Foundation Analysis and Design: Michael Valley, P.E

4-1
4
FOUNDATION ANALYSIS AND DESIGN
Michael Valley, P.E.
This chapter illustrates application of the 2000 Edition of the NEHRP Recommended Provisions to the
design of foundation elements. Example 4.1 completes the analysis and design of shallow foundations for
two of the alternate framing arrangements considered for the building featured in Example 5.2. Example
4.2 illustrates the analysis and design of deep foundations for a building similar to the one highlighted in
Chapter 6 of this volume of design examples. In both cases, only those portions of the designs necessary
to illustrate specific points are included.
The force-displacement response of soil to loading is highly nonlinear and strongly time dependent.
Control of settlement is generally the most important aspect of soil response to gravity loads. However,
the strength of the soil may control foundation design where large amplitude transient loads, such as those
occurring during an earthquake, are anticipated.
Foundation elements are most commonly constructed of reinforced concrete. As compared to design of
concrete elements that form the superstructure of a building, additional consideration must be given to
concrete foundation elements due to permanent exposure to potentially deleterious materials, less precise
construction tolerances, and even the possibility of unintentional mixing with soil.
Although the application of advanced analysis techniques to foundation design is becoming increasingly
common (and is illustrated in this chapter), analysis should not be the primary focus of foundation design.
Good foundation design for seismic resistance requires familiarity with basic soil behavior and common
geotechnical parameters, the ability to proportion concrete elements correctly, an understanding of how
such elements should be detailed to produce ductile response, and careful attention to practical
considerations of construction.
Although this chapter is based on the 2000 Provisions, it has been annotated to reflect changes made to
the 2003 Provisions. Annotations within brackets, [ ], indicate both organizational changes (as a result of
a reformat of all of the chapters of the 2003 Provisions) and substantive technical changes to the 2003
Provisions and its primary reference documents. While the general concepts of the changes are
described, the design examples and calculations have not been revised to reflect the changes to the 2003
Provisions. The most significant change to the foundation chapter in the 2003 Provisions is the addition
of a strength design method for foundations. Another change was made to introduce guidance for the
explicit modeling of foundation load-deformation characteristics. Where they affect the design examples
in this chapter, other significant changes to the 2003 Provisions and primary reference documents are
noted. However, some minor changes to the 2003 Provisions and the reference documents may not be
noted.
FEMA 451, NEHRP Recommended Provisions: Design Examples
4-2
In addition to the 2000 NEHRP Recommended Provisions and Commentary (referred to herein as
Provisions and Commentary), the following documents are either referenced directly or provide useful
information for the analysis and design of foundations for seismic resistance:
ACI 318 American Concrete Institute. 1999 [2002]. Building Code Requirements and
Commentary for Structural Concrete.
ASCE 7 American Society of Civil Engineers. 1998 [2002]. Minimum Design Loads for
Buildings and Other Structures.
Bowles Bowles, J . E. 1988. Foundation Analysis and Design. McGraw-Hill.
Brown 1987 Brown, D. A., L. C. Reese, and M. W. ONeill. 1987. Cyclic Lateral Loading
of a Large-Scale Pile Group, Journal of Geotechnical Engineering, Vol. 113,
No. 11 (November). ASCE.
Brown 1988 Brown, D. A., C. Morrison, and L. C. Reese. 1988. Lateral Load Behavior of
Pile Group in Sand. Journal of Geotechnical Engineering, Vol 114, No. 11,
(November). ASCE.
CRSI Concrete Reinforcing Steel Institute. 1996. CRSI Design Handbook. Concrete
Reinforcing Steel Institute.
FEMA 356 ASCE. 2000. Prestandard and Commentary for the Seismic Rehabilitation of
Buildings, FEMA 356, prepared by the American Society of Civil Engineers for
the Federal Emergency Management Agency.
GROUP Reese, L. C., and S. T. Wang. 1996. Manual for GROUP 4.0 for Windows.
Ensoft.
Kramer Kramer, S. L. 1996. Geotechnical Earthquake Engineering. Prentice Hall.
LPILE Reese, L. C., and S. T. Wang. 1997. Technical Manual for LPILE Plus 3.0 for
Windows. Ensoft.
Martin Martin, G. R., and I. PoLam. 1995. Seismic Design of Pile Foundations:
Structural and Geotechnical Issues. Proceedings: Third International
Conference on Recent Advances in Geotechnical Earthquake Engineering and
Soil Dynamics.
Pender Pender, M. J . 1993. Aseismic Pile Foundation Design Analysis. Bulletin of
the New Zealand National Society for Earthquake Engineering, Vol. 26, No. 1
(March).
PoLam PoLam, I., M. Kapuskar, and D. Chaudhuri. 1998. Modeling of Pile Footings
and Drilled Shafts for Seismic Design, MCEER-98-0018. Multidisciplinary
Center for Earthquake Engineering Research.
Wang & Salmon Wang, C.-K., and C. G. Salmon. 1992. Reinforced Concrete Design .
HarperCollins.
Chapter 4, Foundation Analysis and Design
4-3
Youd Youd, T. L., Idriss, I. M., and et al. 2001. Liquefaction Resistance of Soils:
Summary Report from the 1996 NCEER and 1998 NCEER/NSF Workshops on
Evaluation of Liquefaction Resistance of Soils. Journal of Geotechnical and
Geoenvironmental Engineering (October). ASCE.
Several commercially available programs were used to perform the calculations described in this chapter.
RISA: 3D is used to determine the shears and moments in a concrete mat foundation; LPILE, in the
analysis of laterally loaded single piles; and PCACOL, to determine concrete pile section capacities.
4-4
1
2
7
'
-
4
"
2
5
'
-
0
"
2
5
'
-
0
"
2
5
'
-
0
"
2
5
'
-
0
"
2
5
'
-
0
"
1
'
-
2
"
1
'
-
2
"
1'-2" 25'-0" 25'-0" 25'-0" 25'-0" 25'-0" 25'-0" 25'-0" 1'-2"
177'-4"
N
Figure 4.1-1 Typical framing plan.
4.1 SHALLOW FOUNDATIONS FOR A SEVEN-STORY OFFICE BUILDING, LOS
ANGELES, CALIFORNIA
This example features the analysis and design of shallow foundations for two of the three framing
arrangements for the seven-story steel office building described in Sec. 5.2 of this volume of design
examples. Refer to that example for more detailed building information and for the design of the
superstructure; because Chapter 4 was completed after Chapter 5, some values may differ slightly
between the two chapters.
4.1.1 Basic Information
4.1.1.1 Description
The framing plan in Figure 4.1-1 shows the gravity-load-resisting system for a representative level of the
building. The site soils, consisting of medium dense sands, are suitable for shallow foundations. Table
4.1-1 shows the design parameters provided by a geotechnical consultant. Note the distinction made
between bearing pressure and bearing capacity. If the long-term, service-level loads applied to
foundations do not exceed the noted bearing pressure, differential and total settlements are expected to be
within acceptable limits. Settlements are more pronounced where large areas are loaded, so the bearing
pressure limits are a function of the size of the loaded area. The values identified as bearing capacity are
related to gross failure of the soil mass in the vicinity of loading. Where loads are applied over smaller
areas, punching into the soil is more likely.
4-5
Because bearing capacities are generally expressed as a function of the minimum dimension of the loaded
area and are applied as limits on the maximum pressure, foundations with significantly non-square loaded
areas (tending toward strip footings) and those with significant differences between average pressure and
maximum pressure (as for eccentrically loaded footings) have higher calculated bearing capacities. The
recommended values are consistent with these expectations.
[The 2003 Provisions discuss the settlement and strength limit states in Sec. 7.2.2.2 using slightly
different nomenclature.]
Table 4.1-1 Geotechnical Parameters
Parameter Value
Basic soil properties Medium dense sand
(SPT) N =20
=125 pcf
angle of internal friction =33 deg
Net bearing pressure
(to control settlement due to
sustained loads)
# 4000 psf for B # 20 ft
# 2000 psf for B $ 40 ft
(may interpolate for intermediate dimensions)
Bearing capacity
(for plastic equilibrium strength
checks with factored loads)
2000 B psf for concentrically loaded square footings
3000 B' psf for eccentrically loaded footings
where B and B' are in feet, B is the footing width and B'
is an average width for the compressed area.
Resistance factor, =0.6
[In the 2003 Provisions, the factor for cohesionless soil
is explicitly defined; the value is set at 0.7 for vertical,
lateral, and rocking resistance.]
Lateral properties Earth pressure coefficients
Active, K
A
=0.3
At-rest, K
0
=0.46
Passive, K
P
=3.3
Ultimate friction coefficient at base of footing =0.65
Resistance factor, =0.8
[In the 2003 Provisions, the factor for cohesionless soil
is explicitly defined; the value is set at 0.7 for vertical,
lateral, and rocking resistance.]
The structural material properties assumed for this example are:
f'
c
=4,000 psi
f
y
=60,000 psi
4.1.1.2 Provisions Parameters
The complete set of parameters used in applying the Provisions to design of the superstructure is
described in Sec. 5.2.2.1 of this volume of design examples. The following parameters, which are used
during foundation design, are duplicated here.
Site Class =D
S
DS
=1.0
4-6
Seismic Design Category =D
4.1.1.3 Design Approach
4.1.1.3.1 Selecting Footing Size and Reinforcement
Most foundation failures are related to excessive movement rather than loss of load-carrying capacity.
Settlement control should be addressed first. In recognition of this fact, settlement control should be the
first issue addressed. Once service loads have been calculated, foundation plan dimensions should be
selected to limit bearing pressures to those that are expected to provide adequate settlement performance.
Maintaining a reasonably consistent level of service load bearing pressures for all of the individual
footings is encouraged as it will tend to reduce differential settlements, which are usually of more concern
than are total settlements.
When a preliminary footing size that satisfies serviceability criteria has been selected, bearing capacity
can be checked. It would be rare for bearing capacity to govern the size of footings subjected to sustained
loads. However, where large transient loads are anticipated, consideration of bearing capacity may
become important.
The thickness of footings is selected for ease of construction and to provide adequate shear capacity for
the concrete section. The common design approach is to increase footing thickness as necessary to avoid
the need for shear reinforcement, which is uncommon in shallow foundations.
Design requirements for concrete footings are found in Chapters 15 and 21 of ACI 318. Chapter 15
provides direction for the calculation of demands and includes detailing requirements. Section capacities
are calculated in accordance with Chapters 10 (for flexure) and 11 (for shear). Figure 4.1-2 illustrates the
critical sections (dashed lines) and areas (hatched) over which loads are tributary to the critical sections.
For elements that are very thick with respect to the plan dimensions (as at the pile caps), these critical
section definitions become less meaningful and other approaches (e.g., strut-and-tie modeling) should be
employed. Chapter 21 provides the minimum requirements for concrete foundations in Seismic Design
Categories D, E, and F, which are similar to those provided in prior editions of the Provisions.
For shallow foundations, reinforcement is designed to satisfy flexural demands. ACI 318 Sec. 15.4
defines how flexural reinforcement is to be distributed for footings of various shapes.
Sec. 10.5 of ACI 318 prescribes the minimum reinforcement for flexural members where tensile
reinforcement is required by analysis. Provision of the minimum reinforcement assures that the strength
of the cracked section is not less than that of the corresponding unreinforced concrete section, thus
preventing sudden, brittle failures. Less reinforcement may be used as long as the area of tensile
reinforcement provided is at least one-third greater than that required by analysis. Sec. 10.5.4 relaxes
the minimum reinforcement requirement for footings of uniform thickness. Such elements need only
satisfy the shrinkage reinforcement requirements of Sec. 7.12. Sec. 10.5.4 also imposes limits on the
maximum spacing of bars.
4.1.1.3.2 Additional Considerations for Eccentric Loads
The design of eccentrically loaded footings follows the approach outlined above with one significant
addition consideration of overturning stability. Stability calculations are sensitive to the
characterization of soil behavior. For sustained eccentric loads a linear distribution of elastic soil stresses
is generally assumed and uplift is usually avoided. If the structure is expected to remain elastic when
subjected to short-term eccentric loads (as for wind loading), uplift over a portion of the footing is
acceptable to most designers. Where foundations will be subjected to short-term loads and inelastic
4-7
d/2
(all sides)
(c)
Critical section
for two-way shear
(b)
Critical section
for one-way shear
(a)
Critical section
for flexure
Outside face of concrete
column or line midway
between face of steel
column and edge of
steel base plate (typical)
extent of footing
(typical)
d
Figure 4.1-2 Critical sections for isolated
footings.
(a)
Loading
(b)
Elastic, no uplift
(c)
Elastic, at uplift
(d)
Elastic, after uplift
(e)
Some plastification
(f)
Plastic limit
M
P
Figure 4.1-3 Soil pressure distributions.
response is acceptable (as for earthquake loading), plastic soil stresses may be considered. It is most
common to consider stability effects on the basis of statically applied loads even where the loading is
actually dynamic; that approach simplifies the calculations at the expense of increased conservatism.
Figure 4.1-3 illustrates the distribution of soil stresses for the various assumptions. Most textbooks on
foundation design provide simple equations to describe the conditions shown in parts b, c, and d of the
figure; finite element models of those conditions are easy to develop. Simple hand calculations can be
performed for the case shown in part f. Practical consideration of the case shown in part e would require
modeling with inelastic elements, but offers no advantage over direct consideration of the plastic limit.
(All of the discussion in this section focuses on the common case in which foundation elements may be
assumed to be rigid with respect to the supporting soil. For the interested reader, Chapter 4 of FEMA 356
provides a useful discussion of foundation compliance, rocking, and other advanced considerations.)
4.1.2 Design for Gravity Loads
4-8
Although most of the examples in the volume do not provide detailed design for gravity loads, it is
provided in this section for two reasons. First, most of the calculation procedures used in designing
shallow foundations for seismic loads are identical to those used for gravity design. Second, a complete
gravity design is needed to make the cost comparisons shown in Sec. 4.1.5 below meaningful.
Detailed calculations are shown for a typical interior footing. The results for all three footing types are
summarized in Sec. 4.1.2.5.
4.1.2.1 Demands
Dead and live load reactions are determined as part of the three-dimensional analysis described in Sec. 5.2
of this volume of design examples. Although there are slight variations in the calculated reactions, the
foundations are lumped into three groups (interior, perimeter, and corner) for gravity load design and the
maximum computed reactions are applied to all members of the group, as follows:
Interior: D =387 kips
L =98 kips
Perimeter: D =206 kips
L =45 kips
Corner: D =104 kips
L =23 kips
The service load combination for consideration of settlement is D +L. Considering the load
combinations for strength design defined in Sec. 2.3.2 of ASCE 7, the controlling gravity load
combination is 1.2D +1.6L. Because ASCE 7 load combinations are employed, the alternate strength
reduction factors found in ACI 318 Appendix C must be used. [The 2003 Provisions refer to ACI 318-
02, in which the basic resistance factors have been revised to be consistent with the load combinations in
ASCE 7. These new resistance factors (not those found in the ACI 318 Appendix) are used for seismic
design. This change would affect slightly the results of the example calculations in this chapter .]
4.1.2.2 Footing Size
The preliminary size of the footing is determined considering settlement. The service load on a typical
interior footing is calculated as:
P =D +L =387 kips +98 kips =485 kips.
Since the footing dimensions will be less than 20 ft, the allowable bearing pressure (see Table 4.1-1) is
4000 psf. Therefore, the required footing area is 487,000 lb/4000 psf =121.25 ft
2
.
Check a footing that is 11'-0" by 11'-0":
P
allow
=11 ft(11 ft)(4000 psf) =484,000 lb =484 kips . 485 kips (demand). OK
The strength demand is:
P
u
=1.2(387 kips) +1.6(98 kips) =621 kips.
As indicated in Table 4.1-1, the bearing capacity (q
c
) is 2000 B =2000 11 =22000 psf =22 ksf.
4-9
The design capacity for the foundation is:
P
n
=q
c
B
2
=0.6(22 ksf)(11 ft)
2
=1597 kips o 621 kips. OK
For use in subsequent calculations, the factored bearing pressure q
u
=621 kips/(11 ft)
2
=5.13 ksf.
4.1.2.3 Footing Thickness
Once the plan dimensions of the footing are selected, the thickness is determined such that the section
satisfies the one-way and two-way shear demands without the addition of shear reinforcement. Because
the demands are calculated at critical sections (see Figure 4.1-2) that depend on the footing thickness,
iteration is required.
Check a footing that is 26 in. thick:
For the W14 columns used in this building, the side dimensions of the loaded area (taken halfway
between the face of the column and the edge of the base plate) are about 16 in. Accounting for cover and
expected bar sizes, d =26 - (3 +1.5(1)) =21.5 in.
One-way shear:
. ( )
16
12
11 21.5
11 5.13 172kips
2 12
u
V

