Stewart Makura.ground Improvement Using Compaction Grouting And Dynamic Compaction Techniques

GROUND IMPROVEMENT USING COMPACTION GROUTING AND DYNAMIC COMPACTION TECHNIQUES

By Stewart Makura

2

1. INTRODUCTION There has been a significant increase on the use of geotechnical techniques to provide solutions in problematic ground conditions since the end of the Second World War. This has been largely due to the increasing need to develop derelict and marginal land and the increasing shortage of good ground as well as pressures imposed by the unchecked urban growth. These changes and demands have prompted better understanding of soil behaviour by geotechnical engineers and scientists as well as improvement and development of plant equipment and techniques by specialist contractors.

Soil improvement techniques are of growing importance in the solution of land reclamation, complicated foundation solutions and for remedial purposes in areas where the land has experienced significant settlements beneath existing structures. The most commonly used techniques applied in soil improvement can be divided into two major groups viz. • temporary soil improvement techniques and • permanent soil improvement techniques.

Temporary soil improvement techniques are generally limited to the period of construction and the original soil conditions are reinstated on completion of the construction. These methods are usually applied to provide access and to create reasonable working conditions on a site and are particularly associated with water retention in areas with shallow water tables. Examples of most commonly applied temporary soil improvement techniques include water table lowering, ground freezing and electro-osmosis.

Permanent soil improvement techniques are usually associated with densification, improvements on the bearing capacity, heave reduction and the general improvement of geomechanical properties the soil. Permanent soil improvement techniques can be sub divided into two sections, the first involves techniques, which are applied to the soil to improve the natural state of the soil without the addition of materials. The second technique involves the addition

3

of imported materials natural or synthetic to improve the quality of the soil for the purpose of construction.

The earlier method of soil improvement generally involves energy transfer by means of pounding or vibration. Examples that are commonly used and belong to this category include surface compaction, heavy tamping, and compaction by explosives, deep compaction using vibratory probes, and soil improvement using thermal treatment. Whilst the latter method involves the addition of materials to achieve the desired effect. Most commonly used examples include lime and cement stabilisation, gravel or sand columns, soil replacement, pre loading coupled with vertical drains, grouting of the soil layer and earth reinforcement.

This project describes in detail two techniques applied to achieve permanent soil improvement. These are namely compaction grouting and dynamic compaction.

As discussed in the

paragraph above the first method involves soil improvement by adding materials whilst the latter method does not involve addition of material to the soil.

The first chapter describes and discusses the compaction grouting technique, dealing with the development of the technique, the theoretical aspects, applications, control and assessment of the improvement. Chapter 2 will cover the dynamic compaction technique, its development, theory, applications and assessment of improvement. Chapter 4 presents a case study where the two methods of ground improvements were jointly applied.

1

CHAPTER 1

COMPACTION GROUTING

1.

INTRODUCTION

“Grouting may be defined as the injection of appropriate materials under pressure into certain parts of the earth’s crust through specially constructed holes in order to fill and therefore seal voids, cracks, seams, fissures or other cavities in soils or rock strata.”(Bowen, 1981) The term grouting is also widely applied to sealing of cracks in man made structures such as dams, tunnels and mine shafts.

In an attempt to achieve the intended purpose of the grouting exercise, boreholes are drilled into the formation and the grout is injected under pressure to satisfy conditions specified in the design. Grouting has been put to a variety of applications in civil engineering, which include; •

Reduction of formation permeability under foundations of water retaining structures, to control seepage and loss of stored water.

•

To check the uplift on the structure and /or to prevent the danger of erosion of soil from the foundation.

•

To increase the strength of material below the foundation of heavy structures and or to reduce the deformability of the material in the foundation.

•

Fixing of reinforcing cables in precast and prestressed concrete structures.

•

Fixing of rock prestressing anchors.

•

Lifting and erection of leaning structures and buildings and other applications. (Nonveiller E, 1989)

The intended result from a grouting exercise is to ensure an improvement of the general soil properties thus reducing movement that may adversely affect civil engineering structures. It may also be applied to remedy a situation where settlement has already occurred as a form of

2

repair. There are several methods of grouting which are in use throughout the world. The different types of grouting methods differ largely in the equipment and grout placement techniques as well as the constituents of the grout and the ground conditions. Available techniques can be applied to filling up of fissures or cavities in rock.

The ideal in grouting of soils is to place the correct grout quantity of the correct quality the correct place. Several types of grout are used and these include cement, cement and sand, slag cement, resin gypsum cement, clay asphalt, pulverised (fuel) fly ash (PFA), and a large number of colloidal and low viscosity chemicals. There are numerous techniques applied in soil grouting which include fissure grouting, a technique whereby the grout introduced under pressure to the soil and either creates the fissures or finds weaker zones, which are subsequently filled up with the grout. Fissure grouting is not commonly applied in the grouting of soils, with exceptions to its use in installation of reinforcement anchors and in wad filled fissures found dolomite rock. Pressure grouting through an injection tube and jet flow which disturbs the soil in situ with simultaneous intensive mixing of the grout with a resultant permeation into the existent void spaces is a technique coined “jet grouting”. The jet technique is considered as the ideal method for providing support to existing structures on sands where working space is limited. It is also used to create cut offs below earth dams and in the stabilisation against slope failure by providing impermeable barriers especially below the water level.

The other type of grouting which is the main concern of this chapter is compaction grouting. Compaction grouting is a relatively newer method of grouting which relies not on the infilling, but rather on the densification of suitable soils by displacement in order to overcome settlement and other problems.

The grouting technique and grouting materials to be applied for a specific problem are largely depended on the ground conditions and intended use of the ground. Only then is one in a position to determine the most appropriate technique, also taking into consideration the economic constraints, which may have an effect. It is therefore of utmost importance that a complete geotechnical investigation be completed prior to the decision making.

3

1.1

HISTORICAL DEVELOPMENT OF THE COMPACTION GROUTING

TECHNIQUE

Compaction grouting has been defined by the ASCE Geotechnical Engineering Division’s Committee on Grouting (1980) as “Compaction Grout - Grout injected with less than 25 mm slump. Normally a soil cement with sufficient silt sizes to provides plasticity together with sufficient sand sizes to develop internal friction. The grout generally does not enter soil pores but remains in a homogeneous mass that gives a controlled displacement to compact loose soils, gives controlled displacement for lifting structures or both.”

The concept of compaction grouting entails the controlled injection of grout masses under pressure to radially and laterally displace the surrounding soils without permeating or hydrofracturing the soil. This procedure has been likened to blowing up balloons in the soil with hydraulic pressure. Compaction grouting does not depend on grout entering the void space in the soil, but rather on the displacement and compaction of soils resulting from intrusion of this mass of thick grout. Although some grout may enter large openings in some soils, the operation of this method does not depend on this happening.

The technique of

compaction grouting evolved from the early developments of “mud jacking” which was applied to level settled concrete pavements in the USA.

The Koehring Mudjack was

designed in 1934 to pump a Portland cement silty sand mix for the maitenance of concrete highways. It was widely applied among other things to raise slabs on small columns of very viscous grout to avoid the difficulties of sealing continually growing voids at the edges of such slabs.

It was during similar mudjacking exercises that specialist contractor Robert E Leninhan of Long Beach California, USA who pumped the “mud mix”through pipes driven to various depth who noted that more grout was actually necessary than the calculated amount for the volume displaced. He then concluded that the surrounding soil was being compacted, and the variation in the quantity of grout pumped before the uplift of settled structures under treatment was a function of the in situ density of the soil. Since then an empirical method for grouping grout bulb injections optimally so as to compact low density and compressible soil masses through the process now known as compaction grouting was devised.

4

From the time of invention of the Koehring Mudjack which used a chopping pug mixer to give a uniform mix of plastic materials and the feed system to the piston pumps. This set of equipment had major drawbacks with its pressure capability, which was capable of generating only 1725 kN/m2 which, could not pump the viscous grout further than 7,5 m from the machine. The later model the Railroad Mudjack was able to give pressures of up to 2760 kN/m2 that could permit the grout to be pumped through 30 m. This model was however hampered by its operational procedure whereby the operator adds water as the constituents of the grout mix are fed into the pump. This procedure is liable to human error if correct coordination is not achieved and a slurry grout could be injected in to the soil resulting in both soil shear and loss of control over the grout process.

Since the 1960’s improved compaction grouting equipment has been devised, good examples include the modified concrete mixer and a modified concrete pump with a pumping pressure capacity in the order of 7000 kN/m2 developed by Warner Construction Company in USA. Putzmeister have produced a unit in which the pump and the mixer are incorporated into a single unit with a capability to pump 50 to 60 litres a minute with pressures of up to 700 kN/m2.

Since its inception the compaction grouting technique has been recognised as a radical departure from the permeation and the fracture slurry grouting techniques which are more widespread.

The idea of injecting a thick grout into the ground for the purpose of

compacting these soils was a new and unfamiliar concept but it has gradually proved to be useful and versatile. The technique has gathered wide acceptance in the USA and is also being used in other parts of the world.

A significant number of contracts have been

completed in Southern Africa to date.

Applications of the compaction grouting technique have been largely for the purpose of settlement remedial in structures and pavements.

There have since been newer and

widespread applications in the use of compaction grouting. Such applications include its applications to minimise surface subsidence during tunneling of soft ground that have become standard for subway construction in the U S A. Another application which has become widespread since the 1980s is in its use as a ground improvement technique for new construction sites a few of which have been conducted in Southern Africa. Chapter 3 of this

5

report covers a case study in which the technique was successfully applied for ground improvement at a new development site.

1.2

THE PHILOSOPHY AND THEORY OF COMPACTION GROUTING

The philosophy, theory and practice of compaction grouting are considered empirical. They are based on observations from practising engineers and contractors spanning over a period of just over fifty years. Such observations are recorded in various unpublished project reports. Several papers have been prepared and presented at conferences and are published in journals regarding the mechanics, material requirements equipment, quality control methods and testing (Graf E, 1992).

The process involves injecting into the compressible soil mass, a stiff cement-sand grout through casing or tremie pipe thereby compressing the soil mass with an increasing volume of grout. The grout mass remains in a homogeneous mass with a definite interface with the surrounding soil. The grout mass in turn increases the earth pressures as it increases in volume.

Considerably high pressures are required to successfully pump low slump grout through the casing to attain the desired effect.

Not withstanding the pressure losses along the

transmission lines pressures up to 5 MN/m2 have been applied in compaction grouting. The grout consistency relatively reduces the risk of loss due to leakage of grout into areas for which it is not intended. It also enables the densification of compressible soils to be achieved with minimal chance of hydrofracturing occurring or the mixing of soil which is likely to occur when using more fluid grouts.

Compaction grouting is aimed at providing adequate bearing for overlying soils between grout columns. This is thought to be due to the increase in the lateral stress and enhanced soil shear arching ability that results from the lateral volume displacement from compaction grouting. Several hypothesis and ideas have been postulated concerning the mechanism of compaction grouting. It is generally believed that when stiff grout is injected in to the soil it expands into a mass with a spheroidal or cylindrical shape. When a grout of high consistency is injected through the bottom of an injection hole, the spherical or semi spherical volume is

6

expanded and filled with grout. As the critical pressure is exceeded the surrounding ground is either compressed or displaced toward any free surface. The resultant volume is filled with grout and continues to grow until either the injection process is terminated or until there is no more free surface towards which expansion can proceed (Nonveiller E, 1989). This process is analogous to the blowing up of balloons in soil using hydraulic pressure as discussed earlier.

It has been supported by observations made in tests conducted by Warner

Engineering Services in 1973 where test injections where exposed by excavations.

The grout mass continues to form an expanding cavity within the soil. As the bulb enlarges a series of radial and tangential, stresses develop on its interface with the soil. As grout injection continues the shape of the bulb will be governed largely by the ability of the ground to resist compaction. Displacement, shearing and plastic deformation will occur in the soil adjacent to the grout interface. Theoretical solutions have been proposed for the expansion of cavities in the centre of an infinite mass of soil. The most applicable in the theory of compaction grouting are probably that of Ladany (1963) and Vesic (1972).

According to the solution after Ladany (1963) for the expansion of a spherical cavity in a normally consolidated or over-consolidated clay with an unconfined strength cu =0 the following relationships are found:

For normally consolidated clay For over consolidated clay

p = 2σz . . . . . . . . . . . (1)

4,5σz < p < 6σz . . . . . . . . . . (2).

Where: p = injection pressure σz = overburden effective normal stress at depth z.

