IN-LB Inch-Pound Units

SI International System of Units

Report on Deflection
of Nonprestressed
Concrete Structures
Reported by ACI Committee 435
ACI 435R-20
 First Printing
November 2020
ISBN: 978-1-64195-128-9

Report on Deflection of Nonprestressed Concrete Structures

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 [Link]
[Link]. 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, through electronic subscription,
or reprint and may be obtained by contacting ACI.

Most ACI standards and committee reports are gathered together in the annually revised the ACI
Collection of Concrete Codes, Specifications, and Practices.

American Concrete Institute

38800 Country Club Drive
Farmington Hills, MI 48331
Phone: +1.248.848.3700
Fax: +1.248.848.3701
[Link]
 ACI 435R-20

Report on Deflection of Nonprestressed Concrete

Structures
Reported by ACI Committee 435
Eric S. Musselman, Chair Dylan Freytag, Secretary

Peter H. Bischoff Mamdouh M. El-Badry Adam S. Lubell Debrethann R. Orsak

Allan P. Bommer Amin Ghali Faris A. Malhas Maria A. Polak
Flora A. Calabrese Mayrai Gindy Michael C. Mota Mahmoud M. Reda Taha
Eamonn F. Connolly Shawn P. Gross Hani H. Nassif Andrew Scanlon
Norbert J. Delatte Young Hak Lee Edward G. Nawy Richard H. Scott

Consulting Members
Alex Aswad Satyendra Ghosh Bernard L. Meyers Himat T. Solanki
Finley A. Charney Peter Lenkei Vilas S. Mujumdar Susanto Teng

This report presents a consolidated treatment of initial and CHAPTER 2—NOTATION AND DEFINITIONS, p. 3
time-dependent deflection of nonprestressed reinforced concrete 2.1—Notation, p. 3
members such as simple and continuous beams and one-way and 2.2—Definitions, p. 3
two-way slab systems. It presents the current state of practice of
deflection prediction as well as analytical methods for computer
CHAPTER 3—MATERIAL PROPERTIES, p. 3
use in deflection estimation. Topics include material properties,
3.1—Objective, p. 3
deflection of reinforced concrete one-way flexural members, deflec-
tion of two-way slab systems, and reducing deflection of concrete 3.2—Material properties affecting deflection, p. 4
members. 3.3—Concrete material properties, p. 4
3.4—Reinforcement material properties, p. 9
Keywords: camber; cracking; creep; curvature; deflection; modulus of
rupture; moments of inertia; serviceability; shrinkage; time-dependent CHAPTER 4—DEFLECTION OF REINFORCED
deflection.
CONCRETE ONE-WAY FLEXURAL MEMBERS, p. 9
4.1—General, p. 9
CONTENTS 4.2—Control of deflection, p. 10
4.3—Short-term deflection calculation, p. 11
CHAPTER 1—INTRODUCTION AND SCOPE, p. 2 4.4—Long-term deflection calculation, p. 18
1.1—Introduction, p. 2 4.5—Temperature-induced deflections, p. 21
1.2—Scope, p. 2
CHAPTER 5—DEFLECTION OF A TWO-WAY
SLAB SYSTEM, p. 22
5.1—Introduction, p. 22
ACI Committee Reports, Guides, and Commentaries are 5.2—Deflection calculation methods for two-way slab
intended for guidance in planning, designing, executing, and systems, p. 23
inspecting construction. This document is intended for the use 5.3—Minimum thickness requirements, p. 26
of individuals who are competent to evaluate the significance 5.4—Loads for deflection calculation, p. 28
and limitations of its content and recommendations and who 5.5—Variability of deflections, p. 31
will accept responsibility for the application of the material it
contains. The American Concrete Institute disclaims any and 5.6—Allowable deflections, p. 32
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 ACI 435R-20 supersedes ACI 435R-95(03) and became effective November 2020.
Copyright © 2020, American Concrete Institute.
documents. If items found in this document are desired by
All rights reserved including rights of reproduction and use in any form or by any
the Architect/Engineer to be a part of the contract documents, means, including the making of copies by any photo process, or by electronic or me-
they shall be restated in mandatory language for incorporation chanical device, printed, written, or oral, or recording for sound or visual reproduction
by the Architect/Engineer. or for use in any knowledge or retrieval system or device, unless permission in writing
is obtained from the copyright proprietors.

1
2 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

CHAPTER 6—REDUCING DEFLECTION OF of prestressed concrete are not addressed in this document,
CONCRETE MEMBERS, p. 32 although prestressing can be an effective tool for controlling
6.1—Introduction, p. 32 both short- and long-term deflections.
6.2—Design techniques, p. 32
6.3—Construction techniques, p. 34 1.2—Scope
6.4—Materials selection, p. 35 The principal causes of deflections taken into account
6.5—Summary, p. 36 in this report are those due to elastic deformation, flexural
cracking, creep, shrinkage, and temperature effects. This
CHAPTER 7—REFERENCES, p. 36 document is composed of two introductory chapters and four
Authored documents, p. 36 main chapters that provide information on calculating and
controlling deflections of members constructed using rein-
APPENDIX A—DEFLECTION DESIGN EXAMPLES, forced concrete. The organization of the report is:
p. 39 a) Chapter 1—Introduction and Scope provides back-
Example A.1—Deflection of a simply supported slab, p. 39 ground information on the document.
Example A.2—Age-adjusted deflection of simply b) Chapter 2—Notation and Definitions provides a
supported slab, p. 43 listing of the notation used throughout the document.
Example A.3—Short- and long-term deflection of a four- c) Chapter 3—Material Properties discusses material
span continuous beam, p. 44 properties that affect deflections.
Example A.4—Temperature-induced deflections, p. 48 d) Chapter 4—Deflection of Reinforced Concrete
One-Way Flexural Members discusses behavior of
APPENDIX B—TWO-WAY SLAB DEFLECTION uncracked and cracked members, and time-dependent
EXAMPLES, p. 48 effects. It also includes the relevant code procedures
Example B.1—Deflection design example for long-term and expressions for deflection computation in reinforced
deflection of a two-way slab, p. 48 concrete beams. Numerical examples are included to
Example B.2—Deflection calculation for a flat plate using illustrate the standard calculation methods for simply
the crossing beam method, p. 52 supported and continuous concrete beams.
Example B.3—Minimum thickness calculation, p. 54 e) Chapter 5—Deflection of Two-Way Slab Systems
covers the deflection behavior of reinforced two-way-
CHAPTER 1—INTRODUCTION AND SCOPE action slabs and plates. This chapter gives an overview
of classical and other methods of deflection estimation,
1.1—Introduction such as the crossing beam analogy and the finite element
Design for serviceability is central to the work of structural method for immediate deflection computation. It also
engineers and code-writing bodies. It is also essential to users discusses approaches for determining the minimum thick-
of the designed structures. Increased use of high-strength ness requirements for two-way slabs and plates and gives
concrete and higher-strength reinforcing bars, coupled with a detailed computational example for evaluating the long-
more detailed computer-aided designs, has resulted in lighter term deflection of a two-way reinforced concrete slab.
and more material-efficient and, thus, more flexible structural The chapter emphasizes the uncertainties inherent in esti-
members and systems. This in turn has necessitated better mating deflections of two-way slab systems.
prediction and control of short-term and long-term behavior f) Chapter 6—Reducing Deflection of Concrete
of concrete structures at service loads. Members gives practical and remedial guidelines for
This report presents a consolidated treatment of initial improving and controlling the deflection of reinforced
and time-dependent deflection of nonprestressed reinforced concrete members, hence enhancing their overall long-
concrete members such as simple and continuous beams and term serviceability.
one- and two-way slab systems. It presents current engineering It should be emphasized that the magnitude of actual
practice in design for control of deformation and deflection of deflection in concrete structural members, particularly in
concrete members and includes methods presented in ACI 318 buildings, which are the emphasis and the intent of this
plus selected other approaches suitable for computer-based report, can only be estimated with limited accuracy. This
use in deflection computation. Design examples are given at is because of the large variability in the properties of the
the end of one- and two-way framing chapters showing how constituent materials of these members, the quality control
to evaluate deflection and, thus, control it through adequate exercised in their construction, and the construction methods
design for serviceability. The content of the report as well used. Therefore, for practical considerations, the computed
as the step-by-step examples are intended to familiarize deflection values in the illustrative examples at the end of
practitioners with the current methods for estimating deflec- each chapter should be interpreted with this in mind.
tions in buildings as well as analytical methods suitable for In summary, this single document gives design engineers
computer-based application. The examples apply ACI 318 the key tools for estimating, and thereby controlling through
requirements and a recommended alternative approach with design, the expected deflection in nonprestressed reinforced
a lower cracking moment (to account for shrinkage restraint). concrete building structures. The material presented and the
Methods for predicting initial and time-dependent deflections design examples will help to enhance serviceability when

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 3

used judiciously by the engineer. Designers, constructors, t = time, s

and codifying bodies can draw on the material presented w = uniformly distributed load (load per unit length),
in this document to achieve serviceable deflection of lb/in. (N/mm)
constructed facilities. wc = unit weight of normalweight concrete or equilib-
rium density of lightweight concrete, lb/ft3 (kg/m3)
CHAPTER 2—NOTATION AND DEFINITIONS yt = distance from centroidal axis of gross section,
neglecting reinforcement, to extreme fiber in
2.1—Notation tension, in. (mm)
As = area of nonprestressed tension steel, in.2 (mm2) ∆ = elastic deflection of a beam or slab, in. (mm)
As′ = area of nonprestressed steel in compression zone, ∆cr = additional deflection due to creep, in. (mm)
in.2 (mm2) ∆inc = incremental deflection that occurs after attachment
b = width of the section, in. (mm) on nonstructural elements (includes long-term
bw = web width, in. (mm) deflection ∆LT from sustained loads and immediate
Ct = creep coefficient of concrete at time t, days deflection from the remaining part of live load that
Cu = ultimate creep coefficient of concrete is not sustained, in. (mm)
c = depth of centroidal axis, in. (mm) ∆L = initial (immediate) deflection due to live load, in.
d = distance from the extreme compression fiber to (mm)
centroid of tension reinforcement, in. (mm) ∆LT = deflection from long-term effects, in. (mm)
d′ = distance from the extreme compression fiber to ∆sh = additional deflection due to shrinkage, in. (mm)
centroid of compression reinforcement, in. (mm) ∆sus = initial (immediate) deflection due to sustained load,
db = bar diameter, in. (mm) in. (mm)
E = modulus of elasticity, psi (MPa) εcf = strain due to stress in the concrete
Ec = modulus of elasticity of concrete, psi (MPa) εo = free strain, such as unrestrained shrinkage
fc = stress in concrete, psi (MPa) εsf = strain due to stress in nonprestressed steel
fc′ = specified compressive strength of concrete, psi (MPa) εsh = shrinkage strain of concrete
fcr = stress to cause cracking in concrete, psi (MPa) (εsh)t = shrinkage strain of concrete at time t, days
fr = modulus of rupture of concrete, psi (MPa) (εsh)u = ultimate shrinkage strain of concrete
fres = stress from restraint to shrinkage, psi (MPa) εt = total strain
fs = stress in nonprestressed steel, psi (MPa) ζ = distribution coefficient
fy = specified yield strength of nonprestressed rein- κ = cross section curvature, in.–1 (mm–1)
forcing steel, psi (MPa) κ sh = shrinkage curvature, in.–1 (mm–1)
h = overall thickness of a member, in. (mm) λc = creep multiplier for long-term deflection
hf = flange thickness, in. (mm) λsh = shrinkage warping multiplier for long-term deflection
I = moment of inertia, in.4 (mm4) λt = total multiplier for long-term deflection
Icr = moment of inertia of the cracked section trans- λ∆ = time-dependent multiplier for long-term deflection
formed to concrete, in.4 (mm4) ν = Poisson’s ratio
Ie = effective moment of inertia for computation of ρ = nonprestressed tension reinforcement ratio (As/bd)
deflection, in.4 (mm4) ρ′ = reinforcement ratio for nonprestressed compres-
Ig = moment of inertia for gross concrete section about sion steel (As′/bd)
centroidal axis, neglecting reinforcement, in.4 (mm4) ξ = time-dependent multiplier for deflection
k = depth of compression zone divided by d
ℓ = span length, in. (mm) 2.2—Definitions
ℓn = distance from the inside of the support to the inside Please refer to the latest version of ACI Concrete Termi-
of support, clear span, in. (mm) nology for a comprehensive list of definitions.
M = bending moment, lb-in. (N-mm)
M1 = moment at End 1 of a continuous member, lb-in. CHAPTER 3—MATERIAL PROPERTIES
(N-mm)
M2 = moment at End 2 of a continuous member, lb-in. 3.1—Objective
(N-mm) Deflections in reinforced concrete structures are affected
Ma = maximum service load moment (unfactored) at significantly by numerous material properties, including
stage deflection is computed, lb-in. (N-mm) concrete and reinforcement moduli of elasticity, concrete
Mcr = cracking moment, lb-in. (N-mm) modulus of rupture, creep, and shrinkage. The purposes
MD = moment due to dead load, lb-in. (N-mm) of this chapter are to: 1) briefly address how each of these
ML = moment due to live load, lb-in. (N-mm) material properties affects deflection; and 2) provide brief
Mm = midspan moment, lb-in. (N-mm) guidance on the most common expressions recommended
N = axial member load, lb (N) by various ACI committees for estimation of these param-
n = modular ratio Es/Ec eters during the design process. This chapter is not intended
T = temperature, °F (°C) to provide a comprehensive review of all the material prop-

American Concrete Institute – Copyrighted © Material – [Link]

4 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

erty expressions that have been suggested by researchers, tion. The most common approach for calculating deflection
committees, and design codes. is to use a lower cracking moment computed with a reduced
modulus of rupture to account for the tensile stresses that
3.2—Material properties affecting deflection develop from restraint to shrinkage. For indeterminate struc-
The primary material properties that affect the deflection of tures, the member forces and moments themselves (and thus
reinforced concrete structures are identified in the following. resulting deflections) may be significantly affected by creep
3.2.1 Concrete modulus of elasticity―The elastic modulus and shrinkage. Creep effects can lead to a redistribution of
of concrete (Ec) has a significant impact on deflections in forces among the members in a structure. Shrinkage restraint
reinforced concrete structures, as it is a direct measure of can also develop among the members in a structure, leading
material stiffness. Based on simple mechanics principles to cracking and a subsequent loss in stiffness.
developed for elastic theory, deflection is inversely related 3.2.4 Reinforcement modulus of elasticity―The effect of
to the elastic modulus of the concrete. However, the concrete reinforcement modulus of elasticity on deflections can be
elastic modulus also affects the sectional moment of inertia significant but is limited primarily to the effect on sectional
at the cracked section (Icr) and, thus, the relationship between properties and the restraint of concrete shrinkage. The use of
concrete elastic modulus and deflection is far more complex. a lower modulus reinforcement will lead to smaller sectional
In indeterminate structures, the member forces and moments of inertia for a constant reinforcement ratio,
moments themselves (and thus resulting deflections) will be particularly for the cracked section, which in turn will lead
affected by the modulus of elasticity if it varies across the to larger deflections. On the other hand, a lower modulus
different components of the structure. reinforcement will offer less internal restraint to concrete
3.2.2 Concrete modulus of rupture―The concrete modulus shrinkage and thus will not have as significant an effect in
of rupture (fr) affects deflection primarily in that it estab- reducing the effective cracking moment.
lishes the stress threshold at which a member cracks. Upon
reaching the modulus of rupture, the member (or a portion of 3.3—Concrete material properties
the member) transitions from an uncracked to a cracked state Expressions recommended by ACI committees for esti-
and there is a significant reduction in stiffness accompanied mation of concrete material properties during the design
by an increase in deflection. While most reinforced concrete process are presented in this section. In general, these prop-
members are designed to be cracked at service loads, the erties vary widely among different concretes or even among
extent of cracking along the length of the member will have different batches of concrete produced using a single set of
an impact on the member deflection. mixture proportions. Therefore, these expressions should
For indeterminate structures, the extent of cracking has a only be expected to provide rough estimates for these
significant effect on member stiffness, which in turn affects concrete material properties, with a relatively low degree
internal member forces and moments used in evaluating of accuracy. When a higher level of accuracy in deflection
deflections. calculations is desired, concrete material properties should
3.2.3 Time-dependent concrete properties―Creep and be measured using appropriate ASTM standards on test
shrinkage of concrete have a pronounced effect on the specimens produced using field trial batches.
deflection behavior of reinforced concrete members. Creep 3.3.1 Concrete modulus of elasticity―The modulus of
of concrete will cause additional deformation over time, elasticity is strongly influenced by the constitutive mate-
resulting in increased deflection of flexural members under rials and proportions used and is especially affected by the
sustained load. In many cases, the additional deflection will properties and quantity of the coarse aggregate. An increase
exceed the instantaneous deflection. The consideration of in the modulus of elasticity is commonly associated with
creep is not as simple as using a multiplier based solely on an increase in compressive strength because the slope of
the creep coefficient of the concrete material because creep the ascending branch of the stress-strain diagram becomes
also causes a redistribution of internal stresses over time. steeper for higher-strength concretes, but at a lower rate than
This redistribution of internal stresses can be considered the compressive strength. The value of the secant modulus of
to cause a variation in effective sectional properties (at any elasticity at 0.45fc′ for normal-strength concretes at 28 days
point in the member) with time because the neutral axis and is usually approximately 4 × 106 psi (28,000 MPa), whereas
moment of inertia will change as the compressive strain in for higher-strength concretes with compressive strengths in
the concrete increases with age due to creep. excess of 8000 psi (56 MPa), values as high as 8 × 106 psi
Similarly, shrinkage of concrete will directly lead to time- (56,000 MPa) have been reported (Myers and Yang 2004).
dependent deflections whenever warping is present due ACI 318 recommends Eq. (3.3.1) for computing the
to differential shortening within the depth of the member. modulus of elasticity of concretes with densities in the range
Differential shortening may occur due to an unsymmetric of 90 to 150 lb/ft3 (1445 to 2325 kg/m3) based on the secant
arrangement of internal reinforcement or external environ- modulus at the 0.45fc′ intercept.
mental effects. Restraint to shrinkage, either due to external
restraint or internal restraint of embedded reinforcement, can Ec = 33wc1.5 f c′ (psi)
also have a significant effect on deflection by inducing tensile (3.3.1)
Ec = 0.043wc1.5 f c′ (MPa)
stresses in the concrete that effectively reduce the external
load required to cause cracking and thereby increase deflec-

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 5

Table 3.3.3—Creep and shrinkage ratios from age 60 days to the indicated concrete age (Branson 1977)
Concrete age
Creep, shrinkage ratios 2 months 3 months 6 months 1 year 2 years >5 years
Ct/Cu 0.48 0.56 0.68 0.77 0.84 1.00
(εsh)t/(εsh)u—MC 0.46 0.60 0.77 0.88 0.94 1.00
(εsh)t/(εsh)u—SC 0.36 0.49 0.69 0.82 0.91 1.00
Notes: MC is moist-cured; and SC is steam-cured.

Although Eq. (3.3.1) is considered to provide reason- b) Age of loading: 20 days for both moist and steam-cured
able predictions for concretes in the strength range up to concrete
6000 psi (42 MPa), significant variation is still expected c) Minimum thickness of component: 6 in. (150 mm)
between predicted and measured values. ACI 318 commen- Table 3.3.3 includes creep and shrinkage ratios at different
tary suggests that differences of up to 20 percent between times after loading, where the ratio indicates the percentage
predicted and measured values is to be expected. of ultimate creep or shrinkage that can be expected to have
3.3.2 Concrete modulus of rupture―ACI 318 recom- occurred at a certain age. Most of the time-dependent defor-
mends Eq. (3.3.2a) for computing the modulus of rupture of mation can be expected to occur within the first year.
concrete with different densities ACI Committee 209 (1971, 1982) and ACI 209R recom-
mend a time-dependent model for creep and shrinkage under
f= 7.5λ f c′ (psi) standard conditions as developed by Branson (1971, 1977).
r
(3.3.2a) The term “standard conditions” is defined for a number of
f r = 0.62λ f c′ (MPa) variables related to material properties, the ambient temper-
ature, humidity, and size of members. The standard condi-
where tions for both creep and shrinkage are:
λ = 1.0 for normal-density concrete (145 to 150 lb/ft3 [2325 a) Age of concrete at load application = 3 days (steam),
to 2400 kg/m3]) 7 days (moist)
= 0.85 for normalweight fine aggregate and lightweight b) Ambient relative humidity = 40 percent
coarse aggregate concrete c) Minimum member thickness = 6 in. (150 mm)
= 0.75 for all-lightweight aggregate concrete d) Concrete slump = 3 in. (75 mm)
Alternatively, ACI Committee 435 (1978) recommended e) Fine aggregate content = 50 percent
using Eq. (3.3.2b) for computing the modulus of rupture of f) Air content = 6 percent
concrete with densities wc in the range of 90 to 145 lb/ft3 The coefficient for creep at time t (days) after load appli-
(1445 to 2325 kg/m3). This equation provides a more contin- cation is given by Eq. (3.3.3a)
uous consideration of concrete density than Eq. (3.3.2a) and
yields slightly higher values of fr.  t 0.6 
Ct =  0.6  Cu (3.3.3a)
 10 + t 
f r = 0.65 wc f c′ (psi)
(3.3.2b) where Cu = 2.35KhcKdcKscKfcKaccKtoc = 2.35 for standard
f r = 0.013 wc f c′ (MPa) conditions.
Each K coefficient is a correction factor for conditions
A large degree of scatter can be expected in modulus of other than standard as follows:
rupture test results, so only limited improvement in the accu- Khc = relative humidity factor
racy should be expected when using the aforementioned equa- Kdc = minimum member thickness factor
tions in design calculations. Furthermore, these expressions Ksc = concrete consistency factor
are based on small prism specimens and do not reflect the Kfc = fine aggregate content factor
presence of tensile stresses due to shrinkage restraint caused Kacc = air content factor
by embedded reinforcing bars and supporting elements. Ktoc = age of concrete at load application factor
3.3.3 Time-dependent concrete properties―ACI 318 does Graphic representations and general equations for the
not specify explicit values for the concrete ultimate creep modification factors (K-values) for nonstandard conditions
coefficient Cu and ultimate shrinkage strain (εsh)u. However, are given in Fig. 3.3.3a (Meyers and Thomas 1983).
they can be evaluated from several equations available in For concrete moist-cured for 7 days, the free shrinkage
the literature (ACI Committee 209 1982; Bažant and Panula strain can be estimated using Eq. (3.3.3b) at any time t,
1980; Branson 1977). ACI Committee 435 (1978) suggested where t is the time in days from the end of initial curing
that the average values for Cu and (εsh)u can be estimated as
1.60 and 400 × 10–6, respectively. These values correspond  t 
to the following conditions: (ε=
sh )t   (ε sh )u (3.3.3b)
 35 + t 
a) 70 percent average relative humidity

American Concrete Institute – Copyrighted © Material – [Link]

6 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Fig. 3.3.3a—Creep correction factors for nonstandard conditions, ACI 209R method (Meyers and Thomas 1983).
and for concrete steam-cured for 1 to 3 days, the shrinkage (1977). The presence of chemical admixtures, specifically
strain can be estimated using Eq. (3.3.3c) at any time t, water reducers, has been shown to decrease creep and
where t is the time in days from the end of steam curing shrinkage (Brooks and Al-Qarra 1999), a factor that was
not included in the development of the ACI Committee 209
(1971, 1982) and ACI 209R prediction models. ACI 209.2R
 t 
( ε=
sh ) t   ( ε sh )u
55 + t 
(3.3.3c) outlines other creep and shrinkage functions in addition to
the ACI 209R function based on Branson’s work.
One approach that is employed to include the influence
where (εsh)u = 780 × 10-6 KhsKdsKfsKBsKacs = 780 × 10–6 for of creep in deflection calculations is to use an age-adjusted
standard conditions. modulus of elasticity of the concrete, E¯c (t,t0), which can be
Each K coefficient is a correction factor for other than stan- calculated using Eq. (3.3.3d).
dard conditions. All coefficients are the same as defined for
creep except KBs, which is a coefficient for cement content.
Ec (t0 )
Graphic representation and general equations for the modi- Ec (t , t0 ) = (3.3.3d)
fication factors for nonstandard conditions are given in Fig. 1 + χCt (t , t0 )
3.3.3b (Meyers and Thomas 1983). The aforementioned
procedure, using standard and correction equations and where Ec(t0) is the modulus of elasticity of concrete at time
extensive experimental comparisons, is detailed in Branson t0. Ct(t,t0) is the creep coefficient, which is equal to the creep

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 7

Fig. 3.3.3b―Shrinkage correction factors for nonstandard conditions, ACI 209R method (Meyers and Thomas 1983).
strain in the period t0 to t divided by the instantaneous strain. concretes (ACI 213R). The modulus of elasticity for light-
The aging coefficient χ was introduced by Trost (1967) and weight concrete may be estimated using Eq. (3.3.1), which
Bažant and Kim (1979) and is mainly a function of t0 and t directly considers the weight difference via the parameter
(Ghali et al. 2011). A value of 0.8 can be used in most cases wc. It is important to note however, that the simplified form
because χ varies between 0.7 and 0.9. Tabulated values of χ of Eq. (3.3.1) often used in practice: Ec = 57,000 f c′ (psi)
are given in ACI 209R.
(4700 f c′ [MPa]) is only valid for normal-density concrete
3.3.4 Special concretes―In this section, brief guidance is
and should never be used for lightweight concrete.
provided on the material properties for special concretes.
The modulus of rupture is also proportionally lower for
[Link] Lightweight concrete—Lightweight concrete is
lightweight concrete as compared to normal-density concrete
structural concrete made with lightweight aggregates with
of the same compressive strength. Equations (3.3.2a) or
bulk density less than 70 lb/ft3 (1120 kg/m3) (ACI 213R).
(3.3.2b) may be used to estimate the modulus of elasticity
Extensive information on the material properties of light-
of lightweight concrete. In the case of Eq. (3.3.2a), the equa-
weight concrete is provided in ACI 213R.
tion should be used as presented when the tensile splitting
The modulus of elasticity of lightweight concrete is
strength fct is not specified. Otherwise, the equation should
lower than for normal-density concrete because the moduli
of lightweight aggregates are significantly lower than the be modified by substituting fct /6.7 for f c′ (psi) (1.8fct for
moduli of sand, stone, and gravel used in normal-density f c′ [MPa]) but the value of fct /6.7 (psi) (1.8fct [MPa])

American Concrete Institute – Copyrighted © Material – [Link]

