In-place Pile Foundation for a Tower-building Project

CHAPTER 1

1 Introduction

Pile foundations are used to carry a load and transfer the load of a given structure to the ground bearing, which is found below the ground at a considerable depth. The foundation consists of several piles and pile-caps. Pile foundations are generally long and lean, that transfers the structure load to the underlying soil (at a greater depth) or any rock having a great load-bearing ability.

“The main types of materials used for piles are Wood, steel and concrete. Piles made from these materials are driven, drilled or jacked into the ground and connected to pile caps. Depending upon type of soil, pile material and load transmitting characteristic piles are classified accordingly.” (Pile Foundation Design: A Student Guide by Ascalew Abebe & Dr Ian GN Smith).

The objective of this project is to identify the design & use of a cast-in-place pile foundation, for the tower-building project.

The tower building project is called the Gemini Towers. The purpose of this construction (building) is to facilitate office spaces. This also resides on a rocky area. The building has been designed as per state-of-the-art designing concepts which are basically to attract foreign investors to invest in Oman. The Gemini Building has 1 basement, 1 ground and 19 floors.

Cast-in-place concrete piles are shafts of concrete cast in thin shell pipes, top driven in the soil, and usually closed end. Such piles can provide up to a 200-kip capacity. The chief advantage over precast piles is the ease of changing lengths by cutting or splicing the shell. The material cost of cast-in-place piles is relatively low. They are not feasible when driving through hard soils or rock.

1.1 Aim

The aim of this project is to design and propose cast in-place pile foundation for a tower-building project and study the efficiency for the same. To achieve this aim the following objective has to be achieved.

1.2 Objectives

The objectives of this project are as following:

1. To study the field soil condition, suitability of pile and investigate the soil.
2. To study the advantages and efficiency of using cast-in-place pile for the building.
3. To study the guidelines for the design of cast in-place structure according to BS 8004, 8110, 8002, etc.
4. To design the pile foundation as per the guidelines and the soil conditions (analyse the load, calculate the moment and determine the length and diameter and reinforcement).
5. To use computer structural designing program for performing design (CAD and STAD).

1.3 Methods

The methods followed in preparing this project is by collecting the project plan and the soil investigation report. Then after that, research has been done on in-situ pile foundation type, to identify its characteristics.

The next step is to study the pile designing criteria by referring to BS 8004, 8110 & 8002 codes to understand the guidelines, which shall be followed to accomplish the pile design. For this, the structural loads have to be analysed and identified using ultimate state design method. Then the design is processed depending on the data gathered on soil conditions, design loads and BS code guidelines.

Thus, a proposal for the suitable pile will be prepared by identifying the reasons over the proposal.

The commonest function of piles is to transfer a load that cannot be adequately supported at shallow depths to a depth where adequate support becomes available, also against uplift forces which cause cracks and other damages on superstructure.

Chapter 2 – Literature Review

2 Pile Foundation

“Pile foundations are used extensively in bridges, high-rise buildings, towers and special structures. In practice, piles are generally used in groups to transmit a column load to a deeper and stronger soil stratum. Pile may respond to loading individually or as a group. In the latter case, the group and the surrounding soil will formulate a block to resist the column load. This may lead to a group capacity that is different from the total capacity of individual piles making up the group.” (Adel M. Hanna et al, 2004).

“Pile foundations are the part of a structure used to carry and transfer the load of the structure to the bearing ground located at some depth below ground surface. The main components of the foundation are the pile cap and the piles. Piles are long and slender members which transfer the load to deeper soil or rock of high bearing capacity avoiding shallow soil of low bearing capacity. The main types of materials used for piles are Wood, steel and concrete. Piles made from these materials are driven, drilled or jacked into the ground and connected to pile caps. Depending upon type of soil, pile material and load transmitting characteristic piles are classified accordingly.” (Ascalew Abebe et al, 2005)

2.1 Functions of Piles

The purposes of pile foundations are:

1. to transmit a foundation load to a solid ground.
2. to resist vertical, lateral and uplift load.

