FOUNDATION DESIGN SIMPLY EXPLAINED By JOHN FABER BSc, MICE, MIStructB., MAm.SoaC., M.Soe.C:E,(Fr), MInst.W., A.C.G.L., M.ConaE., Conculting Engineer and FRANK MEAD BSo, AMLCE, AM.LSuuetR,, AMLW. LONDON OXFORD UNIVERSITY PRESS NEW YORK ToRoxTO 1961¥ FOUNDATION DESIGN SIMPLY EXPLAINED It is hoped that elementary principles have been set down clearly. If nothing more, the book should serve to enable architects ‘and engineers, not having a full knowledge of foundation matters, ‘to appreciate the problems likely to arise, and to understand which parts of the work should be entrusted to a specialist, and what information and testing that specialist should reasonably require. ‘The worked examples in Chapters VI, VI, and VIII assume an elementary knowledge of reinforeed concrete design. Students who lack this knowledge are referred to the companion volume, Re- inforced Concrete Simply Hxplained. The cxamples hore are based fon the use of 1:2:4 nominal mix concrete, with mild steel re- inforcemonts. Other conerete mixes and other steels may have advantages in certain cases, though, for foundation design, generally not. ‘The authors acknowledge with thanks permission by Messrs John Wiley & Sons Ino., New York, to reproduce Figs. 16 and 26 from Terzaghi & Peck’s Soil Mechanics in Engineering Practice, ‘They also wish to record their appreciation to Mr Derek Randall, who drew the diagrams, and to Miss Esther Harris who prepared the typeseripts with her usual great enthusiasm, Clappersgate, Jou Fapnr Harpenden, Herts. Frank Mnap Tanuary 1964 rv va var Bo CONTENTS PREFACE yIRST PRINCIPLES ROCKS AND SOILS SIMPLE SITE INVESTIGATIONS ‘THEORY OF SOIL BEHAVIOUR SOIL MECHANICS TESTS SPREAD FOUNDATIONS PILED FOUNDATIONS RBTAINING WALLS MATERIALS AND CONSTRUCTION INDEX 105 m1CHAPTER I FIRST PRINCIPLES § 1. The purpose of a foundation is to enable the load of some particular structure to be applied to the ground in such a manner that the ground is not over-stressed as a result. §2. Tor example, a reinforced-concrete column 12 in. x12 in might be carrying a direct load of 50 tons. If this column were built directly on a thick strata of sound granite, no harm would result: the intensity of prossure applied to the granite would be 50 tons Titxi it and this would be perfectly safe. However, a 50-ton load on the same column built directly on ‘a bed of medium clay would push the column through the clay, and out of sight. This is because the clay, being very much softer than granite, cannot sustain the same intensity of pressure. A more suitable bearing pressure to allow on the clay would be 1 ton/sq. £6, so that a foundation would be required of area 50 tons/sq. ft that is, about 7 {x7 ft. This is indicated at Fig. 1(a). With such a foundation 7 x7 ft, we know that generally speaking each part of the clay under the foundation must. be supporting the underside of the foundation with a pressure inten- sity of 1 ton/sq. ft. (If this were not so, clearly some parts of the clay would be pressing more, and others less, because the total pressure from the ground must balance the total load from the column: and our design assumption is that the clay in question cannot press more than 1 ton/sq. ft.) We see, therefore, that the foundation is being pressed upwards from beneath over an area, very much greater than the area of the column. In consequence, the foundation conerete has certain duties ta perform. TF it is too, thin, it may break either in bending (Fig. 1(b)), or by the column punching its way right through (Big. 1 (c) ‘Thus we see that foundation design is in two parts. First we have to decide on an allowable bearing pressure which the ground can safely sustain. This may vary from about 100 tons/sq. ft for exceptionally sound rocks to 4 fon/sq. ft for silts and very soft2 FOUNDATION DESIGN SIMPLY EXPLAINED yet @ INDEPENDENT SPREAD FOUNDATION. ye" © FAILURE BY BENDING. © Fauve ey SHEAR. Fig. 1. Column spread-foundation. clays. The allowable beating pressure will determine the plan size of our foundation. Then we have to determine the thickness of concrete and amotnt of reinforcement required to ensure that the foundation is strong enough not to fail in bending or by shear. Generally the greatest difficulty in foundation design is to decide ‘on the correct bearing value to allow on the ground. If we are too conservative in our estimate, the cost of the foundations will become excessive. It is important, therefore, to know what FIRST PRINCIPLES a yer ye oo VBeaRING” on SOUND Fig. 2. Choice of foundation level. information should be collected to enable such extravagance to ‘be avoided; and having collected the information, we must know how this is to be interpreted. ‘One purpose of this book is to show how the allowable bearing pressures for various forms of strata can best be determined; and ‘then how to design and calculate suitable foundations so that the ground pressures are limited to the allowable values. § 8. Consider again our 12 in. x12 in, column, loaded to 50 tons, but to be supported on strata as shown in Fig. 2. ‘The upper 15 ft is clay, good for 1 ton/sq. ft, and beneath this is sound granite. Clearly there are two possible foundation arrangements: either we could excavate 15 ft down and sit directly on the granite without4 FOUNDATION DESIGN SIMPLY EXPLAINED any need to spread the load; or alternatively we could provide a foundation 7 ft 7 ft higher up on the clay. The foundation on the clay will require more eonerete and steel reinforcement, but. will entail less excavation work. A number of considerations such as time, convenience, and cost will arise in settling the choice. ‘The example shown in Fig. 2 is fairly clear-cut. Generally the strength of the ground varies less markedly than from 1 ton/sq. ft to 50 tons/sq, ft, and the change may be spread over greater depths than 15 ft. This indicates the need for exploratory boring work before it is possible to decide on the most suitable depth for founding in any given ease. § 4. Sometimes the soil for the uppor part is good for less than 1 ton/sq, ft, and the depth to harder material may be greater than 16 ft, Here it may be very costly to spread the load sufficiently in the upper soil; and equally the eost of excavating to the harder material may be prohibitive. In these circumstances a foundation may be provided more cheaply by driving piles through the soft upper strata so as to reach and derive support from the harder strata below. Oceasion- ally conditions are such that bored piles have advantages over driven piles. All these considerations are discussed more fully in Chapter VIL § 5. Sometimes soft silty ground is found extending to depths of 100-200 ft or more. In such eases even piling may offer no practical solution, the piles pushing into the ground almost as easily at 100 ft depths as at 10 ft depth. In these circumstances a buoyant foundation may be necessary. This, in essence, is like a ship, floating on the water. According to Archimedes’ principle, the buoyancy of a body immersed in a fluid is equal to the weight of the fluid it displaces. If a ship plus its contents weighs 10,000 tons, it must be lowered into the water s0 as to displace 10,000 tons of the water before its downward weight, is just balanced by its upward buoyancy: this is the level at which the ship will float. Similarly with a buoyant foundation. Figure 3 shows a hollow foundation, 15 ft deep (stiffened internally with eross-walls). This is very much lighter than the weight of ground it has been con- structed to replace. Roughly, for every square foot of plan area, the 15 ft of ground removed by excavation weighs about 0-8 ton, whereas the selfweight of the buoyant foundation is equivalent to only about 0-3 ton, so that the net buoyaney effect is 0-8—03=055 ton/sq. ft. ‘This buoyancy effect can be set against a part of the downward FIRST PRINCIPLES 5 Fig. 3. Buoyant foundation. pressure intensity from the superstructure load, so as to relieve the net loading intensity applied to the ground. Suppose the columns are arranged at 16 ft centres in both directions as shown, and each carry a load of 226 tons. ‘Then the gross downward loading intensity is 225 tons 15 X15 From this we can subtract the upward buoyaney effect of =1-0 ton/ag. ft. 0:5 ton/sq, ft. ‘This leaves us a net loading intensity on the ground of only 05 ton/sq. ft. ‘This may well be within the capacity of the silt to sustain,CHAPTER IT ROCKS AND SOILS § 6. Engineers classify the materials which form the outer layers of the world’s surface broadly as rocks and soils. Rocks are hard, brittle masses, whereas soils consist of particles with water or air eantained in the spaces between. Rooks include granite, limestone, sandstone, marble, slate, and others: and it is common knowledge that these materials are so hard, strong, and permanent that they give excellent service as building materials. Soils, on the other hand, include such materials as clay, sand, and gravel. These are not suitable in their natural form as building materials, due to porosity, impermanence, and their inability to stand to any height without support. However, clays may be baked dry to form bricks; and sands and gravels form excellent ingredients for artificial conglomerates (concrete) when suitably cemented together. In terms of agriculture, the best soils are largely composed of organic matter of a peaty vegetable nature: such soils promote the growth of healthy crops. However, these soils are quite unsuitable for building on, because they allow considerable subsidence when. loaded: the organic constituents decompose, and the water is squeezed out. Thus it is important, in engineering, when we talk of soils, to be clear that we mean materials such as clays, sands, and gravels. We are not referring to soils rich in vegetable matter as would delight the farmer or horticulturist. Rooks can stand unsupported to very considerable heights. In nature, the cliffs along the south coast of Sussex are a striking example: and in artificial form we soo railway euttings standing sheer 100 ft or more. But soils cannot stand more steeply than their natural angle of repose,* so that railway cuttings aro sloped back at about 1 in 2, or sometimes more, depending on the nature of the soil. ‘Thie natural anglo of ropovo of a coil depends greatly on its water content. For example, a stiff clay will stand vertically when first exposed but if wetted sufficiently can become so soft as to resemble a paste, or liquid slumy. And though sand will stand well when. *1f tho student will pour dry salt from « container on to table, he wil fad that however eirefully ho docs this, he eannot get the salt to stand more steeply tha certain maximum angle, This is known as the natural angle of repose of the material. ROCKS, AND SOILS 1 appropriately damp, it is limited to a natural angle of repose when dry: and poorly graded sands, when saturated, slump in the fashion of a fluid—quicksand. Our childhood experiments at the seaside will have demonstrated these effects. In broad terms, then, rocks are hard, strong, and generally reliable for building on. But soils are of an un-rigid nature, liable to move or give way when loaded, and are sensitive to external in- fluences. It normally happens, therefore, that good foundations on rock need only be very much smaller than would be required on. soils. However, as a point in favour of soils, rocks are very much. ‘moréexpensive to exeavate, sometimes causing delay and difficulty. ROCKS § 7. Rocks may be divided into three groups, their classification depending on the conditions governing their formation. Thus we have (a) igneous rocks, (b) sedimentary rocks, and (c) metamorphic rocks. Igneous rocks are the direct result of molten minerals from the earth’s interior cooling down at the earth’s crust,* The best-known rocks in this group are granite and basalt. Sedimentary rocks are composed of smallish particles, cemented together, or compacted by the weight of overlying material. In the cases of sandstone and shale, the initial particles are derived from the earlier disintegration and break-down of other rocks. In the cases of limestone and chalk, the particles aro the bony remains of animals and other small creatures. The particles, in each case, are transported by water, or wind, or gravity, and deposited in hori- zontal layers prior to being cemented together or compacted to form rock. The hardness of the rock depends on the cementing material and the degree of compaction, ‘Shale is formed from clays. Under pressure the water is removed, and in the process the material becomes laminated. If lamination does not occur, the rock is known as mudstone. Metamorphic rocks are formed by the action of heat and pres- sure on cither igneous or sedimentary rocks, causing a change in ‘the original rock structure. Slates are derived from shales and possess rather special and charactaristie cleavage (splitting) planes not necessarily related in direction to the bedding planes. Schists have grains running in parallel orientation and are thus very susceptible to shear failure and cleavage along the planes of the grain faces. The most commonly known schist is marble which ‘Phe intarioe of the word is eo hot that everything there is molten. Outside this molten interior ism hard outer shel known a the earths crust.5 FOUNDATION DESIGN SIMPLY EXPLAINED originates from limestone. Gneisses have a structure which alter- nates schistose layers with others in which the erystal orientation is less well defined, giving a banded texture. Additionally the grain sizes are coarser. Gneisses are thus stronger than schists. Fig. 4. Rook bedding planes. BOOK FORMATION § 8. The layers in which rocks are formed aro known as beds, and the plane surfaces which mark interraptions in the action of Fig. 5. Geological fault ROCKS AND SOILS ° deposition are known as bedding planes (see Fig. 4). Bedding planes frequently become parting planes, and are accentuated by the effects of weather. ‘Roughly at right angles to the bedding planes, fractures develop across the beds, due to internal stresses set up during the cooling, ‘or drying-out of the rock, or due to external effects such as pressure or movement. These fractures are called joints. ‘During their history some rock beds have become subject to enormous forces within the carth’s crust which have caused the various strata to become inclined or folded, As a result of this movement and stressing, the rock may have fractured, with one part of the strata sliding over the other. This type of fracture is Known as a fault (see Fig. 5) BEARING CAPACITY OF ROCKS § 9. While the unit strength of the material comprising the rock proper may be very considerable, the effective strength of a rock formation may be greatly influenced by the condition and inelina- ‘tion of the bedding planes and joints, and the position and complexity of any faults. For example, the actual limestone rock indicated in Fig. 6 may well be strong enough to support an applied. pressure of 30 tons/sq. ft, but the safety of the formation here depends not on the strength of the limestone but on the slipper ness of the clay contained in the bedding planes. This clay will have entered the limestone by being washed down from above through a fault or through the joints, and the arrival of further water by the same route will probably lubricate the clay so as to leave little frictional resistance, preventing the upper part of the limestone from sliding down to the right. ‘Thus, while the figures in Table I give safe bearing capacities for rocks evenly bedded and in sound condition, it is clear that it would be dangerous to work to these bearing values without first CLAY IN JOINTS § BEDDING PLANES ‘ig. 6. Slip-planes in limestone.10 FOUNDATION DESIGN SIMPLY EXPLAINED ensuring that the condition of the rock and the bedding planes is satisfactory. Where the strength of the rock itself is in doubt, specimen cores should be drilled out, and tested. ‘Tame 1 Save Branixo Carscres of Rocxs ‘Safe bearing capacity ‘ype of rock tonal. fe Igneous rock and geieooe 0. Limestonos and herd sandstones 20 Schists and slates 20 Shaler and madstones 10 ‘ard block elle 6 Cortain rocks, particularly limestone, chalk and shale, are liable to contain pot-holes, swallow-holes, or other solution channels. ‘These can occur at all depths. St Michael’s Cave in the Rock of Gibraltar is an impressive example: the Cheddar caves are another. Smaller holes 5, 10, or 20 ft across can be sufficient to cause disaster to a local foundation; and it is wise to probe below the bottoms of foundation excavations to seck out such snares that lie in wait for the unwary. Chalk, shales, and mudstones are liable to disintegrate on ex- posure to air and water, and for this reason the bottoms of the excavations in these materials should be covered as quickly as possible with a protective layer of conorete. FORMATION OF SOILS § 10. Soils are formed by the disintegration of rocks, ‘When water, contained in joints or pores in a rock, freezes, it does so with an increase in volume, and this has a wedging action, splitting the rock into smaller pieces. Similarly, rain will wash away loose particles; and dilute acids dissolved in rainwater actually attack certain rocks. Rain also may soften certain strata, producing Iandslides, resulting in shattered rock debris. Rivers and seas erode rooks Tying in their course, and against which they wash, Heat from the sun causes the break-down of even the hardest rocks, because of alternating expansion and contraction; and where the minerals which form the rock have different coefficients of expansion, the rock structure itself breaks down with change of temperature. "This increase in volume is what causes pipes to burst in foaty weather. ROCKS AND SOILS n ESTUARINE Ww BOULDER ed Bfb Roce = Big. 7. Soil beds, Wind-blown and water-borne particles also wear away rocks in much the same way as sandblasting is used to clean metallic surfaces. All these processes lead to the break-down of rocks to smaller particles of soil sizes. These are then earried by winds, rivers, seas, or ice sheets and deposited in beds of varying thicknesses over long periods of time (see Fig. 7); and during this process of transporta- tion the particles are subjected to further abrasion and break-down. SOIL TYPES § LL, Soils are divided into groups according to their particle size, They range from boulders, gravels, sands and silts, down to clays. Often a particular strata is made up of a mixture of these ‘various soil types, so that one meets in practice ‘sandy gravels’, ‘silty clays’, and so on, Soils rely for their strength on two physical properties. One is ‘the property of cohesion the moleoular attraction of emall moist, particles to one another—the sort of gluey effect that causes clay to stick to your boots. The other property is frictional—the rough- ness which prevents adjacent particles from sliding freely past one another—much like the roughness of two pioces of sand-paper being rubbed against one another. For convenience, soils are classified according to whether or notrr FOUNDATION DESIGN SIMPLY EXPLAT [ibe aera vac ala] a Sie _waNacee [TOR pa wT] ET SOT heme nammeqein yyneha rs el E B SHEP eoucoce aS oo Eg Bef scr feacrion [sano Paacrions [eaniec Pascton] "| Fig. 8. Particle-size distribution curves, their strength is due mainly to cohesion. Thus we have cohesive soils and cohesionless soils. Cohesionless soils derive their strength from internal friction. Cohesive soils have particles too small to be seen by the naked eye, and include silts and clays. Cohensionless soils are of larger particle size, and include sands, gravels, and boulders. Figure 8 indicates the proportions of particles of different sizes for three typical soil samples. This manner of representing soil particle sizes is standard, and shows, at a glance, the range and Particle sizes in a given specimen, so that one can tell immediately whether it is predominantly a silt or a sand or a clay, and whether the particle sizes are evenly graded or not. SOIL BEHAVIOUR § 12. Cohesive soils (clays and silts) squeeze under load, and do not recover when the load is removed. This is the effect of part. of the internal water being squeezed out. The strength of cohesive soils depends mainly on the internal effect of molecular attraction, and is not greatly influenced by whether the soil is subjected to external pressure or not. ‘Thus any mniform clay will have a. strength which is likely to remain roughly constant at all depths (except only above the lovel where seasonal effects like summer dryness and winter frost interfere). Cohesionless soils (boulders, gravels, and sands), on the other hand, rely for their strength on the frictional effects of the indivi- dual ‘particles resisting being rubbed past one another. ‘This ROCKS AND SOILS 13 frictional strength depends on how the individual particles vary: in size and shape, how they fit together, and how tightly they are racked. P®When e cohesionless soil has little or no pressure applied to it, its strength is only small.