See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/276212494

Strain Limits vs. Reinforcement Ratio Limits – A Collection of New and Old
Formulas for the Design of Reinforced Concrete Sections

Article  in  Case Studies in Structural Engineering · May 2015

DOI: 10.1016/j.csse.2015.05.001

CITATIONS READS

0 165

1 author:

Carlos E. Orozco
University of North Carolina at Charlotte
37 PUBLICATIONS   198 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Finite Element Methods Applied to Viscoelasticity Problems View project

Using Modern Computer Languages for Engineering View project

All content following this page was uploaded by Carlos E. Orozco on 20 July 2015.

The user has requested enhancement of the downloaded file.

Accepted Manuscript

Strain Limits vs. Reinforcement Ratio Limits – A Collection of New and Old
Formulas for the Design of Reinforced Concrete Sections

Carlos E. Orozco

PII: S2214-3998(15)00007-7
DOI: http://dx.doi.org/10.1016/j.csse.2015.05.001
Reference: CSSE 17

To appear in:

Please cite this article as: C.E. Orozco, Strain Limits vs. Reinforcement Ratio Limits – A Collection of New and
Old Formulas for the Design of Reinforced Concrete Sections, (2015), doi: http://dx.doi.org/10.1016/j.csse.
2015.05.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
 1 Strain Limits vs. Reinforcement Ratio Limits – A Collection of
2 New and Old Formulas for the Design of Reinforced Concrete
3 Sections
4
5 Carlos E. Orozco
6 Associate Professor, Department of Engineering Technology
7 University of North Carolina at Charlotte,
8 9201 University City Boulevard, Charlotte, NC 28223.
9 E-mail: ceorozco@uncc.edu, Phone: (704) 687-5054, Fax: (704) 687-6577
10
11
12
13 Abstract
14 This paper presents a formulation for the design of reinforced concrete flexural members. The
15 formulation yields exactly the same results as the current American Concrete Institute (ACI)
16 design approach but it is based entirely on the concept of reinforcement ratios. This is in contrast
17 to the current ACI approach which relies on strain limits [1]. A formulation based on
18 reinforcement ratios is simpler and more intuitive and therefore has important pedagogical
19 advantages. The formulation presented here can be thought of as an attempt to reconcile the new
20 approach to design introduced by the ACI code in 2002, with the traditional approach to design
21 that was in use from 1963 to 2002. The traditional approach to design of reinforced concrete
22 sections uses the concept of reinforcement ratios. The new ACI approach, referred to here as the
23 unified design method (UDM), requires consideration of rather cumbersome strain limits and/or
24 geometric strain relationships. In this paper, it is shown that the UDM approach can be
25 formulated much in the same way as the traditional approach, as long as a series of formulas
26 involving reinforcement ratios are introduced. These formulas are presented in this paper. Many
27 of them are well known, but some are new. In particular, a new formula for the compression-
28 controlled reinforcement ratio limit, and a new direct procedure for the design of transition-zone
29 sections are presented. The formulation presented in this paper should prove useful both for the
30 instructor in the classroom, and for the practicing structural engineer. Derivation details for many
31 of the formulas in the paper are given and several numerical examples to illustrate their use are
32 provided at the end.
33
34 Keywords: Reinforced Concrete, Unified Design Method, Reinforcement Ratios, Compression
35 Controlled Sections, Transition Zone Sections, Tension Controlled Sections.
Page 1 of 22
36
37 1. Introduction
38 In 2002, the ACI introduced a new approach to the design of structural members subject to
39 bending. This change was introduced through the ACI-318-02 publication [2]. The motivation
40 for the change was to achieve uniformity between the design procedures used for the design of
41 pre-stressed, reinforced, and compression members [3]. As a consequence of this, the new
42 approach to design has been termed unified design method (UDM) by some authors [4, 5]. We
43 will adopt this name in this paper. The UDM introduced the ideas of tension-controlled,
44 compression-controlled, and, transition-zone beam sections. This characterization of reinforced
45 concrete sections requires consideration of strain limits and/or geometric strain relationships. In
46 the author’s opinion, this characterization lacks intuitive and pedagogical appeal. The traditional
47 approach to design of reinforced concrete sections relies instead on the concept of reinforcement
48 ratios. The traditional approach can still be used (as of the ACI-318 2011 edition) but it has been
49 relegated to an appendix in the ACI code; and it is likely to disappear from it in subsequent
50 editions [6]. The objective of this paper is to introduce a series of formulas for the design of
51 flexural members that relies entirely on reinforcement ratios.
52
53 1.1. Tension-Controlled, Transition-Zone, and Compression-Controlled Sections.
54 Fig. 1 shows the usual representation of the flexural stresses and strains in a typical reinforced
55 concrete section. The notation and nomenclature are the usual, with c representing the depth of
56 the neutral axis of the section; d representing the distance from the top of the section to the
57 centroid of the main reinforcement; dt the distance from the top of the beam to the bottom layer
58 of the tensile reinforcement; a, the depth of the compression block; As, the area of the main
59 reinforcement; and fs, the stress in the tension steel.
60 Referring to Fig. 1b, we observe that,
c ε cu
=
61 d t ε cu + ε t (1)
62 On the other hand, from the stress diagram (Fig. 1c), we get the familiar expression for the depth
63 of the compression block a, as:
As f s
a=
64 0.85 f c'b (2)