= =

>172 kips. OK ( ) ( )( )( )
1
1000
0.75 2 4000 11 12 21.5 269kips
n c
V V = = =
Two-way shear:
.
( ) ( )
2
16 21.5
12
621 5.13 571kips
u
V
+
= =
>571 kips. OK ( ) ( ) ( )( )
1
1000
0.75 4 4000 4 16 21.5 21.5 612kips
n c
V V = = + =

4.1.2.4 Footing Reinforcement
Footing reinforcement is selected considering both flexural demands and minimum reinforcement
requirements. The following calculations treat flexure first because it usually controls:
. ( ) ( )
2
16
12
11 1
11 5.13 659ft-kips
2 2
u
M

= =

Try 10 #8 bars each way. The distance from the extreme compression fiber to the center of the top layer
of reinforcement, d =t - cover - 1.5d
b
=26 - 3 - 1.5(1) =21.5 in.
T =A
s
f
y
=10(0.79)(60) =474 kips.
Noting that C =T and solving the expression C =0.85 f'
c
b a for a produces a =1.06 in.
>659 ft-kips. OK
( ) ( )( )( )
1.06 1
2 2 12
0.80 474 21.5 663ft-kips
a
n
M T d = = =
4-10
The ratio of reinforcement provided =10(0.79)/[(11)(12)(21.5)] =0.00278. The distance between bars
spaced uniformly across the width of the footing s =[(11)(12)-2(3+0.5)]/(10-1) =13.9 in.
According to ACI 318 Sec. 7.12, the minimum reinforcement ratio =0.0018 <0.00278. OK
and the maximum spacing is the lesser of 3 26 in. or 18 =18 in. >13.9 in. OK
4.1.2.5 Design Results
The calculations performed in Sec. 4.1.2.2 through 4.1.2.4 are repeated for typical perimeter and corner
footings. The footing design for gravity loads is summarized in Table 4.1-2; Figure 4.1-4 depicts the
resulting foundation plan.
Table 4.1-2 Footing Design for Gravity Loads
Location Loads
Footing Size and Reinforcement;
Soil Capacity
Critical Section Demands
and Design Strengths
Interior D =387 kip
L =98 kip
P =485 kip
P
u
=621 kip
11'-0" 11'-0" 2'-2" deep
10-#8 bars each way
P
allow
=484 kip
P
n
=1597 kip
One-way shear: V
u
=172 kip
V
n
=269 kip
Two-way shear: V
u
=571 kip
V
n
=612 kip
Flexure: M
u
=659 ft-kip
M
n
=663 ft-kip
Perimeter D =206 kip
L =45 kip
P =251 kip
P
u
=319 kip
8'-0" 8'-0" 1'-6" deep
10-#6 bars each way
P
allow
=256 kip
P
n
=614 kip
One-way shear: V
u
=88.1 kip
V
n
=123 kip
Two-way shear: V
u
=289 kip
V
n
=302 kip
Flexure: M
u
=222 ft-kip
M
n
=230 ft-kip
Corner D =104 kip
L =23 kip
P =127 kip
P
u
=162 kip
6'-0" 6'-0" 1'-2" deep
7-#5 bars each way
P
allow
=144 kip
P
n
=259 kip
One-way shear: V
u
=41.5 kip
V
n
=64.9 kip
Two-way shear: V
u
=141 kip
V
n
=184 kip
Flexure: M
u
=73.3 ft-kip
M
n
=80.2 ft-kip
[Use of the new resistance factors in ACI 318-02 would change these results.]
4-11
Corner:
6'x6'x1'-2" thick
Perimeter:
8'x8'x1'-6" thick
Interior:
11'x11'x2'-2" thick
Figure 4.1-4 Foundation plan for gravity-load-resisting system.
4.1.3 Design for Moment-Resisting Frame System
Framing Alternate A in Sec. 5.2 of this volume of design examples includes a perimeter moment resisting
frame as the seismic-force-resisting system. A framing plan for the system is shown in Figure 4.1-5.
Detailed calculations are provided in this section for a combined footing at the corner and focus on
overturning and sliding checks for the eccentrically loaded footing; settlement checks and design of
concrete sections would be similar to the calculations shown in Sec. 4.1.2. The results for all footing
types are summarized in Sec. 4.1.3.4.
4-12
5

