You are on page 1of 11

Studia Geotechnica et Mechanica, Vol. 37, No.

4, 2015
DOI: 10.1515/sgem-2015-0048

PILE LOAD CAPACITY – CALCULATION METHODS

BOGUMIŁ WRANA

Civil Engineering Department, Institute of Structures Mechanics, Soil-Structure-Interaction Branch,
Cracow University of Technology, Warszawska 24, 31-155 Kraków, Poland,
e-mail: wrana@limba.wil.pk.edu.pl

Abstract: The article is a review of the current problems of the foundation pile capacity calculations. The article considers the main
principles of pile capacity calculations presented in Eurocode 7 and other methods with adequate explanations. Two main methods
are presented: α – method used to calculate the short-term load capacity of piles in cohesive soils and β – method used to calculate
the long- term load capacity of piles in both cohesive and cohesionless soils. Moreover, methods based on cone CPTu result are pre-
sented as well as the pile capacity problem based on static tests.

Key words: pile load capacity calculation, Eurocode 7, α – method and β – method, direct methods based on CPTu data

1. INTRODUCTION The pile load capacity on compression (Fig. 1a–1c)
is considered in the article, in particular the sufficient
compressing resistance case (Fig. 1a).
Piles can be either driven or cast in place. Pile
driving is achieved by: impact dynamic forces from
hydraulic and diesel hammers; vibration or jacking.
Concrete and steel piles are most common. Driven
piles which tend to displace a large amount of soil due
to the driving process are called full-displacement
piles. Cast-in-place (or bored) piles do not cause any
soil displacement, therefore, they are non-displace-
ment piles.
Piles may be loaded axially and/or transversely.
The limit states necessary to be considered in the
design of piles are the following (EN-1997-1,
§7.2.(1)P):
• Bearing resistance failure of the pile foundation,
• Insufficient compression resistance of the pile
(Fig. 1a), Fig. 1. Piles load capacity:
• Uplift or insufficient tensile resistance of the pile (a)–(c) on compression, (d), e) on tension,
(Fig. 1f), Figure 2a shows the following main parameters
• Structural failure of the pile in compression (Fig. used in the pile capacity problem:
1b), tension (Fig. 1e), bending (Fig. 1g), buckling • s(Q) – load-settlement top pile data recorded in the
(Fig. 1c) or shear (Fig. 1h), in-situ test on compression,
• Combined failure in the ground, in the pile foun- • sk – characteristic settlement, generally calculated
dation and in the structure, using the assumption on soil behavior as: semi-
• Excessive settlement, heave or lateral movement, infinite elastic, isotropic and homogenous area
• Loss of overall stability, (Boussinesq theory), which gives such larger set-
• Unacceptable vibrations. tlement than the measured one,