Page 2 of 22
65 Comparing (1) and (2), recalling that a = β1c and that the reinforcement ratio is defined as
As
66 ρ≡ , (see e.g., [7]) we conclude that,
bd

0.85β1 f c' ε cu dt
ρ=
fs ε cu + ε t d
67 (3)
68 Eq. (3) is a completely general relationship for singly reinforced concrete beam sections. This
69 equation can be specialized to fit a number of important situations as it will be shown below.
ε cu
C
c a

h d dt d

εs T = As fs
εt
b As
70
71
72 a) b) Strains c) Stresses
73
74 Figure 1. Schematic Representation of Stresses and Strains in a Typical Reinforced Concrete
75 Section.
76
77
78
79 1.2 Tension-Controlled Sections
80 The ACI-318-02 defines a tension-controlled section as a section such that the strain ε t in the
81 lowermost layer of steel is greater than or equal to 0.005. This is to ensure that the main steel
82 yields well before the concrete crushes, providing enough ductility for the section even for
83 seismic zones. This definition is inspired by the yield strain of grade 60 steel (G60 for short1)

1
The grade is equal to the yield strength of the steel in the corresponding units. For instance, in metric units G420

steel has a yield strength of 420 MPa (i.e., f y = 420 MPa). This steel is practically equivalent to G60 steel which

has a nominal yield stregth of 60 ksi (i.e., f y = 60 ksi).

Page 3 of 22
84 which is approximately equal to 0.0022. (The definition of tension-controlled section is the same
85 for other grades of steel despite the fact that ε y is different: for G75 steel for instance, ε y =

86 0.0026). The European code (known as EC2) has similar provisions to guarantee adequate
87 ductility of the sections (see e.g., [8, 9]).
88 Now, by substituting ε cu = 0.003 and ε t = 0.005 in (1), one obtains,
c
89 = 0.375 (4)
dt
90 Eq. (4) is the most commonly used equation to test for tension-controlled sections in the
91 literature (see e.g., [7]). It is the purpose of this paper to replace expressions of the type (4) with
92 equivalent expressions in terms of the reinforcement ratio ρ . To obtain the tension-controlled
93 limit, it suffices to substitute ε t = 0.005 and f s = f y in (3). The expression for the tension-

94 controlled reinforcement ratio is then obtained as3:

2.55β1 f c' d t
95 ρ tcl = (5)
8 fy d

96 Eqn. (5) assumes that the steel at a depth d from the top of the beam (i.e., at the level of the
97 centroid of the main reinforcement) has also yielded when ε t = 0.0054
98
99 1.3 Compression-Controlled Sections and Transition-Zone Sections.
100 Beam sections for which ε t ≤ ε y are classified as compression-controlled sections and are not

101 permitted by the ACI code as of the 2002 edition. Occasionally however, it might still be
102 required to find the moment carrying capacity of one of these sections since it may have been
103 designed in a non-compliant way by mistake, or using a different standard. Or the section may
104 have been designed during the period from 1999 to 2002 in which the ACI code was not clear

2
Actually, ε y = 0.00207 for G60 steel (assuming a modulus of elasticity Es = 29,000 ksi).
3
Ref. [10] suggests a tension-controlled limit of 0.0066 (instead of 0.005) for high-strength steels. All of the
formulas presented in this paper can be easily modified to reflect any desired change in this or any other strain limit.
4
There is a slight possibility that this will not happen. The steel at the level of the main reinforcement will not yield

(despite the fact that εt = 0.005) in the very unusual circumstance that the ratio d/d t is less than

f y + Esε cu
. This value is approximately equal to 0.6336 for G60 steel.
Es ε cu + 0.005Es

Page 4 of 22
105 about whether these sections were allowed or not (see [6], page 127). The strength reduction
106 factor for compression-controlled sections (with ties) is φ = 0.65.
107 Beam sections for which
108 ε y ≤ ε t ≤ 0.005 (6)

109 are considered transition-zone sections. For transition-zone sections with ties, the strength
110 reduction factor φ must be linearly interpolated between the compression-controlled limit value
111 of 0.65 and the tension-controlled limit value of 0.9, as given by the next equation5:
250
112 φ = 0.65 + (ε t − 0.002) (7)
3
fy
113 Now, the compression-controlled limit corresponds to ε t = ε y . By substituting ε t = ε y ≡ in
Es
114 Eq. (1), we obtain,
c Es ε cu
115 = (8)
d t Esε cu + f y

116 which upon substitution of the values ε cu = 0.003 [6], and Es = 29, 000 ksi (200,000 MPa),
117 becomes6,
c 87
118 = (9)
d t 87 + f y

119 in US customary units, and

c 600
120 = (10)
d t 600 + f y

121 in metric units.

122 For G60 (G420) steel the compression-controlled limit corresponds to ε y = 0.00207(0.0210) . By

123 substituting ε y = 0.00207 (0.0210) and ε cu = 0.003 in Eq. (1), we obtain,

c
124 = 0.5918 (11)
dt
125 in US customary units, and

5
This formula is for G60 (G420) steel only. For other steel grades see e.g., [6].
6
These formulas are unit dependent. For the metric formula, stresses must be in MPa. For the US customary units
formula, stresses must be in ksi.