a
t

2
5
'
-
0
"
7 at 25'-0"
N
Figure 4.1-5 Framing plan for moment resisting frame system.
4.1.3.1 Demands
A three-dimensional analysis of the superstructure, in accordance with the requirements for the equivalent
lateral force (ELF) procedure, is performed using the RAMFRAME program. Foundation reactions at
selected grids are reported in Table 4.1-3.
Table 4.1-3 Demands from Moment-Resisting Frame System
Location Load Rx Ry Rz Mxx Myy
A-5 D -203.8
L -43.8
Ex -13.8 4.6 3.8 53.6 -243.1
Ey 0.5 -85.1 -21.3 -1011.5 8.1
A-6 D -103.5
L -22.3
Ex -14.1 3.7 51.8 47.7 -246.9
Ey 0.8 -68.2 281.0 -891.0 13.4
Note: Units are kips and feet. Load Ex is for loads applied toward the east, including appropriately amplified
counter-clockwise accidental torsion. Load Ey is for loads applied toward the north, including appropriately
amplified clockwise accidental torsion.
Sec. 5.2.3.5 of this volume of design examples outlines the design load combinations, which include the
redundancy factor as appropriate. Considering two senses of accidental torsion for loading in each
direction and including orthogonal effects results in a large number of load cases. The detailed
calculations presented here are limited to two primary conditions, both for a combined foundation for
columns at Grids A-5 and A-6: the downward case (1.4D +0.5L +0.32Ex +1.11Ey) and the upward case
4-13
(0.7D +0.32Ex +1.11Ey). [Because the redundancy factor is changed substantially in the 2003
Provisions, the factors in these load combinations would change.]
Before loads can be computed, attention must be given to Provisions Sec. 5.4.5 [5.2.5]. That section
permits foundations of structures . . . to be designed for three-fourths of the foundation overturning
design moment, M
f
. Because the overturning moment in question is the global overturning moment for
the system, judgment must be used in determining which design actions may be reduced. If the seismic-
force-resisting system consists of isolated shear walls, the shear wall overturning moment at the base best
fits that description. For a perimeter moment-resisting frame, most of the global overturning resistance is
related to axial loads in columns. Therefore, in this example column axial loads (Rz) from load cases Ex
and Ey will be multiplied by 0.75 and all other load effects will remain unreduced.
4.1.3.2 Downward Case (1.4D + 0.5L + 0.32Ex + 1.11Ey)
In order to perform the overturning checks a footing size must be assumed. Preliminary checks (not
shown here) confirmed that isolated footings under single columns were untenable. Check overturning
for a footing that is 10 ft wide by 40 ft long by 5 ft thick. Further, assume that the top of the footing is
2 ft below grade (the overlying soil contributes to the resisting moment). (In these calculations the
0.2S
DS
D modifier for vertical accelerations is used for the dead loads applied to the foundation but not for
the weight of the foundation and soil. This is the authors interpretation of the Provisions. The footing
and soil overburden are not subject to the same potential for dynamic amplification as the dead load of the
superstructure, and it is not common practice to include the vertical acceleration on the weight of the
footing and the overburden. Furthermore, for footings that resist significant overturning, this issue makes
a significant difference in design.) Combining the loads from columns at Grids A-5 and A-6 and
including the weight of the foundation and overlying soil produces the following loads at the foundation-
soil interface:
P =applied loads +weight of foundation and soil
=1.4(-203.8 - 103.5) +0.5(-43.8 - 22.3) +0.75[0.32(3.8 +51.8) +1.11(-21.3 +281)]
- 1.2[10(40)(5)(0.15) +10(40)(2)(0.125)]
=-714 kips.
M
xx
=direct moments +moment due to eccentricity of applied axial loads
=0.32(53.6 +47.7) +1.11(-1011.5 - 891.0)
+[1.4(-203.8) +0.5(-43.8) +0.75(0.32)(3.8) +0.75(1.11)(-21.3)](12.5)
+[1.4(-103.5) +0.5(-22.3) +0.75(0.32)(51.8) +0.75(1.11)(281)](-12.5)
=-7258 ft-kips.
M
yy
=0.32(-243.1 - 246.9) +1.11(8.1 +13.4)
=-133 ft-kips. (The resulting eccentricity is small enough to neglect here, which simplifies the
problem considerably.)
V
x
=0.32(-13.8 - 14.1) +1.11(0.5 +0.8)
=-7.49 kips.
V
y
=0.32(4.6 +3.7) +1.11(-85.1 -68.2)
=-167.5 kips.
Note that the above load combination does not yield the maximum downward load. Reversing the
direction of the seismic load results in P =-1173 kips and M
xx
=3490 ft-kips. This larger axial load does
not control the design because the moment is so much less that the resultant is within the kern and no
uplift occurs.
4-14
The soil calculations that follow use a different sign convention than that in the analysis results noted
above; compression is positive for the soil calculations. The eccentricity is:
e =|M/P| =7258/714 =10.17 ft.
Where e is less than L/2, a solution to the overturning problem exists; however, as e approaches L/2, the
bearing pressures increase without bound. Since e is greater than L/6 =40/6 =6.67 ft, uplift occurs and
the maximum bearing pressure is:
max
2 2(714)
4.84ksf
40
3 3(10) 10.17
2 2
P
q
L
B e
= = =

and the length of the footing in contact with the soil is:
.
40
3 3 10.17 29.5ft
2 2
L
L e

= = =

The bearing capacity q
c
=3000 B' =3000 min(B, L'/2) =3000 min(10, 29.5/2) =30,000 psf =30 ksf.
(L'/2 is used as an adjustment to account for the gradient in the bearing pressure in that dimension.)
The design bearing capacity q
c
=0.6(30 ksf) =18 ksf >4.84 ksf. OK
The foundation satisfies overturning and bearing capacity checks. The upward case, which follows, will
control the sliding check.
4.1.3.3 Upward Case (0.7D + 0.32Ex + 1.11Ey)
For the upward case the loads are:
P =-346 kips
M
xx
=-6240 ft-kips
M
yy
=-133 ft-kips (negligible)
V
x
=-7.5 kips
V
y
=-167 kips
The eccentricity is:
e =|M/P| =6240/346 =18.0 ft.
Again, e is greater than L/6, so uplift occurs and the maximum bearing pressure is:
max
2(346)
11.5ksf
40
3(10) 18.0
2
q = =

