chapter 2

Concepts in Structural
Steel Design
2.1 DE SIG N PHIL OSO PH IES

A s discussed earlier, the design of a structural member entails the selection of a

cross section that will safely and economically resist the applied loads. Economy
usually means minimum weight—that is, the minimum amount of steel. This amount
corresponds to the cross section with the smallest weight per foot, which is the one
with the smallest cross-sectional area. Although other considerations, such as ease
of construction, may ultimately affect the choice of member size, the process begins
with the selection of the lightest cross-sectional shape that will do the job. Hav-
ing established this objective, the engineer must decide how to do it safely, which is
where different approaches to design come into play. The fundamental requirement
of structural design is that the required strength not exceed the available strength;
that is,
Required strength # available strength
In allowable strength design (ASD), a member is selected that has cross-sectional
properties such as area and moment of inertia that are large enough to prevent the
maximum applied axial force, shear, or bending moment from exceeding an allow-
able, or permissible, value. This allowable value is obtained by dividing the nominal,
or theoretical, strength by a factor of safety. This can be expressed as
Required strength # allowable strength (2.1)
redsquarephoto/Shutterstock.com

where
nominal strength
Allowable strength 5
safety factor

Strength can be an axial force strength (as in tension or compression members), a

exural strength (moment strength), or a shear strength.

21
www.jamarana.com https://t.me/universitarios
22 Chapter 2: Concepts in Structural Steel Design

If stresses are used instead of forces or moments, the relationship of Equation 2.1
becomes

Maximum applied stress # allowable stress (2.2)

This approach is called allowable stress design. The allowable stress will be in the
elastic range of the material (see Figure 1.3). This approach to design is also called
elastic design or working stress design. Working stresses are those resulting from the
working loads, which are the applied loads. Working loads are also known as service
loads.
Plastic design is based on a consideration of failure conditions rather than working
load conditions. A member is selected by using the criterion that the structure will
fail at a load substantially higher than the working load. Failure in this context means
either collapse or extremely large deformations. The term plastic is used because, at
failure, parts of the member will be subjected to very large strains—large enough to
put the member into the plastic range (see Figure 1.3b). When the entire cross section
becomes plastic at enough locations, “plastic hinges” will form at those locations, cre-
ating a collapse mechanism. As the actual loads will be less than the failure loads by
a factor of safety known as the load factor, members designed this way are not unsafe,
despite being designed based on what happens at failure. This design procedure is
roughly as follows.
1. Multiply the working loads (service loads) by the load factor to obtain the
failure loads.
2. Determine the cross-sectional properties needed to resist failure under
these loads. (A member with these properties is said to have sufficient
strength and would be at the verge of failure when subjected to the factored
loads.)
3. Select the lightest cross-sectional shape that has these properties.
Members designed by plastic theory would reach the point of failure under the fac-
tored loads but are safe under actual working loads.
Load and resistance factor design (LRFD) is similar to plastic design in that
strength, or the failure condition, is considered. Load factors are applied to the ser-
vice loads, and a member is selected that will have enough strength to resist the
factored loads. In addition, the theoretical strength of the member is reduced by the
application of a resistance factor. The criterion that must be satisfied in the selection
of a member is

Factored load # factored strength (2.3)

In this expression, the factored load is actually the sum of all service loads to be
resisted by the member, each multiplied by its own load factor. For example, dead
loads will have load factors that are different from those for live loads. The factored
strength is the theoretical strength multiplied by a resistance factor. Equation 2.3
can therefore be written as

o(loads 3 load factors) # resistance 3 resistance factor (2.4)

 2.2: American Institute of Steel Construc tion Specification 23

The factored load is a failure load greater than the total actual service load, so
the load factors are usually greater than unity. However, the factored strength is
a reduced, usable strength, and the resistance factor is usually less than unity. The
factored loads are the loads that bring the structure or member to its limit. In terms
of safety, this limit state can be fracture, yielding, or buckling, and the factored
resistance is the useful strength of the member, reduced from the theoretical value
by the resistance factor. The limit state can also be one of serviceability, such as a
maximum acceptable de ection.