* This is like all other experiences of friction. An encyclopaedia will casily slide across the floor when no pressure is applied from above; but if someone stands his full weight on the book, the friction increases ro rata to the weight of the person, and the book becomes very much more difficult to move, Similarly, with coheoionloos coils, the frictional strength increases roughly pro rata to the pressure applied, and we find that gravel and sand strata generally get stronger and stronger the deoper they are below the ground surface. BEARING CAPAOTTY OF SOILS § 13. The matter of determining specific values for safe bearing capacities for soils is not easy. Tabulated values can often be ‘Tame TE Sam Branina-Carsciris vox Sons Types of cits | Safe tearing capacity Gohesine sot ‘onslog ‘Veny elif and hard shaly clays we Si cays (andy oF not) a ‘Firm clays (candy o: 108) 12 Soft claye and site i Very ac lays silts fit Cahesinlens ail (ry) Wellgreded sands ant sendy gravels (Gompact) “6 Well gro sons sony ral Le oot) ‘Uniform sands (compact) a4 Uniform anes (loos) i misleading, because similar soils vary from one another with no marked definition, and engineers often disagree on terms such as ‘hard’, ‘stiff’, and ‘soft’ as applied to clays, and ‘compact’, ‘loose’, ‘nd ‘uniform’ aa applied to ands. This ia all disousood ot groator Jength in Chapters IV and V. Nevertheless, Table IT is included here as giving the student a rough guide. “Thin is why it is so dificult to walk over Toose pebbles or very dry sand at the seaside: the sand fs too weak to take the backward thrust from your feet, whieh, in Consequence, sip backwards ‘Damp sand ie not the tame. This it Because the moisture exeates internal attractive forces (eurTuce tensions) which give the sand sotne cohesive strengthu FOUNDATION DESIGN SIMPLY EXPLAINED When cohesionless soils are submerged, the particles receive a hydrostatic buoyancy (like the buoyancy effect on a skin-diver) and this reduces the internal friction strength, roughly halving the bearing capacity of the soil. Therefore with foundations which extend down into, or near, ground-water level, the safe bearing capacity should be taken as half the values given in Table II. The science known as soil mechanics is intended to assist in determining truer values for safe bearing capacities for soils in specific cases; Chapters TV and V are aimed at explaining how soil mechanics can hest be used to this end. This science, nsed wisely, can give a better indication of soil strengths than the arbitrary reading of a table such as Table IT or the casual prodding of @ foundation bottom with the end of an umbrella, But let the student be clear that soil mechanies is not a cut-and- dried exact science. Tt can never tell you the exact bearing capacity of a real soil, because all real soils vary widely in their physical properties at every foot of depth and at each foot of length and breadth across the extent of any building site. Soil mechanics can only tell you about the strength of a very small sample taken at such-and-such depth in one of a few boreholes, and this sample may either be stronger or weaker than the ground only a few feet away, whether deoper, or to one side or another. No use, then, to bow down and worship soil mechanics. No use cither to average the test results, which would virtmally be as senseless as ignoring the poorer results—the weak links in the chain, Wo shalll still need intelligent judgement in knowing what to do with the findings of our tests, because soils in nature are 80 variable, so persistently variablo, that no one true answer for an allowable bearing capacity could ever exist; and indeed, if any soil study pretended that the matter was simple and free from confusion, it would clearly be so insensitive a science as to be untrustworthy. In Chapters IV and V we shall see what soil mechanics has to offer—how we take samples from the ground, how we test them in the laboratory, and what the results mean when we get them. Then we can work through some typical examples of foundation design, and get the feel of the limits of accuracy necessary to enable us to choose one form of fonndation rather than another. We shall find that often we have little need to work closely to the theoretical values for allowable bearing capacities, and that the conception of the foundation anatomy affects the economy of the matter more than any precise mathematical calculation, | | CHAPTER IIT SIMPLE SITE INVESTIGATIONS § 14. Suppose we have to design the foundations for a building on asite where we know the soil is ‘uniform’ in character, and extends without change for a considerable depth. If we have had experi- ence of this sort of soil hefore, we may he able to judge from its feel and appearance how many tons per square foot it will safely carry. For example, you can push the blunt end of a peneil quite easily into a softish’ clay which will bear about 1 ton/sq. ft; it, requires a good firm pressure to get the peneil into 2-ton cla and you will have great difficulty in getting the pencil into a 4-ton clay at all. But these rule-of-thumb methods can mean very little to people without @ good deal of experience behind them. It is only when you have dealt with a number of clays you really know that you can have any confidence in judging the strength of others by comparison. Even then, it is éasy to be misled, and such methods are, at the best, only very approximate. LOADING TESTS §15. A much better method is to apply an actual test-load on a known area of the ground. If we expect our foundations are going to be, say, 4 ft below ground-level, this will be the level for making our test. We dig a hole 4 ft deep, trim the bottom to a level even surface, and place on it a conerete or steel pad of some definite size. 2ft x2 ft is often suitable for tests on olay. Langer pads require such large test-loads as to be rather inconvenient, and smaller pads are too susceptible to local variations or weaknesses in the ground and, apart from giving misleading results, are inelined to tilt and behave awkwardly. On ‘uniform’ sandy soils a test pad 1 ft 1 ft may be sufficient. § 16. Let us consider sn setual test carried out by the authors ona firm clay near Liverpool. The concrete pad was 2 ft x2 ft, cast directly on the clay, and the load was applied using a hydraulic jack pressing against the underside of steel beams supporting about 16 tons of steel billets (seo Fig. 9). ‘First the jack was pumped to 2 tons, and maintained at this load for 24 hours. The amount the concrete pad settled was recorded,6 FOUNDATION DESIGN SIMPLY EXPLAINED STEEL eiLLers oe CONCRETE Fig. 9. Ground loading-test. first just after the 2-ton load was applied, and again at the end of the 24 hours. In this case a micrometer gauge was used for measuring the settlements, though some engineers are content: with the accuracy obtained by using a surveyor’s level. ‘The 2 tons sproad over the area 2 ftx2 ft gave an applied pressure on the clay of 2 tons DRx2 tt and at this pressure the immediate settlement was 0-026 in; the settlement after 24 hours was 0-031 in.—not a very material increase. Next, the jack was pumped to 4 tons, giving an applied pressure of i lons_=1 ton/sq. ft, BHex2tt ania ‘and once more the settlement was measured immediately, and again after a further 24 hours. And so the process was repeated, stage by stage, until the total load applied was 16 tons—equivalent to a pressure of 4 tons/sq, ft. ‘The total settlement was then about 1-24 in,, and the test was discontinued. ‘The results of the test are indieated below: =} ton/sq, ft, “Appl Seonet i } —_ | Temas | fer howe tons fons. fe in in. : 7 ocio ooa 4 1 5-061 0.082, 6 L oul7 oa, 8 2 bars 0-215 10 2 058 0330 2B 3 bats ose it 3 O32 0-388 6 a ost 1387 SIMPLE SITE INVESTIGATIONS n Loa (Tens/sa. Fr) Mey tiyptey 2 pds ah UNE o3 on os Cincnes) oc SETTLEMENT 9 T z a 5 zg 7 o 1 2 8 4 5 6 7 8 ime (pays) Fig. 10. Settlement curves of loading-test. The same results are also indicated graphically in Fig. 10. The dotted curve is made up of a series of kinks, each representing the sudden increase in settlement that occurred immediately after cach load increment, followed by a period of relative lull until the time of the next load inerement. ‘The full-line eurve in Fig. 10 is18 POUNDATION DESIGN SIMPLY BXPLAINED drawn through the points of maximum setilement oceurring after the 24-hour rest-periods, and this curve represents the relationship between load and final settlement for our 2 ft square test pad.* ‘Load-settlement graphs of this type vary in form according to ‘the nature of the soil. Dense or stiff soils show a sudden increase of settlement once the ultimate bearing capacity has been exceeded. ‘Loose or softer soils, on the other hand, display no sudden failure, and the curve gradually bends over little by little. In such cases the failure load is taken as the point where the slope of the curve hecomes more pronounced. In the case of the graph shown in Fig. 10, the ultimate bearing capacity would be regarded as about 24 tons/sq. ft and the allow- able bearing pressure would he about 1} tons/sq. ft. LIMITATION OF LOADING-TRSTS §17. Now loading-tests, as described above, are satisfactory when we know the soil is uniform in character, and extends with- out change over an adequate depth. We can determine whether this is so by the simple methods described in §§ 18-21. q UU “@ test Pao. @) FOUNDATION. Pig. IL. Limitation of loading-test. Where the ground varies at greater depths, the danger of relying only on the results of loading-tests is illustrated m Pig. 11. Clearly, within the soil the influence of loading a small test pad will be relatively local, whereas the effect of loading a larger structure will extend much further. Thus a loading-test may indicate only the * tia not atroty tre that the settlement after a 24-hour Tl would be fina, though for our purposo, and with e test as clomentary as we are describing here, the settlement ‘can be Fogarded ae nese enongh fina, SIMPLE SITE INVESTIGATIONS 10 strength of an upper erust or hard layer, whereas a heavy structure ‘will need to rely on the strengths of the soil layers at far greater depths. Normally it is necessary to check the adequacy of the soil under foundations to a depth of at least one and a half times the width of the loaded area. Thus for a pad foundation 10 fé square, the soil should be checked to a depth of at least 16 ft below the underside of the foundation. For larger structures where the influence of adjacent founda- tions overlap, ar where a. raft foundation is used, it will be neces. sary to explore far deeper. For example, a heavy industrial structure 40 ft wide would require soil examination to a depth of about 60 ft unless rock were struck at some lesser depth, or local Imowledge or records indicated improved strata at increasing depths, ‘In § 35 it is shown how the applied soil stresses diminish as the loads disperse lower down, so that weaker soils may be permissible at greater depths—provided wo know of their existence and satisfy ourselves that they are adequate for the reduced applied stresses to which they will be subject. TRIAL PITS § 18. The most satisfactory way of seeing how the ground varies at different dopths is to dig a number of trial pits. By lining the pits with open timbering, the soil ean be seen in its vingin state and a good general impression of the extent of variation is obtained. Samples can be taken as required and kept for qualitative exami- nation, or for testing as described below, and in Chapter V. Tt generally becomes expensive to dig trial pits more than about 15 ft deop, and below this it is more usual to make borings. SHALLOW BOREHOLES § 19. Tn unconsolidated soils, shallow borings can be made with hand-augers. These are worked either from ground-level, or some- times from the bottoms of trial pits, Trial pits need to be a little larger than would otherwise be required if borings are to be made from the bottom. It is not normal to line the holes of auger borings with steel casings. Such boreholes are usually limited to about 20 ft penetration. For investigations to greater depths, and wherever waterlogged soils are encountered, the boreholes are eased with sectional steel tubing. Boring is then carried out by one of two methods as20 FOUNDATION DESIGN SIMPLY EXPLAINED described in §§ 20 and 21. Generally these boreholes are made 6 ih. diameter. N3 ‘SHELL AND AUGER’ noRING § 20. Boreholes in this country are most frequently made by what is known as percussion boring. This omnibus term covers the use of percussion chisels, clay cutters, augers, and shell ‘balers’ ‘These tool-attachments are all operated from ground-level, using a simple four-legged derrick, fitted with a winch—the whole assembly being known collectively as the ‘shell-and-auger gear’ The tooi-attachments are illustrated in Fig. 12. Percussion chisels, suitably weighted, are repeatedly dropped in the borehole so as to break up the soil. ‘The debris is then lifted from the bottom of the hole using the shell baler: this latter is aucen io corres cise, Baten Fig. 12. Shell-and-auger boring gear, SDIPLE SITE INVESTIGATIONS a fitted with a simple flap valve (much like a foree-pump), and, by repeatedly dropping the shell, the debris passes up through the valve until the chamber is full. The shell is then lifted out of the hole, and the debris recovered at ground-level. Percussion chisels are used mainly for boring in compact sands, gravels, and chalk. The shell (without the chisel) operates satis- factorily in wet sands or blowing soils. For boring in dry sands or silt, and in some clays, the auger tool is used. This is rotated in the borehole on a steel rod, and acts very much like the bit on a carpenter's brace-and-bit, the spoil being collected internally within the tool, instead of falling to waste externally. ‘The clay cutter is used for boring firm clays. The tooth-edged open-ended cylinder is dropped into the olay, and when this is pulled out again, the clay is retained in the cutter. WASH-BORING § 21. The second method of boring is known as wash-boring. A pipe, fitted with a chiselled end-piece, is lowered within the borehole (sec Fig, 13). The chisel is used to loosen the soil, and a high-speed jet of water foreed down the pipe lifts the soil in liquid-suspension to ground-level. Clearly this mothod of boring must greatly disturb the condition — [Syren “Fo ORIN CHiseL END ¢ Fig. 13. Wash-boring gear.2 FOUNDATION DBSIGN SIMPLY EXPLAINED of the soil. Nevertheless the size, shape, and grading of silt or sand particles will be apparent, though all evidence of their natural compaction will be lost. But no other satisfactory method of sampling coarse silts and sands, without disturbing their natural degree of compaction, has yot been developed, and wash-boring is certainly a very quick and simple process. Probably the greatest benefit of wash-boring is derived when the method is used in conjunction with pereussion boring on the same site, The wash-boring gives quick results in checking whether strata Tneated in a few percussion bores extonds consistently over the whole of the site. § 22. Tho simple soil investigations described in §§ 18-21 are qualitative only. The samples obtained are disturbed, and therefore unsuitable as a basis for carrying out any scientific strength-tests in a laboratory. Nevertheless they do give a general indication of the nature of the underlying strata at various depths. For example, the presence of a layer of compressible peat, 20 ft down, would guide an engineer immediately as to the form of foundation he should adopt (and certainly what form of founda- tion he should avoid!). And the knowledge that a soft clay lay underneath what had looked like a good reliable sand would give an immediate warning. And so on. SETTLEMENT OF FOUNDATIONS § 28. Before we pass on to the more scientific approaches described in Chapters IV and V, let us consider for a moment the question of the settlement of foundations. Accurate settlement calculations are difficult to make, and are often uncertain as to their validity: in any case they fall outside the scope of a simple book. But if we revert to our loading-test described in §16 we can get a very good lead as to the likely settlement of a full-size foundation on fairly uniform strata —probably sufficient for many everyday problems. On cohesive soils, the settlement of rectangular foundations is roughly proportional to the breadth of the foundation. Thus in the case of our test already described, if we decide to work to a bearing pressure on the clay of 1} tons/sq. ft, we know that: the settlement of our pad 2 ft square at that preasure is 0-14 in,; and therefore, if our actual foundation works out to be 6 ft square, we can anticipate =0-42 in, a settlement of about 0-14 in. x3 With cohesionless soils (sandy soils) the relationship between settlement and foundation breadth is not as simple. A useful SIMPLE SITE INVESTIGATIONS 2a formula in these ciroumstances, due to Terzaghi & Peck, and intended for foundations built on homogeneous sand, is onlay 0 where A\ is the settlement of a foundation of breadth B, and A is the settlement of a test pad 1-0 ft square under the same intensity of loading. For example, suppose a test pad 1-0 ft square had settled 0°10 in, under the sume pressure as decided on for the design of the full-size foundations. ‘Then the settlement of a foundation 6 ft square on the same sandy soil would be fl 10( 28 2 § 24. Some people decry the very idea of loading-tests on soils, and will argue that a test on a small pad is so limited in application as to be useless or even misleading. These scientifically minded ‘people prefer, curiously, to rely on the results of tests on very much ‘smaller samples (generally 1} in. diameter and 3 in, long) as against the volume of soil influenced hy a test load, say 2 ft across and 3 or 4 ft deep. When we remember that the loading-tesb deals with the soil ‘in situ, with the moisture content and adjoining soil reactions exactly as they would be for the full-size foundation, it is difficult, to share this greater enthusiasm for tests on small samples (neces- sarily disturbed, though reputedly not) which may with luck be typical of the strata they represent, but more probably, by the laws of chance, are either better or worse. Certainly for simple site investigations there is a Jot to be said in favour of the loading-test. It is a relatively cheap operation to conduct; and for works of not too ambitious a scale, where the ground is reasonably uniform, it still takes a lot of beating. If the limitations of the method are properly appreciated, there may be less likelihood of error than with the more scientific methods described in the chapters that follow.CHAPTER IV THEORY OF SOIL BEHAVIOUR § 25. It has been explained in § 11 that soils rely for their strength ‘on two physical properties: cohesion and friction. ‘True clays are cohesive—sticky, while not gritty; and dry sands aro frictional — the particles grating against one another. but having no sticki- ness, It is possible to express these properties of cohesion and friction in the form of a mathematical equation, and there are two good reasons for doing this. Firstly, it enables us to describe the total strength of any given soil in specific terms of its cohesive and frictional properties. It would be clumsy if we had to describe a, silty clay having both cohesive and frictional properties as being s0 much per cent. eohesive plus so much per cent. frictional; indeed the frictional strength of a soil varies acoording to the pressure at the depth considered (see § 12), so that the con- tribution of the friction to the total strength of the soil varies according to the loading and to the depth. These matters are all conveniently taken care of by the use of our one mathematical equation, ‘The second advantage of expressing the properties of our soil by ‘@ mathematical equation is as follows. The tests we are able to make in the laboratory give us results which require a. certain amount of manipulation before we ean separate (for caleulation purposes) the shate of the strength which is due to the cohesion, from the share which is duo to friction. Once we have achieved in ‘our minds this idea of separating artificially the cohesion from the friction, it becomes quite simple to calculate the strengths of soils in all normal situations. Of course, one cannot in practice take a silty clay and separate the cohesion from the friction in the same way as one can take a nut and separate the shell from the kernel; but mathematically we can separate the two, and this is very convenient. § 26. Figure 14(a) shows the simplest way in which a foundation can fail. The diagram does not show the whole of the truth, but it shows a very important part of it. ‘The foundation pushes beneath itself a triangular-shaped wedge of soil, in much the same fashion as a simple snow-plough will push a triangle of snow in front of itself, Thus the soil particles within the triangle move down relative to the soil particles immediately outside the boundaries of the THEORY OF SOIL BEHAVIOUR 26 a ay? r7Ns > e ig. 14, Simple failure of soil under spread-fowndation, triangle. The only reactions which prevent the foundation and the triangle of sol from sinking further are the two forces @. Suppose we replace the boundary plane AB of the moving ‘wedge with an imaginary pieco of plywood. It is clear that part of the effect of g will be to squeeze or compress the plywood as at Fig. 14(c); and the other effect of g will be to try and slide the plys of the wood past one another as indionted wt Fig, 14(d), If we now remove our imaginary piece of plywood, it is clear that the same forces must be acting as between the soil particles which lie on cither side of the boundary plane AB. ‘The sliding effect indicated at Fig. 14(d) is known as shearing. (It is the same action as we take advantage of in the garden when ‘we cut the grass with ‘shears’, one blade of the shears sliding past2 FOUNDATION DESIGN SIMPLY EXPLAINED the other and shearing the grass.) The natural strength of a soil which acts to avoid failure of this kind is known as the shear strength of the soil. 