“A structure can be founded on piles if the soil immediately beneath its base does not have adequate bearing capacity. If the results of site investigation show that the shallow soil is unstable and weak or if the magnitude of the estimated settlement is not acceptable a pile foundation may become considered. Further, a cost estimate may indicate that a pile foundation may be cheaper than any other compared ground improvement costs. Piles can also be used in normal ground conditions to resist horizontal loads. Piles are a convenient method of foundation for works over water, such as jetties or bridge piers.” (Pile Foundation Design: A Student Guide, by Ascalew Abebe & Dr Ian GN Smith, 2003).

2.2 Classification of Piles

2.2.1 Classification of pile with respect to load transmission

• End-bearing.
• Friction-piles.
• Mixture of cohesion piles & friction piles.

2.2.1.1 End bearing piles

This type of piles is designed to transfer the structural load to a stable soil layer which is found at a greater depth below the ground. The load bearing capacity of this stratum is found by the soil penetration resistance from the pile-toe (as in figure 1.2.1.1).

The pile normally has attributes of a normal column, and should be designed as per the guidelines. The pile will not collapse in a weak soil, and this should be studied only if a part of the given pile is unsupported. (Eg: If it is erected on water / air). Load transmission occurs through cohesion / friction, into the soil. At times, the soil around the pile may stick to the pile surface and starts “negative skin friction”. This phenomenon may have an inverse effect on the pile capacity. This is mainly caused due to the soil consolidation and ground water drainage. The pile depth is determined after reviewing the results from the soil tests and site investigation reports.

2.2.1.2 Friction piles (cohesion)

The bearing capacity is calculated from the soil friction in contact with the pile shaft. (as in Figure 1.2.1.2).

2.2.1.3 Mixture of cohesion piles & friction piles.

This is an extended end-bearing pile, when the soil underneath it is not hard, which bears the load. The pile is driven deep into the soil to create efficient frictional resistance. A modified version of the end-bearing pile is to have enlarged bearing base on the piles. This can be achieved by immediately pushing a large portion of concrete into the soft soil layer right above the firm soil layer, to have an enlarged base. Similar result is made with bored-piles by creating a bell / cone at the bottom by the means of reaming tools. Bored piles are used as tension piles as they are provided with a bell which has a high tensile-strength. (as in figure 1.2.1.3)

2.3 Cast-in-Place Pile Foundation

Cast-in-place piles are installed by driving to the desired penetration a heavy-section steel tube with its end temporarily closed. A reinforcing cage is next placed in a tube which is filled with concrete. The tube is withdrawn while placing the concrete or after it has been placed. In other types of pile, thin steel shells or precast concrete shells are driven by means of an internal mandrel, and concrete, with or without reinforcement, is placed in the permanent shells after withdrawing the mandrel.

2.3.1 Advantages

• Length of the pile can be freely altered to cater varying ground conditions. Soil removed during the boring process can be verified and further tests can be made on it.
• Large diameter installations are possible.
• End enlargements up to two or three diameters are possible in clays.
• Pile materials are independent during driving / handling.
• Can be installed to greater depths in the soil.
• Vibration-free and noise-free while installation.
• Can be installed in conditions of very low headroom.
• Ground shocks are completely nil.

2.3.2 Disadvantages

• Susceptible to “necking” or “wasting” in pressing ground.
• Concrete is not pumped under suitable conditions and cannot be inspected.
• The cement on the pile shaft will be washed up, if there is a sudden surge of waster from any pressure caused underground.
• Special techniques need to be used to ensure enlarged pile ends.
• Cannot be easily prolonged above ground-level especially in river and marine structures.
• Sandy soils may loosen due to boring methods and base grouting may be required for gravely soils to improve base resistance.
• Sinking piles may result in ground-loss, leading to settlement of nearby structures.

CHAPTER 3

3 Load Distribution

To a great extent the design and calculation (load analysis) of pile foundations is carried out using computer software. The following calculations are also performed, assuming the following conditions are met:

1. The pile is rigid.
2. The pile is pinned at the top and at the bottom.
3. Each pile receives the load only vertically (i.e. axially applied).
4. The force P acting on the pile is proportional to the displacement U due to compression.

