This commentary discusses some of the background and consideration of Chapter 3Construction requirements, p. 307R3
Committee 307 in developing the provisions contained in Design and 3.3Strength tests
Construction of Reinforced Concrete Chimneys (ACI 30798). The
changes from the previous edition are noted. Two appendices provide the
3.4Forms
derivation of the equations for nominal strength and temperature stresses. 3.5Reinforcing placement
Keywords: chimneys; compressive strength; concrete construction; earth Chapter 4Service loads and general design
quakeresistant structures; formwork (construction); foundations; high criteria, p. 307R3
temperature; linings; loads (forces); moments; openings; precast concrete; 4.1General
quality control; reinforced concrete; reinforcing steels; specifications; 4.2Wind loads
static loads; strength; structural analysis; structural design; temperature; 4.3Earthquake loads
thermal gradient; wind pressure.
4.5Deflection criteria
CONTENTS
Introduction, p. 307R2 Chapter 5Design of chimney shell: Strength
method, p. 307R7
5.1General
Chapter 1General, p. 307R3 5.3Required strength
1.1Scope
5.4Design strength
1.4Reference standards
5.5Nominal moment strength
5.6Design for circumferential bending
Chapter 2Materials, p. 307R3
Chapter 6Thermal stresses, p. 307R9
6.1General
ACI Committee Reports, Guides, Standard Practices, and 6.2Vertical temperature stresses
Commentaries are intended for guidance in planning, de
signing, executing, and inspecting construction. This doc Appendix ADerivation of equations for nominal
ument is intended for the use of individuals who are strength, p. 307R9
competent to evaluate the significance and limitations
of its content and recommendations and who will accept Appendix BDerivation of equations for
temperature stresses, p. 307R13
responsibility for the application of the material it con
tains. The American Concrete Institute disclaims any and Appendix CReferences, p. 307R14
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 307R98 supercedes ACI 307R95 and became effective November 1, 1998.
documents. If items found in this document are desired Copyright 1998, American Concrete Institute.
All rights reserved including rights of reproduction and use in any form or by any
by the Architect/Engineer to be a part of the contract doc means, including the making of copies by any photo process, or by electronic or
uments, they shall be restated in mandatory language for mechanical device, printed, written, or oral, or recording for sound or visual reproduc
tion or for use in any knowledge or retrieval system or device, unless permission in
incorporation by the Architect/Engineer. writing is obtained from the copyright proprietors.
307R1
307R2 ACI COMMITTEE REPORT
The Committee had extensive discussion on the question The load factor for acrosswind forces has been re
of including strength design in the 1979 specification. The duced from 1.40 to 1.20.
decision to exclude it was based on the lack of experimental The vertical strength reduction factor has been re
data on hollow concrete cylinders to substantiate this form of duced from 0.80 to 0.70.
analysis for concrete chimneys. However, the Committee It should be noted that the reduced load factors must be
continued to consider strength design and encouraged used in concert with the revised strength reduction factor and
experiments in this area. the wind and seismic loads specified in ASCE 795.
Shortly after the 1979 edition was issued, the Committee The foregoing revisions are discussed in more detail in the
decided to incorporate strength design provisions and update following commentary.
the wind and earthquake design requirements. Finally, the Committee believes that the ACI 307 standard is
The 1988 edition of ACI 307 incorporated significant particularly unique in its inclusion of specific procedures to
changes in the procedures for calculating wind forces as well calculate wind and seismic forces on chimneys. Consequently,
as requiring strength design rather than working stress. The the Committee feels that the previous Commentary regarding
effects of these and other revisions resulted in designs with these subjects should be retained wherever possible.
relatively thin walls governed mainly by steel area and, in Similarly, the Committee believes that the Commentary
many instances, acrosswind forces. regarding the assumptions and procedures for strength de
The subject of acrosswind loads dominated the attention sign and other recent revisions should also be retained for
of the Committee between 1988 and 1995 and the 1995 stan reference.
dard introduced modified procedures to reflect more recent
A chapterbychapter commentary follows.
information and thinking.
Precast chimney design and construction techniques were
CHAPTER 1GENERAL
introduced as this type of design became more prevalent for 1.1Scope
chimneys as tall as 300 ft (91.4 m). The scope of the 1995 standard was expanded to include
The subject of noncircular shapes was also introduced in 1995. precast chimney shells. Additional information may be
However, due to the virtually infinite array of possible configu found in PCI manuals.1,2 Warnes3 provides further guide
rations, only broadly defined procedures were presented. lines on connection details for precast structures. Additional
Because of dissimilarities between the load factors re information is given in ACI 550R, Design Recommenda
quired by the ACI 307 standard and ACI 318, the Committee tions for Precast Concrete Structures.
added guidelines for determining bearing pressures and
loads to size and design chimney foundations. 1.4Reference standards
In summary, the following highlights the major changes The year of adoption or revision for the referenced stan
that were incorporated into the 1995 standard: dards has been updated.
