You are on page 1of 146

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

Vice Presidents
Sharon L. Wood
Michael J. Schneider
Directors
Roger J. Becker
Dean A. Browning
Jeffrey W. Coleman
Alejandro Durn-Herrera
Robert J. Frosch
Augusto H. Holmberg
Cary S. Kopczynski
Steven H. Kosmatka
Kevin A. MacDonald
Fred Meyer
Michael M. Sprinkel
David M. Suchorski

a journal of the american concrete institute


an international technical society

123 Evaluation of Column Load for Generally Uniform Grid-Reinforced


Pile Cap Failing in Punching, by Honglei Guo
135 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
147 
Inverted-T Beams: Experiments and Strut-and-Tie Modeling, by
N.L.Varney, E. Fernndez-Gmez, D. B. Garber, W. M. Ghannoum, and
O. Bayrak

Past President Board Members


Anne M. Ellis
James K. Wight
Kenneth C. Hover

157 
Energy-Based Hysteresis Model for Reinforced Concrete BeamColumn Connections, by Tae-Sung Eom, Hyeon-Jong Hwang, and
Hong-Gun Park

Executive Vice President


Ron Burg

167 Ductility Enhancement in Beam-Column Connections Using Hybrid


Fiber-Reinforced Concrete, by Dhaval Kheni, Richard H. Scott,
S.K.Deb, and Anjan Dutta

Technical Activities Committee


Ronald Janowiak, Chair
Daniel W. Falconer, Staff Liaison
JoAnn P. Browning
Catherine E. French
Fred R. Goodwin
Trey Hamilton
Neven Krstulovic-Opara
Kimberly Kurtis
Kevin A. MacDonald
Jan Olek
Michael Stenko
Pericles C. Stivaros
Andrew W. Taylor
Eldon G. Tipping

Staff

Executive Vice President


Ron Burg

179 
Behavior and Simplified Modeling of Mechanical Reinforcing Bar
Splices, by Zachary B. Haber, M. Saiid Saiidi, and David H. Sanders
189 
Bond-Splitting Strength of Reinforced Strain-Hardening Cement
Composite Elements with Small Bar Spacing, by Toshiyuki Kanakubo
and Hiroshi Hosoya
199 Wide Beam Shear Behavior with Diverse Types of Reinforcement,
by S. E. Mohammadyan-Yasouj, A. K. Marsono, R. Abdullah, and
M.Moghadasi
209 Effect of Axial Compression on Shear Behavior of High-Strength
Reinforced Concrete Columns, by Yu-Chen Ou and Dimas P. Kurniawan

Engineering
Managing Director
Daniel W. Falconer

221 Experimental Investigations on Prestressed Concrete Beams with


Openings, by Martin Classen and Tobias Dressen

Managing Editor
Khaled Nahlawi

233 Discussion

Staff Engineers
Matthew R. Senecal
Gregory M. Zeisler
Jerzy Z. Zemajtis

Bond-Slip-Strain Relationship in Transfer Zone of Pretensioned Concrete




Publishing Services
Manager
Barry M. Bergin
Editors
Carl R. Bischof
Tiesha Elam
Kaitlyn Hinman
Kelli R. Slayden
Editorial Assistant
Angela R. Matthews

Elements. Paper by Ho Park and Jae-Yeol Cho

Contents cont. on next page


Discussion is welcomed for all materials published in this issue and will appear ten months from
this journals date if the discussion is received within four months of the papers print publication.
Discussion of material received after specified dates will be considered individually for publication or
private response. ACI Standards published in ACI Journals for public comment have discussion due
dates printed with the Standard.
Annual index published online at http://concrete.org/Publications/ACIStructuralJournal.
ACI Structural Journal
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. Publication office: 38800 Country Club Drive, Farmington Hills, MI 48331. Periodicals postage paid at Farmington, MI, and
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. Facsimile (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
ACI Structural Journal


Analysis and Prediction of Transfer Length in Pretensioned, Prestressed

The ACI Structural Journal is an open


forum on concrete technology and papers
related to this field are always welcome.
All material submitted for possible publication must meet the requirements of
the American Concrete Institute Publication Policy and Author Guidelines
and Submission Procedures. Prospective
authors should request a copy of the Policy
and Guidelines from ACI or visit ACIs
website at www.concrete.org prior to
submitting contributions.
Papers reporting research must include
a statement indicating the significance of
the research.
The Institute reserves the right to return,
without review, contributions not meeting
the requirements of the Publication Policy.
All materials conforming to the Policy
requirements will be reviewed for editorial
quality and technical content, and every
effort will be made to put all acceptable
papers into the information channel.
However, potentially good papers may be
returned to authors when it is not possible
to publish them in a reasonable time.

ening Systems. Paper by Nabil Grace and Mena Bebawy

Concrete Members. Paper by Byung Hwan Oh, Si N. Lim, Myung K. Lee, and
SungW. Yoo


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

241

Reviewers in 2014

MEETINGS
MARCH/APRIL
30-2Concrete Sawing & Drilling
Association Convention and Tech Fair,
St. Petersburg, FL, www.csda.org/events/
event_details.asp?id=444478&group

APRIL
13-15BEST Conference Building
Enclosure Science & Technology,
Kansas City, MO, www.nibs.org/?page=best
26-292015 Post-Tensioning Institute
Convention, Houston, TX, www.posttensioning.org/page/pti-convention
26-3057th Annual IEEE-IAS/
PCA Cement Industry Technical
Conference, Toronto, ON, Canada, www.
cementconference.org

11-132015 International Concrete


Sustainability Conference, Miami, FL,
www.concretesustainabilityconference.org
14-16The American Institute of
Architects Convention, Atlanta, GA,
http://convention.aia.org/event/homepage.aspx
14-17The Masonry Society 2015
Spring Meetings, Denver, CO, http://
www.masonrysociety.org/index.
cfm?showincenter=http%3A//www.
masonrysociety.org/html/calendar/index.htm
17-2012th North American Masonry
Conference, Denver, CO, www.
masonrysociety.org/NAMC/index.html
24-26 - Fifth International Symposium on
Nanotechnology in Construction, Chicago,
IL, www.nicom5.org

MAY

JUNE

3-7International Cement Microscopy


Association Annual Conference, Seattle,
WA, www.cemmicro.org

1-37th RILEM Conference on


High Performance Fiber Reinforced
Cement Composites, Stuttgart,
Germany, www.rilem.org/gene/main.
php?base=600040#next_614

4-72015 World of Coal Ash Conference,


Nashville, TN, www.worldofcoalash.org

THE ACI CONCRETE CONVENTION AND EXPOSITION: FUTURE DATES


2015April 12-16, Marriott & Kansas City Convention Center, Kansas City, MO
2015November 8-12, Sheraton Denver, Denver, CO
2016April 17-21, Hyatt & Wisconsin Center, Milwaukee, WI
For additional information, contact:
Event Services, ACI
38800 Country Club Drive, Farmington Hills, MI 48331
Telephone: +1.248.848.3795
e-mail: conventions@concrete.org

ON COVER: 112-S12, p. 136, Fig. 2Reinforced concrete (left) and prestressed concrete (right) specimens
on the E-Defense shake table.

Permission is granted by the American Concrete Institute for libraries and other users registered with the Copyright
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
$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.
The Institute is not responsible for statements or opinions expressed in its publications. Institute publications are not able
to, nor intend to, supplant individual training, responsibility, or judgment of the user, or the supplier, of the information
presented.
Papers appearing in the ACI Structural Journal are reviewed according to the Institutes Publication Policy by individuals
expert in the subject area of the papers.

122

Discussion
All technical material appearing in the
ACI Structural Journal may be discussed.
If the deadline indicated on the contents
page is observed, discussion can appear
in the designated issue. Discussion should
be complete and ready for publication,
including finished, reproducible illustrations. Discussion must be confined to the
scope of the paper and meet the ACI Publication Policy.
Follow the style of the current issue.
Be brief1800 words of double spaced,
typewritten copy, including illustrations
and tables, is maximum. Count illustrations
and tables as 300 words each and submit
them on individual sheets. As an approximation, 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,
and include two good quality copies of the
entire discussion. References should be
complete. Do not repeat references cited
in original paper; cite them by original
number. Closures responding to a single
discussion should not exceed 1800-word
equivalents in length, and to multiple
discussions, approximately one half of
the combined lengths of all discussions.
Closures are published together with
the discussions.
Discuss the paper, not some new or
outside work on the same subject. Use
references wherever possible instead of
repeating available information.
Discussion offered for publication should
offer some benefit to the general reader.
Discussion which does not meet this
requirement will be returned or referred to
the author for private reply.
Send manuscripts to:
http://mc.manuscriptcentral.com/aci
Send discussions to:
Journals.Manuscripts@concrete.org

ACI Structural Journal/March-April 2015

ACI STRUCTURAL JOURNAL

TECHNICAL PAPER

Title No. 112-S11

Evaluation of Column Load for Generally Uniform GridReinforced 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 configurationhinder 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 mechanism 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, respectively; 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 superstructure 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 1The collapse mechanism is assumed so that
the upper-bound solution to punching shear resistance is
obtained using the theory of plasticity, called the upperbound 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

of the oblique sections of the punching cone, and the perimeter 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 coefficient containing the punching-span ratio to correct the
punching shear resistance (the definition of being given
in Eq. (1)).
Type 2The 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 lowerbound 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 structures whose bearing mechanism is well known.
In technical codes, the text and Appendix A of
ACI318-08,1 the CRSI handbook,3 CAN/CSA A23.3-04,4
BSEN 1992-1-1:2004,5 and AS 3600-20016 either adopt or
contain thismethod.
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 wasintroduced.
As a supplement to the ACI 318-08 code, the CRSI handbook3 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 contribution 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

123

Fig. 1Load-transfer mechanism of SSTM for punching


failure of pile caps.

Fig. 3Failure 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.)

Fig. 2Damage mechanism of SSTM for punching failure


of pile caps.
the two-way shear of the pile caps to the superposition of
the one-way shear of two mutually orthogonal deep beams
whose width is equal to the length of the columnedges.
In the theoretical study, Wen7 modeled the punching of pile
caps as the coupling between two orthogonal deep beams,
while Kinnunen and Nylander8 regarded it as a spatial shell.
However, recent studies and practice have proved that it
is more rational to liken the bearing mechanism of punching
failure of pile caps to a spatial STM (SSTM).9-12
Herein, as the basis for the derivation of the column load
of pile cap failing in punching to be conducted later, a brief
introduction to the authors research conclusions is given
asfollows.11,12
Load-transfer mechanism of punching failureAs shown
in Fig. 1, the load-transfer system of the punching failure of
pile caps is analogous to the SSTM, where the compression
struts are used to model the zones of concrete with primarily
unidirectional compressive stresses, while the reinforcements within the range of primarily unidirectional tensile
stresses are approximated by tension ties.11,12
The pile load distribution during the punching failure of
pile caps can approximately adopt the value of the pile caps
in the elastic stage.11
Damage mechanism of punching failureAs shown in
Fig. 2 and 3, the strut is represented as three zones: namely,
Zone III, the shear-compression zone intersecting the column
124

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
column load increased, the oblique crack develops toward
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
III), the column load being jointly borne by Struts A and B.
Part of the column load is transferred to the longitudinal
reinforcement and the uncracked concrete of Zone I by
StrutA, and the other part is transferred to the pile by StrutB.
When punching failure occurs, Strut A is punched out relative to Strut B to have the punching cone formed. It can be
considered that the column load at this moment is jointly
borne by Zones I and III, together with the dowel action of
the bottom longitudinal reinforcement. The two parts are
correlated, and the loss of the punching shear resistance is a
result of the damages in the aforementioned parts occurring
one after another so that, with no additional external load, the
oblique section suddenly collapses. Therefore, the punching
failure of pile caps is either the strut failure, which begins
with the splitting in the midpart of the strut (Zone II) and
ends with shear-compression failure at the two ends of the
strut (Zones I and III) or the yield failure of the tension tie
resulting from insufficient tension tie reinforcement amount.
But the tension tie failure is also accompanied by the strut
failure, so the strut failure is an indication of the loss of the
pile cap punching shear resistance.12
The two basic factors influencing the strut strength are the
punching-span ratio and concrete strength.12 The strengths at
the two ends of the strut are not appreciably different; their
average can be taken as the strut strength.12

ACI Structural Journal/March-April 2015

Fig. 4Effective depth and location of SSTM upper node.


Dimensions of SSTMConstruct 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 value0.6 times that of the pile
(except for the strut between the column and the pile beneath
the column)whatever the distance of pile to column.12
2. As shown in Fig. 4, the upper node of the SSTM is
located at 0.1 times the effective depth vertically downwards
from the column center on the top surface of the pile caps.12
3. As shown in Fig. 5, for simplicity, take a two-pile cap
as an example to illustrate the location of the SSTM lower
node, which is obtained in accordance with the three steps:
1) link upper node A to pile center B to obtain line segment
AB; 2) project AB onto the plane where the longitudinal
reinforcement centroid is located to obtain line segment CD,
while obtaining the projection line L of the pile periphery
onto the same plane; 3) intercept CD with L to obtain line
segment ED, and the midpoint of ED is just lower node F of
the SSTM.12
4. As shown in Fig. 6, the effective range of the tension tie
is twice the pile diameter that is concentric with the lower
node of the SSTM.11
It should be noted that compared with the currently available extensive literature on the bearing mechanism of the
SSTM, investigation on the evaluation of the punching shear
resistance of pile caps with uniform grid reinforcement is as
of yet inadequate. So, based on the previously mentioned
research conclusions about the punching bearing mechanism, further studies will be made along these lines.
RESEARCH SIGNIFICANCE
Many punching shear resistances of pile caps are evaluated by design aids with the rule-of-thumb procedures,
which have at least two drawbacks: 1) the theoretical calculation values either far exceed the experimental ones or,
ACI Structural Journal/March-April 2015

Fig. 5Location of SSTM lower node.

Fig. 6Effective range of tension tie, punching-span,


andAs.
although no larger than the experimental ones, are significantly variable. Hence, the hidden safety risks; and 2) the
theory of STM applied to solve the punching of pile caps
is significantly conservative. This paper focuses on the
derivation of the three-dimensional (3-D) rather than twodimensional (2-D) strut strength, from which the calculating
method for the punching shear resistance of the pile caps
that is, the column load of pile cap failing in punchingis
developed. Careful verification, comparison, and analysis
show that the results obtained in this paper should contribute
to improving the aforementioned situation, and the information presented in this paper should prove useful to organizations that publish design aids for pile caps.
OVERALL CONSIDERATIONS FOR DERIVATION
OF EVALUATION
First, two variables are defined as follows:
1. Punching-span ratio

= w/d (1)

where, as shown in Fig. 4, the effective depth d is the depth


to the centroid of the bottom longitudinal reinforcement. As
shown in Fig. 6, the punching span w is the distance GB1,
where line segment AB1 is obtained by linking column
center A to pile center B1, and point G is obtained through
the interception of AB1 by the periphery of column. If not a
round column, convert its cross section to a circular one of
equal perimeter.
125

Table 1Relationship between g and l, fc' for l 0.95

6.7 MPa (971.5 psi) fc


35 MPa (5075 psi)

35 MPa (5075 psi) fc


50 MPa (7252 psi)

0.95

= 2.89255 0.31042fc*

= 1.059

1.0

= 2.89255 0.31042fc

= 1.056

1.2

= 2.8618 0.30712fc

= 1.04775

1.4

= 2.8782 0.30888fc

= 1.05225

1.6

= 2.8659 0.30756fc

= 1.04775

1.8

= 2.8618 0.30712fc

= 1.04925

2.0

= 2.88025 0.3091fc

= 1.0545

Unit of fc is MPa; is nondimensional. Similarly hereinafter.

Table 2Relationship between g and l, fc' for


0.15 l 0.95
6.7 MPa (971.5 psi) fc
35MPa (5075 psi)

35 MPa (5075 psi) fc


50MPa (7252 psi)

0.75
0.95

= 2.05 (2.341 0.9751)


0.22 (2.398 1.057)
fc

= 0.75 (2.32 0.96)

0.35
0.75

= 2.05 (1.972 0.521)


0.22 (1.9738 0.488)fc

= 0.75 (1.9789 0.493)

0.15
0.35

= 2.05 (2.125 0.973)


0.22 (2.14258 0.945)
fc

= 0.75 (2.18292 1.085)

2. Reinforcement ratio of tension tie


As
(2)
2Dp d

where, as shown in Fig. 6, As is the sum of the crosssectional areas of the longitudinal reinforcements within the
effective range of the tension tie; and Dp is the pile diameter.
As pointed out earlier, the strut failure is an indication
of the loss of the pile cap punching shear resistance. So
the evaluation of punching shear resistance of pile caps is
exactly an evaluation of the strut bearing load, while strut
bearing load F is the cross-sectional area at the strut end S
strut strength fce. It is known from the earlier statement that S
is 0.6 times the cross-sectional area of the pile, and fce is the
average of the strengths at the two ends of the strut. So F can
be expressed as follows

F = S f ce = 0.6R 2

f ce1 + f ce 2
(3)
2

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
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.
DERIVATION FOR fce
Define = fce/fc, where fc is the cylinder compressive
strength of the strut concrete.
126

It is known from the foregoing conclusion that the two


basic factors influencing the strut strength are the punchingspan ratio and concrete strength; thus, = (,fc). To find
the specific expression for , the ADINA nonlinear finite
element (NFE) program, which has successfully evaluated
the punching shear resistance of pier deck13 (pier deck is
similar to pile cap), is adopted. Analysis and derivation of
the expression for are made by referring to mathematical
deduction of the gradual least-square method for multiple
variables (GLSMV).
In selecting the model for computerization, as the purpose
of computerization is simply to derive the strut strength,
there is no need for consideration of the pile number or pile
arrangement other than the choice of the strut. Therefore, a
quarter of the four-pile cap of symmetrical and determinant
pile arrangement is selected, as shown in Fig. 3.
In developing the numerical model, the concrete of pile
caps is divided into four layers, the greater part of which
are 3-D isoparametric elements with eight nodes and three
degrees-of-freedom per node, a few triangular prism-shaped
degenerate elements being taken as transition ones. Where
the pile cap is near the column and pile, the 3 3 3 integration order is adopted, while the 2 2 2 integration
order is used elsewhere. The column and pile are linked to
the pile cap also in the form of 3-D isoparametric elements.
The concrete material model adopted is a nonlinear one
with compression crushing, tensile cutoff with strain softening, and shear stress transferring across the cracks taken
into account.13 The reinforcing bars are represented by truss
elements with two nodes, the constitutive law for which is an
elastic-plastic material model.
As shown in Fig. 3, for the strut between the column
bottom and the pile top in question, take and fc as 0.15
to 2.0 and 6.7 to 50 MPa (971.7 to 7252 psi), respectively.
Thus, a total of 102 cases of combination is investigated.
In the process of analysis, as of the pile caps with uniform
grid reinforcement is in general rather small, no larger than
1.2% at most and has little influence on the strut strength,14,15
it can be maintained at 0.6% throughout.
The relationships between and , fc for 0.95 are
shown in Table 1 as an example.
The expressions in Table 1 are summed up as follows:
1. = a bfc, for 6.7 MPa (971.5 psi) fc 35 MPa
(5075 psi)
2. = c, for 35 MPa (5075 psi) fc 50 MPa (7252 psi)
Obviously, a, b, c are the functions of . Use the leastsquare method once again to obtain

a = 2.90045 0.0170961 2.05 1.41485

b = 0.30668 0.00183469 0.22 1.394

c = 1.05888 0.00459047 0.75 1.41184

In conclusion, for 0.95:

ACI Structural Journal/March-April 2015

1) = 2.05 1.41485 0.22 1.394fc, for 6.7 MPa


(971.5 psi) fc 35 MPa (5075 psi); and

2) = 0.75 1.41184, for 35 MPa (5075 psi) fc 50


MPa (7252 psi).

Similarly, the relationships between and , fc for other


ranges of are obtained in Table 2. Observing the situation
of in each of its ranges shown in Tables 1 and 2 to know:
1. Whatever the range is in, for 6.7 MPa (971.5 psi) fc
35 MPa (5075 psi)

= 2.05f1() 0.22f2()fc

(4)

2. Whatever the range is in, for 35 MPa (5075 psi) fc


50 MPa (7252 psi)

= 0.75f3()

(5)

3. In the same range of , the expressions for f1(), f2(),


and f3() are almost identical, so a unified expression can
be taken

( ) =

f1 ( ) + f 2 ( ) + f 3 ( )
3

Take () out of Eq. (4) and (5), then


= (fc) ()

where, for (fc):


1) (fc) = 2.05 0.22fc, for 6.7 MPa (971.5 psi) fc
35 MPa (5075 psi)
(6a)
2) (fc) = 0.75, for 35 MPa (5075 psi) fc 50 MPa
(7252 psi)
whereas, for ():

1) () = 1.4, for 0.95


2) () = 2.35 , for 0.75 0.95
(6b)

3) () = 1.975 0.5, for 0.35 0.75

4) () = 2.15 , for 0.15 0.35

Observation of Eq. (6b) shows, for 0.15 0.95, that


the slopes of all the fold line segments making up () are
almost identical. So the straight line linked by point = 0.15
and point = 0.95 can be used to represent () in this range
(0.15 0.95) in a unified manner; that is, ultimately

1) () = 2.1125 0.75, for 0.15 0.95
(6c)

2) () = 1.4, for 0.95
Thus, the ultimate expression of the strut strength fce is
ACI Structural Journal/March-April 2015

f ce = f c = ( f c)( ) f c (7)

which, substituted back into Eq. (3), gives the ultimate


bearing load expression of the strut

F = 0.6R 2 f ce = 1.885 R 2 ( f c) ( ) f c (8)

where (fc) and () are found in Eq. (6a) and (6c), respectively. As shown in Eq. (7), fce is a constantly increasing
function of fc, whatever the range fc is in; for 0.15
0.95, fce is a decreasing function of , while for 0.95, fce
is a constant function of .
RESULTS AND DISCUSSION
Table 3 lists the published test data of 98 specimens on
the punching failure of the pile caps with uniform grid reinforcement in literature, whose pile number, pile arrangement, punching-span ratio, concrete strength, and reinforcement 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
ACI 318-08 code1 (ACI STM); 4) the American CRSI handbook 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 reinforcement,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 3Summary of pile cap test results


Specimen

Column size, mm
(diameter or side length)

d, mm

fc, MPa

Reinforcement layout,
No. of bar bar diameter, mm

Bar yield stress fy, MPa

Test column load


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

111.62

31.3

3 5.715 + 3 2.68 each way

:499.4; :410.1

225.7

108.59

41.0

7 5.715 + 4 2.032 each way

:499.4; :480.2

263.5

SS04
SS05

76.2 round

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

117.50

31.5

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

MS16

76.2 round

288.9

Taylor and Clarke (No. of pile: 4; pile arrangement: determinant; pile diameter, 200 mm)
16

A001
A009

200 square

400

20.9

10 10 each way

410

1110

26.8

10 10 each way

410

1450

479

1781

Adebar et al. (No. of pile: 4; pile arrangement: diamond; pile diameter, 200 mm)
9

300 square

445

24.8

9 11.3 one way; 15 11.3 other way

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

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

T423

60 square

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


d, mm

fc, MPa

Reinforcement layout,
No. of bar bar diameter, mm

Bar yield stress fy, MPa

Test column load


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

96

12.5

18 2.8 each way

276.9

78.6

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

276.5

364.6

Specimen

T433
T435

Column size, mm
(diameter or side length)

60 square

Shen (No. of pile: 4; pile arrangement: determinant; pile diameter, 100 mm)
17

T452

150 square

225

9.4

11 8 each way

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

T601
T602

150 square

225

13.5

12 10+20 6 each way

:272.0;:276.3

460.6

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
PC454*

150 square

215

12.2

7 12 each way

280.2

370

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)
S1

220 square

259

15.4

20 12 one way;
18 12 other way

318.6

1250

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

C2-2

150 square

520

7.54

5 8 each way

289.3

559

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


Specimen

Column size, mm
(diameter or side length)

d, mm

fc, MPa

Reinforcement layout,
No. of bar bar diameter, mm

Bar yield stress fy, MPa

Test column load


at failure, kN

Ma22 (No. of pile: 3; pile arrangement: isosceles triangle; pile diameter, 90 mm)
P5
P6

100(one side)
140(other side) rectangle

180

20.1

4 6 each way

340

222.5
226.4

Suzuki et al. (No. of pile: 4; pile arrangement: determinant; pile diameter, 150 mm)
23

TDS3-1
TDM3-1

250 square

300
250

28.0

11 9.53 each way

356

1299

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-3020-70-2

200 square

BDA-4025-70-1

250 square

250

24.6

6 9.53 each way

549
358

350

25.9

8 9.53 each way

1019

Suzuki and Otsuki25 (No. of pile: 4; pile arrangement: determinant; pile diameter, 150 mm)
BPB-3520-1

200 square

290

20.4

9 9.53 each way

353

755

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)
F(Saeed)

150 round

230

20.68

10 12.8 + 6 6.5 each way

150 round

230

27.6

12 12.8 + 6 6.5 each way

, :413

480
560

Blvot and Frmy28 (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

4N3

500 square

920

34.15

4 32 + 4 25 + 8 12 each way

:250.6; :281.2;
:293.1

6500

4N3b

500 square

920

49.3

4 25 + 4 20 + 8 10 each way

:484.5; :446;
:429.5

9000

9A3

150 square

470

34.4

16 12

450.25

1700

Blvot and Frmy (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 theres


no law about the crack distribution of PC454 in the limit
state, its 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.
130

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,
ACI Structural Journal/March-April 2015

Table 4Pe/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

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


Predicting method
TPM

ACI
CSM
ACI
STM

CRSI

JGJ
94-94

Total number of specimens Average Standard deviation Coefficient of variation

Maximum

All calculable specimens

98

1.0179

0.1940

0.1906

Asterisked specimens in
Table 3 removed

96

1.0179

0.1832

0.1800

0.6175

1.4631

All calculable specimens

86

1.1177

0.9079

0.8123

0.1063

3.1890

Asterisked specimens in
Table 3 removed

84

1.1318

0.9126

0.8063

0.1063

3.1890

All calculable specimens

81

2.3789

0.6796

0.2857

1.1479

5.9605

Asterisked specimens in
Table 3 removed

79

2.3929

0.6813

0.2847

1.1479

5.9605

All calculable specimens

51

0.7228

0.3126

0.4325

0.2124

1.4426

Asterisked specimens in
Table 3 removed

50

0.7277

0.3138

0.4313

0.2124

1.4426

All calculable specimens

88

1.2526

0.5066

0.4044

0.4911

2.9343

Asterisked specimens in
Table 3 removed

86

1.2585

0.5087

0.4042

0.4911

2.9343

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


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 specimens 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.
132

Minimum

0.5349 (T441*) 1.501 (PC454*)

Applicability
TPM is capable of evaluating all the specimens in Table3,
so its calculating mode is comparatively unified and not
restricted by the number of piles and the form of pile arrangement. 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. Likewise, 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 applicability of TPM is recommendable.
Further analysis is as follows:
1. As mentioned earlier, it has been anticipated during
computerization that the of pile caps with uniform grid
reinforcement has little influence on its punching shear resistance, 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
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 outnamely, the
punching shear resistance of pile caps with uniform grid
reinforcement is not sensitive to its longitudinal reinforcement ratio, as demonstrated in References 14 and 15. The
aforementioned discussion shows that if longitudinal reinforcement is arranged according to a uniform grid, it is not
necessary to impose restrictions on the reinforcement ratio
of tension tie for TPM.
2. As previously mentioned, during derivation of fce, under
the prerequisite for punching failure, the punching-span
ratio is given a large range of variationnamely, 0.15 to
2.0and concrete strength fc basically contains the whole
range of ordinary concrete strength as wellnamely, 6.7to
50 MPa (971.5 to 7252 psi). Likewise, a large range of variation of 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 practice, 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 , , 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, Canadian Standards Association, Toronto, ON, Canada, 2010, pp. 63-65.
5. BSEN 1992-1-1:2004, Eurocode 2: Design of Concrete Structures-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 Technology, 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. Brea, 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), Industrial 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), Industrial 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 StructuresThe 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 FourPile Deep Pile Cap, Proceedings of High-Rise Buildings and Bridge Foundation 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. Blvot, J. L., and Frmy, R., Semelles sur Pieux, Institute Technique
du Btiment 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 Engineering 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 fr 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

TECHNICAL PAPER

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 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 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 provisions of ACI 318-11 are discussed.
Keywords: collapse; damage; design; full-scale; moment frame; multistory; 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 requirements 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 reinforced 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 difficult to continue using them after the earthquake and resulted
in high repair costs. This experience demonstrates that
further improvements in seismic design of concrete buildings 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, reinforced 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

capacity, and failure mode; 2) to identify suitable computational 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 provisions 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 photograph 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

135

Fig. 1Framing and reinforcing details. (Note: Dimensions are in mm; 1 mm = 0.039 in.)