Unauthenticated

α – adhesion coefficient depending on pile mate.70–0. ⎪ 4 or su / σ v′ > 1. Proposition for α coef- ficient depends on type of pile (Table 1) Table 1. On the stress and undrained shear strength but decreases for other hand. Pile’s side friction (shaftor skin friction) and end bearing Very soft 0–12 1. Usually. Ip.92 5.5 for cu ≥ 70 kPa. WRANA tween their surface and the surrounding soils. UNIT SKIN FRICTION qs(z) Steel piles Medium stiff 24–48 0. capacity of the soil underlying their bases.2 (1984). σ v′ . 1987) suggests values for α as a function of cu as follows ⎧ cu − 25 ⎪1 − 90 for 25 kPa < cu < 70 kPa. d = Qb + Qs = Ab ⋅ qb + ∑A s . SHORT-TERM LOAD Stiff 48–96 0. stress his- Design bearing capacity (resistance) can be de.0 for cu ≤ 25 kPa.96–0. i .48 CAPACITY FOR COHESIVE SOIL Very stiff 96–192 0. Niazi and Mayne [24] presented 25 methods of end-bearing piles are used to transfer most of their estimating pile unit shaft resistance within α-method loads to a stronger stratum that exists at a reasonable and compared them. ⎧ 1 rial and clay type.00–0.00 Soft 12–24 1.2 method. Kolk and Van der Velde sive soil as follows and the interface shear stress qs method [18].86 B. In real. the skin friction is dependent on the effective ⎪⎩ 2 su / σ v′ Unauthenticated Download Date | 3/8/16 7:28 PM . fined as Belowthe main methods estimating skin friction in claysare shown: Rc .00–0. 1984. d . undrained shear strength (NAVFAC DM 7. ity.75–0. method. thus.1. 3. α-METHOD. They showed main differences depth.2) Soil Undrained shear Pile type α consistency strength su [kPa] Fig. α vs.75 concrete piles 5. ⎪2 or su / σ v′ ≤ 1.i ⋅ qs .36–0.48–0.96 Timber and Medium stiff 24–48 0. 1987) The equation by API (1984. with respect to parameters: length effect. ⎪ ⎩ (b) NAVFAC DM 7. Coefficient α is based on the ratio between the pile surface and the surrounding soil is of undrained shear strength and effective stress. end-bearing piles rely on the bearing long piles.00 Soft 12–24 1. the skin friction is assumed to be propor- (c) Equation based on undrained shear strength and tional to the undrained shear strength su.36 Very stiff 96–192 0. it is well suited for As in the API method.70 Stiff 48–96 0. (7) (a) American Petroleum Institute (API.92–0.33 Very soft 0–12 1. (9) ⎪0. of the cohe- effective vertical stress. ⎪ α = ⎨1.19 The method is based on the undrained shear strength of cohesive soils. su. determined as A large database of pile skin friction results was qs ( z ) = α ( z ) su ( z ) (8) analyzed and correlated to obtain α value (Table 2). tory. (d) Simple rules to obtain coefficient α based on where su/ σ v′ proposed standard DNV-OS-J101-2007 su – undrained shear strength. ⎪ su / σ v′ It is usually assumed that ultimate skin friction is α =⎨ (10) 1 independent of the effective stress and depth. progressive failure. plugging effect. In this neglected in the DM 7. effective stress effects are short-term pile load capacity calculations.

The unit skin resistance qs. Skin friction factor dependent on su/ σ v′ su/ σ v′ 0. 4. Measured values of α in relation to normalized strength. μ.7 0.6 1.2 0.52 0.62 0.56 0.9 2.3 0.8 0. β-METHOD.55 0. They showed main differences Unauthenticated Download Date | 3/8/16 7:28 PM .2 2.50 0. 4.41 0. The progres.39 (e) Mechanism controlling friction fatigue. K = .2.4 α 0.1 2. Pile load capacity – calculation methods 87 Table 2. Randolph Figure 4 presents mobilized values of α versus [26] sud/ σ v′ 0 for all piles discussed in this paper.3 2.60 0. [15]. σ v′ – vertical effective stress. estimating pile unit shaft resistance within β-method for all piles (Karlsrud et al.41 0.0 2.41 0.65 0. UNIT BASE RESISTANCE qB degree of softening ξ and the pile compressibility K 2 ⎛ 1 ⎞ For cohesive soils it can be shown.41 0. proposed modification of the AND COHESIONLESS SOILS NGI method by introducing correlation of sud/ σ v′ 0 and Ip with α – coefficient presented by the trend lines shown in Fig.40 0.0 (Skempton [29]).8 1.3 1.0 1.1. all 6.49 0.4 0. between the pile and the surrounding soil is calculated by multiplying the friction factor.1 1. EA – axial stiffnes of pile.6 0. OCR – overconsolidation ratio. The method is based on effective stress analysis and is suited for long-term (drained) analyses of pile load capacity.5 1. Randolph [26] proposed a reduction factor (Rf) which depends on the 5. NGI-05 LOAD CAPACITY FOR COHESIVE Karlsrud et al.2 1.4 2. New data are included herein.40 0.0 4. Δwres – post-peak displacement required to mobi- lize the residual shaft resistance.48 su/ σ v′ 1.5 ≤ 3. with K0 = (1 – sinφ′)(OCR)0. (12) τ peak Δwres bearing capacity coefficient that can be assumed equal to 9.5 3.40 0. Studies Randolph [26] suggested that progressive failure. 4. between the pile and soil by σ h′ qs ( z ) = μσ h′ = μ ( z ) K ( z )σ v′ ( z ) = β ( z )σ v′ ( z ) (14) where at rest pressure coefficient depends on the installation mode. sive failure from the peak (τpeak) to the residual (τres) shaft resistance is shown in Fig.77 0.5 0.0 α 0.39 0. using Ter- R f = 1 − (1 − ξ )⎜1 − ⎟ (11) zaghi’s bearing capacity equation.53 0. sponding α-values. that the unit base ⎝ 2 K⎠ resistance of the pile is where qb = ( su )b N c (13) τ peak π DL2 where (su)b is the undrained shear strength of the co- τ ( EA) pile hesive soil under the base of the pile. LONG-TERM (f) Norwegian Geotechnical Institute. usually K = K0.7 1.47 0. was a possible on the mobilized ultimate shaft friction and corre- mechanism controlling friction fatigue.9 1.42 0.42 0. have shown that the plasticity index has a largeimpact which occurs in strain softening soil.40 0. 6. Niazi and Mayne [24] presented 15 methods of Fig.95 0. and Nc is the ξ = res . UNIT SKIN RESISTANCE qs(z) previous data have been re-interpreted. [15]) and compared them.70 0.

