INTERNATIONAL SOCIETY FOR

SOIL MECHANICS AND

GEOTECHNICAL ENGINEERING

This paper was downloaded from the Online Library of

the International Society for Soil Mechanics and
Geotechnical Engineering (ISSMGE). The library is
available here:

[Link]

This is an open-access database that archives thousands

of papers published under the Auspices of the ISSMGE and
maintained by the Innovation and Development
Committee of ISSMGE.
 3b/10

Bearing Capacity of Piles and Pile Groups

Le Pouvoir Portant des Pieux et des Groupes de Pieux

by Â. K é z d i , Budapest, XI., Budafoki-u. 4, Hungary

Summary Sommaire
The author deduces a semi-empirical law relating skin friction to L’auteur déduit une loi de cisaillement entre les contraintes et les
strain for the case of piles embedded in sand. This new law enables déplacements pour le cas des pieux dans le sable. La nouvelle loi
the loading test diagram for piles to be constructed. Piles sup- permet de construire le diagramme des essais de pieux; la solution
ported exclusively by skin friction and piles with point support are du problème des pieux flottants, des pieux travaillant à la pointe,
dealt with, as simple cases; then the composite problem is solved. est décrite, puis Celle du cas des pieux mixtes. Ce cas général montré
The général case exhibits the nature of local shear failures; there is la nature de la rupture locale de cisaillement; il n’y a pas de discon­
no sharp break at failure, the loading diagram continues to slope. tinuité à la rupture; le diagramme est continu.
The theoretical results are supported by model tests carried out in Les essais sur modèle réduit confirment les résultats théoriques,
the laboratory and in the field. Field test results are used to déter­ Les constantes des équations sont déterminées à l’aide d’un essai sur
mine the constants of the équations. chantier.
The ultimate load of pile groups is determined by large model tests La force portante du groupe de pieux fut déterminée par des essais
in the field. The ultimate load of pile rows and of four piles placed sur grands modèles. La force de rupture de quatre pieux battus et
at the corners of a square is surprisingly greater than four times the rangés sur une ligne, ou en carré est plus grande que quatrefois la
ultimate load of a single pile, and the seulement is considerably force portante d’un seul pieu, ce qui est très surprenant, l’enfoncement
smaller. étant considérablement plus petit.
Research on piles has already shown in détail th a t the so- In the form ula t is the shear stress required for producing a
called dynam ic pile driving form ulae and the static relation- displacement p at norm al stress a ; <j>is the angle o f full shearing
ships based on the earth pressure theories are n o t suitable for résistance; p0 is the shear displacem ent necessary fo r the
determ ining the bearing capacity. The déterm ination o f the development o f full friction; k is a constant depending on the
safe load on the basis o f loading diagram s taken in the course kind and the com pactness o f the soil.
o f a load test also présents difficulties, either because the value F o r example, Fig. 2 présents plots taken during a shear test
o f the ultim ate load could n o t be reached o r because général o f a sandy gravel on a concrete surface and the values are
shear failure does n o t occur: the loading diagram does n o t
become vertical b u t continues to slope; there is no failure
point. A further problem arises w ith the bearing capacity o f
pile groups. Theoretical investigations have so far covered
individual piles; actual observations, however, have shown th at
the seulem ent o f a group o f piles can hardly be correlated with
th a t o f the individual pile. D eform ations require m ore con­
sidération in the investigation o f the bearing capacity o f piles;
pile foundations should be designed on the basis o f allowable
settlement.
T his paper aim s at contributing to the élucidation o f the
controversial problem s in this field by the theoretical déterm ina­
tio n of the loading diagram o f individual piles and by presenting
the results o f systematic model tests.
Essai de cisaillement; et contraintes déplacements
Theoretical Détermination of the Loading Diagram of an
Individual Pile
com puted by équation 1. Similar good agreem ent was found
(1) T o determine the inter-relation between load and settle­ in num erous cases.
m ent a new law o f shear is introduced. This law was deduced Equation 1 satisfies the boundary conditions : if a = 0,
from m easurem ents o f the angle o f shearing résistance between r = 0 ; if p = 0, t = 0 ; if p = p0, r = u tan <f>an d dr/dp = 0.
granular m aterials and other construction materials in a shear T he constant k is determined by the slope o f the tangent to the
test apparatus o f 30 X 30 cm 2 surface and with strain control. curve t = f(p ) a t the origin:
A characteristic experimental resuit is shown in Fig. 1. F o r the
ratio o f stresses r and cr pertaining to full shear the Coulom b / 0t \ At
line is obtained ; on the other hand, if the value o f the shearing b d = 7 ; ^ = Po tan a0 . . . . ( 2)
\ cpl P =0 P 0
stresses associated with a certain horizontal displacement p is
plotted as a function o f the corresponding norm al stresses, (2) Let the loading diagram o f the pile be investigated on the
sim ilar straight lines are obtained. The am ount o f displace­ basis o f this law o f shear. The pile will first be regarded as a
m ent necessary for the development of full friction varies within rigid, weight-less cylindrical colum n em bedded into a homo-
close limits so th a t the sought relation can be approached for geneous granular m aterial; there is no point résistance and
cohesionless soils by the following law o f shear: the pile is floating. U nder the effect o f a given pénétration />,
résistances to shear develop ; their sum is equal to the force P
r = a ta n < £ [ l — exP ( - * ^ ^ ) ] . . . . ( 1)
acting on the pile. In the case of bored piles the earth pressure
46
at rest is acting as a norm al stress on the peripheral surface depth 2, then the vertical strain is dpjdz = P /E A , where E is
(a = K0zy), hence: Y oung’s m odulus for the pile and A is its area o f cross-section,
therefore
P= = / - exp ( - * ^ 7)]* P = E A ^ = EA ^
dz dz d z1
(K = circumference o f pile)
If the shear stress associated with a displacement p is -r(p), then
p = tan <£^1 — exp k— jj ----- (3)
dP - T(p)Kdz
Fig. 3 présents a curve taken in the course of a model test; the dP , .„
point résistance being eliminated. Tz = t(p)K
Pile load kg
4 6 10