Fig. 2Reinforced 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 deformations. 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 equipment. Figure 1 shows dimensions and reinforcement details
of typical members. The test building was designed in accordance 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 longitudinal reinforcement but differed in the spacing of transverse
reinforcement (Fig. 1). A complete set of drawings and specimen 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 1Weight and design forces


(A) Structural elements, kN

RC

Roof

Table 2Material properties of concrete

Fourth
floor

Third
floor

Second
floor

B,
N/mm2

Ec,
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

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

Roof

Fourth
floor

Third
floor

Second
floor

Stair and handrail

Measurement frame

17

17

Equipment

112

Sum

118

14

23

23

Total of (A) and (B), kN

934

867

872

867

Fourth
story

Third
story

Second
story

First
story

Wi, kN

934

1801

2673

3541

Ci = 0.2 Ai

0.29

0.25

0.22

0.20

Qi, kN

273

450

593

708

(B) Non-structural
elements, kN
Steel

Fc,
N/mm2

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 Wi; and Qi is shear at story i; 1
kN = 0.225 kip.

in this section. The building specimen was designed to withstand the seismic lateral forces presented in Table 1 (MLIT
2007) without members exceeding their elastic limits. These
forces, which sum to 20% of the weight of the structure, are
higher than those that would be specified by ASCE 7-10
(Section 12.8.1.3), which caps seismic lateral forces for
a low-rise building to 1/R times the structure weight for a
design basis earthquake, where R is the response modification coefficient (8 for special reinforced concrete moment
frames and 6 for special reinforced concrete shear walls).
The vertical distribution of the design forces, given by the
parameter Ai in Table 1, is similar to the ASCE 7-10 specification (approximate inverted triangular distribution).
Results of material tests are given in Tables 2 and 3. In
subsequent evaluations, the moment and shear strengths
of each member were calculated adopting the compressive
strength of concrete and the yield strength of steel reinforcement obtained by averaging material test results.
To aid in the design of the test specimen, pushover
(nonlinear static) analyses were conducted on line-element
models of the structure. Figure 3 presents pushover results
for the final test specimen details. The analytical model used
for pushover analyses was built following work by Kabeyasawa et al. (1984). The effective flange width of a top slab
was adopted in accordance with the recommendations of
the 2007 MLIT Standard. A vertical distribution defined by
the parameter Ai (Table 1) was adopted for the lateral force
distribution. In the analytical model, inelastic deformations
of beam elements were represented by rotational springs
at the ends of elements. The first and second break points
corresponding to member cracking strength and flexural
strength were assigned in the tri-linear moment-rotation
ACI Structural Journal/March-April 2015

Notes: Fc is specified concrete compressive strength; B is measured concrete compressive strength; and Ec is measured secant modulus of concrete; 1 N/mm2 = 0.145 ksi.

Table 3Material properties of steel


D22

Grade

Anominal, mm2

y, N/mm2

t, N/mm2

Es, kN/mm2

SD345

387

370

555

209

D19

SD345

287

380

563

195

D13

SD295

127

372

522

199

D10

SD295

71

388

513

191

D10

SD295

71

448

545

188

D10

KSS785

71

952

1055

203

Notes: Anominal is nominal area of reinforcing bars; y is measured yield strength of


steel reinforcement; t is measured ultimate strength of steel reinforcement; and Es
is measured elastic modulus of steel reinforcement; 1 mm2 = 0.0016 in.2; 1 N/mm2=
0.145 ksi.

relationship. The secant stiffness corresponding to the flexural strength was calculated in accordance with provisions
of the MLIT standard (2007). Beyond flexural yielding, the
stiffness was reduced to 0.01 times the initial effective stiffness. The pushover analysis indicates that the ultimate baseshear strength of the building specimen is approximately
0.42W (1500 kN [337 kip]) in the frame direction and 0.51W
(1800 kN [405kip]) in the wall direction.
Figure 4 shows the column-beam moment strength ratios.
Reinforcement of the top slab was reflected in the moment
strength of beams in negative bending (top in tension). Effective flange widths of beams were adopted in accordance
with the recommendations of the 2007 MLIT Standard or
ACI318-11, which produced roughly similar flange widths.
Variations of column axial forces due to lateral forces were
estimated from pushover analysis in the Japanese calculations. In the U.S. calculations, a plastic mechanism was
assumed in which hinging of the columns occurs at the foundation and just below the roof, and beam hinging occurs at
column faces at intermediate floors in the frame direction. In
the wall direction, the assumed plastic mechanism considered
hinging of the columns and walls at the foundation, and beam
hinging at column and wall faces. Discrepancies in columnbeam moment strength ratios evaluated using ACI and MLIT
procedures (Fig. 4) can mostly be attributed to differences in
the estimates of axial forces on columns. From the second to
fourth floors, the column-beam moment strength ratios were
slightly below 1.0 for interior columns, while those of exterior columns ranged from approximately 1.0 to 1.87.
Assessment of specimen design in accordance
with U.S. seismic design practice
The structure was assessed in both the x- and y-directions using ACI 318-11 and ASCE 7-10 provisions. The
137

Fig. 3Pushover analysis results. (Note: 1 kN = 0.225 kip.)

Fig. 4Moment 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 spectral acceleration divided by an R factor of 6, as specified in
ASCE 7-10 for a building frame system with special rein138

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 effective 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 Committee41
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. 5Acceleration 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 considered 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 walls 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 stressbased 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 satisfies 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
(100mm [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 satisfies all ACI 318 hoop spacing requirements for the boundary
element. In the upper stories of the wall at AxisA, hoops
in the boundary regions were spaced at 100 mm (3.93 in.)
and therefore did not satisfy the ACI 318-required spacing
of 83mm (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 corresponding 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 acceleration 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 approximately 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 requirement that concrete shear contribution be taken as zero
(ACI318-11, 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
0.22MPa). 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 development 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 transverse reinforcement met spacing and layout requirements of
ACI318-11 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 4Key response values at roof


Test No.

Maximum roof drift*

Residual roof drift

Input wave

x-direction, m/s

y-direction, m/s

x-direction, mm

y-direction, mm

x-direction, mm

y-direction, mm

JMA-Kobe 25%

3.12

6.37

16.9

24.2

0.5

0.4

JMA-Kobe 50%

7.03

11.01

122.4

106.9

1.1

5.4

JMA-Kobe 100%

9.65

14.01

242.7

323.9

6.2

22.5

JR-Takatori 40%

6.46

8.13

240.4

240.8

1.3

7.9

JR-Takatori 60%

8.09

9.99

278.1

414.0

8.0

11.6

Maximum roof acceleration


2

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.

satisfied the 6/5 minimum requirement of ACI 318-11. That


requirement was not satisfied at interior joints.
E-DEFENSE SHAKE-TABLE FACILITY AND
TESTCONDITIONS
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
15m (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 displacement of 1.0 m (39.4 in.) in two horizontal directions simultaneously. It can accommodate a specimen weighing up to
1200 tonnes (1323 tons). In this series of tests, the considered 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).
Fig. 6Maximum interstory drift distribution.
Joint shear demands for both interior and exterior joints
were calculated considering force equilibrium on a horizontal plane at the midheight of the joints, in accordance
with ACI318-11. Joint shear demands calculated including
the contribution of slab flexural tension reinforcement within
the ACI 318 effective flange width were found to be approximately 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
350mm (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

LOADING PROGRAM
Ground motions designated as JMA-Kobe and JR-Takatori, recorded in the 1995 Hyogoken-Nanbu earthquake,
were adopted as the input base motions. The North-Southdirection wave, East-West-direction wave, and verticaldirection wave were input to the y-direction, x-direction, and
vertical direction of the specimen, respectively. The intensity 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
40and 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 favorably with periods estimated using ASCE 7-10 Eq. 12.8-7
(0.44seconds in the frame direction and 0.31 seconds in the
ACI Structural Journal/March-April 2015

Fig. 7Damage 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 collapseprevention 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 interior 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 (0.06in.), 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 (9.8in.) 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-Takatori 60% test.
The corner portion of both wall bases suffered compressive
failure to a height of 300 mm (11.8 in.) and length of 600mm
(23.6 in.) in the JMA-Kobe 100% test. The longitudinal reinforcement 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 deformation 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 beamcolumn joint contributes to the average story drift angle, as
141

Fig. 8Deformation 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 Peaksa
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 hysteretic 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) indicates 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 direction 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. 9Deformation 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 coefficient 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 coefACI Structural Journal/March-April 2015

ficients estimated using the ASCE 7-10 equivalent lateral-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

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.
Figure12 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-dimensional effects, and strain-rate effects.

Fig. 10Hysteretic 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

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 approximately 0.1Ag fc at the first story (where Ag is the column
gross-section area and fc is the measured concrete compressive 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 Standard [2007], CSAA23.3-04 [2004], and NZS 3101 2006
[2006a,b]) account for the effects of axial force on confinement requirements of concrete columns. While these codes
treat the effects of axial forces in different ways, they
generally require less confinement reinforcement for lower
axialforces.
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 relatively low shear stresses. Such observations indicate that
beams under low shear stresses and conforming to the principles 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 highintensity 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 displacements, ACI 318 provisions would have required confined
boundary elements.

ACI Structural Journal/March-April 2015

Fig. 11Distribution of floor lateral force coefficient.

Fig. 12Hysteretic 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 satisfying 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 analysis, 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-sheardemands.
The interior beam-column joints sustained significant
damage during the earthquake simulation tests. Implications for ACI 318 are not readily extracted, however,
because the beam-column joint designs did not satisfy the
ACI318 requirements. Deficiencies included deficient ratios
of column-beam flexural strength ratios and deficient volumetric ratio of joint transverse reinforcement.

ACI Structural Journal/March-April 2015

SUMMARY AND CONCLUSIONS


A full-scale, four-story, reinforced concrete building structure 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 observations 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, longitudinal 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, particularly 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 roughened 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 contributions 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 reinforcement 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 Education, 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 instrumentation 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, Architectural 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 Rehabilitation 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 Engineers, Reston, VA, 636 pp.
CSA A23.3-04, 2004, Design of Concrete Structures, Canadian Standards Association, Mississauga, ON, Canada, 258 pp.
Kabeyasawa, T.; Shiohara, H.; and Otani, S., 1984, U.S.-Japan Cooperative 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 Engineering 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 Engineering Simulation, https://nees.org/warehouse/report/project/1005. (last
accessed Feb. 2014)
NZS3101, Part 1:2006, 2006a, Concrete Structures Standard: Part1
The Design of Concrete Structures, Standards Association of New
Zealand, Wellington, New Zealand, 309 pp.
NZS3101, Part 2:2006, 2006b, Concrete Structures StandardCommentary, Standards Association of New Zealand, Wellington, New Zealand, 397 pp.
Tuna, Z., 2012, Seismic Performance, Modeling, and Failure Assessment 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 BeamsLessons 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

TECHNICAL PAPER

Title No. 112-S13

Inverted-T Beams: Experiments and Strut-and-Tie Modeling


by N. L. Varney, E. Fernndez-Gmez, 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 configuration 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 recommended 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 conditions 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), triggering 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

In the past two decades, many structural design codes


have adopted strut-and-tie modeling (STM) as a more transparent option for the design of deep beams and other structures 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 largescale 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

147

Fig. 1Typical 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 bearings or support reactions are examples of load discontinuities. According to St. Venants 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 approximately one effective member depth d from the load or
geometricchange.
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 procedures. 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 solutions 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 equilibrium with the external forces and the concrete has enough
deformation capacity to accommodate the assumed distribution 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.

148

Fn Fu (1)

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 beams 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 reinforcing 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 reaction, and T stands for tension or tie. A CCC node is one
in which only struts intersect, a CCT node has tie(s) intersecting in only one direction, and a CTT node has ties intersecting 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 analysis. 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 additional 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


inverted-T beams is limited.

Table 1ACI 318-116 and Birrcher et al.4 STM


concrete efficiency factors
ACI 318-11

Recent advances in STM provisions


ACI 318-116 STM examines the strength of struts and
nodes separately. For the strut-node interface, the smaller
of the nominal compressive strength of the end of the strut,
Fns, and of the face of the nodal zone at which the strut acts,
Fnn, is used. These nominal compressive strengths are calculated using their respective cross sectional areas, Acs and Anz,
as shown in following equations. The concrete efficiency
factors, s and n, are summarized in Table 1.

Fns = 0.85s fcAcs (2)

Fnn = 0.85s fcAcs (3)

Birrcher et al. conducted a thorough investigation of rectangular 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 determined by the limiting compressive stress at the faces of the node
by the concrete efficiency factor in the following equation.
4

Fs = 0.85fcAcn (4)

The appropriate concrete efficiency factor, , 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
AppendixA 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 structural 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
ACI Structural Journal/March-April 2015

Strut, fce =
0.85sfc

Node, fce =
0.85nfc

Prismatic

1.00

CCC

1.00

Bottle-shaped*

0.75

CCT

0.80

Tension flange

0.40

CTT

0.60

Birrcher et al.
Node, fcu =
mfc

Bearing face,

CCC

0.85

CCT

0.70

Back face,

Strut-node
interface,
0.45 0.85

Without reinforcement satisfying ACI 318 Section A3.3.3, s = 0.60.

Without reinforcement satisfying AASHTO Article 5.6.3.5, = 0.45.

f c
0.65
20

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 provisions 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 geometry was determined based on the experimental variables
under consideration, and the preliminary loads and corresponding 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 corresponding 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
149

Fig. 3Geometry of CCT Support Node A.

Fig. 4Longitudinal 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 25degrees
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 technically 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 wellestablished 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 25degrees 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
modelA 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 corresponded to the horizontal ledge tie. The centroid of the horizontal 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 investigationOnce 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 experimental 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 recommendations for application to inverted-T beams.7 For step-bystep 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 database.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 applicability 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 (transverse 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. 5Cross-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 lengthThe 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 depthTo fully capture the effect of ledge geometry, 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 ratioTwo 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 reinforcement, v and h, 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 reinforce151

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. Likewise, 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 loadsThe beams in this investigation 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. 6Scaled cross sections of literature specimens with corresponding reference number, current specimens, and in-service
bent caps. (Note: Dimensions in inches; 1 in. = 25.4 mm.)

total force to multiple locations and preventing local ledge


failure. The spacing of the three point loads was representative of in-service girders. They were also used to help quantify 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 depthA review of the literature revealed a significant difference in the size of the in-service bent caps when
compared to the specimens used to calibrate the shear provisions 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 installation 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.
(1067mm) 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 specimens in Table 2 and their respective variables, the following
nomenclature was developed

Sample specimen designation: DC1-42-1.85-03

Fig. 7Experimental variables.


152

ACI Structural Journal/March-April 2015

Fig. 8Typical 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
(1067or 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 materials 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 reinforcement. 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
Austins Phil M. Ferguson Structural Engineering Laboratory. 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 specimens. U-shaped frames applied load evenly to the ledges of
the test specimens. At each support, six 3 in. (76 mm) diameter 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
(222kN) 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 serviceability considerations of the experimental program.7
Each beam was designed with two test regions. Specimens 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 posttensioning 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 properties and the Birrcher et al.4 STMs as implemented for invertedT 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 2Summary of beam details

Reinforcement fy, ksi

Specimen

Support plate,
in.

Load
plate, in.

No. 11

No. 6

No. 5

No. 4

fc, ksi

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

Vcalc, kip Vtest, kip

Vtest/Vcalc

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
154

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
ACI Structural Journal/March-April 2015

Table 3Summary of experimental/calculated


shear capacity
Inverted T-beams, 33 tests

Rectangular deep beams,4


179 tests

Vtest/Vcalc

Birrcher
etal.4

ACI 3186

Birrcher
etal.4

ACI 3186

Minimum

1.05

1.04

0.73

0.87

Maximum

2.17

2.17

4.14

9.80

Mean

1.50

1.57

1.54

1.80

Unconservative

0.0%

0.0%

0.6%

1.7%

Coefficient of
variation*

0.22

0.20

0.28

0.58

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 specimens 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 s 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-ofplane tension. If only web shear failure modes were considered 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 standard 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 evaluated 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

Fig. 9Test setup: specimen at failure of second test region.

Fig. 10Comparison 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 provisions 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 ultimate shear capacity obtained from the test results and
the nominal shear capacity from the STM calculations4,6
revealed conservative strength estimates for every specimen. Furthermore, the Birrcher et al.4 STMs accurately
predicted the web shear failure mode for 28 of the 33 specimens. 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 geometry of inverted-T beams requires the use of a threedimensional STM model or two equivalent and compatible 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 predictions 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 University, 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 Fernndez-Gmez is a structural engineer at
Osseous StructuralEngineering, Ciudad Jurez, Mexico. He received his
BS from Universidad Autnoma 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 International University, Miami, FL. He received his BS from Johns Hopkins
University, Baltimore, MD, in 2009, and his MS and PhD from the University of Texas at Austin in 2011 and 2014, respectively. His research interests 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, Environmental, 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 contributions 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 Association 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 Institute, Farmington Hills, MI, 1999, 56 pp.
3. Schlaich, J.; Schfer, K.; and Jennewein, M., Toward a Consistent 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 Examination of ACI 318 Strut and Tie Modeling Provisions, Symposium
Honoring James O. Jirsas 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.; Fernndez-Gmez, E.; Garber, D.; Bayrak, O.; and Ghannoum, 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 Transportation 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 Structural 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. Schtt, H., ber das Tragvermgen wandartiger Stahlbetontrger,
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 Trger, Deutscher
Ausschuss fr 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 International, 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

TECHNICAL PAPER

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 connections 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 statistically 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 dissipation 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 degradation, strength degradation, and energy dissipation, is significantly 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

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 relationship 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
dynamicanalysis.
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 significantly 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 degradation 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

157

Fig. 1Cyclic response and joint load-transfer mechanism of beam-column connections affected by bar bond-slip and diagonal 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 structures 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
Sucuolu and Acun,33 Sucuolu and Erberik,34 and Kwak
and Kim.35 Ibarra et al.36 proposed the energy-based model
for beam-column connection. Particularly, in the energybased 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, considering the design parameters. Then, the energy-based hysteresis model was defined using the predicted energy dissipation; 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 practical 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 load158

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 properties 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
(40.0to 103 ksi) and db = 9.5 to 35.8 mm (0.37 to 1.41in.),
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 parameters and the proposed method were given in the Applications section.
For parametric study of the existing test results, the energy
dissipation ratio specified in ACI 374.1-0539 was used. As
shown in Fig. 2, is defined as the ratio of the actual energy
dissipation EII per load cycle to the idealized elastic-perfectly plastic energy dissipation Eep: = EII/Eep. Generally,
the 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 loaddisplacement relationship. Thus, in the present study,
according to ACI 374.1-05,39 was defined at the third load
cycle of a relatively large story drift ratio = 3.5%. However,
when the existing test conditions did not satisfy the requirement of ACI 374.1-05,39 the was defined differently: when
the number of load cycles at = 3.5% was less than three,
was calculated for the second load cycle. When a specimen
failed before = 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. 2Definition 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 specimen), was evaluated at a moderate drift ratio of less than
= 3.5%. In the calculation of Eep, the initial stiffness ki
was defined from the envelope curve (refer to Fig. 2).39 The
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 dissipation capacity of beam-column connections is affected by
various design parameters, such as the geometry and reinforcement details of the beams and columns. However, as
discussed in the Introduction and Fig. 1, the energy dissipation 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

20 for ACI 318-11


fy
hc
(1)
20
20 for ACI 352R-02
db 420
1.25 f y

for NZS 3101:2006

3.3 f d f c

For exterior connections

fy
for ACI 318-11

5.4 f c

ldh f y
(2)
for ACI 352R-02

db 6.2 f c
0.24 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 development length of the beam flexural bars anchored inside the
joint in exterior connections. In Eq. (1), f and d are coefficients addressing the direction of the beam flexural bars
and the ductility of beam plastic hinges; and is the coefficient addressing the story drift ratio demand of the joint
(=1.530.29, in %). In Eq. (2), is the overstrength factor
of steel reinforcing bars addressing the strain-hardening
behavior (=1.25); and 1 and 2 are coefficients addressing
the details of hook anchorage and the joint confinement by
transversehoops.
The requirement for the joint shear strength is defined as
follows (refer to Fig. 1(b) and (c)).

V jn V ju (3)

where Vjn = jfcAj (nominal shear strength of the joint);


Vju=T1 + C2 Vc for interior connections; Vju = C1 Vc for
exterior connections (shear demand of the joint); is strength
reduction factor for shear; j 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)(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/12fy) 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 specimens 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
parameters and the energy dissipation ratios for the interior and exterior connections, respectively. In the figures, the
vertical and horizontal axes indicate the 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 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 coefficients 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).
On the other hand, in Fig. 3(c) and 4(c), the parameters,
which include the effects of the story drift ratio (that is, )
and the details of hook anchorage and transverse reinforcement (that is, 12), did not show good correlations with the
values.
159

Fig. 3Variation of energy dissipation ratio according to bond parameters: interior connections.9,A1-A19

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

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


For parametric study for the joint shear strength, from
Eq.(3), Vjn/Vju was chosen as the parameter. The shear
parameters of the interior and exterior connection specimens
are presented in Tables A1 and A2. Figures 5(a) and (b) show
the relationships between the energy dissipation ratios and
the joint shear parameter Vjn/Vju for the interior and exterior
connections, respectively. As presented in Tables A1 and
A2, in the majority of the connection specimens, the sum of
column flexural capacities (that is, Mc) was greater than the
sum of beam flexural capacities (that is, Mb), which indicates that the load-carrying capacity of the specimens was
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,
the R2 values were much less than those of Fig. 3(b) and
4(b), which indicates the correlation between the joint shear
parameter and the energy dissipation capacity was significantly weaker than that of the bond parameters.
The effect of beam reinforcements on the energy dissipation ratio was also investigated (refer to Appendix C).
The results showed that the correlation coefficients in the
statistics were not improved but even worse. This is because,
even for the connection specimens designed in accordance
with the bar bond requirements specified in ACI 318-115 and

160

ACI352R-02,6 the cyclic responses were dominated by the


joint deformations rather than by the beams or columns.
On the basis of the results shown in Fig. 3(b) and 4(b),
the energy dissipation ratios of the interior and exterior
connections were defined as the linear functions of the bond
parameters (hc/db)(fc/fy) and (ldh/db)(fc/fy), respectively,
using the method of least squares.
For interior beam-column connections

= 0.80

f c
hc f c
h
+ 0.053 0.16 c
0.60 (4a)
db f y
db f y

For exterior beam-column connections


= 1.56

f c
ldh f c
l
0.058 0.13 dh
0.35 (4b)
db f y
db f y

In Eq. (4a) and (4b), the upper and lower limits on the bond
parameters (hc/db)(fc/fy) and (ldh/db)(fc/fy) were specified as the minimum and maximum values presented in
TablesA1 and A2, which represents the range of the design
parameter of the existing tests.

