Bearing Capacity of Soils
Bearing Capacity of Soils
Bearing Capacity of Soils
P: (877) 322-5800
F: (877) 322-4774
[email protected]
DEPARTMENT OF THE ARMY EM 1110-1-1905
U.S. Army Corps of Engineers
CECW-EG Washington, DC 20314-1000
Engineer Manual
No. 1110-1-1905 30 October 1992
3. General. This manual is intended as a guide for determining allowable and ulti-
mate bearing capacity. It is not intended to replace the judgment of the design
engineer on a particular project.
WILLIAM D. BROWN
Colonel, Corps of Engineers
Chief of Staff
_______________________________________________________________________________
This manual supersedes EM 1110-2-1903, dated 1 July 1958.
DEPARTMENT OF THE ARMY EM 1110-1-1905
U.S. Army Corps of Engineers
CECW-EG Washington, DC 20314-1000
Engineer Manual
No. 1110-1-1905 30 October 1992
Table of Contents
CHAPTER 1 INTRODUCTION
Purpose and Scope . . . . . . . . . . . . . . 1-1 1-1
Definitions . . . . . . . . . . . . . . . . . 1-2 1-2
Failure Modes . . . . . . . . . . . . . . . . 1-3 1-8
Factors Influencing Bearing Capacity . . . . . 1-4 1-11
i
EM 1110-1-1905
30 Oct 92
ii
EM 1110-1-1905
30 Oct 92
CHAPTER 1
INTRODUCTION
1-1. Purpose and Scope. This manual presents guidelines for calculation of the
bearing capacity of soil under shallow and deep foundations supporting various types
of structures and embankments. This information is generally applicable to
foundation investigation and design conducted by Corps of Engineer agencies.
(1) Definitions, failure modes and factors that influence bearing capacity
are given in Chapter 1.
1-1
EM 1110-1-1905
30 Oct 92
TABLE 1-1
Step Procedure
4 Perform settlement analysis when possible and adjust the bearing pressure
until settlements are within tolerable limits. The resulting design bearing
pressure qd may be less than qa. Settlement analysis is particularly
needed when compressible layers are present beneath the depth of the zone
of a potential bearing failure. Settlement analysis must be performed on
important structures and those sensitive to settlement. Refer to EM
1110-1-1904 for settlement analysis of shallow foundations and embankments
and EM 1110-2-2906, Reese and O’Neill (1988) and Vanikar (1986) for
settlement of deep foundations.
1-2. Definitions.
(1-1a)
where (1-1b)
1-2
EM 1110-1-1905
30 Oct 92
(a) Bearing pressures exceeding the limiting shear resistance of the soil
cause collapse of the structure which is usually accompanied by tilting. A bearing
capacity failure results in very large downward movements of the structure,
typically 0.5 ft to over 10 ft in magnitude. A bearing capacity failure of this
type usually occurs within 1 day after the first full load is applied to the soil.
(c) Calculation of the bearing pressure required for ultimate shear failure
is useful where sufficient data are not available to perform a settlement analysis.
A suitable safety factor can be applied to the calculated ultimate bearing pressure
where sufficient experience and practice have established appropriate safety
factors. Structures such as embankments and uniformly loaded tanks, silos, and mats
founded on soft soils and designed to tolerate large settlements all may be
susceptible to a base shear failure.
(1-2a)
(1-2b)
1-3
EM 1110-1-1905
30 Oct 92
(3) Factors of Safety. Table 1-2 illustrates some factors of safety. These
FS’s are conservative and will generally limit settlement to acceptable values, but
economy may be sacrificed in some cases.
(c) FS in Table 1-2 for deep foundations are consistent with usual
compression loads. Refer to EM 1110-2-2906 for FS to be used with other loads.
1-4
EM 1110-1-1905
30 Oct 92
TABLE 1-2
Structure FS
Retaining
Walls 3
Temporary braced excavations > 2
Bridges
Railway 4
Highway 3.5
Buildings
Silos 2.5
Warehouses 2.5*
Apartments, offices 3
Light industrial, public 3.5
Footings 3
Mats > 3
Deep Foundations
With load tests 2
Driven piles with wave equation analysis 2.5
calibrated to results of dynamic pile tests
Without load tests 3
Multilayer soils 4
Groups 3
(a) A spread footing distributes column or other loads from the structure to
the soil, Figure 1-1a, where B ≤ W ≤ 10B. A continuous footing is a spread footing
where W > 10B.
1-5
EM 1110-1-1905
30 Oct 92
1-6
EM 1110-1-1905
30 Oct 92
through poor soil (loose sands, soft clays, and collapsible materials) into
competent bearing materials. Even when piles or drilled shafts are carried into
competent materials, significant settlement can still occur if compressible soils
are located below the tip of these deep foundations. Deep foundation support is
usually more economical for depths less than 100 ft than mat foundations.
(a) A pile may consist of a timber pole, steel pipe section, H-beam, solid or
hollow precast concrete section or other slender element driven into the ground
using pile driving equipment, Figure 1-2a. Pile foundations are usually placed in
groups often with spacings S of 3 to 3.5B where B is the pile diameter.
Smaller spacings are often not desirable because of the potential for pile
intersection and a reduction in load carrying capacity . A pile cap is necessary to
spread vertical and horizontal loads and any overturning moments to all of the piles
in the group. The cap of onshore structures usually consists of reinforced concrete
cast on the ground, unless the soil is expansive. Offshore caps are often
fabricated from steel.
(b) A drilled shaft is a bored hole carried down to a good bearing stratum
and filled with concrete, Figure 1-2b. A drilled shaft often contains a cage of
reinforcement steel to provide bending, tension, and compression resistance.
Reinforcing steel is always needed if the shaft is subject to lateral or tensile
loading. Drilled shaft foundations are often placed as single elements beneath a
column with spacings greater than 8 times the width or diameter of the shaft. Other
names for drilled shafts include bored and underreamed pile, pier and caisson.
Auger-cast or auger-grout piles are included in this category because these are not
driven, but installed by advancing a continous-flight hollow-stem auger to the
required depth and filling the hole created by the auger with grout under pressure
as the auger is withdrawn. Diameters may vary from 0.5 to 10 ft or more. Spacings
> 8B lead to minimal interaction between adjacent drilled shafts so that bearing
capacity of these foundations may be analyzed using equations for single shafts.
Shafts bearing in rock (rock drilled piers) are often placed closer than
8 diameters.
(d) The length L of a deep foundation may be placed at depths below ground
surface such as for supporting basements where the pile length L ≤ D, Figure 1-2a.
1-7
EM 1110-1-1905
30 Oct 92
1-3. Failure Modes. The modes of potential failure caused by a footing of width B
subject to a uniform pressure q develop the limiting soil shear strength τ s at a
given point along a slip path such as in Figure 1-3a
1-8
EM 1110-1-1905
30 Oct 92
(1-3)
where
From Figure 1-3a, the force on a unit width of footing causing shear is qu times
B, qu B. The force resisting shear is τ s times the length of the slip path
’abc’ or τ s ’abc’. The force resisting shear in a purely cohesive soil is c ’abc’
and in a purely friction soil σntan φ ’abc’. The length of the slip path ’abc’
resisting failure increases in proportion to the width of footing B.
a. General Shear. Figure 1-3a illustrates right side rotation shear failure
along a well defined and continuous slip path ’abc’ which will result in bulging of
the soil adjacent to the foundation. The wedge under the footing goes down and the
soil is pushed to the side laterally and up. Surcharge above and outside the
footing helps hold the block of soil down.
1-9
EM 1110-1-1905
30 Oct 92
(a) Shear failure has been found to occur more frequently under shallow
foundations supporting silos, tanks, and towers than under conventional buildings.
Shear failure usually occurs on only one side because soils are not homogeneous and
the load is often not concentric.
(b) Figure 1-3b illustrates shear failure in soft over rigid soil. The
failure surface is squeezed by the rigid soil.
(3) Horizontal Length of Failure. The length that the failure zone extends
from the foundation perimeter at the foundation depth Lsh, Figure 1-3a, may be
approximated by
(1-5a)
(1-5b)
where D is the depth of the foundation base beneath the ground surface and ψ’ =
45 - φ’/2. Lsh ≈ 1.73(H + D) if φ’ = 30 deg. The shear zone may extend
horizontally about 3B from the foundation base. Refer to Leonards (1962) for an
alternative method of determining the length of failure.
1-10
EM 1110-1-1905
30 Oct 92
1-11
EM 1110-1-1905
30 Oct 92
Figure 1-5. Variation of the nature of bearing capacity failure in sand with
relative density DR and relative depth D/B (Vesic 1963). Reprinted by
permission of the Transportation Research Board, Highway Research Record 39,
"Bearing Capacity of Deep Foundations in Sands" by A. B. Vesic, p. 136
(2) Cohesionless Soil. Bearing capacity of cohesionless soil and mixed "c-φ"
soils increases nonlinearly with increases in the effective friction angle.
1-12
EM 1110-1-1905
30 Oct 92
(1) The bearing capacity of larger footings with a slip path that intersects
a rigid stratum will be greater than that of a smaller footing with a slip path that
does not intersect a deeper rigid stratum, Figure 1-3.
(2) Foundations placed at depths where the structural weight equals the
weight of displaced soil usually assures adequate bearing capacity and only
recompression settlement. Exceptions include structures supported by
underconsolidated soil and collapsible soil subject to wetting.
(1) If the water table is below the depth of the failure surface, then the
water table has no influence on the bearing capacity and effective unit weights γ’D
and γ’H in Equation 1-1 are equal to the wet unit weight of the soils γ D and γ H.
(2) If the water table is above the failure surface and beneath the
foundation base, then the effective unit weight γ’H can be estimated as
(1-6)
where
1-13
EM 1110-1-1905
30 Oct 92
(3) The water table should not be above the base of the foundation to
avoid construction, seepage, and uplift problems. If the water table is above the
base of the foundation, then the effective surcharge term σ’D may be estimated by
(1-7a)
(1-7b)
where
1-14
EM 1110-1-1905
30 Oct 92
CHAPTER 2
2-2. Earthquake and Dynamic Motion. Cyclic or repeated motion caused by seismic
forces or earthquakes, vibrating machinery, and other disturbances such as vehicular
traffic, blasting and pile driving may cause pore pressures to increase in
foundation soil. As a result, bearing capacity will be reduced from the decreased
soil strength. The foundation soil can liquify when pore pressures equal or exceed
the soil confining stress reducing effective stress to zero and causes gross
differential settlement of structures and loss of bearing capacity. Structures
supported by shallow foundations can tilt and exhibit large differential movement
and structural damage. Deep foundations lose lateral support as a result of
liquefaction and horizontal shear forces lead to buckling and failure. The
potential for soil liquefaction and structural damage may be reduced by various soil
improvement methods.
2-3. Frost Action. Frost heave in certain soils in contact with water and subject
to freezing temperatures or loss of strength of frozen soil upon thawing can alter
bearing capacity over time. Frost heave at below freezing temperatures occurs from
2-1
EM 1110-1-1905
30 Oct 92
formation of ice lenses in frost susceptible soil. As water freezes to increase the
volume of the ice lense the pore pressure of the remaining unfrozen water decreases
and tends to draw additional warmer water from deeper depths. Soil below the depth
of frost action tends to become dryer and consolidate, while soil within the depth
of frost action tends to be wetter and contain fissures. The base of foundations
should be below the depth of frost action. Refer to TM 5-852-4 and Lobacz (1986).
a. Frost Susceptible Soils. Soils most susceptible to frost action are low
cohesion materials containing a high percentage of silt-sized particles. These
soils have a network of pores and fissures that promote migration of water to the
freezing front. Common frost susceptible soils include silts (ML, MH), silty sands
(SM), and low plasticity clays (CL, CL-ML).
b. Depth of Frost Action. The depth of frost action depends on the air
temperature below freezing and duration, surface cover, soil thermal conductivity
and permeability and soil water content. Refer to TM 5-852-4 for methodology to
estimate the depth of frost action in the United States from air-freezing index
values. TM 5-852-6 provides calculation methods for determining freeze and thaw
depths in soils. Figure 2-1 provides approximate frost-depth contours in the United
States.
(2) Pressure inject the soil with lime slurry or lime-flyash slurry to
decrease the mass permeability.
2-2
EM 1110-1-1905
30 Oct 92
2-4. Subsurface Voids. A subsurface void influences and decreases bearing capacity
when located within a critical depth Dc beneath the foundation. The critical
depth is that depth below which the influence of pressure in the soil from the
foundation is negligible. Evaluation of Dc is described in section 3-3b.
a. Voids. Voids located beneath strip foundations at depth ratios Dc/B > 4
cause little influence on bearing capacity for strip footings. B is the foundation
width. The critical depth ratio for square footings is about 2.
