Report on Indirect

Method to Obtain Stress­

Strain Response of Fiber­
Reinforced Concrete
(FRC)
Reported by ACI Committee 544

'
I

0:::
00 •

�
�
L[)

u
<( �
� acI •
�
American Concrete Institute
Always advancing
 First Printing
American Concrete Institute
Always advancing
March 2016
ISBN: 978-1-942727-72-9

Report on Indirect Method to Obtain Stress-Strain Response of Fiber-Reinforced

Concrete (FRC)

Copyright by the American Concrete Institute, Farmington Hills, MI. All rights reserved. This material
may not be reproduced or copied, in whole or part, in any printed, mechanical, electronic, film, or other
distribution and storage media, without the written consent of ACI.

The technical committees responsible for ACI committee reports and standards strive to avoid
ambiguities, omissions, and errors in these documents. In spite of these efforts, the users of ACI
documents occasionally find information or requirements that may be subject to more than one
interpretation or may be incomplete or incorrect. Users who have suggestions for the improvement of
ACI documents are requested to contact ACI via the errata website at http://concrete.org/Publications/
DocumentErrata.aspx. Proper use of this document includes periodically checking for errata for the most
up-to-date revisions.

ACI committee documents are intended for the use of individuals who are competent to evaluate the
significance and limitations of its content and recommendations and who will accept responsibility for
the application of the material it contains. Individuals who use this publication in any way assume all
risk and accept total responsibility for the application and use of this information.

All information in this publication is provided "as is" without warranty of any kind, either express or
implied, including but not limited to, the implied warranties of merchantability, fitness for a particular
purpose or non-infringement.

ACI and its members disclaim liability for damages of any kind, including any special, indirect, incidental,
or consequential damages, including without limitation, lost revenues or lost profits, which may result
from the use of this publication.

It is the responsibility of the user of this document to establish health and safety practices appropriate
to the specific circumstances involved with its use. ACI does not make any representations with regard
to health and safety issues and the use of this document. The user must determine the applicability of
all regulatory limitations before applying the document and must comply with all applicable laws and
regulations, including but not limited to, United States Occupational Safety and Health Administration
(OSHA) health and safety standards.

Participation by governmental representatives in the work of the American Concrete Institute and in
the development of Institute standards does not constitute governmental endorsement of ACI or the
standards that it develops.

Order information: ACI documents are available in print, by download, on CD-ROM, through electronic
sub�cription, or reprint and may be obtained by contacting ACI.

Most ACI standards and committee reports are gathered together in the annually revised ACI Manual of
Concrete Practice (MCP).

American Concrete Institute

38800 Country Club Drive
Farmington Hills, MI 48331
Phone: +1.248.848.3700
Fax: +1.248.848.3701
www.concrete.org

Coovriaht American Concrete Institute

 ACI 544.8R-16

Report on Indirect Method to Obtain Stress-Strain

Response of Fiber-Reinforced Concrete (FRC)

Reported by ACI Committee 544

Barzin Mobasher't, Chair Clifford N. MacDonald, Membership SecretarY

Neven Krstulovic-Opara', Secretary

Carina-Maria A Idea Rishi Gupta Max L. Porter Consulting Members

Emmanuel K. Attiogbe George C. Hoff Venkataswamy Ramakrishnan P.N. Balaguru
Nemkumar Banthia John Jones Klaus Alexander Rieder Hiram Price Ball Jr.
Joaquim Oliveira Barros' David A. Lange Pierre Rossi Arnon Bentur
Gordon B. Batson Maria Lopez de Murphy Surendra P. Shah Andrzej M. Brandt
Peter H. Bischoff Michael A. Mahoney Flavia de Andrade Silva James I. Daniel
Jean-Phi Iippe Charron Bruno Massicotte Kay Wille Sidney Freedman
Xavier Destree Christian Meyer Robert C. Zellers Melvyn A. Galinatl
Ashish Dubey James Milligan Lihe Zhang Henry J. Malloy
Mahmut Ekenel Nicholas C. Mitchell Jr. Antoine E. Naaman
Liberato Ferrara Gerald H. Morton *Members of the subcommittee that

Gregor D. Fischer Antoine E. Naaman* authored this report.

IChair of the task group that drafted
Dean P. Forgeron Jeffrey L. Novak
this report.
Antonio Gallovich Giovanni A. Plizzari
:Deceased.

Development of proper design procedures for fiber-reinforced
concrete (FRC) materials requires use of material tensile and
compressive stress strains that reflect the contribution offibers to
the post-cracking behavior. While uniaxial tension tests provide
the most fundamental material properties, conducting closed-loop
tension tests are difficult to accomplish; therefore, methods based
on indirect measurement of tensile properties using flexural tests
are typically used.
This report presents the methodologies that are used for data
reduction and presentation of the flexural test results in terms of
an equivalent tensile stress-strain response for FRC materials.
Existing methods for estimating uniaxial tensile stress-strain
response of strain-softening and hardening FRCs from flexural
beam-test data are introduced. Different approaches applied to
beam tests based on elastic equivalent, curvefitting, or back-calcu-
ACI Committee Reports, Guides, and Commentaries are
intended for guidance in planning, designing, executing, and
inspecting construction. This document is intended for the use
of individuals who are competent to evaluate the significance
and limitations of its content and recommendations and who
will accept responsibility for the application of the material it
contains. The American Concrete Institute disclaims any and
all responsibility for the stated principles. The Institute shall
not be liable for any loss or damage arising therefrom.
Reference to this document shall not be made in contract
documents. If items found in this document are desired by
the Architect/Engineer to be a part of the contract documents,
they shall be restated in mandatory language for incorporation
by the Architect/Engineer.
..... ...
lation of flexural data are introduced. These are divided into two
general categories: elastic equivalent approach or inverse analysis
method. In the elastic equivalent approach, a summary of available
test methods by various code agencies are presented.
Using back-calculation methods, tools based on the finite
element method and analytical closed-o f rm solutions are presented.
An approach is presented that uses closed-form moment-curvature
relationships and obtains load-deflection responses for a beam of
three- or four-point loading. The method is used to obtain equiva­
lent parametric tensile stress and strain relationships for a variety
of FRC materials. The methods are compared against the available
residual strength and also elastically equivalent residual strengths
obtained by different specimen geometries.
Results for a range of FRC materials studied show the back­
calculated post-peak residual tensile strength is approximately 30
to 37 percent of the elastically equivalent flexural residual strength
for specimens with different fiber types and volume fractions.
Keywords: fiber-reinforced concrete; inverse analysis; tensile stress-strain
diagram.
ACI544.8R-16 was adopted and published March 2016.
Copyright © 2016, American Concrete Institute.
All rights reserved including rights of reproduction and use in any form or by any
means, including the making of copies by any photo process, or by electronic or
mechanical device, printed, written, or oral, or recording for sound or visual reproduc­
tion or for use in any knowledge or retrieval system or device, unless permission in
. \V�iting i_s_�bt?i�e�_ from the copyright proprietors.

Coovriaht American Concrete Institute �.

 2 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
CONTENTS
CHAPTER 1-INTRODUCTION AND SCOPE, p. 2
1.1-Introduction, p. 2
1.2-Scope, p. 3
CHAPTER 2-NOTATION AND DEFINITIONS, p. 3
2.1-Notation, p. 3
2.2-Definitions, p. 4
CHAPTER 3-TENSILE AND FLEXURAL TESTING
AND MECHANICAL PROPERTIES, p. 4
3 .1-Tension and flexural testing, p. 4
3 .2-Strain softening and hardening, p. 4
3 .3-Deflection softening and hardening, p. 5
3 .4-Equivalent tensile stress-strain responses, p. 6
3 .5-Inverse analysis methods, p. 6
CHAPTER 4-TEST METHODS, p. 6
4.1-Test and specimen types, p. 6
4.2-Stress-strain diagrams, p. 7
4.3-Stress-strain diagram in RILEM approach, p. 7
4.4-Flexural tensile strength frJI and residual flexural
tensile strengths.fR.1 and/R,4, p. 7
4.5-Relationship between uniaxial tensile stress and
flexural strength, p. 8
4.6-Tensile stress-strain diagrams and strain values s1,
�>2, and s3, p. 8
CHAPTER 5-STRESS-STRAIN CURVES BY
BAC K-CALCULATION APPROACH, p. 10
5 .!-Parametric stress-strain curves, p. 10
5.2-Back-calculation of flexural test data, p. 12
5.3-Comparison with averaged residual strength results,
p. 12
5.4-Comparison with RILEM and JCI methods, p. 13
CHAPTER 6-CONCLUSIONS, p. 1 3
CHAPTER 7-REFERENCES, p. 14
Authored documents, p. 14
A PPENDIX A-SPREADSHEET-BASED INVERSE
ANALYSIS PROCEDURES, p. 1 6
A. l -Simplified strain-softening/hardening fiber-rein-
forced concrete model, p. 16
A.2-Derivation of moment-curvature diagram, p. 17
A.3-Derivation of nominal flexural strength, p. 17
A.4-Simplified moment curvature diagram, p. 19
A.5-Load-deflection response, p. 20
A.6-Example: Three-point bending test, p. 21
CHAPTER 1-INTRODUCTION AND SCOPE
1.1 -lntroduction
This report provides guidelines for obtaining uniaxial
stress-strain curves of fiber-reinforced concrete (FRC) from
Coovriaht Am�te Institute
beam test data. In many FRC systems, the contribution of
fibers is apparent after the concrete cracks and the fibers that
bridge such cracks start to debond and pullout, thus resisting
its opening and generating a force that transfers the loads
across the crack. The magnitude of load carried by the fibers
depends on the opening of the crack width; it can be normal­
ized with respect to the cracked area and referred to as a
residual strength. The role that fibers play in bridging a main
tensile crack is therefore characterized as resisting crack
opening, also referred to as bridging force and is represented
as an average effective stress and described by a tensile
stress-crack width relationship. A majority of FRC mixtures
exhibit a distinct stress-crack width relationship that can
also be integrated with the initial elastic response of the
composite into a combined nonlinear stress-strain response.
The stress-crack width relationship can be represented as
an equivalent stress-strain response by assuming a character­
istic length parameter to smear the crack width into a nominal
strain distribution. If mechanical tests that only focus on the
tensile strength (ASTM C496/C496M) or flexural strength
(ASTM C78/C78M) of the FRC are conducted, this contri­
bution, herein referred to as residual strength, is either inac­
curate, not reported, or reported in terms of parameters that
may not be useful for design or analysis. Furthermore, accu­
rate measurements of tension tests that capture the post-peak
response are difficult to conduct; therefore, many agencies
use flexural tests as an indication of tensile response.
While the tension test theoretically shows the true mate­
rial behavior and the flexural test represents a structural
response, a flexural test is often used as a means of property
measurement. The difference between the tension and flex­
ural test results of many FRC materials is that in a tension
test, the post-peak tensile stress-crack width response does
not influence the maximum load obtained by the member. In
a flexural test, however, the maximum load can be directly
related to the residual stress levels such that the overall
behavior can be affected by the post-peak response.
An alternative method is to calculate post-cracking
behavior using the experimental flexural results and reduce
them into a set of material parameters that are in compliance
with the model assumptions. This topic is the subject of this
report. The testing methodologies are discussed in detail in
committee reports such as ACI 544.3R. The present report
addresses procedures to obtain an effective tensile stress­
strain curve from the experimental results.
Many structural systems that use FRC, such as structural
floors or indeterminate structures, can exhibit an increase in
strength values in proportion to the residual strength, which
is a direct contribution of fibers; however, this parameter is
only measured in a qualitative way using flexural tests such
as ASTM C l 609/C l 609M.
To develop and apply design procedures for FRC mate­
rials, simplified equations are needed to account for the
fiber's contribution to the tensile response, especially after
cracking has occurred. This report addresses methods to
compute the stress transfer after cracking is initiated in a
concrete section.
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 3

1.2-Scope F force component in stress diagram, lb (N)