ACI Structural Journal/March-April 2015

Fig. 6Energy-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-displacementrelationship.
The proposed load-displacement relationship of the rotational 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 developed 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 stiffness 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 monotonic 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 = u M n and M r = r M n (5b)

In Eq. (5b), u can be theoretically determined from


nonlinear section analysis addressing reinforcement details,
load conditions, and actual material strengths. When detailed
analysis is not performed, u 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, r = 0.2 was defined according to ASCE/
SEI 41-06.37
The rotational spring element represents the shear deformation of the joint and the rotation of the beam plastic
hinges. Thus, the yield deformation angle y 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 y = 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 approximated as by = ylp (1.7y/hb)0.5hb = 0.85y (y is the yield
curvature of the beam cross section, lp is the length of the
beam plastic hinge, y is the yield strain of beam flexural
bars, and hb is the overall depth of the beam cross section).
Thus, the yield deformation angle y at EY of the rotational
spring element can be defined as the sum of the elastic shear
deformation (y) of the joint and the yield rotation (by) of the
beam plastic hinge: y = y + by = y + 0.85y.
The deformation angles u at EU and r at ER were determined 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 condition. In the proposed method, on the other hand, the rotational spring elements (Fig. 6(a)) represent the overall plastic
deformation angle of a beam-column connection. Therefore,
u at EU for the rotational spring elements can be defined as
the sum of the y, bu, and ju

u = y + bu + ju (6)

where bu is the maximum plastic rotation angle of the beam;


and ju is the maximum shear angle of the joint. The values
of bu and ju are specified in ASCE/SEI 41-06.37 By the definition, ju excludes the yield rotation y.
For more accurate analysis, the maximum plastic deformation angle u at EU can be determined from other advanced
methods.51 The plastic deformation angle r at ER, which
defines the post-peak descending slope of the envelope
curve, was determined from the existing test results. For
simplicity, r was approximated as r = 2.0u.
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 (mp, Mmp) and CN (mm,
Mmm) denote the positive and negative peak points, respectively, where the unloading and reloading behaviors begin.
C2 (c2, Mc2) and C4 (c4, Mc4) denote the points where the
unloading stiffness significantly decreases, causing pinching
in the cyclic response. C1 (c1, Mc1) and C3 (c3, 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

162

kup = (1 k ) k yp and kun = (1 k ) k yn (7)

where kyp and kyn are secant stiffnesses connecting Point O


and the positive and negative yield points EY, respectively
(Fig.6(b)), and k is the coefficient representing the degradation of the unloading stiffness under cyclic loading. In
the present study, the degradation of the unloading stiffness is defined as the function of the load cycle number, i
(=0,1,2,), accumulated during the entire loadinghistory.38

k = 0.05 i 0.8 (8)

The original definition of k in the reference (OpenSees


manual) is K1 maxK2 + K3 iK4 k,max, and the coefficients
are determined on the basis of test results, users experience,
or engineering judgment. In the present study, the coefficients K1, K2, K3, K4, and k,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 (c1, Mc1) at
PointC1 and (c3, Mc3) at Point C3 are defined as the functions of the predicted energy dissipation ratio of Eq. (4a)
and (4b), as follows.

c1 = mp and c 3 = mn (9)

M c1 = M M mp and M c 3 = M M mn (10)

where (mp, Mmp) and (mn, Mmn) are plastic deformation


angles and moments at the peak points CP and CN, respectively, where the unloading behavior starts; and and M
are coefficients defined as the functions of the energy dissipation ratio , as follows.

= 0.95 + 0.5 (11)

M = 1.5 0.12 (12)

The coefficients and M 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 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 values and drift levels. On the basis
of the results, and M were defined as the linear functions
of , 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 value ranges 0.15 to 0.54, and
M vary from 0.36 to 0.01 and from 0.11 to 0.69, respectively. 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. 7Predicted 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 corresponding to mp and mn, respectively. Because the plastic
deformation angles at CP and CN are greater than those at
CP and CN, the moments corresponding to mp and mn of the
second and third load cycles are less than those of the first
load cycle. Herein, the plastic rotation angles mp and mn at
the modified peak points CP and CN are defined as follows
(refer to Fig. 6(d)).

mp
= (1 + s ) mp and mn
= (1 + s ) mn (13)

The coefficient s is defined by the number of load cycles,


i(=0, 1, 2, ) accumulated during the entire loadinghistory.38

s = 0.1 i 0.2 0.5 (14)

The original definition of s in the reference (OpenSees


manual) is S1 maxS2 + S3 iS44 s,max, and the coefficients
are determined on the basis of test results, users experience,
or engineering judgment. In the present study, the coefficients S1, S2, S3, S4, and s,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 numerical 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 accurately 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 beamcolumn 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 beamcolumn connections, as the function of the energy dissipation 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 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 specimens.9,A3,A4,A11,A19,A22-A24,A26,A28 In all specimens, the column
flexural capacities (that is, Mc) were greater than the beam
flexural capacities (that is, Mb). 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 [=4700fc]
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. 8Predicted cyclic responses versus test results for exterior connections.A22-A24,A26,A28
Table 1Modeling parameters for existing test specimens
Bond resistance
parameter
ldj

Specimens

f c

Modeling parameter

db f y

lq

lM

qy, rad

qbu, rad

qju, rad

qu, rad

Interior

Hwang S3A19
Durrani S3A4
Brooke 4B9
Xian U5A11
Xian U3A11
Dai U1A3

0.162
0.292
0.324
0.356
0.415
0.588

0.182 (0.174)
0.287 (0.303)
0.312 (0.333)
0.338 (0.358)
0.385 (0.391)
0.524 (0.500)

0.327
0.227
0.204
0.179
0.134
0.002

0.153
0.311
0.348
0.387
0.458
0.666

0.0079
0.0052
0.0063
0.0048
0.0054
0.0034

0.025
0.025
0.025
0.025
0.025
0.025

0.015
0.015
0.015
0.015
0.015
0.015

0.0479
0.0452
0.0463
0.0448
0.0454
0.0434

Exterior

Tsonos S2A26
Shiohara L06A28
Ehsani 4A22
Kaku 2A23
Ehsani 2A22
Chutarat SAA24

0.141
0.170
0.215
0.251
0.291
0.341

0.161 (0.184)
0.207 (0.212)
0.277 (0.281)
0.333 (0.342)
0.397 (0.391)
0.474 (0.510)

0.347
0.303
0.237
0.184
0.123
0.050

0.122
0.191
0.296
0.380
0.476
0.591

0.0040
0.0035
0.0041
0.0040
0.0038
0.0035

0.025
0.025
0.025
0.025
0.025
0.025

0.010
0.010
0.010
0.010
0.010
0.010

0.0390
0.0385
0.0391
0.0390
0.0388
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 u = 1.25 and
r = 0.2, Mu and Mr were determined as 1.25My and 0.2My,
respectively. As mentioned, the yield deformation angles y
at EY of the specimens were defined as the sum of the elastic
shear deformation y of the joint and the yield rotation by
(=0.85y) of the beam plastic hinge: y = y + 0.85y. The
y 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.
Table1 presents the y values of the test specimens. The y
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 u of the
connection specimens were determined by using Eq. (6):
u = bu + ju + y. Table 1 presents the maximum beam plastic
rotation angles bu, and the maximum joint shear angles ju
of the connection specimens. bu and ju corresponding to the
164

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 dissipation ratio in Eq. (4a) and (4b). The values of the specimens are presented in Table 1. Detailed calculations for the
envelope curves and the cyclic curves are presented in the
AppendixD. In Table 1, for instance, the predicted 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
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 hysteresis model predicted the cyclic responses of the specimens
with reasonable precision, including the energy dissipation,
pinching, and strength and stiffness degradations during
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 values
to the less-pinched cyclic curves with higher values. In
AppendixB, 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 Shiohara L06 and Ehsani 2 was significantly overestimated. This
is because Kim and LaFaves method49,50 underestimated the
yield deformation angles y. 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 Mnc/Mnb 1.0, the joint shear capacitydemand ratios 0.5 Vn/Vu 4.25, the column depth-beam
bar diameter ratios 14.5 hc/db 37.5 for interior connection, 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.

correlated better with the bond resistance of beam flexural


bars at the joints, than the joint shear resistance. Thus, the
energy dissipation ratios of interior and exterior connections were defined as the linear functions of the bond parameters of beam flexural bars, (hc/db)(fc/fy) and (ldh/db)(fc/fy),
respectively.
2. To simulate the cyclic responses of interior and exterior 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 and the loading
history. The predictions of the proposed method correlated
well with the test results of existing interior and exterior connection specimens. By using the constraint condition 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.

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 interior and 63 exterior beam-column connections, the relationships between the bond resistance of beam flexural bars at
the joints and the energy dissipation capacity were quantified. 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 connections 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 dissipation per load cycle) of interior and exterior connections

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. 22612275. 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 BeamColumn Connections in Monolithic Reinforced Concrete Structures (ACI352R02), American Concrete Institute, Farmington Hills, MI, 2002, 38 pp.

ACI Structural Journal/March-April 2015

AUTHOR BIOS

Tae-Sung Eom is an Assistant Professor in the Department of Architectural 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 Engineering 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 Architecture & 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 reinforced 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 Construction and Environmental Engineering Seoul National University Research
Program, funded by the Ministry of Education & Human Resources Development. The authors are grateful to the authorities for their support.

REFERENCES

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 Earthquake 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 BondSlip 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 Engineering 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 BeamColumn Joints, ASCE Structures Congress, Seattle, WA, 2003.
16. Mitra, N., and Lowes, L. N., Evaluation, Calibration, and Verification 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 Deformation 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 Reinforced Concrete Frames Including Joint Shear Deformation, Engineering 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-Dimensional 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 Relationships for Reinforced Concrete Elements, Earthquake-Resistant Concrete
Structures Inelastic Response and Design, SP-127, S. K. Ghosh, ed., American 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 ForceDisplacement 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, Engineering Structures, V. 32, No. 9, 2010, pp. 2884-2893. doi: 10.1016/j.
engstruct.2010.05.007
33. Sucuolu, 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. Sucuolu, 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, Proceedings, 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 Engineering, 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 BeamColumn 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 Structural 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 Isakovi, 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 Procedures, Federal Emergency Management Agency, Washington, DC, 2005.

ACI Structural Journal/March-April 2015

ACI STRUCTURAL JOURNAL

TECHNICAL PAPER

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 fibersnamely, polypropylene and polyester
were used. The HyFRC prisms exhibited approximately 10to 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 displacement 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 connections 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 beamcolumn joints under seismic actions. The existence of two
shear-resisting mechanismsone involving joint shear
reinforcement and the other a linear concrete strutwere
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. AbdelFattah and Wight5 studied the relocation of plastic hinging
zones for earthquake-resistant design of reinforced concrete
(RC) buildings. Chutarat and Aboutaha6 investigated a solution 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

tested four full-scaled exterior precast beam-column joint


under cyclic loading in order to identify a suitable technique for connecting precast beam and column components.
Nie etal.8 tested six beam-column joints comprising three
interior and three corner joints to develop a new connection system for concrete-filled steel tube composite column
and RC beams. Park and Mosalam9 carried out both experimental 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 buildings 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 compression 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 properties. 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 macrocracks 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 polyolefin 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

167

Table 1Details of fibers


Length, in.
(mm)

Diameter, in.
(mm)

Shape

Density, lb/ft3
(kg/m3)

Polyester (PE)

0.47 to 0.71
(12 to 18)

0.0012 to 0.0014
(0.03 to 0.035)

Straight

85.625 (1370)

400 to 500

Micro

57.92 to 72.4
(400 to 500)

Polypropylene (PP)

0.47 (12)

0.0012 to 0.0014
(0.03 to 0.035)

Straight

56.875 (910)

400 to 500

Micro

57.92 to 72.4
(400 to 500)

Steel (SF1)

2.36 (60)

0.0296 (0.75)

490.625
(7850)

80

Macro

177.37 (1225)

Steel (SF2)

1.38 (35)

0.0216 (0.55)

490.625
(7850)

64

Macro

159.27 (1100)

Types of fibers

Hooked

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 strainhardening, 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 fractions 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 reinforcement for confinement and shear in RC columns. Kumar
etal.19 carried out detailed studies to improve seismic performance 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 transverse 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 prefabricated 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 investi168

Aspect ratio
Tensile strength,
(length/diameter) Classification
ksi (MPa)

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 reinforcement 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 dissipation 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 polypropylene, were used in the concrete mixtures, and the geometrical 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 2Volume fractions of fibers used and average toughness of prisms


Types of fibers (volume fraction, %)
No. of combination

SF1

SF2

PP

PE

Total volume fraction, %

Average toughness,
kip-in. (kN-mm)

1 (plain concrete)

0.221 (24.949)

0.5

0.5

1.00

2.213 (250.039)

0.4

0.4

0.15

0.95

3.197 (361.291)

0.4

0.4

0.2

1.00

2.486 (280.991)

0.5

0.5

0.15

1.15

3.445 (389.317)

0.5

0.5

0.2

1.20

2.849 (321.982)

0.6

0.6

0.15

1.35

3.225 (364.467)

0.4

0.4

0.15

0.95

2.713 (306.551)

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 fraction 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 enhancement 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 specimens, thus leading to a more ductile type of failure pattern.
DETAILS OF BEAM-COLUMN CONNECTIONS
To ascertain the efficacy of different fibers and their combinations 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 fibersused as a control specimen

ACI Structural Journal/March-April 2015

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,
SectionA.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
IS1392024 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 downward 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
169

Fig. 1Beam-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 actu170

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 pushpull 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. 2Strain gauge layout in gauged bar. (Note: Dimensions in mm; 1 mm = 0.0394 in.)

Fig. 5Displacement history. (Note: 1 mm = 0.0394 in.)


Table 3Strength of concrete in beam-column
specimens
Fig. 3Test arrangement.

Fig. 4View of actual test setup.


in the present work. Tests were stopped when the degradation 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 beamcolumn 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

Specimen

Fiber types

Compressive cube
strength, ksi (MPa)

Indirect tensile
strength, ksi (MPa)

None

4.67 (32.22)

0.585 (4.04)

SF1/SF2/PP

4.50 (31.11)

0.847 (5.85)

SF1/SF2/PE

4.89 (33.77)

0.715 (4.94)

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 consistently 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 Specimen1
(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 4Test result of different beam-column specimens


Peak load, kip (kN)
Specimen

Pull

Push

Maximum
displacement, in. (mm)

Drift, %

Toughness, kip-in. (kN-mm)

8.75 (38.935)

7.73 (34.402)

1.84 (46.67)

5.18

236.55 (26,711)

8.69 (38.645)

9.64 (42.819)

2.76 (70.0)

7.77

688.029 (77,691)

7.24 (32.212)

6.863 (30.533)

2.76 (70.0)

7.77

593.66 (67,035)

6.64 (29.537)

7.47 (33.273)

2.43 (61.67)

6.85

410.855 (46,393)

Fig. 6Comparison of damage in joint region.


peak level and toughness in HyFRC specimens are consistently 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 displacements 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 immediately apparent. The addition of polymer fibers was particularly effective. A more detailed comparison of specimen
172

behavior is obtained by comparing the load-displacement


hysteresis loops as shown in Fig. 7. All four specimens
exhibited displacement ductility, but the superior performance of the specimens with fibers is indicated by their
higher final displacements and their reduced strength degradation 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 aformentioned 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. 7Load-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 2.43in. (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 comparable 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 Specimen 4 did. Further, Fig. 10 demonstrates the contribution
of fibers in the test specimens in terms of nominal principal
tensile stresses values developed in the beam-column joint
regions.29 From this figure it is construed that though the
developed nominal principal tensile stresses of all joints
are not substantially different, the decay in stress for Specimen 1 is much more significant than all other specimens.
Specimen 2 and 3 also presented flatter stress patterns than
Specimen 4. Thus, both damage index and nominal principal

ACI Structural Journal/March-April 2015

Fig. 8Envelope curve for beam-column specimens. (Note:


1 kN = 0.225 kip; 1 mm = 0.0394 in.)
tensile stress indicate the influence of fibers in enhancing the
seismic performance of these specimens.
The observation related to displacement ductility and
cumulative energy dissipation of all the four specimens are
discussed in detail to demonstrate the significance of using
HyFRC in the D-region of a beam-column connection.
173

Fig. 11Cumulative energy dissipation of beam-column


specimens. (Note: 1 kN = 0.225 kip; 1 mm = 0.0394 in.)
Fig. 9Comparisons of the Park and Ang32 damage indexes
of the tested specimens.

Fig. 10Nominal 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. Toughness values for all the four specimen types are shown in
Table4 and it is immediately clear that the combination SF1/
SF2/PP of appropriate volume fraction has tremendous potential 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 >


Specimen1 (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
Figure12 shows (using the results for Specimen 1) the necessary 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 inter174

Fig. 12Procedures 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 corresponding 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 Gauge1
(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 reinforcement 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 5Displacement ductility of different beamcolumn specimens


Specimen

y, in. (mm)

u, in. (mm)

u/y

0.172 (4.375)

1.044 (26.5)

6.057

0.125 (3.165)

1.826 (46.335)

14.640

0.156 (3.950)

1.793 (45.5)

11.519

0.210 (5.330)

1.834 (46.58)

8.739

connection zone until the entirety of the U-bar in the connection 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.
(3.33mm). The cracks in the other three specimens also
started approximately at the same location and displacement level. The strain distributions in all three specimens
are shown in Fig. 13. The maximum value of strain in
Specimen1 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 displacements of 1.837 in. (46.67 mm)the largest displacement
sustained by all four specimenspeak strains were approximately (as there was considerable creep) 21,420, 6110,
9900, and 13,300 microstrain for Specimens 1 to 4, respectively. 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 reinforcing bars required to achieve a given displacement, with
hybrid Specimens 2 and 3 proving more effective than Specimen 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 development 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 efficient under seismic loading.
Overall, adding fibers gave increased displacement
ductility coupled with reduced reinforcement strains. It
may further be noted that polypropylene has better ultimate elongation as compared to polyester fibers. Thus, it is
observed that the performance of Specimen 2 (with polypropylene) was relatively better than Specimen 3 (with
ACI Structural Journal/March-April 2015

Fig. 13U-bar strains at 0.131 in. (3.33 mm) displacement.


(Note: 1 mm = 0.0394 in.)
polyester). Further, hybrid fibers (steel and polymer fibers)
are better than the use of only steel fiber, as the inclusion
of steel fibers in the concrete mixture is an effective way of
reducing macrocracking, while polymer fibers are very good
175

Fig. 15U-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.

Fig. 14U-bar strains at 1.837 in. (46.67 mm) displacement. (Note: 1 mm = 0.0394 in.)
at arresting microcracking, thus leading to overall enhancement 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

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 beamcolumn 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 addition of any fiber in concrete and was considered as the control
specimen for comparison, whereas Specimens2 and3 were
ACI Structural Journal/March-April 2015

hybrids, containing both steel and polymer fibers, and Specimen 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 polypropylene 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 reinforcement 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 maintaining 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 interests include studies of beam-column joints.
ACI member Richard H. Scott is a Visiting Professor of structural engineering 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 Structures, 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-performance materials in concrete and system identification based studies in
assessment of structures.

ACKNOWLEDGMENTS

The financial support provided by the Royal Societys 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 Engineering, 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
BarsExperimental 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 Structural 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 Proposition, ACI Materials Journal, V. 92, No. 1, Jan.-Feb. 1995, pp. 48-57.
11. ACI Committee 544, Design Considerations for Steel Fiber Reinforced 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 Composites, 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 Reinforced 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, University of California, Berkeley, Berkeley, CA, 2011.
20. Bedirhanoglu, I.; Ilki, A.; and Kumbasar, N., Precast Fiber Reinforced Cementitious Composites for Seismic Retrofit of Deficient RC
JointsA 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 Standards, 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 Rehabilitation 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 JacketsAn
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 BeamColumn Connection Behavior, ACI Structural Journal, V. 93, No. 3,
May-June 1996, pp. 336-346.

ACI Structural Journal/March-April 2015

ACI STRUCTURAL JOURNAL

TECHNICAL PAPER

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 castin-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

newer and more innovative bridge columns in earthquakeprone 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 halfscale 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 testresults.
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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

179

Fig. 1Mechanical reinforcing bar splice commercially available in the United States.

Fig. 2Uniaxial 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)
Grade60 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 16in.
(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 manufacturers minimum specified torque of 150 lbf-ft
(203N-m).
GC specimens were constructed using No. 8 (D25)
Grade60 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 1Summary of test results


Yield stress, ksi

Ultimate stress, ksi

Strain at rupture, %
Bar

ID

Average Standard deviation Average

Standard deviation Average

Coupler region

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
asGC-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 0.07in./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 44.5mm/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 (8.72and
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
ACI Structural Journal/March-April 2015

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,
181

Table 2Average measured strain rates during


dynamic tests (unit: strain/s)
HC-D

Fig. 3Stress-strain curves from monotonic tests on


HC device (static tests are solid lines; dynamic tests are
dashedlines).

Fig. 4Stress-strain curves from monotonic tests on


GC device (static tests are solid lines; dynamic tests are
dashedlines.)

Fig. 5Evidence 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

GC-D

Stress range

Average

Standard
deviation

Average

Standard
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 groutfilled 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. 6Cyclic 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-Dspecimens.
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
0.01in. (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
manufacturers 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 stressstrain 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).

Fig. 7Relationships between strain in reinforcing bar and


strain over splice.
One GC specimen (GC-C1) was tested under cyclic
loading. The average yield stress was found to be
comparable to static results, but the average ultimate stress
and elongation of the reinforcing bar at rupture were 9% and
65% lower than static tests, respectively (Fig. 6(d) through
(f)). The discrepancies between static and cyclic tests on
the GC device were not caused by premature failure of
the splice. In both cases failure was a result of reinforcing
bar rupture away from the splice. However, in GC-C1, the
reinforcing bar did not exhibit ductile behaviorthat is,
necking of the bar at the point of fracture did not occur rather
the fracture surface was flat. This may have been caused by
a low-cycle fatigue-type effect, but such failures typically
require significantly larger strain reversals.20 There was also
a slight difference between the static and cyclic stress-strain
backbone curves for the coupler region. The slope of the
unloading curves for both the coupler region and reinforcing
bar were similar indicating the reinforcing bars control
the unloading stiffness. Unlike HC-C tests, there was not
a visible instantaneous deformation during the transition
between tensile and compressive loads. The average stressstrain history from the midheight of the cast-iron sleeve
indicates low strains and slight nonlinearity.
Comparison and discussion
It was observed that the presence of the mechanical splice
reduced the deformation capacity over the coupler regions for
both devices. Figure 7 presents representative relationships
between the strain over LCR and the estimated strain over
LSp in terms of the average reinforcing bar strain obtained
from strain gauge measurements. The slope of the curves is
referred to as the splice-bar strain ratio (SR) and indicates
the relative stiffness of the splice to the reinforcing bar. As
mentioned previously, there was little difference between the
static and dynamic behavior of the coupler regions, which
can be seen in Fig. 7. This indicates that strain rates similar
to those expected from an earthquake would not adversely
184

CORRELATION BETWEEN UNIAXIAL AND


BRIDGE COLUMN TESTS
A series of half-scale reinforced concrete bridge column
tests were conducted by the authors to investigate the
seismic performance of precast column-footing joints
with mechanical reinforcing bar splices.8 Two connection
configurations were studied: one employing two layers of HC
splices, referred to as HCNP, and the second using singlelayer GC splices, referred to as GCNP. The reinforcement
details and model geometry were designed assuming the
behavior would be similar to a conventional cast-in-place
(CIP) benchmark column; this is otherwise referred to
as emulative design. That is, a conventional column
model was designed for a target displacement ductility
capacity of C = 7.0 (ultimate displacement/effective yield
displacement) according to Caltrans SDC, and the details
of the plastic hinge region were modified to incorporate the
mechanically spliced connections. The column models had
24 in. (610 mm) diameter cross sections, an aspect ratio of
4.5, and longitudinal and transverse reinforcement ratios
of 1.9% and 1.0%, respectively. The ratio of axial load to
the product of the gross column cross-sectional area and
the specified concrete compressive strength was 0.1. Along
with the precast models, a CIP benchmark model was also
constructed and tested for comparison. Each column was
tested in a cantilever configuration and subjected to slow
cyclic loading at increasing drift levels. Figure 8(a) shows
the pertinent plastic hinge connection details of these three
half-scale column models. A detailed discussion of these
tests can be found in Haber et al.8
Figures 8(b) and (c) show the moment-rotation
relationships for columns with HC and GC splices,
respectively, measured near the base of the column (shown
in Fig. 8(a)). The data, which is only shown up to 5% due to
localized bar buckling thereafter, captures the influence of
the mechanical splices on the hinge rotations. The effect of
gap closure can be clearly observed in the moment-rotation
response of HCNP in the form of a slight pinch in the
unloading branches. The pinch, however, is small and stable
and does not lengthen as deformations increase. Except for
the pinch, the loops are similar in shape and magnitude to
those of the CIP column. This is not surprising given the
relatively short length of the HC device. On the other hand,
ACI Structural Journal/March-April 2015

Fig. 8Behavior 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 stressstrain 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 stressstrain 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. 9Stress-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 linearelastic 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

186

ey

Es
= SRE
Es*

(1)

Fig. 10Analytical model for bridge column with grouted


sleeve column-footing connection.

e*u e*sh Esh


=
=
= SRI
e u e sh Esh*
ey
SRI

SRE e sh

(2)

(3)

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 4.7ksi (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 Mandersmodel.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. 11Comparison 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 columnfooting interface. In this model, the fibers at the columnfooting 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 displacementbased 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. 12Comparison 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 BuildingCode).
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 Evaluation 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 Engineering 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 Conshohocken, 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, Department 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 StressStrain 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

TECHNICAL PAPER

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 hardening 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 engineered cementitious composites (ECCs), show excellent
mechanical behavior characterized by tensile strain hardening 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 degradation 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 conventional RC beams to keep the crack opening under 0.3mm
(0.012in.) 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 strainhardening 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

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 distinguishes 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 characteristics 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

189

Table 1Mechanical properties of PVA fiber

Type

Length,
mm (in.)

Diameter,
mm (in.)

