Geotechnical Engineering

7GEOTECHNICAL

ENGINEERING*

Mohamad H. HusseinPartner, Goble Rausche Likins and Associates, Inc.

Orlando, FloridaPartner, Pile Dynamics, Inc., Cleveland, Ohio

Frederick S. Merritt

Consulting EngineerWest Palm Beach, Florida

I n a broad sense, geotechnical engineer-ing is that branch of civil engineeringthat employs scientific methods to deter-mine, evaluate, and apply the interrela-

tionship between the geologic environment andengineered works. In a practical context, geotech-nical engineering encompasses evaluation, design,and construction involving earth materials.

The broad nature of this branch of civil engi-neering is indicated by the large number andnature of technical committees comprising theGeotechnical Engineering Division of the AmericanSociety of Civil Engineers: (1) computer applica-tions and numerical methods, (2) deep founda-tions, (3) earth retaining structures, (4) embank-ment dams and slopes, (5) engineering geology, (6)environmental geotechniques, (7) geophysicalengineering, (8) geotechnical safety and reliability,(9) grouting, (10) soil improvement and geosyn-thetics, (11) rock mechanics, (12) shallow founda-tions, (13) soil dynamics, and (14) soil properties.

Unlike other civil engineering disciplines,which typically deal with materials whose proper-ties are well defined, geotechnical engineering is

*Updated and revised from Sec. 7, “Geotechnical Engineering,” for Civil Engineers,” 3rd ed., McGraw-Hill Book Company, New Yor

7.1

Copyright (C) 1999 by The McGraw-Hill Cthis product is subject to the terms of its

concerned with subsurface materials whose prop-erties, in general, cannot be specified. Pioneers ofgeotechnical engineering relied on the “observa-tional approach” to develop an understanding ofsoil and rock mechanics and behavior of earthmaterials under loads. This approach wasenhanced by the advent of electronic field instru-mentation, wide availability of powerful personalcomputers, and development of sophisticatednumerical techniques. These now make it possibleto determine with greater accuracy the nonhomo-geneous, nonlinear, anisotropic nature and behav-ior of earth materials for application to engineer-ing works.

Geotechnical engineers should be proficient inthe application of soil and engineering mechanics,subsurface exploration methods, and laboratoryinvestigation techniques. They should have a thor-ough knowledge of relevant geological and geo-physical applications. And they should have broadpractical experience, inasmuch as the practice ofgeotechnical engineering requires more art thanscience. This requirement was clearly expressed byKarl Terzaghi, who made considerable contribu-

Frederick S. Merritt and William S. Gardner, “Standard Handbookk.

ompanies, Inc. All rights reserved. Use ofLicense Agreement. Click here to view.

7.2 n Section Seven

tions to the development of modern soil mechan-ics: “The magnitude of the difference between theperformance of real soils under field conditionsand the performance predicted on the basis of the-ory can only be ascertained by field experience.”

Foundation engineering is the art of selecting,designing, and constructing for engineering worksstructural support systems based on scientific prin-ciples of soil and engineering mechanics and earth-structure interaction theories, and incorporatingaccumulated experience with such applications.

7.1 Lessons from ConstructionClaims and Failures

Unanticipated subsurface conditions encounteredduring construction are by far the largest source ofconstruction-related claims for additional paymentby contractors and of cost overruns. Failures ofstructures as a result of foundation deficiencies canentail even greater costs, and moreover jeopardizepublic safety. A large body of experience has iden-tified consistently recurring factors contributing tothese occurrences. It is important for the engineerto be aware of the causes of cost overruns, claims,and failures and to use these lessons to help mini-mize similar future occurrences.

Unanticipated conditions (changed condi-tions) are the result of a variety of factors. The mostfrequent cause is the lack of definition of the con-stituents of rock and soil deposits and their varia-tion throughout the construction site. Relatedclaims are for unanticipated or excessive quantitiesof soil and rock excavation, misrepresentation ofthe quality and depth of bearing levels, unsuitableor insufficient on-site borrow materials, and unan-ticipated obstructions to pile driving or shaftdrilling. Misrepresentation of groundwater condi-tion is another common contributor to work extrasas well as to costly construction delays and emer-gency redesigns. Significant claims have also beengenerated by the failure of geotechnical investiga-tions to identify natural hazards, such as swellingsoils and rock minerals, unstable natural and cutslopes, and old fill deposits.