Page 5 of 22
 c
126 = 0.5882 (12)
dt
127 in metric units.
128 If ε t = ε y is approximated as 0.002, either of the previous two formulas can be approximated as:

c 3
129 = = 0.6 (13)
dt 5
130 Eq. (13) is the expression most commonly used to test for compression-controlled sections in the
131 literature (see e.g., [6]). The linear interpolation formula for φ (Eq. (7)) can also be written in
132 terms of the c/d t ratio as follows7,

dt 5
133 φ = 0.65 + 0.25( − ) (14)
c 3
134 It is also useful to write this formula in terms of the depth of the compression block a, as,
0.7 0.25β1dt
135 φ= + (15)
3 a
136 It can be concluded now that transition-zone sections are characterized by,
c
137 0.375 ≤ ≤ 0.6 (16)
dt
138 Eqs. (16) is valid for G60 (G420) steel only. A more general, and slightly more accurate
139 expression to characterize transition-zone sections can be obtained by using Eq. (8) instead of
140 Eq. (13) as the upper limit in (16) as follows,
3 c Es ε cu
141 ≤ ≤ (17)
8 d t Esε cu + f y

142 This formula is of course general and unit independent. In what follows, only general formulas
143 will be given. To specialize any of the these formulas either to the US customary system or the
144 metric system, it suffices to replace the term Esε cu with the quantity ``87'' for the US customary
145 system and with the quantity ``600'' for the metric system.
146 In terms of strains, the expression that characterizes transition-zone sections for G60 steel
147 becomes,
148 0.002 ≤ ε t ≤ 0.005 (18)

7
This is for G60 (G420) steel only. For other steel grades see e.g., [6].

Page 6 of 22
149 To ensure enough ductility however, the ACI code further restricts the value of ε t to a minimum
150 value of 0.004. When this value is substituted in Eq. (1), it can be concluded that the value of the
151 maximum allowable c/d t ratio is,

c 3
152 = ≈ 0.4286 (19)
dt 7
153 which means that the range of permitted (or allowed) transition-zone sections is actually,
c
154 0.375 ≤ ≤ 0.4286 (20)
dt
155 which is of course a subinterval of (16).
156 When c/d t = 3/7 is used in conjunction with Eq. (14), the corresponding range of permitted φ
157 values is obtained as,
158 0.8167 ≤ φ ≤ 0.9 (21)
159 When ε t = 0.004, together with the value of ε cu = 0.003, and f s = f y , are substituted into (3), an

160 expression for the maximum permitted value of the reinforcement ratio is obtained as:

* 2.55β1 f c' d t
161 ρ max = (22)
7 fy d

162 where an asterisk is used to distinguish this ratio from the traditional maximum reinforcement
163 ratio ( ρ max = 0.75ρb [7, 11]). The permitted range of transition-zone sections, in terms of
164 reinforcement ratios, can now be written as,
*
165 ρ tcl ≤ ρ ≤ ρ max (23)
166 which is completely equivalent to Eq. (21).
167 Now, by comparing Eq. (5) with Eq. (22), it can be concluded that,
* 8
168 ρ max = ρ tcl (24)
7
169 It is interesting to compare Eq. (24) with the traditional expression for the balanced
170 reinforcement ratio. When this is done, the following relationship is obtained,

* 3 Es ε cu + f y d t
171 ρ max = ρb (25)
7 Esε cu d

172 where ρ b is the traditional balanced reinforcement ratio given by,

0.85β1 f c' E sε cu
173 ρb = (26)
fy E sε cu + f y

Page 7 of 22
174 For G60 steel and for the particular case when d t = d (one layer of main reinforcement steel),
175 Eq. (25) becomes,
*
176 ρ max ≈ 0.73 ρ b (27)
177 which compares very well with the traditional value of the maximum reinforcement ratio [7, 11],
178 ρ max = 0.75ρb (28)

179 1.4 Compression Controlled Limit

180 In order to find an expression for the reinforcement ratio that corresponds to the compression-
181 controlled limit, it is necessary to first establish a relationship between the strain in the
182 lowermost layer of steel ε t and the strain at the centroid of the main steel reinforcement ε s . The
183 reason for this is that when the lowermost steel is exactly at the yield stress (compression-
184 controlled limit), the stress at the centroid of the main steel reinforcement is less than the yield
185 stress.
186 Referring to Fig. 1b, and using similar triangles, we obtain,
d −c
187 εs = εt (29)
dt − c
188 or, in terms of stresses,
d −c
189 fs = ft (30)
dt − c
190 When the lowermost reinforcement is exactly at the yield point, the stress at the centroid of the
191 main reinforcement is then,
d −c
192 f sccl = fy (31)
dt − c

193 Now, making use of Eq. (1), and multiplying numerator and denominator of (37) by Es , we get,

f y d − Es ε cu (d t − d )
194 f sccl = (32)
dt

195 Substituting Eq. (32) into Eq. (3), and multiplying and dividing again by Es , we finally obtain,

0.85β1 f c' d t E sε cu dt
196 ρ ccl = (33)
f y d − E s ε cu (d t − d ) E s ε cu + f y d

