CHAPTER 4

PILES AND
PILING FORMULAS

ALLOWABLE LOADS ON PILES

A dynamic formula extensively used in the United States to determine the

allowable static load on a pile is the Engineering News formula. For piles dri-
ven by a drop hammer, the allowable load is

2WH
Pa ! (4.1)
p"1
For piles driven by a steam hammer, the allowable load is
2WH
Pa ! (4.2)
p " 0.1

where Pa ! allowable pile load, tons (kg)

W ! weight of hammer, tons (kg)
H ! height of drop, ft (m)
p ! penetration of pile per blow, in (mm)

The preceding two equations include a factor of safety of 6.

For a group of piles penetrating a soil stratum of good bearing characteris-
tics and transferring their loads to the soil by point bearing on the ends of the
piles, the total allowable load would be the sum of the individual allowable
loads for each pile. For piles transferring their loads to the soil by skin friction
on the sides of the piles, the total allowable load would be less than the sum on
the individual allowable loads for each pile, because of the interaction of the
shearing stresses and strains caused in the soil by each pile.

LATERALLY LOADED VERTICAL PILES

Vertical-pile resistance to lateral loads is a function of both the flexural stiff-

ness of the shaft, the stiffness of the bearing soil in the upper 4 to 6D length of
shaft, where D ! pile diameter and the degree of pile-head fixity.

105
106 CHAPTER FOUR

The lateral-load versus pile-head deflection relationship is developed from

charted nondimensional solutions of Reese and Matlock. The solution assumes
the soil modulus K to increase linearly with depth z; that is, K ! nh z, where nh !
coefficient of horizontal subgrade reaction. A characteristic pile length T is cal-
culated from

B nh
EI
T! (4.3)

where EI ! pile stiffness. The lateral deflection y of a pile with head free to
move and subject to a lateral load Pt and moment Mt applied at the ground line
is given by
T3 T2
y ! Ay Pt " By Mt (4.4)
EI EI

where Ay and By are nondimensional coefficients. Nondimensional coefficients

are also available for evaluation of pile slope, moment, shear, and soil reaction
along the shaft.
For positive moment,
M ! Am Pt T " Bm Mt (4.5)

Positive Mt and Pt values are represented by clockwise moment and loads

directed to the right on the pile head at the ground line. The coefficients applic-
able to evaluation of pile-head deflection and to the maximum positive moment
and its approximate position on the shaft, z/T, where z ! distance below the
ground line, are listed in Table 4.1.
The negative moment imposed at the pile head by pile-cap or another struc-
tural restraint can be evaluated as a function of the head slope (rotation) from
A#PtT # EI
$Mt ! $ s (4.6)
B# B#T

where #s rad represents the counterclockwise (") rotation of the pile head and
A# and B# are coefficients (see Table 4.1). The influence of the degrees of fixity

TABLE 4.1 Percentage of Base Load Transmitted to

Rock Socket

Er /Ep

Ls /ds 0.25 1.0 4.0

0.5 54* 48 44
1.0 31 23 18
1.5 17* 12 8*
2.0 13* 8 4

*Estimated by interpretation of finite-element solution;

for Poisson’s ratio ! 0.26.
 PILES AND PILING FORMULAS 107

of the pile head on y and M can be evaluated by substituting the value of $Mt
from the preceding equation into the earlier y and M equations. Note that, for
the fixed-head case,

yf !
PtT 3
EI ! A B
Ay $ # y
B# " (4.7)

TOE CAPACITY LOAD

For piles installed in cohesive soils, the ultimate tip load may be computed from
Qbu ! Ab q !Ab Nc cu (4.8)

where Ab ! end-bearing area of pile, ft2 (m2)

q ! bearing capacity of soil, tons/ft2 (MPa)
Nt ! bearing-capacity factor
cu ! undrained shear strength of soil within zone 1 pile diameter above
and 2 diameters below pile tip, psi (MPa)
Although theoretical conditions suggest that Nc may vary between about 8 and
12, Nc is usually taken as 9.
For cohesionless soils, the toe resistance stress, q, is conventionally
expressed by Eq. (4.1) in terms of a bearing-capacity factor Nq and the effective
overburden pressure at the pile tip &'vo
q ! Nq&'vo ( ql (4.9)

Some research indicates that, for piles in sands, q, like fs, reaches a quasi-
constant value, ql , after penetrations of the bearing stratum in the range of 10 to
20 pile diameters. Approximately
ql ! 0.5Nq tan % (4.10)

where % is the friction angle of the bearing soils below the critical depth. Val-
ues of Nq applicable to piles are given in Fig. 4.1. Empirical correlations of soil
test data with q and ql have also been applied to predict successfully end-bearing
capacity of piles in sand.

