You are on page 1of 40

ACI 349.

1R-07

Reinforced Concrete Design


for Thermal Effects on
Nuclear Power Plant Structures

Reported by ACI Committee 349


First printing
June 2007

American Concrete Institute
Advancing concrete knowledge

Reinforced Concrete Design for Thermal Effects


on Nuclear Power Plant 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 occa-
sionally 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. Proper use of this document includes periodically checking for errata at
www.concrete.org/committees/errata.asp 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 regula-
tions, including but not limited to, United States Occupational Safety and Health Administration (OSHA)
health and safety standards.

Order information: ACI documents are available in print, by download, on CD-ROM, 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 ACI Manual of
Concrete Practice (MCP).

American Concrete Institute


38800 Country Club Drive
Farmington Hills, MI 48331
U.S.A.
Phone: 248-848-3700
Fax: 248-848-3701
www.concrete.org

ISBN 978-0-87031-246-05
ACI 349.1R-07

Reinforced Concrete Design for Thermal Effects


on Nuclear Power Plant Structures
Reported by ACI Committee 349

Ronald J. Janowiak*
Chair

Omesh B. Abhat Branko Galunic Charles J. Hookham* Richard S. Orr*


Adeola K. Adediran Partha S. Ghosal* Scott A. Jensen* Bozidar Stojadinovic
Hansraj G. Ashar Herman L. Graves, III Jagadish R. Joshi Barendra K. Talukdar
Ranjit L. Bandyopadhyay Orhan Gurbuz* Richard E. Klingner Donald T. Ward
Peter J. Carrato James A. Hammell Nam-Ho Lee Andrew S. Whittaker
Ronald A. Cook Gunnar A. Harstead* Dan J. Naus* Albert Y. C. Wong
Rolf Eligenhausen Christopher Heinz Dragos A. Nuta Charles A. Zalesiak*
Werner A. F. Fuchs

*
Committee 349 members who were major contributors to the development of this report.

This report presents a design-oriented approach for considering thermal Keywords: cracking (fracturing); frames; nuclear power plants; shells;
effect on reinforced concrete structures. Although the approach is intended to structural analysis; structural design; temperature; thermal effect; thermal
conform to the general provisions of Appendix E of ACI 349, it is not gradient; thermal properties.
restricted to nuclear power plant structures. The general behavior of structures
under thermal effects is discussed together with the significant issues to CONTENTS
consider in reinforcement design. Two types of structuresframes and Chapter 1Introduction, p. 349.1R-2
axisymmetric shellsare addressed. For frame structures, a rationale is 1.1General
described for determining the extent of component cracking that can be
assumed for purposes of obtaining the cracked structure thermal forces and
1.2Thermal effects and structural responses
moments. Stiffness coefficients and carryover factors are presented in graph- 1.3General guidelines
ical form as a function of the extent of component cracking along its length 1.4Analysis techniques
and the reinforcement ratio. Fixed-end thermal moments for cracked compo- 1.5Consideration of thermal effects in analysis
nents are expressed in terms of these factors for: 1) a temperature gradient 1.6Stiffness and deformation effects
across the depth of the component; and 2) end displacements due to a
uniform temperature change along the axes of adjacent components. For the
1.7Summary
axisymmetric shells, normalized cracked section thermal moments are
presented in graphical form. These moments are normalized with respect to Chapter 2Notation and definitions, p. 349.1R-5
the cross-sectional dimensions and the temperature gradient across the 2.1Notation
section. The normalized moments are presented as a function of the internal 2.2Definitions
axial forces and moments acting on the section and the reinforcement ratio.
Use of the graphical information is illustrated by examples.
Chapter 3Frame structures, p. 349.1R-7
3.1Scope
ACI Committee Reports, Guides, Standard Practices, and 3.2Section cracking
Commentaries are intended for guidance in planning, 3.3Component cracking
designing, executing, and inspecting construction. This
document is intended for the use of individuals who are 3.4Cracked component fixed-end moments, stiffness
competent to evaluate the significance and limitations of its coefficients, and carryover factors
content and recommendations and who will accept 3.5Frame design example
responsibility for the application of the material it contains.
The American Concrete Institute disclaims any and all
responsibility for the stated principles. The Institute shall not
be liable for any loss or damage arising therefrom.
Reference to this document shall not be made in contract ACI 349.1R-07 supersedes ACI 349.1R-91 and was adopted and published May 2007.
documents. If items found in this document are desired by the Copyright 2007, American Concrete Institute.
All rights reserved including rights of reproduction and use in any form or by any
Architect/Engineer to be a part of the contract documents, they means, including the making of copies by any photo process, or by electronic or
shall be restated in mandatory language for incorporation by mechanical device, printed, written, or oral, or recording for sound or visual reproduction
the Architect/Engineer. or for use in any knowledge or retrieval system or device, unless permission in writing
is obtained from the copyright proprietors.

349.1R-1
349.1R-2 ACI COMMITTEE REPORT

Chapter 4Axisymmetric structures, p. 349.1R-21 result from cracking of the concrete structure. Thermal
4.1Scope effects should be considered in design for serviceability. The
4.2 |e/d| 0.7 for compressive N and tensile N report discusses conditions under which it can be shown that
4.3General e/d the thermal effects do not adversely affect the safe shutdown
4.4Design examples capacity of the plant. Behavior and general guidance is
addressed in Chapter 1. Chapter 2 addresses notation and
Chapter 5References, p. 349.1R-32 definitions. Chapter 3 addresses frame structures, and
5.1Referenced standards and reports Chapter 4 deals with axisymmetric structures. For frame
5.2Cited references structures, general criteria are given in Sections 3.2 (Section
cracking) and 3.3 (Component cracking). The criteria are
Appendix AExamples in metric, p. 349.1R-33 then formulated for the moment distribution method of
A.1Frame design example from 3.5
structural analysis in Section 3.4. Cracked component fixed-
A.2Design examples from 4.4
end moments, stiffness coefficients, and carryover factors
are derived and presented in graphical form. For axisymmetric
CHAPTER 1INTRODUCTION
1.1General structures, an approach is described for regions away from
ACI 349, Appendix E, provides general considerations in discontinuities, and graphs of cracked section thermal
designing reinforced concrete structures for nuclear power moments are presented.
plants subject to thermal effects. Thermal effects are defined This report is intended to propose simplifications that may
to be the exposure of a structure or component thereof to be used for structural assessments. It will permit exclusion of
varying temperature at its surface or temperature gradient thermal cases with small effect and a reduction of thermal
through its cross section; the resulting response of the effects for a large class of thermal cases without resorting to
exposed structure is a function of its age and moisture sophisticated and complex solutions (Appendix E, 349-06).
content, temperature extreme(s), duration of exposure, and Also, as a result of the report discussion, the design examples,
degree of restraint. The terms force, moment, and and graphical presentation of cracked section thermal moments,
stress apply and are used in this report where a structure is it is hoped that a designer will better understand how thermal
restrained against thermally induced movements. Further effects are influenced by the presence of other loads and the
treatment of these forces, moments, and stresses are resulting concrete response, primarily in the form of
contained in this report as a function of type of structure. cracking, although reinforcement yielding, concrete creep,
The Commentary to Appendix E, Section RE.1.2, of ACI nonlinear concrete stress-strain, and shrinkage are also very
349-06 (ACI Committee 349 2006) instructs the designer to significant in mitigating thermal effects in concrete structures.
consider the following:
1. Linear thermal strain causes stress only under conditions 1.2Thermal effects and structural responses
of restraint, and a portion of such stress may be self-relieving. Thermal effects cause expansion or contraction of the
Mechanisms for relief are: cracking, yielding, relaxation, components in a structural system. If the components are
creep, and other time-dependent deformations; and restrained, which is usually the case, stresses are induced. It is
2. Accident temperature transients may be of such short sufficient to note that there are three types of thermal effects:
duration that the resulting temperature distributions and 1. Bulk temperature change. In this case, the entire structural
corresponding stress changes are not significant. Therefore, component (or segments of the component) is subject to a
these temperature transients may not adversely affect the uniform temperature change;
safe shutdown capacity of the plant. 2. Thermal gradient. A temperature crossfall or thermal
The Commentary to Appendix E, Section RE.3.3, of gradient is caused by different thermal conditions on two
ACI 349-06 addresses three approaches that consider faces of a structure, such as two sides of a wall or the top and
thermal effects in conjunction with all mechanical loads bottom of a beam; and
acting on the structure. One approach is to consider the structure 3. Local thermal exposure. Elevated temperature at a local
uncracked under the mechanical loads and cracked under the surface caused by an external source such as operating
thermal effects. The results of two such analyses are then equipment or piping or an abnormal event such as a fire.
combined. Thermal effects will result in different states of stress and
The Commentary to Appendix E also contains a method of strain in structural components as a function of restraints.
treating temperature distributions across a cracked section. The analysis for thermal effects must distinguish between
In this method, an equivalent linear temperature distribution different types of thermal effects and properly characterize
is obtained from the temperature distribution, which can the structural response accordingly (for example, the degree of
generally be nonlinear. The linear temperature distribution is fixity of end and boundary restraints, component stiffness,
then separated into a pure gradient T and into the difference influence of cracking, and concrete and reinforcing steel strain).
between the mean and base (stress-free) temperatures Tm Tb . Thermal effects can arise from many sources including,
This report discusses approaches for making an assessment but not limited to, process fluid transport; proximity to hot
of thermal effects that are consistent with the aforementioned gasses, steam, or water passage (for example, reactor vessel
provisions. The goal is to present a designer-oriented or steam piping from reactor building to turbine); fire; or
approach for determining the reduced thermal moments that gradients formed when opposing faces of a structure are
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-3

