were defined as the linear functions of the bond
the joint shear parameter Vjn/Vju for the interior and exterior parameters (hc/db)(√fc′/fy) and (ldh/db)(√fc′/fy), respectively,
connections, respectively. As presented in Tables A1 and using the method of least squares.
A2, in the majority of the connection specimens, the sum of For interior beam-column connections
column flexural capacities (that is, ΣMc) was greater than the
sum of beam flexural capacities (that is, ΣMb), which indi- hc f c′  h f c′ 
cates that the load-carrying capacity of the specimens was κ = 0.80 + 0.053  0.16 ≤ c ≤ 0.60 (4a)
db f y  db f y 
determined by the flexural capacities of the beams. Thus,
Vjn and Vju were calculated by using the beam plastic hinge
mechanism (refer to Tables A1 and A2). As shown in Fig. 5, For exterior beam-column connections
the R2 values were much less than those of Fig. 3(b) and
4(b), which indicates the correlation between the joint shear  
ldh f c′ l f c′
parameter and the energy dissipation capacity was signifi- κ = 1.56 − 0.058  0.13 ≤ dh ≤ 0.35 (4b)
cantly weaker than that of the bond parameters. db f y  db f y 
The effect of beam reinforcements on the energy dissi-
pation ratio κ was also investigated (refer to Appendix C). In Eq. (4a) and (4b), the upper and lower limits on the bond
The results showed that the correlation coefficients in the parameters (hc/db)(√fc′/fy) and (ldh/db)(√fc′/fy) were speci-
statistics were not improved but even worse. This is because, fied as the minimum and maximum values presented in
even for the connection specimens designed in accordance Tables A1 and A2, which represents the range of the design
with the bar bond requirements specified in ACI 318-115 and parameter of the existing tests.

160 ACI Structural Journal/March-April 2015

Fig. 6—Energy-based hysteresis model for beam-column connections.
Energy-based hysteresis model
Figure 6(a) shows the proposed lumped plasticity model
for the interior and exterior beam-column connections. The
concept of the lumped plasticity model was proposed by
Birely at al.45 The proposed model comprises the elastic
beam-column elements, rigid elements in the joint region,
and rotational spring elements at the joint interface. If plastic
hinges are expected to develop in columns, additional rigid
and rotational spring elements can be used in the columns.
The elastic beam-column elements simulate elastic flexural
responses of beams and columns. The rigid elements are
used to address the offset effect corresponding to the joint
depth or height. The rotational spring elements at the joint
interface are used to simulate the combined responses of the
beam plastic hinge and joint. The advantage of the proposed
model over the existing lumped plasticity approaches18-24 is
the simplicity: the rotational spring elements describe the
combined cyclic responses of the beam-column connections
rather than the separate responses of the beams and the joint.
Thus, the number of the spring elements can be reduced.
Although simple spring models are used, by using the
constraint condition of energy dissipation (Eq. (4)), the effects
of the bar bond-slip and shear deformation can be directly
addressed in the proposed load-displacement relationship.
The proposed load-displacement relationship of the rota-
tional spring element consists of an envelope curve and
cyclic curves (refer to Fig. 6(b) and (c)). The envelope curve
was developed modifying the backbone curve specified in
ASCE/SEI 41-06.37 In the backbone curve, the parameters,
except peak strength, yield strength, and initial stiffness,
need to be determined empirically on the basis of available
test results. Thus, in the present study, the parameters were
determined on the basis of the existing test results reported
in this paper. On the other hand, the cyclic curve was devel-
oped by modifying the Pinching 4 model of OpenSees.38 To
determine the parameters for the cyclic curve, the proposed
model used a very important constraint condition: the energy
dissipation capacity, which indicates the area of the cyclic
curve. Thus, although the specific unloading/reloading stiff-
ness is not exactly the same as each test result, the shape
ACI Structural Journal/March-April 2015
and area of the cyclic load-displacement relationship can be
predicted without big mistakes.
Figure 6(b) illustrates the envelope curve (that is, a
moment-plastic deformation angle relationship by mono-
tonic loading) defined by modifying the backbone curve of
ASCE/SEI 41-06.37 When the responses of the positive and
negative loadings are different (that is, when the number of
bars are different at the top and bottom of the beam cross
section), the positive and negative envelope curves can be
defined differently. In Fig. 6(b), EY, EU, and ER indicate
characteristic points corresponding to yielding, ultimate,
and residual states of connection, respectively. To define the
yield moment My at EY, the nominal yield moment Mny at
the critical section of beams (that is, at the joint interface) is
used as follows (refer to Fig. 6(b)).
M y ≈ M ny (5a)
The ultimate moment Mu at EU and the residual moment
Mr at ER are defined as functions of the nominal flexural
capacity Mn.
M u = βu M n and M r = β r M n (5b)
In Eq. (5b), βu can be theoretically determined from
nonlinear section analysis addressing reinforcement details,
load conditions, and actual material strengths. When detailed
analysis is not performed, βu can be approximated as 1.25,
considering the tensile stress 1.25fy of reinforcing bars
increased by the cyclic strain-hardening behavior.5 For the
residual moment, βr = 0.2 was defined according to ASCE/
SEI 41-06.37
The rotational spring element represents the shear defor-
mation of the joint and the rotation of the beam plastic
hinges. Thus, the yield deformation angle θy at EY (the yield
point) includes the elastic shear deformation of the joint
and the yield rotation of the beam plastic hinge. According
to Shin and LaFave,46 the elastic shear deformation of
beam-column joints varies within the ranges of γy = 0.002
to 0.01 rad, depending on the design variables. The elastic
161
shear deformation of the joint can be predicted by using
existing elaborate nonlinear analysis methods, such as the
compression field theory and the softened truss model.46-48
Alternatively, the shear deformation of the joint at the yield
point can be approximately estimated by using an empirical
method proposed by Kim and LaFave49 and LaFave and
Kim.50 According to Priestley40 and Paulay and Priestley,44
the yield rotation of a beam plastic hinge can be approxi-
mated as θby = ϕylp ≈ (1.7εy/hb)0.5hb = 0.85εy (ϕy is the yield
curvature of the beam cross section, lp is the length of the
beam plastic hinge, εy is the yield strain of beam flexural
bars, and hb is the overall depth of the beam cross section).
Thus, the yield deformation angle θy at EY of the rotational
spring element can be defined as the sum of the elastic shear
deformation (γy) of the joint and the yield rotation (θby) of the
beam plastic hinge: θy = γy + θby = γy + 0.85εy.
The deformation angles θu at EU and θr at ER were deter-
mined according to ASCE/SEI 41-06.37 However, ASCE/
SEI 41-0637 separately defines the plastic rotation angle at
the beam plastic hinges and the plastic shear angle at the
joint, addressing the reinforcement detail and loading condi-
tion. In the proposed method, on the other hand, the rota-
tional spring elements (Fig. 6(a)) represent the overall plastic
deformation angle of a beam-column connection. Therefore,
θu at EU for the rotational spring elements can be defined as
the sum of the θy, θbu, and θju
θu = θ y + θbu + θ ju (6)
where θbu is the maximum plastic rotation angle of the beam;
and θju is the maximum shear angle of the joint. The values
of θbu and θju are specified in ASCE/SEI 41-06.37 By the defi-
nition, θju excludes the yield rotation θy.
For more accurate analysis, the maximum plastic deforma-
tion angle θu at EU can be determined from other advanced
methods.51 The plastic deformation angle θr at ER, which
defines the post-peak descending slope of the envelope
curve, was determined from the existing test results. For
simplicity, θr was approximated as θr = 2.0θu.
Figure 6(c) shows the cyclic curve of the moment-plastic
deformation angle relationship, connecting six characteristic
points CP, C1, C2, CN, C3, and C4, which are defined such
that the area enclosed by the cyclic curve is the same as the
predicted energy dissipation.31-36 CP (θmp, Mmp) and CN (θmm,
Mmm) denote the positive and negative peak points, respec-
tively, where the unloading and reloading behaviors begin.
C2 (θc2, Mc2) and C4 (θc4, Mc4) denote the points where the
unloading stiffness significantly decreases, causing pinching
in the cyclic response. C1 (θc1, Mc1) and C3 (θc3, Mc3) denote
the points where the reloading stiffness is recovered. To
ease the use in practice, the cyclic curve including stiffness
and strength degradations was proposed by modifying the
Pinching 4 model of OpenSees.38
The unloading behavior continues from points CP to C2
and from points CN to C4, where the moments are zero
(Mc2 = Mc4 = 0; refer to Fig. 6(c)). The unloading stiffness kup
and kun are defined as
kup = (1 − γ k ) k yp and kun = (1 − γ k ) k yn (7)
162
where kyp and kyn are secant stiffnesses connecting Point O
and the positive and negative yield points EY, respectively
(Fig. 6(b)), and γk is the coefficient representing the degra-
dation of the unloading stiffness under cyclic loading. In
the present study, the degradation of the unloading stiff-
ness is defined as the function of the load cycle number, i
(= 0, 1, 2, …), accumulated during the entire loading history.38
γ k = 0.05 ⋅ i ≤ 0.8 (8)
The original definition of γk in the reference (OpenSees
manual) is K1 ∙ δmaxK2 + K3 ∙ iK4 ≤ γk,max, and the coefficients
are determined on the basis of test results, user’s experience,
or engineering judgment. In the present study, the coeffi-
cients K1, K2, K3, K4, and γk,max were defined as 0, 0, 0.05,
1, and 0.8, respectively, from the comparison between the
predicted hysteresis curves and the existing test results.
As shown in Fig. 6(c), the hysteretic energy dissipation per
load cycle EII of the connection is affected by the moments
and deformation angles at Points C1 and C3. Therefore,
the moments and plastic deformation angles (θc1, Mc1) at
Point C1 and (θc3, Mc3) at Point C3 are defined as the func-
tions of the predicted energy dissipation ratio κ of Eq. (4a)
and (4b), as follows.
θc1 = λ θ θ mp and θc 3 = λ θ θ mn (9)
M c1 = λ M M mp and M c 3 = λ M M mn (10)
where (θmp, Mmp) and (θmn, Mmn) are plastic deformation
angles and moments at the peak points CP and CN, respec-
tively, where the unloading behavior starts; and λθ and λM
are coefficients defined as the functions of the energy dissi-
pation ratio κ, as follows.
λ θ = −0.95κ + 0.5 (11)
λ M = 1.5κ − 0.12 (12)
The coefficients λθ and λM are defined such that the energy
dissipation per load cycle EII enclosed by the cyclic curve
(C1-CP-C2-C3-CN-C4) is the same as κ Eep, where Eep is the
energy dissipation by the elastic-perfectly plastic behavior
between CP and CN (refer to Fig. 6(c)). Curve fitting
between the cyclic curves of Fig. 6(c) and the test results was
performed for various κ values and drift levels. On the basis
of the results, λθ and λM were defined as the linear functions
of κ, in Eq. (11) and (12). The validity of Eq. (11) and (12)
was verified in Appendix B. In the existing test results in
Tables A1 and A2, as the κ value ranges 0.15 to 0.54, λθ and
λM vary from 0.36 to –0.01 and from 0.11 to 0.69, respec-
tively. The cyclic curve defined in Eq. (7) through (12) is
applicable to both interior and exterior connections.
Strength degradation can occur during repeated load
cycles between the peak points CP and CN, which is
defined as the cyclic strength degradation in FEMA 440.52
The cyclic strength degradation (that is, a delay in strength
ACI Structural Journal/March-April 2015
Fig. 7—Predicted cyclic responses versus test results for interior connections.9,A3,A4,A11,A19
development) is caused by the bond deterioration of beam
flexural bars, and the concrete crushing at the joint interface.
Modifying the Pinching 4 model of OpenSees,38 the cyclic
strength degradation was addressed as follows. As shown
in Fig. 6(d), the cyclic curves after the first load cycle are
defined with the modified peak points CP′ and CN′ corre-
sponding to θmp′ and θmn′, respectively. Because the plastic
deformation angles at CP′ and CN′ are greater than those at
CP and CN, the moments corresponding to θmp and θmn of the
second and third load cycles are less than those of the first
load cycle. Herein, the plastic rotation angles θmp′ and θmn′ at
the modified peak points CP′ and CN′ are defined as follows
(refer to Fig. 6(d)).
′ = (1 + γ s ) θ mp and θ mn
θ mp ′ = (1 + γ s ) θ mn (13)
The coefficient γs is defined by the number of load cycles,
i(= 0, 1, 2, …) accumulated during the entire loading history.38
γ s = 0.1⋅ i 0.2 ≤ 0.5 (14)
The original definition of γs in the reference (OpenSees
manual) is S1 ∙ δmaxS2 + S3 ∙ iS44 ≤ γs,max, and the coefficients
are determined on the basis of test results, user’s experience,
or engineering judgment. In the present study, the coeffi-
cients S1, S2, S3, S4, and γs,max were defined as 0, 0, 0.1,
0.2, and 0.5, respectively, from the comparison between the
predictions and the existing test results.
The advantages of the proposed model can be summarized
as follows.
1. For simplicity, beam-column connections were
modeled with rotational springs of limited numbers. Thus,
the proposed model can be conveniently used for the numer-
ical analysis of overall moment frames.
2. The proposed model is able to accurately predict
the energy dissipation capacity. In the present study, the
energy dissipation of beam-column connection was accu-
rately defined by the bar bond-slip parameters, as shown in
Fig. 3(b) and 4(b).
3. The proposed model defines the cyclic behavior of beam-
column connections, using a constraint condition of energy
ACI Structural Journal/March-April 2015
dissipation capacity. Thus, the cyclic load-displacement
relationship can be directly and accurately predicted without
big mistakes.
4. The proposed model defines the cyclic behavior of beam-
column connections, as the function of the energy dissipa-
tion capacity. Thus, the proposed model can be conveniently
used for the performance-based design/analysis of structures;
in actual performance design, if a target energy dissipation
ratio κ is determined considering the design parameters, the
beam-column connection model for numerical analysis can
be directly determined according the target value.
Applications
The proposed lumped plasticity model (Fig. 6(a)), using
the energy-based hysteresis moment-rotation relationship,
was applied to existing interior and exterior connection spec-
imens.9,A3,A4,A11,A19,A22-A24,A26,A28 In all specimens, the column
flexural capacities (that is, ΣMc) were greater than the beam
flexural capacities (that is, ΣMb). Thus, the rotational spring
elements were used only for the beams. The cross sections
of beams at the joint interface are shown in Fig. 7 and 8.
The dimensions and modeling parameters of the specimens
are presented in Table 1. To highlight the advantage of the
proposed model, the specimens that exhibited various shapes
in the cyclic responses from significant pinching (that is, low
energy dissipation ratio) to no-pinching (that is, high energy
dissipation ratio) were used in these examples.
As shown in Fig. 6(a), the specimens were modeled
with the elastic beam-column elements, rotational spring
elements, and rigid elements. In the elastic beam-column
elements, 1.0EcIg (Ec is modulus of concrete [=4700√fc′]
and Ig is second-order moment of inertia of the gross cross
section) was used for the flexural rigidity of the beams.
Because the columns of the specimens were not subjected to
axial compression load, the flexural rigidity of the columns
was defined as 0.5EcIg according to ASCE/SEI 41-06,37 and
Paulay and Priestley.44 To address the offset effects by the
joint depths, infinite flexural rigidity was assigned to the
rigid elements.
For the rotational spring elements located at the joint
interface, the moments at the characteristic points of the
163
Fig. 8—Predicted cyclic responses versus test results for exterior connections.A22-A24,A26,A28
Table 1—Modeling parameters for existing test specimens
Bond resistance
parameter Modeling parameter
†
ldj f c′
db f y
Specimens *
k‡ lq lM qy, rad qbu, rad qju, rad qu, rad
Hwang S3A19 0.162 0.182 (0.174) 0.327 0.153 0.0079 0.025 0.015 0.0479
Durrani S3A4 0.292 0.287 (0.303) 0.227 0.311 0.0052 0.025 0.015 0.0452
Brooke 4B9 0.324 0.312 (0.333) 0.204 0.348 0.0063 0.025 0.015 0.0463
Interior
Xian U5A11 0.356 0.338 (0.358) 0.179 0.387 0.0048 0.025 0.015 0.0448
Xian U3A11 0.415 0.385 (0.391) 0.134 0.458 0.0054 0.025 0.015 0.0454
Dai U1A3 0.588 0.524 (0.500) 0.002 0.666 0.0034 0.025 0.015 0.0434
Tsonos S2A26 0.141 0.161 (0.184) 0.347 0.122 0.0040 0.025 0.010 0.0390
Shiohara L06A28 0.170 0.207 (0.212) 0.303 0.191 0.0035 0.025 0.010 0.0385
Ehsani 4A22 0.215 0.277 (0.281) 0.237 0.296 0.0041 0.025 0.010 0.0391
Exterior
Kaku 2A23 0.251 0.333 (0.342) 0.184 0.380 0.0040 0.025 0.010 0.0390
Ehsani 2A22 0.291 0.397 (0.391) 0.123 0.476 0.0038 0.025 0.010 0.0388
Chutarat SAA24 0.341 0.474 (0.510) 0.050 0.591 0.0035 0.025 0.010 0.0385
*
Material and geometric properties are presented in Tables A1 and A2.
†
ldj = hc for interior connections and ldh for exterior connections.
‡
Values are predictions estimated from Eq. (4a) and (4b) and values inside brackets are test results.

envelope curves, My, Mu, and Mr, were determined from
section analysis of the beam cross sections: the nominal
flexural strength Mn was used for My; by using βu = 1.25 and
βr = 0.2, Mu and Mr were determined as 1.25My and 0.2My,
respectively. As mentioned, the yield deformation angles θy
at EY of the specimens were defined as the sum of the elastic
shear deformation γy of the joint and the yield rotation θby
(=0.85εy) of the beam plastic hinge: θy = γy + 0.85εy. The
γy values (0.0014 to 0.0049 rad) were determined by using
the empirical equation proposed by Kim and LaFave49,50
instead of using elaborate nonlinear analysis methods such
as the compression field theory and the softened truss model.
Table 1 presents the θy values of the test specimens. The θy
ranged from 0.0034 to 0.0079 rad, depending on the design
variables such as concrete strength, beam reinforcement
yield strength, beam reinforcement ratio, and joint hoop
ratio. The maximum plastic deformation angles θu of the
connection specimens were determined by using Eq. (6):
θu = θbu + θju + θy. Table 1 presents the maximum beam plastic
rotation angles θbu, and the maximum joint shear angles θju
of the connection specimens. θbu and θju corresponding to the
reinforcement details and load conditions were determined
from ASCE/SEI 41-06.37
The cyclic curves of the rotational spring elements were
determined from the properties of Eq. (7) through (12),
which were defined as the functions of the energy dissipa-
tion ratio κ in Eq. (4a) and (4b). The κ values of the speci-
mens are presented in Table 1. Detailed calculations for the
envelope curves and the cyclic curves are presented in the
Appendix D. In Table 1, for instance, the predicted κ of the
specimen Ehsani 2 was 0.397, which was very close to the
test result 0.391. For the specimen Chutarat SA, the predicted
κ was 0.474, which was very close to the test result 0.510.
This result indicates that the proposed model predicted the
test results with reasonable precision.
Figures 7 and 8 compare the predicted cyclic responses
of the interior and exterior connection specimens with the
test results. As shown in the figures, the proposed lumped
plasticity method using the proposed energy-based hyster-
esis model predicted the cyclic responses of the specimens
with reasonable precision, including the energy dissipation,
pinching, and strength and stiffness degradations during

164 ACI Structural Journal/March-April 2015

cyclic loading. In particular, the energy-based hysteresis
model was applicable to various cyclic curves, from the
significantly-pinched cyclic curves with lower κ values
to the less-pinched cyclic curves with higher κ values. In
Appendix B, the energy dissipation ratios from the proposed
cyclic curves and the test results were quantitatively
compared for the specimens Durrani S3,A4 Xian U5,A11 and
Dai U1.A3 However, in the predictions shown in Fig. 8(d)
through (f), strength degradation occurred earlier than the
test results. The difference between the prediction and the test
result is attributed to the underestimation of the maximum
deformation by ASCE/SEI 41-06.37 As mentioned, the
present study focused on the energy dissipation ratio, while
the maximum deformation was predicted following ASCE/
SEI 41-06.37 In Fig. 8(b) and (e), the initial stiffness of Shio-
hara L06 and Ehsani 2 was significantly overestimated. This
is because Kim and LaFave’s method49,50 underestimated the
yield deformation angles θy. In Fig. 7(a), the deformation
of Hwang S3 under unloading was underestimated because
ASCE/SEI 41-0637overestimated the maximum deformation.
The application of the proposed model is limited to the
joints with transverse hoops and beam reinforcing bars
without lap splices at the joint. Further, it is assumed that
beams show stable flexural behavior without deficiency in
shear strength, and thus the overall cyclic response of the
connections is affected by the bond-slip damage of the joint
region and the flexural damage of the beam end, rather than
the shear damage of the beam. In addition, further research
is required for the connections with columns subjected to
moderate or high compressive load, because the specimens
analyzed in the present study were mostly free from axial
compressive load. The ranges of the design parameters are
limited to those of existing test specimens that were used
to develop the proposed model: the column moment-beam
moment ratio ΣMnc/ΣMnb ≥ 1.0, the joint shear capacity-
demand ratios 0.5 ≤ Vn/Vu ≤ 4.25, the column depth-beam
bar diameter ratios 14.5 ≤ hc/db ≤ 37.5 for interior connec-
tion, and the embedment length-beam bar diameter ratios
9.5 ≤ ldh/db ≤ 28.6 for exterior connection. Regarding other
design parameters, including the reinforcement details of
joints, material and geometric properties of beams, and story
drift ratio, further research is required.
SUMMARY AND CONCLUSIONS
In the present study, a simplified method to model the
beam-column connections subjected to cyclic loading was
investigated. By analyzing the cyclic test results of 69 inte-
rior and 63 exterior beam-column connections, the relation-
ships between the bond resistance of beam flexural bars at
the joints and the energy dissipation capacity were quanti-
fied. On the basis of the results, an energy-based hysteresis
model was proposed by modifying the backbone curves of
ASCE/SEI 41-0637 and the Pinching 4 model of OpenSees.38
For verification, the cyclic responses of the existing connec-
tions predicted by the proposed method were compared with
the test results. The major conclusions of the present study
are summarized as follows.
1. The energy dissipation capacity (or the energy dissi-
pation per load cycle) of interior and exterior connections
ACI Structural Journal/March-April 2015
correlated better with the bond resistance of beam flexural
bars at the joints, than the joint shear resistance. Thus, the
energy dissipation ratios κ of interior and exterior connec-
tions were defined as the linear functions of the bond param-
eters of beam flexural bars, (hc/db)(√fc′/fy) and (ldh/db)(√fc′/fy),
respectively.
2. To simulate the cyclic responses of interior and exte-
rior beam-column connections, an energy-based hysteresis
model was developed such that the area enclosed by the
overall cyclic curve of the connection was the same as the
energy dissipation predicted using the bond parameters. The
unloading/reloading stiffness, pinching, and strength and
stiffness degradations under cyclic loading were defined as
the functions of the energy dissipation ratio κ and the loading
history. The predictions of the proposed method correlated
well with the test results of existing interior and exte-
rior connection specimens. By using the constraint condi-
tion of energy dissipation, the shape and area of the cyclic
load-displacement relationships were predicted without big
mistakes, which is the advantage of the proposed method.
AUTHOR BIOS
Tae-Sung Eom is an Assistant Professor in the Department of Architec-
tural Engineering at Dankook University, Gyeonggi-do, South Korea. He
received his BE, MS, and PhD in architectural engineering from Seoul
National University, Seoul, South Korea. His research interests include the
analysis and design of reinforced concrete structures.
Hyeon-Jong Hwang is an Assistant Professor in the College of Civil Engi-
neering at Hunan University, Hunan, China. He received his BE, MS, and
PhD in architectural engineering from Seoul National University.
ACI member Hong-Gun Park is a Professor in the Department of Architec-
ture & Architectural Engineering at Seoul National University. He received
his BE and MS in architectural engineering from Seoul National University,
and his PhD in civil engineering from the University of Texas at Austin,
Austin, TX. His research interests include the analysis and design of rein-
forced concrete structures.
ACKNOWLEDGMENTS
This research was financially supported by the Basic Science
Research Program through the National Research Foundation of Korea
(NRF), funded by the Ministry of Education, Science, and Technology
(2012R1A1A1003282), and the Integrated Research Institute of Construc-
tion and Environmental Engineering Seoul National University Research
Program, funded by the Ministry of Education & Human Resources Devel-
opment. The authors are grateful to the authorities for their support.
REFERENCES
1. Meinheit, D. F., and Jirsa, J. O., “Shear Strength of Reinforced
Concrete Beam-Column Joints,” Report No. 77-1, Department of Civil
Engineering, Structures Research Laboratory, University of Texas at
Austin, Austin, TX, 1977.
2. Ehsani, M. R., “Behavior of Exterior Reinforced Concrete Beam to
Column Connections Subjected to Earthquake Type Loading,” Report No.
UMEE 82R5, Department of Civil Engineering, University of Michigan,
Ann Arbor, MI, 1982, 275 pp.
3. Leon, R. T., “Interior Joints with Variable Anchorage Lengths,”
Journal of Structural Engineering, ASCE, V. 115, No. 9, 1989, pp. 2261-
2275. doi: 10.1061/(ASCE)0733-9445(1989)115:9(2261)
4. Soleimani, D.; Popov, E. P.; and Bertero, V. V., “Hysteretic Behavior
of Reinforced Concrete Beam-Column Subassemblages,” ACI Journal
Proceedings, V. 76, No. 11, Nov. 1979, pp. 1179-1196.
5. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-11) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2011, 503 pp.
6. Joint ACI-ASCE Committee 352, “Recommendations for Design of Beam-
Column Connections in Monolithic Reinforced Concrete Structures (ACI 352R-
02),” American Concrete Institute, Farmington Hills, MI, 2002, 38 pp.
165
 7. Kitayama, K.; Otani, S.; and Aoyama, H., “Earthquake Resistant
Design Criteria For Reinforced Concrete Interior Beam-Column Joints,”
Pacific Conference on Earthquake Engineering, New Zealand, V. 1, 1987,
pp. 315-326.
8. Hakuto, S.; Park, R.; and Tanaka, H., “Effect of Deterioration of Bond
of Beam Bars Passing through Interior Beam-Column Joints of Flexural
Strength and Ductility,” ACI Structural Journal, V. 96, No. 5, Sept.-Oct.
1999, pp. 858-864.
9. Brooke, N. J.; Megget, L. M.; and Ingham, J. M., “Bond Performance
of Interior Beam-Column Joints with High-Strength Reinforcement,” ACI
Structural Journal, V. 103, No. 4, July-Aug. 2006, pp. 596-603.
10. NZS 3101:2006, “The Design of Concrete Structures,” Standards
New Zealand, Wellington, New Zealand, 2006, 698 pp.
11. BS EN 1998-1:2004, “Eurocode 8: Design of Structures for Earth-
quake Resistance,” British Standards Institution, London, UK, 2004.
12. Lowes, L. N., and Altoontash, A., “Modeling Reinforced-Concrete
Beam-Column Joints Subjected to Cyclic Loading,” Journal of Structural
Engineering, ASCE, V. 129, No. 12, 2003, pp. 1686-1697. doi: 10.1061/
(ASCE)0733-9445(2003)129:12(1686)
13. Elmorsi, M.; Kianoush, M. R.; and Tso, W. K., “Modeling Bond-
Slip Deformation in Reinforced Concrete Beam-Column Joints,” Canadian
Journal of Civil Engineering, V. 27, No. 3, 2000, pp. 490-505. doi: 10.1139/
l99-085
14. Fleury, F.; Reynouard, J. M.; and Merabet, O., “Multi-Component
Model of Reinforced Concrete Joints for Cyclic Loading,” Journal of Engi-
neering Mechanics, ASCE, V. 126, No. 8, 2000, pp. 804-811. doi: 10.1061/
(ASCE)0733-9399(2000)126:8(804)
15. Altoontash, A., and Deierlein, G. D., “A Versatile Model for Beam-
Column Joints,” ASCE Structures Congress, Seattle, WA, 2003.
16. Mitra, N., and Lowes, L. N., “Evaluation, Calibration, and Verifi-
cation of A Reinforced Concrete Beam-Column Joint Model,” Journal
of Structural Engineering, ASCE, V. 133, No. 1, 2007, pp. 105-120. doi:
10.1061/(ASCE)0733-9445(2007)133:1(105)
17. Uma, S. R., and Prasad, A. M., “Seismic Evaluation of R/C
Moment Resisting Frame Structures Considering Joint Flexibility,” 13th
World Conference on Earthquake Engineering Conference Proceedings,
No. 2799, Vancouver, BC, Canada, 2004.
18. El-Metwally, S. E., and Chen, W. F., “Moment-Rotation Modeling of
Reinforced Concrete Beam-Column Connections,” ACI Structural Journal,
V. 85, No. 4, July-Aug. 1988, pp. 384-394.
19. Alath, S., and Kunnath, S. K., “Modeling Inelastic Shear Deforma-
tion in RC Beam-Column Joints,” Proceedings of the 10th Conference on
Engineering Mechanics, University of Colorado at Boulder, Boulder, CO,
1995, pp. 822-825.
20. Kunnath, S. K., “Macromodel-Based Nonlinear Analysis of
Reinforced Concrete Structures,” Structural Engineering Worldwide,
No. T101-5, Elsevier Science, Ltd., Oxford, England, 1998.
21. Ghobarah, A., and Biddah, A., “Dynamic Analysis of Rein-
forced Concrete Frames Including Joint Shear Deformation,” Engi-
neering Structures, V. 21, No. 11, 1999, pp. 971-987. doi: 10.1016/
S0141-0296(98)00052-2
22. Anderson, M.; Lehman, D.; and Stanton, J., “A Cyclic Shear
Stress-Strain Model for Joints Without Transverse Reinforcement,”
Engineering Structures, V. 30, No. 4, 2008, pp. 941-954. doi: 10.1016/j.
engstruct.2007.02.005
23. Birely, A. C.; Lowes, L. N.; and Lehman, D. E., “A Model for The
Practical Nonlinear Analysis of Reinforced-Concrete Frames Including
Joint Flexibility,” Engineering Structures, V. 34, 2012, pp. 455-465. doi:
10.1016/j.engstruct.2011.09.003
24. Magliulo, G., and Ramasco, R., “Seismic Response of Three-Dimen-
sional R/C Multi-Storey Frame Building Under Uni- and Bi-Directional
Input Ground Motion,” Earthquake Engineering & Structural Dynamics,
V. 36, No. 12, 2007, pp. 1641-1657. doi: 10.1002/eqe.709
25. Clough, R. W., “Effects of Stiffness Degradation on Earthquake
Ductility Requirement,” Rep. No. 6614, Struct. and Mat. Res., University
of California, Berkeley, Berkeley, CA, 1966.
26. Otani, S., “Inelastic Analysis of R/C Frame Structures,” Journal of
the Structural Division, ASCE, V. 100, No. 7, 1974, pp. 1433-1449.
27. Saatcioglu, M., “Modeling Hysteretic Force-Deformation Relation-
ships for Reinforced Concrete Elements,” Earthquake-Resistant Concrete
Structures Inelastic Response and Design, SP-127, S. K. Ghosh, ed., Amer-
ican Concrete Institute, Farmington Hills, MI, 1991, pp. 153-198.
28. Takeda, T.; Sozen, M. A.; and Nielsen, N. N., “Reinforced Concrete
Response to Simulated Earthquakes,” Journal of the Structural Division,
ASCE, V. 96, No. 12, 1970, pp. 2557-2573.
29. Song, J. K., and Pincheira, J. A., “Spectral Displacement Demands
of Stiffness- and Strength-Degrading Systems,” Earthquake Spectra, V. 16,
No. 4, 2000, pp. 817-851. doi: 10.1193/1.1586141
166
30. Sivaselvan, M. V., and Reinhorn, A. M., “Hysteretic Models
for Deteriorating Inelastic Structures,” Journal of Engineering
Mechanics, ASCE, V. 126, No. 6, 2000, pp. 633-640. doi: 10.1061/
(ASCE)0733-9399(2000)126:6(633)
31. Eom, T.; Park, H.; and Kang, S., “Energy-Based Cyclic Force-
Displacement Relationship for Reinforced Concrete Short Coupling
Beams,” Engineering Structures, V. 31, No. 9, 2009, pp. 2020-2031. doi:
10.1016/j.engstruct.2009.03.008
32. Eom, T., and Park, H., “Evaluation of Energy Dissipation of
Slender Reinforced Concrete Members and Its Applications,” Engi-
neering Structures, V. 32, No. 9, 2010, pp. 2884-2893. doi: 10.1016/j.
engstruct.2010.05.007
33. Sucuoğlu, H., and Acun, B., “Energy-Based Hysteresis Model for
Flexural Response of Reinforced Concrete Columns,” ACI Structural
Journal, V. 109, No. 4, July-Aug. 2012, pp. 541-549.
34. Sucuoğlu, H., and Erberik, A., “Energy-Based Hysteresis and
Damage Models for Deteriorating Systems,” Earthquake Engineering &
Structural Dynamics, V. 33, No. 1, 2004, pp. 69-88. doi: 10.1002/eqe.338
35. Kwak, H., and Kim, S., “Nonlinear Analysis of RC Beam Subjected to
Cyclic Loading,” Journal of Structural Engineering, ASCE, V. 127, No. 12,
2001, pp. 1436-1444. doi: 10.1061/(ASCE)0733-9445(2001)127:12(1436)
36. Ibarra, L.; Medina, R.; and Krawinkler, H., “Hysteretic Models that
Incorporate Strength and Stiffness Deterioration,” Earthquake Engineering
& Structural Dynamics, V. 34, No. 12, 2005, pp. 1489-1511. doi: 10.1002/
eqe.495
37. ASCE/SEI 41, “Seismic Rehabilitation of Existing Buildings,”
American Society of Civil Engineers, Reston, VA, 2007.
38. Mazzoni, S.; McKenna, F.; Scott, M. H.; and Fenves, G. L.,
“OpenSees Command Language Manual,” University of California,
Berkeley, Berkeley, CA, 2006.
39. ACI Committee 374, “Acceptance Criteria for Moment Frames
Based on Structural Testing and Commentary (ACI 374.1-05),” American
Concrete Institute, Farmington Hills, MI, 2005, 9 pp.
40. Priestley, M. J. N., “Performance Based Seismic Design,” Proceed-
ings, 12th WCEE, No. 2831, Auckland, New Zealand, 2000, pp. 1-22.
41. Park, H., and Eom, T., “A Simplified Method for Estimating the
Amount of Energy Dissipated by Flexure-Dominated Reinforced Concrete
Members for Moderate Cyclic Deformations,” Earthquake Spectra, V. 22,
No. 2, 2006, pp. 459-490. doi: 10.1193/1.2197547
42. Eom, T., and Park, H., “Elongation of Reinforced Concrete
Members Subjected to Cyclic Loading,” Journal of Structural Engi-
neering, ASCE, V. 136, No. 9, 2010, pp. 1044-1054. doi: 10.1061/(ASCE)
ST.1943-541X.0000201
43. Eom, T., and Park, H., “Evaluation of Shear Deformation and Energy
Dissipation of RC Members Subjected to Cyclic Loading,” ACI Structural
Journal, V. 110, No. 5, Sept-Oct. 2013, pp. 845-854.
44. Paulay, T., and Priestley, M. J. N., Seismic Design of Reinforced
Concrete and Masonry Buildings, John Wiley & Sons, Inc., New York,
1992, pp. 768.
45. Birely, A.; Lowes, L.; and Lehman, D., “A Practical Model for Beam-
Column Connection Behavior in Reinforced Concrete Frames,” ATC & SEI
Conference on Improving the Seismic Performance of Existing Buildings
and Other Structures, San Francisco, CA, 2009, pp. 560-571.
46. Shin, M., and LaFave, J. M., “Modeling of Cyclic Joint Shear
Deformation Contributions in RC Beam-Column Connections to Overall
Frame Behavior,” Structural Engineering & Mechanics, V. 18, No. 5, 2004,
pp. 645-669. doi: 10.12989/sem.2004.18.5.645
47. Vecchio, F. J., and Collins, M. P., “The Modified Compression Field
Theory for Reinforced Concrete Elements Subjected to Shear,” ACI Struc-
tural Journal, V. 83, No. 2, Mar.-Apr. 1986, pp. 219-231.
48. Biddah, A., and Ghobarah, A., “Modeling of Shear Deformation
and Bond Slip in Reinforced Concrete Joints,” Structural Engineering &
Mechanics, V. 7, No. 4, 1999, pp. 413-432. doi: 10.12989/sem.1999.7.4.413
49. Kim, J., and LaFave, J. M., “Joint Shear Behavior of Reinforced
Concrete Beam-Column Connections Subjected to Seismic Lateral
Loading,” Newmark Structural Engineering Laboratory-NSEL Report
Series, NSEL-020, 2009.
50. LaFave, J. M., and Kim, J., “Joint Shear Behavior Prediction for
RC Beam-Column Connections,” International Journal of Concrete
Structures and Materials, V. 5, No. 1, 2011, pp. 57-64. doi: 10.4334/
IJCSM.2011.5.1.057
51. Fischinger, M.; Kramar, M.; and Isaković, T., “Cyclic Response of
Slender RC Columns Typical of Precast Industrial Buildings,” Bulletin of
Earthquake Engineering, V. 6, No. 3, 2008, pp. 519-534. doi: 10.1007/
s10518-008-9064-7
52. FEMA, 440, “Improvement of Nonlinear Static Seismic Analysis Proce-
dures,” Federal Emergency Management Agency, Washington, DC, 2005.
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S15
Ductility Enhancement in Beam-Column Connections
Using Hybrid Fiber-Reinforced Concrete
by Dhaval Kheni, Richard H. Scott, S. K. Deb, and Anjan Dutta
In the first part of this study, 36 prisms made of plain concrete,
steel fiber-reinforced concrete, and hybrid fiber-reinforced concrete
(HyFRC) were tested under quasi-static load to account for variability
in fiber specifications. Two types of steel fibers with hooked ends and
two types of polymer fibers—namely, polypropylene and polyester—
were used. The HyFRC prisms exhibited approximately 10 to 15 times
the enhancement in toughness compared to similar plain concrete
prisms up to failure. In the second part of the experimental program,
four types of beam-column connections were tested under cyclic
loading. Test results established that the addition of hybrid fiber in the
joint region of the specimens is effective in enhancing their displace-
ment ductility and energy dissipation capacity. Detailed measurement
of strain distributions along the main reinforcement of the specimens
showed that there was substantial reduction in strain levels in the
specimens with HyFRC in the joint region.
Keywords: beam-column; cyclic loading; damage; gauged bar; hybrid
fiber-reinforced concrete; toughness.
INTRODUCTION
During past devastating earthquakes, it has been noted
that beam-column connections act as one of the weakest
links in moment-resisting reinforced concrete (RC) framed
structures. Behavior of reinforced concrete frame structures
during earthquakes throughout the world has highlighted
the consequences of poor performance of beam-column
connections and it has been observed that exterior connec-
tions suffer more in comparison to interior ones. For some
years, the Indian Institute of Technology Guwahati has
been involved in a major research project to investigate
seismic effects on exterior reinforced concrete beam-column
connections because large parts of India lie in highly active
seismic zones, making issues related to the failure of these
connections of particular relevance.1,2 Some of the important
studies on beam-column connections are reviewed in this
section. Paulay et al.3 examined the behavior of beam-
column joints under seismic actions. The existence of two
shear-resisting mechanisms—one involving joint shear
reinforcement and the other a linear concrete strut—were
postulated and the effects of reversed cyclic loading on these
mechanisms, in both elastic and inelastic range of response,
were discussed. Durrani and Wight4 reported results of an
experimental investigation on the performance of an interior
beam-column joint under earthquake-type loading. Abdel-
Fattah and Wight5 studied the relocation of plastic hinging
zones for earthquake-resistant design of reinforced concrete
(RC) buildings. Chutarat and Aboutaha6 investigated a solu-
tion for relocating potential beam plastic hinge zones by the
use of headed bars in the exterior RC joints. Joshi et al.7
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
tested four full-scaled exterior precast beam-column joint
under cyclic loading in order to identify a suitable tech-
nique for connecting precast beam and column components.
Nie et al.8 tested six beam-column joints comprising three
interior and three corner joints to develop a new connec-
tion system for concrete-filled steel tube composite column
and RC beams. Park and Mosalam9 carried out both exper-
imental and analytical studies to develop a shear strength
model and a moment-rotation relationship (backbone curve)
of unreinforced corner beam-column joint. It was observed
that consideration of the flexibilities for unreinforced joints
is important for seismic assessment of older-type RC build-
ings with unreinforced joints.
The main thrust of the present investigation was to enhance
the displacement ductility of the beam-column connections
rather than seek increases in strength. It is known that the
ductile steel fibers in concrete continue to carry stresses
beyond matrix cracking.10 However, the effect of steel fibers
on the compressive strength of concrete is variable. The range
of increase was from negligible in most cases to 23% for
concrete containing 2% by volume of fiber.11 The compres-
sion stress-strain curves for steel fiber-reinforced concrete
(SFRC) showed that using steel fibers does not necessarily
increase the peak stress dramatically, but the post-peak
descending slope of SFRC is significantly less steep than
that of plain concrete. Ultimate flexural strength generally
increases in relation to the fiber volume fraction and aspect
ratio. Concentrations less than 0.5 volume percent of low
aspect fibers have negligible effect on static strength proper-
ties. However, the gradual and multi-scale nature of concrete
implies that different types of fibers may be combined to get
enhanced response from the structure. The use of both steel
fibers (macrofibers) and polymer fibers (microfibers) was
found to be very effective as both microcracks and macro-
cracks are arrested, leading to enhanced impact strength
and toughness.12 Investigations were carried out12-14 using
different types of steel and polypropylene fibers to determine
the optimal volume fraction of these fibers in concrete. There
is good enhancement in displacement ductility and energy
absorption capacity of a beam having hybrid fiber-reinforced
concrete (HyFRC), where the optimum ratio of steel to poly-
olefin fiber is 0.6:0.4.15 However, incorporation of steel
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-286.R2, doi: 10.14359/51687405, received June 21, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
167
Table 1—Details of fibers
Length, in. Diameter, in.
Types of fibers (mm) (mm) Shape
0.47 to 0.71 0.0012 to 0.0014
Polyester (PE) Straight
(12 to 18) (0.03 to 0.035)
0.0012 to 0.0014
Polypropylene (PP) 0.47 (12) Straight
(0.03 to 0.035)
Hooked
Steel (SF1) 2.36 (60) 0.0296 (0.75)
Steel (SF2) 1.38 (35) 0.0216 (0.55)
fiber decreases the workability considerably. This situation
adversely affects the consolidation of the fresh mixture and
the fiber volume at which this situation is reached depends
on the length and diameter of the fiber. An investigation had
been carried out16 to find the proper dosage of plasticizer
using different combinations of fibers. Parra-Montesinos17
presented an overview of applications of tensile strain-
hardening, high-performance fiber-reinforced cement
composites (HPFRCCs) for earthquake-resistant design
of structural elements such as beam-column connections,
low-rise walls, and coupling beams. Numerous types of
FRCCs reinforced with steel, polymeric, glass, and carbon
fibers were evaluated for structural applications. With regard
to twisted steel fibers, high-performance tensile response
could be achieved with a 1.5 to 2.0% volume fraction.
Ultra-high-molecular-weight polyethylene in volume frac-
tions ranging between 1.5 and 2.0% was found to exhibit
excellent tensile response with multiple cracking patterns.
Zohrevand and Mirmiran18 used two promising materials
to develop a new hybrid system. Engineered cementitious
composites (ECCs) allow optimization of the microstructure
of the material to achieve ultra-high strength, ductility, and
fracture toughness, while fiber-reinforced polymer (FRP)
tubes help to eliminate the need for lateral steel reinforce-
ment for confinement and shear in RC columns. Kumar
et al.19 carried out detailed studies to improve seismic perfor-
mance of bridge columns using self-consolidating HyFRC
(SC-HyFRC) composite. It was observed that SC-HyFRC
columns exhibited better damage resistance and superior
load-carrying capacity in spite of a 50% reduction in trans-
verse reinforcement. Bedirhanoglu et al.20 investigated the
seismic behavior of deficient RC exterior beam-column
joints constructed with low-strength concrete and plain
reinforcing bars before and after retrofitting with prefab-
ricated high-performance fiber-reinforced cementitious
composite (HPERCC) panels. Tests showed that retrofitting
with prefabricated HPFRCC panels provided considerable
enhancement, both in strength and in displacement capacity,
provided that the panels were properly anchored to the joint
core and the slippage of the beam longitudinal bars in the
joint core was prevented. Thus, while the existing literature
clearly demonstrated that the use of hybrid fibers led to the
development of improved energy dissipation and a more
ductile mode of failure of the specimen, a detailed investi-
168
Density, lb/ft3 Aspect ratio Tensile strength,
(kg/m3) (length/diameter) Classification ksi (MPa)
57.92 to 72.4
85.625 (1370) 400 to 500 Micro
(400 to 500)
57.92 to 72.4
56.875 (910) 400 to 500 Micro
(400 to 500)
490.625
80 Macro 177.37 (1225)
(7850)
490.625
64 Macro 159.27 (1100)
(7850)
gation was needed to arrive at the best possible combination
of steel and polymer fibers and their volume fraction due to
likely variations in the locally available product used for the
experimental investigation. The combinations showing best
possible enhancement in toughness as compared to a similar
plain concrete specimen up to failure were selected for use
in beam-column connections.
As part of the extensive main test program, strain-gauged
U-bars manufactured at Durham University, UK, were
incorporated as part of the main beam reinforcement in four
beam-column specimens. The intention was to use the very
comprehensive strain information generated by these bars
to give detailed comparisons of the displacement ductility
and energy dissipation in the three specimens with different
fiber combinations and one conventional reinforced concrete
specimen. Detailed measurement of strain distribution along
the reinforcement of the specimens would also show the
extent of the reductions in strain levels in the main reinforce-
ment in the specimens with HyFRC in the joint region.
RESEARCH SIGNIFICANCE
It is expected that use of HyFRC in the joint region of
beam-column connections would delay crack formation
and crack growth, which in turn would reduce strain level
in the reinforcing bars. This would result in enhancement
of displacement ductility of the connections due to delayed
bond-slip failure. In this study, comprehensive strain
measurements were also made for objective assessment of
reduction in strain in main reinforcing bars and the measured
reinforcing bar strain data were observed to be correlated to
the enhancement of displacement ductility and energy dissi-
pation of the beam-column specimens using HyFRC.
TEST FOR SELECTION OF FIBERS
A detailed experimental exercise was carried out for the
selection of the type and volume fraction of the fibers to be
used in the concrete for the enhancement of toughness.21
Steel fibers of two different aspect ratio (length/diameter)
synthetic polymer fibers, such as polyester and polypro-
pylene, were used in the concrete mixtures, and the geomet-
rical and mechanical properties are shown in Table 1.
Thirty-six specimens of size 5.9 x 5.9 x 27.56 in. (150 x
150 x 700 mm) were cast for the evaluation of toughness
of the HyFRC element having different types of fibers with
ACI Structural Journal/March-April 2015
Table 2—Volume fractions of fibers used and average toughness of prisms
Types of fibers (volume fraction, %)
Average toughness,
No. of combination SF1 SF2 PP PE Total volume fraction, % kip-in. (kN-mm)
1 (plain concrete) — — — — — 0.221 (24.949)
2 0.5 0.5 — — 1.00 2.213 (250.039)
3 0.4 0.4 0.15 — 0.95 3.197 (361.291)
4 0.4 0.4 0.2 — 1.00 2.486 (280.991)
5 0.5 0.5 0.15 — 1.15 3.445 (389.317)
6 0.5 0.5 0.2 — 1.20 2.849 (321.982)
7 0.6 0.6 0.15 — 1.35 3.225 (364.467)
8 0.4 0.4 — 0.15 0.95 2.713 (306.551)
9 0.4 0.4 — 0.2 1.00 2.720 (307.375)
10 0.5 0.5 — 0.15 1.15 2.638 (298.071)
11 0.5 0.5 — 0.2 1.20 2.706 (305.843)
12 0.6 0.6 — 0.15 1.35 2.657 (300.253)

different volume fractions. Twelve different combinations
of the mixtures were considered with three samples for
each combination. The target strength of concrete used for
the plain as well as all the specimens with different types
of fibers is 4.35 ksi (30 MPa). Table 2 shows volume frac-
tion and type of fibers used for each combination along with
average toughness obtained from the test. Flexural toughness
tests were carried out according to ASTM C1609/C1609M.22
Prisms were tested in third-point loading with middle third
of 7.87 in. (200 mm) under constant flexural demand.
Displacement control test was performed and, hence, the test
could be carried out to track the post-crack behavior until
almost complete loss of load-carrying capacity of the prism
specimens. The area under the load-displacement curve was
used to evaluate the total strain energy stored or equivalent
toughness of the specimen.
The 12 combinations (Table 2) considered in the present
study demonstrated that the best toughness was achieved
with a combination of steel fibers (two different aspect
ratios) with 0.5 (SF1) and 0.5 (SF2) volume fraction (%)
and 0.15% polypropylene by volume fraction. The enhance-
ment in toughness was found to be about 15 times that of
the plain concrete specimen. HyFRC, comprising two basic
types of fibers (steel fibers and synthetic polymer fibers
[polypropylene, polyester]), performed very well in general,
as a substantial improvement in toughness was observed. All
these specimens were also observed to undergo much higher
ultimate displacements compared to plain concrete speci-
mens, thus leading to a more ductile type of failure pattern.
DETAILS OF BEAM-COLUMN CONNECTIONS
To ascertain the efficacy of different fibers and their combi-
nations in enhancing ductility in beam-column connections,
four different cases were considered. The beam-column
specimens considered for the experimental investigations
were of a two-thirds scale. The fiber contents in the four
strain-gauged specimens were determined as follows:
Specimen 1: No fibers—used as a control specimen
Specimen 2: SF1 (0.5%) + SF2 (0.5%) + PP (0.15%)—
steel fibers + polypropylene
Specimen 3: SF1 (0.5%) + SF2 (0.5%) + PE (0.15%)—
steel fibers + polyester
Specimen 4: SF1 (0.5%) + SF2 (0.5%)—steel fibers only
The fiber contents shown are percentage by total volume
of the concrete. Specimens 2 and 3, containing two types of
fiber (steel and polymer) were termed “hybrid” specimens.
Concrete with the aforementioned combinations was placed
in the D-region (ACI 318, Section A.123) only, whereas
normal concrete was used in the B-region (ACI 318,
Section A.123), representing remaining parts of the specimen
as shown in Fig. 1(a).
The four beam-column connection specimens were
designed following the provisions of ductile detailing
according to IS 1392024 and satisfying the condition of
strong column-weak beam flexural design. Columns had a
7.87 x 7.87 in. (200 x 200 mm) cross section, and beams
were 9.45 in. (240 mm) deep by 7.87 in. (200 mm) wide.
Confining reinforcement according to the provisions of
IS 1392024 was provided in the joint region. Specimens were
cast using concrete with a target strength of 4.35 ksi (30 MPa)
for 5.9 in. (150 mm) cube. The concrete mixture proportion
was 1:1.84:3.18 with a water-cement ratio (w/c) of 0.59,
and the observed standard deviation in cube test results was
approximately 0.36 ksi (2.5 MPa). Portland cement Type IP
as per ASTM C595/C595M25 was used and the maximum
aggregate size was 0.63 in. (16 mm). The reinforcement was
equivalent to UK Grade B500C,26 which has a yield stress
of 72.46 ksi (500 MPa), and elongation at ultimate load is
7.5%. Details of the specimens and the reinforcement layout
are shown in Fig. 1(b). The beam reinforcement from top
(two side bars of 0.39 in. [10 mm] diameter) was bent down-
ward in the beam-column connection up to the requirement
of development length. Similarly, beam reinforcement from
bottom (two side bars of 0.39 in. [10 mm] diameter) was
bent upward in the beam-column connection. The middle
bar of the beam (0.47 in. [12 mm] diameter) was the gauged
U-bar. Each U-bar contained 31 electric resistance strain

