22 - Bearing Capacity of Shallow

Source: GEOTECHNICAL ENGINEERING
22 Bearing Capacity of Shallow

Foundations
22.1 OVERVIEW
22.1.1 The Functional Foundation

In civil engineering a foundation links a structure with the material on which it
rests. The foundation is designed to transmit the weight of the structure, plus
effects of live loads and wind loads, to the underlying material so that it is not
stressed beyond its safe bearing capacity. If the supporting material is hard rock,
design is simplified, but construction can become more costly because of the
requirement for heavy ripping or blasting. Usually, however, the depth to rock is
such that the foundation rests on soil.
In this chapter we will deal with shallow foundations, also called spread footings
because they spread the load. If the soil is not competent to support the
anticipated load, the soil can be reinforced, or the load can be carried deeper to
more competent soil or rock by means of piles or piers. The soil or rock support
for a foundation is variously referred to as the foundation soil or foundation bed.
22.1.2 Site Evaluation

Site evaluation for a small or inexpensive structure may only involve on-site
inspection to determine the presence of unstable slopes, landslides, limestone
sinks, mine cave-ins, abandoned wells, ravines, trash, pollution, seepage, ponded
water, hog confinements, or other more or less obvious deficiencies.
Less obvious but still exceedingly important are clues that can reveal the presence
of expansive clays—vertical shrinkage cracks in the soil, heaved and distressed
pavements, or cracks in nearby structures. Reference should be made to available
documents including soil maps and reports. Hand borings or probings are made
as needed for soil identification and evaluation.
638 Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
Copyright © 2007 The McGraw-Hill Companies. All rights reserved.
Any use is subject to the Terms of Use as given at the website.
Bearing Capacity of Shallow Foundations
Bearing Capacity of Shallow Foundations 639
Many engineers now are reluctant to make such judgments without detailed on-
site borings and tests because of liability issues. All moderately heavy to heavy
structures require exploration borings, and many loan agencies require borings
and a soil report before approving loans. Soil borings and on-site tests are
discussed in Chapter 26.
Regardless of the detail of the investigation, a final step is to inspect open

basement and footing excavations for buried topsoil or fill that were not
encountered in the borings and which should be removed prior to placing
concrete. Any loose soil that may have fallen into open foundation excavations
must be scooped out, and the firmness of the bare foundation soil can quickly be
verified with a hand probe or penetrometer. Covering up the evidence literally sets
the stage for future complications.
22.1.3 Building Codes

Many cities and other organizations such as state highway departments have
established tables of safe bearing capacities on the basis of experience with the
local soils. The Boston Building Code is an example (Table 22.1).
Building codes are a valuable guide, but are oversimplified. For example, they
typically specify a value or a narrow range of values for safe bearing capacity for
each class of soil without giving any consideration to width, shape, and depth of a
foundation, or to the proximity to other structures, and even the relation to a
groundwater table. The foundation width is particularly important because wider
foundations confine the soil and allow a higher bearing pressure. Building codes
therefore may include an option that design can be performed by a registered
professional engineer.
22.2 FOUNDATION BEARING CAPACITY
22.2.1 Two Kinds of Failures

The most ruinous type of foundation failure results when the underlying soil
shears and is displaced laterally, allowing the foundation to sink into the ground.
This is a classic bearing capacity failure. Fortunately, it is rare. The concept is
relatively straightforward, downward and sideways. The supporting ability of a
soil depends in part on the depth of the foundation; the deeper the foundation, the
greater the ability. The overloaded foundation is like a boat, sinking deeper until
it finds its stable depth. The distinction from a boat is because soils are much
denser than water and possess shear strength. Also eccentric loading and
variations in shear strength inevitably cause the structure to tip one way or the
other. We can call this a Type A failure. Because the soil is shearing, it becomes
weaker through remolding.
Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

640 Geotechnical Engineering
Table 22.1 Allowable bearing

Allowable bearing Class of material pressure, tons/ft2 (kPa)
pressures of 1. Massive igneous rocks and conglomerate, all in
foundation sound condition (sound condition allows minor cracks) 100 (9600)
materials 2. Slate in sound condition (minor cracks allowed) 50 (4800)
(Sec. 725, Boston 3. Shale in sound condition (minor cracks allowed) 10 (960)a
Building Code, 4. Residual deposits of shattered or broken
1970) bedrock of any kind except shale 10 (960)
5. Glacial till 10 (960)
6. Gravel, well-graded sand and gravel 5 (480)
7. Coarse sand 3 (290)
8. Medium sand 2 (190)
9. Fine sand 1 to 2 (95 to 190)b
10. Hard clay 5 (480)
11. Medium clay 2 (190)c
12. Soft clay 1 (95)c
13. Inorganic silt, shattered shale, or any natural
deposit of unusual character not provided for herein b
14. Compacted granular fill 2 to 5 (190 to 480)b

b
15. Preloaded materials
Note: aThe allowable bearing pressures given in this table are intended to prevent bearing
capacities from exceeding the soil shear strength. However, these allowable bearing pressures
do not ensure that the settlements will be within the tolerable limits for a given structure.
b
Value to be fixed by the building official.
c
Alternatively, the allowable bearing pressure shall be computed from the unconfined
compressive strength of undisturbed samples, and shall be taken as 1.5 times that strength
for round and square footings, and 1.25 times that strength for footings with length-width
ratios of greater than four (4); for intermediate ratios interpolation may be used.
The Type B failure also involves sinking but it is slow and not as catastrophic.
This is excessive and/or differential settlement, which are far more common. In a
Type B failure the soil consolidates and becomes stronger instead of shearing and
becoming weaker.
An example of Type B is the famous Leaning Tower, which started to lean soon
after construction began, so the builders compensated for the lean by using
thicker courses of stone on the low side. The correction temporarily relieved the
off-center loading, but meanwhile soil underneath the low side had consolidated
more than under the high side and was slightly stiffer, so tilting pursued in another
direction. The stonemasons played into the game like a juggler balancing a
broomstick, so the Tower zig-zagged upward. When it was completed there could
be no more corrections, so tilting continued and then slowly speeded up as the
load went more off-center. Had the soil become weaker through shearing instead
of stronger through consolidation, the Tower would have fallen over.

22.2.2 Factors Influencing Type A Bearing Capacity

Bearing capacity failures are relatively rare except when an earthquake tem-
porarily liquefies a sandy or silty foundation soil. During a 1964 earthquake in
Niigata, Japan, a cluster of multistory apartment buildings sank and tipped to
such a high angle that the occupants reached safety by crawling out of their
windows and walking down the outside walls.
Another contributor to low shearing strength is excess pore water pressure

caused by rapid loading of a saturated, poorly draining foundation soil. As the
load goes on and pore water pressure goes up, frictional strength goes down in
accordance with the concept of effective stress. This is unlikely to occur where
loading is restricted by a small allowable settlement and therefore is more
common where larger settlements can be tolerated, as under road embankments
or grain bins. Grain bins commonly are clustered on a single concrete pad,
so settlement can be kept on an even keel using the same method as was used for
the Tower, distributing more load to the high side.
A famous example of a bearing capacity failure from development of excess pore

water pressure is the Tranconia grain elevator in Canada, which tipped about 458.
The story does not end there because after the silos were emptied they were
righted by underpinning!
Bearing capacity failures also occur under road, railroad, or dam embankments
if the rate of loading exceeds the capacity of the soil to drain away excess pore
water pressure. In one case a heavily loaded train making a slow first run over a
new railroad track slowed down and laid over on its side as the result of a bearing
capacity failure.
Excess pore water pressures are predictable from laboratory consolidation and
triaxial tests, and are readily measured in the field with wellpoints, more formally
referred to as piezometers. The first stage of failure may be creep, but unlike a
landslide where stress remains relatively constant, as a building or other structure
tips, soil stresses increase under the low side, so a failure can occur with little
warning.
22.2.3 Analytical Approach to Bearing Capacity