2-5. Expansive and Collapsible Soils. These soils change volume from changes in
water content leading to total and differential foundation movements. Seasonal
wetting and drying cycles have caused soil movements that often lead to excessive
long-term deterioration of structures with substantial accumulative damage. These
soils can have large strengths and bearing capacity when relatively dry.
a. Expansive Soil. Expansive soils consist of plastic clays and clay shales
that often contain colloidal clay minerals such as the montmorillonites. They
include marls, clayey siltstone and sandstone, and saprolites. Some of these soils,
especially dry residual clayey soil, may heave under low applied pressure but
collapse under higher pressure. Other soils may collapse initially but heave later
on. Estimates of the potential heave of these soils are necessary for consideration
in the foundation design.
Soils with Liquid Limit (LL) < 35 and Plasticity Index (PI) < 12 have no potential
for swell and need not be tested.
2-3
EM 1110-1-1905
30 Oct 92
(2) Potential Collapse. The potential for collapse should be determined from
results of a consolidometer test as described in EM 1110-1-1904. The soil may then
be modified as needed using soil improvement methods to reduce or eliminate the
potential for collapse.
(2) Slope stability package UTEXAS2 (Edris 1987) may be used to perform an
analysis of the bearing capacity of either the unreinforced or reinforced soil
beneath a foundation. A small slope of about 1 degree must be used to allow the
computer program to operate. The program will calculate the bearing capacity of the
weakest slip path, Figure 1-3a, of infinite length (wall) footings, foundations, or
embankments.
2-4
EM 1110-1-1905
30 Oct 92
(a) Fabric reinforcement helps to hold the embankment together while the
foundation strength increases through consolidation.
(c) A bearing failure and center sag may occur when fabrics with insufficient
tensile strength and modulus are used, when steep embankments are constructed, or
when edge anchorage of fabrics is insufficient to control embankment splitting. If
2-5
EM 1110-1-1905
30 Oct 92
the bearing capacity of the foundation soil is exceeded, the fabric must elongate to
develop the required fabric stress to support the embankment load. The foundation
soil will deform until the foundation is capable of carrying the excessive stresses
that are not carried in the fabric. Complete failure occurs if the fabric breaks.
(2-1)
where
1 + sinφ g
Kp = Rankine passive pressure coefficient,
1 - sinφ g
φg = friction angle of granular material, degrees
γc = moist unit weight of weak clay, kip/ft3
D = depth of the rectangle foundation below ground surface, ft
B = width of foundation, ft
L = length of foundation, ft
Cu = undrained shear strength of weak clay, ksf
(2) The minimum height of the granular column required to support the footing
and to obtain the maximum increase in bearing capacity is 3B.
(3) Refer to Bachus and Barksdale (1989) and Barksdale and Bachus (1983) for
further details on analysis of bearing capacity of stone columns.
2-7. Heaving Failure in Cuts. Open excavations in deep deposits of soft clay may
fail by heaving because the weight of clay beside the cut pushes the underlying clay
up into the cut, Figure 2-2 (Terzaghi and Peck 1967). This results in a loss of
ground at the ground surface. The bearing capacity of the clay at the bottom of the
cut is CuNc. The bearing capacity factor Nc depends on the shape of the cut. Nc
may be taken equal to that for a footing of the same B/W and D/B ratios as
provided by the chart in Figure 2-3, where B is the excavation width, W is the
excavation length, and D is the excavation depth below ground surface.
(2-2)
2-6
EM 1110-1-1905
30 Oct 92
(1) Sheet pile, even if the clay depth is large, will reduce flow into the
excavation compared with pile and lagging support.
2-7
EM 1110-1-1905
30 Oct 92
(2) Driving the sheet pile into a hard underlying stratum below the
excavation greatly reduces the tendency for a heave failure.
2-8. Soil Erosion and Seepage. Erosion of soil around and under foundations and
seepage can reduce bearing capacity and can cause foundation failure.
(1) Scour Around Drilled Shafts or Piles in Seawater. The scour depth may be
estimated from empirical and experimental studies. Refer to Herbich, Schiller and
Dunlap (1984) for further information.
(a) The maximum scour depth to wave height ratio is ≤ 0.2 for a medium to
fine sand.
(2-3)
where Su is in feet.
(2) Scour Around Pipelines. Currents near pipelines strong enough to cause
scour will gradually erode away the soil causing the pipeline to lose support. The
maximum scour hole depth may be estimated using methodology in Herbich, Schiller,
and Dunlap (1984).
(2) Seepage around piles can reduce skin friction. Skin friction resistance
can become less than the hydrostatic uplift pressure and can substantially reduce
bearing capacity. Redriving piles or performing load tests after a waiting period
following construction can determine if bearing capacity is sufficient.
2-8
EM 1110-1-1905
30 Oct 92
CHAPTER 3
SOIL PARAMETERS
(1) Local Building Codes. Local building codes may give presumptive
allowable bearing pressures based on past experience. This information should only
be used to supplement the findings of in situ tests and analyses using one or more
3-1
EM 1110-1-1905
30 Oct 92
methods discussed subsequently because actual field conditions, and hence bearing
pressures, are rarely identical with those conditions used to determine the
presumptive allowable bearing pressures.
(3) Drainage. The ground surface should be sloped to provide adequate runoff
of surface and rainwater from the construction area to promote trafficability and to
minimize future changes in ground moisture and soil strength. Minimum slope should
be 1 percent.
(1) Relative Density and Gradation. Relative density and gradation can be
used to estimate the friction angle of cohesionless soils, Table 3-1a. Relative
density is a measure of how dense a sand is compared with its maximum density.
3-2
EM 1110-1-1905
30 Oct 92
TABLE 3-1
40 34 36 36 38 38 41
60 36 38 38 41 41 43
80 39 41 41 43 43 44
100 42 43 43 44 44 46
(a) ASTM D 653 defines relative density as the ratio of the difference in
void ratio of a cohesionless soil in the loosest state at any given void ratio to
the difference between the void ratios in the loosest and in the densest states. A
very loose sand has a relative density of 0 percent and 100 percent in the densest
possible state. Extremely loose honeycombed sands may have a negative relative
density.
(b) Relative density may be calculated using standard test methods ASTM D
4254 and the void ratio of the in situ cohesionless soil,
(3-1a)
(3-1b)
where
3-3
EM 1110-1-1905
30 Oct 92
The specific gravity of the mineral solids may be determined using standard test
method ASTM D 854. The dry density of soils that may be excavated can be determined
in situ using standard test method ASTM D 1556.
(a) The NSPT value may be normalized to an effective energy delivered to the
drill rod at 60 percent of theoretical free-fall energy
(3-2)
where
NSPT may have an effective energy delivered to the drill rod 50 to 55 percent of
theoretical free fall energy.
(b) Table 3-1 illustrates some relative density and N60 correlations with the
angle of internal friction. Relative density may also be related with N60 through
Table 3-2c.
(c) The relative density of sands may be estimated from the Nspt by (Data from
Gibbs and Holtz 1957)
(3-3a)
(d) The relative density of sands may also be estimated from N60 by
(Jamiolkowski et al. 1988, Skempton 1986)
(3-3b)
(e) The undrained shear strength Cu in ksf may be estimated (Bowles 1988)
(3-4)
3-4
EM 1110-1-1905
30 Oct 92
TABLE 3-2
CN σ’
vo*, ksf
1.6 0.6
1.3 1.0
1.0 2.0
0.7 4.0
0.55 6.0
0.50 8.0
*σ’
vo = effective overburden pressure
Very Loose 0 - 15 0 - 3
Loose 15 - 35 3 - 8
Medium 35 - 65 8 - 25
Dense 65 - 85 25 - 42
Very Dense 85 - 100 42 - 58
(3) Cone penetration test (CPT). The CPT may be used to estimate both
relative density of cohesionless soil and undrained strength of cohesive soils
through empirical correlations. The CPT is especially suitable for sands and
preferable to the SPT. The CPT may be performed using ASTM D 3441.
3-5
EM 1110-1-1905
30 Oct 92
(3-5)
(3-6)
3-6
EM 1110-1-1905
30 Oct 92
where Cu, qc, and the total vertical overburden pressure σvo are in ksf units.
The cone factor Nk should be determined using comparisons of Cu from laboratory
undrained strength tests with the corresponding value of qc obtained from the CPT.
Equation 3-6 is useful to determine the distribution of undrained strength with
depth when only a few laboratory undrained strength tests have been performed. Nk
often varies from 14 to 20.
(4) Dilatometer Test (DMT). The DMT can be used to estimate the
overconsolidation ratio (OCR) distribution in the foundation soil. The OCR can be
used in estimating the undrained strength. The OCR is estimated from the horizontal
stress index KD by (Baldi et al 1986; Jamiolkowski et al 1988)
(3-7a)
(3-7b)
(3-7c)
where
The OCR typically varies from 1 to 3 for lightly overconsolidated soil and 6
to 8 for heavily overconsolidated soil.
(5) Pressuremeter Test (PMT). The PMT can be used to estimate the undrained
strength and the OCR. The PMT may be performed using ASTM D 4719.
(a) The limit pressure pL estimated from the PMT can be used to estimate
the undrained strength by (Mair and Wood 1987)
(3-8a)
(3-8b)
where
pL, σho, and Gs are found from results of the PMT. Equation 3-8b requires an
estimate of the shear strength to solve for Np. Np may be initially estimated as
some integer value from 3 to 8 such as 6. The undrained strength is then determined
from Equation 3-8a and the result substituted into Equation 3-8b. One or two
iterations should be sufficient to evaluate Cu.
3-7
EM 1110-1-1905
30 Oct 92
(b) σho can be used to estimate the OCR from σ’ho/σ’vo if the pore water
pressure and total vertical pressure distribution with depth are known or estimated.
(6) Field Vane Shear Test (FVT). The FVT is commonly used to estimate the in
situ undrained shear strength Cu of soft to firm cohesive soils. This test should
be used with other tests when evaluating the soil shear strength. The test may be
performed by hand or may be completed using sophisticated equipment. Details of the
test are provided in ASTM D 2573.
(3-9)
where
(3-10a)
where
(3-10b)
(c) Anisotropy can significantly influence the torque measured by the vane.
d. Water Table. Depth to the water table and pore water pressure
distributions should be known to determine the influence of soil weight and
surcharge on the bearing capacity as discussed in 1-4d, Chapter 1.
3-8
EM 1110-1-1905
30 Oct 92
3-3. Soil Exploration. Soil classification and index tests such as Atterberg
Limit, gradations, and water content should be performed on disturbed soil and
results plotted as a function of depth to characterize the types of soil in the
profile. The distribution of shear strength with depth and the lateral variation of
shear strength across the construction site should be determined from laboratory
strength tests on undisturbed boring samples. Soil classifications and strengths
may be checked and correlated with results of in situ tests. Refer to EM 1110-2-
1907 and EM 1110-1-1804 for further information.
(2) Location of Borings. Optimum locations for soil exploration may be near
the center, edges, and corners of the proposed structure. A sufficient number of
borings should be performed within the areas of proposed construction for laboratory
tests to define shear strength parameters Cu and φ of each soil layer and any
significant lateral variation in soil strength parameters for bearing capacity
analysis and consolidation and compressibility characteristics for settlement
analysis. These boring holes may also be used to measure water table depths and
pore pressures for determination of effective stresses required in bearing capacity
analysis.
(a) Preliminary exploration should require two or three borings within each
of several potential building locations. Air photos and geological conditions
assist in determining location and spacings of borings along the alignment of
proposed levees. Initial spacings usually vary from 200 to 1000 ft along the
alignment of levees.
3-9
EM 1110-1-1905
30 Oct 92
(1) 10 Percent Rule. The depth of soil exploration for at least one test
boring should be at the depth where the increase in vertical stress caused by the
structure is equal to 10 percent of the initial effective vertical overburden stress
beneath the foundation, Figure 3-2. Critical depth for bearing capacity analysis
Dc should be at least twice the minimum width of shallow square foundations or at
least 4 times the minimum width of infinitely long footings or embankments. The
depth of additional borings may be less if soil exploration in the immediate
vicinity or the general stratigraphy of the area indicate that the proposed bearing
strata have adequate thickness or are underlain by stronger formations.
(2) Depth to Primary Formation. Depth of exploration need not exceed the
depth of the primary formation where rock or soil of exceptional bearing capacity is
located.