Modified procedures for calculating acrosswind loads;
Added requirements for precast concrete chimney col CHAPTER 2MATERIALS
umns; No changes of note have been made in this section.
Added procedures for calculating loads and for design
ing noncircular chimney columns; CHAPTER 3CONSTRUCTION REQUIREMENTS
Deleted exemptions previously granted to smaller chim 3.3Strength tests
neys regarding reinforcement and wall thickness; and Requirements for testing precast concrete units were add
Deleted static equivalent procedures for calculating ed in the 1995 standard.
earthquake forces.
3.4Forms
Shear transfer within precast concrete shells must be con
Synopsis of current revisions
Revisions to the ASCE 795 standard relating to wind and sidered in design especially if the structure has vertical as
seismic forces required that several changes be made to the well as horizontal construction joints.
1995 edition of ACI 307. The following highlights the
changes incorporated into the current standard: 3.5Reinforcing placement
The size, spacing, and location of vertical cores within pre
Sitespecific wind loads are calculated using a 3sec
cast concrete chimney shells will be determined by geometry
gust speed determined from Fig. 61 in ASCE 795 in
and steel area requirements. It is important that the design of
stead of the previously used fastestmile speed.
precast chimneys comply with the minimum spacing require
Sitespecific earthquake forces are calculated using the
ments of ACI 318 when arranging reinforcement within the
effective peak velocityrelated acceleration contours
cores to permit proper bar splicing and concrete placement.
determined from Contour Map 92 in ASCE 795 in
stead of previously designated zonal intensity.
CHAPTER 4SERVICE LOADS
The vertical load factor for alongwind forces has been AND GENERAL DESIGN CRITERIA
reduced from 1.7 to 1.3. 4.1General
The vertical load factor for seismic forces has been re The 1995 Committee reevaluated the previous exemp
duced from 1.87 to 1.43. tions regarding twoface reinforcement and minimum wall
307R4 ACI COMMITTEE REPORT
thickness for chimneys 300 ft (91.4 m) or less in height and Eq. (41) permits the mean hourly speed at height z to be
less than 20 ft (6.1 m) in diameter. Recent information has determined from the basic design speed that is the 3sec
indicated that twoface circumferential reinforcement is nec gust speed at 33 ft (10 m) over open country. The conver
essary to minimize vertical cracking due to radial wind pres sion is based on the relationship recommended by Hollister.5
sures and reverse thermal gradients due to the effects of solar The specified wind loads presume that the chimney is located
heating. Reverse thermal gradients due to solar heating may in open country. In rougher terrains the overall loads will
be more pronounced when the air space between the column be reduced, but for a tall chimney (height on the order of
and lining is purged by pressurization fans and gas tempera 650 ft [198 m]) the reduction is not likely to exceed 20
tures are low. Further, the 1995 Committee believed that percent.
twoface reinforcement should be required in all chimney VR in Eq. (41) is the product of the square root of the im
columns, regardless of size, considering the aggressive envi portance factor I and V, the basic wind speed as charted and
ronment surrounding chimneys. defined in ASCE 795. It should be noted that I can be used
4.1.3.1A minimum wall thickness of 8 in. (200 mm) to vary probability, as well as to classify the importance of
(7 in. [180 mm] if precast) is required to provide for proper the structure. The Committee believes that all chimneys
concrete placement within and around two curtains of rein should be designed to be part of an essential facility classi
forcement. fied as a Category IV structure. The importance factor of
4.1.3.2The 1995 Committee expressed concern re 1.15 for Category IV buildings and structures corresponds to
garding edge buckling of relatively thin walls through re a mean recurrence interval of 100 years. Additional informa
gions where tall openings are present. The simplified tion can be found in ASCE 795.
procedure given in this section will give approximately the The simplified provisions of this standard do not preclude
same results as the procedures of Chapter 10.10 of ACI 318. the use of more detailed methods, and the results of a full dy
If jamb buttresses are used, it is recommended that they be namic analysis employing accepted approaches and recog
poured homogeneously with the section or adequately tied to nizing the flow profile and turbulence levels at a specific site
ensure composite action. may be used in place of the standard provisions. The approx
4.1.7.2Foundation design: The loading combinations imate methods have, however, been tested against more de
in the 1995 version of this article have been deleted. The tailed analyses, using probablistic6,7 and deterministic8
psuedobearing pressure/pile loads shall be computed by approaches. These methods yielded acceptable results.
multiplying the unfactored dead and axial bending loads by 4.2.2 Alongwind loadsThe recommended drag coeffi
their appropriate load factor from Sections 5.3.1 and 5.3.2. cients are consistent with slender chimneys [h/d(h) > 20]
with a relative surface roughness on the order of 104 to 105.
