Professional Documents
Culture Documents
STRUCTURAL STEEL
DESIGN
Dr. Ammar A. Ali
STEEL DESIGN
1. INTRODUCTION
This lecture is based mainly on:
Applied Structural Steel Design, 4th edition, by Spiegel, L. and Limbrunner, G.F.
-
Steels used in construction are generally carbon steels: alloys of iron and carbon.
The carbon content is ordinarily less than 1% by weight.
The chemical composition of steel is varied, according to the properties desired,
such as: strength and corrosion resistance. This achieved by addition of other
alloying elements, such as: silicon, manganese, copper, nickel, chromium, and
vanadium, in very small amounts.
When steel contains a significant amount of any of such alloying elements, it is
referred to as an alloy steel.
The advantages of steel are:
Uniformity of material,
Predictability of properties,
Dimensional stability,
Ease of fabrication, and
Speed of erection.
The disadvantages of steel are:
Susceptibility to corrosion (in most but not all steels)
Loss of strength at elevated temperatures.
Specifications:
Steel Properties:
2
strain = =
where
0
0
L0 = elongation or the change in length between two reference points on the tensile
specimen (in.)
L0 = original length between two reference points (in.)
GW
GL
W
RS G
L0
FR
FR = Fillet radius.
L = Overall length.
RS = Reduced section.
The steel remains elastic as long as it is not stressed past a value slightly higher than
the proportional limit, called the elastic limit.
4
The proportional limit is the point at which the stress-strain relationship becomes
nonlinear.
The proportional limit and the elastic limit are so close that they are often
considered to be the same value.
The yield stress, Fy, will be reached when the strain in the specimen increases
rapidly at constant stress.
The slightly higher stress that exists just after the proportional limit, called the
upper yield. It exists only instantaneously and is unstable.
5
According to the allowable stress design (ASD) method, most structural steel is
designed so that actual stresses in the structural member do not exceed allowable
stresses, well below Fy.
As the steel continues to strain, it reaches a point at which its load-carrying capacity
increases. This phenomenon of increasing strength is termed strain hardening.
The maximum stress to which the test of increasing strength is called the tensile
strength, Fu (ksi).
In the ASD method stresses are limited to some fraction of Fy, but another design
method, called plastic design, allows small but definite areas of members to be
stressed to Fy and strained into the plastic range.
For all practical purposes, in structural steel design, it is only the elastic range and
the plastic range that need be of interest, since the strains in the strain-hardening
range are of such magnitude that the deformation of the structure would be
unacceptable. Thus, an idealized diagram for structural steel is sufficient for
purposes of illustrating the steel stress-strain relationships, as shown in Fig. 1.5.
The strain at the upper limit of the plastic range, p, is approximately 10-15 times
the strain at the yield point, y.
The modulus of elasticity, E (or Youngs modulus), is the slope of the stress-strain
curve in the elastic range:
=
stress
=
strain
6
Youngs modulus E is reasonably constant for structural steel. AISC recommends the
value of E to be 29,000 kips/in2.
The ability to undergo large deformations before failure is called the ductility.
The ductility of steel is important in the safeguarding of lives and property in
unknown and uncertain loading situations such as earthquakes.
Products Available
-
Steels are usually specified according to ASTM (American Society for Testing and
Materials) number.
The ASD manual, Part 1, Table 1, contains a list of ASTM structural steels. Note that
minimum Fy ranges from 32 to 100 ksi and tensile strength Fu ranges from 58 to 130
ksi.
The most widely used structural carbon steel is A36. A36 steel has a yield stress of
36 ksi except for cross sections in excess of 8 in thick.
More recently, the trend has been toward the higher-strength steels, with A572
grade 50 being widely used.
A recently developed steel specification designated A992 covers steel that is similar
to A572 grade 50 but has some important differences. In addition to differences in
chemical composition, A992 has a minimum Fu of 65 ksi, and Fy range of 50-65 ksi,
and a maximum ratio of Fy/Fu of 0.85.
Steels designated as A588 and A242 are weathering steels. These steels are popular
for use in bridges and exposed building frames.
Structural shapes are:
1- W and M shapes (Fig. 1.6) are used primarily as beam and column members.
bf
Flange
tf
slope 0-5%
Web
d
tw
W36 300
depth () 36 in.
= overall depth,
= flange width,
= flange thickness,
= web thickness
M shapes are miscellaneous shapes. They have cross sections that appear to be
exactly like W shapes, but with different dimensions and not widely available.
M8 6.5
depth () 8 in.
2- S shapes (Fig. 1.7) are American standard beams. They have sloping inner faces on
the flanges, relatively thicker webs, and depths that are mostly full inches. They find
some application where heavy point loads are applied to the flanges, such as in
monorails for the support of hung cranes.
bf
Flange
tf
slope 1632%
Web
tw
depth () = 12 in.
weight () = 35 lb/ft
3- HP shapes (Fig. 1.8) are bearing pile shapes and are characterized by a rather
square cross section with flanges and webs of nearly the same or equal thickness (so
that the web will withstand pile-driving hammer blows).
8
bf
tf
Flange
Web
d
tw
depth () 12 in.
weight () = 74 lb/ft
4- C and MC shapes (Fig. 1.9) are American standard channels and miscellaneous
channels. The channel shapes are characterized by short flanges that have sloping
inner surfaces and depths to full inches. Their applications are usually as
components of built-up cross sections, bracing and tie members, and members that
frame openings.
tf
bf
Flange
slope 1632%
Web
tw
weight () = 58 lb/ft
5- Angles (Fig. 1.10) are designated by the letter (L), leg length of each leg, and
thickness. They may be equal or unequal leg angles. Angles are commonly used
singly or in pairs as bracing members and tension members. They are also used as
brackets and connecting members between beams and their supports. Light trusses
and open web joists may also utilize angles for component parts.
1
For unequal leg angles the longer leg is stated first. For example, L9 4 indicates
2
an angle with one leg 9-in. long and one leg 4-in. long, both having thickness of in.
2
Note: Designation does not provide the unit weight of the angle as has been the case with
all shapes before. The weights in pounds per foot is tabulated and can also be calculated by
using the unit weight of steel of 490 lb/ft3.
tf
d
Flange
tw
Web
Tees are used primarily for special beam applications and as components in
connections and trusses.
(b) Square
HSS
(c) Rectangular
HSS
The HSS (Hollow Structural Section) are given in the ASD manual, Part 1, Table 3 and in the
LRFD manual, Part 2, Table 2-3.
A round HSS is designated by nominal diameter and wall thickness, each expressed to three
decimal places, such as HSS5.563 0.258.
Tubes make excellent compression members, although the connections usually involve
some welding. The most common use of tubes is in compression and tension members, but
they also used as beams in some situations. Structural members made of tubes are easier to
clean and maintain than their wide-flange counterparts.
8- Plates
Plates are used in assorted applications such as plate girder elements, gusset plates,
stiffenners, beam bearing plates, column base plates, etc. Plates are avalible in various
1
11
The letters PL followed by thickness, width, and length commonly designates plates. In this
designation, thickness and width are fractional inches, and length in feet and inches.
1
PL 14 1 4
2
The weight of a plate can be taken from tables, when available, or computed by using a unit
weight of stee of 49 lb/ft3.
Design Considerations
1. Safety
2. Economy, beauty, fuctionality, maintainability, permanance, etc.
Safety
The expression of safety is normally made in terms of a factor of safety. The factor of
safety is the ratio of the load (or stress) that causes failure to the maximum load (or stress)
actually allowed in the structure.
In allowable stress design (ASD) the attainment of yield stress in a member is considered to
be analogous to failure.
Although the steel will not actually fail (rupture) at yield, significant and unacceptable
deformations are on the verge of occuring, which may render the structure unusable.
As an example, assume that a member composed of a steel having yield stress Fy has as
specified allowable stress of 0.66 Fy. The factor of safety (F.S.) against yielding would then
be
. . =
" "
=
=
= 1.5
0.66
The LRFD method uses series of factors of safety called load factors when applied to loads
and resistance factors when applied to member strength or resistance. Each factor is the
result of a statistical study of the variability of the particular quantity and reflects the
probability that the specific load or resistance is incorrect. In that design method strength
reduction factors are used to predict a practical strength and load factors ae used to modify
service loads and estabilish design loads for use in design considerations.
F.S. reasons:
12
Loads
Dead loads are static loads that produce vertical forces due to gravity and include the
weight of the steel framework and all materials permanently attached to it and supported
by it.
Live loads include all vertical loads that may be either present on or absent from the
structure.
Generally, lateral loads are considered live loads whether they are permanent or not.
Example of live loads are: snow, people, furniture, stored materials, vheicles (on brodges or
warehouses), cranes, wind, lateral prssure due to earth or stored lquids, and earhquakes.
13
2. TENSION MEMBERS
Although easy to proprtion, however, tension members, and structures in which main loadcarrying members are iin tension, require great care in the design and detailing of their
connections.
A tensile load applied along the longitudinal axis of the member tends to hold the member
in alignment, therby making instability a minor concern.
Of the most concern in the selection of tension members is the choice of the gonfiguration
of the cross section so that the connections will be simple and efficient. Also, the connection
should transmit the load to the member with as little eccentricty as possible.
Examples of tension members may be found in many structures. They include hangers for
catwalks and strorrage bins, truss web and chord members, cables for direct support of
roofs, sag rods, tie rods, and various types of braces.
Most of the common hot-rolled structural steel shapes may be used as tension members.
Tension Member Analysis
The direct stress formula is the basis for tension member analysis. It may be written for
stress,
Pt = axial tensile load capacity (or maximum allowable axial tensile load)
A = cross-sectional area of axially loaded tension member (either gross area Ag, net area An,
or effective net area Ae)
Gross area is the origonal, unaltered cross-sectional area of the member
14
(for plate) =
where
b = width of plate
t = thickness of plate
An, net area, is the cross-sectional area actually available to be stressed in tension (Fig. 2.1).
= (area of holes)
= 2( )
16
") +
16
1
16
"
16
"
(B2)
" = + "
8
For purposes of analysis and design, hole diameters are taken as the fastener diameter plus
1
8
".
yielding and stress distribution. This will result in a uniform stress distribution at ultimate
load.
The allowable tensile stress Ft takes into consideration two types of failure:
1. The mamber may rupture on the least net area. This is the classical and historical
approach to tension member analysis.
2. The tension member may undergo uncontrolled yielding of its gross area without
rupture. Excessive elongation of tension member is undesirable in that it normally
results in deformation of the structure and can lead to failure in other parts of the
structural system.
The types of failure above may be predicted using the following formulas:
1. = 0.50 (on the net area)
D1
These allowable stresses do not apply to pin-connected members (such as eye bars or
plates connected with relatively large pins), threaded steel rods, or flexible tension
members such as cables and wire ropes.
Block Shear
A tearing failure that can occure at end connections along the perimeter of welds or
along the perimeter of a group of bolt holes (Fig 2.2).
Tension area
Tension area
Pt
Pt
Shear area
Shear area
Shear area
Shear area
Tension
area
Pt
16
Depending on the end connection, the failurecould occur in either the tension member
itself or in the member to which is attached (e.g., a gusset plate).
Block shear is characterized by a combination of shear failure along a plane through the
welds or bolt holes and a simulatianeous tension failure along a perpendicular plane.
Block shear strength is calculated from the summation of net shear are Av times the
allowable shear stress Fv and net tension area At times the allowable tensile stress Ft,
where
= 0.30
} 4
and
= 0.50
= +
For purposes of block shear strength calculations for bolted connections, hole diameters
1
for the net area determination aretaken as the fastener diameter plus in. This is the same
approach as that used for tensile net area calculations.
Definitions
In Fig. 2.3, the tensile load P is assumed to be applied parallel to and ciincident with the
longitudinal axis of the member.
Gage lines
Longitudinal axis
d3
d1
E.D.2
E.D.1
s1
s2
d2
17
The bolt holes are located on gage lines that are parallel to the longitudinal axis. The
dimension g between the gage lines is called the gage. The dimension s parallel to the gage
line and taken between centers of bolt holes is called the pitch (or the bolt spacing). The
distance between bolts is a straight-line distance between any two bolts. The edge
distance is the prependicular distance from the center of a hole to the nearest edge.
The analysis of a tension member involves the determination of the indiviual allowable
loads based on the various failure modes. Once these have been determined, the axial load
capacity of the member is taken as the smallest value.
Example:
Find the axial tensile load capacity Pt of the lapped, bolted tension member shown in Fig.
2.4. Bolts are
3
4
in. diameter, and the plate is A36 steel (Fu = 58-80 ksi from the ASD
manual, Part 1, Table 1 and the LRFD manual, Part 2, Table 2-3). Assume that the fasteners
are adequate and do not control the tensile capacity. Pitch, gage, and distance are as
shown.
Pt
7
16 "
Pt
7
16 "
(a)
A
3"
F
Pt
3"
C
3"
G
3"
E
121"
3"
3"
121"
(b)
Pt
= 0.60
7
= 0.60(36) ( ) (12) = 113.4 kips
16
where Av and At are the net shear area and net tension area, respectively. Note that
the hole diameter is taken as
3 1
+ = 0.875 in.
4 8
Case I: failure line FBCDG:
7
= 2 ( ) [7.5 2.5(0.875)] = 4.65 in.2
16
7
[6 2(0.875)] = 1.859 in.2
=
16
= 4.65(0.30)(58) + 1.859(0.50)(58) = 134.8 kips
7
) [7.5 2.5(0.875)] = 2.32 in.2
16
7
= ( ) [9 2.5(0.875)] = 2.98 in.2
16
= 2.32(0.30)(58) + 2.98(0.50)(58) = 126.8 kips
= (
19
Staggered Holes
In some cases the fasteners will be arranged so that the controlling fructure line will be
something other than transverse, as shown in Fig. 2.5. This situation can occur when
fasteners are staggered to accommodate a desired size or shape of connection. There are
two possible failure lines across the width of the plate. These may be defined as lines ABCD
and ABE. For large values of s, line ABE will be the more critical failure line (smaller net
area). For small values of s, line ABCD will be more critical.