and the length of the footing in contact with the soil is:
.
40
3 18.0 6.0ft
2
L

= =

4-15
c
=3000 min(10, 6/2) =9,000 psf =9.0 ksf.
c
=0.6(9.0 ksf) =5.4 ksf <11.5 ksf. NG
Using an elastic distribution of soil pressures, the foundation fails the bearing capacity check (although
stability is satisfied). Try the plastic distribution. Using this approach, the bearing pressure over the
entire contact area is assumed to be equal to the design bearing capacity. In order to satisfy vertical
equilibrium, the contact area times the design bearing capacity must equal the applied vertical load P.
Because the bearing capacity used in this example is a function of the contact area and the value of P
changes with the size, the most convenient calculation is iterative.
By iteration, the length of contact area L' =4.39 ft.
c
=3000 min(10, 4.39) =13,170 psf =13.2 ksf. (No adjustment to L' is needed
as the pressure is uniform.)
c
=0.6(13.2 ksf) =7.92 ksf.
(7.92)(4.39)(10) =348 kips . 346 kips, so equilibrium is satisfied; the difference is rounded off.
The resisting moment, M
R
=P (L/2-L'/2) =346 (40/2 - 4.39/2) =6160 ft-kip . 6240 ft-kip. OK
Therefore, using a plastic distribution of soil pressures, the foundation satisfies overturning and bearing
capacity checks.
The calculation of demands on concrete sections for strength checks should use the same soil stress
distribution as the overturning check. Using a plastic distribution of soil stresses defines the upper limit
of static loads for which the foundation remains stable, but the extreme concentration of soil bearing tends
to drive up shear and flexural demands on the concrete section. It should be noted that the foundation
may remain stable for larger loads if they are applied dynamically; even in that case, the strength demands
on the concrete section will not exceed those computed on the basis of the plastic distribution.
For the sliding check, initially consider base traction only. The sliding demand is:
.
2 2 2 2
( 7.49) ( 167) 167.2kips
x y
V V V = + = + =
As calculated previously, the total compression force at the bottom of the foundation is 346 kips. The
design sliding resistance is:
V
c
= friction coefficient P =0.8(0.65)(346 kips) =180 kips >167.2 kips. OK
If base traction alone had been insufficient, resistance due to passive pressure on the leading face could be
included. Sec. 4.2.2.2 below illustrates passive pressure calculations for a pile cap.
The calculations performed in Sec. 4.1.3.2 and 4.1.3.3 are repeated for combined footings at middle and
side locations. Figure 4.1-6 shows the results.
4-16
Corner:
10'x40'x5'-0" w/
top of footing
2'-0" below grade
Middle:
5'x30'x4'-0"
Side:
8'x32'x4'-0"
Figure 4.1-6 Foundation plan for moment-resisting frame system.
Figure 4.1-7 Framing plan for concentrically
braced frame system.
One last check of interest is to compare the flexural stiffness of the footing with that of the steel column,
which is needed because the steel frame design was based upon flexural restraint at the base of the
columns. Using an effective moment of inertia of 50 percent of the gross moment of inertia and also
using the distance between columns as the effective span, the ratio of EI/L for the smallest of the
combined footings is more than five times the EI/h for the steel column. This is satisfactory for the
design assumption.
4.1.4 Design for Concentrically Braced Frame System
4-17
Framing Alternate B in Sec. 5.2 of this volume of design examples employs a concentrically braced frame
system at a central core to provide resistance to seismic loads. A framing plan for the system is shown in
Figure 4.1-7.
4.1.4.1 Check Mat Size for Overturning
Uplift demands at individual columns are so large that the only practical shallow foundation is one that
ties together the entire core. The controlling load combination for overturning has minimum vertical
loads (which help to resist overturning), primary overturning effects (M
xx
) due to loads applied parallel to
the short side of the core, and smaller moments about a perpendicular axis (M
yy
) due to orthogonal effects.
Assume mat dimensions of 45 ft by 95 ft by 7 ft thick with the top of the mat 3'-6" below grade.
Combining the factored loads applied to the mat by all eight columns and including the weight of the
foundation and overlying soil produces the following loads at the foundation-soil interface:
P =-7,849 kips
M
xx
=-148,439 ft-kips
M
yy
=-42,544 ft-kips
V
x
=-765 kips
V
y
=-2,670 kips
Figure 4.1-8 shows the soil pressures that result from application in this controlling case, depending on
the soil distribution assumed. In both cases the computed uplift is significant. In Part a of the figure the
contact area is shaded. The elastic solution shown in Part b was computed by modeling the mat in RISA
3D with compression only soil springs (with the stiffness of edge springs doubled as recommended by
Bowles). For the elastic solution the average width of the contact area is 11.1 ft and the maximum soil
pressure is 16.9 ksf.
c
=3000 min(95, 11.1/2) =16,650 psf =16.7 ksf.
c
=0.6(16.7 ksf) =10.0 ksf <16.9 ksf. NG
4-18
(a)
Plastic
solution
(b)
Elastic solution
pressures (ksf)
0
4
8
12
16
12.2 ksf
~
Figure 4.1-8 Soil pressures for controlling bidirectional case.
As was done in Sec. 4.1.3.3 above, try the plastic distribution. The present solution has an additional
complication as the off-axis moment is not negligible. The bearing pressure over the entire contact area is
assumed to be equal to the design bearing capacity. In order to satisfy vertical equilibrium, the contact
area times the design bearing capacity must equal the applied vertical load P. The shape of the contact
area is determined by satisfying equilibrium for the off-axis moment. Again the calculations are iterative.
Given the above constraints, the contact area shown in Figure 4.1-8 is determined. The length of the
contact area is 4.46 ft at the left side and 9.10 ft at the right side. The average contact length, for use in
determining the bearing capacity, is (4.46 +9.10)/2 =6.78 ft. The distances from the center of the mat to
the centroid of the contact area are
5.42 ft
18.98 ft
x
y
=
=
c
=3000 min(95, 6.78) =20,340 psf =20.3 ksf.
c
=0.6(20.3 ksf) =12.2 ksf.
(12.2)(6.78)(95) =7,858 kips . 7,849 kips, confirming equilibrium for vertical loads.
(7,849)(5.42) =42,542 ft-kips . 42,544 ft-kips, confirming equilibrium for off-axis moment.
The resisting moment, >148,439 ft-kips. OK
,
7849(18.98) 148,974ft-kips
R xx
M P y = = =
So, the checks of stability and bearing capacity are satisfied. The mat dimensions are shown in Figure
4.1-9.
4-19
Mat:
45'x95'x7'-0"
with top of mat
3'-6" below grade
Figure 4.1-9 Foundation plan for concentrically braced frame system.
4.1.4.2 Design Mat for Strength Demands
As was previously discussed, the computation of strength demands for the concrete section should use the
same soil pressure distribution as was used to satisfy stability and bearing capacity. Because dozens of
load combinations were considered and hand calculations were used for the plastic distribution checks,
the effort required would be considerable. The same analysis used to determine elastic bearing pressures
yields the corresponding section demands directly. One approach to this dilemma would be to compute
an additional factor that must be applied to selected elastic cases to produce section demands that are
consistent with the plastic solution. Rather than provide such calculations here, design of the concrete
section will proceed using the results of the elastic analysis. This is conservative for the demand on the
concrete for the same reason that it was unsatisfactory for the soil: the edge soil pressures are high (that is,
we are designing the concrete for a peak soil pressure of 16.9 ksf, even though the plastic solution gives
12.2 ksf).
[Note that Sec. 7.2.3 of the 2003 Provisions requires consideration of parametric variation for soil
properties where foundations are modeled explicitly. This example does not illustrate such calculations.]
Concrete mats often have multiple layers of reinforcement in each direction at the top and bottom of their
thickness. Use of a uniform spacing for the reinforcement provided in a given direction greatly increases
the ease of construction. The minimum reinforcement requirements defined in Sec. 10.5 of ACI 318 were
discussed in Sec. 4.1.1.3 above. Although all of the reinforcement provided to satisfy Sec. 7.12 of ACI
318 may be provided near one face, for thick mats it is best to compute and provide the amount of
required reinforcement separately for the top and bottom halves of the section. Using a bar spacing of 10
in. for this 7-ft-thick mat and assuming one or two layers of bars, the section capacities indicate in Table
4.1-4 (presented in order of decreasing strength) may be precomputed for use in design. The amount of
4-20
reinforcement provided for marks B, C, and D are less than the basic minimum for flexural members, so
the demands should not exceed three-quarters of the design strength where those reinforcement patterns
are used. The amount of steel provided for Mark D is the minimum that satisfies ACI 318 Sec. 7.12.
Table 4.1-4 Mat Foundation Section Capacities
Mark Reinforcement A
s
(in.
2
per ft) M
n
(ft-kip/ft) 3/4M
n
(ft-kip/ft)
A 2 layers of #10 bars
at 10 in. o.c.
3.05 899 not used
B 2 layers of #9 bars
at 10 in. o.c.
2.40 not used 534
C 2 layers of #8 bars
at 10 in. o.c.
1.90 not used 424
D #8 bars
at 10 in. o.c.
0.95 not used 215
Note: Where the area of steel provided is less than the minimum reinforcement for flexural members as
indicated in ACI 318 Sec. 10.5.1, demands are compared to 3/4 of M
n
as permitted in Sec. 10.5.3.
To facilitate rapid design the analysis results are processed in two additional ways. First, the flexural and
shear demands computed for the various load combinations are enveloped. Then the enveloped results
are presented (see Figure 4.1-10) using contours that correspond to the capacities shown for the
reinforcement patterns noted in Table 4.1-4.
4-21
(a)
M positive
x
C
L
(b)
M negative
x
(c)
M positive y
(d)
M negative y
C
L
C
L
C
L
B
C
D
B
B
C
C
D
D
B
C
D
B
C
C
C B
B
+
669
+
881
B
B
B
B
C
C
C
D
D
D
D
D
D
+
884
444
+
Figure 4.1-10 Envelope of mat foundation flexural demands.
Using the noted contours permits direct selection of reinforcement. The reinforcement provided within a
contour for a given mark must be that indicated for the next higher mark. For instance, all areas within
Contour B must have two layers of #10 bars. Note that the reinforcement provided will be symmetric
about the centerline of the mat in both directions. Where the results of finite element analysis are used in
the design of reinforced concrete elements, averaging of demands over short areas is appropriate. In
Figure 4.1-11, the selected reinforcement is superimposed on the demand contours. Figure 4.1-12 shows
a section of the mat along Gridline C.
4-22
C
L
C
L
(a)
E-W bottom
reinforcement
(b)
E-W top
reinforcement
(c)
N-S bottom
reinforcement
(d)
N-S top
reinforcement
C
L
C
L
A B A B C D C D
A B
A
B
A
B
8'-4"
3'-4"
10'-0"
5'-0"
2'-6"
7'-6"
4
'
-
2
"
4
'
-
2
"
4
'
-
2
"
4
'
-
2
"
Figure 4.1-11 Mat foundation flexural reinforcement.
4-23
3" clear
(typical)
8
"
8
"
Figure 4.1-12 Section of mat foundation.
Figure 4.1-13 presents the envelope of shear demands. The contours used correspond to the design
strengths computed assuming V
s
=0 for one-way and two-way shear. In the hatched areas the shear stress
exceeds and in the shaded areas it exceeds . The critical sections for two-way shear (as 4
c
f 2
c
f
discussed in Sec. 4.1.1.3 also are shown. The only areas that need more careful attention (to determine
whether they require shear reinforcement) are those where the hatched or shaded areas are outside the
critical sections. At the columns on Gridline D, the hatched area falls outside the critical section, so
closer inspection is needed. Because the perimeter of the hatched area is substantially smaller than the
perimeter of the critical section for punching shear, the design requirements of ACI 318 are satisfied.
One-way shears at the edges of the mat exceed the criterion. Note that the high shear stresses are 2
c
f
not produced by loads that create high bearing pressures at the edge. Rather they are produced by loads
that created large bending stresses parallel to the edge. The distribution of bending moments and shears is
not uniform across the width (or breadth) of the mat, primarily due to the torsion in the seismic loads and
the orthogonal combination. It is also influenced by the doubled spring stiffnesses used to model the soil
condition. However, when the shears are averaged over a width equal to the effective depth (d), the
demands are less than the design strength.
In this design, reinforcement for punching or beam shear is not required. If shear reinforcement cannot be
avoided, bars may be used both to chair the upper decks of reinforcement and provide resistance to shear
in which case they may be bent thus: .
4-24
(a) V
x
y
(b) V
Critical section
(typical)
Figure 4.1-13 Critical sections for shear and envelope of
mat foundation shear demands.
4.1.5 COST COMPARISON
Table 4.1-5 provides a summary of the material quantities used for all of the foundations required for the
various conditions considered. Corresponding preliminary costs are assigned. The gravity-only condition
does not represent a realistic case because design for wind loads would require changes to the
foundations; it is provided here for discussion. It is obvious that design for lateral loads adds cost as
compared to a design that neglects such loads. However, it is also worth noting that braced frame systems
usually have substantially more expensive foundation systems than do moment frame systems. This
condition occurs for two reasons. First, braced frame systems are stiffer, which produces shorter periods
and higher design forces. Second, braced frame systems tend to concentrate spatially the demands on the
4-25
foundations. In this case the added cost amounts to about $0.80/ft
2
, which is an increase of perhaps 4 or 5
percent to the cost of the structural system.
Table 4.1-5 Summary of Material Quantities and Cost Comparison
Design Condition
Concrete at Gravity
Foundations
Concrete at Lateral
Foundations
Total Excavation Total Cost
Gravity only
(see Figure 4.1-4)
310 cy at $150/cy
=$46,500
310 cy at $15/cy
=$4,650
$ 51,150
Moment frame
(see Figure 4.1-6)
233 cy at $150/cy
=$34,950
537 cy at $180/cy
=$96,660
800 cy at $15/cy
=$12,000
$143,610
Braced frame
(see Figure 4.1-9)
233 cy at $150/cy
=$34,950
1108 cy at $180/cy
=$199,440
1895 cy at $15/cy
=$28,425
$262,815
4-26
Figure 4.2-1 Design condition: column of concrete moment resisting frame
supported by pile cap and cast-in-place piles.
4.2 DEEP FOUNDATIONS FOR A 12-STORY BUILDING, SEISMIC DESIGN
CATEGORY D
This example features the analysis and design of deep foundations for a 12-tory reinforced concrete
moment-resisting frame building similar to that described in Chapter 6 of this volume of design examples.
4.2.1 Basic Information
4.2.1.1 Description
Figure 4.2-1 shows the basic design condition considered in this example. A 22 pile group is designed
for four conditions: for loads delivered by a corner and a side column of a moment-resisting frame system
for Site Classes C and E. Geotechnical parameters for the two sites are given in Table 4.2-1.
4-27
Table 4.2-1 Geotechnical Parameters
Depth Class E Site Class C Site
0 to 3 ft Loose sand/fill
=110 pcf
soil modulus parameter, k =25 pci
neglect skin friction
neglect end bearing
Loose sand/fill
=110 pcf
neglect skin friction
neglect end bearing
3 to 30 ft Soft clay
=110 pcf
undrained shear strength =430 psf
strain at 50 percent of maximum stress,
50
=0.01
skin friction (ksf) =0.3
neglect end bearing
Dense sand (one layer: 3 to 100 ft depth)
=130 pcf
skin friction (ksf)* =0.3 +0.03/ft # 2
end bearing (ksf)* =65 +0.6/ft # 150
30 to 100 ft Medium dense sand
=120 pcf
skin friction (ksf)* =0.9 +0.025/ft # 2
end bearing (ksf)* =40 +0.5/ft # 100
Pile cap
resistance
300 pcf, ultimate passive pressure 575 pcf, ultimate passive pressure
Resistance factor for capacity checks () =0.75.
Safety factor for settlement checks =2.5.
[In the 2003 Provisions, factors for
cohesive and cohesionless soils are
explicitly defined; for vertical, lateral and
rocking resistance, the values would be 0.8
for the clay layer and 0.7 for the sand
layers.]
*Skin friction and end bearing values increase (up to the maximum value noted) for each additional foot of depth
below the top of the layer. (The values noted assume a minimum pile length of 20 ft.)
The structural material properties assumed for this example are as follows:
f'
c
=3,000 psi
f
y
=60,000 psi
4.2.1.2 Provisions Parameters
Site Class =C and E (both conditions considered in this example)
S
DS
=0.9
Seismic Design Category =D (for both conditions)
4-28
4.2.1.3 Demands
The unfactored demands from the moment frame system are shown in Table 4.2-2.
Table 4.2-2 Gravity and Seismic Demands
Location Load Rx Ry Rz Mxx Myy
Corner D -351.0
L -36.0
Vx 40.7 0.6 142.5 4.8 439.0
Vy 0.8 46.9 305.6 489.0 7.0
ATx 1.2 2.6 12.0 27.4 12.9
ATy 3.1 6.7 31.9 70.2 33.0
Side D -702.0
L -72.0
Vx 29.1 0.5 163.4 3.5 276.6
Vy 0.8 59.3 18.9 567.4 6.5
ATx 0.1 3.3 8.7 31.6 1.3
ATy 0.4 8.4 22.2 80.8 3.4
Note: Units are kips and feet. Load Vx is for loads applied toward the east. ATx is the corresponding
accidental torsion case. Load Vy is for loads applied toward the north. ATy is the corresponding accidental
torsion case.
Using ASCE 7 Load Combinations 5 and 7, E as defined in Provisions Sec. 5.2.7 [4.2.2] (with 0.2S
DS
D =
0.18D and taking =1.0), considering orthogonal effects as required for Seismic Design Category D, and
including accidental torsion, the following 32 load conditions must be considered. [Although the
redundancy factor is changed substantially in the 2003 Provisions, it is expected that this system would
still satisfy the conditions needed for =1.0, so these load combinations would not change.]
1.38D +0.5L 1.0Vx 0.3Vy max(1.0ATx, 0.3ATy)
1.38D +0.5L 0.3Vx 1.0Vy max(0.3ATx, 1.0ATy)
0.72D 1.0Vx 0.3Vy max(1.0ATx, 0.3ATy)
0.72D 0.3Vx 1.0Vy max(0.3ATx, 1.0ATy)
4.2.1.4 Design Approach
For typical deep foundation systems resistance to lateral loads is provided by both piles and pile cap.
Figure 4.2-2 shows a simple idealization of this condition. The relative contributions of these piles and
pile cap depend on the particular design conditions, but often both effects are significant. Resistance to
vertical loads is assumed to be provided by the piles alone regardless of whether their axial capacity is
primarily due to end bearing, skin friction, or both. Although the behavior of foundation and
superstructure are closely related, they typically are modeled independently. Earthquake loads are
applied to a model of the superstructure, which is assumed to have fixed supports. Then the support
reactions are seen as demands on the foundation system. A similar substructure technique is usually
applied to the foundation system itself, whereby the behavior of pile cap and piles are considered
separately. This section describes that typical approach.
4-29
Passive resistance
(see Figure 4.2-5)
p-y springs
(see Figure 4.2-4)
Pile
cap
Pile
Figure 4.2-2 Schematic model of deep foundation system.
P
group
P
p
P
p
+ =
P
group
P
ot
P
ot
M
M
group
V
group
M
group
V
group
V
passive
M
V
O
Figure 4.2-3 Pile cap free body diagram.
4.2.1.4.1 Pile Group Mechanics
With reference to the free body diagram (of a 22 pile group) shown in Figure 4.2-3, demands on
individual piles as a result of loads applied to the group may be determined as follows:
and M =V R, where R is a characteristic length determined from analysis of a
4
group passive
V V
V