4.2Wind loads Some reduction in the drag coefficient Cdr with decreasing h/
4.2.1 GeneralThe basic wind speed V in the current d(h) can be expected but unusually rough (e.g., ribbed) chim
standard has been revised from fastestmile to a 3sec neys would have higher values of Cdr. The variations of Cdr
gust speed to reflect the changes published in ASCE 795. with roughness and aspect ratio are discussed by Basu9 and
Eq. (41) has been modified accordingly. In Eq. (41), 1.47 Vickery and Basu.10
converts wind speed from mph to ft/sec and 0.65 converts 3 The total load per unit length is computed as the sum of
sec gust speed to a mean hourly speed. The revised power the mean component w(z) and the fluctuating component
law coefficient 0.154 (as an approximation of 1/6.5) comes w(z). The dynamic component was evaluated using a
from Table C66 in the Commentary to ASCE 795, for Ex slightly modified form of the gust factor approaches de
posure C and for flexible or dynamically sensitive structures; scribed by Davenport,11 Vickery,6 and Simiu.12 The base
the increase in the exponent increases the calculated pres moment is evaluated using the gust factor approach but the
sures over the chimney height for the same speed. loads producing this moment are approximated by a trian
The 3sec gust speed is always higher than the previous gular distribution rather than a distribution matching the
ly specified fastestmile speed. A fastestmile wind mean. Eq. (46) is a simple empirical fit to values of Gw
speed may be converted to a 3sec gust speed for normal computed as before for a structural damping of 1.5 percent
speeds of interest in chimney design using the following of critical. Except for referencing V as the 3sec gust speed,
equation no revisions have been made to the procedures for calculat
ing alongwind loads.
3sec gust V = 1.0546 (fastest mile V + 11.94) The natural period of the chimney may include the effect
of foundation springs.
The relationship between 3sec gust speed and any other 4.2.3 Acrosswind loadsNo revisions have been made to
averaging time can be found in texts such as Wind Effects on the procedures for calculating acrosswind forces. However,
Structures4 by Simiu and Scanlon. Eq. (48a) has been rewritten for simplification and several
The procedure was determined from simplified dynamic typographical errors were corrected.
analyses that yield equivalent static load distributions. This The 1995 Committee had numerous user comments and
approach requires that a wind speed averaged over a period discussions regarding the procedures included in the 1988
on the order of 20 min to 1 hr be used as a basis for design. standard for acrosswind forces. Virtually all of the com
COMMENTARY ON REINFORCED CONCRETE CHIMNEYS 307R5
mentators felt that the 1988 procedures were unduly conser using the procedures of Section 4.2.3.3 of ACI 30788 or by
vative, especially in the absence of any record of structural the general approach put forth by Vickery.13
failure. As a result of these discussions, and with the avail It should be noted, however, that the procedures for deter
ability of new data and fullscale observations, the proce mining shedding forces are not materially affected by the
dures for calculating acrosswind loads were extensively configuration of the lower third of the shell, which may
revised. range from plumb to any degree of taper.
A general solution for the acrosswind response of circular However, it should also be noted that noncircular shapes
chimneys with any geometry was developed by Vickery.13 may be more sensitive to acrosswind forces and may require
These procedures, based on Vickerys general solution, were analyses beyond the scope of this standard.
simplified to some extent, which requires that their applica Eq. (416) establishes a basis for increasing structural
tion be restricted to certain geometries. Similar models have damping from a minimum of 1.0 percent to a maximum of
provided the basis for vortexinduced forces incorporated by 4.0 percent when the wind speed V exceeds V(zcr). Structural
the National Building Code of Canada, and the ASME/ANSI damping of 1 percent of critical is consistent with measured
STS11992 Steel Stack Standard. values and moderate stress levels with little cracking. Damp
Circular chimneys outside the bounds of these procedures, ing of 4.0 percent, which would be permitted when V = 1.30
or where a flare or strong taper (nozzle) exists for more than V(zcr), is more consistent with damping values permitted in
one diameter near the top, may be conservatively analyzed seismic design.
307R6 ACI COMMITTEE REPORT
Eight sample chimneys were studied using the 1988 pro ty will now be established using the effective peak velocity
cedures and the 1995 procedures. Fatigue damage was also related (EPV) acceleration contours Av, as shown on Contour
considered using the procedures put forth by Vickery.13 It Map 92 in ASCE 795.
was concluded that a casebycase analysis of fatigue in cir EPVrelated acceleration is used because frequencies of
cular chimneys that would require a supplemental working concrete chimney shells are generally lower than about 3 Hz,
stress analysis was not necessary, as fatigue stresses in the and velocityrelated acceleration controls the response.
sample chimneys were within acceptable limits. Table 4.3.2(b) has been revised to reflect the changes nec
Results using the 1988 and the 1995 procedures are compared essary to relate scaling ratios to acceleration contours. Al
in Table 4.2.3. These chimneys were selected from a group of though the probability of seismic acceleration not being
projects where the aspect ratio h/d is at or near 10, where peak exceeded has been revised from 80 to 90 percent, the re
excitation is normally found. Note that for Chimneys 7 and 9 the sponse spectrum shown in Fig. 4.3.2 has not been changed,
critical wind speed exceeds the design wind speed, permitting
since it is comparable to that given in the 1994 UBC for rock
modification of both damping [Eq. (416)] and Ma [Eq. (48a)],
and stiff (firm) soils.
which significantly reduces the base moments.