A
B
Pt
Pt
E
s
2
= +
4
Where wg represents gross width and dh represents the hole diameter to be used for
design.
20
The forgoing formula for wn is convenient to use with members of uniform thickness. If the
formula is multiplied by thickness t it becomes
2
= +
4
Or, since = and = ,
= +
2
4
The latter formula for An is the more useful, since it provides net area directly and is also
applicable with members that do not have uniform thickness (i.e., channels). In a
determination of critical net area where multiple failure lines exist, the critical net area is
the least net area.
Example:
Determine the critical net width wn for the plate shown in Fig. 2.6. Fasteners will be 1-in.diameter bolts.
221"
B
F
Pt
3"
Pt
4"
C
D
221"
2"
2
= +
4
1
= 1 " = 1.13 in.
8
21
Line
ABCD
ABFG
EFG
22
22
EFCD
22
22
ABFCD
wn = 9.19 in.
A zigzag failure line sometimes occurs in a member that has more than one element making
up its cross section, such as an angle.
Example:
Calculate the tensile capacity Pt of the angle shown in Fig. 2.7. Assume that the member to
which the angle is connected does not govern capacity. Likewise, assume that the
3
connection does not govern capacity. Assume A36 steel and -in.-diameter bolts.
L4X3X41
13
4"
4"
1"
22
5 @ 121"
(b)
(a)
D A
1"
14
F
g
1"
12
C
5@
121"
(c)
1. Calculate Pt based on yielding of the gross area. The gross width of the angle is
determined from
= 4 + 3
1
= 6.75 in.
4
2. Next, caclulate Pt based on tensile fracture of the net area. Two net width values are
determined:
= 6.75 2(0.875) +
1.5
= 5.14 in.
4(4)
Case I: in the block shear failure line defined by FEBG, the diagonal line considered to be
2
Therfore,
= + = (0.30 ) + (0.50 )
= 0.25[9(1.5) 4(0.875)] = 2.5 in.2
1.52
] = 0.816 in.2
= 0.25 [4 0.875 +
4(4)
23
Case II: use the block shear failure line defined by FEBC, where EB and BC are considered
tension areas and FE is a shear area:
= 0.25[6 1.5(0.875)] = 1.172 in.2
From which
1.52
] = 1.082 in.2
4(4)
Failure line DEBG could also be checked. It would be found to have tensile capacity slightly
greater than that of line FEBC. Therfore, the tensile capacity Pt of the angle is 36.5 kips,
controlled by yielding of the gross area.
Effective Net Area
For some tension members, such as rolled shapes, that do not have all the elements of the
cross section connected to the supporting members (see Fig. 2.8), the failure load is less
than would be predicted by the product AnFu. The phenomenon to this situation is generally
attributed is called shear lag.
24
The ASD specifications, Section B3 and LRFD specifications, Section D3, accounts for the
effect of shear lag through the use of an effective net area, which is a function of how the
tension member is connected at its ends.
When the load is transmitted directly to each of the cross-section elements by either bolts
or welds, the effective net area Ae is equal to the net area An.
When the load is transmitted by bolts through some, but not all, of the cross-sectional
elemets of the member, the effective net area Ae shall be computed from
= . 3 1
where
2
3 d (min.)
d
d
Case II
Case III
When the load is transmitted by welds through some, but not all, of the cross-sectional
elements of the member, the effective net area Ae shall be computed from
where
= . 3 2
25
The U values for a welded connection is the same as that for a bolted connection except that
Case III is not applicable and the condition as to the number of bolts in Case I and Case II
does not apply.
With respect to weld end connections, the ASD specifications, Section B3, furnishes
effective net area criteria for two special considerations:
a. When a load is transmitted by transverse welds to some, but not all, of the crosssectional elements of W, M, or S shapes and structural tees cut from these shapes,
the effective net area Ae shall be taken as that area of the directly connected
elements.
b. When a load is transmitted by longitudinal welds used alone along both edges of a
flat bar (or plate) axially loaded tension member, the length of each weld shall not
be less than the width of the plate. The effective net area Ae shall be computed by Eq.
(B3-2) using the reduction coefficient U shown in Table 2.2, where
Additionally, for relatively short connection fittings such as splice plates, gusset plates, and
beam-to-column fitting subjected to tensile force, the effective net area shall be taken as the
actual net area except that it shall not be taken as greater than 85% of the gross area.
Therefore, for these short plates and fittings subjected to tension, U does not apply, and
Example:
lines of -in.-diameter bolts in each flange as shown in Fig. 2.9. There are three bolts per
8
26
line, 33-in. pitch, 1 in. edge distance, and A36 steel. Find the tensile load capacity Pt.
2
Assume that the gusset plates and the bolt capacities are satisfactory.
Pt
Gusset
plates
W8 X 24
Bottom
chord
d = 7.93 in.
bf = 6.495 in.
tf = 0.40 in.
=
= 0.60
= 0.60(36)(7.08) = 152.9 kips
=
= 0.50
=
= 4(1.0)(0.40) = 7.08 1.60 = 5.48 in.2
27
2
= 0.67(7.93) = 5.31 in.
3
= 6.495 in. > 5.31 .
= 0.50
= 0.50(58)(0.90)(5.48) = 143.0 kips
3. The block shear consideration involves four blocks as shown in Fig. 2.10, two in
each flange.
= + = (0.30 ) + (0.50 )
= 4(0.4)[7.5 2.5(1.0)] = 8.0 in.2
Therefore,
= 4(0.4) [1.5
1.0
] = 1.6 in.2
2
For this member, Pt = 143.0 kips as controlled by a rupture failure based on the least net
area.
1"
62
1"
12
1"
12
3"
3"
11 "
28
Find the tensile load capacity Pt for the double-angle tension member shown in Fig. 2.11.
All structural steel is A36 (Fu = 58 ksi). Assume that the welds are adequate and do not
control the tensile capacity.
3
8"
Gusset plate
6"
2Ls 3 21 X 3 X 38 (LLBB)
Ag = 4.59 in.2
3"
1"
32
Pt
=
= 0.60
= 0.60(36)(4.59) = 99.1 kips
3. Check block shear in the gusset plate along the perimeter of the welds:
Therefore,
= + = (0.30 ) + (0.50 )
= 0.375(6 + 3) = 3.38 in.2
= 3.5(0.375) = 1.313 in.2
29
For this member, Pt = 96.9 kips as controlled by block shear failure in the gusset plate.
Length Effects
Tension members do not suffer from the problems of instability and buckling that
compression members and beams do. Therefore, length plays a minor role. The ASD
specifications, section B7, suggests upper limits for the slenderness ratios of tension
members.
The recommended upper limit for slenderness ratio l / r is 300 for all tension members.
The recommended upper limit on l / r is preferred, but is not mandatory, and applies to
tension members other than steel rods and cables.
If the slenderness ratio is within the recommended limit, however, there will be some
resistance to undesirable vibrations as well as some resistance to bending and deformation
during shipping and erection handling.
Example:
Maximum preferred l / r = 300. Calculate actual l / r. use the least radius of gyration.
20(12)
=
= 149 < 300
1.61
OK
The minimum required ptoperties as governed by the first two foregoing items involve
strength and slenderness and are easily calculated by using the principles already
discussed.
The third required property (net area) is also easily calculated. Sections, however, are
tabukated on the basis of gross area. The relationship between gross area and net area
depends in part on the thicknessof the material, which is unknown at this point.
The fourth item (block shear strength) depends on the material thickness, the size and type
of fasteners, and the geometry of the connection. If the connection details can be
estabilshed or approximated, a required thickness can be calculated.
Alternatively, the block shear strength cn be checked once the member has been selected
and the connection has been designed.
The fifth consideratrion involves the way the member will fit into and be affected by the
structure of which it is a part.
The selection of a tension member (particularly in trusses) must be based on assumed end
conditions.
After a member is selected and the end connections have been designed, it may be
necessary to revise the selection.
Since block shear plays such an imprtant role, a preliminary understanding of simple
connections for tension members is essenatial.
Two types of connections will be considered here, and they are, lap and butt connections,
as shown in Fig. 2.12.
P
(a) Lap connection
P
P
(b) Butt connection
Fig. 2.12 Types of connection.
31
In most structural connections the bolt is required to prevent the movement of the
connected material in a direction perpendicular to the length of the bolt as shown in Fig.
2.13.
Shear plane
Three types of bolts are given: A307, A325, and A409. These designations are ASTM
material designations. The notations under loading refer to single shear (S) and double
shear (D).
In addition to considering shear failure in the bolts, the members being connected where
they bear on the bolts must be considered. If a material is overly thin, the hole will
elangoate into an oval shape and the connection will be said to have failed in bearing as
shown in Fig. 3.14.
P
Table I-E, part 4 shows the bearing strengths of various diameters of bolts.
The ASD Specifications recommends that the minimum distance between bolts be taken as
2
2 times the bolt diameter (and that 3 times the bolt diameter is preferred).
3
Table J3-5, Part 5 shows minimum edge distances for various bolt sizes and edge
conditions.
Example:
Select the lightest double-angle tension member for member BC, a web member in a light
truss, as shown in Fig. 2.15. The tensile load will be 46 kips. Use A36 steel. The length is
3
13.4 ft. fasteners will be -in.-diameter A325 bolts and will connect the double-angle
3
member to a -in.-thick gusset plate. Assume that the strength of the gusset plate will not
control.
3"
8
Section A-A
The required gross area, based on general yielding of the member, is calculated from
=
46
=
=
= 2.13 in.2
0.60 060(36)
33
34
35
= 300
13.4(12)
, =
=
= 0.54 in.
300
300
Remaining to be considered are (a) minimum required net area based on tensile fructure
and (b0 block shear, both of which depend on the thickness of the member and on
connection details.
3
A check of double-angle properties shows that angles of -in. thickness could provide the
required Ag of 2.13
in2.
1
4
-in. thick angles, compare the allowable load per bolt for
Use 4 bolts.
46
=
= 3.07
bolt allowable 15
2
2 ( ) = 2.0 in.
3
1"
14
3 @ 2"
1"
14
P = 46 kips
2 Ls
Fig. 2.16 Connection detail.
Changes in the angle thickness or changes in the connection, which affect strength, may yet
occur. For instance, if the bolt spacing or edge distance changes in the final connection, the
block shear strength will be affected.
1
Assuming an angle thickness of in., calculate the minimum required gross area based on
4
Therefore,
= = 0.50
= 0.50 ( )
+
(0.5)
46
3 1 1
+ 2 + = 2.30 in.2
0.85(0.5)(58)
4 8 4
1
4
= 2.38 in.2
= 0.891 in.
OK
OK
= + = (0.30 ) + (0.50 )
= 2(0.25)(7.25 3.5(0.875)) = 2.09 in.2
= 2(0.25)(1.25 0.5(0.875)) = 0.406 in.2
Use 2 L3 2 .
4
38
Start
Known: shape, end connection
details, l, Ag, Fy, Fu, least r.
Find Pt based on yielding in the gross section
(Pt = 0.60 Fy Ag)
Start
Determine
. , =
300
Select trial
section
, =
0.60
Assume reduction
coefficient U based on shape
and end connection details
Determine =
Determine Pt based on fracture
in the effective net section
= 0.50
Reselect trial section.
Ensure that min. reqd
r and Ag are provided
Determine Pt based
on block shear
= +
Specify section
to use
End
40
Adjust connection
and/or reselect trial
section. Ensure that
min. reqd r and Ag
are provided
3. Compression Members
Structural members that carry compressive loads are sometimes given names that identify
them as to their function. Compression members that serve as bracing are commonly called
struts. Other compression members may be called posts or pillars. Trusses are composed
of members that are in compression and members that are in tension. These may be either
chord or web members. The main vertical compression members in building frames are
called columns, which are of primary interest here.
Columns are compression members that have their length dimension considerabily longer
than their least cross-sectional dimensions.
Members considered here are subjected to axial loads (concentric loads), that is, the loads
are coincedent with the longitudenal centroidal axis of the member. This is a special case
and one that exists rarely. Where small eccentricities exist, however, it may be
assumedthat an appropraite factor of safety will compensate for the eccentricity, and the
column may be designed as ttough it were axially loaded. Columns may support varying
amounts of axial load and bending moment. If the range of possible combinations of load
and moment supported on columns, then at one end of the range is the axially loaded
column. This column carries no moment. At the other end of the range is the member that
carries only moment with no (or very little) axial load. (As a moment-carrying member, it
could be considered a beam). When a column carries both axial load and moment, it is
called a beam-column.
Commonly used cross sections for steel compression members include most of the rolled
shapes. These and other typical cross sections are shown in Fig. 3.1.
For the W shapes, the cross sections usually used are those that are rather square in shape
and that have nominal depths of 14 in. or less. These shapes are more efficient than others
for supporting compressive loads (the deeper shapes are more efficient when used as
bending members). For larger loads it is common to use a built-up cross section. In
addition to providing increased cross-sectional area, the built-up sections allow a designer
to tailor to specific needs the radius of gyration (r) values about x-x and y-y axes.
the member as a result of the method of manfacture, which involves inequal cooling rates
within the cross section.
(b)
(a)
(c)
(e)
Shapes
(f)
(d)
(g)
(h)
Built-up sections
Fig. 3.1 Compression member cross sections.