ACI Structural Journal/March-April 2015 169

Fig. 1—Beam-column connections. (Note: Dimensions in mm; 1 mm = 0.0394 in.)
gauges installed within a central longitudinal duct and was
positioned as shown in Fig. 2. Bars of this type have been
used in a number of previous investigations at Durham
University and full details of their manufacture can be
found elsewhere.27,28
TEST PROCEDURE
The schematic diagram of the testing arrangement is
shown in Fig. 3, and a photograph of the test rig is shown in
Fig. 4. The column was placed in a horizontal position with
the beam vertical. An axial load of 10% of the gross capacity
of column was applied to the column to simulate gravity
loading. To simulate support condition at both ends of the
column, special roller supports were fabricated. Cyclic load
was applied to the beam by a servo-controlled MTS actu-
170
ator with a loading capacity of ±36.23 ksi (250 kN) and
±4.92 in. (125 mm) maximum stroke. The displacement
history as shown in Fig. 5 was followed with three push-
pull cycles being applied at each increment of amplitude and
each cycle taking 40 seconds to complete. The importance
of loading sequence has not yet been established through
research, and the sequence of large-versus-small excursions
in an element of a structure subjected to a severe earthquake
does not follow any consistent pattern. The number of the
inelastic excursions increases with a decrease in the period
of the structural system, with the rate of increase being very
high for short-period systems. It is to be recognized that
cyclic demands for a structure depend on a great number
of variables and a unique loading history will always be a
compromise.29 Thus, a multi-cycle loading history was used
ACI Structural Journal/March-April 2015
Fig. 2—Strain gauge layout in gauged bar. (Note: Dimensions in mm; 1 mm = 0.0394 in.)
Fig. 3—Test arrangement.
Fig. 4—View of actual test setup.
in the present work. Tests were stopped when the degrada-
tion in load-carrying capacity was in the range of 50 to 60%.
Further, the damage pattern in the connection zone was also
monitored and the test was stopped when the damage in the
connection zone was serious enough to cause concern about
the safety of the testing equipment.
RESULTS
Table 3 lists 28-day concrete strengths for the four beam-
column specimens. It is observed that the compressive
cube strength of Specimen 1 without fibers is slightly more
than that of Specimen 2 with steel fibers and PP fibers. The
compressive cube strength of specimen with polymer fiber
may be less than that of plain concrete.11 The workability
ACI Structural Journal/March-April 2015
Fig. 5—Displacement history. (Note: 1 mm = 0.0394 in.)
Table 3—Strength of concrete in beam-column
specimens
Compressive cube Indirect tensile
Specimen Fiber types strength, ksi (MPa) strength, ksi (MPa)
1 None 4.67 (32.22) 0.585 (4.04)
2 SF1/SF2/PP 4.50 (31.11) 0.847 (5.85)
3 SF1/SF2/PE 4.89 (33.77) 0.715 (4.94)
4 SF1/SF2 4.95 (34.21) 0.607 (4.19)
was influenced by the addition of fibers, which, however,
was improved by the addition of a plasticizer. If improperly
produced, HyFRC may entrap excessive air and may thus
possess low density. Hence, some variation in strength may
also be attributed to variability in workability of HyFRC.
However, the improvement in rate of decay of load from the
peak level and toughness in HyFRC specimens are consis-
tently observed to be very significant, and this is true even
when the compressive strength of HyFRC is lower than that
of plain concrete. Further, it is observed that the difference
in the values of indirect tensile strength between Specimen 1
(without fibers) and Specimen 4 (with steel fibers) is not
significant. In the case of steel fiber-reinforced concrete,
the response is generally linear until the tensile stress
reaches a value slightly higher than the tensile strength of
the plain concrete. The fiber concrete cracks at that point.
The maximum post-cracking strength can either be less
or greater than the cracking stress, which depends on the
volume fraction as well as aspect ratio of steel fibers used.30
While higher volume fraction and aspect ratio is likely to
show higher enhancement in tensile strength, the adopted
volume fraction of 1% and aspect ratio of 60 to 80 could
actually achieve a marginal improvement in tensile strength.
However, the improvement in rate of decay of load from the
171
Table 4—Test result of different beam-column specimens
Peak load, kip (kN)
Specimen Pull Push
1 8.75 (38.935) 7.73 (34.402)
2 8.69 (38.645) 9.64 (42.819)
3 7.24 (32.212) 6.863 (30.533)
4 6.64 (29.537) 7.47 (33.273)
Fig. 6—Comparison of damage in joint region.
peak level and toughness in HyFRC specimens are consis-
tently observed to be very significant, and this is true even
when the compressive strength of HyFRC is lower than that
of plain concrete.
At early load stages, all four specimens performed in a
similar fashion with initial cracking in both the beam and the
connection zones being followed by a plastic hinge forming
in the beam close to the column face. This was as expected
in view of the similarity of their geometry and the design
of their reinforcement layout. However, as beam displace-
ments increased, real differences became apparent between
the behavior of Specimen 1 (no fibers) and the other three
specimens (with fibers), as indicated by the maximum
displacements listed in Table 4. Figure 6 shows photographs
of the connection zones after the final displacement, and the
relative contribution of three selected fiber combinations in
controlling degradation of the connection zone is immedi-
ately apparent. The addition of polymer fibers was partic-
ularly effective. A more detailed comparison of specimen
172
Maximum
displacement, in. (mm) Drift, % Toughness, kip-in. (kN-mm)
1.84 (46.67) 5.18 236.55 (26,711)
2.76 (70.0) 7.77 688.029 (77,691)
2.76 (70.0) 7.77 593.66 (67,035)
2.43 (61.67) 6.85 410.855 (46,393)
behavior is obtained by comparing the load-displacement
hysteresis loops as shown in Fig. 7. All four specimens
exhibited displacement ductility, but the superior perfor-
mance of the specimens with fibers is indicated by their
higher final displacements and their reduced strength degra-
dation as displacements increased. Direct comparisons can
be made by comparing the envelopes of their hysteresis
curves, which are generated by joining all the peak values
of the capacity corresponding to first cycle for each of the
displacement amplitudes. Envelope curves for all four
specimens are shown in Fig. 8, which reinforces the afor-
mentioned observations. Specimen 1 (no fibers) achieved a
final displacement of 1.837 in. (46.67 mm) with significant
load reduction as this displacement was approached. It is
worth mentioning that the damage pattern and load-carrying
capacities of Specimen 1 were in agreement with test results
on similar specimens as part of the ongoing test program
at IIT Guwahati.31 Both Specimens 3 and 4 performed
noticeably better, but Specimen 3 was slightly superior
ACI Structural Journal/March-April 2015
Fig. 7—Load-displacement hysteresis loops. (Note: 1 kN = 0.225 kip; 1 mm = 0.0394 in.)
as it achieved a final displacement of 2.76 in. (70.0 mm)
compared to 2.43 in. (61.67 mm) of Specimen 4. Specimen 2
clearly outperformed the other three specimens with its final
displacement of 2.76 in. (70.0 mm) coupled to a maximum
load of approximately 5.8 ksi (40.0 kN), which was compa-
rable with that for Specimen 1 (control specimen) and better
than the values for Specimens 3 and 4. Table 4 shows the
ultimate load carried by different specimens as obtained
from experimental investigation. Values of damage index by
Park and Ang32 for all four specimens are presented in Fig. 9
for comparison reasons. Different parameters involved
in the evaluation of damage index were estimated as per
Karayannis et al.33 From these results it can be inferred that
the Specimen 2 presented a lowest damage factor, while
Specimen 1 showed highest during the course of loading.
Specimen 3 had a relatively lower damage index than Spec-
imen 4 did. Further, Fig. 10 demonstrates the contribution
of fibers in the test specimens in terms of nominal principal
Fig. 8—Envelope curve for beam-column specimens. (Note:
tensile stresses values developed in the beam-column joint
1 kN = 0.225 kip; 1 mm = 0.0394 in.)
regions.29 From this figure it is construed that though the
developed nominal principal tensile stresses of all joints tensile stress indicate the influence of fibers in enhancing the
are not substantially different, the decay in stress for Spec- seismic performance of these specimens.
imen 1 is much more significant than all other specimens. The observation related to displacement ductility and
Specimen 2 and 3 also presented flatter stress patterns than cumulative energy dissipation of all the four specimens are
Specimen 4. Thus, both damage index and nominal principal discussed in detail to demonstrate the significance of using
HyFRC in the D-region of a beam-column connection.

ACI Structural Journal/March-April 2015 173


Fig. 9—Comparisons of the Park and Ang32 damage indexes
of the tested specimens.
Fig. 10—Nominal principal tensile stresses developed
in beam-column joint region. (Note: 1 MPa = 0.145 ksi;
1 mm = 0.0394 in.)
Cumulative energy dissipation
Cumulative energy dissipation is calculated as the area
under the load-displacement hysteresis loops, which is the
measure of toughness of the specimen before failure. Tough-
ness values for all the four specimen types are shown in
Table 4 and it is immediately clear that the combination SF1/
SF2/PP of appropriate volume fraction has tremendous poten-
tial for use in beam-column connections in seismic zones for
the enhancement of ductile behavior. Figure 11 shows the
cumulative energy dissipation by all the specimens. Specimen
2 has the maximum energy dissipation capacity, whereas the
control specimen (Specimen 1) has much less capacity to
store dissipated energy. The results show the sequence of the
performance of all the specimens as
Specimen 2 (best) > Specimen 3 > Specimen 4 >
Specimen 1 (worst)
Displacement ductility
Displacement ductility, which is the ratio of ultimate
displacement to yield displacement, was calculated for all
specimens from the respective envelope curves according
to the procedure proposed by Shannag and Alhassan.34
Figure 12 shows (using the results for Specimen 1) the neces-
sary construction for estimating beam tip displacements
under yield and ultimate conditions. As shown in this figure,
the yield displacement is calculated as the point of inter-
174
Fig. 11—Cumulative energy dissipation of beam-column
specimens. (Note: 1 kN = 0.225 kip; 1 mm = 0.0394 in.)
Fig. 12—Procedures for displacement ductility calculation.
(Note: 1 kN = 0.225 kip; 1 mm = 0.0394 in.)
section between two straight lines drawn on the envelope
curve. The first line is obtained by extending the line joining
the origin and the point on envelope curve corresponding
to 50% of ultimate load-carrying capacity, while the second
line is the horizontal line corresponding to 80% of ultimate
load-carrying capacity. Ultimate displacement corresponds
to the point of intersection between the horizontal line corre-
sponding to 80% of ultimate load carrying capacity and the
envelope curve at the far end. The displacement ductility
was calculated as the ratio of maximum displacement to
the yield displacement, and the values are listed in Table 5.
Specimen 2 showed the maximum displacement ductility,
while the control specimen had the least and was worse than
any of the other three fiber-reinforced specimens.
Observation on reinforcement strain data
Detailed data pertaining to reinforcement strains were
obtained from the strain-gauged U-bars. The distributions
are plotted along a straightened form of the bar with Gauge 1
(top leg) at the left-hand end and Gauge 31 (bottom leg) at
the right-hand end. Strains are plotted by considering tensile
strains as positive. At low displacement levels, the reinforce-
ment was behaving elastically with peak strains occurring
at the column face due to flexural cracking. There was also
a degree of symmetry between distributions for the push
and pull directions of loading, as would be expected from
the symmetrical reinforcement layout. As displacements
increased, however, tensile stresses spread further into the
ACI Structural Journal/March-April 2015
Table 5—Displacement ductility of different beam-
column specimens
Specimen ∆y, in. (mm) ∆u, in. (mm) ∆u/∆y
1 0.172 (4.375) 1.044 (26.5) 6.057
2 0.125 (3.165) 1.826 (46.335) 14.640
3 0.156 (3.950) 1.793 (45.5) 11.519
4 0.210 (5.330) 1.834 (46.58) 8.739

connection zone until the entirety of the U-bar in the connec-

tion zone was in tension for both the push and pull loading
directions. Finally, the reinforcement yielded, leading to the
development of large residual strains. This behavior, which
was observed in all four specimens, was consistent with results
from beam-column connection tests performed by Scott.35
Specimen 1 showed the appearance of the first crack
at the beam-column face at a displacement of ±0.131 in.
(3.33 mm). The cracks in the other three specimens also
started approximately at the same location and displace-
ment level. The strain distributions in all three specimens
are shown in Fig. 13. The maximum value of strain in
Specimen 1 was marginally higher than those in the other
three specimens. Specimen 1 was tested up to a maximum
displacement of ±1.837 in. (46.67 mm) and the development
of deep crack along the beam-column interface could be
observed; the beam was observed to rotate about this plane
during the final stages of the experiment. Distributions of
strain for displacement of ±1.837 in. (46.67 mm) are shown
for all the specimens in Fig. 14, where the magnitudes of
peak strains indicate improved performance of specimens
with fiber compared to the control specimen. At displace-
ments of 1.837 in. (46.67 mm)—the largest displacement
sustained by all four specimens—peak strains were approx-
imately (as there was considerable creep) 21,420, 6110,
9900, and 13,300 microstrain for Specimens 1 to 4, respec-
tively. Increasing displacements led to increased strains in
Specimens 2 to 4. The data indicated that the addition of
fibers to the concrete mixture reduced the strain level in rein-
forcing bars required to achieve a given displacement, with
hybrid Specimens 2 and 3 proving more effective than Spec-
imen 4, which had steel fibers only. Specimens 2 and 3 were
tested up to ±2.76 in. (70 mm), while Specimen 4 could be
tested up to ±2.43 in. (61.67 mm). The distributions of strain
for the displacement of ±2.43 in. (61.67 mm) are shown for
Specimens 2 to 4 in Fig. 15. Thus, it is again observed that
fibers are very effective in arresting microcracks as well
as macrocracks and led to the growth of relatively lower
order of strain. The strain values were observed to gradually
increase near the connection zone, indicating the develop-
ment of plastic hinges. The magnitude of maximum strains
for Specimen 2 is again observed to be the least (Fig. 15)
and, hence, a HyFRC Specimen 2 is likely to be more effi-
cient under seismic loading. Fig. 13—U-bar strains at 0.131 in. (3.33 mm) displacement.
Overall, adding fibers gave increased displacement (Note: 1 mm = 0.0394 in.)
ductility coupled with reduced reinforcement strains. It polyester). Further, hybrid fibers (steel and polymer fibers)
may further be noted that polypropylene has better ulti- are better than the use of only steel fiber, as the inclusion
mate elongation as compared to polyester fibers. Thus, it is of steel fibers in the concrete mixture is an effective way of
observed that the performance of Specimen 2 (with poly- reducing macrocracking, while polymer fibers are very good
propylene) was relatively better than Specimen 3 (with

ACI Structural Journal/March-April 2015 175


Fig. 14—U-bar strains at 1.837 in. (46.67 mm) displace-
ment. (Note: 1 mm = 0.0394 in.)
at arresting microcracking, thus leading to overall enhance-
ment in toughness. The performance of Specimen 4 (with
steel fibers only) was thus relatively inferior compared with
Specimens 2 and 3. The addition of polymer fibers proved to
176
Fig. 15—U-bar strains at 2.43 in. (61.17 mm) displacement.
(Note: 1 mm = 0.0394 in.)
be a particularly effective way of limiting connection zone
degradation (Fig. 6) thus making joint repair after a seismic
event a more practicable proposition.
CONCLUSIONS
The improvement in displacement ductility of concrete
through the use of different fibers and their combination is
fairly well-known. However, specific applications of HyFRC
in beam-column connections with detailed measurements of
strain development were carried out to further understand
how steel strain is reduced while displacement ductility is
improved. Four tests were performed on exterior beam-
column connections subjected to seismic loading, each of
which contained a strain-gauged U-bar as part of the main
beam reinforcement. Specimen 1 was cast without the addi-
tion of any fiber in concrete and was considered as the control
specimen for comparison, whereas Specimens 2 and 3 were
ACI Structural Journal/March-April 2015
hybrids, containing both steel and polymer fibers, and Spec-
imen 4 contained steel fibers only. Based on the results of
this study, the following conclusions are made:
1. All three specimens with fibers showed marked
improvements in displacement ductility compared with the
control specimen. Hybrid Specimen 2 (steel and polypro-
pylene fibers) performed best while hybrid Specimen 3 (steel
and polyester fibers) performed relatively more poorly. The
use of fibers with better ultimate elongation is attributed to
the relative improvement in performance. Specimens with
hybrid fibers performed better compared to specimens with
only steel fibers, as both macrocracks and microcracks are
better controlled by hybrid fibers. Thus, it is also observed
that Specimens 2 and 3 performed better than Specimen 4
(steel fibers only).
2. The extent of damage of the connection zone in all the
three specimens with fibers is significantly less than in the
control specimen.
3. The results from the strain-gauged bars indicated that
the improved displacement ductility in specimens with
fibers was accompanied by reduced strains in the reinforce-
ment under ultimate displacement conditions. Large residual
strains were developed in all four specimens once the
reinforcement had yielded.
4. The high confinement steel requirement in and around
the connection zone of beam-column connections in seismic
areas may be reduced by the use of HyFRC, while still main-
taining a very high displacement ductility level.
AUTHOR BIOS
Dhaval Kheni is a Postgraduate Student in civil engineering at the Indian
Institute of Technology (IIT) Guwahati, Guwahati, India. He received his
bachelor of engineering from Gujurat University, India. His research inter-
ests include studies of beam-column joints.
ACI member Richard H. Scott is a Visiting Professor of structural engi-
neering at City University London, London, UK, and IIT Roorkee, Roorkee,
India. He received his BSc(Eng) in civil engineering from Queen Mary
College, University of London, London, UK, in 1968; his MSc in concrete
structures and technology from Imperial College London, London, UK, in
1973; and his PhD from Durham University, Durham, UK, in 1985. He is
a member of ACI Committees 435, Deflection of Concrete Building Struc-
tures, and 444, Structural Health Monitoring and Instrumentation. His
research interests include the behavior of reinforced concrete structural
elements and structural health monitoring.
S. K. Deb is a Professor of civil engineering at IIT Guwahati. He received
his bachelor of engineering from Gauhati University, Guwahati, India; his
masters in engineering from Jadavpur University, Kolkata, India; and his
PhD from IIT Roorkee. His research interests include the passive structural
control, system identification, and seismic retrofitting.
Anjan Dutta is a Professor of civil engineering at IIT Guwahati. He
received his bachelor of engineering from Gauhati University; his masters
in engineering from IIT Madras, Chennai, India; and his PhD from IIT
Delhi, Delhi, India. His research interests include the use of high-perfor-
mance materials in concrete and system identification based studies in
assessment of structures.
ACKNOWLEDGMENTS
The financial support provided by the Royal Society’s International Joint
Project Award is gratefully acknowledged.
REFERENCES
1. Choudhury, A. M.; Deb, S. K.; and Dutta, A., “Study on Size Effect of
Fibre Reinforced Polymer Retrofitted Reinforced Concrete Beam-Column
ACI Structural Journal/March-April 2015
Connections under Cyclic Loading,” Canadian Journal of Civil Engi-
neering, V. 40, No. 4, 2013, pp. 353-360. doi: 10.1139/cjce-2012-0041
2. Marthong, C.; Dutta, A.; and Deb, S. K., “Seismic Rehabilitation of
RC Exterior Beam-Column Connections Using Epoxy Resin Injection,”
Journal of Earthquake Engineering, V. 17, No. 3, 2013, pp. 378-398. doi:
10.1080/13632469.2012.738284
3. Paulay, T.; Park, R.; and Priestly, M. J. N., “Reinforced Concrete
Beam-Column Joints under Seismic Actions,” ACI Journal Proceedings,
V. 75, No. 6, June 1978, pp. 585-593.
4. Durrani, A. J., and Wight, J. K., “Behavior of Interior Beam-to-Column
Connections under Earthquake-Type Loading,” ACI Journal Proceedings,
V. 82, No. 3, May-June 1985, pp. 343-349.
5. Abdel-Fattah, B., and Wight, J. K., “Study of Moving Beam Plastic
Hinging Zones for Earthquake-Resistant Design of Reinforced Concrete
Buildings,” ACI Structural Journal, V. 84, No. 1, Jan.-Feb. 1987, pp. 31-39.
6. Chutarat, N., and Aboutaha, R. S., “Cyclic Response of Exterior
Reinforced Concrete Beam-Column Joints Reinforced with Beaded
Bars—Experimental Investigation,” ACI Structural Journal, V. 100, No. 2,
Mar.-Apr. 2003, pp. 259-264.
7. Joshi, M. K.; Murty, C. V. R.; and Jaisingh, M. P., “Cyclic Behaviour
of Precast RC Connections,” Indian Concrete Journal, V. 79, No. 11, 2005,
pp. 43-50.
8. Nie, J.; Bai, Y.; and Cai, C. S., “New Connection System for Confined
Concrete Columns and Beams. I: Experimental Study,” Journal of Struc-
tural Engineering, ASCE, V. 134, No. 12, 2008, pp. 1787-1799. doi:
10.1061/(ASCE)0733-9445(2008)134:12(1787)
9. Park, S., and Mosalam, K. M., “Experimental and Analytical Studies on
Reinforced Concrete Buildings with Seismically Vulnerable Beam-Column
Joints,” Report No. PEER 2012/03, Pacific Earthquake Engineering Research
Center, University of California, Berkeley, Berkeley, CA, 2012.
10. Banthia, N., and Trottier, J. F., “Test Methods for Flexural Toughness
Characterization of Fiber-Reinforced Concrete: Some Concerns and a Prop-
osition,” ACI Materials Journal, V. 92, No. 1, Jan.-Feb. 1995, pp. 48-57.
11. ACI Committee 544, “Design Considerations for Steel Fiber Rein-
forced Concrete (ACI 544.4R),” ACI Structural Journal, V. 85, No. 5,
Sept.-Oct. 1988, pp. 563-580.
12. Banthia, N., and Soleimani, S. M., “Flexural Response of Hybrid
Fiber-Reinforced Cementitious Composites,” ACI Materials Journal,
V. 102, No. 6, Nov.-Dec. 2005, pp. 382-389.
13. Qian, C., and Stroeven, P., “Fracture Properties of Concrete Reinforced
with Steel-Polypropylene Hybrid Fibres,” Cement and Concrete Compos-
ites, V. 22, No. 5, 2000, pp. 343-351. doi: 10.1016/S0958-9465(00)00033-0
14. Banthia, N., and Nandakumar, N., “Crack Growth Resistance of Hybrid
Fibre Reinforced Cement Composites,” Cement and Concrete Composites,
V. 25, No. 1, 2003, pp. 3-9. doi: 10.1016/S0958-9465(01)00043-9
15. Mohankumar, G., and Bangaruchandran, L., “Structural Behavior of
Hybrid Fibre Reinforced Concrete Beams,” Indian Concrete Journal, V. 83,
No. 10, 2009, pp. 14-20.
16. Blunt, J., and Ostertag, C. P., “Performance-Based Approach for
the Design of a Deflection Hardened Hybrid Fibre-Reinforced Concrete,”
Journal of Engineering Mechanics, ASCE, V. 135, No. 9, 2009, pp.
978-986. doi: 10.1061/(ASCE)0733-9399(2009)135:9(978)
17. Parra-Montesinos, G. J., “High-Performance Fiber-Reinforced
Cement Composites: An Alternative for Seismic Design of Structures,” ACI
Structural Journal, V. 102, No. 5, Sept.-Oct. 2005, pp. 668-675.
18. Zohrevand, P., and Mirmiran, A., “Cyclic Behavior of Fibre Rein-
forced Polymer-Encased Engineered Cementitious Composite for Bridge
Columns,” Structures Congress 2010, 2010, pp. 1828-1839.
19. Kumar, P.; Jen, G.; Trono, W.; Panagiotou, M.; and Osterberg, C.,
“Self Compacting Fiber R.C. Composites for Bridge Columns,” Report No.
PEER 2011/106, Pacific Earthquake Engineering Research Center, Univer-
sity of California, Berkeley, Berkeley, CA, 2011.
20. Bedirhanoglu, I.; Ilki, A.; and Kumbasar, N., “Precast Fiber Rein-
forced Cementitious Composites for Seismic Retrofit of Deficient RC
Joints—A Pilot Study,” Engineering Structures, V. 52, 2013, pp. 192-206.
doi: 10.1016/j.engstruct.2013.02.020
21. Govindbhai, K. D.; Deb, S. K.; and Dutta, A., “Studies on Toughness
of Hybrid Fibre-Reinforced Cementitious Composite Beam,” Proceedings
of International Conference on Structural Engineering Construction and
Management, Kandy, Sri Lanka, Dec. 2011.
22. ASTM C1609/C1609M-12, “Standard Test Method for Flexural
Performance of Fiber-Reinforced Concrete (Using Beam with Third-Point
Loading),” ASTM International, West Conshohocken, PA, 2012, 9 pp.
23. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-02) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2002, 443 pp.
24. IS 13920:1993, “Ductile Detailing of Reinforced Concrete Structures
Subjected to Seismic Forces – Code of Practice,” Bureau of Indian Stan-
dards, New Delhi, India, 1993.
177
 25. ASTM C595/C595M-13, “Standard Specification for Blended Hydraulic
Cements,” ASTM International, West Conshohocken, PA, 2013, 13 pp.
26. BS 4449:2005, “Steel for the Reinforcement of Concrete – Weldable
Reinforcing Steel – Bar, Coil and Decoiled Product – Specification,” British
Standards Institution, London, UK, 2005.
27. Scott, R. H., and Beeby, A. W., “Long-Term Tension Stiffening
Effects in Concrete,” ACI Structural Journal, V. 102, No. 1, Jan.-Feb. 2005,
pp. 31-39.
28. Scott, R. H., and Whittle, R. T., “Moment Redistribution Effects in
Beams,” Magazine of Concrete Research, V. 57, No. 1, 2005, pp. 9-20. doi:
10.1680/macr.2005.57.1.9
29. Karayannis, C. G., and Sirkelis, G. M., “Strengthening and Rehabili-
tation of RC Beam-Column Joints Using Carbon-FRP Jacketing and Epoxy
Resin Injection,” Earthquake Engineering & Structural Dynamics, V. 37,
No. 5, 2008, pp. 769-790. doi: 10.1002/eqe.785
30. Karayannis, C. G., “Nonlinear Analysis and Tests of Steel-Fiber
Concrete Beams in Torsion,” Structural Engineering & Mechanics, V. 9,
No. 4, 2000, pp. 323-338. doi: 10.12989/sem.2000.9.4.323
178
31. Choudhury, A. M., “Study of Size Effect of RC Beam-Column
Joints with and without Retrofitting under Cyclic Loading,” PhD thesis, IIT
Guwahati, Guwahati, India, 2010.
32. Park, R., and Ang, A. H. S., “Mechanistic Seismic Damage Model for
Reinforced Concrete,” Journal of Structural Engineering, ASCE, V. 111,
No. 4, 1985, pp. 722-739. doi: 10.1061/(ASCE)0733-9445(1985)111:4(722)
33. Karayannis, C. G.; Chalioris, C. E.; and Sirkelis, G. M., “Local
Retrofit of Exterior RC Beam-Column Joints Using Thin RC Jackets—An
Experimental Study,” Journal of Earthquake Engineering and Structural
Dynamics, V. 37, No. 5, 2008, pp. 727-746. doi: 10.1002/eqe.783
34. Shannag, M. J., and Alhassan, M. A., “Seismic Upgrade of Interior
Beam-Column Subassemblages with High-Performance Fiber-Reinforced
Concrete Jackets,” ACI Structural Journal, V. 102, No. 1, Jan.-Feb. 2005,
pp. 131-138.
35. Scott, R. H., “Intrinsic Mechanisms in Reinforced Concrete Beam-
Column Connection Behavior,” ACI Structural Journal, V. 93, No. 3,
May-June 1996, pp. 336-346.
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S16
Behavior and Simplified Modeling of Mechanical
Reinforcing Bar Splices
by Zachary B. Haber, M. Saiid Saiidi, and David H. Sanders
Bridge seismic design codes do not allow mechanical reinforcing
bar splices in regions expected to undergo significant inelastic
deformations during earthquakes, thus severely limiting precast
and innovative bridge column construction that uses such splices.
The uniaxial behavior of two commercially available mechanical
splices under different loading conditions was investigated
experimentally in this study with emphasis on deformation
response. Tests were performed with static, dynamic, and cyclic
loading. The performance of the splices was satisfactory under
all loading conditions in that bar fracture occurred outside the
splice. Furthermore, the results revealed the effect of the relatively
high stiffness of mechanical couplers. The responses of individual
splices were used to interpret data from a series of cyclic tests on
half-scale bridge columns employing mechanical splices in plastic
hinge zones. Lastly, a simple method was proposed and validated
for modeling these devices in reinforced concrete members.
Keywords: accelerated bridge construction; acceptance criteria; coupler;
ductility; mechanical splice; repair; seismic; shape-memory alloy.
INTRODUCTION
Mechanical reinforcement splices have been used in cast-
in-place concrete construction when long, continuous bars
or reinforcement cages are required. Unlike lap splices,
which can require lengths greater than 30 bar diameters (db),
mechanical splices can be used to join bars at discrete locations.
Some of the mechanical reinforcing bar splices commercially
available in the United States1 are shown in Fig. 1. Bridge
and building design codes use acceptance criteria such
as International Code Council (ICC) AC1332 and ASTM
A1034/A1034M3 to quantify the ability of a splice to transfer
load, withstand load reversals, and resist slip. Furthermore,
some state departments of transportation (DOTs) have
developed their own acceptance criteria.4 After evaluation,
mechanical splices are given a performance classification
compatible with the corresponding code provision of interest,
which is used to restrict placement in a structural member
or limit stress/strain demands on spliced bars. In the United
States, there is one significant difference between bridge
and building code requirements for mechanical splices.
ACI 318-025 allows Type 2 mechanical splices, which must
be able to develop the full tensile strength of the spliced bars
to be placed at any location within a member regardless of
local inelastic demands. On the other hand, bridge design
codes such as the AASHTO Bridge Design Specifications6
and Caltrans Seismic Design Criteria (SDC)7 prohibit
all mechanical splices from being placed in plastic hinge
regions, which are subjected to high inelastic demands. Such
provisions have prevented the use of mechanical splices in
plastic hinges of bridge columns and have been a barrier to
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
newer and more innovative bridge columns in earthquake-
prone areas.
Although previous studies have used mechanical splices
in plastic hinge zones,8 there is little information as to the
deformation characteristics of mechanical splices and their
effects on local and global member behavior. Researchers
have studied the uniaxial behavior of mechanically spliced
bar assemblies, but these studies have focused primarily on
how strength is affected by fatigue loading,9 bar diameter,9
and blast loading rates.10 It was suggested by Haber et al.8
that the length of a splice is a critical factor that affects the
post-yielding flexural behavior of a member. That is, splices
with smaller LSp/db ratios (<4) are less likely to change the
plastic hinging mechanism, where LSp is the length of the
mechanical splice, and splices with larger ratios (>14) may
adversely affect hinge formation and behavior. The objective
of this study was to evaluate the deformation characteristics
of two commercially available mechanical splices under
static, dynamic, and cyclic loading. The correlation between
component- and system-level behaviors was addressed
by comparing uniaxial test results with a series of half-
scale bridge column test results conducted by the authors.
Lastly, experimental test data is used as a foundation for a
simple method to incorporate mechanical splices in member
deformation and capacity calculations.
RESEARCH SIGNIFICANCE
Previous studies have identified a number of applications
for mechanical reinforcing bar splices in plastic hinge
regions. However, there is little understanding as to the local
deformation response of these devices and how that response
can affect the global behavior of a ductile reinforced concrete
member. This paper provides much-needed data and
insight into the local deformation behavior of mechanical
reinforcing bar splices through a series of uniaxial tests.
Using test data, a simple method for analytical modeling of
mechanical splices in with concrete members is proposed
and validated with large-scale experimental test results.
EXPERIMENTAL INVESTIGATION
Specimen details
Two of the commercially available mechanical reinforcing
bar splices, shown in Fig. 1, were investigated in this study:
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-319.R3, doi: 10.14359/51687455, received June 11, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
179
Fig. 1—Mechanical reinforcing bar splice commercially available in the United States.
Fig. 2—Uniaxial test setup and instrumentation plans.
the upset headed coupler (HC) and grouted sleeve coupler
(GC). The HC splice consists of male and female threaded
steel collars that join bar segments with deformed heads,
which are created by heating the bar end and compressing
the heated end with specially designed hydraulic ram. The
force transferring mechanism for the HC splice consists of
compression being transmitted directly through deformed
heads and tension through the threaded collars. The GC
splice has been commonly used in conventional and precast
construction in East Asia11 and the United States.12 At the
precasting plant, reinforcing bars are inserted into the tapered
end of the sleeve and the device is then cast within the concrete
member. On site, the precast element is positioned such that
reinforcing bar dowels protruding from the adjacent member
enter the open sleeve ports. The connection is completed by
pumping a proprietary high-strength (>14 ksi [96.5 MPa])
cementitious grout into the sleeve. Force is transmitted
180
through formation of compression struts in the grout which
transfer force to the sleeve.
HC specimens were constructed using No. 8 (D25)
Grade 60 ASTM A706 bars having an average measured
yield stress, ultimate stress, and percent elongation at rupture
of 67.9 ksi (468 MPa), 95.1 ksi (655 MPa), and 18.2%,
respectively. Specimens were prepared with two 16 in.
(406 mm) headed bar segments joined using the threaded
steel collars described previously. The two collars were
initially tightened by hand followed by a pipe wrench to
the manufacturer’s minimum specified torque of 150 lbf-ft
(203 N-m).
GC specimens were constructed using No. 8 (D25)
Grade 60 ASTM A615 bars having an average measured
yield stress, ultimate stress, and percent elongation at
rupture of 66.8 ksi (460 MPa), 111.3 ksi (767 MPa), and
15.8%, respectively. To construct GC specimens, reinforcing
bars were first placed into the tapered end of the sleeves and
the assembly was tied to a support frame. The prepackaged
high-strength cementitious grout was mixed according to
manufacturer specifications and the sleeves were filled
approximately three-quarters full. The grout was rodded
with a smooth 0.25 in. (6.5 mm) diameter rod to ensure good
consolidation, and the second reinforcing bar was inserted
into the sleeve. The average 28-day compressive strength of
the grout was 15.7 ksi (108 MPa) according to ASTM C109/
C109M-02.
Test setup and loading protocols
All specimens were tested in the Large-Scale Structures
Laboratory (LSSL) at the University of Nevada, Reno
(UNR), using a servo-hydraulic loading frame. The test setup
was developed according to ASTM A1034/A1034M3 and
Caltrans Test Method CT6704 (Fig. 2). Strain was measured
directly from the reinforcing bars using foil-backed resistive
gauges (two on opposite sides per location), and over the
length of the splice using a digital extensometer. For GC test
specimens, a pair of strain gauges was also installed at the
midheight of the sleeve. The extensometer gauge lengths
over the coupler region LCR for HC and GC specimens were
6 and 18 in. (152 and 457 mm), respectively. The clear length
between load frame grips, LClear, was selected as the minimum
specified by CT670, which were 26.5 and 38 in. (673 and
965 mm) for the HC and GC specimens, respectively.
ACI Structural Journal/March-April 2015
Table 1—Summary of test results
Yield stress, ksi Ultimate stress, ksi Strain at rupture, %
Bar Coupler region
ID Average Standard deviation Average Standard deviation Average Standard deviation Average Standard deviation
HC control bar 67.9 3.61 95.1 1.61 18.2 3.08 — —
HC-S 67.2 1.09 95.2 2.71 16.4 2.09 7.70 0.37
HC-D 71.9 0.42 98.1 0.31 15.5 1.09 8.46 0.27
HC-C1 68.5 — 93.3 — 16.9 — 7.80 —
HC-C2 67.7 — 94.6 — 12.6 — 8.48 —
GC control bar 66.8 3.69 111.3 1.61 15.8 0.44 — —
GC-S 66.3 0.66 108.6 1.08 15.9 1.19 5.61 0.35
GC-D 70.4 — 110.8 1.00 16.2 3.61 5.53 0.28
GC-C1 66.1 — 98.7 — 5.59 — 2.69 —
Note: 1 ksi = 6.89 MPa.

Five different loading protocols were used to evaluate
the uniaxial behavior of the spliced bars. Three specimens
per splice type were tested for each protocol with exception
of the cyclic loading tests. The loading protocols and
associated nomenclature were: monotonic static loading
until failure (S), monotonic dynamic loading until failure
(D), slow cyclic loading until failure (C), single-cycle elastic
slip (SCS) loading, and multi-cyclic elastic slip (MCS)
loading. Specimens are identified by splice and loading
type, respectively. For example, a grouted splice specimen
tested under monotonic dynamic loading would be identified
as “GC-D.”
Loading was displacement-controlled for monotonic
static and dynamic tests. The loading rates for monotonic
static tests were determined according to ASTM A370.13
For HC-S specimens, pre- and post-yield displacement
rates were 0.00625 and 0.05 in./s (0.159 and 1.27 mm/s,
respectively. For GC-S specimens, pre- and post-yield
displacement rates were 0.01875 and 0.15 in./s (0.476 and
3.81 mm/s), respectively. The dynamic loading protocol
was selected to subject specimens to strain rates in the range
of those imposed by an earthquake event.14 A target rate
of 0.07 in./in./s was selected knowing that achieved rates
would be approximately 80 to 120% of the target.15 The
corresponding displacement rates, which are based on LClear,
for HC-D and GC-D specimens were 1.575 and 1.75 in./s
(40 and 44.5 mm/s), respectively.
The effect of tension-compression load reversals was
studied by applying cyclic loads. Although the widely used
ICC AC332 test criteria requires cyclic testing, spliced bars
are only subjected to reversals up to five times the specified
yield strain of the bar (5ey). Testing in this study subjected
splices to load reversals beyond this level and was continued
until failure. Cyclic tests were conducted in load control
mode at rates of 1 kip/s (4.45 kN/s) during tensile loading and
at 0.5 kip/s (2.22 kN/s) during compression loading, which
correspond to stress rates of 1.27 and 0.635 ksi/s (8.72 and
4.36 MPa) for tension and compression, respectively. Each
cycle of loading consisted of a single tensile and compression
cycle. For each cycle, the peak tensile load was increased by
increments of 0.2fy from 0.5fy to 1.1fy followed by increments
of 0.1fy thereafter. After the target tension was reached, the
load was reversed until the specimen reached a compression
stress of 20.7 MPa (10.5 kN). Both specimen types had long
unsupported lengths. Thus, a low compression stress target
was selected to prevent buckling.
Elastic slip tests were conducted in accordance with
Caltrans and AASHTO methods, which are used to determine
the permanent relative deformation between the reinforcing
bar and mechanical splice. In single cycle slip tests (SCS),
samples were loaded to an initial stress of 3 ksi (20.7 MPa)
and the elongation over the sample measurement gauge
length (DInitial) was measured. Samples were then stressed
to 30 ksi (207 MPa), held for 30 seconds, and subsequently
destressed to 3 ksi (20.7 MPa). Upon distressing, a final
elongation measurement (DFinal) was recorded. The resulting
slip, DSlip, is defined as the difference between final and initial
elongation measurements. After completing the single cycle
test, samples were subjects to three to five additional cycles
to determine if slip increased with additional loading. This
sequence is referred to as the multi-cycle slip test (MCS).
The maximum permitted elastic slip for splices with No. 8
(D25) bars according to Caltrans and AASHTO are 0.028
and 0.01 in. (0.71 and 0.25 mm), respectively.
EXPERIMENTAL RESULTS
A summary of test results is provided in Table 1, along with
the measured material properties for unspliced reinforcing
bars. On the day of testing, the average measured grout
strength for GC specimens was 18.5 ksi (128 MPa).
Monotonic static tests (S)
As would be expected, the average elongation over LCR
(otherwise referred to as the coupler region) was reduced
due to the presence of the threaded steel collars joining the
deformed heads. The average elongation at failure over
the coupler region was 7.70%, which was 53% less than
that of the reinforcing bar. Figure 3 shows representative
constitutive relationships for HC-S tests. The stress-strain
curve for the coupler region exhibited a stiff initial slope
up to approximately 10 ksi (69 MPa), which subsequently
softened and remained linear up to yielding of steel. Softening
occurs as the precompressive force on the deformed heads,
which is a result of the initial torque on the threaded collars,

ACI Structural Journal/March-April 2015 181


Fig. 3—Stress-strain curves from monotonic tests on
HC device (static tests are solid lines; dynamic tests are
dashed lines).
Fig. 4—Stress-strain curves from monotonic tests on
GC device (static tests are solid lines; dynamic tests are
dashed lines.)
Fig. 5—Evidence of strain penetration into grouted sleeve.
is overcome and the heads separate. Head separation is
only evident up to yielding of steel. However, the heads
continue to separate afterward, but the deformation over
the coupler region is controlled by nonlinear deformation
of the reinforcing bars. Each HC-S specimen failed due to
182
Table 2—Average measured strain rates during
dynamic tests (unit: strain/s)
HC-D GC-D
Standard Standard
Stress range Average deviation Average deviation
0 to yield 0.0175 0.0121 0.0781 0.0073
Yield to ultimate 0.0908 0.0178 0.0924 0.0041
Ultimate to failure 0.0633 0.0200 0.1060 0.0770
ductile reinforcing bar rupture, which occurred away from
and without damage to the splice.
In GC-S tests, the response of the spliced reinforcing
bars was similar to that of the control bars (Fig. 4). The
deformation capacity over the coupler region was reduced
by 65% compared with strain measurements taken from
spliced reinforcing bars due to the presence of the grout-
filled cast-iron sleeve. Unlike the coupler region response in
HC-S specimens, the initial branch of the stress-strain curve
for GC-S specimens was similar to that of the reinforcing bar.
This indicates that the elastic stiffness of the GC assembly is
similar to that of mild steel bars. Similar to the HC-S tests, each
GC-S specimen failed by reinforcing bar rupture away from
and without damage to the splice. Strain measurements from
the midheight of the sleeves indicated that the average strain
in the sleeve at failure was 0.7%, which was three times the
average measured yield strain of steel. Strain measurements
from the sleeve also indicated that the sleeves undergo
nonlinear deformations once the stress in the reinforcing bars
reach approximately 70 ksi (482 MPa).
Axially loaded reinforcing bars that are well-anchored in
cementitious materials undergo localized deformations from
the anchoring material as a result of strain penetration,16
which is typically referred to as “bond-slip.” Although
not explicitly measured, Fig. 5 shows a grout cone failure
surface indicating that strain penetration occurs within the
grouted coupler assembly during loading. It was shown
by Haber et al.17 that bond slip can account for up to 40%
of the deformation over the coupler region (LCR). None of
the GC tests conducted in this study exhibited bar pullout
failure, but other studies have shown this can occur.10 Such a
failure mode would be caused by insufficient grout strength
or improper installation.
Monotonic dynamic tests (D)
Representative stress-strain curves from monotonic
dynamic loading are shown along with the static curves in
Fig. 3 and 4 for HC-D and GC-D tests, respectively. The
average measured strain-rates are listed in Table 2 according
to stress range. Previous tests on mild steel reinforcing bars
loaded at strain-rates similar to those shown in Table 2 have
exhibited increased yield and ultimate stresses by as much
as 30%18; Zadeh and Saiidi15 provide detailed discussion
regarding the behavior of axially loaded reinforcing bars
under high strain-rate loading. Thus, it is not unexpected
that in both HC-D and GC-D tests that the yield and ultimate
stresses were slightly larger than corresponding static tests.
Slight variations among the initial slopes of stress-strain
curves can be observed between static and dynamic tests,
ACI Structural Journal/March-April 2015
Fig. 6—Cyclic test results: HC results (a) through (c); GC results (d) through (f).
which are expected due to differences in clamping forces at
the grips. Previous studies have shown that dynamic loading
does not have a significant effect on the elastic modulus of
mild steel reinforcing bars.19
In HC-D tests, the average yield and ultimate stresses
increased 6% and 3%, respectively, compared with HC-S test
results. In GC-D tests, the average yield and ultimate stresses
increased 6% and 2%, respectively, compared with GC-S
test results. Similar to the static tests, both HC-D and GC-D
specimens exhibited reduced elongation over the coupler
region compared with measurements from the spliced
reinforcing bars. The average elongation at failure over the
coupler region was 45% and 66% lower in HC-D and GC-D
specimens, respectively. All HC-D and GC-D specimens
failed due to ductile reinforcing bar fracture away from and
without damage to the splices. This indicates that increased
yield and ultimate stresses caused by dynamic loading were
sustained by both splice types without an adverse effect on
the failure mode. Similar the GC-S tests, GC-D specimens
were inspected after testing and evidence of strain penetration
into the grouted sleeves was found in all specimens. Lastly,
dynamic loading did not affect the stress-strain curves in the
coupler regions of HC-D and GC-D specimens.
Slip tests
The maximum slip recorded for HC-SCS and GC-SCS
samples were 0.007 and 0.0175 in. (0.178 and 0.044 mm),
respectively. The multi-cycle slip tests for each splice
type did not indicate cumulative slippage with application
of three or more cycles. Both HC and GC splices passed
single- and multi-cycles slip tests according to both Caltrans
and AASHTO maximum slip criteria, which are 0.028 and
0.01 in. (0.71 and 0.25 mm), respectively.
ACI Structural Journal/March-April 2015
Cyclic loading tests
Two HC-C specimens were tested, one with the
manufacturer’s minimum specified torque applied to the
threaded collars, denoted as “HC-C1,” and a second with
collars hand-tight, denoted as “HC-C2.” The measured yield
and ultimate stresses of both specimens were within 2% of
those tested under monotonic static loading. Furthermore,
the elongation over the coupler region was also comparable
with static tests. The cyclic stress-strain response for the
coupler region and the bar assembly of HC-C2 is shown in
Fig. 6(a) and (b), respectively, which indicate that the stress-
strain backbones for both the coupler region and reinforcing
bar are nearly identical to those from monotonic static tests.
Both HC-C specimens failed due to ductile reinforcing bar
rupture away from and without damage to the splice.
Once each peak stress level was reached, the load was
reversed to a target compressive stress of –3 ksi (–20.7 MPa).
During unloading, the slope of the stress-strain curves for the
coupler region and the reinforcing bar were approximately the
same, indicating the reinforcing bars control the unloading
stiffness of the device. Once the load in the bar approached
zero, a distinct, instantaneous deformation occurred. It was
hypothesized that separation of the deformed heads (otherwise
referred to as gap opening) within the steel collars occurred
once precompression from the applied torque was overcome.
Cyclic loading confirms this behavior and a relationship
between peak stress and gap length can be established. The
gap length, Dgap, was defined as the deformation during the
transition between tensile and compressive force within the bar.
There was an approximately linear relationship between the
peak stress in the bar and the gap length between the deformed
heads. The peak stress versus gap length plot (Fig. 6(c)) also
indicates elastic slip limits allowed by Caltrans and AASHTO.
It can be observed that these limits are significantly exceeded
even before yielding of the reinforcing bar. Further discussion
of this behavior is provided in subsequent sections.
183
 affect the characteristic behavior or failure modes of the
splice assembly. However, higher strain rates, such as those
expected in a blast, reduce the ultimate strength and ductility
of mechanically spliced bars.14
The SR plot indicates that the deformation response of
some splices could be as little as 25% (SR = 0.25) of that
of the reinforcing bar throughout the loading history, which
can be observed in test data from the GC device. The HC
splice did not exhibit as large of stiffness increase as the GC
splice. However, the gap-opening behavior is clearly visible
in Fig. 7, which allows the HC splice to deform significantly
compared to the reinforcing bars over a short part of the
loading history. Prior to a strain of 0.02, the majority of
deformation occurred within the coupler region, which is
indicated by a slope exceeding 1:1 (SR = 1) (Fig. 7).