GROUPS OF PILES

A pile group may consist of a cluster of piles or several piles in a row. The
group behavior is dictated by the group geometry and the direction and location
of the load, as well as by subsurface conditions.
Ultimate-load considerations are usually expressed in terms of a group
efficiency factor, which is used to reduce the capacity of each pile in the
108 CHAPTER FOUR

1000

500
Bearing capacity factor, Nq

100

50

10
25° 30° 35° 40° 45° 50°
Angle of internal friction, φ
FIGURE 4.1 Bearing-capacity factor for granular soils related to angle of
internal friction.

group. The efficiency factor Eg is defined as the ratio of the ultimate group
capacity to the sum of the ultimate capacity of each pile in the group.
Eg is conventionally evaluated as the sum of the ultimate peripheral friction
resistance and end-bearing capacities of a block of soil with breadth B, width
W, and length L, approximately that of the pile group. For a given pile, spacing
S and number of piles n,

2(BL " WL) fs " BWg

Eg ! (4.11)
nQu
where fs is the average peripheral friction stress of block and Qu is the single-pile
capacity. The limited number of pile-group tests and model tests available suggest
that for cohesive soils, Eg ) 1 if S is more than 2.5 pile diameters D and for cohe-
sionless soils, Eg ) 1 for the smallest practical spacing. A possible exception
might be for very short, heavily tapered piles driven in very loose sands.
In practice, the minimum pile spacing for conventional piles is in the range
of 2.5 to 3.0D. A larger spacing is typically applied for expanded-base piles.
A very approximate method of pile-group analysis calculates the upper limit
of group drag load, Qgd from

Qgd ! AF*FHF " PHcu (4.12)

where Hf , *f , and AF represent the thickness, unit weight, and area of fill con-
tained within the group. P, H, and cu are the circumference of the group, the
 PILES AND PILING FORMULAS 109

thickness of the consolidating soil layers penetrated by the piles, and their
undrained shear strength, respectively. Such forces as Qgd could only be
approached for the case of piles driven to rock through heavily surcharged,
highly compressible subsoils.
Design of rock sockets is conventionally based on
+ 2
Qd ! +ds Ls fR " ds qa (4.13)
4
where Qd ! allowable design load on rock socket, psi (MPa)
ds ! socket diameter, ft (m)
Ls ! socket length, ft (m)
fR ! allowable concrete-rock bond stress, psi (MPa)
qa ! allowable bearing pressure on rock, tons/ft2 (MPa)
Load-distribution measurements show, however, that much less of the load
goes to the base than is indicated by Eq. (4.6). This behavior is demonstrated
by the data in Table 4.1, where Ls /ds is the ratio of the shaft length to shaft
diameter and Er /Ep is the ratio of rock modulus to shaft modulus. The finite-
element solution summarized in Table 4.1 probably reflects a realistic trend if
the average socket-wall shearing resistance does not exceed the ultimate fR
value; that is, slip along the socket side-wall does not occur.
A simplified design approach, taking into account approximately the compati-
bility of the socket and base resistance, is applied as follows:
1. Proportion the rock socket for design load Qd with Eq. (4.6) on the assump-
tion that the end-bearing stress is less than qa [say qa /4, which is equivalent to
assuming that the base load Qb ! (+/4) d2s qa /4].
2. Calculate Qb ! RQd, where R is the base-load ratio interpreted from Table 4.1.
3. If RQd does not equal the assumed Qb , repeat the procedure with a new qa
value until an approximate convergence is achieved and q ( qa.

The final design should be checked against the established settlement tolerance
of the drilled shaft.
Following the recommendations of Rosenberg and Journeaux, a more realis-
tic solution by the previous method is obtained if fRu is substituted for fR. Ideally,
fRu should be determined from load tests. If this parameter is selected from data
that are not site specific, a safety factor of at least 1.5 should be applied to fRu in
recognition of the uncertainties associated with the UC strength correlations.*

FOUNDATION-STABILITY ANALYSIS

The maximum load that can be sustained by shallow foundation elements at

incipient failure (bearing capacity) is a function of the cohesion and friction
angle of bearing soils as well as the width B and shape of the foundation.