(b) For a building foundation carried to rock 3 to 5 rock corings are usually
required to determine whether piles or drilled shafts should be used. The percent
recovery and rock quality designation (RQD) value should be determined for each rock
core. Drilled shafts are often preferred in stiff bearing soil and rock of good
quality.
(a) Dense sands and gravels and firm to stiff clays with low potential for
volume change provide the best bearing strata for foundations.
(b) Standard penetration resistance values from the SPT and cone resistance
from the CPT should be determined at a number of different lateral locations within
the construction site. These tests should be performed to depths of about twice the
minimum width of the proposed foundation.
3-10
EM 1110-1-1905
30 Oct 92
3-11
EM 1110-1-1905
30 Oct 92
TABLE 3-3
Step Description
6 Estimate Cu/σ’
vo from
(3-11)
where Cu = undrained shear strength and plot with depth, Figure 3-3c.
3-12
EM 1110-1-1905
30 Oct 92
3-13
EM 1110-1-1905
30 Oct 92
CHAPTER 4
SHALLOW FOUNDATIONS
4-1
EM 1110-1-1905
30 Oct 92
(4-1)
where
(1) Net Bearing Capacity. The net ultimate bearing capacity q’u is the
maximum pressure that may be applied to the base of the foundation without
undergoing a shear failure that is in addition to the overburden pressure at depth
D.
(4-2)
4-2
EM 1110-1-1905
30 Oct 92
(1) Bearing Capacity Factors. The Terzaghi bearing capacity factors Nc and
Nq for general shear are shown in Table 4-1 and may be calculated by
TABLE 4-1
φ’ Nq Nc Nγ φ’ Nq Nc Nγ
(4-3a)
(4-3b)
TABLE 4-2
4-3
EM 1110-1-1905
30 Oct 92
(1) Bearing Capacity Factors. Table 4-4 provides the bearing capacity
factors in 2-degree intervals.
Cohesion: ζ c = ζ cs ζ ci ζ cd
Wedge: ζγ = ζ γs ζ γi ζ γd
Surcharge: ζ q = ζ qs ζ qi ζ qd
(4-4b)
(4-4c)
(4-4d)
where
4-4
EM 1110-1-1905
30 Oct 92
4-5
EM 1110-1-1905
30 Oct 92
TABLE 4-4
Nγ
(4-7a)
(4-7b)
4-6
EM 1110-1-1905
30 Oct 92
where
d. Hansen Model. The Hansen model considers tilted bases and slopes in
addition to foundation shape and eccentricity, load inclination, and foundation
depth. Assumptions are based on an extension of Meyerhof’s work to include tilting
of the base and construction on a slope. Any D/B ratio may be used permitting
bearing capacity analysis of both shallow and deep foundations. Bearing capacity
factors Nc and Nq are the same as Meyerhof’s factors. N γ is calculated assuming
ψ = 45 + φ/2. These values of N γ are least of the methods. Correction factors
ζ c, ζ γ , and ζ q in Equation 4-1 are
Cohesion: ζ c = ζ cs ζ ci ζ cd ζ cβ ζ cδ (4-8a)
Wedge: ζγ = ζ γs ζ γi ζ γd ζ γ β ζ γ δ (4-8b)
Surcharge: ζ q = ζ qs ζ qi ζ qd ζ qβ ζ qδ (4-8c)
(1) Restrictions. Foundation shape with eccentricity ζ cs, ζ γs, and ζ qs and
inclined loading ζ ci, ζ γi, and ζ qi correction factors may not be used
simultaneously. Correction factors not used are unity.
(1) Bearing Capacity Factors. Nc and Nq are identical with Meyerhof’s and
Hansen’s factors. N γ was taken from an analysis by Caquot and Kerisel (1953) using
ψ = 45 + φ/2.
(4-9)
Equation 4-9 assumes a local shear failure and leads to a lower bound estimate of
qu. Nq from Equation 4-9 may also be used to calculate Nc and N γ by the
equations given in Table 4-6.
4-7
EM 1110-1-1905
30 Oct 92
4-8
EM 1110-1-1905
30 Oct 92
4-9
EM 1110-1-1905
30 Oct 92
(1) Program CBEAR. Program CBEAR (Mosher and Pace 1982) can be used to
calculate the bearing capacity of shallow strip, rectangular, square, or circular
footings on one or two soil layers. This program uses the Meyerhof and Vesic
bearing capacity factors and correction factors.
(2) Program UTEXAS2. UTEXAS2 is a slope stability program that can be used
to calculate factors of safety for long wall footings and embankments consisting of
multilayered soils (Edris 1987). Foundation loads are applied as surface pressures
on flat surfaces or slopes. Circular or noncircular failure surfaces may be
assumed. Noncircular failure surfaces may be straight lines and include wedges.
Shear surfaces are directed to the left of applied surface loading on horizontal
slopes or in the direction in which gravity would produce sliding on nonhorizontal
slopes (e.g., from high toward low elevation points). This program can also
consider the beneficial effect of internal reinforcement in the slope. qu
calculated by UTEXAS2 may be different from that calculated by CBEAR partly because
the FS is defined in UTEXAS2 as the available shear strength divided by the shear
stress on the failure surface. The assumed failure surfaces in CBEAR are not the
same as the minimum FS surface found in UTEXAS2 by trial and error. FS in Table 1-2
are typical for CBEAR. Program UTEXAS2 calculates factors of safety, but these FS
have not been validated with field experience. UTEXAS2 is recommended as a
supplement to the Terzaghi, Meyerhof, Hansen, and Vesic models until FS determined
by UTEXAS2 have been validated.
4-10
EM 1110-1-1905
30 Oct 92
(1) Depth of Analysis. The maximum depth of the soil profile analyzed need
not be much greater than the depth to the failure surface, which is approximately
2B for uniform soil. A deeper depth may be required for settlement analyses.
(a) If the soil immediately beneath the foundation is weaker than deeper
soil, the critical failure surface may be at a depth less than 2B.
(b) If the soil is weaker at depths greater than 2B, then the critical
failure surface may extend to depths greater than 2B.
(2) Dense Sand Over Soft Clay. The ultimate bearing capacity of a footing in
a dense sand over soft clay can be calculated assuming a punching shear failure
using a circular slip path (Hanna and Meyerhof 1980; Meyerhof 1974)
Wall Footing:
(4-10a)
Circular Footing:
(4-10b)
where
The punching shear coefficient kps can be found from the charts in Figure 4-2
using the undrained shear strength of the lower soft clay and a punching shear
parameter Cps. Cps, ratio of ζ/φ sand where ζ is the average mobilized angle of
shearing resistance on the assumed failure plane, is found from Figure 4-2d using
φ sand and the bearing capacity ratio Rbc. Rbc = 0.5γ sandBN γ /(CuNc). B is the diameter
of a circular footing or width of a wall footing. The shape factor Ss, which
varies from 1.1 to 1.27, may be assumed unity for conservative design.
(3) Stiff Over Soft Clay. Punching shear failure is assumed for stiff over
soft clay.
(a) D = 0.0. The ultimate bearing capacity can be calculated for a footing
on the ground surface by (Brown and Meyerhof 1969)
4-11
EM 1110-1-1905
30 Oct 92
4-12
EM 1110-1-1905
30 Oct 92
Wall Footing:
(4-11a)
(4-11b)
Circular Footing:
(4-11c)
(4-11d)
where
(b) D > 0.0. The ultimate bearing capacity can be calculated for a footing
placed at depth D by
where
Ncw,D may be found using Table 4-7 and Ncw,0 from Equation 4-11b. Refer to
Department of the Navy (1982) for charts that can be used to calculate bearing
capacities in two layer soils.
4-13
EM 1110-1-1905
30 Oct 92
TABLE 4-7
D/B Ncw,D/Ncw,0
0.0 1.00
0.5 1.15
1.0 1.24
2.0 1.36
3.0 1.43
4.0 1.46
h. Correction for Large Footings and Mats. Bearing capacity, obtained using
Equation 4-1 and the bearing capacity factors, gives capacities that are too large
for widths B > 6 ft. This is apparently because the 0.5 B’γ’HN γ ζ γ term becomes
too large (DeBeer 1965; Vesic 1969).
(1) Settlement usually controls the design and loading of large dimensioned
structures because the foundation soil is stressed by the applied loads to deep
depths.
(4-13)
where B > 6 ft.
(1) Bearing pressures produced by eccentric loads that include dead plus
normal live loads plus permanent lateral loads should not exceed qna pressures of
Table 4-8.
(2) Transient live loads from wind and earthquakes may exceed the allowable
bearing pressure by up to one-third.
4-14
EM 1110-1-1905
30 Oct 92
TABLE 4-8
Presumptive Allowable Bearing Pressures for Spread Footings
(Data from Department of the Navy 1982, Table 1, Chapter 4)
Nominal Allowable
Bearing Material In Place Consistency Bearing Pressure
q , ksf
na
Massive crystalline igneous and Hard sound rock 160
metamorphic rock: granite,
diorite, basalt, gneiss,
thoroughly cemented conglomerate
(sound condition allows minor
cracks)
Foliated metamorphic rock: Medium hard sound 70
slate, schist (sound condition rock
allows minor cracks)
Sedimentary rock; hard cemented Medium hard sound 40
shales, siltstone, sandstone, rock
limestone without cavities
Weathered or broken bed rock of Soft rock 20
any kind except highly
argillaceousrock (shale); Rock
Quality Designation less than 25
Compaction shale or other highly Soft rock 20
argillaceous rock in sound
condition
Well-graded mixture of fine and Very compact 20
coarse-grained soil: glacial
till, hardpan, boulder clay
(GW-GC, GC, SC)
Gravel, gravel-sand mixtures, Very compact 14
boulder gravel mixtures (SW, Medium to compact 10
SP, SW, SP) Loose 6
Coarse to medium sand, sand with Very compact 8
little gravel (SW, SP) Medium to compact 6
Loose 3
Fine to medium sand, silty or Very compact 6
clayey medium to coarse sand Medium to compact 5
(SW, SM, SC) Loose 3
Homogeneous inorganic clay, Very stiff to hard 8
sandy or silty clay (CL, CH) Medium to stiff 4
Soft 1
Inorganic silt, sandy or clayey Very stiff to hard 6
silt, varved silt-clay-fine sand Medium to stiff 3
Soft 1
4-15
EM 1110-1-1905
30 Oct 92
(4) For a bearing stratum underlain by weaker material, pressure on the weak
stratum should be less than the nominal allowable bearing pressure given in
Table 4-8
(4-14)
where
(4-15)
where W’
T is the total effective weight of soil and foundation resisting uplift.
(4-16)
where Nc,N γ ,Nq and ζ c,ζ γ ,ζ q are given in Tables 4-3, 4-4, 4-5, or 4-6. If
Hansen’s model is used, then the exponent for ζ γi and ζ qi in Table 4-5
should be changed from 5 to 2 (Bowles 1988).
4-4. In Situ Modeling of Bearing Pressures. In situ load tests of the full size
foundation are not usually done, except for load testing of piles and drilled
shafts. Full scale testing is usually not performed because required loads are
usually large and as a result these tests are expensive. The most common method is
to estimate the bearing capacity of the soil from the results of relatively simple,
less expensive in situ tests such as plate bearing, standard penetration, cone
penetration, and vane shear tests.
4-16
EM 1110-1-1905
30 Oct 92
(4-17a)
where
(2) Strength Increasing Linearly With Depth. The ultimate bearing capacity
of the foundation in cohesionless or cohesive soil with strength increasing linearly
with depth may be estimated by
(4-17b)
(4-18)
where
The results of the plate load test should indicate that q/ρi is essentially
constant. q1 and plate diameter Bp can then be input into the Terzaghi and Peck
chart for the appropriate D/B ratio, which is 1, 0.5 or 0.25 (see Figure 3-3, EM
1110-1-1904). The actual footing dimension B is subsequently input into the same
chart to indicate the allowable foundation bearing pressure.
4-17
EM 1110-1-1905
30 Oct 92
(4) Extrapolation of Test Results. Load tests performed using several plate
sizes may allow extrapolation of test results to foundations up to 6 times the plate
diameter provided the soil is similar. Other in situ test results using standard
penetration or cone penetration data are recommended for large foundation diameters
and depths more than 4Bp.
b. Standard Penetration Test (SPT). The SPT may be used to directly obtain
the allowable bearing capacity of soils for specific amounts of settlement based on
past correlations.