PVA

12.0 (0.472)

0.04
(1.6 103)

Tensile
strength,
N/mm2 (ksi)

Elastic
modulus,
kN/mm2 (ksi)

Fiber volume
fraction, %

w/b

Sandbinder
ratio

1690 (245)

40.6 (5890)

2.0

0.50

0.77

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 characteristics, shear-resistance performance, and bond behavior associated 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 considering 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 longitudinal 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 (1.6 103 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
240kg/m3 (405 lb/yd3) fly ash. Fine sand with a size under
0.2 mm (7.9 103 in.) was used as fine aggregate. The
mixing of SHCC was carried out using a biaxial mixer of
2m3 (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 2Mixture proportion of SHCC


Unit weight, kg/m3 (lb/yd3)
Water
400
(675)

Cement Fly ash


540
(911)

240
(405)

Sand
619
(1044)

Cement: Ordinary portland cement


Fly ash: Type II of Japanese Industrial Standard (JIS A 6202)
Sand: Size under 0.2 mm (7.9 103 in.)
High-range water-reducing admixture: binder 0.65%

types of uniaxial tension tests with several types of cementitious 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 orientation 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 100 x 200 mm (3.94 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 3Material test results of SHCC


Compression test (100 x 200 mm cylinder)

Bending test* (100 x 100 x 400 mm)

Test series

Compressive strength, N/mm (ksi)

Elastic modulus, kN/mm (ksi)

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 4Mechanical 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. 1Assumption of stress distribution of HPFRCC.

Fig. 2Bending test setup.


respectively. The tension test results for the deformed bars
used in this study are listed in Table 4.
PULLOUT BOND TEST
Outline of experiment
The specimen and loading method are shown in Fig. 5. The
specimen was a rectangular SHCC block with a height of
91mm (3.58 in.). One deformed bar with a diameter of 13mm
(0.51 in.) was arranged in the central position of the block.
The unbonded regions were set at both the loaded and free
ends. The embedded length was four times the bar diameter db
in the central part of the specimen. The slits made of foamed
polystyrene were set, as shown in the figure, to simulate the
cover thickness or bar spacing. The dimension of the slit was
the main parameter, and it was set to have a cover thickness C
ACI Structural Journal/March-April 2015

Fig. 3Moment-curvature curve.

Fig. 4Tested reinforcement (D13).


of 0.5, 1.0, and 1.5 times the bar diameter. These thicknesses
are small values compared with the cases for conventional RC
elements. Three identical specimens for each parameter were
tested. Nine specimens were tested in total.
SHCC was cast from the side of the specimen, as shown
in Fig. 5. The casting level of SHCC at both sides of the slits
was raised almost uniformly to avoid damage to the polystyrene slits. The casting was done with careful observations of
the flow of the SHCC around the reinforcement to not have
any voids between the slits.
The monotonic pullout load was applied until the
reinforcement slipped out from the block under the controlled
displacement. Teflon sheets were placed between the specimen and the reaction plate to facilitate lateral displacement
of the block. The LVDT was set to measure slip at the free
end. The loaded end slip is obtained as the summation of the
191

Table 5Pullout test results


Average
C/db
0.5

1.0

1.5

ID

Maximum bond stress b,max,


N/mm2 (ksi)

Slip at maximum bond stress,


smax, mm (in.)

P05-1

7.68 (1.11)

0.499 (0.0196)

P05-2

6.85 (0.99)

0.589 (0.0232)

P05-3

8.57 (1.24)

0.533 (0.0210)

P10-1

8.37 (1.21)

0.455 (0.0179)

P10-2

11.19 (1.62)

0.577 (0.0227)

P10-3

9.70 (1.41)

0.389 (0.0153)

P15-1

12.01 (1.74)

0.303 (0.0119)

P15-2

12.98 (1.88)

0.333 (0.0131)

P15-3

14.73 (2.14)

0.179 (0.0070)

Fig. 5Pullout 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
(1.12to 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, 0.5times the bar diametera higher strength is
obtained than in the case of conventional concrete. Figure6
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. Figure7 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 thickness exhibited a look-alike yielding behavior.
192

Maximum bond stress b,max,


N/mm2 (ksi)

Slip at maximum bond stress,


smax, mm (in.)

7.70 (1.12)

0.540 (0.0213)

9.75 (1.41)

0.474 (0.0186)

13.24 (1.92)

0.272 (0.0107)

Tepfers11 suggested the cylinder models of stress distributions 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) reinforcement. 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.

ACI Structural Journal/March-April 2015

Fig. 6Pullout 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 (0.24in.)
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
1.57in.). Embedded lengths of 16 or 8 times the bar diameter 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.
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
force divided by the surface area of the main bar. The bond
stress-free end slip curves are shown in Fig. 11 through 13.
In all specimens, a sudden decrease in the bond stress was
not observed. The curves show ductile behavior until several
millimeters of slip. From the left-side figures, in which specimens without stirrups are shown, curves obtained from the
corner bars and center ones show similar behavior. On the

ACI Structural Journal/March-April 2015

Fig. 7Bond stress-slip curves from pullout test.

Fig. 8Bond strength of pullout specimen.


other hand, in the right-side figures, corner bars show high
bond stresses in the specimens with stirrups.
Bond strength of beam bond test
Table 7 lists the experimental results for bond strength
(maximum bond stress). Figure 14 shows the relationship
between the average bond strength and stirrup ratio. It is
clearly observed that bond strength increases as the stirrup
193

Table 6List of beam specimens


Main bar

Stirrup
Arrangement

pw, %

Embedded
length

None

0.0

16db

2-D6 at 80

0.4

16db

None

0.0

8db

No.2-R

2-D6 at 40

0.8

16db

No.3-L

None

0.0

16db

2-D6 at 80

0.4

16db

ID

Arrangement

pt, %

No.1-L
No.1-R
No.2-L

No.3-R
No.4-L

5-D13
(C/db = 1.0)

6-D13
(C/db = 0.8)

0.91

1.09

None

0.0

8db

No.4-R

2-D6 at 40

0.8

16db

No.5-L

None

0.0

16db

2-D6 at 80

0.4

16db

No.5-R
No.6-L

7-D13
(C/db = 0.6)

1.27

No.6-R

None

0.0

8db

2-D6 at 40

0.8

16db

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

Fig. 10Examples 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
194

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
ACI Structural Journal/March-April 2015

Fig. 11Bond stress-slip curves from beam test (5-D13).

Fig. 13Bond 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 specimen affects the average bond strength, as listed in Table7.
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)

Fig. 12Bond stress-slip curves from beam test (6-D13).


the experimental results of cantilever-type specimens, while
Orangun and Jirsa formulas were built by the results from
beam specimens. The Orangun and Jirsa formulas include the
effect of bond length, whereas the Morita and Fujii formulas
do not include the term of bond length. Figure 16 shows the
comparisons of bond strength between experimental and
calculated values by two formulas. Clearly, experimental
bond strengths are quite higher than the calculated bond
strengths. The calculated values from the Orangun and Jirsa
formulas are higher than those from Morita and Fujii due to
the difference of bond length influence.
PREDICTION METHOD OF BOND-SPLITTING
STRENGTH OF SHCC
The test results of the pullout bond test indicate that the
stress distribution of SHCC corresponds to the plastic stage
model suggested by Tepfers. This assumption is understandable for SHCC, which retains tensile stress after cracking.
Though the non-uniform tensile stress distribution is
assumed by Chao et al.6 or Asano and Kanakubo,7 a uniform
stress distribution can be assumed in the case of small cover
thickness or bar spacing, as shown in Fig. 17. Furthermore,
the bond stress-slip curves from the pullout bond test show
very ductile behavior over the slip of 1/10 the bar diameter.
ACI Structural Journal/March-April 2015

s db b = t 2C s (1)
bc = s cot
bc = t

(2)

2C  s
cot (3)
db  b

where s is the splitting stress; db is the diameter of deformed


bar; b is embedded length; t is the tensile strength of
SHCC; C is the half-length of bar spacing or cover thickness
(2C = [b Nt db]/Nt); s is the splitting length; bc is the bond
strength provided by SHCC; is the angle between principal
bearing stress and axial direction; b is the width of the beam;
and Nt is the number of main bars.
The confinement effect of the lateral reinforcement has to
be considered to evaluate the bond strength of the specimens
with stirrups. Yasojima and Kanakubo14 proposed a prediction method of bond splitting strength for conventional RC
elements. The bond strength provided by the confinement
effect is predicted by considering compatibility conditions
between the splitting crack opening and deformation of
the lateral reinforcement. The bond strength is given at the
compressive failure of concrete bearing with the rib of the
deformed bar. The identical situation can be considered in
the case of the SHCC element because the tensile behavior
of SHCC has no effect on these phenomena. Equation (4) is
195

Table 7Beam test results


Test variables
ID

Main bar

No.1-L

Bond strength, N/mm2 (ksi)

pw, %

Embedded length

Corner bar (average)

Center bar (average)

Average (all bars)

0.0

16db

5.81 (0.843)

5.94 (0.862)

5.89 (0.854)

0.4

16db

8.98 (1.302)

7.06 (1.024)

7.83 (1.136)

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)

0.4

16db

7.91 (1.147)

6.32 (0.917)

6.85 (0.994)

No.1-R
No.2-L

No.3-R

5-D13

6-D13

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)

0.4

16db

7.78 (1.128)

5.62 (0.815)

6.24 (0.905)

No.4-L

No.5-R
No.6-L

7-D13

No.6-R

0.0

8db

7.65 (1.110)

8.15 (1.182)

8.01 (1.162)

0.8

16db

8.37 (1.214)

5.97 (0.866)

6.66 (0.966)

Fig. 14Bond strength-stirrup ratio relationship.


proposed for the bond strength provided by the lateral reinforcement confinement14

bs = 0.018

Fig. 15Bond strength of corner and center bars.

b pw hr

Est B cot (4)


N t d b 9d w

where bs is the bond strength provided by lateral reinforcement 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; B is the compressive strength of
SHCC; and is the angle between principal bearing stress
and axial direction (= 56 degrees).
Consequently, bond strength b is predicted by Eq. (5)

b = bc + bs (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

196

Fig. 16Comparison 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.

ACI Structural Journal/March-April 2015

Table 8Calculated bond strength


ID

Main
bar

No.1-L
No.1-R

Predicted bond strength, N/mm2 (ksi)

Experimental bond
strength, N/mm2 (ksi)

bc

bs

bc + bs

Experiment/prediction

5.89 (0.854)

6.32 (0.917)

0 (0)

6.32 (0.917)

0.93

7.83 (1.136)

6.32 (0.917)

1.59 (0.230)

7.91 (1.148)

0.99

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

6.85 (0.994)

5.06 (0.734)

1.21 (0.175)

6.27 (0.909)

1.09

No.2-L

No.3-R
No.4-L

5-D13

6-D13

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

6.24 (0.905)

3.79 (0.550)

0.96 (0.139)

4.75 (0.689)

1.31

8.01 (1.162)

7.59 (1.101)

0 (0)

7.59 (1.101)

1.06

6.66 (0.966)

3.79 (0.550)

1.36 (0.197)

5.15 (0.747)

1.29

No.5-R
No.6-L
No.6-R

7-D13

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.
AUTHOR BIOS

Fig. 17Cylinder model for plastic stage.

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 structural 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.

REFERENCES

Fig. 18Comparison 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
ACI Structural Journal/March-April 2015

1. RILEM TC 208-HFC, Strain Hardening Cement Composites: Structural Design and Performance, State-of-the-Art Report of the RILEM Technical 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 Structural 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, Gteborg,
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.

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 Splitting Behavior of ECC, Bond in Concrete 2012, Volume 2Bond 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-Curvature 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

198

b = (co + st) 1.22

(A1)

co = ( 0.307 bi + 0.427 ) B (A2)

st = 24.9 k

Ast
B (A3)
s Nt db

where b is the bond strength for bottom bars (kgf/cm2);


co is the bond strength without stirrup (kgf/cm2); st is
the bond strength increment caused by stirrup (kgf/cm2);
bi = b/N db 1 (side split); B 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

b = co + st (A4)

3C 50db
co = 1.2 + s +
B (A5)
db
 b

st =

Ast wy
500 s db

B (A6)

where b is the bond strength (psi); co is the bond strength


without stirrup (psi); st 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.); b is the bond length (splice
length) (in.); B is the concrete compressive strength (psi);
Ast is the sectional area of pair of stirrup (in.2); wy is the
yield strength of stirrup (psi);and s is stirrup spacing (in.).

ACI Structural Journal/March-April 2015

ACI STRUCTURAL JOURNAL

TECHNICAL PAPER

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 investigate 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, independent 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 deflection, 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 reinforcement; 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, commercial buildings, parking garages, and office buildings, reinforced 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 recommendations 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 mechanisms 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

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 variation 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, ultimate 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 reinforcement 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 910mm
[18.11 to 35.83in.], 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 longitudinal 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers 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 reinforced 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 experimental 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 reinforcement 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 250mm
(9.8 in.) height. The column part was square and with
nominal dimensions of 300 mm (11.8 in.) sides and 300mm
(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 appropriate 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 reinforcementthis 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 independent 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 independent bent-up bars was approximately 240mm (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 75mm
(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 reinforcement was arranged in a band of width equal to and centered
on the column width to compare its effect with the evenly

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 longitudinal 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.) longitudinally 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 Specimen 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 performances 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 specimens were tested and, for each bar stock, randomly selected
samples were used to determine the standard stress-strain
curve of the steel bars.

Fig. 1Specimens configuration and test setup.


ACI Structural Journal/March-April 2015

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 reinforcement. 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 1Geometric properties of specimens


Designation of shear reinforcement
Specimen

bw/d, mm
(in.)

Height h,
mm (in.)

Shear span a,
mm (in.)

Stirrups
a/d

w, % Av, mm (in. ) Sv, mm (in.)


2

Horizontal bars

Independent bent-up bars

Avh, mm (in. )

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. 2Reinforcement cages before casting.


strength of a reinforced concrete beam, Vc, can be written as
(ACI 318-08, Eq. (11-2))11

Vn = Vc + Vs (1)

where Vc is nominal shear strength of concrete, and Vs is


nominal shear strength of web reinforcement.
In the truss analogy contains the shear resistance of a
parallel chord truss and a web-reinforced concrete beam,

202

where concrete struts run parallel to diagonal cracks and


stirrups perform as tension members.26
For a member subject to shear and flexure only, the expression used for shear capacity of the member without shear
reinforcement is (ACI 318-08, Eq. (11-3))

Vc = 0.166 f cbw d (MPa) (2a)

Vc = 2.0 f cbw d (psi) (2b)

ACI Structural Journal/March-April 2015

Table 2Reinforcement characteristics


Steel
bar

Diameter,
mm (in.)

Yield stress fy, Yield strain


MPa (ksi)
y

D-6

5.39 (0.21)

637.7 (92.49)

1.2 103

727.5 (105.5)

D-10

9.69 (0.38)

601.5 (87.24)

1.3 103

700.4 (101.6)

D-12

11.53 (0.45) 617.6 (89.57)

1.8 10

726.4 (105.4)

D-16

15.71 (0.62) 465.6 (67.53)

1.1 10

547.1 (79.35)

3
3

Ultimate strength
fu, MPa (ksi)

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 + 17w u bw d (MPa) (3a)
Mu

V d
Vc = 2.0 f c + 2500w u bw d (psi) (3b)
Mu

where fc is specified compressive strength of concrete; w


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) overestimates the influence of fc and underestimates the influence of w 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))

Vsv =

Av f ys d
sv

(4)

where Av is the area of shear reinforcement; sv is center-tocenter spacing of shear reinforcement measured in direction
parallel to longitudinal tensile reinforcement; and fys is yield
strength of stirrup.
Research6,7 has shown that, with a decrease in the transverse spacing of stirrup legs across the section in wide beams
with substantial flexural reinforcement, the shear behavior is
improved. Other research14 has indicated that the transverse
spacing of web reinforcement shown in ACI318-08 limited
to the lesser of: a) the effective depth d; or b) 600 mm [24
in.] is adequate when the nominal shear stress does not

ACI Structural Journal/March-April 2015

Fig. 3Stress-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))

Vsb =

Avb f yb (sin + cos )d


sb

(5)

where fyb is yield strength of bent-up bar; is the angle


between bent-up bar and longitudinal tensile reinforcement;
Avb is the area of parallel bent-up bars; and sb is center-tocenter 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 propogating 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 independent 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))

VRd , c = [0.18(1 + 200 /d )(100w f ck )1/ 3 ]bw d (MPa) (7a)

VRd , c = [4.96(1 + 7.87 /d )(100w f ck )1/ 3 ]bw d (psi) (7b)

with a minimum shear capacity of


VRd , c min = [0.035(1 + 200 /d )3/ 2

f ck ]bw d (MPa) (8a)

VRd , c min = [0.42(1 + 7.87 /d )3/ 2

f ck ]bw d (psi) (8b)

where characteristic cylinder strength of concrete fck is taken


equivalent to the specified compressive strength of concrete
203

fc. EC2 considers action of a reinforced concrete beam


in shear by the analogous truss with an angle between
22to 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))

VRd , s =

Av
df ys cot (9)
sv

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


Vwd, is given by

Vwd = f yb Asb sin

0.9d (cot + cot )


(10)
sb

where the maximum longitudinal spacing of bent-up bars by


EC2 is limited to 0.6d(1 + cot), 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 reinforcementThe lowest failure load, 401 kN
(90 kip), was for reference Beam WB1 with no shear reinforcement. 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, approximately 2.6 mm (0.1 in.), was less than 50% of the maximum
midpan displacement, 6.9 mm (0.27 in.), for WB2. Independent 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 ACI318-08 and 97% more than EC2.
With a moderate increase in failure load, approximately 26%
greater than WB1, SpecimenWB3 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 displacement of this specimen was 3.1mm (0.12in.), which was
larger than in WB1. Regarding the stirrups used in WB5,
the failure load of 581kN (131kip) for this specimen was
45% higher than the failure load of WB1, but 13% less than
204

Fig. 4Load-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,
5.8mm (0.228in.), was less than that of WB5 and WB2.
Longitudinal flexural reinforcementSpecimen 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 Specimen WB1, it is observed that concentration of the longitudinal flexural reinforcement in the column band increased
ACI Structural Journal/March-April 2015

Table 3Ultimate 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 4Flexural and shear cracking load and displacement


Specimen

Pu,experiment, kN (kip)

Pflexural crack, kN (kip)

Pshear crack, kN (kip)

Pu,experiment Pshear crack,


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 appearance of first shear crack until ultimate load are presented in
Table4. 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 symmetrically 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.
ACI Structural Journal/March-April 2015

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, diagonal 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 WB3with mid-depth horizontal shear
reinforcementalso appeared at a lower load.
Shear reinforcementIn 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
150mm (5.9 in.) longitudinal spacing improved the specimen 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 influenced the first shear crack in WB6 to appear under a 480kN
(108kip) load, which was greater than in WB1 to WB5.
205

Fig. 5Crack 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 indicate that the lack of confining pressure under the loading
plate with a width lesser than the beam causes the crack to
extend.
Longitudinal flexural reinforcementIn Table 1, longitudinal flexural reinforcement ratio w is similar for WB1 and
WB2 (at 1.4%), but approximately two-thirds of the longitudinal 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 (88kip), whereas the first crack for WB4 was at a load
of 430kN (7 kip), approximately a 10% increase. Concentration 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 reinforcement 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 longitudinal 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 concentrated longitudinal flexural reinforcement strain in the middle
bar of the column band, is labelled StWB4. Before 100kN (22
kip) loading, minor digressions from the linear part of loadstrain 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 reinforcement 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 independent bent-up bars as new shear reinforcement, was modeled
by the FE software. In Fig. 7, a summary of the load-midspan 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 numerical modeling can be validated by the experimental results
and used to model other specimens to conduct a parametric
analysis.

Fig. 6Overview on location and variation of middepth


horizontal shear bars and longitudinal flexural bars.

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 independent 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
through failure. The results also indicated that the beam with
ACI Structural Journal/March-April 2015

Fig. 7Load-midspan defection of experimental and FE


analysis for Specimen WB2.
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 Candidate of civil engineeringstructure 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
buildingstructures.
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 engineering 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 reinforced and precast concrete structures, industrialized building systems, and
tallbuildings.

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 Engineering 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 Reinforcement, ACI Structural Journal, V. 86, No. 5, Sept.-Oct. 1989, pp. 507-515.
7. Leonhardt, F., and Walther, R., The Stuttgart Shear Tests 1961, Translation 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 StructuresPart 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 Reinforcement 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 Aggregate 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 Investigation 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 Longitudinal 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. Baant, Z. P., and Kim, J. K., Size Effect in Shear Failure of Longitudinally 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, masters 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, masters
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,
masters research report, Universiti Teknologi Malaysia, Johor, Malaysia,
Dec. 2011, 106 pp.
26. Mrsch, 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

TECHNICAL PAPER

Title No. 112-S19

Effect of Axial Compression on Shear Behavior of HighStrength 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 0.3fc to 0.4fc and compared to eight columns
tested in a previous study with an axial compressive stress of 0.1fc
to 0.2fc. Test results showed that the increase rate of concrete
shear strength tended to decrease with increasing axial compression and reached an upper limit at high axial compression. Most
columns under axial compressive stress of 0.3fc to 0.4fc 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 practical 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 development 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
114,000psi) for main and transverse reinforcement,1 respectively, are now commercially available. However, in shear
design for columns, the current ACI Building Code2 limits
concrete compressive strength fc to 70 MPa (10,000 psi)
(ACI 318-11, Section 11.1.2) due to the lack of test data and
practical experience with fc 70 MPa (10,000 psi). Moreover, 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 0.02fc to 0.81fc, which were reported by Joint
ACI-ASCE Committee 326.8 However, only four specimens
had axial compressive stress higher than 0.15fc, and the fc
values ranged from 14 to 41 MPa (2000 to 6000 psi). For
further assessment of the applicability of the ACI Code equaACI Structural Journal/March-April 2015

tions for Vc to axial compressive stress greater than 0.15fc,


experiments9 were performed using 38 members with fc
ranging from 22 to 27 MPa (3120 to 3950 psi) subjected
to axial compressive stress ranging from zero to 0.7fc. Of
the 38 members tested, 23 were subjected to axial compressive stress exceeding 0.15fc. The ACI code equations for Vc
proved to be conservative for axial compressive stress up to
0.7fc. However, these studies were limited to normal-strength
concrete. Test results10,11 of reinforced concrete beams with
fc 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 fc. Experimental
studies12 of 24 concrete elements with fc ranging from 30 to
87 MPa (4300 to 12,600 psi) under various combinations of
shear and axial compression showed that ACI Code equations 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 shearstrength equation should be limited to 20 MPa (3000 psi).
The equation for shear strength provided by shear reinforcement (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 484to 1454MPa
(70,000to 211,000 psi)3,13,14 showed that shear reinforcement may not be able to develop its yield strength when fyt
700MPa (102,000 psi) and fc < 40 MPa (5800 psi), or when
fyt is very highfor example, fyt = 1454 MPa (211,000psi).
Test results for 42 columns with fyt ranging from 846 to
1447MPa (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 resistshear.
This study tested eight large-scale columns with highstrength 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

209

Table 1Specimen design


Axial compres- Shear reinforcing
Column sion ratio, %
bar spacing, mm
C-1
C-2
C-3

450
30
260

C-4
D-1
D-2
D-3

450
40

D-4

260

Concrete compressive strength Longitudinal reinforcing bar D32 (No. 10) Shear reinforcing bar D13 (No. 4)
fcs, MPa

fc, MPa

70

104.1

100

138.8

70

104.6

100

130.0

70

101.0

100

125.5

70

106.4

100

127.8

fyls, MPa

fyl, MPa

l, %

w, %

fyts, MPa

fyt, MPa

t, %
0.16
0.28

685

735

3.52

1.37

785

862
0.16
0.28

Note: 1 MPa = 145 psi.

Fig. 1Specimen 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 typically carry large axial compression. By using high-strength
concrete and steel, designers can control the column dimensions 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 (60,900psi).
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 material 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

210

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
600x 600 mm (23.62 x 23.62 in.) and a clear height of
1800mm (70.87 in.). The columns were reinforced with
D32 (No. 10) SD685 high-strength deformed bars for longitudinal reinforcement and D13 (No. 4) SD785 high-strength
deformed bars for shear reinforcement, and were cast with
high-strength concrete with two levels of fcs. Table 2 lists the
concrete mixture proportions. Two levels of axial compression ratio30% (Column C series) and 40% (Column D
series)were examined. The axial compression ratio is the
ratio of applied axial compressive load to fcAg. The fc was
obtained from the average of three 150 x 300 mm (6 x12in.)
concrete cylinders. Two levels of shear reinforcement
spacing450 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.
ACI Structural Journal/March-April 2015

Table 2Concrete mixture proportions


Unit weight, kgf/m3
fc, MPa

w/b, %

FS

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 0.1fcAg to 0.2fcAg. Those test
results were used as comparison in the current study to investigate the axial compression effect. Figures 4 and 5 show the
lateral force-displacement relationships for Column C and
D series, respectively. Shear failure occurred before longitudinal reinforcement yielding, as expected in design. It has
been observed that at low axial compression,1 increasing
axial compression from 0.1fcAg to 0.2fcAg enhanced shear
strength. The Vtest at the ultimate condition (peak applied
load), on average, increased by 38% from 0.1fcAg to 0.2fcAg
(Table 3). However, the brittleness increased. The difference
in Vtest between the ultimate and diagonal cracking conditions was, on average, reduced from 32% to 15% when axial
compression was increased from 0.1fcAg to 0.2fcAg (Table 3).
In Series C and D columns with axial compression of 0.3fcAg
and 0.4fcAg, 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 0.2fcAg to 0.3fcAg, and
from 0.3fcAg to 0.4fcAg, respectively (Table3). However,
increasing axial compression had little or no effect on shear
strength. The Vtest at the ultimate condition, on average,
increased by 9% from 0.2fcAg to 0.3fcAg, and on average
by 4% from 0.3fcAg to 0.4fcAg (Table 3). Figures6(a) to
6(d), and 6(e) to 6(h) show cracking and spalling patterns
in Series C and D columns, respectively, at peak applied
load. Cover concrete spalling was minor for Series A and B
columns.1 The extent of spalling increased for Series C and
D columns, particularly at the top and bottom corners.
Figure 7 shows the relationship between the diagonal
crack angle (with respect to column longitudinal axis) and
the drift ratio for each of the Series A, B, C, and D columns.
The diagonal crack angle at a drift level is defined as the
ACI Structural Journal/March-April 2015

Fig. 2Multi-axial testing system (MATS). (Note: 1 mm =


0.0394 in.)