*Rosenberg, P. and Journeaux, N. L., “Friction and End-Bearing Tests on Bedrock for High-
Capacity Socket Design,” Canadian Geotechnical Journal, 13(3).
110 CHAPTER FOUR

The net bearing capacity per unit area, qu, of a long footing is convention-
ally expressed as
qu ! ,f cu Nc " &'voNq " -f *BN* (4.14)

where ,f ! 1.0 for strip footings and 1.3 for circular and square footings
cu ! undrained shear strength of soil
&'vo ! effective vertical shear stress in soil at level of bottom of
footing
-f ! 0.5 for strip footings, 0.4 for square footings, and 0.6 for
circular footings
* ! unit weight of soil
B ! width of footing for square and rectangular footings and
radius of footing for circular footings
Nc, Nq, N* ! bearing-capacity factors, functions of angle of internal
friction %
For undrained (rapid) loading of cohesive soils, % ! 0 and Eq. (4.7) reduces to
qu ! N c' cu (4.15)

where N'c ! ,f Nc. For drained (slow) loading of cohesive soils, % and cu are
defined in terms of effective friction angle %' and effective stress c'u.
Modifications of Eq. (4.7) are also available to predict the bearing capacity of
layered soil and for eccentric loading.
Rarely, however, does qu control foundation design when the safety factor is
within the range of 2.5 to 3. (Should creep or local yield be induced, excessive
settlements may occur. This consideration is particularly important when select-
ing a safety factor for foundations on soft to firm clays with medium to high
plasticity.)
Equation (4.7) is based on an infinitely long strip footing and should be
corrected for other shapes. Correction factors by which the bearing-capacity
factors should be multiplied are given in Table 4.2, in which L ! footing
length.
The derivation of Eq. (4.7) presumes the soils to be homogeneous throughout
the stressed zone, which is seldom the case. Consequently, adjustments may be
required for departures from homogeneity. In sands, if there is a moderate varia-
tion in strength, it is safe to use Eq. (4.7), but with bearing-capacity factors repre-
senting a weighted average strength.
Eccentric loading can have a significant impact on selection of the bearing
value for foundation design. The conventional approach is to proportion the
foundation to maintain the resultant force within its middle third. The footing is
assumed to be rigid and the bearing pressure is assumed to vary linearly as
shown by Fig. 4.2(b). If the resultant lies outside the middle third of the foot-
ing, it is assumed that there is bearing over only a portion of the footing, as
shown in Fig. 4.2(d). For the conventional case, the maximum and minimum
bearing pressures are

qm !
P
BL !1 . 6eB " (4.16)
 PILES AND PILING FORMULAS 111

TABLE 4.2 Shape Corrections for Bearing-Capacity Factors of Shallow

Foundations*

Correction factor
Shape of
foundation Nc Nq Ny

Rectangle† 1" ! BL "! NN " q

c
1" ! BL " tan % 1 $ 0.4 ! BL "
Circle and
square
1" ! NN "q

c
1 " tan % 0.60

*After De Beer, E. E., as modified by Vesic, A. S. See Fang, H. Y., Foundation

Engineering Handbook, 2d ed., Van Nostrand Reinhold, New York.
†
No correction factor is needed for long-strip foundations.

where B ! width of rectangular footing

L ! length of rectangular footing
e ! eccentricity of loading
For the other case Fig. 4.2(c), the soil pressure ranges from 0 to a maximum of

2P (4.17)
qm !
3L(B/2 $ e)

b
b

P
M P M

(a) P (c) e P
e
3( b – e)
2
p2
P1
p
(b) (d)

FIGURE 4.2 Footings subjected to overturning.

112 CHAPTER FOUR

For square or rectangular footings subject to overturning about two principal

axes and for unsymmetrical footings, the loading eccentricities e1 and e2 are
determined about the two principal axes. For the case where the full bearing
area of the footings is engaged, qm is given in terms of the distances from the
principal axes, c1 and c2, the radius of gyration of the footing area about the
principal axes, r1 and r2, and the area of the footing A as

qm !
P
A !1 " erc
1 1
2
1
"
e2c2
r22 " (4.18)

For the case where only a portion of the footing is bearing, the maximum pres-
sure may be approximated by trial and error.
For all cases of sustained eccentric loading, the maximum (edge) pressures
should not exceed the shear strength of the soil and also the factor of safety
should be at least 1.5 (preferably 2.0) against overturning.