197 In Eqs. (31) to (33) the subindex ccl stands for compression-controlled.
198 The range of transition-zone sections can now be written exclusively in terms of reinforcement
199 ratio limits as,

Page 8 of 22
200 ρtcl ≤ ρ ≤ ρ ccl (34)

201 Note that when d t = d , the expression for the compression-controlled limit (Eq. (33)) reduces to
202 the expression for the traditional balanced reinforcement ratio (Eq. (26)).
203 Fig. 2 illustrates schematically the relative values of all reinforcement limits together with their
204 meaning in terms of the behavior of the corresponding sections. It is important to note (as it can
205 be seen in Fig. 2) that for the normally occurring case when d t > d , the value of ρ ccl is larger

206 than ρ b . This means that for many transition-zone sections, the main tensile steel reinforcement

207 does not yield. It is then necessary to solve a quadratic equation for a or for f s to find the
208 moment capacity of the section. This is a well-known procedure but it is outlined in the next
209 section for completeness.
210
211
212 ε t ≥ 0.005 ε y ≤ ε t ≤ 0.005 εt ≤ ε y

213
214 Figure 2. Schematic Representation of the Relative Values of the Reinforcement Limits
215 Presented in this Paper. In this figure ρ min represents the minimum reinforcement ratio for beams

3 f c' 200
216 given by: ρ min = (≥ ) (see e.g., [7]).
fy fy

217

Page 9 of 22
218 2. Determination of the Moment Carrying Capacity of a Section When the Tensile
219 Reinforcement Does not Yield8
220 When ρ > ρb , then f s < f y and the value of f s is unknown. To determine the moment carrying

221 capacity of this type of section it is necessary to first determine the value of f s or the (also
222 unknown) value of the depth of the compression block a. Referring to Fig. 1b, and using similar
223 triangles, it can be easily established that,
β1d
224 f s = E sε cu −1 (35)
a
225 as long as f s ≤ f y . Substituting now the value of f s from Eq. (35) into Eq. (2), the following

226 quadratic equation for a can be readily obtained (see e.g. [6]):
227 (0.85 f c' )a 2 + ( Es ε cu dρ )a − E sε cu ρβ 1d 2 = 0 (36)
228 Once this quadratic equation is solved for a, the stress in the steel can be found from (35). The
229 nominal moment carrying capacity of the section can then be determined by using the well-
230 known equation,
a
231 M n = As f s (d − ) (37)
2
232 Alternatively, instead of (36), a quadratic equation for f s can be derived as follows: Eq. (1) can

233 be written in terms of ε s and d instead of ε t and d t as, (see Fig. 1b),

c ε cu
234 = (38)
d ε cu + ε s
235 Now, by following a procedure similar to the one used to obtain Eq. (3) above, the following
236 formula is obtained (see Eq. (3)):

0.85β1 f c' ε cu
237 ρ= (39)
fs ε cu + ε s
238 By multiplying and dividing the term between parentheses in (39) by Es , we obtain,

0.85β1 f c' Esε cu

239 ρ= (40)
fs Es ε cu + f s

8
These reinforced concrete sections are not allowed by the current ACI-318-11 code since ρ > ρb means also that
*
ρ > ρ max . However, as mentioned before, it might still be necessary to find the moment capacity of these sections

in practice.

Page 10 of 22
240 This is also a completely general expression for the reinforcement ratio of singly reinforced
241 sections. As a matter of fact, if f s = f y is substituted into this equation, the expression for the

242 balanced reinforcement ratio (Eq. (26)) is obtained.

243 Eq. (40) can be rearranged to yield a quadratic equation in f s , as follows,

244 ρf s2 + E sε cu ρf s − 0.85 E sε cu β1 f c' = 0 (41)

245 Solving (41) for f s , and recalling that f s must be positive, we obtain,

Es ε cu 2 0.85Es ε cu β1 f c' Esε cu

246 fs = ( ) + − (42)
2 ρ 2
247 Once the stress in the steel is known, the value of a can be found from Eq. (2), and the nominal
248 moment carrying capacity can be found using Eq. (37).
249
250 3. Design Aspects.
251 Most authors use a trial and error procedure for the design of reinforced concrete sections (see
252 e.g., [6, 7]). The author prefers to use a quadratic formula for the reinforcement ratio ρ . This
253 formula is derived next. Similar formulas can be found in the literature (see e.g., [4], page 100).
254 We start with the relationship for the nominal moment carrying capacity of a singly reinforced
255 section with the tension steel yielding (Eq. (37) with f s = f y ),

a
256 M n = As f y (d − ) (43)
2
257 or, by using Eq. (2) with f s = f y , and multiplying both sides of (43) by φ ,

As f y
258 φM n = φAs f y (d − ) (44)
1.7 f c'b

259 Recognizing now that the most economical design results when φM n = M u , substituting ρbd

260 for As in Eq. (44), and reorganizing terms, we get the following quadratic equation for ρ ,