This report presents ex1stmg methods for estimating fc' uniaxial compressive strength, psi (MPa)
characteristic tensile stress-strain or tensile stress crack /eq equivalent flexural tensile strength parameter, psi
width response of strain-softening fiber-reinforced concrete (MPa)
(FRC) using flexural beam test data. Methods are proposed ft tensile strength, psi (MPa)
for strain-softening FRCs that do not exhibit distributed or fR residual flexural tensile strength parameter, psi
parallel microcracking when tested in flexural loading condi­ (MPa)
tions, and strain-softening FRCs that do exhibit distributed h, net section height, in. (mm)
or parallel microcracking when tested in flexural loading hune = distance between notch tip and cross-section top
conditions. k neutral axis depth ratio
A set of definitions for an equivalent stress-strain diagram L clear span, in. (mm)
(Naaman and Reinhardt 2006; Noghabai 1998) are presented Lf fiber length, in. (mm)
first and followed by calculation procedures for obtaining LP plastic length for crack localization in beams, in.
flexural tensile and residual flexural strengths from beam (mm)
test data. Specific coefficient values for notched beams M moment, in.-lb (N.mm)
(RILEM TC 162-TDF 2003) and third-point beams (NBN Mer moment at first cracking, in.-lb (N.mm)
B 15 238:1992) are validated. Because the coefficient values Me moment in elastic range, in.-lb (N.mm)
are not directly reported for beam types tested per ASTM M;1 moment at an intersection point, in.-lb (N.mm)
C1609/C1609M and ASTM C l 399/C1399M, this report Mn nominal flexural strength, lb-ft (N.mm)
proposes an approach to do so and compares the results Afr, nominal flexural strength at maximum deflection/
with other methods. The relationship is presented in terms curvature, in.-lb (N.mm)
of parameter-based stress coefficients that are determined Per total load at first cracking, lb (N)
using a step-by-step inverse analysis procedure in Appendix vf volume fraction of fiber
A. The report concludes with the relationship between the a normalized transitional strain (1':1,../!':er)
parameters that define the stress-strain diagram and the �erit = normalized critical tensile strain
experimental flexural residual strengths. ���� normalized tensile strain at maximum stress (1':1,/!':cr)
The proposed approach has several drawbacks, as it is for 811 maximum deflection, in. (mm)
a one-dimensional material behavior model, and yields an f:e compressive strain
effective stress-strain response based on the model assump­ !':cr first cracking tensile strain
tions of trilinear tension and bilinear compression. Further­ f:eu maximum compressive strain
more, because the results of back-calculation are size- and f:cy compressive strain at yielding
geometry-dependent, the procedure is presented for test data !':; strain value at a point along the stress strain
obtained from ASTM C1609/C1609M specimens. Other response, in./in. (mm/mm)
geometries and specimen dimensions may need to be corre­ !':peak= strain value corresponding to tensile strength, in./
lated with standard-sized specimens. The method can be in. (mm/mm)
applied to any size specimen and the results can be used for f:t tensile strain
comparative basis. Different beam sizes may yield different !':If.,. transitional tensile strain (a x !':cr)
stress-strain values because of the size effect. The subject 1':111 maximum tensile strain
of the size effect is not addressed in this document because q, curvature, 1/in. (1/mm)
the majority of the flexural test specimens are in the narrow <Per curvature at first cracking, 1/in. (1/mm)
range of 4 x 4 to 6 x 6 in. (100 x 100 to 150 x 150 mm) <Pit curvature at an intersection point, 1/in. (1/mm)
size. Comprehensive sets of data with different fiber types <Pu maximum curvature, 1/in. (1/mm)
and specimen sizes have been analyzed using back-calcula­ 11 ratio of Ee,JEc
tion procedures, and a comparative evaluation of the results A normalized top compressive strain (f:ci!':er)
are discussed in recent publications (Bakhshi et a!. 2014; Aeu normalized maximum compressive strain (f:e,/!':er)
Mobasher et a!. 2014). A111 normalized compressive strain when bottom fiber
reaches maximum tensile strain
CHAPTER 2-NOTATION AND DEFINITIONS ll normalized post-peak residual tensile strength
llcrit = normalized post-peak residual tensile strength for
2.1 -Notation elastic perfectly plastic moment-curvature response
b beam width, in. (mm) Gc compressive stress, psi (MPa)
c1, c2, c3, C; Gcy compressive yield strength, psi (MPa)
constants for definition of tensile stress strain curve G; stress value at a point along the stress strain
d beam depth, in. (mm) response, psi (MPa)
E Young's modulus, psi (MPa) G1 tensile stress, psi (MPa)
Ee Young's modulus, psi (MPa) co compressive-to-tensile strength ratio (f:cyEI!':erE)
Eer equivalent Young's modulus of a cracked tension
member, psi (MPa)
Coovriaht American Concrete Institute �-
 4 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
2.2-Definitions
ACI provides a comprehensive list of definitions through
an online resource, " ACI Concrete Terminology," https://
www.concrete.org/store/productdetail.aspx?ItemiD=CT13 .
Definitions provided herein complement that resource.
bend-over point-tensile stress measured at the onset of
first cracking (in a uniaxial tension test); this point is consid­
ered as tensile yield point or limit of proportionality.
crack bridging-ability of the fibers to extend from one
face of a crack to the opposite side and transfer load during
crack opening to the opposite face.
crack mouth opening displacement-measure of crack
length extension beyond the tip of a notch created at the
center of a beam specimen subjected to flexural loads in a
three-point bending configuration.
deflection hardening-multiple cracking accompanied
by an increase in flexural strength after first cracking due to
strain hardening of the bridging material.
deflection softening-flexural response with a primarily
linear elastic ascending branch up to the first flexural
cracking, followed by an immediate reduction in load with
increasing deflection.
first crack strength-stress corresponding to the load
at first crack for a fiber-reinforced concrete composite in
bending or tension.
flexural toughness-area under the flexural load-deflec­
tion curve obtained from a static test of a specimen up to
a specified deflection, which is an indication of the energy
absorption capability of a material.
limit of proportionality-flexural stress measured at the
onset of first cracking (in a bending test).
microcrack coalescence-process by which microcracks
grow sufficiently to join together and form a larger, perhaps
visible, macrocrack.
microcracking-formation of localized, microscopic
cracks within concrete, approximately 0.04 in. (0.1 mm)
wide up to a few mm (less than 0.16 to 0.20 in. [4 to 5 mm])
long that may not be visible.
progressive collapse-failure of a primary structural
element that results in the failure of adjoining structural
elements, leading to further structural failure.
residual strength-apparent strength in the post-peak
load region of a static load-deflection curve, calculated by
the ratio of the bending moment in the post-cracking stage at
a given deflection by the section modulus of original section.
serviceability limit state-performance criterion for
serviceability of a structure, corresponding to conditions
beyond which the structure might fail to meet technical
requirements such as stress limits, deformation limits,
and crack control and, therefore, declared unfit for routine
loading conditions.
smeared crack-formation of multiple cracks spread
over a distinct area.
stable crack propagation-incremental and sequen­
tial growth of a crack as a result of addition of load on the
specimen
strain softening-process where unstable crack propaga­
tion directly follows crack initiation.
Coovriaht Am�te Institute
strain softening fiber-reinforced-fiber-reinforced
concrete composite with a primarily linear elastic ascending
tensile stress-strain curve up to the same point representing
both the first cracking and peak stress, followed by a reduc­
tion in stress as the strain is increased.
strain hardening fiber-reinforced-fiber-reinforced
concrete composite with a primarily linear elastic ascending
tensile stress-strain curve up to the first cracking, microcrack
coalescence, and a post-elastic portion characterized by an
increase in the stress with a much lower effective stiffness
(representing tension stiffening) up to the maximum stress.
unstable crack propagation-uncontrolled and dynamic
extension and widening of a crack that is not solely depen­
dent on additional load on the specimen.
CHAPTER 3-TENSILE AND FLEXURAL TESTING
AND MECHANICAL PROPERTIES
3.1-Tension and flexural testing
Fiber-reinforced concrete (FRC) materials have been
extensively studied under direct tension loading conditions.
Due to brittleness of the material, problems in conducting
stable tension tests that must be addressed include:
a) Inherent instability of crack propagation from one side
of the specimen to the other that changes the uniformity of
applied deformation
b) Difficulty in control testing protocol and control mode
due to brittleness and sudden crack growth in the material
c) Test results that are significantly influenced by the
support boundary conditions of fixed or free, specimen size,
and notched versus unnotched specimens (Van Mier 1996)
d) Maintaining objectivity in the closed-loop control
conditions, such as cracking symmetry for notched speci­
mens (Mobasher 2011; Hordijk 1991)
3.2-Strain softening and hardening
Many of the physical, durability, and mechanical prop­
erties of fiber-reinforced concrete (FRC) are affected by
the amount, type, bond, and nature of anchorage (straight,
hooked, twisted, or fibrillated) of fibers used. Other criteria
that are affected by the interaction of the fiber on the stress­
strain response include, but are not limited to, strength, strain
capacity, first crack strain, elastic modulus, crack width,
durability, permeability, total energy absorption, fatigue life,
thermal conductivity, and shrinkage and creep characteris­
tics. Therefore, characterization of the tensile stress-strain
response is an important parameter of the analysis and
design of FRC structures.
Strain-softening behavior in FRC represents tensile
response with a primarily elastic ascending stress-strain
curve up to the point that is both the first cracking bend­
over point and also ultimate tensile strength. Response is
followed by unloading due to unstable cracking, and opening
of a crack with partial resistance offered by the debonding
and sliding pullout of fibers. All subsequent stresses remain
smaller than the first cracking (maximum) stress. In a strain­
softening material, therefore, formation of a crack under
loa.d is_directlyfollo'-Yt;d.by its sudden and self-driven propa-
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
Stress
toad
(a) Tensile Response
Fig. 3.2-Classi.fication of fiber-reinforced concretes (FRCs) based on their tensile
response (after Naaman and Reinhardt [2006]). (Notes: BOP is bend over point; UTS is
ultimate tensile strength; MOR is modulus ofrupture; and LOP is limit ofproportionality.)
gation because the material is quasi-brittle with low bridging
stress. Hence, the sample loses its ability to carry load and
the stress-strain response goes through a descending branch,
as shown in Fig. 3 .2. The post-cracking stress remains below
the peak stress in subsequent loading (Naaman and Rein­
hardt 2006); therefore, the softening behavior does not affect
tensile strength, but increases tensile ductility.
Strain hardening behavior in FRC is a behavior repre­
sented by an elastic ascending tensile stress-strain curve up
to the bend-over point, followed by an increasing load as
the strain is increased, as shown in Fig. 3 .2a. The effective
stiffness in the response after the bend-over point level is
much lower than the elastic stiffness (representing tension
stiffening). The response continues at this stiffness up to the
nominal tensile strength (ultimate tensile strength). In this
case, the magnitudes of first cracking and nominal tensile
strength may be distinctly different. Response after the ulti­
mate tensile strength is followed by gradual unloading.
The strain hardening occurs under condition with a suffi­
ciently high fiber content, stiffness, and bond properties
such that the composite tensile strength exceeds the matrix
cracking strength. In this process, after initial formation,
a stable crack propagation stage takes place and the mate­
rial could behave as perfectly-plastic or exhibit hardening
behavior. Because the sample is capable of supporting addi­
tional load, multiple parallel cracks may also be formed and
the tensile and flexural load deformation responses after
cracking are characterized by an ascending load-deforma­
tion branch up to the nominal strength (Fig. 3 .2b).
3.3-Deflection softening and hardening
Flexural behavior may vary significantly depending on
whether fiber-reinforced concretes (FRCs) exhibit tension
strain-softening or strain-hardening response. An approach
to classify FRC based on its tensile mechanical response
from the uniaxial tension can also be extended to flexural
behavior. This classification depends on the efficiency and
amount of fibers that covers a range from strain softening to
hardening (Naaman and Reinhardt 2006).
Coovriaht American Concrete Institute
5
--- Plain Concrete
--- Strain softening Deflection Softening FRC
Strain softening Deflection hardening FRC
Strain Hardening FRC
Deflection 0 mu
(b) Flexural Response
It is noted that many of the characteristics of strain­
softening and strain-hardening responses in tension testing
are also observed when the specimen is tested in flexure.
The responses in these conditions are referred to as deflec­
tion softening and deflection hardening; however, there is
no one-to-one correlation between the two modes. Deflec­
tion hardening occurs for samples that have a sufficiently
high level of fibers to exhibit strain-softening behavior in
tension, but hardening behavior under flexural testing. The
post-crack response in flexure exhibits ascending load with
increasing deflection after the first crack point. For example,
if the volume of the fibers is increased in excess of a level
designated as critical fiber content, then the entire stress­
strain response is also affected because the manner of load
and strain distribution change in the sample. This is because,
as the plain matrix cracks, there are sufficient fibers to carry
the load that is being released by the matrix.
Because the flexural design of FRC elements is
normally governed by the tensile stress-strain response,
the post-cracking behavior has a direct effect on the results
.
(Soranakom and Mobasher 2009b) Proper characteriza­
tion of the load-deformation response and computation of
valid material behavior data from these tests is essential to
successful design applications (Banthia and Trottier 1995a,b;
Mobasher 2011; Cunha 2010). Many structural systems
with strain-softening materials, such as structural floors or
indeterminate structures, can exhibit deflection-hardening
behavior and an increase in strength values in proportion to
the residual strength.
The difference between the tension and flexural test results
of strain-softening materials is that, in a tension test, the post­
peak tensile stress-crack width response does not influence
the maximum load obtained. In a flexural test, the maximum
load attained is related to the softening stress magnitudes.
While a strain-softening FRC can develop deflection soft­
ening or deflection hardening based on the residual stress
level present, FRC with tensile strain-hardening behavior
always exhibits a deflection-hardening response.
�-
 6 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
3.4-Equivalent tensile stress-strain responses
As an alternative to direct measurement of the stress-strain
by uniaxial tension test, a back-calculation approach can be
applied to flexural tests. This is attractive, as flexural tests are
routinely conducted as quality control tests and, therefore, the
data collected can be used to verify the material properties and
further used in the design of the next sections. Back-calcu­
lation of equivalent tension results from flexural tests, also
defined as inverse analysis, is a useful tool because it provides
equivalent tension response needed in member design.
Approaches have been developed by RILEM TC 162-TDF
(2003) and DIN EN 14561:2006-08. The equivalent tensile
response can then be used in the design of flexural members
for applications such as slabs, panels, and other applications
where fiber contribution is introduced using equivalent soft­
ening parameters. Different approaches do not always yield
unique results, as assumptions regarding the shape of the
stress-strain response inherently affect the back-calculated
parameters and, therefore, several test methods for calcu­
lation of the equivalent tensile properties are summarized.
An overview of the existing inverse methods for obtaining
uniaxial tensile stress-strain curves from beam test data
follows in 3 .5.
3.5-lnverse analysis methods
There is no standard specimen geometry for flexural testing
among the various test methods of RILEM TC 162-TDF
(2003), CEB-FIP (2011), and DIN EN 14561:2006-08.
These approaches use notched three-point bending tests
loaded along the midspan. ASTM methods such as ASTM
C1609/C1609M and C l 399/C l 399M, however, recommend
unnotched four-point beam tests. Based on these recom­
mended standards, tensile stress-strain response can be
calculated using prescribed equations. Values of the coef­
ficients of these equations are determined from the experi­
mental'. data using an equivalent elastic or inelastic method.
The inverse-analysis method is optimal for finding values
when specific coefficient values are unknown and include:
a) Technical guidelines for FRC, developed by RILEM
TC 162-TDF (2003) for steel fiber-reinforced concrete
(SFRC) by introducing the concept of equivalent feq and
residualfR flexural tensile strength parameters. These values
were u·sed to derive the stress-strain or the stress-crack width
response of an SFRC (Namman and Reinhardt 2006; DIN
EN 14561:2006-08; RILEM TC 162-TDF 2000a,b).
b) CEB-FIP (2011) also adopts the concept ofresidual flex­
ural tensile strength parameters to derive the constitutive law
that characterizes the tensile behavior of strain-softening and
strain-hardening FRCs. The code proposes two simplified
stress-crack width models based on the rigid-plasticity and
linear-elasticity tensile behavior of FRCs. Correction factors
are applied to scale the equivalent residual flexural strength
at serviceability and nominal states, which are determined
from three-point bending tests conducted based on BS EN
14651:2005+ A1:(2007) (Blanco et a!. 2013; Bakhshi et a!.
2014).
c) German guidelines similar to RILEM TC 162-TDF
(2003) for design of flexural members use the strain compat-
Coovriaht Am�te Institute
ibility analysis to determine the flexural strength of SFRC
members (Teutsch 2004).
d) The UK traditionally followed JCI SF4:1983, but
recently shifted toward the RILEM design method­
ology (Barr and Lee 2004) and methods based on BS EN
14651 :2005+ A1:(2007).
e) The Italian guidelines (di Prisco et a!. 2004) are based
on load-crack tip opening displacements from four-point
bending tests on notched beams using equivalent stresses.
f) The Scandinavian approach is also based on residual
strength parameters (Silfwerbrand 2008).
g) The U.S. design guidelines for flexural members are
based on empirical equations of Swamy et a!. (1975), ASTM
C1609/C1609M, ASTM C1399/C l 399M, Banthia and
Dubey (1999), and Fischer (2004). The residual strength
parameter is obtained using elastic section properties; the
effect of fiber type and concrete composition are not speci­
fied. ACI 544.6R and ACI 544.7R have initiated the use of
residual tensile strength as a parameter in the design.
h) Soranakom and Mobasher (2007c, 2008, 2009a)
proposed closed-form solutions to correlate the flexural load
deflection to a simplified tensile stress strain response-a
procedure applicable to a wide variety of design applica­
tions, such as one- and two-way slabs, beams, slabs-on­
ground, and earth-retaining structures.
In a majority of these cases, the uniaxial compressive
response is determined directly from standard compres­
sion cylinder tests. Flexural tests conducted by BS EN
14651:2005+ A1:(2007) are recommended for characteriza­
tion of tensile behavior. The CEB-FIP (2011) advises against
using uniaxial tensile tests for standard testing of new
mixtures due to the associated difficulty of execution and
interpretation (Blanco et a!. 2013). These procedures can
also be modified for a serviceability level design criteria. The
long-term behavior of cracked FRC under tension, however,
can be developed and required for structural fibers, such as
organic or natural fibers, whose long-term performance is
influenced by creep. CEB-FIP (2011) specifies if certain
requirements about characteristic residual strengths and the
limit of proportionality (BS EN 14651:2005+ A1:(2007))
are fulfilled, then fiber reinforcement can substitute conven­
tional reinforcement at serviceability limit.
CHAPTER 4-TEST METHODS
4.1-Test and specimen types
Various test methods exist for direct or indirect evaluation
of the uniaxial-tensile properties of fiber-reinforced concrete
(FRC). These include uniaxial tensile, wedge splitting, and
beam tests. Common flexural beam tests are performed
under a three- or four-point layout:
a) Three-point load recommended in RILEM TC 162-TDF
(2003), CEB-FIP (2011), and ASTM C293/C293M
b) Four-point load recommended in ASTM C78/C78M
c) Four-point load for FRC in ASTM C1399/C l 399M,
ASTM C1609/C1609M, Teutsch (2004), JCI SF4:1983 , and
NBN B 15 238:1992
Beam tests can be subdivided by beam type:
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 7