Fig. 3Loading protocol.


average angle of dominant diagonal cracks at that drift
level. Crack patterns in Series A and B columns generally started as flexural cracks and became flexure-shear
cracks with increased drift. Further increases in lateral load
produced web-shear cracks with reduced crack angles. In
Series A columns, diagonal crack angles (average of those
of four columns) were 42 degrees (ranging from 40 to
45degrees) and 33 degrees (ranging from 31 to 34 degrees)
at diagonal cracking condition (first appearance of diagonal
shear cracks) and ultimate condition, respectively. In Series
B columns, the angles were 35 degrees (ranging from 31 to
41 degrees) and 24 degrees (ranging from 20 to 28 degrees),
211

Fig. 4Hysteretic 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 Table3
list the drift at ultimate condition and corresponding shear
reinforcement stress, respectively. Except for ColumnC-3,
all the Series C and D columns reached the ultimate condition 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 rein212

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). Redistribution of internal forces after diagonal cracking was successful
for Column C-3, which failed at a higher load than the diagonal 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 SeriesC
and D columns. The ability of the column to control diagonal 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 compression effect is needed in the minimum shear reinforcement
ACI Structural Journal/March-April 2015

Fig. 5Hysteretic behavior of specimens with 40% axial compression: Specimens (a) D-1; (b) D-2; (c) D-3; and (d) D-4.

Fig. 6Crack 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 ultimate condition at a larger drift than the diagonal cracking
condition (SeriesA and B columns and Column C-3), no
columns showed yielding of shear reinforcement at the ultimate condition. This observation is consistent with the literature,13-15 as mentioned previously. Note that test results also
showed that a higher amount of shear reinforcement delayed

ACI Structural Journal/March-April 2015

column shear failure to a larger drift and, hence, increased


shear reinforcement stress at the ultimate condition.
Table 3 also lists experimental shear strength Vtest, steel
shear strength Vs_test, and concrete shear strength Vc_test under
the diagonal cracking and ultimate conditions. The Vtest is the
load (shear) applied to the column. The Vs_test was calculated
using Eq. (1), where st was determined by the stress-drift

213

Table 3Shear strength contributed by concrete and steel from tests


Diagonal cracking condition*

Ultimate condition

Column

fc, MPa

Drift ratio, %

st, MPa

Vtest, kN

Vs_test, kN

Vc_test, kN

Drift ratio,%

st, 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

1255

0.57

243

1578

150

1428

A-2

99.9

0.33

1286

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 was determined by the measured


crack angle (Fig. 7). The Vc_test was calculated using Eq. (2).
Av st d
cot (1)
s

Vs _ test =

Vc_test = Vtest Vs_test (2)

It can be seen from Table 3 that the amount of shear reinforcement 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 compression. As axial compression increased from 0.1fcAg to 0.2fcAg,
from 0.2fcAg to 0.3fcAg, and from 0.3fcAg to 0.4fcAg, the
Vc_test increased in average by 57%, 24%, and 4%, respectively. Moreover, the difference in Vc_test between the diagonal cracking condition and ultimate condition decreased
with increasing axial compression.
EXAMINATION OF ACI 318 SHEAR EQUATIONS
According to the ACI Code, nominal shear strength Vn can
be obtained from two components: shear strength provided
by concrete, Vc, and shear strength provided by steel reinforcement, Vs. In the Code, Eq. (3) to (6) are used to calculate
Vc of reinforced concrete members under axial compression.
Equation (3) is the simplified equation. Equations (4) and
(5) are used for detailed calculation for Vc but should not be
greater than Eq. (6). The Vs can be calculated using Eq. (7).

214

Fig. 7Drift ratio versus diagonal crack angle.

Nu
Vc = 0.17 1 +

13.8 Ag

Nu
Vc = 2 1 +

2000 Ag

f cbw d (MPa)

(3)

f cbw d (psi)

V d
Vc = 0.16 f c + 17w u bw d (MPa)
Mm

(4)

V d
Vc = 1.9 f c + 2500w u bw d (psi)
Mm

ACI Structural Journal/March-April 2015

Fig. 8Damage 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

Vc = 0.29 f cbw d 1 +
Vc = 3.5 f cbw d 1 +

( 4h d ) (5)
8

0.29 N u
(MPa)
Ag
(6)

Nu
(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 effective 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

Vs =

Av f yt d
s

(7)

Although Eq. (3) to (6) are used to estimate nominal shear


strength in the ACI Code, they were originally derived based
on shear corresponding to diagonal cracking.8 The Vc_test at
two conditionsdiagonal cracking and ultimate shear conditionswere compared to the simplified (Eq. (3)) and detailed
(Eq. (6)) shear-strength equations (Table 4). Note that the
ACI Code limit on concrete compressive strength (fc 70
MPa [10,000 psi]) was not applied when using the above ACI
Code equations. Equation (3) yields conservative prediction
for most columns except Columns D-2 and D-4, but becomes
less conservative as axial compression increases. In the
detailed shear strength calculation, the Mm values (Eq. (5))
are negative for all columns. This means that the moment
effect is small and, hence, Vc is independent of moment. In
this case, Vc is governed by Eq. (6). Comparisons with the
test results show that Eq. (6) is not conservative for Vc at the
diagonal cracking and ultimate conditions for all columns.
When Eq. (7) is used with the actual yield strength of
shear reinforcement (Table 1), it does not yield conservative results for Series A and B columns (the last column in
Table4) because, at peak applied load, stress in shear reinforcement was far from yield (Table 3). Equation (7) cannot
be evaluated in Series C and D columns except for Column
ACI Structural Journal/March-April 2015

Fig. 9Stress of shear reinforcement.


C-3 because the shear reinforcement of these columns failed
to permit redistribution of internal forces, as noted previously. In other words, shear reinforcement did not function
effectively in these columns. Equation (7) obtained a conservative estimate for Column C-3 even though shear reinforcement stress was below yield stress (Table 3). The estimate
was conservative because actual shear crack angles were
much smaller than 45 degrees, as assumed in Eq. (7), leading
to a larger actual steel shear strength. As noted previously,
shear reinforcement stress can be further increased at peak
applied load by increasing the amount of shear reinforcement. This should be properly considered when determining
the limit value of shear reinforcement stress to be used in
Eq.(7). Further research is needed to address this issue.
Equation (6) originated from Eq. (8), which was derived
based on the principal stress equation at the point of diagonal
cracking. The effects of axial load and shear were considered. The effect of moment was assumed to be small and,
hence, was neglected.8

Vc =

ft
Nu
bw d 1 +
(8)
F2
f t bw d

215

Table 4Ratio of test results to shear-strength prediction using ACI 318-11 without strength limitation
Diagonal cracking shear strength
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 ft/F2 was set to 0.29fc (MPa) or


3.5fc (psi). The ft was assumed to be 0.62fc (MPa) or
7.5fc (psi). Thus, Eq. (8) becomes Eq. (9).

Vc = 0.29 f cbw d 1 +
Vc = 3.5 f cbw d 1 +

1.6 N u
f cbw d

0.133N u
f cbw d

(MPa)

(9)

(psi)

For simplicity, 1.6/fc MPa (0.133/fc psi) was replaced


by a constant value of 0.29 MPa (0.002 psi), which corresponds to an fc of approximately 30 MPa (4400 psi). Moreover, bwd was approximated by Ag. With these two changes,
Eq. (9) becomes Eq. (6). The simplification made in Eq. (6)
by assuming fc is equal to 30 MPa (4400 psi) leads to an
overestimation when fc exceeds the assumed value. In the
case of fc = 100 MPa (14,500 psi) and Nu/fcAg = 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 literature,1,13,15,17-23 a test database of Vc_test at diagonal cracking
from 77 shear-critical high-strength columns was established (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,
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.

216

Ultimate shear capacity

of Vc_test to Vc predicted by Eq. (3), (6), and (9), respectively. Whereas Eq. (3) yields conservative predictions for
most columns, Eq. (6) yields nonconservative predictions
for 48columns. 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 fc in Fig. 10. Only two relationships corresponding
to fc of 100 and 130 MPa (14,500 and 18,800 psi), respectively, 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 compression 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 0.3fcAg to
0.4fcAg. The Sakaguchi13 data show an upper limit of 0.4fcAg
to 0.5fcAg. The Maruta15 data show that the increase in Vc
substantially slows when axial compression is increased
from 0.3fcAg to 0.6fcAg. 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, ft, caused by the
presence of compressive stress acting in the other principal
direction.24-28 The principal compressive stress increases
with increasing axial compression of the column.
ACI Structural Journal/March-April 2015

Fig. 10Relationship between test data and Vc predictions.


t = ft

(10)

The ft is set equal to 0.5fc (MPa) or 6fc (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 . The c in Eq. (11) is limited
to 0.6fc due to limited biaxial test data for c larger than
0.6fc and because 0.6fc is the highest axial compressive
stress observed in the 77 columns.

= 1 0.85 c for 0 c 0.6 (11)


f
f c

The c at diagonal cracking is calculated using the principal


stress equation (Eq. (12)) with applied shear equal to diagonal cracking shear.
2


Fig. 11Tensile strength degradation of high-strength
concrete.
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
to compressive stress in the other principal direction can be
described by the following equation.

ACI Structural Journal/March-April 2015

N V
N
c = u + u + c (12)
2 Ag
2 Ag bw d

After defining the reduction factor, Vc can be calculated


using Eq. (13), which is based on Eq. (9) with reduced principal tensile strength t (Eq. (10)) substituted for ft.

Vc = 0.29 f cbw d 1 +
Vc = 3.5 f cbw d 1 +

2 Nu
f cbw d
0.17 N u

f cbw d

(MPa)

(13)

(psi)

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 c is only 0.005 to 0.0017fc
for axial compression of 0.1fcAg to 0.6fcAg, respectively.
For simplicity, Nu/Ag may be used instead of c 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 0.1fcAg and
0.6fcAg, respectively. The 15th column in Table A1 shows
predictions obtained by Eq. (13) with 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
fc 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
0.6 (14)
= 1 0.85
for 0
f cAg
f cAg

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 0.29fc (MPa) or 3.5fc (psi)
replaced by 0.25fc (MPa) or 3.0fc (psi) and with an upper
limit of 0.2fcAg on Nu. The 16th column in TableA1 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 0.2fcAg 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).

Vc = 0.25 f cbw d 1 +
Vc = 3.0 f cbw d 1 +

1.6 N u
f cbw d

0.133 N u
f cbw d

(MPa)

(15)

(psi)

Nu shall not be taken greater than 0.2fcAg.


Axial compression reduces principal tensile stress and,
hence, increases Vc. This is the mechanism of axial compression 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 mechanisms increases Vc at low axial compression but the increase
slows down with increasing axial compression and eventually 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 0.1fcAg
to 0.4fcAg, 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 conditions. For most columns under high axial compression
(0.3fcAg and 0.4fcAg), 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 (0.1fcAg
and 0.2fcAg). 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 condition. 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 reinforcement. 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 0.3fcAg to 0.4fcAg.
Test results reported by other researchers have shown an
upper limit from 0.4fcAg to 0.5fcAg.
3. The ACI simplified Vc equation, Eq. (3), yields conservative 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 engineering 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 President of the Taiwan ChapterACI. 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.

Ag
Av
a
bw
d
F2

=
=
=
=
=
=

fc'
fcs'
ft'
fy
fyl
fyls
fyt
fyts
h
M m
Mu
Nu
s
Vc
Vc_test
Vn
Vs
Vs_test
Vtest
Vu
a
q
rl
rt
rw
sc
sst
st

=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=

NOTATION

gross area of concrete cross-section


total cross-sectional area of shear reinforcement
shear span
effective web width of member cross section
effective depth of member cross section
ratio of shear stress at diagonal cracking point to average shear
stress on effective cross section
concrete compressive strength
specified concrete compressive strength
concrete principal tensile strength
yield strength of steel
yield strength of longitudinal reinforcement
specified yield strength of longitudinal reinforcement
yield strength of shear reinforcement
specified yield strength of shear reinforcement
overall height of member cross section
applied moment modified to consider effect of axial compression
applied moment
applied axial load (positive in compression)
spacing of shear reinforcement
nominal shear strength provided by concrete
experimental shear strength provided by concrete
nominal shear strength
nominal shear strength provided by shear reinforcement
experimental shear strength provided by shear reinforcement
experimental shear strength
applied shear
reduction factor
shear crack angle to column longitudinal axis
longitudinal reinforcement ratio
shear reinforcement ratio
longitudinal tension reinforcement ratio
principal compressive stress
shear reinforcement stress
concrete principal tensile strength reduced by principal compressive 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 Engineering 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 Proceedings, 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 Proceedings, 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 Proceedings, 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 Procedures 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 International 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 Symposium 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 Performance of RC Column Using 200 N/mm2 Concrete, Summaries of Technical 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 Structures, 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 Institute 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 B = 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. Johns, 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 Engineering, 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 innovative 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 deformation 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 reinforcement 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.
Keywords: beams; high-strength concrete; integrated floor slab system;
prestressed concrete; shear force; ultimate strength; web openings.

INTRODUCTION
The majority of existing buildings have monofunctional
properties, characterized by inflexible floor layouts that
complicate changes in use (for example, from residential
to office building or vice versa) or they are incompatible to
current architectural requirements or new technical innovations. This often leads to demolishing or substantially
restructuring such buildings long before they reach their
economic lifetime. To exploit the buildings full economic
lifetime, adaptive structural systems with a high degree
of flexibility should be developed. Wide-spanning floor
slab systems with integrated building services can make a
compromising contribution.1-5
These integrated floor slabs provide wide spans for high
flexibility and adaptability to allow for conversions of use
without significant modification of the building structure.
Beside static aspects, the choice of the floor slab structure has
an impact on numerous building properties. Floor slabs not
only fulfill load-bearing and bracing functions, they create
the separation between adjoining functional units and, thus,
influence the planimetry, building services, physical properties of the building, and economic and ecological impact
of the structure. Hence, the profile of requirements6 for integrated floor slabs includes issues from the fields of structural
engineering, architecture, manufacturing, fire protection,
building physics, dismantling, and recycling.
A general approach to design-integrated slab systems
is to break up the conventional additive ceiling assembly
ACI Structural Journal/March-April 2015

Fig. 1Prestressed floor slab system for integration of


building services.
(flooring/supporting structure/building services and installations/suspended ceiling) and to dissolve the compact cross
sections of conventional floor slabs into wide-span, slender
multi-web structures. In Fig. 1, an example of an integrated
floor slab system with a multi-web structure composed of
prestressed concrete beams is shown. The precast floor slab
elements are designed for spans up to 16 m (630 in.) and
a service load of 5 kN/m2 (0.725 psi). The provided space
between the webs is used for the integration of building
services. The dimensions of the cross section are detailed
in Fig. 1. Openings in the web of the girders allow for flexible arrangements in all directions. Ease of access to the
installation floor and a convenient installation and maintenance of the components from above is achieved by the
use of removable cover panels placed on top flanges of the
concrete beams. Revision openings allow for maintenance
and minor modifications. To make use of innovative floorslab concepts, the technical feasibility needs to be proven.
Therefore, the load-bearing and deformation behavior of the
integrated concrete slab5,7 was investigated in a comprehensive experimental study that was aimed at systematically
exploring the impact of large web openings in prestressed
concrete beams. The results of these experimental investigations are presented in this paper.
ACI Structural Journal, V. 112, No. 2, March-April 2015.
MS No. S-2014-188.R, doi: 10.14359/51687302, 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 authors
closure, if any, will be published ten months from this journals date if the discussion
is received within four months of the papers print publication.

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 theoretical 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 weakened by large openings need to carry increased shear forces.
This leads to the formation of a local load-carrying mechanism 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, secondorder 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 compression-shear failure may occur. Contrary to the Failure Mode1
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 calculate 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 experimental results was found.

Vt / b =

At / 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 Schfer10 recommend calculating the distribution of shear force according to the ratio of bending stiffnesses, It/b, of top and bottom chords. Hereafter, the proportion of shear force in the chords results from a linear-elastic
calculation of a framework system, neglecting shear deformation of the beams (Eq. (2)).
Vt / b =

It /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 stiffness of the chords led to better results.
Vt = V Vb (3)

Vt =

I i ,t
I i ,t + I i ,b ( cr )

Vb =

I i ,b
I i ,t + I i ,b

V (5)

V (6)

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 completion 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.

VOm , cr =

ft

N P ,b
Ab

lOm
Vb lO 1
z A + V 2 W
G b
b

(7)

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.

b,eff = b Ib = [15(l,b + l,b) 0.25 + (nb 0.5)2] Ib (8)

V (4)

t,eff = t It = [15(l,t + l,t) + 0.4 (nt 0.5)2] It (9)

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 fulldepth crack appears at the bottom (tensile) chord, the entire
shear has to be carried by the top (compression) chord.

Vb =

Ab I b
Ab I b + At I t

ACI Structural Journal/March-April 2015

nb / t =

Nb / t
(10)
Ab / t f c

where l,t/b (l,t/b) are the flexural tensile (compression) reinforcement 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 stiffness. Beside these complex calculations, simple assumptions
for the distribution of the shear force may be taken. Leonhardt18 suggests the following distribution at failure

223

Fig. 3Dimensions 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-Tshaped concrete beams with prestressed tension chord was
investigated. Based on the initial beam Test DE-1.1, individual 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 concretes 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 (9.8in.), 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
(0.218psi) and a live load of 5.0 kN/m2 (0.725psi) 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 longitudinal reinforcement in the compression chord consisted
of four bars 20 mm (0.79 in.) in diameter. Bars 10 mm
(0.39in.) in diameter were used as longitudinal reinforcement in the tensile chord. Closed stirrups 6 mm (0.24 in.)
in diameter at 160 mm (6.3in.) 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
0.366in.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 1638MPa
(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 111MPa
(9427 and 16,099 psi). Test cylinders (150 x 300mm
[5.9x11.8in.]), cubes (150 x 150 mm [5.9 x 5.9 in.]), and
prisms (700x150x100 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 Youngs 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 1Material properties of reinforcement

Test setup and instrumentation


Two experiments were performed on each specimen.
Interference between the tests was avoided by a sufficiently
large distance between the openings and adequate bedding
conditions. The span of the first three-point bending tests
was 8 m (315 in.). After the failure of each test, the supports
were moved and a second test was carried out, but with a
reduced span of 5500mm (216 in.). To prevent bending
failure, a shear slenderness of a/d = 4.76 was chosen.
The beam details are presented in Fig. 5. The beams were
supported on steel plates with slide bearings. The load was
transferred from the hydraulic jack to the beam through a
steel plate. All beams were monotonically loaded up to the
predicted service load. Afterward, 50 load cycles between 35
and 130% of the service load were executed. To predict the

Bar size,
mm (in.)

A0, mm
(in.2)

fy, MPa
(ksi)

ft, MPa
(ksi)

6 (0.236)

28 (1.10)

605
(87.75)

653
(94.71)

17.3

201.3
(29,196)

10
(0.394)*

78 (3.07)

499
(72.37)

626
(90.79)

19.0

201.3
(29,196)

10
(0.394)

77 (3.03)

610
(88.47)

704
(102.11)

20.0

206.0
(29,878)

20 (0.787)

307
(12.09)

544
(78.90)

645
(93.55)

23.9

200.4
(29,066)

Stirrups.

Longitudinal reinforcement.

A10, %

Es, GPa
(ksi)

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 Youngs modulus.

Table 2Details 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 Youngs 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. 4Reinforcement details of Test DE-1.1.


ACI Structural Journal/March-April 2015

225

Fig. 5Testing procedure, beam spans, and loading.


reduced distance between the opening and support (DE-2.1),
the bending failure in the opening corner occurred in combination with a shear failure in the concrete joint interface
between web and bottom chord. The beam tests with openings 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 ultimate shear forces and main test parameters are summarized
in Table 2.

Fig. 6Test 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 electrical resistance strain gauges. To determine the curvature
at each edge of the opening, the steel strain of the longitudinal 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 instrumentation 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

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 specimen (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 approximately 30% of the predicted load-bearing capacity. Consequently, 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 loadbearing capacities. A sufficient ductility and good loadbearing 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 0.2mm (0.004 to 0.008 in.). The angle of the
ACI Structural Journal/March-April 2015

Fig. 7Arrangement of strain measurements: (a) concrete strains; and (b) reinforcement steel strains.

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

Fig. 9Failure pattern of test specimens.

ACI Structural Journal/March-April 2015

227

Fig. 10Typical crack formation: Test: (a) DE-2.1; (b) DE-2.2; and (c) DE-3.2.

Fig. 11Strain 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 inclination 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 areillustrated.
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 allocated 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. 12Curvature 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 m. In Fig. 12, the
curvature m 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 reinforcement is assumed to hang up and anchor the shear force
carried by the compression chord of the beam. In the literature, 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 reinforcement 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
of the opening, alternative load transfer (arch mechanism)
can be eliminated. Thus, at ultimate load, an appreciable
part of the shear force was carried by the bottom chord. The
theoretical load-bearing capacity of the reinforcement at the
edges of the opening was not reached for Tests DE-2.2 and
DE-3.2 with a higher reinforcement ratio. Considering a sole
transfer of the shear force by the top chord, the load-bearing
capacity of the reinforcement was not attained. Due to the
higher load level, higher stress of the reinforcement at the
opening edges was reached in tests with higher concrete
strength. Comparing the measured strains of reinforcement
ACI Structural Journal/March-April 2015

Fig. 13Loading of vertical reinforcement at opening edges.

Fig. 14Strain of vertical reinforcement at opening edges.


in Fig. 14, these findings are confirmed. Using two bars of
10 mm (0.39 in.) diameter, an ultimate strain of approximately 0.003 was reached in the tests so that the stresses in
the vertical reinforcement reached the yield strength (499
N/mm2 [72.37 ksi] according to Table 1). The strains in the
vertical reinforcement of Test DE-2.2 with a higher reinforcement ratio stayed significantly below this value. Here,
only an ultimate strain of 0.0013 was reached, which corresponds to a tensile stress of 261 N/mm2 (37.9ksi) in the
vertical reinforcement bars.
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 loadbearing capacity, no differences between the tests could be
observed. After that, the deflection of Test DE-2.1 disproportionately 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 reinforcement 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
[15,954psi]) 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. 15Exemplary determination of load-bearing capacity


for Test DE-1.1.

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 theoretical 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 calculated 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 TestDE-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 Rm of a reinforced
interface according to Eurocode 223 to the shear stress in
the interface caused by loading E at ultimate limit state.
The shear resistance Rm 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 experimental 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.

Fig. 16Comparison of acting and resisting shear stress in interface.