8 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

agreement that the drying rates in high-strength concrete

should not exceed f c′ . Work by Nassar (2002) indicates
will be slower when only drying shrinkage is considered.
that Eq. (3.3.2a) may be unconservative for lightweight
Nagataki and Yonekura (1978) reported that the shrinkage of
concretes with higher compressive strengths.
high-strength concrete containing high-range water reducers
ACI 213R reports that while there is a large variation
was less than for lower-strength concrete.
in measured values of drying shrinkage for lightweight
On the other hand, there is general agreement that the
concrete, it appears that low-strength lightweight concrete
creep coefficients for high-strength concretes are lower than
generally has greater drying shrinkage than comparable
for normal-strength concrete (ACI 363R). The creep coef-
normal-density concretes. At higher strengths, some light-
ficient (ratio of creep strain to initial elastic strain under
weight concrete has been found to exhibit lower shrinkage.
sustained axial compression) for high-strength concrete may
Creep coefficients reported in the literature are generally
be as low as one-half that generally associated with low-
higher for lightweight concrete as compared to comparable
strength concrete (Ngab et al. 1981; Nilson 1985). However,
normal-density concrete.
high-strength concretes are often exposed to a higher level
[Link] High-strength concrete—High-strength concrete is
of sustained stress such that the total creep strain may be
defined by ACI 363R as concrete with specified compressive
similar to that for normal strength concrete.
strengths exceeding 8000 psi (56 MPa). Extensive informa-
[Link] Self-consolidating concrete—Self-consolidating
tion on the material properties of high-strength concrete is
concrete is concrete proportioned such that it may be consol-
provided in ACI 363R.
idated without mechanical vibration. Many self-consol-
The modulus of elasticity is generally higher for high-
idating concrete mixtures use admixtures and, in some
strength concrete as compared to normal-strength concrete.
cases, a reduced quantity of coarse aggregate to maximize
Several equations have been proposed by various researchers
stability and minimize segregation and bleeding, which can
for the modulus of elasticity of high-strength concrete, but
lead to differences in hardened material properties relative
a study by ACI 363R found that no single empirical expres-
to conventionally-placed concrete. However, self-consoli-
sion estimated the modulus of elasticity to a high degree of
dating mixtures can also be engineered to have properties
accuracy for all concretes considered in the study. However,
similar to, or better than, those of conventionally placed
Eq. (3.3.4.2a) was found by ACI 363R to be a relatively
concrete (ACI 237R). Information on the material properties
reliable lower-bound expression for normal-density high-
of self-consolidating concrete is provided in ACI 237R.
strength concrete based on most of the test data considered
In some studies, self-consolidating concrete mixtures with
high sand-to-total aggregate ratios (s/a) have been observed
=Ec 40,000 f c′ + 1,000,000 (psi) to have a modulus of elasticity of approximately 10 to 15
(3.3.4.2a)
=Ec 3320 f c′ + 6900 (MPa) percent lower than that of conventional concrete of similar
compressive strength, whereas, in other studies, there has
been little effect observed on the modulus of elasticity (ACI
The values reported by various investigators (ACI 363R)
237R). ACI 237R recommends the use of Eq. (3.3.1) for the
for the modulus of rupture of both low-density and normal-
modulus of elasticity of self-consolidating concrete in appli-
density high-strength concretes range between 7.5 f c′ and cations where the modulus is not critical, and recommends
12 f c′ (psi) (0.62 f c′ and 1.0 f c′ [MPa]). Therefore, Eq. testing to determine the modulus of elasticity when greater
(3.3.2a) can underestimate the modulus of rupture of normal- accuracy is required.
density high-strength concrete, and ACI 363R recommends ACI 237R does not recommend any unique considerations
the use of Eq. (3.3.4.2b) for high-strength concrete instead for the estimation of the modulus of rupture for self-consol-
idating concrete. As such, Eq. (3.3.2a) or (3.3.2b) may be
f r = 11.7 f c′ (psi) used to estimate the modulus of rupture. A study by Sonebi
(3.3.4.2b) and Bartos (2001) found that the modulus of rupture of self-
f r = 0.97 f c′ (MPa) consolidating concrete may actually be higher than that of
conventional concrete with similar proportions.
For high-strength concretes with low water-cementitious ACI 237R notes that several studies have reported the
materials ratio (w/cm) and high binder content, the mecha- drying shrinkage of self-consolidating concrete to be similar
nisms of shrinkage may be more complex than for normal- to or lower than that of conventional concrete with similar
strength concretes, with chemical and autogenous shrinkage compressive strength. When mixture proportions are similar
supplementing the effects of drying shrinkage. There are to those for conventionally placed concrete, self-consoli-
considerably less data available on these mechanisms of dating concrete has been found to have similar creep charac-
shrinkage due to the difficulty in making such measurements. teristics to those for conventional concrete. When a greater
According to ACI 363R, there is no clear trend with respect paste volume is used for stability considerations, a higher
to drying shrinkage of high-strength concrete, except that creep in comparison to conventionally placed concrete
it is generally similar to the shrinkage of normal-strength may occur (ACI 237R). ACI 237R recommends testing to
concretes. Some researchers have reported a relatively high determine the creep coefficient when creep is an important
initial rate of shrinkage for high-strength concrete (Swamy parameter in design.
and Anand 1973), but ACI 363R reports that there is general

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 9

[Link] Fiber-reinforced concrete—Fiber-reinforced values provided by the reinforcement manufacturer should be

concrete consists primarily of hydraulic cements, aggregates, used when available. Ranges of elastic moduli for different
and discrete reinforcing fibers. Common materials used commercially-available FRPs are provided by ACI 440.1R:
as fiber reinforcement include steel, glass, and synthetic a) Glass-fiber-reinforced polymer (GFRP): Ef = 5.1 to
materials, including polypropylene and polyethylene. ACI 7.4 × 106 psi (35,000 to 51,000 MPa)
544.1R provides detailed information on the characteristics b) Carbon-fiber-reinforced polymer (CFRP): Ef = 15.9 to
and behaviors of the common types of fiber reinforcement, 84.0 × 106 psi (110,000 to 580,000 MPa)
and their effect on concrete properties. c) Aramid-fiber-reinforced polymer (AFRP): Ef = 6.0 to
The primary contribution of fibers to concrete is to 18.2 × 106 psi (41,000 to 125,000 MPa)
improve the pre- and post-cracking behavior of the concrete
in tensile regions. Some fibers have the ability to increase CHAPTER 4—DEFLECTION OF REINFORCED
the tensile capacity of the concrete, effectively increasing the CONCRETE ONE-WAY FLEXURAL MEMBERS
modulus of rupture. No general equation has been proposed
to quantify this increase as the effect is highly dependent on 4.1—General
the fiber type, shape, and application. In addition to affecting 4.1.1 Introduction—In many structures, deflection rather
the precracking behavior, fibers also allow for stress transfer than strength is the controlling factor. For these cases,
across a crack. This typically results in a decrease in crack deflection computations will determine the proportions and
width, and a decrease in crack spacing. This behavior configuration of the structural system members. Member
serves to decrease the immediate deflections of the struc- stiffness is a function of short- and long-term mechanical
tural member by reducing the rise of the neutral axis upon properties of the concrete and the extent of flexural cracking.
cracking, thereby reducing the change in the post-cracking Hence, expressions defining the modulus of rupture,
moment of inertia relative to the gross moment of inertia. As modulus of elasticity, creep, shrinkage, and temperature
with the modulus of rupture, no general equation has been effects are important parameters in predicting the deflec-
proposed to quantify this response due to the variation in tion of reinforced concrete members (refer to Chapter 3 for
performance of various fiber types and geometries. a detailed explanation of these properties). The stiffness and
Fibers can also affect the creep properties of concrete. amount of reinforcement also control the member stiffness
Compression creep is not significantly affected; however, after cracking.
long-term flexural creep is affected by the presence of The designer may choose to calculate deflections and
fibers (ACI 544.5R). The long-term deflection response is check that these computed values are less than specified
dependent on the characteristics of the fibers used in the permissible limits. Alternatively, design codes such as ACI
concrete. Longer fibers with higher moduli of elasticity 318 give specified maximum span/depth ratios, with limits of
tend to improve long-term deflection behavior. Specifically, applicability, for which serviceability is assumed to be satis-
fiber-reinforced concrete beams with steel fibers have shown fied and deflections do not need to be explicitly calculated.
less long-term deflections than reinforced concrete beams 4.1.2 Objectives—This chapter covers the initial (imme-
without fibers (Tan et al. 1994). However, fibers with lower diate) and time-dependent deflections at service load levels
moduli of elasticity tend to exhibit more creep behavior over under static conditions for one-way nonprestressed flexural
time (Bernard 2004). concrete members. Details are also provided on deemed-to-
The shrinkage response of concrete is generally not comply limiting span-thickness ratios for which deflections
affected by the presence of fibers. However, the potential do not need to be calculated. This chapter is intended to give
increase in the apparent modulus of rupture and the ability the designer enough basic background to design concrete
of the fibers to distribute cracking may decrease the effect of members that perform adequately under service loads, taking
shrinkage on the deflection behavior of the concrete. into account cracking and both short- and long-term effects.
Information is presented on the effective moment of inertia
3.4—Reinforcement material properties method prescribed in ACI 318 to evaluate deflection. ACI
Deflections are not as sensitive to variation in reinforce- 318 permits use of an expression for the effective moment of
ment material properties in comparison to concrete proper- inertia (Ie) to compute deflection, unless a more comprehen-
ties, so the use of specified or manufacturer-supplied values sive analysis is carried out to determine the member stiffness.
is usually adequate and additional testing on field specimens The existing approach for Ie used by ACI 318-14 and earlier
is not required. The property of interest is the reinforcement editions can underestimate deflection in certain conditions
modulus of elasticity. Typical values are provided in the (Bischoff and Scanlon 2007); hence, an alternative expres-
following. sion for the effective moment of inertia that works well over
3.4.1 Steel reinforcement—ACI 318 specifies using the a wide range of conditions is proposed. Other modifications
value Es = 29 × 106 psi (200,000 MPa) for the modulus of and more refined methods that require detailed calculations
elasticity of nonprestressed reinforcing steel. for estimating deflection are also discussed. This includes the
3.4.2 Fiber-reinforced polymer (FRP) reinforcement— effects of preloading from construction loads and reduced
The modulus of elasticity of FRP reinforcement (Ef) varies flexural stiffness that can result when tensile stresses develop
widely, depending on the type and volume fraction of the in the concrete from restraint to shrinkage.
fiber used in the composite. In general, elastic modulus

American Concrete Institute – Copyrighted © Material – [Link]

10 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Table 4.2.1—Recommended tension reinforcement ratios for nonprestressed one-way members so that
deflection will normally be within acceptable limits (ACI Committee 435 1978)
Members Cross section Normalweight concrete Lightweight concrete
Not supporting or not attached to nonstructural elements likely ρ ≤ 35 percent ρb ρ ≤ 30 percent ρb
Rectangular, T, or box
to be damaged by large deflections ρw ≤ 40 percent ρb ρw ≤ 35 percent ρb
Supporting or attached to nonstructural elements likely to be ρ ≤ 25 percent ρb ρ ≤ 20 percent ρb
Rectangular, T, or box
damaged by large deflections ρw ≤ 30 percent ρb ρw ≤ 25 percent ρb
Notes: For continuous members, the positive region steel ratios only may be used. ρb refers to the balanced steel ratio based on ultimate strength.

Table 4.2.2a—Minimum thickness of nonprestressed beams and one-way slabs unless deflections are
computed (from ACI 318)
Minimum thickness, h
Simply supported One end continuous Both ends continuous Cantilever
Member Members not supporting or attached to partitions or other construction likely to be damaged by large deflections
Solid one-way slabs ℓ/20 ℓ/24 ℓ/28 ℓ/10
Beams or ribbed one-way slabs ℓ/16 ℓ/18.5 ℓ/21 ℓ/8
Notes: Values given should be used directly for members with normalweight concrete and Grade 60 (Grade 420) reinforcement. For other conditions, the values shall be modified
as follows:
a) For fy other than 60,000 psi (420 MPa), the values shall be multiplied by 0.4 + fy/100,000 (0.4 + fy/700).
b) For lightweight concrete having equilibrium density wc in the range 90 to 115 lb/ft3 (1440 to 1840 kg/m3), the values should be multiplied by 1.65 – 0.005wc (1.65 – 0.0003wc)
but not less than 1.09.

Table 4.2.2b—Minimum thickness of beams and one-way slabs used in roof and floor construction (ACI
Committee 435 1978)
Members not supporting or not attached to nonstruc- Members supporting or attached to nonstructural
tural elements likely to be damaged by large deflections elements likely to be damaged by large deflection
Simply One end Both ends Simply One end Both ends
Member supported continuous continuous Cantilever supported continuous continuous Cantilever
Roof slab ℓ/22 ℓ/28 ℓ/35 ℓ/9 ℓ/14 ℓ/18 ℓ/22 ℓ/5.5
Floor slab and roof beam or ribbed
ℓ/18 ℓ/23 ℓ/28 ℓ/7 ℓ/12 ℓ/15 ℓ/19 ℓ/5
roof slab
Floor beam or ribbed floor slab ℓ/14 ℓ/18 ℓ/21 ℓ/5.5 ℓ/10 ℓ/13 ℓ/16 ℓ/4

4.2—Control of deflection (1978), as shown in Table 4.2.1. Other methods for reducing
Beams and one-way slabs are rarely built as isolated deflection are presented in Chapter 6.
members but are part of a monolithic integrated system. 4.2.2 Minimum thickness limitations—Deflections of beams
Excessive deflection of a floor slab may cause dislocations and one-way slabs of normal proportions and supporting
in the partitions it supports or difficulty in leveling furniture typical loads in buildings are normally satisfactory when
or fixtures. Excessive deflection of a beam can damage a the minimum thickness provisions in Table 4.2.2a are met
partition below, and excessive deflection of a spandrel beam or exceeded. This table (ACI 318) applies only to members
or perimeter slab above a window opening could crack the that are not supporting or not attached to partitions or other
glass panels. In the case of roofs or open floors, such as top construction likely to be damaged by excessive deflections.
floors of parking garages, ponding of water can result. Values in Table 4.2.2a were modified by ACI Committee 435
During design, deflection is controlled indirectly through (1978) and are expanded in Table 4.2.2b to include members
reinforcement ratio limitations or minimum thickness that are supporting or attached to nonstructural elements likely
requirements, or is controlled directly through calculating to be damaged by excessive deflections. The thickness limits
a deflection and limiting its value to a maximum based on in Tables 4.2.2a and 4.2.2b may be ignored when computed
functional requirements usually expressed as a fraction of deflections are shown to be satisfactory. The values in Tables
the span length. 4.2.2a and 4.2.2b should be used with care, as they may not
4.2.1 Tension steel reinforcement ratio limitations—One be appropriate for certain situations such as sensitive window
method to minimize deflection of a concrete member in systems, long spans, and high live loads.
flexure is by using a relatively small reinforcement ratio. Due to limitations on the code provisions for minimum
When smaller reinforcement ratios are used, larger (and thickness, several authors have proposed alternative
stiffer) sections are required to meet strength requirements. minimum thickness approaches based on computer simula-
Limiting values of the reinforcement ratio ρ, ranging from tions or engineering beam equations for deflection calcula-
0.25ρb to 0.40ρb were recommended by ACI Committee 435 tion (Grossman 1981, 1987; Rangan 1982; Gardner 2001;
Scanlon and Lee 2006). Grossman (1981, 1987) carried out

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 11

Table 4.2.2c—Maximum permissible computed deflections (ACI 318)

Deflection
Member Condition Deflection to be considered limitation
Flat roofs Not supporting or attached to nonstructural Immediate deflection due to maximum of roof live load, snow and rain ℓ/180*
elements likely to be damaged by large
Floors deflections Immediate deflection due to live load ℓ/360

That part of the total deflection occurring after attachment of nonstructural ℓ/480‡
Roof or Supporting or attached to nonstructural
elements, which is the sum of the time-dependent deflection due to all sustained
floors elements ℓ/240§
loads and the immediate deflection due to any additional live load†
*
Limit not intended to safeguard against ponding. Ponding should be checked by calculations of deflection, including added deflections due to ponded water, and considering long-
term effects of all sustained loads, camber, construction tolerances, and reliability of provisions for drainage.
†
Time-dependent deflection should be determined in accordance with 24.2.4 of ACI 318-14 but should be permitted to be reduced by amount of deflection calculated to occur before
attachment of nonstructural elements. This amount should be calculated on basis of accepted engineering data relating to time-deflection characteristics of members similar to those
being considered.
Limit should be permitted to be exceeded if measures are taken to prevent damage to supported or attached elements.
‡

Limit should not exceed tolerance provided for nonstructural elements.

§

many computer studies to develop a simplified expression ensure that the part of deflection occurring after construction
for the minimum thickness needed to satisfy serviceability of finishes and partitions will be limited to ℓ/500 or 0.78 in.
requirements (Eq. (5.3e) in Chapter 5). Work by Bischoff (20 mm), whichever is less, for spans up to 32.8 ft (10 m).
and Scanlon (2009) indicates that heavily loaded slabs Provision is made in the code for longer spans and for the
(with a service load greater than approximately 200 lb/ft2 amount of tension and compression reinforcement. Service-
[9.6 kPa]) designed using the ACI 318 minimum thickness ability requirements of AS 3600 limit the calculated deflec-
requirements in Table 4.2.2a are not likely to satisfy the tion to a value appropriate to the structure and its intended
corresponding ACI 318 deflection requirements outlined in use, which cannot exceed ℓ/250 for the total deflection and
Table 4.2.2c for that part of the total (incremental) deflection ℓ/500 for the incremental deflection that occurs after the
occurring after attachment of the nonstructural elements; addition or attachment of partitions when provision is made
while beams with the minimum thickness requirements to minimize the effect of movement; otherwise the limit
from Table 4.2.2a are also not likely to satisfy the deflec- is ℓ/1000. BS EN 1992-1-1 limits deflections under quasi-
tion requirements in Table 4.2.2c. Subsequent work by permanent (sustained) loads to ℓ/250 to avoid impairment of
Bischoff and Veysey (2011) provides recommended values the appearance or general utility of a structure. A degree of
of minimum member thickness needed to satisfy the ℓ/240 precamber is permitted. For deflections after construction,
and ℓ/480 deflection limits for one-way slabs and beams. As a limit of ℓ/500 is deemed appropriate for quasi-permanent
the ACI 318 minimum thickness requirements are not always loads and this should avoid damage occurring to adjacent
conservative, it is recommended that deflections be computed parts of the structure.
and compared with the appropriate deflection limits.
Other codes and standards such as BS EN 1992-1-1 give 4.3—Short-term deflection calculation
limiting span/effective depth ratios for flexural members and The short-term or immediate deflection ∆i from flexural
provide limits on computed values of deflection. Gardner curvatures can be computed using a generalized elastic
(2011) provides an extensive comparison of requirements equation for a prismatic member under uniformly distrib-
from numerous code or standard authorities. uted load w.
4.2.3 Computed deflection limitations—The permissible
computed deflections specified in ACI 318 for one-way 5 w 4 5 M 2
=∆i C = K (4.3)
systems are given in Table 4.2.2c, where the span-deflec- 384 Ec I e 48 Ec I e
tion ratios provide for a simple set of permissible deflec-
tions. Where excessive deflection may cause damage to the where M is the moment at the critical section due to service
nonstructural or other structural elements, only that part of loads at the stage deflection is computed. The critical section
the deflection occurring after construction of the affected is usually taken at midspan for simple and continuous
elements (such as partitions) needs to be considered. The members and at the support face for a cantilever. K and C
most stringent deflection limit of ℓ/480 in Table 4.2.2c is are restraint factors that depend on the support fixity and
an example of such a case. Where excessive deflection may loading conditions. For continuous members with uniform
result in a functional problem, such as visual sagging or distributed loading, K = (1.2Mm – 0.2Mo)/Mm, where Mm is
ponding of water, the total deflection should be considered. the moment at midspan and Mo is the total static moment
Project specific requirements, such as for sensitive equip- wℓ2/8. Similarly, C = 1.2Mo/Mm – 0.2. K equals 2.4 and C
ment or curtain wall systems, may require more stringent equals 9.6 for cantilevers. Restraint factor values are given
limits than those listed in Table 4.2.2c. in Table 4.3 for a flexural member subjected to a uniformly
Deflection limits can vary from code to code and represent distributed load w. Effects of cracking and reinforcement
the consensus of experience for typical structures. BS EN on member stiffness are taken into consideration with an
1992-1-1 limits the total deflection to ℓ/250, which should effective moment of inertia (Ie). The modulus of elasticity

American Concrete Institute – Copyrighted © Material – [Link]

12 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Table 4.3—Restraint factors K and C for a uniformly distributed load

Member type (support conditions) K C
Cantilever (fixed end) *
2.4 9.6
Simple span 1.0 1.0
One end continuous with the discontinuous end unrestrained (K = 1.2 – 0.2Mo/Mm) 0.925 for Mm = wℓ /11
2
0.673
One end continuous with the discontinuous end integral with the end support (K = 1.2 – 0.2Mo/Mm) 0.85 for Mm = wℓ /14
2
0.486
Fixed-hinged 0.80
Both ends continuous (K = 1.2 – 0.2Mo/Mm) 0.80 for Mm = wℓ2/16 0.40
Fixed-fixed 0.60 0.20
*
Additional deflection from rotation at support of cantilever should be included.

(Ec) can be obtained from Eq. (3.3.1) for normal-strength where yt is the distance from the centroidal axis to the tension
concrete or Eq. (3.3.4.2a) for high-strength concrete. face of the member, and fr is the modulus of rupture of the
The beam deflection Eq. (4.3) can also be written in terms concrete, as expressed by Eq. (3.3.2a) or (3.3.2b). Use of
of the midspan and member end moments (or midspan and a lower cracking moment to account for shrinkage induced
member end curvatures), which is a convenient form for tension is discussed in [Link] and in 5.2.3 for two-way slabs.
calculating deflection of two way-slabs. For example, ∆i = Cracks develop at several sections along the member
(10Mm – M1 – M2) × ℓ2/(96EcIe), where M1 and M2 are the length. While the cracked moment of inertia Icr applies to
two end moments. Textbooks on reinforced concrete (Nawy the cracked sections, the gross moment of inertia Ig applies
2009; Wight and MacGregor 2009) also provides informa- to the uncracked concrete between these sections. Several
tion on the different methods used to calculate deflection. methods have been developed to estimate the variations in
The approach described previously using Eq. (4.3) to stiffness caused by cracking along the span. These methods
compute deflection is most suitable for hand calculations. Use either modify the flexural rigidity of the member by using an
of Eq. (4.3) implies that deflection of continuous members average or effective moment of inertia (Yu and Winter 1960;
depends on the member stiffness at the critical section, with Branson 1963), make adjustments to the curvature along
no consideration given to the stiffness at the end supports the span and at critical sections (Beeby 1968), compute an
unless a weighted average is taken of stiffness at the midspan average curvature or the ratio M/EI at each location along
and end supports. Other approaches involving more detailed the member span (fib MC2010), or use an incremental evalu-
calculations that may be amenable for implementation into ation of section-curvature (Ghali 1989; Ghali et al. 2011).
computer software include integration of section curvature Deflection can be conservatively estimated by using the
M/EcIe (4.3.3) to account for changes in stiffness along the effective stiffness or curvature at the critical section where
member span. the member stiffness is lowest.
4.3.1 Uncracked members: gross moment of inertia Ig— Studies by Branson (1977, 1985) have shown that the
When the maximum flexural moment at service load in a initial deflections ∆i occurring in a beam or a slab after the
beam or a slab causes a tensile stress less than the modulus of maximum moment Ma has exceeded the cracking moment
rupture (fr), no flexural tension cracks develop at the tension Mcr can be evaluated using an effective moment of inertia Ie
side of the concrete element if the member is not restrained that models a gradual transition from the gross (uncracked)
or the shrinkage and temperature tensile stresses are negli- moment of inertia Ig to the cracked transformed moment of
gible. In such a case, the effective moment of inertia of the inertia Icr.
uncracked transformed section (It) is applicable for deflec- [Link] Branson’s Ie―With the exception of slender walls
tion computations. For design purposes, however, the gross (as defined in 11.8 of ACI 318-14), ACI 318 provides an
moment of inertia Ig, neglecting the reinforcement contribu- expression for the effective moment of inertia Ie originally
tion, can be used with negligible loss of accuracy. The combi- proposed by Branson (1963). This approach was selected
nation of service loads with shrinkage and temperature effects as being sufficiently accurate to control deflections in rein-
due to end restraint or restraint from the internal reinforce- forced concrete unless stiffness values are obtained by a
ment may cause cracking if the tensile stress in the concrete more comprehensive analysis. Branson’s approach is known
exceeds the modulus of rupture. In such cases, 4.3.2 applies. to work well for beams that typically have a reinforcement
4.3.2 Cracked members: effective moment of inertia Ie― ratio greater than 1 percent, but significantly overestimates
Tension cracks occur when the imposed loads cause bending Ie for lightly reinforced slabs where the in-service moment is
moments in excess of the cracking moment, thus resulting often only just greater than the cracking moment (Bischoff
in tensile stresses in the concrete that are higher than its and Scanlon 2007; Gilbert 2007).
modulus of rupture. The cracking moment Mcr may be Branson’s equation for the effective moment of inertia Ie
computed as follows for short-term deflections is as follows

fr I g  M 
3 3

M cr = (4.3.2) M 
yt =I e  cr  I g + 1 −  cr   I cr ≤ I g (4.3.2.1a)
 Ma    M a  
American Concrete Institute – Copyrighted © Material – [Link]
 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 13

Fig. [Link]—Moments of inertia of uncracked and cracked transformed sections.

or in a slightly rearranged form as the level of cracking, expressed as Mcr/Ma. The use of Ie as
a weighted average of Ig and Icr is assumed to give deflec-
M 
3 tion values close to those obtained using a bilinear load-
I e =I cr +  cr  ( I g − I cr ) ≤ I g (4.3.2.1b) deflection response after cracking. Details for computing the
M  a cracked moment of inertia Icr are provided in Fig. [Link].
Deflections have traditionally been computed for each
where Mcr is the cracking moment; Ma is the maximum load level using either form of Eq. (4.3.2.1a) or (4.3.2.1b),
(unfactored) service load moment at the stage for which such as dead load and dead load plus live load. Thus, the
deflections are being considered; Ig is the gross moment of incremental deflection such as that due to live load alone is
inertia of section; and Icr is the moment of inertia of cracked computed as the difference between these values at the two
transformed section. load levels. Another school of thought, however, assumes the
The value Ie provides a transition between the upper and structural member has been preloaded under the full (dead
the lower bounds of Ig and Icr, respectively, as a function of plus live) service load such that subsequent calculations of

American Concrete Institute – Copyrighted © Material – [Link]