1.7 M u f c'
261 f y ρ 2 − 1.7 f c' ρ + =0 (45)
φbd 2 f y
262 whose solution is:

1.7 f c' M u
0.85 f c' − (0.85 f c' ) 2 −
φbd 2
263 ρ= (46)
fy

Page 11 of 22
264 The minus sign before the square root has been chosen in (46) to obtain a reinforcement ratio
265 that is less than one. Eq. (46) is very useful as long as φ is equal to 0.9 (i.e., for tension-
266 controlled sections). If 0.8167 ≤ φ < 0.9 (i.e., transition-zone sections), a different procedure is
267 necessary. This is a new procedure that is described next.
268
269 3.1 Direct Procedure for the Design of Transition-Zone Sections.
270 For transition-zone sections, the value of φ must be interpolated between 0.65 and 0.9. (see Eq.
271 (14) or Eq. (15). This means that the value of φ in Eq. (46) depends implicitly on ρ . For these
272 sections, it is still possible to use Eq. (46) as a first approximation, and then iterate until a
273 sufficiently accurate result is obtained for ρ and φ . Special design aids that rely on a series of
274 charts have also been devised for this case [5, 12]. Ref. [13] also presents an iterative procedure
275 to design transition-zone sections. A new, direct, and exact approach that does not require
276 iterations is presented next.
277
278 It turns out that it is possible to derive an equation that can be solved directly for ρ to obtain the
279 exact combination of ρ and φ for transition-zone sections. Remarkably enough, in spite of the
280 apparent high nonlinearity of the relationship between ρ and φ , the resulting equation is also a
281 quadratic equation. The first step is to use Eq. (2) in conjunction with Eq. (15) and the definition
282 of ρ , to write φ in terms of ρ as:

0.7 0.85β1 f c' d t

283 φ= + (47)
3 4 ρf y d

284 Substituting this value into Eq. (45) we get,

20.4 ρM u f c'
285 f y ρ 2 − 1.7 f c' ρ + =0 (48)
2.8 ρf y bd 2 + 2.55β1 f c'bdd t

286 which, after some algebraic manipulation becomes,

287 2.8 f y2bd 2 ρ 3 + (2.55β1 f c' f y bdd t − 4.76 f c' f y bd 2 ) ρ 2 + (20.4M u − 4.335β1 f c'bdd t ) f c' ρ = 0 (49)

288 This is a cubic equation in ρ with the trivial solution ρ = 0 . We are then left with the following
289 quadratic equation to solve for the nontrivial reinforcement ratio ρ .

1.7 f c' 2.55β1 f c' d t 5.1 f c' M u 1.08375β1 ( f c' ) 2 d t

290 ρ2 − − ρ+ − =0 (50)
fy 2.8 f y d 0.7bd 2 f y2 0.7 f y2 d

Page 12 of 22
291 The value of ρ found from this equation provides the design moment φM n = M u that

292 corresponds to the exact interpolated value of φ given by Eq. (15) (see Example 2 below).
293
294 4. Analysis and Design of T- and Doubly-Reinforced Sections.
295 With only minor modifications, the formulas that are used to analyze and/or design singly
296 reinforced sections can be used to analyze and/or design T- and doubly-reinforced sections.
297 4.1 T-Sections.
298 When dealing with T-sections, it is customary to idealize the actual T-section as two separate
299 sections that, when superimposed, reproduce the original section. These two sections are
300 traditionally called the F-beam (corresponding to the flanges) and the W-beam (corresponding to
301 the web) (Fig. 3). Within this framework, the area of main reinforcement steel that is balanced
302 exactly by the compressive forces on the flanges is denoted by Asf. The area of reinforcement
303 steel that corresponds to the W-beam is then given by,
304 Asw = As − Asf (51)

305 where As is the total area of main reinforcement of the section.

306
307
b b b

hf a
dt d

As As f As w
bw bw bw
308
309
310 Figure 3. Schematic Representation of Stresses and Strains in a Typical T-beam Reinforced
311 Concrete Section.
312
313

Page 13 of 22
314 It is convenient to define reinforcement ratios corresponding to the F-beam and to the W-beam
Asf Asw
315 as ρ f ≡ , and ρ w ≡ . Note that in both of these equations, the denominator involves the
bwd bw d

316 web width bw . This ensures that one can write,

317 ρ = ρw + ρ f (52)

As
318 where ρ = is the total reinforcement ratio of the T-section.
bw d
319 To apply the formulas of reinforcment ratios presented above to the analysis and design of T-
320 sections, it suffices to keep in mind that the quantity that must be compared with the different
321 reinforcement limits is ρ w (since this is the ratio that corresponds to the rectangular portion of

322 the section). For instance, if ρ w ≤ ρtcl , the section is tension-controlled; if ρtcl ≤ ρ w ≤ ρccl the
323 section is in the transition-zone, etc.
324 The remaining equations for the analysis and design of T-sections are standard and can be found
325 in [6].
326
327 4.2 Doubly-Reinforced Sections.
328 The typical notation and nomenclature for doubly-reinforced sections is shown in Fig. 4. In
329 analogy with T-sections, doubly-reinforced sections are also considered to be composed of two
330 parts that superimposed reproduce the original section. These two parts can be called Beam-1
331 and Beam-2. Beam-1 is composed of the compression reinforcement plus the portion of the
332 tensile reinforcement that exactly balances the forces in the compression reinforcement. Beam-2
333 includes the compression block on the concrete and the portion of the tension reinforcement that
334 balances the compressive forces on the concrete (Fig. 4).