However in 1972 Vesic using analytical methods provided a general solution of the problems of expansion of spherical and cylindrical cavities in an ideal soil mass possessing both cohesion (c) and friction (φ) in the Coulomb-Mohr sense. In his solution, Vesic considered the problem of a spherical cavity of initial radius Ri expanded by a uniformly distributed internal pressure p. When this pressure is increased, a spherical zone around the cavity will pass into the state of plastic equilibrium. The plastic zone will expand until the pressure reaches an ultimate value pu. At this moment, the cavity will have a radius, Ru. , and the plastic zone around the cavity will extend to a radius Rp as illustrated in Figure 1. Beyond

7

this radius the rest of the mass remains in a state of equilibrium. It is of engineering application interest to determine the values of pu and Rp..

Figure 1: Expansion of cavity within a soil mass (Vesic, 1972)

In order to determine these parameters it is assumed that the soil in the plastic zone behaves as a compressive plastic solid, defined by Coulomb-Mohr shear strength parameters c and φ, as well as an average volumetric strain ∆.. The volumetric strain can be determined from states of stress in the plastic zone and volume change –stress relationships.

Beyond the

plastic zone the soil is assumed to behave as a linearly deformable, isotropic solid defined by a modulus of deformation E and a Poisson’s ratio ν.

To determine the ultimate pressure, pu , a relationship stating that the change of volume of the cavity is equal the change of volume of the elastic zone plus the change of volume of the plastic zone is used. Denoting the radial displacement of the limit of plastic zone by up this relationship can be written as follows:

R3u – R3i = R3p – (Rp –up)3 + (R3p – R3u)∆ . . . . . . . . . . . . . . . . . . . . . . (3) Where:

8

up = (1+ν ). Rp. (σp –q)/2E . . . . . . . . . . . . . . . . . . . . . . . . . . . (4) in which σp = σr at r = Rp

The value of pu can be computed from the following equation:

pu = cFc + qFq . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .(5) Where Fc and Fq are dimensionless spherical cavity expansion factors given in Figure 2.

Figure 2: Spherical cavity expansion factors.

However when considering a cylindrical cavity the problem is completely analogous to that of the spherical cavity. The main difference being that the cylindrical cavity is axially symmetrical and not spherically symmetrical.

The condition of equal volume becomes: R2u – R2i = - [ (Rp –up)2 - R2p ]+ (R2p – R2u)∆ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (6) Where: up = (1+ν ). Rp. (σp –q) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7) in which σp = σr at r = Rp

Using equation (5) the value of pu can be calculated with the cylindrical cavity expansion factors Fq and Fc obtained from Figure 3.

Figure 3. Cylindrical cavity expansion factors.

These equations and values of the numerical values of cavity expansion and volumetric strain factors allow a relatively simple direct evaluation of ultimate cavity pressure on the basis of physical characteristics of the soil. The solution also further highlights that the principle

9

parameters affecting the ultimate pressure are the initial effective ground stress, strength and volume change characteristics of the soil as well as the rigidity index of the soil.

1.3

DESIGN ASPECTS

1.3.1

Site Investigation and Testing

Compaction is widely accepted as a method of ground improvement that results in increased density in granular soils. The resultant increases in density causes shear strength and bearing capacity increases. To make a decision regarding the design of such a compaction program it is of utmost importance that the soil characteristics that can be modified are understood. The degree of improvement that can be realistically obtained and the optimal methods to acquire. The most economical improvement will also have to be considered (Xanthakos, et al. 1994).

Central to such considerations are the characteristics of the soil to be treated and site conditions. The general characteristics of the soil must be known for engineering analysis. These characteristics which can be found by laboratory and in-situ testing include: •

Soil type

•

Grain size distribution

•

Atterberg limits

•

Moisture contents

•

Unit weight

•

In-situ density

•

Shear strength

•

Compressibility

•

Permeability

•

Location of ground water table

•

In-situ density

•

Vertical and lateral distribution of the soil horizons

10

Due to the empirical nature of compaction grouting and variable soil conditions detailed design work is generally not performed. A sketchy design is normally done based on experience and educated guesses. Execution of the job is based on this initial design with relevant adjustments as the grouting progresses. Normally specifications are performance based and will include restrictions and guidelines. Performance of the task is evaluated with respect to the resultant improvement. Design approaches in compaction grouting are centered around the needed improvement and the specified requirements. It is normally the standard on foundation design projects to calculate predicted settlements and bearing capacity after determination of reliable soil property values. The degree of compaction grouting will endeavor to restrict the induced settlements to within tolerable limits and to attain the allowable bearing capacity.

The amounts of improvement required can be expressed in terms of the following measurements (AASHTO 1990). •

Minimum load-bearing requirement.

•

Minimum or average stiffness.

•

Minimum or average SPT blow count.

•

Minimum or average cone penetration resistance.

•

Minimum or average percent relative density.

•

Minimum or average percent of maximum density.

Design for a ground improvement program can be done in the following steps (Xanthakos et, al. 1994)

1.

Carry out a geotechnical investigation.

2.

With obtained data perform foundation analysis including settlement and bearing capacity.

3.

Check the obtained values against project requirements.

11

4.

If calculated values are lower than project requirements consider soil improvement.

5.

Determine the amount and location of improvement needed.

6.

For shallow problems consider conventional compaction and dynamic compaction. For deep problems consider vibrocompaction, vibroreplacement (stone columns) or compaction grouting.

7.

Evaluate the expected improvement from each considered method and evaluate weather the amount of improvement is acceptable.

8.

If ground improvement is a viable solution, prepare a preliminary design and necessary contract documents.

1.3.2

Compactable soils

All types of soils of soils can be improved using the compaction grouting technique, but the effectiveness for a given amount of effort will significantly vary according to the soil type and density (Warner J, et al. 1973). Tests have shown that in uniform soils, the shape of the resultant bulb is quite regular whereas in non-uniform soils it becomes irregular. The flow of grout in soils has a tendency to follow the path of least resistance and hence improves the weakest zones first. Due to this phenomenon even though there will be some variation in the density of the soil surrounding the injected grout mass, a general improvement in the uniformity of the soil mass will occur. Compaction grouting techniques are most frequently used in fine grained soils where it is most effective although it has been applied in other types of soils with reasonable success. Applications have been effected in trash and rubble fills, loose unsaturated sandy clays, fills composed of soft clay, closure of subsurface erosion paths and to densify collapsed materials in dolomitic formations with karst topography.

Graf E, (1992) states that any soil that can be compacted by squeezing of air or water from the voids by grout pressures developed in the soil may be compacted using compaction grout. Relative compaction of up to 90 % Modified AASHO can easily be attained in most

12

groutable soils.

The following comments are valid for assessing the potential of the

following soil types for compaction grouting.

Clays – when treating clays using compaction grouting the bound water cannot be squeezed from clay particles with the pressures used but free water may be removed from the voids. Care is required when treating such low permeability soils to minimise the effects of excessive pore pressures. In some cases simply reducing the pumping rate will be sufficient to prevent excessive pore pressure build up. However the provision for drainage should be provided where the normal pore pressures are appreciably raised by grout injection. Whilst compaction grouting is probably not the best technique for problems where low permeability soils exist it has been applied on projects where drainage was used and the grouting was done over an extended period.

Silts – have been successfully densified both above and below the water table. Below the water table pore pressures may take a day or more to dissipate.

Silty sands - Silty desert sands subject to hydroconsolidation, silty liquifiable sands and even very clean sands above and below the water table have successfully been densified. These soil types are also suited to treatment by other grouting techniques and economics and the intended use will detect the use of compaction grouting.

1.3.3

Empirical design approach

Due to the relatively recent development of the technique, and the general nature of grouting, definite design methods and procedures for estimating costs have until recently not been well developed. The success of past projects has often depended on trial and error procedures, and the amount of experience of the engineer and the contractor.

Therefore the empirical origins and numerous variables, design approaches used in compaction grouting are largely done based on experience. Common practice has shown that this type of design is characterized by borehole spacing and stage length, both of which are dependent on the anticipated radius of compaction at each injection point.

13

The radius of compaction is thought to be a function of the following (Graf, E. 1992): •

Restraining pressure of the soil, which is directly proportional to weight of soil above the bulb.

•

Total radial force and hence the surface area of the grout bulb.

•

The grout pressure at the bulb, which is for practical purposes is the grout pressure at the collar.

The effective radius of each grout hole will vary according to soil type, moisture and relative density. It is also influenced by the grout injection rate and resulting pressure. Based on experience it has been established that in most cases significant densification of soil will occur within a radius of 0.3m to 1.8m from the interface of the soil and the grout (Warner J, 1992). It is however recognized that some effect will be experienced at greater distances from the point of grout injection.

Borehole spacing is dependent on the soil type and soil conditions, the depth, improvement requirements and judgement. The basic guidelines used for empirical design include (Graf E, 1992): •

Using greater spacing of injection points with increasing depth or load of structure above the compaction zone.

•

For relatively loose soils a greater spacing of injection points can be adopted than in more compact soils because grout bulb would be much larger and hence the radial forces.

•

Due to their lower bearing capacity requirements lighter structures can have greater spacing than for heavier structures.

Normally the boreholes are placed on about 2.4m to 3.6m centers. As a rule alternate primary boreholes are first grouted followed by the secondary boreholes which are normally located at mid centers of the primary boreholes. Injections deeper than 3.6m frequently have horizontal spacing of 2.4m to 4.8m and because of the passive resistance from soil this spacing is usually not increased with depth.

14

With regards to grout injection points down the hole a length of one half the borehole spacing is used, a conservative practice. Some practitioners prefer to use a standard 1m or 2m stage length. The selection of the grouting method will depend on both cost and performance requirements. The down stage method has the advantage of providing good control of the grouting process which is very important when grouting close to the surface or near structures. There are extra costs attached to the down stage methods due to additional drilling costs and a longer construction schedule. The up stage grouting method may be a more cost effective alternative which provides adequate control for most applications. Conditions for the end point refusal are predefined with consideration for the anticipated goal of the grouting. The refusal criteria may include pressure, volume and surface displacements or volumes.

1.3.4

Rational design approach

Schmertmann and Henry (1992) presented a new theory and design for compaction grouting. The theory uses shear enhancement along grout columns, due to lateral stress and density increase to support the soil above and between the grout columns. Laboratory tests were conducted to help support the design theory.

This design approach is considered a

breakthrough and a major improvement to the experienced based empirical design discussed above. The combined use of the two approaches provides a more rationale design process that can also be used for the purposes of cost estimation and verification of completed projects. The rational design approach, its concepts and design analysis are discussed below using Figure 4.

From the cross section in Figure 4 it is clear that: •

High-pressure injections of low slump mortar form a geometric pattern of grout columns on a layer of good bearing.

•

Multiple injections at suitable depth intervals form each column to a height h. Each column helps to displace and hence compact the soil in area s2 between the columns and increases the lateral pressure σh, and soil friction τ at the column/soil boundary. The resultant density and stress increases results in a potential column/soil shear force T and a top bearing R.

15

q↓

d

h

s σh

Grout column

σv

↓↓

bedrock

Figure 4 : Section across treated area

•

Suitable combinations of s column spacing, height h and average diameter d are tried until the overburden soils can be entirely supported by T and R. Any surcharge q generates and additional ∆T and ∆R at the grout columns.

•

Removal of all or part of the overburden load σh or q by erosion or otherwise resulting in arching to the grout columns removes all or part of the risk of surface settlement.

The design analysis aims at achieving a certain factor of safety F, using only effective forces and stresses. Depending on the critical conditions total forces and stresses may also be used (Atkinson, J.H and Bransby P.L, 1978). The following equations were used to define the various forces and stresses considered in evaluating the factor of safety.

The weight requiring support is expressed according to equation 1 Fv ↓ = (γ’z + q) s2 ………………………………………………………………… (where γ’ = effective unit weight of the soil)

The forces available for support Fv↑, are vertical upward forces which consist of T and R plus ∆T and ∆R from any surcharge and are described by the following

(8)

16

equations : 9 to 14. T = τπdh…………………………………………………………………………..

(9)

τ = σh tan φg …………………………………………….…………...…………….

(10)

(Where φg = soil/grout column friction angle)

σh

=

[Ko

+

α

(Kg

–

Ko)]

(σv)

(11)

……………………..………..……………….……… (Where Ko and Kg denote the earth pressure coefficient before and after grouting respectively.) ∆T = Ko q tan φg πdh ………………….…………………………………………

(12)

R = σvc πd2 /4………………….…………………………………………………

(13)

(Where σvc = grouting pressure at top of columns) ∆R = θq πd2 /4 …………………..…………..……………………………………

(14)

(Where θ = stress concentration factor on top of columns, after Boussinesq theory)

The factor of safety is therefore given by : F = Fv ↑ = T + ∆T + R + ∆R Fv ↓

(γz + q)s2

……………………………………………..……..

(15)

However when F < 1 it is recommended that a stress reduction factor SR be used, which is defined by : SR = σ′after σ′before

= [s2 (γz + q) – (T + ∆T + R + ∆R)] / ( s2 - πd2 / 4) (γz + q)

.....………

(16)

(Where σ′ = average vertical normal stress acting between and at the base of the columns.)