Ideal Columns
Eulers formula gives the buckling load Pe for a pin-ended, homogeneous, initially straight,
long column of an elastic material that is concentrically loaded. This is considered to be the
ideal column. The Euler buckling load is expressed as
where
2
2
Tests have verified that Eulers formula accurately prdicts buckling load, where the
buckling stress is less than (approximatly) the proportional limit of the material and
adherence to the basic assumptions is maintened. Since the buckling stress must be
compared with the proportional limit, Eulers formula is commonly written in terms of
stress. This may easily be derived from the preceding buckling load formula, recocnizing
that I = A r2:
=
where
2
()2
r = least radius of gyration of the cross section , where A is the cross-sectional area.
l / r is termed the slenderness ratio.
It is convenient to classify columns into three broad categories according to their modes of
failure as shown in Fig. 3.2.
(a) Short
(b) Intermediate
called a short column. A column that falls between these two extremes will fail by inelastic
buckling when a localized yielding occur. This will be initiated at some point of weakness
or crookedness. This type of column is called an intermediate column. Its failure strength
cannot be determined by using either the elastic buckling criterion of the long column or
the yielding criterion of the short column. It is designed and analyzed by using emperical
formulas based on extensive test results.
Example 1:
Determine the Euler buckling load Pe for an axially loaded W1422 shown in Fig. 3.3. The
column has pinne ends. Assume A36 steel with a proportional limit of 34 ksi. The column
length is 12 ft.
Pe=?
Pinned
Pinned
Pe=?
2
2 (29,000)
=
= 14.93 ksi
()2 [(12 12)/1.04]2
Or
2
2
2 (29,000)(7.00)
= 96.6 kips
(12 12)2
Note that in the preceding example, the use of higher-strength steel would not change the
value of the buckling load. The yield stress Fy plays no role in the Euler formula.
Effective Lengths
Eulers formula gives the buckling load for a column that has pinned ends. A practical
column, in addition to being nonperfect in other aspects may have end conditions (end
supports) that provide restraint of some magnitude and will not allow the column ends to
rotate freely.
The use of Eulers formula may be extended to columns having other than pinned ends
through the use of an effective length. This concept is illustarted in Fig. 3.4, where a column
having rigid (or fixed) ends is shown. The deflected shape of the buckled column is shown
by dashed line.
Pe
l/4
Fixed end
Inflection points
l l/2
l/2
Fixed end
l /4
Pe
(a)
(b)
and the effective length factor K is or 0.50. The effective length is written as Kl, where l is
2
45
the actual length of column. The Eulers formula may be rewritten with the inclusion of the
effective length as
and
For the fixed-ended column,
2
()2
2
(/)2
2
42
=
=
(0.5)2
2
It is seen that the buckling load is increased by a factor of 4 when rigid end supports are
furnished for a column.
Other combinations of column end conditions are covered in the ASD Specifications
Commentary. Table C-C2.1 provides theoretical K values for six idealized conditions in
which joint rotation and translation are either fully realized or nonexistent. Since there is
no perfectly rigid column support and no perfect pin support, the referenced table also
provides recommended design values for K where ideal conditions are approximated.
These values are slightly higher than the ideal values and therefore are conservative (the
predicted Pe will be on the low side).
Eulers formula for buckling load may also be adapted to result in an expression for an
allowable compressive load capacity (Pa). A factor of safety (F.S.) is introduced
=
Example 2:
F. S.
2
=
()2 (F. S. )
A W10 49 column of a36 steel has end conditions that approximate the fixed-pinned
condition (fixed at the bottom, pinned at the top, no sidesway). Assume a proportional limit
of 34 ksi.
(a) If the length of the column is 26 ft, find the allowable compressive load capacity, Pa,
using Eulers formula and factor of safety of 2.0.
46
(b) What is the minimum length of this column at which the Euler formula would still be
valid?
Solution:
From the ASD Specification Commentary, Table C-C2.1, K (for design)= 0.80. For the
W10 49, = 14.4 in.2 , = 93.4 in.4 , and = 2.54 in.
(a) Find fe first and check the applicability of Eulers formula:
=
2
2 (29,000)
=
= 29.6 ksi
(/)2 [0.8(26 12)/2.54]2
29.6(14.4)
=
=
= 213 kips
F. S. F. S.
2.0
Example 3:
2
2 (29,000)
=
=
= 291 in. = 24.3 ft
()2
34(0.82.54)2
Use Eulers formula to select an A-shape column to support an axial load of 50 kips. The
length is 12 ft, and the ends are pinned. Use A36 steel with proportional limit assumed to
be 34 ksi. Check the applicability of Eulers formula. Assume a factor of safety = 3.0 (Note:
this is not the AISC method of column selection).
Solution:
Try W6 20:
2
()2 (F. S. )
Check the applicability of Eulers formula and the capacity of the W6 20:
=
2
2 (29,000)
=
= 31.06 ksi < 34
(/)2 (144/1.50)2
182
=
= 60.8 kips > 50
F. S. 3.0
Since no two practical columns are identical, the failure stresses are expressed to fall
within a range of values for a particular Kl / r value. Columns with Kl / r values to the right
of line A-A have theor failure stresses closely predicted by Eulers formula. They are subject
to elastic buckling where the buckling occurs at a stress less than the proportional limit.
Columns with Kl / r values to the left of line A-A fail by inelastic buckling (yielding occurs),
and a departure of the test data from the curve that represents eulers formula is noted.
The ASD Specifications allowable stress for compression members, as found in Section E2,
may be shown as in Fig. 3.6.
The maximum Kl / r is prefeably limited to 200 for compression members. In the region
labled elastic buckling, the shape of the curve is closely follows the shape of the Euler
48
formula curve. It is esentially the same curve with a factoe of safety applied. Allowable axial
compressive stress section is denoted Fa.
49
The value of Kl / r that seperates elastic buckling from inelastic buckling has been
arbitrarily established as that value at which the Euler buckling stress (fe) is equal to Fy / 2.
This Kl /r value is denotes as Cc. Its value may be determined as follows:
Let fe = Fy / 2 and let Kl / r = Cc; then
From which
2
(/)2
2
=
2 ( )2
2 2
=
Table 4 in the Numerical Values Section of the ASD Specifications list values of Cc for
various values of Fy.
For column Kl / r values less than Cc, Fa is determined by
()2
]
2 2
=
5 3() ()3
3 + 8 8 3
[1
122
=
23(/)2
(2 1)
(2 2)
This is the famialiar Euler formula for buckling stress with a factor of safety of 23 / 12 or
1.92 incorporated.
ASD Manual contains tables that are useful in the determination of Fa; see Tables C-36 and
C-50 in Part 3 entitled Allowable Stresses for Comprssion Members.
Analysis Of Columns Using ASD Specifications:
Example 4:
Find the allowable compressive load capaity Pa for a W12 120 column that has a length of
16 ft. Use A36 steel. The ends are pinned.
50
Solution:
For the W12 120,
3.13
The Kl /r value has been rounded to the nearest whole number from table use.
ASD Manual, Part 3, Table C-36: Fa = 17.33 ksi.
Therefore,
Example 5:
A W10 68 column of A572 grade 50 steel is to carry an axial load of 400 kips. The length
is 20 ft. Determine whether the column is adequate if
(a) The ends are pinned.
(b) The ends are fixed.
Solution:
2.59
And from Table C-50, Fa = 16.29 ksi.
2.59
And from Table C-50, Fa = 22.72 ksi.
Example 6:
Find the compressive axial load capacity for a built-up column that has a cross section as
shown in Fig. 3.7. The steel is A36, the length is 18 ft, and the ends are assumed to be fixedpinned (totally fixed at bottom; rotation free, translation fixed at top).
6.38"
W12 X 65
6.00"
y
3
4"
12.12"
PL
18'
Determine the least moment of inertia for the built-up cross section:
= 533 + 2
= + 2
1
(0.75)(12.12)3 = 756 in.4
12
Notice in the Iy calculation the Ic terms for the plates have been neglected, since they are
very small.
Calculating the radious of gyration:
756
= =
= 4.50 in.
37.3
0.8(18)(12)
=
= 38.4 (use 38)
4.50
From Table C-36 of the ASD Manual, Part 3, gives Fa = 19.35 ksi.
Columns are sometimes braced differently about the major and minor axes, as shown by
column AB in Fig. 3.8. If all connections to the column are assumed to be simple (pinned)
connections, the deflected shapes for buckling about the two axes will be as shown.
may control, depending on which has the associate longer Kl / r ratio. Naturally, if there is
no reasonable certainty that the bracing will not be removed, the columns should be
designed with the bracing neglected.
Example 7:
Find the allowable compressive axial load capacity for a W10 88 that has an unbraced
length of 24 ft with respect to axis x-x and 12 ft with respect to axis y-y. Assume an A572
grade 50 steel member, pin-connected at the top and fixed at the bottom. (Assume that the
column is pin-connected at mid-height).
Solution:
= 2.63 in.
= 4.54 in.
1(12)(12)
=
= 54.8 (top part of column)
2.63
0.8(12)(12)
=
= 43.8 (bottom part of column)
2.63
0.8(24)(12)
=
= 50.8 (top part of column)
4.54
For
Square and rectangular HSS are also commonly used as building columns. These are
manufactured under ASTM A500 in various grades (see ASD Manual, Part I, Table 3). The
tubular members are also relatively efficient and have an advantage in that end-connection
details are simpler than with the pipe columns.
Example 8:
54
Find the allowable compressive load capacity for a 10-in standard steel pipe column that
has an unbraced length of 15 ft. ends are pin-connected (K = 1), and the steel is A53 grade B
(use Fy = 36 ksi).
Solution:
For the 10-in. standard steel pipe (ASD Manual, Part I),
= 11.9 in.2
= 3.67 in.
1(15)(12)
=
= 49
3.67
= 18.44 ksi
Example 9:
Find the allowable compressive load capacity for an HSS8.625 0.322 column that has
unbraced length of 20 ft. The ends are pin-connected and the steel is ASTM A500 grade B.
Solution:
For this round HSS, Fy = 42 ksi (see ASD Manual, Part 1, Table 3)
From the ASD Manual, Part 1,
Therefore,
= 8.40 in.2
= 2.94 in.
1(20)(12)
=
= 82
2.94
From ASD Manual, Part 5, Table 4, Cc = 116.7. Since Fy = 42 ksi and the ASD Manual contains
Fa tables for only Fy = 36 ksi and Fy = 50 ksi, Table 3 in Part 5 will be used. Obtain Cc by
calculation or fromTable in Part 5:
Therefore, since
= 116.7
82
=
= 0.70
116.7
55
Note also that ASD Specification Equation (E2-1) could be used for Fa.
Example 10:
Find the allowable compressive load capacity for a W8 40 with an unbraced length equal
to 26 ft. The member is usedin a wind-bracing system and is pin-connected. Use A36 steel.
Solution:
= 11.7 in.2
= 2.04 in.
1(26)(12)
=
= 153
2.04
= 6.38 ksi
The selection of cross sections for columns is greatly facilitated by the available of design
aids. The allowable axial stress Fa depends on the effective slenderness ratio Kl / r of the
column provided. Therefore, there is no direct solution for a required area or moment of
inertia. If ASD Specifications Equation (E2-2) were knowon to control, a required least r
could be calculated. This is no a practical solution, however.
Most structural steel columns are composed of W shapes, structural tubing, and / or pipes.
The ASD Manual, Part 3, contains allowable axial load tables (referred to as the column
load tables) for the popular column shapes. Allowable loads (Pa) are tabulated as a
function of KL (in feet) and cover the common length L for columns in building frames is
normally taken as the floor-to-floor distance since the floor provide lateral bracing. The
effective length factor K may be determined by using the aids discussed before. For
multistory frames, steel is commonly erected in two-story sections for reason of safety,
convenience, and the need to maintain alignment of the structure. Column splices are
placed just above a floor level.
56
The column load tables, previously discussed, may be used for analysis as well as for
design. For instance, in previous example the allowable compressive load capacity of a
W12 120 of A36 steel was computed to be 612 kips. From the ASD Manual, Part 3, column
load table for the W12 120, with = 1.0 16 ft, the allowable load of 611 kips may be
obtained directly.
The tubular values of allowable loads are with respect to the members minor (or weak)
axis. Although the column load tables are indispendable for the selection of the types of
cross section noted, if built-up sections are required, or a section is desired for which a
column load table is not available, trial-and-error calculation approach will have to be used.
Example 11:
Select the lightest W shape for a column that will support an axial load P of 200 kips. The
length of the column will be 20 ft, and the ends may be assumed to be pinned. Use A992
steel.
Solution:
Using the ASD Manual, Part 3, column load tables, with = 1 20 = 20 ft and
= 200 kips, the following W shapes are observed to be adequate ( ):
14 61
( = 272 kips)
10 49
( = 230 kips)
12 53
8 67
( = 241 kips)
( = 230 kips)
Select the lightest W10 for column AB previously discussed, P = 160 kips. The overall
length (Lx) is 30 ft. The weak axis is braced at midheight (Ly = 15 ft). Assume pinned ends
(K = 1.0) for box axes and A36 steel.
Solution:
Assume that weak axis (y-y axis)will control. Select the column using ASD Manual, Part 3,
column load tables, and then check whether the assumption is correct:
= 15 ft
= 160 kips
57
= 4.27 in. ,
= 1.98 in.
1(30)(12)
=
= 84.3
4.27
1(15)(12)
=
= 90.9
1.98
The larger Kl / r controls, and the assumption of the weak axis controlling was correct. A
W10 39 will be adequate.