(4-19a)
(4-19b)
where
(4-20a)
where qa,1 is the allowable bearing capacity for limiting settlement to 1 inch. The
allowable bearing capacity for any settlement qa may be linearly related to the
allowable settlement for 1 inch obtained from Equations 4-19 assuming settlement
varies in proportion to pressure
(4-20b)
4-18
EM 1110-1-1905
30 Oct 92
where
ρ = settlement, inches
qa,1 = allowable bearing capacity for 1 inch settlement, ksf
c. Cone Penetration Test (CPT). Bearing capacity has been correlated with
cone tip resistance qc for shallow foundations with D/B ≤ 1.5 (Schmertmann 1978).
(4-21a)
(4-21b)
(4-22b)
Units are also in tsf or kg/cm2. Table 4-9 using Figure 4-3 provides a procedure
for estimating qu for footings up to B = 8 ft in width.
d. Vane Shear Test. The vane shear is suitable for cohesive soil because
bearing capacity is governed by short-term, undrained loading which is simulated by
this test. Bearing capacity can be estimated by (Canadian Geotechnical Society
1985)
(4-24)
where
4-5. Examples. Estimation of the bearing capacity is given below for (1) a wall
footing placed on the ground surface subjected to a vertical load, (2) a rectangular
footing placed below the ground surface and subjected to an inclined load, and (3) a
tilted, rectangular footing on a slope and subjected to an eccentric load.
Additional examples are provided in the user manual for CBEAR (Mosher and Pace
1982). The slope stability analysis of embankments is described in the user manual
for UTEXAS2 (Edris 1987). Bearing capacity analyses should be performed using at
least three methods where practical.
4-19
EM 1110-1-1905
30 Oct 92
TABLE 4-9
Step Procedure
(4-23a)
where
(4-23b)
where
(4-23c)
4 Estimate bearing ratio Rk from Figure 4-3 using Rd. The lower bound curve
is applicable to fissured or slickensided clays. The average curve is
applicable to all other clays unless load tests verify the upper bound curve
for intact clay.
(4-23d)
(4-23e)
4-20
EM 1110-1-1905
30 Oct 92
4-21
EM 1110-1-1905
30 Oct 92
(a) Terzaghi Method. Table 4-1 indicates Nc = 5.7, Nq = 1.0 and N γ = 0.00.
The total ultimate capacity qu is
(b) Meyerhof Method. The ultimate bearing capacity of this wall footing
using program CBEAR yields qu = 7.196 ksf. The Hansen and Vesic solutions will be
similar.
(3) Allowable Bearing Capacity. FS for this problem from Table 1-2 is 3.0.
Therefore, qa using Equation 1-2a is qu/FS = 8.000/3 = 2.7 ksf from the Terzaghi
solution and 7.196/3 = 2.4 ksf from CBEAR. The solution using program UTEXAS2 gives
a minimum FS = 2.2 for a circular failure surface of radius 3 ft with its center at
the left edge of the footing.
(4) Recommendation. qa ranges from 2.4 to 2.7 while the proposed design
pressure qd is 4 ksf. qd should be reduced to 2.4 ksf ≤ qa.
4-22
EM 1110-1-1905
30 Oct 92
4-23
EM 1110-1-1905
30 Oct 92
(1) Effective Stress Adjustment. Adjust the unit soil weights due to the
water table using Equation 1-6
ζ qs = ζ γs = 1.15
ζ qi = [1 - (θ/90)]2 = [1 - (11.4/90)]2 = 0.763
ζ qd = ζ γd = 1.115
ζq = 1.15 0.763 1.115 = 0.98
qu = 0.5 B γ’
HN γ ζ γ + σ’D Nq ζ q
= 0.5 3.00 0.08 15.67 0.49 + 0.24 18.40 0.98
= 0.92 + 4.33 = 5.25 ksf
4-24
EM 1110-1-1905
30 Oct 92
qu = 0.5 B γ’
H Nγ ζ γ + σ’
D Nq ζ q
= 0.5 3.00 0.08 15.07 0.376 + 0.24 18.40 0.907
= 0.68 + 4.01 = 4.69 ksf
qu = 0.5 B γ’H N γ ζ γ + σ’
D Nq ζ q
= 0.5 3.00 0.08 22.40 0.441 + 0.24 18.40 1.058
= 1.19 + 4.67 = 5.86 ksf
(5) Program CBEAR. Zero elevation for this problem is defined 3 ft below the
foundation base. Input to this program is as follows (refer to Figure 1-6):
4-25
EM 1110-1-1905
30 Oct 92
Total Net
Method qu, ksf q’
u, ksf
The net bearing capacity is found by subtracting γ D D = 0.12 2 = 0.24 ksf from qu,
Equation 4-2. The resultant applied pressure on the footing is qr = R/(BW) =
10.2/(3 6) = 0.57 ksf. The factor of safety of all of the above methods with
respect to the net bearing capacity is on the order of q’u/qr ≈ 9. The Hansen
method is most conservative.
4-26
EM 1110-1-1905
30 Oct 92
B’ = B - 2eB = 3 - 2 0.5 = 2 ft
W’ = W - 2eW = 5 - 2 1.0 = 3 ft
4-27
EM 1110-1-1905
30 Oct 92
(4) Hansen Method. For φ’ = 26 deg, Nq = 11.85 and N γ = 7.94 from Table
4-4. Nc is not needed since c = 0.0. From Table 4-5,
qu = 0.5 B γ’
H Nγ ζ γ + σ’D Nq ζ q
= 0.5 2.00 0.081 7.942 0.318 + 0.27 11.85 0.730
= 0.205 + 2.335 = 2.54 ksf
(5) Vesic Method. For φ’= 26 deg, Nq = 11.85 and N γ = 12.54 from Table 4-4.
Nc is not needed. From Table 4-6,
4-28
EM 1110-1-1905
30 Oct 92
qu = 0.5 B γ’
H Nγ ζ γ + σ’
D Nq ζ q
= 0.5 2.00 0.081 12.54 0.361 + 0.27 11.85 0.804
= 0.367 + 2.572 = 2.94 ksf
Total Net
Method qu, ksf q’u, ksf
4-29
EM 1110-1-1905
30 Oct 92
4-30
EM 1110-1-1905
30 Oct 92
CHAPTER 5
DEEP FOUNDATIONS
(a) Lateral expansion and rebound of adjacent soil into the bored hole may
decrease pore pressures. Heavily overconsolidated clays and shales may weaken and
transfer some load to the shaft base where pore pressures may be positive. Methods
presented in Section I for calculating bearing capacity in clays may be slightly
unconservative, but the FS’s should provide an adequate margin of safety against
overload.
(b) Rebound of soil at the bottom of the excavation and water collecting at
the bottom of an open bore hole may reduce end bearing capacity and may require
construction using slurry.
(c) Drilled shafts tend to be preferred to driven piles as the soil becomes
harder, pile driving becomes difficult, and driving vibrations affect nearby
structures. Good information concerning rock is required when drilled shafts are
carried to rock. Rock that is more weathered or of lesser quality than expected may
require shaft bases to be placed deeper than expected. Cost overruns can be
significant unless good information is available.
(2) Driven Piles. Driven piles are displacement deep foundation elements
driven into the ground causing the soil to be displaced and disturbed or remolded.
Driving often temporarily increases pore pressures and reduces short term bearing
capacity, but may increase long term bearing capacity. Driven piles are often
constructed in groups to provide adequate bearing capacity. Analysis of the bearing
capacity of driven piles and groups of driven piles is given in Section II.
5-1
EM 1110-1-1905
30 Oct 92
(a) Driven piles are frequently used to support hydraulic structures such as
locks and retaining walls and to support bridges and highway overpasses. Piles are
also useful in flood areas with unreliable soils.
(b) Pile driving causes vibration with considerable noise and may interfere
with the performance of nearby structures and operations. A preconstruction survey
of nearby structures may be required.
(c) The cross-section and length of individual piles are restricted by the
capacity of equipment to drive piles into the ground.
(d) Driven piles tend to densify cohesionless soils and may cause settlement
of the surface, particularly if the soil is loose.
(e) Heave may occur at the surface when piles are driven into clay, but a net
settlement may occur over the longterm. Soil heave will be greater in the direction
toward which piles are placed and driven. The lateral extent of ground heave is
approximately equal to the depth of the bottom of the clay layer.
(a) Applied loads should be sufficiently less than the ultimate bearing
capacity to avoid excessive vertical and lateral displacements of the pile or
drilled shaft. Displacements should be limited to 1 inch or less.
5-2
EM 1110-1-1905
30 Oct 92
(a) Side friction often contributes the most bearing capacity in practical
situations unless the base is bearing on stiff shale or rock that is much stiffer
and stronger than the overlying soil.
5-3
EM 1110-1-1905
30 Oct 92
5-2. Vertical Compressive Capacity of Single Shafts. The approximate static load
capacity of single drilled shafts from vertical applied compressive forces is
(5-1a)
(5-1b)
where
5-4
EM 1110-1-1905
30 Oct 92
a. End Bearing Capacity. Ultimate end bearing resistance at the tip may be
given as Equation 4-1 neglecting pile weight Wp
(5-2a)
where
Methods for estimating end bearing capacity and correction factors of Equation 5-2a
should consider that the bearing capacity reaches a limiting constant value after
reaching a certain critical depth. Methods for estimating end bearing capacity from
in situ tests are discussed in Section II on driven piles.
(1) Critical Depth. The effective vertical stress appears to become constant
after some limiting or critical depth Lc, perhaps from arching of soil adjacent to
the shaft length. The critical depth ratio Lc/B where B is the shaft diameter
may be found from Figure 5-3. The critical depth applies to the Meyerhof and
Nordlund methods for analysis of bearing capacity.
(2) Straight Shafts. Equation 5-2a may be simplified for deep foundations
without enlarged tips by eliminating the N γp term
(5-2b)
or
(5-2c)
Equations 5-2b and 5-2c also compensates for pile weight Wp assuming γ p ≈ γ’L.
Equation 5-2c is usually used rather than Equation 5-2b because Nqp is usually
large compared with "1" and Nqp-1 ≈ Nqp. Wp in Equation 5-1 may be ignored when
calculating Qu.
5-5
EM 1110-1-1905
30 Oct 92
Figure 5-3. Critical depth ratio Lc/B (Data from Meyerhof 1976)
(3) Cohesive Soil. The undrained shear strength of saturated cohesive soil
for deep foundations in saturated clay subjected to a rapidly applied load is c =
Cu and the friction angle φ = 0. Equations 5-2 simplifies to (Reese and O’Neill
1988)
(5-3)
where the shape factor ζ cp = 1 and Ncp = 6 [1 + 0.2 (L/Bb)] ≤ 9. The limiting qbu
of 80 ksf is the largest value that has so far been measured for clays. Cu may be
reduced by about 1/3 in cases where the clay at the base has been softened and could
cause local high strain bearing failure. Fr should be 1.0, except when Bb exceeds
about 6 ft. For base diameter Bb > 6 ft,
(5-4)
where
The undrained strength of soil beneath the base Cu is in units of ksf. Equation
5-3 limits qbu to bearing pressures for a base settlement of 2.5 inches. The
undrained shear strength Cu is estimated by methods in Chapter 3 and may be taken
as the average shear strength within 2Bb beneath the tip of the shaft.
(4) Cohesionless Soil. Hanson, Vesic, Vesic Alternate, and general shear
methods of estimating the bearing capacity and adjustment factors are recommended
for solution of ultimate end bearing capacity using Equations 5-2. The Vesic method
requires volumetric strain data ε v of the foundation soil in addition to the
5-6
EM 1110-1-1905
30 Oct 92
effective friction angle φ’. The Vesic Alternate method provides a lower bound
estimate of bearing capacity. The Alternate method may be more appropriate for deep
foundations constructed under difficult conditions, for drilled shafts placed in
soil subject to disturbance and when a bentonite-water slurry is used to keep the
hole open during drilled shaft construction. Several of these methods should be
used for each design problem to provide a reasonable range of the probable bearing
capacity if calculations indicate a significant difference between methods.
(a) Hanson Method. The bearing capacity factors Ncp, Nqp, and N γp and
correction factors ζ cp, ζ qp, and ζ γp for shape and depth from Table 4-5 may be
used to evaluate end bearing capacity using Equations 5-2. Depth factors ζ cd and
ζ qd contain a "k" term that prevents unlimited increase in bearing capacity with
depth. k = tan-1(Lb/B) in radians where Lb is the embedment depth in bearing soil
and B is the shaft diameter. Lb/B ≤ Lc/B, Figure 5-3.