For mathematical convenience bearing capacity is considered a two-dimensional
problem with a unit dimension normal to the page. The analyses therefore directly
apply to strip footings, and adjustments are required for them to be applied to
square, rectangular, or round footings.
The general procedure is to define an ultimate bearing capacity, which is the

anticipated failure load, and a safe bearing capacity, which is the ultimate capacity
divided by a factor of safety. The usual factor of safety based on load is 3 to 5, but

as will be shown, the corresponding factor of safety in relation to the soil strength
is considerably lower.
The net bearing capacity is the bearing pressure minus that of soil that has been
removed to make the foundation excavation. The net bearing capacity equals the
increase in pressure on the soil at the foundation depth. Bearing capacity is
determined at the base of a foundation and therefore includes the weight of the
foundation element itself. However, because the unit weight of concrete is only
moderately higher than that of the soil removed to make room for the concrete,
the small amount of increase in load usually is ignored.
Many bearing capacity theories have been proposed, and reflect the backgrounds
and biases of the theorists.
22.2.4 Rankine’s Analysis

Rankine viewed bearing capacity in terms of active and passive pressures on
two sides of a retaining wall. He then left out the wall (Fig. 22.1). The foundation
load induces active pressure on one side of the imaginary wall, which in turn
induces passive pressure on the other. The active wedge moves downward and
sideways, and the passive wedge moves upward and sideways, which is consistent
with observations that the ground surface bulges up adjacent to a bearing capacity
failure. The development is as follows, and it does not matter whether, as shown
in the figure, the failure is one-sided or is symmetrical.
Passive side : h ¼ DKp

Active side : qo ¼ h Kp
Substituting for h,
qo ¼ DK2p
where h is the horizontal stress, is the unit weight of the surcharge soil outside
of the foundation, D is the depth of the surcharge, qo is the bearing capacity
pressure at failure, and Kp is the passive coefficient. Substituting the value
Figure 22.1
Rankine’s bearing
capacity theory
uses a simplified
linear failure
geometry and
ignores vertical
shearing
resistance along
vertical surfaces.
for Kp gives
qo ¼ DK2p ð22:1Þ

1 þ sin 2
qo ¼ D ð22:1aÞ
1 þ sin
This relatively simple equation shows the influence of foundation depth, which
is represented by D, but does not include soil cohesion. However, if D ¼ 0 and
there is no cohesion, as in a dry sand, a bearing pressure will sink until D is large
enough for the load to be supported.
Example 22.1
A 100 lb (0.445 kN) person wearing high heels is attempting to walk on a dry sand
beach, so you try to explain the situation. ¼ 258 and ¼ 90 lb/ft3 (14.1 kN/m3). The area
of each heel is 0.25 in.2 (160 mm2). How far must a heel sink into the sand to reach
equilibrium?
Answer: For this friction angle K2p ¼ 6:1. The bearing pressure is qo ¼ 100 lb 7 0.25 in.2 ¼
400 lb/in.2, or 0.445 kN 7 160 mm2 ¼ 0.00278 kN/mm2. Then
3
lb
lb 1
ft
400 2
¼ 90 3 D 6:1
in: ft 1728 in:3
D ¼ 1260 in. ¼ 105 ft, or
kN
kN
13
m
0:00278 ¼ 14:1 D 6:1
mm:2 3
m ð1000 mmÞ3
D ¼ 32 m. And that is why one should not attempt to wear high heels at the
beach.
22.2.5 Critique of Rankine’s Theory

Rankine assumed straight-line boundaries for the shearing soil, which is on the
unsafe side if a curved shear surface is more critical. The model also suggests
vertical shearing stress along the principal plane separating active and passive
wedges, and by definition there can be no shearing stress on a principle plane.
A wedge of soil is pushed down but a larger wedge is pushed up, which does
not balance out. Rankine’s analysis is not reliable for design, but did succeed in
pointing out zones of active and passive pressures that became a guide for
subsequent analyses.
22.2.6 Bell’s Equation

Bell added a cohesion term to Rankine’s analysis, it can be visualized as two
companion triaxial (actually plane strain) test specimens, one oriented vertically
and the other laid on its side, as shown in Fig. 22.2. The horizontally oriented
specimen on the right has a confining stress 3 equal to the surcharge pressure D,
Figure 22.2
Bell’s model is
similar to
Rankine’s but
includes cohesion.
While Bell’s
equation is useful
as a quick check,
the angular slip
surfaces may give
results that are on
the unsafe side.
where is the soil unit weight and D is the depth of the foundation. At failure this
value equals 1 for the left-hand specimen. Mohr circles representing this concept
are shown at the bottom of the figure. Based on this relationship, in 1915
Bell presented the following equation:
qo ¼ D tan2 ð45 þ =2Þ þ 2c tanð45 þ =2Þ½tan2 ð45 þ =2Þ þ 1 ð22:2Þ
where c is the soil cohesion.
A test for any equation is to consider end conditions. If ¼ 0, Bell’s equation

becomes
qo ¼ D þ 4c ð22:2aÞ
This relationship is shown by Mohr circles in Fig. 22.3 and also is known as the
Casagrande-Fadum illustration. If in addition D ¼ 0,
qo ¼ 4c ð22:2bÞ
These relationships apply to saturated clay soils with rapid loading so that there
is no time for drainage, for example under a moving wheel load.
If the unconfined compressive strength is regarded as an undrained test, as

shown by the left-hand Mohr circle in Fig. 22.3, q ¼ 2c, and
qo ¼ 2qu ð22:2cÞ
where qu is the unconfined compressive strength. This equation indicates that
if the unconfined compressive strength is used as an estimate for bearing capacity,
as sometimes is the case for small jobs, the factor of safety is 2.

Figure 22.3
An adaptation of Bell’s model with ¼ 0 equates bearing capacity qo to two
times the unconfined compressive strength qu. A surcharge shifts the Mohr
circles to the right and increases qo by the amount of the surcharge.
Bell’s model is subject to the same objections as Rankine’s with regard to the
linear slip surface, and ignores vertical shear and the unit weight of soil below the
foundation level. Results can provide a simple check on other more sophisticated
analyses.
22.2.7 Critical Edge Pressure and Progressive Failure

Theoretically, under the edge of a foundation as the shearing area decreases to
zero, vertical shearing stress becomes infinite—except of course the soil shears
first. O. K. Fröhlich, a professor of civil engineering at the Technical College in
Vienna and a colleague of Terzaghi, applied elastic theory to map zones of shear
failure in soil under the edges of a foundation. This development defines a plastic
zone that expands as the foundation pressure increases, as shown in Fig. 22.4.
As the plastic zones continue to expand under increasing pressure they coalesce
in the middle, in effect describing a progressive failure.
If the friction angle is zero, Fröhlich’s solution becomes

qo ¼ 3:14c ð22:3Þ
where c is the soil cohesion.
22.2.8 Positive Gains from Perimeter Shear