230

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 highstrength 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 interface 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 was determined from the measured principal strains. Knowing the
shear modulus, the shear stress was calculated considering
Hookes law (Eq. (13)), whereat the impact of global loading
on the shear modulus was considered by Eq. (14) with a
Poissons ratio of 0.2

= G =

Ec
(13)
2(1 + )

M global = x Az Ec =

M global (1 2 )
Az (e x + e x )

(14)

The acting shear force in the top chord results from the integration of the shear stress over the cross section. As reference, 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 reinforcements
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 determined 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.
ACI Structural Journal/March-April 2015

Fig. 17Distribution of applied shear force carried by


topchord.
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 independent 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
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 fr 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 International, V. 19, No. 4, 2009, pp. 404-409. doi: 10.2749/101686609789846948
2. Hegger, J.; Claen, 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 StructuresTall Buildings,
Bridges and Infrastructure, 17th Congress of IABSE, Sept. 17-19, Chicago,
IL, 2008, pp. 230-231.
5. Hegger, J.; Dreen, T.; and Schiel, 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 Spannbetontrgern mit groen Stegffnungen, 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 Proceedings, 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 Schfer, K., Bemessen von Stahlbetonbalken
und -wandscheiben mit ffnungen, Deutscher Ausschuss fr 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 Engineering, 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 groen
ffnungen, Deutscher Ausschuss fr 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 Transverse 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 Mnnig, E., Vorlesungen ber 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 Stahlbetonbau, V. 108, No. 7, July 2013, pp. 462-474.
20. Hegger, J.; Claen, M.; Schaumann, P.; Sothmann, J.; Feldmann, M.;
and Dring, 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 Verbunddbelleisten in schlanken Betongurten, Beton- und Stahlbetonbau, V. 119, No. 12, pp. 882-894.
22. Classen, M., and Hegger, J., Verankerungsverhalten von Verbunddbelleisten 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 groen
Stegffnungen auf das Trag-und Verformungsverhalten von Stahlbetontrgern, 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 Jos R. Mart-Vargas
Professor, ICITECH, Institute of Concrete Science and Technology, Universitat Politcnica de Valncia, 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 strandfree end slip,26 which were available prior to the submission
date. Furthermore, the authors considered one reference by
Balzs14 in their comparisons (Table 3), whereas another
related, complementary study was ignored.27 Moreover,
ACI318-081 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
testmethod.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 ( = 2)
according to Guyons theory,12 whereas Specimens 4, 9, 10,
and 16 displayed a linear bond stress distribution that, in
turn, coincides with the other bound case ( = 3) according to
ACI Structural Journal/March-April 2015

Guyons 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
value ( = 2.67 and = 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 dp is used instead of actual
perimeter (4/3)dp.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. Mart-Vargas, J. R.; Arbelez, C. A.; Serna-Ros, P.; Navarro-Gregori,


J.; and Pallars-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. Mart-Vargas, J. R.; Arbelez, C. A.; Serna-Ros, P.; and CastroBugallo, C., Reliability of Transfer Length Estimation from Strand End
Slip, ACI Structural Journal, V. 104, No. 4, July-Aug. 2007, pp. 487-494.
27. Balzs, 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. 549559.
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. Garca-Taengua, E.; Mart-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. Mart-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. Mart-Vargas, J. R.; Garca-Taengua, E.; Caro, L. A.; and SernaRos,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 Mart-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 structuresPart 1-1: General
Rules and Rules for Buildings. (EN 1992-1-1:2004:E), Comit Europen
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. Mart-Vargas, J. R.; Hale, W. M.; Garca-Taengua, E.; and SernaRos,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. 13Strand strain distribution of Specimen C. (Note:


1mm = 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. Mart-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 discussers 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 Balzs14 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, Comit Euro-International du Bton


(CEB), Bulletin dInformation 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.
Part3: Transmission Length of Prestressing Strand, Stevin Report 25.595-3, 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


behavior of fiber-reinforced polymer (FRP) in fire situations,
but it also presents some puzzles. It is unfortunate that there
apparently were no tests of the reinforcing bars because
both a measured yield stress and a measured stress-strain
curve would have been helpful in understanding some of the
results. The 80 ksi (551 MPa) yield stress implied from the
strain measurements for Beam B-U-O/A seems high but is
plausible for a small bar.
However, even this high stress does not explain the
behavior of this beam. At 10 kip (44.5 kN) force (the reported
yield force), the applied moment is approximately 31.3 ft-kip
(42.4 kN-m), including the small dead-load moment. However,
the computed value of Mn is approximately 25.4 ft-kip
(34.4 kN-m) when fc = 7.3 ksi (50 MPa) and fy = 80 ksi
(551 MPa). At the maximum force of 12.92 kip (57.5 kN),
the applied moment is approximately 40.1 ft-kip (54.4 kN-m).
The difference between the observed yield moment and the
computed nominal capacity seems too large, as does the difference between the observed yield and maximum applied loads.
The computed strain at Mn is less than 0.02, which implies
some strain hardening but probably not 29%.
A possible source of the differences is restrained elongation. The longitudinal restraint applied at the bottom surface
of the beam that is able to increase the moment capacity
from 25 to 40 ft-kip is not too large. The support system is
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 fractured 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 reaction 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 reduction 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.
ACI Structural Journal/March-April 2015

AUTHORS CLOSURE
The authors thank the discusser for his interest in the paper
and the published work. The discussion focuses on three
main points: 1) the moment capacity of the beam; 2) the
rupture of the steel in Beam B-DH-C/F; and 3) the removal
of the entire load in Beam B-CF-G/F. The authors will try to
address all three points.
For the first point, it should be noted that due to the
nature of the fire testing, a special support system was used
throughout the entire experiment, even for beams tested at
ambient temperature. As shown in Fig. 13, the beam was
resting on two 12 in. (305 mm) wide steel plates at its ends.
Therefore, for the purpose of moment calculations, the
effective span should be taken as 10 ft (3.05 m) as the beam
rotated around the interior edges of the support plates. By
considering an effective span of 10 ft (305 mm), the moment
due to dead load will be equal to 0.935 ft-kip (1.27 kN-m).
At steel strain of 3000 , the depth of the NA is 2.26 in.
(57.4 mm) and the theoretical yield moment is 25.72 ft-kip
(34.87 kN-m). By subtracting the moment due to dead load
from the yield moment, the moment due to the concentrated
load becomes 24.79 ft-kip (33.61 kN-m), which is equivalent to a moment due to a concentrated load of 9.91 kip
(44 kN). It is worth noting that the beam was resting freely
on the supports and no significant longitudinal restraint
wasprovided.

Fig. 13Beam B-U-O/A under three-point loading setup.


235

Fig. 14Close-up view of Beam B-U-O/F showing rupture


of bottom steel reinforcement bars after exposure to fire test.

Fig. 15Beam B-DH-GB/F2 split in half after exposure


tofire.
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 experienced 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 1890F (1010 to 1032C), 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 1832F (1000C), the yield strength of steel

236

Fig. 16Beam 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 strengthening 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 significant 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 represented 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 Jos R. Mart-Vargas
Professor, ICITECH, Institute of Concrete Science and Technology Universitat Politcnica de Valncia, 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


( = 2) according to Guyons theory,28 whereas a recent
study27 proposes a linear bond stress distribution which, in
turn, coincides with the other bound case ( = 3) according
to Guyons theory.28 An intermediate case ( = 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 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. Mart-Vargas, J. R.; Arbelez, C. A.; Serna-Ros, P.; NavarroGregori,J.; and Pallars-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. Mart-Vargas, J. R.; Serna-Ros, P.; Navarro-Gregori, J.; and
Pallars,L., Bond of 13 mm Prestressing Steel Strands in Pretensioned
Concrete Members, Engineering Structures, V. 41, 2012, pp. 403-412.
26. Mart-Vargas, J. R.; Arbelez, C. A.; Serna-Ros, P.; and CastroBugallo, 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 Mart-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. Mart-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. Garca-Taengua, E.; Mart-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. Mart-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. Mart-Vargas, J. R.; Garca-Taengua, E.; Caro, L. A.; and SernaRos,P., Measuring Specific Parameters in Pretensioned Concrete Members
Using a Single Testing Technique, Measurement, V. 49, 2014, pp. 421-432.
37. Mart-Vargas, J. R.; Hale, W. M.; Garca-Taengua, E.; and SernaRos,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
themember.
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 (in Guyons theory) of 2 to 3 depending
on the bond stress distribution (that is, constant or linear)
along the transfer length. Intermediate values of 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 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 RCAnamely, 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.

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
discussers 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

Fig. 8Crushing 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

239

Fig. 9Crushing 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 beams 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 aggregates 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.
aniewska-Piekarczyk, Beata
Silesian University of Technology
Gliwice, Poland
Aamidala, Hari Shankar
Parsons Brinckerhoff
Herndon, VA
Abaza, Osama
University of Alaska Anchorage
Anchorage, AK
Abbas, Abdelgadir
Carleton University
Ottawa, ON, Canada
Abbas, Safeer
Western University
London, ON, Canada
Abbasnia, Reza
University of Science and Technology
Tehran, Islamic Republic of Iran
Abdalla, Hany
College of Technological Studies
Shuwaikh, Kuwait
Abdelaziz, Gamal
Benha University
Cairo, Egypt
Abdelaziz, Magdy
Fayoum, Egypt
Abdel-Fattah, Hisham
University of Sharjah
Sharjah, United Arab Emirates
Abdelgader, Hakim
Tripoli University
Tripoli, Libyan Arab Jamahiriya
Abdelrahman, Amr
Heliopolis, Egypt
Abdulla, Nwzad
University of Salahaddin
Erbil, Iraq
Abeyruwan, Helarisi
University of Peradeniya
Peradeniya, Sri Lanka
Abouhussien, Ahmed
Memorial University of Newfoundland
St. Johns, NL, Canada
Aboutaha, Riyad
Syracuse University
Syracuse, NY
Abou-Zeid, Mohamed
American University in Cairo
Cairo, Egypt
Abruzzo, John
Thornton Tomasetti
San Francisco, CA
Abu Yosef, Ali
WDP and Associates
Austin, TX

ACI Structural Journal/March-April 2015

Achillopoulou, Dimitra
Democritus University of Thrace
Xanthi, Greece
Acun, Bora
University of Houston
Houston, TX
Adamczewski, Grzegorz
Warsaw University of Technology
Warsaw, Poland
Adhikary, Bimal
Austin, TX
Afif, Rahma
Damascus University
Damascus, Syrian Arab Republic
Aggelis, Dimitrios
University of Ioannina
Ioannina, Greece
Agustiningtyas, Rudi
Ministry of Public Works
Bandung, Indonesia
Ahmad, Shamsad
King Fahd University of Petroleum & Minerals
Dhahran, Saudi Arabia
Ahmadi, Jamal
University of Science of Technology
Tehran, Islamic Republic of Iran
Ahmed, Ehab
University of Sherbrooke
Sherbrooke, QC, Canada
Ahmed, Zeyad
Michigan Technological University
Houghton, MI
Aire, Carlos
National Autonomous University of Mexico
Mexico, DF, Mexico
Akakin, Tumer
Turkish Ready Mixed Concrete Association
Istanbul, Turkey
Akalin, Ozlem
Plustechno Ltd
Istanbul, Turkey
Akbari, Reza
University of Tehran
Tehran, Islamic Republic of Iran
Akbarnezhad, Ali
The University of New South Wales
Sydney, New South Wales, Australia
Akcaoglu, Tulin
Eastern Mediterranean University
Magusa, Turkey
Akcay, Burcu
Kocaeli University
Kocaeli, Turkey
Aknay, Yuksel
Iron and Steel Institution/Material Researcher Center
Karbuk, Turkey

241

REVIEWERS IN 2014
Akiyama, Mitsuyoshi
Waseda University
Tokyo, Japan
Al-alaily, Hossam
Memorial University of Newfoundland
St. Johns, NL, Canada
Alam, A. K. M. Jahangir
Bangladesh University of Engineering and Technology (BUET)
Dhaka, Bangladesh
Alam, M. Shahria
The University of British Columbia
Kelowna, BC, Canada
Alam, Mahbub
Stamford University Bangladesh
Dhaka, Bangladesh
Al-Attar, Tareq
University of Technology
Baghdad, Iraq
Al-Azzawi, Adel
Nahrain University
Baghdad, Iraq
Albahttiti, Mohammed
Kansas State University
Manhattan, KS
Albuquerque, Albria
Federal Center of Technological Education of Mato Grosso
Cuiab, Mato Grosso, Brazil
Alcocer, Sergio
Institute of Engineering, UNAM
Mexico City, DF, Mexico
Aldea, Corina-Maria
St. Catharines, ON, Canada
Alexander, Mark
University of Cape Town
Cape Town, South Africa
Al-Hadithi, Abdulkader
University of Anbar
Ramadi, Al-anbar, Iraq
Al-Harthy, Ali
Sultan Qaboos University
Al-Khaudh, Oman
Ali, Samia
University of Engineering and Technology, Lahore
Lahore, Punjab, Pakistan
Al-Karkhi, Hassan
Al-Mustansiriya University College of Engineering
Baghdad, Iraq
Alkhairi, Fadi
Arabtech Jardaneh
Amman, Jordan
Allahdadi, Hamidreza
Bangalore, India
Allena, Srinivas
Washington State University Tri-Cities
Richland, WA
Almaral, Jorge
Universidad Autnoma de Sinaloa
Los Mochis, Sinaloa, Mexico
Almeida, Joao
IST
Lisbon, Portugal

242

Al-Mufti, Rafal
Surrey University
Guildford, UK
Almuhsin, Bayrak
University of Technology
Karrada, Baghdad, Iraq
Alsiwat, Jaber
Saudi Consulting Services
Riyadh, Saudi Arabia
Altuniik, Ahmet Can
Karadeniz Technical University
Trabzon, Turkey
Aly, Aly Mousaad
Louisiana State University
Baton Rouge, LA
Amani Dashlejeh, Asghar
Tarbiat Modares University
Tehran, Islamic Republic of Iran
Amani Dashlejeh, Jafar
Bauhaus University of Weimar
Weimar, Germany
Amir, Sana
Delft University of Technology
Delft, South Holland, the Netherlands
Andersson, Ronny
Hollviken, Sweden
Andrade, Jairo
Chatolic University of Rio Grande do Sul
Porto Alegre, RS, Brazil
Andriolo, Francisco
Andriolo Ito Engenharia S/C Ltda
So Carlos, So Paulo, Brazil
Angel, Nelson
Universidad de los Andes
Bogot, Cundinamarca, Colombia
Ansari, Abdul Aziz
Quaid-e-Awam Engineering University
Nawabshah, Sindh, Pakistan
Aoki, Yukari
University of Technology Sydney
Ultimo, New South Wales, Australia
Aragn, Sergio
Holcim (Costa Rica)
San Rafael, Alajuela, Costa Rica
Araujo, Daniel
Federal University of Gois
Goiania, Brazil
Aravinthan, Thiru
University of Southern Queensland
Toowoomba, Queensland, Australia
Arisoy, Bengi
Ege University
Izmir, Turkey
Aristizabal-Ochoa, Jose
National University
Medellin, Antioquia, Colombia
Arockiasamy, Madasamy
Florida Atlantic University
Boca Raton, FL

ACI Structural Journal/March-April 2015


Asaad, Diler
Gaziantep University
Gaziantep, Turkey
Asamoto, Shingo
Saitama University
Saitama, Saitama, Japan
Ashrafy, Mohammad
Islamic Azad University Arak Branch
Kermanshah, Kermanshah, Islamic Republic of Iran
Aslani, Farhad
University of Technology Sydney
Sydney, New South Wales, Australia
Assaad, Joseph
Notre Dame University
Beirut, Lebanon
Atamturktur, Sez
Clemson University
Clemson, SC
Athanasopoulou, Adamantia
Metropolitan College
Xalandri, Attiki, Greece
Avendano, Alejandro
Technological University of Panama
Doral, FL
Aviram, Ady
Simpson Gumpertz & Heger, Inc.
San Francisco, CA
Awida, Tarek
KEO International Consultants
Kuwait
Awwad, Elie
Lebanese University, Branch II
Mount Lebanon, Lebanon
Ayano, Toshiki
Okayama University
Okayama, Japan
Aydin, Abdulkadir Cuneyt
Ataturk University
Erzurum, Turkey
Aydin, Ertug
European University of Lefke
Nicosia, Turkey
Aykac, Sabahattin
Gazi University
Ankara, Turkey
Azad, Abul
King Fahd University of Petroleum & Minerals
Dhahran, Saudi Arabia
Azari, Hoda
The University of Texas at El Paso
El Paso, TX
Aziz, Omar
University of Salahaddin
Erbil, Iraq
Babafemi, Adewumi
Obafemi Awolowo University
Ile-ife, Osun, Nigeria
Bacinskas, Darius
Vilnius Gediminas Technical University
Vilnius, Lithuania

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Bae, Sungjin
Bechtel Corporation
Frederick, MD
Bagge, Niklas
Lule University of Technology
Lule, Sweden
Bai, Shaoliang
Chongqing University
Chonqqing, China
Bai, Yongtao
Kyoto University
Kyoto, Japan
Balakumaran, Soundar
Virginia Center for Transportation Innovation and Research
Charlottesville, VA
Balilaj, Mentor
Polytechnic University of Tirana
Tirana, Albania
Balouch, Sana
University of Dundee
Dundee, UK
Banibayat, Pouya
ARUP
New York, NY
Bani, Davor
Civil Engineering Institute of Croatia
Zagreb, Croatia
Baran, Eray
Middle East Technical University
Ankara, Turkey
Barboza, Aline
Universidade Federal de Alagoas
Maceio, Alagoas, Brazil
Barii, Ivana
Osijek, Croatia
Barragan, Bryan
BASF Construction Chemicals
Treviso, Treviso, Italy
Barroso de Aguiar, Jose
University of Minho
Guimaraes, Portugal
Bartos, Peter
University of Paisley
Paisley, UK
Basava, Vamsi
Malla Reddy Engineering College
Hyderabad, Andhra Pradesh, India
Bashandy, Alaa
Menoufiya University
Shibin El-Kom, Menoufiya, Egypt
Batson, Gordon
Clarkson University
Potsdam, NY
Baty, James
Concrete Foundations Assoc
Mount Vernon, IA
Bayraktar, Alemdar
Trabzon, Turkey

243

REVIEWERS IN 2014
Bayuaji, Ridho
Institut Teknologi epuluh Nopember
Surabaya, East Java, Indonesia
Bebawy, Mena
Lawrence Technological Univeristy
Southfield, MI
Becq-Giraudon, Emilie
Chicago Department of Transportation
Chicago, IL
Beddar, Miloud
Msila University
Msila, Algeria
Bediako, Mark
CSIRBuilding and Road Research Institute
Kumasi, Ashanti, Ghana
Bedirhanoglu, Idris
Dicle University
Diyarbakir, Turkey
Beglarigale, Ahsanollah
Dokuz Eylul University
Izmir, Izmir, Turkey
Behnam, Hamdolah
Hong Kong University of Science and Technology
Hong Kong, China
Behnoud, Ali
Iran University of Science and Tech
Tehran, Islamic Republic of Iran
Belagraa, Larbi
Bordj Bou Arreridj, University Center
Bordj Bou Arreridj, Algeria
Belkowitz, Jon
Stevens Institute of Technology
Freehold, NJ
Belleri, Andrea
University of Bergamo
Dalmine, Italy
Benliang, Liang
Shanghai, China
Bennett, Richard
The University of Tennessee
Knoxville, TN
Bernard, Erik Stefan
TSE P/L
Penrith, Australia
Berry, Michael
Montana State University
Bozeman, MT
Beygi, Morteza
Mazandaran University
Babol - Mazandaran, Islamic Republic of Iran
Bhangal, Malkit
Thapar University
Patiala, Punjab, India
Bharati, Raj
National Institute of Technology Calicut
Calicut, Kerala, India
Bhattacharjee, Bishwajit
Indian Institute of Technology, Delhi
New Delhi, India

244

Bilek, Vlastimil
ZPSV a.s.
Brno, Czech Republic
Bilir, Turhan
Blent Ecevit University
Zonguldak, Turkey
Billah, Abu Hena
The University of British Columbia
Kelowna, BC, Canada
Bimschas, Martin
Regensdorf, Switzerland
Binici, Baris
Middle East Technical University
Ankara, Turkey
Birkle, Gerd
Stantec Consulting Ltd.
Calgary, AB, Canada
Bisby, Luke
University of Edinburgh
Edinburgh, UK
Bisschop, Jan
University of Oslo
Oslo, Norway
Bobko, Christopher
North Carolina State University
Raleigh, NC
Bolhassani, Mohammad
Drexel University
Philadelphia, PA
Bonacci, John
Karins Engineering Group
Sarasota, FL
Bondar, Dali
Tehran, Islamic Republic of Iran
Bondy, Kenneth
Consulting Structural Engineer
West Hills, CA
Bonetti, Rodolfo
Pontificia Universidad Catlica Madre y Maestra
Santo Domingo, Dominican Republic
Borges, Paulo
Federal Centre for Technological Education of Minas Gerais
Belo Horizonte, Minas Gerais, Brazil
Bouras, Rachid
UMMTO
Tiziouzou, Algeria
Bournas, Dionysios
University of Nottingham
Nottingham, UK
Bousias, Stathis
University of Patras
Patras, Greece
Bradberry, Timothy
TXDot Bridge Division
Austin, TX
Braestrup, Mikael
Ramboll Hannemann and Hojlund A/S
Virum, Denmark
Braimah, Abass
Carleton University
Ottawa, ON, Canada

ACI Structural Journal/March-April 2015


Brand, Alexander
University of Illinois
Urbana, IL
Brea, Sergio
University of Massachusetts
Amherst, MA
Brewe, Jared
CTLGroup
Skokie, IL
Broujerdian, Vahid
Iran University of Science and Technology
Tehran, Islamic Republic of Iran
Brown, Michael
Virginia Transportation Research Council
Charlottesville, VA
Brown, Simon
Read Jones Christoffersen Ltd.
Calgary, AB, Canada
Bu, Wensheng
Wardrop Engineering Inc.
Sudbury, ON, Canada
Burak, Burcu
Orta Dogu Teknik Universitesi
Ankara, Turkey
Burris, Lisa
University of Texas at Austin
Austin, TX
Byard, Benjamin
University of Tennessee at Chattanooga
Chattanooga, TN
Bzeni, Dallshad
University of Salahaddin
Erbil, Iraq
Cai-Jun, Shi
Hunan University
Changsha, Hunan, China
Calixto, Jos
UFMG
Belo Horizonte, Brazil
Camero, Hugo
Construdiseos Ingenieros Arquitectos S.A.S.
Bogota D.C., Colombia
Campione, Giuseppe
Universita Palermo
Palermo, Italy
Cano Barrita, Prisciliano
Instituto Politcnico Nacional/CIIDIR Unidad Oaxaca
Oaxaca, Mexico
Canpolat, Fethullah
Yildiz Technical University
Istanbul, Turkey
Cao, Weiqun
Qingdao, China
Capozucca, Roberto
Ancona, Italy
Carino, Nicholas
Chagrin Falls, OH
Carreira, Domingo
Chicago, IL

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Carroll, Chris
University of Louisiana at Lafayette
Lafayette, LA
Carvalho, Alessandra
Pontifical Catholic University of Gois
Goinia, Gois, Brazil
Castles, Bryan
Western Technologies Inc.
Phoenix, AZ
Castro, Javier
Pontificia Universidad Catlica de Chile
Santiago, Chile
Catoia, Bruna
Federal University of So Carlos
So Carlos, So Paulo, Brazil
Cattaneo, Sara
Politecnico di Milano
Milan, Italy
Cavalaro, Sergio Henrique
Universidad Politcnica de Catalua
Barcelona, Spain
avuolu, brahim
Gmhane University
Gmhane, Turkey
Cervenka, Vladimir
Cervenka Consulting
Petriny, Czech Republic
Cetisli, Fatih
Pamukkale University
Denizli, Turkey
Chaallal, Omar
Ecole de Technologie Superiere
Verdun, QC, Canada
Chai, Hwa Kian
Tobishima Corporation
Noda, Chiba, Japan
Chakraborty, Arun
Bengal Engineering And Science University
Howrah, West Bengal, India
Chang, Ta-Peng
NTUST
Taipei, Taiwan, China
Chao, Shih-ho
University of Texas at Arlington
Arlington, TX
Chaudhary, Sandeep
Malaviya National Institute of Technology Jaipur
Jaipur, Rajasthan, India
Chaunsali, Piyush
University of Illinois
Urbana, IL
Chen, Chun-Tao
National Taiwan University of Science and Technology
Taipei, Taiwan, China
Chen, Hua-Peng
The University of Greenwich
Chatham, UK
Chen, Qi
Boral Materials Technology
San Antonio, TX

245

REVIEWERS IN 2014
Chen, Shiming
Tongji University
Shanghai, China
Chen, Wei
Wuhan University of Technology
Wuhan, Hubei, China
Chen, Xia
Changjiang River Scientific Research Institute
Wuhan, China
Cheng, Min-Yuan
National Taiwan University of Science and Technology
Taipei, Taiwan, China
Chi, Maochieh
Wufeng University
Chiayi County, Taiwan, China
Chiang, Chih-Hung
Chaoyang University of Technology
Wufong, Taichung, Taiwan, China
Chindaprasirt, Prinya
Khon Kaen University
Khon Kaen, Thailand
Cho, Jae-Yeol
Seoul National University
Seoul, Republic of Korea
Cho, Soon-Ho
Gwangju University
Gwangju, Republic of Korea
Choi, Chang-Sik
Hnayang University
Seoul, Republic of Korea
Choi, Eunsoo
Hongik University
Seoul, Republic of Korea
Choi, Hyun-Ki
Hanyang University
Seoul, Republic of Korea
Choi, Kyoung-Kyu
Soongsil University
Seoul, Republic of Korea
Choi, Sejin
University of California, Berkeley
Berkeley, CA
Chompreda, Praveen
Mahidol University
Nakornpathom, Thailand
Choong, Kokkeong
Universiti Sains Malaysia
Pulau Pinang, Seberang Perai Selatan, Malaysia
Chorzepa, Migeum
Park Ridge, IL
Chowdhury, Sharmin
Bogazici University
Istanbul, Turkey
Chowdhury, Subrato
Ultra Tech Cement LTD
Mumbai, Maharashtra, India
Christen, Robert
American Engineering Testing Inc.
Port Charlotte, FL

246

Chu Thi, Binh


Hanoi Architectural University
Hanoi, Vietnam
Chung, Deborah
University at Buffalo, the State University of New York
Buffalo, NY
Chung, Jae
University of Florida
Gainesville, FL
Chung, Lan
Dankook University
Seoul, Republic of Korea
Cintra, Danielli
Vitria, Esprito Santo, Brazil
Claisse, Peter
Coventry University
Coventry, UK
Cleland, Ned
Blue Ridge Design Inc.
Winchester, VA
Climent, Miguel
University of Alacant
Alacant, Spain
Coelho, Jano
Altoqi Informatica
Florianopolis, Santa Catarina, Brazil
Colombo, Matteo
Politecnico di Milano
Lecco, Italy
Cordova, Carlos
La Paz, Bolivia
Costa, Ricardo
University of Coimbra
Coimbra, Portugal
Crespi, Pietro
Politecnico of Milan
Milano, Italy
Criswell, Marvin
Colorado State University
Fort Collins, CO
Cueto, Jorge
Universidad de La Salle
Bogota, Colombia
Cumming, Neil
Levelton Engineering Ltd
Richmond, BC, Canada
Dagata, Giuseppe
University of Catania
Catania, Italy
dAndra, Renata
Getafe, Madrid, Spain
Dang, Canh
University of Arkansas
Fayetteville, AR
DArcy, Thomas
Consulting Engineers Group
San Antonio, TX
Darwin, David
University of Kansas
Lawrence, KS

ACI Structural Journal/March-April 2015


Das, Sreekanta
University of Windsor
Windsor, ON, Canada
Das Adhikary, Satadru
National University of Singapore
Singapore
De Brito, Jorge
IST / TUL
Lisbon, Portugal
De Korte, Arin
University of Twente
Enschede, the Netherlands
De Rooij, Mario
TNO
Delft, the Netherlands
De Schutter, Geert
Ghent University
Ghent, Belgium
Deb, Arghya
Indian Institute of Technology, Kharagpur
Kharagpur, West Bengal, India
Decker, Curtis
U.S. Military Academy
West Point, NY
Degtyarev, Vitaliy
Columbia, SC
Delalibera, Rodrigo
University of So Paulo
So Carlos, So Paulo, Brazil
Demir, Serhat
Blacksea Technical Univesity
Trabzon, Turkey
Den Uijl, Joop
Delft University of Technology
Delft, the Netherlands
Deng, Mingke
Xian University of Architecture and Technology
Xian, Shaanxi, China
Deng, Yaohua
Iowa State University
Ames, IA
Devries, Richard
Milwaukee School of Engineering
Milwaukee, WI
Dhinakaran, G.
Sastra University
Thanjavur, India
Dhonde, Hemant
University of Houston
Houston, TX
Di Ludovico, Marco
University of Naples Federico II
Naples, Italy
Diao, Bo
Beihang University
Beijing, China
Dias, W. P. S.
University of Moratuwa
Moratuwa, Sri Lanka