(b) Vesic Method. The bearing capacity factors of Equation 5-2b are
estimated by (Vesic 1977)
(5-5a)
(5-5b)
(5-5c)
(5-5d)
(5-5e)
where
Irr ≈ Ir for undrained or dense soil where ν s ≈ 0.5. Gs may be estimated from
laboratory or field test data, Chapter 3, or by methods described in EM 1110-1-1904.
The shape factor ζ cp = 1.00 and
(5-6a)
(5-6b)
5-7
EM 1110-1-1905
30 Oct 92
where
(5-6c)
(c) Vesic Alternate Method. A conservative estimate of Nqp can be readily
made by knowing only the value of φ’
(5-7)
The shape factor ζ qp may be estimated by Equations 5-6. Equation 5-7 assumes a
local shear failure and hence leads to a lower bound estimate of qbu. A local
shear failure can occur in poor soils such as loose silty sands or weak clays or
else in soils subject to disturbance.
(5-8)
b. Skin Friction Capacity. The maximum skin friction that may be mobilized
along an element of shaft length ∆L may be estimated by
(5-9)
where
Resistance t3o applied loads from skin friction along the shaft perimeter
increases with increasing depth to a maximum, then decreases toward the tip. One
possible distribution of skin friction is indicated in Figure 5-4. The estimates of
skin friction fsi with depth is at best approximate. Several methods of
estimating fsi, based on past experience and the results of load tests, are
described below. The vertical load on the shaft may initially increase slightly
with increasing depth near the ground surface because the pile adds weight which may
not be supported by the small skin friction near the surface. Several of these
methods should be used when possible to provide a range of probable skin friction
values.
5-8
EM 1110-1-1905
30 Oct 92
(1) Cohesive Soil. Adhesion of cohesive soil to the shaft perimeter and the
friction resisting applied loads are influenced by the soil shear strength, soil
disturbance and changes in pore pressure, and lateral earth pressure existing after
installation of the deep foundation. The average undrained shear strength
determined from the methods described in Chapter 3 should be used to estimate skin
friction. The friction angle φ is usually taken as zero.
(a) The soil-shaft skin friction fsi of a length of shaft element may be
estimated by
(5-10)
where
αa = adhesion factor
Cu = undrained shear strength, ksf
5-9
EM 1110-1-1905
30 Oct 92
Local experience with existing soils and load test results should be used to
estimate appropriate αa. Estimates of αa may be made from Table 5-1 in the
absence of load test data and for preliminary design.
TABLE 5-1
(b) The adhesion factor may also be related to the plasticity index PI for
drilled shafts constructed dry by (Data from Stewart and Kulhawy 1981)
(5-11a)
(5-11b)
(5-11c)
where 15 < PI < 80. Drilled shafts constructed using the bentonite-water slurry
should use αa of about 1/2 to 2/3 of those given by Equations 5-11.
(2) Cohesionless Soil. The soil-shaft skin friction may be estimated using
effective stresses with the beta method
(5-12a)
(5-12b)
where
5-10
EM 1110-1-1905
30 Oct 92
(b) Refer to Figure 5-3 to determine the critical depth Lc below which σ’i
remains constant with increasing depth.
(3) CPT Field Estimate. The skin friction fsi may be estimated from the
measured cone resistance qc for the piles described in Table 5-2 using the curves
given in Figure 5-6 for clays and silt, sands and gravels, and chalk (Bustamante and
Gianeselli 1983).
5-11
EM 1110-1-1905
30 Oct 92
TABLE 5-2
a. Drilled Shafts
Cone Resistance
Pile Description Remarks qc, ksf Soil Curve
Drilled shaft Hole bored dry without Tool without teeth; oversize any Clay-Silt 1
bored dry slurry; applicable to blades; remolded soil on
cohesive soil above sides
water table Tool with teeth; immediate > 25 Clay-Silt 2
concrete placement > 94 Clay-Silt 3
any Chalk 1
Immediate concrete placement > 94 Chalk 3
Immediate concrete placement >250 Chalk 4
with load test
Drilled shaft Slurry supports sides; Tool without teeth; oversize any Clay-Silt 1
with slurry concrete placed blades; remolded soil on
through tremie from sides
bottom up displacing Tool with teeth; immediate > 25 Clay-Silt 2
concrete placement > 94 Clay-Silt 3
any Sand-Gravel 1
Fine sands and length < 100 ft >104 Sand-Gravel 2
Coarse gravelly sand/gravel >156 Sand-Gravel 3
and length < 100 ft
Gravel > 83 Sand-Gravel 4
any Chalk 1
Above water table; immediate > 94 Chalk 3
concrete placement
Above water table; immediate >250 Chalk 4
concrete placement with
load test
5-12
EM 1110-1-1905
30 Oct 92
Cone Resistance
Pile Description Remarks qc, ksf Soil Curve
b. Driven Piles
Cone Resistance
Pile Description Remarks qc, ksf Soil Curve
5-13
EM 1110-1-1905
30 Oct 92
Cone Resistance
Pile Description Remarks qc, ksf Soil Curve
5-14
EM 1110-1-1905
30 Oct 92
Cone Resistance
Pile Description Remarks qc, ksf Soil Curve
5-15
EM 1110-1-1905
30 Oct 92
(a) The mean effective vertical stress in a soil layer σ’s such as in a
sand layer below a surface layer, Figure 5-7, may be estimated by
(5-13a)
where
The mean effective vertical stress in the sand layer adjacent to the embedded pile
from Equation 5-13a is
(5-13b)
5-16
EM 1110-1-1905
30 Oct 92
(b) Laboratory strength tests indicate that the average undrained shear
strength of the clay is Cu = 2 ksf. Cone penetration tests indicate an average
cone tip resistance qc in the clay is 40 ksf and in the sand 160 ksf.
(c) Relative density of the sand at the shaft tip is estimated from
Equation 3-5
The effective friction angle estimated from Table 3-1a is φ’ = 38 deg, while
Table 3-1b indicates φ’ = 36 to 38 deg. Figure 3-1 indicates φ’ = 38 deg. The
sand appears to be of medium to dense density. Select a conservative φ = 36 deg.
Coefficient of earth pressure at rest from the Jaky Equation 5-6c is Ko = 1 - sin φ
= 1 - sin 36 deg = 0.42.
(d) The sand elastic modulus Es is at least 250 ksf from Table D-3 in EM
1110-1-1904 using guidelines for a medium to dense sand. The shear modulus Gs is
estimated using Gs = Es/[2(1 + υs)] = 250/[2(1 + 0.3)] = 96 or approximately 100
ksf. Poisson’s ratio of the sand ν s = 0.3.
(2) End Bearing Capacity. A suitable estimate of end bearing capacity qbu
for the pile tip in the sand may be evaluated from the various methods in 5-2a for
cohesionless soil as described below. Hanson and Vesic methods account for a
limiting effective stress, while the general shear method and Vesic alternate method
ignore this stress. The Vesic Alternate method is not used because the sand appears
to be of medium density and not loose. Local shear failure is not likely.
(a) Hansen Method. From Table 4-4 (or calculated from Table 4-5),
Nqp = 37.75 and N γp = 40.05 for φ’ = 36 deg. From Table 4-5,
5-17
EM 1110-1-1905
30 Oct 92
(b) Vesic Method. The reduced rigidity index from Equation 5-5c is
where
(5-5e)
(5-5d)
The shape factor ζ qp = 1.00 when using Equation 5-8. From Equation 5-2c,
Hansen 214
Vesic 89
General Shear 113
5-18
EM 1110-1-1905
30 Oct 92
The Hansen result of 214 ksf is much higher than the other methods and should be
discarded without proof from a load test. The Vesic and General Shear methods give
an average value qbu = 102 ksf.
(a) Cohesive Soil. The average skin friction from Equation 5-10 is
where αa was estimated from Equation 5-11b, αa = 0.9 - 0.01 40 = 0.5 or 0.55 from
Table 5-1. Skin friction from the top 5 ft should be neglected.
(c) CPT Field Estimate. The shaft was bored using bentonite-water slurry.
Use curve 2 of clay and silt, Figure 5-6a, and curve 3 of sand and gravel,
Figure 5-6b. From these figures, fs of the clay is 1.5 ksf and fs of the sand is
2.0 ksf.
(d) Comparison of Methods. Skin friction varies from 1.0 to 1.5 ksf for the
clay and 0.5 to 2 ksf for the sand. Skin friction is taken as 1 ksf in the clay and
1 ksf in the sand, which is considered conservative.
(3) Total Capacity. The total bearing capacity from Equation 5-1a is
Qu = Qbu + Qsu - Wp
where
5-19
EM 1110-1-1905
30 Oct 92
n 2
Qsu = Σ Qsui = Cs ∆L Σ fsi
i=1 i=1
sand clay
Qsu = π B [∆L fs + ∆L fs)
= π 1.5 [15 1 + 10 1)] = 118 kips
(c) Total Capacity. Inserting the end bearing and skin resistance
bearing capacity values into Equation 5-1a is
(4) Allowable Bearing Capacity. The allowable bearing capacity from Equation
1-2b is
Qu 292
Qa = = = 97 kips
FS 3
d. Load Tests for Vertical Capacity. ASTM D 1143 testing procedures for
piles under static compression loads are recommended and should be performed for
individual or groups of vertical and batter shafts (or piles) to determine their
response to axially applied compression loads. Load tests lead to the most
efficient use of shafts or piles. The purpose of testing is to verify that the
actual pile response to compression loads corresponds to that used for design and
that the calculated ultimate load is less than the actual ultimate load. A load
cell placed at the bottom of the shaft can be used to determine the end bearing
resistance and to calculate skin friction as the difference between the total
capacity and end bearing resistance.
(1) Quick Load Test. The "Quick" load test option is normally satisfactory,
except that this test should be taken to plunging failure or three times the design
load or 1000 tons, whichever comes first.
(2) Cost Savings. Load tests can potentially lead to significant savings in
foundation costs, particularly on large construction projects when a substantial
part of the bearing capacity is contributed by skin friction. Load tests also
assist the selection of the best type of shaft or pile and installation depth.
(3) Lower Factor of Safety. Load tests allow use of a lower safety factor of
2 and can offer a higher allowable capacity.
5-20
EM 1110-1-1905
30 Oct 92
(4) Scheduling of Load Tests. Load tests are recommended during the design
phase, when economically feasible, to assist in selection of optimum equipment for
construction and driving the piles in addition to verifying the bearing capacity.
This information can reduce contingency costs in bids and reduce the potential for
later claims.
(a) Load tests are recommended for most projects during early construction to
verify that the allowable loads used for design are appropriate and that
installation procedures are satisfactory.
(b) Load tests during the design phase are economically feasible for large
projects such as for multistory structures, power plants, locks and dams.
(c) When load tests are performed during the design phase, care must be taken
to ensure that the same procedures and equipment (or driving equipment including
hammer, helmet, cushion, etc. in the case of driven piles) are used in actual
construction.
(a) A dial attached to the rod with the stem on the reference beam, Fig-
ure 5-8b, measures the downward movement of the piston. A dial attached to the
pressure pipe measures the upward movement of the pile base. A third dial attached
to the reference beam with stem on the pile top measures the movement of the pile
top. The difference in readings between the top and bottom of the pile is the
elastic compression due to side friction. The total side friction force can be
estimated using Young’s modulus of the pile.
(b) If the pile is tested to failure, the measured force at failure (piston
downward movement is continuous with time or excessive according to guidance in
Table 5-3) is the ultimate end bearing capacity. The measured failure force in the
downward plunging piston therefore provides a FS > 2 against failure considering
that the skin friction capacity is equal to or greater than the end bearing
capacity.
(c) This test can be more economical and completed more quickly than a
conventional load test; friction and end bearing resistance can be determined
separately; optimum length of driven piles can be determined by testing the same
pile at successfully deeper depths. Other advantages include ability to work over
water, to work in crowded and inaccessible locations, to test battered piles, and to
check pullout capacity as well as downward load capacity.
5-21
EM 1110-1-1905
30 Oct 92
(6) Analysis of Load Tests. Table 5-3 illustrates four methods of estimating
ultimate bearing capacity of a pile from data that may be obtained from a load-
displacement test such as described in ASTM D 1143. At least three of these
methods, depending on local experience or preference, should be used to determine a
suitable range of probable bearing capacity. The methods given in Table 5-3 give a
range of ultimate pile capacities varying from 320 to 467 kips for the same pile
load test data.