Steel pile (F-9 cm2)

Measured
<p=32° Computed
y -l-d tons/m3
k0,0-6
I =50cm\K=10-6cm
K -h 5 ;p Q=0-5cm

0-4

pn= 0'5cm
0-5

Fig. 2 Example of shear test Fig. 3 Seulement of pile with free end; model test resuit
Exemple d’essai de cisaillement Enfoncement de pieu avec pointe libre: essais sur modèle
réduit
The load applied to the pile at the head is gradually trans-
mitted to the surroundings through skin friction; at depth z in and finally
the profile it is:
d2P - K
.. (5)
dz2 EA
p , _ c [i - ( - k ^ y - *
In this case deformations resuit only from the elastic com ­
where C = i K 0y ta n <j>K. pression of the material o f the pile: because these are small it
A numerical example is shown in Fig. 4; for différent dis­ may be assumed, with good approxim ation, th at the relation
placements p the curves descr;bed by équation 4, the distribution between the skin friction and the displacement is linear. T hat
of the skin friction, and the loading diagram o f the pile are is r = or tan <f>(p/p0). Then, again in the case o f a bored pile,
presented.
(3) The other lim iting case is the point-bearing pile with a d 2P K v „ ,1
_ = _ t f û r tan 4 ,-p z
flxed pile point, Fig. 5. I f p is the displacement of the pile at

/=70m K ;h20m

Fig. 4 Numerical example: friction pile with free end

Exemple numérique, pieu avec point libre
47
Let and
K v t ,1
— K o y tan<t>- = « 1 + — l3 H------- l6 +
+ 3 .4 + 3 .4 . 6 . 7
An — — l-
d 2p
— apz - 0 (6) 1+ + 2 X J 7 6 /6 + • ' •
dz2
A fter the déterm ination o f the constants the vertical dis­
The général solution o f the differential équation 6 is :
placem ent o f the différent points o f the pile can be computed
from équation 7; and from these displacements the shearing
P = A»(l + 2V + 2 . 3 . 5.6 2 . 3 . 5 . 6 . 8 .9 -) stresses. Finally the p roportion o f the force P acting on the
pile head, which is transm itted to the fixed pile point, can be
calculated.
7Z4 + tz7 + +
3 .4 . 6 . 7 3 .4 .6 .7 .9 .1 0 Fig. 6 shows a num erical example : the displacements of a
(7) / = 10 m long tim ber pile, loaded a t the pile head with P = 4 0 1.
T he displacements o f the pile are so small th a t the application
o f the linear Iaw o f shear is justified. The total skin friction is
scarcely 15 per cent o f P. The distribution o f the skin friction
follows a parabolic law, the force in the pile decreases with
depth as friction becomes mobilized by the elastic displacement
o f the pile.
(4) N o w let us examine the case when the pile penetrates as
a rigid colum n into the soil a t its point. According to G e u z e
(1955) the value o f the point résistance cannot be calculated
from physical characteristics o f the soil because the necessary
physical an d mechanical quantifies are n o t measurable
practically. O n the other hand, according to the empirical
d ata o f K e v e r l i n g -B u is m a n (1955), it m ay be assumed th a t the
point résistance varies in direct p roportion to the depth and
pile diam eter. However, th a t supposition is applicable only
if we consider a sounding rod surrounded with a pipe and
loaded a t its point, w ith the skin friction eliminated. In the
casa o f an em bedded pile, the load transm itted through the
Fig. 5 Stresses acting on the periphery of a pile; fixed pile point periphery increases the vertical stress acting in the plane of
condition pile point and hence the point résistance is in direct proportion
Contraintes latérales sur la surface de pieu; la pointe étant to th a t stress an d n o t to the depth. Let us suppose as an
fixé approxim ation th a t the force transm itted through the skin
friction causes a uniform ly distributed load on a circle in the
T he boundary conditions are : plane o f the pile point having a radius m in = 5, B = 2r). Then.
equilibrium requires,
dp P_
(a) 2 = 0, T7A £0
EA 4P.
P = P p + P . = apB^ly + (9)
(b) z = /, p = 0 (8) n2B 27r) + P s