14 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

both the dead and live load deflection are computed using flexural members when Ig /Icr < 3 (corresponding to steel
the same value of Ie corresponding to full dead plus live load reinforced concrete members with a reinforcing ratio greater
(Grossman 1981). The latter approach is generally assumed than approximately 1 percent). Member stiffness using
to be more realistic for slab construction and provides a Branson’s approach is overestimated when Ig /Icr > 3. The
conservative estimate of long-term deflection. The former ratio Ig/Icr for lightly reinforced concrete slabs (0.2 percent
approach, on the other hand, provides a conservative esti- < ρ < 0.6 percent) can be as high as 12, varies between 7 to
mate of immediate deflection from live load, although this 25 for FRP-reinforced concrete beams and slabs, and ranges
requirement typically does not control design. from 15 to 25 for walls with a central layer of reinforcement
The value of Ie may be determined using Ma at the support (Bischoff and Scanlon 2007). Hence, Branson’s equation
for cantilevers, and at the midspan for simple spans. Equa- (Eq. (4.3.2.1a) or (4.3.2.1b)) can significantly underestimate
tion (4.3.2.1a) shows that Ie is an interpolation between the deflection in these three cases.
well-defined limits of Ig and Icr. This equation was recom- Both the member (cubic power) and section (fourth power)
mended by ACI Committee 435 (1966) and has been used in based expressions for Ie were calibrated from tests (Branson
ACI 318 since 1971 and the AASHTO Bridge Design Speci- 1963) of uniformly loaded beams where the value of Ie was
fications since 1973. Detailed numerical examples using this within 12 percent of Icr for nearly 85 percent of the speci-
method for simple and continuous beams and unshored and mens assessed. The assumed level of service load ranged
shored composite beams are available in Branson (1977). from two to four times the cracking moment. There was
Textbooks on reinforced concrete such as those by Wang and practically no difference (3 to 6 percent at most) between the
Salmon (1992) and Nawy (2009) also provide an extensive member Ie and section based Ie for nearly all the 56 speci-
treatment of the subject. mens considered. Hence, it is difficult to draw any definite
Heavily reinforced members will have an Ie approximately conclusions about the applicability of using either equation
equal to Icr, which may in some cases (flanged members) be for other types of members and loading conditions.
larger than Ig of the concrete section alone. For most prac- [Link] Bischoff’s Ie―Bischoff (2005, 2007) applied basic
tical cases, the calculated Ie will be less than Ig. In both cases, concepts of tension stiffening to flexural members to develop
Ie should be taken as the lesser of the calculated value or Ig a rational expression for Ie given by
when designing for deflection control, unless a justification
can be made for rigorous transformed section computations.
I cr I cr
Equation (4.3.2.1a) or (4.3.2.1b) is intended to= represent Ie = 2
≤ Ig
an averaged value of Ie that accounts for the variation in stiff-  M cr   I cr   M cr   I cr 
1− β 1 − I  1 −  1 − I 
ness along the member. Replacing the cubic power in Eq.  M a   g   M a   g 
(4.3.2.1a) with a fourth power is meant to give the effective
moment of inertia at a section and is useful when integrating (4.3.2.2a)
curvatures at multiple sections along the span length.
with the tension stiffening factor β = Mcr/Ma that varies
4
 M   4 between 1 (at cracking) and 0 when Ma >> Mcr. Setting β
M 
= I e  cr  I g + 1 −  cr   I cr ≤ I g (4.3.2.1c) equal to 1 (instead of Mcr/Ma) gives a bilinear response with
 Ma    M a   constant tension stiffening that represents an upper bound
on member stiffness as shown in Fig. 4.3.2.2a, whereas β
Branson’s equation (Eq. (4.3.2.1a) or (4.3.2.1b)) is an set equal to 0 (instead of Mcr/Ma) provides a lower bound
empirically derived formula that was calibrated to work on member stiffness with no tension stiffening to give the
well for typical steel-reinforced concrete beams having rein- Icr response. Bischoff’s expression for Ie gives a reasonable
forcing ratios greater than approximately 1 percent (Bischoff estimate of deflection provided the correct cracking moment
2005). However, this equation underestimates deflection for is used (Scanlon and Bischoff 2008).
most members reinforced with FRP bars (ACI 440.1R), for Equation (4.3.2.2a) can be rearranged into Eq. (4.3.2.2b)
certain types of slender tilt-up wall panels (Bischoff and that incorporates similar ratios to Eq. (4.3.2.1a). However,
Scanlon 2007), and for steel-reinforced concrete beams comparing the arrangement of terms in the two equations
and slabs with reinforcing ratios below 1 percent (Bischoff highlights the fundamental difference between the two
2005; Gilbert 2006). ACI 440.1R-06 recommends applying approaches. Equation (4.3.2.2b) uses a weighted average of
a correction factor to Eq. (4.3.2.1a) or (4.3.2.1b) for flexibilities whereas Eq. (4.3.2.1a) uses a weighted average
computing deflection of FRP-reinforced concrete members, of stiffnesses. Because deflection is proportional to flex-
while the problem with lightly reinforced slabs has not yet ibility, Bischoff’s approach (Eq. (4.3.2.2a) or (4.3.2.2b)) is
been addressed by ACI 318. However, ACI 318 recognizes applicable over a wider range of situations than Branson’s
the problem with Branson’s equation for tilt-up wall panels, approach (Eq. (4.3.2.1a)). The other primary difference is
and now requires use of a bilinear equation taken from the that the cubic power in Branson’s equation (Eq. (4.3.2.1a))
Uniform Building Code (ICBO 1997). has been replaced by a squared power in Bischoff’s equation
The accuracy of Branson’s equation for Ie is dependent on (Eq. (4.3.2.2a and b)) (refer to Bischoff [2005] for motiva-
the ratio of Ig to Icr (in addition to the level of loading with tions for this).
respect to the cracking moment), and only works well for

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 15

stiffness of the member. Decreasing the cracking moment

has a significant effect on computed values of deflection
for lightly reinforced members such as slabs. Deflection of
moderate to more heavily reinforced members is not signifi-
cantly affected at full-service load as the effective moment
of inertia Ie is closely approximated by Icr at the full service
level of load for these members (where Ma/Mcr > 3).
CSA A23.3 uses a reduced cracking moment for computing
deflection (based on the work of Scanlon and Murray
[1982]) whereas AS 3600 reduces the cracking moment with
a computed value of the shrinkage restraint stress (Gilbert
1999). BS EN 1992-1-1 provisions for computing deflection
can also account for a reduced cracking moment. The use
of a lower cracking moment also accounts for cases where
Fig. 4.3.2.2a—Moment-curvature response for different the service load moment is only slightly less than the unre-
tension stiffening values with Eq. (4.3.2.2a). strained cracking moment as factors such as shrinkage and
temperature can still cause the section to crack over time.
To account for this reduction in cracking load, Gilbert
(1999) recommends reducing the modulus of rupture fr by an
amount equal to the restraint stress fres = 2.5ρEsεsh/(1 + 50ρ),
where εsh is the shrinkage strain in the concrete. AS 3600
uses a modified version of Gilbert’s (1999) expression that
includes the effect of compression steel. For design, Scanlon
and Bischoff (2008) recommend using a reduced cracking
moment equal to two-thirds the value of Mcr (computed
using Eq. (4.3.2) with Eq. (3.3.2a)) when computing deflec-
tions for steel-reinforced concrete using Eq. (4.3.2.2a) for Ie.
ACI 318 already uses (2/3)Mcr for computing out-of-plane
deflection of slender walls. For FRP-reinforced concrete,
the cracking moment should only be reduced by 20 percent
to reflect a lower degree of internal restraint when using
the more flexible FRP reinforcement (Bischoff and Gross
Fig. 4.3.2.2b—Comparison of Ie using Branson and
2011b). Further reductions in the cracking moment may be
Bischoff’s approach with Mcr.
required to account for axial tension induced by restraint to
shrinkage from the supports and adjacent parts of the struc-
ture. For shrinkage stresses approaching fr, or other causes
1  M  1   M cr   1
2 2
1 of large net axial forces, it may be more appropriate and
=  cr  + 1 −   ≥ (4.3.2.2b) numerically stable to modify Ma instead of Mcr.
I e  M a  I g   M a   I cr I g
  Preloading from construction loads prior to installation
of the nonstructural elements can cause additional cracking
Bischoff’s Ie (defined by Eq. (4.3.2.2a) or (4.3.2.2b)) that reduces the member stiffness and increases deflection
is simple and works equally well for concrete beams and arising from the sustained portion of loading. Scanlon and
slabs reinforced with either steel or FRP bars over a wide Bischoff (2008) recommend computing the immediate
range of reinforcing ratios (Bischoff and Scanlon 2007). deflection under sustained loads with an effective moment of
Bischoff’s equation (Eq. (4.3.2.2a)) and Branson’s equation inertia corresponding to the full (dead plus live) service load
(Eq. (4.3.2.1a)) give similar computed values of deflection moment to account for the reduced stiffness. This simplifies
for beams reinforced with steel at reinforcing ratios greater the calculation procedure considerably as only one value of
than 1 percent (where tension stiffening is not significant) as Ie is now considered in deflection calculations.
shown in Fig. 4.3.2.2b. Differences only become significant [Link] Variation in stiffness along member length―Equa-
for flexural members with an Ig/Icr greater than 3 (typically tion (4.3.2.2a) or (4.3.2.2b) effectively provides a section-
for flexural members with ρ < 1 percent for steel-reinforced based expression for the effective moment of inertia. For
concrete). For the reasons given previously, Bischoff’s members with uniform cross section and typical reinforce-
equation (Eq. (4.3.2.2a)) is recommended for computing ment patterns, using either of these equations based on
deflection. the service load moment at the critical section (where the
[Link] Shrinkage restraint and preloading—Restraint to moment is greatest) will provide a conservative estimate of
shrinkage from either the internal reinforcement or adjoining deflection as the member stiffness is lowest at this location.
members and supports induces tensile stresses in the concrete If warranted, a more accurate calculation of deflection can
that decrease the cracking moment and reduce the flexural be carried out by integrating curvatures along the length of

American Concrete Institute – Copyrighted © Material – [Link]

16 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

the member based on the calculated Ie value at each section For beams continuous on one end only
(Bischoff and Gross 2011a). Using this approach gives a
modified expression for Eq. (4.3.2.2a) that includes an addi- Ie = 0.85Ie(m) + 0.15Ie(1) (4.3.2.5c)
tional factor γ to account for the variation in stiffness along
the length of member. ACI 440.1R has adopted this approach Work by Bischoff (2007), however, indicates that deflec-
for design of FRP-reinforced concrete. tion of continuous members can be underestimated when
using a weighted average of Ie, for cases where the member
I cr stiffness is greater at the end supports than at midspan. Once
=Ie 2
≤ Ig ([Link]) again, deflection would be reflected more correctly by taking
M   I 
1 − γ  cr  1 − cr  a weighted average of the inverse stiffness (flexibility) values
 Ma   Ig  (1/Ec Ie). Reasonable values of deflection are obtained by using
the midspan value of the effective moment of inertia provided
The integration factor γ ≥ 1 accounts for the length of the that: 1) the end support moments do not exceed twice the
uncracked regions of the member in addition to the change midspan value; and 2) the value of Ie at the end supports is
in stiffness along the cracked regions. The value of γ depends between one-half and three times the value at midspan.
on the type of loading, boundary conditions, and reinforce- When Ie is calculated as indicated in the previous discus-
ment pattern. Earlier work by Al-Zaid et al. (1991) has also sion, the deflection can be obtained using a generalized
recognized that the Ie value can be affected by the type of form of the elastic equation such as Eq. (4.3), or any other
loading (concentrated or distributed) on the member. Expres- approach used to compute elastic deflection such as the
sions for the factor γ in Table [Link], provided for different moment-area method or conjugate beam method. Numerical
types of beams and loading cases, can be used in place of integration procedures can also be used. The moment-curva-
numerical integration of curvature. Closed-form solutions ture (rotation) response of the member is taken into consid-
for deflection of continuous members with moments at the eration with each approach.
end supports are not readily available and would typically 4.3.3 Section-based moment-curvature approach—The
require numerical integration unless simplifying assump- calculation methods for member deflection described in
tions are made. 4.3.1 and 4.3.2 rely on the use of an average member stiff-
The increase in member stiffness with Eq. ([Link]) ness between the supports. For uncracked members (4.3.1),
depends on the type of load, boundary conditions, reinforcing the gross moment of inertia, Ig, is used, while an average
pattern, reinforcing material, reinforcement ratio, and level effective moment of inertia, Ie, can be used for cracked
of service load relative to the cracking moment. Setting γ members (4.3.2). In the latter case, the parameter Ie can
= 1 results in Eq. (4.3.2.2a), which provides a reasonably account for variations in the amount of cracking along the
conservative estimate of deflection in many cases, particu- member length as well as the tension stiffening of concrete
larly for steel reinforced concrete members subjected to a between the cracks by using integration factors or other
uniformly distributed load (Bischoff and Gross 2011a). simplifications to represent the variation in bending moment.
[Link] Continuous members―For continuous beams and In general, the mathematical complexity in these techniques
one-way slabs, ACI 318 stipulates that Ie may be taken as is intentionally low, using linear-elastic material properties
the average of values obtained for the critical positive and to facilitate hand calculations or simple computer-based
negative moment sections. For prismatic members, Ie may techniques, such as spreadsheets. However, with the avail-
be taken as the value obtained at midspan for continuous ability of personal computers in engineering design offices,
spans. The use of midspan section properties for continuous more accurate analytical procedures can be routinely consid-
prismatic members is considered satisfactory in approxi- ered for the calculation of deflections under complex loading
mate calculations primarily because the midspan rigidity, scenarios, time dependent effects ([Link]), or to directly
including the effect of cracking, has a dominant effect on consider the nonlinear material behavior.
deflections (ACI Committee 435 1978). A numerical section-based approach can directly consider
ACI 318 suggests a simple averaging of Ie values for the the compressive stress-strain response characteristics of
positive and negative moment section if the designer chooses different concrete types within a cross section as well as
to average the effective moment of inertia Ie. the calculation of deflections before and after yielding of
the reinforcement at different locations along the member.
Ie = 0.50Ie(m) + 0.25(Ie(1) + Ie(2)) (4.3.2.5a) Sectional-based analytical methods can also be used to esti-
mate the influence of nonprismatic geometry or variable rein-
where the subscripts m and 1 and 2 refer to midspan and the forcement detailing on the overall deflection of members.
two beam ends, respectively. Numerical modeling is performed by first determining the
Improved results for continuous prismatic members can, nonlinear moment-curvature response of representative cross
however, be obtained using a weighted average, as presented sections along the member span that satisfy compatibility of
in the following equations. axial strains and equilibrium of section forces. The member
For beams continuous on both ends deflection is then computed by numerically integrating the
curvatures (Park and Paulay 1975; Collins and Mitchell
Ie = 0.70Ie(m) + 0.15(Ie(1) + Ie(2)) (4.3.2.5b) 1997; Ghali et al. 2011). Such calculations can be performed

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 17

Table [Link]—Integration factor to account for changes in stiffness along the member (Bischoff and
Gross 2011a)
Beam and loading type Moment and elastic deflection equations Ie = Icr/[1 – γη(Mcr/Ma)2] with η = 1 – Icr/Ig

P
Ma = PL/4
γ = 3 – 2(Mcr/Ma)
∆ = PL3/48EcIe
L/2 L/2

P/2 P/2
Ma = PL/6
γ = 1.7 – 0.7(Mcr/Ma)
∆ = 23PL3/1296EcIe
L/3 L/3 L/3

P/2 P/2 Ma = Pa/2

3(a /L) − 4ζ(a /L)3
∆ = α′PL3/48EcIe γ=
3(a /L) − 4(a /L)3
a a
α′ = 3(a/L) – 4(a/L)3 ζ = 4(Mcr/Ma) – 3
L

w
Ma = wL2/8
1.6ζ3 − 0.6ζ 4
=
γ + 2.4 ln(2 −=
ζ) 1.72 − 0.72( M cr /M a )
( M cr /M a ) 2
∆ = 5wL4/384EcIe
L
ζ = 1 − 1 − M cr /M a

P
Ma = PL
γ = 3 – 2(Mcr/Ma)
∆ = PL3/3EcIe
L

w
Ma = wL2/2
γ = 1 – 2ln(Mcr/Ma)
∆ = wL4/8EcIe
L

using a piece-wise numerical integration approach described strain, εtf. For a reinforced member without prestressing, the
herein, or a finite-element-based approach. strain due to stress in the reinforcement, εsf, is identical to εtf.
One simple method to obtain the variation in curvature With the strain distribution defined, the stress-strain response
along the member axis is to solve the relationship for curva- of the concrete and reinforcement is used to establish the
ture κ = Ma /Ec Ie at selected cross sections, where Ma is the corresponding stresses fc and fs. To approximate tension stiff-
moment at the cross section and Ie is the effective moment of ening of the concrete, an average tensile stress to average
inertia from Eq. (4.3.2.1c) or (4.3.2.2a), depending on use of tensile strain relationship (Vecchio and Collins 1986) can be
Branson’s or Bischoff’s estimate of Ie, respectively. used to estimate the tensile stress after cracking. Finally, for
Alternatively, direct consideration of the nonlinear this overall stress condition over the cross section, the equi-
mechanical properties of the materials in combination with librium conditions by Eq. (4.3.3a) and (4.3.3b) are checked.
the cross section configuration can be made. For a given
loading condition at each selected cross section, the axial N = ∫ f c ⋅ dA + ∫ f s ⋅ dAs (4.3.3a)
strains are established using the well-known hypothesis that A As

plane sections before bending remain plane after bending.

Thus, a linear variation of strains occurs over the height of
the cross section, with subscripts including t to denote total M =− ∫ f c ⋅ y ⋅ dA − ∫ f s ⋅ y ⋅ dAs (4.3.3b)
A As
(Fig. 4.3.3a). For short-term loading with no time-dependent
effects, the strain due to stress in the concrete, denoted as εcf, where N and M are the internal axial force and internal
is determined at each elevation in the cross section. Local bending moment on the cross section due to external
bending of the reinforcing bars is ignored and perfect bond loading, and y is the vertical distance measured up from the
is assumed between the concrete and the bars. Thus, the total centroidal axis. The corresponding curvature κ is the slope
strain in each reinforcing bar is identical to the total axial of the total strain diagram.

American Concrete Institute – Copyrighted © Material – [Link]

18 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Fig. 4.3.3a—Sectional response.

Fig. 4.3.3b—Layer-by-layer of sectional forces.

To simplify the calculations in Eq. (4.3.3a) and (4.3.3b), including virtual work, conjugate beam methods, or the use
the calculations can be performed numerically using a of moment-area theorems.
layer-by-layer approach. As shown in Fig. 4.3.3b, the cross Appendix B of ACI 435R-95(03) provides a detailed
section is discretized into n layers of either constant or approach for computing deflection of reinforced concrete
unequal height. From the linear distribution of total strains, members, with or without prestressing, having a variable
the representative strain due to stress and the stress in each cross -section. Section analysis is carried out to determine
concrete layer can be computed based on the average strain curvature of an uncracked and cracked section (ignoring
in the layer or through further refinement based on the shape concrete in tension). Curvature of the uncracked and cracked
of concrete stress-strain response. Corresponding layer-by- section is then interpolated using fib MC2010 to obtain an
layer forces in the concrete are then solved by considering interpolated value of average curvature at each cross -section
the concrete stresses and the cross section geometry of the that takes tension stiffening of the concrete into account. The
layer. The reinforcement strains due to stress and forces average curvature obtained with this approach at sections
are computed from the overall strain distribution using a along the member span is used to obtain deflection.
similar procedure, noting that the concrete and reinforce-
ment strains are identical at the reinforcement level due to 4.4—Long-term deflection calculation
the perfect bond assumption. Typical stress-strain relation- 4.4.1 Traditional ACI method—Unless values are obtained
ships for concrete and reinforcement are given in Wight by a more comprehensive analysis, the additional long-term
and MacGregor (2009). In place of a nonlinear stress-strain deflection due to the combined effects of creep and shrinkage
model for concrete in tension, Collins and Mitchell (1997) can be calculated for nonprestressed one-way flexural
suggest that tension stiffening can be approximated as a members by applying a multiplier λ∆ to the computed short-
uniform tensile stress of 0.5fcr acting over the part of the term (immediate) deflections caused by the sustained load.
concrete within 7.5db of each reinforcing bar. Setting the Hence, the long-term deflection from the sustained loads is
concrete stress to zero for concrete in tension surrounding obtained by
the reinforcement gives the section response at a crack but
is not useful for computing deflection as this does not take ∆LT = λ∆∆sus (4.4.1a)
tension stiffening into account. Collins and Mitchell (1997)
also note that the layer-by-layer approach for concrete in where ∆sus is the immediate deflection due to sustained loads
compression can be modeled using an equivalent stress- (deflection from dead load ∆D plus any deflection from the
block approach if the concrete stress-strain response has a sustained portion of the live load ∆L,sus), and λ∆ is the time-
parabolic shape. dependent multiplier for a defined duration of time t.
The moment-curvature relationships for different cross
sections along the member can be used to establish the ξ
λ∆ = (4.4.1b)
approximate curvature distribution (Fig. 4.3.3c). The corre- 1 + 50ρ′
sponding slopes and deflections can be obtained based on
numerical integration using a number of different approaches

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 19

Fig. 4.4.1—ACI 318 multipliers for long-term deflections.

This behavior is mainly due to lower creep strain charac-
teristics. However, the influence of compression steel rein-
forcement is less pronounced in high-strength concrete
members. This is because the substantial force transfer from
the compression concrete to compression reinforcement is
greatly reduced for high-strength concrete members, for
which creep is lower than normal-strength concrete. Nilson
(1985) suggested that two modifying factors should be intro-
duced into Eq. (4.4.1b). The first is a material modifier μm
with values equal to or less than 1.0, applied to ξ to account
for the lower creep coefficient. The second is a section
Fig. 4.3.3c—Typical curvature distribution.
modifier μs also having values equal to or less than 1.0, to
where ρ′ is the reinforcement ratio for nonprestressed be applied to ρ′ to account for the decreasing importance
compression steel reinforcement; and ξ is the time-depen- of compression steel in high-strength concrete members.
dent factor from Fig. 4.4.1 (ACI 318). Nilson (1985) introduced a single modifier μ to account for
It is permitted to assume that ξ is equal to: both effects simultaneously, leading to the following simpli-
5 years or more: 2.0 fied equation
1 year: 1.4
6 months: 1.2 µξ
3 months: 1.0 λ∆ = (4.4.1d)
1 + 50µρ′
Caution may need to be exercised with this method,
particularly for lightly reinforced slabs, as the ratio of long-
term deflection to short-term sustained load deflection can where 0.7 ≤ μ = 1.3 – 0.00005f′c ≤ 1.0 with fc′ in psi (0.7 ≤ μ
vary significantly from the ACI 318 λ∆ values (Walkup et = 1.3 – 0.0073fc′ ≤ 1.0 with fc′ in MPa).
al. 2019). This equation results in μ = 1.0 for concrete strengths
The incremental deflection ∆inc that occurs after attach- less than 6000 psi (42 MPa) and provides a reasonable fit
ment of the nonstructural elements includes the long-term of experimental data for higher concrete strengths. More
deflection ∆LT from the sustained loads plus the immediate data is needed, however, particularly for strengths between
deflection from the remaining part of live load (∆L,add = 9000 and 12,000 psi (62 and 83 MPa) and beyond before a
∆L – ∆L,sus) that is not sustained. The portion of long-term definitive statement can be made. Gardner and Zhang (1996)
deflection occurring before attachment of the nonstructural have also proposed alternative long-term deflection multi-
elements may be deducted. pliers based on the relative load level, tensile steel ratio, and
compression steel ratio.
∆inc = λ∆∆sus + ∆L,add (4.4.1c) 4.4.2 ACI Committee 435 modified method (Branson 1963,
1977)—For computing creep and shrinkage deflections sepa-
Note that ACI 440.1R recommends using 60 percent of rately, Branson’s (1963, 1977) Eq. (4.4.2a) and (4.4.2b) are
the time-dependent factor to give λ∆ = 0.6ξ for FRP rein- recommended by ACI Committee 435 (1966, 1978).
forcement (compression reinforcement is not considered for
FRP-reinforced concrete members). ∆cr = λt∆sus (4.4.2a)
Research has shown that high-strength concrete members
exhibit less sustained-load deflections than low-strength where λt = 0.85Ct /(1 + 50ρ′).
concrete members (Luebkeman et al. 1985; Nilson 1985).