CORRELATION BETWEEN UNIAXIAL AND

BRIDGE COLUMN TESTS
A series of half-scale reinforced concrete bridge column
Fig. 7—Relationships between strain in reinforcing bar and tests were conducted by the authors to investigate the
strain over splice. seismic performance of precast column-footing joints
with mechanical reinforcing bar splices.8 Two connection
One GC specimen (GC-C1) was tested under cyclic
configurations were studied: one employing two layers of HC
loading. The average yield stress was found to be
splices, referred to as “HCNP,” and the second using single-
comparable to static results, but the average ultimate stress
layer GC splices, referred to as “GCNP.” The reinforcement
and elongation of the reinforcing bar at rupture were 9% and
details and model geometry were designed assuming the
65% lower than static tests, respectively (Fig. 6(d) through
behavior would be similar to a conventional cast-in-place
(f)). The discrepancies between static and cyclic tests on
(CIP) benchmark column; this is otherwise referred to
the GC device were not caused by premature failure of
as “emulative” design. That is, a conventional column
the splice. In both cases failure was a result of reinforcing
model was designed for a target displacement ductility
bar rupture away from the splice. However, in GC-C1, the
capacity of µC = 7.0 (ultimate displacement/effective yield
reinforcing bar did not exhibit ductile behavior—that is,
displacement) according to Caltrans SDC, and the details
necking of the bar at the point of fracture did not occur rather
of the plastic hinge region were modified to incorporate the
the fracture surface was flat. This may have been caused by
mechanically spliced connections. The column models had
a low-cycle fatigue-type effect, but such failures typically
24 in. (610 mm) diameter cross sections, an aspect ratio of
require significantly larger strain reversals.20 There was also
4.5, and longitudinal and transverse reinforcement ratios
a slight difference between the static and cyclic stress-strain
of 1.9% and 1.0%, respectively. The ratio of axial load to
backbone curves for the coupler region. The slope of the
the product of the gross column cross-sectional area and
unloading curves for both the coupler region and reinforcing
the specified concrete compressive strength was 0.1. Along
bar were similar indicating the reinforcing bars control
with the precast models, a CIP benchmark model was also
the unloading stiffness. Unlike HC-C tests, there was not
constructed and tested for comparison. Each column was
a visible instantaneous deformation during the transition
tested in a cantilever configuration and subjected to slow
between tensile and compressive loads. The average stress-
cyclic loading at increasing drift levels. Figure 8(a) shows
strain history from the midheight of the cast-iron sleeve
the pertinent plastic hinge connection details of these three
indicates low strains and slight nonlinearity.
half-scale column models. A detailed discussion of these
tests can be found in Haber et al.8
Comparison and discussion
Figures 8(b) and (c) show the moment-rotation
It was observed that the presence of the mechanical splice
relationships for columns with HC and GC splices,
reduced the deformation capacity over the coupler regions for
respectively, measured near the base of the column (shown
both devices. Figure 7 presents representative relationships
in Fig. 8(a)). The data, which is only shown up to 5% due to
between the strain over LCR and the estimated strain over
localized bar buckling thereafter, captures the influence of
LSp in terms of the average reinforcing bar strain obtained
the mechanical splices on the hinge rotations. The effect of
from strain gauge measurements. The slope of the curves is
gap closure can be clearly observed in the moment-rotation
referred to as the splice-bar strain ratio (SR) and indicates
response of HCNP in the form of a slight pinch in the
the relative stiffness of the splice to the reinforcing bar. As
unloading branches. The pinch, however, is small and stable
mentioned previously, there was little difference between the
and does not lengthen as deformations increase. Except for
static and dynamic behavior of the coupler regions, which
the pinch, the loops are similar in shape and magnitude to
can be seen in Fig. 7. This indicates that strain rates similar
those of the CIP column. This is not surprising given the
to those expected from an earthquake would not adversely
relatively short length of the HC device. On the other hand,

184 ACI Structural Journal/March-April 2015

Fig. 8—Behavior of RC members with mechanical splices: (a) general column details; (b) moment-rotation relationships for
HCNP; (c) moment-rotation relations for GCNP; and (d) force-drift behavior.
the moment-rotation behavior of the GC column section
differs significantly from that of CIP. The maximum rotation
in the GC section is approximately one-third the maximum
rotation of CIP, and very little plastic rotation is achieved.
This is consistent with observations from uniaxial tests,
which indicated that the deformation capacity of the GC
splice could be as little as 25% of the reinforcing bar.
Although individual characteristics of the mechanical
splices can be observed locally within each column,
the global force-displacement relationships are not as
significantly affected in this case. However, the influence
of the mechanical splices could become more apparent
in the global force-deformation response with different
column geometries and/or reinforcement details. As shown
in Fig. 8(d), the hysteresis loops for three columns are
comparable. GCNP had slightly higher peak loads after
2% drift due presence of GC splices. HCNP also exhibited
slightly higher peak load, but was a result of cementitious
grout present within the hinge zone that was approximately
twice as strong as the concrete for CIP. Furthermore,
the slight pinch observed at the local level can be seen in
global response, but is not significant. It should be noted
that although force-displacement relationships were similar
among the three columns, the presence of the GC splices in
GCNP ultimately shifted plastic hinging to the footing and
above the couplers. This shifted hinge mechanism eventually
caused longitudinal bars to rupture in the footing at 6% drift
due to strain concentrations and numerous load reversals. As
a result, the drift capacity of GCNP was significantly less
than the drift at failure in CIP and HCNP, which was 10%.
Typically, designers do not account for the presence
of mechanical splices in design calculations for ductile
reinforced concrete members. This is primarily because
splices are not typically placed in locations expected to
undergo significant nonlinear deformations. Other reasons
may include that designers assume splice behavior is
approximately the same as the spliced reinforcing bar or that
ACI Structural Journal/March-April 2015
a proven method to model these devices does not exist. Data
provided in this paper indicate that the deformation response
of mechanical splices indeed is different from that of the
reinforcing bar. Furthermore, it was evident that the presence
of the GC splices slightly increased the lateral load capacity
of GCNP and affected the local moment-rotation behavior.
Similar effects could occur using other devices with large
LSp:bd ratios. Thus, modeling techniques are required for
incorporating these devices in analysis procedures. Haber
et al.21 described a method for modeling columns with GC
splices, but required detailed methods and specific material
and geometric properties of the mechanical splice. The
following sections provide a simple method for modeling
mechanical splices and validation with the previously
described column test data.
SIMPLIFIED MODELING OF MECHANICAL
SPLICES IN BRIDGE COLUMNS
Proposed method
It is proposed that mechanical splices can be modeled in
a simplified manner by defining an effective uniaxial stress-
strain relationship for spliced longitudinal reinforcement that
can be employed in fiber section analysis. This model may
be used in most commercially available structural analysis
programs that have moment curvature, lumped-plasticity
frame element, or distributed plasticity frame element
analysis capabilities. Distinct points on the effective stress-
strain curve can be calibrated to match test results in specific
cases or may be based on the relative stiffness between the
splice assembly and the reinforcing bar in a more generalized
approach. The strain component of the effective stress-strain
relation is defined by strains occurring over LSp, which can
be determined from measurements over LCR. This method is
advantageous because it does not require direct knowledge
of material and geometric properties of mechanical splice
assembly, which may be proprietary. That is, the intrinsic
deformation characteristics of the device are captured using
185
Fig. 9—Stress-strain model for reinforcing steel and
proposed splice model.
standard evaluation and acceptance criteria tensile test data.
With respect to the GC device, these include influence of
the cast-iron sleeve on splice stiffness, deformation caused
by unsupported reinforcing bar length, and deformation
resulting from bond slip within the sleeve.
It was observed in Fig. 7 that the GC device exhibited
an approximately linear SR relationship, where the strain
over LSp was approximately 0.25 times that in the bar. Using
this information, an effective stress-strain relationship can
be created by scaling a constitutive stress-strain model for
reinforcing steel to produce a similar SR relationship between
the two stress-strain models. The effective stress-strain
model can be applied in an individual nonlinear uniaxial fiber
with the same cross-sectional area as the spliced reinforcing
bars. Figure 9 presents a typical stress-strain curve for mild
steel reinforcing bars (solid line), which has an initial linear-
elastic branch with slope Es, a yield plateau beginning at the
yield strain ey, a nonlinear strain-hardening branch with a
slope Esh beginning at esh, and a plateau at the ultimate stress
fu and corresponding strain eu. The stress-strain model for
the splice, otherwise referred to as the “Proposed Model”,
is defined using the same constitutive model as the mild
reinforcing steel, but the characteristic parameters are
scaled to achieve the desired SR response. The proposed
curve representing the spliced region LSp is shown with a
dashed line and its parameters are identified with a “*”. The
parameters for the proposed curve are defined according to
Eq. (1) and (2), which correspond to the elastic and plastic
portions of the stress-strain curve, respectively. Once tensile
testing has been performed, and the SR relationship has been
identified, the user can select appropriate elastic SRE and
plastic SRI strain ratios for scaling the stress-strain curve for
the splice model. Equation (3) must be used if the reinforcing
steel model includes a yield plateau to prevent calculation
errors as the stress state approaches yield.
e*y
Es
= = SRE (1)
ey Es*
186
Fig. 10—Analytical model for bridge column with grouted
sleeve column-footing connection.
e*u e*sh Esh
= = = SRI (2)
e u e sh Esh*
SRI ey
≥ (3)
SRE e sh
Validation with half-scale column test results
The proposed model was validated using column model
test results discussed previously for GCNP. Figure 10 shows
the general details of the half-scale test model along with
the key components of the analytical model. The analytical
model was developed in OpenSEES using force-based
distributed plasticity frame-elements with fiber sections at
each integration point. Because of the importance of bond-slip
deformations, the method described by Haber et al.17 was used
to incorporate the influence of the GC splice by including a
rotational spring at the column-footing interface. Two different
fiber-section assignments were used. Section “S1” was used in
the frame element representing the shaft of the column above
the region with GC splices, and Section “S2” was used in the
region with the GC splices. For both sections, constitutive
relationships for unconfined and confined concrete, and
reinforcing bars were defined using available models from
OpenSEES; namely “Concrete01,” “Concrete04,” and
“ReinforcingSteel,” respectively. Each constitutive model
was calibrated using average measured materials properties.
The measured compressive strength of concrete on the day of
test was 4.7 ksi (32 MPa), and the average yield and ultimate
stress of longitudinal steel were 67 and 111 ksi (461 and
765 MPa), respectively. Confined concrete properties were
determined according to Mander’s model.21
GC splices were defined in S2 as single fibers with the same
cross-sectional area as the No. 8 reinforcing bars. The actual
diameter of GC device was approximately 2db – 2.5db, which
corresponds to an area footprint of 4 to 6.25 times the bar
area. Therefore, at each splice location, an equal-area segment
of confined concrete fibers were removed to account for the
ACI Structural Journal/March-April 2015
Fig. 11—Comparison of measured and calculated SR
relationships.
presence of the splice. The constitutive relationship for the GC
fibers was based on the previously described method of scaling
the reinforcing steel stress-strain properties. For this study,
SRE and SRI were selected to be 1.0 and 0.26, respectively.
Strain data from the GCNP column test indicated that yielding
of longitudinal reinforcement first occurred at the column-
footing interface. In this model, the fibers at the column-
footing interface employ effective properties, and therefore
strains in the reinforcing bars cannot be calculated explicitly.
However, by setting SRE = 1.0, the first yield of longitudinal
steel can be approximated. This is a reasonable selection
because GC-S tests indicated that the strain over the splice
was approximately 75% of that from the reinforcing bar prior
to yielding of steel. The ability to reasonably approximate the
yield displacement of the column is critical for displacement-
based designed methods, which require displacement ductility
calculations. Figure 11 shows a comparison between the
measured SR response and that calculated using the proposed
method with SRE = 1.0 and SRI = 0.26.
Comparison with experimental results and
discussion
The frame element model shown in Fig. 12 was subjected
to monotonic lateral displacement twice, once in each
direction because the column reinforcing pattern was not
symmetric. Pushover analysis was conducted with and
without the presence of bond slip due to strain penetration
into the footing. The resulting pushover curves were
compared with the measured hysteretic response. The
calculated initial stiffness with bond-slip included was
approximately the same as the measured stiffness. When
bond slip was excluded, the calculated initial stiffness was
slightly higher than the measured result, which is to be
expected. Nonetheless, the calculated and measured results
were still comparable and the analytical model would be
adequate for design purposes. Circular markers identify the
measured and calculated first yield points of the longitudinal
steel. In the positive direction, there was very little
difference between the measured and calculated results. In
the negative direction, the results were comparable with a
6% and 17% difference in the measured and calculated yield
ACI Structural Journal/March-April 2015
Fig. 12—Comparison between measured and calculated
force-displacement response.
displacement and lateral load, respectively. After 1.0% drift,
the measured data indicated initiation of strain hardening.
There was very good correlation between the post-yielding
parts of the calculated backbone curves and the measured
curve. The calculated response including bond slip deviated
slightly from the measured curved after 3.0% with the
maximum difference between the measured and calculated
lateral loads being 5.5% at 6% drift in the negative direction.
When bond slip was excluded, the maximum difference was
8.6%, which is still reasonable.
In general, the calculated force-displacement response
of the column showed good correlation with the measured
result using an effective constitutive relationship for GC
splices. In this study, a force-based distributed-plasticity
frame element model was used to validate the effective
material method for incorporating mechanical reinforcing
bar splices within a ductile member. However, the proposed
modeling method could also be used in moment curvature or
fiber section lumped plasticity analysis and can be applied to
other constitutive relationships for reinforcing steel using an
approach similar to that described in this paper.
SUMMARY AND CONCLUSIONS
Two commercially available mechanical reinforcing bar
splices, namely an upset headed (HC) splice and a grout-filled
ductile cast-iron sleeve (GC) splice, where evaluated under
uniaxial monotonic static and dynamic loading until failure,
slow reversed cyclic loading until failure, and a series of
elastic slip tests. Analysis of results focused on characterizing
the force-deformation characteristics of each device. Key
observations from uniaxial tests were then correlated with
a series of half-scale bridge column models employing
mechanical splices in flexural plastic hinge zones. Local
moment-rotation and global force-displacement relationships
were presented for columns HCNP and GCNP, containing
mechanical splices and for a conventional cast-in-place
column. Based on observations from uniaxial tests, a simple
method for determining an effective stress-strain model was
proposed for mechanical splices. As an example, effective
uniaxial properties were established for the GC splice
187
using the measured coupler behavior. Effective properties
were implemented in a nonlinear frame-element model for
GCNP, and pushover analysis results were compared with the
measured force-displacement response. Based on the results
of this study, the following conclusions can be made:
Mechanical splices can significantly reduce the
deformation capacity of spliced reinforcing bars by as much
as 75%, which can have a noticeable effect on the local
moment-rotation behavior of a ductile reinforced concrete
member depending on the size and stiffness of the splice
relative to the reinforcing bar.
The characteristic stress-strain behavior for HC and GC
splices is not significantly affected by strain rates expected
during earthquakes. These devices were able to sustain the
higher ultimate stress demands associated with the strain
rate effect on mild steel reinforcing bars without adverse
effect on failure modes.
The gap formation between the heads of the HC splice,
which significantly exceeded the elastic slip limits for
AASHTO and Caltrans bridge design codes, does not have
an adverse effect on the force-displacement behavior of
members containing these splices in plastic hinge zones.
The procedure proposed for determining the effective
stress-strain properties for a mechanical splice can be
employed using test results from standard test methods
and acceptance criteria. Thus, proprietary geometric
specifications and material properties for a mechanical
splice are not required.
The calculated member response using the proposed
model showed very good correlation with the measured
test results with and without incorporation of bond-slip
deformation. The calculate column load and displacement at
first yield of longitudinal bars also were in close agreement
with the measured data.
The proposed effective strain-strain model for mechanical
splices can be easily implemented in available software
packages.
AUTHOR BIOS
ACI member Zachary B. Haber is a Bridge Research Engineer with
Professional Service Industries (PSI) at the Federal Highway Administration
Turner-Fairbank Highway Research Center in McLean, VA. He received
his BS and MS in civil engineering from the University of Central Florida,
Orlando, FL, and received his PhD in civil engineering from the University
of Nevada, Reno, Reno, NV. His research interests include large-scale
testing, advanced materials in civil engineering, and bridge engineering.
M. Saiid Saiidi, FACI, is a Professor of civil and environmental engineering
and the Co-Director of USDOT University Transportation Center on
Accelerated Bridge Construction-Seismic at the University of Nevada, Reno.
He is the Founding and former Chair and a current member of ACI Committee
341, Earthquake-Resistant Concrete Bridges, and a member of Joint
ACI-ASCE Committee 352, Joints and Connections in Monolithic Concrete
Structures. He is also a member of ACI Subcommittee 318-D, Subcommittee
on Flexure and Axial Loads (Structural Concrete Building Code).
David H. Sanders, FACI, is a Professor at the University of Nevada, Reno. He
received his BS from Iowa State University, Ames, IA, and his MS and PhD from
the University of Texas at Austin, Austin, TX. He a member of the ACI Board
of Direction and former Chair of the ACI Technical Activities Committee; ACI
Committee 341, Earthquake Resistant Concrete Bridges; and Joint ACI-ASCE
Committee 445, Shear and Torsion. His research interests include concrete
structures with an emphasis on seismic performance of bridges.
188
ACKNOWLEDGMENTS
The research presented in this document was funded by the California
Department of Transportation (Caltrans) under contracts No. 65A0372
and 65A0425. The support and advice of M. Mahan, R. Bromenschenkel,
M. Keever, and T. Ostrom of Caltrans are appreciated. The authors would like
to thank Headed Reinforcement Corp. (HRC), Splice Sleeve Japan, and Splice
Sleeve North America for donating splices and bars used in this study. Special
thanks are expressed to R. Nelson and C. Lyttle for their help with testing.
REFERENCES
1. ACI Committee 439, “Types of Mechanical Splices for Reinforcing
Bars (ACI 439.3R-07),” American Concrete Institute, Farmington Hills,
MI, 2007, 20 pp.
2. ICC-ES AC133, “Acceptance Criteria for Mechanical Connector
Systems for Steel Reinforcing Bars,” International Code Council Evalua-
tion Service, Whittier, CA, 2010, 9 pp.
3. ASTM A1034/A1034M-10, “Standard Test Methods for Testing
Mechanical Splices for Steel Reinforcing Bars,” ASTM International, West
Conshohocken, PA, 2010, 5 pp.
4. CT670, “Method of Tests for Mechanical and Welded Reinforcing
Steel Splices,” California Department of Transportation, Division of Engi-
neering Services, Sacramento, CA, 2011, 11 pp.
5. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-02) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2002, 443 pp.
6. American Assiciation of State Highway and Transportation Officials
(AASHTO), “AASHTO Guide Specifications for LRFD Seismic Bridge
Design,” second edition, Washington, DC, 2011, 301 pp.
7. California Department of Transportation, “Seismic Design Criteria
(SDC) Version 1.7,” Division of Engineering Services, Sacramento, CA,
2013, 180 pp.
8. Haber, Z. B.; Saiidi, M.; and Sanders, D. H., “Seismic Performance of
Precast Columns with Mechanically Spliced Column-Footing Connection,”
ACI Structural Journal, Vol. 111, No. 3, May-June 2014, pp. 639-650.
9. Paulson, C., and Hanson, J. M., “Fatigue Behavior of Welded and
Mechanical Splices in Reinforcing Steel,” NCHRP Report 10-35, Dec.
1991, 158 pp.
10. Rowell, S. P.; Grey, C. E.; Woodson, S. C.; and Hager, K. P., “High
Strain-Rate Testing of Mechanical Couplers,” Report ERDC TR-09-8, U.S.
Army Corps of Engineers, Washington, DC, Sept. 2009, 74 pp.
11. Aida, H.; Tanimura, Y.; Tadokoro, T.; and Takimoto, K., “Cyclic
Loading Experiment of Precast Columns of Railway Rigid-Frame Viaduct
Installed with NMB Splice Sleeves,” Proceedings of the Japan Concrete
Institute, V. 27, No. 2, 2005, pp. 613-618.
12. Culmo, M. P., “Connection Details for Prefabricated Bridge Elements
and Systems,” Report FHWA-IF-09-010, Federal Highway Administration,
Washington, DC, Mar. 2009, 568 pp.
13. ASTM A370-03a, “Standard Test Methods and Definitions for
Mechanical Testing of Steel Products,” ASTM International, West Consho-
hocken, PA, 2003, 49 pp.
14. Motaref, S.; Saiidi, M. S.; and Sanders, D. H., “Seismic Response
of Precast Bridge Columns with Energy Dissipating Joints,” Report No.
CCEER-11-01, Center for Civil Engineering Earthquake Research, Depart-
ment of Civil Engineering. University of Nevada, Reno, Reno, NV, 2011.
15. Sadrossadat-Zadeh, M., and Saiid Saiidi, M., “Effect of Strain Rate
on Stress-Strain Properties and Yield Propagation in Steel Reinforcing
Bars,” Report No. CCEER-07-02, Center for Civil Engineering Earthquake
Research, Department of Civil Engineering. University of Nevada, Reno,
Reno, NV, 2007.
16. Otani, S., and Sozen, M. A., “Behavior of Multistory Reinforced
Concrete Frames during Earthquakes,” Structural Research Series No. 392,
University of Illinois, Urbana, IL, 1972, 551 pp.
17. Haber, Z. B.; Saiidi, M.; Ou, Y. C.; and Sanders, D. H., “A Method for
Calculating the Seismic Response of Bridge Columns with Grouted Sleeve
Column-Footing Connections,” Proceedings, Seventh National Seismic
Conference on Bridges & Highways, Oakland, CA., May 20-22, 2013.
18. Malvar, L. J, “Review of Static and Dynamic Properties of Steel
Reinforcing Bars” ACI Materials Journal, V. 95, No. 5, Sept.-Oct.1998,
pp. 609-614.
19. Fu, H. C.; Erki, M. A.; and Seckin, M., “Review of Effects of Loading
Rate on Reinforced Concrete,” Journal of Structural Engineering, ASCE,
V. 117, No. 12, Dec. 1991, pp. 3660-3679.
20. Mander, J. B.; Panthaki, F. D.; Kasalanati, A. “Low-Cycle Fatigue
Behavior of Reinforcing Steel,” Journal of Materials in Civil Engineering,
ASCE, V. 6, No. 4, 1994, pp. 453-468.
21. Mander, J. B.; Priestley, M. J. N.; and Park, R., “Theoretical Stress-
Strain Model for Confined Concrete,” Journal of Structural Engineering,
ASCE, V. 114, No. 8, 1988, pp. 1808-1826.
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S17
Bond-Splitting Strength of Reinforced Strain-Hardening
Cement Composite Elements with Small Bar Spacing
by Toshiyuki Kanakubo and Hiroshi Hosoya
Strain-hardening cement composites (SHCCs) show excellent
mechanical behavior that is characterized by tensile strain hard-
ening and multiple fine cracks. A suitable application of SHCC
for bond improvement involves reducing the cover thickness and
bar spacing of the main bars. To investigate the bond behavior of
reinforced SHCC elements and to propose a predicting method
for bond strength, the pullout bond test and beam bond test are
conducted in which small cover thickness and bar spacing are used.
The results of the pullout bond test show that the bond strength
of SHCC is higher than that of conventional concrete, which is
expected by the “partly cracked elastic stage” of the cylindrical
model by Tepfers. It is considered that the tensile stress distribution
of SHCC surrounding main bar corresponds to the “plastic stage.”
The results of the beam bond test also show that the bond strength
of SHCC has a higher value. A prediction methodology is proposed
as the summation of the bond strength exhibited by SHCC and the
confinement of lateral reinforcement.
Keywords: beam bond test; bond splitting; confinement effect; cylinder
model; pullout bond test; tensile strength.
INTRODUCTION
Strain-hardening cement composites (SHCCs), which is
grouped into similar composites such as high-performance
fiber-reinforced cement composites (HPFRCCs) and engi-
neered cementitious composites (ECCs), show excellent
mechanical behavior characterized by tensile strain hard-
ening and multiple fine cracks.1 Examples of practical
applications of SHCC have been reported in the literature.2
SHCC (ECC) is applied in the coupling beams of center core
systems used in high-rise reinforced concrete (RC) buildings.
The coupling beams are designed in compliance with the
following two requirements: 1) no substantial load degrada-
tion at a translational angle as high as 4%; and 2) no cracks
influencing durability with a width greater than 0.3  mm
(0.012 in.) after an earthquake. It is difficult for conven-
tional RC beams to keep the crack opening under 0.3 mm
(0.012 in.) after the elements have deformed at an angle
of 4%. The finely distributed cracking behavior of SHCC
has also been exploited to use SHCC for surface repair of
concrete dams, water channels, and retaining walls.3
The advantages of using SHCC lie in the appropriate
use of its tensile property. The flexural performance of
structural elements can directly be improved by the strain-
hardening and multiple-cracking behavior of SHCC. In the
case of shear elements such as coupling beams and shear
walls, the bridging effect of the fiber in SHCC can transmit
shear stress through multiple cracks. This paper focuses
on improvement of the bond behavior, especially for the
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
steel-reinforced SHCC elements that are associated with
splitting of cover matrix. The use of high-strength materials
causes the increment of reinforcement ratio and increases the
transmission of stress from the reinforcement to the matrix.
Bond failure associated with splitting of concrete cover is
often observed in RC elements that have a large amount of
longitudinal reinforcement under seismic loading. Concrete
cover shows splitting because of ring tension caused by the
bearing force from deformed bars.4 Therefore, the bridging
stress of the fiber can resist the ring tension by restricting the
expansion of the splitting crack. Hence, it is considered that
bond strength and ductility can be improved using SHCC.
Many researchers have studied bond behavior between
deformed bars and fiber-reinforced cementitious composites
(FRCCs), including fiber-reinforced concrete (FRC). For
example, Hota and Naaman5 investigated the bond stress-slip
relationship of deformed bar embedded in FRC. The bond
strength and ductility of FRC show a remarkable increase
compared to conventional concrete. Concerning HPFRCC,
Chao et al.6 also studied the bond stress-slip relationship
of deformed bar embedded in HPFRCC. They mentioned
that the superior bond response in HPFRCC can be directly
related to its tensile strain-hardening behavior, which distin-
guishes it from conventional concrete or conventional FRC.
In fact, the bridging stress distribution along the inner crack
has been considered based on the direct tension characteris-
tics of HPFRCC in the literature.
The other way to appropriately use SHCC for bond
improvement involves reducing the cover thickness and
bar spacing. SHCC, which can provide excellent bond
strength and ductility, has the potential to produce enough
bond response in spite of smaller cover thickness and bar
spacing. Furthermore, SHCC, in which coarse aggregate is
not required, is able to reduce bar spacing when the fibers
in SHCC are distributed uniformly in the element section.
Asano and Kanakubo7 also investigated the bond properties
of SHCC (ECC), focusing on the size effect. The pullout
bond test was conducted using SHCC block specimens in
which slits were inserted to vary the cover thickness. The
test results show a definite increase in bond strength as the
size of the specimen decreases. The smallest tested cover
thickness is 5 mm (0.20 in.), which is smaller than the fiber
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-322.R3, doi: 10.14359/51687228, received May 26, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
189
Table 1—Mechanical properties of PVA fiber
Tensile Elastic
Length, Diameter, strength, modulus,
Type mm (in.) mm (in.) N/mm2 (ksi) kN/mm2 (ksi)
0.04
PVA 12.0 (0.472) 1690 (245) 40.6 (5890)
(1.6 × 10–3)
length of 12 mm (0.47 in.). It is assumed that the smaller
cover affects the orientation of fibers to bridge the matrix
over the section at the slit and shows higher performance
than larger specimens.
The objectives of this study are to investigate the
bond-splitting behavior of reinforced SHCC elements and
to propose a predicting method for the bond strength. The
pullout bond test and the beam bond test are conducted using
small cover thickness and bar spacing. As mentioned in the
literature,6 the predicting method is based on the material
test results of SHCC. The fiber used for SHCC in this paper
is polyvinyl alcohol (PVA) fiber. The volume fraction of
PVA fiber is set to 2.0%.
RESEARCH SIGNIFICANCE
Using SHCC for quake-resistant elements such as coupling
beams, shear walls, and energy-absorption columns provides
excellent structural performance and controlled opening
of the crack width. Tensile behavior of SHCC due to the
bridging effect of fiber affects crack-opening characteris-
tics, shear-resistance performance, and bond behavior asso-
ciated with splitting of the matrix surrounding longitudinal
reinforcement. The evaluation and prediction of bond
strength are two of the important issues for the designing of
reinforced SHCC elements and using SHCC appropriately.
It is also important that the evaluations be conducted consid-
ering the relationship between the material characteristics of
SHCC, such as not only compression behavior but also tensile
behavior, and the structural behaviors of SHCC elements.
SHCC can retain sufficient tensile stress for the ring tension
caused by bearing stress in the matrix surrounding the longi-
tudinal reinforcement. Furthermore, SHCC does not include
coarse aggregates. Thus, the minimum cover thickness and
spacing of the reinforcement can likely be reduced.
SHCC USED AND MATERIAL TEST
PVA fiber 0.04 mm (1.6 × 10–3 in.) diameter was used in
this study. The fiber volume fraction is 2.0%. Table 1 lists the
mechanical properties of PVA fiber used. The water-binder
ratio (w/b) was 0.50, and the unit weight of binder consisted
of 540 kg/m3 (911 lb/yd3) ordinary portland cement and
240 kg/m3 (405 lb/yd3) fly ash. Fine sand with a size under
0.2 mm (7.9 × 10–3 in.) was used as fine aggregate. The
mixing of SHCC was carried out using a biaxial mixer of
2 m3 (2.62 yd3) capacity. The mixture proportion of SHCC
is shown in Table 2.
It is difficult to perform the uniaxial tension test of
HPFRCC with a standardized test method. Many researchers
and institutes have conducted uniaxial tension tests with
various test methods including specimens with different
dimensions and shapes, and different boundary conditions
at the ends of specimens. Kanakubo8 also conducted several
190
Table 2—Mixture proportion of SHCC
Sand- Unit weight, kg/m3 (lb/yd3)
Fiber volume binder
fraction, % w/b ratio Water Cement Fly ash Sand
400 540 240 619
2.0 0.50 0.77
(675) (911) (405) (1044)
Cement: Ordinary portland cement
Fly ash: Type II of Japanese Industrial Standard (JIS A 6202)
Sand: Size under 0.2 mm (7.9 × 10–3 in.)
High-range water-reducing admixture: binder × 0.65%
types of uniaxial tension tests with several types of cemen-
titious composites. The values of characteristics obtained
from each test show differences in spite of using the same
HPFRCC. It is assumed that the size effect due to fiber orien-
tation and distribution causes the difference of bridging effect
of fiber in HPFRCC. Furthermore, tensile behavior is very
sensitive to the boundary conditions of loading, such as the
existence of a secondary moment caused by non-uniformity
of fiber distribution.
On the other hand, not many values of characteristics are
required for structural design. As one example, the perfect
elastic-plastic model is proposed for the tensile stress-strain
curve of HPFRCC in the Japan Society of Civil Engineers
Recommendations.9 The Japan Concrete Institute has a
standard test method to determine the tensile characteristics
of HPFRCC on the basis of the bending test.10 The tensile
strength and ultimate tensile strain can be obtained by simple
reverse calculation from the bending test results given in the
standard. The calculation method for tensile strength and
ultimate tensile strain is introduced based on the assumptions
for stress distribution under the maximum bending moment,
as shown in Fig. 1, which are: 1) the stress distribution on the
compression side is triangular; and 2) the stress distribution
on the tension side is uniform. These assumptions represent
a state in which the strain on the tension edge has reached
the ultimate strain but the stress on the compression edge has
not reached the compressive strength under the maximum
bending moment.10 In this study, this method is applied for
the material test of SHCC.
The material test results of SHCC are listed in Table 3.
A ϕ100 x 200 mm (ϕ3.94 x 7.87 in.) test cylinder was used
for the compression test. The compressive strength was
approximately 45 N/mm2 (6.5 ksi). Figures 2 and 3 show the
bending test setup and measured moment-curvature curves,
respectively. Deflection-hardening and multiple-crack
behaviors are observed. The tensile strength varied from
4.2 to 4.6 N/mm2 (0.62 to 0.67 ksi). The ultimate strains
were 2.86 and 1.25%. According to previous experimental
results,8 the scattering of tensile characteristics of test pieces
is greater than that of the compressive strength. It was
reported that the coefficient of variation is over 30% for
ultimate strain. Furthermore, one of the test pieces for beam
test specimen ruptured at the out of the region of the linear
variable displacement transducers (LVDTs). This test piece
showed small ultimate strain.
A deformed bar (D13) with a specific diameter of 13 mm
(0.51 in.) was used for main bars (Fig. 4). The rib height
and spacing are 1.0 mm (0.039 in.) and 8.9 mm (0.35 in.),
ACI Structural Journal/March-April 2015
Table 3—Material test results of SHCC
Compression test (ϕ100 x 200 mm cylinder) Bending test* (100 x 100 x 400 mm)
Test series Compressive strength, N/mm (ksi)2
Elastic modulus, kN/mm (ksi)
2
Tensile strength, N/mm2 (ksi) Ultimate strain, %
For pullout test 44.5 (6.45) 15.3 (2220) 4.61 (0.669) 2.86
For beam test 46.1 (6.69) 17.5 (2540) 4.24 (0.615) 1.25
*
JCI-S-003-2007. Method of test for bending moment-curvature curve of fiber-reinforced cementitious composites.
Note: 1 mm = 0.0394 in.

Table 4—Mechanical properties of reinforcement

Type Test series Yield strength, N/mm2 (ksi) Tensile strength, N/mm2 (ksi) Elastic modulus, kN/mm2 (ksi)
D13 (13 mm) Pullout 369 (53.5) 543 (78.8) 193 (28,000)
D13 (13 mm) Beam (main bar) 762 (110.5) 952 (138.1) 192 (27,800)
D6 (6 mm) Beam (stirrup) 364 (52.8) 514 (74.5) 185 (26,800)

Note: 1 mm = 0.0394 in.

Fig. 1—Assumption of stress distribution of HPFRCC. Fig. 3—Moment-curvature curve.

Fig. 4—Tested reinforcement (D13).

Fig. 2—Bending test setup.
of 0.5, 1.0, and 1.5 times the bar diameter. These thicknesses
respectively. The tension test results for the deformed bars
are small values compared with the cases for conventional RC
used in this study are listed in Table 4.
elements. Three identical specimens for each parameter were
tested. Nine specimens were tested in total.
PULLOUT BOND TEST
SHCC was cast from the side of the specimen, as shown
Outline of experiment
in Fig. 5. The casting level of SHCC at both sides of the slits
The specimen and loading method are shown in Fig. 5. The
was raised almost uniformly to avoid damage to the polysty-
specimen was a rectangular SHCC block with a height of
rene slits. The casting was done with careful observations of
91 mm (3.58 in.). One deformed bar with a diameter of 13 mm
the flow of the SHCC around the reinforcement to not have
(0.51 in.) was arranged in the central position of the block.
any voids between the slits.
The unbonded regions were set at both the loaded and free
The monotonic pullout load was applied until the
ends. The embedded length was four times the bar diameter db
reinforcement slipped out from the block under the controlled
in the central part of the specimen. The slits made of foamed
displacement. Teflon sheets were placed between the spec-
polystyrene were set, as shown in the figure, to simulate the
imen and the reaction plate to facilitate lateral displacement
cover thickness or bar spacing. The dimension of the slit was
of the block. The LVDT was set to measure slip at the free
the main parameter, and it was set to have a cover thickness C
end. The loaded end slip is obtained as the summation of the

ACI Structural Journal/March-April 2015 191

Table 5—Pullout test results
Average
Maximum bond stress τb,max, Slip at maximum bond stress, Maximum bond stress τb,max, Slip at maximum bond stress,
C/db ID N/mm2 (ksi) smax, mm (in.) N/mm2 (ksi) smax, mm (in.)
P05-1 7.68 (1.11) 0.499 (0.0196)
0.5 P05-2 6.85 (0.99) 0.589 (0.0232) 7.70 (1.12) 0.540 (0.0213)
P05-3 8.57 (1.24) 0.533 (0.0210)
P10-1 8.37 (1.21) 0.455 (0.0179)
1.0 P10-2 11.19 (1.62) 0.577 (0.0227) 9.75 (1.41) 0.474 (0.0186)
P10-3 9.70 (1.41) 0.389 (0.0153)
P15-1 12.01 (1.74) 0.303 (0.0119)
1.5 P15-2 12.98 (1.88) 0.333 (0.0131) 13.24 (1.92) 0.272 (0.0107)
P15-3 14.73 (2.14) 0.179 (0.0070)

Tepfers11 suggested the cylinder models of stress distribu-

Fig. 5—Pullout specimen. (Note: 1 mm = 0.0394 in.)
elongation of the reinforcement and the free end slip under
the assumption that bond stress distributes uniformly among
the embedded region (4db).
Test results of pullout bond test
Table 5 lists the test results of the pullout bond test. The
maximum bond stress varied from 7.70 to 13.24 N/mm2
(1.12 to 1.92 ksi), with 13.24 N/mm2 (1.92 ksi) as the average
value among the same three specimens. As expected, the
maximum bond stress increased as the cover thickness also
increases. Even if the cover thickness is remarkably small—
that is, 0.5 times the bar diameter—a higher strength is
obtained than in the case of conventional concrete. Figure 6
shows examples of the specimens after loading. The splitting
cracks between slits and the bar are observed at the free end of
the specimen in the case of small thickness. From the photos
of the side view, the opening between the slit and the block is
recognized at the loaded end of the specimen. Figure 7 shows
the bond stress-slip relationships. In general, the pullout load
showed a sudden drop in the case of conventional concrete,
when splitting cracks occur. In the case of SHCC, however,
the bond stress-slip relationships showed very ductile
behavior over the slip of 1/10 the bar diameter. Furthermore,
the relationships between specimens with small cover thick-
ness exhibited a look-alike yielding behavior.
tions of concrete around a pulled deformed bar as dividing
into three stages: 1) elastic stage; 2) partly-cracked elastic
stage; 3) and plastic stage. Figure 8 shows the relationships
between the cover thickness and bond strength for the three
stages of the Tepfers models. In the calculation of the bond
strength by these models, the tensile strength of SHCC listed
in Table 3 is adopted as that of the matrix, and the angle
between the bearing principal stress and axial direction is
assumed to be 45 degrees, which is similar to that assumed
in the Tepfers study. As shown in Fig. 8, the obtained bond
strengths of SHCC are equal or higher than those predicted
by the plastic stage assumption. The higher bond strength
was obtained by the specimens with C/db = 0.5. The cover
thickness of these specimens was about half of fiber length.
It is assumed that fiber orientation shows the tendency to
have the similar direction to the perpendicular between the
reinforcement and slits. This may cause the advancement of
a fiber-bridging effect to the splitting crack.
BEAM BOND TEST
Outline of experiment
A bond test for beam specimens with small bar spacing
was conducted to investigate the bond behavior and bond
strength of SHCC. An example of the beam specimens is
shown in Fig. 9. The specimens were designed for observing
the bond behavior in tension-side (bottom-side) reinforce-
ment. Each beam had two test regions (L and R). After the
left-side test region was subjected to three-point bending
using the loading and support positions indicated by void
triangles in the figure, the right-side test region underwent
identical loading. An unbonded zone covered by steel pipes
was arranged at the supported area. The slits along the
perpendicular direction at the loaded and free ends were
set to avoid continuous cracks from the untested zone. The
length between two slits was 208 mm (8.19 in.). SHCC
was cast from the side of the beam, as shown in the figure,
because the SHCC elements could be used as the precast
members. Though the fiber orientation around main bars
could not be observed, it was assumed that the fiber oriented
both in longitudinal and circumferential directions.

192 ACI Structural Journal/March-April 2015

Fig. 6—Pullout specimen after loading.
The list of beam specimens is shown in Table 6. Twelve
loading tests were carried out on six beam specimens. The
test parameters were the number of main bars, stirrup ratio,
and embedded length of the test region. A high-strength D13
deformed bar (refer to Table 4) was used as the main bar
to avoid flexural yielding. The number of main bars was
5, 6, and 7, corresponding to bar spacings of 1.0, 0.8, and
0.6 times the bar diameter db, respectively. A 6 mm (0.24 in.)
diameter deformed bar, D6, was used for the stirrups for half
of the specimens with a spacing of 80 or 40 mm (3.15 or
1.57 in.). Embedded lengths of 16 or 8 times the bar diam-
eter are selected. The embedded length was arranged by the
length of steel pipes, as shown in Fig. 9.
The monotonic load was applied under the controlled
displacement. The LVDTs were set for the all tension bars to
measure the slip at the free end. Strain gauges were placed
on all the tension bars at the leaded end, as shown in Fig. 9. Fig. 7—Bond stress-slip curves from pullout test.
Visible crack observations were recorded in each loading
measurement step.

Test results of beam bond test

Examples of the final crack patterns after loading are
shown in Fig. 10. Multiple fine cracks can be observed along
the main bar in side-view photos showing shear cracks. Axial
and perpendicular cracks also occurred on the bottom side.
In the specimens with seven main bars (No. 5 and No. 6),
cracks along the main bars extended through the section,
and finally a “side-split”-type bond failure was observed. No
yielding of the main bars was observed.
Experimental bond stress was obtained from measured
strain at the loaded end of the main bar. The bond stress is
calculated as the average stress that is obtained by tensile Fig. 8—Bond strength of pullout specimen.
force divided by the surface area of the main bar. The bond other hand, in the right-side figures, corner bars show high
stress-free end slip curves are shown in Fig. 11 through 13. bond stresses in the specimens with stirrups.
In all specimens, a sudden decrease in the bond stress was
not observed. The curves show ductile behavior until several Bond strength of beam bond test
millimeters of slip. From the left-side figures, in which spec- Table 7 lists the experimental results for bond strength
imens without stirrups are shown, curves obtained from the (maximum bond stress). Figure 14 shows the relationship
corner bars and center ones show similar behavior. On the between the average bond strength and stirrup ratio. It is
clearly observed that bond strength increases as the stirrup

ACI Structural Journal/March-April 2015 193

Table 6—List of beam specimens
Main bar Stirrup
Embedded
ID Arrangement pt, % Arrangement pw, % length
No.1-L None 0.0 16db
No.1-R 5-D13 2-D6 at 80 0.4 16db
0.91
No.2-L (C/db = 1.0) None 0.0 8db
No.2-R 2-D6 at 40 0.8 16db
No.3-L None 0.0 16db
No.3-R 6-D13 2-D6 at 80 0.4 16db
1.09
No.4-L (C/db = 0.8) None 0.0 8db
No.4-R 2-D6 at 40 0.8 16db
No.5-L None 0.0 16db
No.5-R 7-D13 2-D6 at 80 0.4 16db
1.27
No.6-L (C/db = 0.6) None 0.0 8db
No.6-R 2-D6 at 40 0.8 16db

Fig. 9—Beam specimen. (Note: Dimensions in mm; 1 mm = 0.0394 in.)

Fig. 10—Examples of crack patterns of beam specimens.

ratio increases. It is considered that this increment is due to
the confinement effect of lateral reinforcement that is also
observed in conventional RC elements. Figure 15 shows
comparisons of bond strengths that are obtained from the
corner bars and center bars. In the specimens with stirrups,
the bond strengths of the corner bars are higher than those of
the center bars. The average ratio of the corner to center bars
is 1.30 and 1.41 for specimens with stirrup ratios of 0.4%
and 0.8%, respectively.
The obtained bond strength is compared with the calculated
strength that is proposed for conventional RC elements—
that is, prediction formulas proposed by Morita and Fujii12
and Orangun and Jirsa.13 These formulas are listed in the
Appendix. Morita and Fujii formulas were built based on

194 ACI Structural Journal/March-April 2015

Fig. 11—Bond stress-slip curves from beam test (5-D13). Fig. 13—Bond stress-slip curves from beam test (7-D13).
This means that the bond stress along the axial direction of
main bar distributes uniformly through the embedded length
in the beam specimen at the maximum bond stress. From the
test results, however, the embedded length of the beam spec-
imen affects the average bond strength, as listed in Table 7.
The bond strength obtained in 8db specimens is an average
of 1.84 times that obtained in 16db specimens. It is assumed
that at the unbonded region between slits (length = 208 mm
[8.19 in.], hereafter called the splitting length), SHCC also
resists the splitting force from the main bars. Equation (1)
shows the equilibrium condition between the tensile strength
of SHCC and the splitting stress. As described in previous
literature, bond stress is given by splitting stress, as shown in
Eq. (2). Substituting Eq. (2) with Eq. (1) gives Eq. (3)

σs · db · ℓb = σt · 2C · ℓs (1)
Fig. 12—Bond stress-slip curves from beam test (6-D13).
the experimental results of cantilever-type specimens, while τbc = σs · cotα (2)
Orangun and Jirsa formulas were built by the results from
beam specimens. The Orangun and Jirsa formulas include the 2C  s
τ bc = σ t ⋅ ⋅ ⋅ cot α (3)
effect of bond length, whereas the Morita and Fujii formulas db  b
do not include the term of bond length. Figure 16 shows the
comparisons of bond strength between experimental and
where σs is the splitting stress; db is the diameter of deformed
calculated values by two formulas. Clearly, experimental
bar; ℓb is embedded length; σt is the tensile strength of
bond strengths are quite higher than the calculated bond
SHCC; C is the half-length of bar spacing or cover thickness
strengths. The calculated values from the Orangun and Jirsa
(2C = [b – Nt · db]/Nt); ℓs is the splitting length; τbc is the bond
formulas are higher than those from Morita and Fujii due to
strength provided by SHCC; α is the angle between principal
the difference of bond length influence.
bearing stress and axial direction; b is the width of the beam;
and Nt is the number of main bars.
PREDICTION METHOD OF BOND-SPLITTING
The confinement effect of the lateral reinforcement has to
STRENGTH OF SHCC
be considered to evaluate the bond strength of the specimens
The test results of the pullout bond test indicate that the
with stirrups. Yasojima and Kanakubo14 proposed a predic-
stress distribution of SHCC corresponds to the “plastic stage”
tion method of bond splitting strength for conventional RC
model suggested by Tepfers. This assumption is understand-
elements. The bond strength provided by the confinement
able for SHCC, which retains tensile stress after cracking.
effect is predicted by considering compatibility conditions
Though the non-uniform tensile stress distribution is
between the splitting crack opening and deformation of
assumed by Chao et al.6 or Asano and Kanakubo,7 a uniform
the lateral reinforcement. The bond strength is given at the
stress distribution can be assumed in the case of small cover
compressive failure of concrete bearing with the rib of the
thickness or bar spacing, as shown in Fig. 17. Furthermore,
deformed bar. The identical situation can be considered in
the bond stress-slip curves from the pullout bond test show
the case of the SHCC element because the tensile behavior
very ductile behavior over the slip of 1/10 the bar diameter.
of SHCC has no effect on these phenomena. Equation (4) is

ACI Structural Journal/March-April 2015 195

Table 7—Beam test results
Test variables Bond strength, N/mm2 (ksi)
ID Main bar pw, % Embedded length Corner bar (average) Center bar (average) Average (all bars)
No.1-L 0.0 16db 5.81 (0.843) 5.94 (0.862) 5.89 (0.854)
No.1-R 0.4 16db 8.98 (1.302) 7.06 (1.024) 7.83 (1.136)
5-D13
No.2-L 0.0 8db 10.03 (1.455) 11.32 (1.642) 10.81 (1.568)
No.2-R 0.8 16db 10.48 (1.520) 7.91 (1.147) 8.94 (1.297)
No.3-L 0.0 16db 5.04 (0.731) 4.76 (0.690) 4.85 (0.703)
No.3-R 0.4 16db 7.91 (1.147) 6.32 (0.917) 6.85 (0.994)
6-D13
No.4-L 0.0 8db 9.45 (1.371) 9.43 (1.368) 9.44 (1.369)
No.4-R 0.8 16db 9.75 (1.414) 6.51 (0.944) 7.59 (1.101)
No.5-L 0.0 16db 4.87 (0.706) 4.48 (0.650) 4.59 (0.666)
No.5-R 0.4 16db 7.78 (1.128) 5.62 (0.815) 6.24 (0.905)
7-D13
No.6-L 0.0 8db 7.65 (1.110) 8.15 (1.182) 8.01 (1.162)
No.6-R 0.8 16db 8.37 (1.214) 5.97 (0.866) 6.66 (0.966)

Fig. 14—Bond strength-stirrup ratio relationship.

proposed for the bond strength provided by the lateral rein- Fig. 15—Bond strength of corner and center bars.
forcement confinement14

b ⋅ pw hr
τ bs = 0.018 ⋅ ⋅ ⋅ Est ⋅ σ B ⋅ cot α (4)
N t ⋅ d b 9d w

where τbs is the bond strength provided by lateral reinforce-

ment confinement; b is the width of beam; pw is the stirrup
ratio; Nt is the number of main bars; db is the diameter of
main bar; hr is rib height; dw is diameter of stirrup; Est is the
elastic modulus of stirrup; σB is the compressive strength of
SHCC; and α is the angle between principal bearing stress
and axial direction (= 56 degrees).
Consequently, bond strength τb is predicted by Eq. (5)

τb = τbc + τbs (5)

Table 8 lists the calculation results using the proposed
method, and Fig. 18 shows the comparison between the
experimental bond strengths and predicted ones by the
proposed method. In the calculations, the angle between the
principal bearing stress and the axial direction is assumed
to be 56 degrees, which was obtained for conventional RC
Fig. 16—Comparison with calculated bond strength.
specimens reported in the literature.14 The rib height of the
D13 main bar is 1.0 mm (0.039 in.). The average of the ratio
of the experimental strengths to predicted strengths is 1.07.
The predicted bond strength shows good agreement with the
experimental results.