Under different conditions it is possible that the strong axis will control and be the buckling
axis for the column. The column load tables then cannot be used directly. Once an initial
section has been selected (based on the assumption that the y-y axis controls), however, a
very rapid analysis check can be made by using the tabulated properties at the bottom of
the column load tables. The procedure is as follows:
2. Compare with KLy. The larger of the two values becomes the controlling KL.
3. With the controlling KL value, find Pa in the appropriate column load table.
Example 13:
Rework the previous example except this time, the weak axis is braced at the third points
so that Ly = 10 ft. Lx remian at 30 ft.
Solution:
= 160 kips
Try a W10 33, Pa = 167 kips (based on weak axis controlling) and rx / ry = 2.16:
30 ft
=
= 13.89 ft
2.16
This is an equivalent weak-axis length (i.e., column length based on weak-axis buckling
that results in the same capacity as does the 30-ft strong-axis buckling length).
Since 13.89 ft Ly, the strong axis controls.
58
Rounding the KL of 13.89 ft to 14 ft and entring the column load table for the W10 33
gives Pa = 142 kips, and
142 kips 160 kips
Try W10 39:
30 ft
=
= 13.89 ft 14 ft
2.16
13.89 ft >
Therefor, the strong axis controls. From the column load table,
Therefor, use a W10 39.
For a plate covering less than the full area of concrete support,
where
2
= 0.35 0.70
1
The base plate must be thick enough that bending in the plate itself will not be critical.
Two-way bending is involved, since as the column pushes down on the base plate, the parts
of the plate not directly under the column itself will tend to curl (or deflect) upward. For all
but the smallest base plates, the required plate thickness may be determined by
considering 1-in.-wide sections of the base plate to act as cantilever beams spanning in
each of two directions, fixed at the edges of a rectangle whose sides are 0.80bf and 0.95d, as
shown in Fig. 3.9.
Notation:
A
m
1"
B or N
P
1"
N 0.95d
tp
A
fp
Section A-A
Assumed
critical planes
for bending
A
n
0.80bf
The column load P is assumed to be uniformly distributed over the top of the base plate in
sufficiently rigid and will distribute the applied load so that the pressure underneath the
base plate is also uniformly distributed.
The most economical column base plate will result if the length N and the width B are
selected so that m = n. This condition is approached when
1 +
Once the length N of the plate has been determined, the required B can be computed from
After m and n can be determined.
required =
required 1
The allowable bending stress in the plate is taken as 0.75 Fy (ASD Specifications, Section
F2). The required thickness of the base plate can be computed from
= 2
= 2
Note that the largest required plate thickness will rsult from the larger value of m or n.
When the base plate is just large enough in area to accommodate the outside envlope of the
column (dimensions d and bf), the values of m and n will be small. The resulting plate
thickness will then be small also, and the assumption of uniform bearing pressure under
the plate is no longer valid. For light loads with this type of base plate, the colmun load is
assumed to be distributed on an H-shaped area of the footing under the plate. This area is
shaded in Fig. 3.10.
For small base plates that more heavily loaded, the required plate thickness may be taken
as
= 2
where
=
Column
CL
Wide flange
column
2L
Column
web
Assumed
loaded area
plate
L
L L
Plan view
Section A-A
21 1
1.0
< 1.0
When is less than 1.0 or q is less than 0.64, the design for lightly loaded base plates
governs. If > 0.64, take as 1.0.
After computing , calculate and determine the required plate thickness from
= 2( )
62
Since the expression for required tp is in the same from as the expression for tp for large
plates, the largest dimensions (, , or ) will control, and expression may be rewritten
as
required = 2
Design a rectangular base plate for a W14 74 column that it is to carry an axial load of
350 kips. Assume that the base plate will cover the full area of a concrete pier of = 3 ksi.
Use A36 steel.
Solution:
required 1 =
350
=
= 333 in2
1.05
= 0.5(0.95 0.80 )
= 0.5[0.95(14.17) 0.8(10.07)] = 2.70 in.
required 1
333
=
= 15.89 in.
20.95
= 16 in.
350
=
= 1.042 ksi < 1.05
16(21)
4. Calculate m, n, and n:
=
0.95 21 0.95(14.17)
=
= 3.77 in.
2
2
0.8 16 0.8(10.07)
=
= 3.97 in.
2
2
14.17(10.07)
=
= 2.99 in.
4
4
5. Check for the case of a lightly loaded small base plate using
=
=
+
4(1.042)(14.17)(10.07)
(14.17 + 10.07)2 (1.05)
Therefore, take = 1.0. Alternativly, the upper limit ( = 1.0) can always be used as a
conservative assumption.
6. Calculate the required plate thickness using c as a larger , , and :
1.042
required = 2 = 2(3.97)
= 1.351 in.
36
7. Refer to the ASD Manual, Part 1, Bars and Plates Products Available for
3
Design an economical rectangular column base plate for a W12 50 column that is to carry
an axial load of 65 kips. All steel is A36. Assume that the base plate will cover the full area
of the concrete support. Here = 3.0 ksi; therefore,
Solution:
1. The required area is
required 1 =
65
=
= 61.9 in2
1.05
2. For the W12 50, = 12.19 in. and = 8.08 in. The area of the rectangular
profile of the column is
= 8.08(12.19) = 98.5 in.2
98.5 in.2 > 61.9 in.2
Therefore, to accommodate the envelope of the W12 50, B and N must be selected to the
full inch dimensions that are larger than bf and d. Use N = 13 in. and B = 9 in. This will
furnish an area of 117 in.2
3. The actual bearing pressure under the plate is
=
4. Calculate m, n, and n:
=
65
=
= 0.556 ksi < 1.05
9(13)
0.95 13 0.95(12.19)
=
= 0.710 in.
2
2
0.8 9 0.80(8.08)
=
= 1.268 in.
2
2
12.19(8.08)
=
= 2.48 in.
4
4
+
4(0.556)(12.19)(8.08)
=
= 0.508 < 0.64
(12.19 + 8.08)2 (1.05)
21 1
21 1 0.508
0.508
= 0.838
0.556
= 2(2.08)
= 0.517 in.
36
16
1 1.
16
in.
required 1 =
2. Select B and N so that m and n are approximately equal (if possible). Use
where
required 1 +
= 0.5(0.95 0.80 )
required =
required 1
required 1
66
0.95
2
0.80
2
5. Check for the case of a lightly loaded small base plate using
< 1.0
21 1
1.0
If q 0.64, the design for a lightly loaded base plate governs ( should be determined). If q
0.64, then = 1.0. The upper limit ( = 1.0) can always be used as a conservative
assumption and will simplify the computations a bit.
6. Calculate the required plate thickness tp using c as the larger of , , and :
= 2
67
Determine L, P, Fy,
bracing conditions,
and end conditions
No
for analysis.
Yes
Specify column
section to use
. Make use of
req d
Calculate
Calculate , , ,
Conservatively
take = 1.0
+
0.64?
Calculate
No
Calculate = 2
= 1.0
1.0
Yes
21 1
1.0
End
4. BEAMS
Beams are among the most common members that one will find in structures. They are
sructural members that carry loads that are applied at right angles to the longitudinal axis
of the member. These lloads cause the beam to bend.
In the process of beam design, the bending moment, which is produced in the beam by the
loads it supports, will be of initial concern. Other effects, such as shear or deflection, may
eventualy control the design of the beam and will have to be checked.
Beams are sometimes called by other names that are indicative of some specialized
fucntion(s):
Girder: a major, or deep, beam that often provides support for other beams,
Spandrel: a beam on the outside perimeter of a building that supports, among other loads,
the exterior wall,
Purlin: a beam thatsupports a roof and frames between or over supports, such as roof
trussesor rigid frames, and
Grit: generally, a light beam that supports only the lightweight exterior sides of a building
(typical in preengineered metal buildings).
Mechanics of Bending
The maximum stress due to flexure (bending) in the beam may be determined by use of the
flexure formula:
where
69
= distance from the nuetral axis to the extreme outside of the cross section
= moment of inertia of the cross section about the bending nuetral axis
= section modulus ( = ) of the cross section about the bending neutral axis.
Fig. 4.1 shows the cross section and depicts the resulting bending stress diagram. The
shape of the diagram is typical for bending stress at any pint along the beam. Several points
should be noted.
fb
Flange
y
Web
x
Compression
Tension
Neutral axis
y
fb
The allowable bending stress, Fb is specified by the ASD Specifications based on the type of
steel that is beong used and other conditions that affect the strength of the beam in
bending.
For equilibrium the internal resisting moment at any point in the beam must be equal to
the external applied moment at the same.
If the external applied moment becomes the maximum allowed, the bending stresses at the
top and bottom of the beam will be equal to the allowable bending stress Fb. Additional
70
moment should not be applied, since this would cause the actual bending stress to exceed
Fb. The internal resisting moment that when the bending stress is Fb is termed MR.
From the flexural formula the resisting moment MR can be calculated by substituting Fb for
fb, and MR for M:
=
Then
(1 1)
where Fy is the material yield stress. For a member to qualify for an alowable bending
stress Fb of 0.66Fy, it must have an axis of symmetry in, and be loaded in, the plane of the
web. An important condition associated with the use of this value for Fb is the lateral
support of the compression flange. The compression flange behaves somewhat like a
column, and it will tend to buckle to the side, or laterally, as the stress increases if it is not
restrained in some way. Varying amounts and types of lateral support may be present as
shown in Fig. 4.2.
The distance between the points of lateral support in inches is denoted l. for convenience
this distance will be denoted Lb when it is in feet.
To qualify for Fb = 0.66Fy, the compression flange of a beam must have adequate lateral
support such that
where
76
and
71
20,000
Another important condition that must be met if the beam cross section is to qualify for Fb
= 0.66Fy deals with the response of the beam in an overload situation. Allowable stress
design assumes failure to occur when Fy is first reached. The beam will not fail at this point
because it has a substantial resrve of strength. If the cross section continues to strain under
increased moment, the outer fibers will further strain, but the stress will remain at Fy. Fy
will be reached by the fibers at levels progressively closer to the neutral axis until virtually
72
the entire cross section is stressed to Fy. When this occurs, the beam has achieved its
plastic moment capacity (this is the basis for plastic design). The cross section, however,
must be proportioned so that no local buckling of the flange or web occur before the plastic
moment capacity is achieved. A cross section that meets this criterion is said to be
compact. The 1989 ASD Specifications, Section B5.1, clasifies steel sections as compact,
noncompact and slender element sections. Only compact sections qualify for Fb = 0.66Fy.
The test for compactness is found in the ASD Specifications, section B5 and table B5.1. the
governing criteria are the width-thickness ratio of the compression flange and compression
web elements of the cross section. These are called the flange criterion and web
criterion, respectivly.
Assuming that there are no axial loads on the beam, and using the defentions for width and
thickness from the ASD Specifications, section B5.1, the two equations required may be
simplifed as follows.
For a section to be considered compact:
65
2
640
Both the flange and the web criteria must be satisfied for a member to be considered
compact.
Example 1:
73
Solution:
= 0.680 in.
= 18.21 in.
= 0.425 in.
11.035
=
= 8.11
2 2(0.68)
65
65
=
= 10.8
36
8.11 < 10.8
18.21
=
= 42.8
0.425
640 640
=
= 106.7
36
The preceding could be shortened by using tabulated quantities from the ASD Manual, part
1, Properties of A shapes, or the Numerical Values furnished in Table 5 of the ASD
Specifications. The tabulted quantities are rounded slighly in some cases.
A faster way to determine compactness of cross section the rolled shapes is to calculate
the value of a hypothetical yield stress Fy that would cause equality in each of the two
criteria. For the flange criterion,
65
=
2
= (
65 2
)
2
74
65 2
) = 64.2 ksi
= (
8.11
This shows that the flange criterion is satisfied provided that Fy does not exceed 64.2 ksi.
This value is termed and is tabulated in the ASD manual, Part 1, as a property of the
W18 76. If the W18 76 is of a steel with Fy in excess of 64.2 ksi, it is not compact by the
flange criterion. Therefore, for compactness by the flange criterion, the following
condition should exist:
Should the calculated value of be in excess of the heighst available , this is reflected by
the tabulation of a dash () for the value. For the web criterion,
640
640 2
)
= (
640 2
) = 224 ksi
= (
42.8
This shows that the W18 76 is compact by the web criterion provide that Fy does not
exceed 224 ksi. Reference to the ASD Manual, Part 1, table 1, shows that shapes with Fy in
excess of 65 ksi are currently not available. Therefore, the W18 76 is compsct by the web
criterion in all steels. This is a general rule. All rolled, W, M, an S shapes tabulated in the
ASD Manual are compact by the web criterion ( when fa = 0). This does not hold true for
built-up section and plate girders. Web noncompactness will cause Fb to be reduced to
0.60Fy (assuming adequate lateral support).
If a shape does not satisfy the flange criterion, it is considered a noncompact shape, and Fb
must be reduced. The variation of Fb for rolled W shapes that have adequate lateral
support is summarized graphically in the Fig. 4.3.
For the range of /2 between 65/ and 95/, the ASD Specifications provides for a
linear reduction in Fb to 0.60Fy according to the following equation:
75
= [0.79 0.002 ( ) ]
2
(1 3)
In determining Fb, the yield stress Fy must be known. The best source for the value of Fy is
the ASD Manual, Part 1, Tables 1 and 2. For a known shape, determine the appropriate
group from Table 2. Then, knowing the steel type, use Table 1 to determine Fy.
Example 2:
Find Fb for the following shapes. Assume adequate lateral support for the compression
flange.
(a) W30 132 of A36 steel.
(b) W12 65 of A242 steel.