Failures of structures during construction areusually related to undesirable subsurface condi-tions not detected before or during construction,faulty design, or poor quality of work. Examplesare foundations supported on expansive or col-

Copyright (C) 1999 by The McGraw-Hill Cthis product is subject to the terms of its L

lapsing soils, on solutioned rock, or over undetect-ed weak or compressible subsoils; foundationdesigns too difficult to construct properly; founda-tions that do not perform as anticipated; and defi-cient construction techniques or materials. Anoth-er important design-related cause of failure isunderestimation or lack of recognition of extremeloads associated with natural events, such as earth-quakes, hurricanes, floods, and prolonged precipi-tation. Related failures include soil liquefactionduring earthquakes, hydrostatic uplift or waterdamage to structures because of a rise in ground-water level, undermining of foundations by scourand overtopping, or wave erosion of earth dikesand dams.

It is unlikely that conditions leading to con-struction claims and failures can ever be complete-ly precluded, inasmuch as discontinuities andextreme variation in subsurface conditions occurfrequently in many types of soil deposits and rockformations. An equally important constraint thatmust be appreciated by both engineers and clientsis the limitations of the current state of geotechni-cal engineering practice.

Mitigation of claims and failures, however, canbe achieved by fully integrated geotechnical inves-tigation, design, and construction quality assur-ance conducted by especially qualified profession-als. Integration, rather than departmentalization ofthese services, ensures a continuity of purpose andphilosophy that effectively reduces the risks asso-ciated with unanticipated subsurface conditionsand design and construction deficiencies. It is alsoextremely important that owners and primedesign professionals recognize that cost savingsthat reduce the quality of geotechnical servicesmay purchase liabilities several orders of magni-tude greater than their initial “savings.”

7.2 Soil and RockClassifications

All soils are initially the products of chemical alter-ation or mechanical disintegration of bedrock thathas been exposed to weathering processes. Soil con-stituents may have been subsequently modified bytransportation processes such as water, wind, andice and by inclusion and decomposition of organicmatter. Consequently, soil deposits may be given ageologic as well as a constitutive classification.

ompanies, Inc. All rights reserved. Use oficense Agreement. Click here to view.

Geotechnical Engineering n 7.3

Rock types are broadly classified by their modeof formation into igneous, metamorphic, and sedi-mentary deposits. The supporting ability (quality)assigned to rock for design or analysis shouldreflect the degree of alteration of the rock mineralsdue to weathering, the frequency of discontinu-ities within the rock mass, and the susceptibility ofthe rock to deterioration upon exposure.

7.2.1 Geologic Classification of Soils

The classification of a soil deposit with respect toits mode of deposition and geologic history is animportant step in understanding the variation insoil type and the maximum stresses imposed onthe deposit since deposition. (A geologic classifica-tion that identifies the mode of deposition of soildeposits is shown in Table 7.1.) The geologic histo-ry of a soil deposit may also provide valuable infor-mation on the rate of deposition, the amount oferosion, and the tectonic forces that may haveacted on the deposit subsequent to deposition.

Geological and agronomic soil maps anddetailed reports are issued by the U. S. Departmentof Agriculture, U. S. Geological Survey, and corre-sponding state offices. Old surveys are useful forlocating original shore lines, stream courses, andsurface-grade changes.

7.2.2 Unified Soil ClassificationSystem

This is the most widely used of the various consti-tutive classification systems and correlates soil typewith generalized soil behavior. All soils are classi-fied as coarse-grained (50% of the particles > 0.074mm), fine-grained (50% of the particles < 0.074mm), or predominantly organic (see Table 7.2).