Such a design concept provides a rational basis for the design of grouted soils that support the overlying materials between and above the soil columns. Whilst this approach provides a

17

very important design method it is rarely feasible to layout a programme in advance that turns out exactly as predicted. Therefore maximum flexibility should remain the constant objective so that the design can be refined as exploratory and drilling data become available for every hole completed (Warner J et al, 1974). It is also considered good practise to produce a thorough programme so the work is not likely to be substantially increased.

1.4

QUALITY CONTROL AND ASSESSMENT

To optimize compaction grout design, quality control becomes a very important aspect of the programme. Attaining a balance between quality control and optimum design provides a means to minimizing cost for the client. Evaluation of such a balance should be determined by the risk of grouting failure and the consequences should failure occur (Huck, PJ and Waller MJ 19??)

For effective quality control an elaborate pre-planned control procedure should be devised. The main aspects which need to be included in the quality control procedure are equipment, materials and grout mix, pumping rates and pressures as well as placement procedures. The Engineer for maintaining control on projects may use various instruments and or control devices. The use of continuous sequential records is paramount for good quality control. Prepared data sheets should be available for the resident engineer and his assistants to collect the required data. To obtain the most benefit from such data, each of the recorded parameters should be thoroughly considered including the intended use of such information.

It is important to take into consideration that stabilization of all intended soils is attained, by ensuring that grout is actually injected into the planned or required areas within the mass. The proper layout of holes on site, location, inclination, depth and sequence of holes. Procedures for grouting each hole until refusal should be planned to ensure proper grout placement. As it is rarely feasible and probably not practical to obtain enough information to lay out a program in advance that turns out exactly as predicted, maximum flexibility and control must be a constant objective (Warner J and Brown DR, 1974). This brings an added advantage, so that adjustments are made, as drilling and grouting data becomes available.

18

To ensure quality control of the highest standard all aspects of the grouting process have to be checked prior to, and during the placement of grout. The main aspects to be included in the quality control programme are discussed in the following sections.

1.4.1

Equipment

Production boreholes are best drilled using percussion rigs, which have substantially greater speed. Drilling equipment should be in sound condition including its accessories such as compressors. Detail of such equipment is covered in numerous drilling texts and will not be discussed in this text. Only qualified contractors should be engaged to undertake drilling with experienced personnel.

For successful grouting individual grout components should be weight batched. Modern weight batching plants assembled on site form standard basic units for storing, proportioning of components of mixes and feeding into the mixing units should be used. The traditional batching by sacks of cement, wheel barrow volumes of sand and water result in errors and should be avoided.

Mixers similar to plaster mixers normally give good results. Sufficient power should however be used to ensure blending of the ingredients into low slump consistencies Pug mixers or screw mixers which can produce the high shear required for the stiff and harsh mixes. Most of appropriate mixers are of the horizontal batch type with blades oriented to provide a chopping type of action. For accurate effective operation of mixers it is highly recommended that a metering supply system for grout ingredients is implemented.

Suitable pumps must be capable of displacing zero slump grouts. It should be able to maintain pressures from 0 – 4200 k Nm-2 and rates of displacement as low as 0.008 m3 (Warner J and Brown DR, 1974). Controls must however be provided which enable the operator to vary the rate of displacement over the range uniformly while continuously pumping. A method of measuring the amount of grout pumped should be provided. One such method is the use of an electronic counter on a piston type of pump, which has been calibrated, to displace a volume of grout per stroke. A force feed mechanism should be in place to prevent cavitation of the very stiff grout.

19

Other miscellaneous tools and equipment, which need to be of the required minimum standard, include high pressure hoses usually 38 to 50mm diameter. Full flow ball cock type of valves should be provided at the injection point. Pressure gauges should be equipped with suitable gauge savers to prevent damage from grout.

1.4.2

Materials and Grout Mix

Constituent materials normally used in compaction grout are water, silty sand and cement. Silts are essential to attain the required water holding consistency that allows the grout to be pumped. Limitations on the silt contents are in the 10 – 25 % of the sand but could be as high as 35% if the sits are coarse. To make the mixture pumpable it may be necessary to add clay, fly ash or similar material. Such additives increase the plasticity but should be less that 1% to avoid the possibility of hydrofracture in the host soil. Some practitioners sometimes add bentonite to achieve pumpability, this is not considered a good practice because of the very high plasticity of bentonite. The range of preferable sand gradation is given in Figure 5.

Figure 5 – Preferred gradation of silty sand (After Warner1992)

The ideal sand material would be natural, rounded material having 100% passing a no. 8 sieve and not more than about 20% finer than 50 microns.

This gradation of the sandy material is probably one of the most important factors in successful application of the compaction grouting technique (Warner J, 1982). Coarser sand result in grout, which is subject to breakdown under high pressure, and tends not to retain water introduced in the mix. It also results in blockages within the transmission pipes. On the other hand too fine sand material provides a grout that is difficult to control and may be of limited durability.

For reasons discussed above excessive amounts of clay should be avoided. The author knows no criterion limiting the plasticity requirement. Such criterion as a plasticity index is recommended for a pressure range to produce a grout mix that will neither hydrofracture the soil or plug the grout system.

20

A most commonly used grout mixture contains about 12% cement which when used with appropriate sand will provide a grout with an unconfined compressive strength of 4200 kN/m2 which is usually more than adequate. Recent research and projects demonstrate that no cement is necessary when compaction grout is used as a soil replacement material. In some instances where low strength grouts are required pozzolan or hydrated lime are used in place of cement.

The significance of low slump can not be over emphasized. It is recommended that the slump should not exceed 50mm. Any grout with a slump exceeding 50mm will result in serious problems of leakage, hydraulic fracturing of the soil and premature surface upheaval tends to occur. In addition, control of the disposition of the grout is lost and the amount injected is reduced. Control of slump alone is not a valid means to assure adequate low mobility grout. Tests conducted in Denver by Warner have indicated that close control of the rheology of grout used for compaction grouting is mandatory if maximum benefits are to be realised from the work. Werner and Brown (1973) also concluded that irrespective of slump or pumpability, grouts that are too mobile could result in hydraulic fracturing of the soil and loss of control of the injection operation. The principal factors affecting both slump and mobility are the shape and gradation of the sand, material, the amount and nature of the fine fraction.

1.4.3

Pumping Rates and Pressures

The rate of pumping and pumping pressures, which are initially based on experience, are interrelated factors which are important in completing a successful compaction grouting exercise. For pumping at depths over 6 to 9 m a pumping rate of 20 litres per minute or less is normally specified for maximum (refusal) pressures of between 2.75 to 3.5 MPa. For depths greater than 4.5m a high pumping rate of 85 litres per minute can be used until the pressure approaches refusal. At which time, the pumping rate should be reduced to achieve optimum compaction. For depths shallower than 3m, pumping rates of up to 56 litres a minute are used but the pressure should be closely monitored and adjusted to attain optimum compaction (Graf E 1992).

21

Tests conducted by Warner Construction Company (Brown DR and Warner J 1973) indicated that : •

For shallow depths very high pressures were needed.

•

In holes where the casing was cemented in place higher maximum pressures (2 100 kN/m2) were recorded relative to the holes where the casing was uncemented ((862 kN/m2).

•

Lower pumping rates significantly increase the grout takes

It can therefore be appreciated from the experiences and recommendations given above that one of the principal controlling factors in compaction grouting is the grout pressure behavior. This phenomenon is usually monitored at both the grout pump and at the borehole collar, and should be continuously recorded and monitored. Such monitoring will continue until refusal is attained or surface disturbance is noted. The key controlling criteria for grout injection is not to permit a rapid pressure build up which is directly related to the grout injection rate and should not exceed the optimal rate. The optimal injection rate, which is determined for specific sites during test runs or initial injections, varies for different sites as it is governed by soil properties. The maximum desirable rate of pressure increase is in the order of 35 to 48 kN/m2 per minute.

Close observation and analysis of pressure behavior can reveal a great deal of information about the existing conditions in the ground. For example by plotting injection pressure against time two typical graphs shown in Figure 6 emerge.

Figure 6: Typical injection pressure behavior

Where the pressure build up is consistent (Figure 6a) the indications are that a relatively uniform condition exists. If instead, large fluctuations prevail a very non-uniform soil condition exists as reflected in Figure 6b. Sudden pressure losses, which are indicated in Figure 6b, are usually related to a break through of grout into a void or an area of relatively loosely packed material. It may also indicate an impending surface disturbance, escape of grout into other substructures or a loss of lateral restraint.

22

A theoretical approach has been proposed by Wong H Y (1974) to provide as a first approximation of grout pressure prediction for compaction grouting. He considered a spherical grout mass of radius a with a centre placed at a distance h below a horizontal soil surface. If the grout pipe is assumed to be cemented to the adjacent material, a loose sand and then the upper limiting pressure coincides with the inception of ground leave. From the model of the reaction of compaction after Graf a truncated soil cone above the grout source will be disturbed by the injection as illustrated in Figure 7.

Figure 7 : Reaction for compaction (After Graf E 1992)

As shown in Figure 7 the potential conical shear failure surface will be inclined at an angle θ to the horizontal. According to the Mohr-Coulomb failure criterion for any point on the conical surface, the maximum and minimum principle stresses ought to lie along the vertical and horizontal directions respectively to give; θ = 45o + θ/2

…………………………….(17)

(where θ is the angle of shear failure)

Assuming that grouting pressure is uniform throughout the mass, the radius of which increases with increase in grouting pressure. The maximum allowable grouting pressure ρo at a certain equivalent radius of the grouted mass can be expressed by the following equation (18). γh (h/a)2 + 3 (h/a) tan θ + 3 tan2 θ

* 1+

1 + 2 (1 – sin φ) cos (180 - θ + φ)

ρo = 3 tan2 θ Where

Cos φ cos θ

γ = total soil bulk density a =radius of the grout column φ = internal angle of friction h = height conical soil mass`

The maximum pressures used to determine refusal take into consideration the pressure losses as a result of friction and the overburden pressure. Pressure guidelines, which are normally, used in fine and medium grained soils, are given below.

23

Table 1: Guidelines of the maximum allowable grout pressures

1.4.4

Depth (m)

Pressure (MPa)

0-5

0,4

5 - 10

1,0

10 - 15

1,5

15 - 20

2,4

20 - 25

3,0

25 - 30

3,5

30 - 35

4,0

Grout Placement

A predetermined sequence for placing grout into boreholes should be followed. The resident engineer based on set criterion can alter such a sequence, which is decided upon during the design, on site. This sequence normally involves grouting of adjacent primary boreholes. This is followed by grouting of secondary holes placed mid distance from the primaries. Care should be taken to ensure that interference with drilling pressure form the drilling rig are minimized.

An important aspect of the monitoring is to keep a record of the pumping pressures both at the pump and at the borehole collar. It is suggested that readings be recorded at predetermined time periods. This part of the grouting is crucial in deciding whether or not refusal has been attained, there by ensuring an optimal use of materials and the grouting process. This information can be analysed by plotting graphs such as Figure 6 which can be of great assistance in determining the processes occurring in the ground.

Monitoring during grouting is essential to verify proper performance of compaction grouting. As a minimum, the grout consistency, injection rate, injection pressure and injected volume in each stage need to be monitored. A review of such data will permit evaluation of the success of the compaction grouting program. Lower injection pressures and higher injection volumes are normally attributed to softer and less dense areas for example. Secondary

24

injections should show higher pressures and lower volumes due to improvement induced by primary injections.

1.5

VERIFICATION OF COMPACTION GROUTING

In an attempt to evaluate and verify the effectiveness of compaction grouting there are several verification methods available to the grouting practitioner. Such methods assess whether or not grouting operations have adequately achieved the pre-planned objectives, which enables the use of compaction grouting with greater confidence. Use of such methods provides a means by which one can measure or observe the process of grouting which is often conducted beneath the surface. Verification of the effectiveness of compaction grouting is best determined by careful observation of the finished product and thoughtful engineering judgment and analysis of physical data obtained during and after the compaction process.

It is therefore important that a verification programme is drawn up prior to conducting compaction grouting. At this stage the problem to be addressed by the grouting should be defined in detail to the satisfaction of all parties. Prior agreements with regard to the criteria which define success or failure have to be drawn up and the data which needs to be collected before and after compaction grouting. This ensures that success or failure of a grouting operation can then be assessed based on some evidence. When the compaction grouting operation is completed records, of grout take, grout injection location, injection depths and other data are a valuable tool. In some instances they provide sufficient indicators of proper grout placement. However parameters such as soil strength can at times be measured to confirm any changes which may result from the grouting. As it is often assumed that if grout sets in the desired location the job is successful, the success of the grouting operation is measured by field procedures which verify the grout location Byle M and Roy Borden (1995). There are several field tests which can be used to infer or verify the actual location of solidified grout. Such methods include geophysical instruments and may require specialised equipment and expert personnel. Depending on the site conditions and application some of these methods can be economical.