2.2 AM E R I C AN I NS T I TUT E O F STEEL

CO NS TRU C T ION S PECIF IC AT ION
Because the emphasis of this book is on the design of structural steel building mem-
bers and their connections, the Speci cation of the American Institute of Steel Con-
struction is the design speci cation of most importance here. It is written and kept
current by an AISC committee comprising structural engineering practitioners, edu-
cators, steel producers, and fabricators. New editions are published periodically, and
supplements are issued when interim revisions are needed. Allowable stress design
has been the primary method used for structural steel buildings since the rst AISC
Speci cation was issued in 1923, although plastic design was made part of the Speci-
cation in 1963. In 1986, AISC issued the rst speci cation for load and resistance
factor design along with a companion Manual of Steel Construction. The purpose of
these two documents was to provide an alternative to allowable stress design, much
as plastic design is an alternative. The current speci cation (AISC, 2016a) incorpo-
rates both LRFD and ASD.
The LRFD provisions are based on research reported in eight papers pub-
lished in 1978 in the Structural Journal of the American Society of Civil Engineers
(Ravindra and Galambos; Yura, Galambos, and Ravindra; Bjorhovde, Galambos,
and Ravindra; Cooper, Galambos, and Ravindra; Hansell et al.; Fisher et al.;
Ravindra, Cornell, and Galambos; Galambos and Ravindra, 1978).
Although load and resistance factor design was not introduced into the AISC
Specification until 1986, it is not a recent concept; since 1974, it has been used in
Canada, where it is known as limit states design. It is also the basis of most European
building codes. In the United States, LRFD has been an accepted method of design
for reinforced concrete for years and is the primary method authorized in the
American Concrete Institute’s Building Code, where it is known as strength design
(ACI, 2014). Current highway bridge design standards also use load and resistance
factor design (AASHTO, 2014).
The AISC Specification is published as a stand-alone document, but it is also
part of the Steel Construction Manual, which we discuss in the next section. Except
for such specialized steel products as cold-formed steel, which is covered by a dif-
ferent specification (AISI, 2012), the AISC Specification is the standard by which
virtually all structural steel buildings in this country are designed and constructed.
Hence the student of structural steel design must have ready access to this docu-
ment. The details of the Specification will be covered in the chapters that follow, but
we discuss the overall organization here.

www.jamarana.com https://t.me/universitarios
24 Chapter 2: Concepts in Struc tural Steel Design

The Specification consists of three parts: the main body, the appendixes,
and the Commentary. The body is alphabetically organized into Chapters A
through N. Within each chapter, major headings are labeled with the chapter des-
ignation followed by a number. Furthermore, subdivisions are numerically labeled.
For example, the types of structural steel authorized are listed in Chapter A,
“General Provisions,” under Section A3, “Material,” and, under it, Section 1,
“Structural Steel Materials.” The main body of the Specification is followed by
Appendixes 1–8. The Appendix section is followed by the Commentary, which gives
background and elaboration on many of the provisions of the Specification. Its orga-
nizational scheme is the same as that of the Specification, so material applicable to a
particular section can be easily located.
The Specification incorporates both U.S. customary and metric (SI) units.
Where possible, equations and expressions are expressed in non-dimensional form
by leaving quantities such as yield stress and modulus of elasticity in symbolic form,
thereby avoiding giving units. When this is not possible, U.S. customary units are
given, followed by SI units in parentheses. Although there is a strong move to met-
rication in the steel industry, most structural design in the United States is still done
in U.S. customary units, and this textbook uses only U.S. customary units.

2.3 LOAD FAC TO R S, R ESIS TAN CE FAC TO R S,

AND  LOAD   CO M B INAT ION S F O R LR FD
Equation 2.4 can be written more precisely as
o i Qi # R n (2.5)
where

Qi 5 a load effect (a force or a moment)

i 5 a load factor
Rn 5 the nominal resistance, or strength, of the component under consideration
5 resistance factor

The factored resistance R n is called the design strength. The summation on

the left side of Equation 2.5 is over the total number of load effects (including,
but not limited to, dead load and live load), where each load effect can be associ-
ated with a different load factor. Not only can each load effect have a different
load factor but also the value of the load factor for a particular load effect will
depend on the combination of loads under consideration. Equation 2.5 can also
be written in the form
Ru # Rn (2.6)
where
Ru 5 required strength 5 sum of factored load effects (forces or moments)
Section B2 of the AISC Specification says to use the load factors and load com-
binations prescribed by the governing building code. If the building code does not
 2.3: Load Factors, Resistance Factors, and  Load  Combinations for LRFD 25

give them, then ASCE 7 (ASCE, 2016) should be used. The load factors and load
combinations in this standard are based on extensive statistical studies and are pre-
scribed by most building codes.
The following load combinations are based on the combinations given in
ASCE 7-16:

Combination 1: 1.4D
Combination 2: 1.2D 1 1.6L 1 0.5(Lr or S or R)
Combination 3: 1.2D 1 1.6(Lr or S or R) 1 (0.5L or 0.5W)
Combination 4: 1.2D 1 1.0W 1 0.5(Lr or S or R)
Combination 5: 0.9D 1 1.0W

where

D 5 dead load
L 5 live load due to occupancy
Lr 5 roof live load
S 5 snow load
R 5 rain or ice load*
W 5 wind load

Note that earthquake (seismic) loading is absent from this list. If earthquake loads
must be considered, consult the governing building code or ASCE 7.
In combinations 3 and 4, the load factor on L should be increased from 0.5 to 1.0
if L is greater than 100 pounds per square foot and for garages or places of public
assembly.
In combinations with wind load, you should use a direction that produces the
worst effect.
Combination 5 accounts for the possibility of dead load and wind load counter-
acting each other; for example, the net load effect could be the difference between
0.9D and 1.0W. (Wind loads may tend to overturn a structure, but the dead load will
have a stabilizing effect.)
As previously mentioned, the load factor for a particular load effect is not the
same in all load combinations. For example, in combination 2 the load factor for the
live load L is 1.6, whereas in combination 3, it is 0.5. The reason is that the live load
is being taken as the dominant effect in combination 2, and one of the three effects,
Lr , S, or R, will be dominant in combination 3. In each combination, one of the
effects is considered to be at its “lifetime maximum” value and the others at their
“arbitrary point in time” values.
The resistance factor for each type of resistance is given by AISC in the Spec-
ification chapter dealing with that resistance, but in most cases, one of two values
will be used: 0.90 for limit states involving yielding or compression buckling and
0.75 for limit states involving rupture (fracture).

*
This load does not include ponding, a phenomenon that we discuss in Chapter 5.

www.jamarana.com https://t.me/universitarios
26 Chapter 2: Concepts in Structural Steel Design

2. 4 SAFET Y FAC TO R S AN D LOAD CO M B IN AT ION S F O R A SD

For allowable strength design, the relationship between loads and strength (Equation 2.1)
can be expressed as
Rn
Ra # (2.7)
V

where

Ra 5 required strength
R n 5 nominal strength (same as for LRFD)
V 5 safety factor
R nyV 5 allowable strength

The required strength Ra is the sum of the service loads or load effects. As with
LRFD, speci c combinations of loads must be considered. Load combinations
for ASD are also given in ASCE 7. The following combinations are based on
ASCE 7-16:

Combination 1: D
Combination 2: D1L
Combination 3: D 1 (Lr or S or R)
Combination 4: D 1 0.75L 1 0.75(Lr or S or R)
Combination 5: D 1 0.6W
Combination 6: D 1 0.75L 1 0.75(0.6W) 1 0.75(Lr or S or R)
Combination 7: 0.6D 1 0.6W

The factors shown in these combinations are not load factors. The 0.75 factor in
some of the combinations accounts for the unlikelihood that all loads in the combi-
nation will be at their lifetime maximum values simultaneously.
Corresponding to the two most common values of resistance factors in
LRFD are the following values of the safety factor V in ASD: For limit states
involving yielding or compression buckling, V 5 1.67.* For limit states involving
rupture, V 5 2.00. The relationship between resistance factors and safety fac-
tors is given by

1.5
V5 (2.8)

For reasons that will be discussed later, this relationship will produce similar designs
for LRFD and ASD, under certain loading conditions.
If both sides of Equation 2.7 are divided by area (in the case of axial load) or
section modulus (in the case of bending moment), then the relationship becomes
f#F

*
The value of V is actually 1 2 ⁄ 3 5 5y3 but has been rounded to 1.67 in the AISC speci cation.
 2.4: Safet y Factors and Load Combinations for ASD 27

where
f 5 applied stress
F 5 allowable stress
This formulation is called allowable stress design.

Example 2.1
A column (compression member) in the upper story of a building is subject to the
following loads:
Dead load: 109 kips compression
Floor live load: 46 kips compression
Roof live load: 19 kips compression
Snow: 20 kips compression
a. Determine the controlling load combination for LRFD and the correspond-
ing factored load.
b. If the resistance factor is 0.90, what is the required nominal strength?
c. Determine the controlling load combination for ASD and the corresponding
required service load strength.
d. If the safety factor V is 1.67, what is the required nominal strength based on
the required service load strength?

