Spacing of soil samples.

It is frequently specified that soil samples should be taken at intervals of 1.5 m and at each change of strata in boreholes While this spacing may be adequate if a large number of boreholes is to be drilled, there can be a serious deficiency in quantitative soil information if the size of the area under investigation warrants only a few boreholes The lack of information is particularly noticeable where structures with shallow foundations are proposed Thus it is quite usual for the first sample to be taken just below the topsoil, say from 0.2 to 0.7 m

The next, at the 1.5 m spacing, is from 1.7 to 2.2 m Exactly the same depths are adopted for all the boreholes on the site It is normal to place foundations in clay at a depth of 0.9 or 1.0 m Thus there is no information on soil shear strength and compressibility at and for a distance of 0.8 m below foundation level, probably within a zone where there are quite large variations in soil characteristics due to the effects of surface desiccation (Fig. 1.2)

Where only a few boreholes are to be sunk it is a good practice to adopt continuous sampling for the first few metres below ground level or to stagger the sampling depths where the 1.5 m spacing is adopted.

Soil sampling - Foundations.

There are two main types of soil sample which can be recovered from boreholes or trial pits

(a) Disturbed samples, as their name implies, are samples taken from the boring tools examples are auger parings, the contents of the split-spoon sampler in the standard penetration test, sludges from the shell or wash-water return, or hand samples dug from trial pits

(b) Undisturbed samples, obtained by pushing or driving a thin-walled tube into the soil, represent as closely as is practicable the in-situ structure and water content of the soil. It is important not to over-drive the sampler as this compresses the contents

It should be recognized that no sample taken by driving a tube into the soil can be truly undisturbed

Disturbance and the consequent changes in soil properties can be minimized by careful attention to maintaining a water balance in the borehole That is, the head of water in the borehole must be maintained, while sampling, at a level corresponding to the piezometric pressure of the pore water in the soil at the level of sampling This may involve extending the borehole casing above ground level or using bentonite slurry instead of water to balance high piezometric pressures

The care in sampling procedure and the elaborateness of the equipment depends on the class of work which is being undertaken, and the importance of accurate results on the design of the works.

BS 5930 recommends five quality classes for soil sampling following a system developed in Germany

The classification system, the soil properties which can be determined reliably from each class, and the appropriate sampling methods are shown in Table 11

In cohesive soils sensitive to disturbance, quality classes 1 and 2 require a good design of sampler such as a piston or thin-walled sampler which is jacked or pulled down into the soil and not driven down by blows of a hammer Class 1 and 2 sampling in soils insensitive to disturbance employs open-drive tube samplers which are hammered into the soils by blows of a sliding hammer or careful hand-cut samples taken from trial pits There is a great difference in cost between piston and open sampling, but the engineer should recognize the value of good quality if this can result in economies in design, for example, good-quality sampling means higher indicated shear strengths, with higher bearing pressures and consequently reduced foundation costs

In certain projects good sampling may mean the difference between a certain construction operation being judged possible or impossible, for example the placing of an embankment on very soft soil for a bridge approach If shear strength as indicated by poor-quality sampling is low, then the engineer may decide it is impossible to use an embanked approach and will have to employ an expensive piled viaduct On the other hand in 'insensitive' clays such as stiff glacial till the sampling procedure has not much effect on shear strength and thick-wall open samplers may give quite adequate information Also, elaborate samplers such as the fixed piston types may be incapable of operation in clays containing appreciable amounts of large gravel

Table 1.1 Quality classification for soil sampling

The presence of discontinuities in the form of pockets or layers of sand and silt, laminations, fissures, and root holes in cohesive soils is of significance to their permeability, which in turn affects their rate of consolidation under foundation loading, and the stability of slopes of foundation excavations The use of large-diameter sample tubes may be justified to assess the significance of such discontinuities or 'fabnc' to the particular foundation problem

The engineer should study the foundation problem and decide what degree of elaborateness in sampling is economically justifiable, and he should keep in mind that in-situ tests such as the vane or cone tests may give more reliable information than laboratory tests on undisturbed samples If in-situ tests are adopted, elaborateness an undisturbed sampling is unnecessary and the 'simple' class is sufficient to give a check on identification of soil types