Therefore, P = k U

Since P = E A

E A = k U

k = (E A ) / U

Where:

P = vertical load component

k = material constant

U = displacement

E = elastic module of pile material

A = cross-sectional area of pile (Figure 3 – load on single pile)

The length L should not necessarily be equal to the actual length of the pile. In a group of piles. If all piles are of the same material, have same cross-sectional area and equal length L, then the value of k is the same for all piles in the group

3.1 Pile foundations: vertical piles only

3.1.1 Neutral axis load

The pile cap is causing the vertical compression U, whose magnitude is equal for all members of the group. If Q (the vertical force acting on the pile group) is applied at the neutral axis of the pile group, then the force on a single pile will be as follows:

Pv = Q / n

Where

Pv = vertical component of the load on any pile from the resultant load Q

n = number of vertical piles in the group (see figure 3.1.2)

Q = total vertical load on pile group

3.1.2 Eccentric Load

If the same group of piles are subjected to an eccentric load Q which is causing rotation around axis z (see fig 3.1b); then for the pile i at distance rxi from axis z:

Ui = rxi . tanθ

∴ Ui = rxi θ => Pi = k . r xi . θ

θ is a small angle ∴ tanθ ≈ θ (see figure 3.1.2).

Pi = force (load on a single pile i).

Ui = displacement caused by the eccentric force (load) Q.

rxi = distance between pile and neutral axis of pile group.

rxi positive measured the same direction as e and negative when in the opposite direction.

e = distance between point of intersection of resultant of vertical and horizontal loading with underside of pile.

(Figure 3.1.2 – Example of a pile foundation – vertical piles)

The sum of all the forces acting on the piles should be zero ⇔

⇔

Mxi = Pi . rxi = k . rxi . θ rxi = k . θ r2xi =>

=> Mxi =

From previous equation,

Mz = ΣMz

Applying the same principle, in the x direction we get equivalent equation. If we assume that the moment MX and MZ generated by the force Q are acting on a group of pile, then the sum of forces acting on a single pile will be as follows:

If we dividing each term by the cross-sectional area of the pile, A, we can establish the working stream σ:

CHAPTER 4

4 Load on Pile

4.1 Introduction

“Piles can be arranged in a number of ways so that they can support load imposed on them. Vertical piles can be designed to carry vertical loads as well as lateral loads. If required, vertical piles can be combined with raking piles to support horizontal and vertical forces.” (Pile Foundation Design: A Student Guide by Ascalew Abebe & Dr Ian GN Smith)

“Often, if a pile group is subjected to vertical force, then the calculation of load distribution on single pile that is member of the group is assumed to be the total load divided by the number of piles in the group.” (Pile Foundation Design: A Student Guide by Ascalew Abebe & Dr Ian GN Smith)

However, if a given pile group is subjected to eccentric vertical load or combination of lateral & vertical load that can start moment force. Proper attention should be given during load distribution calculation.

4.2 Pile Arrangement

“Normally, pile foundations consist of pile cap and a group of piles. The pile cap distributes the applied load to the individual piles which, in turn, transfer the load to the bearing ground. The individual piles are spaced and connected to the pile cap. Or tie beams and trimmed in order to connect the pile to the structure at cut-off level, and depending on the type of structure and eccentricity of the load, they can be arranged in different patterns.” (Pile Foundation Design: A Student Guide by Ascalew Abebe & Dr Ian GN Smith)

(Figure 2.2 – Pile Foundation Design: A Student Guide by Ascalew Abebe & Dr Ian GN Smith))

In this section, considering pile/soil interaction, calculations on the bearing capacity of single piles subjected to compressive axial load has been described. During pile design, the following factors should be taken into consideration:

• Pile material compression and tension capacity.
• Deformation area of pile, bending moment capacity.
• Condition of the pile at the top and the end of the pile.
• Eccentricity of the load applied on the pile.
• Soil characteristics.
• Ground water level.