196 ACI Structural Journal/March-April 2015

Table 8—Calculated bond strength
Predicted bond strength, N/mm2 (ksi)
Main Experimental bond
ID bar strength, N/mm2 (ksi) τbc τbs τbc + τbs Experiment/prediction
No.1-L 5.89 (0.854) 6.32 (0.917) 0 (0) 6.32 (0.917) 0.93
No.1-R 7.83 (1.136) 6.32 (0.917) 1.59 (0.230) 7.91 (1.148) 0.99
5-D13
No.2-L 10.80 (1.566) 12.65 (1.835) 0 (0) 12.65 (1.835) 0.85
No.2-R 8.94 (1.297) 6.32 (0.917) 2.25 (0.326) 8.57 (1.243) 1.04
No.3-L 4.85 (0.703) 5.06 (0.734) 0 (0) 5.06 (0.734) 0.96
No.3-R 6.85 (0.994) 5.06 (0.734) 1.21 (0.175) 6.27 (0.909) 1.09
6-D13
No.4-L 9.44 (1.369) 10.12 (1.468) 0 (0) 10.12 (1.468) 0.93
No.4-R 7.59 (1.101) 5.06 (0.734) 1.71 (0.248) 6.77 (0.982) 1.12
No.5-L 4.59 (0.666) 3.79 (0.550) 0 (0) 3.79 (0.550) 1.21
No.5-R 6.24 (0.905) 3.79 (0.550) 0.96 (0.139) 4.75 (0.689) 1.31
7-D13
No.6-L 8.01 (1.162) 7.59 (1.101) 0 (0) 7.59 (1.101) 1.06
No.6-R 6.66 (0.966) 3.79 (0.550) 1.36 (0.197) 5.15 (0.747) 1.29

of the beam bond test also show that the bond strength of
SHCC has a higher value. A prediction methodology is
proposed as the summation of the bond strength provided
by SHCC and the confinement of lateral reinforcement. The
predicted bond strength shows good agreement with the
experimental results.

Fig. 17—Cylinder model for plastic stage.
AUTHOR BIOS
ACI member Toshiyuki Kanakubo is an Associate Professor at the
Department of Engineering Mechanics and Energy, University of Tsukuba,
Tsukuba, Japan, where he received his PhD. His research interests include
high-performance fiber-reinforced cementitious composites, the structural
behavior of fiber-reinforced polymer reinforced concrete structures, and
bond properties of reinforcement and concrete.

Hiroshi Hosoya is a Research Engineer at the Technical Research Institute,

Okumura Corporation, Japan. He received his DrE from the University of
Tokyo, Tokyo, Japan. His research interests include structural behavior of
high-performance fiber-reinforced cementitious composites, precast struc-
tural systems for high-rise buildings, concrete structures using new types of
materials, and high-strength materials.

ACKNOWLEDGMENTS
The authors wish to express their gratitude and sincere appreciation to the
Techno Material Co., Ltd. for providing SHCC materials.

Fig. 18—Comparison with predicted bond strength.
CONCLUSIONS
To investigate the bond-splitting behavior of reinforced
SHCC elements and to propose a predicting method for
bond strength, a pullout bond test and beam bond test were
conducted with small cover thickness and bar spacing.
The test results of the pullout bond test show that the
bond strength of SHCC is higher than that of conventional
concrete, which is expected by the “partly cracked elastic
stage” of the cylindrical model by Tepfers. It is considered
that the tensile stress distribution of SHCC surrounding the
main bar corresponds to the “plastic stage.” The test results
REFERENCES
1. RILEM TC 208-HFC, “Strain Hardening Cement Composites: Struc-
tural Design and Performance,” State-of-the-Art Report of the RILEM Tech-
nical Committee 208-HFC, SC3, 2013.
2. Kanda, T.; Tomoe, S.; Nagai, S.; Maruta, M.; Kanakubo, T.; and
Shimizu, K., “Full Scale Processing Investigation for ECC Pre-cast Struc-
tural Element,” Journal of Asian Architecture and Building Engineering,
V. 5, No. 2, 2006, pp. 333-340. doi: 10.3130/jaabe.5.333
3. Kunieda, M., and Rokugo, K., “Recent Progress on HPFRCC in Japan;
Required Performance and Applications,” Journal of Advanced Concrete
Technology, V. 4, No. 1, 2006, pp. 19-33. doi: 10.3151/jact.4.19
4. Tepfers, R., “A Theory of Bond Applied to Overlapped Tensile
Reinforcement Splices for Deformed Bars,” Publication 73:2, Division
of Concrete Structures, Chalmers University of Technology, Göteborg,
Sweden, 1973, 328 pp.
5. Hota, S., and Naaman, A. E., “Bond Stress-Slip Response of
Reinforcing Bars Embedded in FRC Matrices under Monotonic and
Cyclic Loading,” ACI Structural Journal, V. 94, No. 5, Sept.-Oct. 1997,
pp. 525-537.
6. Chao, S. H.; Naaman, A. E.; and Parra-Montesinos, G. J., “Local
Bond Stress-Slip Models for Reinforcing Bars and Prestressing Strands
in High-Performance Fiber-Reinforced Cement Composites,” Antoine E.

ACI Structural Journal/March-April 2015 197

Naaman Symposium – Four Decades of Progress in Prestressed Concrete,
FRC, and Thin Laminate Composites, SP-272, American Concrete Institute,
Farmington Hills, MI, 2010, pp. 151-172.
7. Asano, K., and Kanakubo, T., “Study on Size Effect in Bond Split-
ting Behavior of ECC,” Bond in Concrete 2012, Volume 2—Bond in New
Materials and under Severe Conditions, 2012, pp. 855-859.
8. Kanakubo, T., “Tensile Characteristics Evaluation Method for
Ductile Fiber-Reinforced Cementitious Composites,” Journal of Advanced
Concrete Technology, V. 4, No. 1, 2006, pp. 3-17. doi: 10.3151/jact.4.3
9. Japan Society of Civil Engineers, “Recommendations for Design and
Construction of High Performance Fiber Reinforced Cement Composites
with Multiple Fine Cracks (HPFRCC),” JSCE Concrete Engineering Series
82, 2007, pp. 14-15.
10. Japan Concrete Institute, “Method of Test for Bending Moment-Curva-
ture Curve of Fiber-Reinforced Cementitious Composites (JCI-S-003-2007),”
http://www.jci-net.or.jp/j/jci/study/jci_standard/JCI-S-003-2007-e.pdf
11. Tepfers, R., “Lapped Tensile Reinforcement Splices,” Journal of the
Structural Division, ASCE, V. 108, 1982, pp. 283-301.
12. Morita, S., and Fujii, S., “Bond Capacity of Deformed Bars due
to Splitting of Surrounding Concrete,” Bond in Concrete, P. Bartos, ed.,
Applied Science Publishers, London, UK, 1982, pp. 331-352.
13. Orangun, C. O.; Jirsa, J. O.; and Breen, J. E., “A Reevaluation of
Test Data on Development Length and Splices,” ACI Journal Proceedings,
V. 74, No. 3, Mar. 1977, pp. 114-122.
14. Yasojima, A., and Kanakubo, T., “Local Bond Splitting Behavior
of RC Members with Lateral Reinforcement,” 14th World Conference on
Earthquake Engineering, Conference Proceedings, Paper ID 05-03-0033,
2008. (DVD)
APPENDIX
The Morita and Fujii12 calculation formula (notation is
altered from the original) is
τb = (τco + τst) · 1.22 (A1)
τ co = ( 0.307 ⋅ bi + 0.427 ) ⋅ σ B (A2)
198
Ast
τ st = 24.9 ⋅ k ⋅ ⋅ σ B (A3)
s ⋅ Nt ⋅ db
where τb is the bond strength for bottom bars (kgf/cm2);
τco is the bond strength without stirrup (kgf/cm2); τst is
the bond strength increment caused by stirrup (kgf/cm2);
bi = b/N · db – 1 (side split); σB is the concrete compressive
strength (kgf/cm2); k = 1 (side split); Ast is the sectional area
of pair of stirrup (cm2); s is stirrup spacing (cm); Nt is the
number of main bar; and db is the diameter of main bar (cm).
The Orangun and Jirsa calculation formula (the notation is
altered from the original) is
τb = τco + τst (A4)
 3C 50db 
τ co = 1.2 + s + ⋅ σ B (A5)
 db  b 
Ast ⋅ σ wy
τ st = ⋅ σ B (A6)
500 ⋅ s ⋅ db
where τb is the bond strength (psi); τco is the bond strength
without stirrup (psi); τst is the bond strength increment
caused by stirrup (psi); Cs is half of main bar spacing (in.); db
is the diameter of main bar (in.); ℓb is the bond length (splice
length) (in.); σB is the concrete compressive strength (psi);
Ast is the sectional area of pair of stirrup (in.2); σwy is the
yield strength of stirrup (psi);and s is stirrup spacing (in.).
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S18
Wide Beam Shear Behavior with Diverse Types of
Reinforcement
by S. E. Mohammadyan-Yasouj, A. K. Marsono, R. Abdullah, and M. Moghadasi
The shear behavior of six wide beams was was studied to inves-
tigate the effectiveness of various types of shear reinforcement in
improving the shear capacity of wide beams. One specimen each
was provided: without vertical stirrups, with vertical stirrups, inde-
pendent bent-up bars, independent middepth horizontal bars, and
the combination of vertical stirrups and bent-up bars. To study
the effectiveness of longitudinal flexural reinforcement on the
shear capacity of wide beams, an additional specimen without
stirrups, but including approximately two-thirds of longitudinal
flexural reinforcement that were arranged in column band, was
investigated. The performances were measured in terms of deflec-
tion, crack patterns, concrete and steel strains, ultimate load, and
modes of failure. The results showed that independent bent-up bars
increased the shear capacity and ductility of wide beams. It was
revealed that, although independent horizontal bars increased the
shear capacity to some extent, the beam was less ductile through
failure. The results also indicated that the beam with banded main
reinforcement achieved larger failure load.
Keywords: ductility; independent bent-up bars; longitudinal reinforce-
ment; shear reinforcement; slabs; stirrups; wide beams.
INTRODUCTION
Reinforced concrete materials are widely accepted due to
their strength, durability, reduced costs, quality, and ease of
forming into various shapes and sizes to construct structural
members such as beams, slabs, columns, and shear walls.
The use of reinforced concrete wide beams is advantageous
for many reasons. In buildings such as warehouses, commer-
cial buildings, parking garages, and office buildings, rein-
forced concrete wide beams with a width-depth ratio of at
least 2 are used to reduce floor height and facilitate the run
of services under the floor.
There have been many studies on wide beam behavior,
mostly on their support width and transversal spacing of
stirrup legs.1-10 Regarding the particular feature and behavior
of connections in wide beams that the column is narrower
than the beam, some researchers conducted different tests
to investigate the behavior of wide beams under different
load conditions, statically and dynamically. Based on the
results of the research, the use of wide beams for different
regions and conditions is limited. The seismic performance
of wide beams was investigated,1,2,10 with some recom-
mendations to use these members in seismic regions. In
a different loading for shear,3 tests on impact behavior of
reinforced concrete beams for the effect of shear mecha-
nisms revealed that specimens with higher shear capacity
are able to sustain more impacts and absorb higher values
of energy. Abbas et al.4 investigated the structural response
of wide beams and the results indicated that, under high-rate
loading, the beam is capable of withstanding higher values
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
of loading. They discussed that, in the critical early stages,
the values of strain rate for high-rate loading are lower than
the threshold established by experiments relating the varia-
tion in compressive and tensile strength of concrete under
different rates. Previous researchers5-7 found that the rate of
loading can influence the arrangement and distribution of
shear reinforcement in beams. There are some guidelines on
wide beam properties and design in ACI 31811 and Eurocode
2,12 where many codes do not address them directly. In fact,
most of them refer to the special cases of beams or slabs,
which are in concert with wide beams. In recent research,13,14
there have been studies to evaluate and propose a practical
and optimum arrangement of shear reinforcement for these
members and to compare against ACI 318 and Eurocode 2. To
clarify and improve upon the shear strength predicted by ACI
318 and Eurocode 2, and to apply it to reinforced concrete wide
beams, the influence of shear reinforcement distribution and
support width were stuided.13 In addition, the use of two stirrup
legs was banned because the maximum spacing among vertical
legs in a stirrup is suggested to be limited to values close to the
beam depth. Taking into account the large number of stirrup
legs, the small height of wide beams, and thus the difficulty of
stirrup placement, there should be some alternatives to these
beams for shear.
Based on previous research, aggregate size, beam size,
flexural reinforcement, and stirrups influence the shear
strength of reinforced concrete wide beams.8,15-21 It is
accepted that, in the presence of shear reinforcement, ulti-
mate strength capacity is governed either by flexure or
by the web crushing, the least of the shearing, and shear
compression resistance.22 Considering the geometry of wide
beams using a larger number of longitudinal flexural rein-
forcement in comparison to normal beams, shear strength
is highly influenced by longitudinal flexural reinforcement.
It is known that concrete shear strength is decreased when
the longitudinal flexural reinforcement ratio is reduced, and
with an increase of depth (for example, from 460 to 910 mm
[18.11 to 35.83 in.], the equivalent decrease in concrete
shear strength was 18%.16 Lubell et al.19 demonstrated that
in members with no shear reinforcement, both the member
depth and the details of longitudinal flexural reinforcement
influence the shear capacity of the member.
Cracking spacing is influenced by the distrbution of longitu-
dinal flexural reinforcement and its bond effect on concrete.17
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-360.R4, doi: 10.14359/51687299, received May 27, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
199
Due to the short height of wide beams, the spacing between
longitudinal flexural reinforcement and the midheight of the
beams becomes smaller; thereby crack spacing can be affected
by longitudinal flexural reinforcement.
In beam-column connections, researchers suggested that
all beam longitudinal flexural reinforcement should pass
through the beam supports.1 Popov et al.18 tested rein-
forced concrete beam-column-slab cruciform interior joint
subassemblages under simulated seismic loadings. They
confirmed that, in wide beams with narrow interior supports,
the contribution of longitudinal bars outside of the column
confinement to the lateral resistance is significant. It was
also shown that when the supported width is narrower than
the width of the member, the shear capacity of a member
decreases.8
This paper investigates the overall behavior of wide beams
under different methods of reinforcing for shear. Three types
of shear reinforcement that have been previously used23,24
are used in this study, which include: stirrups as normal
reinforcement; independent bent-up bars as a new type of
shear reinforcement for wide beams (which is a focus of this
study); and independent middepth horizontal bars. A column
part added to the beam part in the specimens plays the
role of the reinforced concrete wide beam-internal column
structure.
The main objective of this paper is to find an easier and
more effective way of reinforcing concrete wide beams to
behave under shear. Furthermore, to quantify the effect of
critical design parameters, a numerical model is employed.
The numerical analysis, after being verified by the exper-
imental results, gives a better understanding of specimen
behavior and provides the possibility of change into the
details of the specimens for future design schemes.
RESEARCH SIGNIFICANCE
Many researchers express the importance of shear capacity
of wide beams and suggest guidelines to use stirrups in these
beams. They recommend to increase stirrup legs through the
cross section in wide beams. The need for a large amount
of flexural reinforcement in concrete wide beams provides
anchorage support for independent bent-up bars; however, it
is difficult to place stirrups with more than two legs in these
beams. Independent bent-up bars could be a feasible option as
an innovative shear reinforcement that can contribute to the
shear capacity of wide beams, which can also be developed
into reinforced concrete slabs.
EXPERIMENTAL PROGRAM
This paper presents results of experimental tests on six
reinforced concrete wide beams that were part of a study
on the influence of different types and arrangements of rein-
forcement on concrete wide-beam capacity.25 The specimens
consisted of identical concrete mixtures and configurations,
including a beam part and a column part. From the design
stage, the ratio of beam weight to minimum ultimate load
was negligible; therefore, the column part is projected to
support the real condition of thebeam-internal columns. In
this case, when load was applied on the column part of the
specimen, the supports’ reactions were considered imposed
200
loads on the beam. Details of specimen configuration and
test setup are shown in Fig. 1, and reinforcement cages
before casting can be seen in Fig. 2. In all of the specimens,
the beam part was designed to nominal dimensions of 1820
mm (71.6 in.) length, 750 mm (29.5 in.) width, and 250 mm
(9.8 in.) height. The column part was square and with
nominal dimensions of 300 mm (11.8 in.) sides and 300 mm
(11.8 in.) height. The specimens were supported under the
beam part at ends with a shear span of 550 mm (21.6 in.)
from each support to the face of the column. Shear span
together with other geometric properties of the specimens
are detailed in Table 1. The column parts were reinforced
properly with stirrups and axial reinforcement using appro-
priate anchorage length of bars to resist the applied load and
transfer it to the beam part.
Materials
Ready mixed concrete containing coarse aggregate of
20  mm (0.8 in.) maximum size and a nominal specified
strength of 30 MPa (4.3 ksi) was used. All specimens were
simultaneously cast in plywood formwork and cured under
moist burlap. Standard cylindrical molds 150 mm (6 in.) in
diameter and 300 mm (12 in.) high were cast at the same
time as the specimens and cured for control tests. Concrete
samples were tested, and the average specified compressive
strength of the concrete fc′ was 29 MPa (4.2 ksi).
Steel bars of 6, 10, 12, and 16 mm (0.24, 0.39, 0.47, and
0.63 in.) were used for stirrups, independent bent-up bars,
compression bars, independent middepth horizontal bars,
and flexural reinforcement, respectively. The properties of
reinforcing steel bars are shown in Table 2 and Fig. 3.
Specimens
Each specimen was denoted with WB, an acronym of
“wide beam,” followed by the specimen number. From
the geometric properties of the specimens shown in Table
1, Specimen WB1 was devised with no shear reinforce-
ment—this served as the control specimen. Specimen WB2,
with two parallel independent bent-up bars in cross section,
was designed to investigate the shear capacity of indepen-
dent bent-up bars. An inclination angle equal to 50 degrees
with the longitudinal flexural reinforcement was selected
for independent bent-up bars. This angle was close to 45
degrees, which is perpendicular to the most critical cracks
in shear span. As illustrated in Fig. 1(c), the length of inde-
pendent bent-up bars was approximately 240 mm (9.5 in.).
In some previous studies,23,24 independent bent-up bars
with different anchorage lengths were tested in reinforced
concrete beams, and specimens including independent
bent-up bars with a minimum anchorage length of 75 mm
(2.9 in.) revealed acceptable performance. Accordingly,
an anchorage length of 100 mm (3.9 in.) for independent
bent-up bars was selected in the present study. Specimen
WB3 included independent middepth horizontal bars for
shear reinforcement that were evenly distributed through
the width of the beam part and spanning the entire length of
the specimen. The value of this reinforcement, regarding the
size and number of longitudinal flexural reinforcement, was
approximately 40% of flexural reinforcement. In Specimen
ACI Structural Journal/March-April 2015
WB4, about two-thirds of the longitudinal flexural reinforce-
ment was arranged in a band of width equal to and centered
on the column width to compare its effect with the evenly
Fig. 1—Specimens configuration and test setup.
ACI Structural Journal/March-April 2015
distributed bars across the width in other specimens. Stirrups
with spacing of approximately 150 mm (5.9 in.) for vertical
legs, transversally and longitudinally, were determined as
shear reinforcement in Specimen WB5. From the specimen
configuration in Fig. 1(a), each stirrup in the cross section
was composed of three rectangular stirrups, including two
internal stirrups 152 x 184 mm (6 x 7.2 in.) and one external
stirrup of 672 x 184 mm (26.5 x 7.2 in.). Specimen WB5 was
a sample of a normal wide beam reinforced with only stirrups
as shear reinforcement. In beams, bent-up bars should not
be used as shear reinforcement except in combination with
stirrups.12 Therefore, a combination of independent bent-up
bars and stirrups was adopted in Specimen WB6. In addition
to providing shear capacity, stirrups could keep the longi-
tudinal flexural reinforcement tightly together. Independent
bent-up bars (same as SpecimenWB2) and stirrups of 150
mm (5.9 in.) transversally and 300 mm (11.8 in.) longitudi-
nally were arranged for WB6. Doubled longitudinal spacing
of stirrups in Specimen WB6, relative to Specimen WB5,
was designed to prevent flexural failure prior to shear failure.
Linear variable displacement transducers (LVDTs) were
used to measure vertical displacement of the specimens. To
investigate the forces in steel bars, electrical strain gauges
were installed on flexural bars of Specimens WB1, WB2,
and WB4, and on middepth horizontal shear bars of Spec-
imen WB3. For flexural bars, strain gauges were installed
on the middle bars near the column face and, for middepth
horizontal shear bars, on one of the bars passing through
the column band and another on the bar out of the column
band. The strain gauges on the middepth horizontal shear
bars were installed at a distance of approximately h/2 from
the column face where the highest shear stress in the cross
section was predicted. Before the main test, approximately
10% of predicted failure load was applied and released for
each specimen to check that the supports and equipment were
firm and consistent. A load cell was placed on the column
stub under the machine head to measure the applied load.
The load was applied through the column stub at the middle
of the beam part and support reactions acted as line loads on
each end of the beams. Data from LVDTs, strain gauges, and
load cell were recorded by an electronic data logger.
Items of investigation
After curing the specimens for 28 days under laboratory
conditions, the specimens were tested and their perfor-
mances were measured in terms of midspan displacement,
crack pattern, concrete and steel strain, ultimate load, and
mode of failure. At the same time, cylindrical concrete spec-
imens were tested and, for each bar stock, randomly selected
samples were used to determine the standard stress-strain
curve of the steel bars.
SHEAR CAPACITY THEORY
The beams were reinforced in such a way that flexural
failure was prevented. This implied that a shear failure
mechanism forms first before the yielding of flexural rein-
forcement. In members without shear reinforcement, shear
failure mechanism depends on the tensile strength of the
concrete. Based on modified truss analogy, nominal shear
201
Table 1—Geometric properties of specimens
Designation of shear reinforcement
Stirrups Horizontal bars Independent bent-up bars
bw/d, mm Height h, Shear span a,
Specimen (in.) mm (in.) mm (in.) a/d ρw, % Av, mm (in. ) Sv, mm (in.)
2 2
Avh, mm (in. )
2 2
Avb, mm2 (in.2) Sb, mm (in.)
WB1* 3.6 (3.5) 250 (9.8) 550 (21.6) 2.6 1.4 — — — — —
WB2 3.6 (3.5) 250 (9.8) 550 (21.6) 2.6 1.4 — — — 157 (0.2) 150 (5.9)
WB3 3.6 (3.5) 250 (9.8) 550 (21.6) 2.6 1.4 — — 804 (1.3) — —
WB4 *
3.6 (3.5) 250 (9.8) 550 (21.6) 2.6 1.4 — — — — —
WB5 3.6 (3.5) 250 (9.8) 550 (21.6) 2.6 1.4 170 (0.3) 150 (5.9) — — —
WB6 3.6 (3.5) 250 (9.8) 550 (21.6) 2.6 1.4 170 (0.3) 300 (11.8) — 157 (0.2) 150 (5.9)
*
No shear reinforcement; however, they were different in longitudinal flexural reinforcement distribution.
Notes: bw/d is beam width-depth ratio; a/d is shear span-depth ratio; and Avh is the area of middepth horizontal shear bars.

Fig. 2—Reinforcement cages before casting.

strength of a reinforced concrete beam, Vc, can be written as where concrete struts run parallel to diagonal cracks and
(ACI 318-08, Eq. (11-2))11 stirrups perform as tension members.26
For a member subject to shear and flexure only, the expres-
Vn = Vc + Vs (1) sion used for shear capacity of the member without shear
reinforcement is (ACI 318-08, Eq. (11-3))
where Vc is nominal shear strength of concrete, and Vs is
nominal shear strength of web reinforcement. Vc = 0.166 f c′bw d (MPa) (2a)
In the truss analogy contains the shear resistance of a
parallel chord truss and a web-reinforced concrete beam,
Vc = 2.0 f c′bw d (psi) (2b)

202 ACI Structural Journal/March-April 2015

Table 2—Reinforcement characteristics
Steel Diameter, Yield stress fy, Yield strain Ultimate strength
bar mm (in.) MPa (ksi) εy fu, MPa (ksi)
D-6 5.39 (0.21) 637.7 (92.49) 1.2 × 10–3 727.5 (105.5)
D-10 9.69 (0.38) 601.5 (87.24) 1.3 × 10–3 700.4 (101.6)
D-12 11.53 (0.45) 617.6 (89.57) 1.8 × 10–3
726.4 (105.4)
D-16 15.71 (0.62) 465.6 (67.53) 1.1 × 10–3
547.1 (79.35)

where fc′ is specified compressive strength of concrete; bw is

the web width; and d is the distance from the compression
face to the centroid of longitudinal tensile reinforcement.
Under a more detailed calculation, three variables such as
tensile strength of concrete, the ratio of area of longitudinal
tensile reinforcement, and Vud/Mu are taken into account.
The basic equations then become (ACI 318-08 Eq. (11-5))
 V d
Vc =  0.166 f c′ + 17ρw u  bw d (MPa) (3a)
 Mu 
 V d
Vc =  2.0 f c′ + 2500ρw u  bw d (psi) (3b)
 Mu 
where fc′ is specified compressive strength of concrete; ρw
is the ratio of area of longitudinal tensile reinforcement (As)
to the multiplication of the web width (bw), and the distance
from the compression face to the centroid of longitudinal
tensile reinforcement (d); Vu is the factored shear force at a
section; and Mu is factored moment at a section.
Some researchers27,28 indicate that Eq. (3a) or (3b) over-
estimates the influence of fc′ and underestimates the influ-
ence of ρw and Vu d/Mu; however, to consider the influence of
mid-depth horizontal shear bars on the shear capacity of the
specimen, Eq. (3a) or (3b) can be used in this study.
With the assumption that the diagonal members in the
truss analogy are assumed to be inclined at 45 degrees, shear
reinforcement needs to carry the exceeding shear that causes
inclined cracking. Where shear reinforcement used in the
member is vertical to the longitudinal tensile reinforcement,
nominal shear strength provided by shear reinforcement, Vsv,
is given by (ACI 318-08 Eq. (11-15))
Av f ys d
Vsv = (4)
sv
Fig. 3—Stress-strain of steel bars.
exceed 0.42 f c′ MPa (5 f c′ psi); otherwise, the limits
should be reduced by half.
Wherever a group of parallel bent-up bars is used as shear
reinforcement, Vsb is computed by (ACI 318-08, Eq. (11-16))
Avb f yb (sin α + cos α )d
Vsb = (5)
sb
where fyb is yield strength of bent-up bar; α is the angle
between bent-up bar and longitudinal tensile reinforcement;
Avb is the area of parallel bent-up bars; and sb is center-to-
center spacing of bent-up bars measured in direction parallel
to the longitudinal tensile reinforcement. Beams reinforced
with stirrups and bent-up longitudinal bars should be spaced
at d/2 such that any crack initiating at mid-depth and propo-
gating at approximately 45 degrees is crossed by at least one
row of stirrups or bent-up bars.11 Therefore, to control the
most critical shear cracks by independent bent-up bars, this
limit is important.
In a member using a combination of stirrups and inde-
pendent bent-up bars as shear reinforcement, the term Vs in
Eq. (1) can be written as
Vs = Vsv + Vsb (6)
where Vsv and Vsb are nominal shear strengths of vertical shear
reinforcement and independent bent-up bars, respectively.
In comparison to ACI 318-08,11 the method of shear
design used by Eurocode 212 is the variable strut inclination
method, and the shear capacity of the concrete VRd,c is given
by (EC2, Clause (6.2.2))

where Av is the area of shear reinforcement; sv is center-to- VRd , c = [0.18(1 + 200 /d )(100ρw ⋅ f ck )1/ 3 ]bw d (MPa) (7a)
center spacing of shear reinforcement measured in direction
parallel to longitudinal tensile reinforcement; and fys is yield VRd , c = [4.96(1 + 7.87 /d )(100ρw ⋅ f ck )1/ 3 ]bw d (psi) (7b)
strength of stirrup.
Research6,7 has shown that, with a decrease in the trans- with a minimum shear capacity of
verse spacing of stirrup legs across the section in wide beams
with substantial flexural reinforcement, the shear behavior is
VRd , c min = [0.035(1 + 200 /d )3/ 2 f ck ]bw d (MPa) (8a)
improved. Other research14 has indicated that the transverse
spacing of web reinforcement shown in ACI 318-08 limited
to the lesser of: a) the effective depth d; or b) 600 mm [24 VRd , c min = [0.42(1 + 7.87 /d )3/ 2 f ck ]bw d (psi) (8b)
in.] is adequate when the nominal shear stress does not
where characteristic cylinder strength of concrete fck is taken
equivalent to the specified compressive strength of concrete

ACI Structural Journal/March-April 2015 203

fc′. EC2 considers action of a reinforced concrete beam
in shear by the analogous truss with an angle θ between
22 to 45 degrees to the horizontal for inclined compression
members.29 In this analogy, the bottom chord and vertical
stirrups are the horizontal tension steel and the transverse
tension members, respectively. Where the ultimate shear
force VEd is larger than VRd,c, all shear will be resisted by
the provision of stirrups without direct contribution from the
shear capacity of the concrete. The shear resistance of the
stirrups, VRd,s, is given by (EC2, Clause (6.2.3))

Av
VRd , s = × df ys cot θ (9)
sv

and shear resistance of a multiple system of bent-up bars,

Vwd, is given by

0.9d (cot α + cot θ)

Vwd = f yb Asb sin α × (10)
sb

where the maximum longitudinal spacing of bent-up bars by

EC2 is limited to 0.6d(1 + cotα), where at least 50% of shear
reinforcement should be resisted by stirrups.
In this study, when using a combination of independent
bent-up bars and stirrups, a larger longitudinal spacing of
stirrup legs is used to make a shear-critical, rather than
flexure-critical, member. Safety factors were removed from
design stage formulas, however, to predict the real failure
load under the test stage.

EXPERIMENTAL RESULTS AND DISCUSSION

Load-displacement response
All the specimens were tested and load-displacement
responses of the various specimens are presented in Fig. 4.
Load details and displacements of the specimens are also
shown in Tables 3 and 4.
Shear reinforcement—The lowest failure load, 401 kN
(90 kip), was for reference Beam WB1 with no shear rein-
forcement. Approximately an 11% increase of failure load in
comparison to the load predicted by ACI 318-08 and a 10%
increase to the load predicted by EC2 was observed in this
specimen. Maximum midspan displacement of WB1, approx-
imately 2.6 mm (0.1 in.), was less than 50% of the maximum
midpan displacement, 6.9 mm (0.27 in.), for WB2. Indepen-
dent bent-up bars improved the shear capacity of WB2 to
approximately 51% and it was able to carry a load of 604 kN
(136 kip). The failure load for WB2 was approximately 18%
less than the load from ACI 318-08 and 97% more than EC2.
With a moderate increase in failure load, approximately 26%
greater than WB1, Specimen WB3 carried 507 kN (114 kip)
to failure. In contrast to the predicted load from the design
stage, independent middepth horizontal bars enhanced the
failure load of WB3 up to 30% and 18% higher than ACI
318-08 and EC2, respectively. Maximum midspan displace-
ment of this specimen was 3.1 mm (0.12 in.), which was
larger than in WB1. Regarding the stirrups used in WB5,
the failure load of 581 kN (131 kip) for this specimen was
45% higher than the failure load of WB1, but 13% less than
Fig. 4—Load-displacement responses of specimens.
the load from ACI 318-08 and 20% less from EC2. Using
stirrups of 150 mm (5.9 in.) longitudinal spacing in WB5,
maximum midspan displacement became 5.9 mm (0.232 in.).
Specimen WB6, with a dual system of independent bent-up
bars and stirrups of 300 mm (11.8 in.) longitudinal spacing
in comparison to WB1, revealed an increase in failure load
to approximately 635 kN (143 kip). The failure load by WB6
was the highest load among the specimens. This load was
29% less than the load from ACI 318-08 and 7% less than
EC2. Maximum midspan displacement recorded for WB6,
5.8 mm (0.228 in.), was less than that of WB5 and WB2.
Longitudinal flexural reinforcement—Specimen WB4
failed under 480 kN (108 kip) load, which was 34% larger
than the predicted failure load by ACI 318-08 and 27% larger
than that predicted by EC2. In contrast to reference Spec-
imen WB1, it is observed that concentration of the longitu-
dinal flexural reinforcement in the column band increased

204 ACI Structural Journal/March-April 2015

Table 3—Ultimate capacity and comparison
Specimen Pu,ACI, kN (kip) Pu,EC2, kN (kip) Pu,experiment, kN (kip) Pu,experiment/Pu,ACI Pu,experiment/Pu,EC2
WB1 360 (81) 379 (85) 401 (90) 1.11 1.10
WB2 736 (165) 307 (69) 604 (136) 0.82 1.97
WB3 387 (87) 429 (96) 507 (114) 1.30 1.18
WB4 360 (81) 379 (85) 480 (108) 1.34 1.27
WB5 664 (149) 727 (163) 581 (131) 0.87 0.80
WB6 889 (200) 686 (154) 635 (143) 0.71 0.93

Notes: Pu,ACI is total capacity of each specimen by ACI 318 from design stage; Pu,EC2 represents total capacity of each specimen by Eurocode 2 from design stage; Pu,experiment is
ultimate load that caused failure to each specimen.

Table 4—Flexural and shear cracking load and displacement

Pu,experiment – Pshear crack,
Specimen Pu,experiment, kN (kip) Pflexural crack, kN (kip) Pshear crack, kN (kip) kN (kip) ∆u,experiment, mm (in.)
WB1 401 (90) 216 (49) 392 (88) 9 (2) 2.6 (0.102)
WB2 604 (136) 195 (44) 420 (94) 184 (41) 6.9 (0.272)
WB3 507 (114) 195 (44) 400 (90) 107 (24) 3.1 (0.122)
WB4 480 (108) 210 (47) 430 (97) 50(11) 3.5 (0.138)
WB5 581 (131) 200 (45) 440 (99) 141 (32) 5.9 (0.232)
WB6 635 (143) 180 (41) 480 (108) 155 (35) 5.8 (0.228)

Notes: Pflexural crack is load under which first flexural crack was observed; Pshear crack is load under which first shear crack was observed; and ∆u,experiment is final deflection of the wide
beam midspan at Pu,experiment.

the failure load of WB4 to approximately 79 kN (17.76 kip).
Due to the change in reinforcement concentration, maximum
midspan displacement of WB4, shown to be 3.5 mm (0.138
in.), was greater than the maximum midspan displacement
of WB1, which was 2.6 mm (0.102 in.).
Load-displacement responses of the specimens, however,
indicate that middepth horizontal shear reinforcement
moderately improves the shear capacity and maximum
midspan displacement of wide beams. In addition, the beam
with banded main reinforcement achieves a larger failure
load than the beam with evenly distributed main bars.
Results also showed that independent bent-up bars enhance
the shear capacity of wide beams like stirrups, and an even
higher maximum midspan displacement and final load for
a wide beam using only independent bent-up bars can be
achieved. A combination of independent bent-up bars and
stirrups induce a reasonable performance of the wide beam
with a high shear capacity.
Crack development and mode of failure
Final crack patterns for the specimens are shown in Fig. 5.
To compare against the capacity of the specimens, values
of loads for first flexural crack, first shear crack, and the
value of total load carried by each specimen from appear-
ance of first shear crack until ultimate load are presented in
Table 4. It is important to note that a review of the type of
cracks was based on the visual crack monitoring during the
test process and taking into account the cracks visible to the
eye. In all the specimens, first cracks propagated at midspan
in flexure mode and then developed upward and symmetri-
cally on the left and right sides. Middepth horizontal shear
reinforcement was placed in position with neutral axes and
could not influence flexural capacity of WB3 significantly.
Consequently, loads for first flexural cracks were observed
to be in a close range and were not very different. In all the
specimens, after a certain load for flexural cracks, diag-
onal shear cracks began at an angle of approximately 45
degrees. Specimens WB1 and WB4, with almost the same
load for first flexural crack (216 and 210 kN [49 and 47 kip],
respectively), revealed different load for first shear crack. It
appeared that stirrups in other specimens influenced the first
flexural cracks to initiate at a lower load; however, the first
flexural cracks in WB3—with mid-depth horizontal shear
reinforcement—also appeared at a lower load.
Shear reinforcement—In WB1, only two diagonal shear
cracks occurred; soon after the first shear crack, a brittle
failure occurred due to the second shear crack. In WB3, more
diagonal cracks were observed; however, the last crack,
similar to WB1, caused a brittle failure to the specimen. All
shear cracks in WB1 and WB3 occurred on only one side
of the specimens, started at middepth, and then propagated
to the column face and support. Table 4 shows a higher
load capacity than that of the design prediction. Principal
diagonal shear cracks in WB2, WB5, and WB6 appeared
symmetrically on both the left and right sides of the beam
part after loading, and ductile failure for these specimens
was observed. The resisted load after first shear crack in
WB2 was greater than in other specimens, which indicated
a good shear capacity of bent-up bars. In WB5, stirrups of
150 mm (5.9 in.) longitudinal spacing improved the spec-
imen and had a first shear crack load higher than WB1 to
WB3; however, the load after this crack to failure was less
than those in WB2 and WB6. A combination of bent-up bars
and 300 mm (11.8 in.) longitudinal spacing stirrups influ-
enced the first shear crack in WB6 to appear under a 480 kN
(108 kip) load, which was greater than in WB1 to WB5.

ACI Structural Journal/March-April 2015 205

Fig. 5—Crack development and modes of failure.
From the results for crack patterns and mode of failure, the
use of independent bent-up bars improved the failure mode
of wide beams, where wide beams with a combination of
stirrups and independent bent-up bars exhibit a large number
of smaller cracks and a high resisted load, showing a more
ductile failure. The shear crack extended past the beam
centerline confirming the Lubell et al.8 results, which indi-
cate that the lack of confining pressure under the loading
plate with a width lesser than the beam causes the crack to
extend.
Longitudinal flexural reinforcement—In Table 1, longitu-
dinal flexural reinforcement ratio ρw is similar for WB1 and
WB2 (at 1.4%), but approximately two-thirds of the longi-
tudinal flexural reinforcement of WB4 was concentrated in
the column band. The first shear crack in WB1 with evenly
distributed longitudinal flexural reinforcement occurred at
392 kN (88 kip), whereas the first crack for WB4 was at a load
of 430 kN (7 kip), approximately a 10% increase. Concen-
tration of longitudinal flexural reinforcement improved the
shear capacity of Specimen WB4, but the crack propagation
and mode of failure in WB4 was approximately the same as
WB1.
The anchorage length for longitudinal flexural reinforce-
ment was 200 mm (7.87 in.) to prevent the slippage at
supports. The crack propagation in supports for Specimens
206
WB3 and WB4 with no shear stirrups may indicate slippage
of longitudinal flexural reinforcement at high loads. In other
words, the rule of stirrups in bonding and confining concrete
may improve specimens to prevent slippage of longitu-
dinal flexural reinforcement at support. Moreover, crack
development in Specimen WB2, including independent bent-up
bars as shear reinforcement with no stirrups, also showed no
slippage for longitudinal flexural reinforcement at supports,
which indicates an advantage of independent bent-up bars.
Reinforcement strains
The location of each steel strain gauge is shown in Fig. 6,
as well as the variation in mid-depth horizontal shear bars
and longitudinal flexural bars obtained from electrical strain
gauges. Strain in longitudinal flexural reinforcement of WB1
without shear reinforcement and WB2 with bent-up bars are
denoted by StWB1 and StWB2, respectively. WB4, with concen-
trated longitudinal flexural reinforcement strain in the middle
bar of the column band, is labelled StWB4. Before 100 kN (22
kip) loading, minor digressions from the linear part of load-
strain responses of the specimens were observed, but following
the increase in load, these responses became nonlinear.
The strain profile of StWB2 indicated that, with increasing
the applied load, longitudinal flexural reinforcement yielded.
Results showed that using independent bent-up bars as shear
ACI Structural Journal/March-April 2015
reinforcement increased the shear capacity of the specimen,
but the strain in longitudinal flexural reinforcement did not
change much.
Two other electrical strain gauges, Stouter and Stinner, were
located on middepth horizontal shear reinforcements of
WB3. The total response of Stouter exhibited a linear behavior
and indicated no yield in the horizontal shear reinforcement
that was out of the column area. In comparison to Stouter, a
large deviation in the response of Stinner indicated a higher
strain due to the shear stress that caused yielding of the
longitudinal shear bar in the column area.
Strain gauge StWB4 is compared to StWB1 and, from Fig. 6,
it is shown that a concentration of approximately two-thirds
of longitudinal flexural reinforcement in the column band
induces a higher strain, which could have resulted from
higher stress in the column band.
With regard to the stress-strain relationship, results
showed that, across the width of a wide beam supported
by a column, not supporting the full cross sectional of the
beam, shear stress is not evenly distributed. However, the
higher stress and yielding of longitudinal flexural reinforce-
ment in the column band of the specimen with concentrated
reinforcement indicates stress deviation due to the change of
reinforcement area through the cross section.

Numerical model for parametric study

The license for the ABAQUS, Version 6.9, finite element
(FE) software is available at the Universiti Teknologi
Malaysia.30 Specimen WB2, which included only indepen-
dent bent-up bars as new shear reinforcement, was modeled Fig. 6—Overview on location and variation of middepth
by the FE software. In Fig. 7, a summary of the load-mid- horizontal shear bars and longitudinal flexural bars.
span deflection of experimental and FE analysis for Specimen
WB2 are presented. Results for the numerical modeling that
is close to the experimental results indicate that the numer-
ical modeling can be validated by the experimental results
and used to model other specimens to conduct a parametric
analysis.

FURTHER RESEARCH
Testing wide beam specimens with a larger number and
smaller size of independent bent-up bars in combination
with stirrups is recommended. It is also recommended to
test specimens using independent bent-up bars for punching
shear, if acceptable. The use of independent bent-up bars as
shear reinforcement is faster and easier than other types of
shear reinforcement. In both cases, the results will contribute
new practical guidelines to improve shear capacity of wide
beams and slabs using independent bent-up bars.

CONCLUSIONS
The behavior of reinforced concrete wide beams with
diverse types of reinforcement was investigated under this
experimental study. The results revealed that using indepen-
dent bent-up bars significantly improved the shear capacity
of wide beams. The combination of independent bent-up bars
with stirrups led to higher shear capacity and gradual failure
of the specimen. Independent horizontal bars increased the
shear capacity to some extent, but the beam was less ductile
Fig. 7—Load-midspan defection of experimental and FE
through failure. The results also indicated that the beam with
analysis for Specimen WB2.