a) Unnotched, such as those used by ASTM

b) Notched, such as those used with the three-point load
test by RILEM TC 162-TDF (2003) and CEB-FIP (2011)
Controlling the post-peak response is much more depend­
able using a crack-mouth-opening displacement (CMOD) 3.51 2.0: E [o/oo)
c
in a notched beam as opposed to deflection control used in I I
I I
unnotched beams. The CMOD measured in a notched beam I I
I I
is a monotonic increasing variable, which provides stable I I
I I
testing conditions while deflection includes elastic defor­ I I
mation outside the fracture process zone, which may cause Compression Tension
--
unstable tests if extra caution is not considered (Hordijk
1991 ). Further detailed description of each beam type and Fig. 4. 3a-Stress-strain diagram for FRC in uniaxial
test procedure is presented in the corresponding standards tension and compression (RILEM TC 1 62-TDF 2003).
and, therefore, beyond the scope of this report.

4.2-Stress-strain diagrams
There are various ways to represent the residual strength of
a sample after the point of maximum load. In each case, the
model should include the localization effect within the crack
region where all incremental deformation is concentrated,
while the remainder of the specimen undergoes unloading
in an elastic or inelastic manner. The average strain over the
descending response is affected by the definition of a char­
8[mm]
acteristic length parameter as a material property, as crack
opening is represented by a smeared crack. Existing soft­
ening responses are subdivided into two main categories: a)
1) Stress-crack width definition F[N]
2) Stress-strain approach, where the strain is defined in 1
terms of average displacement over a predetermined char­
acteristic length or, alternatively, with a smeared crack
approach (di Prisco et al. 2004; CEB-FIP 2011)

4.3-Stress-strain diagram in RILEM approach

The stress-strain diagram shown in Fig. 3 .2 is proposed
by RILEM TC 162-TDF (2003). The values that define
this constitutive model are based on a piecewise linearized
response in the post peak domain. Average or characteristic
values are then used later for design purposes.
In Fig. 4.3a, the key points of the compression side of the
diagram are taken directly from the standard compressive
cylinder test. For the tension side of the diagram, the key
points are:
a) cr1 and E1-Tensile strength and corresponding strain
b) cr2 and E2-Stress and strain at the onset of the stable
strain-softening branch
c) cr3 and E3-Stress and strain at the end of the softening
branch
The values of <J; and E; (i = 1, 2, 3) can next be directly
related to beam test data using the approach presented in
RILEM TC 162-TDF (2000a,b; 2002a,b) and Barros et al.
(2005). In this case, <J; (i = 1, 2, 3) are related to the following
characteristic points on the beam-test load displacement
curve (Fig. 4.3b):
a) Average value of the flexural tensile strength, ftmJI, for
example, is the limit of proportionality of the beam response
b) Residual flexural tensile strength at deflection of 1/64
in. (0.46 mm)JR,I
b)
Fig. 4.3b-Load-displacement response ofa test beam used
,
to evaluate: (a) fR.J from FR,1; and (b) A4from F R 4(RILEM
TC 1 62-TDF 2000a, b; 2002a, b; Barros et al. 2005).
c) Residual flexural tensile strength at deflection of 1/8 in.
(3.0 mm)JR,4
Details on how to evaluateftmJI,.JR,�, andfR,4 are presented in
4.4 and details on <J;, E;,ft111J1,./R,1, andfR,4 are provided in 4.5.
4.4-Fiexural tensile strength f1,,1 and residual
flexural tensile strengths fR,1 and fR,4
The flexural tensile strengthftJI and residual flexural tensile
strengthsfR,I andfR,4 values are calculated directly from the
load displacement response of the test beam, which is sche­
matically represented in Fig. 4.3b (DR,I = 1164 in. [0.46 mm]
and 8R,4 = 1/8 in. [3.0 mm]).
4.4.1 Flexural tensile strength f1,R-The flexural tensile
strengthft JI is defined as the maximum flexural tensile stress

Coovriaht American Concrete Institute �-

 8 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)

IF
within the critical section at a load level equal to the limit of
proportionality, FL. The limit of proportionality is the load at
"
the end of the linear-elastic response, as shown in Fig. 4.3b. 4
When there is no clear end-point to the linear-elastic portion <J

of the load-displacement curve, the limit of proportionality hsp

is defined as the highest load up to a deflection of 0.002 in. 4 I <J
4
(0.05 mm) (RILEM TC 162-TDF 2002a,b). ,j <J
L ....1 4
4 l_
<J
Similar to the modulus of rupture J,. of conventional
concrete, the flexural tensile strength j;JI is calculated "
!8
assuming linear stress distribution within the critical section. i _ar-f!.La�ti.c_ b.LQ.c_k-J-_ s
�Li.!!J _ -J-L.i.!!_f!.ar-elasl(c
_b.Ll1.ck--1

When a three-point beam bending test is used, L and b define Non-Linear

span and width of the test beam, respectively; h 11 defines the hin�e

net section height; and the flexural tensile strength j;JI is

calculated as
(4.4.1)
If a notched test beam is used, the net section height h 11
equals the distance between the notch tip and cross section
top (h ) or defined as h sp in Fig. 4.6.1.1. Alternatively, if
u11c , and tension
the loading geometry changes from three-point bending to
four-point bending, the parameters of the equation should be
changed accordingly.
4.4.2 Residual flexural strengths fR,1 and fR,.r-Residual
flexural tensile strengths /R,l and /R,4 are taken from beam
loads FR,l and FR,4 that correspond to beam deflections of
1/64 and 1/8 in. (0.46 and 3 . 0 mm), respectively (RILEM
TC 162-TDF 2002a,b), as shown in Fig. 4.3b. However, at
this deflection level, the stress distribution is no longer linear
within the critical section and the uncracked beam height is
smaller than its value at the limit of proportionality, hunc· The
same relationship to that used for calculating the flexural
tensile strength ft.fl is still used
(4.4.2a)
(4.4.2b)
Fig. 4. 6. 1 . 1-Specimen idealization for model use (Barros
et a!. 2005). Uncracked section is distance between notch tip
and cross section top, h sp·
curve and use them to construct a piecewise response. The
tensile strain value s1 is obtained following Hooke's Law
and using Young's modulus of fiber-reinforced concrete
(FRC), Ec, which is assumed to be the same in compression
(4.6)
Other strain values s2 and s3 depend on the choice of
modeling technique used for the back-calculation of the
properties. A sample model is presented in 4.6.1.
4.6.1 Tensile stress values a 1 , a2, and a3-Relationships
connecting tensile stress values, cr; (i = 1, 2, 3), with flexural­
tensile valuesj;mJIJR,J, and/R,4 vary depending on the type of
test beam used, listed as follows:
1) Notched (RILEM TC 162-TDF 2003) three-point
bending
2) NBN B 15 238:1992
3) Unnotched ASTM C78/C78M and ASTM C293/C293M
4) Unnotched ASTM C1399/C1399M and ASTM C1609/
C1609M
4.6.1.1 RJLEM notched three-point beam -If a notched
three-point bending is used (RILEM TC 162-TDF 2003),
the specimen can be approximated as shown in Fig. 4.6.1.1.
Tensile stress values cr " cr2, and cr3, as defined in Fig. 4.3b,
can now be calculated as