The design response spectrum provided in the standard is
4.2.3.4 Grouped chimneysInteractions between closely
spaced cylindrical objects have been studied in considerable an average elastic response spectrum, normalized for a peak
detail but virtually all the test results are for subcritical val horizontal ground acceleration of 1.00 with 5 percent of crit
ues of Reynolds Numbers and their applicability to chimneys ical damping. It represents a spectrum of 50 percent shape
is highly questionable. However, even with the scale effects bound probability level that the response of the structure dur
introduced by the inequality of the Reynolds Number, the ing an earthquake would not exceed. It is the same spectrum
wind tunnel is presently the only tool that will provide guid that has been adopted for use in the design of steel chimney
ance as to the likely magnitude of interference effects. A re liners for earthquakes by the Task Committee of the American
view of interference effects is given by Zdravkokvich.14 Society of Civil Engineers.20 To obtain the design response
Vickery13 attributes the amplification of shedding forces to spectrum, the normalized spectrum must be scaled down to
increased turbulence and additional buffeting effects, which the effective peak velocity EPV related ground acceleration.
formed the basis for revisions made to this section. The ASCE 795 map for the EPVrelated acceleration co
At centertocenter spacings s, in excess of 2 to 3 diame efficient is used in this standard. This map differs from those
ters, the prime interference effect is related to acrosswind used in the Uniform Building Code, which was based on the
excitation due to shedding. The recommendations in Section
maximum recorded intensity of shaking without regard to
4.2.3.4 are based on the results of Vickery and Daly15 and
the frequency with which earthquake shaking might occur.
were obtained at subcritical values of the Reynolds Number.
The first term in modifier (c) is an enhancement factor to ac The ASCE 795 map, on the other hand, has a more uniform
count for buffeting due to vortices shed by the upstream probability of earthquake occurrence, and is based on those
structure; the second term accounts for smallscale turbu given by the NEHRP (see Reference 21). For example, in
lence. The same reference also contains results for two cyl Zone 4 seismic area, the EPVrelated acceleration is 0.4g and
inders of different size with the upstream structure having a the probability of not exceeding this peak EPV ground accel
diameter 25 percent greater than the diameter d of the other. eration within 50 years is estimated to be 90 percent. This is
In this case the amplification of the response of the down equivalent to a mean recurrence interval of 475 years, or an
wind chimney is roughly 3.4  0.2 s/d for 4 < s/d < 12. The average annual risk of 0.002 events per year. The peak EPV
amplification of shedding for grouped cylinders has also related ground acceleration at a site can be determined either
been noted at full scale16 but the available data is not suffi by using this contour map and the recommended scale fac
cient to quantitatively validate model test results. tors given in Table 4.3.2 or from the specific seismic record
4.2.4 Circumferential bendingThe equation for the pre available at the site. It should be noted that a ductility factor
diction of the circumferential moments is based upon mea of 1.33 is built into the scale factors of Table 4.3.2. For in
sured pressure distributions.17,18 Comparative values for the stance, instead of 0.40, a scale factor of 0.30 is used for a site
bending moments as obtained from different distributions with an Av of 0.4.
are given in Reference 8. The use of a gust factor Gr in this
computation is based upon the assumption that the mean It should also be pointed out that the recommended design
pressure distribution (when expressed in coefficient form) is response spectrum is based on firm soil sites. Soil conditions
also applicable for shortduration gusts. at the firm site consist of bedrock with shear wave velocity
The increase in the loads near the tip is consistent with greater than 2500 ft/sec (762.0 m/sec) or stiff soils with de
observations19 that the drag coefficient increases significant posits less than 200 ft (61.0 m). For chimneys to be built on
ly in this region. shallow and soft or mediumstiff clays and sands, a greater
design response spectrum is anticipated. Guidelines provid
4.3Earthquake loads ed in NEHRP21 to obtain a modified design response spec
4.3.1The seismic intensity for any site within the United trum and the soilstructure interaction may be used.
States had previously been determined by the zonal map In place of a dynamic response spectrum analysis, a time
shown in Fig. 14 and 15 of ASCE 788. ASCE 795 no long history dynamic analysis is permitted, provided a reliable
er references earthquake zones. Sitespecific seismic intensi time history of earthquake ground motion is used.