exposed to differing temperatures (for example, spent fuel load is such that the negative moment reinforcement is near
pool) or cyclic gradients from plant startup and shutdown. yield (for example, at 99% of yield). If the temperature is
Temperature change is manifested under one or more of the increased at the bottom of the beam, introducing a thermal
following transfer mechanisms: gradient over its entire length, this gradient would cause
1. Radiation. The electromagnetic transfer of heat from a additional negative moment. When the negative moments
higher temperature source to a lower temperature surface of increase by 1%, the reinforcement would start yielding. The
the concrete structure, such as from a radiator heating a room negative moments at the ends, however, cannot increase
and the surrounding wall and floor structures; beyond 100% because the reinforcement is already yielded.
2. Convection. The transfer of heat usually by the movement Thus, it can be said that the thermal moments at the ends are
of a liquid or gas across a surface, such as from environmental relieved as soon as the reinforcement yields. Yet, the structure
temperature changes in the air next to a concrete structure; and remains stable because the lateral load does not change.
3. Conduction. The transfer of heat through a solid, such Therefore, simplifying assumptions and approximations in
as from a steam pipeline into the surrounding concrete at a analyses are acceptable. Although it is not feasible to set
penetration. definitive rules regarding these assumptions and approxi-
There are many instances where all three mechanisms are mations, some general guidelines are presented. These
present, such as in the case of a fire acting on a structure. guidelines are based on experience with thermal analyses and
Radiation and convection from the flame itself transfers heat engineering judgment.
to the impinged structure. The surface of the flame radiates Design for thermal effects is primarily for serviceability
heat, which is absorbed by the concrete and reinforcing steel; and should address control of cracking;
finally, heat is transferred away from the flame-impinged area Extreme environmental events (that is, Ess and Wt ) are
by means of conduction through the structure. The structure rare occurrences. Serviceability of the structure may
will also lose heat by means of convection and radiation. not be required after such events. If they occur, high
Response of a structure to thermal effects depends on the stresses are induced in local areas. Under these stresses,
nature of the temperature distribution, end constraints, material concrete cracks and reinforcement may yield, relieving
properties, and mechanical loads. A proper thermal stress thermally induced stresses. Thus, it may be too conserva-
analysis must take these parameters into account. tive to add the thermal stresses based on elastic analysis
Stresses in the concrete and reinforcement occur due to to the stresses due to extreme environmental loads.
restraint of thermal movement and these stresses are generally Consequently, the elastically calculated thermal stresses
self-relieving. These thermal stresses are generally small, as should be reduced considering concrete cracking and
most thermal exposures are within prescribed ACI 349 reinforcement yield before combining them with the
temperature limits. Furthermore, internally generated stresses from extreme environmental loads;
stresses are complex to analyze given the size and geometry In nuclear power structures, the controlling load combi-
of safety-related concrete structures. As such, structural nations are generally those that include Eo or Ess . These
analyses using manual calculations with simplifying, load cases provide sufficient reinforcement to control
conservative assumptions (for example, concrete is cracked) cracking. It would be counterproductive to add reinforce-
are typically considered to be appropriate. Computer-based ment to mitigate thermal effects because the additional
analysis tools may also be used to determine the effects of reinforcement would stiffen the structure, thus
thermal exposure and structure response as illustrated in the increasing the stresses due to thermal effects. This is
following paragraphs. It should be noted that a thorough and unnecessary because thermal effects typically self-
complete computer-based thermal analysis is much more relieve without the need for additional reinforcement. If
complicated than a structural analysis of mechanical loads. additional thermal reinforcement is indicated by the
The difficulty of defining important parameters would also design and analysis, the appropriateness of methods
make such computer thermal analyses controversial because and means should be re-evaluated;
differences in parameter ranges may produce significant Thermal gradients should be considered in the design of
differences in analytical results. Finally, it is known that reinforcement for normal conditions to control cracking.
ambient thermal effects, in all but very unusual situations Thermal gradients less than approximately 100 F (56 C)
(for example major fire events), will have little effect on the need not be analyzed because such gradients will not cause
ultimate strength of a concrete structure. significant stress in the reinforcement or strength deteriora-
tion. It may cause a small incursion into the nonlinear
1.3General guidelines range for extreme events, but such incursions are not likely
Stresses resulting from thermal effects are generally self- to adversely affect the overall behavior of the structure.
relieving, that is thermal forces and moments are greatly Because thermal gradient results in flexural stress, the
reduced or completely relieved once concrete cracks or minimum reinforcement on the tension face should be
reinforcement yields; as a result, thermal effects do not established assuming this face is in flexural tension.
reduce the strength of a section for mechanical loads. For The change in curvature due to thermal gradient is
example, assuming a fixed-end beam under transverse loads approximated by A = T/t. The maximum additional
would produce negative moments at the ends and a positive strain in reinforcement due to the change in curvature can
moment at the center. It is also assumed that the transverse be approximated by c = At = T. This approximation is
349.1R-4 ACI COMMITTEE REPORT

for a beam element; for a plate element, the change of conditions related to plant operations and the natural environ-
curvature will be somewhat higher due to Poissons ment (for example, temperature source and exposure time data).
ratio, that is, A = T/[t(1 )]. This approximation 1.4.1 Hand calculationsMany thermal exposure conditions
conservatively assumes rotation of the section about the can be solved by hand calculations, especially where the
outside compression face. With 6 106 in./in./F initial conditions are relatively uncomplicated or the structure
(1.1 105 mm/mm/C) and T = 100 F (56 C), the considered is simple in configuration and behavior. It is
additional reinforcement strain will be about 600 in./in. possible to look at both steady-state and time-dependent
(600 mm/mm), or 0.0006 in./in. (0.0006 mm/mm), solutions in this manner. Simplified hand-calculation
which corresponds to about 0.3y for 60 ksi (420 MPa) approaches are still valuable today (the examples presented
steel (because y = y/Es = (60 ksi)/(29,000 ksi) = in Chapters 3 and 4 are based on such calculations), particu-
0.002. Therefore, the thermal strain, relative to yield larly in the confirmation of computer-based analysis results.
strain, due to a thermal gradient of T = 100 F (56 C), is 1.4.2 Computer-based analysisUse and capability of
[(0.0006)/(0.002)] = 0.3 or 0.3y). If the reinforcement computers and software in structural design has expanded
strain is equal to 0.9y without the thermal effect, the greatly since the time when the early nuclear power plants
total strain with the thermal gradient will be approxi- were designed. Computer analysis methods can be broadly
mately 1.2y, or about 20% beyond yield. Such an classified by the modeling and algorithm used, with the two
exceedance is inconsequential (Gurfinkel 1972), and most prevalent systems involving: 1) stiffness matrix or
will not reduce the capacity of the concrete section for finite element method consisting of linear elements (for
mechanical loads. example, linear beam elements); and 2) finite element
Similarly, the maximum additional concrete strain can method consisting of two- (for example, plate elements) or
be estimated for a fully constrained component to be three-dimensional elements (for example, brick elements).
also approximately 0.0006 in./in. (0.0006 mm/mm), Both linear elements and finite elements can be used to
which is only about 20% of the maximum design accurately model a concrete structure and its components, to
concrete strain of 0.003. Again, such a small exceed- input applied mechanical loads and effects such as thermal
ance in the extreme fiber of the cross section will not be gradients, and to produce an output of resultant forces,
detrimental to the overall strength of the structure; and moments, stresses, and deformations representative of the
A uniform temperature change (Tm Tb) of 50 F (28 C) modeled structure based on mechanics and code requirements.
or less need not be analyzed. Such a temperature change Programs written around linear elements are particularly
may cause up to about 0.0003 in./in. (0.0003 mm/mm) useful for frame-type structures to determine the transfer of
strain, which is only 10% of the maximum design loads to connected components. Many existing computer
concrete strain of 0.003. (Note: The referenced text is programs contain embedded design code (for example, ACI
intended for concrete in compression. The 10% strain 318) requirements to support proportioning and reinforcement
refers to the concrete ultimate strain, which is generally definition. Such programs allow the introduction of thermal
accepted as 0.003. In real structures, the maximum gradients, with resulting forces and moment computation due
concrete strain will be significantly less than 0.002, which to restraints, and certain programs have the ability to examine
is the strain at the peak concrete stress in the Hognestad nonlinear response. Analyzing the impacts of cracking on
stress-strain curve [Kohli and Gurbuz 1976]. Thus, a prac- structural capacity and stiffness, however, is generally beyond
tical strain limit in the example problem will increase from the capabilities of this software program type.
0.002 to 0.0023, which is still less than the ultimate Finite element analysis (FEA) involves a more detailed
concrete strain.) Minimum reinforcement should be simulation of a structure, which is geometrically divided into
established using the requirements for flexural tension. a mesh of two- or three-dimensional elements for a greater
In the determination of forces and moments resulting from precision in assessment of force and displacement distribu-
thermal effects, it would be theoretically possible to perform tion within the structure. The input model also includes
analyses with precise stiffness and stress-strain information boundary conditions and mechanical loading data. Finite
on all structural materials. Fortunately, however, such a element models can be used for structural analysis and
burdensome effort is not necessary. The aforementioned design, as well as for assessing steady-state and transient
guidelines are general; there may be cases where the effect temperature effects and heat transfer. FEA methods can be
of the thermal effects may be more significant on a critical used for both static and dynamic analysis and refined
part of the structure due to layout of the structure. The engineer assessment of nonlinear response. FEA methods are capable
should review the temperature distribution in the structure of analyzing both steady-state and transient thermal effects
and address the potential effects on a case-by-case basis. in reinforced concrete.
Elastic FEAs can be used with a reduced elastic modulus
1.4Analysis techniques for concrete to account in a very simple manner for the
There are several well-recognized techniques for the analysis various effects of cracking, creep, and yield. Values of
of thermal effects in structures, from both steady-state and tran- 0.50Ec have been used in past practice.
sient (time-dependent) temperature profiles. Determination of a Nonlinear thermal analysis is useful where the material
steady-state or transient design-and-analysis approach is based properties in a structure change with exposure to elevated
on the initial condition of the structure in question and process temperatures, such as those listed in Appendix E.4.2 and
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-5