=
laterally loaded single pile.
, where s is the pile spacing, h is the height of the pile cap,
4
2
group group p passive
ot
V h M M h V
P
s
+ +
=
and h
p
is the height of V
passive
above Point O.
and P =P
ot
+P
p
4
group
p
P
P =
4-30
Site Class E, depth =10 ft
Site Class E, depth =30 ft
Site Class C, depth =10 ft
Site Class C, depth =30 ft
0.0
1
S
o
i
l

r
e
s
i
s
t
a
n
c
e
,

p

(
l
b
/
i
n
.
)
10
100
1,000
10,000
0.2 0.1 0.3
Pile deflection, y (in.)
0.4 0.5 0.6

0.7
100,000
0.9 0.8 1.0
Figure 4.2-4 Representative p-y curves (note that a logarithmic scale is used
on the vertical axis).
4.2.1.4.2 Contribution of Piles
The response of individual piles to lateral loads is highly nonlinear. In recent years it has become
increasingly common to consider that nonlinearity directly. Based on extensive testing of full-scale
specimens and small-scale models for a wide variety of soil conditions, researchers have developed
empirical relationships for the nonlinear p-y response of piles that are suitable for use in design.
Representative p-y curves (computed for a 22 in. diameter pile) are shown in Figure 4.2-4. The stiffness
of the soil changes by an order of magnitude for the expected range of displacements (the vertical axis
uses a logarithmic scale). The p-y response is sensitive to pile size (an effect not apparent is the figure
which is based on a single pile size); soil type and properties; and, in the case of sands, vertical stress,
which increases with depth. Pile response to lateral loads, like the p-y curves on which the calculations
are based, is usually computed using computer programs like LPILE.
4.2.1.4.3 Contribution of Pile Cap
Pile caps contribute to the lateral resistance of a pile group in two important ways: directly as a result of
passive pressure on the face of the cap that is being pushed into the soil mass and indirectly by producing
a fixed head condition for the piles, which can significantly reduce displacements for a given applied
lateral load. Like the p-y response of piles, the passive pressure resistance of the cap is nonlinear. Figure
4.2-5 shows how the passive pressure resistance (expressed as a fraction of the ultimate passive pressure)
is related to the imposed displacement (expressed as a fraction of the minimum dimension of the face
being pushed into the soil mass).
4-31
0.05
0.6
0.2
0
0.1
0.0
0.01
P
/
P
u
l
t

0.5
0.4
0.3
1.0
0.9
0.8
0.7
0.03 0.02 0.04
/H
0.06
Figure 4.2-5 Passive pressure mobilization curve (after FEMA 356).
4.2.1.4.4 Group Effect Factors
The response of a group of piles to lateral loading will differ from that of a single pile due to pile-soil-pile
interaction. (Group effect factors for axial loading of very closely spaced piles may also be developed,
but are beyond the scope of the present discussion.) A useful discussion of this group effect may be
found in PoLam Sec. 2.6.4, from which the following observations are taken:
The pile group effect has been a popular research topic within the geotechnical community for
almost 50 years. At present, there is no common consensus on the approach for group effects.
Full-size and model tests by a number of authors show that in general, the lateral capacity of a pile
in a pile group versus that of a single pile (termed efficiency) is reduced as the pile spacing is
reduced. . . .
[The experimental research reported in Brown 1987, Brown 1988, and other publications] . . .
yielded information that largely corroborated each other on the following aspects:
(1) Most of these experiments first used the single pile data to verify the validity of the widely used
Reeses and Matlocks benchmark p-y criteria and all concluded that the Reese and Matlock p-y criteria
provide reasonable solutions.
(2) The observed group effects appeared to be associated with shadowing effects and the various
researchers found relatively consistent pile group behavior in that the leading piles would be loaded more
heavily than the trailing piles when all piles are loaded to the same deflection. ... All referenced researchers
recommended to modify the single pile p-y curves by adjusting the resistance value on the single pile p-y
curves (i.e. p-multiplier). . . .
The experiments reported by McVay also included data for pile center-to-center spacing of 5D
which showed p-multipliers of 1.0, 0.85, and 0.7 for the front, middle and back row piles,
respectively. For such multipliers, the group stiffness efficiency would be about 95% and group
effects would be practically negligible.
4-32
The basis of the calculation procedure for group effect factors that is shown below is described in Chapter
6 of GROUP. In these expressions, D is the pile diameter and s is the center-to-center spacing between
the piles in question. In the equation for each efficiency factor, where s/D equals or exceeds the noted
upper limit, the corresponding value of is 1.0.
For piles that are side by side with respect to the applied load, a factor to reflect the reduction in
efficiency,
a
, may be calculated as:
.
0.5659
0.5292 for 1 3.28
a
s s
D D

= <

For piles that are in-line with respect to the applied load, a factor to reflect the reduction in efficiency (
b
)
may be calculated as follows:
Leading piles: .
0.2579
0.7309 for 1 3.37
bL
s s
D D

= <

Trailing piles: .
0.3251
0.5791 for 1 5.37
bT
s s
D D

= <

For piles that are skewed (neither in line nor side by side) with respect to the applied load, a factor to
reflect the reduction in efficiency (
s
) may be calculated as:
2 2 2 2
cos sin
s a b
= +
where
a
and
b
are calculated as defined above using s equal to the center-to-center distance along the
skew and setting equal to the angle between the direction of loading and a line connecting the two piles.
If a group contains more than two piles, the effect of each pile on each other pile must be considered. If
the effect of pile j on pile i is called
ji
and it is noted that
ji
=1.0 when j =i (as this is a single pile
condition), the p-reduction factor for any given pile i is
.
1
n
mi ji
j
f
=
=
Because the direction of loading varies during an earthquake and the overall efficiency of the group is the
primary point of interest, the average efficiency factor is commonly used for all members of a group in
the analysis of any given member. In that case, the average p-reduction factor is:
.
1 1
1
n n
m ji
i j
f
n

= =
=

For a 22 pile group thus with s =3D, the group effect factor is calculated as:
3 1
4 2
11
=1.0,
4-33
1

0.0
0.2
2
G
r
o
u
p

e
f
f
e
c
t

f
a
c
t
o
r
0.4
0.6
0.8

1.0
5 3
Group size (piles per side)

4
s =1.5 D
s =2 D
s =3 D
s =4 D
Figure 4.2-6 Calculated group effect factors.
,
0.5659
21
3
0.5292 1.0 0.985
1
a b

= = =

, and
0.2579
31
3
1.0 0.7309 0.970
1
a b

= = =

41
=
a

b
=(1.0)(1.0) =1.0 (because s/D =4.24).
Thus, f
m1
=
11

21

31

41
= (1.00)(0.985)(0.970)(1.00) =0.955 . 0.96.
By similar calculations, f
m2
=0.96, f
m3
=0.79, and f
m4
=0.79.
And finally, .
0.96 0.96 0.79 0.79
0.87
4
m
f
+ + +
= =
Figure 4.2-6 shows the group effect factors that are calculated for square pile groups of various sizes with
piles at several different spacings.
4.2.2 Pile Analysis, Design, and Detailing
4.2.2.1 Pile Analysis
For this design example it is assumed that all piles will be fixed-head, 22-in.-diameter, cast-in-place piles
arranged in 22 pile groups with piles spaced at 66 inches center-to-center. The computer program
LPILE Plus 3.0 is used to analyze single piles for both soil conditions shown in Table 4.2-1 assuming a
4-34
30
25
20
15
10
5
0
-5 0 5 10 15
Shear, V (kip)
D
e
p
t
h