5-3. Capacity to Resist Uplift and Downdrag. Deep foundations may be subject to
other vertical loads such as uplift and downdrag forces. Uplift forces are caused
by pullout loads from structures or heave of expansive soils surrounding the shaft
tending to drag the shaft up. Downdrag forces are caused by settlement of soil
surrounding the shaft that exceeds the downward displacement of the shaft and
increases the downward load on the shaft. These forces influence the skin friction
that is developed between the soil and the shaft perimeter and influences bearing
capacity.
5-22
EM 1110-1-1905
30 Oct 92
TABLE 5-3
πB2
Limit Value 1. Draw a line with slope Ep
4L
(Davisson where B = diameter
1972) of pile, inches;
Ep = Young’s pile
modulus, kips/inch2;
L = pile length, inches
5-23
EM 1110-1-1905
30 Oct 92
ρ
80 Percent 1. Plot load test results as
Q
(Hansen vs. ρ
1963)
2. Draw straight line through
data points
5. Ultimate deflection is
ρu = b/a
5-24
EM 1110-1-1905
30 Oct 92
evaluating vertical capacity for uplift and downdrag loads are given in Reese and
O’Neill (1988).
(1) Force Distribution. Deep foundations may resist pullout forces by shaft
skin resistance and resistance mobilized at the tip contributed by enlarged bases
illustrated in Figure 5-9. The shaft resistance is defined in terms of negative
skin friction fn to indicate that the shaft is moving up relative to the soil.
This is in contrast to compressive loads that are resisted by positive skin friction
where the shaft moves down relative to the soil, Figure 5-4. The shaft develops a
tensile stress from pullout forces. Bearing capacity resisting pullout may be
estimated by
(5-14a)
(5-14b)
where
(2) End Bearing Resistance. Enlarged bases of drilled shafts resist pullout
and uplift forces. qbu may be estimated using Equation 5-2c. Base area Ab
resisting pullout to be used in Equation 5-1b for underreamed drilled shafts is
(5-15)
where
Bb = diameter of base, ft
Bs = diameter of shaft, ft
(b) Cohesionless Soil. Nqp of Equation 5-2 can be obtained from Equation
5-7 of the Vesic alternate method where ζ qp is given by Equations 5-6.
5-25
EM 1110-1-1905
30 Oct 92
(3) Skin Resistance. The diameter of the shaft may be slightly reduced from
pullout forces by a Poisson effect that reduces lateral earth pressure on the shaft
perimeter. Skin resistance will therefore be less than that developed for shafts
subject to compression loads because horizontal stress is slightly reduced. The
mobilized negative skin friction fni may be estimated as 2/3 of that evaluated for
compression loads fsi
(5-16a)
(5-16b)
where
Cs = shaft circumference, ft
∆L = length of pile element i, ft
fsi = positive skin friction of element i from compressive loading
using Equations 5-10 to 5-12
5-26
EM 1110-1-1905
30 Oct 92
subject to heave. This soil is often within the top 7 to 20 ft of the soil profile
referred to as the depth of the active zone for heave Za. The shaft located within
Za is sometimes constructed to isolate the shaft perimeter from the expansive soil
to reduce this uplift thrust. The shaft base and underream resisting uplift should
be located below the depth of heaving soil.
(1) Force Distribution. The shaft moves down relative to the soil above
neutral point n, Figure 5-10, and moves up relative to the soil below point n.
The negative skin friction fn below point n and enlarged bases of drilled shafts
resist the uplift thrust of expansive soil. The positive skin friction fs above
point n contributes to uplift thrust from heaving soil and puts the shaft in
tension. End bearing and skin friction capacity resisting uplift thrust may be
estimated by Equations 5-14.
(2) End Bearing. End bearing resistance may be estimated similar to that for
pullout forces. Ncp should be assumed to vary from 0 at the depth of the active
zone of heaving soil to 9 at a depth 2.5Bb below the depth of the active zone of
heave. The depth of heaving soil may be at the bottom of the expansive soil layer
or it may be estimated by guidelines provided in TM 5-818-7, EM 1110-1-1904, or
McKeen and Johnson (1990).
(3) Skin Friction. Skin friction from the top of the shaft to the neutral
point n contributes to uplift thrust, while skin friction from point n to the
base contributes to skin friction that resists the uplift thrust.
5-27
EM 1110-1-1905
30 Oct 92
(1) Force Distribution. The shaft moves up relative to the soil above point
n, Figure 5-11, and moves down relative to the soil below point n. The positive
skin friction fs below point n and end bearing capacity resists the downward
loads applied to the shaft by the settling soil and the structural loads. Negative
skin friction fn above the neutral point contributes to the downdrag load and
increases the compressive stress in the shaft.
(2) End Bearing. End bearing capacity may be estimated similar to methods
for compressive loads given by Equations 5-2.
(3) Skin Friction. Skin friction may be estimated by Equation 5-9 where the
positive skin friction is given by Equations 5-10 to 5-12.
(1) Load-Transfer Principle. Vertical loads are transferred from the top of
the shaft to the supporting soil adjacent to the shaft using skin friction-load
transfer functions and to soil beneath the base using consolidation theory or base
load-transfer functions. The total bearing capacity of the shaft Qu is the sum of
the total skin Qsu and base Qbu resistances given by Equations 5-1.
5-28
EM 1110-1-1905
30 Oct 92
(b) The program should be used to provide a minimum and maximum range for the
load-displacement behavior of the shaft for given soil conditions. A listing of
AXILTR is provided to allow users to update and calibrate this program from results
of field experience.
(2) Base Resistance Load Transfer. The maximum base resistance qbu in
Equation 5-1b is computed by AXILTR from Equation 5-2a
(5-17)
where
c = cohesion, psf
Ncp = cohesion bearing capacity factor, dimensionless
Nqp = friction bearing capacity factor, dimensionless
σ’
L = effective vertical overburden pressure at the pile base, psf
5-29
EM 1110-1-1905
30 Oct 92
TABLE 5-4
L
None Pullout QDL - P Straight: Qsur + Wp Qsur = πBs ∫ f-
sdL
0
Underream: Smaller of L
Qsub = πBb ∫ τ sdL
Qsub + Wp 0
Qsur + Qbur + Wp π
Qbur =b q-bu B2s) (B2
4
π
Wp s =L γ p B2
4
Ln
Swelling Uplift Qus Straight: Qsur + Wp Qus = πBs ∫ f-
sdL
soil thrust 0
Underream: L
πBs ∫ f-
Qsur =sdL
Qsur + Qbur + Wp Ln
π
Qbur =b q-bu B2s) (B2
4
Ln
Settling Downdrag Qd + Qsud Qsur + Qbu
Qsud = πBs ∫ fndL
soil
0
L
πBs ∫ f-
qsur =sdL
Ln
Nomenclature:
Bb Base diameter, ft QDL Dead load of structure, pounds
Bs Shaft diameter, ft P Pullout load, pounds
f-s Maximum mobilized shear strength, psf Qsub Ultimate soil shear resistance of
fn Negative skin friction, psf cylinder diameter Bb and length equal
L Shaft length, ft to depth of underream, pounds
Ln Length to neutral point n, ft Qsud Downdrag, pounds
qbu Ultimate base resistance, psf Qsur Ultimate skin resistance, pounds
Qbu Ultimate base capacity, pound Qus Uplift thrust, pounds
Qbur Ultimate base resistance of upper Qd Design load, Dead + Live loads, pounds
portion of underream, pounds Wp Shaft weight, pounds
τs Soil shear strength, psf γp Unit shaft weight, pounds/ft3
Correction factors ζ are assumed unity and the N γp term is assumed negligible.
Program AXILTR does not limit σ’
L.
(a) Nqp for effective stress analysis is given by Equation 5-7 for local
shear (Vesic Alternate method) or Equation 5-8 for general shear.
5-30
EM 1110-1-1905
30 Oct 92
Ncp for total stress analysis is assumed 9 for general shear and 7 for local
shear; Nqp and total stress friction angle φ are zero for total stress analysis.
(c) End bearing resistance may be mobilized and base displacements computed
using the Reese and Wright (1977) or Vijayvergiya (1977) base load-transfer
functions, Figure 5-12a, or consolidation theory. Ultimate base displacement for
the Reese and Wright model is computed by
(5-18)
where
The ultimate base displacement for the Vijayvergiya model is taken as 4 percent of
the base diameter.
(4) Skin Resistance Load Transfer. The shaft skin friction load-transfer
functions applied by program AXILTR are the Seed and Reese (1957) and Kraft, Ray,
and Kagawa (1981) models illustrated in Figure 5-12b. The Kraft, Ray, and Kagawa
model requires an estimate of a curve fitting constant R from
(5-19a)
5-31
EM 1110-1-1905
30 Oct 92
5-32
EM 1110-1-1905
30 Oct 92
where
(5-19b)
where νs is the soil Poisson’s ratio. A good value for νs is 0.3 to 0.4.
(a) Load-transfer functions may also be input into AXILTR for each soil layer
up to a maximum of 11 different functions. Each load-transfer function consists of
11 data values consisting of the ratio of the mobilized skin friction/maximum
mobilized skin friction fs/f-s correlated with displacement as illustrated in
Figure 5-12b. The maximum mobilized skin friction f- s is assumed the same as the
maximum soil shear strength. The corresponding 11 values of the shaft displacement
(or shaft movement) in inches are input only once and applicable to all of the load-
transfer functions. Therefore, the values of fs/f- s of each load transfer function
must be correlated with the given shaft displacement data values.
5-33
EM 1110-1-1905
30 Oct 92
(b) Swell or settlement occurs depending on the difference between the input
initial void ratio and the final void ratio determined from the swell and
compression indices, the swell pressure, and the final effective stress for each
soil element. The method used to calculate soil swell or settlement of soil
adjacent to the shaft is described as Method C of ASTM D 4546.
(c) The depth of the active zone Za is required and it is defined as the
soil depth above which significant changes in water content and soil movement can
occur because of climate and environmental changes after construction of the
foundation. Refer to EM 1110-1-1904 for further information.
5-4. Lateral Load Capacity of Single Shafts. Deep foundations may be subject to
lateral loads as well as axial loads. Lateral loads often come from wind forces on
the structure or inertia forces from traffic. Lateral load resistance of deep
foundations is determined by the lateral resistance of adjacent soil and bending
moment resistance of the foundation shaft. The ultimate lateral resistance Tu
often develops at lateral displacements much greater than can be allowed by the
structure. An allowable lateral load Ta should be determined to be sure that the
foundation will be safe with respect to failure.
a. Ultimate Lateral Load. Brom’s equations given in Table 5-5 can give good
results for many situations and these are recommended for an initial estimate of
ultimate lateral load Tu. Ultimate lateral loads can be readily determined for
complicated soil conditions using a computer program such as COM624G based on beam-
column theory and given p-y curves (Reese, Cooley, and Radhakkrishnan 1984). A p-y
curve is the relationship between the soil resistance per shaft length (kips/inch)
and the deflection (inches) for a given lateral load T.
5-34
EM 1110-1-1905
30 Oct 92
TABLE 5-5
Brom’s Equations for Ultimate Lateral Load (Broms 1964a, Broms 1964b, Broms 1965)
Short
L ≤ Lc (5-22a)
(5-22b)
Long
L ≥ Lc
(5-22c)
(5-24a)
Short
L ≤ Lc
(5-24b)
Long (5-24c)
L ≥ Lc
5-35
EM 1110-1-1905
30 Oct 92
(5-23a)
Short
L ≤ Lcs
(5-23b)
Inter-
mediate (5-23c)
Lcs ≤ L
L ≥ Lcl
(5-23d)
(5-23e)
Long
L ≥ Lcl
(5-25a)
Short
L ≤ Lcs
(5-25b)
5-36
EM 1110-1-1905
30 Oct 92
(5-25c)
Inter-
mediate
Lcs ≤ Lcl
L ≥ Lcl (5-25d)
(5-25e)
Long
L ≥ Lcl
e. Nomenclature
Bs = diameter of pile shaft, ft
Cu = undrained shear strength, kips/ft2
c = distance from centroid to outer fiber, ft
e = length of pile above ground surface, ft
1.5Bs + f = distance below ground surface to point of maximum bending moment in cohesive soil, ft
f = distance below ground surface at point of maximum bending moment in cohesionless soil, ft
fy = pile yield strength, ksf
Ip = pile moment of inertia, ft4
Kp = Rankine coefficient of passive pressure, tan2(45 + φ’/2)
L = embeded pile length, ft
Lc = critical length between long and short pile, ft
Lcs = critical length between short and intermediate pile, ft
Lcl = critical length between intermediate and long pile, ft
Ma = applied bending moment, positive in clockwise direction, kips-ft
My = ultimate resisting bending moment of entire cross-section, kips-ft
T = lateral load, kips
Tu = ultimate lateral load, kips
Tul = ultimate lateral load of long pile in cohesionless soil, kips
Z = section modulus Ip/c, ft3
Zmax = maximum section modulus, ft3
Zmin = minimum section modulus, ft3
γ = unit wet weight of soil, kips/ft3
φ’ = effective angle of internal friction of soil, degrees
5-37
EM 1110-1-1905
30 Oct 92
(1) Considerations.