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Diaz Loya, Eleazar
Louisiana Tech University
Ruston, LA
Ding, Yining
Dalian, China
Diniz, Sofia Maria
Universidade Federal de Minas Gerais
Be lo Horizonte, Brazil
Do, Jeongyun
Kunsan National University
Kunsan, Jeonbuk, Republic of Korea
Dogan, Unal
Istanbul Technical University
Istanbul, Turkey
Dolan, Charles
University of Wyoming
Laramie, WY
Dongell, Jonathan
Pebble Technologies
Scottsdale, AZ
Dontchev, Dimitar
University of Chemical Technology and Metallurgy
Sofia, Bulgaria
Dotreppe, Jean-Claude
Universit of Liege-Mehanique Genie Civil
Liege, Belgium
Du, Hongjian
National University of Singapore
Singapore
Du, Lianxiang
The University of Alabama at Birmingham
Birmingham, AL
Du, Jinsheng
Beijing Jiao Tong University
Beijing, China
Du, Yingang
Anglia Ruskin University, UK
Chelmsford, UK
Dutta, Anjan
Indian Institute of Technology Guwahati
Guwahati, Assam, India
Ebead, Usama
Qatar University
Doha, Qatar
Eid, Rami
University of Sherbrooke
Sherbrooke, QC, Canada
El Meski, Fatima
American University of Beirut
Beirut, Lebanon
El Ragaby, Amr
University of Manitoba
Winnipeg, MB, Canada
El Sayed, Mohamed
University of Windsor
Windsor, ON, Canada
Elamin, Anwar
University of Nyala
Nyala, Sudan

247

REVIEWERS IN 2014
Elbatanouny, Mohamed
University of South Carolina
Columbia, SC
El-Dash, Karim
College of Technological Studies
Kuwait
El-Dieb, Amr
Ain Shams University
Abbasia, Cairo, Egypt
El-Hawary, Moetaz
Kuwait Institute for Scientific Research
Safat, Kuwait
El-Maaddawy, Tamer
United Arab Emirates University
Al-Ain, Abu Dhabi, United Arab Emirates
El-Metwally, Salah
University of Hawaii at Manoa
Honolulu, HI
El-Refaie, Sameh
El-Gama City, Mataria, Cairo, Egypt
El-Salakawy, Ehab
University of Manitoba
Winnipeg, MB, Canada
El-Sayed, Ahmed
University of Sherbrooke
Sherbrooke, QC, Canada
Elfgren, Lennart
Lule University of Technology
Lule, Sweden
Elhashmy, Awad
Cairo, Egypt
Elkady, Hala
NRC
Giza, Egypt
Elnady, Mohamed
Mansoura University
Vancouver, BC, Canada
Elmenshawi, Abdelsamie
University of Calgary
Calgary, AB, Canada
Elsayed, Tarek
Cairo, Egypt
Emamy Farvashany, Firooz
Perthpolis Pty Ltd
Perth, Western Australia, Australia
Erdem, T.
Izmir Institute of Technology
Izmir, Turkey
Ergn, Ali
Afyonkarahisar, Turkey
Esmaeily, Asad
Kansas State University
Manhattan, KS
Esmaili, Omid
University of California, Irvine
Irvine, CA
Esperanza, Menendez
IETCC-CSIC
Madrid, Spain
Etman, Emad
El-Mahalla El-Kobra, Egypt

248

Evangelista, Lus
Instituto Superior de Engenharia de Lisboa
Lisbon, Portugal
Faleschini, Flora
University of Padova
Padova, Italy
Fantilli, Alessandro
Politecnico di Torino
Torino, Italy
Fardis, Michael
Patras, Greece
Farghaly, Ahmed
University of Sherbrooke
Sherbrooke, QC, Canada
Faria, Duarte
Faculdade de Cincias e Tecnologia
Caparica-Lisbon, Portugal
Farrokhi, Farhang
Zanjan, Islamic Republic of Iran
Farrow, William
Lebanon, NJ
Farzam, Masood
Tabriz, Islamic Republic of Iran
Fathi, Hamoon
Sanandaj Branch, Islamic Azad University
Sanandaj, Kurdistan, Islamic Republic of Iran
Feldman, Lisa
University of Saskatchewan
Saskatoon, SK, Canada
Felekoglu, Burak
Dokuz Eylul University
Izmir, Turkey
Fernndez Montes, David
Madrid, Spain
Fernndez Ruiz, Miguel
Ecole Polytechnique Federale De Lausanne
Lausanne, Vaud, Switzerland
Ferrara, Liberato
Politecnico di Milano
Milan, Italy
Ferrier, E.
Universit Lyon 1
Villerubanne, France
Folino, Paula
University of Buenos Aires
Buenos Aires, Argentina
Foraboschi, Paolo
Universita IUAV di Venezia
Venice, Italy
Fouad, Fouad
University of Alabama at Birmingham
Birmingham, AL
Fradua, Martin
Feld, Kaminetzky & Cohen, P.C.
Jericho, NY
Francois, Buyle-Bodin
University of Lille
Villeneuve dAscq, France
Freyne, Seamus
Manhattan College
Riverdale, NY

ACI Structural Journal/March-April 2015


Fuchs, Werner
University of Stuttgart
Stuttgart, Germany
Fuentes, Jose Maria
Polytechnic University of Madrid
Madrid, Spain
Furlong, Richard
Austin, TX
Gabrijel, Ivan
University of Zagreb
Zagreb, Croatia
Gajdosova, Katarina
Bratislava, Slovakia
Galati, Nestore
Structural Group Inc.
Elkridge, MD
Gallegos Mejia, Luis
Fundacion Padre Arrupe de El Salvador
Soyapango, San Salvador, El Salvador
Gamble, William
University of Illinois
Urbana, IL
Gao, Xiangling
Tongji University
Shanghai, China
Garber, David
Florida International University
Miami, FL
Garcez, Estela
Universidade Federal de Pelotas
Pelotas, RS, Brazil
Garcia-Taengua, Emilio
Queens University of Belfast
Belfast, UK
Gedik, Yasar
Istanbul Technical University
Istanbul, Turkey
Gesoglu, Mehmet
Gaziantep University
Gaziantep, Turkey
Gettu, Ravindra
Indian Institute of Technology Madras
Chennai, India
Ghafari, Nima
Laval University
Quebec, QC, Canada
Ghali, Amin
University of Calgary
Calgary, AB, Canada
Ghanem, Hassan
Texas A&M University
College Station, TX
Ghasemzadeh, Farnam
North Carolina State University
Raleigh, NC
Ghezal, Acha
Ecole de Technologie de Montreal
Montreal, QC, Canada
Ghoddousi, Parviz
Iran University of Science and Technology
Tehran, Islamic Republic of Iran

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Giaccio, Craig
AECOM
Melbourne, Victoria, Australia
Girgin, Canan
Yildiz Technical University
Istanbul, Turkey
Goel, Rajeev
CSIR-Central Road Research Institute
Delhi, India
Gke, H. Sleyman
Ege University
Izmir, Turkey
Gongxun, Wang
Hunan University of Science and Technology
Xiangtan, China
Gonzales Garcia, Luis Alberto
Lagging SA
Lima, Peru
Gonzlez, Javier
University of Basque Country
Bilbao, Basque Country, Spain
Gonzlez-Valle, Enrique
Madrid, Spain
Goudarzi, Nabi
Edmonton, AB, Canada
Grandi, Davor
University of Rijeka
Rijeka, Croatia
Gribniak, Viktor
Vilnius Gediminas Technical University
Vilnius, Lithuania
Gu, Xiang-Lin
Tongji University
Shanghai, China
Guadagnini, Maurizio
The University of Sheffield
Sheffield, UK
Guan, Garfield
Cambridge, UK
Guimaraes, Giuseppe
Pontificia Universidade Catlica do Rio de Janeiro
Rio de Janeiro, Brazil
Gulec, Cevdet
Thornton Tomasetti, Inc.
Los Angeles, CA
Gneyisi, Erhan
Gaziantep University
Gaziantep, Turkey
Guo, Honglei
Wuhan Polytechnic University
Wu Han City, Hu bei Province, China
Guo, Liping
Southeast University
Nanjing, Jiangsu Province, China
Guo, Zixiong
Huaqiao University
Quanzhou, Fujian, China
Gupta, Ajay
M.B.M. Engineering College
Jodhpur, Rajasthan, India

249

REVIEWERS IN 2014
Gupta, Rishi
Vancouver, BC, Canada
Gupta, Supratic
Indian Institute of Technology, Delhi
New Delhi, India
Haddad, Rami
Jordan University of Science and Technology
Irbid, Jordan
Haddadin, Laith
United Nations
New York, NY
Hadje-Ghaffari, Hossain
John A. Martin & Assoc.
Los Angeles, CA
Hagenberger, Michael
Ohio State University
Columbus, OH
Haggag, Hesham
Cairo, Egypt
Hamilton, Trey
University of Florida
Gainesville, FL
Hammood, Oday
University Technology Malaysia
Skudai, Johor Buhro, Malaysia
Han, Dongyeop
University of Texas at Austin
Austin, TX
Harajli, Mohamed
American University of Beirut
Beirut, Lebanon
Harbec, David
Universit de Sherbrooke
Sherbrooke, QC, Canada
Hariri-Ardebili, Mohammad Amin
University of Colorado
Boulder, CO
Harries, Kent
University of Pittsburgh
Pittsburgh, PA
Harris, Devin
University of Virginia
Charlottesville, VA
Harris, G. Terry
Green Cove Springs, FL
Hasan, Sahar
Higher Institute for Engineering and Technology
Alexandria, Egypt
Hashemi, Shervin
Seoul National University
Seoul, Republic of Korea
Hasnat, Ariful
University of Asia Pacific
Dhaka, Bangladesh
Hassan, Assem
Toronto, ON, Canada
Hassan, Mohamed
University of Sherbrooke
Sherbrooke, QC, Canada

250

Hassan, Wael
University of California, Berkeley
Berkeley, CA
Hassani, Abolfazl
Tarbiat Modares University
Tehran, Islamic Republic of Iran
He, Zhiqi
Southeast University
Nanjing, Jiangsu, China
Heinzmann, Daniel
Lucerne University of Applied Sciences and Arts
Horw, Switzerland
Helal, Yasser
University of Sheffield
Sheffield, UK
Helmy, Huda
Applied Science International
Durham, NC
Hemalatha, T.
CSIR-Structural Engineering Research Centre
Chennai, Tamil Nadu, India
Henry, Richard
University of Auckland
Auckland, New Zealand
Herrera, Angel
Rio Piedras, Puerto Rico
Hindi, Riyadh
Saint Louis University
St. Louis, MO
Ho, Johnny
The University of Hong Kong
Hong Kong, China
Hochstein, Daniel
Manhattan College
Riverdale, NY
Hoehler, Matthew
Encinitas, CA
Hoff, George
Hoff Consulting Inc.
Clinton, MS
Holschemacher, Klaus
HTWK Leipzig
Leipzig, Germany
Hong, Sung-Gul
Seoul National University
Seoul, Republic of Korea
Hosny, Amr
North Carolina State University
Raleigh, NC
Hossain, Mustaque
Kansas State University
Manhattan, KS
Hoult, Neil
Toronto, ON, Canada
Hrynyk, Trevor
University of Texas at Austin
Austin, TX
Hu, Jiong
Texas State University
San Marcos, TX

ACI Structural Journal/March-April 2015


Hu, Nan
Tsinghua University
Beijing, China
Huang, Yishuo
Chaoyang University of Technology
Wufeng, Taichung, Taiwan, China
Huang, Zhaohui
Brunel University
London, UK
Huang, Chang-Wei
Chung Yuan Christian University
Chung Li, Taiwan, China
Huang, Chung-Ho
Dahan Institute of Technology
Hualien, Taiwan, China
Huang, Jianwei
Southern Illinois University Edwardsville
Edwardsville, IL
Huang, Xiaobao
GM-WFG/GM-N American Project Center
Warren, MI
Huo, Jingsi
Hunan University
Changsha, Hunan, China
Husain, Mohamed
Zagazig University
Zagazig, Egypt
Husem, Metin
Karadeniz Technical University
Trabzon, Turkey
Huynh, Minh Phuoc
Ho Chi Minh City University Transport
Ho Chi Minh, Vietnam
Ibell, Tim
University of Bath
Bath, UK
Ibrahim, Amer
Baquba, Iraq
Ichinose, Toshikatsu
Nagoya Institute of Technology
Nagoya, Japan
Ikponmwosa, Efe
University of Lagos
Akoka, Lagos, Nigeria
Ince, Ragip
Firat University Engineering Faculty
Elazig, Turkey
Ipek, Sleyman
Gaziantep University
Gaziantep, Turkey
Irassar, Edgardo
National University of Central Buenos Aires
Olavarria, Buenos Aires, Argentina
Islam, Md.
Chittagong University of Engineering & Technology (CUET)
Chittagong, Bangladesh
Issa, Mohamed
National Center for Housing and Building Resarch
Giza, Egypt

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Issa, Mohsen
University of Illinois at Chicago
Chicago, IL
Izquierdo-Encarnacin, Jose
Porticus
San Juan, Puerto Rico
Jaari, Asaad
Dera, Dubai, United Arab Emirates
Jain, Mohit
Nirma University
Ahmedabad, Gujarat, India
Jain, Shashank
Delhi Technological University (DTU)
New Delhi, India
Jalal, Mostafa
PWUT
Tehran, Islamic Republic of Iran
Jan, Song
Bechtel Corp.
Houston, TX
Jang, Seung Yup
Korea Railroad Research Institute
Uiwang, Gyongggi-do, Republic of Korea
Jansen, Daniel
California Polytechnic State University
San Luis Obispo, CA
Jawaheri Zadeh, Hany
Miami, FL
Jayapalan, Amal
Exponent Failure Analysis Associates
Menlo Park, CA
Jeng, Chyuan-Hwan
National Chi Nan University-Taiwan
Puli/Nantou, Taiwan, China
Jiang, Jiabiao
W R Grace (Singapore) Pte Ltd
Singapore
Johnson, Gaur
University of Hawaii
Honolulu, HI
Jozi, Draan
Split, Croatia
Kaklauskas, Gintaris
Vilnius Gediminas Technical University
Vilnius, Lithuania
Kan, Yu-Cheng
Chaoyang University of Technology
Taichung County, Taiwan, China
Kanagaraj, Ramadevi
Kumaraguru College of Technology
Coimbatore, Tamilnadu, India
Kanakubo, Toshiyuki
University of Tsukuba
Tsukuba, Japan
Kandasami, Siva
Bristol, UK
Kang, Thomas
Seoul National University
Seoul, Republic of Korea

251

REVIEWERS IN 2014
Kankam, Charles
Kwame Nkrumah University of Science & Technology
Kumasi, Ghana
Kansara, Kunal
Mouchel Infrastructure Services
Bristol, UK
Kantarao, Velidandi
Central Road Research Institute
New Delhi, India
Karayannis, Christos
Democritus University of Thrace
Xanthi, Greece
Karbasi Arani, Kamyar
University of Naples Federico II
Napoli, Campagna, Italy
Kawamura, Mitsunori
Kanazawa, Ishikawa, Japan
Kazemi, Mohammad
Sharif University of Technology
Tehran, Islamic Republic of Iran
Kazemi, Sadegh
University of Alberta
Edmonton, AB, Canada
Kenai, Said
Universit de Blida
Blida, Algeria
Khan, Mohammad
King Saud University
Riyadh, Saudi Arabia
Khan, Sadaqat
Universiti Teknologi Petronas
Tronoh, Perak, Malaysia
Khuntia, Madh
Dukane Precast Inc.
Naperville, IL
Kianoush, M. Reza
Ryerson University
Toronto, ON, Canada
Kim, Jang Hoon
Ajou University
Suwon, Republic of Korea
Kim, Sang-Woo
Kongju National University
Cheonan, Chungnam, Republic of Korea
Kim, Woo
Chonnam National University
Kwangju, Republic of Korea
Kim, Yail Jimmy
University of Colorado Denver
Denver, CO
Kirgiz, Mehmet
Hacettepe University
Ankara, Turkey
Kishen, Chandra
Indian Institute of Science
Bangalore, Karnataka, India
Kishi, Norimitsu
Muroran Institute of Technology
Muroran, Japan

252

Klein, Gary
Wiss, Janney, Elstner Associates, Inc.
Northbrook, IL
Klemczak, Barbara
Silesian Technical University
Gliwice, Poland
Ko, Lesley Suz-Chung
Holcim Group Support Ltd.
Holderbank, AG, Switzerland
Koehler, Eric
University of Texas at Austin
Austin, TX
Koenders, Eddy A. B.
Delft University of Technology
Delft, the Netherlands
Konsta-Gdoutos, Maria
Northwestern University
Evanston, IL
Kotsovos, Gerasimos
National Technical University of Athens
Athens, Greece
Kotsovos, Michael
Athens, Greece
Kreger, Michael
Purdue University
West Lafayette, IN
Ksiek, Mariusz
Wrocaw University of Technology
Wroclaw, Poland
Kumar, Pardeep
University of California, Berkeley
Berkeley, CA
Kumar, Rakesh
Central Road Research Institute
Delhi, India
Kumaravel, S.
Annamalai University
Cuddalore, Tamilnadu, India
Kupwade-Patil, Kunal
Massachusetts Institute of Technology
Cambridge, MA
Kurtis, Kimberly
Georgia Institute of Technology
Atlanta, GA
Kusbiantoro, Andri
Universiti Malaysia Pahang
Gambang, Pahang, Malaysia
Kuyucular, Adnan
Pamukkale University
Kinikli-Denizli, Turkey
Kwan, Albert
The University of Hong Kong
Hong Kong, China
Lai, James
La Caada, CA
Lai, Jianzhong
Nanjing University of Science and Technology
Nanjing, Jiangsu, China
Laldji, Said
Universit de Sherbrooke
Sherbrooke, QC, Canada

ACI Structural Journal/March-April 2015


Lam, Eddie
The Hong Kong Polytechnic Universiy
Hong Kong, China
Larbi, Kacimi
University of Sciences and Technology of Oran
Oran, Algeria
Laskar, Aminul
National Institute of Technology
Silchar, Assam, India
Laterza, Michelangelo
University of Basilicata
Potenza, Italy
Latifee, Enamur
Clemson University
Clemson, SC
Law, David
RMIT University
Melbourne, Victoria, Australia
Lawler, John
Wiss, Janney, Elstner Associates, Inc.
Northbrook, IL
Lawrence, Adrian
Gainesville, FL
Lee, Chi King
Nanyang Technological University
Singapore
Lee, Chung-Sheng
University of California, San Diego
La Jolla, California
Lee, Deuck Hang
University of Seoul
Seoul, Republic of Korea
Lee, Douglas
Douglas D. Lee and Associates
Fort Worth, TX
Lee, Heui Hwang
Arup
San Francisco, CA
Lee, Hung-Jen
National Yunlin University of Science and Technology
Douliu, Yunlin, Taiwan, China
Lee, Jung-Yoon
Sung Kyun Kwan University
Suwon, Republic of Korea
Lee, Nam Ho
SNC-Lavalin Nuclear
Oakville, ON, Canada
Lee, Seong-Cheol
KEPCO International Graduate School (KINGS)
Ulsan, Republic of Korea
Lei, Aizhong
China Institute of Water Resources and Hydropower Research
Beijing, China
Lepage, Andres
University of Kansas
Lawrence, KS
Lequesne, Remy
University of Kansas
Lawrence, KS

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Li, Fumin
China University of Mining and Technology
Xuzhou, Jiangsu, China
Li, Long-yuan
University of Plymouth
Plymouth, UK
Li, Wei
Wenzhou University
Wenzhou, Zhejiang, China
Li, Xinghe
University of New Hampshire
Durham, NH
Li, Yi-An
National Taiwan University
Taipei, Taiwan, China
Lignola, Gian Piero
University of Naples Federico II
Naples, Italy
Lin, Wei-Ting
Ilan, Taiwan, China
Lin, Zhibin
Fargo, ND
Liu, Jun
Beijing, China
Liu, Junshan
Sargent Lundy LLC
Chicago, IL
Liu, Shuhua
Wuhan University
Wuhan, Hubei, China
Liu, Zhao
Southeast University
Nanjing, Jiangsu, China
Liu, Xuejian
University of Texas at Arlington
Arlington, TX
Liu, Yanbo
Florida Atlantic University
Boca Raton, FL
Liu, Ze
China University of Mining & Technology, Beijing
Beijing, China
Lo, T. Y.
City University of Hong Kong
Hong Kong, China
Long, Nguyen
Kosice, Slovakia
Long, Xu
NanyangTechnological University
Singapore
Loo, Yew-Chaye
Gold Coast, Australia
Lopes, Anne
Furnas Centrais Eletricas Sa Aparecida De Goiania
Goias, Brazil
Lopes, Sergio
University of Coimbra
Coimbra, Portugal
Lpez-Almansa, Francisco
Technical University of Catalonia
Barcelona, Spain

253

REVIEWERS IN 2014
Lotfy, Abdurrahmaan
Lafarge Canada Inc.
Toronto, ON, Canada
Lounis, Zoubir
National Research Council
Ottawa, ON, Canada
Lubell, Adam
Read Jones Christoffersen Ltd.
Vancouver, BC, Canada
Ludovit, Nad
Alfa 04
Kosice, Slovakia
Luo, Baifu
Harbin, China
Lushnikova, Nataliya
National University of Water Management and
NatureResourcesUse
Rivne, Ukraine
Ma, Zhongguo
University of Tennessee
Knoxville, TN
MacDonald, Kevin
Cemstone Concrete Products Co.
Mendota Heights, MN
Machida, Atsuhiko
Saitama University
Saitama, Japan
Macht, Jrgen
Kirchdorf, Austria
Maekawa, Koichi
University of Tokyo
Tokyo, Japan
Maganti, Ravindra
D.M.S. S.V.H. College of Engineering Machilipatnam
Andhra Pradesh, India
Magliulo, Gennaro
University of Naples Federico II
Naples, Italy
Maguire, Marc
Utah State University
Paradise, UT
Mahfouz, Ibrahim
Cairo, Egypt
Mahrenholtz, Christoph
Berlin, Germany
Mahrenholtz, Philipp
Frankfurt, Germany
Malik, Adnan
University of New South Wales
Sydney, Australia
Mander, John
Texas A&M University
College Station, TX
Manso, Juan
University of Burgos
Burgos, Castilla - Len, Spain
Mari, Antonio
Universitat Politecnica de Catalunya
Barcelona, Spain

254

Marikunte, Shashi
Southern Illinois University
Carbondale, IL
Mart-Vargas, Jos
Universitat Politcnica de Valncia
Valencia, Spain
Martinelli, Enzo
University of Salerno
Fisciano, Italy
Maruyama, Ippei
Nagoya University
Nagoya, Aichi, Japan
Maslehuddin, Mohammed
King Fahd University of Petroleum and Minerals
Dhahran, Saudi Arabia
Matta, Fabio
University of South Carolina
Columbia, SC
Maximos, Hany
Pharos University in Alexandria
Alexandria, Egypt
Mbessa, Michel
University of Yaound I - ENSP
Yaound, Center, Cameroon
McCarter, John
Heriot Watt University
Edinburgh, UK
McDonald, David
USG Corp
Libertyville, IL
McLeod, Heather
Kansas Department of Transportation
Topeka, KS
Meda, Alberto
University of Bergamo
Bergamo, Italy
Medallah, Khaled
Saudi Aramco IKPMS
Al Khobar, Saudi Arabia
Meddah, Mohammed Seddik
Kingston University London
Kingston, UK
Mehanny, Sameh
Cairo University
Cairo, Egypt
Meinheit, Donald
Wiss, Janney, Elstner Associates, Inc.
Chicago, IL
Melo, Jos
University of Aveiro
Aveiro, Portugal
Meng, Tao
Institution of Building Materials
Hangzhou, Zhejiang, China
Menon, Devdas
Indian Institute of Technology
Chennai, Tamilnadu, India
Mermerda, Kasm
Hasan Kalyoncu University
Gaziantep, Turkey

ACI Structural Journal/March-April 2015


Meshgin, Pania
University of Colorado Boulder
Boulder, CO
Milestone, Neil
Callaghan Innovation
Lower Hutt, New Zealand
Minehane, Michael
RPS Group Ltd.
Cork, Ireland
Mlynarczyk, Alexandar
Wiss, Janney, Elstner Associates, Inc.
Princeton Junction, NJ
Mo, Yi-Lung
University of Houston
Houston, TX
Mohamed, Ashraf
Alexandria University
Alexandria, Egypt
Mohamed, Nayera
Assiut University
Assiut, Egypt
Mohammadyan Yasouj, Seyed Esmaeil
UTM University
Johor, Malaysia
Mohammed, Tarek
University of Asia Pacific
Dhaka, Bangladesh
Mohd Zain, Mumammad Fauzi
Universiti Kebangsaan Malaysia
Bangi, Malaysia
Mokarem, David
Virginia Polytechnic University
Blacksburg, VA
Mondal, Bipul
Chittagong University of Engineering & Technology
Chittagong, Bangladesh
Montejo, Luis
North Carolina State University
Raleigh, NC
Moradi, Hiresh
Amirkabir University of Technology
Tehran, Islamic Republic of Iran
Moretti, Marina
University of Thessaly
Athens, Greece
Moser, Robert
U.S. Army Engineer Research and Development Center
Vicksburg, MS
Mostafaei, Hossein
University of Toronto
Toronto, ON, Canada
Mostofinejad, Davood
Isfahan University of Technology
Isfahan, Islamic Republic of Iran
Muciaccia, Giovanni
Politecnico di Milano
Milan, Italy
Mulaveesala, Ravibabu
Indian Institute of Technology Ropar
Rupnagar, India