4.3 The behaviour of piles under load

Piles are designed in line with the calculations based on load bearing capacity. It is based on the application of final axial-load, as per the given soil conditions at the site, within hours after the installation.

This ultimate load capacity can be determined by either:

1. The use of empirical formula to predict capacity from soil properties determined by testing. or
2. Load test on piles at the site.

When increasing compressive load is applied on the pile, the pile soil system reacts in a linear elastic way to point A on the above figure (load settlement). The pile head rebounds to the original level if the load realises above this point.

“When the load is increase beyond point A there is yielding at, or close to, the pile-soil interface and slippage occurs until point B is reached, when the maximum skin friction on the pile shaft will have been mobilised. If the load is realised at this stage the pile head will rebound to point C, the amount of permanent settlement being the distance OC. When the stage of full mobilisation of the base resistance is reached (point D), the pile plunges downwards without any farther increase of load, or small increases in load producing large settlements.” (Pile Foundation Design: A Student Guide).

4.4 Geotechnical design methods

In order to separate their behavioural responses to applied pile load, soils are classified as either granular / noncohesive or clays/cohesive. The generic formulae used to predict soil resistance to pile load include empirical modifying factors which can be adjusted according to previous engineering experience of the influence on the accuracy of predictions of changes in soil type and other factors such as the time delay before load testing.

From figure 4.1b, the load settlement response is composed of two separate components, the linear elastic shaft friction Rs and non-linear base resistance Rb. The concept of the separate evaluation of shaft friction and base resistance forms the bases of “static or soil mechanics” calculation of pile carrying capacity. The basic equations to be used for this are written as:

Q = Qb + Qs – Wp

Rc = Rb + Rs – Wp

Rt = Rs + Wp

Where:

Q = Rc = the ultimate compression resistance of the pile.

Qb = Rb = base resistance.

Qs = Rs = shaft resistance.

Wp = weight of the pile.

Rt = tensile resistance of pile.

In terms of soil mechanics theory, the ultimate skin friction on the pile shaft is related to the horizontal effective stress acting on the shaft and the effective remoulded angle of friction between the pile and the clay and the ultimate shaft resistance Rs can be evaluated by integration of the pile-soil shear strength πa over the surface area of the shaft.

τa = Ca + σ n tanφ a

Where: σn = Ks σv

∴ τa = Ca + KS σv tanφa

where:

p = pile perimeter

L = pile length

φ = angle of friction between pile and soil

Ks = coefficient of lateral pressure

The ultimate bearing capacity, Rb, of the base is evaluated from the bearing capacity theory:

Ab = area of pile base.

C = undrained strength of soil at base of pile.

NC = bearing capacity factor.

CHAPTER 5

5 Calculating the resistance of piles to compressive loads

5.1 Cast in Place Piles – Shaft resistance

These piles are installed by drilling through soft overburden onto a strong rock the piles can be regarded as end-bearing elements and their working load is determined by the safe working stress on the pile shaft at the point of minimum cross-section, or by code of practice requirements. Bored piles drilled down for some depth into weak or weathered rocks and terminated within these rocks act partly as friction and partly as end-bearing piles.

The author Duncan C. Wyllie, gives a detailed account of the factors governing the development of shaft friction over the depth of the rock socket. The factors which govern the bearing capacity and settlement of the pile are summarized as the following:

1. The length to diameter ratio of the socket.
2. The strength and elastic modulus of the rock around and beneath the socket.
3. The condition of the side walls, that is, roughness and the presence of drill cuttings or bentonite slurry.
4. Condition of the base of the drilled hole with respect to removal of drill cuttings and other loose debris.
5. Layering of the rock with seams of differing strength and moduli.
6. Settlement of the pile in relation to the elastic limit of the side-wall strength.
7. Creep of the material at the rock/concrete interface resulting in increasing settlement with time.

The effect of the length/diameter ratio of the socket is shown in Figure 5.1a, for the condition of the rock having a higher elastic modulus than the concrete.