American Concrete Institute – Copyrighted © Material – [Link]

20 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

 (ε ) 
∆ sh = k sh κ sh  2 = k sh  Ash sh t   2 (4.4.2b) is the stress in the tension reinforcement calculated on the
 h  basis of a cracked section; and σsr is the stress in the tension
reinforcement calculated on the basis of a cracked section
Ct and (εsh)t may be determined from Eq. (3.3.3a) through under the loading conditions causing first cracking.
(3.3.3c) and Table 3.3.3. Note that σsr /σs may be replaced by Mcr/M for flexure or
Ncr/N for pure tension, where Mcr is the cracking moment
1/ 2 and Ncr is the cracking force. Research (Scott and Beeby
 ρ − ρ′ 
A= 0.7(ρ − ρ′ )  for ρ – ρ′ ≤ 3.0 percent 2005) recommends the modification of using β = 0.5 when
 ρ 
sh
calculating short-term deflections because loss of tension
= 0.7ρ1/3 for ρ′ = 0 stiffening occurs much more rapidly than was previously
= 1.0 for ρ – ρ′ > 3.0 percent thought to be the case.
Deformations due to loading may be assessed using the
The reinforcement ratios ρ and ρ′ are computed at the tensile strength and the effective modulus of elasticity of the
support section for cantilevers and at the midspan sections concrete. In general, it is recommended that using the mean
for simple and continuous spans. value of the axial tensile strength of the concrete is likely to
The shrinkage deflection constant ksh is as follows: give the best estimate of the behavior.
Cantilevers = 0.50 The effects of creep may be included by using an effective
Simple beams = 0.13 modulus of elasticity for the concrete given by Eq. (4.4.3.1c)
Spans with one end continuous (multi spans) = 0.09
Spans with one end continuous (two spans) = 0.08 Ec,eff = Ecm/(1 + Cu) (4.4.3.1c)
Spans with both ends continuous = 0.07
Separate computations of creep and shrinkage are prefer- where Cu is the final value of the creep coefficient. This is
able when part of the live load is considered as a sustained dependent on the age of the concrete at the time of loading
load. and the relative humidity of the exposure conditions.
4.4.3 Other methods—Other methods for time-dependent Shrinkage curvatures are assessed using Eq. (4.4.3.1d)
deflection calculation in reinforced concrete beams and
one-way slabs are available in the literature. They include κsh = εcsne(S/I) (4.4.3.1d)
several methods listed in ACI Committee 435 (1966),
Appendix B of ACI 435R-95(03) described by Ghali et al. where κsh is the curvature due to shrinkage; εcs is the free
(2011), fib MC2010, and BS EN 1992-1-1. Other methods shrinkage of concrete; S is the first moment of area of the
described in 5.2.4 are also suitable for calculation of time- reinforcement about the centroid of the section (section
dependent deflection. This section highlights BS EN 1992- modulus); I is the second moment of area of the section
1-1 (Eurocode 2), and briefly describes the section curvature (moment of inertia); and ne is the effective modular ratio
approach, referring the reader to the literature for details. equal to Es /Ec,eff (which allows for the effects of creep). S and
[Link] BS EN 1992-1-1 (Eurocode 2) method—BS EN I should be calculated for the uncracked and cracked condi-
1992-1-1 uses the concept of the partially cracked section. tions and the final curvature assessed using Eq. (4.4.3.1a).
According to BS EN 1992-1-1, members that are expected to BS EN 1992-1-1 observes that the most rigorous method
crack, but may not be fully cracked, will behave in a manner for assessing deflections is to compute curvatures at frequent
intermediate between the uncracked and fully cracked condi- sections along the member and then calculate the deflec-
tions. For members subjected mainly to flexure, an adequate tion by numerical integration. However, it will frequently
prediction of behavior is given by Eq. (4.4.3.1a). be acceptable to compute the deflection twice, assuming the
whole member to be uncracked and then fully cracked, and
α = ζαII + (1 – ζ)αI (4.4.3.1a) then interpolate using Eq. (4.4.3.1a).
[Link] Time-dependent change in strain and curvature at
where α is the deformation parameter considered. This a section—As described in 4.3.3, the section-based moment-
parameter may be, for example, a strain, a curvature, or curvature method is a general approach to determine the
a rotation but, as a simplification, α may also be taken member deflection by considering the calculated response
as a deflection; αI and αII are the values of the parameter at representative cross sections. The cross section response
calculated for the uncracked and fully cracked conditions, directly considers the stress-strain characteristics of the
respectively; and ζ is a distribution coefficient (allowing for materials and can account for differences between the total
tensioning stiffening at a section) given by Eq. (4.4.3.1b) strain εt and the strain due to stress of the concrete and rein-
forcement, εcf and εsf, respectively. Therefore, this approach
ζ = 1 – β(σsr/σs)2 (4.4.3.1b) lends itself to consider the time-dependent influences on
deflection including shrinkage, creep, and modeling the
(ζ = 0 for uncracked sections) deflection of members at different construction stages.
where β is a coefficient taking account of the influence of the The time-dependent action of creep in concrete results
duration of the loading or of repeated loading on the average in gradual changes of deflection under sustained loading.
strain (β = 1.0 for a single short-term loading and β = 0.5 Equation (3.3.3d) presents a relationship to determine the
for sustained loads or many cycles of repeated loading); σs
American Concrete Institute – Copyrighted © Material – [Link]
 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 21

age-adjusted modulus of concrete in compression by means

of a creep coefficient. The resulting modulus can be used to
determine an effective stress-strain response of the concrete
in compression for direct use within the section moment-
curvature calculation methods from Section 4.3. Note that
while there may be small sustained tensile loads in the
concrete before cracking, or between cracks after cracking,
any tension-induced creep will have a negligible influ-
ence on the bending response and can typically be ignored.
The deflection at any given period should account for the
sequence of loading and the amount of creep that results Fig. [Link]—Consideration of shrinkage strains.
over each time interval of a given loading scenario.
Other time-dependent effects of concrete can result in joints. Before performing an analysis for temperature effects,
deflections but do not rely on sustained loading. As one it is necessary to select design temperature gradients. Martin
example, uniform shrinkage over the cross section of an (1971) summarizes design temperatures that are provided in
unrestrained (statically determinate) member will result various national and foreign codes.
in a shortening of the member without external forces. ACI 435.7R outlined procedures for estimating changes in
However, even though this is externally free shrinkage, only stiffness and temperature-induced deflections for reinforced
the concrete shrinks while the reinforcement does not. To concrete members. The following expressions are taken
maintain strain compatibility between the materials, internal from that report.
stresses will result, and these should be accounted for in 4.5.1 Temperature gradient on statically determinate
the force equilibrium at the cross section. This can result in (unrestrained) member―With temperature distribution T(y)
changes to the cracking moment Mcr and can also induce on the cross section, thermal strain at a distance y from the
deflections that will depend in part on the reinforcement bottom of the section can be expressed by
configuration. One approach to calculate these effects is to
separate the total strain at each point in the cross section (εt) εth(y) = αT(y) (4.5.1a)
into a strain due to stress εcf and the other free strains εo. where α is the coefficient of thermal expansion.
This is illustrated conceptually in Fig. [Link] where the total To restrain the movement due to temperature T(y), a stress
strains are used to determine the member curvature κ. The is applied in the opposite direction to εth(y)
part of Fig. [Link] representing the strains due to stress are
used within the force equilibrium Eq. (4.3.3a) and (4.3.3b). f(y) = EcαT(y) (4.5.1b)
4.4.4 Finite element method―Many finite element models
have been developed to account for time-dependent deflec- The net restraining axial force and moment are obtained
tions of reinforced concrete members. The sophistication by integrating over the cross-section depth
and complexity of these methods varies widely. Some
approaches are similar to section curvature methods (Hirsch h
2009) whereas other approaches attempt to model the full P= ∫ fdA= ∫ [ αEcT ( y )b( y ) ] dy (4.5.1c)
nonlinear time-dependent complexity of reinforced concrete A 0

as a heterogeneous three-dimensional continuum. The more h

complex approaches are only justifiable when a high degree ∫ [ αEcT ( y )b( y )( y − yt ) ] dy
∫ f ( y − yt )dA =
M = (4.5.1d)
A 0
of precision is required for special structures and only when
substantially accurate creep and shrinkage data are avail-
able. In special cases, such information on material proper- where yt represents the distance to the centroidal axis
ties is warranted and may be obtained experimentally from measured from the bottom of the section. To obtain the
tests of actual materials to be used. Even with tests of the total strains on the unrestrained cross section, P and M are
actual materials to be used, assumptions regarding tension applied in the opposite direction to the restraining force and
stiffening and other behavior still need to be made to get moment. Assuming plane sections remain plane, axial strain
reasonable predictions of deflection. εa and curvature κ are given by

4.5—Temperature-induced deflections P αh
Variations in ambient temperature can significantly affect ε=
a = ∫ [T ( y )b( y ) ] dy (4.5.1e)
AEc A o
deformations of reinforced concrete structures. Deflections
occur in flexural members when a temperature gradient M αh
occurs between its opposite faces. It has become standard
=
κ = ∫ [T ( y )b( y )( y − yt ) ] dy (4.5.1f)
Ec I I 0
practice to evaluate thermal stresses and displacements in
tall building structures. Movements of bridge superstruc-
tures and precast concrete members are also computed for The net stress distribution on the cross section is given by
the purpose of design of support bearings and expansion

American Concrete Institute – Copyrighted © Material – [Link]

22 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

P M ( y − yt )
fn ( y) = ± − Ec αT ( y ) (4.5.1g)
A I

and assumes the member remains uncracked.

For a linear temperature gradient varying from 0 to ∆T, the
curvature is given by
α∆T
κ= (4.5.1h)
h
Fig. 4.5.3—Consideration of thermal gradients.
In the case of a uniform vertical temperature gradient stresses in the concrete and the reinforcing steel are solved
constant along the length of a member, deflections for simply based on the εcf values. To ensure force equilibrium between
supported (∆ss) and cantilever (∆cant) beams are calculated as the external loading causing M and the internal forces using
Eq. (4.3.3a) and (4.3.3b), the total strain diagram should be
κ 2 α∆T  2 solved iteratively by varying the curvature κ and the strain at
∆ ss= = (4.5.1i) one elevation, normally the extreme fiber strain on the flex-
8 h 8
ural compression side. Refer to Collins and Mitchell (1997)
κ 2 α∆T  2 for techniques appropriate for spreadsheet implementation
∆ cant = = (4.5.1j) and Ghali et al. (2011) for techniques suitable for finite
2 h 2
element software.
The deflection-to-span ratio is given by
CHAPTER 5—DEFLECTION OF A TWO-WAY
∆ α∆T 
= (4.5.1k) SLAB SYSTEM
 h k
where k = 8 for simply supported beams and 2 for cantilever 5.1—Introduction
beams. Two-way slab systems are widely used for roof and floor
4.5.2 Effect of restraint on thermal movement—If a construction in reinforced concrete buildings. Because
member is restrained from deforming under the action slabs typically have large span-depth ratios, the resulting
of temperature changes, internal stresses are developed. flexibility means that deflection control is usually a major
Cracking that occurs when tensile stresses exceed the factor in design. Efforts to minimize slab thickness to reduce
concrete tensile strength reduces the flexural stiffness of construction costs can lead to problems associated with
the member and results in increased deflections under deflection of slab systems.
subsequent loading. Consequently, significant temperature ACI 318 specifies minimum thickness requirements for
effects should be taken into account in determining member control of two-way slab deflections. If the slab thickness
stiffness for deflection calculation. The calculation of the equals or exceeds the specified minimum thickness, deflec-
effective moment of inertia should be based on maximum tions need not be computed and serviceability in terms of
moment conditions. deflection control is deemed to be satisfied.
In cases where stresses are developed in the member due Experience has shown that, in most cases, the use of ACI
to restraint of axial deformations, the induced stress due to minimum thickness equations produces slabs that behave
axial restraint should be included in the calculation of the in a satisfactory manner under service loads. ACI 318
cracking moment in a manner analogous to that for including does permit a smaller-than-minimum thickness to be used
the prestressing force in prestressed concrete beams. if deflections are calculated and shown to be within limits
4.5.3 Section moment-curvature method for temperature- specified by the standard. However, the calculation of slab
induced deflections―As noted in [Link], the moment-curva- deflections requires careful consideration of many factors if
ture method can be used to predict the deflection of members realistic estimates of deflections are to be obtained. ACI 318
that experience deflection due to internal restraint. Thus, it is provides limited guidance on deflection calculation. This
a suitable method to consider deflections caused by tempera- chapter provides a detailed discussion of the many factors
ture-induced deflections, including thermal gradients. that need to be considered.
Figure 4.5.3 illustrates the case of applying a linear thermal This chapter reviews calculation of two-way slab deflec-
gradient over the height of a member, where the tempera- tions. Methods of calculating slab deflections are presented.
ture at the top surface is increased while the bottom surface Effects of two-way action, cracking, creep, and shrinkage
is kept constant. The member is also subjected to applied are considered. The minimum thickness equations in ACI 318
external loads, resulting in bending moment M at the cross are examined and proposals for alternative methods of speci-
section considered. A distribution of total strains εt is given fying minimum thickness are reviewed.
in Fig. 4.5.3. The free thermal strains and the strain due to
stress are also illustrated in Fig. 4.5.3. Note that the internal

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 23

5.2—Deflection calculation methods for two-way Rangan (1976) and Scanlon and Murray (1982) described
slab systems calculation procedures in which the column and middle
5.2.1 Immediate deflection of uncracked slabs—In this strips are treated as continuous beams; the middle strip is
section, calculation procedures are given for immediate considered to be supported at its ends by column strips that
deflections based on three approaches, namely classical run perpendicular to the middle strip. Results obtained by
solutions, simplified crossing beam analogies, and finite the crossing beam approach have been found to compare
element analysis. favorably with results obtained by finite element analyses
[Link] Classical solutions—Immediate deflection of (Scanlon and Murray 1982; Chang and Hwang 1996).
uncracked two-way slab systems loaded uniformly can be For two-way systems supported on a rectangular layout of
determined using plate bending theory for elastic thin plates. columns, the ACI 318 method of design for strength involves
Load-deflection response is governed by the plate equation dividing the slab into column strips and middle strips in each
(Timoshenko and Woinowsky-Krieger 1959) of the two orthogonal directions. The total static moment
for each span Mo = wuℓ2ℓn2/8, is divided between positive
and negative moment regions and then between column and
∂4 ∆ ∂4 ∆ ∂4 ∆ w
4
+2 2 4 + 4 = (5.2.1.1a) middle strips, using either the direct design method or the
∂x ∂x ∂y ∂y d equivalent frame method, where wu is the intensity of load
per unit area, ℓn is the effective span, and ℓ2 is the dimen-
where x, y are the orthogonal coordinate axes of the middle sion in the perpendicular direction. The distribution of
surface; ∆ is the deflection of the plate; w is the transverse moments approximates the elastic distribution for the given
load per unit area; D is the flexural rigidity per unit width, loading and therefore can be used to obtain an estimate of
Eh3/12(1 – ν2); E is the modulus of elasticity; h is the plate immediate deflections. However, the applied moments for
thickness; and ν is Poisson’s ratio. strength design use factored loads whereas the moments for
The relationship between moments and curvatures is calculation of immediate deflections use service loads that
given by are unfactored. The bending moments required to calculate
deflections or curvatures of column and middle strips may
be taken as the bending moments determined for factored
M x  1 v 0  κx 
  loads according to ACI 318 multiplied by the ratio of service
Eh3   
M y  v 1 0   κ y  (5.2.1.1b) load to factored load.
  12(1 − ν2 ) 
0 0 2(1 + ν)  κ xy  Figure [Link] shows a rectangular panel in a column-
 M xy  supported two-way slab system. The dotted areas represent a
set of crossing beams from which column strip deflection ∆c
For rectangular plates with uniformly distributed loads,
and middle strip deflection ∆m can be obtained. Each beam
the solution of Eq. (5.2.1.1b) leads to an expression for
can be treated as a strip of unit width for which end moments,
maximum deflection in the form
midspan moment, and flexural rigidity properties can be
obtained. Note that, by definition, end moments are those
αw 4 αw 4 at the faces of supports, such as column or column capital
=∆ = (5.2.1.1c)
D Eh3 /12(1 − ν2 ) faces, and that the beam span is the clear span between the
faces of such supports.
where ℓ is the longer span length; w is the uniform transverse Once the end moments and midspan moment have been
load; and α is a coefficient depending on the boundary condi- obtained for a column or middle strip, the deflection for
tions and aspect ratio. the strip can be calculated using the elastic beam deflection
It is noted from Eq. (5.2.1.1c) that the influence of Pois- equation
son’s ratio ν on deflections is quite small. Typical values of
ν for concrete fall in the range between 0.15 and 0.25. The 5  2n
term (1 – ν2), in the flexural rigidity D, falls in the range 0.94
=∆
48 EI
[ M m + 0.1(M1 + M 2 )] (5.2.1.2a)
to 0.98. The error involved in neglecting Poisson’s ratio is,
therefore, approximately 2 to 6 percent. where ℓn is the clear span; M1 and M2 are the end moments
Solutions of the plate equation for various geometries per unit width; and Mm is the midspan moment per unit width.
and support conditions have been given by Timoshenko and Positive Mm, M1, or M2 produce tension at bottom fiber.
Woinowsky-Krieger (1959) and by Jensen (1938). Because Using this procedure, the deflection of each column strip
closed-form solutions of the plate equation are available for (∆c) and of each middle strip (∆m) can be calculated. The
only a limited number of cases, alternative solution proce- midpanel deflection ∆mp is obtained by adding the column
dures are required for most practical situations. and middle strip deflections.
[Link] Crossing beam methods—Several approaches
have been developed in which the two-way slab system is ∆mp = ∆c + ∆mp (5.2.1.2b)
considered an orthogonal one-way system, thus allowing
deflection calculations by beam analogy. Some of the earlier
approaches were summarized in ACI 435.6R.

American Concrete Institute – Copyrighted © Material – [Link]

24 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

accounted for in the analysis for κ at each section. In the

absence of prestressing, simplification can be made by use
of multipliers and graphs (Ghali 1989; Ghali et al. 2011) that
also account for the cracking, creep, and shrinkage effects.
[Link] Finite element method—The finite element
method can be used to analyze plates with irregular support
and loading conditions. Effects of beams and columns can
be included, and general-purpose computer programs are
available for elastic analysis of plate systems. Computer
programs that account for cracking, creep, and shrinkage
have been under development since the 1970s (Scanlon
and Murray 1974; Gilbert and Warner 1978; ASCE Task
Committee 1982; Hirsch et al. 2012) and commercial soft-
ware is now available.
Although the method is becoming increasingly popular
in engineering practice, some skill is required in selecting
appropriate finite element and material models, developing
an appropriate mesh, preparing computer input data, and
interpreting the results.
5.2.2 Effect of cracking—The procedures outlined previ-
ously are applicable to linear elastic isotropic plate systems
and should be modified for concrete slabs to include the
Fig. [Link]―Crossing beam approach. effect of cracking on flexural stiffness. An early application
of classical anisotropic theory to analysis of two-way rein-
For cantilever slabs, the deflection equation for cantile- forced concrete slabs is given in the text by Timoshenko and
vers should be used and the rotation at the support should Woinowsky-Krieger (1959). More recently, procedures have
be included. been proposed for including cracking in finite element anal-
An earlier version of the equivalent frame method for ysis and in the crossing beam analogies for two-way slabs.
calculating deflections proposed by Vanderbilt et al. (1965) The effective moment of inertia Ie concept, developed
considers the midpanel deflection as the sum of a column originally by Branson (1963) for beams, can be applied
strip deflection, cantilever deflection of a portion of the directly to the column and middle strips in the crossing beam
middle strip extending from the column strip, and the analogies described in [Link] for elastic uncracked plates.
midpanel deflection of a simply supported rectangular plate. In Eq. (5.2.1.2a), the cross section stiffness EI becomes EcIe
The procedure developed by Nilson and Walters (1975), using the usual averaging procedures given in ACI 318 for Ie,
based on the equivalent frame method, is similar to the calculated at both positive and negative moment locations.
method outlined previously, except that a reference deflec- Kripanarayanan and Branson (1976) presented an extension
tion is calculated for the total panel width. Deflections for of the Nilson and Walters (1975) equivalent frame proce-
column and middle strips are then obtained from this refer- dure to include cracking using the Ie procedure. An alterna-
ence deflection using lateral distribution factors based on tive expression for effective moment of inertia proposed by
relative M/EI values. A numerical example (Nawy 2009) Bischoff (2005, 2007) is discussed in [Link].
calculating the expected deflection limits using this proce- Finite element models for cracking in two-way slabs
dure is given in Appendix B. The resulting values are appli- based on layered plate modeling and concrete stress-strain
cable instead of the minimum thickness values from ACI relationships accounting for tension stiffening have been
318-14 Table [Link]. available since the early 1970s (Scanlon and Murray 1974;
Ghali (1989) calculates the deflection at midspan of a Gilbert and Warner 1978). More recently, general purpose
column or middle strip from values of curvature calculated computer programs accounting for tension stiffening have
on the basis of compatibility and equilibrium at the midspan become commercially available. A simple generalization of
and supports using the relationship the effective moment of inertia concept to two-way systems
was suggested by Scanlon and Murray (1982) and imple-
2 mented in a modified version of a linear elastic plate bending
=
∆ (κ L + 10κ c + κ R ) (5.2.1.2c)
96 finite element by Graham and Scanlon (1986). An extension
of this approach has recently been presented by Vollum
where κL, κc, and κR are the curvatures calculated from (2011).
analysis of sections at the left end, center, and right end of 5.2.3 Restraint cracking—In two-way reinforced concrete
column and middle strips, respectively, and ℓ is the distance slabs built monolithically with supporting column and wall
between the two ends. This relationship is based on the elements, in-plane shortening due to shrinkage, and thermal
assumption that variation of curvature over the length ℓ is effects is restrained. The restraint is provided by a combina-
parabolic. The effects of cracking, creep, and shrinkage are tion of factors, including embedded reinforcement, attach-

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 25

ment to structural supports, and lower shrinkage rates of For slab systems in which significant restraint to in-plane
previously placed adjacent panels when slab panels are deformations may be present, Scanlon and Murray (1982)
placed at different times. Nonlinear distribution of free recommended that a reduced effective modulus of 4 f c′ (psi)
shrinkage strains across the cross section may also be a
(0.33 f c′ [MPa]) be used to compute the effective moment of
contributing factor.
inertia Ie based on Branson’s equation. Scanlon and Bischoff
Service load moments in two-way slabs are often of the
same order of magnitude as the code-specified cracking (2008) recommended a value of 5 f c′ (psi) (0.41 f c′ [MPa])
moment Mcr. Deflection calculations made using the code- (equivalent to 2/3Mcr) along with Bischoff’s equation for the
specified modulus of rupture will often result in an uncracked effective moment of inertia provided in Eq. (4.3.2.2a).
section being used when cracking may actually be present 5.2.4 Long-term deflections—Long-term deflections
due to a combination of flexural stress and restraint stress. can be estimated by applying a long-term multiplier to the
ACI 318 specifies the modulus of rupture for deflection calculated immediate deflection. Values for the long-term
multiplier are specified in design codes such as ACI 318,
calculations as 7.5 f c′ (psi) (0.62 f c′ [MPa]). Laboratory
where a value of 2/(1 + 50ρ′) is applied to the immediate
test data, summarized in ACI 209R, indicate values ranging
deflection caused by the sustained load considered. As
from 6 to 12 f c′ (psi) (0.5 to 1.0 f c′ [MPa]). discussed in [Link], numerous authors have suggested that
For slab sections with low reinforcement ratios, the ACI 318 long-term multiplier is too low for application
approaching minimum reinforcement, the difference to two-way slab systems, being based on poor correlation
between cracked and uncracked flexural stiffness is signifi- between reported calculated long-term deflections and field-
cant. It is important, therefore, to account for effects of measured deflections. It may be that much of the discrepancy
any restraint cracking that may be present. Unfortunately, between calculated and measured deflections is due to the
the extent of restraint cracking is difficult to predict. To effect of restraint cracking described previously. Addition-
account for restraint cracking in two-way slabs, Rangan ally, shrinkage warping is more significant in shallow slab
(1976) suggested that column strip deflections be based on systems than in deeper beam sections. Shrinkage warping
the moment of inertia of a fully cracked section Icr and that occurs where the centroids of the steel reinforcement and
middle strip deflections be based on (Ig + Icr)/2. Good agree- the uncracked portion of the concrete do not coincide, and
ment was reported between calculated and field measured the shrinkage of concrete, combined with the dimensional
deflections. stability of the reinforcement, warps the member causing
A more general approach was proposed by Scanlon and deflections to occur.
Murray (1982). They suggested that the effect of restraint Two approaches are presented next for estimating long-
cracking be included by introducing a restraint stress fres that term deflections—namely, by detailed computations and by
effectively reduces the modulus of rupture for calculating the ACI multiplier methods.
Mcr—that is [Link] Detailed calculations—Effects of creep deflection
and shrinkage warping may be considered separately using
fe I g procedures outlined in ACI 209R, based on the work of
M cr = (5.2.3) Branson et al. (1970) and Branson and Christiason (1971).
yt
Deflection due to creep is obtained from
where fe = fr – fres.
∆cr = krCt∆I (5.2.4.1a)
A value of 4 f c′ (psi) (0.33 f c′ [MPa]), or approxi-
mately half of the value specified in ACI 318, was proposed where Ct is the time-dependent creep coefficient representing
for the modulus of rupture used to compute the reduced creep strain at any time t, in days, after load application; kr
cracking moment. This approach was investigated by Tam is the factor to account for compression reinforcement and
and Scanlon (1986) and has produced good correlation neutral axis shift; and ∆i is the immediate deflection due
between calculated deflection and reported mean field- to dead load plus sustained live load, including effects of
measured deflections (Jokinen and Scanlon 1985; Graham cracking.
and Scanlon 1985). Equation (3.3.3a) gives the ACI 209R equation for Ct. ACI
Ghali (1989) has also used the idea of a reduced modulus 209R provides typical values of factors applying to moist-
of rupture and demonstrated the calculation of restraint stress cured concrete loaded at 7 days or later (refer to Chapter 3
due to reinforcement in the presence of uniform shrinkage. for details). For slabs loaded before 7 days, these values may
An additional consideration is that the moments used be used for first approximations.
in design for strength are based on some redistribution of In a two-way slab, shrinkage occurs in all directions.
moments. The distribution of design moments does not The shrinkage deflection should therefore be calculated for
reflect the high peaks of moment adjacent to columns that orthogonal column and middle strips, and the results added
occur in uncracked slabs. Deflection calculations based to give the total midpanel shrinkage deflection. Although
on moment distributions used for design therefore tend to there may be a contribution to shrinkage warping from
underpredict the effects of cracking on flexural stiffness. nonuniform shrinkage strains through the slab cross section,

American Concrete Institute – Copyrighted © Material – [Link]

26 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Table [Link]—Multipliers recommended by different authors

Source Modulus of rupture, psi Immediate Creep λc Shrinkage λsh Total λt

Sbarounis (1984) 7.5 f c′ 1.0 2.8 1.2 5.0

Branson (1977) 7.5 f c′ 1.0 2.0 1.0 4.0

7.5 f c′ 1.0 2.0 2.0 5.0

Graham and Scanlon (1986)
4 f c′ 1.0 1.5 1.0 3.5

Hossain et al. (2011) 7.5 f c′ 1.0 3.0 4.0

ACI 318 7.5 f c′ 1.0 2.0 3.0

Note: 7.5 f c′ (psi) = 0.62 f c′ (MPa) and 4 f c′ (psi) = 0.33 f c′ (MPa).

there are insufficient experimental data available to make multiplier is not the most accurate method for predicting the
specific recommendations for deflection calculations. deflections associated with shrinkage warping as they are not
Shrinkage warping deflection for a beam is given by directly proportional to the immediate deflection. However,
this approach has been adopted by numerous researchers
∆sh = kshκshℓ2 (5.2.4.1b) and codes (including ACI 318) due to its simplicity and
where ksh is the coefficient depending on end conditions: convenience.
= 11/128 (one end continuous) As a first approximation, the additional deflection at inter-
= 1/16 (both ends continuous) mediate time intervals due to sustained loads can be calcu-
= 1/8 (simple beam) lated using the values for ξ (Table [Link].3 of ACI 318-14)
= 1/2 (cantilever) multiplied by the factor (λt – 1)/2.
κsh is the shrinkage curvature It is recommended that, in cases where restraint stresses
= 0.7εsh(t)ρ1/3/h, singly reinforced member are expected to have an insignificant effect, the multi-
= 0.7εsh(t)(ρ – ρ′)1/3((ρ – ρ′)/ρ)1/2/h, doubly reinforced plier for sustained-load deflection be increased from 2
member to 4, as recommended by Sbarounis (1984) and Graham
Shrinkage warping deflections can also be determined using and Scanlon (1986). In this case, the ACI 318 value
the equivalent tension force method outlined in ACI 209R. for modulus of rupture would be used. In cases where
The total deflection at any time is obtained by adding restraint stresses are likely to have a significant effect on
immediate deflection due to sustained load, creep deflec- cracking, for example, large slab areas and stiff lateral
tion due to sustained load, shrinkage warping deflection, and restraint elements such as structure walls and columns, it is
deflection due to the part of the live load that is transient. recommended that a reduced modulus of rupture given by fr
Sophisticated finite element models have been developed = 4 f c′ (psi) (0.33 f c′ [MPa]) be used along with a long-
(ASCE Task Committee 1982) to account for time-depen- term sustained-load multiplier of 2.5.
dent deformations of two-way slabs caused by creep and Values recommended in ACI 209R for ultimate creep and
shrinkage. These models are generally used for research shrinkage coefficients are Cu = 2.35 and (εsh)u = 780 × 10–6,
purposes and are considered too complex for normal design respectively, at standard conditions as discussed in Chapter
applications, particularly when the high variability of creep 3. Sbarounis (1984) has suggested that, at standard condi-
and shrinkage properties is considered. tions, the long-term multipliers be modified if the concrete
[Link] ACI multiplier—While deflection calculations properties are known, and better estimates of ultimate creep
can be made for long-term effects of creep and shrinkage, C̄u and shrinkage (ε sh )u are available. Thus,
as outlined previously, the use of a multiplier applied to the
immediate deflection provides a simple calculation proce-
Cu (ε )
dure that is adequate for most purposes. This approach is λ t = (1 + λ c ) + λ sh sh u (5.2.4.2b)
used in ACI 318, in which a sustained-load multiplier of up Cu (ε sh )u
to 2 is applied to the immediate deflection of a member with
no compression reinforcement. Several authors have recom- 5.3—Minimum thickness requirements
mended increasing this factor for two-way slabs, as indicated Because of the complexities involved in calculating
in Table [Link]. The variation among the proposals indicates two-way slab deflections, engineers have preferred to
that there is considerable uncertainty in predicting long-time control deflections by giving minimum slab thickness as a
deflections. The total long-time multiplier is expressed as function of span length. Equations such as those in 8.3 of
ACI 318-14, as shown in Table 5.3, are based on experience
λt = 1 + λc + λsh (5.2.4.2a) gained over many years. The ACI 318 equations express
minimum thickness in terms of clear span between columns,
where λc is the multiplier for creep and λsh is the multiplier steel yield strength, and flexural stiffness of edge beams.
for shrinkage warping. Expressing shrinkage warping as a The minimum thickness values are modified for the effects

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 27

Table 5.3—Minimum thickness of nonprestressed two-way slabs without interior beams, in. (ACI 318-14
Table [Link])*
Without drop panels† With drop panels†
Exterior panels Exterior panels
Interior Interior
fy‡ Without edge beams With edge beams §
panels Without edge beams With edge beams§ panels
40,000 psi (280 MPa) ℓn/33 ℓn/36 ℓn/36 ℓn/36 ℓn/40 ℓn/40
60,000 psi (420 MPa) ℓn/30 ℓn/33 ℓn/33 ℓn/33 ℓn/36 ℓn/36
75,000 psi (520 MPa) ℓn/28 ℓn/31 ℓn/31 ℓn/31 ℓn/34 ℓn/34
For two-way construction, ℓn is the length of clear span in the long direction, measured face-to-face of supports (in.).
*

Drop panels as given in ACI 318-14 Section 8.2.4.