COMMENTARY ON REINFORCED CONCRETE CHIMNEYS 307R7
In the design of a chimney for horizontal earthquake forces, is below a certain value. A total vertical steel ratio in the
only one horizontal direction need be considered. Unlike chimney cross section less than that per the minimum re
building structures, chimneys are generally axisymmetric, and quirement of ACI 318 for flexural members is permitted.
the orthogonal effects from two horizontal earthquakes acting Even when the maximum concrete compressive strain m
simultaneously in the two principal directions are negligible. is less than 0.003, the stress block is still considered rectan
The effect of the vertical component of the earthquake on gular. However, in these instances, the stress level is modi
the chimney has been determined to be of no design signifi fied by a correction factor called the parameter Q. See
cance. An extensive time history analysis made by the Com commentary on Section 5.5.1.
mittee shows that the vertical stresses from dead load and
horizontal seismic excitation are increased by at most a few
percent by the effects of vertical seismic excitations. This is 5.3Required strength
5.3.1The Committee noted that the fastestmile provi
principally because the psa responses to vertical and hori
zontal acceleration do not occur simultaneously. sions in the 1988 edition of ACI 307 resulted in an increase
Design based on SRSS of vertical and horizontal earthquake in wind moments of between 0 and 50 percent when com
forces will be unduly conservative. Therefore, the inclusion of pared with ACI 30779. The use of a 3sec gust wind speed
vertical seismic effects is not recommended by the Committee. results in further increases in axial bending moments, which
For cases in which the chimney lining (brick, steel, or oth are 10 to 20 percent higher than moments calculated using
er materials) is supported by the concrete chimney shell, ei fastestmile speeds. Since the Committee has no data or in
ther at the top of the chimney shell or at other intermediate formation concerning axial bending failures of chimney
points, a dynamic analysis including both concrete shell and shells designed using previously established procedures, it
liner should be used. Appropriate damping values should be was decided that the load factor for alongwind loads can be
used for the liner depending on its construction (e.g., 1.5 per safely reduced from 1.7 to 1.3 when 3sec gust wind
cent for steel liners, 4.0 percent for brick liners, and 2.0 per speeds are used. It should be noted that a wind load factor of
cent for fiber reinforced plastic liners). 1.3 is consistent with that recommended by ASCE 795.
Similarly, the Committee has determined that the wind
4.5Deflection criteria load factor for along, plus acrosswind loads can be reduced
The incorporation of the strength design method into the from 1.4 to 1.2.
standard will generally result in chimneys with thinner walls It should be noted that the vertical load factor reductions
in the lower portion and with higher deflections. The Com incorporated in the current standard must be accompanied by
mittee felt that deflections under service loads should be a decrease in the strength reduction factor from 0.80 to
checked and that the deflections of chimneys designed by the 0.70, as described in Article 5.4.1. The net effect of the revi
strength method should not vary greatly from the deflections sion to the vertical load factors, coupled with the change in
of existing chimneys designed by the working stress method. the strength factor, is relatively minor. Table 5.3.1 summa
Limiting deflections also serves to reduce the effects of sec rizes the effects of the revisions on 12 sample chimney shells
ondary bending moments. over a range of wind speeds. The geometry of the chimneys
However, the procedures in the ACI 307 1988 edition studied is as follows
were found to be too restrictive for shorter chimneys and
were modified in the 1995 standard. The deflection limit is
Chimney no. Height, ft TOD, ft BOD, ft
compared against the deflection calculated using uncracked
1 250 13.50 19.75
concrete sections and a fixed base.
Operation, access for inspection, lining type, etc., as well 2 275 28.00 28.00
as wind or earthquakeinduced deflection, should be consid 3 325 15.00 20.00
ered when establishing shell geometry. 4 375 20.00 32.00
5 425 35.00 39.00
CHAPTER 5DESIGN OF CHIMNEY SHELL: 6 485 47.67 53.50
STRENGTH METHOD 7 534 51.09 61.55
5.1General 8 545 33.00 55.00
Several significant revisions were made to this section,
9 613 73.00 73.00
most notably the load factors specified in 5.3 and the strength
reduction factor specified in 5.4. Portions of previous com 10 700 60.00 78.00
mentary are, however, retained for reference. 11 773 43.00 70.00
5.1.2 The maximum compressive strain in the concrete is 12 978 73.00 114.78
assumed to be 0.003, or the maximum tensile strain in the
steel is assumed to be the fracture limit of 0.07, whichever is 5.3.2The Committee has determined that, based on the
reached first. If the steel fracture limit is reached first, the required use of velocityrelated acceleration contours cou
maximum concrete strain computed from the linear strain di pled with a reevaluation of the ductility inherent in chimney
agram is below 0.003. This deviates from the design assump shells, a decrease in the ratio of the load factor to the strength
tions of ACI 318. For a given total vertical steel ratio, this reduction factor for earthquake forces from 2.34 to 2.04 is
may occur when the ratio of the vertical load to the moment warranted.