Solution:
(a) All W shapes in a36 steel are compact except for the W6 15, since its is less than
36 ksi. From the ASD Manulal, Part 1, Table 1, all shapes in A36 steel have Fy = 36
ksi. Thus
76
(b) From the ASD Manual, Part 1, Table 2, W12 65 is found in Group 2. From Table 1,
Fy = 50 ksi. From the ASD Manual properties for the W12 65, = 43.0 ksi. Since
> , the member is not a compact shape in A242 steel. Another check for
compactness is as follows. The following quantities can also be found in the ASD
Manual, Part 1, Properties of W Shapes, and Table5 of the ASD Specifications
(Numerical Values):
= 9.9
2
65
65
=
= 9.2
50
95
95
=
= 13.4
50
95
65
<
<
2
= [0.79 0.002 ( ) ]
2
77
A W21 44 beam is to span 24 ft on simple supports as shown in Fig. 4.4. Assume full
lateral support and A36 steel. The load shown is superimposed load, meaning that it does
not include the weight of the beam. Determine whether the beam is adequate by
(a) Comparing the computed bending stress with the allowable bending sress.
(b) Comparing the computed applied moment with the resisting moment MR.
1.75 kips/ft
(superimposed load)
24'-0"
From the properties tables, for the W21 44, = 81.6 in.3 Moment may be determined by
shear force and moment diagram or by formula (see the ASD Manual, Part 2, Beam
Diagrams and Formulas) for review).
The total load should include the weight of the beam:
1.75
kips/ft
+0.044
kip/ft
1.794
kips/ft
(applied load)
(beam weight)
2 1.794(24)2
Applied moment =
=
= 129.2 ft. kips
8
8
=
129.2(12)
=
= 19.00 ksi
81.6
78
From the ASD Manual, Part 1, = (). Therefore, since > , the member is
compact in A36 steel, and the allowable bending stress Fb is 24.0 ksi. Therefore,
<
(b) The applied moment m has been determined to be 129.2 ft.kips. The resisting
moment MR may be calculated from the flexure formula:
24(81.6)
= =
= 163.2 ft. kips
12
Therefore,
<
Example 4:
A W18 40 beam spans 20 ft span, as shown in Fig. 4.5. Assume A992 steel. The
comprssion flange is supported laterally at the quarter points where equal concentrated
load P are applied. Therefore, Lb = 5 ft. Determine the allowable value for each load P (kips).
indicates lateral support
for compression flange
5'-0"
5'-0"
5'-0"
5'-0"
= 5.67
= 50 ksi
79
= 68.4 in.3
= ()
Test for adequate lateral support by comparing Lb with Lc. From the ASD Manual, Part 2,
Allowable Stress Design Selection Table, Lc = 5.4 ft. Lb = 5 ft. Since Lb Lc, this beam has
adequate lateral support. = (), implying that is high enough that this shape is always
compact. Therefore,
= 0.66 = 33.0 ksi
= =
33.0(68.4)
= 188.1 ft. kips
12
The resisting moment is 188.1 ft.kips. The applied moment due to the beams own weight
and the moment due to the three equal loads P cannot exceed the resisting moment MR. the
applied moment due to beamweight is
2 0.04(20)2
=
=
= 2.0 ft. kips
8
8
The moment due to the concentrated loads may be determined by using the shear and
moment diagrams shown in Fig. 4.6 , or aids such as those found in the ASD Manual, Part 2,
Beam Diagrams and Formulas.
P
3P
2
5'-0"
5'-0"
5'-0"
3P
2
P
2
P
2
5'-0"
3P
2
SFD
3P
2
10 P
BMD
Thus,
= 10 ft. kips
10 = 186.1
186.1
= 18.61 kips
10
Therefore, the maximum allowable value of each concentrated load P on this beam is 18.61
kips.
It should be noted that the weight of the beam itself in both of the forgoing examples has
been a very minor part of the total load carried. This is generally true; nevertheless, it
should always be considered. As one becomes more experienced, the effect of beam weight
is easier to estimate. Various shortcuts and rules of thumbs are used by designers to
simplfy the inclusion of weight of structure in design problems. In analysis problems,
where the cross sectionis known, inclusion of the beam weight is a simple matter.
The following procedure for beam analysis for moment only is general and typical for the
various shapes that may be used for beams, primarily wide-flange sections and, to a lesser
extent, other sections. The precise method of solution will depend on the nature of the
particular problem, the known conditions, and the information sought.
1. Determine Fy. Use the ASD Manual, Part 1, Tables 1 and 2.
2. Check the adequacy of lateral support. See the ASD Specifications, Section F1. If
lateral support is inadequate, Fb will be calculated as in next section.
3. Check the compactness of the cross section. Use from the table of properties in
the ASD Manual, Part 1.
4. Using the preceding information, determine Fb.
5. If the applied loads are known, the applied moment can be found. Draw shear and
moment diagrams or use beam formulas from the ASD Manual, Part 2.
6. If the magnitude of the applied loads is unknown, write an expression for the
applied moment in terms of the unknown loads. This can then be equated to the
resisting moment of the beam.
81
Unloaded
position
Deflected
position
Deflected and
buckled
position
A
l
(a)
The ASD Specifications, Section F1.3, establishes empirical expressions for Fb for the
inadequate lateral support situation. Tension and compression allowable bending stresses
are treated separately. The tension Fb is reduced. For typical rolled shapes, this of no
consequence because the shapes are symmetrical and the lower Fb of the two values will
control.
The ASD Specifications provides three empirical equations for the reduced compression Fb.
The mathematical expressions that give an exact prediction of the buckling strength of
beams are too complex for general use. Therefore, the ASD specifications equations only
approximate this strength for purposes of determining a reasonable Fb. The Fb that is
finally used is the larger of the Fb values determined from the applicable equations. The
first two ASD specifications Equations (F1-6) and (F1-7), gives the Fb value when the
lateral bending resistance of the compression flange provides the lateral buckling
resistance. The third, ASD specifications Equation (F1-8), gives Fb when the torsional
resistance of the beam section provides the primary resistance to lateral buckling. In no
case should Fb be greater than 0.60Fy for beams that have inadequate lateral support. The
equations that will be applicable depend on the value of the ratio , where
= distance between points of lateral support for the compression flange (in.)
= radius of gyration of a section comprising the compression flange plus one-third of the
compression web area taken about an axis in the plane of the web (in.) as shown in Fig. 4.8.
y
tf
d d
2 2 - tf
1 d
3(2-tf)
Compression
Neutral axis
y
83
( )2
2
]
= [
3 1530 103
=
where
170 103
( )2
12 103
( )
(F1-6)
(F1-7)
(F1-8)
= a liberalizing modifying factor whose value is between 1.0 and 2.3 that accounts for a
moment gradient over the span and a decrease in the lateral buckling tendancy; Cb may be
conservatively taken as 1.0; see ASD Specifications, Section F1.3, for details.
= depth of cross section (in.)
102 103
??
Yes
= 0.66
No
510 103
??
Yes
No
Note that one will use ASD Specifications Equations (F1-6) and (F1-8) or ASD Specifications
Equations (F1-7) and (F1-8). The larger resulting Fb is used. Note that Table 5 of the
Numerical Values Section of the ASD Specifications provides the following numerical
equivelants for A36 steel (Fy = 36 ksi):
84
Example 5:
102 103
= 53
510 103
= 119
A W21 50 shown in Fig. 4.9, spans 36 ft on simple span. The compression flange is
laterally supported at the third points. A36 steel is used. Determine Fb for this beam.
indicates lateral support
for compression flange
12'-0"
12'-0"
12'-0"
= 5.96
= 6.53 in.
= 1.60 in.
To qualify for Fb = 0.66Fy, the unbraced length of the compression flange must be equal, or
less than,
76
and
20,000
( )
As required in the ASD Specifications, Section F1.1. For A36 steel, these expressions
become
85
12.7 and
556
As was mentioned previously, the lesser of these two expressions is designated Lc (in feet).
If Lb Lc, the beam will qualify for Fb = 0.66Fy.
= 12 ft
= 6.9 ft
(from Table 5 of the Numerical Values section of the ASD Specifications). Next, compute the
slenderness ratio of the compression flange:
144
=
= 90.0
1.60
53 = 53
119 = 119
53 <
< 119
( )2
2
2
36(90)2
]
[
] 36 = 17.14 ksi
= [
3 1530 103
3 1530(103 )(1.0)
86
= 17.14 ksi
The ASD Specifications equations for Fb for beams that have inadequate lateral support are
also applicable to built-up members and plate girders, provide that they have an axis of
symmetry in the plane of the web.
Example 6:
Determine the allowable superimposed uniformly distributed load that may be placed on
the W21 50 of previous example.
Solution:
17.41(94.5)
= 135.0 ft. kips
12
8 8(135.0)
=
= 0.833 kip/ft
2
362
When the beams own weight is subtracted, the allowable superimposed load is
833 50 = 783 lb/ft = 0.783 kip/ft
The basis for moment design is to provide a beam that has a moment capacity (MR) equal
to or greater than the anticipated maximum applied moment M. The flexure formula is used
to determine a required sectiom modulus S:
required =
The section modulus on which the selection will be based is assumed to be the strong-axis
section Sx. The allowable Stress Design Selection Table (Sx Table) in the ASD Manual, Part 2,
can be use to make this selection. It lists common beam shapes in order odf decreasing
87
section modulus. This table also lists the resisting moment MR of each section. The value of
MR is calculated using an allowable stress Fb of 23.8 ksi (or 23.76 ksi) rather than the
rounded value of 24.0 ksi. This may cause some small inconsistencies in calculations and
results.
Example 7:
Select the lightest W shape for the beam shown in Fig. 4.10. Assume full lateral support (Lb
= 0) and A36 stee. Consider moment only.
Solution:
The beam reactions are determined from the load diagram; the shear and moment
diagrams are drawn.
6k
3 kips/ft
14'- 0"
14'- 0"
(a)
45 k
6k
V
(b)
336 ft-kips
M
(c)
required =
336(12)
=
= 168.0 in.3
24
From the ASD Manual, Part 2 (Allowable Stress Design Selection Table), select a W24 76
with an Sx = 176 in3. It is the lightest W shape that will furnish the required section
modulus. Note that the section selected weighs 76 lb/ft. Add in the effect of the beam
weight:
2 0.076(28)2
additional =
=
= 7.45 ft. kips
8
8
new total = 336 + 7.45 = 343 ft. kips
new required =
343(12)
=
= 171.5 in3
24
The W24 76 is satisfactory, since 176 in3 171.5 in3 required. Also, now check the
assumed Fb; the W24 76 is compact, since > , and has adequate lateral support;
therefore the assumed Fb is satisfactory. Use W24 76.
Example 8:
The beam shown in Fig. 4.11 is to be of A36 steel. Note the lateral support conditions. Select
the lightest W shape. Consider moment only.
Solution:
For this design, an estimated beam weight of 40 lb/ft (0.04 kip/ft) has been added to the
given uniform load. (This estimate may be based on anything from an educated guess to a
rough design worked quickly on scrap paper).
Establish Fb.
Fy is 36 ksi, and a compact shape that has adequate lateral support will be assumed.
From which
required =
129.0(12)
=
= 64.5 in.3
24
Select a W16 40 with Sx = 64.7 in3. This is a compact shape, since > . Check the
adequacy of lateral support (l = 60 in or Lb = 5 ft) by comparing Lb with Lc. From the ASD
89
Manual, Part 2, Lc = 7.4 ft. Therefore, since Lb Lc, the beam has adequate lateral support,
and the assumed Fb is satisfactory. The assumed beam weight is satisfactory.
Use .
10 k
5k
4.04
kips/ft
5'
5'
5'
5'
(a)
35.27
V
(kips)
25.20
15.07
5.07
5.0
15.13
25.13
(b)
45.33
129.0
M
(ft.kips)
(c)
75.4
The shape of plot is typical for a compact cross section. A plot of MR versus Lb, would have
the same form, since MR = FbSx and Sx is constant for the cross section.
A family of these curves is found in the ASD Manual, Part 2. These curves make up a very
valuable design aid for beams and should be used whenever possible. The lightest beam
section may be selected directly by using only the applied moment M (ft-kips) and the
unbraced length of compression flange Lb (ft). Note that the vertical axis for these curves is
allowable moment. This is the same as resisting moment MR. The curve values of
maximum MR (for the case where Lb Lc) are calculated using an allowable bending stress
Fb of 24.0 ksi. Therefore, these maximum M values do not correlate exactly with those
tabulated in the Allowable Stress Design Selection Table. The differences are very small.
The two other terms are:
Lc = Maximum unbraced length (ft) of the compression flange at which the allowable
bending stress may be taken at 0.66Fy (for compact shapes), or as determined by ASD
Specifications Equation (F1-3) or (F2-3) (when applicable), Lc is the smaller value obtained
from
76
and
91
20,000
( )
Lu = Maximum unbraced length (ft) of the compression flange at which the allowable
bending stress may be taken at 0.60Fy.
With Fb = 0.60Fy and Cb = 1.0, the value of Lu (in feet) for most shapes is given as
20,000
( )
from the ASD Specifications Equation (F1-8). For a few shapes, Lu (in feet) is given as
102,000
12
from ASD Specifications Equation (F1-6). Lu is taken as the larger value obtained from these
two expressions. For lengths greater than Lc but not greater than Lu, Fb may be taken as
0.60Fy. In no case is Lc taken as greater than Lu. Lc and Lu are unique for each shape and also
vary with Fy. They are tabulated properties and may be found in the ASD Manual, Part 2,
Allowable Stress Design Selection Table, and in the Beam Tables of Part 2, for Fy of 36 ksi
and 50 ksi. Their use greatly facilitates the determination of whether or not lateral support
is adequate. The precise values of Lc and Lu, however, are of no consequence when using
the beam curves.