†

For fy between the values given in the table, minimum thickness shall be determined by linear interpolation.
‡

§
Slabs with beams between columns along exterior edges. Exterior panels shall be considered to be without edge beams if αf is less than 0.8. The value of αf for the edge beams shall
be calculated in accordance with ACI 318-14 Section [Link].
Note: 1 in. = 25.4 mm.

of drop panels and discontinuous edges. ACI 318 permits the tural elements. Two such approaches are described in the
use of thinner slabs if deflections are computed and found to following paragraphs.
satisfy the specified maximum permissible values. Gilbert (1985) extended to two-way slabs an expression
An extensive evaluation of the current ACI minimum developed by Rangan (1982) for maximum allowable span-
thickness equations was reported by Thompson and Scanlon depth ratio for beams. Rangan’s equation involves rear-
(1988). The study was based on finite element analyses of ranging the basic equation for beam deflection calculations
more than 300 slabs covering a range of thickness values,
aspect ratios, edge beam dimensions, construction loads, and ∆ = ∆L,add + ∆sus (5.3a)
other parameters. The main conclusions of the study were as
follows: where ∆L,add is the deflection due to the variable part of
a) Calculated deflections for slabs designed according to live load (kwL,addℓ4)/EcIe; ∆sus is the total deflection due to
the minimum thickness requirements of ACI 318 were sustained load (ws) including sustained part of live load (=
within the permissible limits, when the calculations λ∆(kwsℓ4/Ec Ie)); k is a constant based on support and load
were based on the ACI 318-specified value of 7.5 f c′ conditions; and λ∆ is the long-term multiplier.
(psi) (0.62 f c′ [MPa]) for modulus of rupture, and the Replacing Ie by αbd3, where the term α gives an approxi-
ACI 209R equations for creep and shrinkage deflection. mation for Ie /bd3 as a function of the reinforcement ratio ρ,
Construction loads due to shoring and reshoring were also Eq. (5.3a) can be rewritten as
considered. 3
b) When the calculations were based on a reduced modulus ∆ k ( wL , add + λ ∆ ws )
= (5.3b)
of rupture to account for restraint cracking, the ACI 318  Ec αbd 3
limit of ℓ/480 on incremental deflection was exceeded for
slab panels with aspect ratios less than 1.5. An increase of If ∆/ℓ is given as the maximum permissible deflection-to-
10 percent over the current minimum thickness value for span ratio, the corresponding maximum span-to-effective
square panels was suggested to obtain calculated deflec- depth ratio can be obtained from
tions within the allowable limits. The suggested increase 1/3
in minimum thickness decreases linearly to zero for a  ∆ αbEc 
< k1  (5.3c)
panel with an aspect ratio equal to 1.5. d   wL , add + λ ∆ ws 
The results of this study suggest that the ACI minimum
thickness equations will provide satisfactory serviceability where k1 is a combination of factors to account for support
in most cases, confirming the satisfactory performance of conditions and effect of beam flanges.
slabs designed and built according to the requirements in Gilbert extended Eq. (5.3c) by adding a slab system factor
ACI 318 prior to the 1989 edition. When more stringent than k2 to account for two-way action—that is
normal deflection limits are required, or when nonstandard
1/3
loads are applied, a more detailed analysis should be used.  ∆ αbEc 
Other means to increase the slab stiffness, such as the addi- < k1k2  (5.3d)
d   wL , add + λ ∆ ws 
tion of beams, can also be considered. Attempts have been
made to develop criteria for span-depth ratios or minimum The factor k2 was developed for a variety of conditions
thickness of slabs that explicitly include the effects of such from parameter studies using a sophisticated finite element
parameters as live-to-dead load ratio, permissible deflec- model. Equation (5.3d) involves an iterative procedure
tion-to-span ratio, effect of cracking, sustained load level, because the reinforcement ratio required to determine α and
and time between construction and installation of nonstruc- the dead load is initially unknown.

American Concrete Institute – Copyrighted © Material – [Link]

28 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

A somewhat simpler expression for beams was devel- (∆inc)allow is required incremental deflection limit
oped by Grossman (1981, 1987); it was based on numerous This equation can be specialized for flat plates as
computer-generated beam deflection calculations. Gross-
man’s minimum thickness equation is given by 1

 n  ∆ inc  1200 Ec  3
=   (in.-lb)
hmin >
 
Cd ≥

(5.3e) h    allow λ ∆ ws + wL , add 
c∆ 24 1
(5.3h)
 n  ∆ inc  8300 × 10 Ec 3 3
Correction factors are given for variations in support =    (SI)
conditions, d/h, fy, and concrete density. The term c was h    allow λ ∆ ws + wL , add 
developed from the computer analyses and depends on the
load levels and construction methods used. For heavily- An example taken from Scanlon and Lee (2006) is
loaded members, a limiting value of c = 4320 was proposed presented in Section B.3 to illustrate the application of the
by Grossman (1981) for heavily loaded members. Smaller proposed equation.
thicknesses can be obtained if the required reinforcement Gardner (2011) compared span-depth approaches from
ratio for less heavily loaded members is known and is used several codes and standards as well as those proposed by
to obtain a larger revised value of c from Grossman’s data. various authors. Lee and Scanlon (2007) presented a frame-
The term Cd, given by work based on utility theory to compute minimum thickness
considering deflection variability and uncertainty in estab-
λ′D + L lishing deflection limits.
Cd = (5.3f)
D+L
5.4—Loads for deflection calculation
accounts for both the live-load-to-dead-load ratio, L/D, and ACI 318 stipulates that calculated deflections must not
the net long-term multiplier λ′ for deflections occurring after exceed certain permissible values, expressed as fractions of
installation of partitions in buildings. span length. Components of deflections to be considered are
Although developed for beams only, the equation could immediate live load deflection and incremental deflection,
be extended to two-way systems using a slab system factor including that due to live load, after installation of nonstruc-
similar to that given by Gilbert (1985). tural elements. The live load component of deflection is
Scanlon and Lee (2006) extended Rangan’s (1982) orig- normally considered as that due to total load minus that due to
inal proposal, incorporating an approximation for effective dead load. Under monotonic loading, two effective moment of
moment of inertia, to produce a unified minimum thickness inertia values should be used to calculate the deflections at the
equation that can be applied to beams, one-way slabs, edge- two different load levels, as shown in Fig. 5.4a.
supported slabs, flat plates, and flat slabs. The general equa- For multistory slab construction, the loads imposed on
tion is given by the slab during construction may be the largest use of the
slab’s capacity throughout the life of the structure (Grundy
1
and Kabaila 1963). For these slabs, the extent of cracking is
n  ∆ inc  2400k DP Ec (b /12)  3
= β ES   (in.-lb) determined by the construction loads resulting from shoring

h    allow Kk AR k SS (λ ∆ ws + wL , add )  and reshoring procedures. Under these conditions, all values
(5.3g)
1 of immediate deflection should be calculated using the effec-
n  ∆ inc  16, 640k DP Ec b 3 tive moment of inertia corresponding to the construction
= β ES    (SI)
h    allow Kk AR k SS (λ ∆ ws + wL , add )  load level, as illustrated in Fig. 5.4b. Because the construc-
tion procedure is usually unknown during design, the full
where ws is sustained load (psf, Pa [slabs]; plf, N/m [beams]) specified dead load plus live load is recommended as a first
wL,add is additional live load (psf, Pa [slabs]; plf, N/m approximation to the maximum load during construction.
[beams]) This calculation procedure usually results in a smaller live
βES = 1 except βES = (long span)/(short span) ≤ 2.0 for edge- load deflection and larger dead load deflection, with corre-
supported slabs spondingly larger sustained load deflection.
K is deflection coefficient depending on support conditions A typical load-time history is shown in Fig. 5.4c for a slab
(= 5 for simply supported, = 1.4 for both ends continuous, = in a multistory structure. During construction, where multiple
2 for one end continuous, and = 48 for fixed end cantilver) levels of shoring equipment is used, the load on the slab
kDP = 1 except kDP = 1.35 for slab with drop panels increases as new slabs are placed above. When construction
kss = 1 except kss = 1.35 for column supported two-way slab above is no longer supported by the slab under consideration,
systems the load decreases to a value corresponding to the slab self-
kAR = 1 except kAR = 0.2 + 0.4β for edge-supported slabs weight plus an allowance for superimposed dead load and
b = 12 in. (1000 mm) for one-way and two-way slabs sustained portion of live load (load level at t1, in Fig. 5.4c).
= beam width (= web width, bw for T-beams), in. (mm) A simple procedure to determine slab loads during
construction was proposed by Grundy and Kabaila (1963).
More refined analysis procedures reported subsequently (for

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 29

example, Liu et al. [1985] and Aguinaga-Zapata and Bažant

[1986]) give results that are similar to the original Grundy
and Kabaila procedure. The maximum load during construc-
tion, including loads due to shoring and reshoring plus an
allowance for construction live load, can be estimated using
the following relationship

w = k1k2Rwslab + wCL/N (5.4a)

where k1 is the allowance for error in theoretical load ratio R;

k2 is the allowance for weight of formwork; R is the applied
load/slab dead load ratio and the load ratio calculated by the
Grundy and Kabaila procedure; wslab is the slab dead load;
wCL is the construction live load; and N is the number of
shored and reshored levels.
Gardner (1985) recommends k1 = k2 = 1.1. The construc-
tion live load may be taken as 50 lb/ft2 (2.4 kPa) as recom-
mended by ACI 347R. The factor k1 accounts for errors in
computing R due to variations in stiffnesses between the
Fig. 5.4a—Definition of δD and δL for monotonic moment- stories in the supporting system. The factor R has been
deflection curve. shown to vary from 1.2 to 2.2, depending primarily on the
number of stories of shores and reshores in the system. If
the shoring system to be used is unknown, a value of R =
1.5 can be used in the calculation as recommended by ACI
347.2R. Instead of multiplying by a factor k2 for formwork
weight, the formwork weight can be additive to the equation.
A value of 10 lb/ft2 (0.48 kPa) is considered a reasonable
allowance for most formwork systems, but this may need
to be confirmed against the actual weight of formwork if
deflection calculations need to be more precise. At time t2
in Fig. 5.4c, a slight increase in the sustained load occurs as
nonstructural elements are installed. The variable portion of
live load may be considered as applied intermittently there-
after. One application of live load is shown at time t3.
An analysis procedure based on this type of loading
history and ACI 209R creep and shrinkage functions was
developed by Graham and Scanlon (1986), using the prin-
ciple of superposition. Effects of partial creep recovery were
considered. Analyses were also made for the simplified
load-time history shown in Fig. 5.4d with the corresponding
displacement-time history. Long-term sustained load deflec-
tions were obtained using multipliers calibrated with the
more complex history of Fig. 5.4c. Resulting multipliers are
Fig. 5.4b—Definition of δD, δL, and δconst when construction included in Table [Link].
loads exceed specified dead plus live load. Based on the procedures suggested by Sbarounis (1984)
and Graham and Scanlon (1986), the following approach

Fig. 5.4c—Schematic load-time history.

American Concrete Institute – Copyrighted © Material – [Link]
30 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Fig. 5.4d—Simplified load-time history and corresponding deflection-time history.

based on the simplified load-time history can be used to esti-
mate long-term deflections in multistory slab systems.
(t2 − t1 )0.60
a) Estimate the maximum construction load expected, =λ1 λt (5.4e)
wconst(max), based on usual procedures for multistory 10 + (t2 − t1 )0.60
construction.
f) Calculate ultimate sustained load deflection.
b) Calculate the corresponding immediate construction
load deflection ∆const.
∆2 = λt∆sus (5.4f)
c) Calculate the live load deflection by scaling the
construction load deflection. where λt is long-term multiplier (Eq. 4.4.1b).
g) Calculate the deflection due to the variable portion of
wL E (Const ) live load, that is, that portion of live load not assumed as
=∆L ⋅ ∆ const ⋅ c (5.4b)
wconst (max) Ec ( L) sustained.

where Ec(Const) and Ec(L) are modulus of elasticity values ∆L,add = K∆L (from Step 3) (5.4g)
at application of construction load and live load, respec-
tively, which can be calculated using the equations in 3.3.1. where
d) Scale the construction load deflection to the sustained
load level. Sustained load includes dead load plus any variable live load wL , add
=K =
portion of the live load assumed to be sustained. total live load wL

wsus E (Const ) h) Calculate increment in deflection after installation of

=
∆ sus ∆ const c (5.4c)
wconst (max) Ec ( Sust ) nonstructural elements.

∆inc = ∆2 + ∆L,add – ∆1 (5.4h)

where Ec(Sust) is the modulus of elasticity at the time
sustained load is applied (for example, at end of construc- i) Compare calculated deflections with appropriate
tion period). permissible values.
e) Calculate sustained load deflection at time of installa- Scanlon and Suprenant (2011) presented a simplified
tion of nonstructural elements. method for estimating time-dependent deflections under a
prescribed construction loading time history. The method
∆1 = λ1∆sus (5.4d) is based on a generalization of creep superposition applied
to time-dependent deflections calculated using the crossing
where λ1 is multiplier corresponding to time interval t1 to
beam method. Hirsch et al. (2012) showed that this approach
t2. The time function given in Eq. (3.3.3a) can be used to
produces results that compare favorably with available
determine λ—that is,
experimental data and results obtained by more complex
finite element analysis.
American Concrete Institute – Copyrighted © Material – [Link]
 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 31

Fig. 5.5a—Field-measured deflections for 40 nominally identical slab panels in 28-story building (Jokinen and Scanlon 1985).
5.5—Variability of deflections
ACI 435.4R reported that the variability of actual deflec-
tions under nominally identical conditions is often large. For
simply supported beams under laboratory conditions, it was
reported that, using Branson’s I-effective procedure, there
is approximately a 90 percent probability that actual deflec-
tions of a particular beam will range between 80 and 130
percent of the calculated value. The variability of deflections
in the field can be even greater.
Based on Monte Carlo simulation, Ramsay et al. (1979)
indicated that the coefficient of variation for immediate
deflection of beams ranged from 25 to 50 percent. The major
source of variation was found to be flexural stiffness and
tensile strength of concrete, particularly when the service
load moment is close to the calculated cracking moment.
Other sources of variability include variations of slab thick-
ness and effective depth of reinforcing steel, and creep and
shrinkage properties. Choi et al. (2004) extended the work of
Ramsay et al. (1979) to include variability of time-dependent Fig. 5.5b—Histogram of 1-year deflection (Jokinen and
concrete properties and demonstrated that time-dependent Scanlon 1985).
deflections have higher variability than immediate deflec-
A histogram of 1-year deflections, shown in Fig. 5.5b,
tions, as expected.
indicates a coefficient of variation of 29.9 percent for these
Jokinen and Scanlon (1985) presented results of an
slabs and a range of deflections from approximately the
analysis of field-measured two-way slab deflections for a
mean minus 50 percent to the mean plus 70 percent. Calcu-
28-story office tower in Edmonton, Alberta, Canada. Figure
lated deflections at 1 year based on three assumed values
5.5a shows a plot of deflection versus time measurements
of modulus of rupture, and long-term multipliers proposed by
for 40 nominally identical slab panels. The high variability
Jokinen and Scanlon (1985), are shown in Fig. 5.5b. These
is evident, both during the construction period (first 35 days)
results indicate that the best estimate of the mean deflection
and at approximately 1 year thereafter.
was obtained using a reduced modulus of rupture of 4 f c′

American Concrete Institute – Copyrighted © Material – [Link]

32 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

expected deflection that enable design engineers to propor-

(psi) (0.33 f c′ [MPa]). The calculated deflection based on the
tion building structures to meet both strength and service-
ACI 318 specified value, 7.5 f c′ (psi) (0.62 f c′ [MPa]), was ability requirements. The result could be more economical
found to lie at the low end of the range of measured deflections. structures compared to those designed with unnecessarily
The calculations included effects of construction loads. conservative deflection-control approaches. The discussion
Sbarounis (1984) reported on deflection measurements taken assumes that a competent design is prepared in accordance
after 1 year on 175 bays of a multistory building in Chicago, with ACI 318 and construction follows good practices.
IL. Measured deflections had a mean value of 1.35 in. (34.3 To properly evaluate options for reducing deflection, a
mm) and a coefficient of variation of 21.2 percent. The range design engineer should know the level of stress in the member
in measured deflections was 0.53 to 2.16 in. (13.5 to 54.9 mm), under consideration, particularly whether the member is
that is, from the mean minus 60 percent to the mean plus 60 uncracked, partially cracked, or fully cracked. Heavily
percent. Calculated values were close to the mean deflection. reinforced members tend to be fully cracked because of the
Numerous case studies of large deflections reported in heavy loads they are subjected to. In this chapter, only two
the literature have been summarized in ACI 435.8R. These limiting conditions are considered: uncracked members and
case studies, including examples from Australia, Scotland, fully cracked members. If the maximum applied moment in
Sweden, and the United States, highlight the large number of the positive region is more than twice the cracking moment,
factors that can cause variability in in-place slab deflections. the member may be considered as fully cracked (Ie = Icr)
with little loss of accuracy. Frequently, a member is only
5.6—Allowable deflections partially cracked (Ie significantly greater than Icr) and the
ACI 318 specifies limits on maximum permissible computed statements about both limiting conditions are not strictly
deflections for live load deflection, and incremental deflec- applicable. In these situations, engineering judgement and
tion occurring after installation of nonstructural elements. No appropriate calculations should be made to assess the actual
limit is specified for total deflection. However, on the assump- serviceability conditions of the member. Chapters 4 and 5
tion that nonstructural elements will be installed shortly after of this report discuss methods for computing deflections of
construction of the slab, the specified limit on incremental partially cracked members.
deflection indirectly controls the total deflection. In addition to the stress conditions, there may be physical
The specified deflection limits apply to calculated deflections. or nonstructural constraints on the use of some options, such
By calculating deflections based on unfactored design loads and as limits on increasing concrete dimensions. All options
expected material properties, the calculated deflection should should be evaluated in terms of cost because some may
be interpreted as an estimate of the mean deflection. Recog- increase the cost, and some may have offsetting consider-
nizing the variability of actual deflections as measured in the ations that reduce the cost, while still others may have little
field, some variation from the calculated deflection is to be effect on cost. For each option presented, there is a discussion
expected. If the calculated deflection is close to the allowable on the effect of implementation on deflection, the approxi-
deflection, there is a high probability that the actual deflection mate range of potential reduction of deflection, and appro-
will exceed the allowable limit on computed deflection. priate situations in which the option should be considered.
Because the deflection limits are based on experience and The options are arranged in three groups: design techniques,
past practice resulting in generally satisfactory behavior, construction techniques, and materials selection.
it should be assumed that the variability of actual deflec-
tions is accounted for indirectly in the specified limits. The 6.2—Design techniques
specified limits in ACI 318 are based on previous experi- 6.2.1 Increasing section depth―Increasing the depth may
ence. The primary concern of ACI 318 is public safety. The not be possible after schematic design of the building has
serviceability provisions are of a general nature intended to been established because such dimensional changes may
provide adequate serviceability for most design situations. affect the architectural and mechanical work. There are many
Individual cases may require more stringent requirements instances, however, where beam depth can be increased. For
than the limited treatment given in ACI 318. Guidance on both uncracked and cracked sections, the increase in section
appropriate deflection limits for a range of applications was stiffness from an increase in depth is more than proportional.
given in ACI 435.3R. The moment of inertia of an uncracked rectangular section
is proportional to the section height cubed, and is given by
CHAPTER 6—REDUCING DEFLECTION OF Eq. (6.2.1a). The moment of inertia of a cracked rectan-
CONCRETE MEMBERS gular section with no axial loads is proportional to the depth
squared, and is given by Eq. (6.2.1b).
6.1—Introduction
Building structures designed by limit states approach may uncracked Ig = bh3/12 (6.2.1a)
have adequate strength but unsatisfactory serviceability
response. Namely, they may exhibit excessive deflection. cracked (nonprestressed) Icr = nAs(1 – k)jd2 (6.2.1b)
Thus, the size of flexural members is, in many cases, deter-
mined by deflection considerations rather than by strength. For preliminary design purposes, the cracked moment of
This chapter proposes design procedures for reducing the inertia can be assumed to be proportional to the area of the

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 33

Table [Link]—Effect of increasing d and As on cracked moment of inertia

A. Initial B. Increase d C. Increase d and decrease As
fcʹ 4000 psi (28 MPa) 4000 psi (28 MPa) 4000 psi (28 MPa)
fy 60,000 psi (410 MPa) 60,000 psi (410 MPa) 60,000 psi (410 MPa)
n 10 10 10
b 10 in. (250 mm) 10 in. (250 mm) 10 in. (250 mm)
h 18 in. (460 mm) 20 in. (510 mm) 20 in. (510 mm)
d 15.5 in. (390 mm) 17.5 in. (440 mm) 17.5 in. (440 mm)
As 1.0 in.2 (650 mm2) 1.0 in.2 (650 mm2) 0.88 in.2 (570 mm2)
Mu 790 kip-in. (89 kN-m) 900 kip-in. (100 kN-m) 790 kip-in. (89 kN-m)
Icr 1610 in. (6.7 × 10 mm )
4 8 4
2100 in. (8.7 × 10 mm )
4 8 4
1890 in.4 (7.9 × 108 mm4)

reinforcement times the depth of the reinforcement squared from the concrete to the compression reinforcement, which
(Asd2). For the cross sections in the example in [Link], using does not itself creep. The closer the reinforcement is to the
this assumption leads to errors of less than 5 percent. compression face of the member relative to the neutral axis
If, by increasing the depth, the concrete tensile stress in depth, the more effective reinforcement is in reducing creep
a member is reduced sufficiently so that it changes from deflection. Thus, compression reinforcement is more effec-
a cracked or partially cracked member to an uncracked tive in deeper than in shallower beams or slabs if the concrete
member, the stiffness could increase dramatically. The cover to the compression face of the member is of constant
uncracked stiffness can be as much as three times the value. For some very shallow members, due to the require-
cracked stiffness (Grossman 1981). Caution should be exer- ments of minimum bar cover, compression reinforcement
cised when assuming a member is uncracked, as shrinkage could be at or near the neutral axis and be almost totally
and thermal stresses may lessen the load that is required to ineffective in reducing long-term creep deflection.
crack the member. Shrinkage warping occurs where the centroids of the steel
[Link] Example: Effect of depth on cracked moment of reinforcement and the uncracked portion of the concrete do
inertia—This example shows the effect of increasing the not coincide, and the shrinkage of concrete, combined with
depth of a member on the cracked moment of inertia of a the dimensional stability of the reinforcement, warps the
cross section. Cross Section A is the initial cross section member in a fashion similar to a piece of bimetal subject
before changes. In Cross Section B, the depth is increased, to temperature variations. Compression reinforcement
but all other parameters are kept constant (and moment reduces shrinkage warping because it brings the centroid
capacity is increased); in Cross Section C, the depth is of the tension and compression reinforcement closer to the
increased the same as Cross Section B, but the area of rein- centroid of the uncracked concrete. While compression rein-
forcement is decreased to give the same moment capacity as forcement reduces shrinkage and warping of all flexural
Cross Section A (Table [Link]). members, it is especially effective for T-beams where the
6.2.2 Increasing section width—This option is not appli- neutral axis is close to the compression face and far from the
cable to slabs or other members with physical constraints on tension reinforcement. If the T-beam has a thin slab subject
their width. Where beams cannot be made deeper because to higher-than-normal shrinkage because of its high surface-
of floor-to-floor height limitations, but can be made wider, to-volume ratio, then compression reinforcement will be
the increase in stiffness is proportional to the increase in more effective than for a rectangular beam. This will be true
width if the member is uncracked. If the member is cracked for ribbed slabs or joist systems as well.
and remains cracked after increasing the width, the increase 6.2.4 Addition of tension reinforcement—For uncracked
in stiffness is very small. However, if a cracked member members, addition of tension reinforcement has hardly any
becomes uncracked because the width is increased, its stiff- effect on deflection. For fully cracked members, the addi-
ness increases appreciably, possibly by as much as a factor tion of tension reinforcement reduces both immediate and
of three (Grossman 1981). long-term deflection almost in proportion to the increase
6.2.3 Addition of compression reinforcement—Compres- in the steel reinforcement area. This can be seen from the
sion reinforcement has some effect on immediate deflection. cracked moment of inertia Icr, defined in Eq. (6.2.1b). For
It can influence Icr, and as a result Ie, and the corresponding all practical purposes, Icr = 0.9Asnd2 because the variation
initial deflection, although the influence may be small. It in the term (1 – k)j is usually small. For example, if the total
can, however, significantly reduce the incremental deflection deflection of a cracked member is 1.5 in. (38 mm), as in the
that occurs due to creep and shrinkage. previous example, increasing the tension reinforcement by
The increases in deflection that occur over time have 50 percent will reduce the deflection to approximately 1.1 in.
two components: creep deflection and shrinkage warping. (28 mm). However, the increased reinforcement area should
Compression reinforcement reduces deflection because still be less than the maximum permitted by code to ensure
concrete creep tends to transfer the compression force ductile behavior. This option is most effective for lightly