It will be seen that if it is desired to utilize base resistance as well as socket friction the socket length should be less than four pile diameters. The high interface stress over the upper part of the socket will be noted.

The condition of the side walls is an important factor. In a weak rock such as chalk, clayey shale, or clayey weathered marl, the action of the drilling tools is to cause softening and slurrying of the walls of the borehole and, in the most adverse case, the shaft friction corresponds to that typical of a smooth-bore hole in soft clay. In stronger and fragmented rocks the slurrying does not take place to the same extent, and there is a tendency towards the enlargement of the drill hole, resulting in better keying of the concrete to the rock. If the pile borehole is drilled through soft clay this soil may be carried down by the drilling tools to fill the cavities and smear the sides of the rock socket. This behaviour can be avoided to some extent by inserting a casing and sealing it into the rock-head before continuing the drilling to form the rock socket, but the interior of the casing is likely to be heavily smeared with clay which will be carried down by the drilling tools into the rock socket.

As mentioned in Duncan C. Wyllie, suggests that if bentonite is used as a drilling fluid the rock socket shaft friction should be reduced to 25% of that of a clean socket unless tests can be made to verify the actual friction which is developed.

It is evident that the keying of the shaft concrete to the rock and hence the strength of the concrete to rock bond is dependent on the strength of the rock. Correlations between the unconfined compression strength of the rock and rock socket bond stress have been established by Horvarth(4.50), Rosenberg and Journeaux(4.51), and Williams and Pells(4.52). The ultimate bond stress, fs, is related to the average unconfined compression strength, quc, by the equation:

Where

α = reduction factor relating to, quc as shown in Figure 5.1b

β = correction factor associated with cut-off spacing in the mass of rock as shown in Figure 5.1c.

The curve of Williams and Pells in Figure 5.1b is higher than the other two, but the β factor is unity in all cases for the Horvarth and the Rosenberg and Journeaux curves. It should also be noted that the α factors for all three curves do not allow for smearing of the rock socket caused by dragdown of clay overburden or degradation of the rock.

The β factor is related to the mass factor, j, which is the ratio of the elastic modulus of the rock mass to that of the intact rock as shown in Figure 5.1d. If the mass factor is not known from loading tests or seismic velocity measurements, it can be obtained approximately from the relationships with the rock quality designation (RQD) or the discontinuity spacing quoted by Hobbs (4.53) as follows:

5.2 End Bearing Capacity

Sometimes piles are driven to an underlying layer of rock. In such cases, the engineer must evaluate the bearing capacity of the rock. The ultimate unit point resistance in rock (Goodman, 1980) is approximately.

N = tan2 (45 + / 2)

qu = unconfined compression strength of rock

= drained angle of friction

Table 5.2a

Table 5.2b

The unconfined compression strength of rock can be determined by laboratory tests on rock specimens collected during field investigation. However, extreme caution should be used in obtaining the proper value of qu, because laboratory specimens usually are small in diameter. As the diameter of the specimen increases, the unconfined compression strength decreases – a phenomenon referred to as the scale effect. For specimens larger than about 1 m (3f) in diameter, the value of qu remains approximately constant.

There appears to be fourfold to fivefold reduction of the magnitude of qu in the process. The scale effect in rock is caused primarily by randomly distributed large and small fractures and also by progressive ruptures along the slip lines. Hence, we always recommend that:

The above table (Table 5.2a) lists some representative values of (laboratory) unconfined compression strengths of rock. Representative values of the rock friction angle are given in the above table (Table 5.2b).

A factor of safety of at least 3 should be used to determine the allowable point bearing capacity of piles. Thus:

CHAPTER 6

6 Pile Load Test (Vesic’s Method)

A number of settlement analysis methods for single piles are available. These methods may be broadly classified into three categories:

1. Elastic continuum methods
2. Load–transfer methods
3. Numerical methods

Examples of such methods are the elastic methods proposed by Vesic (1977) and Poulos and Davis (1980), the simplified elastic methods proposed by Randolph and Wroth (1978) and Fleming et al. (1992), the nonlinear load–transfer methods proposed by Coyle and Reese (1966) and McVay et al. (1989), and the numerical methods based on advanced constitutive models of soil behaviour proposed by Jardine et al. (1986). In this paper, three representative methods are adopted for the calibration exercise: the elastic method proposed by Vesic (1977), the simplified analysis method proposed by Fleming et al. (1992), and a nonlinear load–transfer method (McVay et al. 1989) implemented in program FB-Pier (BSI 2003).