Page 14 of 22
 A’s ε cu
ε’s c a
C
d’
d dt d

εs T = As fs
εt
b As

C1 = A’s f s’ C2
a a

T1 = As1 f y T 2 = As2 f y

335 Beam 1 Beam 2

336
337 Figure 4. Schematic Representation of Stresses and Strains in a Typical Doubly-Reinforced
338 Concrete Section.
339
340 The only difficulty with the analysis and design of doubly-reinforced sections is the fact that
341 sometimes the compression reinforcement does not yield at failure. When this is the case, it is
342 necessary to determine the value of the stress f s' in the compression reinforcement at failure.
343 Whether the compression reinforcement yields or not depends on the amount of tension
344 reinforcement in the section (see e.g., [6]). This leads to the procedure outlined below.
345
346 The areas of steel reinforcement corresponding to the original beam section and to Beams 1 and
As' f s'
347 2 are related by As = As1 + As 2 . Now, by the definition of Beam-1 (see Fig. 4), As1 = and
fy

As' f s' A'

348 therefore As 2 = As − . It is useful to define also, ρ ' = s , and write,
fy bd

f s'
349 ρ2 = ρ − ρ ' (53)
fy

As
350 where ρ = , as usual.
bd

Page 15 of 22
351 It can be shown (by a procedure similar to the one outlined in Section 1.1), that the limiting
352 reinforcement ratio to determine whether the compression reinforcement has yielded at failure is
353 (see e.g., [6]).

0.85β1 f c' E sε cu d'

354 ρ cry = (54)
fy E sε cu − f y d

355 where the subindex cry stands for ``compression reinforcement yielding''. Now, it is important to
356 emphasize that the quantity that must be compared with ρ cry (and with any other reinforcement

357 ratio limit for doubly reinforced sections) is ρ 2 . The difficulty with this approach is of course

358 that to be able to calculate ρ 2 , f s' must be known (see Eq. (53)). A way around this problem is

359 to approximate ρ 2 as ρ~2 ≈ ρ − ρ ' , and perform the test using ρ~2 instead. This approximation

360 does not pose any problems since ρ 2 ≥ ρ~2 . The test can then be performed as follows: if ρ~2 >

361 ρ cry , the compression reinforcement yields. This of course means that f s' = f y ; if ρ~2 < ρ cry , the

362 compression reinforcement does not yield. This means that f s' < f y , and the stress in the

363 compression reinforcement f s' is unknown. When the compression reinforcement yields, the
364 nominal moment carrying capacity of a doubly-reinforced section can be determined by simply
365 adding the contributions from Beam-1 and Beam-2 in the standard way (see e.g., [6]).
366
367 When the compression reinforcement does not yield, it is necessary to determine a (the depth of
368 the compression block) by means of the following quadratic equation:
369 (0.85 f c'b)a 2 − ( As f y − As' Es ε cu )a − As' Es ε cu β1d ' = 0 (55)

370 The procedure to derive this equation is analogous to the one used in Section 2 to obtain Eq.
371 (36).
372 Once Eq. (55) is solved for a, the stress in the compression steel can be found as:

β1d '
373 f s' = Es ε cu 1 − (80)
a

374 which is analogous to Eq. (35). The nominal moment carrying capacity of a doubly-reinforced
375 section can then be determined in the standard way.
376 Alternatively, when the compression reinforcement does not yield, it can be shown that the stress
377 in the compression steel can be found as:

Page 16 of 22
 Es ε cu ρf y Eε ρf ρf 0.85β1 f c' d '
378 f s' = ( + ' ) − ( s cu + y' ) 2 − Esε cu ( 'y − ) (82)
2 2ρ 2 2ρ ρ ρ 'd
379 The derivation of this equation is similar to that of Eq. (42). It should be emphasized that when
380 classifying, analyzing, or designing doubly-reinforced sections, the quantity that must be
381 compared with the reinforcement limits is ρ 2 (or ρ~2 ) since this is the ratio that corresponds to
382 the rectangular portion of the section. For instance, if ρ 2 ≤ ρ tcl the section is tension-controlled.

383 if ρ tcl ≤ ρ 2 ≤ ρ ccl the section is in the transition zone, etc.

384
385 5. Numerical Examples
386 The following three numerical examples illustrate the use of some of the formulas and concepts
387 presented above. Example 1 illustrates the use of Equation (33); Example 2, that of Equation
388 (50); and Example 3, the design procedure for T- and doubly-reinforced sections.
389 5.1 Example 1.
390 The purpose of this example is to illustrate the use of Equation (33) and to demonstrate its

391 accuracy. To do this, we will show that for a beam with ρ = ρccl , the ratio c/dt is exactly equal
392 to 0.5918 as indicated by Equation (12).
393
394 The compression controlled reinforcement ratio is given by Equation (33). For the dimensions of
'
395 the beam shown in Fig. 5, and assuming G60 steel and fc = 3.6 ksi, Equation (33) yields:

396 ρccl =0.03776

397 Now, the balanced reinforcement ratio for this beam (see Equation (26)) is:

398 ρb =0.02566
399 On the other hand, the reinforcement ratio for this beam is (Area of a #8, 0.79 in2; area of a #10,
400 1.27 in2):
401 ρ = 0.037738
402 In other words, the reinforcement of the beam shown in Fig. 5 is, in practical terms, equal to the
403 compression controlled limit. In addition, since this reinforcement ratio is larger than the

404 balanced reinforcement ratio ( ρb =0.02566 ), the tension reinforcement does not yield at failure.
405 Therefore, to find the value of the depth of the compression block it is necessary to solve
406 Equation (36) for a, as illustrated next.