'The shearing resistance of the soil is indicated at Fig. 14(e) by the arrows marked s, The effect of squeezing or compression shown at Fig. 14(c) is also indicated at Fig. 14(e) by the arrows marked p. This is the total normal pressure across the shear plane, When the soil fails under real foundations, the mode of soil failure is more complicated than indicated. by the simple triangle at Fig, 14(q). This is discussed moro in §§ 29 and 30; nevertheless the principle established that the failure of the ground arises from a shear failure of the soil is entirely correct. Satisfactory founda- tions have to be designed so that the shear stresses in the soil are kept everywhere within limits that the soil ean safely sustain. covLoms's EQUATION § 27. Now the shear strength of a soil is made up partly of cohesion and partly of friction. Let us denote these by symbols as follows: the shearing resistance of the soil, c=the cohesion of the soil, the angle of internal friction of the soil, the total normal pressure across the shear plane. Then we can express the shearing resistance of the soil by the mathematical equation s=e-p.tan p. -.. @ ‘This equation was first put forward by Coulomb, and is perfectly simple in its meaning, and perfectly simple in its method of use. First let us understand its meaning. Going back to Fig. 14, if our foundation had been sitting on a plastic clay, the whole of the support from the soil would have been due to cohesion, because such clays have no frictional strength. ‘Then the total shear strength of the soil would have been equal to the cohesion; so that for a cohesive soil (3) In ulher words, the plan @ of equation (2), which relates to friction, is equal to zero, Suppose, on the other hand, our foundation had beon sitting on dry sand. ‘Then there would have been no cohesive strength, 80 that in equation (2) we have c equal to zero; and for a frictional soil -p.tan 9. rr) THEORY OF SOIL BEHAVIOUR ” Now ¢ is the angle of internal friction of the soil, and for a dry sand this has the same numerical value as its natural angle of repose (see als6 footnote at page 6). And p is the pressure across the plano of sliding; compare the analogy at § 12 of the encyclo- paedia sliding across the floor. Thus the greater the friction (p), ‘the greater the shear strength, so that sharp angular particles give a stronger soil than a heap of ball-besrings which would tend to slump or collapse; and the tighter the particles are wedged together by the pressure normal to the shear plane, the more difficult it is for the particles to slide past one another, and so, again, the stronger will be the soil. Many soils are neither solely cohesive nor solely frictional. They rely for their strength on a mixture of both. We have already referred to silty clay as an example. Here the strength of the soil is due partly to cohesion and partly to friction, and the total shearing resistance of such a soil is given by the whole of equation (2), namely s=c-+p.tan @. ‘This general equation may be used as follows. Suppose the cohesive strength of a soil is 15 Ibs/sq. in., the angle of internal friction is 25 degrees, and the total normal pressure across the shear plane is 20 Ihs/sq. in; ie, 16 Ibs/sq. in.; p= 25°; p=20 Ibs/ sq. in, Then 3=15+ (20 xtan 25°), ‘We find tan 25° by looking up in a book of mathematical tables ‘the natural tangent of 25 degrees, and find tan 25° equals 0-466. 15-+(20 x 0-466) 15-4.9°32 24-32 Ibs/sq. in, ‘This value of 24-32 Ibs/sq. in. is the resistance of the soil to fail in the manner indicated in Fig. 14 with the triangular wedge of soil shearing its way down past the supporting soil. If we know how many Ibs load from our structure we have to support, and how many square inches there are at the boundary plano of owr triangular wedge, then we can work out whether the applied shear stress from the stricture is greater than the shear strength of our soil. Tf, in our particular case, the shear stress resulting from the applied load is greater than 24:32 Ibs/sq. in., the soil will fail in shear and the foundation will settle: on the other hand, if tho shear stress is less than 24-32 Ibs/sq. in., the soil will not fail, In practice we allow some factor to give ourselves a margin of ‘Therefore28 FOUNDATION DESIGN SIMPLY EXPLAINED safety to cover many uncertainties, such as the variability of the soil, the inexactness of our calculations for the loading, and so on. We also work the problem in rather a different manner, as will be seen later in this chapter. But the elementary principles given above are fundamental, and if the student has understood the idea we are after by expressing the strength of the soil by Coulomb's equation, then the basis of soil mechanics will have been grasped. In the foregoing numerical example we have tacitly assumed that values of ¢, p, and p were available to us. Indeed, such values can be determined by scientific tests which will he deseribed in Chapter V. However, before we learn how these values are obtained it is nocessary to know more of how soils actually fail under foundations (more than Fig. 14) and what factors we should allow to give us reasonable margins of safety. The remainder of the present chapter is devoted to these matters. ‘We shall come back to Coulomb’s equation again at the begin- ning of Chapter V, where we discuss the measurements of ¢ and g: also at Chapter VITT where we deal with the pressures occurring at the faces of retaining walls. § 28. In practical design work it is convenient to establish what unit intensity of pressure we should allow under our foundations. Engineers designing full-size foundations normally work to units of tons per square foot, as already indicated at ‘Tables I and II. On the other hand, scientists in the laboratory more usually work to Ibs per square inch. It makes no difference which units we work to so long as we describe them carefully and do not get them muddled up. It is convenient to remember that 2,240 Ibs 14 sq. in. ‘The symbols normally used to denote pressure intensities are as follows: q—unit intensity of pressure under foundation, qa=safo bearing capacity of the soil, q=ultimate bearing capacity of the soil. 1 ton/sq. f= 55 Ibs/sq. in, Tho first step is to find the ultimate bearing capacity of a soil Later we can consider what would be a safe bearing capacity for that soil COURSIVE SOILS § 29. When a foundation pushes down a triangular wedge of soil as shown in Fig. 14, the boundary planes AB and OB get pushed away, and rotate about contres very approximately at A and C, THEORY OF SOL BEHAVIOUR 2» Fig. 15, Ultimate failure of soit under spread-foundation. with the effect that the ground to the sides of the foundation is pushed sideways and upwards as indicated in Fig. 15. This matter has been studied carefully by a number of investigators who have calculated the minimum shear resistance along the curved lines in Fig. 15 in'relation to the unit intensity of pressure under the foundation and the weight of soil lifted at the sides. In practice the matter becomes a. little complicated by the roughness of the underside of the foundation, For foundations at ground-level on plastio clays (Le. non- frictional soils, where p=0), it has been shown that approximately = (6) In other words, if we are told by the laboratory that a clay soil has a cohesive strength of 15 Ibsjsq. in. (i.e. e=15 Ibs/sq. in), then ‘we know that the ultimate bearing eapacity of that clay is by 15 u-8x 5-8 tona/sq, ft. To this it would. be necessary to apply a suitable load factor to convert from ultimate bearing capacity to safe bearing capacity, Where foundations are constructed below ground-level—and they usually are—an allowance may be made to the ultimate bearing capacity of the soil for the effect of overburden. This allowance is made by adding to the value given at equation (5) an amount equal to the depth of the foundation multiplied by the bulk-density of the soil. Thus if =the depth of the foundation below surface level, and y=the bulk-density of the soil, ‘equation (5) becomes a= boty 2 @) Suppose then in our previous example the foundation were to be30 FOUNDATION DESIGN SIMPLY EXPLAINED constructed 6 ft below surface level, and the bulk-density of the soil were 110 Ibsjeu. ft, then 6x15, 0xs 15-5 "2240 5-80-25 =6-05 tons/sq. ft." COHESIVE AND FRICTIONAL SOILS § 30. With soils which combine cohesive and frictional qualities, the matter becomes more complicated. Nevertheless Terzaghi 4 Peck Inave found a mathematieal solution for the special ease of a strip footing and modified this very slightly in the light of experimental results to suit square foundations of side B, on dense or stiff soil, as follows: GYHVBCN 4 yeNg + 0-4yBN,, SeeeeGEE (7) Ne Ng and N, are dimensionless quantities known as bearing capacity coefficients, and depend only on the value of g. The values of these coefficients have been caleulated for various p values and are given in Fig. 16. ws ; st Psckitebhs pts | Ashe tT PRES L S01 a we 2~| : a3 ; 3 $ sot wot $ ° ° VALUES OF Ne G Ng ED VALLES OF Ny 16. Tereaghi & Peck's bearing-capacity coefficients for all soil types. ‘Thus the ultimate bearing capacity of a soil having c=6 Ibs/ sq. in. and p=15°, for a foundation 8 £% square at a dopth of 5 fb would be Fi 63411404 -8 tons/sq. fb. * The increase of 025 ton [eq ft may nobook very much, Bu Jer in $38 we shall ‘see thatthe 68 tons/oq, ft hus to be reduood hy a load fastor of 3, whesoas the O28 fonjeq. f dows not. | | | THEORY OF SOIL BEHAVIOUR at Equation (7) applies only for foundations on stiff or dense soils. Where soils are soft or loose, appreciable settlement of the founda- tion is likely to ocour before general shear failure takes place, and ‘under these circumstances ‘Terzaghi & Peck have put forward the alternative equation for square foundations: Y= 0-8TN + y2N y+ O4yBN, tH) with modified bearing capacity coefficients as shown dotted in Fig. 16. FRICTIONAL SOILS § 31. §§ 29 and 80 give respectively direct methods of calculating the ultimate bearing capacities of truly cohesive soils, and soils which combine the properties of cohesion and friction. ‘Theso methods rely on our first knowing the values of and 9 for the soils concerned: these are determined in the laboratory by tests on undisturbed samples. The undisturbed samples of both these types, of soil can be taken from boreholes in the manner described later in § 37. ‘However, it is impracticable to take undisturbed samples of traly frictional materials (sands and gravels) because they tend to spill, and the very act of sampling such materials causes distur- ‘baneo to their natural degree of compaction. Therefore no method of calculating the strengths of cohesionless materials is given here, 1s in practice it is more satisfactory to rely on penetration tests carried out on such soils in situ in the boreholes. ‘This is described fally at § 45. CARTWHEEL FAILURE § 82. The form of soil failure described at § 29, and which applies also at § 30, is cortainly the most common and would normally be mot, for example, at individual column foundations for a building. However, cases have occurred where large structures supported at quite modest bearing pressures have sought out a large plane of weakness extending to a considerable depth so as to produce a cartwheel failure as indicated in Fig. 17. Here it is not the founda- tions of individual columns that have failed, but the larger mass of ground supporting the building as a whole. The instability has arisen from the weight of the structure on ono sido of some centro of rotation being greater than the weight of ground on the other side of the samo centre acting in combination with the shear resistance of the ground against rotational failure along the are indicated.32 FOUNDATION DESIGN SIMPLY EXPLAINED ea z Fig. 17. Cartwheel failure of soil under large structure. Analysis of this type of failure was first suggested by Fellenius. We use the following symbols: the ultimate load from the struoture, the eccentricity of @ about the contre of rotation, the weight of the soil mass within the are of rotation, the eccentricity of W about the centre, ‘the radius of the aro, the length of the are, =the shear strength of the soil. ‘Then by taking moments about the centre of rotation, we have Q.c=W.as.L.R Ww oteLR, Oy ‘The analysis is made by trying various centres of rotation until a contre is found which gives the least value for Q. A check against cartwheel failure should be made in eases where the load applied to the soil is considerable and extensive, and whore the shear strength of the soil is known to be poor and largely non-frictional. For example, large grain-silos have been known to fail in this way. Whore the soil strength varies at different depths, the different shear strengths along the are of rotation may Le laken int account quite simply by this method whenee @ ‘ULTIMATE’ AND ‘SAP’ BEARING CAPACITIES § 38. Having now determined by the methods given in §§ 29, 30, and 32 what is the theoretical ultimate bearing capacity of the soil THEORY OF SOIL BEHAVIOUR 38 on which we propose to construct our foundation, we have to know what would be the safe bearing capacity of that soil, bearing in mind that our laboratory tests which gave us our values of ¢ and are likely to be in error to some extent, the samples on which ‘these tests are made are unlikely to be truly representative of the whole of the soil at our site, and different rates of sottlement of different parts of our structure may lead to parts of the founda- tions being loaded more heavily than we had allowed for when making our calculations. The safe bearing capacity of a soil is determined by applying a suitable load factor to the ultimate bearing eapacity. Thus if we denote the safe bearing capacity as ge, we have =¥ w=] where F is the load factor. Where the expression for ultimate bearing capacity takes into account the depth the foundation is below surface level, the load factor is of course not applied to this part of the equation, ‘Thus for clay soils we have, from equation (6), ao= Etre. EEE CIO) And for soils which combine cohesive and frictional qualities we have, from equation (7), a Beet ye(We + OG BN tr... (1) ‘The load factor against ultimate shear failure is normally taken as 3 § 34. Safe bearing capacity moans just exactly what it says, and nothing more. It means that foundations designed to bear on the ground at this pressure are safe, and the ground will not collapse due to the development of shear failure. But sometimes it is not sufficient to know that the foundation is safe. In cohesive soils a perfectly safe foundation may sink or settle over a period of years more than we can accept. Some build- ings have large sliding doors which would become jammed if the roof settled towards the floor more than a certain amount; certain buildings which are constructed with more floors at one end than ‘the other would be quite unsatisfactory if the tall part settled more than the lower part. 'The student should be able to think of other cases where excessive settlement could become an em- barrassment. ‘The full matter of settlement calculation is complicated andou FOUNDATION DESIGN SIMPLY RXPLAINED comes outside the soope of this book. Nevertheless approximate methods of determining sottlements have already been given in § 23. ‘The student is warned that where foundations are built on cohesive soils, it may be necessary to work to lower bearing pressures than the saje bearing capacities. Such reduced bearing pressures would then be known as allowable bearing pressures. DISTRIBUTION OF STRESS WITH DEPTH § 36, In this chapter we have seen so far how the shear strength of a soil is dependent on cohesion and friction. The cohesion is given directly by ¢, and the friction is determined from g. We shall seo in Chapter V how we get ¢ and g for various soils. Also in the Present chapter we have seen at §§ 29 to 33 how the safe bearing capacity for a soil under a foundation is determined from ¢ and p. ‘We are then well on our way to being able to design some real foundations. But before going further with this it will be well to think back to § 17 and Fig. 11, where we referred to the influence a loaded foundation has on the ground at various depths below. It is one thing to know that the soil immediately beneath a founda- tion is strong enough, but it is necessary to check that the ground at all layers further down is also strong enough for the duty that ‘will be required of it. It is necessary, therefore, to understand how the applied pressure intensity under a foundation diminishes at increased depths. Then wo can decide how deep to go with our exploratory boreholes, and what minimum strength of soil we shall, require lower down. An approximate estimate of how the stress distributes under a foundation can be made on the basis of simple dispersion, assuming a constant angle of say 30° as shown in Fig. 18. The vertical stress Fig. 18. Stress distribution by simple dispersion. THEORY OF SOIL BEHAVIOUR 3% ‘p-on a horizontal plane at depth z below a strip foundation is then given by or Pg Te (a2) where q=the applied pressure at foundation level, and B=the breadth of the foundation. A more exact method is the authors’ adaptation of a graphical method first proposed by Housel (see Fig. 19). From the two edges of the strip foundation, lines 4B, AC, and AD are drawn down to any horizon BCD where we wish to know how the stress intensities vary. Lines AB are vertical, AC slope at 2 to 1, and AD slope at 1 to 1, as indicated. 'The maximum stress at the depth 2 is then determined on the basis that the total applied load from the foundation be divided by the width CC. This maximum stress ‘therefore equals B Bye and is assumed to extend over the width BB: thus the points B’ are marked above B to some suitable scale. Points DB" are then joined, and the complete line of stross-distribution is given by the line DB'B'D. This graphical method of determining stresses at any depths is generally sufficiently accurate for most practical purposes. Now both the methods illustrated above at Figs. 18 and 19 relate to strip foundations, where the dispersion ean occur only in two directions. With square foundations the dispersion occurs in four directions (ie. it spreads out from each of the four sides of the foundation), s0 that the stresses diminish more rapidly for a given increase of depth. Pe (13) 4 WALL is Fig. 19. Graphical method of determining stress dispersion,36 FOUNDATION DESIGN SIMPLY EXPLAINED ‘This is indicated very clearly at the graphs in Fig. 20, where Dulbs of constant vertical soil stress have been plotted. Each bulb of constant stress has been drawn by joining up a number of points where the stress has the same value—be it 0-20g, 0-40q, or what- ever other proportion of g may be chosen. Thus, in the same way that each contour-line on a map is everywhere at some constant height above sea-level, so, in Fig. 20, each line of constant stress indicates everywhere positions whero the vertical stresses in the soil are constant. 4 4 caeceweall | I psererca — aererer ||| prererer— | EN (@) savas FouNpation. Fig. 20. Bulbs of constant vertical soil-stress. (@) STRIP FOUNDATION. Very often in foundation design we limit, the extent of our investigation of the soil to such a depth that the stresses from the foundation have diminished to about a fifth of the values imme- diately under the foundation itself, ie. to the level of the 0-20q stress-bulb. At Fig. 20(a), for a square foundation, this is seen to be about one and a half times the breadth of the foundation B. For a strip foundation, Fig. 20(b) shows that tho same stress distxibu- tion is not achieved until roughly double this depth, i.e. when the depth is about three times the foundation-breadth B. For significant structures whore the loads from the individual columns will spread out so as to link up with one another and form all-embracing bulbs of stress under the structure as a whole, it is usual to carry the site investigations to a depth equal to about, twvice the plan dimension of the foundations, and to a width of about one and a half times. The reasons for this are clearly seen by reference to Fig. 20. CHAPTER V SOIL MECHANICS TESTS §86. In tho previous chapter we have shown that the bearing capacity of a soil depends on its shear strength, which in tum depends on the summation of two internal properties, namely cohesion and friction, We have seen that the relationship between the shear strength and the cohesion and friction ean be expressed in a simple mathematical equation ‘ ctp.tan p. If now we can measure values of ¢ and » for the particular soil on which we wish to construct a foundation, we can then determine the bearing capacity of the soil, using such formulae as we have given in the previous chapter at equations (5) and (7); and the object of the present chapter will be to explain how these values. ofc and g are measured. UNDISTURBED SAMPLDS § 87. Most methods used for measuring ¢ and p involve testing in a laboratory. The first thing to do, therefore, is to get un- disturbed samples of the soil from the site to the laboratory. Normally this is done by taking the samples at convenient depths during the process of sinking the boreholes described in § 20. The ‘auger tool is used just prior to sampling in order to reduce the amount of local disturbance in the soil. Sampling is achieved by driving a thin-walled sampler tube (Shelby tube), of about 4 in. internal diameter, into the soil. On recovery, the ends of the tube are sealed as quickly as possible with wax, capped, and taped. Later, the sample, sealed in its tube, is sent to the laboratory for testing. When the 4 in, undisturbed samples are received at the laboratory, 1} in. diameter specimens, are cut from these for the actual testing Tt is impossible to aver-emphasize the fact that the number af undisturbed samples taken at any site should be considerable. All real soils vary widely within themselves, even though they appear by visual inspection to be uniform; and at most sites the strata vary from depth to depth, and from position to position across the site, Thus it would be most unwise to base the design for the foundations of an important structure on the results of tests on38 FOUNDATION DESIGN SIMPLY EXPLATNED only two or three samples taken at random depths in each of only ‘two or three boreholes. It is unlikely that so few samples would reveal adequately the strength (and wealmesses) of the enormous mass of soil on which we are to rely to support any extensive structure. Note that it is of course impossible to obtain truly ‘undisturbed’ samples of any soil, because the very penetration of the sampling tube into the soil causes disturbance. With cohesive soils (lays) which are not too hard or friable, the disturbanee is normally of a minor character, and the errors which arise are generally Tess than the variation between the soil as actually sampled and other soils elsewhere on the site which it is hoped the sample may reasonably represent. But with sands and gravels the diserepancies are more pro- nounced, partly because the sampling tube has to push individual particles to one side in order to achieve penetration at all, and partly because the samples tend to spill out of the tube, interfering with the natural degree of compaction. For these reasons it is generally more reliable, and certainly simpler, to determine the strengths of sandy soils and gravels by penctration tests as described in § 45. HESIVE SOILS §88. Plastic clays are the simplest soils to consider under laboratory conditions. They come very close to being truly cohesive, ie, they rely for their strength almost. entirely on cohesion and derive negligible benefit from internal friction forees. ‘Thus a clay is soft and smooth to the touch. Hsventially, it is not gritty. If We take a small circular prism of clay of unit oross-sectional area, as shown at Fig. 21(a), and load this miniature column as indicated, it will eventually collapse, as shown either at Figs. 21 (b) or 21(c). At Fig, 21()) the two ends of our prism have sought out the weakest plane between them, and are sliding past one another at: an anglo of 45°. At Fig, 21 (c) the clay is more plastic and tends to squeeze up and outwards into the shape of a, harrel: then one end starts to penetrate the swollen centre of the other, until failure ocours in this manner. Kither way, failure is by part of the material sliding against the other at an angle of 45°. ‘These diagonal planes can be seen developing while the load on the prism is being built up to its maximum value at failure, and very real forces persist as between the clay on one side of the failure SOIL MECHANICS TESTS 30 @ Pig. 21. Failure of clay prism under azcial load. plane and the clay on the other side. These forces are the cohesive strength of the clay and, as clays have negligible internal friction strength, the ehear slrengih here is made up entirely by the cohesive strength of the clay. Thus if s=tho shearing resistance of the clay, and c=the cohesive strength of the clay, ‘wo can represent the relationship mathematically as This is the same as equation (3) whieh we derived from Coulomb's equation in the previous chapter. If the force applied on the end of our prism of unit, eross- sectional area is taken as qu, the effectiveness of this on a plane at 46° will be less, actually. % . And the area of the oblique plane on which this force acts is greater than unity, actually V3. Therefore the shear stress on the plane is fore But as we have (4) In other words, the shear strength of the clay (which is due to its cohesive strength) is equal to half the stress applied at the end of our little column, UNCONFINED COMPRESSION TEST §39. In the laboratory it is customary to make the miniature column we test 1] in. diameter and 8 in. long, ‘This is then referred to as our specimen.