ACI Structural Journal/March-April 2015 207

banded main reinforcement achieved a larger failure load
than did the beam with evenly distributed main bars.
AUTHOR BIOS
ACI member Seyed Esmaeil Mohammadyan-Yasouj is a PhD Candi-
date of civil engineering–structure at Universiti Teknologi Malaysia,
Johor Bahru, Malaysia, where he received his MS. His research interests
include analysis and design of reinforced concrete structures, industrialized
building systems, and research to practical guidelines on the construction
of concrete structures.
Abdul Kadir Marsono is an Associate Professor of civil engineering at
Universiti Teknologi Malaysia. He received his MPhil from Heriot-Watt
University, Edinburgh, UK, and his PhD from University of Dundee,
Dundee, UK. His research interests include industrialized building
systems, nonlinear analysis, and reinforced concrete shear walls of tall
building structures.
Ramli Abdullah is a Senior Lecturer and an Associate Professor of civil
engineering at Universiti Teknologi Malaysia. He received his MS from the
University of Strathclyde, Glasgow, UK, and his PhD from Heriot-Watt
University. His research interests include reinforced concrete structures.
ACI member Mostafa Moghadasi is an Assistant Professor of civil engi-
neering at Bu-Ali Sina University, Hamedan, Iran. He received his MSc and
PhD in structural engineering from Amirkabir University of Technology
(Tehran Polytechnic), Tehran, Iran, and Universiti Teknologi Malaysia,
respectively. His research interests include nonlinear behavior of rein-
forced and precast concrete structures, industrialized building systems, and
tall buildings.
ACKNOWLEDGMENTS
The authors would like to acknowledge the support from the faculty of
civil engineering at Universiti Teknologi Malaysia.
REFERENCES
1. Stehle, J. S.; Abdouka, K.; Goldsworthy, H.; and Mendis, P., “The
Seismic Performance of Reinforced Concrete Frames with Wide Band
Beams,” Second International Symposium on Earthquake Resistant Engi-
neering Structures, WIT Press, Catania, Italy, 1999, pp. 113-122.
2. Benavent-Climent, A., “Shaking Table Tests of Reinforced Concrete
Wide Beam-Column Connections,” Earthquake Engineering & Structural
Dynamics, V. 34, No. 15, 2005, pp. 1833-1839. doi: 10.1002/eqe.507
3. Saatci, S., and Vecchio, F. J., “Effects of Shear Mechanisms on Impact
Behavior of Reinforced Concrete Beams,” ACI Structural Journal, V. 106,
No. 1, Jan.-Feb. 2009, pp. 78-86.
4. Abbas, A. A.; Pullen, A. D.; and Cotsovos, D. M., “Structural
Response of RC Wide Beams Under Low-Rate and Impact Loading,”
Magazine of Concrete Research, V. 62, No. 10, 2010, pp. 723-740. doi:
10.1680/macr.2010.62.10.723
5. Hsiung, W. W., and Frantz, G. C., “An Exploratory Study of the Shear
Strength of Wide Reinforced Concrete Beams with Web Reinforcement,”
Research Report CE 83-151, Department of Civil Engineering, University
of Connecticut, Storrs, CT, 1983.
6. Anderson, N. S., and Ramirez, J. A., “Detailing of Stirrup Reinforce-
ment,” ACI Structural Journal, V. 86, No. 5, Sept.-Oct. 1989, pp. 507-515.
7. Leonhardt, F., and Walther, R., “The Stuttgart Shear Tests 1961,” Trans-
lation No. 111, Cement and Concrete Association, London, UK, 1964, 134 pp.
8. Lubell, A. S.; Bentz, E. C.; and Collins, M. P., “One-Way Shear in
Wide Concrete Beams with Narrow Supports,” ASCE Structural Congress,
Crossing Borders, Reston, VA, 2008.
9. Shuraim, A. B., “Transverse Stirrup Configurations in RC Wide
Shallow Beams Supported on Narrow Columns,” Journal of Structural
208
Engineering, ASCE, V. 138, No. 3, 2012, pp. 416-424. doi: 10.1061/
(ASCE)ST.1943-541X.0000408
10. Gentry, T. R., and Wight, J. K., “Wide Beam-Column Connections
under Earthquake-Type Loading,” Earthquake Spectra, V. 10, No. 4, 1994,
pp. 675-703. doi: 10.1193/1.1585793
11. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-08) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2008, 473 pp.
12. Eurocode 2, “Design of Concrete Structures—Part 1-1: General
Rules and Rules for Buildings (EN1992-1-1),” European Committee for
Standardization, Brussels, Belgium, Dec. 2004, 451 pp.
13. Serna-Ros, P.; Fernandez-Prada, M. A.; Miguel-Sosa, P.; and Debb,
O. A. R., “Influence of Stirrup Distribution and Support Width on the Shear
Strength of Reinforced Concrete Wide Beams,” Magazine of Concrete
Research, V. 54, No. 3, 2002, pp. 181-191. doi: 10.1680/macr.2002.54.3.181
14. Lubell, A. S.; Bentz, E. C.; and Collins, M. P., “Shear Reinforce-
ment Spacing in Wide Members,” ACI Structural Journal, V. 106, No. 2,
Mar.-Apr. 2009, pp. 205-214.
15. Sherwood, E. G.; Bentz, E. C.; and Collins, M. P., “Effect of Aggre-
gate Size on Beam-Shear Strength of Thick Slabs,” ACI Structural Journal,
V. 104, No. 2, Mar.-Apr. 2007, pp. 180-190.
16. Tompos, E. J., and Frosch, R. J., “Influence of Beam Size, Longitudinal
Reinforcement, and Stirrup Effectiveness on Concrete Shear Strength,” ACI
Structural Journal, V. 99, No. 5, Sept.-Oct. 2002, pp. 559-567.
17. Zakaria, M.; Ueda, T.; Wu, Z.; and Meng, L., “Experimental Inves-
tigation on Shear Cracking Behavior in Reinforced Concrete Beams with
Shear Reinforcement,” Journal of Advanced Concrete Technology, V. 7,
No. 1, 2009, pp. 79-96. doi: 10.3151/jact.7.79
18. Popov, E. P.; Cohen, J. M.; Thomas, K.; and Kasai, K., “Behavior of
Interior Narrow and Wide Beams,” ACI Structural Journal, V. 89, No. 6,
Nov.-Dec. 1992, pp. 607-616.
19. Lubell, A. S.; Bentz, E. C.; and Collins, M. P., “Influence of Longi-
tudinal Reinforcement on One-Way Shear in Slabs and Wide Beams,”
Journal of Structural Engineering, ASCE, V. 135, No. 1, 2009, pp. 78-87.
doi: 10.1061/(ASCE)0733-9445(2009)135:1(78)
20. Bažant, Z. P., and Kim, J. K., “Size Effect in Shear Failure of Longi-
tudinally Reinforced Beams,” ACI Journal Proceedings, V. 81, No. 5,
Sept.-Oct. 1984, pp. 456-468.
21. Collins, M. P., and Kuchma, D., “How Safe Are Our Large, Lightly
Reinforced Concrete Beams, Slabs, and Footings?” ACI Structural Journal,
V. 96, No. 4, July-Aug. 1999, pp. 482-490.
22. Placas, A., and Regan, P. E., “Shear Failure of Reinforced Concrete
Beams,” ACI Journal Proceedings, V. 68, No. 10, Oct. 1971, pp. 763-773.
23. Guan, A. T., “The Influence of the Anchorage of Independent
Bent-Up Bar on Its Shear Capacity,” master’s research report, Universiti
Teknologi Malaysia, Johor, Malaysia, Apr. 2008, 112 pp.
24. Nabilah, N. B., “The Effectiveness of Independent Bent-up Bars with
Insufficient Anchorage and Inclined Links as Shear Reinforcement,” master’s
thesis, Universiti Teknologi Malaysia, Johor, Malaysia, Apr. 2010, 123 pp.
25. Mohammadyan-Yasouj, S. E., “The Influence of Different Types and
Arrangements of Reinforcement on Capacity of Concrete Wide Beams,”
master’s research report, Universiti Teknologi Malaysia, Johor, Malaysia,
Dec. 2011, 106 pp.
26. Mörsch, E., Concrete-Steel Construction (Der Eisenbetonbau),
Translation of the Third German Edition by E. P. Goodrich, McGraw-Hill,
New York, 1909.
27. Joint ACI-ASCE Committee 426, “Shear Strength of Reinforced
Concrete Members (ACI 426R-74),” Proceedings, ASCE, V. 99, June 1973,
pp. 1148-1157.
28. Kani, G. N. J., “Basic Facts Concerning Shear Failure,” ACI Journal
Proceedings, V. 63, No. 6, June 1966, pp. 675-692.
29. Mosley, B.; Bungey, J.; and Hulse, R., Reinforced Concrete Design to
Eurocode2, Palgrave Macmillan, New York, 2007, 408 pp.
30. ABAQUS, ABAQUS manual, Version 6.9, Pawtucket, RI, 2009.
ACI Structural Journal/March-April 2015
 ACI STRUCTURAL JOURNAL
Title No. 112-S19
Effect of Axial Compression on Shear Behavior of High-
Strength Reinforced Concrete Columns
by Yu-Chen Ou and Dimas P. Kurniawan
To observe the effect of axial compression on the shear behavior
of high-strength reinforced concrete columns, eight shear-critical
high-strength columns were tested under cyclic shear with an axial
compressive stress of 0.3fcʹ to 0.4fcʹ and compared to eight columns
tested in a previous study with an axial compressive stress of 0.1fcʹ
to 0.2fcʹ. Test results showed that the increase rate of concrete
shear strength tended to decrease with increasing axial compres-
sion and reached an upper limit at high axial compression. Most
columns under axial compressive stress of 0.3fcʹ to 0.4fcʹ failed at
the same drift as diagonal cracking. This suggests the minimum
shear reinforcement equations of the ACI Building Code need to
include the effect of axial compression. Based on a test database of
77 high-strength columns and the biaxial behavior of high-strength
concrete, this study proposes concrete shear-strength equations
incorporating the weakening effect of axial compression.
Keywords: axial compression; columns; cyclic loading; diagonal cracking;
double curvature; high-strength concrete; high-strength reinforcement;
reinforced concrete; shear.
INTRODUCTION
The advantages of high-strength concrete combined
with high-strength steel have been demonstrated in prac-
tical use. They are commonly used in high-rise buildings
to reduce the dimensions of columns in lower stories to
increase available floor area and to relive reinforcement
congestion. Advanced technology has enabled the develop-
ment of high-strength materials in Taiwan. High-strength
concrete with specified compressive strength up to 100 MPa
(14,500 psi) and high-strength deformed reinforcement with
specified yield strength of 685 and 785 MPa (100,000 and
114,000 psi) for main and transverse reinforcement,1 respec-
tively, are now commercially available. However, in shear
design for columns, the current ACI Building Code2 limits
concrete compressive strength fcʹ to 70 MPa (10,000 psi)
(ACI 318-11, Section 11.1.2) due to the lack of test data and
practical experience with fcʹ ≥ 70 MPa (10,000 psi). More-
over, the yield strength of shear reinforcement (fyt) is limited
to 420 MPa (60,900 psi) (ACI 318-11, Section 11.4.2) to
control diagonal crack width and to ensure yielding of shear
reinforcement before shear failure.3
The equations for shear strength provided by concrete
(Vc) of the ACI Code2 for nonprestressed members subject
to axial compression were developed based on the results
of studies4-7 of 67 specimens under axial compressive stress
ranging from 0.02fcʹ to 0.81fcʹ, which were reported by Joint
ACI-ASCE Committee 326.8 However, only four specimens
had axial compressive stress higher than 0.15fcʹ, and the fcʹ
values ranged from 14 to 41 MPa (2000 to 6000 psi). For
further assessment of the applicability of the ACI Code equa-
ACI Structural Journal/March-April 2015
TECHNICAL PAPER
tions for Vc to axial compressive stress greater than 0.15fcʹ,
experiments9 were performed using 38 members with fcʹ
ranging from 22 to 27 MPa (3120 to 3950 psi) subjected
to axial compressive stress ranging from zero to 0.7fcʹ. Of
the 38 members tested, 23 were subjected to axial compres-
sive stress exceeding 0.15fcʹ. The ACI code equations for Vc
proved to be conservative for axial compressive stress up to
0.7fcʹ. However, these studies were limited to normal-strength
concrete. Test results10,11 of reinforced concrete beams with
fcʹ values ranging from 21 to 93 MPa (3000 to 13,500 psi)
showed that the degree of conservatism of the ACI Code
equations for Vc reduced with increasing fcʹ. Experimental
studies12 of 24 concrete elements with fcʹ ranging from 30 to
87 MPa (4300 to 12,600 psi) under various combinations of
shear and axial compression showed that ACI Code equa-
tions for Vc were nonconservative for highly axially loaded
reinforced concrete elements. The study recommended that
the axial compression term Nu/Ag in the simplified ACI shear-
strength equation should be limited to 20 MPa (3000 psi).
The equation for shear strength provided by shear rein-
forcement (Vs) of the ACI Code2 was developed based on the
truss analogy.8 A limit of 414 MPa (60,000 psi) was imposed
for fyt because test data showed that shear reinforcement
with high fyt was not able to develop its yield strength. Test
results of 87 beams with fyt ranging from 484 to 1454 MPa
(70,000 to 211,000 psi)3,13,14 showed that shear reinforce-
ment may not be able to develop its yield strength when fyt
≥ 700 MPa (102,000 psi) and fcʹ < 40 MPa (5800 psi), or when
fyt is very high—for example, fyt = 1454 MPa (211,000 psi).
Test results for 42 columns with fyt ranging from 846 to
1447 MPa (123,000 to 210,000 psi)13-15 showed that many
of the columns did not show yielding of shear reinforcement.
The presence of axial compression appeared to decrease the
effectiveness of shear reinforcement to resist shear.
This study tested eight large-scale columns with high-
strength steel and high-strength concrete. The columns
were tested under double-curvature cyclic loading with high
axial compression to simulate seismic loading conditions
in typical lower-story columns in high-rise buildings. Test
results of the eight columns and test data from literature
were then used to examine the effects of axial compression
on shear strength of high-strength concrete columns.
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2013-365.R2, doi: 10.14359/51687300, received May 1, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.
209
Table 1—Specimen design
Concrete compressive strength Longitudinal reinforcing bar D32 (No. 10) Shear reinforcing bar D13 (No. 4)
Axial compres- Shear reinforcing
Column sion ratio, % bar spacing, mm fcs′, MPa fc′, MPa fyls, MPa fyl, MPa ρl, % ρw, % fyts, MPa fyt, MPa ρt, %
C-1 70 104.1
450 0.16
C-2 100 138.8
30
C-3 70 104.6
260 0.28
C-4 100 130.0
685 735 3.52 1.37 785 862
D-1 70 101.0
450 0.16
D-2 100 125.5
40
D-3 70 106.4
260 0.28
D-4 100 127.8
Note: 1 MPa = 145 psi.

Fig. 1—Specimen design: (a) Specimens C-1, C-2, D-1, and D-2; (b) Specimens C-3, C-4, D-3, and D-4; and (c) cross section.
(Note: 1 mm = 0.0394 in.)
RESEARCH SIGNIFICANCE
Columns in the lower stories of high-rise buildings typi-
cally carry large axial compression. By using high-strength
concrete and steel, designers can control the column dimen-
sions and increase the available floor area. However, current
ACI Code equations for shear-strength limit concrete
compressive strength to 70 MPa (10,000 psi) and limit the
yield strength of shear reinforcement to 420 MPa (60,900 psi).
This study tested high-strength concrete columns under high
axial compression and compared test results with those of
high-strength columns reported in the literature to develop
shear-strength equations for designing columns with mate-
rial strengths that exceed the ACI limitations.
EXPERIMENTAL PROGRAM
Specimen design and test setup
Eight large-scale columns were tested. Table 1 lists the
design parameters of the columns. The columns were tested
approximately 1 year after fabrication. Figure 1 illustrates
the specimen design. The locations of strain gauges installed
in the longitudinal and shear reinforcement are also shown
in Fig. 1. Further details of instrumentation can be found
elsewhere.16 The columns had a square cross section of
600 x 600 mm (23.62 x 23.62 in.) and a clear height of
1800 mm (70.87 in.). The columns were reinforced with
D32 (No. 10) SD685 high-strength deformed bars for longi-
tudinal reinforcement and D13 (No. 4) SD785 high-strength
deformed bars for shear reinforcement, and were cast with
high-strength concrete with two levels of fcsʹ. Table 2 lists the
concrete mixture proportions. Two levels of axial compres-
sion ratio—30% (Column C series) and 40% (Column D
series)—were examined. The axial compression ratio is the
ratio of applied axial compressive load to fcʹAg. The fcʹ was
obtained from the average of three 150 x 300 mm (6 x 12 in.)
concrete cylinders. Two levels of shear reinforcement
spacing—450 mm (17.72 in.) and 260 mm (10.24 in.)—
were studied with shear reinforcement ratios of 0.16% and
0.28%, respectively. The spacings were selected to ensure
shear failure before longitudinal reinforcement yielding.

210 ACI Structural Journal/March-April 2015

Table 2—Concrete mixture proportions
Unit weight, kgf/m3
fc′, MPa w/b, % C S FS W FA CA HRWRA SL, cm
70 29 285 200 25 143 789 936 8.16 70 ± 5
100 23 350 300 50 150 654 866 14 70 ± 5

Notes: W is water; B is binder; C is cement; S is slag; FS is fly ash; FA is fine aggregate; CA is coarse aggregate; HRWRA is high-range water-reducing admixture; and SL is slump;
1 MPa = 145 psi; 1 kgf = 2.2046 lbf; 1 m = 39.37 in.; 1 cm = 0.394 in.

The columns were tested using the multi-axial testing

system (MATS) (Fig. 2) at the National Center for Research
on Earthquake Engineering (NCREE), Taiwan. Rotations of
the top and bottom ends of the column were restrained by
fixing the top and bottom blocks of the column to the testing
system. During testing, axial compression was applied first
and maintained constant using force control throughout the
testing. Displacement-controlled lateral cyclic loading was
then applied with the loading history, as shown in Fig. 3,
until the columns lost axial load capacity.

TEST RESULTS AND DISCUSSION

Crack pattern and general behavior
This study was a continuation of earlier tests1 of A and
B series high-strength concrete columns (with the same
specimen designs as C and D series) subjected to low axial
compression ranging from 0.1fcʹAg to 0.2fcʹAg. Those test
results were used as comparison in the current study to inves-
tigate the axial compression effect. Figures 4 and 5 show the
lateral force-displacement relationships for Column C and
Fig. 2—Multi-axial testing system (MATS). (Note: 1 mm =
D series, respectively. Shear failure occurred before longi-
0.0394 in.)
tudinal reinforcement yielding, as expected in design. It has
been observed that at low axial compression,1 increasing
axial compression from 0.1fcʹAg to 0.2fcʹAg enhanced shear
strength. The Vtest at the ultimate condition (peak applied
load), on average, increased by 38% from 0.1fcʹAg to 0.2fcʹAg
(Table 3). However, the brittleness increased. The difference
in Vtest between the ultimate and diagonal cracking condi-
tions was, on average, reduced from 32% to 15% when axial
compression was increased from 0.1fcʹAg to 0.2fcʹAg (Table 3).
In Series C and D columns with axial compression of 0.3fcʹAg
and 0.4fcʹAg, respectively, increasing axial compression also
increased brittleness. The difference in Vtest between the
ultimate and diagonal cracking conditions was, on average,
reduced from 15% to 1%, and from 1% to 0%, when axial
compression was increased from 0.2fcʹAg to 0.3fcʹAg, and
from 0.3fcʹAg to 0.4fcʹAg, respectively (Table 3). However,
increasing axial compression had little or no effect on shear Fig. 3—Loading protocol.
strength. The Vtest at the ultimate condition, on average, average angle of dominant diagonal cracks at that drift
increased by 9% from 0.2fcʹAg to 0.3fcʹAg, and on average level. Crack patterns in Series A and B columns gener-
by –4% from 0.3fcʹAg to 0.4fcʹAg (Table 3). Figures 6(a) to ally started as flexural cracks and became flexure-shear
6(d), and 6(e) to 6(h) show cracking and spalling patterns cracks with increased drift. Further increases in lateral load
in Series C and D columns, respectively, at peak applied produced web-shear cracks with reduced crack angles. In
load. Cover concrete spalling was minor for Series A and B Series A columns, diagonal crack angles (average of those
columns.1 The extent of spalling increased for Series C and of four columns) were 42 degrees (ranging from 40 to
D columns, particularly at the top and bottom corners. 45 degrees) and 33 degrees (ranging from 31 to 34 degrees)
Figure 7 shows the relationship between the diagonal at diagonal cracking condition (first appearance of diagonal
crack angle (with respect to column longitudinal axis) and shear cracks) and ultimate condition, respectively. In Series
the drift ratio for each of the Series A, B, C, and D columns. B columns, the angles were 35 degrees (ranging from 31 to
The diagonal crack angle at a drift level is defined as the 41 degrees) and 24 degrees (ranging from 20 to 28 degrees),

ACI Structural Journal/March-April 2015 211

Fig. 4—Hysteretic behavior of specimens with 30% axial compression: Specimens (a) C-1; (b) C-2; (c) C-3; and (d) C-4.
respectively. Due to high axial compression, web-shear
cracks in Series C and D columns were major cracks with
little or no flexural or flexure-shear cracks. The diagonal
crack angle at diagonal cracking condition was the same as
that at peak applied load. For Series C columns, the diagonal
crack angle was 17 degrees (ranging from 14 to 19 degrees)
at both diagonal cracking and peak applied load. For Series
D columns, the angle was 15 degrees (ranging from 12 to
16 degrees). Figure 8 shows the damage distribution of each
column at test end.
Shear contribution of steel and concrete
Figure 9 shows the relationship between the column drift
and the maximum stress of shear reinforcement for each
column. The shear reinforcement stress increased slowly in
the early drift and increased rapidly after diagonal cracks
formed. As axial compression increased, the formation of
diagonal cracks tended to be delayed, but after diagonal
cracking, shear reinforcement stress increased more rapidly.
The third and fourth columns in Table 3 list the drift at diagonal
cracking condition and corresponding shear reinforcement
stress, respectively. The eighth and ninth columns in Table 3
list the drift at ultimate condition and corresponding shear
reinforcement stress, respectively. Except for Column C-3,
all the Series C and D columns reached the ultimate condi-
tion due to sudden, explosive failure of compression zone at
the same drift as diagonal cracking. Redistribution of internal
forces after diagonal cracking was not successful. Shear rein-
212
forcement stress increased drastically before failure without
increasing the peak applied load. This type of failure mode is
similar to that for members with very small amounts of shear
reinforcement. In such cases, shear reinforcement stress at
ultimate condition was set equal to that at diagonal cracking
condition in the calculation of Vs_test (Table 3). Redistribu-
tion of internal forces after diagonal cracking was successful
for Column C-3, which failed at a higher load than the diag-
onal cracking load. Note that Column C-3 had the highest
ratio of shear reinforcement capacity to diagonal cracking
load (Vs to Vtest at diagonal cracking) among all the Series C
and D columns. The ability of the column to control diag-
onal cracking to allow redistribution likely increases with
increasing ratio of shear reinforcement capacity to diagonal
cracking load. Columns C-1 and C-2 failed once diagonal
cracks formed. Thus, strain measurements were unavailable
after diagonal cracking (Fig. 9).
For columns with low axial compression (Series A and
B columns), redistribution of internal forces after diagonal
cracking was successful. The ultimate condition occurred
at a larger drift than the diagonal cracking. This allowed
further formulation of diagonal cracks and, hence, increased
shear reinforcement stress at the ultimate condition (Table
3). The foregoing observations suggest that as the axial
compression increases, more shear reinforcement is needed
to ensure a successful redistribution of forces after diagonal
cracking. In other words, a term to include axial compres-
sion effect is needed in the minimum shear reinforcement
ACI Structural Journal/March-April 2015
Fig. 5—Hysteretic behavior of specimens with 40% axial compression: Specimens (a) D-1; (b) D-2; (c) D-3; and (d) D-4.

Fig. 6—Crack pattern at peak applied load for Specimens: (a) C-1; (b) C-2; (c) C-3; (d) C-4; (e) D-1; (f) D-2; (g) D-3; and
(h) D-4.
equations of the ACI Code. For columns that had the ulti- column shear failure to a larger drift and, hence, increased
mate condition at a larger drift than the diagonal cracking shear reinforcement stress at the ultimate condition.
condition (Series A and B columns and Column C-3), no Table 3 also lists experimental shear strength Vtest, steel
columns showed yielding of shear reinforcement at the ulti- shear strength Vs_test, and concrete shear strength Vc_test under
mate condition. This observation is consistent with the liter- the diagonal cracking and ultimate conditions. The Vtest is the
ature,13-15 as mentioned previously. Note that test results also load (shear) applied to the column. The Vs_test was calculated
showed that a higher amount of shear reinforcement delayed using Eq. (1), where σst was determined by the stress-drift

ACI Structural Journal/March-April 2015 213

Table 3—Shear strength contributed by concrete and steel from tests
Diagonal cracking condition* Ultimate condition†
Column fc′, MPa Drift ratio, % σst, MPa Vtest, kN Vs_test, kN Vc_test, kN Drift ratio,% σst, MPa Vtest, kN Vs_test, kN Vc_test, kN
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)
A-1 92.5 0.35 19 1264 9 1255 0.57 243 1578 150 1428
A-2 99.9 0.33 6 1286 3 1283 0.53 235 1638 150 1488
A-3 96.9 0.32 16 1279 13 1266 0.75 359 1772 413 1359
A-4 107.1 0.33 14 1298 10 1288 0.79 418 1781 447 1334
B-1 108.3 0.45 18 1862 10 1852 0.59 223 2078 165 1913
B-2 125.0 0.41 20 2007 11 1996 0.50 183 2298 195 2103
B-3 112.9 0.40 16 2098 17 2081 0.54 214 2418 411 2007
B-4 121.0 0.42 18 2103 14 2089 0.64 380 2528 522 2006
C-1 104.1 0.42 28 2036 45 1991 0.42 28 2036 45 1991
C-2 138.8 0.60 28 2958 39 2919 0.60 28 2958 39 2919
C-3 104.6 0.38 28 2153 56 2097 0.70 602 2210 1140 1070
C-4 130.0 0.62 32 3018 68 2950 0.62 32 3018 68 2950
D-1 101.0 0.37 30 2239 46 2193 0.37 30 2239 46 2193
D-2 125.5 0.46 24 2486 36 2450 0.46 24 2486 36 2450
D-3 106.4 0.45 32 2355 77 2278 0.45 32 2355 77 2278
D-4 127.8 0.44 28 2547 92 2455 0.44 28 2547 92 2455
*
First appearance of diagonal shear cracks.
†
Peak applied load.
Notes: 1 MPa = 145 psi; 1 kN = 0.224 kip.

relationship (Fig. 9), and θ was determined by the measured

crack angle (Fig. 7). The Vc_test was calculated using Eq. (2).

Av σ st d
Vs _ test = cot θ (1)
s

Vc_test = Vtest – Vs_test (2)

It can be seen from Table 3 that the amount of shear rein-

forcement did not substantially affect the Vc_test value at
diagonal cracking. On the other hand, axial compression has
a positive effect on Vc_test at diagonal cracking but the effect
appeared to reach an upper limit under high axial compres-
sion. As axial compression increased from 0.1fcʹAg to 0.2fcʹAg,
from 0.2fcʹAg to 0.3fcʹAg, and from 0.3fcʹAg to 0.4fcʹAg, the
Vc_test increased in average by 57%, 24%, and –4%, respec- Fig. 7—Drift ratio versus diagonal crack angle.
tively. Moreover, the difference in Vc_test between the diag-
onal cracking condition and ultimate condition decreased  Nu 
Vc = 0.17 1 +  f c′bw d (MPa)
with increasing axial compression.  13.8 Ag  (3)

EXAMINATION OF ACI 318 SHEAR EQUATIONS  Nu 

Vc = 2 1 +  f c′bw d (psi)
According to the ACI Code, nominal shear strength Vn can  2000 Ag 
be obtained from two components: shear strength provided
by concrete, Vc, and shear strength provided by steel rein-
forcement, Vs. In the Code, Eq. (3) to (6) are used to calculate  V d
Vc of reinforced concrete members under axial compression. Vc =  0.16 f c′ + 17ρw u  bw d (MPa)
Equation (3) is the simplified equation. Equations (4) and  Mm  (4)
(5) are used for detailed calculation for Vc but should not be  V d
greater than Eq. (6). The Vs can be calculated using Eq. (7). Vc = 1.9 f c′ + 2500ρw u  bw d (psi)
 Mm 

214 ACI Structural Journal/March-April 2015

Fig. 8—Damage distribution at end of test: Specimens (a) C-1; (b) C-2; (c) C-3; (d) C-4; (e) D-1; (f) D-2; (g) D-3; and (h) D-4.

M m = M u − Nu
( 4h − d ) (5)
8

0.29 N u
Vc = 0.29 f c′bw d 1 + (MPa)
Ag (6)
Nu
Vc = 3.5 f c′bw d 1 + (psi)
500 Ag

The Mu is taken as: 1) moment at distance d from the section

of maximum moment when the ratio of shear span to effec-
tive depth is greater than 2; or 2) moment at the center of
shear span when the ratio of shear span to effective depth is
less than 2.8

Av f yt d
Vs = (7)
s
Fig. 9—Stress of shear reinforcement.
Although Eq. (3) to (6) are used to estimate nominal shear
C-3 because the shear reinforcement of these columns failed
strength in the ACI Code, they were originally derived based
to permit redistribution of internal forces, as noted previ-
on shear corresponding to diagonal cracking.8 The Vc_test at
ously. In other words, shear reinforcement did not function
two conditions—diagonal cracking and ultimate shear condi-
effectively in these columns. Equation (7) obtained a conser-
tions—were compared to the simplified (Eq. (3)) and detailed
vative estimate for Column C-3 even though shear reinforce-
(Eq. (6)) shear-strength equations (Table 4). Note that the
ment stress was below yield stress (Table 3). The estimate
ACI Code limit on concrete compressive strength (fcʹ ≤ 70
was conservative because actual shear crack angles were
MPa [10,000 psi]) was not applied when using the above ACI
much smaller than 45 degrees, as assumed in Eq. (7), leading
Code equations. Equation (3) yields conservative prediction
to a larger actual steel shear strength. As noted previously,
for most columns except Columns D-2 and D-4, but becomes
shear reinforcement stress can be further increased at peak
less conservative as axial compression increases. In the
applied load by increasing the amount of shear reinforce-
detailed shear strength calculation, the Mm values (Eq. (5))
ment. This should be properly considered when determining
are negative for all columns. This means that the moment
the limit value of shear reinforcement stress to be used in
effect is small and, hence, Vc is independent of moment. In
Eq. (7). Further research is needed to address this issue.
this case, Vc is governed by Eq. (6). Comparisons with the
Equation (6) originated from Eq. (8), which was derived
test results show that Eq. (6) is not conservative for Vc at the
based on the principal stress equation at the point of diagonal
diagonal cracking and ultimate conditions for all columns.
cracking. The effects of axial load and shear were consid-
When Eq. (7) is used with the actual yield strength of
ered. The effect of moment was assumed to be small and,
shear reinforcement (Table 1), it does not yield conserva-
hence, was neglected.8
tive results for Series A and B columns (the last column in
Table 4) because, at peak applied load, stress in shear rein-
forcement was far from yield (Table 3). Equation (7) cannot ft ′ Nu
Vc = bw d 1 + (8)
be evaluated in Series C and D columns except for Column F2 f t ′ bw d

ACI Structural Journal/March-April 2015 215

Table 4—Ratio of test results to shear-strength prediction using ACI 318-11 without strength limitation
Diagonal cracking shear strength Ultimate shear capacity

Vc _ test Vc _ test Vc _ test Vc _ test Vs _ test

Column VEq ( 3) VEq ( 6 ) VEq ( 3) VEq ( 6 ) VEq ( 7 )

(1) (2) (3) (4) (5) (6)

A-1 1.61 0.81 1.83 0.93 0.43
A-2 1.49 0.76 1.72 0.88 0.43
A-3 1.55 0.79 1.67 0.85 0.68
A-4 1.44 0.74 1.49 0.76 0.74
B-1 1.69 0.89 1.75 0.92 0.47
B-2 1.39 0.77 1.46 0.82 0.56
B-3 1.53 0.85 1.48 0.82 0.68
B-4 1.42 0.80 1.36 0.77 0.86
C-1 1.23 0.74 1.23 0.74 —
C-2 1.27 0.82 1.27 0.82 —
C-3 1.30 0.77 0.66 0.39 1.89
C-4 1.39 0.88 1.39 0.88 —
D-1 1.15 0.73 1.15 0.73 —
D-2 0.97 0.66 0.97 0.66 —
D-3 1.12 0.72 1.12 0.72 —
D-4 0.95 0.65 0.95 0.65 —

Based on test data,8 ftʹ/F2 was set to 0.29√fc′ (MPa) or
3.5√fc′  (psi). The ft′ was assumed to be 0.62√fc′ (MPa) or
7.5√fc′ (psi). Thus, Eq. (8) becomes Eq. (9).
1.6 N u
Vc = 0.29 f c′bw d 1 + (MPa)
f c′bw d (9)
0.133N u
Vc = 3.5 f c′bw d 1 + (psi)
f c′bw d
For simplicity, 1.6/√fc′ MPa (0.133/√fc′ psi) was replaced
by a constant value of 0.29 MPa (0.002 psi), which corre-
sponds to an fcʹ of approximately 30 MPa (4400 psi). More-
over, bwd was approximated by Ag. With these two changes,
Eq. (9) becomes Eq. (6). The simplification made in Eq. (6)
by assuming fcʹ is equal to 30 MPa (4400 psi) leads to an
overestimation when fcʹ exceeds the assumed value. In the
case of fcʹ = 100 MPa (14,500 psi) and Nu/fcʹAg = 0.4, design
parameters for Columns D-2 and D-4, the simplification
causes a 30% overestimation of Vc.
Based on experimental results (Table 3) and the litera-
ture,1,13,15,17-23 a test database of Vc_test at diagonal cracking
from 77 shear-critical high-strength columns was estab-
lished (Table A1 in the Appendix*). Because the value of
Mm (Eq. 5) is negative for all 77 columns, Eq. (6) governs
detailed shear-strength calculation instead of Eq. (4). The
12th, 13th, and 14th columns of Table A1 show the ratio
The Appendix is available at www.concrete.org/publications in PDF format,
*
of Vc_test to Vc predicted by Eq. (3), (6), and (9), respec-
tively. Whereas Eq. (3) yields conservative predictions for
most columns, Eq. (6) yields nonconservative predictions
for 48 columns. The number of nonconservative results is
greatly reduced to 17 if Eq. (9) is used.
Figure 10 shows the relationship between Vc at diagonal
cracking and axial compression for each of the 77 columns.
The figure also shows the Vc predicted by various models.
Note that each model generates different relationships for
different fcʹ in Fig. 10. Only two relationships corresponding
to fcʹ of 100 and 130 MPa (14,500 and 18,800 psi), respec-
tively, which cover most data, are shown for each model.
Figure 10 also shows that, although Eq. (3) is conservative
for most columns, the linear correlation with axial compres-
sion described by Eq. (3) significantly differs from behavior
revealed by the test data. The test data indicate that Vc
increases with axial compression, but the rate of increase
tends to decrease. At high axial compression, Vc appears
to reach an upper limit. For instance, Vc test data from this
study show an upper limit at axial compression of 0.3fcʹAg to
0.4fcʹAg. The Sakaguchi13 data show an upper limit of 0.4fcʹAg
to 0.5fcʹAg. The Maruta15 data show that the increase in Vc
substantially slows when axial compression is increased
from 0.3fcʹAg to 0.6fcʹAg. Equations (6) and (9) are better for
capturing the increasing trend of test data but cannot reflect
the upper-limit phenomenon at high axial compression.
To address the aforementioned Vc behavior under varying
axial compression, Eq. (8) was modified to include the
reduction in principal tensile strength, ftʹ, caused by the
presence of compressive stress acting in the other principal

appended to the online version of the published paper. It is also available in hard copy
from ACI headquarters for a fee equal to the cost of reproduction plus handling at the direction.24-28 The principal compressive stress increases
time of the request. with increasing axial compression of the column.

216 ACI Structural Journal/March-April 2015

Fig. 10—Relationship between test data and Vc predictions.
σt = αft′ (10)

The ftʹ is set equal to 0.5√fc′ (MPa) or 6√fc′ (psi) based on

earlier studies29-32 and test data shown in Fig. 11.26,28 Based
on regression analysis of test data (Fig. 11), Eq. (11) is
proposed for reduction factor α. The σc in Eq. (11) is limited
to 0.6fc′ due to limited biaxial test data for σc larger than
0.6fc′ and because 0.6fc′ is the highest axial compressive
stress observed in the 77 columns.

 σ  σ
α = 1 − 0.85 c  for 0 ≤ c ≤ 0.6 (11)
 f ′
c  f c′

The σc at diagonal cracking is calculated using the principal

stress equation (Eq. (12)) with applied shear equal to diag-
onal cracking shear.

2 2
N  N   V 
σ c = u +  u  +  c  (12)
2 Ag  2 Ag   bw d 

Fig. 11—Tensile strength degradation of high-strength After defining the reduction factor, Vc can be calculated
concrete. using Eq. (13), which is based on Eq. (9) with reduced prin-
cipal tensile strength σt (Eq. (10)) substituted for ftʹ.
Figure 11 shows test data for concrete under biaxial loading
for high-strength concrete26,28 (69 to 100 MPa [10,000 to
14,500 psi]). The decrease in principal tensile strength due 2 Nu
Vc = 0.29α f c′bw d 1 + (MPa)
to compressive stress in the other principal direction can be α f c′bw d (13)
described by the following equation.
0.17 N u
Vc = 3.5α f c′bw d 1 + (psi)
α f c′bw d

ACI Structural Journal/March-April 2015 217

 Note that before Vc can be calculated using Eq. (13), the
Vc is required as input in Eq. (12). Thus, the aforementioned
calculation requires an iterative procedure. However, the
difference between Nu/Ag and σc is only 0.005 to 0.0017fc′
for axial compression of 0.1fc′Ag to 0.6fc′Ag, respectively.
For simplicity, Nu/Ag may be used instead of σc in Eq. (11)
to eliminate iteration. The new equation is Eq. (14). The
simplification made in Eq. (14) increases the Vc by only
0.41% and 1.91% for axial compression of 0.1fc′Ag and
0.6fc′Ag, respectively. The 15th column in Table A1 shows
predictions obtained by Eq. (13) with α obtained by Eq. (14).
Conservative results are obtained for all columns except for
Columns A-2 and A-4, in which the ratios of measured to
predicted strength are 0.96 and 0.98, respectively. Figure 10
graphically compares predictions obtained by Eq. (13) with
fcʹ of 100 and 130 MPa (14,500 and 18,800 psi) with the test
data. It can be seen that predictions obtained by Eq. (13)
approximate the lower bound of test data.
 Nu  Nu
α = 1 − 0.85  for 0 ≤ ≤ 0.6 (14)
 f c′Ag  f c′Ag
For simplicity in design, Eq. (13) combined with Eq. (14) can
be conservatively approximated by Eq. (15). Equation (15)
is Eq. (9) with the coefficient 0.29√fc′ (MPa) or 3.5√fc′ (psi)
replaced by 0.25√fc′ (MPa) or 3.0√fc′ (psi) and with an upper
limit of 0.2fc′Ag on Nu. The 16th column in Table A1 shows
the predictions obtained by Eq. (15); conservative results
are obtained for all columns except for Column A-4, in
which the ratio of measured to predicted strength is 0.99.
Figure 10 shows that Eq. (15) reaches an upper limit at an
axial compression of 0.2fc′Ag and remains constant with
increasing axial compression. This study proposes to replace
the ACI Code Eq. (6) with Eq. (13) combined with Eq. (14)
or with Eq. (15).
1.6 N u
Vc = 0.25 f c′bw d 1 + (MPa)
f c′bw d (15)
0.133 N u
Vc = 3.0 f c′bw d 1 + (psi)
f c′bw d
Nu shall not be taken greater than 0.2fc′Ag.
Axial compression reduces principal tensile stress and,
hence, increases Vc. This is the mechanism of axial compres-
sion on Vc of the ACI Code equations (Eq. (3) to (6)). On
the other hand, axial compression reduces principal tensile
strength and, hence, decreases Vc. This is the mechanism
this research proposes to add to Vc (Eq. (13) combined with
Eq. (14) or Eq. (15)). The combined effect of the two mecha-
nisms increases Vc at low axial compression but the increase
slows down with increasing axial compression and eventu-
ally reaches an upper limit at high axial compression.
CONCLUSIONS
The effect of axial compression on shear behavior of
high-strength reinforced concrete columns was examined by
218
testing eight large-scale shear-critical columns under high
axial compression and then comparing the results with the
other eight columns tested in an earlier study under low axial
compression. A test database of Vc at diagonal cracking from
77 high-strength columns was established and analyzed with
various Vc models considering axial compression effect. The
main conclusions are summarized as follows.
1. The results of tests on the 16 high-strength columns
showed that with increasing axial compression from 0.1fcʹAg
to 0.4fcʹAg, shear cracking patterns gradually changed from
flexure-shear cracks to web-shear cracks with the average
diagonal crack angle at the ultimate condition changed from
33 to 15 degrees. Moreover, increasing axial compression
increased brittleness, reducing the difference in measured
responses between diagonal cracking and ultimate condi-
tions. For most columns under high axial compression
(0.3fcʹAg and 0.4fcʹAg), redistribution of internal forces after
diagonal cracking was not successful. The columns failed
at the same drift as diagonal cracking. The redistribution
was more successful under low axial compression (0.1fcʹAg
and 0.2fcʹAg). The ultimate condition occurred at a larger
drift than the diagonal cracking condition. Based on these
observations, a term to consider axial compression effect is
needed in the minimum shear reinforcement equations of the
ACI Code. For columns that had the ultimate condition at a
larger drift than the diagonal cracking condition, no columns
showed yield in shear reinforcement at the ultimate condi-
tion. Stress in shear reinforcement at the ultimate condition
tended to decrease with increasing axial compression and
tended to increase with an increasing amount of shear rein-
forcement. Further research is needed to recommend a limit
on shear reinforcement stress in shear strength design.
2. The Vc increased as axial compression increased.
However, the rate of increase tended to decrease with
increasing axial compression and reached an upper limit at
high axial compression. The tested columns showed an upper
limit on Vc at an axial compression of 0.3fcʹAg to 0.4fcʹAg.
Test results reported by other researchers have shown an
upper limit from 0.4fcʹAg to 0.5fcʹAg.
3. The ACI simplified Vc equation, Eq. (3), yields conserva-
tive predictions for most columns in the test database. However,
the linear relationship between Vc and axial compression
described by Eq. (3) significantly differed from the behavior
revealed by the test database. Predictions by the ACI detailed
Vc equations were not conservative for most columns in the
test database. This study proposes new Vc equations (Eq. (13)
with Eq. (14) or Eq. (15)) to replace the upper limit equation of
the ACI detailed Vc equations (Eq. (6)). Comparisons between
the proposed equations and test database show conservative
predictions for most columns.
AUTHOR BIOS
Yu-Chen Ou is an Associate Professor of civil and construction engi-
neering at the National Taiwan University of Science and Technology,
Taipei, Taiwan. He received his PhD from the University of Buffalo, the
State University of New York at Buffalo, Buffalo, NY. He is the Vice Presi-
dent of the Taiwan Chapter – ACI. His research interests include reinforced
concrete structures and earthquake engineering.
ACI member Dimas P. Kurniawan is a Research Assistant of civil and
construction engineering at the National Taiwan University of Science
ACI Structural Journal/March-April 2015
and Technology. He received his BS from Bandung Institute of Technology,
Bandung, Indonesia, and his MS from the National Taiwan University of
Science and Technology.
ACKNOWLEDGMENTS
The authors would like to thank National Center for Research on
Earthquake Engineering (NCREE), Taiwan, and the Excellence Research
Program of National Taiwan University of Science and Technology for their
financial support.
NOTATION
Ag = gross area of concrete cross-section
Av = total cross-sectional area of shear reinforcement
a = shear span
bw = effective web width of member cross section
d = effective depth of member cross section
F2 = ratio of shear stress at diagonal cracking point to average shear
stress on effective cross section
fc' = concrete compressive strength
fcs' = specified concrete compressive strength
ft' = concrete principal tensile strength
fy = yield strength of steel
fyl = yield strength of longitudinal reinforcement
fyls = specified yield strength of longitudinal reinforcement
fyt = yield strength of shear reinforcement
fyts = specified yield strength of shear reinforcement
h = overall height of member cross section
M m = applied moment modified to consider effect of axial compression
Mu = applied moment
Nu = applied axial load (positive in compression)
s = spacing of shear reinforcement
Vc = nominal shear strength provided by concrete
Vc_test = experimental shear strength provided by concrete
Vn = nominal shear strength
Vs = nominal shear strength provided by shear reinforcement
Vs_test = experimental shear strength provided by shear reinforcement
Vtest = experimental shear strength
Vu = applied shear
a = reduction factor
q = shear crack angle to column longitudinal axis
rl = longitudinal reinforcement ratio
rt = shear reinforcement ratio
rw = longitudinal tension reinforcement ratio
sc = principal compressive stress
sst = shear reinforcement stress
st = concrete principal tensile strength reduced by principal compres-
sive stress in perpendicular direction
REFERENCES
1. Ou, Y. C., and Kurniawan, D. P., “Shear Behavior of Reinforced
Concrete Columns with High-Strength Steel and Concrete,” ACI Structural
Journal, V. 112, No. 1, Jan.-Feb. 2015, 12 pp.
2. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-11) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2011, 503 pp.
3. Lee, J. Y.; Choi, I. J.; and Kim, S. W., “Shear Behavior of Reinforced
Concrete Beams with High-Strength Stirrups,” ACI Structural Journal,
V. 108, No. 5, Sept.-Oct. 2011, pp. 620-629.
4. Morrow, J., and Viest, I. M., “Shear Strength of Reinforced Concrete
Frame Member without Web Reinforcement,” ACI Journal Proceedings,
V. 53, No. 3, Mar. 1957, pp. 833-869.
5. Baldwin, J. W. J., and Viest, I. M., “Effect of Axial Compression on
Shear Strength of Reinforced Concrete Frame Members,” ACI Journal
Proceedings, V. 55, No. 11, Nov. 1958, pp. 635-654.
6. Baron, M. J., and Siess, C. P., “Effect of Axial Load on Shear Strength
of Reinforced Concrete Beams,” Structural Research Series, Civil Engi-
neering Studies, University of Illinois, Urbana, IL, June 1956, 72 pp.
7. Diaz de Cossio, R., and Siess, C. P., “Behavior and Strength in Shear
of Beams and Frames without Web Reinforcement,” ACI Journal Proceed-
ings, V. 56, No. 2, Feb. 1960, pp. 695-736.
8. Joint ACI-ASCE Committee 326, “Shear and Diagonal Tension,” ACI
Journal Proceedings, V. 59, Jan., Feb., and Mar. 1962, pp. 1-30, 277-334,
352-396.
9. Mattock, A. H., and Wang, Z., “Shear Strength of Reinforced Concrete
Members Subject to High Axial Compressive Stress,” ACI Journal Proceed-
ings, V. 81, No. 3, May-June 1984, pp. 287-298.
ACI Structural Journal/March-April 2015
10. Mphonde, A. G., and Frantz, G. C., “Shear Tests of High- and
Low-Strength Concrete Beams without Stirrups,” ACI Journal Proceed-
ings, V. 81, No. 4, July-Aug. 1984, pp. 350-357.
11. Elzanaty, A. H.; Nilson, A. H.; and Slate, F. O., “Shear Capacity of
Reinforced Concrete Beams Using High-Strength Concrete,” ACI Journal
Proceedings, V. 83, No. 2, Mar.-Apr. 1986, pp. 290-296.
12. Gupta, P. R., and Collins, M. P., “Evaluation of Shear Design Proce-
dures for Reinforced Concrete Members under Axial Compression,” ACI
Structural Journal, V. 98, No. 4, July-Aug. 2001, pp. 537-547.
13. Sakaguchi, N.; Yamanobe, K.; Kitada, Y.; Kawachi, T.; and Koda,
S., “Shear Strength of High-Strength Concrete Members,” Second Inter-
national Symposium on High-Strength Concrete, SP-121, W. T. Hester, ed.,
American Concrete Institute, Farmington Hills, MI, 1990, pp. 155-178.
14. Watanabe, F., and Kabeyasawa, T., “Shear Strength of RC Members
with High-Strength Concrete,” High-Strength Concrete in Seismic Regions,
SP-176, C. W. French and M. E. Kreger, eds., American Concrete Institute,
Farmington Hills, MI, 1998, pp. 379-396.
15. Maruta, M., “Shear Capacity of Reinforced Concrete Column Using
High Strength Concrete,” Invited Lecture in the 8th International Sympo-
sium on Utilization of High-Strength and High-Performance Concrete,
Tokyo, Japan, Oct. 27-29, 2008, pp. 403-408.
16. Kurniawan, D. P., “Shear Behavior of Reinforced Concrete Columns
with High Strength Steel and Concrete under Low Axial Load,” MS thesis,
National Taiwan University of Science and Technology, Taipei, Taiwan,
2011, 341 pp.
17. Takami, S., and Yoshioka, K., “Shear Strength of RC Columns
Using High-Strength Concrete,” Summaries of Technical Papers of Annual
Meeting, Structures IV, Architectural Institute of Japan, Tokyo, Japan,
1997, pp. 25-26. (in Japanese)
18. Takaine, Y.; Nagai, S.; Maruta, M.; and Suzuki, N., “Shear Perfor-
mance of RC Column Using 200 N/mm2 Concrete,” Summaries of Tech-
nical Papers of Annual Meeting, Structures IV, Architectural Institute of
Japan, Tokyo, Japan, 2010, pp. 295-296. (in Japanese)
19. Kuramoto, H., and Minami, K., “Experiments on the Shear Strength
of Ultra-High Strength Reinforced Concrete Columns,” Proceedings of the
Tenth World Conference on Earthquake Engineering, Madrid, Spain, July
1992, pp. 3001-3006.
20. Aoyama, H., Design of Modern Highrise Reinforced Concrete Struc-
tures, Imperial College Press, London, UK, 2001, 442 pp.
21. Shinohara, Y.; Kubota, T.; and Hayashi, S., “Shear Crack Behaviors
of Ultra-High-Strength Concrete Columns (Part 1 and Part 2),” Summaries
of Technical Papers of Annual Meeting, Structures IV, Architectural Insti-
tute of Japan, Tokyo, Japan, 2008, pp. 605-608. (in Japanese)
22. Akihiko, N.; Kuramoto, H.; and Koichi, M., “Shear Strength and
Behavior of Reinforced Concrete Columns Using High-Strength Concrete
of σB = 1200 kgf/cm2 (Part 1 and Part 2),” Proceedings of Architectural
Institute of Japan, 1990, pp. 53-60. (in Japanese)
23. Sibata, M.; Kanasugi, H.; Uwada, M.; Ooyama, H.; and Yamashita,
Y., “Experimental Study on Shear Behavior of Reinforced Concrete
Columns Using High-Strength Shear Reinforcement of 8000 kgf/cm2 Grade
(Part 4),” Summaries of Technical Papers of Annual Meeting, Structures IV,
Architectural Institute of Japan, 1997, pp. 7-8. (in Japanese)
24. McHenry, D., and Karni, J., “Strength of Concrete under Combined
Tensile and Compressive Stress,” ACI Journal Proceedings, V. 54, No. 4,
Apr. 1958, pp. 829-839.
25. Kupfer, H.; Hilsdorf, H. K.; and Rusch, H., “Behavior of Concrete
under Biaxial Stresses,” ACI Journal Proceedings, V. 66, No. 8, Aug. 1969,
pp. 656-666.
26. Hussein, A., “Behavior of High-Strength Concrete under Biaxial
Loading Conditions,” PhD thesis, Memorial University of Newfoundland,
St. John’s, NL, Canada, Apr. 1998, 245 pp.
27. Hussein, A., and Marzouk, H., “Behavior of High-Strength Concrete
under Biaxial Stresses,” ACI Materials Journal, V. 97, No. 1, Jan.-Feb.
2000, pp. 27-36.
28. Hampel, T.; Speck, K.; Scheerer, S.; Ritter, R.; and Curbach, M.,
“High-Performance Concrete under Biaxial and Triaxial Loads,” Journal of
Engineering Mechanics, ASCE, V. 135, No. 11, 2009, pp. 1274-1280. doi:
10.1061/(ASCE)0733-9399(2009)135:11(1274)
29. Paulay, T., and Priestley, M. J. N., Seismic Design of Reinforced Concrete
and Masonry Buildings, John Wiley & Sons, Inc., New York, 1992, 768 pp.
30. Zheng, W.; Kwan, A. K. H.; and Lee, P. K. K., “Direct Tension Test of
Concrete,” ACI Materials Journal, V. 98, No. 1, Jan.-Feb. 2001, pp. 63-71.
31. Tureyen, A. K., and Frosch, R. J., “Concrete Shear Strength: Another
Perspective,” ACI Structural Journal, V. 100, No. 5, Sept.-Oct. 1996,
pp. 609-615.
32. Sezen, H., and Moehle, J. P., “Shear Strength Model for
Lightly Reinforced Concrete Columns,” Journal of Structural Engi-
neering, ASCE, V. 130, No. 11, 2004, pp. 1692-1703. doi: 10.1061/
(ASCE)0733-9445(2004)130:11(1692)
219
 NOTES:

220 ACI Structural Journal/March-April 2015

 ACI STRUCTURAL JOURNAL TECHNICAL PAPER
Title No. 112-S20

Experimental Investigations on Prestressed Concrete

Beams with Openings
by Martin Classen and Tobias Dressen

Due to the needs of sustainability, there are efforts to develop inno-

vative integrated floor slabs that feature wide spans, and enable
a variable arrangement of building services in the construction
height to allow for buildings with adaptive floor layouts and high
flexibility of use. These integrated floor slab concepts usually
require large web openings in the structural bending elements. The
influence of openings on the load-bearing capacity and deforma-
tion behavior of double-T-shaped concrete beams with prestressed
tension chord was investigated within six beam tests. The main test
parameters were concrete strength, amount of vertical reinforce-
ment at the edges of the opening, and location of the opening in the
longitudinal direction. Proper arrangement and dimensioning of
reinforcement, the load-carrying capacity of concrete beams with
openings can attain approximately the same load-carrying capacity
as concrete beams without openings. At ultimate limit state, the
global shear force of beams with openings is mainly carried by the
compression chord. Fig. 1—Prestressed floor slab system for integration of
Keywords: beams; high-strength concrete; integrated floor slab system; building services.
prestressed concrete; shear force; ultimate strength; web openings.
(flooring/supporting structure/building services and installa-
tions/suspended ceiling) and to dissolve the compact cross
INTRODUCTION
sections of conventional floor slabs into wide-span, slender
The majority of existing buildings have monofunctional
multi-web structures. In Fig. 1, an example of an integrated
properties, characterized by inflexible floor layouts that
floor slab system with a multi-web structure composed of
complicate changes in use (for example, from residential
prestressed concrete beams is shown. The precast floor slab
to office building or vice versa) or they are incompatible to
elements are designed for spans up to 16 m (630 in.) and
current architectural requirements or new technical inno-
a service load of 5 kN/m2 (0.725 psi). The provided space
vations. This often leads to demolishing or substantially
between the webs is used for the integration of building
restructuring such buildings long before they reach their
services. The dimensions of the cross section are detailed
economic lifetime. To exploit the buildings’ full economic
in Fig. 1. Openings in the web of the girders allow for flex-
lifetime, adaptive structural systems with a high degree
ible arrangements in all directions. Ease of access to the
of flexibility should be developed. Wide-spanning floor
installation floor and a convenient installation and main-
slab systems with integrated building services can make a
tenance of the components from above is achieved by the
compromising contribution.1-5
use of removable cover panels placed on top flanges of the
These integrated floor slabs provide wide spans for high
concrete beams. Revision openings allow for maintenance
flexibility and adaptability to allow for conversions of use
and minor modifications. To make use of innovative floor-
without significant modification of the building structure.
slab concepts, the technical feasibility needs to be proven.
Beside static aspects, the choice of the floor slab structure has
Therefore, the load-bearing and deformation behavior of the
an impact on numerous building properties. Floor slabs not
integrated concrete slab5,7 was investigated in a comprehen-
only fulfill load-bearing and bracing functions, they create
sive experimental study that was aimed at systematically
the separation between adjoining functional units and, thus,
exploring the impact of large web openings in prestressed
influence the planimetry, building services, physical prop-
concrete beams. The results of these experimental investiga-
erties of the building, and economic and ecological impact
tions are presented in this paper.
of the structure. Hence, the profile of requirements6 for inte-
grated floor slabs includes issues from the fields of structural
engineering, architecture, manufacturing, fire protection, ACI Structural Journal, V. 112, No. 2, March-April 2015.
building physics, dismantling, and recycling. MS No. S-2014-188.R, doi: 10.14359/51687302, received May 27, 2014, and
reviewed under Institute publication policies. Copyright © 2015, American Concrete
A general approach to design-integrated slab systems Institute. All rights reserved, including the making of copies unless permission is
obtained from the copyright proprietors. Pertinent discussion including author’s
is to break up the conventional additive ceiling assembly closure, if any, will be published ten months from this journal’s date if the discussion
is received within four months of the paper’s print publication.