Using the findings of the detailed geotechnical investigation surprises may be avoided. Byle, MJ and Borden, RH (1995) have suggested that when considering a monitoring

25

programme one should consider the possible conditions that could affect the grouting. Such conditions include the possibility of groundwater to dilute the grout mix, groundwater flow rates may lead to turbulence and related erosion. Dissolved chemicals in the groundwater may affect the grout. Discontinuities in the soil or rock mass could direct the grout to an unplanned location. In general unknown conditions may affect the grouting programme in unexpected ways. Hence parameters during the progress of the work enables previously unknown conditions to be identified and treated appropriately.

Prior to carrying out a verification programme, it is important to clearly define the properties to be measured. Methods available include direct measures of performance of the grouted system or may include methods for verifying proper construction. The verification programme goals should be set appropriately for the project. It should be highlighted that in most cases only a qualitative assessment of grouting performance is needed but in some instances accurate quantitative assessment of grouting may be needed.

To ascertain the success of a grouting programme it is necessary for the owner, design engineer and contractor to agree before hand upon the success and failure parameters. In determination of the method of verification the cost of the method must be balanced against the benefit to be gained. The following should be addressed prior to making the final choice of the method to be applied: •

What is the purpose of the grouting?

•

What measurable changes are expected due to grouting?

•

What methods can measure these changes?

•

How will the verification test results be evaluated?

•

What will be the acceptance criteria?

•

What are the consequences of failure?

•

What is the cost of the verification?

•

How will the verification fit into the construction sequence?

On addressing these issues a well defined verification programme can be devised. Such a programme will be cost effective, provide an appropriate level of assurance, provide results in a reasonable time frame. The methods to be applied should be consistent with project

26

goals and should incorporate the verification in the design and constractability evaluation. They should be written into the specification to provide the required level of quality assurance. Another consideration is the interpretations to the grouting which may be needed to allow for before and after testing to measure improvement.

The test grouting section should include the same verification programme intended for production grouting. Sufficient time should be budgeted for completion and evaluation of the test grouting.

1.5.1 Verification methods

A variety of methods are available for measuring the performance of grouting either directly or indirectly. These methods are used to determine a change in some property of the sub surface after grouting or to detect the presence of grout. Some of these methods are non intrusive and/or non destructive and can be applied without disturbing or damaging the grouted area. Such methods are generally indirect and require interpretation of the desired information from some other measured property. The intrusive methods can either collect samples for viewing or other analysis, or measure some in-situ properties. Some methods can be used while grouting is underway and others are sensitive to disturbances caused by ongoing construction and can only be used when all operations have stopped.

These methods can be subdivided into mechanical, chemical geophysical and hydraulic methods. These methods are discussed in detail by Byle MJ and Borden RN (1995) and are summarised in Table 2 Suitability of some of the methods for applications in verifying different grouting methods are shown in Table 2.

TABLE 1 : A SUMMARY OF VERIFICATION METHODS

(i)

Load testing – e.g. plate load test

(ii)

Penetration resistance – e.g. cone penetrometer test (CPT)

(iii)

Probing and sampling – e.g. Shelby tube, SPT

(iv)

Excavation / coring

(v)

Modulus test – e.g. flat dilatometer tests (DMT)

27 Mechanical

(vi)

Extensiometers – e.g. linear variable differential transformers

Methods

(vii)

Tiltmeters

Chemical Methods

(viii)

Optical surveying of survey points

(ix)

Fluid levels

(x)

Rotating laser levels

(xi)

Density test – nuclear density gauge (troxler)

(xii)

Shear strength test – laboratory

(i)

pH Indicators

(ii)

Chemical Dyes

(i)

Seismic

(ii)

Conductivity

Geophysical

(iii)

Electrical

Methods

(iv)

Electromagnetic

(v)

Acoustic emissions

(vi)

Gravimetric

(vii)

Magnetic

Hydraulic Methods

(i)

Laboratory methods – constant head, falling head

(ii)

Field methods – soakaway, pumping tests

28

CHAPTER 2

DYNAMIC COMPACTION

2.1

INTRODUCTION

Dynamic compaction is a process of improving weak soils by controlled high energy tamping. The process of dynamic compaction is dependant on the type of soil subjected to treatment and on the energy input. The process by varying soil type and energy input is capable of improving soils to substantial depth. The mechanism of compaction is simply to reduce the air voids, thus increasing the density and improving the overall engineering properties of the soil. Applications to which dynamic compaction has been applied include: • • • 2.2

Controlling of settlement Increasing the soils bearing capacity Preventing reducing the risk of liquefaction

HISTORICAL DEVELOPMENT OF DYNAMIC COMPACTION

The principle of dropping heavy weights on the ground surface to improve soils at depth dates back several centuries. Chinese drawings such as that in Figure 2.1 are evidence for these suggestions (Menard and Broise, 1976). There are also reports that this method was used during the construction of the Roman Empire (Kerisel, 1985). Figure 2.1

More recently the principle of dynamic compaction has been related to the principle of Proctor compaction that was presented by Proctor for the first time about 60 years ago. The advent of current high energy tamping was first performed on a regular basis in France in the 1970’s. This was mainly due to the development of the Menard compactor shown in Figure 2.2. This was followed shortly by the use of large crawler cranes, which characterise deep dynamic compaction today. Figure 2.2: Menard Compactor

The physical performance of dynamic compaction is fairly simple, defined by a crane of sufficient capacity, dropping a suitable size of weight in a virtual free fall from a certain drop height. For most applications standard crawler cranes, slightly modified for safety reasons,

29

with a single lifting rope attached to the top of the weight are used. An example of such cranes is shown in Figure 2.3.

Figure 2.3 : Crawler crane used for dynamic compaction

According to Moseley (1993) the majority of British and American contracts utilise weights within the range of 6 - 20 tonnnes dropped from heights of up to 20m. In America weights of up to 33 tons and 30m drop height have been used. Specialist frames based on the Menard compactor fitted with quick release mechanisms have been utilised to drop weights of up to 50 tonnes. Menard built equipment to drop 170 tonnes from a 22m height at Nice Airport in France. In South Africa weights in the order of 8 to 15 tonnes dropped from heights in the order of 10 to 20m are common (Everett and Friedlander, 1990). The weights used for the impacts are typically constructed using box steel and concrete, toughened steel plate or reinforced mass concrete where durability is the prime requirement. Different sizes and shapes of the weights (pounders) have been developed. Narrow weights have been specifically used for driving materials. An example of a weight is shown in Figure 2.4. Figure 2.4: Pounder

2.3

THE PHILOSOPHY AND THEORY OF DYNAMIC COMPACTION

2.3.1

Terminology Development of dynamic compaction throughout the world has resulted in a large number of terms, some of which can have different meanings in different areas. To avoid confusion the following terms which have been mainly adopted in the United Kingdom with South African adaptations will be used in this dissertation (Moseley, 1993). Effective depth of influence: Maximum depth at which significant improvement is attained. Zone of major improvement: Typically half to two thirds of the effective depth. Drop energy: Energy per blow, i.e. mass multiplied by drop height. Pass: The performance of each grid pattern over the whole treatment area. Total energy: Sum of energy of each pass i.e. number of drops multiplied by drop energy divided by respective grid areas (normally expressed in ton metre/metre2).

30

Recovery period: The period of time required between passes to allow the excess pore water pressures to dissipate to a low enough level for the next past. Induced settlement: The average reduction in the general site level as a result of the treatment.

Threshold energy: Energy input beyond which no further improvement can be practically achieved. Overtamping A condition in which the threshold energy has been exceeded, causing remolding and dilation of the soil. Shape test: Detailed measurement of imprint volume and surrounding heave or drawdown effect which permits comparison of overall volumetric change with increasing energy input. Imprint : The crater formed by the weight at the point of impact.

2.3.2

Energy waves and effective dept of influence When the pounder is released from the drop height, it imparts energy into the ground by means of shock waves. Two types of waves are observed at each impact. : • •

Volume or Love waves Interface or Rayleigh waves (Varaskin, 1981)

The volume waves fall into two categories, compression and shear waves: Compression or primary waves (P), which are propagated in the liquid, phase similar to sound waves travelling through the air. They initially increase and then decrease the pore pressure resulting in the initial dislocation of the solid or semi solid skeleton. Shear or secondary waves (S), which are slower, can travel through the soil completing the dislocation. By shearing the soil particles, the soil is rearranged into a denser state. The effects of the P and S waves are illustrated in Figures 2.5(a) and Figure 2.5(b), which are shown below.

Figure 2.5(a): Effect of (P) wave

Figure 2.5(b): Effect of (S) wave

31

On the other hand the Rayleigh waves have two components the horizontal or longitudinal component and a vertical component. Whilst the horizontal waves are principally responsible for the rupture of the existing soil structure and break up at the surface of the soil layer. The vertical, traverse waves are responsible for easier compaction of the soil in deeper layers. The permeability of the soil in the upper part of the layer to be compacted is greatly improved. This makes quick evacuation of pore water much easier, while a new settlement (compaction) is made possible by the changing shear stresses to the traverse waves.

An illustration of the effect of Rayleigh waves is shown in Figure 2.6.

Figure 2.6: Rayleigh Waves

Woods (1968) postulates that Rayleigh waves comprise 67% of the total energy whilst shear waves make up 26% with the remaining 7% due to compression. The effective compaction energy calculated per blow, Edv is given by Wallays (1993) as: Edv = ηMh

…………………………………………………..(19)

Where M = mass of drop weight h = drop height of pounder The yield coefficient η is a combination of the yield of the drop movement, yield of the blow its self and the yield strain as well as the plastic strain of the upper layer. Repeated testing and analysis by practitioners has obtained a total yield η, with an average value of η = 4/27.

Accordingly Wallays (1993) has calculated the corresponding settlement, s max of each layer to be compacted as: s max = σ max B2 A

Where σ max = Where

B A,F Ed

γ eff

…………………………………………………(20)

1 B2

ηMh A

F2 ηMhA

+

F 2ηMhA

= largest horizontal traverse size of drop weight (M) = factors which can be calculated from B and Ed (dynamic modulus elasticity of the layer) = dynamic modulus of elasticity of the layer = effective volume weight of the soil and the thickness of the layer to be compacted.

Menard originally proposed that the effective depth of treatment was related to the energy input expression. d = WH Where

W H d

………………………………………………………………(21) = weight in tons = drop height in metres = depth in metres

32

This was later modified by a factor 0.5 by Leonard et al (1980) for granular coarse soils, and factors of 0,375 to 0,7 by Mitchell and Katti (1981) for granular and cohesive soils respectively. A more widely acceptable formula for calculating effective depth is : d = k1 k2 (MH)

………………………………………………(22)

Where k, kz are factors ranging between 0.55 and 0.95 depending on shape of pounder, type of lifting device, material type and presence of water.

Moseley (1993) suggests that the range of effective depth of influence varies with initial soil strength, soil type and energy input as illustrated in Figure 2.7. Figure 2.7: Depth of influence (Moseley, 1993)

There are several factors, which affect the effective depth of influence. They include the type and competence of the surface layers, position of the water table and number of drop repetitions at each point. The kinetic energy at the point of impact is also a major factor in determining the effective depth of influence. The shape of improvement in the ground tends to be similar to the Boussinessq distribution of stresses for a circular foundation. A modification of energy levels for each pass can be used design a treatment scheme to a specific soil profile and engineering requirements. A variation in the soil profile can influence the effective depth, for example a solid “plug” of very dense or stiff material can form beneath the impact location and hinder the improvement to depth. A weak surface and a high water table can also limit the performance of sufficient number of stress impulses to induce minimal improvements to underlying layers. Nevertheless, knowledge of the effective depth of influence of any stress impulse is a vital factor in both the planning of treatment operations and the potential for transmission of vibrations. 2.4

DESIGN AND CONSTRUCTION

The use of detailed design work is generally not performed when using dynamic compaction techniques. As with compaction grouting and other ground improvement techniques specifications are normally written on a performance basis including restrictions and guidelines. Performance is evaluated in terms of a measure of improvement, how such improvement is to be measured will be specified. Design aspects, which have been discussed in Section 1.3 in the preceding chapter, are generally applicable to dynamic compaction. There are however certain design steps which should be undertaken to ensure the success of dynamic compaction. These steps include, but are not limited to: • • •

Performing site investigations. Developing settlement influence curves. Conceptualising an initial dynamic compaction programme.

33

•

2.4.1

Anticipated response of cohesive and granular soils to treatment.