Coarse-grained soils are categorized by theirparticle size into boulders (particles larger than 8in), cobbles (3 to 8 in), gravel, and sand. For sands(S) and gravels (G), grain-size distribution is iden-tified as either poorly graded (P) or well-graded(W), as indicated by the group symbol in Table 7.2.The presence of fine-grained soil fractions (under50%), such as silt and clay, is indicated by the sym-bols M and C, respectively. Sands may also be clas-sified as coarse (larger than No. 10 sieve), medium(smaller than No. 10 but larger than No. 40), or fine(smaller than No. 40). Because properties of thesesoils are usually significantly influenced by relativedensity Dr , rating of the in situ density and Dr is an


important consideration (see Art. 7.4).Fine-grained soils are classified by their liquid

limit and plasticity index as organic clays OH or siltsOL, inorganic clays CH or CL, or silts or sandy siltsMH or ML, as shown in Table 7.2. For the silts andorganic soils, the symbols H and L denote a highand low potential compressibility rating; for clays,they denote a high and low plasticity. Typically, theconsistency of cohesive soils is classified from pock-et penetrometer or Torvane tests on soil samples.The consistency ratings are expressed as follows:

Soft—under 0.25 tons/ft2

Firm—0.25 to 0.50 tons/ft2

Stiff—0.50 to 1.0 ton/ft2

Very stiff—1.0 to 2.0 tons/ft2

Hard—more than 2.0 tons/ft2

Table 7.1 Geologic Classification of Soil Deposits

Classification Mode of Formation

AeolianDune Wind deposition (coastal and desert)Loess Deposition during glacial periods

AlluvialAlluvium River and stream depositionLacustrine Lake waters, including glacial lakesFloodplain Floodwaters

ColluvialColluvium Downslope soil movementTalus Downslope movement of rock debris

GlacialGround Deposited and consolidated by glaciers

moraineTerminal Scour and transport at ice front

moraineOutwash Glacier melt waters

MarineBeach or Wave deposition

barEstuarine River estuary depositionLagoonal Deposition in lagoonsSalt marsh Deposition in sheltered tidal zones

ResidualResidual Complete alteration by in situ weathering

soilSaprolite Incomplete but intense alteration and

leachingLaterite Complex alteration in tropical environmentDecom- Advanced alteration within parent rock

posed rock


Well-graded gravels, gravel-sandmixtures, little or no fines

Poorly graded gravels or gravel-sand mixtures, little or no fines

Silty gravels, gravel-sand-siltmixtures

Clayey gravels, gravel-sand-claymixtures

Wide range in grain sizes andsubstantial amounts of allintermediate particle sizes

Predominantly one size, or arange of sizes with someintermediate sizes missing

Nonplastic fines or fines withlow plasticity (see ML soils)

Plastic fines (see CL soils)

Atterberg limits belowA line or PI < 4

Atterberg limits aboveA line with PI > 7

Soils above A line with4 < PI < 7 are border-line cases, require useof dual symbols

Clean gravels (littleor no fines)

Gravels with fines(appreciableamount of fines)

GW

GP

GM

GC

D60 /D10 > 41 < D30 /D10D60 < 3D10 , D30 , D60 = sizes corresponding to 10, 30, and

60% on grain-size curve

Not meeting all gradation requirements for GW

Table 7.2 Unified Soil Classification Including Identification and Descriptiona

Typical Name Field IdentificationProceduresbMajor Division Group

Symbol Laboratory Classification Criteriac

A. Coarse-grained soils (more than half of material larger than No. 200 sieve)d

1. Gravels (more than half of coarse fraction larger than No. 4 sieve)e

Well-graded sands, gravellysands, little or no fines

Poorly graded sands or gravel-sands, little or no fines

Silty sands, sand-silt mixtures

Clayey sands, sand-clay mixtures

Wide range in grain sizes andsubstantial amounts of allintermediate particle sizes

Predominantly one size, or arange of sizes with someintermediate sizes missing

Nonplastic fines or fines withlow plasticity (see ML soils)

Plastic fines (see CL soils)

Atterberg limits belowA line or PI < 4

Atterberg limits aboveA line with PI > 7

Soils with Atterberg lim-its above A line while4 < PI < 7 are border-line cases; require use ofdual symbols

Clean sands (littleor no fines)

Sands with fines(appreciableamount of fines)

SW

SP

SM

SC

D60 /D10 > 61 < D30 /D10D60 <3

Not meeting all gradation requirements for SW

2. Sands (more than half of coarse fraction smaller than No. 4 sieve)e

7.4

Copyright (C) 1999 by The McGraw-Hill Companies, Inc. All rights reserved. Use ofthis product is subject to the terms of its License Agreement. Click here to view.