4.5-Relationship between uniaxial tensile stress (4.6. l . l a)

and flexural strength
The flexural tensile strengthj;JI and residual flexural tensile (4.6. l . l b)
strengths/R,I and/R,4 can be directly related to the key tensile
stress parameters cr; (i = 1, 2, 3) that define the stress-strain (4.6.1.1c)
diagram shown in Fig. 4.3b. This relationship is a function
of the specimen type and size, and introduced through stress where d is the depth of the beam's cross section, and ftmJI
coefficients C1, C2, and C3, the values of which can be calcu­ is the average value of the flexural tensile strength, j;JI . In
lated from test data using an inverse analysis procedure. RILEM TC 162-TDF (2003) and Barros et a!. (2005), there
is a size factor k11 that is also introduced, and the C; stress
4.6-Tensile stress-strain diagrams and strain coefficients and strain values can be assumed as (Barros et
values E1, E2, and E3
a!. 2005)
An inverse analysis procedure is used to determine the
entire stress-strain response from the experimental tests. A c, = o.52, c2 = o.36, c3 = o.n,
more simple approach is to identify a few key points of the s2 = 0.12%, s3 = 10.4% (4.6.1.1d)
Coovriaht Am�te Institute
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
4.6.1.2 Belgian NBN unnotched four-point beam­
Nemegeer et a!. (2003) used the deformation-controlled
bending test defined by NBN B 15 238:1992 to evaluate
FRC material properties. The test is a four-point bending
test, with a span of 18 in. (457 mm), with beam width and
height of 6 in. (150 mm).
Contrary to the RILEM approach presented in the previous
section where case loads are recorded at a deflection of 1/64
and 1/8 in. (0.46 and 3.0 mm) (Barros et a!. 2005), NBN B
15 23 8: 1992 records loads at a deflection of 1116 and 1/8 in.
(1.5 and 3 mm)-for example, FR.J s and FR.3 0, respectively.
Corresponding residual flexural strengths, fR-Is and fR_30 , are
calculated as
(4.6.1.2a)
(4.6.1.2b)
This approach is based on an equivalent elastic analysis
and assumes that once a beam has cracked, the depth of
the compressive zone can be estimated at approximately
10 percent of the beam height. The residual flexural tensile
stress in the post-cracking range can be estimated as 37
percent of the corresponding linear-elastic stress. Based on
these assumptions, resulting tensile stress values 0 1 , 02, and
as defined in Fig. 4.3a, are
= 0.21ifj! (in MPa)
0"1 (4.6.1.2c)
(4.6.1.2d)
03 = 0.37fR-30 (4.6.1.2e)
(4.6.1.2f)
The strain values £2 and £3 can be used to represent the
serviceability and also the nominal strength levels.
The C; parameters and the strain values of the points that
define the tensile behavior of FRC can also be determined
by inverse analysis using a finite element method approach
(Sena-Cruz et a!. 2004; Mazaheripour et a!. 2012), a cross
section layer approach detailed in Barros and Sena-Cruz
(2001), or an iterative procedure with a closed-form solu­
tion, such as the one presented in Appendix A (Soranakom
and Mobasher 2008). In the inverse analysis, the 0-£ tensile
diagram is obtained by fitting, with the minimum error, the
average force-deflection, or crack-mouth-opening displace­
ment (CMOD) curve recorded in the experimental notched
beam bending tests.
4.6.1.3 ASTM beam tests: ASTM C78/C78M four-point
flexural and ASTM C293/C293M three-point beams­
ASTM C78/C78M and ASTM C293/C293M are used to
obtain the flexural strength using unnotched third-point and
center-point loading, respectively. These beam dimensions
Coovriaht American Concrete Institute
9
and general loading geometry have also been used within
a different testing protocol to obtain the post-crack tensile
response. There are no standardized procedures for correla­
tion of these ASTM test methods with the uniaxial stress­
strain response of FRC. Stress coefficients C1 , C2, and C3
are not reported for nominal strength values from ASTM
C78/C78M and ASTM C293/C293M, partly because these
tests were not intended to measure the deflection; therefore,
calculation of stress-strain components are not possible. No
relationship exists that connects tensile stress values, 0; (i =
1, 2, 3), to load-displacement behavior of unnotched beams.
If, however, one measured the deformation during the test,
an inverse analysis procedure that can be used to relate
tensile stress values, 0; (i = 1, 2, 3), to load-displacement
behavior of unnotched beams. Due to the strain-softening
behavior, to measure the post-peak deflection response, tests
are required to be conducted under closed-loop deflection­
controlled as described in Chapter 5.
4.6.1.4 Unnotched ASTM CJ399/CJ399M and ASTM
Cl 609/Cl609M beams-Both open-loop and closed-loop
tests have been developed for measuring the deflection in
a flexural sample. The average residual strength (ARS),
proposed by Banthia and Dubey (1999), is obtained through
ASTM C l 399/C1399M, which is an open-loop test. ASTM
C1399/C1399M is conducted by means of an open-loop
load-controlled test machine on a precracked specimen. A
steel plate is placed underneath a concrete beam and the spec­
imen is loaded under four-point bending until the concrete
cracks. The purpose of using steel plate in the precracking
process is so the load transferred onto the specimen does
not diminish after cracking. The operator is expected to
stop the precracking loading process immediately after the
cracking begins. The steel plate is removed and the cracked
specimen is reloaded for post-crack flexural loads at deflec­
tion levels of 0.02, 0.03, 0.04, and 0.05 in. (0.5, 0.75, 1.0,
and 1.25 mm) for specimens 4 x 4 x 14 in. (100 x 100 x 350
mm). The equivalent stress results are averaged to represent
ARS value. This parameter has been used as a method to
compare different material formulations and could be used
as a representative tensile strength measure for design, after
consideration of appropriate reduction factors. Chapter 5
discusses the correlation of the ARS value with the residual
tensile strength value. The procedures for computation of
this reduction factor are provided.
ASTM C 1609/C1609M tests an unnotched specimen
under deflection control to compute the post-peak flexural
load-deflection response and uses a method similar to ARS
proposed in ASTM C1399/C1399M, wherein the post­
crack range is expressed as an equivalent residual strength
measure. Figure 4.6.1.4 shows the setup schematics for one
of two linear variable differential transformers (LV DTs)
used as closed-loop feedback control parameters (ASTM
C1609/C1609M).
As the post-peak response is averaged, the residual load
is used to calculate an elastically equivalent flexural stress
using the section modulus of the uncracked beam where
the moment is divided by the equivalent elastic section
modulus. The problem with this test is that the formation
�-
 10 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE O F FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
Fig. 4. 6. 1. 4-Crack growth in FRC samples under four­
point bending test (Mobasher et a!. 2011).
of localized cracks may vary within the middle third range
of the specimen and the deflection control might not be an
accurate measure of localization. Also, during closed-loop
testing of unnotched specimens, support conditions in terms
of friction, sliding, and rotational capacities become critical
and can cause scatter in test results. Additionally, the fixity
of the roller supports plays a dominant role in the results
(Wille and Parra-Montesinos 2012). No reported values for
the stress coefficients c � , c2, and c3 exist based on nominal
strength values from ASTM C1399/C1399M and ASTM
C1609/C1609M.
CHAPTER 5-STRESS-STRAIN CURVES BY
BACK-CALCULATION APPROACH
5.1 -Parametric stress-strain curves
The approach proposed by RILEM TC 162-TDF (2003)
uses flexural data and techniques such as finite element
method to obtain approximate values of the material models
using a trial-and-error procedure. This approach should be
repeated for different fiber types and sample geometries as
well as loading geometries to obtain the back-calculated
values for tensile stress-strain approach. For example, the
methodology discussed (Vandewalle and DuPont 2003) is
applicable to steel fibers with a specific aspect ratio, fiber
length, and fiber volume. If the properties for other fiber
types or combinations are measured, the procedures would
have fo be repeated, which may be impractical for rudi­
mentaty design procedures. While all the back-calculation
procedures essentially use a similar approach of changing
the input variables to a simple or complicated model to find
the optimized output, the procedure can be streamlined if
a set of the calculations are automated using parameter­
ized variables and constants that can be easily changed. At
the satne time, a parameterized load-deflection procedure,
or a moment-curvature procedure, allows for a fast tum­
around on the tedious computations that would otherwise
Coovriaht Am�te Institute
have to be calculated by the finite element method, or by
hand computation. The proposed approach can be used
with ASTM C1609/C1609M, JCI SF4:1983 , RILEM TC
162-TDF (2003), and BS EN 10 14651:2005+ A1:(2007) to
process the experimental data into equivalent stress-strain
responses. It furthermore allows the conversion and modifi­
cation of ASTM test results for engineering calculations and
can be used as a method to justify the finite element calcula­
tions and empirical reduction factors used by RILEM TC
162-TDF (2003).
The simplified stress-strain tensile model can be related
to the flexural data using a closed-form solution of moment­
curvature response and load-deflection calculation of fiber­
reinforced concrete (FRC) as proposed by Soranakom and
Mobasher (2007a,b,c) and Soranakom et a!. (2008). This
approach relates the measured flexural load-deflection
response of FRC test beams to tensile stress-strain response
through an iterative process that is conducted in a spread­
sheet setting (Mobasher et a!. 2015).
The constitutive model using nondimensional parameter­
ized strain-softening and strain-hardening FRC is shown in
Fig. 5.1a. The compression zone is linear elastic-perfectly­
plastic (Fig. 5.1 b(b)) defined by a yield point (Eey' C5ey) as it
maintains at the constant yield stress C5cy until the ultimate
strain Ecu ·
The tension strain £1 is described by a nondimensional
strain measure 0 using a piecewise linear response with an
elastic range E up to the tensile strength point Ecn followed
by a residual tensile strength f.i.E'Ecn also referred to as ap in
Fig. 5.1a(a). Three strain measures define the first cracking
(Ecr), transition strain (£11-u ), and the ultimate tensile strain
level (E,u). Ultimate tensile and compressive strains limit the
ultimate flexural strength. With this approach, all strains and
stresses are normalized in terms of the cracking strain Ecr
and cracking strength C5cr = EerE. Corresponding normalized
strain parameters a, p, 0111 , co, 'A, and Acz" as well as stress
parameter fl, are defined in Fig. 5.1a and Eq. (5.1a). The
ultimate tensile strain level Ptu is used as the limiting state,
whereas first cracking tensile strain Ecr and tensile modulus E
are used as intrinsic material parameters to normalize other
parameters
£cy
co = - ; a = £/rn ; 1-'� - s... '·
-
£cr Ecr Ecr
(5. l a)
R Etu 'l Ecu .
P tu - ' 1\.cu
Ecr Ecr
Nomenclature used is:
a) Cracking strength in tension, acn is defined as EcrE
b) Post-cracking strength in tension, defined using scalar
parameter fl, is ap = fl£C/E
c) Transition point in the post-crack strain in tension,
defined using scalar parameter a, is Etrn = aEcr
d) Yield strength in compression, C5cy = COEcrE
e) Y ield strain in compression, Eey = WEer
f) Ultimate tensile strain £1u, defined using scalar param­
eter p,ll, is Ctu = PtuCcr
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 11

(a) (b)

Ed =yE

L.-.;..._
._ _____ ...
.... .,; £t
Ecr Etrn=accr Etu=PtuEcr
Fig. 5. 1a-FRC model used by Soranakom and Mobasher (200 7a,b,c) and Soranakom et
al. (2008), where (a) is the tensile response and (b) is the compression response.

cctop=Accr fc1
f''''''''f ''''''''=''''''''''''''''''''''''''''
2. 1 Fc1
� hc1 � kd �
················X ······· Y.· · · · · · · · t · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·
: :
1 � ht1 �d 1
............. ... ......... . ....)( ...................� ........... f� 1
. . . . .......-
...-+
. ----1-.
Ft2
.

2 � ht2 2 -+-----'._...
-"'----...... .................Y. ..................Y. .................. .
ft2
(a)

cctop=Accr
'='''''''''�''''''''='''''''''''''''''''''''''''' fc1
3. 1 Fc1

(b)
Fig. 5. 1 b-Strain and stress diagrams a t the post-crack stage: (a) strain distribution; and
(b) stress distribution (Soranakom and Mobasher 200 7a, b,c). (Refer to Stages 2. 1 and 3. 1
in Table A.3afor post crack stage.)

g) Ultimate compressive strain Ecu = AcuEcr

A relationship has been proposed between the approach by
Soranakom and Mobasher (2007a,b,c) and the approach by (5. l b)
BS EN 1465 1 :2005+A l :(2007)