American Concrete Institute – Copyrighted © Material – [Link]

34 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

reinforced solid and ribbed slabs. The option of adding concrete has reached its design strength (refer to 6.3.1).
more tension reinforcement is not available or is limited for Construction loads should never be allowed to exceed
heavily reinforced beams unless compression reinforcement design loads, with applicable reductions included, without
is also added to balance the increase in tension bar area and proper prior engineering evaluation. Also, the total construc-
ensure code ductility requirements are satisfied. tion load should not be permitted to exceed the maximum
6.2.5 Revision of structure geometry—Common solu- load-carrying capacities of the early age concrete slabs. ACI
tions to reduce deflections include increasing the number 347.2R provides information on construction load analysis
of columns to reduce the length of the spans, adding cross and shoring/reshoring design.
members to create two-way systems, and increasing the Construction of formwork and shoring should ensure that
size of columns to provide more end restraint to flexural a sag or negative camber is not built into the slab. Experi-
members. ence indicates that frequently the apparent deflection varies
6.2.6 Revision of deflection limit criteria—If the predicted widely between slabs of identical design and construction.
deflection of a member is excessive, the deflection limits may Some reasons for this may be that such slabs were not all
be reexamined to determine if they are unnecessarily restric- built level or at the specified grade or the method and timing
tive. If experience or analysis indicates that those limits of form stripping was not uniformly applied. Also, construc-
(refer to Chapter 4) can be relaxed, then other action might tion loads may not have been applied uniformly.
not be required. Many building codes do not set absolute 6.3.4 Delay of the first loading—This allows the concrete
limits on deflection. An engineer may be able to determine to gain more strength before loading or helps to reach its
that the building occupancy, or construction conditions, such design strength. Both the modulus of elasticity Ec and the
as a sloping roof, do not require the normal deflection limits. modulus of rupture fr will be increased. An increase in Ec
increases the flexural stiffness. An increase in the modulus
6.3—Construction techniques of rupture value fr reduces the amount of cracking or even
6.3.1 Concrete curing to allow gain in strength—Deflec- allows the member to remain uncracked with an increase in
tion response is affected more by concrete strength at the time flexural stiffness EI as noted in the following section.
of loading, than by final concrete strength. If the construction 6.3.5 Delay in installation of deflection-sensitive elements
schedule makes early loading of the concrete likely or desir- or equipment—A delay in equipment installation will have
able, then measures to ensure high strength at first loading or no effect on immediate or total deflection (except as previ-
construction loading can be effective. Consider for example ously noted in 6.3.1), but incremental deflection occurring
a concrete member with design compressive strength fc′ of after installation of deflection-sensitive components will be
4000 psi (28 MPa), that would be uncracked as designed. reduced. The deflection occurring before a deflection-sensi-
If loaded when the concrete’s actual strength is 2500 psi tive component is installed will have no effect on the compo-
(17.2 MPa), it could be excessively cracked due to a lower nent. For example, if the additional long-term deflection is
modulus of rupture at the time of loading. Even though its 1 in. (25 mm), and installation of partitions is delayed for
final load-carrying capacity may be deemed satisfactory, the 3 months, the incremental deflection will be approximately
cracked member may still deflect several times more than a 0.50 in. (13 mm), or approximately one-half as much as the
similar uncracked one. Furthermore, the modulus of elas- total deflection.
ticity of a 4000 psi (28 MPa) concrete is higher than that of 6.3.6 Location of deflection-sensitive equipment to avoid
2500 psi (17.2 MPa) concrete (refer to 6.4 for the effects of deflection problems—Equipment such as printing presses,
material selection on these parameters). scientific equipment, and the like should remain level and
6.3.2 Concrete curing to reduce shrinkage and creep— should be located at midspan where the change in slope is
Immediate deflection will not be greatly affected by concrete very small with the increase in deflection. On the other hand,
curing, but additional time-dependent deflection will be because the amplitude of vibration is highest at midspan,
reduced. Assuming the time-dependent component of vibration-sensitive equipment may be best located near the
deflection is evenly divided between shrinkage and creep, supports.
if shrinkage is reduced 20 percent by adequate curing, the 6.3.7 Provision of architectural details to accommodate
additional time dependent deflection due to shrinkage and expected deflection—Partitions that abut columns, as an
creep will be reduced by 10 percent. The effect will be most example, may show the effect of deflection by separating
pronounced on members subject to high shrinkage such as horizontally from the column near the top even though the
those with a high surface/volume ratio (smaller members), partition is not cracked or otherwise damaged. Architec-
those with thin flanges, and structures in arid atmospheres or tural details should accommodate such movements. Like-
members that are restrained. The effect of adequate curing wise, windows, walls, partitions, and other nonstructural
on creep is similar to its effect on shrinkage. elements supported by or located under deflecting concrete
Curing may also reduce shrinkage warping due to members should be provided with slip joints to accom-
nonuniform shrinkage through the depth of the member. If modate the expected deflections or differential deflections
shrinkage at the top of a slab is higher than at the bottom due between concrete members above and below the nonstruc-
to premature drying, the slab will deflect downward. tural elements.
6.3.3 Control of shoring and reshoring procedures―The 6.3.8 Building camber into floor slabs—Built-in
shoring loads may be imposed on the floor slabs before the camber has no effect on the computed deflection of a slab.

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 35

Cambering, however, is effective for the installation of parti- levers is particularly sensitive to misplacement of the top
tions and equipment, if the objective is to have a level floor reinforcing bars. Deflection could increase in continuous
slab after deflection takes place. For best results, deflec- members if the reduction in strength at negative moment
tion should be carefully calculated using the appropriate regions results in redistribution of moments.
modulus of concrete Ec value and the correct moment of
inertia I. Overestimating the deflection value can lead the 6.4—Materials selection
designer to specify unreasonable overcambering. Hence, the 6.4.1 Selection of materials for mixture design that reduce
pattern and value of cambering at several locations should shrinkage and creep or increase the moduli of elasticity
be specified and the results monitored during construction. and rupture—Materials having an effect on these proper-
Procedures can be revised as necessary for slabs that are to ties include aggregates, cement, silica fume, and admix-
be constructed at a later date. tures. Lower water-cement ratio (w/c), a lower slump, and
6.3.9 Ensuring that top bars are not displaced down- changes in other materials proportions can reduce shrinkage
ward—Downward displacement of top bars always reduces and creep or increase the moduli of elasticity or rupture.
strength. The effect on deflection in uncracked members is Shrinkage-reducing admixtures may also be used.
minimal, but its effect on cracked members, namely, those 6.4.2 Use of concretes with a higher modulus of elas-
that are heavily loaded, is in proportion to the square of the ticity—Using ACI 318 procedures, the stiffness of an
ratio of change in effective depths for cantilevers but much uncracked member increases in proportion to the elastic
less for continuous spans. This reduced effect in continuous modulus, which varies in proportion to the square root of
members is because the flexural stiffness and resulting the cylinder strength (ACI 318-14 Section 19.2.2). The stiff-
deflection of the member is determined primarily by member ness of a cracked section is affected little by a change in the
stiffness at the midspan section. Thus, the deflection of canti- modulus of elasticity.

Table 6.5—Deflection reducing options

Effect on section stiffness
Option Uncracked Cracked
Design techniques
(d*/d)2 or (d*/d)
1. Deeper members (h*/h)3 for rectangular beams; less for T-beams
If change to uncracked section, up to 300 percent
2. Wider members (b*/b) Small unless changed to uncracked section
3. Add As′ Up to 50 percent for ∆LT. No effect for ∆i. Up to 50 percent for ∆LT. No effect for ∆i.
4. Add As No effect As*/As
5. Add prestress Reduces dead load deflection to nearly zero Reduces dead load deflection to nearly zero and member to uncracked
6. Structural geometry Large effect Large effect
7. Revise criteria See text See text
Construction techniques
8. Cure: f ′c Same as higher Ec and fr Same as high Ec and fr and could change to uncracked section.
9. Cure εsh and εcr For long-term deflection (εsh*/εsh) and (εcr*/εcr) For long-term deflection (εsh*/εsh) and (εcr*/εcr).
10. Shoring Large effect, see text Large effect, see text
11. Delay first loading Same as higher Ec and fr Same as high Ec and fr and could change to uncracked section.
12. Delay installation Up to 50+ percent depending on time delay Up to 50+ percent depending on time delay
13. Locate equipment See text See text
14. Architectural details See text See text
15. Camber See text See text
16. Top bars No effect Up to (d*/d)2 for cantilevers
Materials
17. Materials See 6.4 See 6.4
18. Mixture design See 6.3.2 and 6.4.3 See 6.3.2 and 6.4.3
19. Higher Ec (Ec*/Ec) or (fc′*/fc′)0.5 Significant
20. Higher fr None Significant
21. Use fiber reinforcement See 6.3.2 and 6.4.4 See 6.3.2 and 6.4.4
Parameters that have been changed to reduce deflection.
*

American Concrete Institute – Copyrighted © Material – [Link]

36 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Because the stiffness of a flexural member is EI, both ACI 435.7R-85(97)—State-of-the-Art Report on Temper-
the modulus of elasticity and the moment of inertia affect ature-Induced Deflections of Reinforced Concrete Members
stiffness significantly. Concrete with a higher modulus of (no longer in print)
elasticity will reduce deflections. The actual modulus of ACI 435.8R-85(97)—Observed Deflection of Reinforced
elasticity may vary significantly from the ACI 318 code Concrete Slab Systems, and Causes of Large Deflections
prediction. If historic mixture data is not available, modulus ACI 440.1R-15—Guide for the Design and Construction
of elasticity tests can be performed on trial batches. of Concrete Reinforced with FRP Bars
6.4.3 Use of concretes with a higher modulus of rupture— ACI 544.1R-96(09)—Report on Fiber Reinforced Concrete
Concrete with a higher modulus of rupture does not neces- ACI 544.5R-10—Report on the Physical Properties and
sarily increase the stiffness of uncracked members and highly Durability of Fiber Reinforced Concrete
cracked members. However, stiffness of partially cracked
members increases because of the reduction of the degree British Standards Institution
of cracking. The increase in stiffness (decrease in deflection) BS EN 1992-1-1:2004+A1:2014—Eurocode 2: Design of
depends on steel reinforcement percentage, the increase in Concrete Structures. General Rules and Rules for Buildings
modulus of rupture, and the magnitude of applied moment.
6.4.4 Addition of short discrete fibers to concrete mixture— Canadian Standards Association
Adding short discrete fibers to the concrete mixture (fiber- CSA A23.3:2014—Design of Concrete Structures
reinforced concrete) has been reported to reduce shrinkage
and increase the cracking strength, both of which might International Conference of Building Officials (merged into
reduce deflection (Alsayed 1993). Changes in the material International Code Council)
behavior depend on the type and amount of fibers used. ICBO 1997—Uniform Building Code

6.5—Summary International Federation for Structural Concrete (fib)

Table 6.5 summarizes some of the preventive measures fib MC2010—Model Code for Concrete Structures 2010
needed to reduce or control deflection. This table can serve as
a general guide to the design engineer but is not all inclusive, Standards Australia
and engineering judgement should be exercised in the choice of AS 3600-2009—Design and Construction of Concrete
the most effective parameters that control deflection behavior. Building Structures

CHAPTER 7—REFERENCES Authored documents

Committee documents are listed first by document number ACI Committee 209, 1971, “Effects of Creep, Shrinkage,
and year of publication followed by authored documents and Temperature Effects in Concrete Structures,” Designing
listed alphabetically. for Effects of Creep Shrinkage Temperature in Concrete
Structures, SP-27, American Concrete Institute, Farmington
American Concrete Institute Hills, MI, pp. 51-93.
ACI 209R-92(08)—Prediction of Creep, Shrinkage, and ACI Committee 209, 1982, “Prediction of Creep,
Temperature Effects in Concrete Structures Shrinkage, and Temperature Effects in Concrete Structures,”
ACI 209.2R-08—Guide for Modeling and Calculating Designing for Creep & Shrinkage in Concrete Structures: A
Shrinkage and Creep in Hardened Concrete Tribute to Adrian Pauw, SP-76, B. B. Goyal, ed., American
ACI 213R-14—Guide for Structural Lightweight-Aggre- Concrete Institute, Farmington Hills, MI, pp. 193-300.
gate Concrete ACI Committee 435, 1966, “Deflections of Reinforced
ACI 237R-07(19)—Self-Consolidating Concrete Concrete Flexural Members,” ACI Journal Proceedings, V.
ACI 318-14—Building Code Requirements for Structural 63, No. 6, June, pp. 637-674.
Concrete and Commentary ACI Committee 435, 1978, “Proposed Revisions by
ACI 347R-14—Guide to Formwork for Concrete Committee 435 to ACI Building Code and Commentary
ACI 347.2R-17—Guide for Shoring/Reshoring of Provisions on Deflections,” ACI Journal Proceedings, V. 75,
Concrete Multistory Buildings No. 6, June, pp. 229-238.
ACI 363R-10—Report on High-Strength Concrete Aguinaga-Zapata, M., and Bažant, Z. P., 1986, “Creep
ACI 435R-95(03)—Control of Deflection in Concrete Deflections in Slab Buildings and Forces in Shores During
Structures Construction,” ACI Journal Proceedings, V. 83, No. 5,
ACI 435.3R-68(89)—Allowable Deflections (no longer in Sept.-Oct., pp. 719-726.
print) Al-Zaid, R.; Al-Shaikh, A. H.; and Abu-Hussein, M.,
ACI 435.4R-72(89)—Variability of Deflections of Simply 1991, “Effect of Loading Type on the Effective Moment
Supported Reinforced Concrete Beams (no longer in print) of Inertia of Reinforced Concrete Beams,” ACI Structural
ACI 435.6R-74(89)—Deflection of Two-way Reinforced Journal, V. 88, No. 2, Mar.-Apr., pp. 184-190.
Concrete Floor Systems: State-of-the-Art Report (no longer Alsayed, S. H., 1993, “Flexural Deflection of Reinforced
in print) Fibrous Concrete Beams,” ACI Structural Journal, V. 96,
No. 1, Jan.-Feb., pp. 72-76.

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 37

ASCE Task Committee, 1982, Finite Element Analysis of Branson, D. E., and Christiason, M. L., 1971, “Time-
Reinforced Concrete, ASCE, New York, 553 pp. Dependent Concrete Properties Related to Design-Strength
Bažant, Z. P., and Kim, S., 1979, “Approximate Relax- and Elastic Properties, Creep and Shrinkage,” Symposium on
ation Function for Concrete,” Journal of the Structural Divi- Creep, Shrinkage, and Temperature Effects, SP-27, Amer-
sion, V. 105, No. 12, Dec., pp. 2695-2705. ican Concrete Institute, Farmington Hills, MI, pp. 257-277.
Bažant, Z. P., and Panula, L., 1980, “Creep and Shrinkage Branson, D. E.; Meyers, B. L.; and Kripanarayanan, K.
Characterization for Analyzing Prestressed Concrete Struc- M., 1970, “Time-Dependent Deformation of Noncomposite
tures,” PCI Journal, V. 25, No. 3, May-June, pp. 86-122. and Composite Prestressed Concrete Structures,” Highway
doi: 10.15554/pcij.05011980.86.122 Research Record, No. 324, pp. 15-43.
Beeby, A. W., 1968, “Short Term Deformations of Rein- Brooks, J. J., and Al-Qarra, H., 1999, “Assessment of
forced Concrete Members,” Cement and Concrete Associa- Creep and Shrinkage of Concrete for the Flintshire Bridge,”
tion Technical Report TRA 408, London, UK, Mar. Structural Engineering, V. 77, No. 5, Mar., pp. 21-26.
Bernard, E. S., 2004, “Creep of Cracked Fibre Reinforced Chang, K.-Y., and Hwang, S.-J., 1996, “Practical Estima-
Shotcrete Panels,” Shotcrete: More Engineering Develop- tion of Two-way Slab Deflections,” Journal of Structural
ments, Taylor & Francis Group, London, UK, pp. 47-57. Engineering, V. 122, No. 2, pp. 150-159. doi: 10.1061/
Bischoff, P. H., 2005, “Reevaluation of Deflection (ASCE)0733-9445(1996)122:2(150)
Prediction for Concrete Beams Reinforced with Steel and Choi, B.-S.; Scanlon, A.; and Johnson, P. A., 2004, “Monte
Fiber Reinforced Polymer Bars,” Journal of Structural Carlo Simulation of Immediate and Time-dependent Deflec-
Engineering, V. 131, No. 5, pp. 752-767. doi: 10.1061/ tions of Reinforced Concrete Beams and Slabs,” ACI Struc-
(ASCE)0733-9445(2005)131:5(752) tural Journal, V. 101, No. 5, Sept.-Oct., pp. 633-641.
Bischoff, P. H., 2007, “Rational Model for Calculating Collins, M. P., and Mitchell, D., 1997, Prestressed
Deflection of Reinforced Concrete Beams and Slabs,” Concrete Structures, Prentice-Hall Inc., Englewood Cliffs,
Canadian Journal of Civil Engineering, V. 34, No. 8, pp. NJ, 766 pp.
992-1002. doi: 10.1139/l07-020 Gardner, N. J., 1985, “Shoring, Reshoring, and Safety,”
Bischoff, P. H., and Gross, S. P., 2011a, “Equivalent Concrete International, V. 7, No. 4, Apr., pp. 28-34.
Moment of Inertia Based on Integration of Curvature,” Gardner, N. J., 2001, “Span/Thickness Limitations for
Journal of Composites for Construction, V. 15, No. 3, pp. Deflection Control,” Code Provisions for Deflection Control
263-273. doi: 10.1061/(ASCE)CC.1943-5614.0000164 in Concrete Structures, SP-203, E. G. Nawy and A. Scanlon,
Bischoff, P. H., and Gross, S. P., 2011b, “Design Approach eds., American Concrete Institute, Farmington Hills, MI, pp.
for Calculating Deflection of FRP Reinforced Concrete,” 95-114.
Journal of Composites for Construction, V. 15, No. 4, pp. Gardner, N. J., 2011, “Span/Thickness Limits for Deflec-
490-499. doi: 10.1061/(ASCE)CC.1943-5614.0000195 tion Control,” ACI Structural Journal, V. 108, No. 4, July-
Bischoff, P. H., and Scanlon, A., 2007, “Effective Moment Aug., pp. 453-460.
of Inertia for Calculating Deflections of Concrete Members Gardner, N. J., and Zhang, J., 1996, “Controlling Deflec-
Containing Steel Reinforcement and FRP Reinforcement,” tion Serviceability by Span/Depth Limits and Long-Term
ACI Structural Journal, V. 104, No. 1, Jan.-Feb., pp. 68-75. Deflection Multipliers for Reinforced Concrete Beams,”
Bischoff, P. H., and Scanlon, A., 2009, “Span-Depth Recent Developments in Deflection Evaluation of Concrete,
Ratios for One-Way Members Based on ACI 318 Deflection SP-161, American Concrete Institute, Farmington Hills, MI,
Limits,” ACI Structural Journal, V. 106, No. 5, Sept.-Oct., pp. 165-196.
pp. 617-626. Ghali, A., 1989, “Prediction of Deflections of Two-Way
Bischoff, P. H., and Veysey, S., 2011, “Minimum Thick- Floor Systems,” ACI Structural Journal, V. 86, No. 5, Sept.-
ness Requirements Needed to Satisfy Deflection Limits for Oct., pp. 551-562.
One-Way Reinforced Concrete Construction,” CSCE 2011 Ghali, A.; Favre, R.; and Elbadry, M., 2011, Concrete
General Conference, Ottawa, ON, Canada. Structures: Stresses and Deformations: Analysis and Design
Branson, D. E., 1963, “Instantaneous and Time-Dependent for Serviceability, fourth edition, CRC Press, Boca Raton,
Deflections of Simple and Continuous Reinforced Concrete FL, 646 pp.
Beams,” HPR Publication No. 7, Part 1, Alabama Highway Gilbert, R. I., 1985, “Deflection Control of Slabs Using
Department, U.S. Bureau of Public Roads, Aug., pp. 1-78. Allowable Span to Depth Ratios,” ACI Journal Proceedings,
Branson, D. E., 1971, “Compression Steel Effect on Long- V. 82, No. 1, Jan.-Feb., pp. 67-77.
Time Deflections,” ACI Journal Proceedings, V. 68, No. 8, Gilbert, R. I., 1999, “Deflection Calculation for Rein-
Aug., pp. 555-559. forced Concrete Structures—Why We Sometimes Get It
Branson, D. E., 1977, Deformation of Concrete Struc- Wrong,” ACI Structural Journal, V. 96, No. 6, Nov.-Dec.,
tures, McGraw-Hill Book Co., Advanced Book Program, pp. 1027-1032.
New York, 546 pp. Gilbert, R. I., 2006, “Discussion of “Reevaluation of Deflec-
Branson, D. E., 1985, Handbook of Concrete Engineering, tion Prediction for Concrete Beams Reinforced with Steel
second edition, M. Fintel, ed., Van Nortrand Reinhold Co., and Fiber Reinforced Polymer Bars,” by Peter H. Bischoff,”
New York, pp. 53-78. Journal of Structural Engineering, V. 132, No. 8, pp. 1328-
1330. doi: 10.1061/(ASCE)0733-9445(2006)132:8(1328)

American Concrete Institute – Copyrighted © Material – [Link]

38 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Gilbert, R. I., 2007, “Tension Stiffening of Lightly Beams,” Research Report No. 85-2, Department of Struc-
Reinforced Concrete Slabs,” Journal of Structural Engi- tural Engineering, Cornell University, Ithaca, NY, Feb.
neering, V. 133, No. 6, pp. 899-903. doi: 10.1061/ Martin, I., 1971, “Effects of Environmental Conditions on
(ASCE)0733-9445(2007)133:6(899) Thermal Variations and Shrinkage of Concrete Structures in
Gilbert, R. I., and Warner, R. F., 1978, “Tension Stiffening the United States,” Designing for Effects of Creep Shrinkage
in Reinforced Concrete Slabs,” Journal of the Structural Temperature in Concrete Structures, SP-27, American
Division, V. 104, No. 12, pp. 1885-1900. Concrete Institute, Farmington Hills, MI, pp. 279-300.
Graham, C. J., and Scanlon, A., 1985, “Deflection of Meyers, B. L., and Thomas, E. W., 1983, Handbook
Concrete Slabs under Construction Loading,” Deflections of Structural Concrete, McGraw-Hill, New York, pp.
of Concrete Structures, SP-86, G. Sabnis, ed., American 11.1-11.33.
Concrete Institute, Farmington Hills, MI, pp. 215-230. Myers, J. J., and Yang, Y., 2004, “Field and Laboratory
Graham, C. J., and Scanlon, A., 1986, “Long-Time Multi- Performance of Prestressed High Performance Concrete
pliers for Estimating Two-Way Slab Deflections,” ACI Girders for Missouri Bridge Structures,” Report No. 04-49,
Journal Proceedings, V. 83, No. 5, Sept.-Oct., pp. 899-908. Center for Infrastructure Engineering Studies, Missouri
Grossman, J. S., 1981, “Simplified Computations for University of Science and Technology, Rolla, MO, June, pp.
Effective Moment of Inertia (Ie) and Minimum Thickness to 96-98, 102.
Avoid Deflection Computations,” ACI Journal Proceedings, Nagataki, S., and Yonekura, A., 1978, “Studies of the
V. 78, No. 6, Nov.-Dec., pp. 423-439, and author’s closure, Volume Changes of High-Strength Concretes with Super-
V. 79, No. 5, Sept.-Oct. 1982, pp. 414-419. plasticizer,” Journal, Japan Prestressed Concrete Engi-
Grossman, J. S., 1987, Building Structural Design Hand- neering Association, V. 20, pp. 26-33.
book, R. N. White and C. G. Salmon, eds., John Wiley & Nassar, A. J., 2002, “Investigation of Transfer Length,
Sons, New York, pp. 699-786. Development Length, Flexural Strength and Prestress
Grundy, P., and Kabaila, A., 1963, “Construction Loads on Loss Trend in Fully Bonded High-Strength Lightweight
Slabs with Shored Formwork in Multi-Story Buildings,” ACI Prestressed Girders,” Virginia Polytechnic Institute and
Journal Proceedings, V. 60, No. 12, Dec., pp. 1729-1738. State University, May 15, 136 pp.
Hirsch, J., 2009, “Accurate Long-Term Deflection Predic- Nawy, E. G., 2009, Reinforced Concrete: A Fundamental
tion in Flat Slabs Using Linear Elastic Global Analysis,” 24th Approach, sixth edition, Pearson Education, NJ, 915 pp.
Biennial Conference of the Concrete Institute of Australia, Ngab, A. S.; Nilson, A. H.; and Slate, F. O., 1981,
Sydney, Australia, 2009, 11 pp. “Shrinkage and Creep of High-Strength Concrete,” ACI
Hirsch, J.; Calabrese, F.; Connolly, E.; and Bommer, A., Journal Proceedings, V. 78, No. 4, July-Aug., pp. 255-261.
2012, “Practical Deflection Prediction of Concrete Slabs,” Nilson, A. H., 1985, “Design Implications of Current
SP-284, American Concrete Institute, Farmington Hills, MI. Research on High-Strength Concrete,” High-Strength
(CD-ROM) Concrete, SP-87, H. G. Russell, ed., American Concrete
Hossain, T. R.; Vollum, R.; and Ahmed, S. U., 2011, Institute, Farmington Hills, MI, pp. 85-117.
“Deflection Estimation of Reinforced Concrete Flat Plates Nilson, A. H., and Walters Jr., D. B., 1975, “Deflection of
Using ACI Method,” ACI Structural Journal, V. 108, No. 4, Two-Way Floor Systems by the Equivalent Frame Method,”
July-Aug., pp. 405-413. ACI Journal Proceedings, V. 72, No. 5, May, pp. 210-218.
Jensen, V. P., 1938, “Solutions for Certain Rectangular Park, R., and Paulay, T., 1975, Reinforced Concrete Struc-
Slabs Continuous Over Flexible Supports,” Bulletin No. tures, John Wiley & Sons, Inc., New York, 769 pp.
303, University of Illinois Engineering Experiment Station, Ramsay, R. J.; Mina, S. A.; and Macgregor, J. G., 1979,
Urbana, IL. “Monte Carlo Study of Short Time Deflections of Rein-
Jokinen, E. P., and Scanlon, A., 1985, “Field-Measured forced Concrete Beams,” ACI Journal Proceedings, V. 76,
Two-Way Slab Deflections,” Proceedings, Annual Confer- No. 8, Aug., pp. 897-918.
ence, CSCE, SK, Canada, May. Rangan, B. V., 1976, “Prediction of Long-Term Deflec-
Kripanarayanan, K. M., and Branson, D. E., 1976, “Short- tions of Flat Plates and Slabs,” ACI Journal Proceedings, V.
Time Deflections of Flat Plates, Flat Slabs, and Two-Way 73, No. 4, Apr., pp. 223-226.
Slabs,” ACI Journal Proceedings, V. 73, No. 12, Dec., pp. Rangan, B. V., 1982, “Control of Beam Deflections by
686-690. Allowable Span-Depth Ratios,” ACI Journal Proceedings,
Lee, Y. H., and Scanlon, A., 2007, “Deflection Control of V. 79, No. 5, Sept.-Oct., pp. 372-377.
Concrete Structures Based on Utility Theory,” ACI Struc- Sbarounis, J. A., 1984, “Multi-Story Flat Plate Buildings:
tural Journal, V. 104, No. 1, Jan.-Feb., pp. 60-67. Construction Loads and Immediate Deflections,” Concrete
Liu, X.; Chen, W. F.; and Bowman, M. D., 1985, International, V. 6, No. 2, Feb., pp. 70-77.
“Construction Load Analysis for Concrete Structures,” Scanlon, A., and Bischoff, P. H., 2008, “Shrinkage
Journal of Structural Engineering, V. 111, No. 5, pp. 1019- Restraint and Loading History Effects on Deflection of Flex-
1036. doi: 10.1061/(ASCE)0733-9445(1985)111:5(1019) ural Members,” ACI Structural Journal, V. 105, No. 4, July-
Luebkeman, C. H.; Nilson, A. H.; and Slate, F. O., 1985, Aug., pp. 498-506.
“Sustained Load Deflection of High Strength Concrete Scanlon, A., and Lee, Y. H., 2006, “Unified Span-to-Depth
Ratio Equation for Nonprestressed Concrete Beams and