Site Investigations The importance of carrying out a geotechnical site investigation can not be overemphasized. This subject is addressed in detail in Chapter One and will not be discussed in detail here. However, issues to be considered include soil profiles and their geological origin, groundwater level and flow regime, in situ density and shear strength characteristics, the degree of saturation and other geological anomalies. It is important to remember that thorough planning must be done prior to the investigation, bearing in mind the intended use of the results of the investigation.

2.4.2

Settlement influence Settlement influence curves are drawn up after a testing programme. The procedure involves penetration and heave control tests in which the penetration/heave of the soil is measured after successive blows to optimise energy input. A typical result of the settlement influence test is given in Figure 2.8.

Figure 2.8 : Result of settlement influence

Since the effective depth of compaction is limited by the drop energy as discussed. It should therefore be recognized that the maximum effective depth of influence is achieved when the energy is spread over a big enough area to absorb the effect in the compacted zone. The compacted layer acts as a shield to the lower horizons preventing further transmission of the energy input. In this case a compacted “raft” is formed over the underlying compressible materials. It is therefore important at the testing stage to determine the depth to which improvement is required. This information will be used to determine the correct drop height and pounder combination for the primary pass. According to Everett and Friedlander (1990) when a pounder is dropped at the point of compaction the horizontal spread of the compacted bulb results in a cylindrical cone or pyramid of compacted material as illustrated in Figure 2.9. Figure 2.9

Phases in compaction

In the primary pass an optimum spacing of the pounding positions should be determined. This spacing should ensure an overlap of the foot of the resultant bulb with the bulbs from adjacent prints as illustrated in Figure 2.10(a). After completion of the primary pass which may be done in two or more passes the secondary pass is mainly aimed at compacting the wedge between the primary bulbs. It is usually at mid point between the existing primary pass imprints. The effect of the drop energy for the secondary pass whilst primarily compacting the wedge between primary bulbs will also overlap with deeper potions of the primary bulbs, enhancing compaction in this zone. Figure 2.10(b) illustrates the effect of the

34

secondary pass. On completion of the secondary pass a phase known as blanket ironing is carried out over the entire site to compact the remaining gaps as illustrated in Figure 2.10(c). It should be borne in mind, that different variations in drop height and pounder weight are applied to meet the required effective depth of influence, as predetermined in the settlement influence tests. Depending on the required depth of improvement and the soil profile a “full” penetration dynamic compaction treatment may be required or a “raft” treatment dynamic compaction may be required. The former involves effecting dynamic compaction to the full depth of the compressible material. In this instance the compacted material is driven to “refuse” on competent material. In the latter however the upper compressible materials are compacted reducing differential settlements while considerable total settlement may still occur. These two options are shown in Figure 2.11.

Figure 2.11a: Dynamic compaction of entire layer Figure 2.11b: Partial compaction of layer (raft compaction)

2.4.3

Initial dynamic compaction programme Based on available information and knowledge of dynamic compaction theory an initial programme can be drawn up for a specific site. Such a programme will include and address the following aspects. • • • • • •

2.4.4

Determine depth of improvement and material type Determine drop energy required (drop weight and height combination) Determine horizontal spread and grid pattern Conduct a full scale dynamic compaction trial section complete with compliance tests to optimise initial assessment. Draw up programme to include optimised grid pattern showing positions of primary, secondary and blanketing passes. Optimised pounder weight and drop weight for each pass. Testing programme to include all tests to be conducted subsequent to and after dynamic compaction treatment.

Response of materials to be compacted There are fundamental differences in the responses of different materials to the processes involved with high energy impacts. As discussed earlier the process of dynamic compaction is normally done in three principle phases, primary secondary and ironing which attain three respective effective depths of improvement. The deepest layer is treated using a wider grid pattern, a middle layer using an intermediate grid and finally the blanket ironing pass for the near surface layer. The main differences in response are found between granular and cohesive soils, which may be treated and are discussed below.

2.4.4.1 Granular Soils

35

In dry granular materials the improvement of engineering properties is attained through the physical displacement of particles and low frequency excitation. The result is a reduction in void ratio and an increase in relative density thereby providing improved load bearing and enhanced settlement characteristics. Careful planning and determination of the depth of influence from the first pass is required. This is mainly due to the formation of a hardened layer that inhibits penetration of stress impulses to deeper layer. This layer however plays an important role in providing superior settlement performance beneath foundation bases. In places where the depth to the water table is shallow and within the compactable layer, a high proportion of dynamic impulses is transferred to the pore water. If surface impacts increase to a critical stage a significant rise in pore water pressure may induce liquefaction. According to a theory postulated by Menard the induced liquefaction is similar to that caused by earthquakes. Whereby the soil is subjected to high shear strains and looses its shear strength due to seismic shaking and pore pressure build up that reduces effective stress in the soil. The low frequency vibrations caused by further stress impulses will then reorganise the particles into a denser state. The character of the related ground motion (acceleration and frequency) is determined by the grading of the soil and the in-situ stress conditions. Dissipation of the pore pressures, together with the effective surcharge of the liquefied layer of the soils above, results in further increases in relative density. However such dissipation will occur over time varying from 1 – 2 days for well graded sand and gravel, to 1 – 2 weeks for sandy silts. This time dependant response should be determined in the test programme. Mitchell and Solymar (1984) have suggested that chemical bounding or high residual stresses within the soil matrix may result in longer term improvement. Alternatively when treating soils which have shallow water tables the dynamic compaction is conducted in such a manner that the liquefied state is avoided. This may not be possible in deep loose sandy deposits with high water tables. Such treatment is designed to provide compaction by displacement without dilation or high excess pore pressures. This can be attained by using a smaller number of drops from a lower drop height. The method requires substantially lower drop energy input than the liquefaction approach. Laboratory testing has shown that in order to achieve maximum density, the lowest number of stress impulses to the required energy input will provide the optimum result. Saturated granular soils normally will require higher treatment energy than if the soils were essentially dry. A typical volumetric response for granular soils is illustrated in Figure 2.12. Figure 2.12: Volumetric response-granular soils

In granular soils where the individual particles are weak, such as calcareous pedocretes crushing tends to occur during treatment. A similar response affects, ash clinker and slag. When these materials are dry the effect of such breakdown is not particularly significant but below the water table the higher proportion of fines developing due to this breakdown results in a rapid change from a granular to a pseudo-cohesive soil response which is discussed in the next section. The presence of very dense layers within the soil profile can cause anomalous volumetric response results. These dense layers tend to absorb the energy impulse as discussed above.

36

However in situations where they are found at shallow depth, the dynamic compaction process will breakup these layers. At greater depths the energy levels required to break up these layers may be beyond the capabilities of the equipment on site. According to AASHTO (1990) the method of dynamic compaction works best on granular soils where the degree of saturation is low, the permeability of the soil mass is high and drainage is good. Care need to be exercised for treatment of soils with significant recovery periods between passes.

2.4.4.2 Cohesive Soils The response of cohesive soils to dynamic compaction is more complex than that discussed above for granular soils. The response is based on the conventional consolidation theory. In general imposing a static load on cohesive soils will expel soil water to induce consolidation and increase strength. The rate at which this occurs is dependent upon the imposed load, coefficient of consolidation and length of drainage paths. In the case of dynamic compaction the surcharge is virtually instantaneous. This imposes stresses that are transferred to the pore water pressure, creating zones of positive pore water pressure gradients, which induce water to rapidly, drain from the soils matrix. The effect is further accelerated by the formation of additional drainage paths by shear and hydraulic fracture. The resultant consolidation therefore occurs much more rapidly than would be the case in static loading. It can therefore be perceived compaction due the dynamic compaction literally squeezes water out the soil to effectively pre-load the ground. A typical volumetric response for cohesive soils is illustrated in Figure 12.13.

Figure 12.13: Volumetric response – cohesive soils

When the soils are above the watertable, the cohesive soils tend to be of relatively low moisture content and the drainage path is relatively short. As such treatment is relatively straightforward and rapid. However where the cohesive soils occur below the watertable, a much larger reduction in moisture content is generally required in the presence of a smaller available pore pressure gradient and longer drainage paths. Such conditions result in threshold energy being achieved more rapidly and many passes which require greatly extended contract periods (due to longer recovery periods) in comparison to normal productivity. The resultant improvement is generally nominal, in thick layers of relatively weak saturated clays and silts. Practical tests conducted with additional measures such as drainage trenches filled with sand or wick drains have not produced significant improvement. For predominantly clay-type fills above the water table, the clay lumps can be considered as large weak particles of an almost granular response. A major improvement occurs in these soils as the voids collapse to provide a more compact structure. The strength of these lumps and sensitivity of the clay also play an important role. Material types and different degrees of weathering can also give rise to a wide variation of responses on a site. Experienced observation is required to define such locations. Mudstone, siltstone, shale and some dolomite fragments can breakdown to material of clayey response particularly when exposed to seasonal wetting and drying.

37

If the clay-type fills discussed above occur below the water table, the voided structure allows relatively higher mobility of water causing relatively lower excess pore water pressures and shorter recovery periods in comparison to intact clays. The compaction of large clay lumps would be achieved mainly by collapse of voids between them as in granular soils. Monitoring of excess pore pressures by mean of piezometer is useful but it can be problematic in partially saturated soils. It is clear that dynamic compaction treatment of clayey materials requires very close, experienced site control. During treatment heave develops around the imprint. In some cases the heave can build up to such an extent that it can exceed the volume of the imprint. Therefore particular care has to be exercised in the timing of successive passes to permit adequate recovery as excess pore pressures dissipate to avoid excessive remolding of the soil. If excessive heave around an imprint starts to occur, it is essential that treatment at this position be stopped. It is better in such soft areas to use twice the number of lighter energy input passes adopting a “softly softly” approach. Alternative soil improvement methods can also be considered. The considerations discussed above apply when attempting to provide treatment to a significant depth where surface layers are cohesive. The strength of these surface soils can be reduced in the short term and time should be spent improving a disturbed matrix to the desired properties. Treatment of cohesive soils will nearly always require a large number of passes when compared to granular material. Efficient treatment is achieved by attempting to provide as much improvement as quickly as possible whilst recognising that the response period will affect the speed of treatment operations. West (1976) has highlighted that cohesive soils continue to improve for a significant period after treatment as excess pore pressures dissipate. Due to the complexities associated with cohesive soil improvement using dynamic compaction it is clear that this method is not best suited for such materials. Hence when considering improvement of cohesive soils this technique should be considered in comparison to methods such as use of stone columns which may be better suited.

2.5

QUALITY CONTROL AND ASSESSMENT Quality control is an essential part of the dynamic compaction programme to ensure that the desired result is achieved, economically without compromising on quality. Aspects of the programme which require attention, site monitoring and control are: • • • •

Safety Induced settlement Environmental considerations Other groundwater, organic content etc.

2.5.1 Safety

38

Since standard cranes are not designed to repeatedly drop heavy weights, the cranes used for dynamic compaction need to be properly modified. Such modification ensures that between 80% and 95% of the potential energy is released at the point of impact. This normally involves modifications to the crane boom cables, drum clutch and outriggers which are necessary to safely attain these high potential energy levels. The cranes also need to be supported by a free draining working surface. The thickness of such surface will depend upon the type of ground being treated. If the thickness of the top 1m is granular, no imported material is generally required. But where the material is cohesive, imported materials are usually necessary particularly when the ground is wet. If there are two or more cranes on site, they should be sufficiently separated to avoid interference and to promote safe working conditions. A distance of at least 30m is usually considered to be safe. Similarly subsequent operations by other contractors may have to be delayed. Flying debris are characteristic of dynamic compaction treatment especially on granular soils. The programme should either have some flexibility to permit treatment to be performed within 50m of features such as roads or buildings, only when the conditions permit its safe operation. Alternatively safety screens could be employed. These however need to be continually moved to be close to the point of impact, otherwise the materials will fly over the top of the screen. Heavy duty membranes are mainly used for this purpose although polythene sheeting, wood hoarding are sometimes used. The later tend to be easily broken by the velocity of impact of the ejected particles. The continual movement of screens can adversely affect productivity and should be incorporated in the programming.

2.5.2

Induced settlement The settlement induced on impact needs to be recorded and monitored. A surveyor is normally employed to assist with measurements, which are often taken after every pass. The measurements include, depth and volumes of imprints, change in elevation of the compacted surface. This information is collated daily to assist in identifying areas that may require additional compaction and to determine the relative increase in compaction achieved. Comparison of results from different phases and passes of prints gives valuable information on the horizontal spread of compaction. The information received is vital for verifying the optimization of the energy input as defined from the test programme. It allows for adjustments of energy input for possible discrepancies recorded on site. Considering that induced settlement is dependant on the total energy input and that initial shape test are conducted when the soil is in its loosest state and that simple extrapolation is often used to analyse these results thereby overestimating the amount of induced settlement. Daily site monitoring is conducted during the contract not only to verify the initial tests but also to allow for adjustments of energy input for possible discrepancies recorded on the site.