A good practice, recommended by Rowei2 is to adopt continuous sampling in the first boreholes drilled on a site An open-drive sampler with an internal split sleeve is used to enable the samples to be split longitudinally for examination of the soil fabric The critical soil layers can be identified and the appropriate class of sampling or zn-situ testing adopted

BS 5930 gives details and dimensions of five types of soil samplers for use in boreholes

These are Thin-walled samplers

General-purpose 100mm diameter open-tube sampler

Split-barrel standard penetration test sampler

Thin-walled stationary piston sampler

Continuous sampler

Thin-walled samplers which are pushed rather than hammered into the soil cause the minimum of moisture content changes and disturbance to the fabric of the soil Sample diameters are generally 75—100 mm, but tubes up to 250 mm can be provided for special purposes The thin-wall sampler is suitable for use in very soft to soft clays and silts

One type of thin-walled sampler, not described in BS 5930, is the Laval sampler developed in Canada for sampling soft clays ' It has been shown to provide samples of a quality equal to those obtained by conventional hand-cut block sampling The tube is hydraulic- ally pushed into a mud-supported borehole to recover samples 200 mm in diameter and 300 mm long The tube is over-cored before withdrawal

The general-purpose 100 mm diameter open-tube sampler was developed in the UK as a suitable device for sampling the very stiff to hard clays, gravelly glacial till, and weak weathered rocks such as chalk and marl

In this respect the detachable cutting shoe is advantageous It can be discarded or reconditioned enabling many reuses of the equipment However, the relatively thick-walled tube and cutting shoe do cause some disturbance of the fabric of the soil and moisture content changes within the sample The equipment is suitable for geotechmcal category 2 investigations

The split-barrel standard penetration test sampler is used to make the in-situ soil test The tube has an internal diameter of 35 mm and recovers a disturbed sample suitable for classes 3 and 4 in Table 11 Some indication of layering or laminations can be seen when the sampler is taken apart

Thin-walled stationary piston samplers are suitable for quality class 1 m Table 11 and for geotechnical category 3 investigations Diameters range from 75 to 100 mm with special types up to 250 mm They recover good samples of very soft to soft clays and silts, and sandy soils can sometimes be recovered Special thin- wall piston samples are used in stiff clays

The Delft continuous sampler is an example of this type It is made in 29 and 66 mm diameters with a penetration generally up to 18 m, but samples up to 30 m can be recovered in favourable soil conditions

It is designed to be pushed into the ground using the 200 kN thrust of the standard cone penetration test sounding machine (see below)

The samples from the 66 mm tubes are retained in plastic liners which can be split longitudinally to examine the stratification and fabric of the soil

Figure 1.2 Lack of information on shear strength at foundation level due to adoption of uniform sampling depths in all boreholes

Trial pits and borings - Soils - Foundations.

Trial pits are generally used for geotechnical category 1 investigations They are useful for examining the quality of weathered rocks for shallow foundations Trial pits extended to trenches provide the most reliable means of assessing the state of deposition and characteristics of filled ground

Hand and mechanical auger borings are also suitable for category 1 investigations in soils which remain stable in an unlined hole When carefully done augering causes the least soil disturbance of any boring method Light cable percussion borings are generally used in British practice The simple and robust equipment is well suited to the widely varying soil conditions in Britain, including the very stiff or dense stony glacial soils, and weathered rocks of soil-like consistency

Large-diameter undisturbed samples (up to 250 mm) can be recovered for special testing

Rotary open hole drilling is generally used m USA, Middle East, and Far Eastern countries The rotary drills are usually tractor or skid-mounted and are capable of rock drilhng as well as drilling in soils Hole diameters are usually smaller than percussion-drilled holes, and sample sizes are usually limited to 50 mm diameter

Bentonite slurry or water is used as the drilling fluid, but special foams have been developed to assist in obtaining good undisturbed samples

Wash borings are small-diameter (about 65 mm) holes drilled by water flush aided by chiselling Sampling is by 50 mm internal diameter standard penetration test equipment (see below) or 50—75 mm open-drive tubes