Inorganic silts and very finesands, rock flour, silty or clayeyfine sands, or clayey silts withslight plasticity

Inorganic clays of low to mediumplasticity, gravelly clays, sandyclays, silty clays, lean clays

Organic silts and organic siltyclays of low plasticity

Inorganic silts, micaceous ordiatomaceous fine sandy or siltysoils, elastic silts

Inorganic clays of high plasticity,fat clays

Organic clays of medium to highplasticity

Plasticity chart laboratory classifica-tions of fine-grained soils com-pares them at equal liquid limit.Toughness and dry strengthincrease with increasing plasticityindex (PI)

Silts and clays withliquid limit lessthan 50

Silts and clays withliquid limit morethan 50

ML

CL

OL

MH

CH

OH

Typical Name

Identification Proceduresf

Major Division GroupSymbol Laboratory Classification Criteriac

B. Fine-grained soils (more than half of material larger than No. 200 sieve)d

Information required for describing coarse-grained soils:

For undisturbed soils, add information on stratification, degree of compactness, cementation, moisture conditions, and drainage characteristics. Give typical name;indicate approximate percentage of sand and gravel; maximum size, angularity, surface condition, and hardness of the coarse grains; local or geological name andother pertinent descriptive information; and symbol in parentheses. Example: Salty sand, gravelly; about 20% hard, angular gravel particles, 1/2 in maximum size;rounded and subangular sand grains, coarse to fine; about 15% nonplastic fines with low dry strength; well compacted and moist in place; alluvial sand; (SM).

Dry Strength(Crushing

Characteristics)

Dilatancy(Reaction

to Shaking)

Toughness(Consistency

Near PL)

C. Highly organic soils

Peat and other highly organicsoils

None to slight

Medium tohigh

Slight tomedium

Slight tomedium

None to veryhigh

Medium tohigh

Quick to slow

None to veryslow

Slow

Slow to none

None

None to veryslow

None

Medium

Slight

Slight tomedium

High

Slight tomedium

Pt Readily identified by color, odor, spongy feel, andoften by fibrous texture

7.5


Table 7.2 (Continued)

Field identification procedures for fine-grained soils or fractions:g

Dilatancy (Reaction to Shaking)

After removing particles larger than No. 40sieve, prepare a pat of moist soil with a volume ofabout 1/2 in3. Add enough water if necessary tomake the soil soft but not sticky.

Place the pat in the open palm of one hand andshake horizontally, striking vigorously against theother hand several times. A positive reaction con-sists of the appearance of water on the surface ofthe pat, which changes to a livery consistency andbecomes glossy. When the sample is squeezedbetween the fingers, the water and gloss disappearfrom the surface, the pat stiffens, and finally itcracks or crumbles. The rapidity of appearance ofwater during shaking and of its disappearanceduring squeezing assist in identifying the charac-ter of the fines in a soil.

Very fine clean sands give the quickest andmost distinct reaction, whereas a plastic clay hasno reaction. Inorganic silts, such as a typical rockflour, show a moderately quick reaction.

Dry Strength (Crushing Characteristics)

After removing particles larger than No. 40 sieve,mold a pat of soil to the consistency of putty, addingwater if necessary. Allow the pat to dry completelyby oven, sun, or air drying, then test its strength bybreaking and crumbling between the fingers Thisstrength is a measure of character and quantity ofthe colloidal fraction contained in the soil. The drystrength increases with increasing plasticity.

High dry strength is characteristic of clays ofthe CH group. A typical inorganic silt possessesonly very slight dry strength. Silty fine sands andsilts have about the same slight dry strength butcan be distinguished by the feel when powderingthe dried specimen. Fine sand feels gritty, whereasa typical silt has the smooth feel of flour.