(a) Lateral load failure may occur in short drilled shafts and piles, which
behave as rigid members, by soil failure and excessive pile deflection and in long
piles by excessive bending moment.
(5-26)
where
Shafts which carry insignificant axial loads such as those supporting overhead signs
can be placed at this minimum depth if their lateral load capacity is acceptable.
(c) Cyclic loads reduce the support provided by the soil, cause gaps to
appear near the ground surface adjacent to the shaft, increase the lateral
deflection for a given applied lateral load and can reduce the ultimate lateral load
capacity because of the loss of soil support.
(d) Refer to ASTM D 3966 for details on conducting lateral load tests.
(2) Broms’ Closed Form Solution. Broms’ method uses the concept of a
horizontal coefficient of subgrade reaction and considers short pile failure by flow
of soil around the pile and failure of long piles by forming a plastic hinge in the
pile. Refer to Broms (1964a), Broms (1964b), Broms (1965), and Reese (1986) for
estimating Tu from charts.
(3) Load Transfer Analysis. The method of solution using load transfer p-y
curves is also based on the concept of a coefficient of horizontal subgrade
reaction. A fourth-order differential equation is solved using finite differences
and load transfer p-y curves.
5-38
EM 1110-1-1905
30 Oct 92
(b) Program COM624G has provided excellent agreement with experimental data
for many load test results.
(1) Minimum and maximum values of the expected soil modulus of subgrade
reaction should be used to determine a probable range of lateral load capacity.
This modulus may be estimated from results of pressuremeter tests using the Menard
deformation modulus (Reese 1986), estimates of the elastic soil modulus with depth,
or values given in Table 5-6b.
(5-27)
5-39
EM 1110-1-1905
30 Oct 92
TABLE 5-6
e 1/4
Short Free 4T (1 + 1.5 ) E
u L sl
Head y = β =
o E L c 4E I
β L < 1.5 sl p p
c
E = pile lateral elastic modulus, ksf
p
Short Fixed T I = pile moment of inertia, ft4
u p
β L < 0.5 y =
c o E L E = modulus of subgrade reaction, ksf
sl sl
Stiff 1 - 2 3 - 6
Very Stiff 2 - 4 6 - 13
Long Fixed T β
u c Hard > 4 > 13
Head y =
o E
β L > 1.5 sl
c
1/5
3 E I
T β p p
u β =
y = F k
o y E I
p p
k = constant relating elastic soil modulus with depth,
E = kz, kips/ft3
s
Representative Values for k
2 3
C , kips/ft k, kips/ft
u
Static Cyclic
0.25 - 0.5 50 20
0.50 - 1.0 170 70
1.0 - 2.0 500 200
2.0 - 4.0 1700 700
4.0 - 8.0 5000 2000
Values for F
y
L
F
β y
2 1.13
3 1.03
4 0.96
5 0.93
5-40
EM 1110-1-1905
30 Oct 92
TABLE 5-7
Calculation of EpIp:
Using ACI Code Equation 10.8 (approximate)
5-41
EM 1110-1-1905
30 Oct 92
(2) Ultimate Lateral Load. Broms equations in Table 5-5a for a free head
pile in cohesive soil may be used to roughly estimate Tu. The ultimate bending
moment resistance My using data in Table 5-7 is
This shaft with L = 20 ft is considered long. From Equation 5-27c, the ultimate
lateral load Tu is
(3) Allowable Lateral Load. From Table 5-6b, the ultimate lateral deflection
yo is
or 0.26 inch. On the basis of Equation 5-27 the design displacement will be
(10/68.4) 0.26 or 0.04 inch, which is less than the specified allowable deflection
ya = 0.25 inch. The trial dimensions are expected to be fully adequate to support
the design lateral load of 10 kips. Additional analysis using COM624G should be
performed to complete a more economical and reliable design.
5-5. Capacity of Shaft Groups. Drilled shafts are often not placed in closely
spaced groups because these foundations can be constructed with large diameters and
can extend to deep depths. The vertical and lateral load capacities of shaft
foundations are often the sum of the individual drilled shafts. The FS for groups
should be 3.
5-42
EM 1110-1-1905
30 Oct 92
a. Axial Capacity. The axial capacity of drilled shafts spaced ≥ 8Bs will
be the sum of the capacities of individual shafts. If drilled shafts are
constructed in closely spaced groups where spacing between shafts is < 8Bs, then the
capacity of the group may be less than the sum of the capacities of the individual
shafts. This is because excavation of a hole for a shaft reduces effective stresses
against both the sides and bases of shafts already in place. Deep foundations where
spacings between individual piles are less than 8 times the shaft width B also
cause interaction effects between adjacent shafts from overlapping of stress zones
in the soil, Figure 5-13. In situ soil stresses from shaft loads are applied over a
much larger area leading to greater settlement and bearing failure at lower total
loads.
where (5-28a)
5-43
EM 1110-1-1905
30 Oct 92
Eg should be > 0.7 for spacings > 3Bs and 1.0 for spacings > 8Bs. Eg should vary
linearly for spacings between 3Bs and 8Bs. Eg = 0.7 for spacings ≤ 2.5Bs.
(5-28b)
where
L = depth of penetration, ft
W = horizontal length of group, ft
B = horizontal width of group, ft
Cua = average undrained shear strength of the cohesive soil in which the
group is placed, ksf
Cub = average undrained shear strength of the cohesive soil below the
tip to a depth of 2B below the tip, ksf
The presence of locally soft soil should be checked because this soil may cause some
shafts to fail.
(c) The ultimate bearing capacity of a group in a strong clay soil overlying
weak clay may be estimated by assuming block punching through the weak underlying
soil layer. Group capacity may be calculated by Equation 5-28b using the undrained
strength Cub of the underlying weak clay. A less conservative solution is
provided by (Reese and O’Neill 1988)
(5-29)
where
5-44
EM 1110-1-1905
30 Oct 92
(1) Widely Spaced Drilled Shafts. Shafts spaced > 7Bs or far enough apart
that stress transfer is minimal and loading is by shear, the ultimate lateral load
of the group Tug is the sum of individual shafts. The capacity of each shaft may
be estimated by methodology in 5-4.
(2) Closely Spaced Drilled Shafts. The solution of ultimate lateral load
capacity of closely spaced shafts in a group requires analysis of a nonlinear soil-
shaft system
(5-30)
where
Refer to Poulos (1971a, Poulos (1971b), and Reese (1986) for detailed solution of
the lateral load capacity of each shaft by the Poulos-Focht-Koch method.
(3) Group Behavior as a Single Drilled Shaft. A pile group may be simulated
as a single shaft with diameter Cg/π where Cg is the circumference given as the
minimum length of a line that can enclose the group. The moment of inertia of the
group is n Ip where Ip is the moment of inertia of a single shaft. Program
COM624G may be used to evaluate lateral load-deflection behavior of the simulated
single shaft for given soil conditions. Comparison of results between the Poulos-
Focht-Koch and simulated single pile methods was found to be good (Reese 1986).
5-6. Effects of Pile Driving. Driving piles disturbs and substantially remolds
soil. Driving radially compresses cohesive soils and increases the relative density
of cohesionless soils near the pile.
a. Cohesive Soil. Soil disturbance around piles driven into soft or normally
consolidated clays is within one pile diameter. Driving into saturated stiff clays
closes fissures and causes complete loss of stress history near the pile (Vesic
1969).
5-45
EM 1110-1-1905
30 Oct 92
(1) Driving in Dense Sand and Gravel. Driving in dense sand and gravel can
decrease pore pressures from soil dilation and temporarily increase soil shear
strength and pile load capacity. Shear strength can increase substantially and may
exceed the capacity of pile driving equipment to further drive the piles into the
soil. Pore pressures increase after driving and cause the shear strength to
decrease and reduce the pile load capacity. This effect is soil relaxation.
5-7. Vertical Capacity of Single Driven Piles. The vertical capacity of driven
piles may be estimated using Equations 5-1 similar to drilled shafts
(5-1a)
(5-1b)
where
5-46
EM 1110-1-1905
30 Oct 92
required capability of the driving equipment and to establish pile driving criteria.
Refer to program GRLWEAP (Goble Rausche Likins and Associates, Inc. 1988) for
details of wave equation analysis. Pile driving formulas are also recommended to
quickly estimate the ultimate bearing capacity.
(5-2a)
where
Equation 5-2a may be simplified for driven piles by eliminating the N γp term
(5-2b)
or
(5-2c)
Equations 5-2b and 5-2c also adjust for pile weight Wp assuming γ p ≈ γ’L.
Equation 5-2c is usually used because omitting the "1" is usually negligible.
Bearing capacity does not increase without limit with increasing depth. Refer to
Figure 5-3 to determine the critical depth Lc below which effective stress remains
constant using the Meyerhof and Nordlund methods.
(1) Cohesive Soil. The shear strength of cohesive soil is c = Cu, the
undrained strength, and the effective friction angle φ’ = 0. Equation 5-2a leads
to
(5-2d)
5-47
EM 1110-1-1905
30 Oct 92
(a) Meyerhof Method. Figure 5-15 illustrates the bearing capacity factors to
be used with Equation 5-2b (Meyerhof 1976). The range between "low" and "high"
factors in Figure 5-15 should account for soil conditions such as loose or dense
sands, overconsolidation ratio of clays, and soils with different degrees of
compressibility. The correction factors ζ cp and ζ qp in Equation 5-2b are unity.
Ncp and Nqp are estimated as follows:
1. If φ’ < 30° and L > Lc/2, then use Ncp,high and Nqp,high
directly from curves of Figure 5-15
(5-31a)
(5-31b)
(5-31c)
Refer to Vanikar (1986) for further applications using the Meyerhof method.
(b) Nordlund Method. This semi-empirical method considers the shape of the
pile taper and the influence of soil displacement on skin friction. Equations for
calculating ultimate capacity are based on load test results that include timber,
steel H, pipe, monotube, and Raymond steptaper piles. Ultimate capacity is
determined by, Figure 5-16
5-48
EM 1110-1-1905
30 Oct 92
5-49
EM 1110-1-1905
30 Oct 92
(5-32a)
where
φ’ may be estimated from Table 3-1. Point resistance qbu = αfNqpσ’LAp should not
exceed q Ap where q is given by Equation 5-31c. αf and Nqp may be found from
Figure 5-17, K from Figure 5-18, δ from Figure 5-19 for a given φ’ and Cf may
be found from Figure 5-20. Equation 5-32a for a pile of uniform cross-section (ω =
0) and length L driven in a homogeneous soil with a single friction angle φ and
single effective unit weight is
(5-32b)
where A is the pile cross-section area, Cs is the pile perimeter and σ’ m is the
mean effective vertical stress between the ground surface and pile tip, ksf. Table
5-8 provides a procedure for using the Nordlund method (Data from Vanikar 1986).
5-50
EM 1110-1-1905
30 Oct 92
5-51
EM 1110-1-1905
30 Oct 92
5-52
EM 1110-1-1905
30 Oct 92
5-53
EM 1110-1-1905
30 Oct 92
TABLE 5-8
Step Procedure
Add Qsui of each soil layer to obtain Qsu, Qsu = ∑ Qsui of each
layer
14 Compute ultimate total capacity, Qu = Qbu + Qsu
5-54
EM 1110-1-1905
30 Oct 92
(3) Field Estimates From In Situ Soil Tests. The ultimate end bearing
capacity of soils may be estimated from field tests if laboratory soil or other data
are not available.
(a) SPT Meyerhof Method. End bearing capacity may be estimated from
penetration resistance data of the SPT by (Meyerhof 1976)
(5-33)
where NSPT is the average uncorrected blow count within 8Bb above and 3Bb below
the pile tip. Lb is the depth of penetration of the pile tip into the bearing
stratum. qbu is in units of ksf.