E.4.3 of ACI 349-06. For instance, the specific heat capacity effects, strain-softening effects (treatment of negative strain),
of steel rises with elevated temperature exposure, whereas and out-of-plane shear. Selection of software for use in
mechanical properties decrease. Concrete compressive thermal analysis should consider compatibility with modeling
strength typically decreases, particularly where the exposure concrete (certain programs are suited for homogeneous
is either long-term or locally severe. Other nonlinear materials only), ability to handle strain-softening, user
schemes will involve the change in radiative and convective capability and software experience base, and status of
properties as temperatures vary. verification with respect to 10CFR50 Appendix B requirements
(Office of the Federal Register 2002).
1.5Consideration of thermal effects in analysis
Structural analysis for most nuclear plant concrete structures 1.7Summary
involves the determination of whether or not thermally Thermal effects can cause forces and moments in a
induced forces or moments may occur as a result of restraint. reinforced concrete structure due to restraint of thermal
A typical prerequisite step in completing the structural expansion and contraction. Because concrete resistance to
analysis and design of a structure is the determination of tension is low, the concrete cracks under the thermally
temperature boundary conditions, temperature ranges (bulk induced tensile stresses. As concrete cracks, these tensile
temperature change), and gradients. Considerations include: stresses are relieved. Although rarely a design concern from
Thermal conduction. The assessment of temperatures at a strength perspective, the serviceability of a structure
given locations in concrete structures. The temperature exposed to thermal effects may ultimately be of concern, and
distribution could be calculated either as a steady state such cases should be correctly identified in the assessment
or a transient, depending on the exposure; and considered in the design.
Temperature-induced stress. The temperature distribution Accurate analysis of thermal effects, mechanical loads,
can be used to determine the elastic thermal stresses restraint, concrete cracking, and other stress-relieving effects
throughout the structure; and are difficult at best. The use of a simplified procedure should
Fire protection. Thermal analysis can be used to establish be adequate to calculate the thermal effects. If further
the temperature/time relationship influencing a structure examination is warranted, computer programs are available
subjected to fire. The analysis can take account of fire to deal with the issue more effectively. Specific design
barriers and protective materials such as intumescent methods associated with frame and axisymmetric structures
coatings and fireboards. are contained in this report.
Thermal effects in combination with mechanical loads often
lead to structural cracking of concrete that could adversely CHAPTER 2NOTATION AND DEFINITIONS
affect corrosion protection of the reinforcement. The thermal 2.1Notation
As = area of tension reinforcement within width b, in.2
assessment should consider the possibility that, under certain
(mm2)
extreme conditions, thermal movements could significantly
affect the behavior of the structure. In components or systems As = area of compression reinforcement within width
that include variable temperature exposure, mechanical loads, b, in.2 (mm2)
and restraint against displacement, the support requirements a = length of cracked end of component at which
for each of the loading schemes may be mutually exclusive. A stiffness coefficient and carryover factor are
structure designed to restrict displacement-induced stresses determined, for example, in a end-cracked beam
due to vibration or bracing may provide greater end fixity than (Fig. 3.3 through 3.6) or interior-cracked beam
desired, resulting in elevated thermally induced stresses. A (Fig. 3.7 through 3.10), a is length of uncracked
combined thermal and structural analysis using FEA end of component at which stiffness coefficient
programs is typically needed to balance the requirements. and carryover factor are determined, in. (mm)
b = width of rectangular cross section, in. (mm)
1.6Stiffness and deformation effects CO = cracked component carryover factor from end of
As structural components such as concrete slabs and walls the component to opposite end
are exposed to significant levels of stressing and cracking as COAB = cracked component carryover factor from End A
a result of restrained thermal deformations, reductions in to End B
stiffness due to said cracking and thermal creep can cause COBA = cracked component carryover factor from End B
rapid decay in the restraint forces developed, and deformation to End A
can increase. Simplified approaches, such as use of cracked DF = distribution factor
section properties and moment of inertia, can be used to d = distance from extreme fiber of compression face
calculate the resultant deflections using empirical formulas to centroid of tension reinforcement, in. (mm)
per ACI 207.2R, 209R, and 435.7R, and Fu and Daye (1991). d = distance from extreme fiber of compression face
Nonlinear finite element analysis procedures may also be to centroid of compression reinforcement, in. (mm)
employed to investigate the theoretical response and Ec = modulus of elasticity of concrete, psi (MPa)
deformations in such components. Simulations of the Eo = load effects of operating basis earthquake (OBE)
component response can be obtained. Important to achieving or related internal moments and forces, including
accurate results are the consideration of tension-stiffening OBE-induced piping and equipment reactions
349.1R-6 ACI COMMITTEE REPORT

Es = modulus of elasticity of reinforcing steel, psi MT = thermal moment due to T, MT = M M, in.-lb


(MPa) (mmN)
Ess = load effects of safe shutdown earthquake (SSE) or N = internal axial force at section centerline due to
related internal moments and forces, including factored mechanical loads, including factored
SSE-induced piping and equipment reactions axial force due to Tm Tb, kips (N)
e = eccentricity of internal force N on the rectangular n = modular ratio = Es /Ec
section, measured from section centerline, in. Tb = base (stress-free) temperature, F (C)
(mm) Tm = mean temperature, F (C)
F = flexural coefficient, in.-lb: bd2/12,000 (SI: bd2/ t = thickness of rectangular section, in. (mm)
1000) (ACI Committee 340 1997) Wt = loads generated by design basis tornado (DBT), or
fc = final cracked section extreme fiber compressive related internal moments and forces
stress resulting from internal section forces M, N, = concrete coefficient of thermal expansion, in./in./F
and T, psi (MPa) (mm/mm/C)
fc = specified compressive strength of concrete, psi = transverse displacement difference between ends
(MPa) of cracked component due to Tm Tb acting on
fcL = cracked section extreme fiber compressive stress adjoining components, in. (mm)
resulting from internal forces M and N, psi (MPa) T = linear temperature gradient, F (C)
fr = modulus of rupture of concrete, psi (MPa) c = final cracked section strain at extreme fiber of
fy = specified yield strength of reinforcing steel, psi compression face = cL + cT
(MPa) cL = cracked section strain at extreme fiber of
Icr = cracked section moment of inertia about centroid compression face resulting from M, N, and T
of cracked rectangular section, in.4 (mm4) cT = concrete strain in compression face due to T
Ig = uncracked section moment of inertia (excluding (Fig. 4.1)
reinforcement) about centerline of rectangular = final cracked section curvature change = L + T
section, in.4 (mm4) A = curvature due to thermal gradient
j = ratio of distance between centroid of compression L = cracked section curvature change resulting from
and centroid of tension reinforcement to depth d internal forces M and N
K = cracked component stiffness at End a (pinned), T = cracked section curvature change required to
with opposite end fixed return free thermal curvature T/t to 0
KA = cracked component stiffness at End A (pinned), = Poissons ratio of concrete
with opposite end fixed = ratio of tension reinforcement = As /bd
KB = cracked component stiffness at End B (pinned), = ratio of compression reinforcement = As /bd
with opposite end fixed
Ku = strength coefficient for resistance = fc (1 0.59), 2.2Definitions
where = fy/fc , psi (MPa) (ACI Committee 340 base temperaturethe temperature at which a concrete
1997) member is cured. This is the temperature at which it is
k = ratio of depth of triangular compressive stress assumed the material is free of thermal stresses.
block to depth d, resulting from internal section mechanical loadloads that cause stresses in elements to
forces M, N, and T maintain equilibrium, such as gravity, earthquake and wind
kd = neutral axis location on section due to N, M, and loads, and pressures. Loads due to geometric constraints
T (Fig. 4.1), in. (mm) (including thermal, settlement, creep, and shrinkage effects)
kL = ratio of depth of triangular compressive stress are not considered mechanical loads as the resulting stresses
block to depth d, resulting from internal section are relieved when the constraint is removed. For example, an
forces M and N axial load applied to a column will cause stresses. On the
kLd = neutral axis location on section due to N and M other hand, uniform temperature increase will not cause
(Fig. 4.1), in. (mm) stresses unless the column is restrained from axial deformation.
ks = dimensionless stiffness coefficient = KL/EcIg Another example: internal pressure in a cylinder will cause
L = total length of component, in. (mm) stresses in the walls and ends. If the cylinder is subjected to
LT = cracked length of component, in. (mm) uniform temperature increase, no stresses will be induced
M = internal moment at section centerline due to unless the expansion is restrained.
factored mechanical loads, including factored safety-related concrete structuressafety-related concrete
moment due to Tm Tb, in.-lb (mmN) structures are structures that are designed to remain functional
M = final internal moment at section centerline under the design loads. These structures support, house, or
resulting from M and T, in.-lb (mmN) protect nuclear safety class systems or components.
Mcr = cracking moment = bt2fr /6, in.-lb (mmN) secondary stressa self-limiting normal or shear stress
MFE = cracked component fixed-end moment due to T that is caused by the constraint of a structure and that is
or Tm Tb , at end a, in.-lb (mmN) expected to cause minor distortions that would not result in
Mu = moment capacity of section, ft-kips (mmN) a failure of the structure.
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-7

self-relievingthe process by which excessive internal 1. Concrete compression stress is taken to be linearly
pressure (stress) is automatically relieved. proportional to strain over the component cross section;
temperature distributionthe variation of the total 2. For an uncracked section, the moment of inertia is Ig,
temperature across a section at a point in time. The temperature where Ig is based on the gross concrete dimensions and the
distribution across a section can vary with time as well as reinforcement is excluded. For a cracked section, the
varying along the length of the member. For such variations, moment of inertia is Icr , where Icr is referenced to the centroidal
the engineer should evaluate the effects of temperature distribu- axis of the cracked section. In the formulation of Icr , the
tion at a number of sections and for a number of time durations. compression reinforcement is excluded and the tension
reinforcement is taken to be located at the tension face; that
CHAPTER 3FRAME STRUCTURES is d = t is used; and
3.1Scope 3. The axial force on the section due to mechanical loads
The thermal effect on the frame is assumed to be represented and thermal effects is assumed to be small relative to the
by temperatures that vary linearly through the thicknesses of moment (e/d 0.5). Consequently, the extent of section
the components. The linear temperature distribution for a cracking is taken as that which occurs for a pure moment
specific component must be constant along its length. Each acting on the section.
such distribution can be separated into a gradient T and into
The first assumption is strictly valid only if the extreme
a temperature change with respect to a base (stress-free)
fiber concrete compressive stress due to combined mechanical
temperature Tm Tb .
loads and thermal effects does not exceed 0.5fc . At this
Frame structures are characterized by their ability to
stress, the corresponding concrete strain is around 0.0005 in./in.
undergo significant flexural deformation under these thermal
(0.0005 mm/mm). For extreme fiber concrete compressive
effects. They are distinguished from the axisymmetric
strains greater than 0.0005 in./in. (0.0005 mm/mm) but less
structures discussed in Chapter 4 by the ability of their
than 0.001 in./in. (0.001 mm/mm), the differences are
structural components to undergo rotation, such that the free
insignificant between a cracked section thermal moment
thermal curvature change of T/t is not completely
based on the linear assumption adopted herein versus a
restrained. The thermal moments in the components are
nonlinear concrete stress-strain relationship such as that
proportional to the degree of restraint. In addition to frames,
described in Fu and Daye (1991), Gurfinkel (1972), and
slabs and walls may fall into this category.
Kohli and Gurbuz (1976). Consequently, cracked component
The aforementioned rotational feature is automatically
thermal moments given by Eq. (3-3) and (3-4) are sufficiently
considered in a structural analysis using uncracked component
accurate for concrete strains not exceeding 0.001 in./in.
properties. An additional reduction of the component thermal
(0.001 mm/mm).
moments, however, can occur if component cracking is taken
into account. Sections 3.2 and 3.3 of this chapter describe For concrete strains greater than 0.001 in./in. (0.001 mm/mm),
criteria for the cracking reduction of component thermal the equations previously identified will result in cracked
moments. These criteria can be used as the basis for an analysis component thermal moments that are greater than those
of the structure under thermal effects, regardless of the method based on the nonlinear theory. In this regard, the thermal
of analysis selected. In Section 3.4, these criteria are applied to moments are conservative. They are, however, still reduced
the moment distribution analysis method. from their uncracked values. This cracking reduction of
thermal moments can be substantial, as discussed in Section 4.1,
There are frame and slab structures that can be adequately
idealized as frames of sufficient geometric simplicity to lend which also incorporates Assumption 1.
themselves to moment distribution. Even if an entire frame Formulation of the thermal moments based on a linear
or slab structure does not permit a simple idealization, concrete stress-strain relationship allows the thermal
substructures can be isolated to study thermal effects. Often, moments to be expressed simply by the equations in Chapter 3,
with the use of computer programs for the analysis of or by the normalized thermal moment graphs of Chapter 4.
complex structures, a feel for the reasonableness of the Such simplicity is desirable in a designer-oriented approach.
results is attainable only through less complex analyses Regarding Icr in Assumption 2, the assumptions for the
applied to substructures. The moment distribution method compression and tension reinforcement result in the simple
for thermal effects is applicable for this work. This design expression of (6jk2)Ig for Icr if the axial load is small, as specified
approach uses cracked component stiffness coefficients and in Assumption 3. The use of (6jk2)Ig will overestimate the
carryover factors. These depend on the extent of component cracked section moment of inertia of sections, for which e/d
cracking along its length due to mechanical loads, as 0.5, either with or without compression reinforcement. For
discussed in Section 3.3. a component with only tension reinforcement typically located
at d = 0.9t, the actual cracked section moment of inertia is
3.2Section cracking overestimated by 35% (Ig = (1/12)bt3; Ig1 = (1/12)b(0.9t)3;
Simplifying assumptions are made as follows for the Ig/Ig1 1.35), regardless of the amount of reinforcement. For
purpose of obtaining the cracked section thermal moments a component with equal amounts of compression and tension
and the section (cracked and uncracked) stiffnesses. The reinforcement, located at d = 0.1d and d = 0.9t, its actual
fixed-end moments, stiffness coefficients, and carryover cracked section moment of inertia is also overestimated. The
factors of Section 3.4 are based on these assumptions: overestimation will vary from 35% at the lower reinforce-
349.1R-8 ACI COMMITTEE REPORT