T h at is where
A \ = e0 the vertical displacement o f the head of the embedded P s = K qB t t - tan </>y[l - exp k — 3— j j
pile; P0

Displacement cm Skin friction-C ton/m 2 pz ton

0-1 0-2 0-3 0-4 0 0-4 0-8 10 0 10 20 30 40
f /
/
Embedded pile- >
Y /
// N
/ ,/ \ j

! / /
/ Free -standin 1 pile
C
.O
// ' rto r ü
Point re~iistance
.O
an 0=0-7 .C
70 / y z 1-7ton/m3
Point résistance
Timber pile; 1=10m v K=1-20rr p 0 - 0-5 cm
a E =100,000 kg/cm2
Fig. 6 Numerical example; pile point fixed
Exemple numérique, la pointe du pieu est fixé
48
The load-displacem ent curve o f the pile can be drawn from takes place, the am ount being equal to the volume of the pile.
équation 9 ; Fig. 7 shows an example. The diagram, as in the In conséquence it is no longer the static pressure th at acts on
case of actual test diagrams, continues to slope and there is no the surface of the pile, but a greater one, depending on the
gênerai shear failure. The relation shows that point résistance diameter of the pile and on the initial compactness o f the sand.
and skin friction depend on each other; both are functions of
displacement. W hen p > p 0 the skin friction is fully mobilized 0 20 40 60 eo 100 %

|, Brown
Pile load tons
sand
ï—__ _
t
\ Yellow
* sand
t
t

Sand Silt
100 -

80-
>
5 60-
,o>
É40- vJ>> th r

20-
0. '
70-e 0-5 0-2 0-1 0 0S005 0 02
Parlicle size D mm
Fig. 9 Physical data of the soil at the test pile site
Fig. 7 Load-settlement curve of a pile with both point résistance Caractéristiques physiques du sol du chantier des essais
and skin friction
Enfoncement d’un pieu avec résistance à la pointe et frotte­
ment latérale

and cannot increase; point résistance is then proportional to

pénétration. The factor of proportionality a, is a pure number,
a function of the angle o f internai friction and a modulus
of elasticity. The establishm ent of a strict mathematical rela-
tionship seems impossible for the tim e being. It m ust be
emphasized th a t équation 9 applies only to the case where

Fig. 8 Stresses acting on the periphery of a driven pile

Contraintes latérales sur la surface d’un pieu battu

cohesionless soil surrounds the pile. Piles driven through clay Fig. 10 The shortest pile dismantled
soils with the pile points bearing in sand or gravel call for in­ Le pieu le plus court démonté
vestigation o f the three-dimensional failure beneath pile points ;
in this case the skin friction is of less im portance (see case The distribution o f the norm al stress will be according to Fig. 8 ;
treated in sub-section (3) and proportionality between point near to the surface the passive stress State will be approached,
résistance and displacement cannot be assumed. and in the deeper layers an approxim ately constant norm al
(5) Around driven piles substantial com paction o f the soil stress will be added to the static pressure. If the radius o f the
49
cylinder, within which displacements take place in conséquence take place. O n unloading the settlem ent increased a little,
o f driving, is denoted by R (this value being proportional to the and subsequently elastic rebound occurred. The force corre-
radius r o f the pile) the volume strain is e = r2IR 2 = 1//?2, from sponding to m axim um pénétration was regarded as the ultimate
which the pressure acting on the surface o f the pile is already load. The constants in the theoretical équation o f the loading
calculable by the Theory o f Elasticity or by m aking use o f the curve were com puted from one o f the experim ental curves;
findings o f S t e i n f e l d (1953).