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 39

Slabs,” ACI Structural Journal, V. 103, No. 1, Jan.-Feb., pp. plier,” ACI Structural Journal, V. 116, No. 2, Mar.-Apr., pp.
142-148. 89-100.
Scanlon, A., and Murray, D. W., 1974, “Time Depen- Wang, C. K., and Salmon, C. G., 1992, Reinforced
dent Reinforced Concrete Slab Deflections,” Journal of the Concrete Design, fifth edition, Harper Collins, 1030 pp.
Structural Division, V. 100, pp. 1911-1924. Wight, J. K., and MacGregor, J. G., 2009, Reinforced
Scanlon, A., and Murray, D. W., 1982, “Practical Calcula- Concrete Mechanics and Design, fifth edition, Pearson Pren-
tion of Two-way Slab Deflections,” Concrete International, tice Hall, Upper Saddle River, NJ, 1112 pp.
V. 4, No. 11, Nov., pp. 43-50. Yu, W. W., and Winter, G., 1960, “Instantaneous and
Scanlon, A., and Suprenant, B., 2011, “Spreadsheet for Long-Term Deflections of Reinforced Concrete Beams
Estimating Two-way Slab Deflections Including Construc- Under Working Loads,” ACI Journal Proceedings, V. 57,
tion Load Effects,” Concrete International, V. 33, No. 7, No. 1, Jan., pp. 29-50.
July, pp. 29-34.
Scott, R. H., and Beeby, A. W., 2005, “Long Term Tension APPENDIX A—DEFLECTION DESIGN EXAMPLES
Stiffening Effects in Concrete,” ACI Structural Journal, V. Numerical examples are presented to demonstrate some of
102, No. 1, Jan.-Feb., pp. 31-39. the calculation procedures outlined in Chapter 4. Examples
Sonebi, M., and Bartos, P. J. M., 2001, “Performance of are presented for a simply supported slab and for a four-span
Reinforced Columns Cast with Self Compacting Concrete,” continuous beam, as well as an example for temperature-
Fifth CANMET/ACI Conference on Recent Advances in induced deflections.
Concrete Technology-Proceeding, SP-200, V. M. Malhotra,
ed., American Concrete Institute, Farmington Hills, MI, pp. Example A.1—Deflection of a simply supported
415-432. slab
Swamy, R. N., and Anand, K. L., 1973, “Shrinkage and This example compares short- and long-term deflec-
Creep of High Strength Concrete,” Civil Engineering and tion computed using both Branson’s (1963) and Bischoff’s
Public Works Review (London), V. 68, No. 807, Oct., pp. (2005, 2007) expression for Ie. Results are presented for a
859-865, 867-868. simply supported one-way slab with an 18 ft (5.5 m) span.
Tam, K. S. S., and Scanlon, A., 1986, “Deflection of The slab has a thickness of 10 in. (254 mm) that does not
Two-Way Slabs Subjected to Restrained Volume Change quite satisfy the ACI 318 minimum thickness requirements
and Transverse Loads,” ACI Journal Proceedings, V. 83, in Table 4.2.2a (ℓ/20 for a simply supported slab).
No. 5, Sept.-Oct., pp. 737-744. Loading consists of a dead load wD = 125 lb/ft2 (6 kPa)
Tan, K. H.; Paramasivam, P.; and Tan, K. C., 1994, from the member self-weight and live load wL = 100 lb/ft2
“Instantaneous and Long-Term Deflections of Steel Fiber (4.8 kPa). Part of the live load wL,sus = 20 lb/ft2 (1 kPa) is
Reinforced Concrete Beams,” ACI Structural Journal, V. 91, sustained. The concrete has a specified compressive strength
No. 4, July-Aug., pp. 364-393. of 4000 psi (27.6 MPa) and is reinforced with Grade 60
Thompson, D. P., and Scanlon, A., 1988, “Minimum (Grade 420) steel bars. Strength requirements are satisfied
Thickness Requirements for Control of Two-Way Slab using No. 4 (12.7 mm diameter) bars spaced at 7 in. (178
Deflections,” ACI Structural Journal, V. 85, No. 1, Jan.- mm) on center to give a reinforcing ratio ρ = 0.32 percent.
Feb., pp. 12-22. Maximum deflection is computed using the traditional
Timoshenko, S., and Woinowsky-Krieger, S., 1959, ACI 318 approach (4.4.1) and assumes the slab has been
Theory of Plates and Shells, McGraw-Hill Book Co., New preloaded up to the full dead plus live load during the
York, NY. construction period. Restraint to shrinkage is also consid-
Trost, H., 1967, “Auswirkungen des Superposition- ered by using a reduced cracking moment equal to two-
sprinzips auf Kriech- und Relaxationsprobleme bei Beton thirds Mcr ([Link]).
und Spann beton,” Beton- und Stahlbetonbau, V. 61, pp.
230-238. Solution:
Vanderbilt, M. D.; Sozen, M. A.; and Siess, C. P., 1965,
“Deflections of Multiple-Panel Reinforced Concrete Floor Material properties and bending moment values:
Slabs,” ASCE Proceedings, V. 91, No. ST4, Aug., pp. 77-101.
E= = 3.6 × 106 psi (24,900 MPa)
57,000 f c′ = 57,000 4000
Vecchio, F. J., and Collins, M. P., 1986, “The Modi- c

fied Compression-Field Theory for Reinforced Concrete

Elements Subjected to Shear,” ACI Journal Proceedings, V. Es = 29 × 106 psi (200,000 MPa)
83, 2, Mar.-Apr., pp. 219-231.
Vollum, R. L., 2011, “Influence of Construction Loading modular ratio n = Es/Ec = (29 × 106)/(3.6 × 106) = 8.04
on Long-Term Slab Deflection,” Andy Scanlon Symposium
on Serviceability and Safety of Concrete Structures, SP-284, modulus of rupture f r =
= 7.5 f c′ 7.5
= 4000 474 psi (3.3
American Concrete Institute, Farmington Hills, MI, pp. 1-22. MPa)
Walkup, S.; Musselman, E.; and Gross, S., 2019, “Effect
of Service Load Levels on Long-Term Deflection Multi- Positive moment at midspan = wℓ2/8 for a 1 ft (305 mm)
wide strip of slab:

American Concrete Institute – Copyrighted © Material – [Link]

40 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

MD = 125 × 182/8 = 5062.5 ft-lb = 60,800 in.-lb (6.87 kN-m)

ML = 100 × 182/8 = 4050 ft-lb = 48,600 in.-lb (5.49 kN-m)
MD + ML = 9112.5 ft-lb = 109,400 in.-lb (12.37 kN-m)

Gross section properties:

yt = h/2 = 10/2 = 5 in. (127 mm)
Ig = bh3/12 = 12 × 103/12 = 1000 in.4 (416.2 × 106 mm4)
Mcr = fr Ig/yt = 474 × 1000/5 = 94,800 in.-lb (10.7 kN-m)

Cracked section properties:

As = (12/7) × 0.20 = 0.343 in.2/ft (726 mm2/m) width of slab
(effective number of No. 4 [12.7 mm] bars in a 1 ft [305 mm] wide strip of slab = 12/7=1.71)
d = h – cover – db/2 = 10 – 0.75 – 0.5/2 = 9 in. (229 mm)
ρ = As/bd = 0.343/(12 × 9) = 0.317 percent
nρ = 8.04 × 0.317/100 = 0.0255

c
= ( ) (
(nρ) 2 + 2nρ − nρ=
d (0.0255) 2 + 2 × 0.0255 − 0.0255= )
9 1.82 in. (46 mm)

bc 3 1.823
I cr = + nAs (d − c) 2 =12 × + 8.04 × 0.343(9 − 1.82) 2 = 166 in.4 (69.1 × 106 mm 4 )
3 3

Ig/Icr = 6.0

Effective moment of inertia Ie for full dead plus live load (using Mcr):
Ma = MD + ML = 109,400 in.-lb (12.37 kN-m)
Mcr/Ma = 94,800/109,400 = 0.867
(Mcr/Ma)2 = (0.867)2 = 0.752
(Mcr/Ma)3 = (0.867)3 = 0.652
1 – Icr/Ig = 1 – 166/1000 = 0.834
Ie,Branson = (Mcr/Ma)3Ig + [1 – (Mcr/Ma)3] Icr = 0.652 × 1000 + (1 – 0.652)166 = 710 in.4 (296 × 106 mm4)
Ie,Bischoff = Icr/[1 – (Mcr/Ma)2(1 – Icr/Ig)] = 166/(1 – 0.752 × 0.834) = 445 in.4 (185 × 106 mm4)

The value for Ie is approximately 40 percent less when computed using Bischoff’s expression.

Effective moment of inertia Ie for full dead plus live load (using 0.67Mcr to account for shrinkage restraint):
Ma = MD + ML = 109,400 in.-lb (12.37 kN-m)
Mcr/Ma = 0.67 × 0.867 = 0.581
(Mcr/Ma)2 = (0.581)2 = 0.338
(Mcr/Ma)3 = (0.581)3 = 0.196
Ie,Branson = (Mcr/Ma)3Ig + [1 – (Mcr/Ma)3]Icr = 0.196 × 1000 + (1 – 0.196)166 = 329 in.4 (137 × 106 mm4)
Ie,Bischoff = Icr/[1 – (Mcr/Ma)2(1 – Icr/Ig)] = 166/(1 – 0.338 × 0.834) = 231 in.4 (96.2 × 106 mm4)

Ie decreases by approximately one-half when using the reduced cracking moment. This is expected because tension stiffening
has a significant influence on member stiffness whenever Ma ≤ 2Mcr.

Calculation of deflection using Mcr:

Immediate deflection due to live load (using Mcr):

Using Branson’s expression for Ie gives

5M L  2 5 × 48, 600 × (18 × 12) 2
=
∆L = = 0.092 in. (2.34 mm)
48 Ec I e 48 ×3.6 × 106 × 710
 18 × 12
∆ L 0.092 in. (2.34 mm) ≤ =
= = 0.6 in. (15.2 mm) for floors ∴ OK
360 360

Using Bischoff’s expression for Ie gives

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 41

5M L  2 5 × 48, 600 × (18 × 12) 2

=
∆L = = 0.147 in. (3.73 mm)
48 Ec I e 48 × 3.6 × 106 × 445
 18 × 12
∆ L 0.147 in. (3.73 mm) ≤ =
= = 0.6 in. (15.2 mm) for floors ∴ OK
360 360

Immediate (short-term) deflection under sustained loads (using Mcr):

Msus = MD + 0.2ML = 60,800 + 0.2 × 48,600 = 70,520 in.-lb (7.97 kN-m)

Using Branson’s expression for Ie gives

5M sus  2 5 × 70,520 × (18 × 12) 2
=
∆ sus = = 0.134 in. (3.40 mm)
48 Ec I e 48 × 3.6 × 106 × 710

Using Bischoff’s expression for Ie gives

5M sus  2 5 × 70,520 × (18 × 12) 2
=
∆ sus = = 0.213 in. (5.41 mm)
48 Ec I e 48 × 3.6 × 106 × 445

Long-term and incremental deflection (using λ∆ = 2 and Mcr):

Using Branson’s expression for Ie gives

∆inc = λ∆∆sus + ∆L,add = 2.0 × 0.134 + 0.8 × 0.092 = 0.341 in. (8.66 mm)

 18 × 12
∆=
inc 0.341 in. (8.66 mm) ≤ = = 0.9 in. (22.9 mm) ∴ OK
240 240

Using Bischoff’s expression for Ie gives

∆inc = λ∆∆sus + ∆L,add = 2.0 × 0.213 + 0.8 × 0.147 = 0.543 in. (13.8 mm)
 18 × 12
∆
= inc 0.543 in. (13.8 mm) ≤ = = 0.9 in. (22.9 mm) ∴ OK
240 240
Deflection calculated using Bischoff’s expression for Ie gives approximately 60 percent more deflection than the value
computed with Branson’s approach. The immediate (ℓ/360) and long-term (ℓ /240) deflection requirements are satisfied with
both approaches for this example when the full cracking moment Mcr is used in deflection calculations.

Calculation of deflection using 0.67Mcr:

Immediate deflection due to live load (using 0.67Mcr):

Using Branson’s expression for Ie gives

5M L  2 5 × 48, 600 × (18 × 12) 2
=
∆L = = 0.199 in. (5.05 mm)
48 Ec I e 48 × 3.6 × 106 × 329
 18 × 12
∆ L 0.199 in. (5.05 mm) ≤ =
= = 0.6 in. (15.2 mm) for floors ∴OK
360 360

Using Bischoff’s expression for Ie gives

5M L  2 5 × 48, 600 × (18 × 12) 2
=
∆L = = 0.283 in. (7.19 mm)
48 Ec I e 48 × 3.6 × 106 × 231
 18 × 12
∆ L 0.283 in. (7.19 mm) ≤ =
= = 0.6 in. (15.2 mm) for floors ∴OK
360 360

American Concrete Institute – Copyrighted © Material – [Link]

42 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Short-term deflection under sustained loads (using 0.67Mcr):

Using Branson’s expression for Ie gives

5M sus  2 5 × 70,520 × (18 × 12) 2
=
∆ sus = = 0.289 in. (7.35 mm)
48 Ec I e 48 × 3.6 × 106 × 329

Using Bischoff’s expression for Ie gives

5M sus  2 5 × 70,520 × (18 × 12) 2
=
∆ sus = = 0.412 in. (10.47 mm)
48 Ec I e 48 × 3.6 × 106 × 231

Long-term and incremental deflection (using λ∆ = 2 and 0.67Mcr):

Using Branson’s expression for Ie gives

∆inc = λ∆∆sus + ∆L,add = 2.0 × 0.289 + 0.8 × 0.199 = 0.738 in. (18.75 mm)
 18 × 12
∆=
inc 0.738 in. (18.75 mm) ≤ = = 0.9 in. (22.9 mm) ∴ OK
240 240

Using Bischoff’s expression for Ie gives

∆inc = λ∆∆sus + ∆L,add = 2.0 × 0.412 + 0.8 × 0.283 = 1.05 in. (26.7 mm)
 18 × 12
∆=inc 1.05 in. (26.7 mm) ≤ = = 0.9 in. (22.9 mm) ∴ N.G.
240 240
Using a reduced cracking moment equal to 0.67Mcr doubles the computed value of deflection for this example. Deflection
calculated using Bischoff’s expression for Ie gives approximately 40 percent more deflection than the value computed with
Branson’s approach. Long-term deflection requirements for this slab are no longer satisfied using Bischoff’s expression for Ie,
and a slightly thicker slab of 10.5 in. (267 mm) thickness is required.

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 43

Example A.2—Age-adjusted deflection of simply supported slab

Deflection calculations for the slab in Example A.1 are repeated to estimate the long-term age-adjusted values due to creep of
the concrete. For this example, assume that the slab remains fully shored (that is, completed unloaded) until an age of 28 days
after casting, at which time the loading becomes the member self-weight and the full service live load of wL = 100 lb/ft2 (4.8 kPa),
which is sustained. The concrete has achieved a compressive strength of 4000 psi (27.6 MPa) at 28 days and, for simplicity
in this example, further maturity of the concrete is assumed to have ceased at this same age. The total long-term deflection at
an age 5 years later is to be determined using the age-adjusted modulus method described in 3.3.3. Use the approximation for
Ie using Bischoff’s expression (Eq. (4.3.2.2a)) and a cracking moment based on (2/3)Mcr to account for restrained shrinkage.

Solution:
Similar to Example A.1, the relevant loads and section properties can be determined as follows:

Short-term material properties:

Ec = 57,000 f c′ = 57,000 4000 = 3.6 × 106 psi (24,900 MPa)
Es = 29 × 106 psi (200,000 MPa)

The total positive moment at midspan = wℓ2/8 for a 1 ft (300 mm) wide strip of slab

MD + ML = 9112.5 ft-lb = 109,400 in.-lb (12.37 kN-m)

Creep coefficient and age-adjusted modulus

From Eq. (3.3.3a), the creep coefficient under standard conditions described in 3.3 at t = 1825 (5 years) can be estimated as
 t 0.6 
Ct 
= × 2.35=
× 1.0 2.116
 10 + t 0.6 

Using a typical aging coefficient value of χ = 0.8, the age adjusted modulus from Eq. (3.3.3d) becomes
Ec (28) 3.6 × 106
Ec (1825, 28)
= = = 1.34 × 106 psi (9220 MPa)
1 + χCt (1825, 28) 1 + 0.8 × 2.116

The modular ratio n and Icr associated with the aged adjusted modulus are
n =Es /Ec =(29 × 106 )/ (1.34 × 106 ) =21.7
nρ = 21.7 × 0.317/100 = 0.069

c
= ( (nρ) 2 + 2nρ − nρ=
d) ( )
(0.069) 2 + 2 × 0.069 − 0.069=
9 2.78 in. (70.6 mm)

Icr = bc3/3 + nAs(d – c)2 = 12 × 2.783/3 + 21.7 × 0.343(9 – 2.78)2 = 373 in.4 (155.3 × 106 mm4)
Ig/Icr = 2.68

The age-adjusted effective moment of inertia Ie for full dead plus live load using Bischoff’s expression and using 0.67Mcr to
account for shrinkage restraint
I cr 373
I e , Bischoff
= = = 473 in.4 (196.9 × 106 mm 4 )
  M   I   1 − 0.338 × 0.627
2

1 −  cr  1 − cr  
M Ig  
  a   
The corresponding total deflection at midspan becomes
5 M D+ L ⋅ 2 5 109,400 ⋅ (18 × 12) 2
∆ D+ L = ⋅ = ⋅ = 0.84 in. (21.3 mm)
48 ( Ec I e ) 48 (1.34 × 106 × 473)

Thus, the age adjusted deflection is approximately 1.3 times larger than the short-term (instantaneous) deflection value of
0.64 in. (16.2 mm) from Example A.1.

American Concrete Institute – Copyrighted © Material – [Link]

44 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Example A.3—Short- and long-term deflection of a four-span continuous beam

A reinforced concrete beam supporting a 4 in. (100 mm) slab is continuous over four equal spans ℓ = 36 ft (11 m) as shown
in Fig. A.3a (Nawy 2009). It is subjected to a uniformly distributed load wD = 700 lb/ft (10.22 kN/m), including its self-weight
and a service load wL = 1200 lb/ft (17.52 kN/m). The beam has the dimensions b = 14 in. (356 mm), d = 18.25 in. (464 mm) at
midspan, and a total thickness h = 21.0 in. (533 mm). The end span is reinforced with four No. 9 (28.6 mm diameter) bars at
midspan at the bottom fibers and six No. 9 (28.6 mm diameter) bars at the top fibers of the support section (at the first interior
support). Use the traditional ACI 318 method for long term effect discussed in 4.4.1.
Calculate the maximum deflection of the continuous beam using the ACI 318 method.

Given:
fc′ = 4000 psi (27.6 MPa), normalweight concrete
fy = 60,000 psi (414 MPa)

50 percent of the live load is sustained for 36 months on the structure.

Fig. A.3a—Details of continuous beam (Nawy 2009; image courtesy of Pearson/Prentiss Hall). (Note: 1 in. = 25.4 mm; 1 lb/
ft = 14.6 N/m.)

Solution—ACI method:
Note: All calculations are rounded to three significant figures.

Material properties and bending moment values

Ec = 57,000 f c′ = 57,000 4000 = 3.6 × 106 psi (24,900 MPa)

Es = 29 × 106 psi (200,000 MPa)
modular ratio n = Es/Ec = (29 × 106)/(3.6 × 106) = 8.1
modulus of rupture fr = 7.5 f c′ = 7.5 4000 = 474 psi (3.3 MPa)
For the end span, the positive moment = 0.0772wℓ2

+MD = 0.0772 × 700(36.0)2 × 12 = 840,000 in.-lb (94.9 kN-m)

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 45

+ML = 0.0772 × 1200(36.0)2 × 12 = 1,440,000 in.-lb (162.7 kN-m)

+(MD + ML) = 0.0772 × 1900(36.0)2 × 12 = 2,280,000 in.-lb (257.7 kN-m)
For the negative moment at the first interior support = 0.1071wℓ2
–MD = 0.1071 × 700(36.0)2 × 12 = 1,170,000 in.-lb (132.2 kN-m)
–ML = 0.1071 × 1200(36.0)2 × 12 = 2,000,000 in.-lb (226.1 kN-m)
–(MD + ML) = 0.1071 × 1900(36.0)2 × 12 = 3,170,000 in.-lb (358.3 kN-m)

Effective moment of inertia Ie

Figure A.3b shows the theoretical midspan and support cross sections to be used for calculating the gross moment of inertia Ig.

Fig. A.3b—Gross moment of interia Ig cross sections. (Note: 1 in. = 25.4 mm; 1 in.2 = 645 mm2.)

1. Midspan section:

width of T-beam flange = bw + 16hf = 14.0 + 16 × 4.0 = 78 in. (1981 mm)

Uncracked section properties

Depth from compression flange to the elastic centroid is
78(4 × 2) + 14 × (21 − 4) × 12.5
( A1 y1 + A2 y2 )/( A1 + A2 )
y= 6.54 in. (166.1 mm)
=
78 × 4 + 14 × 17
yt = h – ȳ = 21.0 – 6.54 = 14.5 in. (368.3 mm)
Ig = [78(4)3]/12 + 78 × 4(6.54 – 4/2)2 + [14(21 – 4)3]/12] + 14(21 – 4)(14.5 – [21 – 4]/2)2 = 21,000 in.4 (8.741 × 109 mm4)
Mcr = frIg /yt = (474 × 21,000)/14.5 = 686,000 in.-lb (77.5 kN-m)

Cracked section properties

Depth of centroidal axis (As = four No. 9 [28.6 mm] bars = 4.0 in.2 [2580 mm2]):
The location c of the centroidal axis for the cracked section is obtained by taking the moment of area of the transformed
section about the top (compression) face. Assuming c lies within the flange thickness,
bf c2/2 + nAsc – nAsd = 0
78c2/2 + 8.1 × 4 × c – 8.1 × 4 × 18.25 = 0

This gives c = 3.5 in. ≤ hf = 4 in. (89 mm ≤ 102 mm)

Hence, the centroidal axis is inside the flange and the flange section is analyzed as a cracked rectangular section.