Housel (1956), at the University of Michigan, performed plate bearing tests with
different diameters of plates, and discovered that the smaller the plate, the higher
the bearing pressure at failure. This conflicts with the more widely accepted
Terzaghi theory that is discussed below, which indicates the opposite. Housel
suggested that there is an additional resistance to loading as a result of shearing
resistance at the edges. However, plate load test areas are small compared
with foundation areas, and therefore magnify the perimeter effect. Housel’s
Figure 22.4
As shearing stress is infinite at the edges of a foundation, Fröhlich applied elastic theory and
theorized progressive development of a plastic zone as a load approaches that will cause
a bearing capacity failure. This sketch is for a cohesive soil. (Modified after Jumikis, 1969).
observation nevertheless provides some useful insight into interpretation of plate

load test results that without correction will be on the unsafe side for a foundation
bearing capacity.
Housel proposed an empirical relationship based on plate bearing test results:

W ¼ Pm þ An ð22:4Þ
where W is the total foundation load, P is the perimeter, A is the bearing area, and
m and n are constants. Dividing by A gives
P
qo ¼ mþn ð22:5Þ
A
where q is the bearing capacity. For a square or circular bearing area P/A ¼ 4/B
where B is the plate width. Then
4m
qo ¼ þ n ¼ perimeter effect þ area pressure ð22:6Þ
B
In order to separately evaluate perimeter and area effects, tests are required using
two or more different plate sizes. If a bearing capacity is obtained by means other
than a plate load test, no correction is made because the influence of perimeter
shear diminishes with increasing size of a bearing area. Perimeter shear was
observed in model experiments by Vesic (1975), and became the basis for a
bearing capacity theory based on vertical punching proposed by Janbu.
22.2.9 Loss of Strength from Perimeter Shear

Terzaghi also recognized that perimeter shear must occur, and became concerned
about the reduction in strength that occurs durng shearing, particularly in a
sensitive soil. The amount of strength lost is difficult to predict as it varies

depending on the soil sensitivity and extent of shearing, and Terzaghi suggested an
arbitrary one-third reduction in c and in tan for foundations on sensitive clay.
He also applied the same reduction to loose sands because perimeter shear can
lead to excessive settlement.
22.2.10 When a Plate Bearing Test Should Not Be

Relied upon for Design
A plate bearing test essentially is a model footing, but the model is distorted
because the soil properties are not appropriately scaled. (This is the basis for
conducting bearing tests in a centrifuge.) There are two reasons why plate tests
can be unreliable for design: (1) a small plate emphasizes the role of perimeter
shear, discussed above, but more importantly, (2) the depth of influence is limited
by the small diameter of the plate. The latter effect is shown in Fig. 22.5.
A typical example where a plate load test would give misleading results derives
from the tendency for a clay soil to dry out and gain apparent cohesive strength at
the ground surface. This objection can be overcome by testing in a pit, but costs
go up, and if the test depth is deeper than the footing depth the influence of
surcharge pressure becomes exaggerated.
The same objection to the plate load test applies to settlement predictions, as the
depth of the pressure bulb is substantially reduced by the small plate width
Figure 22.5
A plate load test
does not
accurately model a
much larger
foundation,
particularly if there
are changes in the
soil with depth.

compared with that of a foundation. The most common use of plate bearing tests
is to measure the modulus of pavement elements.
22.3 TERZAGHI BEARING CAPACITY THEORY
22.3.1 Prandtl-Terzaghi Theory

Again demonstrating a benefit from cross-fertilization of ideas, Terzaghi’s bearing
capacity theory derives from a mathematical analysis for penetration of metal by
a punch. Prandtl was a professor at the University of Göttingen in Germany
who is most famous for his contributions to flow patterns in aerodynamics.
He adapted a similar flow pattern to plastic deformation by defining a curved
transition zone to link Rankine’s active and passive zones, and noted that it is
possible to do so without corrupting principal stresses with shear stresses if the
shape is that of a logarithmic spiral. The contours depend on the friction angle
(Fig. 22.6(a)): the higher the friction angle, the wider the spiral and the higher
the bearing capacity.
Terzaghi modified Prandtl’s theory by including a surcharge load (Fig. 22.6(b)).

The Prandtl geometry assumes a smooth base for the punch so that the bottom
is a principle plane, and Teraaghi hypothesized that because soil tends to expand
laterally in accordance with Poisson’s ratio, confining friction should be mobilized
along the base. This in turn should change the shape of the shear curves, as shown
in Fig. 22.6(b). By acting to reduce lateral expansion, the influence of base
roughness should tend to increase the bearing capacity even though the shear
areas are larger.
The mathematical solution is complex, and Terzaghi took a critical step to make
the equations usable by engineers. The simplified equation defines ‘‘bearing
capacity factors’’ that depend on the angle of internal friction and are compiled
into tables. Terzaghi (1943) emphasized that the solution is not exact, and that it
requires a generous factor of safety. He also suggested the arbitrary one-third
reduction in bearing capacity for sensitive soils.
Terzaghi’s theory has been modified by later workers but still is the most
commonly used method for predicting shallow foundation bearing capacity.
The equation is as follows:
B
qo ¼ N þ cNc þ DNq
2 ð22:7Þ
ðqo ¼ width factor þ cohesion factor þ surcharge or depth factorÞ
where qo is the bearing capacity in force per unit area, is the soil unit weight,
B is the foundation width, c is soil cohesion, D is depth of the foundation, and the

Figure 22.6
Bearing capacity
failure geometries:
(a) Prandtl-
Terzaghi smooth
base; (b) Terzaghi
rough base;
(c) Meyerhoff.
N values are bearing capacity factors. An important feature of the equation is that
it shows respective contributions from foundation width, cohesion, and surcharge,
as indicated in the line under the equation.
Terzaghi bearing capacity factors are given in Table 22.2 and depend only on the
soil friction angle and whether the base is smooth or rough. The equations for Nc
and Nq are analytical and exact and may be programmed, but the equations for
N were determined by curve fitting and comparisons to values determined by
testing. While commercial programs are available for calculating bearing capacity,

Table 22.2 Terzaghi rough base Smooth base

Bearing capacity
Nb Ncc Ndq Ne Ncc Nfq
factors for use in
eq. (22.7) 0 0.0 5.7 1.0 0.0 5.1 1.0
5 0.5 7.3 1.6 0.5 6.5 1.6
10 1.2 9.6 2.7 1.2 8.3 2.5
15 2.5 13 4.4 2.6 11 3.9
20 5.0 18 7.4 5.4 15 6.4
25 9.7 25 13 11 21 11
30 20 37 22 22 30 18
35 42 58 41 48 46 33
40 100 96 81 110 75 64
45 300 170 170 270 130 130
Local shear:
0 0.0 5.7 1.0 0.0 5.1 1.0
5 0.3 6.7 1.4 0.3 6.0 1.3
10 0.7 8.0 1.9 0.7 7.0 1.8
15 1.2 9.6 2.7 1.2 8.3 2.5
20 2.0 12 3.8 2.1 10 3.4
25 3.2 14 5.3 3.4 12 4.6
30 5.0 18 7.4 5.4 15 6.4
B
Note: aq0 ¼ N þ cNc þ DNq .
2
b
After Bowles (1968). Approximate equation fit by the authors: N ¼ 1.1 (Nq 1)
tan 1.3.
c
Nc ¼ (Nq 1) cot (exactly).
d eð1:5Þ tan
Nq ¼ ðexactlyÞ:
2 cos2 ð45 þ =2Þ
e
Equation fit by Vesic (1975): N ¼ 2(Nq þ 1) tan .
f
Nq ¼ etan tan2 (45 þ /2) (exactly).
it is a simple matter to make a spreadsheet based on these values or their

approximate formulas.
Table 22.2 also gives Terzaghi bearing capacity factors that are arbitrarily reduced
for local shear, to be used for sensitive clays.
Example 22.2
A clay soil has ¼ 108, c ¼ 19.2 kPa (400 lb/ft2), and ¼ 15.7 kN/m3 (100 lb/ft3). (a) Assume
a rough footing surface and evaluate the relative influences of footing width and depth on
supporting capacity of a footing 1.22 m (4 ft) wide and 1.52 m (5 ft) deep. (b) What would be
the effect of removing the surcharge on one side of the footing? What would be the influence
from using a smooth footing?