Toughness (Consistency Near PL)

After particles larger than the No. 40 sieve areremoved, a specimen of soil about 1/2 in3 in size ismolded to the consistency of putty. If it is too dry,water must be added. If it is too sticky, the speci-men should be spread out in a thin layer andallowed to lose some moisture by evaporation.Then, the specimen is rolled out by hand on asmooth surface or between the palms into a threadabout 1/8 in in diameter. The thread is then foldedand rerolled repeatedly. During this manipulation,the moisture content is gradually reduced and thespecimen stiffens, finally loses its plasticity, andcrumbles when the plastic limit (PL) is reached.

After the thread crumbles, the pieces should belumped together and a slight kneading action con-tinued until the lump crumbles.

The tougher the thread near the PL and thestiffer the lump when it finally crumbles, the morepotent is the colloidal clay fraction in the soil.Weakness of the thread at the PL and quick loss ofcoherence of the lump below the PL indicate eitherorganic clay of low plasticity or materials such askaolin-type clays and organic clays that occurbelow the A line.

Highly organic clays have a very weak andspongy feel at PL.

7.6


Information required for describing fine-grained soils:

For undisturbed soils, add information on structure, stratification, consistency in undisturbed and remolded states, moisture, and drainage conditions. Give typicalname; indicate degree and character of plasticity; amount and maximum size of coarse grains; color in wet condition; odor, if any; local or geological name and otherpertinent descriptive information; and symbol in parentheses. Example: Clayey silt, brown; slightly plastic; small percentage of fine sand; numerous vertical root holes;firm and dry in place; loess; (ML).

a Adapted from recommendations of U.S. Army Corps of Engineers and U.S. Bureau of Reclamation. All sieve sizes United States standard.b Excluding particles larger than 3 in and basing fractions on estimated weights.c Use grain-size curve in identifying the fractions as given under field identification.

For coarse-grained soils, determine percentage of gravel and sand from grain-size curve. Depending on percentage of fines (fractions smaller than No. 200 sieve), coarse-grained soils areclassified as follows:

Less than 5% fines: GW, GP, SW, SPMore than 12% fines: GM, GC, SM, SC5% to 12% fines: Borderline cases requiring use of dual symbols

Soils possessing characteristics of two groups are designated by combinations of group symbols; for example, GW-GC indicates a well-graded, gravel-sand mixture with clay binder.d The No. 200 sieve size is about the smallest particle visible to the naked eye.e For visual classification, the 1/4 in size may be used as equivalent to the No. 4 sieve size.f Applicable to fractions smaller than No. 40 sieve.g These procedures are to be performed on the minus 40-sieve-size particles (about 1/64 in).

For field classification purposes, screening is not intended. Simply remove by hand the coarse particles that interfere with the tests.

7.7Copyright (C) 1999 by The McGraw-Hill Companies, Inc. All rights reserved. Use ofthis product is subject to the terms of its License Agreement. Click here to view.

7.8 n Section Seven

7.2.3 Rock Classification

Rook, obtained from core samples, is commonlycharacterized by its type, degree of alteration(weathering), and continuity of the core. (Whereoutcrop observations are possible, rock structuremay be mapped.) Rock-quality classifications aretypically based on the results of compressivestrength tests or the condition of the core samples,or both. Rock types typical of igneous depositsinclude basalt, granite, diorite, rhyolite, andandesite. Typical metamorphic rocks includeschist, gneiss, quartzite, slate, and marble. Rockstypical of sedimentary deposits include shale,sandstone, conglomerate, and limestone.

Rock structure and degree of fracturing usual-ly control the behavior of a rock mass that has beensignificantly altered by weathering processes. It isnecessary to characterize both regional and localstructural features that may influence design offoundations, excavations, and underground open-ings in rock. Information from geologic publica-tions and maps are useful for defining regionaltrends relative to the orientation of bedding, majorjoint systems, faults, and so on.