“0 FOUNDATION DBSIGN SIMPLY EXPLAINED ‘The apparatus used is known as the unconfined-compression test apparatus and is indicated in Fig. 22. The specimen is placed between two plates 4 and B. Plate B is fixed in position; but plate A, which is supported by a calibrated spring, can be pulled upwards by turning the screwed wheel on the top of the frame. ‘The amount of load applied to the end of the specimen is measured 7 MANOWHEEL & Fixeo PURTE SPECIMEN AL wovagus, SUAPE Fig. 22. Apparatus for unconfined-compression test. by the amount the calibrated spring extends, and this is plotted directly by the apparatus on a chart which also shows the amount the specimen is squeezed. The longitudinal squeeze causes the specimen to swell laterally; and as the applied compressive stress ge equals 12%, this lateral swelling has to be taken into the area’ reckoning. § 40. The use of our result from the unconfined compression test may now be demonstrated as follows. ‘The ultimate bearing capacity of a cohesive soil was given at equation (5) as W But we have just shown at equation (14) that one, 2 ‘Therefore Y=8xG =Bq0. SOIL MECHANICS a But in § 33 we showed that the safe bearing capacity of a soil ga is, given by Therefore ‘Thus, if for example the unconfined compression strength of our soil had been, say, 15+5 Ibs/sq. in., then we should have known that the safe bearing capacity of the soil would be qo=15°6 Ibs|sq. in.=1 ton/sq. ft. SHEAR-VANE TEST § 41. It occasionally happens that a clay is so soft and sensitive that it is likely to undergo considerable change when disturbed. For such clays the unconfined compression test is not entirely suitable, and the shear strength of the clay is determined in situ in the borehole. This is done by foreing the cruciform vane, shown in Fig. 28, into the clay about 3 ft below the bottom of the bore- hole casing, and twisting the control rod at a specified rate. ‘The cohesive strength of the clay is then given by the equation 2 F (i + » z (15) z tho applied twist or torque, she diameter of the vane, the height of the vane, the cohesion of the clay 4 Fig. 23, Attachment for shear-vane test.a FOUNDATION DESIGN SIMPLY EXPLAINED SHWAR-ROX TEST § 42. Before describing the second laboratory test commonly in use, known as the triazial compression test, it is convenient to refer to the shear-box test, because this latter demonstrates well the prineiple of graphical representation of Coulomb's equation. In practice the shear-box test is seldom used nowadays, excopt for ‘testing coarse sands and gravels in rather particular circumstances. ‘The apparatus for the shear-box test is shown at Fig. 24(a) Fasentially, it eonsists of a hox split in two parts about a horizontal centre, The soil specimen is packed into the box, and one part of the box is then moved horizontally in relation to the other while different normal loads are applied across the plane of rupture as indicated, Records are made of the horizontal shearing foree SHEARING. SHEARING, Fouce. soi saMLe @ apparatus, y 4 Pins CUNY NORMAL STRESS (@ cRAPH OF RESULTS. Fig. 24, Shear-bor test. SOIL MECHANICS TESTS 4“ required to produce rupture for the different normal loads applied in each case. ‘Three typical test results are indicated at Fig. 24 (0), where the shearing strength is plotted against the normal pressure. Case 1 is for a plastic clay, which is entirely cohesive, and therefore the shear strength is constant, and quite independent of the normal stress applied. Caso 2 is for a dry sand, which, being truly frictional, has a shear strength direotly proportional to the normal stress, and therefore equal to zero when the normal stress is zero. Case 3 is for a svil which combines cuhesive and frictional properites. Each of the soils indicated at Fig. 24(}) represents graphically Coulomb's equation s=c-+p.tan 9, as the student should now be able to demonstrate for himself. TRIAXIAL COMPRESSION TEST § 43. The triaxial compression test is used for testing soils which combine cohesive and frictional properties. The apparatus used is indicated at Fig. 25(a). The cylindrical specimen is the same size as used for the un- confined compression test, i.e. 1f in. diameter x3 in. long, but this time it is contained in a rubber sheath and surrounded by water housed in a perspex container. A hydraulic pump is used to build up the pressure of the water to any value required, and the vertical load is then applied to the end of the specimen until failure occurs, Let p,=the total vertical pressure, and p,=the lateral pressure, ‘Then for any test the result is plotted as shown at Fig. 25(d), ‘p, and p, being marked off along the base of the graph, anda Mohr semicircle constructed above the base, and passing through p, and. ‘p, A number of tests are made in this way, and a number of Mohr semicircles drawn as indicated. A line is then drawn just touching the tops of the semicircles (the tangent to the semicircles), and it can be demonstrated that this line represents Coulomb's equation in the same way as the lines we saw at Fig. 24(). ‘This is a most convenient result, and oan be used for testing all suils: which have sulflivient culwsion to enable undisturbed samples to be taken and brought to the laboratory. Indeed, with the triaxial compression-test apparatus one can determine values of c and g for most soils other than sands and gravels. ‘Note that the unconfined compression test described in § $9 is merely a special case of the triaxial compression test, but suitable only for cohesive soils where the frictional contribution is negligible.“ FOUNDATION DESIGN SIMPLY EXPLAINED | ARIAL LOAD =} waren once PREssuRE, SPECIMEN RUBBER SHEATH, [—rensrex. covrnnee, @ Avparatus Pa Pr sTRess, ® car oF ResuLTs, Pig. 25. Priaxial-compression test. § 44, Our results from the triaxial compreasion test are made use of as follows. Vor a frictionless soil we apply our value of ¢ in equation (5). For other soils we apply our values for c and g in equation (7). In either ease the result we obtain is the ultimate bearing capacity of the soil, to which it is necessary to apply a suitable load factor to convert to the safe bearing capacity as described at § 33. In cases where cartwheel failure needs to be considered, we apply our values for ¢ and g in Coulomb's equation so as to deter- mine « (the shear strength of the soil); and then we carry out the analysis proposed by Fellenius using equation (9). SOIL MECHANICS TESTS 46 STANDARD PENETRATION TEST $45. For sands and gravels where it is impracticable to obtain undisturbed samples, as referred to previously (§ 37), the bearing capacity of the soil ean be determined from penetration tests made at the bottom of boreholes. After all debris from the boring operations has been moved, a standard tool 2 in. diameter is driven into the soil, using 140-Ib hammer, dropping 30 in, for every blow. The number of blows required to drive the tool a. distanee of 1 ft, into the undisturbed soil is known as the penetration value, N, of the soil ab the depth. of the test. ‘Terzaghi has related the number of blows per foot to the allow- able bearing capacity of sands for footings of different breadths. This relationship is given in Fig. 26 for foundations on dry or moist sands. The likely maximum settlement at these bearing capacities would be 1 in. For saturated sands the allowable bearing pressures should be reduced by 50 per cent. for foundations near the surface, and by 38-per cent. for foundations constructed at a depth below the surface equal to the breadth of the foundation. 1 NOTE! WATER TABLE” MORE THAN 28 FT BELOW a FOUNDATION 1 very pense ‘ 7 oe BEARING PRESSURE TONS/Sa.FT 2 z tecee | ee Ze g°o 5 SOO WIDTH OF FOUNDATION, B FT 26. Perzaghi’s relationship between standard penetration. test values and allowable bearing eapacities for footings on sands.6 FOUNDATION DESIGN SIMPLY EXPLAINED As an example of the use of the standard penetration test, suppose we had to carry a load of 300 tons on a dry sand where ‘the penetration value WV was thirty blows per foot. We see from ‘Fig. 26 that a foundation 10 ft wide would be suitable for an allowable pressure of 3 tons/sq. ft, so that a foundation 10 ft square would carry a load of 10 £10 ft x3 tons/eq. f ‘This would be satisfactory. 300 tons, CLASSIFICATION TESTS § 46. Before leaving this chapter on soil mechanics tests, it would not be out of the way to refer briefly to classification tests. ‘These do not give us specific data for enabling the bearing capacity of'a given soil to be determined, but they do indicate the condition of a soil as it has been found én situ. ‘Thus, cohesionless soils (sandy soils) are normally described as being found in a ‘dense’, ‘medium’, or ‘loose’ condition. 'The vague- ness of these descriptions can be éliminated as follows. ‘The voids ratio of a soil is defined by the ratio volume of voids volume of soil particles the voids ratio of the soil in its natural state, voids ratio of the soil when rearranged into its loosest state, the voids ratio of the soil when rearranged into its densest state, then the relative density of the soil as found is given by the formula (16) Eta When y, is less than 4, the soil is said to be ‘loose’; when y, is between } and 3, the soil is regarded as ‘medium’; and when 7, is greater than § the soil is known as ‘dense’. imilarly, cohesive soils (clayey soils) are deseribed according to their plasticity—the ease with which they may be moulded to shape by manipulation in the hand—the plasticity depending upon ‘the water content of the soil. Thus we have the liguid limit, which is the minimum moisture content at which the soil will flow under its own weight; the addition of further water would change the soil from its plastio state into a liquid. And at the other extreme we have the plastic limit, which is the minimum moisture content at which the soil can be rolled into a thread } in. diameter without: SOIL MECHANICS TESTS “ breaking; any less water would change the soil from its plastic state into a solid. ‘These are known as the consistency limits, Now, if w=the water content of the soil as found, ‘the water content at the plastic limit, to,=the water content at the liquid limit, thon the liquidity index of the soil as found is given by the expression = ww, ‘When 4 approaches unity, the soil is pretty well liquid; and when J approaches zero, the soil is verging on the hard solid state. (17)CHAPTER VI SPREAD FOUNDATIONS § 47. The present chapter seeks to show how spread foundations of different types may be designed; and a number of worked ‘examples are included. But first it will be appropriate to consider what different types of spread foundations are available, and why one type is used in preference to another. ‘The simplest form of spread foundation is 2 continuous sirip Joundation for supporting a wall. This type of foundation is used for domestic dwellings, and for larger buildings where load-bearing walls are employed. For framed buildings, where the superstructure loads reach the foundations through separate columns, the simplest foundation is provided by an independent pad or base under each column. Sometimes the adjacent columns in each row come relatively close to one another, in which case it is a convenience to link the bases between columns so as to form column strip-foundations. This may facilitate excavation work, particularly when the foundations have to be fairly deep. Where columns and walls oceur together at, the same structure, column strip-foundations serve to carry both the columns and the walls. ‘Where column loads are considerable, or the bearing capacity of the soil is poor, independent column bases may become so large ‘as nearly to touch one another. Tt then becomes economical to link up the bases to form one continuous slab or raft, because the stiffer bonding characteristics of a continuous slab of this form make for ‘economy in the reinforced concrete work. Further economies arise with raft construction because the excavation work is simpler, and the amount of edge shuttering is considerably reduced. Often the bending and fixing of steel reinforcement is also simplified. Where ‘the areas of independent column-foundations exceed half the total area of the building, it normally becomes economical to provide a raft foundation. Where continuous retaining walls are required at, the sides of a, bnilding basement, and where external asphalt, protection is involved, thete are further advantages in using « raft. ‘Where the bearing capacity of the soil is so poor that even a normal raft could not distribute the load sufficiently, there may be a solution in constructing a buoyant foundation, In essence this, is a deep hollow raft, itself lighter than the weight of soil removed SPREAD FOUNDATIONS. 0 in the excavation. Thus the ground under the foundation experi- ences a relief by the removal of the soil, and this relief is available on the credit side of the soil’s bearing-capacity account, and stands available for meeting the demands of supporting the loads from the superstructure. This was described in greater detail at § 5. § 48. In designing spread foundations it is necessary to ensure ‘that the undersides of the foundations are not too near the ground surface, This is because soils swell and heave, depending on the prevailing temperature and weather conditions. Clays swell when wetted, and shrink again on drying. These effects die out deeper down where the clay reaches a stable condi- tion, beyond the influence of seasonal variations. Foundations in clay constructed 3 ft below ground-level are generally satisfactory. Saturated sands and chalk suffer from what is known as frost eave. This is the lifting of the ground due to the expansion of ‘water in the soil when it freezes to form ice. Normally foundations constructed 2 ft below ground-level are deep enough to eseape the effects of frost, except on very wet sites where the water table is close to the surface, when it is better to keep the foundations about 3 ft down. $49. Certain soils disintegrate or spoil on exposure to the air, In particular, shales crumble; clays shrink and crack in dry ‘weather; and clays and chalk soften in wet weather, particularly if puddled by men walking on them. Sands and gravels tend to ‘work loose if disturbed by machinery or men walking, Tt is good practice, as 800n as the excavation for a foundation has been completed, to put down a layer of weak plain concrete 2or 3 in. thick to protect the formation. § 50. The design calculations which occupy the remainder of this book assume a knowledge of reinforced concrete design up to the standard given in the first three chapters of the companion volume, Reinforced Concrete Simply Explained.* The same simple symbols are used here as given in that book. Principally these are as follows A area of reinforcing steel, radius arm (or lever arm), readith of concrete, d=efieotive depth of member, to centre of reinforcement, S=shear foree, hear stress s D tensile stress in the reinforcement. * Oxford University Press: Pith Bait (989)0 FOUNDATION DESIGN SIMPLY EXPLAINED The materials used in the reinforeed concrete designs in this book are as follows. The concrete is 1:2:4 nominal mix, with a permissible compressive stress due to bending of 1,000 Ibs/sq. in., ‘and a permissible shear stress of 100 Ibs sq. in. The reinforcements are mild steel with a permissible tensile stress of 20,000 Ibs/sq. in. The elementary reinforced concrete formulae which will be ‘used frequently are as follows. The area of tensile reinforcement required in a member subject to bending is given by Mt 4-E os a us) ‘The shear stress in a member is given by 8 or O8dxE" (as) Note also that the compressive stress in a member subject to M bending will not exceed 1,000 Ibs/sq. in. so long as 575 does nob exceed 184, Alll calculations in this book are made with a slide rule, and do not pretend to an accuracy greater than 1 per cent. Since the inaccuracy of our estimates of the strengths of the soil and the concrete are likely to be many times greater than this, the small inaccuracies of the slide fule have no practical significance. ‘The safe bearing capacities taken in the caleulations would in practice be determined by methods as given earlier in this book, using cither the soil mechanics tests described in Chapter V, or ‘using the results of a tost-load as described in Chapter IIL. Alterna- tively, for approximate or preliminary calculations, the tables of safe bearing capacities given in Chapter IT may be used. WALL STRIP-FOUNDATION §51. Tho simplest form of spread foundation to design is a continuous strip for supporting a wall. This has bending in only one direction, More often then not, wall foundations are proportioned un- ‘economically by out-dated rule-of-thumb methods. The provision of stepped footings, to widen the brickwork at the lower courses, is quite unnecessary now that we have good-quality cements, making for concretes strong in shear, and reinforeing-steel capable of taking all tensile bending stresses. A single example will demonstrate the matter. SPREAD FOUNDATIONS 5 A TONS/TE RN. = 270g ans —} @ bce Fig. 27. Wall strip-foundation. ‘The wall shown in Fig. 27 carries a load of 2} tons for every foot run. The ground is good for a safe bearing pressure of only { ton/sq, ft, so that the foundation width required is 2:5 tons/ft 8 tonjsq, fe *'39 A suitable width to construct the foundation would be 3 ft 6 in. The maximum bending moment per foot run in a ease like this is approximately, wh rb, In reinforced concrete design work it is usually convenient to work in pounds and inches, s0 we have (2}x 2,240) x42, Se M: (20) Mt: 9,400 in, Ibs per fe ran. ‘Tho aroa of reinforcing steel xequined is a 56d xt If we make the foundation 12 in. thick, the effective depth will be 12in, less the thickness of eoncrete cover required outside the bars to give the necessary protection and less half the diameter of the iven hy the formula, Agwe FOUNDATION DESIGN SIMPLY EXPLAINED bare themselves. Allowing 1} in. for the concrete cover, d in our case will be approximately 10} in, Then 29,400 A= 5-86 X 10} x 20,000 0165 oq. in, per fe width, ‘This is conveniently provided by 2 in. diamoter bars arranged at 8 in, contres (area—0-165 aq, in.) In addition, three jin, bars are arranged longitudinally, and wired to the cross bars, and. this makes for a tigid mesh which helps keep the bars properly in position while the concrete is being placed and consolidated. INDEPENDENT BASE, LOADED CONCENTRICALLY § 52. Let us now design the foundation for an independent column such as we might got at the bottom of a framed building. This could be part of a block of offices or flats or the like. The column is 18 in. x 18 in. cross-section, and earties a load of 120 tons. ‘The safe bearing capacity of the soil is 2-5 tons/eq. ft met. (With concentrically loaded foundations near ground-level we normally work to net pressures, ie. pressures which ignore the terms yz in equations (10) and (11); then we ean ignore also the self-weight of the foundation, neglecting the difference in weight between the concrete foundation and the earth it replaces.) ‘The area of foundation required is therefore 120 tons : 3 tons|aq. 8884 f which is given by a foundation 7 6x7 ft Tn a case like this, where the foundation is bending in two directions at right angles, the total bending moment across the foundation in any direction may be taken as approximately (21) ‘Then, in our case, uM (220%2,240) x(7 x12) =1,880,000 in. Ibs. ‘Thio requires an aroa of reinforoing atocl 0-860 xt 1,880,000 0-86 X16 x 20,000 =68 sq, in, SPREAD FOUNDATIONS 53 ‘This is provided by twelve J in. diameter bars (area=7-2 sq. in, ‘The same amount of bending occurs in both directions at right angles, so it is necessary to provide the reinforcements both ways too. ‘To check that the compressive stress in the concrete does not exceed the permissible value of 1,000 Ibs/sq. in. for our 1:2:4 mix, we have to check that 7 does not exceed 184, We have M _ 1,880,000 a= SAX 16 which is therefore satisfactory. Tt now remains only to check the shear strength of the founda- tion. Shear failure in a ease like this occurs invariably by_ the column punching a pyramid-shaped piece of concrete out of the bottom of the foundation, as shown dotted in Fig. 28. The sloping sides of the pyramid are at 45°, and where these strike the level of the reinforcements is regarded as the critical section for shear: in other words, the internal compression forces prevent shear failure occurring nearer to the column than this. Tn our case the critical section for shoar is 9 in.