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Mullapudi, Taraka Ravi
MMI Engineering
Houston, TX
Munoz, Jose
Federal Highway Administration
McLean, VA
Muttoni, Aurelio
Swiss Federal Institute of Technology
Lausanne, Switzerland
Nabavi, Esrafil
Rezvanshahr, Guilan, Islamic Republic of Iran
Nafie, Amr
Cairo, Egypt
Nair, Priya
Cochin University of Science and Technology
Kochi, Kerala, India
Naish, David
California State University, Fullerton
Fullerton, CA
Najimi, Meysam
University of Nevada, Las Vegas
Las Vegas, NV
Nakamura, Hikaru
Nagoya University
Nagoya, Aichi, Japan
Nam, Boo Hyun
University of Central Florida
Orlando, FL
Negrutiu, Camelia
Technical University of Cluj Napoca
Cluj Napoca, Cluj, Romania
Neves, Lus
University of Coimbra
Coimbra, Portugal
Ng, Ivan
Drainage Services Department
Hong Kong, China
Nichols, John
Texas A&M University
College Station, TX
Niemuth, Mark
Lafarge
Alpharetta, GA
Nimityongskul, Pichai
Asian Institute of Technology
Pathumthani, Thailand
Nishiyama, Minehiro
Kyoto University
Kyoto, Japan
Noor, Munaz
Bangladesh University of Engineering and Technology
Dhaka, Bangladesh
Noshiravani, Talayeh
EPFL
Lausanne, Switzerland
Ochotorena, Richard
Permasteelisa Group
Hong Kong Island, Hong Kong, China
Oh, Byung
Seoul National University
Seoul, Republic of Korea

255

REVIEWERS IN 2014
Ohtsu, Masayasu
Kumamoto University
Kumamoto, Japan
Okeil, Ayman
Louisiana State University
Baton Rouge, LA
Okelo, Roman
Dallas, TX
Olanitori, Lekan
Federal University of Technology, Akure
Akure, Ondo State, Nigeria
Ombres, Luciano
University of Calabria
Cosenza, Italy
Omran, Ahmed
University of Sherbrooke
Sherbrooke, QC, Canada
Orakcal, Kutay
Bogazici University
Istanbul, Bebek, Turkey
Orr, John
University of Bath
Bath, UK
Ortega, J.
University of Alacant
Alacant, Alicante, Spain
Ortiz-Lozano, Jose
Autonomous University of Aguascalientes
Aguascalientes, Mexico
Orton, Sarah
University of Missouri Columbia
Columbia, MO
Osifala, Kehinde
Somolu, Lagos, Nigeria
Otieno, Mike
University of the Witwatersrand
Johannesburg, Gauteng, South Africa
Ouzaa, Kheira
USTO
ORAN, Algeria
Ozturan, Turan
Bogazici University
Istanbul, Turkey
Ozturk, Ali
Dokuz Eylul University
Izmir, Buca, Turkey
Pacheco, Alexandre
Universidade Federal do Rio Grande do Sul (UFRGS)
Porto Alegre, RS, Brazil
Palieraki, Vasiliki
National Technical University of Athens
Athens, Zografou, Greece
Palmisano, Fabrizio
Politecnico di Bari
Bari, Italy
Pan, Wang Fook
Segi University
Petaling Jaya, Selangor, Malaysia
Pang, Sze Dai
National University of Singapore
Singapore

256

Pantazopoulou, Stavroula
Demokritus University of Thrace
Xanthi, Greece
Pape, Torill
University of Newcastle
Callaghan, New South Wales, Australia
Parsekian, Guilherme
Federal University of So Carlos
So Carlos, So Paulo, Brazil
Pauletta, Margherita
University of Udine
Tavagnacco, Udine, Italy
Paulotto, Carlo
Acciona S.A.
Alcobendas, Spain
Pavlikova, Milena
CTU
Prague, Czech Republic
Pellegrino, Carlo
University of Padova
Padova, Italy
Peng, Cao
Harbin Institute of Technology
Harbin, Heilongjiang, China
Peng, Jianxin
Institute of Bridge Engineering
Changsha, Hunan, China
Pereira, Eduardo
University of Minho
Guimaraes, Portugal
Perez Caldentey, Alejandro
Universidad Politcnica de Madrid
Madrid, Spain
Persson, Bertil
Bara, Sweden
Phillippi, Don
Diamond Pacific
Rancho Cucamonga, CA
Piccinin, Roberto
Hilti, Inc.
Tulsa, OK
Pocesta, Ylli
Debar, The Former Yugoslav Republic of Macedonia
Ponnada, Markandeya
MVGR College of Engineering
Vizianagaram, Andhra Pradesh, India
Potnoor, Naveen
Sasan, Madhya Pradesh, India
Potter, William
Florida Department of Transportation
Tallahassee, FL
Pourazin, Khashaiar
Pars Ab Tadbir Consulting Engineers Co.
Tehran, Islamic Republic of Iran
Prasad, Saurabh
University of California, San Diego
La Jolla, CA
Prashanth, P.
SJCE
Mysore, Karnataka, India

ACI Structural Journal/March-April 2015


Prasittisopin, Lapyote
Oregon State University
Corvallis, OR
Puertas, F.
Eduardo Torroja Institute
Madrid, Spain
Puthenpurayil Thankappan, Santhosh
Granite Construction Company
Abu Dhabi, United Arab Emirates
Putra Jaya, Ramadhansyah
Universiti Teknologi Malaysia
Skudai, Johor Bahru, Malaysia
Qasrawi, Hisham
The Hashemite University
Zarqa, Jordan
Qian, Kai
Nanyang Technological University
Singapore
Qiangqiang, Zhang
Harbin Institute of Technology
Harbin, Heilongjiang, China
Quaranta, Giuseppe
Sapienza University of Rome
Rome, Italy
Quiroga, Pedro
Escuela Colombiana de Ingenieria
Bogota, Colombia
Rafi, Muhammad
NED University of Engineering and Technology
Karachi, Sindh, Pakistan
Raikar, Chetan
Structwel Designers and Consultants Pvt. Ltd.
Navi Mumbai, India
Ramamurthy, K.
IIT Madras
Chennai, Tamilnadu, India
Ramaswamy, Ananth
Indian Institute of Science
Bangalore, Karnataka, India
Ramos, Antnio
Faculdade de Cincias e Tecnologia
Monte de Caparica, Portugal
Rao, Hanchate
JNTU College of Engineering
Anantapur, India
Rashed, Youssef F.
Giza, Egypt
Rasol, Mezgeen
Dohuk Polytechince University
Zakho, Duhok, Iraq
Ray, Indrajit
Purdue University Calumet
Hammond, IN
Regan, Paul
Trigram
London, UK
Restrepo, Jose
University of California-San Diego
La Jolla, CA

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Riad, Khaled
Ain Shams University
Cairo, Egypt
Riding, Kyle
Kansas State University
Manhattan, KS
Rinaldi, Zila
University of Rome Tor Vergata
Rome, Italy
Rivard, Patrice
Universit de Sherbrooke
Sherbrooke, QC, Canada
Rivero-Angeles, Francisco
Mexico, DF, Mexico
Rizk, Emad
Memorial University of Newfoundland
St. Johns, NL, Canada
Rizwan, Syed Ali
University of Engineering and Technology
Lahore, Punjab, Pakistan
Rodriguez, Mario
National Autonomous University of Mexico
Mexico City, DF, Mexico
Sabouni, Faisal
Architectural Consulting Group
Abu Dhabi, United Arab Emirates
Sadeghi Pouya, Homayoon
Coventry University
Coventry, UK
Saedi, Houman
Tabiat Modares & Tabriz University and TSML CO.
Tehran, Islamic Republic of Iran
Safan, Mohamed
Menoufia University
Shebeen El-Kom, Menoufia, Egypt
Sagaseta, Juan
University of Surrey
Guildford, Surrey, UK
Sagues, Alberto
University of South Florida
Tampa, FL
Sahamitmongkol, Raktipong
CONTEC, SIIT, Thammasat University and MTEC
Pathumthani, Thailand
Sahmaran, Mustafa
Gazi University
Ankara, Turkey
Saibabu, S.
CSIR-Structural Engineering Research Centre
Chennai, Tamilnadu, India
Sajedi, Fathollah
University of Malaya
Kuala Lumpur, Selangor, Malaysia
Saka, Mehmet
Middle East Technical University
Ankara, Turkey
Salem, Hamed
Cairo University
Giza, Egypt
Salib, Sameh
Markham, ON, Canada

257

REVIEWERS IN 2014
Salinas-Basualdo, Rafael
National University of Engineering
Lima, Peru
Sallam, Hossam El-Din
Zagazig University
Zagazig, Sharkia, Egypt
Sanada, Yasushi
Toyohashi University of Technology
Toyohashi, Japan
Snchez, Isidro
University of Alicante
Alicante, Spain
Sanchez, Leandro
So Paulo, Brazil
Santos, Srgio
Instituto de Ps-Graduao
Goinia, Gois, Brazil
Saqan, Elias
American University in Dubai
Dubai, United Arab Emirates
Sarker, Prabir
Curtin University of Technology
Bentley, Western Australia, Australia
Sato, Yuichi
Kyoto University
Kyoto, Japan
Scanlon, Andrew
Pennsylvania State University
University Park, PA
Schileo, Giorgio
Sheffield Hallam University
Sheffield, UK
Schindler, Anton
Auburn University
Auburn, AL
Semaan, Hassnaa
Ottawa Hills, OH
Sener, Siddik
Istanbul Bilgi University
Instanbul, Eyup, Turkey
Sengul, Ozkan
Istanbul Technical University
Istanbul, Turkey
Sengupta, Amlan
Indian Institute of Technology Madras
Chennai, Tamil Nadu, India
Serna-Ros, Pedro
Universidad Politecnia de Valencia
Valencia, Spain
Setin, Jess
University of Cantabria
Santander, Cantabria, Spain
Shafigh, Payam
Kuala Lumpur, Malaysia
Shafiq, Nasir
University Technology Petronas
Tronoh, Perak, Malaysia
Shah, Attaullah
Allama Iqbal Open University
Islamabad, Pakistan

258

Shah, Santosh
Dharmsinh Desai University
Nadiad, Gujarat, India
Shannag, M. Jamal
King Saud University
Riyadh, Saudi Arabia
Shao, Yixin
McGill University
Montreal, QC, Canada
Sharifi, Yasser
Vali-e-Asr University of Rafsanjan
Rafsanjan, Islamic Republic of Iran
Sharma, Akanshu
Institute of Construction Materials
Stuttgart, Germany
Shashikala, A. P.
National Institute of Technology Calicut
Calicut, Kerala, India
Shawky, Mostafa
Alexandria, Egypt
Shehata, Medhat
Ryerson University
Toronto, ON, Canada
Sheikh, Shamim
University of Toronto
Toronto, ON, Canada
Sherif, Alaa
Helwan University
Cairo, Egypt
Sherman, Matthew
Simpson Gumpertz and Heger
Melrose, MA
Sherwood, Edward
Carleton University
Ottawa, ON, Canada
Shi, Xianming
Washington State University
Pullman, WA
Shi, Xudong
Tsinghua University
Beijing, China
Shi, Yilei
Rockville, MD
Shing, Pui-Shum
University of California, San Diego
La Jolla, CA
Shivali, Ram
Central Soil and Materials Research Station
New Delhi, India
Silfwerbrand, Johan
KTH Royal Institute of Technology
Stockholm, Sweden
Singh, Harvinder
Guru Nanak Dev Engineering College
Ludhiana, Punjab, India
Sinn, Robert
Thornton Tomasetti
Chicago, IL
Sisman, Can
Namk Kemal University
Tekirdag, Turkey

ACI Structural Journal/March-April 2015


Smadi, Mohammad
Jordan University of Science and Technology
Irbid, Jordan
Sobhan, Khaled
Florida Atlantic University
Boca Raton, FL
Sobhani, Jafar
Building and Housing Research Center
Tehran, Islamic Republic of Iran
Soliman, Ahmed
Western University
London, ON, Canada
Soltani, Amir
Purdue University Calumet
Hammond, IN
Soltani, Masoud
Tarbiat Modares University
Tehran, Islamic Republic of Iran
Song, Xin
Zhejiang, China
Sossou, Gnida
Kwame Nkrumah University of Science and Technology
Kumasi, Ghana
Souza, Rafael
Universidade Estadual de Maring
Maring, Paran, Brazil
Sylev, Altug
Yeditepe University
Istanbul, Turkey
Spinella, Nino
University of Messina
Messina, Italy
Spyridis, Panagiotis
Institute for Structural Engineering
Vienna, Austria
Stein, Boris
Twining Laboratories
Long Beach, CA
Strang, Fred
New Brunswick Department of Transportation
Fredericton, NB, Canada
Strauss, Alfred
University of Natural Resources and Life Sciences
Vienna, Austria
Su, Yu-Min
National Kaohsiung University of Applied Sciences
Sanmin, Taiwan, China
Sujjavanich, Suvimol
Kasetsart University
Bangkok, Thailand
Suksawang, Nakin
Florida Institute of Technology
Melbourne, FL
Sullivan, Patrick
Sullivan and Associates
Rickmansworth, UK
Suraneni, Prannoy
ETH Zrich
Zrich, Switzerland
Switonski, Aleksander
Bydgoszcz, Poland

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Tabatabai, Habib
University of Wisconsin Milwaukee
Milwaukee, WI
Tadayon, Mohammadhosein
University of Tehran
Tehran, Islamic Republic of Iran
Tae, Ghi Ho
Leader Industrial Co.
Seoul, Republic of Korea
Tahir, Muhammad
UET Taxila
Taxila, Punjab, Pakistan
Tahmasebinia, Faham
University of Wollongong
Wollongong, New South Wales, Australia
Talbot, Caroline
Cleveland, OH
Tan, Kefeng
Southwest University of Science and Technology
Sichuan, China
Tanacan, Leyla
Istanbul, Yesilkoy, Turkey
Tang, Chao-Wei
Cheng-Shiu University
Niaosong District, Kaohsiung City, Taiwan, China
Tang, Liqun
South China University of Technology
Guangzhou, Guangdong, China
Tang, Pei
Eindhoven, the Netherlands
Tangtermsirikul, Somnuk
Sirindhorn International Institute of Technology
Patumthani, Thailand
Tank, Tejenadr
Pandit Deendayal Petroleum University
Gandhinagar, Gujarat, India
Tankut, Tugrul
Middle East Technical University
Ankara, Turkey
Tanner, Jennifer
University of Wyoming
Laramie, WY
Tao, Shi
Zhejiang University of Technology
Hangzhou, China
Tapan, Mcip
Yuzuncu Yil University
Van, Turkey
Tarighat, Amir
Tehran, Islamic Republic of Iran
Tassios, Theodosios
Athens, Greece
Tastani, S. P.
Demokritus University of Thrace
Xanthi, Greece
Tavares, Maria
UERJ-State University of Rio de Janeiro
Rio de Janeiro, Brazil
Tavio
Sepuluh Nopember Institute of Technology (ITS)
Surabaya, East Java, Indonesia

259

REVIEWERS IN 2014
Tawana, M. M.
Tongji University
Shanghai, China
Tawfic, Yasser
Minia University
Minia, Egypt
Taylor, Peter
National Concrete Pavement Technology Center
Ames, IA
Tazarv, Mostafa
University of Nevada, Reno
Reno, NV
Tegos, Ioannis
Salonica, Greece
Tehrani, Fariborz
California State University, Fresno
Fresno, CA
Tepfers, Ralejs
Ralejs Tepfers Consulting
Gteborg, Sweden
Tharmarajah, Gobithas
Belfast, UK
Thiagarajan, Ganesh
University of Missouri - Kansas City
Kansas City, MO
Thokchom, Suresh
Manipur Institute of Technology
Imphal, India
Thomas, Adam
Europoles gmbh
Neumarkt, Germany
Thompson, Phillip
Palm Desert, CA
Thorne, A.
Center of Engineering Materials and Structures
Guilford, UK
Thorstensen, Rein Terje
University of Agder
Grimstad, Norway
Tian, Ying
University of Nevada, Las Vegas
Las Vegas, NV
Tjhin, Tjen
Buckland and Taylor Ltd.
North Vancouver, BC, Canada
Tolentino, Evandro
Centro Federal de Educao Tecnolgica de Minas Gerais
Timteo, Minas Gerais, Brazil
Topu, lker
Eskiehir Osmangazi University
Eskiehir, Turkey
Torrenti, Jean-Michel
Chevilly-Larue, France
Tosun, Kamile
Dokuz Eylul University
Izmir, Turkey
Toubia, Elias
University of Dayton
Dayton, OH

260

Triantafillou, Thanasis
University of Patras
Patras, Greece
Tripura, Deb
NIT Agartala
Agartala, India
Trost, Burkhart
University of Applied Sciences and Arts Northwestern
Switzerland
Muttenz, Switzerland
Tsonos, Alexander
Aristotle University of Thessaloniki
Thessaloniki, Greece
Tsubaki, Tatsuya
Yokohama National University
Yokohama, Japan
Tsuruta, Hiroaki
Kansai University
Suita, Japan
Tuchscherer, Robin
Northern Arizona University
Flagstaff, AZ
Turanli, Lutfullah
Middle East Technical University
Ankara, Turkey
Turk, A. Murat
Istanbul Kultur University
Istanbul, Turkey
Tutikian, Bernardo
Unisinos
Porto Alegre, Rio Grande Do Sul, Brazil
Uygunoglu, Tayfun
Afyon Kocatepe University
Afyonkarahisar, Turkey
Uzal, Burak
Abdullah Gul University
Kayseri, Turkey
Vakhshouri, Behnam
University of Technology Sydney
Sydney, New South Wales, Australia
Van Deurzen, John
Van Deurzen and Associates PA
Overland Park, KS
Varum, Humberto
University of Porto
Porto, Portugal
Vasovic, Dejan
University of Belgrade
Belgrade, Serbia
Vatani Oskouei, Asghar
Shahid Rajaee University (BHRC)
Tehran, Islamic Republic of Iran
Vaz Rodrigues, Rui
EPFL
Lausanne, Switzerland
Vazquez-Herrero, Cristina
La Corua, Spain
Velzquez Rodrguez, Sergio
Universidad Panamericana
Zapopan, Jalisco, Mexico

ACI Structural Journal/March-April 2015


Vellalapalayam Nallagounder, Vijayakumar
Bannari Amman Institute of Technology
Erode, Tamilnadu, India
Velu, Saraswathy
CECRI
Karaikudi, Tamil Nadu, India
Venkatesh Babu, D. L.
Kumaraguru College of Technology
Coimbatore, Tamil Nadu, India
Venkiteela, Giri
New Jersey Department of Transportation
Trenton, NJ
Vercher, Jose
Polytechnic University of Valencia
Valencia, Spain
Vichit-Vadakan, Wilasa
CTLGroup
Skokie, IL
Villar Cocia, Ernesto
Central University of Las Villas
Santa Clara, Cuba
Vimonsatit, Vanissorn
Curtin University
Perth, Australia
Vintzileou, Elizabeth
National Technical University of Athens
Athens, Greece
Viviani, Marco
HEIG-VD
Yverdon les Bains, Switzerland
Vogel, Thomas
Institute of Structural Engineering
Zurich, Switzerland
Waldron, Christopher
University of Alabama at Birmingham
Birmingham, AL
Wan, David
Old Castle Precast Inc.
South Bethlehem, NY
Wang, Chang-Qing
Tongji University
Shanghai, China
Wang, Chong
Brisbane, Queensland, Australia
Wang, Huanzi
San Jose, CA
Wang, Kejin
Iowa State University
Ames, IA
Wang, Vincent
James Cook University
Townsville, Queensland, Australia
Wang, Xuhao
Ames, IA
Wang, Zhen Yu
Harbin Institute of Technology
Harbin, Heilongjiang, China
Watkins, Melanie
Michigan Technological University
Houghton, MI

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Wehbe, Nadim
South Dakota State University
Brookings, SD
Wei, Ya
University of Michigan
Ann Arbor, MI
Wei-Jian, Yi
Changsha, China
Weiss, Jason
Purdue University
West Lafayette, IN
Wen, Qingjie
China University of Mining & Technology
Xuzhou, Jiangsu, China
Werner, Anne
Southern Illinois University Edwardsville
Edwardsville, IL
Wheeler, Andrew
University of Western Sydney
Sydney, New South Wales, Australia
Wilson, William
Universite de Sherbrooke
Sherbrooke, QC, Canada
Windisch, Andor
Karlsfeld, Germany
Wood, Richard
University of California, San Diego
La Jolla, CA
Wu, Chenglin
Missouri S&T
Rolla, MO
Wu, Hui
Beijing, China
Wu, Hwai-Chung
Wayne State University
Detroit, MI
Wu, Yu-You
Dania Beach, FL
Wu, Yu-Fei
City University of Hong Kong
Kowloon, Hong Kong, China
Wu, Yu-You
Dania Beach, Florida
Xia, Jin
Zhejiang University
Hangzhou, Zhejiang, China
Xiang, Tianyu
Chengdu, Sichuan, China
Xiao, Yan
Hunan University
Changsha, Hunan, China
Xie, Guoshuai
Wuhan University
Wuhan, Hubei, China
Xin-hua, Cai
Wuhan University
Wuhan, Hubei, China
Xu, Aimin
ARRB Group
Melbourne, Victoria, Australia

261

REVIEWERS IN 2014
Xuan, D.X.
The Hongkong Polytechnic University
Kowloon, Hong Kong, China
Yahia, Ammar
Universit de Sherbrooke
Sherbrooke, QC, Canada
Yamada, Kanji
Akita Prefectural University
Yurihonjo, Japan
Yanez, Fernando
IDIEM University of Chile
Santiago, Chile
Yang, Kai
Belfast, UK
Yang, Keun-Hyeok
Kyonggi University
Suwon, Kyonggi-do, Republic of Korea
Yang, Kuochen
National Kaohsiung First University of Science and Technology
Kaohsiung, Taiwan, China
Yang, Xinbao
Olathe, KS
Yassein, Mohamed
Doha, Qatar
Yatagan, Serkan
Istanbul Technic University Architecture Faculty
Istanbul, Turkey
Yazbeck, Fouad
Readymix Abu Dhabi
Abu Dhabi, United Arab Emirates
Yazc, emsi
Ege University
zmir, Turkey
Yekrangnia, Mohammad
Sharif University of Technology
Tehran, Islamic Republic of Iran
Yerramala, Amarnath
Dundee University
Dundee, UK
Yeung, Jaime
Yue Xiu Concrete Co Ltd
Hong Kong, China
Yildirim, Hakki
Istanbul, Turkey
Yilmaz, Bulent
Bilecik Seyh Edebali University
Bilecik, Turkey
Ylmaz, Ali
KT
Trabzon, Turkey
Yindeesuk, Sukit
University of Illinois at Urbana-Champaign
Urbana, IL
Yourtcu, Erhan
Concrete Technology
zmir, Turkey
Yoon, Hyeong Jae
Taisei Corporation
Tokyo, Japan

262

Yoon, In-Seok
Induk University
Seoul, Republic of Korea
Yoon, Young-Soo
Korea University
Seoul, Republic of Korea
Youkhanna, Kanaan
University of Dohuk
Duhok, Iraq
Yu, Baolin
Michigan State University
East Lansing, MI
Yu, Haiyong
Shanghai Research Institute of Building Sciences
Shanghai, China
Yu, Jiangtao
Research Institute of Civil Engineering and Disaster Reduction
Shanghai, China
Yu, Tzu-Yang
University of Massachusetts Lowell
Lowell, MA
Yuan, Jiqiu
PSI, Turner-Fairbank Highway Research Center, FHWA
McLean, VA
Yuan, Xiaohui
Wuhan University of Technology
Wuhan, China
Yksel, Isa
Bursa Technical University
Bursa, Turkey
Yun, Hyun Do
Daejeon, Republic of Korea
Zaidi, S. Kaleem
Aligarh Muslim University
Aligarh, UP, India
Zaki, Adel
SNC - Lavalin
Montreal, QC, Canada
Zandi Hanjari, Kamyab
Chalmers University of Technology
Gothenburg, Sweden
Zanuy, Carlos
Universidad Politcnica de Madrid
Madrid, Spain
Zdiri, Mustapha
National Engineering School of Tunis
Tunis, Rades Tunis, Tunisia
Zeris, Christos
National Technical University of Athens
Zografou, Greece
Zhang, Jieying
National Research Council Canada
Ottawa, ON, Canada
Zhang, Jun
Tsinghua University
Beijing, China
Zhang, Peng
Karlsruhe Institute of Technology (KIT)
Karlsruhe, Germany

ACI Structural Journal/March-April 2015


Zhang, Y. X.
The University of New South Wales
Canberra, Australian Capital Territory, Australia
Zhang, Yamei
Southeast University
Nanjing, China
Zhang, Xiaogang
Shenzhen University
Shenzhen, Guangdong, China
Zhang, Xiaoxin
Universidad de Castilla-La Mancha
Ciudad Real, Spain
Zhao, Jian
University of Wisconsin Milwaukee
Milwaukee, WI
Zheng, Herbert
Gammon Construction Limited
Hong Kong, China
Zheng, Jianjun
Zhejiang University of Technology
Hangzhou, China

ACI Structural Journal/March-April 2015

REVIEWERS IN 2014
Zheng, Yu
Dongguan University of Technology
Dongguan, Guangdong, China
Zhou, Changdong
Beijing Jiaotong University
Beijing, China
Zhou, Wei
Harbin Institute of Technology
Harbin, China
Zhou, Xiangming
Brunel University
Uxbridge, UK
Ziehl, Paul
University of South Carolina
Columbia, SC
Zilch, Konrad
Technische Universitat Munchen
Munich, Germany

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.

You might also like