Ultimate Load Capacity of an Individual Pile

Large m odel tests carried out in the field were used to in-
vestigate the ultim ate load capacity. Concrete piles o f dif-
P/le load P ton

Ultimate

Point

4 xultimate
load of the
single pile

4xpoint résistance
of the single

20 30 40 50 60 cm
Distance between pile centres,d

Fig. 13 Ultimate loads of pile groups

Force de rupture des groupes de pieux

Fig. 11 Load-settlement curve of single piles and hence the loading diagram s pertaining to the other pile
Diagramme de charge des pieux individuels lengths were calculated. F o r piles o f 1 -20 an d 1 -60 m length
the com puted and m easured d ata showed good agreement. In
ferent lengths (/ = 0-40; 0-80; 1 -20; 1 -60 and 2 0 m) o f 10 x 10 the case o f the shorter piles the assum ptions do n o t hold ; when
cm 2 cross-sectional area were driven into hom ogeneous fine
grained sand the physical d ata o f which is given in Fig. 9. The

Pile length m Fig. 14 Settlement of pile groups

Fig. 12 Ultimate loads of piles with différent length Enfoncement des groupes de pieux
Force de rupture des pieux avec des longueurs différentes
/ < 8 B no real pile effect occurred. In practical cases a load­
pile points could be loaded separately. The shortest pile is ing test is the best way o f determ ining the constants.
show n dism antled in Fig. 10. The ultim ate load interpreted according to the foregoing is
Two characteristic loading diagrams are shown in Fig. 11. show n in Fig. 12 as a function o f the pile length. Measured
The shape o f the curve corresponds to the case given under sub­ values show here the same regularity as laboratory model tests
section 4 in the previous section; général shear failure did n ot ( K e z d i , 1955); beyond a critical depth, which is about 8 times

50
the pile diam eter; the relationship can be represented by a the adjacent piles increases the vertical stresses and failure side-
straight line. ways is less likely to occur. The group action discontinues for
piles driven a t d = 6 B centres.
Bearing Capacity of Pile Groups Fig. 14 shows the magnitude of loads causing given pile
There are few experimental data or little theoretical research settlements. The less the distance between pile centres, the
on the bearing capacity o f pile groups in the literature. The smaller will be the settlement under a given load; a t d = 6B
qualitative statem ent generally accepted is that in the case o f a between centres the loading diagram approaches the line giving
point-bearing pile the bearing capacity o f the pile group is the fourfold loading of the individual pile as a function of its
n times th at o f the individual pile, while in the case of friction settlement. The groups thus behaved very favourably from
piles it is less than th a t value. the point of view of settlement, m ost probably due to soil com-
To further the solution o f the problem field tests in addition paction caused by the driving.
to laboratory experiments were extended to the investigation of The above facts, o f course, cover only the case when there is
pile groups. F o u r piles (/ = 2-0 m ; A = 10 x 10 cm 2) were no more compressible layer beneath pile points. In th at case,
driven in a row, and a t the corners o f a square, and loading the vertical stresses transm itted at the level o f a close group of
diâgrams were taken a t différent pile spacings. The relation pile points will penetrate to much greater depths than below a
between ultim ate load o f the group and the pile centres is shown single pile and cause greater settlement.
in Fig. 13; the horizontal line dénotés the fourfold bearing
capacity o f the single pile. Both in the case of piles driven in a The author wishes to express his sincere thanks to Research
row and at the corners o f a square, the bearing capacity of the Engineer H . Herzog fo r carrying out the experiments and super-
group is substantially higher than th a t; only at d = 6B does the vizing the field tests.
group capacity reduce to th a t o f the single pile. A t d = 2B the
increase is m ore than 100 per cent. References
The explanation is th a t in the case o f piles driven closely to G e u z e , E. C. W. A. (1955). Fortschritte unserer Kenntnisse über
one another the soil core enclosed between the piles is sub­ die Tragfâhigkeit von Pfahlspitzen. Vortràge der Baugrundtagung
1954 in Stuttgart. Hamburg
stantially com pacted, and moves along with the piles and a K é z d i , Â . (1955). Über die Tragfâhigkeit und Setzung von Pfahl-
larger loaded surface develops in the plane of pile points. gründungen. Gedenkbuch fiir Professor J. Jâky. Budapest
Therefore the law established for sand is valid : the ultimate load S t e i n f e l d , K . (1953). Ueber den râumlichen Erdwiderstand.
increases with the w idth o f the loaded area. Superposition of Mitteilungen der Hannoverschen Versuchsanstalt fur Grundbau und
Wasserbau, Heft 3
piles is therefore favourable from the point o f view of ultimate K e v e r l i n g -B u i s m a n , A. S. (1955). Vortràge der Baugrundtagung 1954
bearing capacity; in the plane of the pile point the loading of in Stuttgart. Hamburg

51