Rock-quality indices determined from inspec-tion of rock cores include the fracture frequency (FF)and rock-quality designation (RQD). FF is the num-ber of naturally occurring fractures per foot of corerun, whereas RQD is the cumulative length of natu-rally separated core pieces, 4 in or more in dimen-sion, expressed as a percentage of the length of corerun. The rock-quality rating also may be based onthe velocity index obtained from laboratory and insitu seismic-wave-propagation tests. The velocityindex is given by (Vs /Vl)

2, where Vs and Vl representseismic-wave velocities from in situ and laboratorycore measurements, respectively. Proposed RQDand velocity index rock-quality classifications and insitu deformability correlations are in Table 7.3. A rel-ative-strength rating of the quality rock cores repre-sentative of the intact elements of the rock mass, pro-posed by Deere and Miller, is based on the uniaxialcompressive (UC) strength of the core and its tan-gent modulus at one-half of the UC.

(D. U. Deere and R. P. Miller, “Classification andIndex Properties for Intact Rock,” Technical ReportAFWL-TR-65-116, Airforce Special Weapons Cen-ter, Kirtland Airforce Base, New Mexico, 1966.)

Inasmuch as some rocks tend to disintegraterapidly (slake) upon exposure to the atmosphere,


the potential for slaking should be rated from lab-oratory tests. These tests include emersion inwater, Los Angeles abrasion, repeated wetting anddrying, and other special tests, such as a slaking-durability test. Alteration of rock minerals due toweathering processes is often associated withreduction in rock hardness and increase in porosi-ty and discoloration. In an advanced stage ofweathering, the rock may contain soil-like seams,be easily abraded (friable), readily broken, andmay (but will not necessarily) exhibit a reducedRQD or FF. Rating of the degree of rock alterationwhen logging core specimens is a valuable aid inassessing rock quality.

7.3 Physical Properties of SoilsBasic soil properties and parameters can be subdi-vided into physical, index, and engineering cate-gories. Physical soil properties include density,particle size and distribution, specific gravity, andwater content.

The water content w of a soil sample representsthe weight of free water contained in the sampleexpressed as a percentage of its dry weight.

The degree of saturation S of the sample is theratio, expressed as percentage, of the volume offree water contained in a sample to its total volumeof voids Vν.

Porosity n, which is a measure of the relativeamount of voids, is the ratio of void volume to thetotal volume V of soil:

Velocity DeformabilityClassification RQD Index Ed /Et*

Very poor 0–25 0–0.20 Under 0.20Poor 25–50 0.20–0.40 Under 0.20Fair 50–75 0.40–0.60 0.20–0.50Good 75–90 0.60–0.80 0.50–0.80Excellent 90–100 0.80–1.00 0.80–1.00

* Ed = in situ deformation modulus of rock mass; Et = tangentmodulus at 50% of UC strength of core specimens.

Source: Deere, Patton and Cording, “Breakage of Rock,” Proceed-ings, 8th Symposium on Rock Mechanics, American Institute ofMining and Metallurgical Engineers, Minneapolis, Minn.

Table 7.3 Rock-Quality Classification andDeformability Correlation



(7.1)

The ratio of Vν to the volume occupied by the soilparticles Vs defines the void ratio e. Given e, thedegree of saturation may be computed from

(7.2)

where Gs represents the specific gravity of the soilparticles. For most inorganic soils, Gs is usually inthe range of 2.67 + 0.05.

The dry unit weight γd of a soil specimen withany degree of saturation may be calculated from

(7.3)

where γw is the unit weight of water and is usuallytaken as 62.4 lb/ft3 for fresh water and 64.0 lb/ft3 forseawater.

The particle-size distribution (gradation) ofsoils can be determined by mechanical (sieve)analysis and combined with hydrometer analysis ifthe sample contains a significant amount of parti-cles finer than 0.074 mm (No. 200 sieve). The soilparticle gradation in combination with the maxi-mum, minimum, and in situ density of cohesion-less soils can provide useful correlations with engi-neering properties (see Arts. 7.4 and 7.52).

Index Definition*

Plasticity Str

Liquidity Co

Shrinkage Sh

Activity Sw

*Wl = liquid limit; Wp = plastic limit; Wn = moisture content, %; Ws =

Table 7.4 Soil Indices


7.4 Index Parameters for SoilsIndex parameters of cohesive soils include liquidlimit, plastic limit, shrinkage limits, and activity.Such parameters are useful for classifying cohesivesoils and providing correlations with engineeringsoil properties.