ment ratio (n = n = 0.02) down to 15% at the higher to be cracked wherever tensile stresses are produced by the
values (n = n = 0.12). mechanical loads if these stresses would be increased by the
The use of (6jk2)Ig for cracked sections and the use of Ig thermal effects. The addition of thermal moments that are of
for uncracked sections are further discussed relative to opposite sign to the mechanical moments that exceed Mcr ,
component cracking in Section 3.3. however, may result in a final section that is uncracked.
Regarding Assumption 3, the magnitude of the thermal Therefore, for simplicity, the component is considered to be
moment depends on the extent of section cracking as uncracked for the thermal effect analysis wherever along its
reflected by Icr. Icr depends on the axial force N and moment length the mechanical moments and thermal moments are of
M. The relationship of Icr/Ig versus e/d, where e = M/N, is opposite sign.
shown in Fig. 3.1. The eccentricity e is referred to the section Two types of cracked components will result: 1) end-
centerline. In Fig. 3.1, it is seen that for e/d 1, Icr is practically cracked; and 2) interior-cracked. The first type occurs for
the same as that corresponding to pure bending. For e/d cases where mechanical and thermal moments are of like
0.5, the associated Icr is within 10% of its pure bending sign at the component ends. The second type occurs where
value. Most nonprestressed frame problems are in the e/d these moments are of like sign at the interior of the component.
0.5 category. Consequently, for these problems, it is accurate Stiffness coefficients, carryover factors, and fixed-end
within 10% to use the pure bending value of (6jk2)Ig for Icr. thermal moments are developed for these two types of
This is the basis of Assumption 3. components in Section 3.4. A comprehensive design
example is presented in Section 3.5.
3.3Component cracking The aforementioned simplification of considering the
Ideally, a sophisticated analysis of a frame or slab structure component to be uncracked wherever the mechanical and
subjected to both mechanical and thermal effects might thermal moments are of opposite sign is conservative due to the
consider concrete cracking and the resulting changes in fact that the initial portion of a thermal effect, such as T, will
component properties at many stages of the load application. actually act on a section that may be cracked under the mechan-
Such an analysis would consider the sequential application ical loads. Consequently, the fixed-end moment due to this part
of the loads and effects, and cracking would be based on the of T will be that due to a component completely cracked
modulus of rupture of the concrete fr. The loads would be along its length. Once the cracks close, the balance of T will
applied incrementally to the structure. After each load act on an uncracked section. Consideration of this two-phase
increment, the section properties would be revised for those aspect makes the problem more complex. The conservative
portions of the components that exhibit extreme fiber tensile approach adopted herein removes this complexity. Some of the
stresses in excess of fr. The properties of the components for conservatism, however, is reduced by the use of Ig for the
a given load increment would reflect the component uncracked section (Assumption 2) rather than its actual
cracking that had occurred under the sum of all preceding uncracked section stiffness, which would include reinforcement,
load increments and effects. In such an analysis, the internal and is substantially greater than Ig for n 0.06.
forces and moments would result in component cracking, The fixed-end moments depend not only on the cracked
with only the results from thermal effects considered relieved. length LT , but also on the location of the cracked length a
The type of analysis summarized previously is consistent along the component. This can be seen from a comparison of
with the approach in Item 2 of Section RE.3.3 of the the results for an end-cracked component and an interior-
Commentary to Appendix E of ACI 349-06. An approximate cracked component for the same value of LT . The method
analysis, but one that is generally conservative for the discussed in Section 3.4 accounts for this. This approach is
thermal effects, is also suggested in the Commentary to more applicable for the determination of the thermal
Appendix E of ACI 349 (Item 3 of Section RE.3.3) as an moments than the use of an effective moment of inertia for
alternative. This alternate analysis considers the structure to the entire component length. The concept of a single effective
be uncracked under the mechanical loads and to be cracked moment of inertia for purposes of component deflection
under the thermal effects. The results of an analysis of the calculation has resulted in Eq. (9-10) of ACI 349-06. This
uncracked structure under mechanical loads are combined equation is empirically based and, as such, accounts for:
with the results of an analysis of the cracked structure under 1) partially cracked sections along the component; and 2) the
the thermal effects. A simplified method of analysis is existence of uncracked sections occurring between flexural
discussed that will yield cracked component thermal cracks. These two characteristics are indirectly provided for
moments that are conservative for most practical problems. (to an unknown extent) by the use of (6jk2)Ig, which over-
The extent of cracking that the components experience estimates the cracked section moment of inertia by the
under the total mechanical load (including the specified load amount described previously.
factors) forms the basis for the cracked structure used for the
thermal effects analysis. Cracking will occur wherever the 3.4Cracked component fixed-end moments,
mechanical load moments exceed the cracking moment Mcr . stiffness coefficients, and carryover factors
The addition of thermal moments that are the same sign as The thermal moments due to the linear temperature
mechanical moments will increase the extent of cracking gradient T and those resulting from the expansion or
along the component length. Recognizing this, in many contraction of the axis of the component Tm Tb are considered
cases, it is conservative for design to consider the component separately. For each type of thermal effect, fixed-end
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-9

Fig. 3.1Effect of axial force on cracked section moment of inertia (no compression reinforcement).
349.1R-10 ACI COMMITTEE REPORT

where ks is the dimensionless stiffness coefficient that is a


function of LT /L and a/LT . Likewise, CO can be expressed
as a function of LT /L and a/LT .
Figures 3.3 through 3.6 show k and CO for selected values
of LT /L and a/LT that should cover most practical problems.
In these figures, ks is given at the end, which is cracked a
distance a, and CO is the carryover factor from this end to the
opposite end. Intermediate values of ks and CO can be deter-
mined by linear interpolation of these curves.
Fig. 3.2(a)T fixed-end moments: member cracked at
ends by mechanical loads. For a component cracked a distance LT in its interior, ks
and CO are determined from Fig. 3.7 through 3.10. ks is the
stiffness coefficient at the end that is uncracked a distance a.
CO is the carryover factor from this end to the opposite end.
Based on the previous discussion, the T fixed-end
moment at the a end of the component can be expressed as

2
E c Tbt k s ( 1 CO )
M FE = ------------------------
- --------------------------- (3-3)
12 2

For the purpose of determining the mean temperature


effects, it is necessary to develop the Tm Tb fixed-end
Fig. 3.2(b)Tm Tb fixed-end moment: member cracked at moment, which is shown in Fig. 3.2(b) for a component
ends by mechanical loads. cracked at its ends.
The Tm Tb fixed-end moment at the end cracked a
moments, stiffness coefficients, and carryover factors were distance a is
obtained for two types of cracked components: 1) end-
cracked; and 2) interior-cracked. The first type applies to Ec Ig
cases where mechanical and thermal effects produce moments - k s ( 1 + CO )
M FE = ------------- (3-4)
2
L
of like sign at component ends. The second type applies to cases
where mechanical and thermal effects produce moments of like
sign in the interior of the component. where ks and CO are same as that defined previously. The
These factors are presented for the case of an end-cracked displacement is produced by Tm Tb acting on an adjacent
component in Fig. 3.2(a) component. The comprehensive design example of Section 3.5
illustrates this.

M A = TL
--------------- K ( 1 CO AB ) 3.5Frame design example
2t A
(3-1) The continuous frame shown in Fig. 3.11 is given with all
TL components 1 ft wide x 2 ft thick and 3 in. cover on the
M B = --------------- K B ( 1 CO BA )
2t reinforcement. The load combination to be considered is U
= D + L +To +Ess.
Although shown only for a component cracked at the ends, The mechanical loading consists of
the aforementioned expressions for MA and MB also apply to
a component cracked in its interior. WD = 406 lb/ft
In Eq. (3-1), TL/2t = the angle change of the component
ends with the rotational restraints removed; KA = the stiffness WL = 680 lb/ft
of the component at A with B fixed (4EcIg /L for uncracked
component); KB = the stiffness of the component at B with A on component BC, and a lateral load of 3750 lb at joint C due
fixed (4EcIg /L for the uncracked component); COAB = the to Ess.
carryover factor from A to B (1/2 for uncracked component); The thermal gradient To results from 130 F interior and
and COBA = the carryover factor from B to A (1/2 for 50 F exterior temperature. Thus, Tm = (130 + 50)/2 = 90 F.
uncracked component). The base temperature Tb is taken as 70 F. For this condition,
The expressions for K and CO can be derived from Tm Tb = 90 F 70 F = +20 F and T = 80 F (hot interior,
moment-area principles. Also, K can be expressed as cold exterior).
The material properties are fc = 3000 psi and Ec = 3.12
Ec Ig 106 psi, fy = 60,000 psi and Es = 29 106 psi; and = 5
K = ---------
-k (3-2)
L s 106 in./in./F. Also, n =Es /Ec = 9.3.
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-11

Fig. 3.3End-cracked beam, ks and CO for LT = 0.1L.