0 (round and square) Driven displacement tapered Using Terzaghi’s bearing capacity equation. 1. β = μ (z)K(z) = tan δ (z)K(z). The main methods esti- mating skin frictionare shown below: (a) according to NAVFAC DM 7. Shear surface around the base of a pile: data.7 for compression. the 1.25 for tension (for uplift piles) β = 0.2 for φ ′ = 28°.5–2.2(1984).2. δ.5 Driven displacement piles 6.1.0 to very stiff clays with OCR of 40.35 for φ ′ = 33°.3–0.0–1. Kp = tan 2(45 +φ ′/2) (d) Karlsrud [16] Karlsrud [16] proposed to take into account the plasticity index Ip in β-method. for driven piles β = 0.75. Tables 3 and 4. Karlsrud [16] compression) under tension) Driven H-piles 0.9 0. d.35 for compression.3 piles unit base resistance at the base of the pile can be cal- Driven jetted piles 0. L.15. K. UNIT BASE RESISTANCE qb 1. φ ′. Nc = (Nq – 1)cotφ ′. Lateral earth pressure coefficient (K) Fig.5–1. 37°. Author Proposition of β value cb′ – cohesion of the soil under the base of the pile. Chart for determination of β-values K (piles under K (piles Pile type dependent on OCR and Ip. 35°.4–0. 5. OCR. Kp and K0: K = (K0 + Ka Kp)/3 where: K0 = (1 – sinφ ′). σ v′ . definition of the angle η (Janbu [13]) Unauthenticated Download Date | 3/8/16 7:28 PM . 35°. WRANA between them with respect to parameters: σ r′ .3–0. 6. ID. Figure 5 shows dia- gram of β-values from as low as 0.0 1. 0.0–1. C = 0.0 0. Table 3.5 for tension (uplift piles) (c) Average K method Earth pressure coefficient K can be averaged from Ka.6–1. which is the upper range of available pile Fig.4 in diameter) qb = (σ v′ )b N q + cb′ N c (15) where (b) proposition value of β = μ (z)K(z) can be estimated (σ v′ ) b – vertical effective stress at the base of the according to the following propositions: pile. Ka = tan2(45 – φ ′/2). Ip. Pile skin friction angle (δ) Pile type Pile-soil interface friction angle (δ) Steel piles 20o Timber piles 3/4 φ′ Concrete piles 3/4 φ′ Table 4.88 B. for bored piles coefficients Nq and Nc for various soils β = Ctan(φ′ – 5) Kraft and Lyons [19] C = 0.2.15 to 0. 37°.6 culated Bored piles (less than 60 cm 0.5 0.10 to 0. McClelland [21] β = 0.045 for low- plastic NC clays to about 2. su.7 0. 0. Values of bearing capacity factor Nq for driven piles Meyerhof [22] (a) Janbu [13] presented equations to estimate capacity β = 0. 0.