Wash probings are used in over-water soil investigations They consist of a small-diameter pipe jetted down and are used to locate rock head or a strong layer overlain by loose or soft soils, for example in investigations for dredging There is no positive identification of the soils and sampling is usually impracticable

Exploration in soils: Investigation methods

Methods of determining the stratification and engineering characteristics of subsurface soils are as follows

Trial pits

Hand auger borings

Site investigations and soil mechanics

Mechanical auger borings

Light cable percussion borings

Rotary open hole drilling

Wash borings

Wash probings

Dynamic cone penetration tests

Static cone penetration tests

Vane shear tests

Pressuremeter tests

Dilatometer tests

Plate bearing tests

Detailed descriptions of the above methods as used in British practice are given in BS 5930

Site Investigations. Brief comments on the applicability of these methods to different soil and site conditions are given in the following Sections 1 42—1 45

1.3 Borehole layout

Whenever possible boreholes should be sunk close to the proposed foundations This is important where the bearing stratum is irregular in depth For the same reason the boreholes should be accurately located in position and level in relation to the proposed structures

Where the layout of the structures has not been decided at the time of making the investigation a suitable pattern of boreholes is an evenly spaced grid of holes For extensive areas it is possible to adopt a grid of boreholes with some form of zn-situ probes, such as dynamic or static cone penetration tests, at a closer spacing within the borehole grid EC 7 recommends, for category 2 investigations, that the exploration points forming the grid should normally be at a mutual spacing of 20—40 m Trial pits for small foundations, such as strip foundations for houses, should not be located on or close to the intended foundation position because of the weakening of the ground caused by these relatively large and deep trial excavations

The required number of boreholes which need to be sunk on any particular location is a difficult problem which is closely bound up with the relative costs of the investigation and the project for which it is undertaken

Obviously the more boreholes that are sunk the more is known of the soil conditions and greater economy can be achieved in foundation design, and the risks of meeting unforeseen and difficult soil conditions which would greatly increase the costs of the foundation work become progressively less However, an economic limit is reached when the cost of borings outweighs any savings in foundation costs and merely adds to the overall cost of the project

For all but the smallest structures, at least two and preferably three boreholes should be sunk, so that the true dip of the strata can be established

Even so, false assumptions may still be made about stratification

The depth to which boreholes should be sunk is governed by the depth of soil affected by foundation bearing pressures The vertical stress on the soil at a depth of one and a half times the width of the loaded area is still one-fifth of the applied vertical stress at foundation level, and the shear stress at this depth is still appreciable Thus, borings in soil should always be taken to a depth of at least one to three times the width of the loaded area

In the case of narrow and widely spaced strip or pad foundations the borings are comparatively shallow (Fig 1 1(a)), but for large raft foundations the borings will have to be deep (Fig 1 1(b)) unless rock is present within the prescribed depth Where strip or pad footings are closely spaced so that there is overlapping of the zones of pressure the whole loaded area becomes in effect a raft foundation with correspondingly deep borings (Fig 11(c)) In the case of piled foundations the ground should be explored below pile-point level to cover the zones of soil affected by loading trans- mitted through the piles EC 7 recommends a depth of five shaft diameters below the expected toe level It is usual to assume that a large piled area in uniform soil behaves as a raft foundation with the equivalent raft at a depth of two-thirds of the length of the piles (Fig 1 1(d))

Figure 1.1 Depths of boreholes for various foundation conditions The 'rule-of-thumb' for a borehole depth of one and a half times the foundation width should be used with caution

Deep fill material could be present on some sites and geological conditions at depth could involve a risk of foundation instability

Where foundations are taken down to rock, either in the form of strip or pad foundations or by piling, it is necessary to prove that rock is in fact present at the assumed depths Where the rock is shallow this can be done by direct examination of exposures in trial pits or trenches, but when borings have to be sunk to locate and prove bedrock it is important to ensure that boulders or layers of cemented soils are not mistaken for bedrock This necessitates percussion boring or rotary diamond core drilling to a depth of at least 3 m in bed-rock in areas where boulders are known to occur On sites where it is known from geological evidence that boulders are not present a somewhat shallower penetration into rock can be accepted In some areas boulders larger than 3 m have been found, and it is advisable to core the rock to a depth of 6 m for important structures

Mistakes in the location of bedrock in boreholes have in many cases led to costly changes in the design of structures and even to failures.