2.5.3

Environmental considerations Dynamic compaction employs large, highly visible equipment, which may be imposing on the surroundings. This may be a concern especially if the process will be conducted over an extended. The actual process of soil improvement creates noise and vibration. The

39

acceptable levels of noise and vibration vary from one standard to another, and different countries have different criterion. Noise is generated by impact, lifting and idling. Impact noises attain levels at typically 115 to 120dB at source, which last for about 1% of the lifting cycle. Lower noises produced by lifting and idling meet most environmental limitations at distances of greater than 50m from operation. When conducting dynamic compaction it should be considered that large plate glass windows can sometimes act diaphragms and can change the noise characteristics. Echoes, wind direction and crane exhaust should also be taken into consideration. Ground vibrations are by far the most important environmental consideration. A vibration can be defined by either two of, particle velocity, amplitude, or maximum particle acceleration. The level of vibration transmitted through the ground is an imprecise science, because of the heterogeneous characteristics of soils. The larger vibrations tend to be associated with granular soils and lower vibrations occur in cohesive soils. A higher water table also tends to produce larger vibrations. In places where the soils being treated is directly underlain by relatively dense sand, gravel or rock it tends to transmit vibrations to larger than normal distances with relatively little attenuation. Pre-existing dense surfaces or buried layers can have similar effects causing the transmission of higher than anticipated vibration levels. Since dynamic compaction results in improved inter particle contact of the soils, vibration levels tend to increase towards the end of the treatment operations even though the final impact energy levels are substantially lower than those performed in earlier passes. The most commonly used measure of vibrations is the particle velocity (Vr) of the Rayleigh wave which travels at the air soil interface. The structural effects may be predicted from Table 2.1 Table 2.1 Vr

Structural effects

< 4mm/sec

No structural damage. Annoyance to occupants.

4mm/sec < Vr < 8mm/sec

Damage may occur to sensitive or previously fissured structures.

8mm/sec < Vr

Damage to ordinary structures.

30mm/sec < Vr

Damage to highly rigid structures.

In general these values in Table 2.1 may be applied safely for most structures. Lower values must be adopted for buildings in poor conditions or environmentally sensitive situations such as schools, hospitals and computer installations. Services and activities must be considered on an individual basis depending on their age condition and importance.

40

When vibrations generated by dynamic compaction become a problem their effect can be reduced by reducing the drop height and compensating by increasing the number of drops per imprint. This method will reduce both the impact energy and the effective depth of influence. Alternatively a smaller pounder may be dropped several times through the same drop height with a similar effect. A more popular method, which does not compromise the effective depth of influence, is the excavation of a trench. The effect of such a trench on the propagation of particle velocity is illustrated in Figure 12.14. Figure 12.14: Effect of trench on Rayleigh wave propagation (After Plochigen)

People are generally sensitive in detecting vibrations and have a psychological reaction in believing that damage has been caused even though measurements indicate that the values below threshold levels. A public participation exercise can help overcome concern amongst local residents. It is generally advisable to conduct crack and damage surveys prior to treatment using dynamic compaction.

2.5.4

Other Low permeability soils under saturated or near saturated conditions result in high excess pore water pressures when treated by dynamic compaction as discussed above. Installation of piezometer would be required to monitor such changes and to ensure adequate dissipation of excess pore water pressures. Dynamic compaction often becomes less effective when soils permeability is less than 10-7m/s. Allowing for pore water pressure dissipation usually imposes time restrictions, which often make dynamic compaction unsuitable. Where high organic content exists in fill materials decomposition may result in settlements in the long term. This is a matter of concern, but it should be borne in mind that readily degradable organic materials decompose rapidly if left in a loose state. Therefore if a landfill 5 to 10 years old is compacted, a considerable amount of organic material will already have composed. Furthermore, after compaction, the flow of air to the material is substantially reduced by compaction. Decomposition of the remaining organic material is therefore generally of a nominal nature. In cases where refuse is assorted containing drums, concrete and organic rubble, generally 50 - 100% more total energy is required than in inert soil. Detailed surface observations are very important for such applications. Impact energy may be increased to attain a state in which no further practical induced settlement will occur.

2.6

VERIFICATION

41

Chapter 3

3.1

INTRODUCTION AND BACKGROUND INFORMATION

A geotechnical investigation for the proposed chipping plant building was undertaken in March 1997. Compaction grouting and dynamic compaction were adopted as mitigation for the instability risk presented by the underlying dolomitic conditions. The unravelling of the overburden by solution activity in karst dolomitic formations has caused many problems in existing structures and has resulted in high foundation costs on new construction. Methods normally applied to remedy karst terrain such as slurry grouting, deep foundations and piled foundations have sometimes solved these problems successfully. The method of compaction grouting provides an alternative to more predictably seal the dolomite cavities and stabilise the overburden soils with a degree of economic certainty and relative cost savings. The advancement in the design approach in compaction grouting reduces the potential cost overruns, which are synonymus with the unknowns in dolomite formations.

Following geotechnical investigations on the site it was decided that the site conditions could best be remedied using a combination of the two techniques as discussed in Chapter One and Chapter Two. This chapter presents the procedures followed in the implementation of the ground improvement techniques on a particular case study. It covers interpretations of the findings of the geotechnical investigations, procedures and considerations in the design and implementation phases as well as field monitoring and testing the effectiveness of the applied methods.

The proposed chipping plant building comprises a single storey masonry and steel structure covering an area of approximately 4900 m2. The columns for the structure are on 6 m centres with anticipated maximum column loads of up to 900 kN which have a large dynamic loading component. The building will house heavy chipping and compacting equipment, which is to be linked by a series of conveyor belts. The chipping plant will produce wooden chips for mixing with dried sludge in the process of making manure.

3.2

GEOTECHNICAL INVESTIGATIONS

Various field and laboratory investigations comprising excavation and profiling of auger holes, exploratory boreholes and in situ penetrometer tests were undertaken as part of the geotechnical investigations. Prior to the detailed geotechnical investigation for the proposed building, use of a soil mattress similar to that employed on the adjacent screening plant building was anticipated. However the occurrence of a depression and related settlement of the screening plant building which are

42

attributed to the presence of unstable dolomitic conditions prompted the detailed investigations. Details of the geotechnical investigation are presented in Geotechnical Reports 30128/G2/1997 and 12977/G1/1997 prepared by Africon and are therefore only presented here in brief.

A generalised profile of the strata encountered across the site comprises an upper layer of sandy clayey chert gravel fill up to 2,8 m in thickness. It is underlain by firm to stiff silty and sandy residual soils derived from tillite, shale and sandstone of Karoo origin. The soils are generally reddish grey mottled red brown, and this sequence is encountered at varying depths of between 4 m and 12 m. Underneath the Karoo sediment layer is a thick dolomite residuum horizon comprising variable amount of chert and soft manganiferrous clay (wad). The occurrence of the aforesaid layer is widespread across the site and extends to depths between 35 m and 50 m below ground level. In places the dolomite residuum is found between bands of Karoo sediments. Which could be interpreted as dolomite floaters that have decomposed through time. Substantial sample and air loss which were recorded whilst drilling through this horizon confirm the porous nature of the residuum and suggest the presence of cavities.

Highly irregular weathered dolomite and chert bedrock was encountered at depths varying between 35 and 50 m. Physical cavities were not found during the exploratory drilling, but sample and air loss noted in these boreholes are considered to be an indication that cavities and fissures are present in the strata. The water table is found at a depth of about 14 m from the ground surface.

From these geotechnical investigation findings and the events at the adjacent screening building it was concluded that the soil profile is susceptible to sub-surface erosion and subsidence which normally leads to the formation of sinkholes and dolines. Potential trigger mechanisms leading to instability could be the ingress of water, or vibrations caused by heavy machinery.

Parts of the chipping plant will exert considerable dynamic as well as static loads on the soil profile. The underlying residual dolomite is of utmost concern as it contains relatively soft gravely sandy clay wad that is found below the current water level. The residual dolomite layer is prone to sub -surface erosion by water percolating through the soil profile from the surface and removing the often loose gravel to boulder size chert fragments in the wad matrix transporting it into cavernous areas occurring lower down the profile. It is also prone to sudden collapse type settlement, which can be triggered by dynamic as well as static loads. This can lead to the formation of sinkholes and dolines.

Several founding solutions were considered and these included piling and use of a soil raft/mattress. The piling option using percussion drilled pre-cast driven piles on bedrock (auger drilling refuses at shallow depths on chert boulders) has been considered. It is not considered a practical solution, as the

43

average depth to bedrock dolomite is about 35 m below the surface. This option is likely to be very expensive and would still not eliminate all risk as any sinkhole or collapse type of subsidence related movement within the dolomitic residuum. Furthermore long slender piles in dolomite residuum become susceptible to snapping when the residuum is eroded.

The soil mattress option was also considered and the anticipated mattress would need to satisfy the following conditions,

• to adequately dissipate the heavy loads and bridge subsidence movements of up to 300 mm occurring over areas extending to between 8 m to 10 m in diameter (medium sinkhole/doline) similar to that, which occurred at the screening plant building, •

and still only allow deflections of a magnitude small enough to allow continued operation of the chippers in the proposed building.

These requirements can only be met with a high quality mattress of considerable thickness and adequate stiffness, which can not easily be attained, by an unreinforced soil mattress.

Following considerable evaluation of the options discussed above it was decided to employ compaction grouting of the dolomite residuum and dynamic compaction in the overburden as a proactive measure. The combined use of these methods would solve the deeper and shallower problems respectively. The soil types encountered are considered to be compactable according to the criterion discussed in Section 1.3.2. Compaction grouting is achieved by the injection of non-plastic thick cement grout introduced under pressure at the desired location in a borehole to form a bulb of material, which displaces and compresses the residual dolomite. The main aim of the compaction grouting method is to reduce and perhaps eliminate the risk of structural damage due to dolomite related subsidence. This is attained through, • significant reduction in the soil permeability which in turn reduces the risk of sub-surface erosion due to the ingress of water. • the filling of any cavities which may occur as well as the increasing of the stiffness of the soft or loose material occurring in this zone to restrict the nature and magnitude of ground movement which may occur. • the resultant increase in the lateral pressure and hence the density and shear strength and bearing capacity of the residual dolomite found below the covering layer up to a depth of 25 m.

Dynamic Compaction (DC) is a process employed to densify materials by means of high-energy impacts of a free falling heavy drop weight. The method entails dropping a weight (typically varying

44

from 8 to 15 tonnes) in free fall from heights (typically from 10 to 22 m) on to the ground to be compacted. The shock waves and high stresses induced in the granular soils result in a series of effects viz. •

a densification of the soil by both shear and compression;

•

a partial liquefaction of the grain structure which allows a better re-arrangement of the particles

•

the creation of tension cracks that work as preferential drainage paths through which pore water can percolate more easily, resulting in rapid consolidation.

Due to the envisaged sporadic occurrence of clayey soils and moderate to high loads, it was decided to install stone columns to serve as foundations for some structures. The purpose of the stone columns is to transmit the loads to the underlying competent ground created by the compaction grouting procedures. The technique of placing stone columns is a process whereby granular material is used to dynamically replace the soft soils in the overburden. The method entails dropping a weight (typically varying from 8 to 15 tonnes) in free fall from heights (typically from 10 to 22 m), onto granular rock material placed on top of the soft soil at the point of impact of the falling weight.

The granular

material will be driven into the soil and additional granular material is placed in the area of impact before the weight is dropped again. The process is repeated until no additional granular material can be driven into the soil at this point. A series of effects similar to those stated for the DC works are also experienced when using this technique.

The procedures followed in the design, application monitoring and verification of the two methods of improvement used on this project are discussed in turn.

2.

COMPACTION GROUTING

2.1

Compaction grouting design and initial trial grouting

Specifications compiled for the compaction grouting at the chipping plant building were based on experiences gained in areas underlain by dolomite residuum particularly the stabilisation of road P451 near Westonaria. Quantities of materials were estimated based on grout takes of 0,5 m3 being assumed per metre over the borehole length between 8 m and 25m below the surface. A primary borehole spacing of 6 m was adopted with provisions to incorporate secondary and tertiary boreholes when and if required. These initial estimates were arrived at using the empirical design approach.

45

The new rational design approach after Schmertmann and Henry (1992) discussed in Section 1.3.4 was also applied to fine tune the design derived from the empirical design approach.

Using the following soil parameters obtained from the geotechnical investigations and applying equations 8 to 20 which are presented in Chapter 1. γ’wad = 14.50 kPa (natural)

σvc = 3MPa

γ’wad = 18 kPa (saturated)

θ = 0.33

γ’fill = 18.5 kPa (saturated)

α = 0.5

φg = 40o

Kg =0.8

Ko =0.5 By means of trial and error whilst varying the dimensions of the compaction grout columns d, h and s defined in section 1.3.4. The factors of safety presented in Table 3.1 were obtained. From the table several combinations satisfy the requirement of a safety factor in the order of between 1,5 and 2. The dimensions suggested from the empirical design approach were also tested and they yielded a safety factor of 1,8 it was then decide to adapt these dimensions for the final design of the compaction grouting layout.