(g) Poisson’s ratio (v). • Skin friction tends to increase with depth and just Fig. Variation of other parameters with depth has not been researched thoroughly. also would reduce where η is an angle defining the shape of the shear with depth. [27]) • Skin friction does not increase linearly with depth as was once believed. CRITICAL DEPTH FOR SKIN FRICTION to 28°. Friction angle φ ′ vs. Table 5. see Table 5. The typical experimental 6. Unauthenticated Download Date | 3/8/16 7:28 PM . 8. The end bearing surface around the tip of a pile as shown in Fig. higher friction angle values are not war- ranted. Variation of skin friction (Randolph et al. (e) The dilation angle of soil (ψ). This is because water jets tend to loosen the (SANDY SOILS) soil.3. PARAMETERS THAT AFFECT variation of skin friction with depth in a pile as evi- THE END BEARING CAPACITY dence for critical depth is shown in Fig. bearing capacity of a pile with regard to relative den- (b) Values of bearing capacity factor Nq according to sity (ID) and vertical effective stress σ v′ (Randolph et NAVFAC DM 7. Figure 7 attempts to formulate the end dense sands. The capacity does not increase at the same rate as the in- angle η ranges from π/3for soft clays to 0. experimental data do not support the old theory with a constant skin friction below the critical depth. 7. Fig. (f) Shear modulus (G). This depth was named a critical depth. Hence. φ′ should be limited 6. which is a function of the friction angle. End bearing capacity of a pile above the tip of the pile to attain its maximum with regard to relative density (ID) and effective stress value. Skin friction would drop rapidly after that. The following parameters affect the end bearing capacity: (c) Effective stress at pile tip. [27]) Remarks: • As one can see. al. [17]). Pile load capacity – calculation methods 89 N q (tan φ ′ + 1 + tan 2 φ ′ ) 2 exp(2η tan φ ′) friction angle decreases with depth. (Randolph et al.2(1984). 6. It is known that the the field data.4.58π for creasing depth. Most of these parameters have been bundled into • No satisfactory theory exists at present to explain the bearing capacity factor Nq. 8. (d) Friction angle at pile tip and below (φ ′). Hence Nq. Skin friction should increase with depth and it be- comes a constant at a certain depth. Nq φ ′ [°] 26 28 30 31 32 33 34 35 36 37 38 39 40 Nq for driven piles 10 15 21 24 29 35 42 50 62 77 86 120 145 Nq for bored piles 5 8 10 12 14 17 21 25 30 38 43 60 72 If water jetting is used.