+16 i from the column centre, so that the total perimeter at the critical shear section is 4 sides x (2X25 in.) each side=200 in, 8, i20t id ’ S. coud tA i2-1g wavs : x I a oe freimse sent yo! _sousne Fig. 28. Independent base, concentrically loaded.oy FOUNDATION DESIGN SIMPLY EXPLATNED ‘The base of our pyramid, of course, bears directly on the ground at the safe bearing pressure of 2-5 tons/sq. ft, 80 that of our total 120 tons, 50 in. x50 i 144 presents no shear across the critical section. However, the balance 2:5 tons/sq. ft=43-4 tons, of 120—43-4=76-6 tons does produce a shear on our perimeter of 200 in, ‘Now the shear stress is given by the formula aoa ee o-86d xb So in our ase we have 76:6 x2,240 0-86 16 x 200 =62 Ibs/sq. in. which, being less than the permissible shear stress of 100 Ibs/sq. in. for 1:2:4 concrete, is satisfactory. § 53. A more exact mothod of caleulating the bending moment in the previous example would be as follows. The critical seetion for bending is assumed to be at the face of the column, ie. at XX in Fig. 28. The upward force to the right of XX is 7 {6x 2-75 f6 x25 tons/sq. f: 80 that the bending moment at XX is M=(48 x 2,240) Ibs x 16} in. 770,000 in, Ibs. ‘This is seen to be only about 5 per cent. less than the 1,880,000 in. Ths obtained by the simplo formula (21). However, the reason for showing the more exact method of ealeulation is that, when we ‘come to eccentrically loaded bases, there is no alternative simple approach, and the fuller calculation becomes necessary. If students find §§ 54 and 55 for eccentrically loaded bases difficult to follow, it is suggested they should pass them by at the first reading of the book. 8 tons, INDEPENDENT BASE, LOADED RCOBNTRICALLY, WITHIN MIDDLB-THIRD § 64. In the provions example we designed a foundation for a column carrying a direct central load, where the load from the column was applied centrally on the foundation and centrally on SPREAD FOUNDATIONS, 5 the ground. Frequently in practice this happy state of affains doos not occur. For example, if the column is carrying a long slender beam, the beam will deflect and in this way will bend the top of the column; and depending on whether the column carries on above the level of the beam, and how stiff the column is, part of the bending induced in the column will be transferred down to the foundation. Bending at the feet of columns arises also from hori- zontal sway effects, such as from wind-forces and overhead-crane surge, Whore foundations have to porform the combined duty of resisting both a direct load and an applied bending moment at the same time, we can consider the effect as being the same as if the same direct load were applied at somo definite eccentricity. Thus the foundation would be eccenirically loaded. Suppose the direct load is W, and the bending moment is HY, then the eccentricity is given by I. In considering the centrally loaded foundation at § 62, we ignored the self-weight of the foundation, and worked to the net bearing capacity of the soil at ground-level. However, with founda- tions which are required to eater for out-of-balance bending effects, ‘the weight of the foundation is normally included in the calculation because it assists in providing the necessary stabilizing balance. ‘Then we have to work to the gross bearing capacity of the soi. Let us take a definite example. A column 12 in, 12 in, carries a direct load of 40 tons and a bending moment of 600 in. tons. It is required to design a suitable foundation to support this column on ground which has a safe gross bearing capacity of 2-25 tons/aq. ft. A rectangular base will give stability against, bending better than a square base. We will try a base 9 ft x4 fox2 ft thick, as shown in Fig. 29, ‘Then W=column load 40 tons weight of base 5 tons ‘Therefore the eccentricity is el ft, ‘This is Zess than one-sixth the length of the base—(} 9 ft—1-5 ft) —so that the resultant lino of action comes within the middlo-third of the base, This is indicated by the arrow in Fig. 29. When the56 FOUNDATION DESIGN SIMPLY BXPLAINED ot Lone (24-6! wee) Fig. 29. Independent base, loaded eccentrically, within middle-third. resultant comes within the middle-third of the base, the whole of the underside of the base is in compression and there is no tendency for uplift to oceur anywhere. In such a case, the macimum pressure on the soil under one end of a foundation of size Dx B is given by the formula Proax (22a) In our case Pam 8 toms (@ exh) FBX O FE of +25 (1-074) -18 tons /sq. ft. Similarly the minimum pressure is given by __W_(;_& a, SPREAD FOUNDATIONS sr so that in our ease ate ss on ( —Sxrnt) FXG OT 0-33 ton/sq. ft. These maximum and minimum pressure intensities of 218 and 0-38 ton/sq, ft are plotted to some suitable scale, as indicated in Fig. 29. The pressure distribution under the foundation is then indicated by the straight line joining the two points so plotted. ‘At the critical section for bending (XX in Fig, 20), the pressure intensity may then be sealed off, and is 1-36 tons/sq. ft. The bending moment at XX is due to the sum of the rectangular and triangular pressure distributions acting to the left of XX, so that M=(4 fe x4 fe 1-86 tons/sq, fe) x2 ft He fea thx OS? ton sq. ft) x207 fe 3517-561 fb tons =1,640,000 in. Ibs. ‘This requires an area of reinforcing steol 1,640,000 0-86 x22 X 20,000 4-4 sq. in, ‘Phis is provided by six 1 in. diameter bars (area=4-7 sq. in.). ‘To check the compressive stress in the eonerete we have BM _ 1,640,000 _ 4 ba 48 x22 which, being less than 184, is satisfactory. INDEPENDENT BASE, LOADED BCCRNTRIGALLY, OUTSIDE MIDDEE-THTRD § 55. The eccentricity in the case we have just considered was such that the resultant line of action came within the middle-third of the baso. Thus there was a positive compression under the full extent of the base; or in other words, ?niq could never be negative. Sometimes it happens that the eccentricity is such that the line of action of the resultant comes outside the middle-third. ‘Then the formulae (224) and (228) do not apply. The following example will demonstrate the matter.8 FOUNDATION DESIGN SIMPLY EXPLAINED A column 12 in. x12 in, carries a direct Joad of 30 tons and a bonding moment of 1,200 in, tons. ‘The gross bearing capacity of the supporting soil is 2-25 tons/sq. ft as before. We will try @ base 10 ft x5 fx 2} ft thiok, as shown in Fig. 30, ‘Then Wecolumn oad 30 tons weight of base 7-5 tons 37.5 tons ‘Therefore the eccentricity is M _ 1,200 in. tons Wo" 37-6 tons ‘This is more than one-sixth the Iength of the base (x10 ft 66 ft), so that the resultant comes outside the middle-third, and is 2-33 ft from the edge of the base, as indicated in Fig, 30. 82 in.= 2-66 fb. 1200 TONS wt 12t} sa. cou. at pure Es Fig. 30. Independent base, loaded eccentrically, outside middle-third. SPREAD FOUNDATIONS. oo Assuming a triangular distribution of stress, the length of base under pressure is BX233=6-99 ft, giving an average pressure intensity of 37-5 tons B99 fixe 107 tons/sq. fe and a mazimum edge-pressure of twice this, i. 2x 107=2-14 tons/sa. fb. ‘Tho pressure distribution is thon plotted to some suitable seale, as shown in Fig. 30. The pressure at XX is measured off, and is, 0-77 ton/sq. ft, and the bending moment at XX is M= (44 2X6 0-77 ton/aq. ft) x2} fe 4-(44 66 Bex “9S tonsyeq. £8) x3 fe -0-4-46°5—85-5 ft. tons 2,300,000 in. Ibs. This requires an area of reinforeing steel 2,300,000 Prov which is provided by seven 1 in. diameter bars (area=6-5 sq. in.) Because of the pressure distribution indicated in Fig. 30, it is clear that the right-hand end of the base is not pressing down on tthe soil at all, and indeed the weight of this part of the base is being used as’ counterweight against an uplifting tendency. In other words, the right-hand end of the base is tending to drop down from the remainder, and for this reason it is advisable to provide a little reinforcement slong the top of the base. In the ‘present example two 1 in. diameter bars would bo suitable. When the resultant comes outside the middle-third of the base, it is advisable to check what margin the foundation has for stability. Suppose, in our case, the eccentricity were increased, by matters beyond our control or knowledge, somo 60 por cont., i.e, from 2:66 ft to 4-0 ft. Then the maximum edge-pressure in the soil would be oy 375 tons Powe =2% ST HXE fe ‘This would be well within the ultimate capacity of a soil whose safe = tons/sq. ft.oo FOUNDATION DESIGN SIMPLY EXPLAINED bearing capacity is 2-25 tons/sq. ft, so that a reasonable margin for stability is ensured. ‘A point of interest worth noting is as follows. If, in our caleula- tion, we had ignored the self-weight of the, base, we should have an eocentricity of 1,200 44, A 79 40 in = 3:38 fe, giving maximum edge-pressure of 30 tons Poux—2% 901-66 feeb TE! tonsa: Me as against the 2-14 tons/sq. f6 we obtained previously. And strictly this should be related to a net bearing pressure for the soil (without ‘the yz term) which, 2} ft higher up, would be 2:14 tons/sq. ft. Clearly, then, the more accurate calculation, taking into account the term yz and the self-weight of the base, is well worth doing in ‘cases where eccentric loads have to be considered. COLUMN STRIP-FOUNDATION § 56. The next type of spread foundation we shall design is a column strip-foundation. The columns of the building are arranged at 14 ft centres in one direction and 22 ft centres in the other direction. Each column is 14 in, X14 in, and carries a concentric load of 90 tons, The safe bearing capacity of the soil is 0-75 ton/sq. ft net. If we were to use independent column bases, these would need an area of 20, 120 ag. ft, requiring a base size of 11 fe x11 ft For these dimensions, a suitable thickness would be 2 ft 6 in.; otherwise the deflections would become excessive. However, with a continuous strip-fomdation, where we have the greater stifiness arising from reverse bending, a thickness of only 1 ft 9 in, will suffice, clearly making for a considerable economy in concrete, And, for each column, the perimeter of excavation-side requiring timbering or other protection is only 28 ft, whereas with independent bases it would be 44 {t—over 50 per cent, more. SPREAD FOUNDATIONS a ‘The area of base required per column for a strip-foundation is {G— 120 aq, ft as before, The minimum width will therefore be 120 5.6 £4; and for convenience we shall make our foundation 8 fe 9 in, wide. ‘The longitudinal bending moments in a regular strip-foundation of this form, where reverse bending occurs, may be taken approxi- mately as 7 wh Peiaert ‘These bending moments occur under the columns (producing tension in the bottom of the foundation) and midway between the columns (producing tension in the top of the foundation). In our case (23) (902,240) x (14 12) 16 2,110,000 in, Ibs. This requires an area of stool 2,110,000 0-86 x 185 x 20,000 66 sq. in. ‘This is provided by six { in. diameter bars, lapped as shown in Fig. 81 to give twelve bars in the bottom of the foundation under the columns, and twelve bars in the top of the foundation midway between the columns (area each=7-2 sq. in.). (In certain circumstances there may be a case for reducing the amount of top reinforcement between the columns by some 30 per cent.: but this is not a matter for a simple book.) Mt: showing that the compressive stress in the conerete due to bending is quite low. ‘The transverse bending moment can be calculated, using equa- tion (20), whenee 2,640,000 in. Ibs,ee FOUNDATION DESIGN SIMPLY EXPLAINED roquiring an area of stool “Apa 2:640,000 0-86 x19 x 20,000 1 sq. in, ‘This is provided by § in. diameter stirrups at 6 in. centres (6-6 sq. in. per 14 ft bay), which serve conveniently to hold the 7 in. top reinforcements in place while the foundation is being conereted, Tie SECTION A-A ay Wa c corteron [Pp ates one one = Ht AMO eg om Mee) Ik Fig. 81. Column strip-foundation. We check for shear in the same way as we did at §52. The critical section for shear is 7 in.+19 in.=20 in, from the column centre, giving a shear perimeter of 4 (2x 26)—208 in. ‘The shear across the critical section is “8 tons, which is well below the permissible value of 100 Ibs/sq. in., and therefore satisfactory. SPREAD FOUNDATIONS 6 RAPT § 57. For our next example, consider a building with columns arranged at 12 ft contres in both directions. ‘The columns are 14 in, x14 in., and each carries a load of 75 tons. ‘The safo bearing capacity of the ground is 0-75 ton/sq. ft net. Tf we were to use independent column bases, these would need an area of te =100 sq, f6, requixing bases 10 ft x10 f6, whieh at 12 ft centres would be only 2 f& spart. This is therefore a ease where a raft foundation would be the most economical. We will try a fiat slab 18 in, thick, as shown in Fig, 32. ‘Let us design an internal bay of such a flat-plato slab. Bending takes place in two directions at right angles; but for convenience ‘we consider first the bending in one direction separately as follows. As in the previous example the effects of bending will be to produce tension in the bottom of the slab under the columns, and tension in the top of the slab across the line parallel and midway between the columns. ‘Now. the total bending moment per bay in one direction, producing tension in the bottom of the slab across the line joining the columns is WL ua", (24a) which in our case gives (75 x 2,240) x (12 12) 16 =1,510,000 in, Ibs, requiring a total area of bottom stecl of 1,510,000 TBO X TH} X 20,000 M= 5-7 sq. in. ‘This area of steel is not distributed uniformly over the total width of the bay, since the majority of the bending ocours in the more heavily loaded half-bay width straddling the line of the columns. ‘Phorefore across the 6 ft width under the columns we provide 16 por cont. of the 6-7 sq. in., equals 4-3 sq. in.; and across the remaining 6 ft width we provide the balanee, being 5°7—4:3 =I-4 sq. in. These areas are met respectively by ten } in. diameter bars (area=4-4 sq. in.) and four in, diamoter bars (area 1-76 sq. in.)oo FOUNDATION DESIGN SIMPLY EXPLAINED qo: ' . 3 ig onas, suds das iu : : é é I - g Hitt i tet Bt —PLAN— ro 12h! cou, cas. pre ae isa cou} a-Mid som wave 498 3 Tor & Borrot — SECTION AT COLUMNS — Fig. 82. Raft foundation. lo UlBTOTAL BOTH WAYS For the bending producing tension in the top of the slab across the line midway between the columns, the total bending moment per bay in one direction is, WD 36" (24n) SPREAD FOUNDATIONS 6 which in our ease gives (TX 2,240) x (12 12) 26 30,000 in, Tbs, requiring a total area of top steel of Ay £30,000 5 sq. in, ‘BG x 15} 20,000 ‘This also is apportioned unequally between the half-bay widths, though not in the same proportions. The 6 ft width straddling the column-lines takes 60 per cent. of the 3:5 sq. in., equals 21 sq. in. and the remaining 6 ft width has the balance of 3-5: 14 sq, in, These are met respectively by five } in. diameter bars (area +2 sq. in.) and four } in. diameter bars (area=1-76 sq. in.) In our case, where the columns are spaced at equal intervals in both directions, the amount of bending in the slab will be equal in both directions. ‘Therefore the calculation given above for re- inforeement in one direction applies equally for the reinforcement required in the slab in the other direction at right angles. ‘The student should now be able to check for himself the shear stress in the raft at the columns. ‘The above calculation applies strictly only to the internal bays of the raft, At the external bays, the amount of reinforcement may require some adjustment. Sometimes it becomes necessary to increase by about 25 per cent. the amount of reinforcement in the top of the raft in the direction at right angles to the raft edge. This depends on how far the raft projects beyond the lines of the outer- most columns, and also on whether the loads in the outer columns are less than the loads in the inner columns—which frequently they are, BUOYANT FOUNDATION § 68. Consider now a building where the columns are arranged at 15 ft centres in both directions, with column loads each of 170 tons, and the ground good for a safo bearing pressure of 0-5 ton/sa. ft. Clearly the load from the column alone produces a pressure 170 St ee TeeTs TOT ton/eq, f, which itself is in exeess of the safe bearing value of 0:5 ton/sq. ft. Therefore, if we are to provide any form of spread foundation, this must be a buoyant foundation. intensity of6 FOUNDATION DESIGN SIMPLY EXPLAINED ‘The approximate self-weight of such a foundation will be 0-25 tonjsq, ft. The student ean check this for himself Inter. ‘The total pressure applied to the soil will therefore be from column loads from weight of foundation 0-75 tonjsq. fe 0-25 ton/sq. ft Bul as the soll can only carry safely 0-5 tonjsq. ft we must excavate away the remaining 0-5 ton/sq. ft; and if the soil weighs 120 Ibs/eu, ft, this will entail excavating 9:6 ft, giving a relief of 9-5 x 120 _0.5 ton/sq. ft as required. 2,240 ‘Thus the form of our buoyant foundation will be as shown in Fig. 33, with the base slab 12 in. thick, the top slab 6 in. thick, and the cross-walls arranged on the lines of the columns and everywhere 12 in. thick. ‘The spaces between the walls are left unfilled. Note that although the net loading intensity on tho ground is only 0-5 ton/sq. ft, the actual pressure producing bending in the base slab is from total applied load Jess self-weight of base slab 1-00 ton/sq. ft 0-07 ton/sq. f 93 ton/sq. ft. "This is equivalent to a load W of 15 ft x15 ft x0-098 ton/sq. fb =210 tons. Now for an internal bay of a base slab such as this, where spans in both directions are equal, the bending moments per 15 ft bay width are WL M= yy wnder each wall, and =... (25) Mu WE pidvoay between valle, 2 26) ‘Thus the moment under each 15 ft length of wall, producing tension in the bottom of the base, is 210 x 2,240) 30 2,820,000 in, Ibs, Eee (1512) SPREAD FOUNDATIONS or shot PMB AT Ia ees tL tg ar tens Lay'gs wr 12! cas, eoTToM BARS. BARS. Sagar Inch. Bottom 2a g ar ID. ‘Bortow Lag ar ifers. soTTom SHOWN IN ONLY ONE DIRECTION IN) OTHER DIRECTION TO BE SMLAR. PLAN OF. BOTTOM SLAB iso! cou. cas | Lette eans | mas, tarney re act bans bags BES } é CRs ee Hr Meas, SECTION Fig. 33. Buoyant foundation.08 FOUNDATION DESIGN SIMPLY EXPLAINED requiring an area of reinforcing steel of 2,820,000 A ‘86 X OF x 20,000 17-2 sq. in.; and the moment midway between walls, producing tension in the top of the slab is M _(210.x 2,240) x(15 x12) uaa 2,100,000 in. Ibs, requiring an area of reinforcing steel of a 2,100,000 0-86 0} x 20,000 =128 9q. in, ‘Those requirements are met by providing a general mat of reinforcement, top and bottom, of fifteen } in. bars ab 12 in. centres (area=6°6 sq. in. each face). The top bars are then lapped. midway between the walls to give thirty bars at 6 in. centres (giving an area of 13-2 sq. in. as against 19-8 eq. in, required). And the bottom bars are lapped under the walls to give thirty bars at 6 in, contres, and a further fifteen bars are added, making a total of forty-five bars under the walls (giving an area of 19-9 sq. in. as against 17-2 sq. in, required), If there were to be a shear failure of the base slab, this would be 9 in. from the faces of the cross-walls, so that the plan area of slab shearing through would be 12 ft 6 in. x12 ft 6 in., and the shear foree on this area would be 12-5 ft x 125 £6 <0-98 ton/sq. f-—=145 tons. ‘This would be resisted on a perimeter of 4 x 12-5 ft=50 ft, so that ‘the shear stress in the concrete would be 145 x 2,240 0-86 x95 x (50x12) =66 Ibs/sq. in. and therefore perfectly satisfactory. Tt remains now only to design the cross-walls for vertical bending. Tho upward load causing bending in each wall is 19 £7} ft 0-93 ton/sq. ft=105 tons, which will produce a bending moment of (105 x 2,240) x (15x 12) SPREAD FOUNDATIONS 0 ‘This will require an area of reinforcing stecl of 7 2,640,000 0-86 x 112 X 20,000 =1:36 eq. in. and this is provided by two 1 in, diameter bars (are: ) in the top and bottom of each wall. The vertical shear force at the end of each cross-wall is 105 : “J =525 tons, producing a shear stress in the concrete of 52-5 X2,240 86x 112 x 12 100 Ths sq. in, which would just be satisfactory. The calculation given above for reinforcements in the base slab applies strictly only to the internal bays of the foundation, At external bays the reinforcements midway between the walls may need to be increased by as much as 75 per cent., depending mainly ‘on whether the edge-column loads are reduced or not. 7 84.CHAPTER VIL PILED FOUNDATIONS §59. Where ground of poor bearing capacity extends down a considerable distance, and a harder stratum is reached at depths of 20 ft or more, it usually pays to provide a piled foundation. Suppose the load in the column in Fig, 34 is 200 tons, and the upper stratum to the depth of 26 t is soft clay good for an allowable bearing pressure of 1 ton/sq. ft, and below this is a compact well- graded ballast. If this column load were to be carried on the upper clay it would require a foundation with an area of 200 sq. ft, ic. a foundation 14 ft square and, say, 3} ft thick. The same load, however, could be carried on four piles, each 30 ft long, driven down into the ballast so as to carry 50 tons each pile; and the load from the column would be transferred to the piles by a pile-cap, only 6 f& square and 2} ft thick. ‘The advantages of the piled @_ sprean- ® _muec- FOUNDATION. FOUNDATION. Fig. 34. Spread and piled foundations, PILED FOUNDATIONS a foundation in these circumstances are that it will not settle or deflect appreciably; it occupies considerably less space than a spread foundation, and normally it will be cheaper. Tn the previous example, if the allowable bearing pressure on the clay had been only } ton/sq. ft, the area required for a spread. foundation would have been 200 tons $sq. fe requiring a foundation 20 f6 square, This would have become very. expensive. Indeed, if adjacent columns had been closer together ‘than 20 ft, a bearing foundation would have been impossible, unless a buoyant foundation wore used which again would increase ‘the cost further. ‘The further alternative, of