(
f
t
)
Site Class C
Site Class E
Figure 4.2-7 Results of pile analysis shear versus
depth (applied lateral load is 15 kips).
30
25
20
15
10
5
0
-1000 -500 0 500
Moment, M (in.-kips)
D
e
p
t
h

(
f
t
)
Site Class C
Site Class E
Figure 4.2-8 Results of pile analysis moment versus
depth (applied lateral load is 15 kips).
length of 50 ft. Pile flexural stiffness is modeled using one-half of the gross moment of inertia because of
expected flexural cracking. The response to lateral loads is affected to some degree by the coincident
axial load. The full range of expected axial loads was considered in developing this example, but in this
case the lateral displacements, moments, and shears were not strongly affected; the plots in this section
are for zero axial load. A p-multiplier of 0.87 for group effects (as computed at the end of Sec. 4.2.1.4) is
used in all cases. Figures 4.2-7, 4.2-8, and 4.2-9 show the variation of shear, moment, and displacement
with depth (within the top 30 ft) for an applied lateral load of 15 kips on a single pile with the group
reduction factor. It is apparent that the extension of piles to depths beyond 30 ft for the Class E site (or
about 25 ft for the Class C site) does not provide additional resistance to lateral loading; piles shorter than
those lengths would have reduced lateral resistance. The trends in the figures are those that should be
expected. The shear and displacement are maxima at the pile head. Because a fixed-head condition is
assumed, moments are also largest at the top of the pile. Moments and displacements are larger for the
soft soil condition than for the firm soil condition.
4-35
30
25
20
15
10
5
0
-0.1 0.0 0.1 0.2 0.3
Displacement (in.)
D
e
p
t
h

(
f
t
)
Site Class C
Site Class E
Figure 4.2-9 Results of pile analysis displacement
versus depth (applied lateral load is 15 kips)
The analyses performed to develop Figures 4.2-7 through 4.2-9 are repeated for different levels of applied
lateral load. Figures 4.2-10 and 4.2-11 show how the moment and displacement at the head of the pile are
related to the applied lateral load. It may be seen from Figure 4.2-10 that the head moment is related to
the applied lateral load in a nearly linear manner; this is a key observation. Based on the results shown,
the slope of the line may be taken as a characteristic length that relates head moment to applied load.
Doing so produces the following:
R =46 in. for the Class C site
R =70 in. for the Class E site
4-36
0
0 5 20 25
Applied lateral load, V (kip)
H
e
a
d

m
o
m
e
n
t
,

M

(
i
n
-
k
i
p
)
Site Class C
Site Class E

10 15 30
400
800
1200
1600
Figure 4.2-10 Results of pile analysis applied
lateral load versus head moment.
Site Class C
Site Class E
0.0
0
0.2 0.4 0.6 0.8
Head displacement, (inch)
A
p
p
l
i
e
d

l
a
t
e
r
a
l

l
o
a
d
,

V

(
k
i
p
)
5
10
15
20
25
30
Figure 4.2-11 Results of pile analysis
head displacement versus applied lateral
load.
A similar examination of Figure 4.2-11 leads to another meaningful insight. The load-displacement
response of the pile in Site Class C soil is essentially linear. The response of the pile in Site Class E soil
is somewhat nonlinear, but for most of the range of response a linear approximation is reasonable (and
useful). Thus, the effective stiffness of each individual pile is:
k =175 kip/in. for the Class C site
k =40 kip/in. for the Class E site
4.2.2.2 Pile Group Analysis
The combined response of the piles and pile cap and the resulting strength demands for piles are
computed using the procedure outlined in Sec. 4.2.1.4 for each of the 32 load combinations discussed in
Sec. 4.2.1.3. Assume that each 22 pile group has a 9'-2" 9'-2" 4'-0" thick pile cap that is placed 1'-6"
below grade.
Check the Maximum Compression Case under a Side Column in Site Class C
Using the sign convention shown in Figure 4.2-3, the demands on the group are:
P =1097 kip
M
yy
=93 ft-kips
V
x
=10 kips
M
yy
=659 ft-kips
V
y
=69 kips
From preliminary checks, assume that the displacements in the x and y directions are sufficient to
mobilize 15 percent and 30 percent, respectively, of the ultimate passive pressure:
( )
1
,
1000
18 48 48 110
0.15(575) 11.0kips
12 2(12) 12 12
passive x
V

= + =

and
( )
1
,
1000
18 48 48 110
0.30(575) 22.1kips
12 2(12) 12 12
passive y
V

= + =

4-37
and conservatively take h
p
=h/3 =16 in.
Since V
passive,x
>V
x
, passive resistance alone is sufficient for this case in the x direction. However, in order
to illustrate the full complexity of the calculations, reduce V
passive,x
to 4 kips and assign a shear of 1.5 kips
to each pile in the x direction. In the y direction the shear in each pile is:
.
69 22.1
11.7kips
4
V

= =
The corresponding pile moments are:
M =1.5(46) =69 in.-kips for x-direction loading
and
M =11.7(46) =538 in.-kips for y-direction loading.
The maximum axial load due to overturning for x-direction loading is:
10(48) 93(12) 4(69) 16(4)
13.7kips
2(66)
ot
P
+ +
= =
and for y-direction loading (determined similarly) P
ot
=98.6 kips.
The axial load due to direct loading is P
p
=1097/4 =274 kips.
Therefore the maximum load effects on the most heavily loaded pile are:
P
u
=13.7 +98.6 +274 =386 kips
.
2 2
(69) (538) 542in.-kips
u
M = + =
The expected displacement in the y direction is computed as:
=V/k =11.7/175 =0.067 in., which is 0.14% of the pile cap height (h).
Reading Figure 4.2-5 with /H =0.0014, P/P
ult
. 0.34, so the assumption that 30 percent of P
ult
would be
mobilized was reasonable.
4.2.2.3 Design of Pile Section
The calculations shown in Sec. 4.2.2.2 are repeated for each of the 32 load combinations under each of
the four design conditions. The results are shown in Figures 4.2-12 and 4.2-13. In these figures, circles
indicate demands on piles under side columns and squares indicate demands on piles under corner
columns. Also plotted are the P-M design strengths for the 22-in.-diameter pile sections with various
amounts of reinforcement (as noted in the legends). The appropriate reinforcement pattern for each
design condition may be selected by noting the innermost capacity curve that envelops the corresponding
demand points. The required reinforcement is summarized in Table 4.2-4, following calculation of the
required pile length.
4-38
-300
-200
-100
0
100
200
300
400
500
600
700
800
0 500 1000 1500 2000 2500
Moment, M (in-kip)
A
x
i
a
l

l
o
a
d
,

P

(
k
i
p
)
8-#7
8-#6
6-#6
6-#5
Side
Corner
Figure 4.2-12 P-M interaction diagram for Site Class C.
-300
-200
-100
0
100
200
300
400
500
600
700
800
0 500 1000 1500 2000 2500
Moment, M (in-kip)
A
x
i
a
l

L
o
a
d
,

P

(
k
i
p
)
8-#7
8-#6
6-#6
6-#5
Side
Corner
Figure 4.2-13 P-M interaction diagram for Site Class E.
4-39
4.2.2.4 Pile Length for Axial Loads
For the calculations that follow, recall that skin friction and end bearing are neglected for the top three
feet in this example. (In these calculations, the pile cap depth is ignored effectively assuming that piles
begin at the ground surface. Because the soil capacity increases with depth and the resulting pile lengths
are applied below the bottom of the pile cap, the results are slightly conservative.)
4.2.2.4.1 Length for Settlement
Service loads per pile are calculated as P =(P
D
+P
L
)/4.
Check pile group under side column in Site Class C, assuming L =47 ft:
P =(702 +72)/4 =194 kips.
P
skin
=average friction capacity pile perimeter pile length for friction
=0.5[0.3 +0.3 +44(0.03)](22/12)(44) =243 kips.
P
end
=end bearing capacity at depth end bearing area
=[65 +44(0.6)](/4)(22/12)
2
=241 kips.
P
allow
=(P
skin
+P
end
)/S.F. =(243 +241)/2.5 =194 kips =194 kips (demand). OK
Check pile group under corner column in Site Class E, assuming L =43 ft:
P =(351 +36)/4 =97 kips.
P
skin
=[friction capacity in first layer +average friction capacity in second layer] pile perimeter
=[27(0.3) +(13/2)(0.9 +0.9 +13[0.025])](22/12) =126 kips.
P
end
=[40 +13(0.5)](/4)(22/12)
2
=123 kips.
P
allow
=(126 +123)/2.5 =100 kips >97 kips. OK
4.2.2.4.2 Length for Compression Capacity
All of the strength-level load combinations (discussed in Sec. 4.2.1.3) must be considered.
As seen in Figure 4.1-12, the maximum compression demand for this condition is P
u
=390 kips.
P
skin
=0.5[0.3 +0.3 +47(0.03)](22/12)(47) =272 kips.
P
end
=[65 +47(0.6)](/4)(22/12)
2
=246 kips.
P
n
=(P
skin
+P
end
) =0.75(272 +246) =389 kips . 390 kips. OK
Check pile group under corner column in Site Class E, assuming L =64 ft:
As seen in Figure 4.2-13, the maximum compression demand for this condition is P
u
=340 kips.
4-40
P
skin
=[27(0.3) +(34/2)(0.9 +0.9 +34[0.025])](22/12) =306 kips.
P
end
=[40 +34(0.5)](/4)(22/12)
2
=150 kips.
P
n
=(P
skin
+P
end
) =0.75(306 +150) =342 kips >340 kips. OK
4.2.2.4.3 Length for Uplift Capacity
Again, all of the strength-level load combinations (discussed in Sec. 4.2.1.3) must be considered.
As seen in Figure 4.2-12, the maximum tension demand for this condition is P
u
=-1.9 kips.
P
skin
=0.5[0.3 +0.3 +2(0.03)](22/12)(2) =3.8 kips.
P
n
=(P
skin
) =0.75(3.8) =2.9 kips >1.9 kips. OK
Check pile group under corner column in Site Class E, assuming L =52 ft.
As seen in Figure 4.2-13, the maximum tension demand for this condition is P
u
=-144 kips.
P
skin
=[27(0.3) +(22/2)(0.9 +0.9 +22[0.025])](22/12) =196 kips.
P
n
=(P
skin
) =0.75(196) =147 kips >144 kips. OK
4.2.2.4.4 Graphical Method of Selecting Pile Length
In the calculations shown above, the adequacy of the soil-pile interface to resist applied loads is checked
once a pile length is assumed. It would be possible to generate mathematical expressions of pile capacity
as a function of pile length and then solve such expressions for the demand conditions. However, a more
practical design approach is to precalculate the capacity for piles for the full range of practical lengths and
then select the length needed to satisfy the demands. This method lends itself to graphical expression as
shown in Figures 4.2-14 and 4.2-15.
4-41
80
70
60
50
40
30
20
10
0
0 100 200 300 400 500 600 700
Design resistance (kip)
P
i
l
e