Table3. 1: Determination of factors of safety (FOS) that can be achieved with compaction grouting

d h FOS when s =1.5 m FOS when s = 4m

3 25 99

3 15 63

3 5 35

1.5 17 29

1.5 15 26

1.5 10 17

1 15 15.6

1 20 21

1 25 28

13

8

4.9

4.4

3.58

2.5

2.2

3

3.9

FOS when s = 6m

6.1

3.9

2.1

1.8

1.25

1.1

0.98

1.3

1.7

FOS when s = 2.2 10m

1.4

0.8

0.6

0.57

0.4

0.35

0.4 7

0.62

A preliminary grid with primary borehole positions located at 6 m centres was designed for the site. In this grid secondary boreholes were positioned equidistant from the adjacent primary boreholes. The idea behind the use of secondary boreholes was to two fold. Firstly to check that adequate compaction is achieved by the grouting done in the adjacent primary boreholes and secondly to augment compaction in areas where the primary boreholes could not attain the intended degree of compaction due to the presence of large void spaces.

A part of the site where geotechnical investigations predicted the presence of extensive voids and was identified as a trial section. In the trial section initially twelve primary boreholes along rows A and B were drilled to 25 m depth and grouted using the upstage grouting technique. At each stage either

46

grout was injected into the surrounding ground until the maximum allowable grout pressure corresponding with the depth as given in the guidelines in Table 1, from section 1.4.3 was recorded at the borehole collar or until 0.5 m3 of grout had been discharged at the stage.

Grout acceptance results observed in the boreholes within the test section indicated the adjacent primary boreholes recorded grout takes in excess of 2,5 m3 per stage below 25 m depth with no significant improvements in the grout pressure. At this point secondary boreholes were drilled at equidistant locations from the primary boreholes. This was an attempt to ensure that no zones of soft residual dolomite would remain untreated. The average grout takes per stage in the secondary boreholes was just below 0,5 m3 before the refusal pressures were attained. Based on the grouting results obtained from the trial section recommendations pertaining to the maximum grout injection per stage were revised to 2,5 m3.

Cube tests were conducted on the grout to determine the average compressive strength of the initial grout mix. Results of 3,1 MPa and 9,3 MPa after 7 days and 28 days of curing respectively were obtained. These strengths are much higher than the required strengths and were therefore reviewed and the cement content was reduced to half a bag of cement (25 kg). Further cube test results obtained after using the reduced cement content yielded an average compressive 28 day strength of 2 MPa.

The upstage grouting technique which is described in more detail below was employed. This technique was preferred over the alternative downstage technique which works from the top downwards and involves considerable additional drilling costs. The preferred technique also offers the advantage of yielding a full depth profile of the strata, which enables the areas that require treatment to be recognised more easily.

2.1

Drilling

The rotary percussion drilling method using down the hole hammers was used. It was assumed that a primary borehole spacing of 6 m would be sufficient to intersect all the possible grykes and cavities of any significance. The applied grout pressures were considered adequate to compact the materials within a 3 m radius of each borehole. Positions of the boreholes are shown on **Drawing 12977/G1/1**.

Primary boreholes were drilled along rows A to X at 6 m intervals. Secondary boreholes were positioned between consecutive rows where they were required at 3 m equidistant from all adjacent

47

boreholes. A total of 4 771 m of 102 mm diameter boreholes was drilled by means of the rotary percussion method using down the hole hammers. However in places (less than 2% of the site) where difficulty was experienced when drilling, especially where significant clay was encountered alternative drilling techniques such as tricone and or drag bit were used.

The contract was successfully completed using a Sonda B50LS fully hydraulic track rig supplied by Rodio shown in plate 1. Whilst this rig is generally slower than the conventional down the hole hammers rotary percussion rigs it managed to stay ahead of the grouting exercise. The air pressure was supplied to the rig using an Ingersoll-Rand 750 compressor with a 17 bar pressure capacity supplied by Compair.

Redrilling was employed primarily to clear chert gravel and wad material stuck at the bottom of the tremie pipes prior to grouting. Three of the initial boreholes were redrilled after grouting to assess the effectiveness of the grouting operation.

5.1

DRILLING AND GROUND PROFILES

Of the 4 771 m drilled during this programme 528 m were drilled in 23 secondary boreholes, whilst the rest were in the 179 primary boreholes. The engineer on site decided upon the positions of these boreholes after analysis of the grout takes recorded in the adjacent boreholes using the criteria described in section 4.5.2. An average of four boreholes was recorded for the duration of the contract, the rate of drilling was haphazard with a range of between six and one boreholes drilled per day. The rate of drilling was dictated by breakdown of the drilling rig, slower rate of grouting and hence availability of tremie pipes.

Samples could only be collected for 50% of drilled meterage on average. There was a widespread loss of sample below the water table and in areas where air loss occurred in thick sections of leached dolomite residuum. However part of the sample loss was due to the low air pressure used for the drilling on some of the days when the compressor was not working to its full capacity. In instances were samples could not be recovered the profile was inferred from the drilling records relying largely on the penetration times kept by the driller. Detailed geological profiles are kept at Africon offices and summaries are presented on Drawing 12977/G1/2c.

48

Plate 1: Sonda B50LS hydraulic track mounted percussion rig

4.3

LOGGING OF DRILLING SAMPLES

Samples of drill cuttings recovered for every metre of advance of the borehole were bagged and tagged. The driller also recorded the rate of penetration for each metre, as well as information pertaining to air loss, presence and absence of water and general drilling conditions. An engineering geologist logged each borehole and a profile of the encountered geological strata was prepared. Plate 2 shows sample of the typical ground profiles encountered on the site, which were recovered in the drilling process.

Since the drilling information was instrumental in deciding whether the tremie pipe used for the grouting was installed to the correct depth in the borehole and also in determining if secondary boreholes were required. The logging of the sampled chips was conducted daily to keep pace with the drilling.

49

Plate 2: Typical samples of the geological strata recovered during drilling

4.4

TREMIE PIPE (CASING)

To ensure grout was pumped to the desired depth 76 mm diameter tremie pipes where installed in the boreholes prior to the grouting operation. The tremie pipe was available in variable lengths up to 3 m and was installed using the drilling rig. In some boreholes immediate collapse occurred in places were the formation was broken. In these instances the percussion drill was employed to push the tremie pipe to the required depths. A total of 4599 m of tremie pipe was installed using this method.

2.3

DYNAMIC COMPACTION AND STONE COLUMNS The area treated by dynamic compaction and stone columns measures approximately 3 500 m2 and is shown in Drawing S/2/18/948. Trial tests were to be conducted to determine the best weight and drop height relationship to attain the specified minimum stiffness. A minimum field stiffness requirement

50

of 15 MPa was prescribed for the area earmarked for Dynamic Compaction and 30 to 50 MPa as shown on Drawing S/2/18/948 was prescribed in the area earmarked for Stone Column work.

51

4.

WORKING METHODS

4.1

LAYOUT OF WORKS

The mixer and pump used for the grouting operation was positioned firstly in the northern end and later near the centre of the site where it could be used to service all boreholes within 50 m of its reach. On completion of the grouting operations, dynamic compaction and installation of the stone columns followed, firstly with the improvements on the ground surface and then in the excavations.

4.5

GROUTING General

4.5.1

The primary objective of the grouting exercise was to fill up cavities, densify the strata thereby increasing its bearing capacity, stiffness and reducing the permeability of the dolomitic residuum and unconsolidated material above the dolomite bedrock. In principle such an operation restricts the nature and magnitude of ground movements that may result due to dolomite related instability. The success of the grouting technique is dependent on the fact that the cement grout is successfully placed below the water table by an upstage grouting technique. Feasibility studies and other completed stabilisation projects have shown that a grout mix can be designed which is sufficiently strong, erosion resistant and stable with respect to segregation when injected into water filled cavities.

The principles established in previous stabilisation projects formed the basic modus operandi for the grouting contract.

4.5.2

Grouting technique

The upstage grouting technique involved: • drilling of the borehole to its predetermined depth of 25 m, this was later revised to a depth of 26 m to cater for the spoil which usually accumulates in the last meter of the borehole;. • installing of tremie pipe to the bottom of the borehole;. • lifting the tremie pipe to the required height and; • pumping grout until refusal pressures are attained or a maximum of 2,7 m3 was pumped.

52

• repeating the lifting of tremie pipe in one meter stages and pumping until the base of the covering layer (about 8 m depth) is reached.

The pumping pressure of the grout was closely monitored during the operation using a pressure gauge installed at the collar of the borehole being grouted. The maximum pressures used to determine refusal take into consideration the pressure losses as a result of friction and the overburden pressure. Table 1 shows the pressure guidelines used during the contract.

53

Table 1: Guidelines of the maximum allowable grout pressures

Depth (m)

Pressure (MPa)

0-5

0,4

5 - 10

1,0

10 - 15

1,5

15 - 20

2,4

20 - 25

3,0

25 - 30

3,5

30 - 35

4,0

A maximum volume injected per stage in each borehole was initially set at 2 m3 but later revised to 2,7 m3 after the test grouting and the associated measurement complications.

The grouting was undertaken in rows of primary boreholes at six meter centres ensuring a distance of at least 20 m was maintained from the drilling rig to minimise the adverse effect of the air pressures used for drilling. Constant monitoring and analyses of the grouting results was maintained to decide on the requirement and location of secondary boreholes. Criterion for deciding on the location of secondary boreholes was largely dependent on the grout takes recorded on the adjacent boreholes.

Following the grout acceptance results observed in the test grouting conducted in the initial rows, it was decided that if the adjacent primary boreholes recorded grout takes in excess of 20m3 with no significant decrease in the grout volume above the 20 m depth, a secondary borehole would be located equidistant from the primary boreholes. This was to ensure that no zones of soft residual dolomite remain untreated.

4.5.3

Grout mix and plant

The original grout mix comprised the following:

water

40 kg

7,8 %

Cement

50 kg

9,8 %

sand

420 kg

82,3 %

The first batch of cube test results obtained had average compressive strengths of 3,1 MPa and 9,3 MPa after 7 days and 36 days of curing respectively. These strengths are much higher than the required strengths and were therefore reviewed and the cement content was reduced to half a bag of

54

cement (25 kg). Further cube test results obtained after using the reduced cement content yielded an average compressive 28 day strength of 2 MPa. The plant used in the grouting operation comprised a Putzmeister grout pump, a concrete mixer and a hydraulic grout mixer. These plant are shown in plates 3 and 4. A hydraulic casing extractor shown in plate 5 was utilised to extract the tremie pipe at the end of grouting of each stage.

Ingredients of the grout mix were mixed on site using a concrete mixer. Quantities were used in the ratios as given above. For ease of measurement, three wheelbarrows of sand were mixed with a bag of cement and the water content was varied according to the moisture content. After thoroughly mixing the grout it was poured into the pump for dispatch to the boreholes.

The grout placement procedure entailed the following: • setting up of scaffold and the extractor jack placed in position locking onto the tremie pipe. • the delivery hose was connected to the tremie pipe and some water was pumped through to ensure the pipe was not blocked. • the grout was then poured from the mixer into the pump from which it was discharged into the borehole via the hose. • grout was pumped into the borehole until either the desired pressure (Table 1) or the stipulated 2,7 m3 was attained. • the hydraulic extractor was used to lift the tremie pipe to the next stage and the pumping procedure was repeated until a depth of 8 m from the surface was reached. • on completion of grouting of each of the holes the tremie pipe was removed, washed and moved to equip the next available borehole.

Two pressure gauges were used to monitor the pressure of the grout flowing through the hose. One pressure gauge was located at the exit point from the pump whilst the other was at the collar of the borehole. The readings on these two gauges were constantly monitored. The reading at the gauge located at the pump was calculated to be twice that at the borehole collar due to losses incurred along the hose with major variations occurring when either the hose blocked or when refusal has been attained.

55

Plate 3: Concrete mixer

Plate 4: Putzmeister grout pump

56

Plate 5: Hydraulic casing extractor

4.6

DYNAMIC COMPACTION AND STONE COLUMNS

The work was conducted by Geofranki using a P& H 1055 80 tonne crawler crane. Two types of pounders were used, namely the 14 tonne penetration pounder and a 12 tonne ironing pounder for the stone columns and dynamic compaction respectively. The dynamic compaction was conducted by dropping the 12 tonne ironing pounder through a free fall distance of 20 m. A single pass was done at each of the points, which were marked out on the ground at 1m centres. The distances between the adjacent points are approximately half the diameter of the ironing pounder, which ensures that there is at least a 50 % overlap of the area, being ironed. These relationships were arrived at following the trial ironing conducted by Geofranki on the site to ensure that the specified field stiffness is attained. The photograph in plate 6, shows the crane and ironing pounder just before the pounder is released.