349.1R-12 ACI COMMITTEE REPORT

Fig. 3.4End-cracked beam, ks and CO for LT = 0.2L.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-13

Fig. 3.5End-cracked beam, ks and CO for LT = 0.4L.


349.1R-14 ACI COMMITTEE REPORT

Fig. 3.6End-cracked beam, ks and CO for LT = 0.6L.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-15

Fig. 3.7Interior-cracked beam, ks and CO for LT = 0.1L.


349.1R-16 ACI COMMITTEE REPORT

Fig. 3.8Interior-cracked beam, ks and CO for LT = 0.2L.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-17

Fig. 3.9Interior-cracked beam, ks and CO for LT = 0.4L.


349.1R-18 ACI COMMITTEE REPORT

Fig. 3.10Interior-cracked beam, ks and CO for LT = 0.6L.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-19

Fig. 3.11Uncracked frame moments (ft-kips).

The reinforcement in the frame consists of two No. 8 bars


at each face in all components. This results in = As /bd = (2
Component End LT /L, ft/ft a/LT , ft/ft
0.79)/12(24 3) = 1.58/(12 21) = 0.0063 and n =
AB A 11.8/20 = 0.59 0
9.3(0.0063) = 0.059. = fy/ fc = 0.0063(60/3) = 0.126; and
AB B 0.59 1
Ku = fc (1 0.59) = 3000 0.126(1 0.59 0.126) = 378
BC B (5.3 + 3.4)/30 = 0.29 5.3/8.7 = 0.61
0.92566 = 349.8. The section capacity is Mu = KuF =
BC C 0.29 3.4/8.7 = 0.39
(0.9)(349.8) (12)(21)2/12,000 = 138.8 ft-kips.
CD C 17.2/20 = 0.86 1
Mechanical loadsAn analysis of the uncracked frame
CD D 0.86 0
results in the component moments (ft-kips) as follows.
Moments acting counterclockwise on a component are
denoted as positive. These values were obtained by moment All components are the end-cracked type. Figures 3.4
distribution, and moments due to Ess include the effect of through 3.6 are used to obtain the coefficients ks and CO,
frame sidesway. which are shown in Table 3.1.
The expressions from Section 3.4 for fixed-end moment
Moment (ft-kips) (FEM) are evaluated as
AB: 52.3 2
BA: 76.0 E c Tbt k s
1. T FEM = ------------------------
- ---- ( 1 CO )
BC: +76.0 12 2
CB: 46.0
6 6 2
CD: +46.0 ( 3.12 10 ) ( 5 10 ) ( 80 ) ( 12 ) ( 24 ) k
= ------------------------------------------------------------------------------------------- ----s ( 1 CO )
DC: +7.5 12 2
These are shown in Fig. 3.11.
The maximum mechanical load moment of 76 ft-kips is T FEM = 59.9ks(1 CO)/2 ft-kips
less than the section capacity of 138.8 ft-kips. Therefore, the
frame is adequate for mechanical loads. Ec Ig
- ( ) ( k s ) ( 1 + CO )
2. Tm Tb FEM = ---------
2
Thermal effects (T = 80 F and Tm Tb = 20 F) L

6 3
The T = 80 F having hot interior and cold exterior is = 3.12 10 ( 24 ) - ( ) ( k ) ( 1 + CO )
------------------------------------- s
expected to produce thermal stresses that are tensile on the 2
( 20 12 )
exterior faces of all components. These stresses will add to
the existing exterior face tensile stresses due to the mechanical Tm Tb FEM = 62.4()(ks)(1 CO) ft-kips
loads. Hence, the LT and a values are arrived at from the
mechanical load moment diagram in Fig. 3.11. 3. = total unrestrained change of length of component BC
349.1R-20 ACI COMMITTEE REPORT

Table 3.1Cracked frame coefficients and thermal moments on components


Thermal FEMs, ft-kips
Distributed thermal
T-80 Tm Tb = 20 Distributed thermal moments and mechanical
L T /L a/L T ks Ky /L *
Component End CO DF FEM FEM Total FEM moments, ft-kips moments, ft-kips
AB A 0.59 0 3.40 0.41 0.17 1.0 +60.1 4.79 +55.31 +36.0 16.3
AB B 0.59 1 2.00 0.70 0.10 0.56 17.97 3.39 21.36 39.9 115.9
BC B 0.29 0.61 2.40 0.43 0.08 0.44 +41.7 0 +41.7 +39.9 +115.9
BC C 0.29 0.39 2.65 0.38 0.088 0.48 49.2 0 49.2 37.0 83.0
CD C 0.86 1 1.90 0.57 0.095 0.52 +24.5 +3.72 +28.2 +37.0 +83.0
CD D 0.86 0 2.39 0.47 0.120 1.0 37.9 +4.38 33.5 33.2 25.7
*Corrected for sidesway.
Notes: DFi = (ksiEiIgi/Li)/(ksiEiIgi/Li); symbols with component i are indicated with subscript i.

Fig. 3.12Final frame moments (ft-kips).

= (Tm Tb)L Ec Ig
- ( 8.19 )
= ---------
3
L
= (5 106)(20)(30 12)
Shear stiffness at C
= 0.036 in.
Ec Ig
Distribute 0.036 in. to Ends B and C of Components AB - [ k sC ( 1 + CO C ) + k sD ( 1 + CO D ) ]
= ---------
3
and CD, respectively, in inverse proportion to the shear L
stiffness at these ends.
Ec Ig
Shear stiffness at B - [ 1.90 ( 1.57 ) + 2.38 ( 1.47 ) ]
= ---------
3
L
Ec Ig
- [ k sA ( 1 + CO A ) + k sB ( 1 + CO B ) ]
= --------- Ec Ig
- ( 6.48 )
= ---------
3
L L
3

Ec Ig Ec Ig
- [ 3.4 ( 1.41 ) + 2.00 ( 1.70 ) ]
= --------- Sum of shear stiffness at B and C = ---------
- (14.67)
3 3
L L
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-21

Fig. 4.1Section under M, N, T.

B = 0.036 in. (6.48/14.67) = 0.036 0.44 = 0.016 in. temperature distribution is separated into a gradient T and
C = 0.036 in. (8.19/14.67) = 0.036 0.56 = 0.020 in. into a uniform temperature change Tm Tb.
To demonstrate the effect of cracking on the thermal Generally, for most axisymmetric structures, a uniform
moments, the fixed-end thermal moments for the uncracked temperature change (Tm Tb) produces significant internal
frame are obtained from the final expressions in 1 and 2 section forces (moment included) only at the externally
using ks = 4, CO = 1/2, and = (1/2)(Tm Tb)L , with L restrained boundaries of the structure where movement due
being the length of Component BC. A moment distribution to thermal change is prevented, or in regions where Tm Tb
is performed, and the resulting distributed moments are varies fairly rapidly along the structure. The magnitude and
added to the mechanical moments. The combined moments extent of these discontinuity forces depend on the specific
are shown in Fig. 3.11 for comparison with the cracked geometry of the structure and on the external restraint
frame moments. provided. If cracking occurs in this region, a prediction of the
cracking reduction of the discontinuity forces is attainable
The fixed-end thermal moments for the cracked frame are
through a re-analysis using cracked section structural
obtained using the aforementioned values for B and C and
properties. A discussion of such an analysis is not within the
by referring to Table 3.1 for ks and CO. These fixed-end
scope of this report. Therefore, forces resulting from an
moments and the resulting distributed thermal moments are
analysis for the Tm Tb part of the thermal effect are considered
given in Table 3.1. The distributed thermal moments include
to be included with corresponding factored mechanical
the effect of sidesway, which occurs because the frame is
forces. These combined axial forces and moments are
unsymmetrically cracked.
denoted as N and M.
Combined loadsThe final frame moments are shown The gradient T produces internal section forces (moment
in Table 3.1 and Fig. 3.12. These can be compared with included) at externally restrained boundaries, and also away
Fig. 3.11 to see the effect of the cracking reduction of from these discontinuities. At discontinuities, the most
thermal moments. significant internal force is usually the moment, primarily
Although not shown, the component axial forces were resulting from the internal restraint rather than the external
evaluated to confirm that section cracking still corresponds boundary restraint. Away from discontinuities, the only
to the pure bending condition of Assumption 3. Recall that e/d significant forces due to T are thermal moments caused by
should be at least 0.5 for this condition. For Components AB the internal restraint provided by the axisymmetric geometry
and CD, the axial forces result primarily from the mechanical of the structure. The cracking reduction of thermal moments
loads and are compressive. For Component BC, the axial that result from internal restraint is the subject of this chapter.
force is compressive and includes the compression due to the Due to the axisymmetric geometry of the subject structures,
20 F increase on the component. the free thermal curvature change T/t is fully restrained.
This restraint produces a corresponding thermal moment
CHAPTER 4AXISYMMETRIC STRUCTURES whose magnitude depends on the extent of cracking the
4.1Scope section experiences. This in turn depends on T, the other
Axisymmetric structures include shells of revolution such section forces N and M, and the section properties (Fig. 4.1).
as shield buildings or, depending on the particular geometry, With the ratio M/N denoted as e, referenced to the section
primary and secondary shield walls. In the structural analysis, centerline, and the distance from the concrete compression
the structure is considered to be uncracked for all mechanical face to the tension reinforcement denoted as d, two cases of
loads and for part of the thermal effects. The thermal effect e/d are identified in Sections 4.2 and 4.3.
is assumed to be represented by a temperature that is distributed The results in Sections 4.2 and 4.3 include the effect of
linearly through the wall of the structure. The linear compression reinforcement. For this reinforcement, a modular
349.1R-22 ACI COMMITTEE REPORT