Answer: (a) From Table 22.2 for a rough footing N ¼ 1.2, Nc ¼ 9.6, and Nq ¼ 2.7. Therefore
B
qo ¼ N þ cNc þ DNq ð22:7Þ
2
qo ¼ ½ð15:7 kN=m3 Þð1:22 mÞ 1:2 þ ð19:2 kN=m2 Þ 9:6 þ ð15:7 kN=m3 Þð1:52Þ 2:7
¼ 11:5 þ 184 þ 64:4 ¼ 260 kN=m2 or
qu ¼ ½ð100 lb=ft3 Þð4 ftÞ 1:2 þ ð400 lb=ft2 Þ 9:6 þ ð100 lb=ft3 Þð5Þ 2:7
¼ 240 þ 3840 þ 1350 ¼ 5430 lb=ft2
It will be noted that these are foundation pressures per unit length of the foundation,
and must be multiplied by the width to obtain a supporting capacity in lb/ft or kN/m.
The relative proportions of the bearing pressure attributed to foundation width, soil
cohesion, and foundation depth are 4.4%, 71%, and 25%, respectively.
(b) Removing the surcharge simply means dropping the third term in the equation, which
will reduce the ultimate bearing capacity by 25%.
(c) Substituting bearing capacity factors for a smooth footing gives
qo ¼ ½ð100Þð4Þ 1:2 þ 400 8:3 þ ð100Þð5Þ 2:5 ¼ 4810 lb=ft2 ð230 kPaÞ
a reduction of 12%.
22.3.2 Pore Water Pressure

Calculations normally are on an effective stress basis—that is, the friction
angle and cohesion used in the formulas are adjusted to compensate for any
development of excess pore water pressure. This is done on the Mohr-Coulomb
shear diagram.
Example 22.3
The soil in the above example is slightly overconsolidated. Assume that prior to shearing
Ko ¼ 0.4 and the Bishop pore pressure parameter is 0.5 (section 18.9.8). Recalculate the
bearing capacity.
Answer: The simplified Bishop pore pressure equation states

u ¼ Að1 3 Þ þ 3 ð18:21Þ
Making appropriate substitutions,
u ¼ ð0:5Þð1 0:41 Þ þ 0:41 ¼ 0:71

That is, unless loading is very slow or drainage is enhanced, 70% of the bearing pressure
is predicted to go to pore water pressure, which should substantially reduce the bearing
capacity. Terzaghi’s effective stress equation states
f0 ¼ c0 þ ðn uÞ tan 0
ð18:7Þ
¼ c0 þ 0:3ðn Þ tan 0
It can be argued that the cohesion term should be relatively unaffected, in which case the
friction angle used in the bearing capacity equation should be reduced by an amount that

gives the reduced tangent, or

¼ tan1 ð0:3 tan 10 Þ ¼ 3
This value is used to determine new bearing capacity factors, which will be quite low.
Drainage can be aided if the foundation is placed on a layer of sand that is allowed to freely
drain.
23.3.3 Allowable Bearing Capacity

Equation (22.7) gives a bearing capacity at failure, and must be divided by a factor
of safety to obtain an allowable bearing capacity that is designated by qa. The
factor based on load-carrying capacity normally is 3 to 5. This may appear to be
extremely conservative but, as will be shown, it is less conservative than it looks.
22.3.4 Meyerhof’s Modification

The Prandtl-Terzaghi geometry assumes that the ground surface adjacent to
a loaded foundation is a principal plane so that the passive wedge rises at an angle
is 45 /2 with horizontal. Observations of actual failures in the field in model
tests indicate that the surface is not flat but curves upward as shown in
Fig. 22.6(c). Meyerhof (1951), at the Technical University of Nova Scotia,
recalculated the bearing capacity factors assuming that the log spiral continues to
the ground surface, and also varied the basal triangle to obtain the most critical
shear angle. The base angle thus determined is about 1.2 compared with for
Terzaghi rough base. While there is only a modest effect on bearing capacity
factors, Meyerhoff ’s analysis probably depicts a more realistic model for what
actually happens.
22.3.5 Eccentric Loading

Although foundations ideally should be symmetrically loaded, there are situations
where eccentric loading is unavoidable, such as for the foundation for a retain-
ing wall. Asymmetrical loading also can occur where a foundation is next to
a property line, or is a combined footing supporting two loads such as a column
load and a wall load.
With reference to Fig. 22.7, eccentricity e is defined as the distance from the
centroid of a bearing area to the central axis of load. As can be seen in the figure,
when that distance exceeds B/6, where B is the footing width, the contact area is
reduced, and if e ¼ 0.5B, the load is centered over the edge of the footing so that
the contact area in effect becomes a line. Using that as a guide, Hansen suggested
that the footing width be reduced as follows for use in bearing capacity
calculations:
B0 ¼ B 2e ð22:8Þ

Example 22.4
An MSE wall is 0.6 times as thick as it is high, and the friction angle of soil behind
the wall is 258. Assume that the unit weight of the soil in the wall and behind
the wall are the same. What is the eccentricity of the foundation load and the effective
width?
Answer: From the friction angle Ka ¼ 0.4. With reference to Fig. 22.7, summing moments
around O gives
0:6H2 e ¼ ð0:4H2 =2ÞðH=3Þ
e ¼ 0:11H; where B ¼ 0:6H and e ¼ 0:185B
B0 ¼ B 2ð0:185BÞ ¼ 0:44B
This will reduce the width factor in the bearing capacity equation.
Example 22.5
The foundation soil in the above example is silt with ¼ 258 and when saturated, c ¼ 0.
¼ 15.7 kN/m3 (100 lb/ft3). The height of the wall is 20 ft (6.1 m). What is the factor of
safety for foundation bearing capacity?
Answer: The average bearing pressure is q ¼ 6.1(15.7) ¼ 95.8 kN/m2 (2000 lb/ft2).
B
qo ¼ N þ cNc þ DNq
2
¼ ½ð15:7Þð0:6Þð6:1Þ 9:7 þ 0 þ 0
¼ 279 kN=m2
The factor of safety without considering the eccentric loading therefore is FS ¼
279/95.8 ¼ 2.9, which is marginally acceptable. However, if the eccentric loading is taken
into consideration, B0 ¼ 0.44B and qo ¼ 0.44(279) ¼ 123 kN/m2. This reduces the factor of
safety to 123/95.8 ¼ 1.3, which is not acceptable and could invite a progressive failure.
22.3.6 Footing on a Slope