When using the curves, any shape represented by a curve that is found above or to the
right of a particular MR and Lb combination is a shape that is satisfcatory. The lightest
adequate shape is represented by a solid line. The dashed lines represent shapes that are
also adequate but are heavier. The ASD Manual contains beam curves for Fy = 36 ksi and Fy
= 50 ksi. These curves may be used for the range of Lb from 0 up to the maximum value
indicated in the curves.
Example 9:
Select the lightest W shape for the beam shown in Fig. 4.13. The depth of the beam is
limited to 36 in. maximum. Consider the following cases:
(a) A36 steel, Lb = 4 ft.
(b) A36 steel, Lb = 16 ft.
(c) A572 grade 50 steel, Lb = 20 ft.
Solution:
(a)
92
Fy = 36 ksi. Determine the applied moment for a simply supported, uniformly loaded,
single-span beam. Neglect the beam weight temporarly. Thus
=
2 7.0(32)2
=
= 896 ft. kips
8
8
32'
From the beam curves, MR for W36 150 with Lb = 4 ft is 1008 ft-kips:
Use
(c) Fy = 36 ksi. M = 896 ft-kips from part (a). Lb = 16 ft. From the beam curves, select a
W36 160. For Lb of 16 ft, MR = 960 ft-kips for this shape. Add the moment due to
the beam weight:
2 0.16(32)2
additional =
=
= 20.5 ft. kips
8
8
Use
(c) Assume that Fy = 50 ksi. M = 896 ft-kips from part (a). Lb = 20 ft.
From the beam curves, select a W36 150 with MR = 960 ft-kips.
93
2 0.15(32)2
additional =
=
= 19.2 ft. kips
8
8
Check Fy. From the ASD Manual, Part 1, Table 1 and 2, the W36 150 is a Group 2 shape
OK
and Fy = 50 ksi.
Use
When the curves are not applicable (e.g., when Fy equals some value other than 36 ksi or 50
ksi), Fb must be assumed and subesquently verified.
Example 10:
Rework the previous example. Select the lightest W shape for the beam shown. The framing
system used indicates the use of an Lb = 9.0 ft. The steel is to be A572 Grade 60 (Fy = 60 ksi).
Solution:
Assume that Fb = 0.66Fy = 0.66(60) = 40 ksi. Using the moment calculated in previous
example (a), which included a 150-lb/ft beam weight,
required =
915(12)
=
= 275 in.3
40
From the ASD Manual, Part 2 (Allowable Stress Design Selection Table), select a W30 108
(Sx = 299 in.3).
Now verify Fb. The section is comapct since = () (ASD Manual, Part 1), which indicates
that it is in excess of 65 ksi. Therefore, > . Since Lc is not tabulated for a stell with Fy =
60 ksi, it must be computed and compared with Lb = 9 ft. With reference to the ASD
Specifications, Section F1.1, Lc is the smaller of
76
and
76(10.475)
60
= 102.8 in
20,000
20,000
=
= 88.9 in
( ) 3.75(60)
94
Therefore,
88.9
= 7.41 ft
12
Since Lb Lc, Fb must be reduced to at least 0.60Fy and possibly lower. This will affect the
required Sx with a subsequent change in the section selected. Assume the new Fb to be
0.60(60) = 36 ksi.
=
915(12)
= 305 in.3
36
From the ASD manual, Part 2, select a W30 116 (Sx = 329 in.3). For this shape, a
calculation for Lc (similar to the preceding) will show that Lc = 8.3 ft.
Therefore, Lb is still greater than Lc. Now determine Lu as the larger of
and
20,000
20,000
=
= 8.3 ft
( ) 12(3.75)(60)
102,000
102,000 2.64
( )=
(
) = 9.1 ft
12
60
12
The beam weight included in the design moment is on the conservative side, since the
assumed 150 lb/ft is greater than the actual beam weight of 116 lb/ft. No modifications of
the calculations are necessary, however.
Use
Note 1: Other beam design aids are available, such as the table of Allowable Loads on
Beams also found in the ASD Manual, Part 2.
Note 2: The curves found in the ASD Manual, Part 2, may also be used for analysis problems
as well as for design. If the beam size and unbraced length (Lb) are known, the allowable
moment may be obtained from the curves.
Example 11:
95
Entering the curves for a W36 150 with an Lb = 24 ft, the total allowable moment = 680
ft-kips. The moment due to beams own weight is
0.15(24)2
=
= 10.8 ft. kips
8
Since the applied moment can equal the allowable moment, as a limit,
=
2
8
8 8(669)
=
= 9.29 kips/ft
2
242
4. After the section has been selected, recompute the design moment, including the
effect of the weight of the section. Check to ensure that the section selected is still
adequate.
5. Check any assumptions that may have been made concerning Fy or Fb.
6. Be sure that the solution to the design problem is plainly stated.
96
Shear in Beams
Except under very special loading conditions, all beams are subjected to shear as well as
moment. In the normal process of design, beams are selected on the basis of the moment to
be resisted and then checked for shear. Shear rarely controls a design unless loads are very
heavy (and, possibly, close to the supports) and/or spans are very short. From strength of
materials, the shear stress that exists within a beam may be determined from the general
shear formula
=
where
= shear stress on a horizontal plane located with reference to the nuetral axis (ksi).
= vertical shear force at that particular section (kips).
= statical moment of area between the plane under consideration and the outside of the
section, about the nuetral axis (in.3).
This formula furnishes us with the horizontal shear stress at apoint, which is equal in
intensity to the vertical shear stress at the same point in a beam.
Example 12:
A W16 100 is subjected to a vertical shear of 80 kips. Determine the maximum shear
stress and plot the distribution of shear stress for the entire cross section.
Solution:
Design dimensions for the W16 100 are shown in Fig. 4.14. Other properties are
= 29.4 in.2
= 1490 in.4
The maximum shear stress will be at the neutral axis (where Q is maximum). Taking Q for
the shaded area about the neutral axis, having
= 10.425(0.985) 7.5 +
0.985
7.5
+ 7.5(0.585) = 98.5 in.3
2
2
97
maximum =
80(98.5)
=
= 9.04 ksi
1490(0.585)
10.425 "
0.42 ksi
7.53 ksi
0.985 "
7.500 "
9.04 ksi
16.970 "
0.585"
Note that the flanges resist very low shear stresses. Even though the areas of the flanges
are large, it is the web that predominantly resists the shear in wide-flange beams. For this
reason the ASD Specifications allows the use of an average web shear approach for the
shear stress determination:
=
where
This method is approximate compared with the theoritically correct general shear formula
and assumes that the shear is resisted by the rectangular area of the web extending the full
depth of the beam. For the W16 100 of previous example,
=
80
=
= 8.06 ksi
16.97(0.585)
98
This is less than the shear stress of 9.04 ksi calculated by the general shear stress formula
and could be considered unsafe.
Allowable shear stresses are set intentially low to account for the fact the computed
average shear stress will be lower than the actual shear stress.
In rolled beams, at locations other than end connections, the ASD Specifications, Section F4,
establishes the allowable shear stress Fv as follows. For
380
where
The allowable shear stress is based on the overall depth of the beam d times the web
thickness and is taken as
= 0.40
(F4 1)
380
limit when Fy =
36 ksi. However, for the higher-yield-strength steels, this is frequently not the case.
For
>
380
, the allowable shear stress is based on the clear distance between flanges times
( ) 0.40
2.89
(F4 2)
where Cv is a function of the distance between transverse stiffners and is computed from
or
45,000
when < 0.8
( )2
190
= 4.00 +
= 5.34 +
where
5.34
< 1.0
2
()
4.00
> 1.0
2
()
Example 13:
The W18 50 beam shown in Fig. 4.15, has been designed for moment. The uniform load
includes the beam weight. Check the beam for shear. Assume A36 steel and full lateral
support.
20 k
4'
50 k
3'
60 k
3.07 kips/ft
3'
4'
78.8
69.6
19.6
V
(kips) 20
7.3
32.3
52.7
61.9
= 0.355 in.
100
= 0.570 in.
78.8
=
= 12.34 ksi
17.99(0.355)
Next, calculating the allowable shear stress (ASD Specifications, Section F4):
17.99 2(0.570)
=
= 47.5
0.355
380 380
=
= 63.3
36
47.5 < 63.3
Therefore,
Which is rounded to 14.5 ksi in the ASD Specifications, Numerical Values. Thus
OK for shear.
<
The concept of shear capacity is also useful. The shear capacity may be determined by
multiplying its web area ( ) by the allowable shear stress Fv. The ASD Manual calls this
the maximum web shear and designates it V.
For the W18 50 of previous example,
This may be readily computed with the maximum applied shear to verify that the beam is
satisfactory. The maximum permissible web shear (V) is a tabulated quantity. Refer to the
Allowable Uniform Load Tables in the ASD Manual, Part 2, where V for each section is
tabulated for Fy = 36 ksi and 50 ksi.
Deflection
When a beam is subjected to a load that creates bending, the beam must sag or deflect, as
shown in Fig. 4.16. Although a beam is safe for moment and shear, it may be
unsatisfactorily because it is too flexible. Therefore, the consideration of the deflection of
beams is another part of the beam design process.
101
Deflection
To counteract the sag in a beam, and upward bend or camber may be given to the beam.
This is commonly done for longer beams to cancel out the dead load deflection and,
sometimes, part of the live load deflection. One production method involves cold bending of
the beam by applying a point load with a hydraulic press or ram. For shorter beams, which
are not intentionally cambered, the fabricator will process the beam so that any neutral
sweep within accepted tolerance will be placed so as to counteract expected deflection.
Deflection criteria are based on some maximum limit to which the deflection of the beam
must be held. This is generally in terms of some fraction of the span length. For the
designer this invloves a calculation of the expected deflection for the beam in question, a
determination of the appropraite limit of deflection, and a comparison of the two.
Various methods are available to calculate the deflection. For common beams and loadings,
the ASD Manual, Part 2, Beam Diagrams and Formulas, contains deflection formulas. The
use of some of these will be illustrated in subesquent examples.
The deflection limitations of specifications and codes are usually in the form of suggested
guidelines because the strength adequacy of the beam is not at stake. Traditionaly, beams
that have supported plastered cielings have been limited to maiximum live load
deflections of span/360. This is a requirement of the ASD Specifications, section L3.1. The
span/360 deflection limit is often used for live load deflections in other situations. It is
common practice, and in accordance with some codes, to limit maximum total deflection
(due to live load and dead load) to span/240 for roofs and floors that support other than
plastered cielings.
The ASD Specifications Commentary, Section L3.1, contains guidelines of another nature. It
suggests:
102
1. The depth of fully stressed beams and girders in floors should, if practicable, be not
less than Fy/800 times the span.
2. The depth of fully stressed roof purlins should, if practicable, be not less than
Fy/1000 times the span, except in the case of flat roofs.
Further, it recommends that where human comfort is the criterion for limiting motion, as in
the case of vibrations, the depth of a steel beam supporting large, open floor areas free of
20
of the span.
Since the moment of inertia increases with the square of the depth, the guidelines for
minimum beam depth limit deflections in a general way. The ASD Specifications
Commentary, Section K2, also contains a method for checking the flexibilty of roof systems
when ponding, the retention of water on flat roofs, is a consideration.
Example 14:
Select the lightest W shape for the beam shown in Fig. 4.17. Assume full lateral support and
A572 grade B steel. Consider moment, shear, and deflection. Maximum allowable deflection
for total load is to be span/360.
10 k
10 k
2.0 kips/ft
8'-0"
16'-0"
8'-0"
From the beam curves, assuming that Fy = 50 ksi, select a W21 62 that has an MR of 349 ftkips. From the ASD Manual, Part 1, Tables 1 and 2, Fy = 50 ksi. Check the additional moment
due to the beams own weight:
103
0.062(32)2
=
= 7.94 ft. kips
8
Check the shear: For this beam, the maximum shear occurs at, and is equal to, the reaction.
Therefore,
maximum shear =
2(10) + 32(2.062)
= 43.0 kips
2
From the Allowable Uniform Load Tables, Part 2, maximum permissible web shear for the
W21 62 is 168 kips. Therefore, the shear strength is ok.
Check deflection (): From the ASD Manual properties tables, I for the W21 62 is 1330
in4:
maximum allowable =
span 32(12)
=
= 1.07 in.
360
360
From formula in the ASD Manual, Part 2, Beam Diagrams and Formulas, the actual expected
deflection may be calculated. Note that the units are kips and inches.
54 (32 42 )
=
+
384
24
10(8)(12)3
5(2.062)(32)4 (12)3
[3(32)2 4(8)2 ]
+
=
384(29,000)(1330) 24(29,000)(1330)
= 1.26 + 0.42 = 1.68 in. > 1.07 in.
The moment of inertia (I) must be increased. Select a larger beam with an I value of
1.68
(1330) = 2088 in.4
1.07
From the ASD Manual, Part 2, Moment of Inertia Selection Tables, select W24 76 (I = 2100
in.4). This shape has a higher Sx and greater shear capacity than the W21 62. Therefore,
moment and shear are satisfactory.
Use .
104
Holes in Beams
Beams are normally found as elements of a total structural system rather than as indivdual,
isolated entities. They are peretrated by mechanical and electrical systems, are envloped by
nonstructural elements, and must be connected to other structural members. They must
sometimes be cut to provide clear areas. The problem of holes in beams is a common one.
Among the more evident of holes in beams (or, generally, any decrease in cross-sectional
area) is capacity reduction. Two such reductions may be readily identified. Holes in beam
webs reduce the shear capacity. Holes in beam flanges reduce the moment capacity.
The ASD Specifications is not specific concerning a recommended design procedure for
beams with web holes. The common procedure of reducing shear capacity in direct
proportion to web area reduction is an oversimplification for other than small holes. Where
larger openings occur, it is common practice to reinforce the beam web by welding
stiffening members around the perimeter of the hole.