307R8 ACI COMMITTEE REPORT
Table 5.3.1Comparison of alongwind design forced concrete beams came to be accepted after extensive
moments* comparative study between the analytical results using this
stressstrain relationship and the test data. The acceptability
Chimney of the rectangular stress block was based on the closeness be
no. 90(3sg)/70(fm) 120(3sg)/100(fm) 150(3sg)/130(fm) tween the results of the analyses and the tests, comparing the
1 1.054 0.973 0.940 following: a) concrete compression; and b) moment of the
2 1.058 0.976 0.944 compression about the neutral axis (for a rectangular section
3 1.062 0.980 0.947
this is equivalent to the distance of the center of gravity of
4 1.065 0.983 0.950
5 1.069 0.988 0.955 the compression stress block from the neutral axis).
6 1.072 0.991 0.958 The preceding comparative study was based on the limited
7 1.073 0.993 0.960 test data available on reinforced concrete members of hollow
8 1.074 0.993 0.960 circular sections subjected to axial and transverse loads.22
9 1.079 0.998 0.965
Another special problem in arriving at the compressive
10 1.082 1.00 0.967
11 1.084 1.002 0.969 stress block for the analysis of reinforced concrete chimneys
12 1.090 1.008 0.976 was the fact that the maximum concrete compressive strain
*{Values of [1.3 M(3sg)/0.7]/1.7 M(fm)/0.8] for sample chimneys} is less than 0.003 when the fracture limit of steel is reached.
That is, the compressive stress block is not fully developed
(see commentary on Section 5.1.2). Thus, the previous at
The load factor for determining the circumferential tempts at specifying the rectangular stress block for chimney
strength required to resist wind load has not been revised, al cross sections needed to be modified.
though the reinforcement necessary to satisfy the higher mo
A numerical study was undertaken by the 1988 Committee
ments may increase up to 15 percent on largediameter
to find an equivalent rectangular stress block for the calcula
chimneys. However, the Committee believes that this addi
tion of the strength of chimney cross sections.
tional reinforcement is justified to minimize vertical crack
ing of chimney shells. For a given value of , the results of the rectangular con
crete compression stress block, expressed by dimensionless
5.4Design strength modifications of (a) and (b) previously stated, were com
5.4.1In the calculation of limitstate bending moments, pared with the corresponding results using a more exact con
allowance needs to be made for the moment caused by the crete stressstrain relationship23 given by Hognestad24 using
weight of the chimney in its deflected shape. The deflection a limiting strain of 0.003. The comparisons were made for
will be less than that calculated by standard methods due to hollow circular sections without openings and with single
the stiffening effect of the concrete in the cracked tension openings with values of of 10, 20, and 30 deg.
zone. The stiffening effect needs to be investigated further. It was concluded that for values of above 20 deg, or
The strength reduction factor for vertical strength has been
when the limiting strain of concrete is reached first, an equiv
reduced from 0.80 to 0.70. A factor of 0.70 was chosen be
alence between the two approaches is reached if the stress
cause it was found that the deadload axial stress on the gross
level of the rectangular compression block is reduced by a
area exceeds 0.10 fc in the lower portions of some sample
chimneys. The effects of this revision are discussed more factor of 0.89. For values of below about 20 deg, a further
fully in Section 5.3. correction is required, leading to the values of the parameter
The formulas are also derived for cross sections with one Q defined in Section 5.5.1.
or two openings in, or partly in, the compression zone. No re Thus the correction factor, or the parameter Q, achieves a
duction in the forces and moments due to reinforcing steel is close equivalence between the resulting values of (a) and (b)
made to allow for the reduction in the distance of the addi previously stated for the thereby corrected rectangular
tional vertical reinforcement on each side of the opening, stress block and the stress block based on the Hognestad
provided per Section 4.4.6. stressstrain relationship, yet retains the simplicity of the
rectangular stress block.
5.5Nominal moment strength 5.5.6 Due to thermal exposure of the concrete chimneys,
The formulas for the nominal moment strength of chimney
the temperature drop across the wall reduces the nominal
cross sections are obtained based on the design assumptions
strength of chimney sections. This effect is accounted for by
of ACI 318, except as modified under Section 5.1.2 of this
standard. The derivations of the formulas are given in reducing the specified yield strength of steel and specified
Appendix A. compressive strength of concrete.
The formulas are derived for circular hollow cross sections The derivation of equations is included in Appendix A.
with a uniform distribution of vertical reinforcing steel
around the circumference. 5.6Design for circumferential bending
5.5.1 The parameter QThe use of a rectangular com 5.6.2 The commentary on Section 5.5.6 applies equally to
pression stress block for rectangular and Tshaped rein this section.
COMMENTARY ON REINFORCED CONCRETE CHIMNEYS 307R9
= Es/fy t fc
APPENDIX ADERIVATION OF EQUATIONS FOR
NOMINAL STRENGTH
Equations for the nominal strength of concrete chimney = Ke t fc
sections, with and without openings, are derived in this Ap
therefore
pendix.
The factored vertical load Pu and the corresponding nom
inal moment strength Mn are expressed in dimensionless [ ( ) cos sin + sin ]
S 1 = 2 m K e t rtf c 
form, as given in Section 5.5.1 by Eq. (52) and (510), re ( 1 cos )
spectively.