It is easy to assume—incorrectly—that a bearing capacity calculated according to
the above considerations also applies to footings located on a slope. However, as
Figure 22.7
Example of
eccentric
foundation loading
by an MSE wall.

can be seen in Fig. 22.8, an important effect of a slope is to take away surcharge
on the downhill side. There also is a reduced length of the slip surface, which
affects the cohesion term. Modified bearing capacity factors were calculated by
Bowles (1988) assuming that the slope surface is a principal plane. As a guide, the
surcharge term can be omitted and a more generous factor of safety adopted.
Another alternative is to analyze the system as a slope stability problem using a
method of slices, discussed in the following section.
22.3.7 Progressive Failure and Upper- and

Lower-Bound Solutions
All of the above solutions approximate ‘‘slip-line solutions’’ based on limit equi-
librium, or the maximum load that can be sustained prior to failure. All of the
stated geometries are approximations because they do not include the influence
of self-weight of the soil on normal stress vectors and therefore the shape of the
failure surface. This can be approached by the method of Sokolovski, which
shows distorted stress fields. The lowest bearing capacity that can be obtained by
adjusting the slip lines is an ‘‘upper-bound solution.’’ This means that a pressure
in excess of that amount is likely to cause failure.
These solutions also assume simultaneous development of maximum shearing

resistance all along the slip surfaces, but failure probably is progressive as shown
in Fig. 22.4. This problem is mitigated by applying a generous factor of safety that
hopefully will limit shearing to a very small portion of the potential slip surface.
Another approach is to consider energy relationships such that a system must

fail. This is called a ‘‘lower-bound solution.’’ The energy generated equals the
weight of a structure times the amount of settlement, and the energy used up
depends on stress-stain relationships in the soils. The ultimate achievement is if
Figure 22.8
(a) Too high an
eccentricity e can
initiate liftoff and
reduce the bearing
area. (b) Footing
on a slope has a
reduced surcharge
contribution to
bearing capacity.
Its effect on slope
stability can be
analyzed by a
method of slices.

the upper- and lower-bound solutions agree with one another. As a practical
matter, inconsistencies and anisotropy of the soil, particularly sedimentary
layering, can overshadow the effects of these adjustments.
22.3.8 Rectangular, Square, and Circular Footings

Column loads normally are supported on square footings, but both analytical and
numerical (finite element) solutions struggle with three dimensions. Qualitatively
an equidimensional footing will increase the role of cohesion because the shear
surface is increased, and reduce the influence from friction angle for the same
reason, by diluting the normal stress on the shear surface. This is shown in Fig. 22.9.
The two-dimensional Terzaghi equation has been extended to three dimensions

empirically on the basis of load tests. As anticipated, Nc is increased and N is
decreased. Tests by De Beer in Belgium indicate that the surcharge factor also is
increased, apparently through better lateral confinement. In the case of small
footings the Prandtl smooth-base geometry appears most appropriate. Bearing
capacity factors suggested by Vesic (1975) are presented in Table 22.3.
Factors for rectangular foundations can be obtained by interpolation of the base

width-to-length ratio, B/L: for infinitely long footings, B/L ¼ 0, with factors given
in Table 22.2, and for circular or square footings, B/L ¼ 1 (Table 22.3).
Example 22.6
Estimate bearing capacity factors for a footing that is twice as long as it is wide, with ¼ 208.
Answer: From Tables 22.2 and 22.3 for a smooth base,
B/L N Nc Nq
Long 0 5.4 15 6.4

Square 1 3.2 21 8.7
Interpolation 0.5 4.3 18 7.6
Figure 22.9
Square or circular foundations create a larger shear area that exaggerates
the influence of cohesion and dilutes the effect of friction angle. These
effects are compensated by adjusting the bearing capacity factors.
Table 22.3 Nbs Nccs Ndqs

Bearing capacity
0 0.00 6.2 1.0
factors for circular or
5 0.27 8.0 1.7
square smooth-base
10 0.73 11 2.9
foundations,
15 1.6 15 5.0
modified after Vesic,
20 3.2 21 8.7
(1975)
25 6.5 31 16
30 13 49 29
35 29 79 57
40 66 140 120
45 160 270 270
B
Note: a q0 ¼ N þ cNc þ DNq .
2
b
Ns ¼ 0:6N .

c Nq
Ncs ¼ Nc 1 þ .
Nc
d
Nqs ¼ Nq ð1 þ tan Þ.
22.4 ANALYSIS USING A METHOD OF SLICES
22.4.1 Getting the Big Picture

It is not generally appreciated that bearing capacity and slope stability are two
statements of the same problem—bearing capacity equals slope stability with a
surcharge and no slope. The same analytical methods therefore can be used
for bearing capacity as for slope stability. Influences of soil layering or building on
a slope are easily included in the analysis. There also is a justification for the log
spiral because self-weight of the soil plays a subordinate role to foundation
pressure. The Terzaghi slip geometry therefore can be selected for analysis.
Example 22.7
In Example 22.2, (a) what is the design load with a design factor of safety of 3?
(b) Determine the factor of safety based on soil shear strength using the simple method
of slices.
Answer: (a) qa ¼ 5430 3 ¼ 1810 lb/ft2 or

¼ 260 3 ¼ 87 kpa
(b) The diagram is shown in Fig. 22.10. Results for individual slices with qa as above are
as follows. Note that the acting forces are negative for slices 4 and 5 because of the

Figure 22.10
(Left) Bearing capacity by the Terzaghi method for a rough base, Example 22.2; (right) the same
problem solved by the ordinary method of slices, Example 22.7.
reverse slope of the slip surface.
Slice No. 1 2 3 4 5 Total
Sa, lb 2910 864 73 377 913 2557

Sr, lb 1191 863 769 745 1456 5024
FS 0.41 1.0 10.5 – – 2.0
22.4.2 Will the Real Factor of Safety Please Stand Up?

In the above example there is a substantial disagreement in regard to the factors
of safety: 3.0 from a bearing capacity analysis and 2.0 from the ordinary method
of slices, which, as previously discussed, tends to overestimate the factor of safety.
Who is right? This is not just an academic argument as there are substantial
consequences. A resolution is needed.
Both answers are right, because it depends on whether one is up looking down, or
down looking up, which is always a source of disagreement. Both calculations
assume the same Terzaghi rough base geometry, so how can both be correct?
We will pause a moment for quiet contemplation and prayer if it helps.
The explanation is in the definition of a factor of safety. In the bearing capacity

equations it represents an allowable foundation load, but the foundation load is
only part of the acting forces that include the weight of the soil. A slope stability
factor of safety is based on the total weight in relation to the shearing strength of
the soil. This in effect adds a constant amount to both the numerator and
denominator, as the bearing capacity factor of safety is 3 7 1 ¼ 3 and the shear
strength factor is (3 þ 1) 7 (1 þ 1) ¼ 2. Bearing capacity factors of safety are easy
to calculate and impress clients; engineers are more concerned with a possibility
of failure.