(b) CPT Meyerhof method. End bearing capacity may be estimated from
cone penetration resistance data by (Meyerhof 1976)
(5-34)
(c) CPT B & G method. End bearing capacity may also be estimated from cone
penetration resistance data by (Bustamante and Gianeselli 1983)
(5-35)
where
TABLE 5-9
k
c
Soil Driven Pile Drilled Shaft
Clay - Silt 0.600 0.375
Sand - Gravel 0.375 0.150
Chalk 0.400 0.200
(d) CPT 1978 FHWA-Schmertmann method (modified). End bearing capacity may be
estimated by (Schmertmann 1978)
(5-36)
5-55
EM 1110-1-1905
30 Oct 92
where qc1 and qc2 are unit cone resistances determined by the procedure described
in Figure 5-21. For example, qc1 calculated over the minimum path is as follows:
(4) Scale Effects. Ultimate end bearing capacity qbu tends to be less for
larger diameter driven piles and drilled shafts than that indicated by Equations 5-
33 or 5-34 or Equation 5-2b using Equations 5-31 to estimate Ncp or Nqp (Meyerhof
1983). Skin friction is independent of scale effects.
(a) Sands. The reduction in end bearing capacity has been related with a
reduction of the effective angle of internal friction φ’ with larger diameter deep
foundations. End bearing capacity qbu from Equation 5-2 should be multiplied by a
reduction factor Rbs (Meyerhof 1983)
(5-37a)
5-56
EM 1110-1-1905
30 Oct 92
for B > 1.64 ft. The exponent m = 1 for loose sand, 2 for medium dense sand, and
3 for dense sand.
(b) Clays. The reduction factor Rbc appears related to soil structure and
fissures. For driven piles in stiff fissured clay, Rbc is given by Equation 5-37a
where m = 1. For bored piles
(5-37b)
(a) Alpha method. The skin friction of a length of pile element may be
estimated by
(5-10)
where
αa = adhesion factor
Cu = undrained shear strength, ksf
Local experience with existing soils and load test results should be used to
estimate appropriate αa. Estimates of αa may be made from Table 5-10 in the
absence of load test data and for preliminary design.
(5-38a)
where
5-57
EM 1110-1-1905
30 Oct 92
TABLE 5-10
5-58
EM 1110-1-1905
30 Oct 92
(2) Cohesionless Soil. The soil-shaft skin friction may be estimated using
effective stresses
(5-12a)
(5-12b)
where
(b) The effective vertical stress σ’i approaches a limiting stress at the
critical depth Lc, then remains constant below Lc. Lc may be estimated from
Figure 5-3.
(3) CPT Field Estimate. The skin friction fsi may be estimated from the
measured cone resistance qc for the piles described in Table 5-2b using the curves
given in Figure 5-6 for clay and silt, sand and gravel, and chalk (Bustamante and
Gianeselli 1983).
5-59
EM 1110-1-1905
30 Oct 92
(1) Description. The pile driving and soil system consists of a series of
elements supported by linear elastic springs and dashpots which have assumed
parameters, Figure 5-23. Characteristics of commonly used pile hammers and piles
are available in the program data files driving systems. Input parameters include
the dynamic damping constants for each dashpot, usually in units of seconds/inch,
ultimate soil resistance Qu in kips and quake in fractions of an inch for each
spring. Each dashpot and spring represent a soil element. The quake of the pile is
its displacement at Qub. Input data for Qub, quake, and ultimate skin resistance
of each element Qsui are usually assumed. Actual load distribution data are
5-60
EM 1110-1-1905
30 Oct 92
normally not available and require results of instrumented load tests. Standard
values are available in the user’s manual for soil input parameters.
(2) Analysis. The wave equation analysis provides a relationship between the
pile capacity and the driving resistance in blows per inch (or blows per foot if
needed). This relationship can be developed for different pile lengths and then
used in the field when the pile has been driven sufficiently to develop the required
capacity. The wave equation can also be used to develop relationships between
driving stresses in the pile and penetration resistance for different combinations
of piles and pile driving equipment.
(3) Application. The wave equation analysis is used to select the most
suitable driving equipment to ensure that the piles can develop the required
capacity and select the minimum pile section required to prevent overstressing the
pile during driving.
d. Pile Driving Formulas. Pile driving formulas, Table 5-11, although not as
good as wave equation analysis, can provide useful, simple estimates of ultimate
pile capacity Qu and they can be obtained quickly. The allowable bearing capacity
can be estimated from Equations 1-2 using FS in Table 5-11. Two or more of these
methods should be used to provide a probable range of Qu.
(a) The mean effective vertical stress in the sand layer adjacent to the
embedded pile σ’s and at the pile tip σ’
L is limited to 1.8 ksf for the Meyerhof
and Nordlund methods. Otherwise, σ’ L is 2.4 ksf from Equation 5-13b.
5-61
EM 1110-1-1905
30 Oct 92
TABLE 5-11
1/2
12EhEr 144EhErL
Danish , Cd = , inches 3 - 6
S + Cd 2AEp
12Wrh
Engineering Drop Hammers: 6
S + 1.0
News Record
24Er
Other Hammers: 6
S + 0.1
Nomenclature:
A = area of pile cross-section, ft2
Eh = hammer efficiency
Ep = pile modulus of elasticity, ksf
Er = manufacturer’s hammer-energy rating (or Wrh), kips-ft
h = height of hammer fall, ft
L = pile length, inches
S = average penetration in inches per blow for the last 5 to 10 blows
for drop hammers and 10 to 20 blows for other hammers
Wr = weight of striking parts of ram, kips
Wp = weight of pile including pile cap, driving shoe, capblock and
anvil for double-acting steam hammers, kips
(b) The average undrained shear strength of the upper clay layer is Cu = 2
ksf. The friction angle of the lower sand layer is estimated at φ’ = 36 deg. Cone
penetration test results shown in Figure 5-21 indicate an average cone tip
resistance qc = 40 ksf in the clay and 160 ksf in the sand.
(2) End Bearing Capacity. A suitable estimate of end bearing capacity qbu
for the pile tip in the sand may be evaluated from the various methods for
cohesionless soil as described below.
5-62
EM 1110-1-1905
30 Oct 92
(b) Nordlund Method. The procedure in Table 5-8a may be used to estimate end
bearing capacity.
(c) Hansen Method. From Table 4-5 (or calculated from Table 4-4) Nqp =
37.75 and N γ p = 40.05 for φ’ = 36 deg. From Table 4-5,
(d) Vesic Method. The reduced rigidity index from Equation 5-5c is
Ir 57.3
Irr = = = 42.6
1 + ε v Ir 1 + 0.006 57.3
where
1 - 2 µs σ’ 1 - 2 0.3 2.4
(Equation 5-5e): εv = L
= = 0.006
2(1 - µs) Gs 2(1 - 0.3) 100
Gs 100
(Equation 5-5d): Ir = = = 57.3
σ’
L tanφ’ 2.4 tan36
5-63
EM 1110-1-1905
30 Oct 92
3 0.685 0.494
Nqp = e 3.852 42.6
3-0.588
Nqp = 1.244 1.984 3.852 6.382 = 60.7
19.475
Nqp = = 47.24
0.412
The shape factor ζ qp = 1.00 when using Equation 5-8. From Equation 52c,
qbu = σ’L Nqp ζ qp = 2.4 47.24 1.00 = 113.4 ksf
qbu = kc qc
5-64
EM 1110-1-1905
30 Oct 92
(h) CPT FHWA & Schmertmann. The data in Figure 5-21 can be used with this
method to give qbu = 162.9 ksf as in the example illustrating this method in
paragraph 5-7a.
Meyerhof 124
Nordlund 97
Hansen 214
Vesic 89
General Shear 113
CPT Meyerhof 124
CPT B & G 60
CPT FHWA & Schmertmann 163
These calculations indicate qbu from 60 to 214 ksf. Discarding the highest
(Hansen) and lowest (CPT B & G) values gives an average qbu = 118 ksf. Scale
effects of Equations 5-37 are not significant because B < 1.64 ft.
(a) Cohesive Soil. The average skin friction using the Alpha method
from Equation 5-10 is
where αa = 1.2 - 0.3Cu = 0.6 from Table 5-11 and L/B < 20. The average skin
friction using the Lambda method from Equation 5-38a and using L = 15 ft for the
penetration of the pile only in the clay is
A reasonable average value of skin friction is 1.4 ksf for the clay.
(b) Cohesionless Soil. The average skin friction from Equation 5-12a
using σ’
s limited to 1.8 ksf is
5-65
EM 1110-1-1905
30 Oct 92
where βf is found from Figure 5-5 using φ’ = 36 deg. The Nordlund method of
Table 5-8b provides an alternative estimate
(c) CPT Field Estimate. The driven pile is described as "steel" from
Table 5-2b. Curve 1 of Clay and Silt, Figure 5-6a, and curve 1 of Sand and Gravel,
Figure 5-6b, should be used. From these figures, fs of the clay is 0.7 ksf and
fs of the sand is 0.7 ksf.
(d) Comparison of Methods. Skin friction varies from 0.7 to 1.4 ksf
for the clay and 0.7 to 1.6 ksf for the sand. Skin friction is taken as 1.0 ksf in
the clay and 1 ksf in the sand.
(3) Ultimate Total Capacity. The total bearing capacity from Equation 5-1a
is
Qu = Qbu + Qsu - Wp
where
πB2 π 1.52 150
Wp = L γ conc = 30 = 8 kips for the pile weight
4 4 1000
n 2
Qsu = Σ Qsui = Cs ∆L Σ fsi
i=1 i=1
5-66
EM 1110-1-1905
30 Oct 92
(c) Inserting end bearing and skin resistance bearing capacity values into
Equation 5-1a,
The minimum and maximum values of qbu and fs calculated above could be used to
obtain a range of Qu if desired.
(4) Allowable Bearing Capacity. The allowable bearing capacity from Equation
1-2b using FS = 3 is
Qu 337
Qa = = = 112 kips
FS 3
5-8. Lateral Load Capacity of Single Piles. Evaluation of lateral load capacity is
treated similarly to that for single drilled shafts in 5-4. Lateral load capacity
may be determined by load tests, by analytical methods such as Broms’ equations or
p-y curves and by arbitrary values. Most piles are placed in groups where group
capacity controls performance.
a. Load Tests. Lateral load tests are economically justified for large
projects and may be performed as described in ASTM D 3966.
b. Analytical Methods.
(1) Program COM624G using p-y curves are recommended for complicated soil
conditions.
(2) Broms’ equations in Table 5-5 can give useful estimates of ultimate
lateral loads for many cases.
c. Arbitrary Values.
5-9. Capacity of Pile Groups. Driven piles are normally placed in groups with
spacings less than 8 times the pile diameter or width 8Bs and joined at the ground
surface by a concrete slab referred to as a pile cap. The capacity of the pile
group can be greater than the sum of the capacities of the individual piles because
driving compacts the soil and can increase skin friction and end bearing capacity.
FS for pile groups should be 3.
5-67
EM 1110-1-1905
30 Oct 92
TABLE 5-12
Recommendations for Allowable Lateral Pile Loads (Data from Vanikar 1986)
Allowable Allowable
Pile Deflection, in. Lateral Load, kips Reference
(1) Optimum Spacing. Piles in a group should be spaced so that the bearing
capacity of the group ≥ sum of the individual piles. Pile spacings should not be
less than 2.5Bs. The optimum pile spacing is 3 to 3.5Bs (Vesic 1977) or greater
than 0.02L + 2.5Bs where L is the pile length in feet (Canadian Geotechnical
Society 1985).
(3) Cohesionless Soil. Group capacity should be taken as the sum of the
individual piles.
(1) Widely Spaced Piles. Where piles are spaced > 7Bs or far enough apart
that stress transfer is minimal and loading is by shear, the ultimate lateral load
of the group Tug is the sum of individual piles. The capacity of each pile may be
estimated by methodology in 5-4.
5-68
EM 1110-1-1905
30 Oct 92
(2) Closely Spaced Piles. The solution of ultimate lateral load capacity of
closely spaced pile groups require analysis of a nonlinear soil-pile system. Refer
to Poulos (1971a), Poulos (1971b), and Reese (1986) for detailed solution of the
lateral load capacity of each pile by the Poulos-Focht-Koch method.
(1) Program CPGA. Pile Group Analysis computer program CPGA is a stiffness
analysis in three-dimensions assuming linear elastic pile-soil interaction and a
rigid pile cap (Hartman et al 1989). Program CPGA uses matrix methods to
incorporate position and batter of piles and piles of different sizes and materials.
Computer program CPGG displays the geometry and results of program CPGA (Jaeger,
Jobst, and Martin 1988).
(2) Program CPGC. Pile Group Concrete computer program CPGC develops the
interaction diagrams and data required to investigate the structural capacity of
prestressed concrete piles (Strom, Abraham, and Jones 1990).
5-69