Also a procedure to account for the temperature effects in or
the vertical and horizontal directions is outlined.
Forces are designated as follows: S1 = 2m Ke t rtfc Q
MDS = design moment strength of the section
S2 = 2( )t rtfy
Mn = nominal moment strength of the section
Mu = factored moment acting on the section but
P = total force in the concrete compressive
stress block t fy = t fc
Pu = factored vertical load acting on section
P, S1, S2, = moments of P, S1, S2, S3, S4 about neutral
S2 = 2( )rtt fc
S 3 , S 4 axis, respectively
S1 = tensile force where steel stress is below
yield point, from to P = 2( n1)rt 0.85fc
S2 = tensile force where steel stress is at yield
point, from to = 1.7rtfc( n1)
S3 = compressive force in steel where stress is
below yield point, from to = 1.7rtfc
S4 = compressive force in steel where stress is
at yield point, from 0 to where
= capacityreduction factor = n 1
S4 = 2trtfy S1 = m r2 tfc Ke t J1
= 2trtfc where
Pu = P + S 3 + S 4 S 1 S 2 or
Ke = Es/fy 2 2
r ( cos cos )
t = tfy/fc
S3 = 2

r ( 1 cos )
m E s t rtd
2 2
r ( cos cos )
2
2 m K e t r tf c
S1 = 2

r ( 1 cos )
m E s t rtd = 
( 1 cos )

( cos 2 cos sin + cos ) d
2 2
2
2 m Es r t t
( cos 2 cos cos + cos ) d
2 2 2
=  2 m K e t r tf c sin 2
( 1 cos )   +  2 cos sin + cos
2
= 
( 1 cos ) 2 4
2
2 m K e t r tf c sin 2 
 cos 2 cos sin +
2 2
= 
 +  2 m K e t r tf c
( 1 cos ) 2 4 = 

( 1 cos )
2 m K e t r tf c
2 [(1/2)(  ) + (1/4)(sin2 sin2) 2cos(sin sin)
= 
 + ( )cos2]
( 1 cos )
[( )cos2 2cos(sin sin) + (1/2)( ) Let
+ (1/4)(sin 2  sin 2)]
J3 = 2[ ]/(1 cos)
Let J = [ ] or
= (  )cos2 + 2 sin cos  2 cos sin J3 = [  + sin cos sin cos 4cos(sin sin)
+ (1/2)sin cos (1/2)sin cos + (1/2)( ) + 2( )cos2]/(1 cos)
or therefore
n1 = 2 =
307R12 ACI COMMITTEE REPORT
= At ultimate, effect on fc is
1
 = 
( 1 + 1 ) 1 + 1
= ratio, outside steel area to total steel area
1 1
 = 

( 1 + 1 ) 1 + 1
= ratio, inside steel area to total steel area
Find the value of a that satisfies this equation. = ratio of total area of vertical outside face reinforce
ment to total area of concrete chimney shell at sec
M about PTS, Mn = {PCB[2  (a/2)] + PCS(22  1)}t tion under consideration
1 = ratio of inside face vertical reinforcement area to
MDS = Mn Mu (A7) outside face vertical reinforcement area
Note: For compression on outside
( c 1 + 2 ) n
fy(c) = fy f STV 
 f CTV
c
fc(c) = fc = te(c  1 + 2)TxnEc
te = te Tx /t c = n(1 + 1) +
307R14 ACI COMMITTEE REPORT
2
2. PCI Design HandbookPrecast and Prestressed Concrete, Pre
[ n ( 1 + 1 ) ] + 2 n [ 2 + 1 ( 1 2 ) ] stressed Concrete Institute, 3rd Edition, 1985.
3. Warnes, C. E., Precast Concrete Connection Details for All Seismic
Zones, Concrete International, V. 14, No. 11, Nov. 1992, pp. 3644.
The derivation for the equations for the maximum horizontal
4. Simiu, E., and Scanlon, R. H., Wind Effects on Structures, 2nd edition,
stresses in concrete and steel due to a temperature drop only, John Wiley and Sons, 1986.
across the concrete wall with two layers of reinforcement, is 5. Hollister, S. C., Engineering Interpretation of Weather Bureau
similar to that for the vertical temperature stresses Records for Wind Loading on Structures, Wind Loads on Buildings and
Replace with Structures, Building Science Series, No. 30, National Bureau of Standards,
Washington, D.C., 1969, pp. 151164.
1 with 1 6. Vickery, B. J., On the Reliability of Gust Loading Factors, Wind
f CTV with f CTC Loads on Buildings and Structures, Building Science Series, No. 30,
fSTV with fSTC National Bureau of Standards, Washington, D.C., 1969, pp. 93104.
c with c 7. Vickery, B. J., and Basu, R. I., Simplified Approaches to the Evalua
tion of the AcrossWind Response of Chimneys, Journal of Wind Engi
2 with 2 neering and Industrial Aerodynamics, V. 14, Amsterdam, 1985, pp. 153
then 166.