Coovriaht American Concrete Institute �-

 12 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE O F FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
z
-"'
"0
"'
0
...J
Polymeric Fiber (Type A)
0 0
0 0.1
Strain, mm/mm
(a)
Fig. 5.2-(a) Effect ofage on back-calculated tensile stress-strain response, and a represen­
tative curvature-deflection relationship; and (b) effect ofage on experimental and simulated
load-deflection responsefor polymericfibers (Bakhshi et al. 2014). (Note: 1 MPa = 145 psi.)
Appendix A presents the procedures for calculation of the
closed-form moment curvature and load-deflection curve.
This approach is extended to FRC elements that can also
include longitudinally steel- and fiber-reinforced polymer
bars, where £2 can be higher than £ 1 , and Gp higher than Gc,.
to simulate strain-hardening FRCs (Mobasher 201 1 ; Taheri
et al. 20 1 1 ).
5.2-Back-calculation of flexural test data
Figure 5.2 presents the back-calculated tensile stress­
stain response and flexural load-deflection response of
macro-synthetic fibers (Bakhshi et al. 2014). The effect of
curing time on the tensile data is obtained from the flexural
load-deflection results and plotted in Fig. 5.2(a). The initial
tensile response is linear elastic up to approximately 87, 1 74,
and 232 psi (0.6, 1 .2, and 1 .6 MPa) for 8-, 1 6-, and 36-hour
samples as the first crack stage, respectively. After cracking,
the load is transferred to the fibers, and crack bridging
results in strain-softening response. Back-calculated tensile
stress-strain responses show that after an average strain level
of approximately 0.0025 in./in. (mm/mm), the residual
strength of the macro-synthetic fiber composites reaches
a constant value. The post-crack residual strength at this
plateau zone increases from approximately 50 to 80 and
1 00 psi (0.35 to 0.55 to 0.7 MPa), respectively due to curing
from 8 to 1 6 and 36 hours. A maximum curvature-versus­
deflection plot is also shown in Fig. 5.2(a) and represents
a linear response. As shown in Fig. 5 .2(b), the simulated
r<e'sponse for the load-deflection response shows a good
correlation with the experimental data. The parameter 11,
representing the normalized post-crack strength of the mate­
rials, is obtained as 0.52, 0.44, and 0.46, respectively for 8-,
1 6-, and 36-hour cured samples. Note that all these samples
ate classified as strain softening, with parameter 11 defined as
the normalized post-crack residual tensile strength.
Coovriaht Am�te Institute
Deflection, in
0 0.02 0.04 0.06
0 .
05 1 .5
Deflection, mm
(b)
The sudden drop in the resistance after cracking refers to
deflection-softening response. According to the model by
Soranakom and Mobasher (2008, 2009b ), assuming that
there is only a single segment to the softening response (for
example, a = 1 ), the transition from deflection softening
to deflection hardening takes place at a critical post-crack
tensile strength of
(5 .2)
The critical post-crack tensile strength level !lcrit maintains
a residual strength equivalent to the cracked strength level. A
value of llcrit = 0.35 is obtained at a compressive strength to
tensile strength ratio co = 9, and characterizes two subclasses
of materials-deflection softening (!l < 0.35) and deflection
hardening (!l > 0.35).
5.3-Comparison with averaged residual strength
results
ASTM C l 399/C 1 399M and ASTM C 1 609/C 1 609M can be
used by means of spreadsheet-based inverse analysis proce­
dures (Soranakom and Mobasher 2007a,b,c, Mobasher et al.
2015). Correlation of residual strength from ASTM C 1 609/
C 1 609M and the back-calculated method is addressed in the
following. To illustrate the differences in the interpretation of
test data between an equivalent elastic and strain-softening
approach, the stress distribution during the late stage of a
macro-fiber sample composite is presented across the cross­
sectional depth in Fig. 5.3 (Soranakom and Mobasher 2 0 1 0).
Using a strain-softening material model captures the incre­
mental movement of the neutral axis, magnitude of compres­
sive stress, and residual tensile stress. The simulation of Fig.
5.3 on polypropylene fiber-reinforced concrete (FRC) reveals
a discrepancy between the present method and average
residual stress (ARS) in accordance with ASTM C 1 399/
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 13

C 1 3 99M or ASTM C 1 609/C 1 609M, which is primarily due
to a difference in stress distribution between the two test
methods (Mobasher 20 1 1 ).
Note that specimen size effect has considerable influence
on the measured residual flexural strength parameters; for
example, ASTM C l 399/C l 399M currently only allows the
testing of 4 x 4 x 14 in. ( 1 00 x 1 00 x 350 mm) sized beams.
ASTM C1 609/C 1 609M, however, allows the use of both 4
X 4 X 14 in. ( 1 00 X 100 X 350 mm) and 6 X 6 X 20 in. ( 1 50 X
1 50 x 500 mm) specimen sizes, depending on the length of
the constituent fibers. Size effect causes smaller beam sizes
to show higher residual flexural strength parameters.
Stress, MPa
0 4
2 eter ap as defined in Fig. A. 1 (a).
c E
N.A. (k) by the Present Method
..c
+-'
5.4-Comparison with RILEM and JCI methods
Similar comparisons of back-calculation data can also be
made with equivalent models obtained from the RILEM TC
1 62-TDF (2003) method. Results indicate the linear elastic­
based residual strength approaches overestimate the residual
tensile strength by as much as three times. The inconsistency
is in residual strength calculation using the linear elastic
approach; the neutral axis is assumed fixed at the centroid
of uncracked section and the stress distribution is linear,
leading to nominal flexural stress levels that are far greater
than tensile strength. Based on this inconsistency, the use
of a consistent and verifiable reduction factor when using
the ARS method in design and analysis of fiber-reinforced
concrete (FRC) sections is needed. Although discussions of
the ASTM C 1 609/C 1 609M test method are beyond the scope
of this report, computation of a residual strength value based
8 on nominal flexural stiffness is marginally associated with
60 post-crack tensile strength or tensile residual strength param­
JCI SF4: 1983 recommends calculating an equivalent flex­
E ural strength from flexural toughness and specimen geometry.

0..
..c
+-'
The Soranakom and Mobasher (2007c) approach requires
Q) Stress Distribution 20 0.. an adjustment factor of 1/3 . 1 1 = 0.32 and 1/3.2 = 0.3 1 , as
-c Q)
Softening Zone -c shown in Fig. A. 1 , which correlates with the recommended
0)
c
0 0 0 g>
0
values by RILEM TC 1 62-TDF (2003), JCI SF4: 1 983, and
cu Barros et a!. (2005) for a variety of FRC materials that use
cu
�
Q) polymeric fibers (Bakhshi et a!. 20 1 3). The comparison
(..) -20
c b etween the residual strength parameter obtained from the
cu -1 ARS Method, c
+-'
cu present method and the experimental values reported based
(/) --- Linear Elastic
i5 -40 � on the aforementioned testing protocol are shown in Fig. 5 .4,
---
Present Method,
0
-2 Elastic Softening which shows a linear correlation with a slope of 3 . 1 to 3.2
(Mobasher et a!. 2014).
���_.--��_.-�_.� -60
0 400 800 1 200 1 600
Stress, psi CHAPTER 6-CONCLUSIONS
The available methodologies that are used for data reduc­
Fig. 5.3-Comparison of stress distribution in the present tion and presentation of the flexural test results in terms of
back-calculation approach with the ARS method (Soranakom an equivalent tensile stress-strain response for fiber-rein­
and Mobasher 2010; Bakhshi et a!. 2013). forced concrete (FRC) materials are reviewed. Due to the
variable nature of the number of parameters and fiber geom-
�o.,., psi

50 1 00 1 50 200 50 1 00 1 50 200

4
t,._ , = 3.104 �0� ob=3.22 �ocr
"'
[L R' = 0.981 ·u;
R2= 0.966 c.
�3
� 400 �
w
...J II?
()
!g_ 2 0 �
0 0 0 HPFRC Steel 0 0 0 0 HPFRC Steel
* * * Macro Polymer '11 '11 '11 Macro Polymer
0 0 0 Giass 0 0 0 Giass
-- Linear Fit -- Linear Fit