Some general rules may be atated with regard to web holes. They should be located away
from areas of higher shear. For uniformly loaded beams, web holes near the center of the
span will not be critical. Holes should be centered on the nuetral axis to avoid high bending
stresses. The holes should be round or have rounded corners (to rectangular holes) to
avoid stress concentations. The cutting of web holes in the field should not be allowed
without the approval of the designer.
With respect to holes in the flanges of beams, it is the moment capacity that is affected. The
cross-sectional property that giverns moment capacity is moment of inertia I. The web of a
wide-flange beam contributes very little to the moment of inertia, and the effect of web
holes on moment capacity may be neglected. The effect of flange holes, however, is to
reduce the moment of inertia. The calculation of the reduction is accomplished by
subtracting from the gross moment of inertia the quantity Ad 2 for each hole, where
A = cross-sectional area of the hole (diameter flange thickness) (in.2)
d = distance from the nuetral axis to the centorid of the hole (in.)
The neutral axis shifts very little where holes exist is only one flange and may be assumed
to remain at the centroid of the gross cross section.
It is the conservative practice of some designers to consider both flanges to have holes (in
symmetrical pattern) even though only one does.
105
It is generally agreed that flange holes for bolts do not reduce the moment capacity of
beams to the extent indicated by the erduced moment of inertia as described before. ASD
Specifications, in Section B10, states that no reduction in moment of inertia shall be made
for bolt holes in either flange providedthat
= net flange area
0.5 0.6
(B10 1)
(B10 2)
Then the flexural properties of the member shall be based on an effective tension flange
area Afe, where
Example 15:
(B10 3)
Using the ASD Specifications, determine the resisting moment MR for a W18 71 shown in
Fig. 4.18, that has two holes punched in each flange for 1-in.-diameter bolts. Fb = 24 ksi.
Assume A36 steel. (Fy = 36 ksi, Fu = 58 ksi).
1 81 " hole
(for design)
8.83 "
17.47 "
9.24 "
0.81 "
7.635 "
Ix = 1170 in4
Sx = 127 in3
First check whether the moment of inertia (I) must be reduced on the ASD Specifications,
Section B10. For one flange,
= 7.635(0.81) = 6.18 in.2
=
Since
The reduction for holes must be considered, and the effective flange area is
=
5
5 58
= ( ) (4.36) = 5.85 in.2
6
6 36
from which
net =
net 1119
=
= 121.1 in.3
9.24
reduced = =
24(121.1)
= 242 ft. kips
12
calculating the percent reduction in the resisting moment, noting that MR for the gross
section (from the ASD Manual beam curves) is 254 ft-kips, yields
254 242
(100) = 4.7%
254
107
A beam that is subjected to concentrated leads applied normal to the flange and symmetric
to the web must be checked to ensure that a localized failure of the web does not occur.
The ASD Specifications, Section K1, establishes requirements for beam webs under
compression due to concentrated loads. When the stapulated requirements are exceeded,
the webs of the beams should be reinforced or the length of bearing increased.
Two cinditions are considered: web yielding and web crippling. The type of deformation
failure expected is illustrated in Fig. 4.19. Practical and commonly used bearing lengths N
are usually large eniugh to prevent this type of failure from occuring.
Localized deformation
at toe of fillet
Web deformation
Reaction or
concentrated load
(a) Web yielding
The assumption is made that the load spreads out so that the critical area for stress,
which occurs at the toe of the fillet, has a length of (N + 2.5k) or (N + 5k) for end reactions
and interior loads, respectively, and a width of tw. The dimension k, which locates the toe of
the fillet, is tabulated for various shapes in the ASD Manual,Part 1.
The controlling equations for web yielding are as follows:
1. For interior loads (defined as applied at a distance from the end of the member that
is greater than the depth d of the member),
108
0.66
( + 5)
0.66
( + 2.5)
(K1 2)
(K1 3)
Where are R is the applied concentrated load and N is the length of bearing.
N
tw
N+5k
Toe of fillet
N+2.5k
With respect to web crippling, the ASD Specifictaions, Section K1, places limits on the
compressive concentated load.
1. For interior loads (defined as applied at a distance from the end of the member that
is greater than d/2), the limiting load R may be taken as
= 67.5
= 34
1.5
1 + 3
1.5
1 + 3
109
(K1 4)
(K1 5)
For unreinforced webs, both web yielding and web crippling shouls be checked under all
concentrated loads and at points where the beam is supported by walls or pedestals or at
columns when the connection is a seated type. If web stiffeners ae providid and extend at
least one-half the web depth, Equations (K1-4) and (K1-5) need not be checked.
Example 16:
A W24 55 beam of A36 steel has an end reaction of 70 kips and is supported on a plate
such that N = 6 in. Check the beam for web yielding and web crippling.
Solution:
= 0.395 in.
= 0.505 in.
= 36 ksi
= 1
16
in.
= 23.57 in.
Check web yielding [using Equation (K1-3) for end reactions]. The compressive stress at
the toe of the fillet is
70
=
= 19.1 ksi
( + 2.5) 0.395[6 + 2.5(1.31)]
Check web crippling [using Equation (K1-5) for end reactions]. The maximum compressive
force is
= 34
1.5
1 + 3
= 34 (0.395)2 1 + 3
6
0.395 1.5
0.505
36
= 55 kips
23.57 0.505
0.395
110
Since 55 kips 70 kips, the beam is inadequate with respect to web crippling. Bearing
stiffeners must be provided, or the length of bearing must be increased.
The web yielding equations [ASD Specifications, Equations (K1-2) and (K1-3)] may be
expressed in different forms. To determine allowable load (based on allowable web
yielding stress):
For end reactions,
= 0.66 ( )( + 2.5)
= 0.66 ( )( + 5)
minimum =
2.5
0.66 ( )
minimum =
5
0.66 ( )
Rather than use the web yielding and web criplling equations as previously described, this
checking process can be significantly simplified by using the ASD Manual, Part 2, Allowable
Uniform Load Tables. The data furnished in the tables are summarized as follows:
1
1. The tabulated value R (kips) is the maximum end reaction for 3 in. of bearing
2
length (N = 3 in). For other values of N, the maximum end reaction with respect to
2
= 1 + 2
where R1 and R2 are costants tabulated for each shape in the Allowable Uniform
Load Tables.
111
2. In a similar manner the maximum end reaction with respwct to web crippling is
computed from
= 3 + 4
where R3 and R4 are also tabulated constants.
Using the tabular values greatly simp;ifies the calculations involving web yielding and web
crippling.
Example 17:
Rework previous example using the tabukated values to determine the maximum end
reaction and compare with the end reaction of 70 kips.
Solution:
From the Alowable Uniform LoadTables, for the W24 55 of A36 steel, for a bearing length
N of 6 in., the maximum end reaction is computed with respect to web yielding and web
crippling.
Web yielding:
Web crippling:
= 1 + 2
= 30.8 + 6(9.39) = 87.1 kips > 70
= 3 + 4
= 36 + 6(3.17) = 55 kips < 70
allowable bending stress Fb. An assumption is made that the pressure developed under the
plate is uniformly distributed.
For a plate covering less than the full area of concrete support,
where
2
= 0.35 0.7
1
2 = maximum area of the portion of the supporting surface that is geometrically similar to
and concentric with the loaded area (in.2).
The moment at the critical section for a 1-in.-wide strip of depth tp (in.), which acts like a
cantilever beam is
= (actual bearing pressure)(area)(moment arm)
113
= ( + 1)
2
=
2
2
2
2
( 2 ) ( 2 ) 3
=
=
=
1( 3 )
2
12
3 2
required =
3 2
= 2
required =
0.75
A procedure for the design of beam bearing plates is given in the ASD Manual, Part 2.
Example 18:
A W16 50 is to be supported on a concrete wall, as shown in Fig. 4.22. = 3000 psi. The
beam reaction is 55 kips. Design a bearing plate for the beam. Assume a 2-in. edge distance
from the edge of the plate to the edge of the wall (maximum N = 6 in.). All steel is A36.
R = 55 kips
N+2.5k
k
k
(N)
6" max.
10" wall
114
Solution:
= 1
16
From the Allowable Uniform Load Tables, for the W16 50,
1 = 29.6 kips
3 = 37.9 kips
2 = 9.03 kips/in.
4 = 3.28 kips/in.
1 55 29.6
=
= 2.81 in.
9.03
2
3 55 37.9
=
= 5.21 in.
4
3.28
2. Since the area of the support and the bearing area (A1 and A2) are unknown,
conservatively assume that
= 0.35 = 0.35(3) = 1.05 ksi
3. The required support area is
required 1 =
55
=
= 52.4 in.2
1.05
required =
1 52.4
=
= 8.73 in.
55
=
= 1.02 ksi < 1.05
6(9)
9.0
=
1.31 = 3.19 in.
2
2
115
1.02
required = 2 = 2(3.19)
= 1.07 in.
36
See the ASD Manual, Part 1, Bars and Plates - Product Availibilty, for information on
plate availablity.
It may be possible, if reactions are small, to support a beam in a bearing situation without
the use of a bearing plate. Bearing pressure, web yielding, web crippling, and flange
bending are the considerations. The critical section for flange bending is again assumed to
be at a distance k from the center of the section.
Example 19:
R = 25 kips
8"
n
12"
= 36 ksi
= 8.99 in.
= 0.68 in.
= 1
16
= 1.44 in.
25
= 0.35 ksi
8.99(8)
116
2. Check the maximum end reaction using data from the Allowable Uniform Load
Tables. Based on web yielding,
= 1 + 2
= 37.6 + 8(10.5) = 121.6 kips > 25
3. The bending stress in the flange may be determined by using the formula developed
previously for fb in the bearing plate (the flange acts exactly as the plate does):
8.99
= =
1.44 = 3.06 in.
2
2
=
3 2
2
3(0.35)(3.06)2
=
= 21.26 ksi
(0.68)2
Therefore, this beam may be used on a bearing length of 8 in. without a bearing plate.
117
Start
Known:
Shape, type of span,
Fy, Fu, Lb, Lc, Lu, Sx, Fy
Yes
(Compact)
No
(Not Compact)
See ASDS No
Appendix B
65
Yes
= 0.66
95
Yes
(Non-Compact)
No
Yes
Yes
No
Fb from ASD
Specifications Eq.
(F1-3)
= 0.66
End
118
Start
Estimate the
beam weight
Select trial
section so that Yes
M
allowable M
Assume Fb.
No Use Fb =
0.66Fy for
initial trial
Use ASDM
beam
curves?
Determine
Lc, Lu, Fy
No
Reselect
trial
section
based on
reqd
shear
capacity
and reqd
moment
capacity
Estimated
beam
weight
ok?
Yes
No
1
Reselect
trial
section
based on
reqd I,
reqd
shear
capacity
and reqd
moment
capacity
1
V Shear
capacity?
Yes
= 0.66
Yes
= 0.60
Deflection
ok?
Yes
Compact
No
Select trial
section
based on
latest reqd
Sx
No
Noncompact
95
65
No
Yes
No
Calculate
req' d
Slender
element
No section.
See ASDS
Appendix
B
Yes
Yes
Specify
section to
use
Calculate req' d =
> reqd
Yes
End
119
No
5. BEAM COLUMNS
A structural member that is subjected to varying amounts of both axial compression and
bending moment is commonly termed a beam-column.
Eccentric
load
(a)
Symmetrical
dead load
(b)
= distance from the neutral axis to the extreme outside of the cross section.
120
= moment of inertial of the cross section about the bendong nuetral axis.
If bending occurs with respect to both axes, the expression becomes
=
= + +
With the negative signs neglected. Dividing both sieds by fmax, yields
1=
+
+
With this arrangement, if any two of the computed stresses become zero, the correct
allowable stress is approached either as an axially loaded column or as a beam subjected to
bending about either axis.
Analysis of Beam-Columns
The previous expression developed is the basis for ASD Specifications Equation (H1-3):
where
+
+
1.0
(H1-3)
It applies to members subjected to both axial compression and bending stresses when
/ is less than or equal to 0.15.
121
When / > 0.15, the secondary moment due to the member deflection may be of a
significant magnitude. This secondary moment results from a lateral deflection initially
caused by the bending moment, as shown in Fig. 5.2. The product of this deflection and
axial load ( , sometimes called P-delta moment) causes further bending and creates
secondary stresses that normally are not considered in individual beam or column analysis
and design.
The effect of the secondary moment may be approximated by multiplying fbx and fby by an
amplification factor
1
(C-H1-1)
1 ( )
12 2
23
12
2
23
In this expression, K is the effective length factor in the plane of bending, lb is the unbraced
length in the plane of bending, and rb is the corresponding raduis of gyration.
P
section of the ASD Specifications or through the use of the properties tabulated at the
bottom of the column load tables in the ASD Manual, Part3.
122
Under some combinations of loading, it was found that this amplification factor
overetimated the effect of secondary moment. To compensate for this condition, the
ampification factor was modofied by a reduction factor Cm.
With the introduction of the two factors, ASD Specifications Equation (H1-3) was modified
for the case when / > 0.15 and expressed as follows:
+
+
1.0
1
1
In which
1
2
= 0.6 0.4
(H1-1)
is the ratio of the smaller end moment to the larger end moment.
1
2
is taken as
positive if the moments tend to cause reverse curavature and negative if they tend to cause
single curvature. Examples of Cm values are shown in Fig. 5.3.
P
M2
M2
M2
M1
M1
P
M1 = M2
M1 = 0
M1 = 21 M2
Cm = 1.0
Cm = 0.6
Cm = 0.4
(a)
(b)
(c)
column ends move out of align,ent, causing an additional secondary moment from the axial
load.