d
e
p
t
h

(
f
t
)
Compression
Tension
Figure 4.2-14 Pile axial capacity as a function of length for Site Class C.
80
70
60
50
40
30
20
10
0
0 100 200 300 400 500 600 700
Design resistance (kip)
P
i
l
e

d
e
p
t
h

(
f
t
)
Compression
Tension
Figure 4.2-15 Pile axial capacity as a function of length for Site Class E.
4.2.2.4.5 Results of Pile Length Calculations
4-42
Detailed calculations for the required pile lengths are provided above for two of the design conditions.
Table 4.2-3 summarizes the lengths required to satisfy strength and serviceability requirements for all
four design conditions.
Table 4.2-3 Pile Lengths Required for Axial Loads
Piles Under Corner Column Piles Under Side Column
Condition Load Min Length Condition Load Min Length
Site Class C
Compression 331 kip 43 ft Compression 390 kip 50 ft
Uplift 133 kip 40 ft Uplift 1.9 kip 5 ft
Settlement 97 kip 19 ft Settlement 194 kip 47 ft
Site Class E
Compression 340 kip 64 ft Compression 400 kip 71 ft
Uplift 144 kip 52 ft Uplift 14.7 kip 14 ft
Settlement 97 kip 43 ft Settlement 194 kip 67 ft
The design results for all four pile conditions are shown in Table 4.2-4. The amount of longitudinal
reinforcement indicated in the table is that required at the pile-pile cap interface and may be reduced at
depth as discussed in the following section.
Table 4.2-4 Summary of Pile Size, Length, and Longitudinal Reinforcement
Piles Under Corner Column Piles Under Side Column
Site Class C
22 in. diameter by 43 ft long 22 in. diameter by 50 ft long
8-#6 bars 6-#5 bars
Site Class E
22 in. diameter by 64 ft long 22 in. diameter by 71 ft long
8-#7 bars 6-#6 bars
4.2.2.6 Pile Detailing
Provisions Sec. 7.4.4 and 7.5.4, respectively, contain special pile requirements for structures assigned to
Seismic Design Category C or higher and D or higher. In this section, those general requirements and the
specific requirements for uncased concrete piles that apply to this example are discussed. Although the
specifics are affected by the soil properties and assigned site class, the detailing of the piles designed in
this example focuses on consideration of the following fundamental items:
1. All pile reinforcement must be developed in the pile cap (Provisions Sec. 7.4.4).
2. In areas of the pile where yielding might be expected or demands are large, longitudinal and
transverse reinforcement must satisfy specific requirements related to minimum amount and
maximum spacing.
3. Continuous longitudinal reinforcement must be provided over the entire length resisting design
tension forces (ACI 318 Sec. 21.8.4.2 [21.10.4.2]).
4-43
The discussion that follows refers to the detailing shown in Figures 4.2-16 and 4.2-17.
4.2.2.6.1 Development at the Pile Cap
Where neither uplift nor flexural restraint are required, the development length is the full development
length for compression (Provisions Sec. 7.4.4). Where the design relies on head fixity or where
resistance to uplift forces is required (both of which are true in this example), pile reinforcement must be
fully developed in tension unless the section satisfies the overstrength load condition or demands are
limited by the uplift capacity of the soil-pile interface (Provisions Sec. 7.5.4). For both site classes
considered in this example, the pile longitudinal reinforcement is extended straight into the pile cap a
distance that is sufficient to fully develop the tensile capacity of the bars. In addition to satisfying the
requirements of the Provisions, this approach offers two advantages. By avoiding lap splices to field-
placed dowels where yielding is expected near the pile head (although such would be permitted by
Provisions Sec. 7.4.4), more desirable inelastic performance would be expected. Straight development,
while it may require a thicker pile cap, permits easier placement of the pile caps bottom reinforcement
followed by the addition of the spiral reinforcement within the pile cap. Note that embedment of the
entire pile in the pile cap facilitates direct transfer of shear from pile cap to pile, but is not a requirement
of the Provisions.
4-44
(4) #5
#4 spiral at
9 inch pitch
(6) #5
#4 spiral at
9 inch pitch
(6) #5
#4 spiral at
4.5 inch pitch
4" pile
embedment
Section A
Section B
Section C
C
B
A
2
1
'
-
0
"
2
3
'
-
0
"
6
'
-
4
"
Figure 4.2-16 Pile detailing for Site Class C (under side column).
4-45
(4) #7
#4 spiral at
9 inch pitch
(6) #7
#5 spiral at
3.5 inch pitch
(8) #7
#5 spiral at
3.5 inch pitch
4" pile
embedment
Section A
Section B
Section C
C
B
A
3
2
'
-
0
"
2
0
'
-
0
"
1
2
'
-
4
"
Figure 4.2-17 Pile detailing for Site Class E (under corner column).
4-46
4.2.2.6.2 Longitudinal and Transverse Reinforcement Where Demands Are Large
Requirements for longitudinal and transverse reinforcement apply over the entire length of pile where
demands are large. For uncased concrete piles in Seismic Design Category D at least four longitudinal
bars (with a minimum reinforcement ratio of 0.005) must be provided over the largest region defined as
follows: the top one-half of the pile length, the top 10 ft below the ground, or the flexural length of the
pile. The flexural length is taken as the length of pile from the cap to the lowest point where 0.4 times the
concrete section cracking moment (see ACI 318 Sec. 9.5.2.3) exceeds the calculated flexural demand at
that point. [A change made in Sec. 7.4.4.1 of the 2003 Provisions makes it clear that the longitudinal
reinforcement must be developed beyond this point.] For the piles used in this example, one-half of the
pile length governs. (Note that providing a given reinforcement ratio means that the reinforcement in
question must be developed at that point. Bar development and cutoff are discussed in more detail in
Chapter 6 of this volume of design examples.) Transverse reinforcement must be provided over the same
length for which minimum longitudinal reinforcement requirements apply. Because the piles designed in
this example are larger than 20 in. in diameter, the transverse reinforcement may not be smaller than 0.5
in. diameter. For the piles shown in Figures 4.2-16 and 4.2-17 the spacing of the transverse reinforcement
in the top half of the pile length may not exceed the least of: 12d
b
(7.5 in. for #5 longitudinal bars and
10.5 in. for #7 longitudinal bars), 22/2 =11 in., or 12 in.
Where yielding may be expected, even more stringent detailing is required. For the Class C site, yielding
can be expected within three diameters of the bottom of the pile cap (3D =3 22 =66 in.). Spiral
reinforcement in that region must not be less than one-half of that required in ASCE 318 Sec. 21.4.4.1(a)
of ACI 318 (since the site is not Class E, F, or liquefiable) and the requirements of Sec. 21.4.4.2 and
21.4.4.3 must be satisfied. Note that Sec. 21.4.4.1(a) refers to Eq. (10-6) [10-5], which often will govern.
In this case, the minimum volumetric ratio of spiral reinforcement is one-half that determined using ACI
318 Eq. (10-6) [10-5]. In order to provide a reinforcement ratio of 0.01 for this pile section, a #4 spiral
must have a pitch of no more than 4.8 in., but the maximum spacing permitted by Sec. 21.4.4.2 is 22/4 =
5.5 in. or 6d
b
=3.75 in., so a #4 spiral at 3.75 in. pitch is used.
For the Class E site, the more stringent detailing must be provided within seven diameters of the pile cap
and of the interfaces between strata that are hard or stiff and strata that are liquefiable or are composed of
soft to medium-stiff clay (Provisions Sec. 7.5.4). The author interprets within seven diameters of . . .
the interface as applying in the direction into the softer material, which is consistent with the expected
location of yielding. Using that interpretation, the Provisions does not indicate the extent of such
detailing into the firmer material. Taking into account the soil layering shown in Table 4.2-1 and the pile
cap depth and thickness, the tightly spaced transverse reinforcement shown in Figure 4.2-17 is provided
within 7D of the bottom of pile cap and top of firm soil and is extended a little more than 3D into the firm
soil. Because the site is Class E, the full amount of reinforcement indicated in ACI 318 Sec. 21.4.4.1
must be provided. In order to provide a reinforcement ratio of 0.02 for this pile section, a #5 spiral must
have a pitch of no more than 3.7 in. The maximum spacing permitted by Sec. 21.4.4.2 is 22/4 =5.5 in. or
6d
b
=5.25 in., so a #5 spiral at 3.5 in. pitch is used.
4.2.2.6.3 Continuous Longitudinal Reinforcement for Tension
Table 4.2-3 shows the pile lengths required for resistance to uplift demands. For the Site Class E
condition under a corner column (Figure 4.2-17), longitudinal reinforcement must resist tension for at
least the top 52 ft (being developed at that point). Extending four longitudinal bars for the full length and
providing widely spaced spirals at such bars reflect the designers judgment (not specific requirements of
the Provisions). For the Site Class C condition under a side column (Figure 4.2-16), design tension due
to uplift extends only about 5 ft below the bottom of the pile cap. Therefore, a design with Section C of
Figure 4.2-16 being unreinforced would satisfy the Provisions requirements, but the author has decided to
extend very light longitudinal and nominal transverse reinforcement for the full length of the pile.
4-47
(2) #6 top bars
(3) #6 bottombars
#4 ties at 7" o.c.
2" clear
at sides
3" clear at
top and bottom
Figure 4.2-18 Foundation tie section.
4.2.3 Other Considerations
4.2.3.1 Foundation Tie Design and Detailing
Provisions Sec. 7.4.3 requires that individual pile caps be connected by ties. Such ties are often grade
beams, but the Provisions would permit use of a slab (thickened or not) or calculations that demonstrate
that the site soils (assigned to Site Class A, B, or C) provide equivalent restraint. For this example, a tie
beam between the pile caps under a corner column and a side column will be designed. The resulting
section is shown in Figure 4.2-18.
For pile caps with an assumed center-to-center spacing of 32 ft in each direction, and given P
group
=1121
kips under a side column and P
group
=812 kips under a corner column, the tie is designed as follows.
As indicated in Provisions Sec. 7.4.3, the minimum tie force in tension or compression equals the product
of the larger column load times S
DS
divided by 10 =1121(0.90)/10 =101 kips.
The design strength for five #6 bars is A
s
f
y
=0.8(5)(0.44)(60) =106 kips >101 kips. OK
According to ACI 318 Sec. 21.8.3.2 [21.10.3.2], the smallest cross-sectional dimension of the tie beam
must not be less than the clear spacing between pile caps divided by 20 =(32'-0" - 9'-2")/20 =13.7 in.
Use a tie beam that is 14 in. wide and 16 in. deep. ACI 318 Sec. 21.8.3.2 [21.10.3.2] further indicates that
closed ties must be provided at a spacing of not more than one-half the minimum dimension =14/2 =7 in.
Assuming that the surrounding soil provides restraint against buckling, the design strength of the tie beam
concentrically loaded in compression is:
P
n
=0.8[0.85f'
c
(A
g
- A
st
) +f
y
A
st
]
=0.8(0.65)[0.85(3)(16)(14) +60(5)(0.44)] =366 kips >101 kips. OK
4.2.3.2 Liquefaction
For Seismic Design Categories C, D, E and F, Provisions Sec. 7.4.1 requires that the geotechnical report
address potential hazards due to liquefaction. For Seismic Design Categories D, E and F, Provisions Sec.
7.5.1 and 7.5.3 [7.5.1 and 7.5.2] further require that the geotechnical report describe the likelihood and
potential consequences of liquefaction and soil strength loss (including estimates of differential
settlement, lateral movement, and reduction in foundation soil-bearing capacity) and discuss mitigation
measures. [In the 2003 Provisions, Sec. 7.5.2 also requires that the geotechnical report describe lateral