excavating 26 ft down to the ballast and building up in piers, would almost certainly be ruled out on considerations of cost. § 60. Other situations where piled foundations are necessary are indicated in Fig. 36. If a normal spread foundation were used at Case (a), the peat layer would squeeze, over a period of time, to perhaps only half its | vere ry il I 100 sc. ft, @ ® Fig. 35. Examples of piled foundations,Es FOUNDATION DESIGN SIMPLY EXPLAINED original thickness. Piles must therefore be driven through the peat to the sand strata beneath: then no additional load will be imposed on the peat, and no reliance placed upon it. At Case (b) the intense loading at wharf level would surcharge the wall, tending to push it over into the river. There is also the risk of the far greater vertical loading on one side of the wall causing a cartwheel slip failure as indicated. If piles are driven to Tell Below the lava of the wall footing, hoth thown dangore ate iminated. $61. Tintil ahont fifty years ago, piles were generally af timber, carefully selected by the engineers, often by visiting the forests. Since then it has become common practice to use concrete piles, generally reinforced with mild steel bars, though in some cases preeast piles are now prestressed using cold-drawn steel wire. Conerete piles have the advantages over timber piles of greater strength and permanence. They can be either precast (and driven into the ground subsequently), or the concrete can be cast in. situ in a shaft previously formed in the ground in the required position. PRECAST PILES § 62. Typical precast piles are indicated in Fig. 36. Precast piles have the following advaritages over cast-in-sitn piles. They are manufactured in clean moulds under carefully controlled eondi- tions above ground-level so that good construction is assured, When, later, they are driven into the ground, the effect of the striking hammer is a greater test of strength than the piles will subsequently have to endure in carrying theit permanent applied loads; in this way, any faults in the manufacture of the piles are shown up in the driving, so that the finished piles are, in a sense, each pre-tested. OAST-IN-SITU PILES 63. Cast-in-situ piles are formed in a variety of ways by different specialist firms. Often they aro about 16 in. or 21 in. diameter, depending on the load to be carried. Generally the shaft, in the gronnd is formed hy a steel the ronghly the same diameter as the finished pile. With driven east-in-situ piles, the tube ie fitted with an expendable pointed shoe at its bottom end, and driven until the required depth or resistance is achieved. With bored cast-in-situ piles, the shaft is formed by boring into the ground and subsequently using the tube as a ler which follows down behind the boring tools. PILED FOUNDATIONS 8 L ——— UNG AT SERS Un OPPS (HNL ATTEND <0 3 UPTING HOLE Furr Hou SiZe_| woRkINc LOADS. jo'sauise] 30 TONS ice [RECOMENDED] vino aENeoRcenENT 20 te a Mae so] oe tbh wee 40 “ sot 50. te toe fod eo] te |} Hee Fig. 36. Details of precast piles. After the shaft has been formed, whether by driving or by boring, “a cage of reinforcement is lowered into the tube, the shaft filled with conerete, and the tube then slowly withdrawn. With some systems the conerete is vibrated or hammered as the tube is lifted. This ensures that the concrete flows outwards to fill the hole made by the tube, together with any voids at the sides of the shaft where soft pockets or porous materials exist in the ground. Cast-in-situ piles have advantages over precast piles in that they can be constructed immediately following the placing of a contract without the delay necessary for the manufacture and curing of precast piles. In addition it is not necessary to know exactly the lengths required for the piles, as the tubo ean be taken down untilu FOUNDATION DESIGN SIMPLY EXPLAINED the required stratum is reached, and then just sufficient concrete placed as will fill the shaft up to the required level. Thus the construction of cast-in-situ piles ean be started more quickly than precast piles, and in certain circumstances there is a greater economy of materials. Another advantage of cast-in-situ piles is that with certain proprietary systems it is possible to construct the piles to very considerable depths. Thus the authors have used piles 110 ft long penetrating silty clay to get support on hard boulder clay beneath. Procast piles of this length would have been impracticable due to the difficulties of handling and driving. Bored piles have the advantage of eausing very little vibration to the surrounding ground or near-by structures; they are formed quite simply by using a three-legged rig only about 12 ft high, whereas driven piles normally require a special pile-driving frame of height varying from about 30 to 70 ft or more, depending on the length of pile required. Thus bored piles can more easily be constructed on sites where access is restricted; and although the ‘work of boring may go more slowly than with piles that are driven, it may be cheaper to use bored piles on smaill jobs where only a few piles are required. The fundamental disadvantages of bored piles, as compared with driven piles, are as follows. The very act of driving piles tightens the soil particles and creates a wedging action on the pile; but with bored piles, this advantage is lost because the shaft is formed in the ground after the soil has been bored out, Secondly, the driving-resistance of driven piles can give some indication of the load-carrying capacity of the finished pile; but with bored piles no evidence of this kind is available, Disadvantages common to all forms of cast-in-situ piles are as follows. It is difficult to ensure that the reinforcement is held everywhere in the centre of the pile, so that the protection of the reinforcements against corrosion is uncertain, And in grounds of a soft and plastic nature, there is a risk that the ground in the lower layers may squeeze into the concrete before it has hardened, and in this way reduce the effective cross-sectional area of the pile. Another objection is that certain soils are liable to. attack concrete, particularly if if is not ‘thoroughly consolidated. Precast piles do not suffer these unfortunate features, and the reinforcements and concrete are placed above ground-level in ‘moulds of precise dimensions under fully controlled conditions, ‘and the piles are not driven into the ground until the conorete is, well matured, PILED FOUNDATIONS 6 HANDLING AND DRIVING PILES § 64, In the driving of precast piles, heavier equipment is required, and greater care needs to be exercised, than with cast- in-situ piles. It is important that precast piles should be lifted only at their fifth points, as shown in Fig. 37, otherwise the piles are likely to be damaged in handling. If the piles are not lifted as shown, they. would have to be constructed of stouter proportions, which would Fig. 87. Stinging of precast piles. greatly increase their weight, as well as their cost and difficulty of handling, ‘Tho efficiency of pile-driving depends on the relation between the weight of the pile and the weight of the hammer, so that the lighter the pile the more efficiently it is driven. This is a practical point in favour of east-in-situ driven piles, where the weight of the steel tube is much less than the weight of a precast conerete pile. END-BEARING AND FRICTION PILES § 65. Some piles are driven through soft ground to derive most of their support by bearing on a harder stratum such as ballast or rook lower down, ‘These piles ate known as end-bearing piles. Other piles are driven in circumstances where no well-defined hard layer can be reached at any reasonable depth, and these piles rely for support on side resistance over a considerable proportion of their length. These piles are known as friction piles Often, in practice, piles are supported by a combination of the two effects. ‘The load-bearing capacity of end-bearing piles driven down to6 FOUNDATION DESIGN SIMPLY EXPLAINED ballast or sand can generally be forecast within reasonable limits of accuracy by pile-driving formulae, as given later. But the carrying capacity of frietion piles in clays and silts cannot be calculated reliably in the same way; and it.is preferable, in these cases, either to rely on calculations based on the frictional resistance between the ground and the pile, or even better to apply a test-load to one or more of the piles some while after they have been driven. TEST-PILES § 66. Where a single site involves the sinking of a large number of piles it is advisable, if time permits, to construct a few full-scale test-piles under normal working conditions. Loading-tests on such piles givo a better indication of the carrying capacity of the works-piles than any form of calculation, Piles can be load-tested in one of three ways, as indicated in Fig. 38. ‘The first method is to construct platform centrally over the pile and build up the necessary load on the platform. This method is not entirely satisfactory, because if the load gots out of balaneo, the pile is liable to suffer, and there is always the fear of the platform tilting and the load falling off with possible danger to ‘the workmen. ‘The second method is to support a suitable mass of pig-iron or sandbags on a simple bridge of steel beams, and to load tho pile by jacking down from the undersides of the beams. This method has the objection of the cost of bringing in and building up the load on the steel beams. ‘The third method is to avoid the provision of any special mass for providing the load, but to tie down to other piles constructed in the ground near by. Often four such anchor-piles will be sufficient for the purpose of tailing down the load necessary for testing one pile. While the test-load is being built up, in stages, to the full amount, measurements of the depression of the head of the pile are taken with great care, using either a surveyor’s level, or, preferably, a micrometer dial gauge clamped to a long heam supported from the ground well away from the area affected by the pile movement. ‘The ultimate resistance of works-piles is most satisfactorily determined by actual tests to failure of test-piles in this way. The likely ultimate resistance of other piles on the same site can then be estimated by comparing (in the case of driven piles) their PILED FOUNDATIONS 7 ie wregnanuce pan OF moodart 5 ta ih 1} anction th ina it Vsb rest ewe. © Pig. 38. Test-louding of piles. resistance to driving, or (in the ease of bored piles) the nature of strata through which they are bored. A word of warning. 'Test-piles tell you the resistance of a single pile, which will stress the ground in the manner indicated at Fig. 89(a). However, where many piles under a large building or8 FOUNDATION DESIGN SIMPLY EXPLAINED aa © ® Big. 39. Limitation of load-testing piles. structure are all loaded at the same time, the stressed zone in the ground extends to a greater depth below the bottoms of the piles as indicated at Fig. 39(b); and if there should be a soft soil layer at this greater depth, the group of piles would suffer, whereas the single pile would not be affected. For this reason test-borings should always be made on every piling site to seck out any soft layers, and clearly these borings should extend well beyond the depth intended for the piles. Piles designed on the basis of test-londs should normally be tested to twice the load they are required to carry as their safe working duty. On the other hand, where reliance is placed on a g00d pile-driving formula, a load factor of 24 to 3 is suitable for end-bearing piles driven to ballast or sand, but a higher factor is, necessary for piles in clays or silts. PILE-DRIVING FORMULAE § 67. All pilo-driving formulae are based on the assumption that the bearing capacity of a pile under static conditions bears a, simple relationship to its resistance to driving for the last foot or two. ‘This is obviously a little irrational. Tt largely ignores the nature of the ground through which the pile was driven to got down to its finished lovel. It takes very little account of the effect of friction PILED FOUNDATIONS on the sides of the pile, and this frietion tends only to develop later on as the ground settles back against the pile over a period of some weeks, And of course there can be no appreciation of any soft layer in the ground below the toes of the piles. ‘Thus no pile-driving formulae should be used without also considering the findings of a soil investigation along the lines already described in Chapters IV and V. And no pile-driving formula will give information as good as an actual load-test on & finished pile. Nevertheless, pile-driving formulae do give some guidance, provided the limitations of the formulae are realized. We may use ‘the following notation: R=the ultimate driving-resistance of the pile (tons), W=the weight of the hammer (tons), =the height of drop of the hammer (inches), and ‘=the final set of the pile per blow (inches). ‘Thon from considerations of kinetic energy Wxh=Rx8 so that Ba tony 2. In practice this simple equation is found to be far too great a simplification of the matter. It ignores the energy losses involved in the temporary clastic compression of the pile and the helmet and the packing; also the temporary compression of the ground; and also the bouneing of the hammer on the pile. If it were not for these losses, the pile would penetrate into the ground some distance greater than «. ‘Thus @ very much more rational formula was developed by Hiley in 1925 as follows: + + (28) whore c=the sum of the temporary elastic compressions in the pile, packings, and ground (inches), and n=the efficiency of the blow of the hammer, W+Pe WEP where P=the weight of the pile, helmet, ete., and he coefficient: of restitution of the materials under impact. ‘Tho temporary compressions in the pile and ground are found by actual measurements made while the pile is being driven, and the80 FOUNDATION DESIGN SIMPLY EXPLAINED temporary compression of various forms of packing has now been established experimentally by laboratory tests. These valuos are given in the Civil Engineering Code of Practice, No. 4, published by the Institution of Civil Engineers (1954). ‘This may all seem rather a complicated mathematical treatment to apply to anything as unrefined as the driving of conerete limbs weighing about 5 tons, with a 3-ton hammer, through 50 ft or more of variable geological strata, A second piling formula worthy of note is due to Oscar Faber, and expressed as follows: a Spa tons see (29) the diameter of the pile (inches), and =a factor depending on the length and elastic moduli of the pile and any packings. ‘Faber found that generally z has a value not greatly different from 0-02; 50 that, by ignoving the small term , we arrive at a simplified version of equation (29) as follows: _Wh 8+0-02h where R tons. + + + (80) ‘The merits of Faber’s formula are that, over a lifetime's ex- perience, he found that his calculated values agreed more closely with the results of actual pile-tests than any other formula he knew; and, requiring no reference to tables for constants, the simplified Version gives a ready tool for quick and simple calculation, No pile-driving formula ean be regarded in any sense as seienti- fieally accurate. Many reasons for this have been given earlier. ‘Under these ciroumstances, probably the simple Faber formula, aiven at equation (30), provides as reliable an estimate of the ultimate resistance of a driven pile as any other formula. §68. As an example of the use of a pile-driving formula, consider the piles 30 ft long shown in Fig. 4. Piles driven ta frictional soils such as ballast are better suited to ealeulation based on pile-driving formulae than piles driven to cohesive soils; and, in such a case, a load factor of 2} would be adequate using any of the formulae given at equations (28), (29), or (30). Suppose the piles were driven with a 3-ton hammer, dropping 4 ft until a final sot was reached of five blows per in. of penetration, PILED FOUNDATIONS a ‘Then using equation (30) we have pe Wh F008 3x48 2+ 02 xa) 24 tons, ‘This would be the ultimate resistance of each pile. As the actual load to be carried per pile is 50 tons, the load factor would be 124 tons 50 tons which would be satisfactory, BEARING CAPACITY OF PILES ON BASIS OF SOIL TESTS § 69. The bearing capacity of piles driven in cohesive soils carmot reliably be determined by the use of any pile-driving formula, There are three main reasons for this. First, the sideways shake of the pile, as it is being driven, makes an enlarged hole so that the pile derives but little frictional support, at the time of driving: yet over a period of some weeks this gap will close in as the clay squeezes home and holds snugly against the pile. Secondly, moisture, trapped temporarily in the quickness of the driving process, may lubricate the sides of the pile: yet. in time this water will percolate away, allowing the full adhesive strength to develop between the clay and the pile. Both these considerations will make for poor sets, giving the appearance of an unrealistically low bearing capacity for the pile. ‘The third confusion is due to a deceptively high end-resistance arising from the low permeability of the soil building up short- term pressures at the instant of impact. Such pressures will dissipate quickly after the effect of driving, but, of course, are always present to cause confusion at the actual moment of striking. For these reasons the ultimate bearing capacity of piles driven in cohesive soils has to be determined either by test-loading (as described in § 66), or by calculation based on the shear strengths of the soils. Shear-strength calculations, furthermore, are the only means of assessing the bearing capacity of bored’ piles, where clearly no driving data will exist. Calculation, under these eireum- stances, is carried out as follows. ‘The ultimate hearing capacity of the pile will be QU G+Oy se BY che ultimate end-bearing capacity, and the ultimate side-friction capacity.8 FOUNDATION DESIGN SINPLY EXPLAINED And if Ay=the cross-sectional area of the pile, the surface area of the pile, per unit length, and J=the length of the pile in a soil of cohesion ¢, then we have Q=9 Aye Soo (aa) and Gf =0°6 loc. see (3) Equation (32) is nothing more than equation (5), but with the factor 6 increased to 9 due to the greater containing influences present locally and at the depth of the point of a pile. Equation (33) is nothing more than the shear strength of the soil acting on the surface area of the pile, and multiplied by a factor 0°5 to allow for the adhesion between ‘the disturbed soil and the pile being less efficient than the full internal shear strength of the soil itself. § 70. As an example of this soil mechanics approach, consider the bored pile shown in Fig. 40. The strata consists of 16 ft of made ground, underlain by 20 ft of medium elay (c=8 Ibs/sq. in), and hot MAGE | GROUND. MEUM cLAY Fig. 40. Example of bored pile through various strata, PILED FOUNDATIONS 83 underlain again by stiff clay (¢-=28 Ibs/sq. in.). The pile is required to support a working load of 40 tons. We will try a pile 16 in, diameter and 55 ft long. ‘From equation (32) we have yg yy EX IG 28 =9 x EXIGE 28 C= faa * 155 ‘From equation (33), we have for the upper clay = 22-7 tons. yf 08 162018 8 = a xt 8 21:6 tons, Similarly, we have for the lower clay : m16,, 28 015x208 5 28, = 0-5 x20 x78 5 28. Therefore total ultimate bearing capacity Q, 119-8 tons. With the actual load to be carried of 40 tons, the load-factor would be 119.8 0 which would be satisfactory for a pile in stiff clay caleulated in this manner, F: PILE-CAPS § 71. Referring back to Fig. 34 again, it will be noted how the oad from the superstructure column is transferred to the four piles by means of a pile-cap. The pile-cap has to be designed so as to be strong enough not to fail by bending at the middle. A check has also to be made that the shear stresses at the column and at the piles are not excessive: though frequently the thickness selected to give the required stiffness, and to give adequate anchorage for the column. and pile reinforcements, leads to fairly low shear stresses in the concrete. A suitable detail for the pile-eap would be as indicated in Fig. 41. Tho piles are at 3 ft 6 in. centres in both directions, and the pile-cap is 6 ft 0 in, square and 2 ft 6 in. thick. Tn a case like this, where the foundation is bending in two directions at right angles between four points of support, the total bending moment across the cap in any direction may be taken approximately as i yaw Gl (34)8 FOUNDATION DESIGN SIMPLY EXPLAINED — saas orm was | | 26" Pig. 41. Pile-cap over four piles. ‘Then, in our case “yp (2002,240) x (9-5 12) x68 360,000 in, Ibs requiring an area of reinforeing steel 2,350,000 O86 x27 x 20,000 05 sq, in. which is provided by seven I in. diameter bars (area=5-5 sq, in.) in both directions In this case the load from the column will spread directly on to the piles, so there is no possibility of shear failure. CHAPTER VIII RETAINING WALLS § 72. Excavations in rock ean be finished to a vertival face so that the rock stands sheer for a considerable height. Soils behave differently and cannot stand more steeply than their natural angle of repose. This frequently is inconvenient (as for example in certain rail and road euttings, or at basements to buildings), and then it becomes necessary to construct a retaining wall behind which the soil will stand, supported artificially by leaning against the back of the wall. § 73. In designing retaining walls we are faced with four inter- related probloms. First we have to estimate what forces will be created on the back of the wall by the soil which is to be retained, ‘Next, we have to select a suitable profile for the wall and its base, and determine whether the wall will be stable so as not to fail by overturning. Then we have to check whether the bearing pressure under the base is suitable and whether the base is deep enough to prevent the wall sliding forward. And finally we have to analyse the wall itself in terms of reinforcement required and stresses in, the concrete. ‘These stages each require a certain amount of experience. This applies particularly to the choice of wall profile which can be determined only by a process of trial and correction. Before attempting to give any mathematical formulae for deter- mining the lateral pressures which ean act on retaining walls, it is perhaps better to appreciate from a common-sense approach the likely limits over which these pressures ean vary. We shall also consider to what extent these pressures can be controlled by simple design features, such as drainage precautions and the like. ACTIVE PRESSURES § 74. A stiff clay can be excavated to a vertical faco, and will stand like this for quite a while, depending on its moisture content and other physical properties. However, on exposure to the effects, of drying air, the clay will lose its cohesive nature and will crumble. On the other hand, an excess of moisture will so weaken the olay ‘that it slumps Tike a liquid. Suppose the liquid clay were entirely without internal friction forces: then it would produce a hydrostatic pressure of yz, where88 FOUNDATION DESIGN SIMPLY EXPLAINED y=the bulk density of the soil, and z—the depth below surface level. ‘Ifwe take the bulk density of the liquefied clay as 90 Ibs/ou. ft this ‘would give us a lateral pressure of 902 Ibs/sq. ft. Clays which are more dense than this are sufficiently plastic to develop internal shear forces. ‘Thus the maximum active lateral pressure from a liquid clay, acting at depth z below the surface level, would be pa=902; but clays which aro stiffer s0 as to be in the plastie range press very much loss, and it ie thon frequently assumed that tho active pressure is pe=302. ‘Now consider a sand. In its dry state this will stand as a heap at an angle of repose of about 30°. If we now fill with dry sand the space between this heap and the back of our retaining wall (Fig. 42), it is clear that half the additional sand will press against the back of the wall and half will press on top of the original heap. ‘Thus one-third of the weight of the sand is active against the wall, 02, or approximately * Ibsen. ft, s0 that p, Suppose, now, the sand were saturated by filling behind the wall ‘with water. Then the pressure on the wall would be increased by the hydrostatic pressure of the water, so that the total active pressure would be pe—302-4-62-45-—924z, ‘Thus we see that the practical limits for the active pressures behind retaining walls, whether from clays or sands, will come within the range of about to pe= 902, - + + (35) jee Fig. 42. Sand pressing on back of retaining wall. RETAINING WALLS s Tt is true that hard clays may stand for a time so as to produce no active pressure at all, but no engineer would dare to rely on this as a basis for designing a permanent wall. WATER PRESSURE §75. The point arises now as to whether the soil should be regarded as dry (p.