+
+
1.0
0.6
(H1-2)
If only one acis of bendign is involved in a problem, one of the terms will equal zero with
the remaining formula still applicable.
In determining Fb for the interaction equations, the compactness of the beam-column must
be established. As be discussed before, the web compactness of a beam is based on fa = 0.
This applies to a beam subjected to bending only, with no axial load. With a baem-column
(fa is not zero), the web compactness must be checked by using equations from ASD
Specifications, table B5.1, for Webs in combined flexural and axial compression. This is
124
simplified in the ASD Manual through the use of , which is defined as the theoritical
maximum yield stress (ksi) based on the depth-thickness ratio of the web below which a
particular shape may be considered compact for any condition of combined bending and
axial stresses.
is tabulated in the properties table of part 1. It is determined in the same way that is
determined, as discussed before. If > , the member is not comact based on the web
criteron, and Fb cannot exceed 0.60Fy.
Example 1:
An A36 steel W625 column is subjected to an eccentric load of 32 kips, as shown in Fig
5.5. The column has an unbraced length of 15 ft and may be assumed to have pinned ends.
Bracing prevents sidesway. Determine whether the column is adequate.
C
Lcolumn
P = 32 k
W 16 beam
W6 X 25
Seat angle
e = 5 21"
Fig. 5.5
Solution:
For the W625
= 7.34 in.2
= 16.7 in.3
= 1.52 in.
= 2.70 in.
= 1 (pinned ends)
=
125
1. Replacing the eccentric load with a concentric load and a couple (moment),
= 32 kips
32
=
= 4.36 ksi
7.34
1(15)(12)
=
= 118.4
1.52
Rounding to 118, from the ASD Manual, Part 3, Table C-36, obtain
= 10.57 ksi
176
=
= 10.5 ksi
16.7
5. Fbx is a function of the actal unbraced length. Determine whether the member is
adequately or inadequately braced. The actual unbraced length Lb = 15 ft. Lc and Lu
may be obtained from the column load tables in the ASD Manual, Part 3:
= 6.4 ft
= 20.0 ft
Therefore, < < . The W625 is compact, since > , and the allowable
bending stress is
= 0.60 = 21.6 ksi
126
1(15)(12)
=
= 66.7
2.70
Rounding to 67, from Table 8 in the Numerical Values section of the ASD
Specifications obtain
= 33.27 ksi
+
1.0
(1 )
4.36
0.6(10.5)
+
1.0
10.57 (1 4.36/33.27)(21.6)
0.41 + 0.34 = 0.75 < 1.0
+ 1.0
0.6
4.36 10.5
+
1.0
21.3 21.6
Example 2:
An A572 (Fy = 50 ksi) W12136 column supports beams framing into it, as shown in Fig.
5.6. The connections are moment connections. The column supports an axial load of 600
kips, which includes the beam reactions at its top. Due to unbalanced floor loading,
moments of 80 ft-kips rach are applied in opposite directions at the top and bottom of
columns as shown. Sidesway is prevented by a bracing system. Ky = 1.0 and Kx is estimated
to be 0.9. Determine whether the member is adequate.
Solution:
= 39.9 in.2
127
= 186 in.3
= 3.16 in.
= 5.58 in.
= 0.9
= 1.0
=
P = 600 kips
M = 80 ft-kips
14'-0"
M = 80 ft-kips
P = 600 kips
Fig. 5.6
1. P = 600 kips, M = 80 ft-kips.
2. = = 60039.9 = 15.0 ksi.
3. Fa is a function of the largest slenderness ratio:
1(14)(12)
=
= 53.2
3.16
0.9(14)(12)
=
= 27.1
5.58
Rounding to 53, from the ASD Manual, Part 3, Table C-50,
= 23.88 ksi
4. = = 80(12)186 = 5.16 ksi.
128
5. is a function of , and .
= 14 ft
= 11.1 ft
= 38.3 ft
Since < < , and since W12136 is compact > , the allowable
bending stress is
= 0.60 = 30 ksi
2
= 0.6 0.4(1)
= 1.0
= 0.6 0.4
is a function of Klb/rb, which in this case Kxl/rx. Note that this is with respect to
the x-x axis. Using the properties tabulated in the ASD Manual, Part 3, the column
load table for the W12136, gives
( )2
From which
102
= 323
323(10)2
= 203 ksi
[0.9(14)]2
Note that may also be obtained using Kxl/rx and Table 8 of the Numerical Values
section of the ASD Specifications.
+
1.0
(1 )
15.0
1.0(5.16)
+
1.0
23.88 (1 15.0/203)(30.0)
0.63 + 0.19 = 0.82 < 1.0
+ 1.0
0.6
15.0 5.16
+
1.0
30.0 30.0
Design of Beam-Columns
The use of the interaction formulas furnishes a convenient means of beam-column analysis.
These may also be used for beam-column design. A trial section must first be selected,
however. After the selection is made, the problem becomes one of analysis. In essence, the
design process is one of trial and error, since no simple design procedure exists whereby a
most economical member can be selected in one quick step.
The ASD Manual furnishes a method of design whereby a trial section may be attained
using an equivelant axial load in conjunction with the ASD Manual axial load table of Part 3.
Using the ASD Manual approach to determine a trial section, it is referred to the ASD
Manual, Part 3, Table B. The equivalent axial load, for design purposes, is designated Peff:
where
= 0 + +
= factor taken from the ASD Manual, Part 3, column load tables
The procedure for selection of a trial sections is as follows:
1. With the known value of KL (in feet), select a value of m from the first
approximation section of Table B and assume that U = 3.
2. Solve for Peff.
3. From the column load table in Part 3 of the ASD Manual, select a trial section to
support Peff.
130
By using this trial section, the beam-column may then be analyzed in the manner discussed
previously using the ASD manula interaction equations.
Example 3:
Using A36 steel and the ASD Specifications, select a wide-flange column for the
conditionsshown in Fig. 5.7. The column has pinned ends and sidesway is prevented.
Bending occurs with respect to the strong (x-x) axis.
P = 85 kips
M = 20 ft-kips
13'-0"
M = 20 ft-kips
P = 85 kips
Fig. 5.7
Solution:
From the ASD Manual, Part 3, Table B, with KL = 13 ft, select a value of m = 2.25 from the
first approximation portion. Since My = 0, the expression for the effective axial load
becomes
eff = 0 +
= 85 + 20(2.25)
= 130 kips
131
From the column load table of the ASD Manual, Part 3, select a W831 (allowable load Pa is
143 kips). From Table B again, select a value of m = 2.85 from the Subsequent
Approximations portion of the table.
eff = 85 + 20(2.85)
= 142 kips
Since 142 kips 143 kips, the W831 remains as the trial section should be checked using
interaction equations. The checking procedure is identical to that in Examples 1 and 2. ASD
Specifications Equation (H1-1) results in a value of 0.77, and Equation (H1-2) results in a
value of 0.82. The W831 is therefore satisfactory.
Example 4:
Using A36 steel and the ASD Specifications, select a wide-flange column for the conditions
shown in Fig. 5.8. Architectural requirements indicate the use of a W8, if possible. The
column is pinned at both ends. Bending occurs with respect to both axes. Sidesway is
prevented in both directions.
Solution:
Replace the eccentric loads with concentric loads and couples (moments). A 4-in.
eccentricity is assumed for strong-axis bending:
= 58 kips
From the ASD Manual, Part 3, Table B, with KL = 16 ft, select a value of m = 2.2 from the
first approximation portion. Let U = 3; tehrefore,
eff = 0 + +
120
84
(2.2) + (3)(2.2)
= 58 +
12
12
= 58 + 22 + 46.2 = 130 kips
From the ASD Manual, Part 3, column load table, select a W835 (allowab;e load Pa is 141
kips). From Table B again, select a value of m = 2.6 from the subsequent approximation
portion and a value of U = 2.59 from the column load table.
eff = 58 +
120
84
(2.6) + (2.6)(2.59)
12
12
132
= 58 + 26 + 47 = 131 kips
Since 131 kips 141 kips, the W835 remains as the trial section should be checked using
interaction equations.
= 10.3 in.2
= 31.2 in.3
= 2.03 in.
= 3.51 in.
= 1.0
= 10.6 in.3
=
133
= 8.5 ft
= 22.6 ft
Since < < , and since W835 is compact > , the allowable bending
stress is
= 0.60 = 21.6 ksi
( )2
From which
102
128(10)2
162
134
= 128
= 42.7
= 50.0 ksi
42.7(10)2
162
= 16.7 ksi
+
+
1.0
(1 ) (1 )
0.60(3.85)
0.60(7.92)
5.63
+
+
1.0
13.60 (1 5.63/50.0)(21.6) (1 5.63/16.7)(27.0)
0.41 + 0.12 + 0.27 = 0.80 < 1.0
+ +
1.0
0.6
5.63 3.85 7.92
+
+
1.0
21.6 21.6 27.0
The effective length of the member depends on the restraints against relative rotation and
lateral movement (sidesway) imposed at the ends of the member. The portal frame shown
in Fig. 5.9 is having rigidly connected beam to the supporting columns, and no lateral
movement allowed. The effective length factor K for the columns can have values range
from 0.5 for ends fixed against rotation to 1.0 for pinned ends.
When the frame depends on its own stiffness for resistance to sidesway, as shown in Fig.
5.10, K will have a value larger than 1.0. Fig. 6.10a shows the deformed shape due to
vertical load. The frame will deflect to the side (sidesway) so as to equalize the moments at
the tops of the columns. Sidesway may also be caused by a laterally applied force, as shown
in Fig. 5.10b. As a rule, columns free to translate in a sidesway mode are appreciably
weaker than columns of equal length braced against sidesway. Also of importance is that
the magnitude of the sidesway of a column is directly affected by the stiffness of the other
members in the frame; or, the magnitude of joint rotation is directly affected by the
stiffness of the members framing into the joint.
To simplify the determination of the effective length factor K, alignment charts are
furnished in the ASD Manual, Part 3 (Figure 1), for the two cases of sidesway prevented
and sidesway not prevented.
135
(a)
(b)
Fig. 5.10 Loaded unbraced frame.
The use of the charts requires an evaluation of the relative stiffness of the members of the
frame at each of the column. The stiffness ratio or relative stiffness of the members rigidly
connected at each joint may be expressed as
and
( )
( )
136
where A and B subscripts refer to the joint at which the erlative stiffness is being
determined; the c and g subscripts refer to the column and girder or beam, respectivly; I is
the moment of inertia, and l is the unsupported length of member. The I / l terms are taken
with respect to an axis normal to the plane of buckling under consideration. Having
determined GA and GB, the appropraite chart may be used to determine K.
For column ends supported by, but not rigidly connected to, a footing or foundation, GB is
theoritically infinity, but unless the joint is designed as a true friction-free pin, GB may be
taken as 10 for practical designs. If the column end is rigidly attached to a properly
designed footing, GB may be taken as 1.0. Fig. 5.11 shows how the G values would be
determined for a given column AB.
Ic1
A
Ig1
(1 1 ) + (2 2 )
1 1 + 2 2
= 1.0
Ig2
Ic2
Rigidly
attached
Compute the effective length factor K for each of the columns in the frame shown in Fig.
5.12 using ASD Manual alignment charts. Preliminary sizes of each member are furnished.
Sidesway is not prevented. Webs of the wide-flange shapes are in the plane of the frame.
137
Solution:
Shape
W1453
W1453
W1850
W2162
W1453
W1453
I/l
3.22
3.76
2.22
3.69
3.76
3.22
W18X50
W21X62
W14X53
W14X53
14'-0"
l (in.)
168
144
360
360
144
168
W14X53
12'-0"
I (in.4)
541
541
800
1330
541
541
W14X53
Member
AB
BC
CD
BE
DE
EF
30'-0"
( )
( )
Pinned end
10.0
3.76
2.22
1.69
3.76 + 3.22
3.69
3.76
2.22
=
138
1.89
1.69
E
F
3.76 + 3.22
3.69
Fixed end
1.89
1.0
K
2.08
1.54
1.54
1.44
Because of the smaller effective length factors used for frames where sidesway is
prevented, it is advisable to provide lateral support wherever possible. This may be
accomplished with diagonal bracing, shear walls, or attachment to an adjacent structure
having adequate lateral stability or by floor slabs or roof decks secured horizontally by
walls or bracing systems.
The determination of the K factors utilizing the alignment charts based on several
assumptions. Two of the principal assumptions are that all columns in a story buckle
simultaneously and that all columns behavior is purely elastic. Either or both of these
conditions may not exist in an actual structure, and as a result, the use of the alignment
charts will produce overly conservative designs.
The ASD Manual, Part 3, contains a design procedure to reduce the K factor value by
multiplying the elastic G value by a stiffness reduction factor. Then, using the alignment
charts as discussed previously, an inelastic K factor is obtained. The stiffness reduction
factor is obtained from Table A in Part 3 of the ASD Manual.
Example 6:
Compute the inelastic K factors for columns BC and DC of previous example. Consider
behavior in the plane of the frame only. Use Figure 1 and Table A of the ASD Manual, Part ,
for the inelastic K factor procedure. Preliminary sizes are shown in Fig. 5.13. Sidesway in
the plane of the frame is not prevented. Webs of the W-shape members are in the plane of
the frame. Use A36 steel. Assume that the columns support loads of 245 kips as shown in
Fig. 5.13.
139
W21X62
W14X53
C
W14X53
12'-0"
245 kips
30'-0"
245
= 15.7 ksi
15.6
inelastic
(top )
= 0.621(1.69) = 1.05
(bottom )
= 0.621(1.89) = 1.17
140