In Vesic’s method, the settlement of a pile under vertical loading, S, includes three components:

S = S1 + S2 + S3

Where:

• S1 is the elastic pile compression.
• S2 is the pile settlement caused by the load at the pile toe.
• S3 is the pile settlement caused by the load transmitted along the pile shaft.

If the pile material is assumed to be elastic, the elastic pile compression can be calculated by:

S1 = (Qb + ξQs)L / (ApEp)

Where Qb and Qs are the loads carried by the pile toe and pile shaft, respectively; Ap is the pile cross-section area; L is the pile length; Ep is the modulus of elasticity of the pile material; and ξ is a coefficient depending on the nature of unit friction resistance distribution along the pile shaft. In this work, the distribution is assumed to be uniform and hence ξ = 0.5. Settlement S2 may be expressed in a form similar to that for a shallow foundation.

S2 = (qbD / Esb) (1-v2)Ib

Where:

• D is the pile width or diameter
• qb is the load per unit area at the pile toe qb = Qb /Ab
• Ab is the pile base area
• Esb is the modulus of elasticity of the soil at the pile toe
• ѵ is Poisson’s ratio
• Ib is an influence factor, generally Ib = 0.85

S3 = (Qs / pL) (D / Ess) (1 – ѵ2) Is

Where:

• p is the pile perimeter.
• Ess is the modulus of elasticity of the soil along the pile shaft.
• Is is an influence factor.
• The influence factor Is can be calculated by an empirical relation (Vesic 1977).
• Is = 2 + 0.35 √(L/D)

With Vesic’s method, both Qb and Qs are required. In this report, Qb and Qs are obtained using two methods. In the first method (Vesic’s method I), these two loads are determined from a nonlinear load–transfer method, which will be introduced later.

In the second method (Vesic’s method II), these two loads are determined using empirical ratios of Qb to the total load applied on pile Q based on field test data. Shek (2005) reported load–transfer in 14 test piles, including 11 piles founded in soil and 3 piles founded on rock. The mean ratios of Qb /Q for the piles founded in soil and the piles founded on rock are summarized in Table 3 and applied in this calibration exercise. The mean values of Qb /Q at twice the design load and the failure load are very similar. Hence, the average of the mean values is adopted for calibration at both twice the design load and the failure load.

In the Fleming et al. method, the settlement of a pile is given by the following approximate closed-form solution (Fleming et al. 1992):

Where:

n = rb / r0, r0 and rb are the radii of the pile shaft and pile toe, respectively (for H-piles, πro2 = πrb2 = Dh, h is the depth of the pile cross-section)

ξG = GL/Gb, GL is the shear modulus of the soil at depth L, and Gb is the shear modulus of the soil beneath the pile toe.

ρ = Gave/GL, Gave is the average shear modulus of the soil along the pile shaft

p is the pile stiffness ratio

p = Ep / GL;

ζ = ln{[0.25 +(2.5ρ(1 – v) –0.25) ξG] L/r0};

É¥L = (2/)1/2(L/r0). If the slenderness ratio L/r0 is less than 0.5p1/2 (L/r0) the pile may be treated as effectively rigid and eq. [7] then reduces to:

If the slenderness ratio L/r0 is larger than 3πp1/2, the pile may be treated as infinitely long, and eq. [7] then reduces to:

In this case, GL’ is the soil shear modulus at the bottom of the active pile length Lac, where Lac = 3r0p1/2.

In the nonlinear load–transfer method implemented in FB-Pier, the axial –Z curve for modelling the pile–soil interaction along the pile is given as (McVay et al. 1989)