Critical depth: dc – critical depth. ESTIMATING PILE LOAD CAPACITY • Critical depth for medium dense sand = 15 D. fcd = K σ c′ tanδ. Then let us assume that the pile was driven fur. the end bearing capacity was assumed to increase till 2. INTRODUCTION Owing to the difficulties and the uncertainties in assessing the pile capacity on the basis of the soil strength-deformation characteristics. same critical depth concept adopted for skin friction Friction angle tends to decrease with depth.5 m and unit skin friction was meas. as suggested by the skin friction equation: nitely. 9 (Rajapakse [28]). 11. CRITICAL DEPTH FOR END BEARING using critical depth theory of the past. approximations were assumed for the critical depth: • Critical depth for loose sand = 10 D (D is the pile diameter or the width).2 gives maximum value of skin • Critical depth for dense sand = 20 D. friction angle.5 m is less in the second case. friction and end bearing capacity is achieved after 20 The critical depth concept is a gross approximation diameters within the bearing zone. CAPACITY (SANDY SOILS) Reasons for limiting skin friction Pile end bearing capacity in sandy soils is related The following reasons have been offered to explain to effective stress. Unit base bearing capacity and critical depth Let us assume that a pile was driven to a depth of 3 m and unit skin friction was measured at a depth of The following approximations were assumed for 1. NAVFAC DM 7. engineers use the 1. 10. Due to lack of a valid theory. 11. K value is a function of the soil friction angle (φ ′). such as ef- pile depth. engineers are still 6. bearing capacity does not increase with depth indefi- nitely.5. as forthe end bearing capacity. Example of unit skin friction distribution (Rajapakse [28]) Fig. fcd – unit skin friction cone penetration test (CPT) is one of the most fre- at critical depth. It has been reported that diameter or the width). It is clear that there is a connection stress levels due to readjustment of sand particles. the most fre- quently followed design practice is to refer to the for- mulae correlating directly the pile capacity compo- nents of qb and qs to the results of the prevalent in situ tests. Experimental data indicate that end why skin friction does not increase with depth indefi. and relative density. Hence. • Critical depth for medium dense sand = 15 D. between end bearing capacity and skin friction since 3. the critical depth within the bearing zone: ther to a depth of 4. quently used investigation tools for pile load capacity σ c′ – effective stress at critical depth evaluations. • Critical depth for loose sand = 10 D (D is the pile ured at the same depth of 1.5 m.5 m. Fig. BASED ON CPT RESULTS • Critical depth for dense sand = 20 D. 7. fective stress. 9. see Fig. Skin friction equation does not hold true at high the critical depth. Within the domain of these in situ methods. On the other hand. As shown in Fig. Ever since the first use of CPT in geo- Unauthenticated Download Date | 3/8/16 7:28 PM . WRANA • Due to lack of a better theory. the Fig. The following that cannot be supported by experimental evidence. unit skin friction at 1.90 B.1. 7. two processes are vastly different in nature. Reduction of local shaft friction with increasing the same soil properties act in both cases. K value decreases with depth (Kulhawy [20]).

accomplish axial pile capacity analysis from CPT data: (a) rational (or indirect) methods and (b) direct methods. CPT based evaluations of pile capacity (Niaziand Mayne [24]) • Direct CPT methods – used the similarity of the cone resistance with the pile unit resistances. ID). σ v′ 0 .. • Pure empirical methods – initial formulations were based solely on cone resistance (qc) derived from mechanical cone penetrometers. [2]. Ardalan et al. which eliminates the need to supplement the field data with laboratory testing and to calcu- late intermediate values. δ.2. the additional estimated parameters are mini-pile foundation. Upper and lower empirical values uring sleeve friction ( fs). φ ′. The indirect methods apply strip-footing components of qb and qs. compressibility and rigidity of the surrounding soil medium affect the pile and the cone work in a similar manner. pressibility and strain softening. and neglect soil com- As commonly reported (e. d. the additional channels meas. (a) qs(z). L. Pile load capacity – calculation methods 91 technical investigations. These methods Cai et al. Fig. 12. such as K. [5]. su. bearing capacity theories. sleeve friction ( fs). the cone penetrometer can be considered as a mini-pile foundation as noted by Ar- dalan et al. and obtained from cone data to estimate bearing ca- shoulder pore water pressure (u2) and the pile capacity pacity. Several methods mod- ify the resistance values to consider the difference in diameter between the pile and the cone. Subse- quently. (b) qb (after Kempfert and Becker [17]) Unauthenticated Download Date | 3/8/16 7:28 PM . research efforts have ad. DIRECT CPT METHODS In one viewpoint. with the introduction of the electrical cone penetrometer.g. CPT readings cone resistance (qc) or more proper such as friction angle and undrained shear strength corrected cone resistance (qt). and rigidity affect the pile and the cone in equal measure. and Nq. there are two main approaches to are rarely used in engineering practice.This concepthas led to the development of many direct CPT methods. and porewater pressures of different piles in coarse grained soils for: (u1 and u2) were considered. K. [2] and Eslami and Fellenius (1997). 13. This has resulted in plethora of taken into account (σr. [6]). soil compressibil- ity. correlative relationships being developed between the • Indirect CPT methods – employ soil parameters. Fig. The mean effective stress. The in- fluence of mean effective stress. • Semi-empirical methods – with the purely CPT pa- vanced the very elementary idea of considering it as rameters. 7. Some methods may use the cone sleeve friction in deter- mining unit shaft resistance.