Icr = [78(3.5)3/3] + 8.1 × 4(18.25 – 3.5)2 = 8160 in.4 (3.40 × 109 mm4)

American Concrete Institute – Copyrighted © Material – [Link]

46 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Mcr/Ma ratio for dead load, dead plus sustained live load, and dead plus live load:
Mcr/MD = 686,000/840,000 = 0.82
Mcr/MD+0.5L = 686,000/(840,000 + 0.5 × 1,440,000) = 0.44
Mcr/MD+L = 686,000/2,280,000 = 0.30

Effective moment of inertia for midspan section:

Ie = (Mcr/Ma)3Ig + [1 – (Mcr/Ma)3]Icr ≤ Ig
Ie,D = 0.55 × 21,000 + 0.45 × 8160 = 15,200 in.4 (6.33 × 109 mm4)
Ie,D+0.5L = 0.085 × 21,000 + 0.915 × 8160 = 9250 in.4 (3.85 × 109 mm4)
Ie,D+L = 0.027 × 21,000 + 0.973 × 8160 = 8500 in.4 (3.54 × 109 mm4)

2. Support section:
Uncracked section properties
Ig = bwh3/12 = 14(21)3/12 = 10,800 in.4 (4.50 × 109 mm4)
yt = 21/2 = 10.5 in. (266.7 mm)
474 × 10.80
Mcr = frIg/yt = = 488,000 in.-lb (55.2 kN-m)
10.5
Cracked section properties

Depth of centroidal axis:

As = six No. 9 (28.6 mm) bars = 6.0 in.2 (3870 mm2)
As′ = two No. 9 (28.6 mm) bars = 2.0 in.2 (1290 mm2)
d = 21.0 – 3.75 = 17.25 in. (438 mm)

A similar calculation for the centroidal axis depth c gives a value c = 7.58 in. (192.5 mm)

Icr = bc3/3 + nAs(d – c)2 + (n – 1)As′(c – d′)2 = 6900 in.4 (2.87 × 109 mm4)

Mcr/Ma ratio for dead load, dead plus sustained live load, and dead plus live load
Mcr/MD = 488,000/1,170,000 = 0.42
Mcr/MD+0.5L = 488,000/(1,170,000 + 0.5 × 2,000,000) = 0.225
Mcr/MD+L = 488,000 / 3,170,000 = 0.154

Effective moment of inertia for support section:

Ie,D = 0.07 × 10,800 + 0.93 × 6900 = 7170 in.4 (2.98 × 109 mm4)
Ie,D+0.5L = 0.01 × 10,800 + 0.99 × 6900 = 6940 in.4 (2.89 × 109 mm4)
Ie,D+L = 0.003 × 10,800 + 0.997 × 6900 = 6910 in.4 (2.88 × 109 mm4)

3. Average effective Ie for continuous span:

average: Ie = 0.85Im + 0.15Iec
Ie,D = 0.85 × 15,200 + 0.15 × 7170 = 14,000 in.4 (5.83 × 109 mm4)
Ie,D+0.5L = 0.85 × 9250 + 0.15 × 6940 = 8900 in.4 (3.70 × 109 mm4)
Ie,D+L = 0.85 × 8500 + 0.15 × 6910 = 8260 in.4 (3.44 × 109 mm4)

4. Deflection calculations
Short-term deflection (maximum deflection of end span):

∆ = 0.0065wℓ4/EI

For all practical purposes, assume the span length ℓ equals the clear span ℓn
0.065w(36 × 12) 4 1 w  9 w 
∆ = 6
× 5.240 in.  3.822 × 10 I mm
3.6 × 10 × I e 12 Ie e

Initial dead-load deflection:

∆D = 5.240(700)/14,000 = 0.26 in. (6.6 mm)

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 47

Initial live-load deflection (∆L = ∆L+D – ∆D):

5.240(1900) 5.240(700)
∆ L= − = 1.21 − 0.26= 0.95 in. (24.1 mm)
8260 14, 000

Initial 50 percent sustained live-load deflection (∆L,S = ∆0.5L+D – ∆D):

5.240(1300) 5.240(700)
∆ L,S = − = 0.77 − 0.26= 0.51 in. (12.9 mm)
8900 14, 000

Long-term deflection:
ρ′ = As′/bd = 0 (at midspan in this case)
multiplier λ∆ = ξ/(1 + 50ρ′)

From Fig. 4.4.1,

ξ = 1.75 for 36-month sustained load
ξ = 2.0 for 5-year loading

Therefore, λ∞ = 2.0 and λt = 1.75

The incremental deflection (∆inc = λ∞∆D + λt∆LS + ∆L) is

∆inc = 2.0 × 0.26 + 1.75 × 0.51 + 0.95 = 1.41 + 0.95 = 2.36 in. (59.9 mm)

Deflection requirements (Table 4.2.2c)

ℓ/180 = (36 × 12)/180 = 2.4 in. (61.0 mm) > ∆L = 0.95 in. (24.1 mm) (OK)
ℓ /360 = 1.2 in. (30.5 mm) > ∆L = 0.95 in. (24.1 mm) (OK)
ℓ/480 = 0.9 in. (22.9 mm) < ∆LT = 2.36 in. (59.9 mm) (N.G.)
ℓ/240 = 1.8 in. (45.7 mm) < ∆LT = 2.36 in. (59.9 mm) (N.G.)

Hence, the continuous beam is limited to floors or roofs not supporting or attached to nonstructural elements such as partitions.
For the alternative, compute the effective moment of inertia Ie as described in [Link]:
I cr
Ie ≤ Ig
1 − ( M cr /M a ) 2 (1 − I cr /I g )

Midspan section:
1 – Icr/Ig = 1 – 8160/21,000 = 0.61
Ie,D = 8160/(1 – 0.67 × 0.61) = 13,800 in.4 (5.74 × 109 mm4) (9 percent smaller)
Ie,D+0.5L = 8160/(1 – 0.19 × 0.61) = 9230 in.4 (3.84 × 109 mm4) (within 1 percent)
Ie,D+L = 8160/(1 – 0.09 × 0.61) = 8630 in.4 (3.59 × 109 mm4) (1.5 percent larger)

Support section:
1 – Icr/Ig = 1 – 6900/10,800 = 0.36
Ie,D = 6900/(1 – 0.18 × 0.36) = 7380 in.4 (3.07 × 109 mm4) (3 percent larger)
Ie,D+0.5L = 6900/(1 – 0.05 × 0.36) = 7020 in.4 (2.92 × 109 mm4) (1 percent difference)
Ie,D+L = 6900/(1 – 0.02 × 0.36) = 6950 in.4 (2.89 × 109 mm4) (0.5 percent difference)

Average effective Ie for continuous span (Ie = 0.85Im + 0.15Iec):

Ie,D = (0.85 × 13,800) + (0.15 × 7380) = 12,800 in.4 (5.33 × 109 mm4)
Ie,D+0.5L = (0.85 × 9230) + (0.15 × 7020) = 8900 in.4 (3.70 × 109 mm4)
Ie,D+L = (0.85 × 8630) + (0.15 × 6950) = 8380 in.4 (3.49 × 109 mm4)

Midspan deflection:
∆D = 5.240(700)/12,800 = 0.29 in. (7.4 mm)
∆L = 5.240(1900)/8380 – 5.240(700)/12,800 = 1.19 – 0.29 = 0.90 in. (22.9 mm)
∆L,S = 5.240(1300)/8900 – 5.240(700)/12,800 = 0.77 – 0.29 = 0.48 in. (12.2 mm)
∆inc = 2.0 × 0.29 + 1.75 × 0.48 + 0.90 = 1.42 + 0.90 = 2.32 in. (58.9 mm)

American Concrete Institute – Copyrighted © Material – [Link]

48 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

There is very little difference between the two approaches for Ie used to compute deflection of the beam end span. Simply
using the cracked moment of inertia at midspan (Icr = 8160 in.4 [3.40 × 109 mm4]) for this example gives ∆inc = 2.35 in. (59.7 mm),
which is also not very different from the computed value using Ie.

Example A.4—Temperature-induced deflections

These design examples illustrate the calculation procedures for temperature-induced deflections.

Example (a): Simply supported vertical wall panel with linear temperature gradient through the wall thickness
∆T = 40°F (22.2°C)
α = 5.5 × 10–6 in./in./°F (9.9 × 10–6 mm/mm/°C)
h = 4 in. (101 mm)

a) Single-story span: ℓ = 12 ft (3.66 m)

∆ = (5.5 × 10-6 × 40 × 1442)/(4 × 8) = 0.14 in. (3.6 mm)

b) Two-story span: ℓ = 24 ft (7.32 m)

∆ = (5.5 × 10-6 × 40 × 2882)/(4 × 8) = 0.57 in. (14.5 mm)

Example (b): Simply supported T-section with linear temperature gradient over depth
∆T = 40°F (22.2°C)
α = 5.5 × 10-6 in./in./°F (9.9 × 10-6 mm/mm/°C)
h = 36 in. (914 mm)
ℓ = 60 ft (18.3 m) simply supported
∆ = (5.5 × 10–6 × 40 × 7202)/(36 × 8) = 0.40 in. (10 mm)

Example (c): Simply supported T-section with constant temperature increase over flange depth, web temperature
remains constant
I = 69,319 in.4 (2.88 × 1010 mm4)
yt = 26.86 in. (682 mm)
∆T = 40°F (22.2°C)
α = 5.5 × 10–6 in./in./°F (9.9 × 10–6 mm/mm/°C)
h = 36 in. (914 mm)
hf = 3 in. (76 mm)
b = 96 in. (2438 mm)
ℓ = 60 ft (18.3 m)
36 36
κ = (α/I) ∫ T ( y ) b ( y ) ( y – yt )dy = (α/I) ∫ (40 × 96)( y − 26.86)dy
0 33

= (88,013 × 5.5 × 10-6)/69,319 = 6.98 × 10-6 1/in. (2.75 × 10-7 1/mm)

∆ = (κℓ2)/8 = (6.98 × 10-6 × 7202)/8 = 0.45 in. (11.4 mm)

APPENDIX B—TWO-WAY SLAB DEFLECTION EXAMPLES

Example B.1—Deflection design example for long-term deflection of a two-way slab

The following example (Nawy 2009) illustrates the application of the equivalent frame approach developed by Nilson and
Walters (1975), along with the modulus of rupture and long-term multiplier given in ACI 318.
A 7 in. (178 mm) slab of a five panel by five panel floor system spanning 25 ft (7.62 m) in the E-W direction and 20 ft (6.10
m) in the N-S direction is shown in Fig. B.1. The panel is monolithically supported by beams 15 x 27 in. (381 x 686 mm) in the
E-W direction and 15 x 24 in. (381 x 610 mm) in the N-S direction. The floor is subjected to a time-dependent deflection due
to an equivalent uniform working (service) load intensity w = 450 lb/ft2 (21.5 kPa).
Material properties of the floor are:
fc′ = 4000 psi (27.6 MPa)
fy = 60,000 psi (414 MPa)
Ec = 3.6 × 106 psi (24,900 MPa)
Assume:
1. Net moment from adjacent spans
E-W: Support 1: 20 × 103 (27.1 × 103); Support 2: 5 × 103 ft-lb (6.8 × 103 N-m)

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 49

N-S: Support 1: 40 × 103 (54.2 × 103); Support 4: 20 × 103 ft-lb (27.1 × 103 N-m)
2. Equivalent column stiffness, Kec = 400Ec lb-in. per radian (6555Ec N-m/rad) in both directions.
3. Cracked moment of inertia:
E-W: Icr = 45,500 in.4 (18.94 × 109 mm4)
N-S: Icr = 32,500 in.4 (13.53 × 109 mm4)
Find the maximum central deflection of the panel due to the long-term loading and determine if its magnitude is acceptable
if the floor supports sensitive equipment, which can be damaged by large deflections.

Fig. B.1—Long-term deflection of two-way multi-panel slab on beams, equivalent frame calculation method (Nawy 2009;
image courtesy of Prentice Hall). (Note: 1 in. = 25.4 mm; 1 ft = 0.305 m.)

Solution:
Note: All calculations are rounded to three significant figures.
Calculate the gross moments of inertia (in.4 [mm4]) of the sections in Fig. B.1, namely, the total equivalent frame Ics in Part
(b), the column strip beam Ic in Part (c), and the middle strip slab Is in Part (d). These variables are:

American Concrete Institute – Copyrighted © Material – [Link]

50 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Ics, in.4 Ic, in.4 Is, in.4 Icr, in.4

E-W 63,600 53,700 3430 45,500
N-S 47,000 40,000 4290 32,500
Note: 1 in.4 = 416,230 mm4.

Next, calculate factors α1ℓ2/ℓ1 and α2ℓ1/ℓ2. In both cases, they are greater than 1.0. Hence, the factored moments coefficients
(percent) obtained from the tables in 8.10 of ACI 318-14.
Column strip (+ and –) Middle strip (+ and –)
E-W 81.0 19.0
N-S 67.5 32.5

E-W direction deflections (span = 25 ft [7.62 m])

Long-term w = 450 lb/ft2 (21.55 kPa)

450 × 20(25) 4 × 123
∆ 25
′ = 0.069 in. (1.8 mm)
384 × 3.6 × 106 × 63, 600
63, 600
∆ c 0.069 × 0.81
= = 0.066 in. (1.7 mm)
53, 700
63, 600
∆ s 0.069 × 0.19 ×
= = 0.243 in. (6.2 mm)
3430

Rotation at End 1 is
M1 20 × 103 × 12
θ=
1 = = 1.67 × 10−4 rad
K ec 400 × 3.6 × 106

and the rotation at End 2 is

M2 5 × 103 × 12
θ=
2 = = 0.42 × 10−4 rad
K ec 400 × 3.6 × 106

where θ is the rotation at one end if the other end is fixed.

∆″ is the deflection adjustment due to rotation at supports 1 and 2 = θℓ/8

∆″ = [(1.67 + 0.42) × 10-4 × 300]/8 = 0.008 in. (0.2 mm)

Therefore,
net ∆cx = 0.066 + 0.008 = 0.074, say, 0.07 in. (1.8 mm)
net ∆sx = 0.243 + 0.008 = 0.251, say, 0.25 in. (6.4 mm)

N-S direction deflections (span = 20 ft [6.10 m])

450 × 25(20) 4 × 123
∆ 20
′ = 0.048 in. (1.2 mm)
384 × 3.6 × 106 × 47, 000
47,000
∆ c 0.048 × 0.675 ×
= = 0.038 in. (0.96 mm)
40,000
47,000
∆ s 0.048 × 0.325 ×
= = 0.171 in. (4.3 mm)
4290
M1 40 × 103 × 12
rotation θ1= = = 3.3 × 10−4 rad
K ec 400 × 3.6 × 106
M4 20 × 103 × 12
rotation θ=
4 = = 1.67 × 10−4 rad
K ec 400 × 3.6 × 106

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 51

n θ 2 (3.3 + 1.67)10−4 × 240

∆= = = 0.015 in. (0.38 mm)
8 8
Therefore,
net ∆cy = 0.038 + 0.015 = 0.053 say 0.05 in. (1.3 mm)
net ∆sy = 0.171 + 0.015 = 0.186 say 0.19 in. (4.8 mm)
total central deflection ∆ = ∆sx + ∆cy + ∆cx
∆E-W = ∆sx + ∆cy = 0.25 + 0.05 = 0.30 in. (7.6 mm)
∆N-S = ∆sy + ∆cx = 0.19 + 0.07 = 0.26 in. (6.6 mm)

Hence, the average deflection at the center of the interior panel 1/2(∆E-W + ∆N-S) = 0.28 in. (7.1 mm).

Adjustment for cracked section:

Use Branson’s effective moment of inertia equation,

3
 M cr    M 3
=Ie  I g + 1 −  cr   I cr
 M a    M a  

Calculation of ratio Mcr/Ma

Mcr = frIg/yt
where fr is the modulus of rupture of concrete; yt is the distance of center of gravity of section from outer tension fibers; E-W
(240 in. [6096 mm] flange width): yt = 21.5 in. (546 mm); N-S (300 in. [7620 mm] flange width): yt = 19.2 in. (488 mm).

fr = 7.5 f c′ = 7.5 4000 = 474 psi (3.3 MPa)

Hence,
474 × 63,600 1
M cr (E-W) = × = 1.17 × 105 ft-lb (159 kN-m)
21.5 12
474 × 47,000 1
M cr (N-S) = × = 0.97 × 105 ft-lb (132 kN-m)
19.2 12
interior panel Ma = wsℓ2/16 = [20 × 450(25)2]/16 for E-W
= 3.52 × 105 ft-lb (477 kN-m)
= [25 × 450(20)2]/16 for N-S
= 2.81 × 105 ft-lb (381 kN-m)
Note that the moment factor 1/16 is used to be on the safe side, although the actual moment coefficients for two-way action
would have been smaller.
E-W effective moment of inertia Ie
Mcr/Ma = (1.17 × 105)/(3.52 × 105) = 0.332
(Mcr/Ma)3 = 0.037
Ie = 0.037 × 63,600 + (1 – 0.037)45,500 = 46,200 in.4 (19.2 × 109 mm4)

N-S effective moment of inertia Ie

Mcr/Ma = (0.97 × 105)/(2.81 × 105) = 0.345
(Mcr/Ma)3 = 0.041
Ie = 0.041 × 47,000 + (1 – 0.041)32,500 = 33,100 in.4 (13.8 × 109 mm4)

Adjusted central deflection for cracked section effect = 1.40 × 0.28 = 0.39 in. (10 mm)
ℓ/∆ = [25 × 12]/0.39 = 769 > 480 allowed in Table 4.2.2c.

Hence, the long-term central deflection is acceptable.

American Concrete Institute – Copyrighted © Material – [Link]

 52 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Example B.2—Deflection calculation for a flat plate using the crossing beam method
An edge panel of 6 in. (150 mm) flat plate with multiple panels in each direction is shown in Fig. B.2.
The plate is supported on 16 × 16 in. (406 × 406 mm) columns. The slab is designed for an unfactored live load of 60 lb/ft2
(2.87 kPa) in addition to its self-weight of 75 lb/ft2 (3.59 kPa). Assume that the slab is subjected to significant in-plane restraint.
Check the live load deflection and incremental deflection at midpanel if nonstructural components are installed 1 month after
removal of shoring.
Material properties are:
fc′ = 3000 psi (20.7 MPa)
fy = 60,000 psi (414 MPa)
Ec = 3.12 × 106 psi (21,500 MPa)

Fig. B.2—Plan of flat plate edge panel, beam crossing calculation method. (Note: 1 in. = 25.4 mm; 1 ft = 0.305 m.)

Using Eq. (5.2.1.2a), deflection of column and middle strips can be obtained from
5  2n
=δ [ M m + 0.1( M 1 + M 2 )]
48 Ec I e
in which moments and Ie are computed for a strip of unit width.

The midpanel deflection is computed as the sum of the column strip deflection in the N-S direction and the middle strip
deflection in the E-W direction. Moments at unfactored load level due to dead plus live load are given in Table B.2.

Table B.2—Calcualtion of Ie for column and middle strips

N-S column strip E-W middle strip
Ext. Neg. (M1) Pos. (Mm) Int. Neg. (M2) Neg. (M1) Pos. (Mm) Neg. (M2)
Moments (Ma), ft·kip/ft 2.45 2.93 4.94 0.72 (<Mcr) 0.63 (<Mcr) 0.72 (<Mcr)
(Mcr/Ma)3 0.154 0.09 0.02 — — —
1 – (Mcr/Ma) 3
0.846 0.91 0.98 — — —
Icr, in. 4
26.8 34.3 46.7 — — —
Ie, in.4
56.0 50.6 50.1 216 (= Ig) 216 (= Ig) 216 (= Ig)
Ie, average 52.2 216
Note: 1 in.4 = 416,230 mm4; 1 ft·kip/ft = 4.448 kN-m/m.

Cracking moment (Mcr)

fr = 4 f c′ (significant restraint) = 219 psi (1.51 MPa)
Mcr = frIg/yt
Ig = 1/12(12)(6)3 = 216 in.4 (89.9 × 106 mm4)
yt = 6/2 = 3 in. (76 mm)

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 53

Mcr = [(219)(216)/3](1/12,000) = 1.314 ft·kip/ft (5.844 kN-m/m)

Effective moment of inertia Ie

M    M 3
=I e  cr  I g + 1 −  cr   I cr
 Ma    M a  

(Refer to Table B.2 for tabulated values)

Average Ie for column strip = 52.2 in.4 (21.7 × 106 mm4)

Average Ie for middle strip = 216 in.4 (89.9 × 106 mm4)

Column strip deflection

(5) (16.7 × 12) 2
∆c × [2.93 + 0.1(−2.45
= − 4.94)](12,000) 0.67 in. (17.0 mm)
(48) (3,120,000)(52.2)

Middle strip deflection

(5) (12.7 × 12) 2
=∆m × [ 0.63 + 0.1(−0.72
= − 0.72) ] (12,000) 0.02 in. (0.51 mm)
(48) (3,120,000)(216)

Mid-panel deflection

∆ = ∆c + ∆m = 0.69 in. (17.5 mm)

Live load deflection

 wL  (60)
=∆L  = 0.69 = (0.69) 0.31 in. (7.9 mm)
 wD + wL  (135)

Span length on diagonal

= 16.7 2 + 12.7 2 = 20.94 ft = 251 in. (6380 mm)

Permissible live load deflection

= ℓ/360 = 251/360
= 0.70 in. (17.8 mm)
> 0.31 in. (7.9 mm) … OK for short-term deflection

Incremental deflection
Use long-term multiplier = 2.5 applied to sustained load deflection.
Assume sustained load = 75 + 20 = 95 lb/ft2 (4.55 kPa)
Additional long-term deflection = (2.5)(0.49) = 1.23 in. (31.2 mm)
Long-term deflection at 1 month = (0.5/2.0)(1.23) = 0.31 in. (7.9 mm)
Incremental deflection = 1.23 – 0.31 = 0.92 in. (23.4 mm)
Additional live load deflection = (40/135)(0.69) = 0.20 in. (5.1 mm)
Total = 1.12 in. (28.4 mm)
Permissible deflection = ℓ/480 = 251/480 = 0.53 in. (13.5 mm) < 1.12 in. (28.4 mm) ... N.G. for long-term deflection.

Hence, camber the slab or revise the design if nonstructural components are supported.

American Concrete Institute – Copyrighted © Material – [Link]

54 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

Example B.3—Minimum thickness calculation

The following example (Scanlon and Lee 2006) presents an alternative to the ACI 318 provisions for minimum thickness of
a two-way system.
An interior panel of a floor system is shown in Fig. B.3. Equation (5.3g) is applied to six options, as summarized in Table B.3.
The compressive strength of the concrete is 4000 psi (27.6 MPa) and the modulus of elasticity is 3.605 × 106 psi (24,900 MPa).

Table B.3—Factors and member dimensions: design examples

Equation factors Beam
Slab thickness,
Case βES kDP kAR kSS hs, in. bw, in. hb, in.
1. One-way slab, short direction, narrow beam 1.0 1.0 1.0 1.0 7 12 24
2. One-way slab, short direction, wide beam 1.0 1.0 1.0 1.0 6 36 18
3. One-way slab, long direction, narrow beam 1.0 1.0 1.0 1.0 9 12 22
4. Flat plate 1.0 1.0 1.0 1.35 10 ― ―
5. Flat slab 1.0 1.35 1.0 1.35 9 ― ―
6. Two-way slab on beams 1.27 1.0 0.708 1.0 6 16 22
Note: 1 in = 25.4 mm.

Fig. B.3—Interior panel of a two-way system. (Note: 1 ft = 0.305 m.)

Case 1: One-way slab, short direction, narrow beam (bw = 12 in. [305 mm])
Estimate slab thickness: h = ℓ/28 = 20 × 12/28 = 8.5 in. (216 mm)
Slab self-weight = (8.5/12) × 150 = 106 psf (5075 Pa)
Clear span ℓn = 20 – 1 = 19 ft (5.79 m)
Sustained load Ws = 106 + 15 + 30 = 151 psf (7230 Pa)
Additional live load, wL(add) = 40 psf (1915 Pa)

Slab:
Factors: βES = 1.0, kDP = 1.0, kAR = 1.0, kSS = 1.0, K = 1.4
1
  12   3
 2400 × 3605000   
n 1  12 
1⋅ 
=  =33.5
h  480 (1.4)( 2 × 151 + 40) 
 

h = 19 × 12/33.5 = 6.8 in. → Use h = 7 in. (178 mm)

(Note: a second iteration with the new depth gives h = 6.51 in. [165 mm])
(ACI 318 gives h = ℓ/28 = 8.5 in. [216 mm] (for slab not supporting partitions)

American Concrete Institute – Copyrighted © Material – [Link]

 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20) 55

Beam:
Clear span = 25 – 1 = 24 ft (7.32 m)
Tributary width = 20 ft (6.10 m)
Slab self-weight = (7/12) × 150 = 87.5 psf (4190 Pa)
Estimate weight of beam web = 400 plf (5840 N/m)
Sustained load = (87.5 + 15 + 30) × 20 + 400 = 3050 plf (44,500 N/m)
Additional live load = 40 × 20 = 800 plf (11,700 N/m)
1
  12   3
n  1 2400 × 3605000  12  
1⋅ 
=  = 12.3
h  480 (1.4)( 2 × 3050 + 800) 
 

h = 24 × 12/12.3 = 23.4 in. (594 mm) → Use h = 24 in. (610 mm)

(ACI 318 gives h = ℓ/21 = 11 in. [279 mm] (for member not supporting partitions)

Cases 2 through 5
The calculation procedure is essentially the same for each case. A first estimate of slab thickness can be taken as ℓ/28 in all
cases. At most one iteration will be required for convergence.

Case 6: Two-way slab on beams (bw = 16 in. [406 mm])

Clear span (long) = 25 – 1.33 = 23.67 ft (7.21 m)
Clear span (short) = 20 – 1.33 = 18.67 ft (5.69 m)
βES = 23.67/18.67 = 1.27
kAR = 0.2 + 0.4 × 1.27 = 0.708
kDP = 1.0, kSS = 1.0, K = 1.4

Slab:
Estimate slab depth = 24 × 12/30 = 9.6 in. (244 mm)
Slab self-weight = (9.6/12) × 150 = 120 psf (5750 Pa)
Sustained load ws = 120 + 15 + 30 = 165 psf (7900 Pa)
Additional live load, WL(add) = 40 psf (1915 Pa)

1
  12   3
 2400 × 3605000   
n 1  12 
1.27 ⋅ 
=  =46.50
h  480 ( 0.708)(1.4)( 2 × 165 + 40) 
 

h = (23.67 × 12)/46.50 = 6.1 in. (155 mm)

Try h = 6 in. (152 mm). Second iteration gives h = 5.7 in. (145 mm) → Use h = 6 in. (152 mm)

Beam (long span):

Clear span = 25 – 1 = 24 ft (7.32 m)
Tributary width = 20 ft (6.10 m)
Estimate beam web weight = 400 plf (5840 N/m)
Sustained load, WS = (75 + 15 + 30) × 20 + 400 = 2800 plf (40900 N/m)
Additional live load, WL(add) = 40 × 20 = 800 plf (11700 N/m)

1
  16   3
 2400 × 3, 605, 000   
n 1  12 
1⋅ 
=  = 13.9
h  480 (1.4)( 2 × 2800 + 800) 
 

American Concrete Institute – Copyrighted © Material – [Link]

56 REPORT ON DEFLECTION OF NONPRESTRESSED CONCRETE STRUCTURES (ACI 435R-20)

h = 24 × 12/13.9 = 20.7 in. (526 mm) → Use h = 22 in. (559 mm)

Beam (short span):

Clear span = 20 – 1 = 19 ft (5.79 m)
Tributary width = 25 ft (7.62 m)

1
n  1 2400 × 3, 605, 000(12/12)  3
1⋅ 
=  = 11.8
h  480 (1.4)(2 × 3400 + 1000) 

h = 19 × 12/11.8 = 19.3 in. (490 mm) → Use h = 22 in. (559 mm) (to match long span)

American Concrete Institute – Copyrighted © Material – [Link]

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 to discuss and share concrete knowledge and fellowship.

American Concrete Institute

38800 Country Club Drive
Farmington Hills, MI 48331
Phone: +1.248.848.3700
Fax: +1.248.848.3701
[Link]
38800 Country Club Drive
Farmington Hills, MI 48331 USA
+1.248.848.3700
[Link]

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

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

Individuals interested in the activities of ACI are encouraged to explore the

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

9 781641 951289