—30z) or saturated (pa=902). Where proper drainage facilities are provided, and build-up of water ‘pressure cannot occur, it is safe to design for active pressures of Pe= 302, However, such drainage facilities are sometimes impracticable, as for example in certain eases of basements or pits. In the case of clay the matter then becomes more difficult to assess. If.a wall is constructed with a shuttered face, and then back- filled against with clay, there can be no doubt that any water in ‘the ground, or water percolating down from above, will press upon ‘the back of the wall. But if the wall is constructed hard against an oxeavated clay face, tho imperviousness of the clay should prevent, water ever finding its way to the back face of the wall. However, under these circumstances, the upper 3 to 5 feot of the clay may shrink away from the wall owing to seasonal shrinkage, in which case the full water pressure may act for the depth of the shrinkage movement, And then how much further down may the water seep subsequently? One thing is clear. When the water gets between the clay and the wall, only the water can press on the wall—not the water plus the clay. Alternatively, if the eondition of the clay is such that its active pressure is greater than p.=62-4z, then the clay will push the water away, and only the active pressure of the clay need be taken into account in the design ealeulations. DRAINAGE § 76. Normally it is possible for engineers to provide suitable drainage facilities at retaining walls to prevent, or at any rate to limit, the possible build-up of water pressure. This may not be so with’ basements or pits, but generally is ro for free-standing, retaining walls, as shown in Fig. 43. Drainage material placed behind the back of the wall will prevent the build-up of water, which can then be run away through weepholes or by a longitudinal back-drain to suitable outlet points. Water from weepholes should be collected, or run away, £0 as to provent: the soil under the toe of the base being88 FOUNDATION DESIGN SIMPLY EXPLAINED softened. Back-drains should be built over a continuous concrete filing to prevent the water running down to foundation level where it could seep underneath and lubricate the base and so set off a sliding failure. In cohesive soils it may be worth giving umbrella protection to the soil at the back of the wall by concrete paving laid to falls to drainage gulleys. ‘Where extensive hill-side behind the back of a wall is likely to give a considerable run-off of surface water, this water should be DRAINAGE RUBBLE WEEPHOLES PAVING Fig. 43. Drainage at retaining wall. intercepted by a continuous drainage ditch dug in the hill-side behind the wall. To prevent the water, cut off by this ditch, seeping down into the soil acting on the back of the retaining wall, the ditch should be kept away from the wall a distance at least ‘equal to the wall height. SUROHARGE § 77. The pressures given at equation (35) are due only to the ground behind the wall, and assume that this is finished off to a horizontal surface. However, when the ground is built up at a slope, or where imposed loadings oceur on top of the ground, the pressures on the back of the wall will be increased quite consider- ably, and we have to take account of this in our calculations. Common examples of imposed loadings behind retaining walls are the storage of goods at a wharf or loading platform, or traffic at the top of a cutting or on the approaches to a bridge. The effect of such imposed loading may conveniently be eonsidered as equiva lent to an extra depth of soil behind the wall, Thus, in Fig. 44, if RETAINING WALLS 50 w=the imposed load on the ground behind the wall, y=the bulk density of the soil, and 24=the equivalent extra depth of soil, we have a . +++ (86) ‘The extra pressure due to 2, may then be determined by using the simple equation (35), or (with caution) the more elegant formulae given later at equations (37) or (38). c— Fig. 44. Surcharge of retaining wall, MATHEMATICAL THEORIES FOR PRESSURES ON WALLS § 78. While not being disposed greatly to favour the use of soil mechanics formulae as applied to the calculations of lateral pressures on retaining walls, the authors feel it would be only right to include here an indication of these formulae. First we will vonsider the active pressure pushing on the back of the wall. Later we will consider the passive pressure at the front which prevents the wall from being pushed forwards. (a) Active Pressure At § 27 we discussed at Jength Coulomb's general equation for ‘the relationship between shear strength, cohesion, and friction in soils, This equation is saetp tan 9. We saw also, at Fig. 25(2), how this equation ean be represented for the condition at failure, by a line drawn as a tangent to the ‘Mohr semicircles,0 FOUNDATION DESIGN SIMPLY BXPLAINED ‘Now, mathematically, it can be shown quite simply that if p,=the major stress, acting vertically, and ‘py=the horizontal stress, ae, 2 2)tetmn (16-8), And sinoe here p,=yz, we have the active horizontal pressure on the back of the wall equal to z tan? (45—£)—: pete pemyet (* #)—2¢tan G %). - (7) ‘his formula was first published by Bell. ‘For cohesionless sols, where c=0, the seoond term on the right- hand side becomes zero, 60 that ye tant (« 2 then py=p, tan? («* Pe ) +. (88a) ‘This special case of the formule is generally attributed to Rankine, By a little mathematical manipulation this may be rewritten (for ‘those who prefer it) as , (1asing’ p= (22). e+ (888) Hither of equations (38) or (388) will give the same numerical result, and for truly cohesionless soils may be regarded as giving realistic results. Frequently these will be of the order of pa=302, as we had derived at equation (35) for dry sands. For cohesive soils equation (37) will ‘give negative pressure~ values down to a certain minimum depth; this is based on the cohesive strength enabling the soil to stand sheer but, as already explained, it would be imprudent to rely on this in any design calculations. Accordingly, sound as it may be theoretically, equa- tion (37) needs tempering with a little common sense, particularly over the range of the smaller depth values. (8) Passive Pressure ‘Now we must consider the passive pressure at the front of the wall foundation whieh prevents the wall being pushed forward ‘The lateral pressure required to move a soil forward is very much greater than the active pressure a soil will exert on the back of a wall, Thus passive pressures are many times greater than active pressures, Therefore to calculate what passive resistance we can rely upon at the front of a retaining wall, we need to derive fresh formulae. RETAINING WALLS a ‘Turning again to Coulomb's equation and Fig. 25(6), it can be demonstrated mathematically that if p:=the major stress, acting horizontally, and py=the vertical stress, thon ay tnt (154-2) tan (454-2). Putting pp=y2, we have the passive horizontal pressure at the front of the wall equal to rrayetont (1648) 120m (1618). + «00 This formula also is due to Bell. For cohesionless soils, where c: hand side becomes zero, so that Pompe tan? («+2): (40a) ‘This is Rankine’s formula for passive pressure, and may also be expressed she second term on the right- Ls rm ( (405) Equations (40a) and (408) are both reasonably reliable for use with traly cohesionless soils. Equation (39), however, needs using with a great deal of care, particularly as passive resistance is normally sought at relatively shallow depths where cohesive soils are liable to shrink away from their duty as a result, of seasonal drying ont. MODES OF FAILURE OF RETAINING WALLS (a) Failure by Rotation § 79. Ibis indicated at Fig. 45(a) how the total active force P on the back of a free-standing wall combines veetorially with the weight of the wall W to produce an inclined resultant thrust B. If this resultant comes beyond the width of the base, the wall will fail by immediate overturning. If, on the other hand, the resultant thrust comes well in from tho edgo of tho base, and the base is of euitable width, the wall will be stable against rotation, However, it is necessary to check that the maximum bearing pressure under the toe of the base does not exceed the safe bearing capacity of the soil, much as we did for eccentrically loaded foundations at §§ 54 and 66: otherwise the toc will settle and the wall will fail by rotation in this way (see Fig. 45(2)).© FOUNDATION DESIGN SIMPLY EXPLAINED Fig. 45. Railure of wall by rotation. Generally, in designing retaining walls of this type, the resultant thrust comes outside the width of the middle-third. This does not matter. But it is necessary in all problems of stability such as this to check that there is a margin of stability of at least 1} on the ‘ultimate bearing capacity of the soil, just as we did in the worked example at § 55. In other words, it is not sufficient to design the wall for stability with the normal assessment of the active pressure behind the wall, and check that the safe bearing capacity of the soil under the toe of the wall is not exceeded: it is also necessary to check that a 50 per cent. increase of the active pressure will not cause failure by the ultimate bearing capacity of the soil being exceeded. (6) Failure by Forward Movement ‘Two things provent a free-standing wall from moving forward— the passive resistance of the soil at the front of the base, and the friction between the soil and the underside of the base (see Fig 46(a)). ‘Phe method of calculating the passive resistance—and the un- certainty of this in the ease of eohesive soils—have already been deseribed at §78(b). Where a considerable passive resistance is required this is often best achieved by constructing a projeot- ing key under the base of the wall, near the heel, as shown at Fig, 46(0).. ‘Phe base friction on cohesionless soils (sands and gravels) can be taken as Ry=W tang. - (41) RETAINING WALLS 98 Passe. geSibrance LS PRTIONAL key © eo Fig. 46, Failure of wall by forward movement. On cohesive soils the base friction is calculated as Ry=els oe 4) where L=the length of the base under pressure. ¢ is of course the ‘cohesion of the soil; and normally this is limited in equation (42) to about 7 Ibs/sq. in., because at this point the adhesion between the clay and the concrete becomes the limiting factor, no matter what the actual shear strength of the clay may be. (0) Cartwheel Failure Failure by cartwheel action may oceur in poor strength non- frictional soils, as indicated in Fig. 47. Here it is not the soil immediately at the wall which fails, but a much larger mass of soil which fails and carries the wall with it. ‘The disturbing influence in this case is the weight of ground SHEAR RESISTANCE Pig. 47. Cartwheel failure of wall.of POUNDATION DESIGN SIMPLY BXPLAINED behind the wall causing @ greater moment about some centre of rotation than the weight of ground in front of the wall acting in combination with the shear resistance of the soil along the are of, failure, . ‘The method of analysis follows the same pattern of procedure as already outlined at § 32. WALL WITH BASE IN FRONT § 80. Ae a practical example of retaining-wall design, consider the problem indicated at Fig, 48 a). A wall is required to hold back 12 ft of clay at the side of a new cutting. The cohesive strength of the clay is 6 Ibs/sq. in., and its bulk density is 120 Ibsfeu. £6. For this purpose a profile is chosen with the base extending from the front face of the wall (compare this with the different arrangement at Fig. 49(a)). The advantages of the presont arrangement are (a) that we do not have to excavate the soil situated to the right of the wall, with consequent saving of cost, and (b) the clay to the right is undisturbed, and not likely to lead to any great build-up of water pressure. ‘The 12 in. horizontal heel projection can, with eare, be out into ‘the vertical clay face; and this heel greatly enhances the stability of the wall, giving a useful tailing-down effect with a leverage arm of greatest advantage, For convenience, the whole of the calculation is made for a vertical slice of the wall 1 ft wide. (a) Stabitity ‘The forces acting on the structure are indicated at Fig. 48(b), and the calculation is most conveniently set out in tabular fashion as below: Forces Vertical | Horizontal Be o Ws (ally 1,635 = Wa (hase) L125 7 WF, (cll over heal) 180 7 (oil over base) 13320 = Pal 3 ‘Total moment | =| m0 RETAINING WALLS 95, rr © cantwueen. 450 “chp in oo 25 ZOO Ww. v8. #8 5/8 at wh cxs, Val xr iwhews sie! fr Sens. sib ux ait easy PEO as daar SF cas, : I ir gaia! Pig. 48, Retaining wall with base in front.96 FOUNDATION DESIGN SIMPLY EXPLAINED ‘The vertical loads are each worked out by multiplying the eross- sectional area shown in the diagram by the unit weight of the material. Thus the weight of the concrete base, for example, is W,=7} fox 1 £6 150 Ibs/ou, £1,126 Ibs. For the soil weights, the unit weight (or bulk density) is taken as 120 Ibs/ou. ft, s0 that, for example, W,=14 fx 1 ft x 120 Ibs/ou. ft= 1,680 Ibs. ‘The active horizontal pressure on the back of the wall is taken from equation 85 aa pa=80z. Therefore at 16 ft depth the pressure will be Pa=30 x 15=450 Ibs sq. fb, and this will reduce up the wall to zero pressure at the top. With a triangular distribution of pressure like this the total active force on the back of the wall will be (P, 450 Ibs jsq. fe x 15 ft See and this will act at a point one-third of the way up the 15 ft triangle, ie. 6 ft above the point marked O in the diagram. Now we take moments of all these applied forces (vertical and horizontal) about any centre, here chosen for convenience at O. The area about which each of the forces acts is indicated in the table; the last column gives the moment, being the product of each force by its respective lever arm. ‘Thus the total anti-clockwise moment is 30,530 ft Ibs. Now, for equilibrium this momont has to be balanced by an equal and opposite moment of 30,530 ft Ibs acting in a clockwise sense. This clockwise moment will comprise the product of the vertical reaction Ry under the base by the distance « of its line of action from the same centre O. We know the amount of Ry must equal the sum of all the applied vertical forces W,, Wa, Wo, and Wy. Therefore 3,380 Ibs, By X0=30,630 30,530 80,580 _ 5.3 f so that pray 8 ‘Thns Ree ( 22 from the too of the hase, and clearly outside the middle-third. Assuming a triangular distribution of stress under the base (as we did at § 55), the length of base under pressure is 3x2-2—6-6 ft, giving an average pressure intensity of 39 tonjsq. ft and a maximum edge-pressure of 0-78 ton/sq. ft (=1,750 Ibs/sq. ft). RETAINING WALLS a Ibsjsq. in., the safe bearing capacity, from For our clay of o= ‘equation (10), is, =0774018 =0-98 ton/sq. f. "The maximum edge-pressure of 0:78 ton/sq. ft is therefore eatis- factory. Note that if the active force on the back of the wall were inereased by 50 per cent., the anti-clockwise moment would become 88,980 ft Ibs, so that # would be S805 —6-8 ft. Thus Re would be (7-5—68)=0-7 ft from the toe of the base. Then the 5,750 _ 2401 129 tons! sq, ft, and the maximum edge-pressure would be 2-46 tons/eq. ft. ‘The ultimate capacity of the clay, from equation (6), is 8x6 4 120x3 165 12,240" =2:324016 2-48 tons/q. ft, so there is an adequate margin of stability. In practieo, before Rp could get as near to the too as 0-7 fe, the heel of the base would try to rise very slightly, and in doing this ‘would have to lift a far greater weight of lay than wo took in the caloulation for Ws, actually spreading back from the face of the wall at about 30°. ‘Thus the margin of stability is greater than has been indicated above, (©) Forward Movement ‘The total active horizontal force Px is 3,380. This has to be resisted by the passive resistance of the soil, plus the friction under the base. It would bo unwise with clay to rely on the passive resistance of the ground in front of the base, owing to seasonal shrinkage. However, the ground in front of the key is better protected, and dooper, so that we can safely allow for this in our calculation, ‘Then, from equation (39), we have the passive horizontal pressure Pymye tant (1948) 42 tan (* r ) average pressure under the base would be98 FOUNDATION DESIGN SIMPLY EXPLAINED But for a non-fictional clay soil, g=zero, 50 tan( 46-49) ‘Therefore Pym yz420 120 x3B)+(2 x6 x 144) 20-4-1,728 =2,148 Ibs. ‘The base friction force, from equation (42), is Ry=eb (6x 144) x(3 2-2) ‘Therefore the total resistance to forward movement 5,700 Ibs. 7,848 Ibs. ‘Thus the factor of safety against forward movement is 48 _ oy 3,380 and this is adequate. (¢) Cartwheel Failure The circle associated with the least factor of safety (or the highest shear stress in the clay) is indicated at Fig. 48(c). The out- of-balance weight of soil cousing tendeney to rotate is 22,000 Ibs acting at a distance of 9 ft to the right of the centre of rotation, ‘This tendency is resisted by the shear stress in the soil, along the are 40-2 ft long, acting at a radius of 197 ft from the contre of rotation. ‘Thus 2,000 Ibs x9 ft—=40-2 fbx 19-7 ft xshear stress, ‘Therefore the actual shear stress is 200% _ 959 thejeq. ft 2x107 =1-78 Ibs/sq. in. ‘Thus the factor of safety against cartwheel failure is which is satisfactory. (@ Reinforced Concrete Design ‘The maximum bending stresses in the wall will oecur at the top of the haunch. Hore 30 13200 lbs/eq. #, 20 that the total horizontal force above the haunch causing bending is 820% 18 =2,540 Ibs. The bending moment is therefore ua2.10x(8 7 2) 152,000 in, Ibs, RETAINING WALLS 0 requiring an area of reinforeing steel A 188,000 0-77 sq. in. 0-86 X10 x 20,000, ‘This is provided by in. diameter bars at 4} in. centres (area ‘80 sq. in.). Higher up the wall, the bending moment reduces rapidly as shown at Fig. 48(d); and part-way up we can stop half the bars, leaving § in. at 9 in. contres; further up again we stop half the remainder, leaving § in. bars at 18 in, contres (see Fig. 48(@)) ‘At the top of the haunch M _ 132,000 ba? 12x 10® which is satisfactory. And the concrete shear stress is given by s 2,540 7 (86x10) XI =25 Ibs/sq. in, which also is satisfactory. ‘The base is analysed in much the same way as the wall. ‘The maximum bending stresses occur at the left edge of the haunch. ‘We have already calculated the pressure distribution under the base, and this is indicated at Fig. 48(8). ‘The bending moment at, the haunch due to the pressure is therefore (4-5 £6 660 Ibs/sq. ft) x2-25 fe (es fo 1190 Tl ") xo =5,680-+8,050 = 13,780 ft Ibs 'Dhis is partially opposed by the bending moment due to the self-weight of the base, and the soil over, namely (45 £6390 Ibs/sq, ft) x 226 ft 3,950 ft Ibs jiving a net bending moment at the base haunch—= _ 9,780 ft Ibs mae a 17,000 in. Ibs ‘This is slightly loss than the bending moment we calculated for the wall, and it is satisfactory to run the same reinforcements from the wall round the corner and along the base (see Fig. 48(c)). Half ‘tho reinforcements are stopped part-way along, where the bending moment has sufficiently reduced.100 FOUNDATION DESIGN SIMPLY EXPLAINED ‘The shear force at the base haunch is— Due to underside prossure: 4-5 ft x 560 Ibs/sq. ft 2,520 Ibs 465 ft 1100 Ihsjog, £02,680 Ibs 5,200 Ibs Less due to selfweight and soil over 465 fe x90 Thajjq. a= 1,750 Ibs 50 Ibs ‘The concrete shear stress is therefore 3,450 (a0) xB which is satisfactory. 