AXIAL-LOAD CAPACITY OF SINGLE PILES

Pile capacity Qu may be taken as the sum of the shaft and toe resistances, Qsu
and Qbu, respectively.
The allowable load Qa may then be determined from either Eq. (4.12) or (4.13):

Qsu " Qbu

Qa ! (4.19)
F
Q Q
Qa ! su " bu (4.20)
F1 F2

where F, F1, and F2 are safety factors. Typically, F for permanent structures
is between 2 and 3, but may be larger, depending on the perceived reliability
of the analysis and construction as well as the consequences of failure.
Equation (4.13) recognizes that the deformations required to fully mobilize
Qsu and Qbu are not compatible. For example, Qsu may be developed at dis-
placements less than 0.25 in (6.35 mm), whereas Qbu may be realized at
a toe displacement equivalent to 5 to 10 percent of the pile diameter. Conse-
quently, F1 may be taken as 1.5 and F2 as 3.0, if the equivalent single safety
factor equals F or larger. (If Qsu/Qbu/ 1.0, F less than 2.0 is usually considered
as a major safety factor for permanent structures.)

SHAFT SETTLEMENT

Drilled-shaft settlements can be estimated by empirical correlations or by

load-deformation compatibility analyses. Other methods used to estimate settle-
ment of drilled shafts, singly or in groups, are identical to those used for piles.
 PILES AND PILING FORMULAS 113

These include elastic, semiempirical elastic, and load-transfer solutions for

single shafts drilled in cohesive or cohesionless soils.
Resistance to tensile and lateral loads by straight-shaft drilled shafts should
be evaluated as described for pile foundations. For relatively rigid shafts with
characteristic length T greater than 3, there is evidence that bells increase the
lateral resistance. The added ultimate resistance to uplift of a belled shaft Qut
can be approximately evaluated for cohesive soils models for bearing capacity
[Eq. (4.14)] and friction cylinder [Eq. (4.15)] as a function of the shaft diameter
D and bell diameter Db.*
For the bearing-capacity solution,

+
Qul ! (D2b $ D2)Nc 0cu " Wp (4.21)
4
The shear-strength reduction factor 0 in Eq. (4.14) considers disturbance
effects and ranges from 1#2 (slurry construction) to 3#4 (dry construction). The cu
represents the undrained shear strength of the soil just above the bell surface,
and Nc is a bearing-capacity factor.
The failure surface of the friction cylinder model is conservatively assumed
to be vertical, starting from the base of the bell. Qut can then be determined for
both cohesive and cohesionless soils from

Qul ! +b L fut " Ws " Wp (4.22)

where fut is the average ultimate skin-friction stress in tension developed on the
failure plane; that is, fut ! 0.8cu for clays or K & vo
' tan % for sands. Ws and Wp
represent the weight of soil contained within the failure plane and the shaft
weight, respectively.

SHAFT RESISTANCE IN COHESIONLESS SOILS

The shaft resistance stress fs is a function of the soil-shaft friction angle 1,

degree, and an empirical lateral earth-pressure coefficient K:

fs ! K & v'o tan 1 ( fl (4.23)

At displacement-pile penetrations of 10 to 20 pile diameters (loose to dense

sand), the average skin friction reaches a limiting value fl. Primarily depending
on the relative density and texture of the soil, fl has been approximated conser-
vatively by using Eq. (4.16) to calculate fs.
For relatively long piles in sand, K is typically taken in the range of 0.7 to
1.0 and 1 is taken to be about % $ 5, where % is the angle of internal friction,

*Meyerhof, G. G. and Adams, J. I., “The Ultimate Uplift Capacity of Foundations,” Canadian
Geotechnical Journal, 5(4):1968.
114 CHAPTER FOUR

degree. For piles less than 50 ft (15.2 m) long, K is more likely to be in the
range of 1.0 to 2.0, but can be greater than 3.0 for tapered piles.
Empirical procedures have also been used to evaluate fs from in situ tests,
such as cone penetration, standard penetration, and relative density tests. Equa-
tion (4.17), based on standard penetration tests, as proposed by Meyerhof, is
generally conservative and has the advantage of simplicity:

N
fs ! (4.24)
50

where N ! average standard penetration resistance within the embedded length

of pile and fs is given in tons/ft2.*

*Meyerhof, G. G., “Bearing Capacity and Settlement of Pile Foundations,” ASCE Journal of
Geotechnical Engineering Division, 102(GT3):1976.