o �-�-�-�-�-�� o o �-�-�-�-�-�� o
0 0.5 1 .5 0 0.5 1 .5

�a.... MPa

Fig. 5. 4-Correlation of residual strength from ASTM C1 609/C1 609M, RILEM, and
JCI-SF4, and the Soranakom and Mobasher (2010) approach requires an adjustment
factor of3. 1 to 3.2. 25.
Coovriaht American Concrete Institute �-
 14 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE O F FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
etry and types, a uniform material model is presented that is
applicable to different types of specimens made with steel,
glass, and polymeric FRC. The method is also compared
against the available residual strength and also elastically
equivalent residual strengths obtained by different specimen
geometries.
CHAPTER 7-REFERENCES
Committee documents are listed first by document number
and year of publication followed by authored documents
listed alphabetically.
American Concrete Institute
ACI 544.3R-08-Guide for Specifying, Proportioning,
and Production of Fiber-Reinforced Concrete
ACI 544.6R- 1 5-Report on Design and Construction of
Steel Fiber-Reinforced Concrete Elevated Slabs
ACI 544.7R- 1 6-Report on Design and Construction of
Fiber-Reinforced Precast Concrete Tunnel Segments
ASTM International
ASTM C78/C78M- 1 5-Standard Test Method for Flex­
ural Strength of Concrete (Using Simple Beam with Third­
Point Loading)
ASTM C293/C293M- 1 0-Standard Test Method for
Flexural Strength of Concrete (Using Simple Beam with
Center-Point Loading)
ASTM C496/C496M- l l -Standard Test Method for Split­
ting Tensile Strength of Cylindrical Concrete Specimens
ASTM C 1 399/C 1 399M- 1 0(201 5)-Standard Test
Method for Obtaining Average Residual-Strength of Fiber­
Reinforced Concrete
ASTM C 1 609/C 1 609M- 1 2-Standard Test Method for
Flexural Performance of Fiber-Reinforced Concrete (Using
Beam with Third-Point Loading)
Belgian standards
NBN B 1 5 23 8 : 1 992-Tests on Fibre Reinforced
Concrete-Bending Test on Prismatic Samples
BritishStandards Institution
BS EN 1465 1 :2005+A 1 :(2007)-Test Method for
Metallic Fibre Concrete Measuring the Flexural Tensile
Strength (Limit of Proportionality (LOP), Residual)
EuropJan standards
DIN EN 14561 :2006-08-Chemical Disinfectants and
Antiseptics - Quantitative Carrier Test for the Evaluation of
Bactericidal Activity for Instruments Used in the Medical
Area - Test Method and Requirements (Phase 2, Step 2)
Japan Concrete Institute
JCI SF4: 1 983-Standard Test Method for Flexural Strength
and Flexural Toughness of Fiber Reinforced Concrete
Authored documents
Bakhshi, M.; Laungrungrong, B . ; Bonakdar, A.; Mobasher,
B.; Borror, C. M.; and Montgomery, D. C., 20 1 3 , "Econom-
Coovriaht Am�te Institute
ical Concrete Mix Design Utilizing B lended Cements,
Performance-Based Specifications, and Pay Factors," Final
Report 633, FHWA-AZ-1 3-633, May, 1 1 1 pp.
Bakhshi, M.; Barsby, C.; and Mobasher, B., 2014,
"Comparative Evaluation of Early Age Toughness Parame­
ters in Fiber Reinforced Concrete," Materials and Structures,
V. 47, No. 5, pp. 853-872. doi: 1 0. 1 6 1 7/s 1 1 527-0 1 3-0098-1
Banthia, N., and Dubey, A., 1 999, "Measurement of Flex­
ural Toughness of Fiber-Reinforced Concrete Using a Novel
Technique-Part 1 : Assessment and Calibration," A CI
Materials Journal, V. 96, No. 6, Nov.-Dec., pp. 65 1 -656.
Banthia, N., and Trottier, J. F., 1 995a, "Test Methods for
Flexural Toughness Characterization of Fiber Reinforced
Concrete: Some Concerns and a Proposition," A CI Mate­
rials Journal, V. 92, No. 1 , Jan.-Feb., pp. 48-57.
Banthia, N., and Trottier, J. F., 1 995b, "Concrete Rein­
forced with Deformed Steel Fibers, Part II: Toughness Char­
acterization," A CI Materials Journal, V. 92, No. 2, Mar.­
Apr., pp. 146-1 54.
Barr, B., and Lee, M. K., 2004, "FRC Guidelines in the
UK, with Emphasis on SFRC in Floor Slabs," Proceeding
of the North American/European Workshop on Advances in
Fiber Reinforced Concrete, BEFIB 2004, Bergamo, Italy,
Sept., pp. 29-3 8.
Barros, J. A. 0., and Sena-Cruz, J., 200 1 , "Fracture
Energy of Steel Fibre Reinforced Concrete," Mechanics of
Composite Materials and Structures, V. 8, No. 1 , Jan.-Mar. ,
pp. 29-45. doi: 1 0. 1 080/1 07594 1 0 1 4598 1 5
Barros, J . A . 0 . ; Cunha, V . M. C. F.; Ribeiro, A . F.; and
Antunes, J. A. B., 2005, "Post-Cracking Behaviour of Steel
Fibre Reinforced Concrete," Materials and Structures, V.
38, No. 1 , Jan.-Feb., pp. 47-5 6. doi: 1 0 . 1 007/BF02480574
Blanco, A.; Pujadas, P. ; de Ia Fuente, A.; Cavalaro,
S.; and Aguado, A., 20 1 3, "Application of Constitutive
Models in European Codes to RC-FRC," Construction &
Building Materials, V. 40, Mar, pp. 246-259. doi: 1 0. 1 0 1 6/j .
conbuildmat.20 1 2.09.096
CEB-FIP, 20 1 1 , "fib Model Code for Concrete Structures
201 0," Comite Euro-Intemational du Beton-Intemational
Federation for Structural Concrete, Lausanne, Switzerland,
Oct., 434 pp.
Cunha, V. M. C. F., 20 1 0, "Steel Fibre Reinforced
Self-Compacting Concrete: From Micro-Mechanics to
Composite Behaviour," PhD thesis, University of Minho,
Braga, Portugal, 467 pp.
di Prisco, M.; Toniolo, G.; Plizzari, G. A.; Cangiano, S.;
and Failla, C., 2004, "Italian Guidelines on SFRC in Fiber­
Reinforced Concrete: From Theory to Practice," Proceeding
of the North American/European Workshop on Advances in
Fiber Reinforced Concrete, BEFIB 2004, Bergamo, Italy,
Sept., pp. 3 9-72.
Fischer, G., 2004, "Current U.S. Guidelines on Fiber Rein­
forced Concrete and Implementation in Structural Design,"
Proceeding of the North American/European Workshop
on Advances in Fiber Reinforced Concrete, BEFIB 2004,
Bergamo, Italy, Sept., pp. 1 3-22.
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)
Hordijk, D. A., 1 99 1 , "Local Approach to Fatigue of
Concrete," dissertation, Delft University of Technology, the
Netherlands, 2 1 6 pp.
Mazaheripour, H.; Barros, J. A. 0.; Soltanzadeh, F.; and
Gonvalves, D. M. F., 20 1 2, "Interfacial Bond Behaviour of
GFRP Bar in Self-Compacting Fiber Reinforced Concrete,"
Eighth RJLEM International Symposium on Fibre Rein­
forced Concrete: Challenges and Opportunities, J. Barros,
ed. Sept., 1 2 pp.
Mobasher, B., 201 1 , Mechanics ofFiber and Textile Rein­
forced Cement Composites, first edition, CRC Press, Boca
Raton, FL, 450 pp.
Mobasher, B.; Krauss, S.; Dey, V.; Barsby, C.; Bakhshi, M.;
and Bonakdar, A., 201 1 , "Flexural Testing of MasterFiber™
MAC Matrix Reinforced Concrete," Report to BASF, 3 1 pp.
Mobasher, B.; Bakhshi, M.; and Barsby, C., 2014, "Back­
calculation of Residual Tensile Strength of Regular and
High Performance Fiber Reinforced Concrete from Flex­
ural Tests," Construction & Building Materials, V. 70, pp.
243-253. doi: 1 0. 1 0 1 6/j.conbuildmat.20 14.07.037
Mobasher, B.; Yao, Y. ; Soranakom, C; and Dey, V. , 20 1 5,
"A Spreadsheet-Based Inverse Analysis Procedure for Flex­
ural Specimens-Strain Softening/Hardening Behavior,"
doi: 1 0. 1 3 1 40/RG.2 . 1 .4409.2882
Naaman, A. E., and Reinhardt, H. W., 2006, "Proposed
Classification ofHPFRC Composites Based on Their Tensile
Response," Materials and Structures, V. 39, No. 5 , June, pp.
547-555. doi: 1 0. 1 6 1 7/s l l 527-006-9 1 03-2
Nemegeer, D.; Vanbrabant, J.; and Stang, H., 2003, "Brite
Euram Program on Steel Fibre Concrete Subtask: Dura­
bility: Corrosion Resistance of Cracked Fibre Reinforced
Concrete," Test and Design Methods for Steel Fibre Rein­
forced Concrete - Background and Experiences, Proceed­
ings of the RILEM TC 1 62-TDF Workshop, B. Schniitgen
and L. Vandevalle, eds., pp. 47-66.
Noghabai, K., 1 998, "Effect of Tension Softening on the
Performance of Concrete Computational Studies," PhD
thesis, Department of Civil and Mining Engineering, Divi­
sion of Structural Engineering, Liilea University of Tech­
nology, Sweden, pp. 458.
RILEM TC 1 62-TDF, 2000a, "Recommendations of
RILEM TC 1 62-TDF: Test and Design Methods for Steel
Fibre Reinforced Concrete: Bending Tests," Materials and
Structures, V. 33, No. 225, Jan.-Feb., pp. 3-5 .
RILEM TC 1 62-TDF, 2000b, "Recommendations of
RILEM TC 1 62-TDF: Test and Design Methods for Steel
Fibre Reinforced Concrete: a-£ Design Method," Materials
and Structures, V. 33, No. 226, Mar., pp. 75-8 1 .
RILEM T C 1 62-TDF, 2002a, "Recommendations of
RILEM TC 1 62-TDF: Design of Steel Fibre Reinforced
Concrete Using a-£ Method: Principles and Applications,"
Materials and Structures, V. 35, No. 249, June, pp. 262-278.
, RILEM TC 1 62-TDF, 2002b, "Final Recommendation
of RILEM TC 1 62-TDF: Test and Design Methods for
Steel Fibre Reinforced Concrete: Bending Test," Materials
' and Structures, V. 35, No. 253, Nov. , pp. 579-582. doi:
:1 0. 1 6 1 71 1 3 884
Coovriaht American Cohcrete Institute
15
RILEM TC 1 62-TDF, 2003, "Final Recommendation of
RILEM TC 1 62-TDF: Test and Design Methods for Steel
Fibre Reinforced Concrete: a-£ Design Method," Mate­
rials and Structures, V. 36, No. 262, Oct., pp. 560-567. doi:
1 0 . 1 6 1 7/1 4007
Sena-Cruz, J. M.; Barros, J. A. 0 . ; Ribeiro, A. F.; Azevedo,
A. F. M.; and Camoes, A. F. F. L., 2004, "Stress-Crack
Opening Relationship of Enhanced Performance Concrete,"
9th Portuguese Conference on Fracture, ESTSerubal,
Portugal, pp. 395-403 .
Silfwerbrand, J., 2008, "Codes for SFRC Structures­
A Swedish Proposal," Tailor Made Concrete Structures,
Walraven and Stoelhorst, eds., pp. 553-558.
Soranakom, C., and Mobasher, B., 2007a, "Closed-Form
Moment-Curvature Expressions for Homogenized Fiber­
Reinforced Concrete," ACI Materials Journal, V. 1 04, No. 4,
July-Aug., pp. 3 5 1 -359.
Soranakom, C., and Mobasher, B., 2007b, "Flexural
Modeling of Strain Softening C Strain Hardening Fiber
Reinforced Concrete," High Performance Fiber Reinforced
Cement Composite: HPFRCC 5 , H. W. Reinhardt and A. E.
Naaman, eds., RILEM Publications SARL, Cachan, France,
pp. 1 5 5 - 1 64.
Soranakom, C., and Mobasher, B., 2007c, "Closed
Form Solutions for Flexural Response of Fiber Reinforced
Concrete Beams," Journal of Engineering Mechanics, V.
1 33, No. 8, Aug., 8 pp.
Soranakom, C., and Mobasher, B., 2008, "Correlation
of Tensile and Flexural Responses of Strain Softening
and Strain Hardening Cement Composites," Cement and
Concrete Composites, V. 30, No. 6, July, pp. 465-477. doi:
1 0 . 1 0 1 6/j .cemconcomp.2008.0 1 .007
Soranakom, C., and Mobasher, B., 2009a, "Design Flex­
ural Analysis and Design of Textile Reinforced Concrete,"
Textile Reinforced Structures: Proceedings ofthe 4nd Collo­
quium on Textile Reinforced Structures (CTRS4) und zur 1.
Anwendertagung, SFB 528, Technische Universitiit Dresden,
Eigenverlag, pp. 273-288.
Soranakom, C., and Mobasher, B., 2009b, "Flexural
Design of Fiber Reinforced Concrete," A CI Materials
Journal, V. 1 06, No. 5, Sept.-Oct., pp. 461 -469.
Soranakom, C., and Mobasher, B., 20 1 0, "Flexural
Analysis and Design of Strain Softening Fiber Reinforced
Concrete," Antoine E. Naaman Symposium - Four Decades
of Progress in Prestressed Concrete, FRC, and Thin Lami­
nate Composites, SP-272, G. J. Para-Montesinos, and P.
Balaguru, eds., American Concrete Institute, Farmington
Hills, MI, pp. 173- 1 87.
Soranakom, C.; Yekani-Fard, M.; and Mobasher, B., 2008,
"Development of Design Guidelines For Strain Softening
Fiber Reinforced Concrete," Seventh International Sympo­
sium ofFiber Reinforced Concrete: Design and Applications
BEFIB, R. Gettu, ed., Sept., pp. 5 1 3-523.
Swamy, R. N.; Mangat, P. S.; and Rao, C. V. S. K., 1 975,
"The Mechanics of Fiber Reinforced Cement Matrices,"
Fiber Reinforced Concrete, SP-44, American Concrete Insti­
tute, Farmington Hills, MI, pp. 1 -28.
�-
 16 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE O F F I B E R-REINFORCED CONCRETE (ACI 544.8R-16)
strain-hardenin g
J.
�,.,
(a) �c,\
� /
/ I p=J.L
cr_ _EcrE
,__
/ ___
crcr=EcrE ····· �C',.. l i
� l
?� 1
/ : crp=J.LEcrE
i
strain softenin g
Ecr Etrn=aEcr
Fig. A . l-Full option material models for both strain-hardening and strain-softening
FRC: (a) compression model; and (b) tension model.
Taheri, M.; Barros, J. A. 0 . ; and Salehian, H. R., 20 1 1 ,
"A Design Model for Strain-Softening and Strain-Hard­
ening Fiber Reinforced Elements Reinforced Longitu­
dinally with Steel and FRP Bars," Composites. Part B,
Engineering, V. 42, No. 6, pp. 1 630- 1 640. doi: 1 0. 1 0 1 6/j .
compositesb.20 1 1 . 04.009
Teutsch, M., 2004, "German Guidelines on Steel Fiber
Concrete," Proceeding of the North American/European
Workshop on Advances in Fiber Reinforced Concrete,
BEFIB 2004, Bergamo, Italy, Sept., pp. 23-28.
Van Mier, J., 1996, Fracture Processes of Concrete:
Assessment of Material Parameters for Fracture Models,
CRC Press, 464 pp.
Vandewalle, L., and DuPont, D., 2003, "Bending Test
and Interpretation," International RILEM Workshop on Test
and Design Methods for Steelfibre Reinforced Concrete, B.
Schniitgen and L. Vandewalle, eds., RILEM Publications
SARL, pp. 1 - 1 3 .
Wille, K., and Parra-Montesinos, J . G . , 20 1 2, "Effect of
Beam Size, Casting Method, and Support Conditions on
Flexural Behavior of Ultra-High-Performance Fiber-Rein­
forced Concrete," A CJ Materials Journal, V. 1 09, No. 3,
May-June, pp. 3 79-388.
APPE NDIX A-SPREADSHEET-BASED INVERSE
A NALYSIS PROCEDURES
This appendix presents the steps in the development of
closed-form moment curvature and load deflection of a
flexural beam. The material properties are assumed based
on two main parameters: the elastic stiffness and the first
crack strain. All other material parameters for the tensile and
compression stress strain response are expressed in terms of
parametric components.
Methods are based on an inverse analysis of beam data
experimentally validated. This appendix presents a complete
spr�adsheet-based inverse analysis for center- or third-point
loaded test beams and specific relationships for specific
beam types (RILEM TC 162-TDF 2003). Relationships can
be used for other beam types by determining stress coef­
ficients C�, C2, and C3 using one of the described inverse
ana.Jysis methods.
�te Institute
Coovrioht Am
···················· )!i····
l
..f.
i
J.L>O
O<J.L< l
II
I Et
�------��� Ec
Etu=13tuEcr Ecy=roecr
A.1 -Simplified strain-softening/hardening fiber­
reinforced concrete model
For the inverse analysis procedure, a general strain hard­
ening tensile and an elastic perfectly plastic compression
model as derived by Soranakom and Mobasher (2007a,b,c;
2008; 2009b) is shown in Fig. A. I . Tensile response is
defined by tensile stiffness E, first crack tensile strain Ec,.,
cracking tensile strength ac,. = EBen ultimate tensile strength
Epeak. and post-crack modulus Ecr· The softening range is
shown as a constant stress level f.t.E'Ecr- The compression
response is defined by the compressive strength acy. defined
as royEEcr· To simplify material characteristics of strain-hard­
ening fiber-reinforced concrete (FRC) and generate closed
form design equations, classical assumptions such as plane
section remain plane, linear strain distribution, and small
deformations are made. By ignoring the post-peak ranges
in tension, the closed-form equations can be simplified
further to the idealized bilinear tension and elastic compres­
sion models, as shown in Fig. A . 1 (a) and (b). To reduce
the complexity of material response, one can disregard the
post-peak tensile response and plasticity in the compression
region. It has been shown that the difference in compressive
and tensile modulus has a negligible effect on the ultimate
moment (Soranakom and Mobasher 2008). By defining all
parameters as normalized, with respect to minimum number
of variables, closed-form derivations are obtained. Applied
tensile and compressive strains at bottom and top fibers, 13
and A., are defined as
� = S. , A = s._ (A. I a)
£cr Ecr
Material parameters required for the simplified models are
summarized as follows. Parameters, a, J.L, TJ, ro, are defined,
respectively, as representing normalized, tensile strain at peak
strength, post-crack modulus, and compressive yield strain
Epeak = crE (Jcy = (J cy
a = -- ' T] - ' co = -- - (A. l b)
Ec,. E EEcr ()cr
 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 17

cctop=A-ccr
t''''''''t "'''''t''"'''''"''''"'''''''''' fc1
(2 . 1 ) Fc1
i hc1 i kd i
················ X ······· ;>j(· · · · · · · · t · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·

: :� d
1 � ht1 1
ccr ............. ················· ·X ··················· � ············ ft.1
. . . . . . . . .-
..-
..-1---JC...•
..�

2 � ht2 :
2 -+-----l� Ft2
"'------11 . . . . . . . . . . . . . .. . .Y...................Y................... .
ctbot=�ccr ft2
(a) Strain distribution (b) Stress distribution

cctop=A-ccr
t''''''''t'''''''t''''''''''''''''''''''''''' fc1
(3. 1 ) Fc1
� hc1 � kd � � Yc1
·· ·············· X � f ······· ········ ·····························

A
········ X········ ··

:: Yt1 :
A
: h t1 : d
:
ft1 ........... � Yt2 �
Ft1 :
�
: Yt3
1
................. x. . . . . . . . . . . . . . . . . . . �. . . . . . . . . . . . . . . . .
�
t )r.
................. �. -�........... ..j.. . .... .... .. .. .ft.:?... . �.. . ._,. Ft2 �
� ht3 ft3 3
. . ....... .� . . . . . . . . . . . . . . . . . . � . . . . . . . . . . . . . . . . ..... .
� v• Ft3
"'------11 . ......