Page 17 of 22
407
408 For the material properties given above, and the dimensions and parameters of the beam shown
409 in Fig. 5, Equation (36) yields:
410 3.06 a 2 + 59.10084 a − 904.2429 = 0

a
411 The solution of this equation is a = 10.06 in. Therefore, c = = 11.835 in. Therefore:
β1
c
412 = 0.59177
dt
413 which corresponds almost exactly to the compression controlled limit of 0.5918 given by
414 Equation (12).
415

416 Alternatively, Equation (42) can be used to solve for the stress in the steel f s , as follows.

417 For ρ = 0.03774, Equation (42) yields f s = 45.316 ksi. Using now Equation (3), a is found as:
418 a = 10.06 in; and c is then c = 11.8347 in. Therefore:
c
419 = 0.5917
dt
420
(508 mm)

18 in
20 in

(457 mm)

5 #10 + 4 #8
14 in

421 (356 mm)

422
423 Figure 5. Beam Section For Numerical Example 1.
424
425 5.3 Example 2.
426 The purpose of this example is to illustrate the use of Equation (50) for the design of transition-
427 zone sections and to verify its validity.
428 Suppose that it is necessary to find the reinforcement of a section with the dimensions shown in

429 Figure 6, for the following data: M u =338 kips-ft; fc' = 4 ksi, and f y = 60 ksi.

Page 18 of 22
430 A quick check, using Equation (46) (with φ =0.9) yields ρ ≈ 0.02014 . On the other hand, the
431 tension-controlled and maximum reinforcement ratios are, respectively:

432 ρtcl =0.02007 and ρmax

*
= 0.02294
433 Therefore, the section is very likely in the transition-zone and φ is less than 0.9. Using now
434 Equation (50) to find ρ , the following quadratic equation is obtained:

435 ρ2 − 0.056ρ + 0.00073992 =0

436 which yields: ρ = 0.02136 . This corresponds to an area of steel As = 5.383 in 2 . For this area of

437 steel, the values of a, φ , and M n , are as follows (using Equations (3), (15), and (37)):

438 a = 6.7852 in; φ = 0.86 ; Mn =393.16 kips-ft

439 which corresponds to: ϕMn =338.12 kips-ft ≈ Mu

440 The design can be completed by choosing 4 # 7 bars plus 3 # 9 bars to yield a total area of steel

441 As = 5.4 in 2 (area of a #7, 0.6 in2; area of a #9, 1.00 in2), and a reinforcement ratio ρ = 0.02143
*
442 ( < ρmax ). This area of steel and reinforcement ratio correspond to φ = 0.858 , Mn = 394.1kips-

443 ft, and, ϕMn =338.0 kips-ft. The final design is shown in Fig. 6.
444
(508 mm)

18 in
20 in

(457 mm)

3 #9 + 4 #7
14 in

445 (356 mm)

446
447 Figure 6. Beam Section For Numerical Example 2.
448
449
450 5.5 Example 3.
451 The purpose of this example is to illustrate the procedures for the design of T- and doubly-
452 reinforced sections outlined in Section 4 of this paper. The beam section for this example is

Page 19 of 22
 '
453 shown in Fig. 7. This T-section is to be designed for the following data: Mu=360 kips-ft; fc =

454 4.0 ksi, and f y = 60 ksi. The first step is to check whether the section behaves as a rectangular

455 section. To this end, Equation (46) with b = 14 in is used. This Equation yields ρ = 0.02185.

456 This corresponds to As =5.506 in 2 . Therefore,

As f y
457 a= = 6.94 in > hf
0.85 f c'b
458 This means that the section must be designed as a T-section. The following values from
459 Equations (5) and (22) will be used in what follows:
*
460 ρtcl = 0.02007; ρmax = 0.02294
461 Following now the procedure outlined in Section 4.1 for the design of T-sections, the following
462 values are obtained:

463 Asf = 0.9067 in 2 ; Mnf = 72.53 kips-ft; Muf = 65.28 kips-ft (assuming φ = 0 .9 );

464 Muw = 294.72kips-ft

465 For Muw = 294.72 kN.m and b =bw, Equation (46) yields ρw= 0.02633. This value is larger than
*
466 ρmax (=.02294) and therefore not permitted by the ACI code. This means that the web portion of
467 the beam (the W-beam) must be doubly-reinforced (assuming that the dimensions of the section