22.4.3 Converting Factors of Safety

A bearing capacity factor of safety can be converted to an equivalent factor of
safety against shear failure by substituting reduced values for cohesion, c, and
for tan into the bearing capacity equation and calculating the corresponding qo.
The correct answer is obtained by trial and error, dividing c and tan by a
trial shear strength factor of safety, determining the allowable, and comparing
it with that obtained from the bearing capacity equation. The trial value is
correct when the allowables are the same.
Example 22.8
In Example 22.7 the load factor of safety is 3 and qa ¼ 1810 lb/ft2 (87 kPa). What is the
corresponding shear strength factor of safety?
Answer: An answer of 2.0 already has been obtained by the method of slices. This may be
tested by dividing c and tan by 2, which gives allowable strength parameters of a ¼ 5.048
and ca ¼ 200 lb/ft2. Substituting appropriate values in eq. (22.5) gives
qa ¼ 0:5ð100Þð4Þ 0:5 þ 200 7:3ð100Þð5Þ 1:6

¼ 200 þ 1460 þ 80 ¼ 1740 lb=ft2 ð83 kPaÞ
This is fairly close to the value obtained with a load factor of safety of 3. Let us therefore
try 1.9, noting that the major influence will be in the cohesion term:
qa ¼ 0:5ð100Þð4Þ 0:5 þ 210 7:3 þ ð100Þð5Þ 1:6

¼ 200 þ 1530 þ 80 ¼ 1810 lb=ft2 ð83 kPaÞ
which agrees with the allowable foundation pressure from the bearing capacity
analysis and indicates an equivalency of the two factors of safety, respectively 3
and 1.9. The shear strength factor of safety of 1.9 therefore is the equivalent of a
load factor of safety of 3.
22.4.4 Importance of Surcharge

The depth of a foundation directly affects the amount and weight of the
surcharge, whose function is to keep the soil that is underneath a foundation from
shearing and squeezing out. Excavation that inadvertently removes surcharge can
have a dramatic and devastating effect, Fig. 22.11.
22.5 BEARING CAPACITIES FOR SPECIAL CASES
22.5.1 Saturated, Undrained Clay

A special case of bearing capacity that applies only to rapid loading is a ‘‘ ¼ 0’’
analysis. Because N ¼ 0 and Nq ¼ 1 for both the rough and smooth base
geometries, eq. (22.7) reduces to
Rough base: q ¼ 5:7c þ D ð22:9aÞ

Smooth base: 1 ¼ 5:14c þ D ð22:9bÞ

If ¼ 0, the unconfined compressive strength is qu ¼ 2c. Substituting this
relationship for c gives
Rough base: q ¼ 2:85qu þ D ð22:10aÞ
Smooth base: q ¼ 2:57qu þ D ð22:10bÞ
A common procedure for small structures on saturated clay is to assume that the
allowable net bearing capacity equals the unconfined compressive strength, which
according to the above relationships automatically includes a strength-based net
factor of safety between about 2.6 and 2.8. Because the friction angle is zero there
is no width factor.
For equidimensional footings on saturated clays with zero drainage, Nc from

Table 22.3 is 6.2, or
Equidimensional smooth base: qa ¼ 3:1qu þ D ð22:10cÞ
Figure 22.11
Cracks around the
window openings
testify to
settlement of the
front wall of this
old building when
surcharge was
removed next to
the foundation.
The building was
temporarily
restrained from
collapse by
cantilevered floor
joists, plus some
sand hastily piled
back against the
foundation.

Use of qu as an allowable net bearing capacity (exclusive of surcharge) for a

square or circular footing therefore gives an inferred factor of safety of 3.1.
Bearing capacity factors may be interpolated for rectangular footings.
The developed factor of safety will be significantly higher than indicated by

eq. (22.10) if partial drainage allows internal friction to become mobilized.
22.5.2 Bearing Capacity of Cohesionless Sand

Without the cohesion term, eq. (22.7) reduces to

B
q ¼ N þ DNq ð22:11Þ
2
Substituting the appropriate N values for ¼ 258 gives

q ¼ ð3:25B þ 16DÞ
For a pneumatic tire q equals the inflation pressure.
22.5.3 Other Bearing Capacity Theories

The failure geometries assumed in an assortment of bearing capacity theories are
shown in Fig. 22.12, with solutions for a friction angle of zero and no surcharge.
The one-sided circular failure suggested by Krey was modified by Wilson, who
solved for the most critical location of the circle, and results are close to those
obtained with two-sided geometrics of the Prandtl-Terzaghi model. Perhaps most
striking is that despite the wide variations in assumed failure geometry, with the
exception of the Rankine geometry the results all fall in a range that is less than
20 percent.
22.5.4 One-Sided and Two-Sided Models

The Prandtl-Terzaghi-Meyerhoff models are symmetrical, but a foundation
loaded to failure inevitably tips one way or the other. This would appear to argue
for support from only one side. However, an analogy may be drawn to a board
supported by ropes at the ends and loaded in the middle: Both ropes support the
load until one breaks.
22.6 THE TYPE B PROBLEM, SETTLEMENT
22.6.1 Overview
Foundation designs must be checked for bearing capacity as outlined above, but in
most cases settlement controls. This will depend on the structure: for example,
settlement is not critical for a light pole so long as it does not tilt, whereas settlement

Figure 22.12
Assumed failure
geometries and
solutions from
various bearing
capacity theories
with ¼ 0 and
D ¼ 0.
should be minimized for a floor supporting sensitive equipment. Differential

settlement, where part of a foundation system settles more than another part, can
rack or destroy a structure, and is evidenced by uneven floors, cracking and faulting
of floors and walls, and doors and windows that stick or drag. Geotechnical
engineers frequently are asked to do the forensics on differential settlement
problems. The maximum differential settlement occurs when one part of a
foundation settles and another does not. Suggestions for allowable maximum and
differential settlements for various structures are indicated in Table 16.3.
The larger the area covered by a structure, the more likely it will be to encounter
variations in the soils. One way to minimize differential settlement is to tie all
foundations together as a continuous mat (or raft) foundation. Such a foundation
must be sufficiently reinforced to bridge over soft spots. Another advantage of
a mat foundation is that a bearing capacity failure becomes unlikely because of
the large width factor. However, a mat generally is considerably more expensive
than separated column and/or wall foundations.
Building codes mentioned earlier in this chapter emphasize bearing capacity over
settlement.
22.6.2 Structures on Clay

Settlement of structures on clay normally is predicted from laboratory
consolidation tests by methods discussed in Chapters 16 and 17. Particularly
important is the preconsolidation pressure, as foundation pressures that are less
than that value will be limited to elastic settlement. The preconsolidation pressure
should not be exceeded for foundations on quick clays.
A relatively new procedure involves manipulating the preconsolidation pressure

by increasing the lateral stress, which is discussed in Chapter 24.
If foundation pressure exceeds the preconsolidation pressure, a rough estimate

of settlement can be made on the basis of an empirical relationship between
the compression index and the liquid limit (eq. (16.9a)). A recent trend has
been toward the use of in-situ tests to predict settlement, which is discussed in
Chapter 26.
The influence of overlapping stress bulbs from adjacent footings should not be
overlooked in predicting settlement. This can become critical if a heavy building
is founded on shallow footings next to an existing building on shallow footings,
and is a reason to select a deep or intermediate foundation for the newer building.
22.6.3 Structures on Sand

Sand is very difficult to sample without causing disturbance, so if consolidation
tests are performed they use reconstituted samples. This difficulty has led to
an increasing use of in-situ tests to predict settlement.
Earlier predictions of settlement on sand were based on Standard Penetration

Test (SPT) blow counts (N) and width of the footing, and did not consider
the distribution of compressive strain with depth. (It will be recalled that clay is
divided into arbitrary layers, and the amount of consolidation is predicted for
each layer based on an elastic distribution of vertical stress.) The SPT and other
in-situ tests will be discussed in relation to prediction of settlement. Details of the
test are presented in Chapter 26.
The preferred procedure is that suggested by Schmertmann and discussed in