ACI Structural Journal/March-April 2015 221

Fig. 2—(a) Global internal forces at opening; and (b) free-body diagram with local internal forces.
RESEARCH SIGNIFICANCE
While significant research on reinforced concrete beams
with large web openings has been carried out in the past,
there are few test data on prestressed concrete beams
available. For that reason, experimental investigations on
prestressed beams with filigree cross sections and a wide
bottom flange made of high-strength concrete are presented
in this paper. Furthermore, the applicability of different theo-
retical approaches on the tested beams is demonstrated and
conclusions for a safe design of prestressed concrete beams
with large web openings are derived.
THEORETICAL BASICS
Load-bearing behavior of prestressed concrete
beams with large web opening
Concrete chords in prestressed concrete beam webs weak-
ened by large openings need to carry increased shear forces.
This leads to the formation of a local load-carrying mech-
anism in the web opening, which has a significant impact
on the global behavior of these slender beams in terms of
their ultimate strengths and deformations. The structural
behavior of the bottom and top concrete chords in the area of
a web opening is characterized by combined moment-shear
stresses. In the literature,8 different mechanical models are
given, which usually suggest a simple Vierendeel truss as
a static system of the opening area. Commonly, the chords
are assumed to be restrained to the adjacent, homogeneous
beam sections. Figure 2 shows the opening with its global
internal forces as well as the simplified static system with
the local internal forces. To calculate the local stresses of the
concrete chords and provide adequate construction design, a
mechanical relationship between global internal forces and
local shear forces in the concrete chords needs to be derived.
Besides these local shear forces in the chords, the root of the
secondary bending moment needs to be determined with its
position in the longitudinal direction (point of contraflexure).
Although in some investigations9,10 significant variations
regarding this point of contraflexure have been detected, for
usual loading states, its position generally can be assumed to
lie in the center of the web opening, as presented in Fig. 2.
222
The normal forces in the concrete chords depend on the
global acting moment and the prestressing force, as shown in
Fig. 2. For slender chords with high normal forces, second-
order effects have to be taken into account. In the adjacent,
homogeneous parts of the beam, the internal shear forces are
transferred by compression and tension struts.10
Based on previous experimental investigations of concrete
beams with web openings, different failure modes were
defined. Kennedy and Abdalla11 stated four different failure
modes for prestressed beams with large web openings:
1. Bending failure of the chords: To realize plastic bending
hinges at the edges of the opening, a sufficient rotation
capacity of the bottom and top chord has to be ensured.
Assuming an adequate design for shear, plastic hinges occur
by reaching the yield surface of the interaction diagram
between bending moment and normal force.12
2. Compression-shear failure: Due to compression forces
caused by the global and local secondary bending moment
at the ends of the compression chord, a sudden compres-
sion-shear failure may occur. Contrary to the Failure Mode 1
listed previously, the transverse reinforcement of the chord
does not yield. Thus, no plastic hinges arise.
3. Shear failure in opening chords: High shear forces may
cause a local shear failure of the slender concrete chords.
Usually, a local bending-shear failure can be stated.
4. Tension failure: Analogous to the compression chord,
tensile forces due to the global bending moment and
secondary moments are acting on the bottom chord. If these
forces cause tensile stresses in the chord that exceed the
concrete tensile strength, cracks occur over the full depth of
the bottom chord. This failure typically appears at openings
in regions of high global bending moment.
To provide a safe design of prestressed concrete beams with
web openings, a suitable failure criterion has to be defined.
According to Neff and Ehmann13 failure occurs with the first
plastic hinge in one of the opening corners. Mansur and Tan14
allow for interaction of plastic hinges at all of the four edges
at failure state. Plastic hinges can only occur if premature
failure due to one of the other failure modes (2, 3, and 4 listed
previously) is eliminated. This can be ensured by an adequate
arrangement of stirrups and longitudinal reinforcement.
ACI Structural Journal/March-April 2015
Distribution of shear force between bottom and
top chord
Knowing the magnitudes of internal forces, a simple
design of the concrete chords according to corresponding
design codes can be performed. In the following, different
approaches to calculate the distribution of local shear forces
to the chords are presented. Predominantly, these approaches
use geometrical properties of the concrete chords to calcu-
late the carried proportions of internal shear force.
Nasser et al.8 as well as Salam and Harrop15 suggest
a distribution of shear force depending on the ratio of the
cross-sectional areas at the opening (Eq. (1)). This approach
has no mechanical background and is limited to the
uncracked state. However, a good accordance with experi-
mental results was found.
At / b
Vt / b = ⋅ V (1)
At + Ab
where Vt/b is the shear force acting in the top and bottom
chord; and At/b is the cross-sectional area of the top and
bottom chord.
Hottmann and Schäfer10 recommend calculating the distri-
bution of shear force according to the ratio of bending stiff-
nesses’, It/b, of top and bottom chords. Hereafter, the propor-
tion of shear force in the chords results from a linear-elastic
calculation of a framework system, neglecting shear defor-
mation of the beams (Eq. (2)).
It /b
Vt / b = ⋅ V (2)
It + Ib
Kennedy and Abdalla11 developed an empirical approach
that considers both cross-sectional area as well as bending
stiffness of the chords (Eq. (3) and (4)). It was calibrated by
finite-element calculations. For I-beams, good accordance
with numerical results was observed. For T-beams, however,
the distribution of shear force according to the bending stiff-
ness of the chords led to better results.
Vt = V – Vb (3)
Ab ⋅ I b
Vb = ⋅ V (4)
Ab ⋅ I b + At ⋅ I t
Due to cracking of the concrete chords at ultimate limit
state (secondary bending moments), the stiffness of the
cross sections in the web opening corners is reduced. The
decrease of stiffness should be considered in the calculation
of shear force distribution between the bottom and top chord.
Barney16 described an approach using effective stiffnesses
Ii,t/b. In the cracked state of the concrete, the shear force
is distributed according to Eq. (5) and (6). The approach
usually provides conservative results (Vb + Vt > V). If a full-
depth crack appears at the bottom (tensile) chord, the entire
shear has to be carried by the top (compression) chord.
ACI Structural Journal/March-April 2015
I i ,t
Vt = ⋅ V (5)
I i ,t + I i ,b ( cr )
I i ,b
Vb = ⋅ V (6)
I i ,t + I i ,b
where Ii,b(cr) is the effective bending stiffness of the cracked
bottom chord.
Kennedy and Abdalla17 also suggest a similar method. In
the uncracked state, the shear force is distributed depending
on the ratio of cross-sectional areas and bending stiffnesses
of the chords (Eq. (3) and (4)). It is assumed that concrete
cracking starts in the bottom chord due to additional tensile
forces caused by the global bending moment. After comple-
tion of the crack propagation, the bottom chord is unable to
carry additional shear forces. Consequently, the local shear
force of the bottom chord remains constant, while additional
shear forces have to be carried by the top chord. Assuming
a concrete tensile strength ft, the acting global shear force
VOm,cr at cracking of the bottom chord is calculated by Eq.
(7) for the static system presented in Fig. 2.
N P ,b
ft −
Ab
VOm , cr = (7)
 lOm Vb lO 1 
 z A + V ⋅ 2 ⋅ W 
G b b
where NP,b is the normal force in the bottom chord due to
prestressing forces; lOm is the distance between the midspan
of the opening and support; and Wb is the section modulus of
the bottom chord.
The corresponding shear force of the bottom chord results
from Eq. (4). In the case of full-depth cracks, the entire shear
force is carried by the top chord.
Neff13 developed a concept to calculate the distribution of
shear forces in the chords based on research by Ehmann,13
which also takes into account the influence of the local
normal forces in the bottom and top chord, depending on the
global bending moment and the prestressing force.
Ιb,eff = χb · Ib = [15(ρl,b + ρ′l,b) – 0.25 + (nb – 0.5)2] · Ib (8)
Ιt,eff = χt · It = [15(ρl,t + ρ′l,t) + 0.4 – (nt – 0.5)2] · It (9)
Nb / t
nb / t = (10)
Ab / t � f c
where ρl,t/b (ρ′l,t/b) are the flexural tensile (compression) rein-
forcement ratios of top and bottom chord, respectively. In
addition to the aforementioned methods, an approach for the
effective stiffness of the chord at failure was developed taking
into account the effects of normal forces on the bending stiff-
ness. Beside these complex calculations, simple assumptions
for the distribution of the shear force may be taken. Leon-
hardt18 suggests the following distribution at failure
223
Fig. 3—Dimensions of cross section and web opening between load introduction and support. (Note: Dimensions in mm (in.).)
Vt = 0.8 ~ 0.9 · V (11)
Vb = 0.1 ~ 0.2 · V (12)
The presented approaches have different levels of complexity
and usability. In the following, the specific qualities and
fields of application of the different formulas are evaluated
and recommendations are deduced:
• Equations (1), (11), and (12) are simple. They can be
used for reinforced concrete beams and allow for an
initial rough estimate of the distribution of shear force
to the chords.
• Equations (2), (3), and (4) may be applied to reinforced
concrete beams with openings in Uncracked State I.
Equations (3) and (4) are recommended for I-sections,
whereas Eq. (2) can be used for T-beams.
• For reinforced concrete beams in Cracked State II,
more sophisticated approaches such as Eq. (5) to (7)
are recommended to determine the distribution of shear
force to the chords.
• Equations (8) to (10) have been developed for prestressed
concrete beams. Here, the longitudinal stress state in the
chords is taken into account.
EXPERIMENTAL PROGRAM
In six beam tests, the influence of large web openings on
the load-bearing and deformation behavior of double-T-
shaped concrete beams with prestressed tension chord was
investigated. Based on the initial beam Test DE-1.1, indi-
vidual parameters were varied in each of the other five tested
beams. In Test DE-2.1, the space between the support and
opening was reduced to 350 mm (13.8 in.). DE-2.2 had a
greater amount of vertical reinforcement at the edge of the
opening. With Test DE-3, the influence of higher concrete
strengths has been investigated, and high-strength concrete
with a cylinder strength of 110 MPa (15,954 psi) instead of
normal-strength concrete’s 65 MPa (9427 psi) was used.
Test DE-1.2 served as a reference test without web opening.
Test specimens, material properties, and beam
fabrication
For practical reasons, the experiments were performed
at spans between 8 and 5.5 m (315 and 217 in.). The test
specimens had an overall length of 8300 mm (328 in.) and a
double-T-shaped cross section. The openings had a length of
600 mm (23.6 in.) and a height of 250 mm (9.8 in.), which
represents approximately half of the overall beam depth.
The dimensions of the cross section and the web opening
224
are presented in Fig. 3. The double-T-shaped concrete beams
featured several arrangements of reinforcement (prestressing
tendons, longitudinal reinforcement, and stirrups). To
ensure practical conditions, the reinforcement and prestress
were designed for an additional dead load of 1.5 kN/m2
(0.218 psi) and a live load of 5.0 kN/m2 (0.725 psi) acting on
a span of 16 m (630 in.). The calculation and dimensioning
of reinforcement is based on current standards. The shear
force reinforcement in the web of the concrete beam was
realized by vertically positioned single reinforcement bars
6 mm (0.24 in.) in diameter at 160 mm (6.3 in.). The longi-
tudinal reinforcement in the compression chord consisted
of four bars 20 mm (0.79 in.) in diameter. Bars 10 mm
(0.39 in.) in diameter were used as longitudinal reinforce-
ment in the tensile chord. Closed stirrups 6 mm (0.24 in.)
in diameter at 160 mm (6.3 in.) were arranged in the top
and bottom chord, respectively. Furthermore, a vertical
reinforcement consisting of single bars 10 mm (0.39 in.)
in diameter were placed close to the edges of the opening.
The amount of vertical reinforcement at the opening edge
was designed to anchor the entire acting shear force (As,A,S
and As,A,L = VEdfyd  = 157 mm2 [0.243 in.2]). In some tests,
As,A,L was varied between 157 and 236 mm2 (0.243 and
0.366 in.2). The overall concrete cover amounted to 20 mm
(0.79 in.). The specimens were prestressed by 10 tendons
with a diameter of 11 mm (0.43 in.) and a concrete cover
of 50 mm (1.97 in.). Additional reinforcement was placed
at the ends of the specimens to avoid splitting failure due
to prestressing forces. The arrangement of reinforcement is
detailed in Fig. 4.
In all specimens, German reinforcing steel BSt 500 and
prestressing steel St1570/1770 were used. The material
properties of the used reinforcement steels were measured
in tensile tests on three reinforcement samples that were
taken from the same charge of reinforcement (Table 1). The
tendons (prestressing steel) had a yield stress fsy of 1638 MPa
(237.6 ksi) measured at a strain of 0.1% and tensile strength
fsu of 1890 MPa (274.1 ksi). The concrete strength was
varied between cylinder strengths of 65 and 111 MPa
(9427 and 16,099 psi). Test cylinders (150 x 300 mm
[5.9 x 11.8 in.]), cubes (150 x 150 mm [5.9 x 5.9 in.]), and
prisms (700 x 150 x 100 mm [27.6 x 5.9 x 3.9 in.]) were cast
from each mixture to determine the concrete compression
and tensile strengths as well as Young’s modulus. Table 2
summarizes the properties of the used concretes.
The specimens were produced in two steps. After
prestressing of the tendons, the bottom chord was concreted.
The concrete facing was roughened at the joint between
ACI Structural Journal/March-April 2015
the bottom chord and the web. On the following day, the
web and the top chord were fabricated in a second step.
The concrete was mixed at the laboratory of the Institute
of Structural Concrete at RWTH Aachen. The specimens
remained in the formwork covered with a polyethylene sheet
for 1 day and were kept in the laboratory environment until
testing at an age of 28 days. After a hardening time of 6 days,
the specimens were prestressed with a prestressing force P0
of 0.79 MN (177.6 kip), which is equivalent to a tension in
the tendons of 1000 MPa (145 ksi).

Table 1—Material properties of reinforcement Test setup and instrumentation

Bar size, A0, mm 2
fy, MPa ft, MPa Es, GPa Two experiments were performed on each specimen.
mm (in.) (in.2) (ksi) (ksi) A10, % (ksi) Interference between the tests was avoided by a sufficiently
605 653 201.3 large distance between the openings and adequate bedding
6 (0.236) 28 (1.10) 17.3 conditions. The span of the first three-point bending tests
(87.75) (94.71) (29,196)
10 499 626 201.3
was 8 m (315 in.). After the failure of each test, the supports
78 (3.07) 19.0 were moved and a second test was carried out, but with a
(0.394)* (72.37) (90.79) (29,196)
reduced span of 5500 mm (216 in.). To prevent bending
10 610 704 206.0
(0.394)†
77 (3.03)
(88.47) (102.11)
20.0
(29,878) failure, a shear slenderness of a/d = 4.76 was chosen.
The beam details are presented in Fig. 5. The beams were
307 544 645 200.4
20 (0.787)
(12.09) (78.90) (93.55)
23.9
(29,066)
supported on steel plates with slide bearings. The load was
transferred from the hydraulic jack to the beam through a
*
Stirrups.
steel plate. All beams were monotonically loaded up to the
†
Longitudinal reinforcement.
predicted service load. Afterward, 50 load cycles between 35
Notes: A0 is cross-sectional area at beginning of tensile test; fy is yield strength; ft is
tensile strength; A10 is strain at failure; and Es is Young’s modulus. and 130% of the service load were executed. To predict the

Table 2—Details of test specimens and test results

Test fc,cyl, MPa (psi) fct,sp, MPa (psi) Ec, GPa (ksi) As,A,L, mm2 (in.2) lOm, mm (in.) Vmax, kN (kip) Vcalc, kN (kip) Vcalc/Vmax
DE-1.1 65.7 (9529) 3.84 (556.9) 35.8 (5192) 157 (0.243) 950 (37.4) 113 (25.4) 122 (27.4) 1.08
DE-1.2 64.2 (9311) 3.78 (548.2) — — — 107 (24.0) — —
DE-2.1 68.3 (9906) 3.64 (527.9) 36.1 (5236) 157 (0.243) 650 (25.6) 101 (22.7) 116 (26.1) 1.15
DE-2.2 68.8 (9979) 4.16 (603.4) 37.1 (5381) 236 (0.366) 950 (37.4) 117 (26.3) 120 (27.0) 1.03
DE-3.1 110.6 (16,041) 6.08 (881.8) 49.6 (7194) 157 (0.243) 950 (37.4) 135 (30.3) 167 (37.5) 1.24
DE-3.2 110.3 (15,998) 6.14 (890.5) 47.5 (6889) 236 (0.243) 950 (37.4) 143 (32.1) 167 (37.5) 1.17

Notes: fc,cyl is cylinder compression strength; fct,sp is splitting tensile strength; Ec is Young’s modulus of concrete; As,A,L is vertical reinforcement at opening edges; lOm is distance
between support and midspan of opening; Vmax is maximum shear force; and Vcalc is calculated load-bearing shear force for local bending failure.

Fig. 4—Reinforcement details of Test DE-1.1.

ACI Structural Journal/March-April 2015 225

Fig. 5—Testing procedure, beam spans, and loading.
Fig. 6—Test setup of Test DE-2.
deflections of the beams at the service load level, different
approaches were used.18,19 From a loading of 70% of the
predicted load-bearing capacity, the beams were loaded
under displacement control until failure. Figure 6 shows the
test setup for Test DE-2.
The deflections of the beams at the load introduction were
measured using a linear variable transducer (WAxx). The
steel and concrete strains at surface were measured by elec-
trical resistance strain gauges. To determine the curvature
at each edge of the opening, the steel strain of the longi-
tudinal reinforcement (SLxx) and the strain of concrete in
compression (using strain gauges) and tension (using linear
transducers) were measured. The strains of the stirrups
(SBxx) were recorded at the four corners of the opening,
the midspan of the top chord, and in the web. The strains at
the concrete surface were measured in three directions in the
midspan of the top chord. The arrangement of the instrumen-
tation is shown in Fig. 7.
TEST RESULTS AND DISCUSSION
Failure characteristic
In the following, results of the performed beam tests are
described. The global failure of each beam test (except
reference Test DE-1.2) was initiated by a local failure in the
area of the web opening. Tests DE-1.1 and DE-2.2 failed
by exceeding the bending capacity of the top chord in one
web-opening corner (Failure Mode 1). In tests with higher
concrete strength (DE-3.1 and DE-3.2) and the test with
226
reduced distance between the opening and support (DE-2.1),
the bending failure in the opening corner occurred in combi-
nation with a shear failure in the concrete joint interface
between web and bottom chord. The beam tests with open-
ings reached the magnitude of the load-bearing capacity of
the reference test (DE-1.2), which failed by a transverse
shear crack in the slender web of the cross section. The ulti-
mate shear forces and main test parameters are summarized
in Table 2.
Load-deflection characteristics
The load-deflection curves of all tests are presented in
Fig.  8. To eliminate the influence of different spans, the
deflections were related to the span and plotted over the
applied shear force V. Initially, the first test on each spec-
imen (DE-1.1, DE-2.1, and DE-3.1) had a small impact on
the load averted opening, on which the second beam test
(DE-1.2, DE-2.2, and DE-3.2) was performed. In fact, the
averted opening was preloaded by a shear force of approxi-
mately 30% of the predicted load-bearing capacity. Conse-
quently, certain initial cracks of low width were detected in
the tensile chord and in the joint between the web and the
bottom chord. Due to that preloading impact, the stiffness
of each second beam test (DE-2.2 and DE-3.2) was lower
compared to the first one. The tests with higher concrete
strength (DE-3.1 and DE-3.2) had higher stiffness and load-
bearing capacities. A sufficient ductility and good load-
bearing behavior was observed throughout the entire test
series. The load-bearing capacities, however, were reached
at different vertical deflections.
Cracking characteristics
The cracking behavior of the beam is strongly influenced
by the secondary bearing mechanism of the web opening.
The crack patterns of the tested beams after failure are
presented in Fig. 9. Caused by secondary bending moments
at the opening edges, the first bending cracks appeared in
the top chord of the concrete beam. The cracking of the
top chord started simultaneously on its lower surface at
the opening edge next to the load introduction and on its
upper surface at the opening edge next to the support. With
increasing load, transverse shear cracks occurred in the
slender web of the beam around the opening. The shear
cracking began approximately at the service load level.
At the moment of their formation, the shear cracks had a
width of 0.1 to 0.2 mm (0.004 to 0.008 in.). The angle of the
ACI Structural Journal/March-April 2015
Fig. 7—Arrangement of strain measurements: (a) concrete strains; and (b) reinforcement steel strains.

Fig. 8—Shear force V over related deflection f/l3: Test: (a) DE-1; (b) DE-2; and (c) DE-3.

Fig. 9—Failure pattern of test specimens.

ACI Structural Journal/March-April 2015 227

Fig. 10—Typical crack formation: Test: (a) DE-2.1; (b) DE-2.2; and (c) DE-3.2.
Fig. 11—Strain distribution at opening corners for Test DE-1.1.
shear cracks in the web varied between 36 and 45 degrees
in Tests DE-1.1 to DE-2.2. For a higher concrete strength,
shear crack angles of 31 degrees (DE-3.1) and 35 degrees
(DE-3.2) were measured. With increasing load, lower shear
crack angles (~25 degrees) were found in the web. Near the
support of the beam, the diagonal shear cracks of the web
encroached upon the bottom chord (Fig. 10(a)).
The high compressive stresses in the top chord redound to
the formation of a very flat concrete strut, which allows for
the transfer of shear forces above the opening. Consequently,
at the ultimate limit level, shear cracks with a very flat incli-
nation occurred in the top chord of the beam, which grew
from the opening edge toward load introduction. Here, the
cracking developed in an almost horizontal direction. In Fig.
10(b), the horizontal cracks of the top chord are illustrated.
Finally, the redistribution of the shear force in the region
of the web opening effected high pullout forces in the
vertical reinforcement next to the openings edge (As,A,S and
As,A,L). This phenomenon causes concentrated stresses in
the adjacent concrete. Consequently, a gradual separation
in the interface between concrete web and bottom chord
occurred, which finally resulted in a concrete pryout allo-
cated at the edge of the opening (Fig. 10(c)). This behavior
228
was observed in Tests DE-2.1, DE-3.1, and DE-3.2, and is
well known from experimental investigations on filigree
prestressed composite beams.20-22 Besides concrete pryout,
vertical cracks occurred at the edge of the web opening in
the ultimate limit state, which resulted from the activation of
the vertical reinforcement. In Fig. 10(b), the vertical cracks
at the opening edge are shown.
Strain of opening corners
In Fig. 11, the strain distribution of the four opening
corners is plotted at different load levels for Test DE-1.1.
The strain distribution was determined by the measured
compression and tension strain at the concrete surfaces as
well as the strain of longitudinal reinforcement. The global
bending moment and prestressing force led to compression
normal forces in the top chord. The strains resulting from
the release of pretension have not been measured separately,
as their ordinates are small compared to the strains resulting
from bending. At ultimate limit state, a compression strain
of 0.002 was reached in the opening of Corners 1 and 2
(Fig. 11). Nevertheless, significant tensile stresses occurred
in the top chord, resulting from the secondary bending
moments. In the bottom chord, which was exposed to tensile
ACI Structural Journal/March-April 2015
Fig. 12—Curvature at opening corners over shear load (Test DE-3.2).
stresses by global bending, the compression zone only
amounted to 20 to 30 mm (0.79 to 1.18 in.), and significantly
smaller magnitudes of compression strain were measured.
The strain distributions in the corners of the web opening
allow for the calculation of the curvatures κm. In Fig. 12, the
curvature κm of each opening corner is plotted over the shear
force V (for Test DE-3.2). At a load level of 60 to 70% of
the load-bearing capacity, a significant increase of curvature
was observed. At least at one corner of the web opening, the
formation of a plastic hinge occurred at failure state.

Influence of vertical reinforcement at opening edges

At the edges of the web opening, the shear force is
distributed to the concrete chords above and below the web
opening. To prevent failure of the connection between chords
and the solid beam, vertical reinforcement at the opening
edges needs to be considered. Generally, the vertical rein-
Fig. 13—Loading of vertical reinforcement at opening edges.
forcement is assumed to hang up and anchor the shear force
carried by the compression chord of the beam. In the liter-
ature, different approaches14,16,18 for the dimensioning of
this vertical reinforcement are given. In Tests DE-1.1 and
DE-3.1, the vertical reinforcement was dimensioned to carry
the entire applied shear force at ultimate limit state (As,A,S
and As,A,L = VEd/fyd; 2 x 10 mm [3.9 in.] diameter reinforce-
ment bars). In Tests DE-2.2 and DE-3.2 the degree of vertical
reinforcement furthermore was increased to 150% (3 x 10
mm [3.9 in.] diameter bars). Figure 13 shows the ratio of
ultimate shear force Vmax and maximum tensile force of the
vertical reinforcing bars at yield stress. Both tests with low
reinforcing ratios led to load factors higher than 1.0 (1.15
and 1.37 for Tests DE-2.2 and DE-3.2, respectively). Taking
into account the crack pattern and the geometrical properties
Fig. 14—Strain of vertical reinforcement at opening edges.
of the opening, alternative load transfer (arch mechanism)
can be eliminated. Thus, at ultimate load, an appreciable in Fig. 14, these findings are confirmed. Using two bars of
part of the shear force was carried by the bottom chord. The 10 mm (0.39 in.) diameter, an ultimate strain of approxi-
theoretical load-bearing capacity of the reinforcement at the mately 0.003 was reached in the tests so that the stresses in
edges of the opening was not reached for Tests DE-2.2 and the vertical reinforcement reached the yield strength (499
DE-3.2 with a higher reinforcement ratio. Considering a sole N/mm2 [72.37 ksi] according to Table 1). The strains in the
transfer of the shear force by the top chord, the load-bearing vertical reinforcement of Test DE-2.2 with a higher rein-
capacity of the reinforcement was not attained. Due to the forcement ratio stayed significantly below this value. Here,
higher load level, higher stress of the reinforcement at the only an ultimate strain of 0.0013 was reached, which corre-
opening edges was reached in tests with higher concrete sponds to a tensile stress of 261 N/mm2 (37.9 ksi) in the
strength. Comparing the measured strains of reinforcement vertical reinforcement bars.

ACI Structural Journal/March-April 2015 229

Influence of openings near supports and influence
of concrete strength
The influence of openings with low distances to supports
was investigated by comparing Test DE-2.1 with DE-1.1
and DE-2.2, respectively. The load-bearing capacity of the
test with openings near the support of the beam was 15%
lower than that of both other tests. At up to 80% of the load-
bearing capacity, no differences between the tests could be
observed. After that, the deflection of Test DE-2.1 dispropor-
tionately increased with a small increase of applied load. At
the same time, longitudinal cracks at the joint between web
and bottom chord as well as the end of the beam occurred.
In the following, the shear force in the joint between web
and bottom chord was mostly carried by the vertical rein-
forcement in terms of dowel action. However, a combined
failure was observed. On the one hand, significant slip in the
reinforced joint between web and bottom chord occurred,
which led to shear failure in interface. On the other hand, the
bending moment capacity of the top chord was reached at the
same time in terms of yielding of longitudinal reinforcement.
The use of higher-strength concrete (fc = 110 MPa
[15,954 psi]) led to a 20% higher load-bearing capacity
compared to tests with a concrete strength of fc = 67 MPa
(9717 psi). Just as Test DE-2.1 with reduced distance
between support and opening, in Tests DE-3.1 and DE-3.2,
a significant slip between web and bottom chord was found
(shear failure in the concrete joint).
Fig. 15—Exemplary determination of load-bearing capacity
for Test DE-1.1.
Fig. 16—Comparison of acting and resisting shear stress in interface.
230
Comparison of experiments and theoretical
approaches
To quantify the influence of shear failure in the concrete
joint interface, and local bending failure of the concrete
chords, the experimental results are compared to theoretical
load-bearing capacities. For that reason, appropriate theo-
retical failure criteria have to be checked for both observed
failure modes. The criterion, which provides the smaller
carrying capacity, is decisive for the design and identifies
the relevant failure mode.
The criterion for bending failure of the chords uses
bending moment/normal force-interaction diagrams. First of
all, the distribution of shear force to the chords was calcu-
lated by Eq. (5) and (6), depending on their effective flexural
stiffness. The point of contraflexure was considered to lie
at midspan of the opening, and the combined moment and
normal force of the chords was calculated in relation to the
applied load P. Plotting the loading path in the interaction
diagram of each chord, the load-bearing capacity was found
at the intersection point of loading path and the interaction
surface. Therefore, the load-bearing capacity for bending
failure of the chords is predicted, while all other failure
modes are excluded. In Fig. 15, the procedure is presented
for the top chord of Test DE-1.1. A good accordance between
predicted load-bearing capacity and maximum experimental
load was found for Tests DE-1.1 and DE-2.2 (refer to Vcalc,
Table 2), which had a local bending failure of the chords.
In Tests DE-2.1, DE-3.1, and DE-3.2, a failure of the
concrete joint interface between web and bottom chord
occurred. To quantify the influence of failure in the concrete
joint, a second failure criterion had to be checked. For that
reason, the maximum resistant shear stresses were compared
to the acting shear stresses in the concrete joint. Figure 16
shows the ratio of the shear resistance τRm of a reinforced
interface according to Eurocode 223 to the shear stress in
the interface caused by loading τE at ultimate limit state.
The shear resistance τRm was estimated by using the mean
values of the material properties (Table 1), assuming a rough
interface. For Tests DE-2.1, DE-3.1, and DE-3.2 with exper-
imental failure of the concrete interface, values lower than
1.0 were calculated. Therefore, the results of the calculation
confirmed the experimental observations.
Generally, the failure of the joint between web and bottom
chord reduces the load-carrying capacity compared to the
carrying capacity of beams with local bending failure.
ACI Structural Journal/March-April 2015
Consequently, for prestressed beams with openings and
concreting joints, the well-known failure modes by Kennedy
and Abdalla11 should be supplemented by a shear failure
criterion of the concrete joint interface, especially for high-
strength concrete. Neglecting this failure may cause a deficit
in safety because the calculated capacities Vcalc for a local
bending failure exceed the experimental capacities for inter-
face failure by approximately 20% (Table 2).

Experimental distribution of shear force in chord

members
To determine the shear force acting in top and bottom
chords, respectively, the measurements of the strain gauge
rosette applied at midspan of the top chord was evaluated.
According to Twelmeier,24 the shearing strain γ was deter-
mined from the measured principal strains. Knowing the
shear modulus, the shear stress was calculated considering
Hooke’s law (Eq. (13)), whereat the impact of global loading
on the shear modulus was considered by Eq. (14) with a
Poisson’s ratio ν of 0.2

Ec
τ = γG = γ (13)
2(1 + ν)

M global (1 − ν2 )
M global = σ x Az ⇒ Ec = (14)
Az (e x + νe x )
The acting shear force in the top chord results from the inte-
gration of the shear stress over the cross section. As refer-
ence, the tensile forces in the vertical reinforcement at the
load-allocated edge of the opening were calculated by the
measured strains. In the case of distinctive crack formation,
the sum of these forces equaled the shear force carried by
the top chord. The results of these calculations are presented
in Fig. 17. With increasing load, both methods led to the
same values. As expected, the strains of the reinforcement
were small at low load levels because the beam remained
in an uncracked state. Thus, the shear force of the top chord
was underestimated by the evaluation of the reinforcement’s
strains. The values calculated by the strain gauge rosette on
the concrete surface remained nearly constant over the entire
load range. Similar results were already detected in tests by
Twelmeier24 and Tan.25
Due to the measured compression strain with the strain
gauge applied in a –45-degree direction in Test DE-1.1, a
significantly lower value of the shear force acting in the
top chord was detected. At higher load levels, however,
the values calculated by the strains of the reinforcement lie
between the values of the other tests.
Higher degrees of vertical reinforcement at the edge of
the opening (Tests DE-2.2 and DE-3.2) led to higher shear
forces carried by the top chord compared to the other tests
with only two bars 10 mm (0.4 in.) in diameter as vertical
reinforcement. With the exception of Test DE-3.1, the
measured shear force carried by the top chord was deter-
mined to be approximately 90% of the acting shear force in
ultimate limit state. These results fall between the predicted
values in References 13 and 18.
Fig. 17—Distribution of applied shear force carried by
top chord.
CONCLUSIONS
The results of the experimental investigations confirm the
capability and technical feasibility of the developed floor
slab system as a structural element in multiple-use buildings.
Based on the results of the experimental investigation on
beams with large web openings, the following conclusions
can be drawn:
1. In the uncracked stage (service load), a considerable part
of the global shear force is carried by the bottom (tension)
chord. However, at failure, approximately 90% of the shear
force is carried by the top (compression) chord indepen-
dent of the position of the opening (M/V ratio), the concrete
strength, and the cross section of the vertical reinforcement
at the edge of the opening (shear concentration factor).
2. The load-bearing capacity of the beams is reached with
the formation of a plastic hinge at one of the opening corners
of the top chord. A ductile failure could be stated for all
tests. The use of higher-strength concrete led to an increase
of the load-bearing capacity, but at a subproportional rate
compared to concrete tensile and compressive strength.
3. The load-bearing capacity of the joint between the web
and the bottom chord, due to the manufacturing process, may
be decisive for the global load-bearing capacity when using
high-strength concrete and openings close to the support.
4. For the investigated conditions, the arrangement of
openings between the transition length of the prestressing
steel had no adverse influence on load-bearing capacity
and deformation behavior. All beams with openings under
moment-shear loading led to higher deformations when

ACI Structural Journal/March-April 2015 231

compared to solid beams. The increase is small at service
load level.
5. An increase of the cross-sectional area of the vertical
reinforcement at the edge of the opening led to higher shear
force carried by the top chord for the entire load range. The
failure, however, is not affected if a sufficient amount of
reinforcement is used. The cross-sectional area of vertical
reinforcement was calculated for the anchorage of the entire
acting shear force.
AUTHOR BIOS
Martin Classen is a Research Engineer at the Institute of Structural
Concrete, RWTH Aachen University, Aachen, Germany, where he received
his degree in structural engineering in 2011. His research interests include
the development of integrated floor slab solutions and the structural
behavior of concrete and composite construction.
Tobias Dressen is a Structural Engineer at Kempen&Krause, Aachen,
Germany. He received his PhD in structural engineering from RWTH
Aachen University in 2011. His research interests include the sustainability
of concrete structures and its structural realization.
ACKNOWLEDGMENTS
This paper comprises results of a comprehensive research program titled
“Sustainable Building with Concrete” under coordination of the Deutscher
Ausschuss für Stahlbeton e. V, which was jointly funded by the German
Federal Ministry of Education and Research (BMBF) and third-party
donors. Their support is gratefully acknowledged.
REFERENCES
1. Frangi, A.; Fontana, M.; and Mensinger, M., “Innovative Composite Slab
System with Integrated Installation Floor,” Structural Engineering Interna-
tional, V. 19, No. 4, 2009, pp. 404-409. doi: 10.2749/101686609789846948
2. Hegger, J.; Claßen, M.; Gallwoszus, J.; Schaumann, P.; Weisheim,
W.; Sothmann, J.; Feldmann, M.; Pyschny, D.; Bohne, D.; and Hargus,
S., “Multifunctional Composite Slab System with Integrated Building
Services,” Stahlbau, V. 83, No. 7, 2014, pp. 452-460. doi: 10.1002/
stab.201410170
3. Classen, M.; Gallwoszus, J.; and Hegger, J., “Load-Bearing Behavior
of an Integrated Composite Floor System,” Bauingenieur, V. 89, No. 3,
Mar. 2014, pp. 91-101.
4. Kolleger, J.; Kainz, A. E.; and Burtscher, S. L., “Slab with Integrated
Installations,” Creating and Renewing Urban Structures—Tall Buildings,
Bridges and Infrastructure, 17th Congress of IABSE, Sept. 17-19, Chicago,
IL, 2008, pp. 230-231.
5. Hegger, J.; Dreßen, T.; and Schießl, P. et al., “Nachhaltiges Bauen im
Lebenszyklus,” Bauingenieur, V. 84, July-Aug. 2009, pp. 304-312.
6. Dressen, T., and Classen, M., “Experimentelle Untersuchung an Span-
nbetonträgern mit großen Stegöffnungen,” Bauingenieur, V. 89, No. 9,
Sept. 2014, pp. 359-369.
7. Classen, M.; Gallwoszus, J.; Hegger, J.; Papakosta, A.; Kuhnhenne,
M.; Psychny, D.; and Feldmann, M., “Sustainability Assessment of Long
Span Floor Systems,” Bauingenieur, V. 89, No. 3, Mar. 2014, pp. 125-133.
232
8. Nasser, K. W.; Acavalos, A.; and Daniel, H. R., “Behavior and Design
of Large Openings in Reinforced Concrete Beams,” ACI Journal Proceed-
ings, V. 64, No. 1, Jan. 1967, pp. 25-33.
9. Pessiki, S., and Thompson, J. M., “Experimental Investigation of
Precast, Prestressed Inverted Tee Girders with Large Web Openings,” PCI
Journal, V. 51, No. 6, 2006, pp. 2-17.
10. Hottmann, H. U., and Schäfer, K., “Bemessen von Stahlbetonbalken
und -wandscheiben mit Öffnungen,” Deutscher Ausschuss für Stahlbeton,
Heft 459, Beuth Verlag, Berlin, Germany, 1996.
11. Kennedy, J. B., and Abdalla, H., “Static Response of Prestressed
Girders with Openings,” Journal of Structural Engineering, ASCE, V. 118,
No. 2, 1992, pp. 488-504. doi: 10.1061/(ASCE)0733-9445(1992)118:2(488)
12. Mansur, M. A.; Tan, K.-H.; and Lee, S.-L., “Collapse Loads
of R/C Beams with Large Openings,” Journal of Structural Engi-
neering, ASCE, V. 110, No. 11, 1984, pp. 2602-2618. doi: 10.1061/
(ASCE)0733-9445(1984)110:11(2602)
13. Schnellenbach-Held, M.; Ehmann, S.; and Neff, C., “Untersuchung
des Trag- und Verformungsverhaltens von Stahlbetonbalken mit großen
Öffnungen,” Deutscher Ausschuss für Stahlbeton, Heft 566, Beuth Verlag,
Berlin, Germany, 2007.
14. Mansur, M. A., and Tan, K.-H., Concrete Beams with Openings:
Analysis and Design, CRC Press, Boca Raton, FL, 1999, 224 pp.
15. Salam, A., and Harrop, J., “Prestressed Concrete Beams with Trans-
verse Circular Holes,” Journal of the Structural Division, ASCE, V. 105,
1979, pp. 635-652.
16. Barney, G. B.; Corley, W. G.; Hanson, J. M.; and Parmalee, R. A.,
“Behavior and Design of Prestressed Concrete Beams with Large Web
Openings,” PCI Journal, V. 22, No. 6, 1977, pp. 32-61. doi: 10.15554/
pcij.11011977.32.61
17. Abdalla, H., and Kennedy, J. B., “Design of Prestressed Concrete Beams
with Openings,” Journal of Structural Engineering, ASCE, V. 121, No. 5,
May 1995, pp. 890-898. doi: 10.1061/(ASCE)0733-9445(1995)121:5(890)
18. Leonhardt, F., and Mönnig, E., Vorlesungen über Massivbau: Dritter
Teil – Grundlagen zum Bewehren im Stahlbetonbau, Springer Verlag,
Berlin, Germany, 1977, 246 pp.
19. Dressen, T., and Classen, M., “Deformation of Reinforced and
Prestressed Concrete Beams with Large Web Openings,” Beton- und Stahl-
betonbau, V. 108, No. 7, July 2013, pp. 462-474.
20. Hegger, J.; Claßen, M.; Schaumann, P.; Sothmann, J.; Feldmann, M.;
and Döring, B., “Integrated Composite Floor-Slab-Systems for Sustainable
Steel Structures,” Stahlbau, V. 82, No. 1, 2013, pp. 11-17. doi: 10.1002/
stab.201301645
21. Classen, M.; Gallwoszus, J.; and Hegger, J., “Einfluss von Querrissen
auf das Schubtragverhalten von Verbunddübelleisten in schlanken Beto-
ngurten,” Beton- und Stahlbetonbau, V. 119, No. 12, pp. 882-894.
22. Classen, M., and Hegger, J., “Verankerungsverhalten von Verbund-
dübelleisten in schlanken Betongurten,” Bautechnik, V. 91, No. 12, 2014,
pp. 869-883.
23. European Committee for Standardization, “Eurocode 2: Design of
Concrete Structures, Part 1.1: General Rules and Rules for Buildings,”
Brussels, Belgium, Dec. 2004, pp. 96-99.
24. Twelmeier, H.; Dallmann, R.; Fischer, T. et al., “Einfluss von großen
Stegöffnungen auf das Trag-und Verformungsverhalten von Stahlbetonträ-
gern,“ Report of the Institute of Structural Analysis, Technical University of
Brunswick, Brunswick, Germany, 1985, pp. 120-123.
25. Tan, K. H.; Mansur, M. A.; and Huang, L.-M., “Reinforced
Concrete T-Beams with Large Web Openings in Positive and Negative
Moment Regions,” ACI Structural Journal, V. 93, No. 3, May-June 1996,
pp. 277-289.
ACI Structural Journal/March-April 2015
 DISCUSSION
Disc. 111-S41/From the May-June 2014 ACI Structural Journal, p. 503
Bond-Slip-Strain Relationship in Transfer Zone of Pretensioned Concrete Elements. Paper by Ho Park and
Jae-Yeol Cho
Discussion by José R. Martí-Vargas
Professor, ICITECH, Institute of Concrete Science and Technology, Universitat Politècnica de València, Valencia, Spain
The discussed paper presents an interesting study on a
novel bond-slip-strain relationship for a strand in the transfer
zone of a pretensioned, prestressed concrete member, as well
as equations for the distributions of the bond stress, slip,
and strand strain. To this end, test specimens with various
test variables, such as concrete compressive strength,
cross section size, cover thickness, strand diameter, curing
method, and debonded region, were fabricated, and strand
strains were measured in the pretensioning and detensioning
process. The authors should be complimented for providing
a detailed paper that is useful for calculating transfer length,
end slip, and maximum bond stress in the transfer zone. The
discusser would like to address the following comments and
questions for the authors’ consideration and response.
1. There are several relevant, complete, and recent
references that have not been considered by the authors,
such as extensive collections of equations for transfer
length, including comparatives and new equations
considering concrete compressive strength25 and strand-
free end slip,26 which were available prior to the submission
date. Furthermore, the authors considered one reference by
Balázs14 in their comparisons (Table 3), whereas another
related, complementary study was ignored.27 Moreover,
ACI 318-081 is used and referenced by the authors, whereas
there is a later edition.28
2. The authors detail that variations in cover depth, cross
section dimensions, and the level of prestressing force
have very little effect on bond characteristics, whereas
a recent study29 has found that concrete cover may also
markedly affect transfer length in pretensioned members.
It has been stated that bond strength reduces as concrete
cover increases,30 whereas increases in bond strength with
increased concrete cover thickness by using pullout tests
have been also reported,31 and the same conclusion was
drawn for prestressing strands32 by using the ECADA
test method.33
3. In addition to the 16 specimens listed in Table 2, more
specimens were originally fabricated, which were excluded
from the analysis because: 1) only concrete strains were
measured; or 2) the specimens seemed to have bond
deficiencies. The authors detail that only strand strains were
measured for the test specimens analyzed in this work.
To offer a better understanding, can the authors provide
additional details on how the bond stress distributions were
obtained for the excluded specimens in which only concrete
strains were measured? Furthermore, it seems that some
specimens were excluded for their lower bond stress values
compared to Specimen 4. It is noteworthy that bond stress
distributions, as offered by Specimens B and D, are also
possible, which would correspond to a bound case (α = 2)
according to Guyon’s theory,12 whereas Specimens 4, 9, 10,
and 16 displayed a linear bond stress distribution that, in
turn, coincides with the other bound case (α = 3) according to
ACI Structural Journal/March-April 2015
Guyon’s theory.12 Specimen A seems to show a combination
of both uniform and linear bond stress distributions, which
would correspond to an intermediate case with a certain
α value (α = 2.67 and α = 2.44 have been also proposed
in theoretical and experimental studies26,27,34). Finally,
Specimen C depicts an atypical bond stress distribution.
What was the transfer length for Specimen C?
4. The measurements on the cut-end sides were also
excluded from the analysis, not to consider an effect of
dynamic impact. However, in light of the differences in
transfer lengths at both ends of one similar specimen, as
shown in Fig. 2, the discusser encourages the authors to
carry out further studies and to include measurements on the
cut end because these measurements are more unfavorable
for strength capacities, and code equations usually do not
reflect manufacturing methods.
5. The authors have assumed that the distribution of
concrete and strand strains has a similar shape. However,
it is noteworthy that the distribution of concrete strains
presents a certain retardation in relation to the distribution of
strand strains. Transfer length is defined as the distance over
which the strand should be bonded to concrete to develop
the effective prestress in prestressing steel.28 This effective
stress is completely transferred to concrete when concrete
stresses are assumed to take a linear distribution, which
occurs outside dispersion length.35 As the authors obtained
transfer length by applying the 95% average maximum
strain (AMS) method9 to the curves of the tendon strain (as
shown in Fig. 3) instead of curves of the concrete strain, it
seems that the authors determine a shorter transfer length.
6. To obtain a bond-slip-strain relationship for a strand, a
basic form of the equation is adopted from a previous bond
model,17 devised for reinforced concrete. The bond stress
in the model is composed of a slip function and a strain
function. It seems that the slip function is obtained from
Fig. 4. Furthermore, the authors detail that the slip at a point
in a member is obtained by integrating the relative difference
of concrete and the strand strain. As only strand strains were
measured and the authors applied the 95% AMS method9 to
the tendon strain curves, can the authors detail how the slips
were obtained?
7. The bond stress was derived from the equilibrium
condition along the strand according to Eq. (6). The discusser
notes that nominal perimeter πdp is used instead of actual
perimeter (4/3)πdp.25,36
8. The authors conclude that slip distribution presents an
initial value in the transfer zone, which corresponds to the
initial value of bond stress at the begining of transfer length.
However, as seen in Fig. 7(a), an initial bond stress value
is observed, whereas an initial slip value is not observed
in Fig. 7(c). The discusser notes that the slip distribution
along the transfer length shown in Fig. 7(c) seems to
qualitatively agree with the slip distribution model recently
233
derived from the experimental tests by the ECADA test
method37 and the theoretical curves obtained from finite
element analyses.29 None of these proposals includes an
initial slip value.
REFERENCES
25. Martí-Vargas, J. R.; Arbeláez, C. A.; Serna-Ros, P.; Navarro-Gregori,
J.; and Pallarés-Rubio, L., “Analytical Model for Transfer Length Prediction
of 13 mm Prestressing Strand,” Structural Engineering and Mechanics,
V. 26, No. 2, 2007, pp. 211-229.
26. Martí-Vargas, J. R.; Arbeláez, C. A.; Serna-Ros, P.; and Castro-
Bugallo, C., “Reliability of Transfer Length Estimation from Strand End
Slip,” ACI Structural Journal, V. 104, No. 4, July-Aug. 2007, pp. 487-494.
27. Balázs, L. G., “Transfer Length of Prestressing Strand as a Function
of Draw-in and Initial Prestress,” PCI Journal, V. 38, No. 2, Mar.-Apr.
1993, pp. 86-93.
28. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-11) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2011, 503 pp.
29. Oh, B. H.; Lim, S. N.; Lee, M. K.; and Yoo, S. W., “Analysis and
Prediction of Transfer Length in Pretensioned, Prestressed Concrete
Members,” ACI Structural Journal, V. 111, No. 3, May-June 2014,
pp. 549‑559.
30. Ichinose, T.; Kanayama, Y.; Inoue, Y.; and Bolander, J. E., “Size
Effect on Bond Strength of Deformed Bars,” Construction and Building
Materials, V. 18, 2004, pp. 549-558.
31. García-Taengua, E.; Martí-Vargas, J. R.; and Serna-Ros, P.,
“Statistical Approach to Effect of Factors Involved in Bond Performance
of Steel Fiber-Reinforced Concrete,” ACI Structural Journal, V. 108, No. 4,
July-Aug. 2011, pp. 461-468.
32. Martí-Vargas, J. R.; Caro, L. A.; and Serna-Ros, P., “Size Effect on
Strand Bond and Concrete Strains at Prestress Transfer,” ACI Structural
Journal, V. 111, No. 2, Mar.-Apr. 2014, pp. 419-429.
33. Martí-Vargas, J. R.; García-Taengua, E.; Caro, L. A.; and Serna-
Ros, P., “Measuring Specific Parameters in Pretensioned Concrete Members
Using a Single Testing Technique,” Measurement, V. 49, 2014, pp. 421-432.
34. Viula, D.; Lucio, V.; Pinho, G.; and Martí-Vargas, J. R., discussion
of “Pull-out and Push-in Tests of Bonded Steel Strands,” Magazine of
Concrete Research, V. 65, No. 18, 2013, pp. 1128-1131.
35. CEN, “Eurocode 2: Design of concrete structures—Part 1-1: General
Rules and Rules for Buildings. (EN 1992-1-1:2004:E),” Comité Européen
de Normalisation, Brussels, Belgium, 2004.
36. Tabatabai, H., and Dickson, T., “The History of the Prestressing
Strand Development Length Equation,” PCI Journal, V. 38, No. 5, Sept.-
Oct. 1993, pp. 64-75.
37. Martí-Vargas, J. R.; Hale, W. M.; García-Taengua, E.; and Serna-
Ros, P., “Slip Distribution Model along the Anchorage Length of
Prestressing Strands,” Engineering Structures, V. 59, 2014, pp. 674-685.
AUTHORS’ CLOSURE
The authors would like to thank the discusser for his
interest in the paper and valuable comments. The authors’
item-by-item response is presented in the following text:
1. Over last decades, many equations for transfer lengths
have been proposed by various researchers. Because it was
impossible to consider all of the equations in the paper, the
most representative and frequently cited equations were
chosen to be compared with the experimental results and the
proposed model. Equations incorporating the term of end
slip26,27 were not included in the comparison because the
measured end slip data were found to be unreliable. There is
no difference in the calculation of transfer lengths according
to the 2008 and the 2011 editions of ACI 318.1,28
2. As mentioned by the discusser, concrete cover depth
greatly affects transfer lengths in the pretensioned concrete
members. The authors stated that cover depth has little effect
on bond characteristics if there are no splitting cracks. Many
researchers indicated that the influence of cover depth on
transfer length is reduced with increasing cover depth. Den
Uijl15 stated that transfer length decreases with increasing
cover depth but no further reduction occurs beyond the cover
depth of 3 to 4dp. Oh and Kim11 reported that the reduction
234
Fig. 13—Strand strain distribution of Specimen C. (Note:
1 mm = 0.039 in.)
rate of transfer length decreases when the cover depth is
increased. In the equation proposed by Oh et al.,29 transfer
length is inversely proportional to cover depth. It means that
the effect of cover depth rapidly diminishes with the increase
in cover depth. Ichinose et al.30 concluded that the size effect
on bond strength is reduced with increasing confinement due
to larger cover depth. Martí-Vargas et al.32 also showed that
the increasing rate of average bond stress is reduced when
cross section is increased.
3. The specimens for which only concrete strains were
measured were excluded from the analysis because the bond
stress distributions could not be obtained. Specimens A to
D were excluded to develop a basic bond model for the
favorable bond condition. The proposed model is capable
to consider different types of bond stress distributions by
modifying the coefficient a0. As the value of the coefficient
a0 becomes smaller, the resulting bond stress approaches a
uniform distribution. Strand strain distribution and transfer
length of Specimen C is provided in Fig. 13.
4. The authors agree with the discusser’s opinion on the
importance of transfer length at the cut end.
5. The assumption of a similar shape between the
distributions of concrete and strand strains was made based
on the measured strain data. In another paper,18 the authors
showed that both strain distributions had a very similar shape
for the given specimen configurations. It can be justified by
the following references. According to CEB-FIP MC90,38 the
difference between the transfer length and the dispersion
length vanishes for a cross section with the total depth less
than 80% of transfer length. In the authors’ experiment, the
depth of the test specimens was less than one-fourth of the
measured transfer lengths. Buckner39 and Den Uijl40 have
demonstrated by means of finite element analysis that the
deviation of strain distributions at concrete surface and at the
strand was not significant for members with a cross section,
which is small in relation to the transfer length.
6. The concrete strain distribution was generated based on
the assumption of a similar shape between the distributions
of concrete and strand strains. The ratio of concrete to strand
strain at any point was given as Eq. (8). The slip distribution
along the strand was obtained from Eq. (9) and (10).
7. The authors followed the approach given in Balázs14 and
Cousins et al.2,41 In their analyses, the nominal perimeter of
strand was used for the calculation of bond stress.
ACI Structural Journal/March-April 2015
 8. The coefficient a5, representing the initial value of slip,
satisfies a mathematical requirement. It is also consistent
with the physical explanation for the initial value of bond
stress. The value of the coefficient a5 is very small, as shown
in Fig. 7(c).
REFERENCES
38. CEB-FIP, “Model Code 1990,” Comité Euro-International du Béton
(CEB), Bulletin d’Information 213/214, London, UK, 1990, 437 pp.
39. Buckner, C. D., “An Analysis of Transfer and Development Lengths
for Pretensioned Concrete Structures,” FHWA-RD-94-049, Virginia
Military Institute, Lexington, VA, Dec. 1994.
40. Den Uijl, J. A., “High Performance Concrete in Prefab Industry.
Part 3: Transmission Length of Prestressing Strand,” Stevin Report 25.5-
95-3, TU Delft, Delft, the Netherlands, 1995, 65 pp.
41. Cousins, T. E.; Johnston, D. W.; and Zia, P., “Transfer and
Development Length of Epoxy Coated and Uncoated Prestressing Strand,”
PCI Journal, V. 35, No. 4, July-Aug. 1990, pp. 92-103.