The liquid limit of cohesive soils represents anear liquid state, that is, an undrained shearstrength about 0.01 lb/ft2. The water content atwhich the soil ceases to exhibit plastic behavior istermed the plastic limit. The shrinkage limit rep-resents the water content at which no further vol-ume change occurs with a reduction in water con-tent. The most useful classification and correlationparameters are the plasticity index Ip , the liquidityindex Il , the shrinkage index Is , and the activity Ac.These parameters are defined in Table 7.4.

Relative density Dr of cohesionless soils may beexpressed in terms of void ratio e or unit dryweight γd:

(7.4a)

(7.4b)

Dr provides cohesionless soil property and param-eter correlations, including friction angle, perme-ability, compressibility, small-strain shear modulus,cyclic shear strength, and so on.

Correlation

ength, compressibility, compactibility, and so forth

mpressibility and stress rate

rinkage potential

ell potential, and so on

shrinkage limit; µ = percent of soil finer than 0.002 mm (clay size).


7.10 n Section Seven

(H. Y. Fang, “Foundation Engineering Handbook,”2nd ed., Van Nostrand Reinhold, New York.)

7.5 Engineering Properties ofSoils

Engineering soil properties and parameters describethe behavior of soil under induced stress and envi-ronmental changes. Of interest to most geotechnicalapplications are the strength, deformability, andpermeability of in situ and compacted soils. ASTMpromulgates standard test procedures for soil prop-erties and parameters.

7.5.1 Shear Strength of CohesiveSoils

The undrained shear strength cu , of cohesive soilsunder static loading can be determined by severaltypes of laboratory tests, including uniaxial com-pression, triaxial compression (TC) or extension(TE), simple shear, direct shear, and torsion shear.The TC test is the most versatile and best under-stood laboratory test. Triaxial tests involve applica-tion of a controlled confining pressure σ3 and axialstress σ1 to a soil specimen. σ3 may be held constantand σ1 increased to failure (TC tests), or σ1 may beheld constant while σ3 is decreased to failure (TEtests). Specimens may be sheared in a drained orundrained condition.

The unconsolidated-undrained (UU) triaxialcompression test is appropriate and commonlyused for determining the cu of relatively good-qual-ity samples. For soils that do not exhibit changes insoil structure under elevated consolidation pres-sures, consolidated-undrained (CU) tests followingthe SHANSEP testing approach mitigate the effectsof sample disturbance.

(C. C. Ladd and R. Foott, “New Design Proceduresfor Stability of Soft Clays,” ASCE Journal of GeotechnicalEngineering Division, vol. 99, no. GT7, 1974.)

For cohesive soils exhibiting a normal claybehavior, a relationship between the normalizedundrained shear strength cu /σ′νo and the overcon-solidation ratio OCR can be defined independent-ly of the water content of the test specimen by

(7.5)


where cu is normalized by the preshear verticaleffective stress, the effective overburden pressureσ′νo , or the consolidation pressure σ′lc triaxial testconditions. OCR is the ratio of preconsolidationpressure to overburden pressure. The parameter Krepresents the cu /σ′νo of the soil in a normally con-solidated state, and n primarily depends on thetype of shear test. For CU triaxial compressiontests, K is approximately 0.32 + 0.02 and is lowestfor low plasticity soils and is a maximum for soilswith plasticity index Ip over 40%. The exponent n isusually within the range of 0.70 + 0.05 and tendsto be highest for OCR less than about 4.

In situ vane shear tests also are often used toprovide cu measurements in soft to firm clays. Testsare commonly made on both the undisturbed andremolded soil to investigate the sensitivity, theratio of the undisturbed to remolded soil strength.This test is not applicable in sand or silts or wherehard inclusions (nodules, shell, gravel, and soforth) may be present. (See also Art. 7.6.3.)