468 cannot be changed). To this end, let ρ2 = ρtcl =0.02007 . This corresponds to As2 =3.6126 in 2 .
469 Following now the procedure outlined in Section 4.2 for the design of doubly-reinforced
470 sections, the following quantities are obtained (see e.g., [6]):
471
472
473
474 a = 6.3752 in

475 Mn2 = 267.6 kips-ft

476 Mu2 = 240.84 kips-ft

477 Mu1 = 53.88 kips.ft

478 Mn1 = 59.87 kips.ft

Page 20 of 22
479
'
This results in a compression steel area As = 0.75 in 2 . ρ2 must now be checked against ρcry to
480 find out whether the compression reinforcement yields or not. Using Equation (54) we obtain

481 ρcry = 0.01724. Therefore ρ2 > ρcry and the compression reinforcement yields. The total W-
'
482 beam reinforcement is then Asw = As 2 + As = 4.36 in 2 . And the total tensile reinforcement for the

483 entire T-section is As = Asw + Asf = 5.27 in 2 . Choose 5 # 8 plus 3 # 6 bars to yield a total

484 reinforcement area of As = 5.27 in 2 (area of a #6, 0.44 in2; area of a #8, 0.79 in2). Choose 4 # 3
485 plus 1 # 5 bars for the compression reinforcement (area of a #3, 0.11 in2; area of a #5, 0.31 in2).
'
486 This corresponds to an area of compression reinforcement As = 0.75 in 2 . The final design is
487 shown in Fig. 7.
488

489
490
491 Figure 7. Beam Section For Numerical Example 3.
492
493 6. Concluding Remarks.
494 A series of ideas and formulas for the flexural design of reinforced concrete sections have been
495 presented. They are and attempt to reconcile the so-called unified design method (UDM) with
496 the ACI traditional approach to design. The UDM relies on strain limits for the characterization
497 of the behavior of reinforced concrete sections. The traditional ACI approach to design relies on
498 the simple concept of reinforcement ratios. The formulas and ideas presented in this paper
499 bridge the gap between these two approaches by allowing the treatment of reinforced concrete
500 sections in a much simpler way. These formulas and ideas should also prove useful in the
501 classroom since the concept of reinforcement ratios is more intuitive and therefore more
502 pedagogically appealing than that of strain limits. Many of the formulas presented in this paper
Page 21 of 22
503 are well known, but some are new. In particular, the formula for the compression-controlled
504 reinforcement ratio limit and the procedure for the design of transition-zone sections are new.
505
506 References
507 1. ACI Committee 318. Building Code Requirements for Structural Concrete (ACI-318-11) and
508 Commentary. American Concrete Institute, Farmington Hills, MI, 2011, 465 pp.
509 2. ACI Committee 318. Building Code Requirements for Structural Concrete (ACI-318-02) and
510 Commentary. American Concrete Institute, Farmington Hills, MI, 2002, 443 pp.
511 3. Ghosh, S. K. (2004), “ Significant Changes in the 2005 ACI Code Including changes
512 Affecting Precast/prestressed Concrete – Part 1” , PCI Journal. Vol. 49, Issue 5, pp. 94-99
513 4. Hassoun M. N. Structural Concrete: Theory and Design. 2nd edition, Prentice Hall, Upper
514 Saddle River, NJ, 2002, 772 pp.
515 5. Munshi A. J. “Design of Reinforced Concrete Flexural Sections by Unified Design
516 Approach”. ACI Structural Journal. Vol. 95, No. 5. September 1998. pp. 618-625.
517 6. McGregor J. G., and Wight J. K. Reinforced Concrete: Mechanics and Design. 4th edition.
518 Pearson Prentice Hall, 2005, 1132 pp.
519 7. Nawy E. G. Reinforced Concrete: A Fundamental Approach. 5th edition, Prentice Hall, 2002,
520 848 pp.
521 8. Hawileh, R. A., Malhas, F. A., and Rahman, A. “Comparison Between ACI-318-05 and
522 Eurocode 2 (EC2-94) in Flexural Concrete Design”. Structural Engineering and Mechanics, vol.
523 32, No. 6, 2009, pp. 705-724.
524 9. Matthys, S. Taerve, L. “Evaluation of Ductility Requirements in Current Design Guidelines
525 for FRP Strengthening”. Cement and Concrete Composites, Vol. 28, Issue 10, 2006, pp. 845-
526 856.
527 10. Mast, R. F., Dawood, M. Rizcalla, S. H., and Zia P. “Flexural Strength Design of Concrete
528 Beams Reinforced with High-Strength Steel Bars”. ACI Structural Journal. Vol. 105, Issue 5,
529 pp. 570-577.
530 11. ACI Committee 318, Building Code Requirements for Structural Concrete (ACI-318-08) and
531 Commentary. American Concrete Institute, Farmington Hills, MI, 2008, 465 pp.
532 12. Munshi J. A. ``Design of Prestressed Flexural Sections by Unified Design Approach''. PCI
533 Journal. Vol. 44, No. 5. September 1999, pp. 72-81
534 13. Aschheim, M., Hernandez-Montes, E., and Gil-Martin L. M. (2008). “Optimal Domains for
535 Flexural and Axial Loading”. ACI Structural Journal. Vol. 105, Issue 6, 2008. pp. 720-728.

Page 22 of 22

View publication stats