Disc. 111-S44/From the May-June 2014 ACI Structural Journal, p. 537

Fire Protection for Beams with Fiber-Reinforced Polymer Flexural Strengthening Systems. Paper by Nabil
Grace and Mena Bebawy
Discussion by W. L. Gamble
FACI, Professor Emeritus, University of Illinois, Urbana, IL

This paper adds useful information and data about the AUTHORS’ CLOSURE
behavior of fiber-reinforced polymer (FRP) in fire situations, The authors thank the discusser for his interest in the paper
but it also presents some puzzles. It is unfortunate that there and the published work. The discussion focuses on three
apparently were no tests of the reinforcing bars because main points: 1) the moment capacity of the beam; 2) the
both a measured yield stress and a measured stress-strain rupture of the steel in Beam B-DH-C/F; and 3) the removal
curve would have been helpful in understanding some of the of the entire load in Beam B-CF-G/F. The authors will try to
results. The 80 ksi (551 MPa) yield stress implied from the address all three points.
strain measurements for Beam B-U-O/A seems high but is For the first point, it should be noted that due to the
plausible for a small bar. nature of the fire testing, a special support system was used
However, even this high stress does not explain the throughout the entire experiment, even for beams tested at
behavior of this beam. At 10 kip (44.5 kN) force (the reported ambient temperature. As shown in Fig. 13, the beam was
yield force), the applied moment is approximately 31.3 ft-kip resting on two 12 in. (305 mm) wide steel plates at its ends.
(42.4 kN-m), including the small dead-load moment. However, Therefore, for the purpose of moment calculations, the
the computed value of Mn is approximately 25.4 ft-kip effective span should be taken as 10 ft (3.05 m) as the beam
(34.4 kN-m) when fc′ = 7.3 ksi (50 MPa) and fy = 80 ksi rotated around the interior edges of the support plates. By
(551 MPa). At the maximum force of 12.92 kip (57.5 kN), considering an effective span of 10 ft (305 mm), the moment
the applied moment is approximately 40.1 ft-kip (54.4 kN-m). due to dead load will be equal to 0.935 ft-kip (1.27 kN-m).
The difference between the observed yield moment and the At steel strain of 3000 µɛ, the depth of the NA is 2.26 in.
computed nominal capacity seems too large, as does the differ- (57.4 mm) and the theoretical yield moment is 25.72 ft-kip
ence between the observed yield and maximum applied loads. (34.87 kN-m). By subtracting the moment due to dead load
The computed strain at Mn is less than 0.02, which implies from the yield moment, the moment due to the concentrated
some strain hardening but probably not 29%. load becomes 24.79 ft-kip (33.61 kN-m), which is equiv-
A possible source of the differences is restrained elonga- alent to a moment due to a concentrated load of 9.91 kip
tion. The longitudinal restraint applied at the bottom surface (44 kN). It is worth noting that the beam was resting freely
of the beam that is able to increase the moment capacity on the supports and no significant longitudinal restraint
from 25 to 40 ft-kip is not too large. The support system is was provided.
not well described other than Fig. 2 showing a roller under
one end of the beam.
Beam B-DH-C/F is reported to have collapsed due to frac-
tured reinforcement. The statement that “the reinforcement
of the beam melted” cannot be true because the melting point
of steel is much higher than either the fire temperature or the
steel temperature at the time of collapse. This suggests some
flaw in the reinforcement, or perhaps some exothermic reac-
tion involving the epoxy. Was a sudden flare-up observed?
Beam B-CF-G/F was loaded to a significant load for
only 15 minutes, and then unloaded. It is not clear whether
the remaining load was only the dead load of the member
and the loading column or some larger load. This load reduc-
tion was described as equivalent to the removal of the live
load, but this is not consistent with actual reinforced concrete
buildings. In most reinforced concrete buildings, the dead
load exceeds the live load, so a representative case might
involve removing half of the applied load, but not a major
fraction. The fact that this beam was able to resist major
loads after cooling is an important piece of information. Fig. 13—Beam B-U-O/A under three-point loading setup.

ACI Structural Journal/March-April 2015 235

Fig. 14—Close-up view of Beam B-U-O/F showing rupture
of bottom steel reinforcement bars after exposure to fire test.
Fig. 15—Beam B-DH-GB/F2 split in half after exposure
to fire.
For the second point, Beam B-DH-C/F did split in half due
to rupture of bottom steel reinforcement, as shown in Fig. 6.
It should be noted that it was not the only beam that experi-
enced this kind of steel fracture/rupture. Beam B-U-O/F had
no FRP wraps and yet experienced steel rupture, as shown
in Fig. 14. Beams B-DH-GB/F2 and B-CF-GB/F also split
in half, as shown in Fig. 15 and Fig. 16, respectively. The air
temperature at the time of failure in all these beams ranged
from 1850 to 1890°F (1010 to 1032°C), which conformed
to ASTM E119 time-temperature curve. While it is possible
that the steel had some flows, according to Purkiss (2007)
and multiple other references and fire design codes, at
temperature of 1832°F (1000°C), the yield strength of steel
236
Fig. 16—Beam B-CF-GB/F split in half after exposure to
fire/loading event.
is only 4% of its ambient yield strength. Also, its modulus
of elasticity at that temperature is also approximately 4.5%
of its ambient value. Therefore, the rupture of the steel even
under the self-weight of the beam at that point of the test
was not unexpected. The wide cracks exposed the steel to
the air temperature and subsequently the reinforcement lost
almost all of its structural strength and fractured/ruptured.
There were flares at failure in all beams with FRP strength-
ening systems due to the ignition of the epoxy, but the
control system of the furnace regulated the air temperature
by cutting off the fuel input and increasing the flow of the
exhaust system. Consequently, the air temperature followed
exactly the ASTM E119 curve and the ignition of the epoxy
adhesive hardly influenced the conditions of the test.
For the third point, Beam B-CF-G/F was loaded to a signif-
icant load for only 15 minutes and the remaining load was
the self-weight of the beam plus the weight of the loading
column, which was resting on the beam. It is true that the
dead load may exceed the live load in reinforced concrete
structures. However, the strengthening system is usually
designed to only sustain the live load (or part of it). The
strengthening system is applied to existing structures, which
are already loaded with at least the dead load. Therefore,
unless the existing beam is jacked prior to the application of
the strengthening system, the FRP strengthening system will
not resist the dead loads. Consequently, in case of fire, there
is a good chance that the FRP strengthening system will not
be stressed if the live load manages to escape early as repre-
sented in the loading scenario of Beam B-CF-G/F.
REFERENCES
Purkiss, J. A., 2007, “Fire Safety Engineering,” Design of Structures,
second edition, Elsevier Ltd.
ACI Structural Journal/March-April 2015
Disc. 111-S45/From the May-June 2014 ACI Structural Journal, p. 549
Analysis and Prediction of Transfer Length in Pretensioned, Prestressed Concrete Members. Paper by Byung
Hwan Oh, Si N. Lim, Myung K. Lee, and Sung W. Yoo
Discussion by José R. Martí-Vargas
Professor, ICITECH, Institute of Concrete Science and Technology Universitat Politècnica de València, Valencia, Spain
Based on three-dimensional finite element analyses and
experimental tests conducted to obtain transfer lengths in
pretensioned, prestressed concrete members, the discussed
paper explores the important factors that affect transfer
length and proposes a realistic prediction equation for a more
rational design of pretensioned members. In particular, it has
been found that transfer length decreases with increased
cover depth, and also with increased concrete compressive
strength. Then these two parameters, as well as prestress
magnitude and strand diameter, have been included in a
new equation for transfer length determination. The authors
should be congratulated for producing a detailed paper.
Some findings are interesting for the discusser, who would
like to address the following comments for the authors’
consideration and response.
1. Regarding the references related with this paper, the
discusser would like to point out that: a) ACI 318-0217 is
referenced by the authors, whereas there are later editions
available prior to the paper submission date23; and b)
there are several relevant, complete, and recent references
that were not considered by the authors, such as extensive
collections of equations for transfer length, including
comparatives and new equations considering concrete
compressive strength24,25 and strand free-end slip.26
2. The authors detail that bond stress-slip relations have
been obtained from the measurement of strains. In particular,
the strand-to-concrete slip has been obtained from measuring
concrete strains at two adjacent points using Eq. (5). It is
noteworthy that, at a point within transfer length, slip is
obtained by integrating the relative differences of the concrete
and strand strains when prestress is transferred.8,27 However,
it seems that Eq. (5) considers only the differences between
the concrete strains at two locations, regardless of strand
strains. To offer a better understanding, can the authors
provide additional details on how slips were obtained?
3. The authors state that there are good correlations between
the transfer lengths and end slip values in pretensioned
members and, hence, it may be possible to calculate transfer
length from the end slip value. The discusser notes that
there are earlier studies on this possibility,28,29 that most
experimental standards26 are based on this method, and that
it has been proposed as a simple nondestructive assurance
procedure by which quality bond can be monitored within
precasting plants.11 Furthermore, several researchers have
conducted experimental studies to obtain transfer length
from the strand free-end slip in hollow-core slabs, beams,
piles, and prisms, while some authors11,30 have established
an allowable free-end slip as the strand-end slip, which
results in a transfer length equal to that computed by the
ACI 318 provisions for transfer length. Despite all these
previous works, the authors have compared their data with
only the Logan22 equation, which showed no agreement with
the good regression equation between transfer length and
strand-end slip (Eq. (11a)) obtained by the authors. It seems
that Logan22 considered a uniform bond stress distribution
ACI Structural Journal/March-April 2015
along transfer length, which coincides with a bound case
(α = 2) according to Guyon’s theory,28 whereas a recent
study27 proposes a linear bond stress distribution which, in
turn, coincides with the other bound case (α = 3) according
to Guyon’s theory.28 An intermediate case (α = 2.44)26,31 has
been also proposed. Therefore, in light of the interest of
these topics, the discusser suggests further analyses to be
addressed to obtain: 1) the corresponding α value from the
authors’ experimental data; and 2) as there is a wide range
of strand-end slips that correspond to the same transfer, and
vice versa, the quantification of the number of cases out the
allowable free-end slip and the predicted transfer length
according to the main codes is recommended.32
4. The authors found that concrete cover may also
markedly affect transfer length in pretensioned members,
whereas a recent study27 provides details that variations
in cover depth, the dimensions of the cross section, and
the level of prestressing force have very little effect on
bond characteristics. It has been stated that bond strength
reduces as concrete cover increases,33 whereas increases
in bond strength with greater concrete cover thickness by
using pullout tests34 have also been reported, and the same
conclusion was obtained for prestressing strands35 by using
the ECADA test method.36
5. The discusser notes that variation in the slip values
along the distance from the end of the pretensioned members
shown in Fig. 15, which were obtained from finite element
analyses, qualitatively agrees well with the slip distribution
model recently derived from experimental tests.37 However,
it seems that the end slip values obtained by the authors are
higher than in other studies,27,37 which perhaps is related with
the specific procedure used, as mentioned in remark No. 2.
REFERENCES
23. ACI Committee 318, “Building Code Requirements for Structural
Concrete (ACI 318-11) and Commentary,” American Concrete Institute,
Farmington Hills, MI, 2011, 503 pp.
24. Martí-Vargas, J. R.; Arbeláez, C. A.; Serna-Ros, P.; Navarro-
Gregori, J.; and Pallarés-Rubio, L., “Analytical Model for Transfer Length
Prediction of 13 mm Prestressing Strand,” Structural Engineering and
Mechanics, V. 26, No. 2, 2007, pp. 211-229.
25. Martí-Vargas, J. R.; Serna-Ros, P.; Navarro-Gregori, J.; and
Pallarés, L., “Bond of 13 mm Prestressing Steel Strands in Pretensioned
Concrete Members,” Engineering Structures, V. 41, 2012, pp. 403-412.
26. Martí-Vargas, J. R.; Arbeláez, C. A.; Serna-Ros, P.; and Castro-
Bugallo, C., “Reliability of Transfer Length Estimation from Strand End
Slip,” ACI Structural Journal, V. 104, No. 4, July-Aug. 2007, pp. 487-494.
27. Park, H., and Cho, J. Y., “Bond-Slip-Strain Relationship in Transfer
Zone of Pretensioned Concrete Elements,” ACI Structural Journal, V. 111,
No. 3, May-June 2014, pp. 503-513.
28. Guyon, Y., Pretensioned Concrete: Theoretical and Experimental
Study, Paris, France, 1953, 711 pp.
29. Thorsen, N., “Use of Large Tendons in Pretensioned Concrete,” ACI
Journal Proceedings, V. 53, No. 6, June 1956, pp. 649-659.
30. Wan, B.; Harries, K. A.; and Petrou, M. F., “Transfer Length of
Strands in Prestressed Concrete Piles,” ACI Structural Journal, V. 99,
No. 5, Sept.-Oct. 2002, pp. 577-585.
31. Viula, D.; Lucio, V.; Pinho, G.; and Martí-Vargas, J. R., discussion
of “Pull-out and Push-in Tests of Bonded Steel Strands,” Magazine of
Concrete Research, V. 65, No. 18, 2013, pp. 1128-1131.
237
 32. Martí-Vargas, J. R., and Hale, W. M., “Predicting Strand Transfer
Length in Pretensioned Concrete: Eurocode versus North American Practice,”
Journal of Bridge Engineering, ASCE, V. 18, No. 12, 2013, pp. 1270-1280.
33. Ichinose, T.; Kanayama, Y.; Inoue, Y.; and Bolander, J. E., “Size
Effect on Bond Strength of Deformed Bars,” Construction and Building
Materials, V. 18, 2004, pp. 549-558.
34. García-Taengua, E.; Martí-Vargas, J. R.; and Serna-Ros, P.,
“Statistical Approach to Effect of Factors Involved in Bond Performance
of Steel Fiber-Reinforced Concrete,” ACI Structural Journal, V. 108, No. 4,
July-Aug. 2011, pp. 461-468.
35. Martí-Vargas, J. R.; Caro, L. A.; and Serna-Ros, P., “Size Effect on
Strand Bond and Concrete Strains at Prestress Transfer,” ACI Structural
Journal, V. 111, No. 2, Mar.-Apr. 2014, pp. 419-429.
36. Martí-Vargas, J. R.; García-Taengua, E.; Caro, L. A.; and Serna-
Ros, P., “Measuring Specific Parameters in Pretensioned Concrete Members
Using a Single Testing Technique,” Measurement, V. 49, 2014, pp. 421-432.
37. Martí-Vargas, J. R.; Hale, W. M.; García-Taengua, E.; and Serna-
Ros, P., “Slip Distribution Model along the Anchorage Length of
Prestressing Strands,” Engineering Structures, V. 59, 2014, pp. 674-685.
AUTHORS’ CLOSURE
The authors would like to thank the discusser for providing
good comments. The following are the appropriate responses
to those comments.
1. First, the discusser commented on the references in the
paper. First of all, the authors would like to thank the discusser
for the addition of more references relevant to the present
paper. Those added references would be good resources for
readers as well as the authors. As for the recent reference
of ACI 318-11,23 the provision for the transfer length is the
same as the previous version of ACI 318.17 Therefore, it does
not affect the content of the paper.
2. Second, the discusser commented on the determination
of slip values. The authors measured the strains of strand
and concrete due to pretensioning, as shown in Fig. 4 in the
paper. The slip was determined, as the discusser pointed
out, by integrating the differences between the change of
strand strain and the concrete strain.8 The authors would like
to clarify that the explanation in the paper was somewhat
mixed with the determination of transfer length, which was
determined from the variation of concrete strains along
the member.
3. Third, the discusser commented on the transfer length
in terms of end slip values, citing the authors’ statement that
“there are good correlations between the transfer lengths
and end slip values in pretensioned members and, hence, it
may be possible to calculate transfer length from the end slip
value.” The discusser also pointed out that previous studies
reported the value α (in Guyon’s theory) of 2 to 3 depending
on the bond stress distribution (that is, constant or linear)
along the transfer length. Intermediate values of α were
also reported from the previous studies.8,26,31 The discusser
suggested that the authors perform further analyses to obtain
the appropriate value of α from the authors’ experimental
data. In the authors’ opinion, it is very likely that the real
distribution of bond stress may be nonlinear, which is
different from the previous assumption of constant or linear
bond stress distribution. Therefore, the authors plan to
further study this subject to clarify the real behavior (and
distribution) of bond stress and slip. As for the scatter of
the data on the end slip values that correspond to the same
238
transfer (for example, as can be seen from Fig. 20 of the paper
and from References 26 and 32), the causes may come from
the variability of many factors such as material strengths,
surface conditions of strands, accuracy of measurement
of prestressing force and slip, and time-dependant effect
(that is, elapsed time after prestress transfer). All of these
factors may partially affect the measured values of slip or
transfer length. Therefore, great care must be paid to reduce
the variability in experiments. A comprehensive analysis of
transfer lengths and slip values considering the variability
of test variables is another valuable subject that needs to be
researched in the future.
4. Next, the discusser commented on the effects of concrete
cover on transfer length and bond. It seems that the discusser
doubts the effect of concrete cover on transfer length. The
discusser also indicated that contradictory results on bond
strength exist as the concrete cover increases.33-36 However,
it can be clearly seen from Fig. 3 and Table 1 that the
measured transfer lengths from experiments decrease with
an increase of concrete cover. This effect is again confirmed
by the results of finite element analyses as shown in
Table 3 and Fig. 16. It is obvious that the concrete cover
plays a key role in transferring steel forces to concrete.38 It
is well-known that shallow cover may cause splitting failure
in reinforced and prestressed concrete due to inadequate
bond capacity.38 As shown in Fig. 16, the rate of increase
of transfer length due to the increase of concrete cover may
decrease as the concrete cover increases further. This means
that the effect of concrete cover may diminish if a sufficient
cover depth is secured, which may be far larger than usual
cover depths. This is because it may reach sufficient bond
capacity after a certain required cover depth.
5. Finally, the discusser commented on the slip distribution
along the transfer length and the end slip values, and
mentioned that the slip distribution data of the authors’
study qualitatively agree well with the results of another
study.37 The discusser also indicated that the end slip values
obtained by the authors seem higher than in other studies.
However, Fig. 20 clearly shows that the present study
includes not only high values but also low values in end
slip. It is generally known that the end slip values are greatly
affected by the design parameters of pretensioned members
such as cover depth, concrete strength, and magnitude of
prestressing forces. In the authors’ study, the cover depth
varies from a very low value to a high value. This may give
a wide range of end slip values depending on cover depth.
Furthermore, the prestress magnitude in the authors’ study
ranges from a very low (0.40fpu) to a high value (0.75fpu),
which may induce low as well as high slip values. The
combination of all these design parameters may give a wide
range of slip values in pretensioned members. This may be
the reason why the slip values in the authors’ study range
from low to high values, as shown in Fig. 15 and 20.
REFERENCES
38. Nilson, A. H.; Darwin, D.; and Dolan, C. W., Design of Concrete
Structures, Chapter 5, 14th edition, McGraw-Hill Co. Inc., 2010, pp. 168-207.
ACI Structural Journal/March-April 2015
Disc. 111-S50/From the May-June 2014 ACI Structural Journal, p. 607
Flexural Testing of Reinforced Concrete Beams with Recycled Concrete Aggregates. Paper by Thomas H.-K.
Kang, Woosuk Kim, Yoon-Keun Kwak, and Sung-Gul Hong
Discussion by Bhupinder Singh
Associate Professor, Department of Civil Engineering, Indian Institute of Technology Roorkee, Roorkee, India
The authors should be complimented for carrying out a
thorough and a meticulous investigation of flexural behavior
of reinforced concrete beams made with recycled concrete
aggregates (RCAs). The discusser invites the authors to
comment on the following issues:
1. The replacement levels of the coarse RCAs used by
the authors are too closely spaced and, at the considered
replacement levels variations in mechanical and structural
properties, are unlikely to manifest themselves distinctly.
2. Only one stand-alone specimen has been tested by the
authors for each parameter under investigation. Toward
ensuring repeatability of results, it is desirable that at least
two companion specimens should be tested and their results
compared before arriving at any conclusion.
3. The authors have not mentioned in the paper how
grading of their RCA particles compares with that of the
coarse natural aggregate (NA) particles. Were the RCA
particles graded to confirm to a particular range given, for
example, in any of the current design codes? Further, no
mention has been made in the paper of the source of the
waste concrete from which the RCA was derived. Was the
waste concrete obtained from a demolition project or was it
sourced from waste laboratory specimens?
4. Table 2 does not give information with respect to two
important physical properties of RCA—namely, residual
mortar content and aggregate crushing value. The residual
mortar content controls water absorption of the RCA
particles and the aggregate crushing value is a good indicator
of strength. The moisture state of the RCA particles used in
the concrete mixtures was not mentioned in the paper. To
maintain a nominally constant free water-cement ratio (w/c)
across comparable concrete mixtures, one would expect the
RCA particles to be used in the saturated surface-dry (SSD)
state. The authors are invited to clarify.
5. The usual silica fume dosage in a concrete mixture is
approximately 10% of the weight of cement. The dosage of
43% used by the authors is unusual. Such a dosage would
severely impact concrete workability and require the use of
high-range water reducer (HRWR) dosages.
Fig. 8—Crushing tests of natural aggregate (first test = 18.6 kN [4.2 kip]; second test =
17.5 kN [3.9 kip]).
ACI Structural Journal/March-April 2015
6. The beam specimens with 2-D10 tension reinforcement
would in effect be doubly reinforced because of the presence
of 2-D10 hanger bars near the top face of the beam, which
will effectively act as compression reinforcement. Because
of the presence of this compression reinforcement, this
beam is likely to show very large ductility compared to other
beams and, hence, its behavior would be an outlier.
7. In Fig. 7, the characteristic (or the measured) flexural
strength of the RCA beams have been compared with
predictions of the ACI Code, which are based on factored
material strengths. Such a comparison would be biased
toward giving conservative flexural strength predictions for
the RCA beams. If a suitable strength reduction factor is
applied to the experimental results plotted in Fig. 7, then it
will be seen that many of them would fall below predictions
of the ACI Code. It would be interesting to know how the
experimental results plotted in Fig. 7 would compare with
ACI Code predictions made using characteristic material
strengths. In the backdrop of such an exercise, the authors
may like to revisit some of the conclusions in the paper.
AUTHORS’ CLOSURE
The authors would like to thank the discusser for the
interest in the paper and comments. Responses to the
discusser’s comments are selectively provided, as the
discusser suggested and as already explained in the paper.
In response to Comment 3, the used RCA has a solid
volume percentage for shape determination of 58.3%,
exceeding the minimum value (55%) specified by Korean
Standard (KS F 2573; Korean Standard 2011), but slightly
lower than that (60.1%) of the used natural aggregate. The
authors note that the RCA was obtained from a commercial
company that is no longer in business. To the best of the
authors’ knowledge, this company acquired the aggregates
from demolition projects.
In response to Comment 4, the authors agree that the
residual mortar content affected the water absorption rate
of RCA. Though it was reported that “the larger water
absorption rate of RCA affected the total water content of the
239
 Fig. 9—Crushing tests of recycled concrete aggregate (first test = 16 kN [3.6 kip]; second
test = 16.1 kN [3.7 kip]).
concrete and its compressive strength,” the authors found
that there was little correlation between the water absorption
rate and the beam’s flexural strength. It is also confirmed that
all the aggregates were generally in the SSD moisture state.
This was done by spraying aggregates with water at 9 am,
laying out to dry in indoor spaces until 6 pm, and packing it
in a bag to use the next day.
The authors agree that aggregate strength does become
important in high-strength concrete, although the strength
of an aggregate is rarely tested; hence, the authors
had reported in Table 2 that “the strength of RCA is
approximately 90% of natural coarse aggregate’s strength”
(not significant). Aggregate compressive strengths typically
vary from 65 to 270 MPa (9.4 to 39.2 ksi), which depends
on the aggregate type. Because it is hard to measure the
crushed aggregate cross-sectional area, the authors tried to
240
compare the strengths of NA and RCA aggregates, as shown
in Fig. 8 and 9, minimizing the eccentricity.
In response to Comment 5, the authors agree that the
mixture proportion is not practical and a large amount of
HRWR was used. However, the authors can assure that the
reported slump values are correct and that the mechanical
test results were used only for assessment of mechanical
behavior. Finally, in reference to Comment 7, the discusser
seemed to mistakenly interpret the graph. As indicated in
the paper, the measured material properties of the steel and
concrete were used to calculate the nominal flexural strength
(not specified properties).
REFERENCES
KS F 2573, 2011, “Recycled Aggregate for Concrete,” Korean Industrial
Standards, Seoul, Korea.
ACI Structural Journal/March-April 2015
  REVIEWERS IN 2014
In 2014, the individuals listed on these pages served as technical reviewers of papers offered for publication
in ACI periodicals. A special “thank you” to them for their voluntary assistance in helping ACI maintain
the high quality of its publication program.

£aźniewska-Piekarczyk, Beata Achillopoulou, Dimitra

Silesian University of Technology Democritus University of Thrace
Gliwice, Poland Xanthi, Greece
Aamidala, Hari Shankar Acun, Bora
Parsons Brinckerhoff University of Houston
Herndon, VA Houston, TX
Abaza, Osama Adamczewski, Grzegorz
University of Alaska Anchorage Warsaw University of Technology
Anchorage, AK Warsaw, Poland
Abbas, Abdelgadir Adhikary, Bimal
Carleton University Austin, TX
Ottawa, ON, Canada Afif, Rahma
Abbas, Safeer Damascus University
Western University Damascus, Syrian Arab Republic
London, ON, Canada Aggelis, Dimitrios
Abbasnia, Reza University of Ioannina
University of Science and Technology Ioannina, Greece
Tehran, Islamic Republic of Iran Agustiningtyas, Rudi
Abdalla, Hany Ministry of Public Works
College of Technological Studies Bandung, Indonesia
Shuwaikh, Kuwait Ahmad, Shamsad
Abdelaziz, Gamal King Fahd University of Petroleum & Minerals
Benha University Dhahran, Saudi Arabia
Cairo, Egypt Ahmadi, Jamal
Abdelaziz, Magdy University of Science of Technology
Fayoum, Egypt Tehran, Islamic Republic of Iran
Abdel-Fattah, Hisham Ahmed, Ehab
University of Sharjah University of Sherbrooke
Sharjah, United Arab Emirates Sherbrooke, QC, Canada
Abdelgader, Hakim Ahmed, Zeyad
Tripoli University Michigan Technological University
Tripoli, Libyan Arab Jamahiriya Houghton, MI
Abdelrahman, Amr Aire, Carlos
Heliopolis, Egypt National Autonomous University of Mexico
Abdulla, Nwzad Mexico, DF, Mexico
University of Salahaddin Akakin, Tumer
Erbil, Iraq Turkish Ready Mixed Concrete Association
Abeyruwan, Helarisi Istanbul, Turkey
University of Peradeniya Akalin, Ozlem
Peradeniya, Sri Lanka Plustechno Ltd
Abouhussien, Ahmed Istanbul, Turkey
Memorial University of Newfoundland Akbari, Reza
St. John’s, NL, Canada University of Tehran
Aboutaha, Riyad Tehran, Islamic Republic of Iran
Syracuse University Akbarnezhad, Ali
Syracuse, NY The University of New South Wales
Abou-Zeid, Mohamed Sydney, New South Wales, Australia
American University in Cairo Akcaoglu, Tulin
Cairo, Egypt Eastern Mediterranean University
Abruzzo, John Magusa, Turkey
Thornton Tomasetti Akcay, Burcu
San Francisco, CA Kocaeli University
Abu Yosef, Ali Kocaeli, Turkey
WDP and Associates Akınay, Yuksel
Austin, TX Iron and Steel Institution/Material Researcher Center
Karbuk, Turkey

ACI Structural Journal/March-April 2015 241

 REVIEWERS IN 2014
Akiyama, Mitsuyoshi Al-Mufti, Rafal
Waseda University Surrey University
Tokyo, Japan Guildford, UK
Al-alaily, Hossam Almuhsin, Bayrak
Memorial University of Newfoundland University of Technology
St. John’s, NL, Canada Karrada, Baghdad, Iraq
Alam, A. K. M. Jahangir Alsiwat, Jaber
Bangladesh University of Engineering and Technology (BUET) Saudi Consulting Services
Dhaka, Bangladesh Riyadh, Saudi Arabia
Alam, M. Shahria Altunişik, Ahmet Can
The University of British Columbia Karadeniz Technical University
Kelowna, BC, Canada Trabzon, Turkey
Alam, Mahbub Aly, Aly Mousaad
Stamford University Bangladesh Louisiana State University
Dhaka, Bangladesh Baton Rouge, LA
Al-Attar, Tareq Amani Dashlejeh, Asghar
University of Technology Tarbiat Modares University
Baghdad, Iraq Tehran, Islamic Republic of Iran
Al-Azzawi, Adel Amani Dashlejeh, Jafar
Nahrain University Bauhaus University of Weimar
Baghdad, Iraq Weimar, Germany
Albahttiti, Mohammed Amir, Sana
Kansas State University Delft University of Technology
Manhattan, KS Delft, South Holland, the Netherlands
Albuquerque, Albéria Andersson, Ronny
Federal Center of Technological Education of Mato Grosso Hollviken, Sweden
Cuiabá, Mato Grosso, Brazil Andrade, Jairo
Alcocer, Sergio Chatolic University of Rio Grande do Sul
Institute of Engineering, UNAM Porto Alegre, RS, Brazil
Mexico City, DF, Mexico Andriolo, Francisco
Aldea, Corina-Maria Andriolo Ito Engenharia S/C Ltda
St. Catharines, ON, Canada São Carlos, São Paulo, Brazil
Alexander, Mark Angel, Nelson
University of Cape Town Universidad de los Andes
Cape Town, South Africa Bogotá, Cundinamarca, Colombia
Al-Hadithi, Abdulkader Ansari, Abdul Aziz
University of Anbar Quaid-e-Awam Engineering University
Ramadi, Al-anbar, Iraq Nawabshah, Sindh, Pakistan
Al-Harthy, Ali Aoki, Yukari
Sultan Qaboos University University of Technology Sydney
Al-Khaudh, Oman Ultimo, New South Wales, Australia
Ali, Samia Aragón, Sergio
University of Engineering and Technology, Lahore Holcim (Costa Rica)
Lahore, Punjab, Pakistan San Rafael, Alajuela, Costa Rica
Al-Karkhi, Hassan Araujo, Daniel
Al-Mustansiriya University College of Engineering Federal University of Goiás
Baghdad, Iraq Goiania, Brazil
Alkhairi, Fadi Aravinthan, Thiru
Arabtech Jardaneh University of Southern Queensland
Amman, Jordan Toowoomba, Queensland, Australia
Allahdadi, Hamidreza Arisoy, Bengi
Bangalore, India Ege University
Allena, Srinivas Izmir, Turkey
Washington State University Tri-Cities Aristizabal-Ochoa, Jose
Richland, WA National University
Almaral, Jorge Medellin, Antioquia, Colombia
Universidad Autónoma de Sinaloa Arockiasamy, Madasamy
Los Mochis, Sinaloa, Mexico Florida Atlantic University
Almeida, Joao Boca Raton, FL
IST
Lisbon, Portugal

242 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Asaad, Diler Bae, Sungjin
Gaziantep University Bechtel Corporation
Gaziantep, Turkey Frederick, MD
Asamoto, Shingo Bagge, Niklas
Saitama University Luleå University of Technology
Saitama, Saitama, Japan Luleå, Sweden
Ashrafy, Mohammad Bai, Shaoliang
Islamic Azad University – Arak Branch Chongqing University
Kermanshah, Kermanshah, Islamic Republic of Iran Chonqqing, China
Aslani, Farhad Bai, Yongtao
University of Technology Sydney Kyoto University
Sydney, New South Wales, Australia Kyoto, Japan
Assaad, Joseph Balakumaran, Soundar
Notre Dame University Virginia Center for Transportation Innovation and Research
Beirut, Lebanon Charlottesville, VA
Atamturktur, Sez Balilaj, Mentor
Clemson University Polytechnic University of Tirana
Clemson, SC Tirana, Albania
Athanasopoulou, Adamantia Balouch, Sana
Metropolitan College University of Dundee
Xalandri, Attiki, Greece Dundee, UK
Avendano, Alejandro Banibayat, Pouya
Technological University of Panama ARUP
Doral, FL New York, NY
Aviram, Ady Banić, Davor
Simpson Gumpertz & Heger, Inc. Civil Engineering Institute of Croatia
San Francisco, CA Zagreb, Croatia
Awida, Tarek Baran, Eray
KEO International Consultants Middle East Technical University
Kuwait Ankara, Turkey
Awwad, Elie Barboza, Aline
Lebanese University, Branch II Universidade Federal de Alagoas
Mount Lebanon, Lebanon Maceio, Alagoas, Brazil
Ayano, Toshiki Barišić, Ivana
Okayama University Osijek, Croatia
Okayama, Japan Barragan, Bryan
Aydin, Abdulkadir Cuneyt BASF Construction Chemicals
Ataturk University Treviso, Treviso, Italy
Erzurum, Turkey Barroso de Aguiar, Jose
Aydin, Ertug University of Minho
European University of Lefke Guimaraes, Portugal
Nicosia, Turkey Bartos, Peter
Aykac, Sabahattin University of Paisley
Gazi University Paisley, UK
Ankara, Turkey Basava, Vamsi
Azad, Abul Malla Reddy Engineering College
King Fahd University of Petroleum & Minerals Hyderabad, Andhra Pradesh, India
Dhahran, Saudi Arabia Bashandy, Alaa
Azari, Hoda Menoufiya University
The University of Texas at El Paso Shibin El-Kom, Menoufiya, Egypt
El Paso, TX Batson, Gordon
Aziz, Omar Clarkson University
University of Salahaddin Potsdam, NY
Erbil, Iraq Baty, James
Babafemi, Adewumi Concrete Foundations Assoc
Obafemi Awolowo University Mount Vernon, IA
Ile-ife, Osun, Nigeria Bayraktar, Alemdar
Bacinskas, Darius Trabzon, Turkey
Vilnius Gediminas Technical University
Vilnius, Lithuania

ACI Structural Journal/March-April 2015 243

 REVIEWERS IN 2014
Bayuaji, Ridho Bilek, Vlastimil
Institut Teknologi epuluh Nopember ZPSV a.s.
Surabaya, East Java, Indonesia Brno, Czech Republic
Bebawy, Mena Bilir, Turhan
Lawrence Technological Univeristy Bülent Ecevit University
Southfield, MI Zonguldak, Turkey
Becq-Giraudon, Emilie Billah, Abu Hena
Chicago Department of Transportation The University of British Columbia
Chicago, IL Kelowna, BC, Canada
Beddar, Miloud Bimschas, Martin
M’sila University Regensdorf, Switzerland
M’sila, Algeria Binici, Baris
Bediako, Mark Middle East Technical University
CSIR—Building and Road Research Institute Ankara, Turkey
Kumasi, Ashanti, Ghana Birkle, Gerd
Bedirhanoglu, Idris Stantec Consulting Ltd.
Dicle University Calgary, AB, Canada
Diyarbakir, Turkey Bisby, Luke
Beglarigale, Ahsanollah University of Edinburgh
Dokuz Eylul University Edinburgh, UK
Izmir, Izmir, Turkey Bisschop, Jan
Behnam, Hamdolah University of Oslo
Hong Kong University of Science and Technology Oslo, Norway
Hong Kong, China Bobko, Christopher
Behnoud, Ali North Carolina State University
Iran University of Science and Tech Raleigh, NC
Tehran, Islamic Republic of Iran Bolhassani, Mohammad
Belagraa, Larbi Drexel University
Bordj Bou Arreridj, University Center Philadelphia, PA
Bordj Bou Arreridj, Algeria Bonacci, John
Belkowitz, Jon Karins Engineering Group
Stevens Institute of Technology Sarasota, FL
Freehold, NJ Bondar, Dali
Belleri, Andrea Tehran, Islamic Republic of Iran
University of Bergamo Bondy, Kenneth
Dalmine, Italy Consulting Structural Engineer
Benliang, Liang West Hills, CA
Shanghai, China Bonetti, Rodolfo
Bennett, Richard Pontificia Universidad Católica Madre y Maestra
The University of Tennessee Santo Domingo, Dominican Republic
Knoxville, TN Borges, Paulo
Bernard, Erik Stefan Federal Centre for Technological Education of Minas Gerais
TSE P/L Belo Horizonte, Minas Gerais, Brazil
Penrith, Australia Bouras, Rachid
Berry, Michael UMMTO
Montana State University Tiziouzou, Algeria
Bozeman, MT Bournas, Dionysios
Beygi, Morteza University of Nottingham
Mazandaran University Nottingham, UK
Babol - Mazandaran, Islamic Republic of Iran Bousias, Stathis
Bhangal, Malkit University of Patras
Thapar University Patras, Greece
Patiala, Punjab, India Bradberry, Timothy
Bharati, Raj TXDot Bridge Division
National Institute of Technology Calicut Austin, TX
Calicut, Kerala, India Braestrup, Mikael
Bhattacharjee, Bishwajit Ramboll Hannemann and Hojlund A/S
Indian Institute of Technology, Delhi Virum, Denmark
New Delhi, India Braimah, Abass
Carleton University
Ottawa, ON, Canada

244 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Brand, Alexander Carroll, Chris
University of Illinois University of Louisiana at Lafayette
Urbana, IL Lafayette, LA
Breña, Sergio Carvalho, Alessandra
University of Massachusetts Pontifical Catholic University of Goiás
Amherst, MA Goiânia, Goiás, Brazil
Brewe, Jared Castles, Bryan
CTLGroup Western Technologies Inc.
Skokie, IL Phoenix, AZ
Broujerdian, Vahid Castro, Javier
Iran University of Science and Technology Pontificia Universidad Católica de Chile
Tehran, Islamic Republic of Iran Santiago, Chile
Brown, Michael Catoia, Bruna
Virginia Transportation Research Council Federal University of São Carlos
Charlottesville, VA São Carlos, São Paulo, Brazil
Brown, Simon Cattaneo, Sara
Read Jones Christoffersen Ltd. Politecnico di Milano
Calgary, AB, Canada Milan, Italy
Bu, Wensheng Cavalaro, Sergio Henrique
Wardrop Engineering Inc. Universidad Politécnica de Cataluña
Sudbury, ON, Canada Barcelona, Spain
Burak, Burcu Çavuşoğlu, İbrahim
Orta Dogu Teknik Universitesi Gümüşhane University
Ankara, Turkey Gümüşhane, Turkey
Burris, Lisa Cervenka, Vladimir
University of Texas at Austin Cervenka Consulting
Austin, TX Petriny, Czech Republic
Byard, Benjamin Cetisli, Fatih
University of Tennessee at Chattanooga Pamukkale University
Chattanooga, TN Denizli, Turkey
Bzeni, Dallshad Chaallal, Omar
University of Salahaddin Ecole de Technologie Superiere
Erbil, Iraq Verdun, QC, Canada
Cai-Jun, Shi Chai, Hwa Kian
Hunan University Tobishima Corporation
Changsha, Hunan, China Noda, Chiba, Japan
Calixto, José Chakraborty, Arun
UFMG Bengal Engineering And Science University
Belo Horizonte, Brazil Howrah, West Bengal, India
Camero, Hugo Chang, Ta-Peng
Construdiseños Ingenieros Arquitectos S.A.S. NTUST
Bogota D.C., Colombia Taipei, Taiwan, China
Campione, Giuseppe Chao, Shih-ho
Universita Palermo University of Texas at Arlington
Palermo, Italy Arlington, TX
Cano Barrita, Prisciliano Chaudhary, Sandeep
Instituto Politécnico Nacional/CIIDIR Unidad Oaxaca Malaviya National Institute of Technology Jaipur
Oaxaca, Mexico Jaipur, Rajasthan, India
Canpolat, Fethullah Chaunsali, Piyush
Yildiz Technical University University of Illinois
Istanbul, Turkey Urbana, IL
Cao, Weiqun Chen, Chun-Tao
Qingdao, China National Taiwan University of Science and Technology
Capozucca, Roberto Taipei, Taiwan, China
Ancona, Italy Chen, Hua-Peng
Carino, Nicholas The University of Greenwich
Chagrin Falls, OH Chatham, UK
Carreira, Domingo Chen, Qi
Chicago, IL Boral Materials Technology
San Antonio, TX

ACI Structural Journal/March-April 2015 245

 REVIEWERS IN 2014
Chen, Shiming Chu Thi, Binh
Tongji University Hanoi Architectural University
Shanghai, China Hanoi, Vietnam
Chen, Wei Chung, Deborah
Wuhan University of Technology University at Buffalo, the State University of New York
Wuhan, Hubei, China Buffalo, NY
Chen, Xia Chung, Jae
Changjiang River Scientific Research Institute University of Florida
Wuhan, China Gainesville, FL
Cheng, Min-Yuan Chung, Lan
National Taiwan University of Science and Technology Dankook University
Taipei, Taiwan, China Seoul, Republic of Korea
Chi, Maochieh Cintra, Danielli
Wufeng University Vitória, Espírito Santo, Brazil
Chiayi County, Taiwan, China Claisse, Peter
Chiang, Chih-Hung Coventry University
Chaoyang University of Technology Coventry, UK
Wufong, Taichung, Taiwan, China Cleland, Ned
Chindaprasirt, Prinya Blue Ridge Design Inc.
Khon Kaen University Winchester, VA
Khon Kaen, Thailand Climent, Miguel
Cho, Jae-Yeol University of Alacant
Seoul National University Alacant, Spain
Seoul, Republic of Korea Coelho, Jano
Cho, Soon-Ho Altoqi Informatica
Gwangju University Florianopolis, Santa Catarina, Brazil
Gwangju, Republic of Korea Colombo, Matteo
Choi, Chang-Sik Politecnico di Milano
Hnayang University Lecco, Italy
Seoul, Republic of Korea Cordova, Carlos
Choi, Eunsoo La Paz, Bolivia
Hongik University Costa, Ricardo
Seoul, Republic of Korea University of Coimbra
Choi, Hyun-Ki Coimbra, Portugal
Hanyang University Crespi, Pietro
Seoul, Republic of Korea Politecnico of Milan
Choi, Kyoung-Kyu Milano, Italy
Soongsil University Criswell, Marvin
Seoul, Republic of Korea Colorado State University
Choi, Sejin Fort Collins, CO
University of California, Berkeley Cueto, Jorge
Berkeley, CA Universidad de La Salle
Chompreda, Praveen Bogota, Colombia
Mahidol University Cumming, Neil
Nakornpathom, Thailand Levelton Engineering Ltd
Choong, Kokkeong Richmond, BC, Canada
Universiti Sains Malaysia D’agata, Giuseppe
Pulau Pinang, Seberang Perai Selatan, Malaysia University of Catania
Chorzepa, Migeum Catania, Italy
Park Ridge, IL d’Andréa, Renata
Chowdhury, Sharmin Getafe, Madrid, Spain
Bogazici University Dang, Canh
Istanbul, Turkey University of Arkansas
Chowdhury, Subrato Fayetteville, AR
Ultra Tech Cement LTD D’Arcy, Thomas
Mumbai, Maharashtra, India Consulting Engineers Group
Christen, Robert San Antonio, TX
American Engineering Testing Inc. Darwin, David
Port Charlotte, FL University of Kansas
Lawrence, KS