(c) Strain distribution (d) Stress distribution

Fig. A. 2-Strain and stress diagrams at the post-crack stage (Ranges 2. 1 and 3.1 in Table
A. Ja).

A.2-Derivation of moment-curvature diagram

For typical strain-hardening fiber-reinforced concrete (A.3a)
(FRC), the compressive strength is higher than the tensile
strength. Thus, flexural strength is controlled by the weaker
tension, and compressive stress is normally low in the elastic F,l = ( 1 - k) & = 3. ( 1 - k)
�
; (A.3b)
range. For this reason, the elastic compression model (Fig. bhcrcr h 3 �
A.2(b)) is used. For the development of design equations,
the accompanying compressive stress developed in a beam
section is limited to the yield compressive stress <Jcy = 0.85/c ' � = (1 - k) ( � - 1) ( 11 � - 11 + 2) .
at compressive yield strain �>cy' where fc' is the uniaxial bhcr cr 2�
compressive strength. (A.3c)
Y1 2 = 211 � 2 - 11 � - 11 + 3� + 3
( 1 - k)
A.3-Derivation of nominal flexural strength h 3� ( 11 � - 11 + 2 )
Nominal flexural strength of a beam section, according to
the imposed tensile strain at the bottom fiber (£1 = �£c,.) can where F and y are the force and its centroid, respectively;
be derived based on the assumed linear strain distribution, subscripts c 1 , t 1 , t2 designate Compression Zone 1 , Tension
as shown in Fig. A. 1 (a). By using material models described Zone 1 , and Tension Zone 2, respectively; and b and h are
in Fig. A. 1 (a) and (b), the corresponding stress diagram is the width and the height of the beam, respectively. The
obtained, as shown in Fig. A.2(b ), in which the stress distribu­ neutral axis parameter k is found by solving the equilibrium
tion is subdivided into a Compression Zone 1 , and Tension of net internal forces equal to zero, Fc 1 + F,1 + F12 = 0 and
Zones 1 and 2. Force components and their centroidal distance expressed in Eq. (A.3d)
to the neutral axis in each zone can be expressed as Eq. (A.3a)
through (A.3c)
(A.3d)

Coovriaht American Concrete Institute �-

 18 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE O F FIBER-REINFORCED CONCRETE (AC1 544.8R-16)

Table A.3a-Neutral axis parameter k, normalized moment M ', and normalized curvature <1> for each stage
of normalized tensile strain at bottom fiber, p
Stage k M' = M,/Mc,. <j> = <j>/<J>a

j'
for y = I
I 2 2� [(y - l)k,3 + 3k,2 - 3k, + I ] �
k - M', - <I>', =
0 < � ::0 1 I - -l + fi 1 - k, 2(1 - k, )
for y * I
-l + y

2
�
, (2 - C2 1 )k21' + 3C21k21 - 3C21k21 + C21
2 M2 1 = �'
D21 - D21 � l - k21
2. 1 k21 = 2 �
! < � :S a D2 1 - � <I>'21 =
2(1 - k 1 )
O < A. :S w (2�3 - 3�2 + I )11 + 3�2 - I 2
D21 = 11(�2 -2� + I ) + 2� - I c2 1 =
�2

�
3 3 2
, (2� - C3 1 )k3 1 + 3C3 1k3 1 - 3C3 1 k3 1 + C3 1
D" - D" �2 M3 1 =
k,, = ' l - k3 1
3.1 D" - � �
a < � ::; �"' <I>'3 1 =
2(1 - k3 1 )
O < A. :S w (2a3 - 3a2 + 1)11 - 3Jl(a2 - �2) + 3a2 - I
D3, = lj ( a2 -2a + 1 ) + 2!1(� - a) + c" =
2a - 1 �2

where c1 = 11 CW - 2B + 1 ) + 2B - 1 . strain capacity a-that need to be estimated for ultimate

The nominal moment M,, is obtained by taking the first
moment of force about the neutral axis, M11 = Fc1Yc1 +
F11y11 + Fay12, and express it as a product of the normal­
ized nominal moment M,, and the cracking moment Mer as
follows in Eq. (A.3e) and (A.3f)
(A.3e)
- 2k + 1 2�e
M' = c e + (A.3f)
2 �2 1-k
2 top compressive strain and bottom tensile strain as follows
where c2 = cl + 2 CI B - B .
If the full stress-strain response is desired, then the loca­
tion of neutral axis and effective moment are obtained under
the definitions provided in Table A.3a. In this table, the deri­
vations of all potential combinations for the interaction of
tensile and compressive response are presented.
Analysis of Eq. (A.3f) using a parametric analysis indi­
cates that the contribution of fibers is mostly apparent in the
post-cracking tensile region, where the response continues to
increase after cracking (Fig. A. 1 (a)). The post-crack modulus
Ecr is relatively flat with values of11 = 0.0-0.4 for a majority
of cement composites. The tensile strain at peak strength,
Bpeak. is relatively large compared to the cracking tensile
strain Ben and may be as high as a = 1 00 for polymeric-based
fiber systems. These unique characteristics cause the flexural
strength to continue increasing after cracking. To simplify
the problem for a typical strain-hardening FRC, the contri­
bution of post-peak tensile strength is ignored; therefore,
the flexural strength is associated with the tensile strain at
peak strength. Also, the effect of post-crack tensile response
parameter ll can be ignored for a simplified analysis. The
parameters for back-calculation are therefore limited to two
variables-post-crack stiffness 11 and post-crack ultimate
moment for the design purposes.
According to bilinear tension and elastic compression
models shown in Fig. A. 1 (a) and Fig. A. 1 (b), the ultimate
moment is obtained when the normalized tensile strain at
the bottom fiber (B = E/Ecr) reaches the tensile strain at peak
strength (a = BpeakiBCI) . Equations (A.3f) to (A.3h) for ulti­
mate moment, however, are applicable for the compressive
stress in elastic region only. The elastic condition should be
checked by computing the normalized compressive strain
developed at the top fiber 'A and compared to the normal­
ized yield compressive strain co. The general solutions for
all the cases are presented in Table A.3a. Using the strain
diagram in Fig. A.2(a) to find the relationship between the
� _£
_,_ (A.3g)
kh (1 - k)h
Substitute f.c = ABcr and Ct = Beer in Eq. (A.3g), then limit
the maximum compressive strain to the yield compressive
strain Ecy = COEcr. Finally, the condition can be expressed in a
normalized form as
k (A.3h)
'A = - � ::;; co
1-k
Case 2. 1 of Table A.3a occurs when the tensile behavior is
elastic-plastic while the compressive behavior is still elastic.
The general solutions presented in Table A.3a allow for the
computation of the location of neutral axis as a function of
applied tensile strain B in Eq. (A.3i)

Coovriaht Am�te Institute

 INDIRECT M ETHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 19

Table A.3b-Location of neutral axis, moment, and moment-curvature response of a strain-hardening

composite material with y = 1; 11 = 0.0001 - 0.5

11 A, ( ��J
k=
M'(k) M'(<jl)
0.5 0.5(�2 + I - 2�) + 2� - I -0.773 + o . J 08 x w-'k-6 0.507 + 0.686$
0.2 0.2(�2 + I - 2�) + 2� - I o.654 + o.5 1 6 x w-2k-6 1 . 1 0 5 + 0.383$
0. 1 0. 1 (�2 + I - 2�) + 2� - I 1 .276 + o.289 x w-2k-6 1 .461 + 0.234$
0.05 0.05(�2 + I - 2�) + 2� - I 1 .645 + 0. 1 632 x w-2k-6 1 .720 + 0.1401$
0.0 1 0.0 1 (�2 + 1 - 2�) + 2� - 1 0.852 + 0.456k-l 1 .342 + 0.3 7 1 -J<jl
0.0001 O.OOO i epl + 1 - 2�) + 2� - 1 3 . 1 77 - 3.068k 3.02 1 - 2.047/-Jq,

�
-E
�
� 0.3
Q)
E
0
z �

-� 0.2
"0 "0
Q)
.!::::!
Cii Cii
E E
� 0.1
'--
0
�,=0.05 z
e--e--€> ,=0.00
� ,=0. 1 0

0
<!1- - � - � ,=0.01
...... ,=0.20
o ���--��������
0 10 20 30 40 50 0 10 20 30 40 50
(a) Normalized Curvature, <I>' (b) Normalized Curvature, <I>'

Fig. A. 3-Effect of (a) depth ofneutral axis on the moment capacity of a section; and (b)
the moment curvature response in the range 2. 1.

This equation can be simplified by assuming equal tension
and compression stiffuess (y = 1 ). For an elastic perfectly plastic
tension material (TJ = 0), the equation reduces to Eq. (A.3j)
(A.3j)
Table A.3b presents the case ofy = 1 for different values of
post-crack stiffness, T] = 0.5, 0.2, 0. 1 , 0.05, 0.0 1 , and 0.00 1 .
Note that the neutral axis is a function � and can be used in
calculation of the moment, or the moment-curvature rela­
tionship. These general responses, which are shown in Fig.
A.3 (a) and (b), show that with an increase in applied tensile
strain, the neutral axis compression zone decreases. Note,
however, that this decrease is a function of a post-crack
tensile stiffness factor. Although the moment-curvature
relationship in this range is ascending, its rate is a function
of the post-crack tensile stiffness. One can use a curve fit
approach and estimate moment-neutral axis depth as well
as the moment curvature values for distinct values of T], as
shown in Table A.3b. The equations in Table A.3b repre­
sent the magnitude of the moment and moment-curvature
response generated for any rectangular cross section using
basic tensile material parameters in the 2 . 1 range, as shown
in Table A.3b, and are applicable within 1 percent accuracy
·
ofthe dosedcform results.
Coovriaht American Concrete Institute
The proper way to use Table A.3b is to generate a column
of incrementally increasing strain values as �- The next step
is to assume a post-crack stiffuess value such as T] = 0. 1 . The
depth of neutral axis is then computed first by calculating the
value of parameter A as a function of each individual value
of �- Using that value of k, the moment and also the moment
curvature relationship are obtained in a column wise format.
The relationship between the moment and curvature as a
function of the neutral axis can be plotted as two x-y column
parameters.
A.4-Simplified moment curvature diagram
Using a simplification approach to represent the normal­
ized moment-curvature diagrams using a bilinear curve,
the load-versus-deflection curve for deflection-hardening
and deflection-softening can be obtained. In the simplified
models, the intersection points (<Pit', Mit') of the linear elastic
response and the linear post-crack response is found by using
a regression equation (Eq. (A.4)). The regression equation is
dimensionless and independent of the unit used.
Mil' = 0.7425Mu' + 0 . 1 739 and <Pit' = Mit' (AA)
The conversion to a bilinear curve is described in detail by
Soranakom and Mobasher (2008).
�-
 20 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE O F FIBER-REINFORCED CONCRETE (AC1 544.8R-16)

p PI PI
(a)
l l
s s
L L
Mu

�be
. . ... �be
•
. ••
(C) · . .. .
.. . •• •
.. •• . . . ..
. · · . . . . ... . . . . . .. . . . . . ... . . . .
· ·· · ·· .
··
·· I ···
· ·· · .. . ··
1 .: 2 .: ···
· ·· . .
· .
··

s s I S/2 1
�u

I
Mbcr S
M"
(1-IMbcr
I
M11 J
i
2

�u

Fig. A. 5-Three- andfour-point bending test: (a) loading setup; (b) moment distribution;
(c) curvature distribution at first bilinear cracking; (d) curvature distribution at ultimate
momentfor high normalizedpost-peak tensile strength (/1 > f.1criJ; and (e) curvature distri­
bution at ultimate momentfor low normalized post-peak tensile strength (/1 < f.1criJ -

A.5-Load-deflection response Lz
0 = -- ·

Figure A.5(a) shows the loading and geometry for three­ II 24M 2 II

and four-point bending tests. Figure A.5(b) show moment

[ (2 M,; - Mu Mbcr - M;c,. ) <i>" + ( M,; + M.. Mbo ) <i>bcr ]
(A.5b)
distributions at cracking (Mbcr) and ultimate levels (M11) .
With the area moment method, the corresponding curvature
diagrams shown in Fig. A.5(c) to A.5(e) are divided into
several areas and taken around the left support to obtain
L M bcr L L
the midspan deflection 8. A set of equations for calculating o = <i> �� " (2L - L
) + .. <i> ( - 2L )
II

midspan deflection of three-point bending at first bilinear 8 p l 2Mbcr p (A.5c)

cracking, at ultimate strength when material has 1-l > 1-lcrit, and ll < llcrit
at ultimate strength when material has 1-l < llcrit are presented
in Eq. (A.5a) to Eq. (A.5c).
Similarly, a set of equations for four-point bending can be
written as
(A.5a)
(A.5d)

Coovriaht Am�te Institute

 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16) 21

= L2 critical normalized post-peak tensile strength f.tcrit = 0.345 is

8, -­
2 1 6M,� calculated by Eq. (A.6a).