Chapter 16, Section 16.10.
22.7 PROPORTIONING FOOTINGS FOR EQUAL SETTLEMENT
22.7.1 Overview
A goal is to minimize differential settlement, but column loads usually differ in
a single structure, and bearing pressures and the geometry are not the same for
column footings and wall footings. Differential settlement is almost assured when
part of a structure is supported on shallow foundations and part is on piles.
Differential settlement also is unavoidable when parts of a structure on shallow
foundations are built at different times, because of the nonlinear time factor
controlling the rate of settlement.
A common guide is to limit maximum settlement to one inch (25 mm), which also
limits differential settlement to one inch.
22.7.2 Proportioning Footing Areas

Differential settlement can be reduced if different foundation bearing areas are
proportioned to give approximately the same amount of settlement. The
apportionment is based on the dead load plus a percentage of the live and wind
loads, depending on whether the live load will be in place over substantial periods
of time. For simplicity, settlement often is assumed to be proportional to pressure
regardless of the size of the footing—in other words, double the weight means
doubling the area of the footing. This is an obvious oversimplification because it
omits any depth effect, but normally is adequate so long as loads are not highly
variable. If adjacent loads do vary by a large amount, consideration should be
given to using a combined footing, discussed in the following paragraphs.
Settlement predictions also assume that footings are free-standing and are not
affected by rigidity of the superstructure. This is the design settlement, which may
or may not be attained by the structure, depending on its flexibility. The designer
ordinarily assumes that the full predicted design settlements will occur, then
examines the structure to determine if the resulting stresses and differential
movements can be tolerated.
22.7.3 Settlement of Combined Footings and Rafts

Footings that support more than one column load or column plus wall loads are
referred to as combined footings. If possible the combined load must act through
the centroid of the bearing area, or the footing will tilt. Two unequal column loads
may be supported by a footing that is trapezoidal in plan. Combined footings are
useful next to property lines so that a footing area extends between adjacent
columns instead of extending out symmetrically around each one.
The extreme case of a combined footing is a raft or mat foundation that extends
under an entire structure and supports all of the columns and walls. Mats are used
on soft soils because bearing pressures are greatly reduced compared to those of
individual footings.
Rafts are sometimes incorrectly referred to as ‘‘floating foundations,’’ implying

that they derive their support from buoyancy, in which case settlement would be
zero. A true floating foundation requires deep excavation, so the weight of the
structure equals the weight of the soil removed. Partially floating foundations
have been used in very soft, compressible soils, as in Mexico City.
Most rafts are reinforced concrete mats with considerable structural rigidity that
limits differential settlement to about one-half of the design settlement, so the
total allowable settlement often is larger, about 50 mm (2 in.). The raft must have
sufficient reinforcement to bridge softer areas. The pressure bulb from a raft
extends much deeper because of the larger width, and this also has an averaging
effect.
Example 22.9
Two columns with centerlines 3 ft apart are to be supported by a single footing.
One column carries 28 tons and the other one-half of that amount. The allowable bearing
capacity is 2 ton/ft2. Design a trapezoidal footing.
Answer: The first step is to determine the base area: A ¼ (28 þ 14) 7 2 ¼ 21 ft2. This is the
area of a square 4.6 ft on a side, which may be used as a guide for selecting an arbitrary
length dimension h shown at the right in Fig. 22.13.
Next, the location of the centroid of the load is obtained by summing moments around
a reference point exterior to the columns. If the point is selected 1 ft from the heavier
column, m(28 þ 14) ¼ 28(1) þ 14(4), and m ¼ 2.0 ft, as shown in the left-hand sketch.
The area of a trapezoid is (Tuma, 1970):
A ¼ hða þ bÞ ¼ 21 ft2
Let a trial h ¼ 6 ft; then 21 ¼ 6(a þ b) and (a þ b) ¼ 3.5 ft. For a rectangular
footing a ¼ b ¼ 1.75 ft. This configuration is shown in the left half of the right-
hand figure.
For a nonrectangular footing, (a þ b) ¼ 3.5, so a reduction in one means an increase in the

other. The centroid of a trapezoid is located a distance f from the longer parallel edge,
Figure 22.13
Example 22.9:
design of a
trapezoidal
footing.

which is defined by
h a þ 2b
f¼ ¼ 2ð3:5 þ bÞ=3:5
3 aþb
A symmetrical footing will extend the same distance from both columns, which is
(6 – 3)/2 ¼ 1.5 ft. Adding the distance from one column to the centroid gives f ¼
2.5 ft. Then b ¼ 0.875 ft and a ¼ 2.625 ft. This configuration is shown at the right in the figure.
22.7.4 Minimizing Settlement with Intermediate and Deep

Foundations
Deep foundations, piles and piers, minimize settlement by transferring loads
downward to stiffer soil layers, as discussed in the next chapter. The supporting
capacity of a deep foundation is predicted, and then normally is checked with full-
scale load tests of sample production piles or piers, which gives a direct measure of
the amount of settlement that can be expected to take place under the load of the
structure.
Another approach is to modify the soil to reduce its modulus. This may be done
by surcharging to increase the preconsolidation pressure so that it exceeds the
anticipated foundation loads, as discussed elsewhere in this book. For granular
soils that are remote from other structures, ‘‘deep dynamic compaction,’’ or
repeatedly dropping a heavy weight, can achieve the same goal. A built-up
foundation soil can incorporate lateral restraint by including horizontal tension
members, as is done for MSE walls. Another approach that is gaining in
popularity is to laterally compact the foundation soil with rammed-in aggregate
piers, which in this book is considered an intermediate foundation and is discussed
in Chapter 24.
22.8 SUMMARY
Shallow foundations are designed to meet two criteria: bearing capacity and
settlement. Foundations on or near slopes impose surcharge loads that must not
cause a landslide. Another aspect that is discussed in a later chapter is safety from
earthquakes. It is axiomatic that the geotechnical investigation that is not made
prior to construction often will be conducted later.
Problems
22.1. Define bearing capacity of a foundation soil. Define allowable bearing
value.
22.2. Explain the difference between gross and net bearing capacity.