ratio of 2n (Ferguson 1965) is used for purposes of simplifying and the corresponding thermal moment is
the determination of the cracked section thermal moment.
Although not all the loads that comprise the section forces N 3 2
E c Tbd ( jk )
and M will necessarily be long-term, the selection of 2n for M T = -------------------------------------- (4-4)
compression reinforcement is consistent with design practice. 2t ( 1 )
The results in Sections 4.2 and 4.3 are based on a linear
stress-strain relationship for the compressive concrete. The where j = 1 k/3.
basis of this assumption was discussed in Section 3.2. From Equation (4-2) for MT for a doubly reinforced section
this discussion, the cracked section thermal moments can be reduces to MT for a singly reinforced section (Eq. (4-4)),
considered to represent upper-bound values when compared with the substitution of (1.1)d for t in Eq. (4-4), and 0 for n
with those that would result from a nonlinear stress-strain in Eq. (4-2). In addition, the substitution of k2/2 for n(1 k) in
concrete relationship. Nevertheless, the thermal moments Eq. (4-2) must be made.
herein do offer a reduction from their uncracked values. The The thermal moments given by Eq. (4-2) and (4-4) are
extent of this reduction is shown in Fig. 4.2 through 4.8. presented in Fig. 4.2 for the special case: t = (1.1)d for both
sections, and n = n and d/d = 0.10 for the doubly reinforced
4.2|e/d| 0.7 for compressive N and tensile N section. For values of less than , linear interpolation
For this range of e/d, a parametric study based on the between the two curves should yield sufficiently accurate
results of Section 4.3 indicates that the cracked thermal results. From Fig. 4.2, it is seen that the cracked section thermal
moment MT is not strongly influenced by the axial force as moment is substantially reduced from its uncracked value.
expressed by the ratio N/(bdEcT). A practical range of The thermal moment MT occurs at the centerline of the
N/(bdEcT) from 0 to 300 was used in this study. There- section. MT should be multiplied by its code-specified load
fore, for ranges of e/d and N/(bdEcT) specified herein, factor before it is added to the moment M.
MT can be calculated from the neutral axis location
corresponding to N = 0. 4.3General e/d
The |e/d| lower limit of 0.70 is conservative for tensile N Depending on e/d, the extent of section cracking and the
and higher n values. Actual |e/d| lower limits for tensile N thermal moment may be significantly affected by the actual
are given in Fig. 4.9. As long as the actual |e/d| value for values of N and M. A theory for the investigation of a doubly
tensile N exceeds this lower-limit curve, the thermal reinforced rectangular section is presented. The Poissons
moments given in this section are applicable. effect increases the section thermal moment due to T by an
For doubly reinforced rectangular sections, the cracked amount 1/(1 ). Although this effect is not shown in the
section neutral axis is kd. For N = 0 following derivations, it is included in the final results (Fig. 4.2
through 4.8).
k =
2
( 2n + n ) + 2 [ 2n ( d d ) + n ] ( 2n + n ) (4-1) It is assumed for the section that the final curvature change
is equal to L , the curvature change due to N and M plus
in which = As /bd, = As /bd, and n = Es /Ec . Also, d is T , the curvature change due to T. L and T are additive
the distance from the concrete compression face to the when the cold face of the section corresponds to the tension
compression reinforcement As . A modular ratio of 2n is used face under M. Therefore
for the compression reinforcement.
The corresponding thermal moment for a section in which = L + T (4-5)
t = (1.1)d is
The curvatures before and after the application of T are
Tbd
2
3 shown in Fig. 4.1.
M T = E c -------------------- { 0.152k + 1.818n [ ( d d ) k ] ( d d ) (4-2)
1 The thermal curvature change T is

+ 0.909n ( 1 k ) } T = T/t

The expression for MT given by Eq. (4-2) is obtained where T is always taken as positive.
from the results of Section 4.3 in the following manner. For Using this and c = kd and cL = L kL d in Eq. (4-5) gives
sections in which |e/d| 0.7, the location of the neutral axis
does not change under the application of T, and this results c /kd = cL /kL d + T/t (4-6)
in kL = k. This substitution for kL is made in Eq. (4-7) and (4-9).
The resulting expression for fc given by Eq. (4-7) is used in
For the case where the concrete stress is a linear function
Eq. (4-11). MT is then obtained by subtraction of Eq. (4-9)
of strain, Eq. (4-6) becomes
from Eq. (4-11).
For singly reinforced rectangular sections and N = 0
fc /(Eckd) = fcL /(EckLD) + T/t
2
k = ( n ) + 2n n (4-3) or
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-23

Fig. 4.2Cracked section thermal moment for |e/d| 0.70.

fc = [fcL/(EckLd) + T/t]Ec kd (4-7) Before T


1. The internal axial force N is equal to the resultant of the
stresses produced by N and M
To maintain equilibrium of the section both before and after
the application of T, the following conditions occur: N = 1/2fcLbkLd + 2nbdfcL[(kL d/d)/kL ] + nbdfcL[(kL 1)/kL] (4-8)
349.1R-24 ACI COMMITTEE REPORT

Fig. 4.3Thermal moments for compressive N, n = 0.02.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-25

Fig. 4.4Thermal moments for compressive N, n = 0.04.


349.1R-26 ACI COMMITTEE REPORT

Fig. 4.5Thermal moments for compressive N, n = 0.06.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-27

Fig. 4.6Thermal moments for compressive N, n = 0.08.


349.1R-28 ACI COMMITTEE REPORT

Fig. 4.7Thermal moments for compressive N, n = 0.10.


DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-29

Fig. 4.8Thermal moments for compressive N, n = 0.12.


349.1R-30 ACI COMMITTEE REPORT

Fig. 4.9e/d limits.

2. The internal moment M is equal to the internal moment of After T


the stresses (about the section centerline) produced by N and M. 1. The internal axial force N is equal to the resultant of the
stresses produced by N, M, and T
M = 1/2fcLbkLd(t/2 kLd/3) + 2nbdfcL [(kL d/d)/kL] (4-9)

[(t/2) d] + nbdfcL [(1 kL)/kL][d (t/2)] N = 1/2fcbkd + 2[(k d/d)/k] + nbdfc[(k 1)/k] (4-10)
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-31

2. There exists an internal centerline moment (moment fact that the increase in section flexural stiffness (EI) due to
about centerline for internal forces) M of the stresses inclusion of the reinforcement is greater than the loss of
produced by N, M, and T. section flexural stiffness that results from the relatively minor
cracking associated with the low e/d value. The net effect is
M = 1/2fcbkd[(t/2) (kd/3)] + 2nbdfc[(k d/d)/k] (4-11) to give a larger actual cracked section stiffness than that
obtained for the gross uncracked concrete section alone.
[(t/2) d] + qnbdfc[(k 1)/k][d (t/2)] If the designer finds these cracked section thermal
moments to be unacceptably high, a potential reduction may
be available through the use of an approach incorporating a
3. The internal thermal moment MT at the section center-
nonlinear representation of the concrete stress-strain
line is equal to M M.
behavior. Such an approach is described in Fu and Daye
In Eq. (4-8) through (4-11), the tension and compression (1991), Gurfinkel (1972), and Kohli and Gurbuz (1976).
reinforcement have been expressed as = As/bd and =
As /bd, respectively. A modular ratio of 2n has been used for 4.4Design examples
the compression reinforcement. Also, the reinforcement 4.4.1 Example 1: Compressive N and |e/d| > 0.70, =
stresses have been expressed in terms of the concrete stress. b = 12 in., t = 36 in., d = 32.7 in., d = 3.3 in., As = 2 in.2, As
From Eq. (4-8), fcL can be expressed in terms of N, kL , and = 3 in.2, Es = 29 106 psi, Ec = 4 106 psi, = 0.2, N =
the section properties. The use of this in Eq. (4-7) allows fc 50 kips compression, M = 100 ft-kips, T = 80 F
to be written in terms of N, kL , k, EcT, and the section prop-
erties. Substitution of this expression for fc into Eq. (4-10) n = Es /Ec = 29/4 = 7.25, e = M/N = 100/50 = 2 ft = 24 in.,
results in a quadratic equation in k that is solved in terms of = 2/12 36 = 0.0046, n = 0.0046 7.25 = 0.033, =
the section properties, kL , and the quantity N/bdEcT. 3/12 36 = 0.0069, n = 0.0069 7.25 = 0.050
By dividing Eq. (4-9) by Eq. (4-8), however, kL can be
written in terms of the section properties and e, where e = M/ Cracked thermal moment MT
N. Thus, k is determined for a specified section e and N/ e/d ratio: e/d = 24/32.7 = 0.733 > 0.70; therefore, Section 4.2
bdEcT. The aforementioned results also allow MT to be results should be used.
determined from these specified quantities. Because t/d = 1.1 and d/d = 0.10, Fig. 4.2 should be used.
The equilibrium equations, appearing as Eq. (4-8) through Because , however, interpolate between the two curves.
(4-11), are based on a triangular concrete stress distribution. From Fig. 4.2 for n = 0.050, (1 )MT /(bd2EcT)
The two extremes of the stress distribution are at kL = 0.10 should be read as
and kL = 1.0. The range 1.0 kL 0.10 should cover many
practical situations not involving prestressed sections. For kL 0.0303 for = 0
outside this range, that is, the entire section being under
tension (kL 0.10) or compression (kL 1.0), a similar set of 0.0320 for =
equilibrium equations based on a rectangular stress distribution
would be required. Interpolate for = 0.0046
Special case
(1 )MT /(bd2EcT) = 0.0303 + (0.0320
= , t/d = 1.1, and d/d = 0.10 0.0303)(0.0045/0.0069) = 0.0314

For this case, cracked section thermal moments were


(1 )MT = 0.0314(12)(32.7)2(4 106) (5.5 106)(80)
calculated for a n range of 0.02 to 0.12 and N/bdEcT
ranging within 300. For the case of compressive N, Fig. 4.3 = 709,000 in.-lb
through 4.8 apply. Alternatively, for compressive N and e/d
0.7, Fig. 4.2 may be used with reasonable accuracy. For the MT = 886,000 in.-lb or 73.9 ft-kips
case of tensile N, only Fig. 4.2 applies.
As previously discussed, the thermal moments are valid Check using equations
only for 1.0 kL 0.10. Associated with these limits are e/d
values that are indicated in Fig. 4.9. 2
kN=0 = ( 2n + n ) + 2 [ 2n ( d d ) + n ]
Also presented in Fig. 4.3 through 4.8 are the uncracked
thermal moments based on both gross section (neglecting (2n + n) = [(2 0.033 + 0.050)2 + 2(2 0.033 0.10
reinforcement) and actual section (including concrete and + 0.050)]1/2 (2 0.033 + 0.050) = 0.239
reinforcement). It is seen that the cracked section thermal
moments are always less than the uncracked thermal (1 )MT /(bd2EcT) = 0.152k3 + 1.818n(d/d k)
moments obtained for the actual section. For the combination of (d/d) + 0.909n(l k) = 0.152(0.239)3 + 1.818(0.034)
higher n values and lower e/d values, however, the cracked (0.1 0.239)(0.10) + 0.909(0.050) (1 0.239) = 0.0317
section thermal moments exceed the uncracked thermal
moments based on a gross concrete section. This is due to the MT = 73.9(0.0317/0.0314) = 74.6 ft-kips 73.9 ft-kips, OK
349.1R-32 ACI COMMITTEE REPORT