It is sometimes the practice, when preparing borehole records, to define rockhead or bedrock as the level at which auger or percussion boring in weak rock has ceased and coring in stronger rock has commenced

This practice is quite wrong The decision to change to core drilling may have nothing to do with the strength of the rock It may depend on the availability of a core drill at any given time or on the level at which the borehole has reached at the end of the morning or after- noon's work Rockhead or bedrock should be defined as the interface between superficial deposits and rock, irrespective of the state of weathering of the latter

Direct exposure of rock in trial pits or trenches is preferable to boring, wherever economically possible, since widely spaced core drillings do not always give a true indication of shattering, faulting, or other structural weakness in the rock Where rock lies at some depth below ground level, it can be examined in large-diameter boreholes drilled by equipment described. Because of the cost, this form of deep exploration is employed only for important structures EC 7 recommends that where the possibility of base uplift in excavations is being investigated the pore-water pressures should be recorded over a depth below ground-water level equal to or greater than the excavation depth Even greater depths may be required where the upper soil layers have a low density When boreholes are sunk in water-bearing ground which will be subsequently excavated, it is important to ensure that they are backfilled with concrete or well-rammed puddle clay If this is not done the boreholes may be a source of considerable inflow of water into the excavations

In a report on an investigation for a deep basement structure in the Glasgow area the author gave a warning about the possibility of upheaval of clay at the bottom of the excavation, due to artesian pressure in the under-lying water-bearing rock After completing the basement the contractor was asked whether he had had any trouble with this artesian water The answer was that 'the only trouble we had with water was up through your borehole' In another case, large bored piles with enlarged bases were designed to be founded within an impervious clay layer which was underlain by sand containing water under artesian pressure The risks of somewhat greater settlement due to founding in the compressible clay were accepted to avoid the difficulty of constructing the piles in the underlying, less compressible sand However, considerable difficulty was experienced in excavating the base of one of the piles because of water flowing up from the sand strata through an unsealed exploratory borehole

Site investigations of foundation failures.

From time to time it is necessary to make investigations of failures or defects in existing structures The approach is somewhat different from that of normal site investigation work, and usually takes the form of trial pits dug at various points to expose the soil at foundation level and the foundation structure, together with deep trial pits or borings to investigate the full depth of the soil affected by bearing pressures A careful note is taken of all visible cracking and movements in the superstructure since the pattern of cracking is indicative of the mode of foundation movement, e g by sagging or hogging It is often necessary to make long-continued observations of changes in level and of movement of cracks by means of tell-tales Glass or paper tell-tales stuck on the cracks by cement pats are of little use and are easily lost or damaged The tell-tales should consist of devices specially designed for the purpose or non-corrodible metal plugs cemented into holes drilled in the wall on each side of the crack and so arranged that both vertical and horizontal movements can be measured by micrometer gauges Similarly, points for taking levels should be well secured against removal or displacement They should consist preferably of steel bolts or pins set in the foundations and surrounded by a vertical pipe with a cover at ground level The levels should be referred to a well-established datum point at some distance from the affected structure, ground movements which may have caused foundation failure should not cause similar movement of the levelling datum

The Building Research Establishment in Britain has developed a number of devices such as tiltmeters and borehole extensometers for monitoring the movements of structures and foundations

A careful study should be made of adjacent structures to ascertain whether failure is of general occurrence, as in mining subsidence, or whether it is due to localized conditions

The past history of the site should be investigated with particular reference to the former existence of trees, hedgerows, farm buildings, or waste dumps The proximity of any growing trees should be noted, and information should be sought on the seasonal occurrence of cracking, for example if cracks tend to open or close in winter or summer, or are worse in dry years or wet years Any industrial plant in which forging hammers or presses cause ground vibrations should be noted, and inquiries should be made about any construction operations such as deep trenches, tunnels, blasting, or piling which may have been carried out in the locality