[ (2 3M,� - 4M,Mba - 4M:a )<P, + (4M,� + 4M,Mba )<Pba· ] (A.5e) 96

. =�= · = 0.345 (A.6a)
llcra 3co - 1 3(9.6) - 1

o, 5L2 <jl, M,L2 <1>&cr Because f! = 0.30 < f.tcrir, the worksheet shows Type 2
= +
72 27Mbcr (Naaman and Reinhardt 2006). According to the normalized
(A.5f)
ultimate tensile failure �''" the corresponding normalized
top compressive strain Aw is checked, which shows that the
Load step P; can be back-calculated from a discrete point compressive strain fails at Stage 3 (co < 'A).
i along the moment curvature diagram
r:l =
1-' w 200 > (co2 + 2f! - l ) = (9 . 62 + 2 x 0 . 3 - l ) = 1 5 2 _ 9 (A.6b)
2 x 0.3
P; = ----j-
2M
for <!>; = 0 . . .<jl, (A.5g)
2f.l

'A,, < Acu means that the ultimate tensile strain will reach
where S is a spacing between the support and loading point; S = the failure before the compressive strain crushing. The
L/2 for three-point bending; and S = L/3 for four-point bending. smaller of'Aw (= 1 1 .03) and Acu (= 3 0) is used as a normalized
ultimate compressive strain 'A, in the calculation of neutral
A.6-Example: Three-point bending test
axis depth ratio ku by Eq. (A.6c).
(Note: An excel spreadsheet with associated uses manual
that computes the material properties for three- and four­
2 11"-
point bending samples is available for download (Mobasher k" = = 0.052 (A.6c)
-(1)2 + 2A.( (!) + ll) + 2J.! - 1
et a!. 20 1 5).)
Determine the moment-curvature diagram and load deflec­
tion response of a beam size 4 x 4 in. ( 1 00 x 1 00 mm) tested (3J.!A.2 + 3coA. 2
- 3Jl - co3 + 2)k 2
M'(A. , k , co, ) =
under three-point bending at clear span L = 12 in. (300 mm) . ll A. 2 (A.6d)
Assume that the material post crack stiffness parameter T) = 0; - 3J.!(2k - 1) = 0.868
therefore, after cracking, the response is directly in the soft­
ening range. Assume the plastic length for crack localized
"''(A., k , (!), Jl) = }:__ =
zone under the point load LP = 4 in. ( 1 00 mm) The ultimate .
1 1 .03
= 1 05 . 52 (A.6e)
uniaxial compressive strength.fc' = 4300 psi (30 MPa), uniaxial 'I' 2k 2(0.052)
tensile strength ac, = 360 psi (2.50 MPa), and post-peak tensile
strength aP = 1 00 psi (0.75 MPa). The ultimate compressive The normalized ultimate moment M,/ and curvature cp,/ are
strain Ecu = 0.003 and the ultimate tensile strain Ew = 0.02. The calculated by Eq. (A.6d) and (A.6e) as 0.868 and 1 05.52,
Young's modulus E = 3,600,000 psi (25,000 MPa). respectively. The intersection points (<Pit' and Mu') of the
Using the specified material properties, determine the normal­ linear elastic and the linear post-crack response for bilinear
ized material parameters. Assuming no nonlinear response, a = moment curvature diagram is found by regression Eq. (A. 1 2).
1 and T) = 0. These parameters are calculated as follows: The normalized cracking moment curvature is always equal
Cracking strain: Ecr = ac,JE = 2.5/25000 = 0.0001 to 1 (<Per'= Mer' = 1). Because the material is deflection soft­
Normalized post peak tensile strength: f! = a/acr ening, the normalized reduced cracking moment Mcr2 ' is
0.75/2.5 = 0.30 needed and calculated by Eq. (A6.f) as 0.8 1 8.
Normalized ultimate tensile strain: �'" = E11/Ecr
0.02/0.000 1 = 200 <!>;, = M(, = 0.7425M,; + 0. 1 739
Normalized ultimate compressive strain: Acu = Ecz/Ecr (A.6f)
= 0.7425(0.868) + 0 . 1 739 = 0.8 1 8
0.003/0.0001 = 30
Assumed compressive yield stress: acy = 0.8 x .fc ' = 0.8 x The four controlling points for normalized moment and
30 = 24 MPa (3480 psi) curvature and dimensionalized moment and curvature, in
Compressive to tensile strain ratio: co = acJ(&cr) = 24/2.5 = 9.6 addition to the computed load and deflection values for these
Using the parameters computed previously, and input control points, are obtained as shown in Table A.6a.
parameters such as test method = 3 (three-point bending) or The moment-curvature response is recovered by multi­
4 (four-point bending), b = 4 in. ( 1 00 mm), d = 4 in. ( 1 00 plying the cracking moment curvature to their normalized
mm), L = 12 in. (300 mm), LP = 4 in. ( 1 00 mm), and E = moment curvature using equations in Table A.3a. The deflec­
3 ,600,000 psi (25,000 MPa). tions of three-point bending for the deflection softening f.l <
According to the material parameters' input, tensile and f.tcrit are calculated (Eq. (A.5b) and (A.5c)) and the load calcu­
compressive stress-strain responses are calculated. The lated (Eq. (A.5f)). Note that the deflections at the cracking are
t��.. sa�� but the load step due to Mbcr and Mcr2 are different.
Coovriaht American Concrete Institute �-
 22 INDIRECT METHOD TO OBTAIN STRESS-STRAIN RESPONSE OF FIBER-REINFORCED CONCRETE (AC1 544.8R-16)

Table A.6a-Normalized and dimensionalized moment and curvature with computed load and deflection
values

<!>' M' <j>, in.-' (mm· ' ) M, lb. i n (N.m) 6, in. (mm) P, lb (kN)
0.000 0.000 0 (0) 0 (0) 0 (0) 0 (0)
1 .000 1 .000 5 .08E-05 (2.000E-06) 3688 (4 1 6.7) 0.00059 (0.0 1 5) 1 249 (5500)
1 .000 0. 8 1 8 5 .08E-05 (2.000E-06) 301 8 (340.9) 0.00059 (0.0 1 5) 1 022 (4500)
1 05.52 1 0.868 0.0054 (2. 1 1 0E-04) 3200 (36 1 .5) 0.052 1 ( 1 .323) 1 084 (4800)

Table A.6b-Diagram of accurate moment curvature

Accurate moment-curvature diagram
Stage 'A k <!>' M' <j>, mm-1 <j>, in.-' M, N.m M, lb.in.

I 0.00 0.500 0.000 0.000 O.OOE+OO 0 0.0 0

I 1 .00 0.500 1 .000 1 .000 2.ooE-o6 5 .08E-05 4 1 6.7 3688
2 1 . 86 0.267 3.480 0.772 6.96E-o6 0.0002 32 1 .5 2846
2 3.15 0. 1 66 9.5 1 0 0.802 1 .9oE-o5 0.0005 334.3 2960
2 5 . 30 0. 1 03 25.725 0.837 5 . 1 5 E-o5 0.00 1 3 348.8 3087
2 7.45 0.075 49.644 0.854 9.93E-o5 0.0025 355.8 3 1 49
2 9.60 0.059 8 1 .267 0.864 1 .63E-o4 0.0041 359.9 3 1 86
3 9.97 0.057 87.334 0.865 1 .75E-o4 0.0044 360.4 3 1 90
3 1 0.34 0.055 93.401 0.866 1 .87E-o4 0.0048 360.8 3 1 94
3 1 0.70 0.054 99.468 0.867 I .99E-o4 0.005 1 3 6 1 .2 3200
3 1 1 .07 0.052 I 05.535 0.867 2. 1 1 E-04 0.0054 3 6 1 .4 3200

Deflection, 6 (in)
0 0.01 0.02 0.03 0.04 0.05 0.06
(a) 6000
�
:v=-­
1200
]j 0.8 5000
E 1 000
0
::E 0.6 4000

�-� 0.4
""" � 800 �
": 3000 0..
"""' 600 ""
"'
0 0
0 -l
@ o
.,Bilinear Approx. -l
z 0.2
2000
-- Exact Soln. 400

0 ����-����- 1000 200

20 40 60 � I'O�
80-L� O�I20
Normalized Curvature, q, 0 I 0
0.3 0.6 0.9 1 .2 1.5
Deflection, 6 (mm)
Curvature, q, (in- 1 )
0 0.0012 0.0024 0.0036 0.0048 0.006
Fig. A . 6b-Computed load and deflection values for the
(b)
4000 example case of a beam presented in Table A . 6a.
400
Alternatively, the accurate moment curvature diagram is
E' 3000 :5
2S 3oo g found using equations for M1 ', M21 ', and M31 ' listed in Table
::E A.3a or approximate values represented in the M'(cp) column
2000 i and A.3b. Figure AS compares the accurate moment-curvature
6 response and their approximate bilinear models. The steps for
@ o
E> Bilinear Approx. 1 000
� computation of load-deflection response, according to bilinear
1 00
-- Exact Soln. models, are shown in Fig. A.6a(a) and Fig. A.6a(b). For the
0
four-point bending test, the procedure is similar, except LP is
�
o�
- �5�
E� o�
-o�5 o.�
oo
�o�
,� .o�
o� oo�I�
5� .o�
o� oo�
2�o�
.oo�5 not needed; users can enter any value or simply zero. An accu­
Curvature, q, (mm·1)
rate expression for moment curvature can be developed using
the closed-form equations identifying each of the three zones of
Fig. A. 6a-(a) Normalized moment-curvature response;
response (Table A.6b) and Fig. A.6a(a) as the moment-curva­
and (b) its corresponding dimensionalized moment-response
ture response and Fig. A.6a(b), which represents the normal­
using the simplified and exact procedures for a rectangular
ized moment-curvature. Figure A.6b shows the load deflection
cross section.
results based on the data presented in Table A6.a ..
Coovriaht Am�te Institute
Coovriaht American Concrete Institute
Coovriaht American Concrete Institute
 American Concrete Institute
Always advancing

As ACI begins its second century of advancing concrete knowledge, its original chartered purpose
remains "to provide a comradeship in finding the best ways to do concrete work of all kinds and in
spreading knowledge." In keeping with this purpose, ACI supports the following activities:

Technical committees that produce consensus reports, guides, specifications, and codes.

Spring and fall conventions to facilitate the work of its committees.

Educational seminars that disseminate reliable information on concrete.

Certification programs for personnel employed within the concrete industry.

Student programs such as scholarships, internships, and competitions.

Sponsoring and co-sponsoring international conferences and symposia.

Formal coordination with several international concrete related societies.

Periodicals: the ACI Structural Journal, Materials Journal, and Concrete International.

Benefits of membership include a subscription to Concrete International and to an ACI Journal. ACI
members receive discounts of up to 40% on all ACI products and services, including documents, seminars
and convention registration fees.

As a member of ACI, you join thousands of practitioners and professionals worldwide who share
a commitment to maintain the highest industry standards for concrete technology, construction,
and practices. In addition, ACI chapters provide opportunities for interaction of professionals and
practitioners at a local level.

American Concrete Institute

38800 Country Club Drive
Farmington Hills, MI 48331
Phone: +1.248.848.3700
Fax: +1.248.848.3701
www .concrete.org

Coovriaht American Concrete Institute

 American Concrete Institute
Always advancing

38800 Country C l u b Drive

Farmington H i l l s , Ml 48331 USA
+1 . 248.848.3700
www.con crete.org

The A m e r i c a n C o n c rete I n stitute (AC I ) is a l e a d i n g a utho rity a n d resou rce

worldwide for the deve l o p m e nt a n d d i stri bution of c o n s e n s u s-based

sta n d a rd s a n d tec h n i c a l reso u rces, ed ucati o n a l pro g r a m s , and certifi cati o n s

for i n d ivid u a l s a n d o rga n i zati o n s i nvo lved i n c o n c rete d es i g n , c o n st r u ct i o n ,

a n d materia l s , w h o s h a re a co m m itment to p u rs u i n g the best u s e o f c o n crete.

I n d ivi d u a l s i nterested in the a ctivities of ACI a re e n c o u ra g e d to explore the

ACI website fo r m e m b e rs h i p o p p o rtu n ities, c o m m ittee activities, and a wide

va ri ety of c o n crete reso u rces. As a vo l u nteer m e m be r-d riven o rg a n izati o n ,

ACI i nvites p a rt n e rs h i p s a n d w e l c o m e s a l l con crete p rofess i o n a l s w h o w i s h to

be p a rt of a res pecte d , c o n n ecte d , soc i a l g ro u p t h at p rovides a n op portu n ity

for p rofessi o n a l g rowth, n etwo r k i n g a n d e nj oy m e nt.

9
1 1 1 11 11 1 1 I
781942 727729
Coovriaht American Concrete Institute