246 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Das, Sreekanta Diaz Loya, Eleazar
University of Windsor Louisiana Tech University
Windsor, ON, Canada Ruston, LA
Das Adhikary, Satadru Ding, Yining
National University of Singapore Dalian, China
Singapore Diniz, Sofia Maria
De Brito, Jorge Universidade Federal de Minas Gerais
IST / TUL Be lo Horizonte, Brazil
Lisbon, Portugal Do, Jeongyun
De Korte, Ariën Kunsan National University
University of Twente Kunsan, Jeonbuk, Republic of Korea
Enschede, the Netherlands Dogan, Unal
De Rooij, Mario Istanbul Technical University
TNO Istanbul, Turkey
Delft, the Netherlands Dolan, Charles
De Schutter, Geert University of Wyoming
Ghent University Laramie, WY
Ghent, Belgium Dongell, Jonathan
Deb, Arghya Pebble Technologies
Indian Institute of Technology, Kharagpur Scottsdale, AZ
Kharagpur, West Bengal, India Dontchev, Dimitar
Decker, Curtis University of Chemical Technology and Metallurgy
U.S. Military Academy Sofia, Bulgaria
West Point, NY Dotreppe, Jean-Claude
Degtyarev, Vitaliy Université of Liege-Mehanique Genie Civil
Columbia, SC Liege, Belgium
Delalibera, Rodrigo Du, Hongjian
University of São Paulo National University of Singapore
São Carlos, São Paulo, Brazil Singapore
Demir, Serhat Du, Lianxiang
Blacksea Technical Univesity The University of Alabama at Birmingham
Trabzon, Turkey Birmingham, AL
Den Uijl, Joop Du, Jinsheng
Delft University of Technology Beijing Jiao Tong University
Delft, the Netherlands Beijing, China
Deng, Mingke Du, Yingang
Xi’an University of Architecture and Technology Anglia Ruskin University, UK
Xi’an, Shaanxi, China Chelmsford, UK
Deng, Yaohua Dutta, Anjan
Iowa State University Indian Institute of Technology Guwahati
Ames, IA Guwahati, Assam, India
Devries, Richard Ebead, Usama
Milwaukee School of Engineering Qatar University
Milwaukee, WI Doha, Qatar
Dhinakaran, G. Eid, Rami
Sastra University University of Sherbrooke
Thanjavur, India Sherbrooke, QC, Canada
Dhonde, Hemant El Meski, Fatima
University of Houston American University of Beirut
Houston, TX Beirut, Lebanon
Di Ludovico, Marco El Ragaby, Amr
University of Naples Federico II University of Manitoba
Naples, Italy Winnipeg, MB, Canada
Diao, Bo El Sayed, Mohamed
Beihang University University of Windsor
Beijing, China Windsor, ON, Canada
Dias, W. P. S. Elamin, Anwar
University of Moratuwa University of Nyala
Moratuwa, Sri Lanka Nyala, Sudan

ACI Structural Journal/March-April 2015 247

 REVIEWERS IN 2014
Elbatanouny, Mohamed Evangelista, Luís
University of South Carolina Instituto Superior de Engenharia de Lisboa
Columbia, SC Lisbon, Portugal
El-Dash, Karim Faleschini, Flora
College of Technological Studies University of Padova
Kuwait Padova, Italy
El-Dieb, Amr Fantilli, Alessandro
Ain Shams University Politecnico di Torino
Abbasia, Cairo, Egypt Torino, Italy
El-Hawary, Moetaz Fardis, Michael
Kuwait Institute for Scientific Research Patras, Greece
Safat, Kuwait Farghaly, Ahmed
El-Maaddawy, Tamer University of Sherbrooke
United Arab Emirates University Sherbrooke, QC, Canada
Al-Ain, Abu Dhabi, United Arab Emirates Faria, Duarte
El-Metwally, Salah Faculdade de Ciências e Tecnologia
University of Hawaii at Manoa Caparica-Lisbon, Portugal
Honolulu, HI Farrokhi, Farhang
El-Refaie, Sameh Zanjan, Islamic Republic of Iran
El-Gama City, Mataria, Cairo, Egypt Farrow, William
El-Salakawy, Ehab Lebanon, NJ
University of Manitoba Farzam, Masood
Winnipeg, MB, Canada Tabriz, Islamic Republic of Iran
El-Sayed, Ahmed Fathi, Hamoon
University of Sherbrooke Sanandaj Branch, Islamic Azad University
Sherbrooke, QC, Canada Sanandaj, Kurdistan, Islamic Republic of Iran
Elfgren, Lennart Feldman, Lisa
Luleå University of Technology University of Saskatchewan
Luleå, Sweden Saskatoon, SK, Canada
Elhashmy, Awad Felekoglu, Burak
Cairo, Egypt Dokuz Eylul University
Elkady, Hala Izmir, Turkey
NRC Fernández Montes, David
Giza, Egypt Madrid, Spain
Elnady, Mohamed Fernández Ruiz, Miguel
Mansoura University Ecole Polytechnique Federale De Lausanne
Vancouver, BC, Canada Lausanne, Vaud, Switzerland
Elmenshawi, Abdelsamie Ferrara, Liberato
University of Calgary Politecnico di Milano
Calgary, AB, Canada Milan, Italy
Elsayed, Tarek Ferrier, E.
Cairo, Egypt Université Lyon 1
Emamy Farvashany, Firooz Villerubanne, France
Perthpolis Pty Ltd Folino, Paula
Perth, Western Australia, Australia University of Buenos Aires
Erdem, T. Buenos Aires, Argentina
Izmir Institute of Technology Foraboschi, Paolo
Izmir, Turkey Universita IUAV di Venezia
Ergün, Ali Venice, Italy
Afyonkarahisar, Turkey Fouad, Fouad
Esmaeily, Asad University of Alabama at Birmingham
Kansas State University Birmingham, AL
Manhattan, KS Fradua, Martin
Esmaili, Omid Feld, Kaminetzky & Cohen, P.C.
University of California, Irvine Jericho, NY
Irvine, CA Francüois, Buyle-Bodin
Esperanza, Menendez University of Lille
IETCC-CSIC Villeneuve d’Ascq, France
Madrid, Spain Freyne, Seamus
Etman, Emad Manhattan College
El-Mahalla El-Kobra, Egypt Riverdale, NY

248 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Fuchs, Werner Giaccio, Craig
University of Stuttgart AECOM
Stuttgart, Germany Melbourne, Victoria, Australia
Fuentes, Jose Maria Girgin, Canan
Polytechnic University of Madrid Yildiz Technical University
Madrid, Spain Istanbul, Turkey
Furlong, Richard Goel, Rajeev
Austin, TX CSIR-Central Road Research Institute
Gabrijel, Ivan Delhi, India
University of Zagreb Gökçe, H. Süleyman
Zagreb, Croatia Ege University
Gajdosova, Katarina Izmir, Turkey
Bratislava, Slovakia Gongxun, Wang
Galati, Nestore Hunan University of Science and Technology
Structural Group Inc. Xiangtan, China
Elkridge, MD Gonzales Garcia, Luis Alberto
Gallegos Mejia, Luis Lagging SA
Fundacion Padre Arrupe de El Salvador Lima, Peru
Soyapango, San Salvador, El Salvador González, Javier
Gamble, William University of Basque Country
University of Illinois Bilbao, Basque Country, Spain
Urbana, IL González-Valle, Enrique
Gao, Xiangling Madrid, Spain
Tongji University Goudarzi, Nabi
Shanghai, China Edmonton, AB, Canada
Garber, David Grandić, Davor
Florida International University University of Rijeka
Miami, FL Rijeka, Croatia
Garcez, Estela Gribniak, Viktor
Universidade Federal de Pelotas Vilnius Gediminas Technical University
Pelotas, RS, Brazil Vilnius, Lithuania
Garcia-Taengua, Emilio Gu, Xiang-Lin
Queen’s University of Belfast Tongji University
Belfast, UK Shanghai, China
Gedik, Yasar Guadagnini, Maurizio
Istanbul Technical University The University of Sheffield
Istanbul, Turkey Sheffield, UK
Gesoglu, Mehmet Guan, Garfield
Gaziantep University Cambridge, UK
Gaziantep, Turkey Guimaraes, Giuseppe
Gettu, Ravindra Pontificia Universidade Católica do Rio de Janeiro
Indian Institute of Technology Madras Rio de Janeiro, Brazil
Chennai, India Gulec, Cevdet
Ghafari, Nima Thornton Tomasetti, Inc.
Laval University Los Angeles, CA
Quebec, QC, Canada Güneyisi, Erhan
Ghali, Amin Gaziantep University
University of Calgary Gaziantep, Turkey
Calgary, AB, Canada Guo, Honglei
Ghanem, Hassan Wuhan Polytechnic University
Texas A&M University Wu Han City, Hu bei Province, China
College Station, TX Guo, Liping
Ghasemzadeh, Farnam Southeast University
North Carolina State University Nanjing, Jiangsu Province, China
Raleigh, NC Guo, Zixiong
Ghezal, Aïcha Huaqiao University
Ecole de Technologie de Montreal Quanzhou, Fujian, China
Montreal, QC, Canada Gupta, Ajay
Ghoddousi, Parviz M.B.M. Engineering College
Iran University of Science and Technology Jodhpur, Rajasthan, India
Tehran, Islamic Republic of Iran

ACI Structural Journal/March-April 2015 249

 REVIEWERS IN 2014
Gupta, Rishi Hassan, Wael
Vancouver, BC, Canada University of California, Berkeley
Gupta, Supratic Berkeley, CA
Indian Institute of Technology, Delhi Hassani, Abolfazl
New Delhi, India Tarbiat Modares University
Haddad, Rami Tehran, Islamic Republic of Iran
Jordan University of Science and Technology He, Zhiqi
Irbid, Jordan Southeast University
Haddadin, Laith Nanjing, Jiangsu, China
United Nations Heinzmann, Daniel
New York, NY Lucerne University of Applied Sciences and Arts
Hadje-Ghaffari, Hossain Horw, Switzerland
John A. Martin & Assoc. Helal, Yasser
Los Angeles, CA University of Sheffield
Hagenberger, Michael Sheffield, UK
Ohio State University Helmy, Huda
Columbus, OH Applied Science International
Haggag, Hesham Durham, NC
Cairo, Egypt Hemalatha, T.
Hamilton, Trey CSIR-Structural Engineering Research Centre
University of Florida Chennai, Tamil Nadu, India
Gainesville, FL Henry, Richard
Hammood, Oday University of Auckland
University Technology Malaysia Auckland, New Zealand
Skudai, Johor Buhro, Malaysia Herrera, Angel
Han, Dongyeop Rio Piedras, Puerto Rico
University of Texas at Austin Hindi, Riyadh
Austin, TX Saint Louis University
Harajli, Mohamed St. Louis, MO
American University of Beirut Ho, Johnny
Beirut, Lebanon The University of Hong Kong
Harbec, David Hong Kong, China
Université de Sherbrooke Hochstein, Daniel
Sherbrooke, QC, Canada Manhattan College
Hariri-Ardebili, Mohammad Amin Riverdale, NY
University of Colorado Hoehler, Matthew
Boulder, CO Encinitas, CA
Harries, Kent Hoff, George
University of Pittsburgh Hoff Consulting Inc.
Pittsburgh, PA Clinton, MS
Harris, Devin Holschemacher, Klaus
University of Virginia HTWK Leipzig
Charlottesville, VA Leipzig, Germany
Harris, G. Terry Hong, Sung-Gul
Green Cove Springs, FL Seoul National University
Hasan, Sahar Seoul, Republic of Korea
Higher Institute for Engineering and Technology Hosny, Amr
Alexandria, Egypt North Carolina State University
Hashemi, Shervin Raleigh, NC
Seoul National University Hossain, Mustaque
Seoul, Republic of Korea Kansas State University
Hasnat, Ariful Manhattan, KS
University of Asia Pacific Hoult, Neil
Dhaka, Bangladesh Toronto, ON, Canada
Hassan, Assem Hrynyk, Trevor
Toronto, ON, Canada University of Texas at Austin
Hassan, Mohamed Austin, TX
University of Sherbrooke Hu, Jiong
Sherbrooke, QC, Canada Texas State University
San Marcos, TX

250 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Hu, Nan Issa, Mohsen
Tsinghua University University of Illinois at Chicago
Beijing, China Chicago, IL
Huang, Yishuo Izquierdo-Encarnación, Jose
Chaoyang University of Technology Porticus
Wufeng, Taichung, Taiwan, China San Juan, Puerto Rico
Huang, Zhaohui Jaari, Asaad
Brunel University Dera, Dubai, United Arab Emirates
London, UK Jain, Mohit
Huang, Chang-Wei Nirma University
Chung Yuan Christian University Ahmedabad, Gujarat, India
Chung Li, Taiwan, China Jain, Shashank
Huang, Chung-Ho Delhi Technological University (DTU)
Dahan Institute of Technology New Delhi, India
Hualien, Taiwan, China Jalal, Mostafa
Huang, Jianwei PWUT
Southern Illinois University Edwardsville Tehran, Islamic Republic of Iran
Edwardsville, IL Jan, Song
Huang, Xiaobao Bechtel Corp.
GM-WFG/GM-N American Project Center Houston, TX
Warren, MI Jang, Seung Yup
Huo, Jingsi Korea Railroad Research Institute
Hunan University Uiwang, Gyongggi-do, Republic of Korea
Changsha, Hunan, China Jansen, Daniel
Husain, Mohamed California Polytechnic State University
Zagazig University San Luis Obispo, CA
Zagazig, Egypt Jawaheri Zadeh, Hany
Husem, Metin Miami, FL
Karadeniz Technical University Jayapalan, Amal
Trabzon, Turkey Exponent Failure Analysis Associates
Huynh, Minh Phuoc Menlo Park, CA
Ho Chi Minh City University Transport Jeng, Chyuan-Hwan
Ho Chi Minh, Vietnam National Chi Nan University-Taiwan
Ibell, Tim Puli/Nantou, Taiwan, China
University of Bath Jiang, Jiabiao
Bath, UK W R Grace (Singapore) Pte Ltd
Ibrahim, Amer Singapore
Baquba, Iraq Johnson, Gaur
Ichinose, Toshikatsu University of Hawaii
Nagoya Institute of Technology Honolulu, HI
Nagoya, Japan Jozić, Dražan
Ikponmwosa, Efe Split, Croatia
University of Lagos Kaklauskas, Gintaris
Akoka, Lagos, Nigeria Vilnius Gediminas Technical University
Ince, Ragip Vilnius, Lithuania
Firat University Engineering Faculty Kan, Yu-Cheng
Elazig, Turkey Chaoyang University of Technology
Ipek, Süleyman Taichung County, Taiwan, China
Gaziantep University Kanagaraj, Ramadevi
Gaziantep, Turkey Kumaraguru College of Technology
Irassar, Edgardo Coimbatore, Tamilnadu, India
National University of Central Buenos Aires Kanakubo, Toshiyuki
Olavarria, Buenos Aires, Argentina University of Tsukuba
Islam, Md. Tsukuba, Japan
Chittagong University of Engineering & Technology (CUET) Kandasami, Siva
Chittagong, Bangladesh Bristol, UK
Issa, Mohamed Kang, Thomas
National Center for Housing and Building Resarch Seoul National University
Giza, Egypt Seoul, Republic of Korea

ACI Structural Journal/March-April 2015 251

 REVIEWERS IN 2014
Kankam, Charles Klein, Gary
Kwame Nkrumah University of Science & Technology Wiss, Janney, Elstner Associates, Inc.
Kumasi, Ghana Northbrook, IL
Kansara, Kunal Klemczak, Barbara
Mouchel Infrastructure Services Silesian Technical University
Bristol, UK Gliwice, Poland
Kantarao, Velidandi Ko, Lesley Suz-Chung
Central Road Research Institute Holcim Group Support Ltd.
New Delhi, India Holderbank, AG, Switzerland
Karayannis, Christos Koehler, Eric
Democritus University of Thrace University of Texas at Austin
Xanthi, Greece Austin, TX
Karbasi Arani, Kamyar Koenders, Eddy A. B.
University of Naples Federico II Delft University of Technology
Napoli, Campagna, Italy Delft, the Netherlands
Kawamura, Mitsunori Konsta-Gdoutos, Maria
Kanazawa, Ishikawa, Japan Northwestern University
Kazemi, Mohammad Evanston, IL
Sharif University of Technology Kotsovos, Gerasimos
Tehran, Islamic Republic of Iran National Technical University of Athens
Kazemi, Sadegh Athens, Greece
University of Alberta Kotsovos, Michael
Edmonton, AB, Canada Athens, Greece
Kenai, Said Kreger, Michael
Université de Blida Purdue University
Blida, Algeria West Lafayette, IN
Khan, Mohammad Książek, Mariusz
King Saud University Wrocław University of Technology
Riyadh, Saudi Arabia Wroclaw, Poland
Khan, Sadaqat Kumar, Pardeep
Universiti Teknologi Petronas University of California, Berkeley
Tronoh, Perak, Malaysia Berkeley, CA
Khuntia, Madh Kumar, Rakesh
Dukane Precast Inc. Central Road Research Institute
Naperville, IL Delhi, India
Kianoush, M. Reza Kumaravel, S.
Ryerson University Annamalai University
Toronto, ON, Canada Cuddalore, Tamilnadu, India
Kim, Jang Hoon Kupwade-Patil, Kunal
Ajou University Massachusetts Institute of Technology
Suwon, Republic of Korea Cambridge, MA
Kim, Sang-Woo Kurtis, Kimberly
Kongju National University Georgia Institute of Technology
Cheonan, Chungnam, Republic of Korea Atlanta, GA
Kim, Woo Kusbiantoro, Andri
Chonnam National University Universiti Malaysia Pahang
Kwangju, Republic of Korea Gambang, Pahang, Malaysia
Kim, Yail Jimmy Kuyucular, Adnan
University of Colorado Denver Pamukkale University
Denver, CO Kinikli-Denizli, Turkey
Kirgiz, Mehmet Kwan, Albert
Hacettepe University The University of Hong Kong
Ankara, Turkey Hong Kong, China
Kishen, Chandra Lai, James
Indian Institute of Science La Cañada, CA
Bangalore, Karnataka, India Lai, Jianzhong
Kishi, Norimitsu Nanjing University of Science and Technology
Muroran Institute of Technology Nanjing, Jiangsu, China
Muroran, Japan Laldji, Said
Université de Sherbrooke
Sherbrooke, QC, Canada

252 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Lam, Eddie Li, Fumin
The Hong Kong Polytechnic Universiy China University of Mining and Technology
Hong Kong, China Xuzhou, Jiangsu, China
Larbi, Kacimi Li, Long-yuan
University of Sciences and Technology of Oran University of Plymouth
Oran, Algeria Plymouth, UK
Laskar, Aminul Li, Wei
National Institute of Technology Wenzhou University
Silchar, Assam, India Wenzhou, Zhejiang, China
Laterza, Michelangelo Li, Xinghe
University of Basilicata University of New Hampshire
Potenza, Italy Durham, NH
Latifee, Enamur Li, Yi-An
Clemson University National Taiwan University
Clemson, SC Taipei, Taiwan, China
Law, David Lignola, Gian Piero
RMIT University University of Naples Federico II
Melbourne, Victoria, Australia Naples, Italy
Lawler, John Lin, Wei-Ting
Wiss, Janney, Elstner Associates, Inc. Ilan, Taiwan, China
Northbrook, IL Lin, Zhibin
Lawrence, Adrian Fargo, ND
Gainesville, FL Liu, Jun
Lee, Chi King Beijing, China
Nanyang Technological University Liu, Junshan
Singapore Sargent Lundy LLC
Lee, Chung-Sheng Chicago, IL
University of California, San Diego Liu, Shuhua
La Jolla, California Wuhan University
Lee, Deuck Hang Wuhan, Hubei, China
University of Seoul Liu, Zhao
Seoul, Republic of Korea Southeast University
Lee, Douglas Nanjing, Jiangsu, China
Douglas D. Lee and Associates Liu, Xuejian
Fort Worth, TX University of Texas at Arlington
Lee, Heui Hwang Arlington, TX
Arup Liu, Yanbo
San Francisco, CA Florida Atlantic University
Lee, Hung-Jen Boca Raton, FL
National Yunlin University of Science and Technology Liu, Ze
Douliu, Yunlin, Taiwan, China China University of Mining & Technology, Beijing
Lee, Jung-Yoon Beijing, China
Sung Kyun Kwan University Lo, T. Y.
Suwon, Republic of Korea City University of Hong Kong
Lee, Nam Ho Hong Kong, China
SNC-Lavalin Nuclear Long, Nguyen
Oakville, ON, Canada Kosice, Slovakia
Lee, Seong-Cheol Long, Xu
KEPCO International Graduate School (KINGS) Nanyang Technological University
Ulsan, Republic of Korea Singapore
Lei, Aizhong Loo, Yew-Chaye
China Institute of Water Resources and Hydropower Research Gold Coast, Australia
Beijing, China Lopes, Anne
Lepage, Andres Furnas Centrais Eletricas Sa Aparecida De Goiania
University of Kansas Goias, Brazil
Lawrence, KS Lopes, Sergio
Lequesne, Remy University of Coimbra
University of Kansas Coimbra, Portugal
Lawrence, KS López-Almansa, Francisco
Technical University of Catalonia
Barcelona, Spain

ACI Structural Journal/March-April 2015 253

 REVIEWERS IN 2014
Lotfy, Abdurrahmaan Marikunte, Shashi
Lafarge Canada Inc. Southern Illinois University
Toronto, ON, Canada Carbondale, IL
Lounis, Zoubir Martí-Vargas, José
National Research Council Universitat Politècnica de València
Ottawa, ON, Canada Valencia, Spain
Lubell, Adam Martinelli, Enzo
Read Jones Christoffersen Ltd. University of Salerno
Vancouver, BC, Canada Fisciano, Italy
Ludovit, Nad Maruyama, Ippei
Alfa 04 Nagoya University
Kosice, Slovakia Nagoya, Aichi, Japan
Luo, Baifu Maslehuddin, Mohammed
Harbin, China King Fahd University of Petroleum and Minerals
Lushnikova, Nataliya Dhahran, Saudi Arabia
National University of Water Management and Matta, Fabio
Nature Resources Use University of South Carolina
Rivne, Ukraine Columbia, SC
Ma, Zhongguo Maximos, Hany
University of Tennessee Pharos University in Alexandria
Knoxville, TN Alexandria, Egypt
MacDonald, Kevin Mbessa, Michel
Cemstone Concrete Products Co. University of Yaoundé I - ENSP
Mendota Heights, MN Yaoundé, Center, Cameroon
Machida, Atsuhiko McCarter, John
Saitama University Heriot Watt University
Saitama, Japan Edinburgh, UK
Macht, Jürgen McDonald, David
Kirchdorf, Austria USG Corp
Maekawa, Koichi Libertyville, IL
University of Tokyo McLeod, Heather
Tokyo, Japan Kansas Department of Transportation
Maganti, Ravindra Topeka, KS
D.M.S. S.V.H. College of Engineering Machilipatnam Meda, Alberto
Andhra Pradesh, India University of Bergamo
Magliulo, Gennaro Bergamo, Italy
University of Naples Federico II Medallah, Khaled
Naples, Italy Saudi Aramco IKPMS
Maguire, Marc Al Khobar, Saudi Arabia
Utah State University Meddah, Mohammed Seddik
Paradise, UT Kingston University London
Mahfouz, Ibrahim Kingston, UK
Cairo, Egypt Mehanny, Sameh
Mahrenholtz, Christoph Cairo University
Berlin, Germany Cairo, Egypt
Mahrenholtz, Philipp Meinheit, Donald
Frankfurt, Germany Wiss, Janney, Elstner Associates, Inc.
Malik, Adnan Chicago, IL
University of New South Wales Melo, José
Sydney, Australia University of Aveiro
Mander, John Aveiro, Portugal
Texas A&M University Meng, Tao
College Station, TX Institution of Building Materials
Manso, Juan Hangzhou, Zhejiang, China
University of Burgos Menon, Devdas
Burgos, Castilla - León, Spain Indian Institute of Technology
Mari, Antonio Chennai, Tamilnadu, India
Universitat Politecnica de Catalunya Mermerdaş, Kasım
Barcelona, Spain Hasan Kalyoncu University
Gaziantep, Turkey

254 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Meshgin, Pania Mullapudi, Taraka Ravi
University of Colorado Boulder MMI Engineering
Boulder, CO Houston, TX
Milestone, Neil Munoz, Jose
Callaghan Innovation Federal Highway Administration
Lower Hutt, New Zealand McLean, VA
Minehane, Michael Muttoni, Aurelio
RPS Group Ltd. Swiss Federal Institute of Technology
Cork, Ireland Lausanne, Switzerland
Mlynarczyk, Alexandar Nabavi, Esrafil
Wiss, Janney, Elstner Associates, Inc. Rezvanshahr, Guilan, Islamic Republic of Iran
Princeton Junction, NJ Nafie, Amr
Mo, Yi-Lung Cairo, Egypt
University of Houston Nair, Priya
Houston, TX Cochin University of Science and Technology
Mohamed, Ashraf Kochi, Kerala, India
Alexandria University Naish, David
Alexandria, Egypt California State University, Fullerton
Mohamed, Nayera Fullerton, CA
Assiut University Najimi, Meysam
Assiut, Egypt University of Nevada, Las Vegas
Mohammadyan Yasouj, Seyed Esmaeil Las Vegas, NV
UTM University Nakamura, Hikaru
Johor, Malaysia Nagoya University
Mohammed, Tarek Nagoya, Aichi, Japan
University of Asia Pacific Nam, Boo Hyun
Dhaka, Bangladesh University of Central Florida
Mohd Zain, Mumammad Fauzi Orlando, FL
Universiti Kebangsaan Malaysia Negrutiu, Camelia
Bangi, Malaysia Technical University of Cluj Napoca
Mokarem, David Cluj Napoca, Cluj, Romania
Virginia Polytechnic University Neves, Luís
Blacksburg, VA University of Coimbra
Mondal, Bipul Coimbra, Portugal
Chittagong University of Engineering & Technology Ng, Ivan
Chittagong, Bangladesh Drainage Services Department
Montejo, Luis Hong Kong, China
North Carolina State University Nichols, John
Raleigh, NC Texas A&M University
Moradi, Hiresh College Station, TX
Amirkabir University of Technology Niemuth, Mark
Tehran, Islamic Republic of Iran Lafarge
Moretti, Marina Alpharetta, GA
University of Thessaly Nimityongskul, Pichai
Athens, Greece Asian Institute of Technology
Moser, Robert Pathumthani, Thailand
U.S. Army Engineer Research and Development Center Nishiyama, Minehiro
Vicksburg, MS Kyoto University
Mostafaei, Hossein Kyoto, Japan
University of Toronto Noor, Munaz
Toronto, ON, Canada Bangladesh University of Engineering and Technology
Mostofinejad, Davood Dhaka, Bangladesh
Isfahan University of Technology Noshiravani, Talayeh
Isfahan, Islamic Republic of Iran EPFL
Muciaccia, Giovanni Lausanne, Switzerland
Politecnico di Milano Ochotorena, Richard
Milan, Italy Permasteelisa Group
Mulaveesala, Ravibabu Hong Kong Island, Hong Kong, China
Indian Institute of Technology Ropar Oh, Byung
Rupnagar, India Seoul National University
Seoul, Republic of Korea

ACI Structural Journal/March-April 2015 255

 REVIEWERS IN 2014
Ohtsu, Masayasu Pantazopoulou, Stavroula
Kumamoto University Demokritus University of Thrace
Kumamoto, Japan Xanthi, Greece
Okeil, Ayman Pape, Torill
Louisiana State University University of Newcastle
Baton Rouge, LA Callaghan, New South Wales, Australia
Okelo, Roman Parsekian, Guilherme
Dallas, TX Federal University of São Carlos
Olanitori, Lekan São Carlos, São Paulo, Brazil
Federal University of Technology, Akure Pauletta, Margherita
Akure, Ondo State, Nigeria University of Udine
Ombres, Luciano Tavagnacco, Udine, Italy
University of Calabria Paulotto, Carlo
Cosenza, Italy Acciona S.A.
Omran, Ahmed Alcobendas, Spain
University of Sherbrooke Pavlikova, Milena
Sherbrooke, QC, Canada CTU
Orakcal, Kutay Prague, Czech Republic
Bogazici University Pellegrino, Carlo
Istanbul, Bebek, Turkey University of Padova
Orr, John Padova, Italy
University of Bath Peng, Cao
Bath, UK Harbin Institute of Technology
Ortega, J. Harbin, Heilongjiang, China
University of Alacant Peng, Jianxin
Alacant, Alicante, Spain Institute of Bridge Engineering
Ortiz-Lozano, Jose Changsha, Hunan, China
Autonomous University of Aguascalientes Pereira, Eduardo
Aguascalientes, Mexico University of Minho
Orton, Sarah Guimaraes, Portugal
University of Missouri Columbia Perez Caldentey, Alejandro
Columbia, MO Universidad Politécnica de Madrid
Osifala, Kehinde Madrid, Spain
Somolu, Lagos, Nigeria Persson, Bertil
Otieno, Mike Bara, Sweden
University of the Witwatersrand Phillippi, Don
Johannesburg, Gauteng, South Africa Diamond Pacific
Ouzaa, Kheira Rancho Cucamonga, CA
USTO Piccinin, Roberto
ORAN, Algeria Hilti, Inc.
Ozturan, Turan Tulsa, OK
Bogazici University Pocesta, Ylli
Istanbul, Turkey Debar, The Former Yugoslav Republic of Macedonia
Ozturk, Ali Ponnada, Markandeya
Dokuz Eylul University MVGR College of Engineering
Izmir, Buca, Turkey Vizianagaram, Andhra Pradesh, India
Pacheco, Alexandre Potnoor, Naveen
Universidade Federal do Rio Grande do Sul (UFRGS) Sasan, Madhya Pradesh, India
Porto Alegre, RS, Brazil Potter, William
Palieraki, Vasiliki Florida Department of Transportation
National Technical University of Athens Tallahassee, FL
Athens, Zografou, Greece Pourazin, Khashaiar
Palmisano, Fabrizio Pars Ab Tadbir Consulting Engineers Co.
Politecnico di Bari Tehran, Islamic Republic of Iran
Bari, Italy Prasad, Saurabh
Pan, Wang Fook University of California, San Diego
Segi University La Jolla, CA
Petaling Jaya, Selangor, Malaysia Prashanth, P.
Pang, Sze Dai SJCE
National University of Singapore Mysore, Karnataka, India
Singapore

256 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Prasittisopin, Lapyote Riad, Khaled
Oregon State University Ain Shams University
Corvallis, OR Cairo, Egypt
Puertas, F. Riding, Kyle
Eduardo Torroja Institute Kansas State University
Madrid, Spain Manhattan, KS
Puthenpurayil Thankappan, Santhosh Rinaldi, Zila
Granite Construction Company University of Rome Tor Vergata
Abu Dhabi, United Arab Emirates Rome, Italy
Putra Jaya, Ramadhansyah Rivard, Patrice
Universiti Teknologi Malaysia Université de Sherbrooke
Skudai, Johor Bahru, Malaysia Sherbrooke, QC, Canada
Qasrawi, Hisham Rivero-Angeles, Francisco
The Hashemite University Mexico, DF, Mexico
Zarqa, Jordan Rizk, Emad
Qian, Kai Memorial University of Newfoundland
Nanyang Technological University St. John’s, NL, Canada
Singapore Rizwan, Syed Ali
Qiangqiang, Zhang University of Engineering and Technology
Harbin Institute of Technology Lahore, Punjab, Pakistan
Harbin, Heilongjiang, China Rodriguez, Mario
Quaranta, Giuseppe National Autonomous University of Mexico
Sapienza University of Rome Mexico City, DF, Mexico
Rome, Italy Sabouni, Faisal
Quiroga, Pedro Architectural Consulting Group
Escuela Colombiana de Ingenieria Abu Dhabi, United Arab Emirates
Bogota, Colombia Sadeghi Pouya, Homayoon
Rafi, Muhammad Coventry University
NED University of Engineering and Technology Coventry, UK
Karachi, Sindh, Pakistan Saedi, Houman
Raikar, Chetan Tabiat Modares & Tabriz University and TSML CO.
Structwel Designers and Consultants Pvt. Ltd. Tehran, Islamic Republic of Iran
Navi Mumbai, India Safan, Mohamed
Ramamurthy, K. Menoufia University
IIT Madras Shebeen El-Kom, Menoufia, Egypt
Chennai, Tamilnadu, India Sagaseta, Juan
Ramaswamy, Ananth University of Surrey
Indian Institute of Science Guildford, Surrey, UK
Bangalore, Karnataka, India Sagues, Alberto
Ramos, António University of South Florida
Faculdade de Ciências e Tecnologia Tampa, FL
Monte de Caparica, Portugal Sahamitmongkol, Raktipong
Rao, Hanchate CONTEC, SIIT, Thammasat University and MTEC
JNTU College of Engineering Pathumthani, Thailand
Anantapur, India Sahmaran, Mustafa
Rashed, Youssef F. Gazi University
Giza, Egypt Ankara, Turkey
Rasol, Mezgeen Saibabu, S.
Dohuk Polytechince University CSIR-Structural Engineering Research Centre
Zakho, Duhok, Iraq Chennai, Tamilnadu, India
Ray, Indrajit Sajedi, Fathollah
Purdue University Calumet University of Malaya
Hammond, IN Kuala Lumpur, Selangor, Malaysia
Regan, Paul Saka, Mehmet
Trigram Middle East Technical University
London, UK Ankara, Turkey
Restrepo, Jose Salem, Hamed
University of California-San Diego Cairo University
La Jolla, CA Giza, Egypt
Salib, Sameh
Markham, ON, Canada

ACI Structural Journal/March-April 2015 257

 REVIEWERS IN 2014
Salinas-Basualdo, Rafael Shah, Santosh
National University of Engineering Dharmsinh Desai University
Lima, Peru Nadiad, Gujarat, India
Sallam, Hossam El-Din Shannag, M. Jamal
Zagazig University King Saud University
Zagazig, Sharkia, Egypt Riyadh, Saudi Arabia
Sanada, Yasushi Shao, Yixin
Toyohashi University of Technology McGill University
Toyohashi, Japan Montreal, QC, Canada
Sánchez, Isidro Sharifi, Yasser
University of Alicante Vali-e-Asr University of Rafsanjan
Alicante, Spain Rafsanjan, Islamic Republic of Iran
Sanchez, Leandro Sharma, Akanshu
São Paulo, Brazil Institute of Construction Materials
Santos, Sérgio Stuttgart, Germany
Instituto de Pós-Graduação Shashikala, A. P.
Goiânia, Goiás, Brazil National Institute of Technology Calicut
Saqan, Elias Calicut, Kerala, India
American University in Dubai Shawky, Mostafa
Dubai, United Arab Emirates Alexandria, Egypt
Sarker, Prabir Shehata, Medhat
Curtin University of Technology Ryerson University
Bentley, Western Australia, Australia Toronto, ON, Canada
Sato, Yuichi Sheikh, Shamim
Kyoto University University of Toronto
Kyoto, Japan Toronto, ON, Canada
Scanlon, Andrew Sherif, Alaa
Pennsylvania State University Helwan University
University Park, PA Cairo, Egypt
Schileo, Giorgio Sherman, Matthew
Sheffield Hallam University Simpson Gumpertz and Heger
Sheffield, UK Melrose, MA
Schindler, Anton Sherwood, Edward
Auburn University Carleton University
Auburn, AL Ottawa, ON, Canada
Semaan, Hassnaa Shi, Xianming
Ottawa Hills, OH Washington State University
Sener, Siddik Pullman, WA
Istanbul Bilgi University Shi, Xudong
Instanbul, Eyup, Turkey Tsinghua University
Sengul, Ozkan Beijing, China
Istanbul Technical University Shi, Yilei
Istanbul, Turkey Rockville, MD
Sengupta, Amlan Shing, Pui-Shum
Indian Institute of Technology Madras University of California, San Diego
Chennai, Tamil Nadu, India La Jolla, CA
Serna-Ros, Pedro Shivali, Ram
Universidad Politecnia de Valencia Central Soil and Materials Research Station
Valencia, Spain New Delhi, India
Setién, Jesús Silfwerbrand, Johan
University of Cantabria KTH Royal Institute of Technology
Santander, Cantabria, Spain Stockholm, Sweden
Shafigh, Payam Singh, Harvinder
Kuala Lumpur, Malaysia Guru Nanak Dev Engineering College
Shafiq, Nasir Ludhiana, Punjab, India
University Technology Petronas Sinn, Robert
Tronoh, Perak, Malaysia Thornton Tomasetti
Shah, Attaullah Chicago, IL
Allama Iqbal Open University Sisman, Can
Islamabad, Pakistan Namık Kemal University
Tekirdag, Turkey

258 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Smadi, Mohammad Tabatabai, Habib
Jordan University of Science and Technology University of Wisconsin – Milwaukee
Irbid, Jordan Milwaukee, WI
Sobhan, Khaled Tadayon, Mohammadhosein
Florida Atlantic University University of Tehran
Boca Raton, FL Tehran, Islamic Republic of Iran
Sobhani, Jafar Tae, Ghi Ho
Building and Housing Research Center Leader Industrial Co.
Tehran, Islamic Republic of Iran Seoul, Republic of Korea
Soliman, Ahmed Tahir, Muhammad
Western University UET Taxila
London, ON, Canada Taxila, Punjab, Pakistan
Soltani, Amir Tahmasebinia, Faham
Purdue University Calumet University of Wollongong
Hammond, IN Wollongong, New South Wales, Australia
Soltani, Masoud Talbot, Caroline
Tarbiat Modares University Cleveland, OH
Tehran, Islamic Republic of Iran Tan, Kefeng
Song, Xin Southwest University of Science and Technology
Zhejiang, China Sichuan, China
Sossou, Gnida Tanacan, Leyla
Kwame Nkrumah University of Science and Technology Istanbul, Yesilkoy, Turkey
Kumasi, Ghana Tang, Chao-Wei
Souza, Rafael Cheng-Shiu University
Universidade Estadual de Maringá Niaosong District, Kaohsiung City, Taiwan, China
Maringá, Paraná, Brazil Tang, Liqun
Söylev, Altug South China University of Technology
Yeditepe University Guangzhou, Guangdong, China
Istanbul, Turkey Tang, Pei
Spinella, Nino Eindhoven, the Netherlands
University of Messina Tangtermsirikul, Somnuk
Messina, Italy Sirindhorn International Institute of Technology
Spyridis, Panagiotis Patumthani, Thailand
Institute for Structural Engineering Tank, Tejenadr
Vienna, Austria Pandit Deendayal Petroleum University
Stein, Boris Gandhinagar, Gujarat, India
Twining Laboratories Tankut, Tugrul
Long Beach, CA Middle East Technical University
Strang, Fred Ankara, Turkey
New Brunswick Department of Transportation Tanner, Jennifer
Fredericton, NB, Canada University of Wyoming
Strauss, Alfred Laramie, WY
University of Natural Resources and Life Sciences Tao, Shi
Vienna, Austria Zhejiang University of Technology
Su, Yu-Min Hangzhou, China
National Kaohsiung University of Applied Sciences Tapan, Mücip
Sanmin, Taiwan, China Yuzuncu Yil University
Sujjavanich, Suvimol Van, Turkey
Kasetsart University Tarighat, Amir
Bangkok, Thailand Tehran, Islamic Republic of Iran
Suksawang, Nakin Tassios, Theodosios
Florida Institute of Technology Athens, Greece
Melbourne, FL Tastani, S. P.
Sullivan, Patrick Demokritus University of Thrace
Sullivan and Associates Xanthi, Greece
Rickmansworth, UK Tavares, Maria
Suraneni, Prannoy UERJ-State University of Rio de Janeiro
ETH Zürich Rio de Janeiro, Brazil
Zürich, Switzerland Tavio
Switonski, Aleksander Sepuluh Nopember Institute of Technology (ITS)
Bydgoszcz, Poland Surabaya, East Java, Indonesia

ACI Structural Journal/March-April 2015 259

 REVIEWERS IN 2014
Tawana, M. M. Triantafillou, Thanasis
Tongji University University of Patras
Shanghai, China Patras, Greece
Tawfic, Yasser Tripura, Deb
Minia University NIT Agartala
Minia, Egypt Agartala, India
Taylor, Peter Trost, Burkhart
National Concrete Pavement Technology Center University of Applied Sciences and Arts Northwestern
Ames, IA Switzerland
Tazarv, Mostafa Muttenz, Switzerland
University of Nevada, Reno Tsonos, Alexander
Reno, NV Aristotle University of Thessaloniki
Tegos, Ioannis Thessaloniki, Greece
Salonica, Greece Tsubaki, Tatsuya
Tehrani, Fariborz Yokohama National University
California State University, Fresno Yokohama, Japan
Fresno, CA Tsuruta, Hiroaki
Tepfers, Ralejs Kansai University
Ralejs Tepfers Consulting Suita, Japan
Gøteborg, Sweden Tuchscherer, Robin
Tharmarajah, Gobithas Northern Arizona University
Belfast, UK Flagstaff, AZ
Thiagarajan, Ganesh Turanli, Lutfullah
University of Missouri - Kansas City Middle East Technical University
Kansas City, MO Ankara, Turkey
Thokchom, Suresh Turk, A. Murat
Manipur Institute of Technology Istanbul Kultur University
Imphal, India Istanbul, Turkey
Thomas, Adam Tutikian, Bernardo
Europoles gmbh Unisinos
Neumarkt, Germany Porto Alegre, Rio Grande Do Sul, Brazil
Thompson, Phillip Uygunoglu, Tayfun
Palm Desert, CA Afyon Kocatepe University
Thorne, A. Afyonkarahisar, Turkey
Center of Engineering Materials and Structures Uzal, Burak
Guilford, UK Abdullah Gul University
Thorstensen, Rein Terje Kayseri, Turkey
University of Agder Vakhshouri, Behnam
Grimstad, Norway University of Technology Sydney
Tian, Ying Sydney, New South Wales, Australia
University of Nevada, Las Vegas Van Deurzen, John
Las Vegas, NV Van Deurzen and Associates PA
Tjhin, Tjen Overland Park, KS
Buckland and Taylor Ltd. Varum, Humberto
North Vancouver, BC, Canada University of Porto
Tolentino, Evandro Porto, Portugal
Centro Federal de Educação Tecnológica de Minas Gerais Vasovic, Dejan
Timóteo, Minas Gerais, Brazil University of Belgrade
Topçu, İlker Belgrade, Serbia
Eskişehir Osmangazi University Vatani Oskouei, Asghar
Eskişehir, Turkey Shahid Rajaee University (BHRC)
Torrenti, Jean-Michel Tehran, Islamic Republic of Iran
Chevilly-Larue, France Vaz Rodrigues, Rui
Tosun, Kamile EPFL
Dokuz Eylul University Lausanne, Switzerland
Izmir, Turkey Vazquez-Herrero, Cristina
Toubia, Elias La Coruña, Spain
University of Dayton Velázquez Rodríguez, Sergio
Dayton, OH Universidad Panamericana
Zapopan, Jalisco, Mexico

260 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Vellalapalayam Nallagounder, Vijayakumar Wehbe, Nadim
Bannari Amman Institute of Technology South Dakota State University
Erode, Tamilnadu, India Brookings, SD
Velu, Saraswathy Wei, Ya
CECRI University of Michigan
Karaikudi, Tamil Nadu, India Ann Arbor, MI
Venkatesh Babu, D. L. Wei-Jian, Yi
Kumaraguru College of Technology Changsha, China
Coimbatore, Tamil Nadu, India Weiss, Jason
Venkiteela, Giri Purdue University
New Jersey Department of Transportation West Lafayette, IN
Trenton, NJ Wen, Qingjie
Vercher, Jose China University of Mining & Technology
Polytechnic University of Valencia Xuzhou, Jiangsu, China
Valencia, Spain Werner, Anne
Vichit-Vadakan, Wilasa Southern Illinois University Edwardsville
CTLGroup Edwardsville, IL
Skokie, IL Wheeler, Andrew
Villar Cociña, Ernesto University of Western Sydney
Central University of Las Villas Sydney, New South Wales, Australia
Santa Clara, Cuba Wilson, William
Vimonsatit, Vanissorn Universite de Sherbrooke
Curtin University Sherbrooke, QC, Canada
Perth, Australia Windisch, Andor
Vintzileou, Elizabeth Karlsfeld, Germany
National Technical University of Athens Wood, Richard
Athens, Greece University of California, San Diego
Viviani, Marco La Jolla, CA
HEIG-VD Wu, Chenglin
Yverdon les Bains, Switzerland Missouri S&T
Vogel, Thomas Rolla, MO
Institute of Structural Engineering Wu, Hui
Zurich, Switzerland Beijing, China
Waldron, Christopher Wu, Hwai-Chung
University of Alabama at Birmingham Wayne State University
Birmingham, AL Detroit, MI
Wan, David Wu, Yu-You
Old Castle Precast Inc. Dania Beach, FL
South Bethlehem, NY Wu, Yu-Fei
Wang, Chang-Qing City University of Hong Kong
Tongji University Kowloon, Hong Kong, China
Shanghai, China Wu, Yu-You
Wang, Chong Dania Beach, Florida
Brisbane, Queensland, Australia Xia, Jin
Wang, Huanzi Zhejiang University
San Jose, CA Hangzhou, Zhejiang, China
Wang, Kejin Xiang, Tianyu
Iowa State University Chengdu, Sichuan, China
Ames, IA Xiao, Yan
Wang, Vincent Hunan University
James Cook University Changsha, Hunan, China
Townsville, Queensland, Australia Xie, Guoshuai
Wang, Xuhao Wuhan University
Ames, IA Wuhan, Hubei, China
Wang, Zhen Yu Xin-hua, Cai
Harbin Institute of Technology Wuhan University
Harbin, Heilongjiang, China Wuhan, Hubei, China
Watkins, Melanie Xu, Aimin
Michigan Technological University ARRB Group
Houghton, MI Melbourne, Victoria, Australia

ACI Structural Journal/March-April 2015 261

 REVIEWERS IN 2014
Xuan, D.X. Yoon, In-Seok
The Hongkong Polytechnic University Induk University
Kowloon, Hong Kong, China Seoul, Republic of Korea
Yahia, Ammar Yoon, Young-Soo
Université de Sherbrooke Korea University
Sherbrooke, QC, Canada Seoul, Republic of Korea
Yamada, Kanji Youkhanna, Kanaan
Akita Prefectural University University of Dohuk
Yurihonjo, Japan Duhok, Iraq
Yanez, Fernando Yu, Baolin
IDIEM University of Chile Michigan State University
Santiago, Chile East Lansing, MI
Yang, Kai Yu, Haiyong
Belfast, UK Shanghai Research Institute of Building Sciences
Yang, Keun-Hyeok Shanghai, China
Kyonggi University Yu, Jiangtao
Suwon, Kyonggi-do, Republic of Korea Research Institute of Civil Engineering and Disaster Reduction
Yang, Kuochen Shanghai, China
National Kaohsiung First University of Science and Technology Yu, Tzu-Yang
Kaohsiung, Taiwan, China University of Massachusetts Lowell
Yang, Xinbao Lowell, MA
Olathe, KS Yuan, Jiqiu
Yassein, Mohamed PSI, Turner-Fairbank Highway Research Center, FHWA
Doha, Qatar McLean, VA
Yatagan, Serkan Yuan, Xiaohui
Istanbul Technic University Architecture Faculty Wuhan University of Technology
Istanbul, Turkey Wuhan, China
Yazbeck, Fouad Yüksel, Isa
Readymix Abu Dhabi Bursa Technical University
Abu Dhabi, United Arab Emirates Bursa, Turkey
Yazıcı, Şemsi Yun, Hyun Do
Ege University Daejeon, Republic of Korea
İzmir, Turkey Zaidi, S. Kaleem
Yekrangnia, Mohammad Aligarh Muslim University
Sharif University of Technology Aligarh, UP, India
Tehran, Islamic Republic of Iran Zaki, Adel
Yerramala, Amarnath SNC - Lavalin
Dundee University Montreal, QC, Canada
Dundee, UK Zandi Hanjari, Kamyab
Yeung, Jaime Chalmers University of Technology
Yue Xiu Concrete Co Ltd Gothenburg, Sweden
Hong Kong, China Zanuy, Carlos
Yildirim, Hakki Universidad Politécnica de Madrid
Istanbul, Turkey Madrid, Spain
Yilmaz, Bulent Zdiri, Mustapha
Bilecik Seyh Edebali University National Engineering School of Tunis
Bilecik, Turkey Tunis, Rades Tunis, Tunisia
Yılmaz, Ali Zeris, Christos
KTÜ National Technical University of Athens
Trabzon, Turkey Zografou, Greece
Yindeesuk, Sukit Zhang, Jieying
University of Illinois at Urbana-Champaign National Research Council Canada
Urbana, IL Ottawa, ON, Canada
Yoğurtcu, Erhan Zhang, Jun
Concrete Technology Tsinghua University
İzmir, Turkey Beijing, China
Yoon, Hyeong Jae Zhang, Peng
Taisei Corporation Karlsruhe Institute of Technology (KIT)
Tokyo, Japan Karlsruhe, Germany

262 ACI Structural Journal/March-April 2015

  REVIEWERS IN 2014
Zhang, Y. X. Zheng, Yu
The University of New South Wales Dongguan University of Technology
Canberra, Australian Capital Territory, Australia Dongguan, Guangdong, China
Zhang, Yamei Zhou, Changdong
Southeast University Beijing Jiaotong University
Nanjing, China Beijing, China
Zhang, Xiaogang Zhou, Wei
Shenzhen University Harbin Institute of Technology
Shenzhen, Guangdong, China Harbin, China
Zhang, Xiaoxin Zhou, Xiangming
Universidad de Castilla-La Mancha Brunel University
Ciudad Real, Spain Uxbridge, UK
Zhao, Jian Ziehl, Paul
University of Wisconsin – Milwaukee University of South Carolina
Milwaukee, WI Columbia, SC
Zheng, Herbert Zilch, Konrad
Gammon Construction Limited Technische Universitat Munchen
Hong Kong, China Munich, Germany
Zheng, Jianjun
Zhejiang University of Technology
Hangzhou, China

ACI Structural Journal/March-April 2015 263

 NOTES:

264 ACI Structural Journal/March-April 2015

ACI
STRUCTURAL
J O U R N A L
J O U R N

The American Concrete Institute (ACI) is a leading authority and

resource worldwide for the development and distribution of
consensus-based standards and technical resources, educational
programs, and certifications for individuals and organizations involved
in concrete design, construction, and materials, who share
a commitment to pursuing the best use of concrete.

Individuals interested in the activities of ACI are encouraged to

explore the ACI website for membership opportunities, committee
activities, and a wide variety of concrete resources. As a volunteer
member-driven organization, ACI invites partnerships and welcomes
all concrete professionals who wish to be part of a respected,
connected, social group that provides an opportunity for professional
growth, networking and enjoyment.