57

Plate 6: P & H 1055 crawler (80 tonne) crane with a hexagonal 12 tonne ironing pounder.

The stone columns were installed using the same crane as shown above, but the ironing pounder was replaced by the heavier 14 tonne penetration pounder. The procedure for the stone columns was as follows; •

An excavation to a nominal depth of 1m was excavated from the surface,

•

Measured volumes of stone were deposited at the bottom of the hole and,

•

The penetration pounder was dropped from a height of 19 m, this was repeated ten times for the first pass.

•

More stone was added and the pounder dropped three times for the second pass.

•

The procedure was repeated until, the stone could no longer be pushed into the soil, which was normally indicated by the bulging of the ground at the edges of the stone columns.

58

An illustration of the installation of a stone column is shown in plate 7.

Plate 7: Stone column before penetration is induced by the 14 tonne pounder.

4.7 4.7.1

QUALITY CONTROL, AND SUPERVISION Drilling Records of the drilling, casing, installation of tremie pipe and occurrence of groundwater were maintained throughout the contract. The compiled geological profiles were used to determine the grouting sequence i.e. ensuring that areas with suspected cavities and leached dolomite residuum were grouted before areas which contained deep Karoo bedrock.

4.7.2

Grouting Quality control consisted of cube strengths and slump tests to measure the consistency of the grout. Random cube samples were taken during the contract and sent for crushing at a laboratory to ascertain the strength of the grout once it had set. Results of the cube tests are discussed in section 4.5.3. Slump tests were periodically undertaken and randomly at the request of the engineer to ensure that the grout complied with the specification of between 25 and 100 mm slump.

4.7.3

Dynamic Compaction and Stone Columns Quality control for these works was in the form of records of the number of passes conducted at each of the points marked on the ground for the dynamic compaction. Volume of stone for each pass and the number of blows delivered at each pass were also recorded. Plate load tests were conducted at the request of the engineer, these covered both the stone columns and the areas treated by the ironing. These tests were conducted to evaluate the stiffness attained by the DC and Stone Columns.

59

4.7.4

Monitoring Prior to commencing with the drilling and grouting, a surveyor installed several monitoring pegs and the existing cracks in the near-by substation were mapped. Readings were taken every fortnight to check on the movement that could be induced by the compaction grouting. No movement was recorded over the period of the contract.

An accelerometer was used to monitor the particle acceleration caused by the waves induced into the ground by the pounding action of the DC and SC work. In addition, monitoring pegs were installed and these were monitored regularly by a surveyor. Throughout the contract, the cracks in the substation were also monitored and no significant movement was recorded.

5.

RESULTS

5.2

GROUT ACCEPTANCES

Based on experiences on compaction grouting contracts completed by ourselves and the contractor it was assumed that unless a cavity or a wad filled cavity was intersected during drilling then grout takes would be relatively low or in the region of the measurements predicted at the tender stages. It was further assumed that the grout takes of the individual boreholes would decrease with the successive stages of grouting, viz. that primary boreholes would have a higher grout take relative to the secondary boreholes.

The results of the grout takes were plotted on the borehole layout plan and zoned according to the total grout takes recorded for each borehole. The different zones are defined below and shown in Drawing 12977/G1/1. It is interesting to note that the emerging pattern on the drawing implies a configuration of typical subsurface dolomite morphology, viz. wad filled cavities and dolomite rock head normally found in the West Rand. The distribution of boreholes considered to have ‘low grout takes’ (less than 12 m3) is coincident with the zone where Karoo conglomerate rock extends to depths of between 16 m and 20 m. The Karoo conglomerate is usually underlain by dolomitic residuum which largely comprises abundant chert rubble with minor or no wad. This material is distributed across the site, but are more prevalent on

60

the western and southern extremes of the treated area. The cross section along line D, Drawing 12977/G1/2b illustrates the distribution of grout, which is typical of the ‘low grout takes’ zone. Areas in which boreholes had grout takes between 12 m3 and 20m3 are considered to be in the zone of ‘moderate grout takes’. The geological profiles of the boreholes drilled in such zones indicate that the Karoo rocks are generally not as thick as the zone of low grout takes. Whilst the underlying dolomitic residuum comprised alternating sub horizons of abundant chert with minor wad and abundant wad with minor chert. The latter sub horizons were responsible for up to 50 % of the grout takes in this zone and are usually located below 20 m from the ground surface. The cross section along line J, Drawing 12977/G1/2c illustrates the distribution of grout, which is typical of the ‘moderate grout takes’ zone. The third zone is defined by areas where boreholes recorded grout takes in excess of 20 m3 and is regarded as a zone of ‘high grout takes’. Grout takes of up to 33 m3 were recorded in this zone. The typical geological profile in this zone comprised a surfcial cover of fill, transported material and Karoo conglomerate up to 10 m thick underlain by a wad dominated dolomite residuum. Drilling in this zone was characterised by extensive sample and air loss, which are indicative of the presence of carvenous ground. Over 75 % of the grout takes recorded in this zone are confined below 20 m depth measured from the ground surface. The cross section along line A, Drawing 12977/G1/2a illustrates the distribution of grout, which is typical of the ‘high grout takes’ zone.

Twenty three secondary boreholes were drilled and grouted. The function of the secondary boreholes was twofold. Firstly to ensure that adequate compaction and filling of cavities and wad filled cavities is attained in areas where the specified grouting pressures are not met in primary boreholes. And secondly, as a measure to evaluate the effectiveness of the grouting completed in the adjacent primary boreholes.

Grout takes in the secondary boreholes show that the grout takes below 20 m are similar to those recorded in the adjacent primary boreholes. This was particularly true in the zone of ‘high grout takes’ which are in excess of 20 m3 and is considered to be an indication that inadequate compaction was attained below 20 m in this zone. However significant decreases of up to 60 % in the intake of grout were recorded between depths of 12 m and 20 m in the same boreholes, which indicates that considerable compaction has been attained at these depths.

5.3

GROUT TAKES PREDICTIONS

61

At the tender stage there were difficulties in accurately predicting the quantity of grout that could be injected in the area to be stabilised. The estimated average of 8,5 m3 per borehole was arrived at based on the intakes recorded on previous contracts completed by the contractor and Africon. However when the grouting programme commenced an average of 19,71 m3 per borehole was recorded in the areas which were expected to have high takes according to findings of the geotechnical investigation. As the grouting progressed it became apparent that the earlier high grout takes were anomalous and an average grout take of 12,27 m3 per borehole was adopted. The latter average was retained and successfully used for projections regarding the grout volumes and the cost implications.

5.4

RESULTS OF THE DC AND SC

Detailed results of the work conducted by Geofranki are included in Appendix A of this report. A summary of the volume of stone used and the number of blows delivered for each pass are given in Table 2.

62

Table 2 : Summary of stone volumes and blow counts

Location

Pass 1 Vol. 3

Pass 2 Blow

Vol.

Pass 3 Blow

3

Vol.

Pass 4 Blow

3

Blow

3

(m )

s

Vol.

Pass 6 Blow

3

(m )

s

Vol.

Blow

3

(m )

s

(m )

s

(m )

2.86

10

-

-

-

-

-

-

-

-

-

-

3.13

10

-

-

-

-

-

-

-

-

-

-

CV 42

6

10

2.73

3

-

-

-

-

-

-

-

-

Drum

4.15

10

-

-

-

-

-

-

-

-

-

-

0

20

-

-

-

-

-

-

-

-

-

-

7.39

7

7.02

7

-

-

-

-

-

-

-

-

3.26

3

1.9

1

3.29

10

17

3

Feeder

s

Vol.

Pass 5

(m )

s

Hopper 1 Feeder Hopper 2

Chipper Knuckle Boom Building

5 Perimeter

9.1

15

7.8

6

Note: Values given here are average values for the stone columns located at these areas.

From the results shown in Table 2 it is clear that there was a variation in the amount of stone used as well as the blows delivered. The variation is mainly due to the variations in the depth to a layer highly weathered Karoo sediments on which the stone columns rest. In the knuckle boom area generally no stone was used as the excavations refused on the weathered Karoo sediments, attempts were however made to install stone columns without much success and where hence abandoned after 20 blow counts. Meanwhile on the other hand, in the areas around the CV42 (perimeter) substantial amounts of stone were used to create the stone columns. This is mainly because in this section the unconsolidated soil and fill are much thicker than elsewhere on the site. Plate load bearing tests were conducted at the positions indicated on Drawing S/2/19/248. The tests were done using large diameter plates (1 m) at the surface and on top of stone columns. For most of the tests the crane was used to act as kentledge with the exception of test number three which was conducted at the bottom of the chipper drum excavation. Plate 8 shows a test being conducted on a stone column. Results of the plate bearing test are summarised in Table 3 and the detailed results are included in Appendix A. Table 3: Plate Load Test Results

Location

Test Number

Secant Modulus (MPa)

Ironing at surface

1

71,65

Stone Column

2

158,44

Bottom of Drum chipper

3

25,28

63

Ironing at surface (heave area)

4

24,09

Bottom of excavation at CV 031

5

12,10

Plate 8: Plate Bearing Test on a Stone Column

6.

CONCLUSION The unique technique of compaction grouting has been applied to reduce the risk and magnitude of dolomite related subsidence of the Chipping Plant building at Olifantsvlei Sludge Handling facility. This method was opted for in favour of piling and soil mattress as discussed in the report.

The compaction grouting work conducted at this site has shown that large quantities of grout are required to fill up cavities and to compact soft wad, which occupies most of these voids. At the chipping plant site the majority of boreholes intersecting such ground conditions recorded very high grout takes to attain adequate compaction. This scenario was paramount in zones of high grout takes, which are mainly located below 20 m depth. Compaction was verified by the significant reduction in the grout takes recorded in the secondary boreholes.

64

The design philosophy employed for the compaction grouting entailed grouting of the unconsolidated material between 8 m and 25 m depth. It is considered that the compaction achieved between 8 m and 25 m will adequately reduce the permeability of the strata and hence minimise the ingress of surface water which is one of the main causes of dolomite related subsidence. Excavations for footings, which are underway at present, have revealed that some storm water had been retained in the stone columns. This is a clear indication that the permeability of the soil has been significantly reduced. This layer of compacted ground is of adequate thickness and stiffeness which, forms a raft capable of spanning across any dolomite related subsidence that may occur below it.

The results of the stiffness measured on the stone columns and the areas improved by dynamic compaction are higher than those required to cater for the design loads. It is therefore concluded that the foundations of the structure are adequate for the intended loading configuration.

65

66

APPENDIX B

DRAWINGS

67

Byle MJ and Borden RH (1995) “Verification of Geotechnical Grouting”, Geotechnical Special Publication No 57, ASCE, October 23-27, 1995.

Schmertmann JH and Henry JF (1992) “A Design Theory for Compaction Grouting”, Grouting, Soil Improvement and Geotsynthetics, RH Borden ed., ASCE Geotechnical Special Publication No. 30 pp 275 to 287.

Warner J (1992), “Compaction Grout; Rheology vs Effectiveness”, Grouting Soil Improvement and Geosynthetics, RH Borden ed., ASCE Geotechnical Special Publication No 30 pp 229-239.

Brown DR and Warner J, (1973) “Compaction Grouting” Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 99, No. SM8, Proc. Paper 9908, August 1973.

Warner J, and Brown DR, (1974) “Planning and Performing Compaction Grouting”, Journal of the Geotechnical Engineering Division, ASCE Vol. 100, No. GT6, Proc. Paper 10606, June 1974, pp 653-666.

Xanthakos PP (1994) “Ground Control and Improvement”, John Wiley & Sons Inc New York, NY.

Nonveiller E (1989) “Grouting Theory and Practice”, Developments in Geotechnical Engineering, 57 Elsevier Science Publishers.

Graf ED (1992), “Compaction Grout “ (1992), Grouting, Soil Improvement and Geotsynthetics, RH Borden ed., ASCE Geotechnical Special Publication No 30, pp 275-287.

Warner J (1982), “Compaction grouting – the first thirty years”, Proceedings of the Conference on Grouting in Geotechnical Engineering, WAH Baker ed., ASCE pp 694-707.

Stilley AN (1982), “Compaction Grouting for Foundation Stabilisation” Proceedings of the Conference on Grouting in Geotechnical Engineering, WH Baker ed., ASCE pp 923-937.

68

Vesic, AS, (1972) “Expansion of Cavities in Infinite Soil Mass”, Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 98 No. SM3, Proc. Paper 8790 pp 265-290. Bowen R (1981), “Grouting in Engineering Practice” 2nd Edition; Applied Science Publishers Ltd, London.