Table 4.1Summary of results for Examples 1, 2, 3, and 4


Temperature Compression Tension Cracked thermal Total moment
Axial force Moment M, differential reinforcement2
reinforcement moment M ,
2 T (M + MT),
Example N, kips Type ft-kips T, F As , in. As , in. ft-kips ft-kips e/d Use figure
1 50 Compression 100 80 2 3 73.9 173.9 0.733 > 0.70 Use Fig. 4.2
2 50 Tension 100 80 3 3 75.3 175.3 0.733 > 0.70 Use Fig. 4.2
3 100 Compression 100 80 3 3 95.3 195.3 0.367 < 0.70 Use Fig. 4.4,
0.367 > 0.25 4.5, and 4.9

4 60 Tension 100 80 3 3 75.3 175.3 0.612 < 0.70 Use Fig. 4.2
0.612 > 0.575 and 4.9

Concrete and reinforcement stresses are calculated from a Concrete and reinforcement stresses are calculated from a
cracked section investigation with N = 50 kips compression cracked section investigation with N = 100 kips compression
and M = 173.9 ft-kips at section centerline. and M = 195.3 ft-kips at section centerline.
4.4.2 Example 2: Tensile N and |e/d| > 0.70, = 4.4.4 Example 4: Tensile N and |e/d| < 0.70
b = 12 in., t = 36 in., d = 32.7 in., d = 3.3 in., As = As = 3.0 in.2, Same as Example 3, except N = 60 kips tension.
Es = 29 106 psi, Ec = 4 106 psi, = 0.2, N = 50 kips e = 100 ft-kips/(60 kips) = 1.67 ft = 20 in., e/d = 20/32.7
tension, M = 100 ft-kips, and T = 80 F = 0.612 or |e/d| = 0.612
From Fig. 4.9 for n = 0.05, the lower limit on |e/d| is
n = 7.25, e = 100/50 = 2 ft = 24 in., = = 0.0069, n 0.575 for the tensile N case. Because 0.612 > 0.575, Fig. 4.2
= n = 0.050 should be used.
Cracked thermal moment MT From Fig. 4.2, with n = 0.05, read (1 )MT /(bd2EcT)
e/d = 24/32.7 = 0.733, |e/d| = 0.733 > 0.7; therefore, = 0.032
Fig. 4.2 should be used. MT = 0.032(1.25)(12)(32.7)2(4)(5.5)(80) = 75.3 ft-kips
From Fig. 4.2 with n = 0.050 and = curve, MT = 75.3 ft-kips
Concrete and reinforcement stresses are calculated from a
cracked section investigation with N = 60 kips tension and M
(1 )MT /(bd2EcT) = 0.032 = 175.3 ft-kips at section centerline.
4.4.5 Summary of results for Examples 1, 2, 3, and 4
MT = (1.25)(12)(32.7)2(4)(5.5)(80)(0.032) = 903,000 in.-lb A summary of results for Examples 1, 2, 3, and 4 can be
MT = 75.3 ft-kips found in Table 4.1.
Concrete and reinforcement stresses are calculated from a
cracked section investigation with N = 50 kips tension and M CHAPTER 5REFERENCES
= 175.3 ft-kips at section centerline. 5.1Referenced standards and reports
The standards and reports listed below were the latest
4.4.3 Example 3: Compressive N and |e/d| < 0.70
editions at the time this document was prepared. Because
Same section as Example 2.
these documents are revised frequently, the reader is advised
N =100 kips compression, M = 100 ft-kips, T = 80 F to contact the proper sponsoring group if it is desired to refer
e = 100 ft-kips- = 1 ft = 12 in.
------------------------- to the latest version.
100 kips
e/d = 12/32.7 = 0.367 < 0.70; therefore, Fig. 4.2 should not American Concrete Institute
be used. 207.2R Effect of Restraint, Volume Change, and Reinforce-
From Fig. 4.9 for n = 0.05, lower limit on |e/d| should be ment on Cracking of Mass Concrete
read as 0.25. Because 0.367 > 0.25, Fig. 4.4 and 4.5 should 209R Prediction of Creep, Shrinkage, and Temperature
be used. Effects in Concrete Structures
318 Building Code Requirements for Structural
Concrete
N/(bdEcT) = 100,000/(12 32.7 4 106 5.5 106
349 Code Requirements for Nuclear Safety-Related
80) = 0.145
Concrete Structures
435.7R Report on Temperature-Induced Deflections of
For N/(bdEcT) = 0.145 and e/d = 0.367, find MT. By Reinforced Concrete Members
interpolation between e/ds: The above publications may be obtained from the
n = 0.04 (Fig. 4.4) at e/d = 0.367; 0.035 by interpolation following organization:
n = 0.06 (Fig. 4.5) at e/d = 0.367; 0.046 by interpolation
For n = 0.05, (1 )MT /bd2EcT =1/2(0.035 + 0.046) American Concrete Institute
= 0.0405 PO Box 9094
MT = 0.0405(1.25)(12)(32.7)2(4)(5.5)(80) = 1,143,000 in.-lb Farmington Hills, MI 48333-9094
MT = 95.3 ft-kips www.concrete.org
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-33

5.2Cited references Mechanical loadsAn analysis of the uncracked frame


ACI Committee 340, 1997, ACI Design Handbook, SP- results in the component moments (m-N) as follows.
17(97), American Concrete Institute, Farmington Hills, Moments acting counterclockwise on a component are
Mich., 482 pp. denoted as positive. These values were obtained by moment
ACI Committee 349, 2006, Code Requirements for Nuclear distribution, and moments due to Ess include the effect of
Safety-Related Concrete Structures (ACI 349-06) and frame sidesway.
Commentary, American Concrete Institute, Farmington Hills,
Mich., 149 pp. Moment (m-N)
Ferguson, P. M., 1965, Reinforced Concrete Fundamentals, AB: 70,900
2nd Edition, John Wiley & Sons, Inc. BA: 103,000
Fu, C. C., and Daye, M. D., eds., 1991, Computer Analysis
BC: +103,000
of Effects of Creep, Shrinkage, and Temperature Changes on
CB: 62,400
Concrete Structures, SP-129, American Concrete Institute,
Farmington Hills, Mich., 191 pp. CD: +62,400
Gurfinkel, G., 1972, Thermal Effects in Walls of Nuclear DC: +10,200
Containment-Elastic and Inelastic Behavior, Proceedings, These are shown in Fig. A.1.
First International Conference on Structural Mechanics in The maximum mechanical load moment of 103,000 m-N
Reactor Technology (Berlin, 1971), V. 5-J, Commission of is less than the section capacity of 189,000 m-N. Therefore,
the European Communities, Brussels, pp. 277-297. the frame is adequate for mechanical loads
Kohli, T., and Gurbuz, O., 1976, Optimum Design of
Reinforced Concrete for Nuclear Containments, Including Thermal effects (T = 44 C and Tm Tb = 11 C)
Thermal Effects, Proceedings, Second ASCE Specialty
The T = 44 C having hot interior and cold exterior is
Conference on Structural Design of Nuclear Plant Facilities
expected to produce thermal stresses that are tensile on the
(New Orleans, 1975), V. 1-B, American Society of Civil
exterior faces of all components. These stresses will add to
Engineers, New York, 1976, pp. 1292-1319.
the existing exterior face tensile stresses due to the mechanical
Office of the Federal Register, 2002, Quality Assurance
loads. Hence, the LT and a values are arrived at from the
Program Requirements for Nuclear Power Plants, Title 10
mechanical load moment diagram in Fig. A.1.
of the Code of Federal Regulations, Part 50, Appendix B.

APPENDIX AEXAMPLES IN METRIC Component End LT /L, m/m a/LT , m/m


A.1Frame design example from 3.5 AB A 3.6/6.1 = 0.59 0
The continuous frame shown in Fig. A.1 is given with all
AB B 0.59 1
components 305 mm wide x 610 mm thick and 75 mm cover
BC B (1.6 + 1.0)/9.1 = 0.29 1.6/2.6 = 0.61
on the reinforcement. The load combination to be considered
BC C 0.29 1.0/2.6 = 0.39
is U = D + L +To +Ess .
CD C 5.2/6.1 = 0.86 1
The mechanical loading consists of
CD D 0.86 0

WD = 5.9 kN/m
All components are the end-cracked type. Figures 3.4
through 3.6 are used to obtain the coefficients ks and CO,
WL = 9.9 kN/m which are shown in Table A.1.
The expressions from Section 3.4 for fixed-end moment
on component BC, and a lateral load of 17 kN at joint C due (FEM) are evaluated as
to Ess.
The thermal gradient To results from 54 C interior and 2
E c Tbt k s
10 C exterior temperature. Thus, Tm = (54 + 10)/2 = 32 C. 1. T FEM = ------------------------
- ---- ( 1 CO )
The base temperature Tb is taken as 21 C. For this condition, 12 2
Tm Tb = 32 C 21 C = +11 C and T = 44 C (hot interior,
4 6 2
cold exterior). ( 2.15 10 ) ( 9 10 ) ( 44 ) ( 305 ) ( 610 ) k
= ------------------------------------------------------------------------------------------------- ----s ( 1 CO )
The material properties are fc = 21 MPa and Ec = 2.15 12 2
104 MPa, fy = 420 MPa and Es = 2 105 MPa; and = 9
106 mm/mm/C. Also, n =Es /Ec = 9.3. T FEM = 81ks(1 CO)/2 m-kN
The reinforcement in the frame consists of two No. 25 bars
at each face in all components. This results in = As /bd = (2 Ec Ig
- ( ) ( k s ) ( 1 + CO )
2. Tm Tb FEM = ---------
510)/(305[610 75]) = 0.0063 and n = 9.3(0.0063) = L
2
0.059. = fy / fc = 0.0063(420/21) = 0.126; and Ku = fc (1
0.59) = 21 0.126(1 0.59 0.126) = 2.4. The section 4 3
( 2.15 10 ) ( 305 ) ( 610 )
capacity is Mu = KuF = (0.9)(2.4) (305)(535)2/1000 = = ------------------------------------------------------------ ( ) ( k s ) ( 1 + CO )
2
189,000 m-N. ( 6100 )
349.1R-34 ACI COMMITTEE REPORT

Table A.1Cracked frame coefficients and thermal moments on components


Thermal FEMs, m-kN
Distributed thermal
T-44 Tm Tb = 11 Distributed thermal moments and mechanical
L T /L a/L T ks Ky /L *
Component End CO DF FEM FEM Total FEM moments, m-kN moments, m-kN
AB A 0.59 0 3.40 0.41 0.17 1.0 +81.5 6.5 +75.0 +48.8 22.1
AB B 0.59 1 2.00 0.70 0.10 0.56 24.4 4.6 29.0 54.1 157
BC B 0.29 0.61 2.40 0.43 0.08 0.44 +56.5 0 +56.5 +54.1 +157
BC C 0.29 0.39 2.65 0.38 0.088 0.48 66.7 0 66.7 50.2 113
CD C 0.86 1 1.90 0.57 0.095 0.52 +33.2 +5.0 +38.2 +50.2 +113
CD D 0.86 0 2.39 0.47 0.120 1.0 51.4 +5.9 45.5 45.0 34.8
*Corrected for sidesway.
Notes: DFi = (ksiEiIgi/Li)/(ksiEiIgi/Li); symbols with component i are indicated with subscript i.