8. Rumman, W. S., Reinforced Concrete Chimneys, Handbook of
fCTC = tecTxEc Concrete Engineering, 2nd Edition, Mark Fintel, ed., Van Nostrand Rein
hold Co., New York, 1985, pp. 565586.
9. Basu, R. I., AcrossWind Responses of Slender Structures of Circular
fSTC = te(2  c)TxEs CrossSection to Atmospheric Turbulence, PhD thesis, Faculty of Engi
neering Science, University of Western Ontario, London, Ontario, 1982.
10. Vickery, B. J., and Basu, R. I., Response of Reinforced Concrete
c = n(1 + 1) + Chimneys to Vortex Shedding, Engineering Structures, V. 6, No. 4, Guild
ford, Oct. 1984, pp. 324333.
11. Davenport, A. G., Gust Loading Factors, Proceedings, ASCE, V.
2 93, ST3, June 1967, pp. 1134.
[ n ( 1 + 1 ) ] + 2 n [ 2 + 1 ( 1 2 ) ]
12. Simiu, E.; Marshall, R. D.; and Haber, S., Estimation of Along
Wind Building Response, Proceedings, ASCE, V. 103, ST7, July 1977,
pp. 13251338.
APPENDIX CREFERENCES
13. Vickery, B., AcrossWind Loading on Reinforced Concrete Chim
C.1Recommended references neys of Circular Cross Section, Boundary Layer Wind Tunnel Report,
American Concrete Institute BLWT31993, University of Western Ontario, Dec. 1993.
30769 Specification for the Design and Construction of 14. Zdravkokvich, M. M., Review of Flow Interference Effects between
Reinforced Concrete Chimneys Two Cylinders in Various Arrangements, Journal of Fluids Engineering,
30788 Standard Practice for the Design and Construc V. 99, 1977, p. 618.
15. Vickery, B. J., and Daly, A., Wind Tunnel Modelling as a Means of
tion of CastinPlace Reinforced Concrete Predicting the Response of Chimneys to Vortex Shedding, Engineering
Chimneys Structures, V. 6, No. 4, Guildford, Oct. 1984, pp. 363368.
318 Building Code Requirements for Structural 16. Ruscheweyh, H., Problems with InLine Stacks: Experience with
Concrete FullScale Objects, Engineering Structures, V. 6, No. 4, Guildford, Oct.
1984, pp. 340343.
50554 Standard Specification for the Design and Con 17. Dryden, H. H., and Hill, G. C., Wind Pressure on Circular Cylinders
struction of Reinforced Concrete Chimneys and Chimneys, Research Paper No. 221, National Bureau of Standards,
550R93 Design Recommendations for Precast Concrete Washington, D.C., 1930. Also, NBS Journal of Research, V. 5, Sept. 1930.
Structures 18. ASCE Task Committee on Wind Forces, Wind Forces on Struc
tures, Transactions, ASCE, V. 126, Part II, 1961, pp. 11241198.
American Society of Civil Engineers
19. Okamoto, T., and Yagita, M., Experimental Investigation Flow
ASCE 788 Minimum Design Loads for Buildings and Past a Circular Cylinder of Finite Length Placed Normal to a Uniform
Other Structures (formerly ANSI A58.1) Stream, Bulletin, Japan Society of Mechanical Engineers (Tokyo), No.
ASCE 795 Minimum Design Loads for Buildings and 16, 1973, p. 805.
20. Task Committee on Steel Chimney Liners, Design and Construction
Other Structures of Steel Chimney Liners, American Society of Civil Engineers, New York,
1975, 226 pp.
American Concrete Institute 21. NEHRP 1994 Recommended Provisions for the Development of
P.O. Box 9094 Seismic Regulations for New Buildings Prepared by the Building Seismic
Safety Council.
Farmington Hills, Mich. 483339094 22. Mokrin, Z. A. R., and Rumman, W. S., Ultimate Capacity of
Reinforced Concrete Members of Hollow Circular Sections Subjected to
American Society of Civil Engineers Monotonic and Cyclic Bending, ACI JOURNAL, Proceedings V. 82, No. 5,
1801 Alexander Bell Drive Sept.Oct. 1985, pp. 653656.
23. Rumman, W. S., and Sun, R. T., Ultimate Strength Design of
Reston, Va. 20191 Reinforced Concrete Chimneys, ACI JOURNAL, Proceedings V. 74, No.
4, Apr. 1977, pp. 179184.
C.2Cited references 24. Hognestad, E., Study of Combined Bending and Axial Load in
1. PCI Manual for Structural Design of Architectural Precast Concrete, Reinforced Concrete Members, Bulletin No. 399, Engineering Experi
Prestressed Concrete Institute, 1977. ment Station, University of Illinois, Urbana, 1951, 128 pp.