loads on foundations, increases in lateral pressures on retaining walls, and flotation of embedded
4-48
structures.] During the design of the structure, such measures (which can include ground stabilization,
selection of appropriate foundation type and depths, and selection of appropriate structural systems to
accommodate anticipated displacements [and forces in the 2003 Provisions]) must be considered.
Commentary Section 7.4.1 contains a calculation procedure that can be used to evaluate the liquefaction
hazard, but readers should refer to Youd for an update of the methods described in the Commentary.
[Sec. 7.4.1 of the 2003 Commentary has been updated to reflect Youd and other recent references.]
4.2.3.3 Kinematic Interaction
Piles are subjected to curvature demands as a result of two different types of behavior: inertial interaction
and kinematic interaction. The term inertial interaction is used to describe the coupled response of the
soil-foundation-structure system that arises as a consequence of the mass properties of those components
of the overall system. The structural engineers consideration of inertial interaction is usually focused on
how the structure loads the foundation and how such loads are transmitted to the soil (as shown in the pile
design calculations that are the subject of most of this example) but also includes assessment of the
resulting foundation movement. The term kinematic interaction is used to describe the manner in which
the stiffness of the foundation system impedes development of free-field ground motion. Consideration
of kinematic interaction by the structural engineer is usually focused on assessing the strength and
ductility demands imposed directly on piles by movement of the soil. Although it is rarely done in
practice, the first two sentences of Provisions Sec. 7.5.4 require consideration of kinematic interaction for
foundations of structures assigned to Seismic Design Category D, E, or F. Kramer discusses kinematic
and inertial interaction and the methods of analysis employed in consideration of those effects, and
demonstrates that the solution to the entire soil-structure interaction problem is equal to the sum of the
solutions of the kinematic and inertial interaction analyses.
One approach that would satisfy the requirements of the Provisions would be as follows:
1. The geotechnical consultant performs appropriate kinematic interaction analyses considering
free-field ground motions and the stiffness of the piles to be used in design.
2. The resulting pile demands, which generally are greatest at the interface between stiff and soft
strata, are reported to the structural engineer.
3. The structural engineer designs piles for the sum of the demands imposed by the vibrating
superstructure and the demands imposed by soil movement.
A more practical, but less rigorous, approach would be to provide appropriate detailing in regions of the
pile where curvature demands imposed directly by earthquake ground motions are expected to be
significant. Where such a judgment-based approach is used, one must decide whether to provide only
additional transverse reinforcement in areas of concern to improve ductility or whether additional
longitudinal reinforcement should also be provided to increase strength. The third sentence of Provisions
Sec. 7.5.4, which defines a specific instance in which this second method is to be employed to define
areas requiring additional transverse reinforcement, helps to make an argument for general application of
this practical approach.
4.2.3.4 Design of Pile Cap
Design of pile caps for large pile loads is a very specialized topic for which detailed treatment is beyond
the scope of this volume of design examples. CRSI notes that most pile caps are designed in practice by
various short-cut rule-of-thumb procedures using what are hoped to be conservative allowable stresses.
Wang & Salmon indicates that pile caps frequently must be designed for shear considering the member
as a deep beam. In other words, when piles are located inside the critical sections d (for one-way action)
4-49
or d/2 (for two-way action) from the face of column, the shear cannot be neglected. They go on to note
that there is no agreement about the proper procedure to use. Direct application of the special
provisions for deep flexural members as found in ACI 318 is not possible as the design conditions are
somewhat different. CRSI provides a detailed outline of a design procedure and tabulated solutions, but
the procedure is developed for pile caps subjected to concentric vertical loads only (without applied
overturning moments or pile head moments). Strut-and-tie models (as described in Appendix A of the
2002 edition of ACI 318) may be employed, but their application to elements with important three-
dimensional characteristics (such as pile caps for groups larger than 21) is so involved as to preclude
hand calculations.
4.2.3.5 Foundation Flexibility and Its Impact on Performance
4.2.3.5.1 Discussion
Most engineers routinely use fixed-base models. Nothing in the Provisions prohibits that common
practice; the consideration of soil-structure interaction effects (Provisions Sec. 5.8 [5.6]) is permitted
but not required. Such fixed-base models can lead to erroneous results, but engineers have long assumed
that the errors are usually conservative. There are two obvious exceptions to that assumption: soft soil
site-resonance conditions (e.g., as in the 1985 Mexico City earthquake) and excessive damage or even
instability due to increased displacement response.
Site resonance can result in significant amplification of ground motion in the period range of interest. For
sites with a fairly long predominant period, the result is spectral accelerations that increase as the
structural period approaches the site period. However, the shape of the general design spectrum used in
the Provisions does not capture that effect; for periods larger than T
0
, accelerations remain the same or
decrease with increasing period. Therefore, increased system period (as a result of foundation flexibility)
always leads to lower design forces where the general design spectrum is used. Site-specific spectra may
reflect long-period site-resonance effects, but the use of such spectra is required only for Class F sites.
Clearly, an increase in displacements, caused by foundation flexibility, does change the performance of a
structure and its contents raising concerns regarding both stability and damage. Earthquake-induced
instability of buildings has been exceedingly rare. The analysis and acceptance criteria in the Provisions
are not adequate to the task of predicting real stability problems; calculations based on linear, static
behavior cannot be used to predict instability of an inelastic system subjected to dynamic loading. While
Commentary Sec. 5.2.8 [4.5.1]indicates that structural stability was considered in arriving at the
consensus judgment reflected in the drift limits, such considerations were qualitative. In point of fact,
the values selected for the drift limits were selected considering damage to nonstructural systems (and,
perhaps in some cases, control of structural ductility demands). For most buildings, application of the
Provisions is intended to satisfy performance objectives related to life safety and collapse prevention, not
damage control or post-earthquake occupancy. Larger design forces and more stringent drift limits are
applied to structures assigned to Seismic Use Group II or III in the hope that those measures will improve
performance without requiring explicit consideration of such performance. Although foundation
flexibility can affect structural performance significantly, the fact that all consideration of performance in
the context of the Provisions is approximate and judgment-based has made it difficult to define how such
changes in performance should be characterized. Explicit consideration of performance measures also
tends to increase engineering effort substantially, so mandatory performance checks are often resisted by
the user community.
The engineering framework established in FEMA 356 is more conducive to explicit use of performance
measures. In that document (Sec. 4.4.3.2.1 and 4.4.3.3.1), the use of fixed-based structural models is
prohibited for buildings being rehabilitated for the Immediate Occupancy Performance Level that are
4-50
sensitive to base rotations or other types of foundation movement. In this case the focus is on damage
control rather than structural stability.
4.2.3.5.2 Example Calculations
To assess the significance of foundation flexibility, one may compare the dynamic characteristics of a
fixed-base model to those of a model in which foundation effects are included. The effects of foundation
flexibility become more pronounced as foundation period and structural period approach the same value.
For this portion of the example, use the Site Class E pile design results from Sec. 4.2.2.1 and consider the
north-south response of the concrete moment frame building located in Berkeley (Sec. 6.2) as
representative for this building.
Stiffness of the Structure. Calculations of the effect of foundation flexibility on the dynamic response of
a structure should reflect the overall stiffness of the structure (e.g., that associated with the fundamental
mode of vibration), rather than the stiffness of any particular story. Table 6-2 shows that the total weight
of the structure is 36,462 kips. Table 6-5 shows that the calculated period of the fixed-base structure is
2.50 seconds, and Table 6-4 indicates that 80.2 percent of the mass participates in that mode. Using the
equation for the undamped period of vibration of a single-degree-of-freedom oscillator, the effective
stiffness of the structure is:
( )
2
2
2 2
4 (0.802)36,462 386.1
4
478 kip/in.
2.50
M
K
T
= = =
Foundation Stiffness. As seen in Figure 6-1 there are 36 moment frame columns. Assume that a 22 pile
group supports each column. As shown in Sec. 4.2.2.1, the stiffness of each pile is 40 kip/in. Neglecting
both the stiffness contribution from passive pressure resistance and the flexibility of the beam-slab system
that ties the pile caps, the stiffness of each pile group is 4 40 =160 kip/in. and the stiffness of the entire
foundation system is 36 160 =5760 kip/in.
Effect of Foundation Flexibility. Because the foundation stiffness is more than 10 times the structural
stiffness, period elongation is expected to be minimal. To confirm this expectation the period of the
combined system is computed. The total stiffness for the system (springs in series) is:
1 1
441 kip/in.
1 1 1 1
478 5760
combined
structure fdn
K
K K
= = =
+ +
Assume that the weight of the foundation system is 4000 kips and that 100 percent of the corresponding
mass participates in the new fundamental mode of vibration. The period of the combined system is
[ ] (0.802)(36,462) (1.0)(4000) 386.1
2 2 2.78 sec
441
M
T
K

+
= = =
which is an 11percent increase over that predicted by the fixed-base model. For systems responding in
the constant-velocity portion of the spectrum, accelerations (and thus forces) are a function of 1/T and
relative displacements are a function of T. Therefore, with respect to the fixed-based model, the
combined system would have forces that are 10 percent smaller and displacements that are 11 percent
larger. In the context of earthquake engineering, those differences are not significant.

Foundation Analysis and Design: Michael Valley, P.E

Uploaded by

Copyright:

Available Formats

Foundation Analysis and Design: Michael Valley, P.E

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Foundation Analysis and Design: Michael Valley, P.E

Uploaded by

Copyright:

Available Formats

4-1

You might also like