Fig. A.1Uncracked moment frames (m-N).

Tm Tb FEM = 3.3()(ks)(1 CO) m-kN Ec Ig


- ( 8.19 )
= ---------
3
3. = total unrestrained change of length of Component BC L
= (Tm Tb)L Shear stiffness at C

= (9 106)(11)(9100) Ec Ig
- [ k sC ( 1 + CO C ) + k sD ( 1 + CO D ) ]
= ---------
3
= 0.90 mm L

Distribute 0.90 mm to Ends B and C of Components AB Ec Ig


and CD, respectively, in inverse proportion to the shear - [ 1.90 ( 1.57 ) + 2.38 ( 1.47 ) ]
= ---------
3
L
stiffness at these ends.

Shear stiffness at B Ec Ig
- ( 6.48 )
= ---------
3
L
Ec Ig
- [ k sA ( 1 + CO A ) + k sB ( 1 + CO B ) ]
= ---------
3
L Ec Ig
Sum of shear stiffness at B and C = ---------
3
- (14.67)
L
Ec Ig
- [ 3.4 ( 1.41 ) + 2.00 ( 1.70 ) ]
= --------- B = 0.90 mm (6.48/14.67) = 0.90 0.44 = 0.40 mm
3
L C = 0.90 mm (8.19/14.67) = 0.90 0.56 = 0.50 mm
DESIGN FOR THERMAL EFFECTS ON NUCLEAR POWER PLANT STRUCTURES 349.1R-35

Fig. A.2Final frame moment (m-N).

To demonstrate the effect of cracking on the thermal b = 305 mm, t = 914 mm, d = 831 mm, d = 84 mm, As =
moments, the fixed-end thermal moments for the uncracked 1290 mm2, As = 1935 mm2, Es = 200,000 MPa, Ec =
frame are obtained from the final expressions in 1 and 2 27,600 MPa, = 0.2, N = 222,000 N compression, M =
using ks = 4, CO = 1/2, and = (1/2)(Tm Tb)L , with L 136 106 mm-N, T = 44 C
being the length of Component BC. A moment distribution is
performed, and the resulting distributed moments are added to n = Es/Ec = 200/27.6 = 7.25, e = M/N = 1.36 108/2.22
the mechanical moments. The combined moments are shown 105 = 610 mm, = 1290/(305 914) = 0.0046, n = 0.0046
in Fig. A.1 for comparison with the cracked frame moments. 7.25 = 0.033, = 1935/(305 914) = 0.0069, n = 0.0069
The fixed-end thermal moments for the cracked frame are 7.25 = 0.050
obtained using the aforementioned values for B and C and
by referring to Table A.1 for ks and CO. These fixed-end Cracked thermal moment MT
moments and the resulting distributed thermal moments are e/d ratio: e/d = 610/831 = 0.73 > 0.70; therefore, Section 4.2
given in Table. A.1. The distributed thermal moments results should be used.
include the effect of sidesway, which occurs because the Because t/d = 1.1 and d/d = 0.10, Fig. 4.2 should be used.
frame in unsymmetrically cracked. Because , however, interpolate between the two curves.
Combined loadsThe final frame moments are shown From Fig. 4.2 for n = 0.050, (1 )MT /(bd2EcT)
in Table A.1 and Fig. A.2. These can be compared with Fig. A.1 should be read as
to see the effect of the cracking reduction of thermal
moments. 0.0303 for = 0
Although not shown, the component axial forces were
evaluated to confirm that section cracking still corresponds 0.0320 for =
to the pure bending condition of Assumption 3. Recall that e/d
should be at least 0.5 for this condition. For Components AB Interpolate for = 0.0046
and CD, the axial forces result primarily from the mechanical
loads and are compressive. For Component BC, the axial (1 )MT /(bd2EcT) = 0.0303 + (0.0320
force is compressive and includes the compression due to the 0.0303)(0.0045/0.0069) = 0.0314
11 C increase on the component.
(1 )MT = [0.0314(305)(831)2(27,600) (9.9 106)(44)]/
A.2Design examples from 4.4
A.2.1 Example 1: Compressive N and |e/d| > 0.70, = 1000 = 79.5 106 mm-N
349.1R-36 ACI COMMITTEE REPORT

Table A.2Summary of results for Examples 1, 2, 3, and 4


Temperature Compression Tension Cracked thermal Total moment
Axial force Moment M, differential reinforcement
2
reinforcement moment M ,
2 T (M + MT),
Example N, kN Type m-kN T, C As , mm As , mm m-kN m-kN e/d Use figure
1 222 Compression 136 44 1290 1935 99.4 236 0.733 > 0.70 Use Fig. 4.2
2 222 Tension 136 44 1935 1935 101 238 0.733 > 0.70 Use Fig. 4.2
3 445 Compression 136 44 1935 1935 128 265 0.37 < 0.70 Use Fig. 4.4,
0.37 > 0.25 4.5, and 4.9

4 267 Tension 136 44 1935 1935 101 238 0.612 < 0.70 Use Fig. 4.2
0.612 > 0.575 and 4.9

MT = 99.4 106 mm-N or 99.4 m-kN N = 445,000 N compression, M = 136 106 mm-N, T =
44 C,
Check using equations
e = 136,000,000 mm-N
----------------------------------------------- = 305 mm
445,000 N
2 e/d = 305/831 = 0.37 < 0.70; therefore, Fig. 4.2 should not
kN=0 = ( 2n + n ) + 2 [ 2n ( d d ) + n ]
be used.
(2n + n) = [(2 0.033 + 0.050)2 + 2(2 0.033 0.10 From Fig. 4.9 for n = 0.05, lower limit on |e/d| should be
+ 0.050)]1/2 (2 0.033 + 0.050) = 0.239 read as 0.25. Because 0.37 > 0.25, Fig. 4.4 and 4.5 should be
used.
(1 )MT /(bd2EcT) = 0.152k3 + 1.818n(d/d k)
(d/d) + 0.909n(l k) = 0.152(0.239)3 + 1.818(0.034) N/(bdEcT) = 445,000/(305 831 27,600 9.9 106
(0.1 0.239)(0.10) + 0.909(0.050) (1 0.239) = 0.0317 44) = 0.145

MT = 99.4 m-kN(0.0317/0.0314) = 100 m-kN 99.4 m-kN, For N/(bdEcT) = 0.145 and e/d = 0.37, find MT. By
OK interpolation between e/ds:
n = 0.04 (Fig. 4.4) at e/d = 0.37; 0.035 by interpolation
Concrete and reinforcement stresses are calculated from a n = 0.06 (Fig. 4.5) at e/d = 0.37; 0.046 by interpolation
cracked section investigation with N = 222 kN compression For n = 0.05, (1 )MT /bd2EcT =1/2(0.035 + 0.046)
and M = 236 m-kN at section centerline. = 0.0405
A.2.2 Example 2: Tensile N and |e/d| > 0.70, = MT = 0.0405(1.25)(305)(831)2(27,600)(9.9 106)(44) = 128
b = 305 mm, t = 914 mm, d = 831 mm, d = 84 mm, As = As = 106 mm-N or 128 m-kN
1935 mm2, Es = 200,000 MPa, Ec = 27,600 MPa, = 0.2, N Concrete and reinforcement stress are calculated from a
= 222,000 N tension, M = 136 106 mm-N, and T = 44 C cracked section investigation with N = 445 kN compression
and M = 265 m-kN at section centerline.
n = 7.25, e = 1.36 108/2.22 105 = 610 mm, = = A.2.4 Example 4: Tensile N and |e/d| < 0.70
0.0069, n = n = 0.050 Same as Example 3, except N = 267,000 N tension.
Cracked thermal moment MT e = 136 106 mm-N/(267,000 N) = 509 mm, e/d = 509/
e/d = 610/831 = 0.73, |e/d| = 0.73 > 0.7; therefore, 831 = 0.612 or |e/d|= 0.612
Fig. 4.2 should be used. From Fig. 4.9 for n = 0.05, the lower limit on |e/d| is
From Fig. 4.2 with n = 0.050 and = curve, 0.575 for the tensile N case. Because 0.612 > 0.575, Fig. 4.2
should be used.
From Fig. 4.2, with n = 0.05, read (1 )MT /(bd2EcT)
(1 )MT /(bd2EcT) = 0.032 = 0.032
MT = 0.032(1.25)(305)(831)2(27,600)(9.9 106)(44) =
MT = (1.25)(305)(831)2(27,600)(9.9 106)(44)(0.032) = 101 106 mm-N or 101 m-kN
101 106 mm-N or 101 m-kN. Concrete and reinforcement stresses are calculated from a
Concrete and reinforcement stresses are calculated from a cracked section investigation with N = 276 kN tension and M
cracked section investigation with N = 222 kN tension and M = 238 m-kN at section centerline.
= 238 m-kN at section centerline. A.2.5 Summary of results for Examples 1,2, 3, and 4
A.2.3 Example 3: Compressive N and |e/d| < 0.70 Table A.2 shows the summary of results for Examples 1,
Same section as Example 2. 2, 3, and 4.

American Concrete Institute
Advancing concrete knowledge

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 and the ACI Materials Journal, and Concrete International.

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

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

American Concrete Institute


38800 Country Club Drive
Farmington Hills, MI 48331
U.S.A.
Phone: 248-848-3700
Fax: 248-848-3701
www.concrete.org
Reinforced Concrete Design for Thermal Effects
on Nuclear Power Plant Structures

The AMERICAN CONCRETE INSTITUTE


was founded in 1904 as a nonprofit membership organization dedicated to public
service and representing the user interest in the field of concrete. ACI gathers and
distributes information on the improvement of design, construction and
maintenance of concrete products and structures. The work of ACI is conducted by
individual ACI members and through volunteer committees composed of both
members and non-members.

The committees, as well as ACI as a whole, operate under a consensus format,


which assures all participants the right to have their views considered. Committee
activities include the development of building codes and specifications; analysis of
research and development results; presentation of construction and repair
techniques; and education.

Individuals interested in the activities of ACI are encouraged to become a member.


There are no educational or employment requirements. ACIs membership is
composed of engineers, architects, scientists, contractors, educators, and
representatives from a variety of companies and organizations.

Members are encouraged to participate in committee activities that relate to their


specific areas of interest. For more information, contact ACI.

www.concrete.org


American Concrete Institute
Advancing concrete knowledge