3-4 Ibs/sq. in WALL WITH BASE BEHIND § 81. Consider now the wall indicated at Fig. 49(a). This is required to retain 20 ft of dry sand fill where the ground is to be built up, as might be required to provide car-parking facilities in front of an hotel building constructed on a site sloping down to the sea-front. The sand weighs 112 Ibs/ou. ft and has an angle of internal friction of 35°. The design of the wall is to cater for a ‘uniform imposed surcharge loading of 2 ewts/sq. ft. In this case it is convenient to build the wall with its base on the side the material is to be filled. ‘The weight of the fill then gives enhanced stability against rotation, and inoreases the friction under the base against forward movement. ‘Thus the base at Fig. 49(a) is shorter in proportion to the base at Fig, 48(a), and consequently cheaper. (a) Stabitity ‘With frictional soils we can reasonably rely on the mathematical equation (38n) for determining the active horizontal pressure on ‘the back of the wall. In our case this yields 1—sin von (Tie) Lo =u (Fyn) =30:32, RETAINING WALLS ips 24 lsa.rr adp Lee] @ Promue Wg we shen, tat § oe ens, Ug ax totcas, ig ae tens Bar toters, Lsret g ar ssteas. @ nemroncement FF Zeza00 mts. © Moments, Fig. 49. Retaining wall with base behind. wo02 FOUNDATION DESIGN SIMPLY EXPLAINED Tt is seen that this is not far from our approximate formula of 30z derived at equation (35) for a dry sand. ‘The surcharge of 2 cwts/sq. fi is equivalent to 224 % =T1a =2 ft of soil. (Equation 36) This will produce a uniform pressure, extending the full height on the back of the wall, of pa=30'3 x2 60-6 Ihs/sq, ft; and a total foreo on 23 f height of wall of 23 x 60:6—1,300 Ibs. ‘The active pressure from the soil will vary from zero at the top to 80-3%23=700 Ibs|sq. ft at 23 ft dopth, giving a total foreo on the back of the wall of 180% *8 3,000 tbs, Tho stability of the wall is then checked in tabular fashion as in the previous example, thus: Anticlocbioe ‘hout 0 fits Wr (wally S300 We (hase) 1 al ver bn) We (aurcharge) al over too) P, (oid 1,200 (Gurcharge) 16,100 otal forces ‘Total moment 202,000 ‘The distance « of the line of vertical reaction R, from 0 is therefore 202,900 f¢ Ib: 26,770 ‘This is 2-4 f¢ from the edgo of tho av ao that the average intensity. 26,770 2240X3 X24 edge pressure is 2x 1-65=830 tons/sq. The allowable pressure on this sand is obtained by using equation (11) and putting e=0. Thus we have of pressure is =1-65 tons/sq. ft, and the maximum RETAINING WALLS 109 ly2(M_~1)-+0-4yBN,] ES 4 fe Ort 112 x 10 x50 ~t 220 [uexexaonn =0-7 tons/aq. ft, ‘This exoceds the actual pressure of 3-30 tons/sq. ft and is therefore satisfactory, (0) Forward Movement ‘The total active horizontal force Ps is 9,390 Ibs. ‘The base friction foree, using equation (41), is =18,600 Ibs. ‘The factor of safety against forward movement, relying only on the base friction is, therefore, 18,600 P. indicating that no key under the base is required in this ease. (c) Cartwheel Failure ‘There is no risk of cartwheel failure with a good frictional soil of this nature. The student ean readily check this for himself. (@ Reinforced Concrete Design At the top of the wall-haunch the pressure due to the soil will be 30:3 x 20—606 Ibs/sq, ft, giving a total force above this level of 08 x20 6,000 Ibs. ‘The total pressure at this level due to surcharge will be 20 x 60-6 =1,212 Ibs, The bending moment is therefore 060 x 20X12 | 212 20x12 Ma 85,000-4-135,000 120,000 in. Ibs,104 FOUNDATION DESIGN SIMPLY EXPLAINED ‘This requires an area of steel of A, —_ 90 sq. in, x19 X 20,000 which is provided by 1 in. diameter bars at 6 in, centres which are stopped off in stages up the height of the wall as in the previous example, M _ 620,000 STs 8 _ 6,060-+1,212 ab 0-86 x19 x12 =143 which is satisfactory. 17 Ibs/sq. in., which also is satisfactory. CHAPTER 1X MATERIALS AND CONSTRUCTION § 82. Having now shown how foundations may be selected in regard to type and size to suit the ground conditions, and how the concrete thicknesses and reinforcements may be determined, it would not be out of the way to refer here to the materials that should he-used for the reinforced conerete work, and what precautions should be taken in the construction. If proper care is taken in the choice of materials and in the construction of reinforced eonerete foundations, the work may be {as near permanent as normal commercial conditions are likely to require. However, poor materials and poor workmanship can lead. to early deterioration and failure of the foundations, no matter how carefully and successfully the designs may have been prepared. § 83. For long life, concrete needs to be dense, with a complete absence of porosity, voids, and honeycombing. To achieve dense concrete, a number of requirements have to be met. The aggregates have themselves to be impervious. ‘They have also to be suitably graded from the largest size of stones down to the smallest. Sufficient cement has to be used in the mix to cover the surface of every aggregate particle, And sufficient water has to be added to set off the chemical reaction which causes the cement later to set and harden, and also to lubricate the aggregate particles so that they can be properly compacted to ensure a complete absence of voids. ‘A surplus of water makes for very weak conerete; and indeed the strength of concrete depends more than anything else on the ‘catio between the water and cement used in the mix. The higher ‘the proportion of water the weaker the mix, and the higher the proportion of cement the stronger the mix, As the cement and ‘the water both assist in lubricating the aggregate particles to enable the concrete to be properly compacted, it is a matter of some judgement ta know how much of the lubrication should be achieved by the use of cement and how much by the use of water. Clearly, cement is expensive while water is cheap. ‘The compaction of the particles is greatly facilitated if these are of a round nature as found with natural gravels and shingles. Angular particles which are obtained by crushing massive rocks such as limestone or granite are the most difficult to compact, and consequently require greater lubrication to achieve an equal degree Sea SeEE Rr ee eer eet eee eer oor eooe ear oor oP EGR eeP eee eee cree coc eee CLO eeeeeeeee eC |106 FOUNDATION DESIGN SIMPLY EXPLAINED of compaction and permanence. Thus rounded aggregates, or irregular near-round aggregates, make for concretes which are more easily compacted for a given amount of cement, and there- fore lend themselves to well-compacted and permanent conerete at the least cost. ‘This, of course, would not be true if one were working in an area where crushed stone was readily available in the locality and could bbe delivered cheaply, whereas rounded aggregates could only be procured by transporting them 50 miles or more. In this event, it might well be cheaper to use the angular aggregate with the expense of additional cement. ‘Nowadays it is normal on most works to compact the concrete, using vibrating tools. ‘This outs down on manual labour, and enables the concrete to be compacted with less water. And since the strength of the concrete depends almost entirely on the water/cement ratio, it enables the amount of cement to be reduced also. ‘Thus, in skilled hands, the cost of vibrating equipment is repaid in saving of manpower and saving of cement. ‘For normal work, the most suitable mix of concrete is the nominal 1:2:4 mix which means 1 part of cement to 2 parts of sand and 4 parts of stone, all measured by volume. If we take the weight of compacted cement as 90 Ibs/eu. ft, this mix is equivalent to 112 Ibs of cement to 2} cu. ft of sand and 6 eu. ft of stone. These are convenient measurements, because eement in bags is delivered in 1121b units, and it is convenient to gauge the sand and stone 50 as to suit a complete bagful of cement. ‘With non-vibrated conerete the water/cement ratio for the 1:2:4 eonerete mix should not exceed about 0-6 by weight. With vibrated concrete this can be reduced considerably. ‘A usefull practical test for controlling on site the amount of water required to the mix is made as follows. An open-ended metal former like a loud-hailer, being a frustum of a cone 12 in. high, 8 in, bottom diameter, and 4 in. top diameter, is stood with its base on a steel plate and filled with the concrete to be tested, the conereto being thoroughly compacted by tamping with a steel rod. The cone is then immediately lifted carefully upwards, so that ‘the conerete in the cone slumps under ite own weight. ‘The amount the conerete reduces in height from the original 12 in. is known as the slump of the eonerete, For unvibrated conerete @ slump of about 3 in, or 4 in. is normally satisfactory. Where mechanical vibrators are used, a slump of 1 in. to 2 in. is satisfactory. Conerete with a zero slump has ¢o little water that for outside work it is generally unsatis- factory and leads to imperfect compaction of the materials A I SS MATERIALS AND CONSTRUCTION wor Conerete with a 6-in. slump is so renny as to lead to segregation of the materials, leakage of grout through the formwork, and ‘consequent honeycombing and great weakness of the conorete, §84. For normal foundation work one would use Ordinary Portland coment of the slow-setting variety. Most cement-works in Great Britain grind their Ordinary Portland cement so fine today that for foundation work there is very little benefit to be achieved in using the so-called Rapid-hardening Portland cement. ‘Twenty or thirty youre ago thoro waa a greater advantago in using Rapid-hardening Portland cement when the quality of Ordinary Portland cement was not as high as it is today. Certain soils, particularly clays, are liable to contain sulphates in such a form as will attack concrete made with Ordinary Port- land cement. Ordinary Portland cement is satisfactory where the sulphate concentrations are not greater than 0-5 per eent. in the soil, or 100 parts in 100,000 parts of the ground water. Beyond this it is necessary to use Sulphate-resisting Portland cement, which is satisfactory for sulphate concentrations up to 2 per eent. in the soil or 600 parts in 100,000 parts of ground water. Beyond these concentrations it becomes necessary to use High Alumina cement. But this is a very special and difficult material ‘to use, and quite outside the scope of a simple book. In such cases the advice of the coment manufacturers should be sought. ‘Whatever cements and concrete mixtures are used, the results will only be satisfactory if the conerete is properly compacted, as already disoussed in § 83. It is the sulphate solutions in the ground water which destroy the cement by percolating through the concrete, and suecessful permanent work is only achieved by making the concrete so dense that ground solutions cannot find their way into the concrete with sufficient freedom to enable the sulphate concentrations readily to renew themselves for further attack. § 85. The setting and hardening of concrete is the result of a chemical reaction between the cement and the water. When the water is first mixed in with the other ingredients, the cement particles are never wetted right through, owing to the very lange surface area they prosent in relation to their minute weight. But over a period of several weeks the excess water in the mix will work its way through into the core of each individual cement particle, in this way completing the hardening process, and so increasing the strength of the concrete. However, if at any stage in the hardening process the conerete is allowed completely to dry out, a hard erust forms at the outer part of the cement particles for as far as these have already been wetted, and this crustos FOUNDATION DESIGN SIMPLY EXPLAINED prevents further chemical action taking place later if the concrete is subsequently wetted. Therefore it is of the greatest importance to ensure that conerete is not allowed to dry out in the early stages of its life. This is particularly so for the first 10 or 14 days, over which period the concrete should be kept covered with sacking or other proprietary materials which retain the concrete in a wet condition. ‘This wetting of the concrete after the time of the setting of the cement (generally 10 hours or less from the time water was first added te the mix), in no way rotarda the proccss of hardening, since, ac has already been explained, this latter is a chemical process and has nothing to do with any drying-out of the conerete. § 86. What does very much affect the rate at which concrete achieves its strength is the temperature. Concrete will harder. 50 per cont. more quickly at 60°F than it will at 36°F, and the hardening process is almost completely checked when the tempera- ‘ture drops to freezing level. For this reason concrete should never be placed when the air temperature is below 35°F, And for practical speedy work, where shuttering has to be removed with tho least possible delay, it is normally advisable to heat the concrete materials to a temperature of 60°F whenever the air temperature is below 40°F. When the chemical process of the cement setting has once started, this reaction will generate its own heat. Where the concrete is 2 ft thick or more, this heat will be suflicient to assist in the rate of hardening of the concrete, especially if the eonerete is covered with straw or sacking or other blanketing to retain the heat which is being generated. However, with very thin concrete members, severely exposed to the effects of rain and wind, the heat generated by tho chemical reaction is dissipated to such an extent as to be ineffective in this regard. Special attention has then to be given to heating the materials and protecting the finished work from frost attack. § 87. Tho quality of steel most suitable for foundation work is commercial mild steel with @ breaking strength of 28 to 33 tons) sq. in., and a yield point of about 15 tons/sq. in., depending on the diameter of the bar. Mild steel can be used in tension to a design stress of 20,000 Ibs/sq. in. Commercial mild steel is the cheapest form of reinforeing steel on the market, generally the most readily obtainable, and certainly the most easy and satisfactory to bend, Some firms and individuals recommend the use of steels of higher tensile strength. While it is true that these steels may be designed to work at higher stresses, with a consequent saving in weight of reinforcement, it has to be remembered that such stocls MATERIALS AND CONSTRUCTION 109 are more expensive to obtain, They are also very much more difficult to bend to the shape required, and the authors know of many cases where steels of this class have broken off short in the ‘process of bending. The bars which actually break can of course be replaced; but one cannot help wondering how many bars which do not break are in fact near to the point of so doing. ‘High-tensile steels have approximately the same elastic proper- ties as mild steel. This means that a high-tensile steel bar under a given load will stretch roughly the same amount as a mild steel bar under the came load. Now we have already explained that the permanence of eonerete work, particularly in foundations, depends on solid dense concrete which will not allow ground waters to percolate in, It is therefore important that the strain or stretching of the reinforeoments should be such as to avoid the formation in the conerete of any significant cracks. But as the strain of the reinforcement: is directly proportional to the stress applied to it, it is clearly undesirable for the permanence of the conerete work to stress the steel highly, even though the steel itself may be capable of earrying such high stresses. ‘There is, therefore, no ease for the use of high-tensile steel in reinforced conerete foundation work. § 88. Where dense concrete is used, as described in § 83, the steel reinforcements will be adequately protected against rusting if they havo 1} in. of conorete cover separating them from the external face of the concrete work, If the reinforcements are allowed to come closer to the face of the conorete than this, there is a likelihood of moisture seeping through the concreto—particularly as the very presence of the reinforcement makes it more difficult to compact the concrete properly in this position. In rusting, the reinforcements oxidize to produce a far greater volume of material, and this pushes the thin cover of concrete off the main body of the foundation, so that farther rusting can take place freely and at a very much greater rate. This in tum leads to a greater inerease in volume of the reinforcements, and may even cause actual splitting of the founda- tion. Certainly the area of full-strength steel will be diminished, as also will be the bond between the steel and the conerete. Thus it is important for pormanent work to onouro that minimum, cover of 1} in. is provided to all reinforcements. If now, on the other hand, an excess of cover is provided, then the reinforcements will act less effectively in resisting bending moments than has been assumed in the calculations. This will lead to higher stresses in the steel, with consequent increased strains or strotching, And since the steel is now some considerable distanceno FOUNDATION DESIGN SIMPLY. EXPLATNED in from the face of the conerete, a given amount of stretch of the steel will cause a greater stretch or cracking of the concrete at the face of the foundation. Thus it is important to see also that the reinforcements do not have an excess of cover over and above the 1} in, In practico this means that reinforeements should be placed with extreme care so as to achieve the 1} in, cover to within a tolerance of about } in. either way. Different contractors have different ways of doing this, but it is important to seo that the matler is given close attention, INDEXINDEX Active pressure, 85, 88 ‘Aahesion of scl ‘ander retaining walls, 88 to piles, 82 lovable bearing pressure, 1 et wea. 34 ‘Anglo of repose, 6,27 ‘Atlan hy moipbatoe fn ground. 107 Basalt 7 ‘Base fiction under retaining walls, 82 Beating capacity of piles, 76, 78, 81 Fook, 16 ols 18-14 ‘Bearing eapacity Sle, 13, 20 ot soq, 33 sltitaate, 20 o se. emma sb, 1 1 94 Dlsnes, ‘Bells forma for earth proseure, 90, 91 Bending moments in ‘buoyant foundations, 6, 68 Indapendent bass, 62, 57, 50 pile eaps, 84 afte, 63, 64 rotsining Walls, 98, 108 Bored pile, 73, 7 ‘design, 8 Borehole, 19 ‘Buoyant foundation, 4, 48 ‘eign, 63 Bly of tres in sil, 80 Cartwheel fare “ofretsining wal, 9 ‘of epread foundation, $1 CCastin-sitn piles, 72-14 Chalk, 7 Choiod ‘of foundation, 3-6, 48 of pl 10-0 Fotaining wall, Classfoation of sols, 12,12, 46 Caisivn eile I ek 20g. ohne (ne Independent base) Golan stip design, 60-62 ‘Goncentsieally Toaded column base, 52 Goneret, 108 et se ‘compestion, 108, 107 feover, 100 imix, 106 Feinforvoments, 108 fetting and hardening, 107-8 (Corrosion of reinforcements, 109 CCoslomb's equation, 20, Cover to reinforeements, 108 ‘Depth of foundation, 49 Dal of ‘and pile esps (se Piles) Peteining "walls (ao, Roteining wall design) spread foundstion (ee Foundations) Deterioration of exposed soll, 40 Distribation of stress in soil, 3 Disturbed eol samples, 22 Drainage of taining walls, 87 Driven piles, 78-14 ‘sign, 80 oconsialy loaded column bass, 64-60 age premore neh oan Ender retaining wll 9 99, adsbncing ples, 13 ‘cepety of eaplorati, 4 15 6 0a. hy 18 selend, 36 Fabor's formula for piles, 80 actors of exfety ice Load faotor) ‘Value, earbwheel, 31 ‘of retaining walls, OL ct se ‘tonlls, 2h seq i eal 9th, 8 Fist principles of design, 1 et 20g. Foundations 1 et 289. “ecrangement and choice, $5, 48 buayanty 4 48 depth, 4 Dll ore Piles) aft, 8 ies, 2 pra, 48 ot 09. ip, 60-82. Foundation design, 48 ot oq. ‘bigvant, 69-80 Exdependant "beso, concentrically Toad, 52-54 independent base, eccentrically Touded, oe aly 2-00, ‘sip, 60-62, 60-62 Rao, soy, 2609 Frition piles, 78 npecity of, 81na FOUNDATION DESIGN SIMPLY BXPLAINED Feiotion under base of retaining wall, 82 Frictional tolls, 11 of 04, Frost heave, 40 avis, § Granite, ‘Ground exploration (G00 Exploration) Hiley’e formula for piles, 79 8 rocks, 7 Tedetign of consaaroly loadod, 52-64 ‘dsig of eccontrioaly Tondo, 64-80 Intensey of preseure, 1 net, Teverlgation of sito, 15 ot 609. ‘ores, 19 Tending teat, 16 {eal pt Soins, rosk, 8, 9 Limestone, 7 Tiquidity index, 47 Tad factor ‘ols, 33 piles, 78, 80 ‘cat on soll, 15 et seq. Timitations of, 18 relation to settlement, 22 Losding test on ples, 16 Materials and construction, 105 et sea, -Metaméxphie rocks, 7 Medstons 7 Not loading intensity, 5 Passive pressure, 00 Penetration test, 45 ‘Peronssion boring, 20 Pile caps, 8 Piles, 4 10 et soa. ‘baring eapaclty, 81 bored, 73, 74 castin sifu, 72, 74 Sriven, 73, 74 fanvang formota, 8 fend-bearing snd fotion, 75 ‘handling and driving, 78 load factors fr, 78, 80 precast, 72, 74 et, 76 Pile, design of ‘bored pile, 82 Adsiven pile, 78, 75, 0, Pia til 19, fi 46 ‘resae ie Pes) ‘rowurdon ring wall (co Retaining val Raft foundation, $8, Rano formal for earth res, 90,91 formula orth pressure, 90, einem 108 alae denny of el «6 Fataining wall 89 et so. se tae 88 disinags fhiure by forward movement, 92 flare by roan, failure, cartwheel action, 98 Rocks, 6,7 et se beating espacity, 9, 10 ‘elding plane, 8 faults, joint, types, 7 Sand, 1 Sencetne, 7 Schis, 7 Sedimentary rocks, 7 Sst ot ple 0 Setcecent, 2,39 Setlement curve loading test 17 Shale? Shel and auger boring, 20 Shear in foundetonn 9 et seg Shearetcength ofsll 25 cb ang, 38 ot 29, Bhear-box fet, 42 rane test Side tion on ple, 75,81 Sas of foundation, 2 Site, Bomp of eonrete, 108 Soll preanre clive 8 Fosse, 80 sol nts 3? ots, ‘pnctaton 45 Ties bon i het wand, 4 {Wail eomproson, 43 tnconfined compresion, 89 Sul 6 10 et eege Tebing capes, 18 Dabevlinr f 1;2 e.g Soe continued cokecive, 12,28, 38 Aistebtion of tore 34, {allure 2 ef seq formation of, 10 tional, 12, 31 sample, 22,37 {ents (ova Sai ets) Spb Taundations, 48 ations, 48 ot seg. ‘buoyant, 48 re Seip, 48 Stabilty of rotaining walls, 82 Standard penoteaton test, 45 Sires distribution, soi, 34 Strip foundation, 48 sign, 50-82 ‘Submergence of soils, 14,45, 86 Balphates in sols, 107 INDEX ne Surcharge ‘of foundations, 29, ‘of reteining wall, 88, 102 ‘Torengh, 80, 45 ‘Teta nol, 37 ob 2g. on pls, 76 ‘thot of si behaviour, 24 et sg, ‘rial pit 19 ‘Piaxalsomprosson test, 43 ‘Ubimate beating capacity nek toile 38 ot sg. ‘uonined compression tat, 39 adisturbed sol sample, 37 Vane test on soll 4 Voids ratio, 46 Wash boring, 21 Water presstire on retaining walls, 86, 87 ‘Weepholes, 87