22.3. What two conditions must be met when determining the bearing capacity
of a shallow foundation? Are there other requirements that are discussed
in other chapters?
22.4. What is meant by the term ‘‘differential settlement’’?
22.5. A square footing is to be founded on a stratum of medium clay. What
is the allowable bearing capacity according to the Boston Building Code?
22.6. The unconfined compression strength of the clay in the previous problem
is 1.5 tons/ft2 (144 kPa). Based on this information alone, can you give
an approximate bearing capacity? Is this in relation to shear failure,
settlement, or both?
22.7. The unconfined compressive strength of a sand is zero. Based on this
information alone, can you estimate the bearing capacity?
22.8. A dry dune sand has a friction angle of 258 and a unit weight of 95 lb/ft3.
Use the Rankine formula to estimate the depth of a footprint exerting
200 lb/ft2 of pressure.
22.9. The answer in the previous problem is subjected to an experimental proof.
Do you expect settlement to be more, less, or exactly as predicted? Explain.
22.10. Explain in simple terms why it is not advisable to wear cowboy boots
or high heels when playing beach volleyball.
22.11. Explain two versions of factor of safety with regard to bearing capacity.
Which is higher? Which is more scientific?
22.12. Recalculate the bearing capacity in Example 22.1 if the friction angle
is 208, and compare the allocation to width, cohesion, and surcharge.
Give qa based on a load factor of safety of 3.
22.13. A concrete wall exerting 10,000 lb/ft is to be founded at a depth of 4 ft in
soil having a cohesion of 400 lb/ft2, a friction angle of 168, and a unit
weight of 100 lb/ft3. What width footing will be required for a load factor
of safety of 3? Round upward to the nearest foot and calculate the
reduction in factor of safety during high water.
22.14. The wall in the previous example is subjected to a horizontal force of
2000 lb at 40% of the wall height. Calculate eccentricity and adjust
the footing width as necessary.
22.15. Determine an allowable bearing capacity for a column footing 4 ft square
on soil having a unit weight of 100 lb/ft3, cohesion of 800 lb/ft2, and
friction angle of 278. The column is interior in a warehouse building so
there is no significant surcharge. Why should a bond breaker be
incorporated between the outside of the footing and a concrete floor?
22.16. What is the maximum permissible differential settlement for a bell tower
90 ft high with a foundation width of 22 ft? What would be the resulting
tilt angle?
22.17. The maximum settlement of a 3 ft 3 ft footing on sand is not to exceed
1 in. The average STP blow count is 16 blows per foot. Disregard any

influence from surcharge and determine the maximum load on this

footing.
22.18. The footing of Problem 22.17 is founded 3 ft deep in soil having a unit
weight of 110 lb/ft3. What is the maximum load?
22.19. The friction angle in Problem 22.18 is 258. Determine the factor of safety
against a bearing capacity failure and adjust the footing size if necessary
for FS ¼ 3.
22.20. When should Schmertmann’s C2 correction be used?
22.21. Redesign the trapezoidal footing in Example 22.9 with the larger column
load increased 50%.
22.22. The footing of Example 22.9 is on sand having an average STP N ¼
24 blows per foot. Estimate settlement at each end of the footing.
22.23. Calculate the minimum cohesion for a clay soil to support a D8 crawler
tractor weighing 227 kN (51,000 lb) with a track width of 560 mm (22 in.)
and a total contact area of 3.5 m2 (5445 in.2). Assume a soil unit weight of
15.7 kN/m3 (100 lb/ft3), an undrained friction angle of zero, and zero
depth of sinking.
22.24. Construct a computer spreadsheet for bearing capacity identifying contri-
butions from width, cohesion, and surcharge terms.
References and Further Reading

Bowles, J. E. (1988). Foundation Analysis and Design. McGraw-Hill, New York.
Housel, W. R. (1956). ‘‘A Generalized Theory of Soil Resistance.’’ ASTM Special Technical
Publication 26, 13–29.
Holtz, R. D. (1991). ‘‘Stress Distribution and Settlement of Shallow Foundations.’’ In
H. Y. Fang, ed., Foundation Engineering Handbook, 2nd ed., Van Nostrand Reinhold,
New York, pp. 166–222.
Jumikis, A. R. (1969). Theoretical Soil Mechanics. Van Nostrand Reinhold, New York.
Lambe, T. W., and Whitman, R. V. (1969). Soil Mechanics. John Wiley & Sons, New York.
Meyerhof, G. G. (1951). ‘‘The Ultimate Bearing Capacity of Foundations.’’ Ge´otechnique
5, 301–332.
Peck, R. B., Hanson, W. B., and Thornburn, T. H. (1974). Foundation Engineering, 2nd ed.
Wiley, New York.
Robertson, P. K., Campanella, R. G., and Wightman, A. (1983). ‘‘SPT-CPT Correlations.’’
ASCE J. Geotech. Eng. 109(11), 1449–1459.
Schmertmann, J. H. (1970). ‘‘Static Cone to Compute Static Settlement over Sand.’’ ASCE
J. Geotech. Eng. 96(SM3), 1011–1043.
Schmertmann, J. H., Hartman, J. P., and Brown, P. R. (1978). ‘‘Improved Strain Influence
Factor Diagrams.’’ ASCE J. Geotech. Eng. Div. 104(GT8), 1131–1135.
Terzaghi, K. (1943). Theoretical Soil Mechanics. John Wiley & Sons, New York.
Tuma, J. T. (1970). Engineering Mathematic Handbook. McGraw-Hill, New York.
Vesic, A. C. (1975) ‘‘Bearing Capacity of Shallow Foundations.’’ In H. F. Winterkorn
and H. Y. Fang, eds., Foundation Engineering Handbook, Van Nostrand Reinhold,
New York, Ch. 3.
Wu, T. H. (1976). Soil Mechanics. Allyn & Bacon, Boston.


22 - Bearing Capacity of Shallow

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

22 - Bearing Capacity of Shallow

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

22 - Bearing Capacity of Shallow

Uploaded by

Copyright:

Available Formats

Source: GEOTECHNICAL ENGINEERING

22 Bearing Capacity of Shallow

22.1.1 The Functional Foundation

22.1.2 Site Evaluation

Bearing Capacity of Shallow Foundations 639

Regardless of the detail of the investigation, a final step is to inspect open

22.1.3 Building Codes

22.2 FOUNDATION BEARING CAPACITY

22.2.1 Two Kinds of Failures

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

640 Geotechnical Engineering

Table 22.1 Allowable bearing

14. Compacted granular fill 2 to 5 (190 to 480)b

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Bearing Capacity of Shallow Foundations 641

22.2.2 Factors Influencing Type A Bearing Capacity

Another contributor to low shearing strength is excess pore water pressure

A famous example of a bearing capacity failure from development of excess pore

22.2.3 Analytical Approach to Bearing Capacity

The general procedure is to define an ultimate bearing capacity, which is the

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

642 Geotechnical Engineering

22.2.4 Rankine’s Analysis

Passive side : h ¼ DKp

Bearing Capacity of Shallow Foundations 643

22.2.5 Critique of Rankine’s Theory

22.2.6 Bell’s Equation

644 Geotechnical Engineering

A test for any equation is to consider end conditions. If ¼ 0, Bell’s equation

If the unconfined compressive strength is regarded as an undrained test, as

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Bearing Capacity of Shallow Foundations 645

22.2.7 Critical Edge Pressure and Progressive Failure

If the friction angle is zero, Fröhlich’s solution becomes

22.2.8 Positive Gains from Perimeter Shear

646 Geotechnical Engineering

observation nevertheless provides some useful insight into interpretation of plate

Housel proposed an empirical relationship based on plate bearing test results:

22.2.9 Loss of Strength from Perimeter Shear

Bearing Capacity of Shallow Foundations 647

sensitive soil. The amount of strength lost is difficult to predict as it varies

22.2.10 When a Plate Bearing Test Should Not Be

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

648 Geotechnical Engineering

22.3 TERZAGHI BEARING CAPACITY THEORY

22.3.1 Prandtl-Terzaghi Theory

Terzaghi modified Prandtl’s theory by including a surcharge load (Fig. 22.6(b)).

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Bearing Capacity of Shallow Foundations 649

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

650 Geotechnical Engineering

Table 22.2 Terzaghi rough base Smooth base

it is a simple matter to make a spreadsheet based on these values or their

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)

Bearing Capacity of Shallow Foundations 651

22.3.2 Pore Water Pressure

Answer: The simplified Bishop pore pressure equation states

Answer: (a) qa ¼ 5430 3 ¼ 1810 lb/ft2 or