Solution Even though a load may not be acting directly on a member, it can still cause a load
effect in the member. This is true of both snow and roof live load in this example.
Although this building is subjected to wind, the resulting forces on the structure are
resisted by members other than this particular column.
a. The controlling load combination is the one that produces the largest factored
load. We evaluate each expression that involves dead load, D; live load resulting
from occupancy, L; roof live load, Lr ; and snow, S.

Combination 1: 1.4D 5 1.4(109) 5 152.6 kips

Combination 2: 1.2D 1 1.6L 1 0.5(Lr or S or R). Because S is larger than Lr
and R 5 0, we need to evaluate this combination only once,
using S.
1.2D 1 1.6L 1 0.5S 5 1.2(109) 1 1.6(46) 1 0.5(20)
5 214.4 kips
Combination 3: 1.2D 1 1.6(Lr or S or R) 1 (0.5L or 0.5W). In this combination,
we use S instead of Lr , and both R and W are zero.
1.2D 1 1.6S 1 0.5L 5 1.2(109) 1 1.6(20) 1 0.5(46)
5 185.8 kips
Combination 4: 1.2D 1 1.0W 1 0.5L 1 0.5(Lr or S or R). This expression re-
duces to 1.2D 1 0.5L 1 0.5S, and by inspection, we can see that
it produces a smaller result than combination 3.

www.jamarana.com https://t.me/universitarios
28 Chapter 2: Concepts in Structural Steel Design

Combination 5 does not apply in this example, because there is no wind load to
counteract the dead load.

Answer Combination 2 controls, and the factored load is 214.4 kips.

b. If the factored load obtained in part (a) is substituted into the fundamental
LRFD relationship, Equation 2.6, we obtain
Ru # R n
214.4 # 0.90R n
R n $ 238 kips

Answer The required nominal strength is 238 kips.

c. As with the combinations for LRFD, we will evaluate the expressions involving
D, L, Lr, and S for ASD.

Combination 1: D 5 109 kips. (Obviously this case will never control when
live load is present.)
Combination 2: D 1 L 5 109 1 46 5 155 kips
Combination 3: D 1 (Lr or S or R). Since S is larger than Lr , and R 5 0, this
combination reduces to D 1 S 5 109 1 20 5 129 kips
Combination 4: D 1 0.75L 1 0.75(Lr or S or R). This expression reduces to
D 1 0.75L 1 0.75S 5 109 1 0.75(46) 1 0.75(20) 5 158.5 kips
Combination 5: D 1 0.6W. Because W is zero, this expression reduces to
combination 1.
Combination 6: D 1 0.75L 1 0.75(0.6W) 1 0.75(Lr or S or R). Because W is
zero, this expression reduces to combination 4.
Combination 7: 0.6D 1 0.6W. This combination does not apply in this
example, because there is no wind load to counteract
the dead load.

Answer Combination 4 controls, and the required service load strength is 158.5 kips.
d. From the ASD relationship, Equation 2.7,

Rn
Ra #
V
Rn
158.5 #
1.67
Rn $ 265 kips

Answer The required nominal strength is 265 kips.

 2.5: Probabilistic Basis of Load and Resistance Fac tors 29

Example 2.1 illustrates that the controlling load combination for LRFD may not con-
trol for ASD.
When LRFD was introduced into the AISC Specification in 1986, the load fac-
tors were determined in such a way as to give the same results for LRFD and ASD
when the loads consisted of a dead load and a live load equal to three times the dead
load. The resulting relationship between the resistance factor and the safety factor V,
as expressed in Equation 2.8, can be derived as follows. Let R n from Equations 2.6
and 2.7 be the same when L 5 3D. That is,

Ru
5 Ra V

1.2D 1 1.6L
5 (D 1 L)V

or
1.2D 1 1.6(3D)
5 (D 1 3D)V

1.5
V5

2.5 PRO BA B ILI S TI C BA SI S O F LOAD

AN D R ESIS TAN CE FAC TO R S
Both the load and the resistance factors speci ed by AISC are based on probabi-
listic concepts. The resistance factors account for uncertainties in material proper-
ties, design theory, and fabrication and construction practices. Although a complete
treatment of probability theory is beyond the scope of this book, we present a brief
summary of the basic concepts here.
Experimental data can be represented in the form of a histogram, or bar graph,
as shown in Figure 2.1, with the abscissa representing sample values, or events,
and the ordinate representing either the number of samples having a certain value
or the frequency of occurrence of a certain value. Each bar can represent a single

FIGURE 2.1

www.jamarana.com https://t.me/universitarios