Drained shear strength of cohesive soils isimportant in design and control of constructionembankments on soft ground as well as in otherevaluations involving effective-stress analyses.Conventionally, drained shear strength τf isexpressed by the Mohr-Coulomb failure criteria as:

(7.6)

The c′ and φ′ parameters represent the effectivecohesion and effective friction angle of the soil,respectively. σ′n is the effective stress normal to theplane of shear failure and can be expressed interms of total stress σn as (σn–ue), where ue is theexcess pore-water pressure at failure. ue is inducedby changes in the principal stresses (∆σ1, ∆σ3 ). Forsaturated soils, it is expressed in terms of the pore-water-pressure parameter Af at failure as:

(7.7)

The effective-stress parameters c′, φ′, and Af arereadily determined by CU triaxial shear testsemploying pore-water-pressure measurements or,excepting Af , by consolidated-drained (CD) tests.

After large movements along preformed failureplanes, cohesive soils exhibit a significantly reduced(residual) shear strength. The corresponding effec-tive friction angle φ′r is dependent on Ip. For many



cohesive soils, φ′r is also a function of σ′n. The φ′rparameter is applied in analysis of the stability ofsoils where prior movements (slides) have occurred.

Cyclic loading with complete stress reversalsdecreases the shearing resistance of saturatedcohesive soils by inducing a progressive buildup inpore-water pressure. The amount of degradationdepends primarily on the intensity of the cyclicshear stress, the number of load cycles, the stresshistory of the soil, and the type of cyclic test used.The strength degradation potential can be deter-mined by postcyclic, UU tests.

7.5.2 Shear Strength of CohesionlessSoils

The shear strength of cohesionless soils under sta-tic loading is conventionally interpreted fromresults of drained or undrained TC tests incorpo-rating pore-pressure measurements. The effectiveangle of internal friction φ′ can also be expressedby Eq. (7.6), except that c′ is usually interpreted aszero. For cohesionless soils, φ′ is dependent ondensity or void ratio, gradation, grain shape, andgrain mineralogy. Markedly stress-dependent, φ′decreases with increasing σ′n , the effective stressnormal to the plane of shear failure.

Fig. 7.1 Chart for determining friction a


In situ cone penetration tests in sands may beused to estimate φ′ from cone resistance qc records.One approach relates the limiting qc values direct-ly to φ′. Where qc increases approximately linearlywith depth, φ′ can also be interpreted from theslope of the curve for qc – σνo vs. φ′νo , where σνo =total vertical stress, σ′νo = σνo – u, and u = pore-water pressure. The third approach is to interpretthe relative density Dr from qc and then relate φ′ toDr as a function of the gradation and grain shape ofthe sand.

Relative density provides good correlation withφ′ for a given gradation, grain shape, and normalstress range. A widely used correlation is shown inFig. 7.1. Dr can be interpreted from standard pene-tration resistance tests (Fig. 7.12) and cone penetra-tion resistance tests (see Arts. 7.6.2 and 7.6.3) or cal-culated from the results of in situ or maximum andminimum density tests.

Dense sands typically exhibit a reduction inshearing resistance at strains greater than thoserequired to develop the peak resistance. At rela-tively large strains, the stress-strain curves of looseand dense sands converge. The void ratio at whichthere is no volume change during shear is calledthe critical void ratio. A volume increase duringshear (dilatancy) of saturated, dense, cohesionless

ngles for sands. (After J. H. Schmertmann.)



soils produces negative pore-water pressures and atemporary increase in shearing resistance. Subse-quent dissipation of negative pore-water pressureaccounts for the “relaxation effect” sometimesobserved after piles have been driven into dense,fine sands.

Saturated, cohesionless soils subject to cyclicloads exhibit a significant reduction in strength ifcyclic loading is applied at periods smaller than thetime required to achieve significant dissipation ofpore pressure. Should the number of load cycles Ncbe sufficient to generate pore pressures thatapproach the confining pressure within a soilzone, excessive deformations and eventually fail-ure (liquefaction) is induced. For a given confiningpressure and cyclic stress level, the number ofcycles required to induce initial liquefaction Nc1increases with an increase in relative density Dr.Cyclic shear strength is commonly investigated bycyclic triaxial tests and occasionally by cyclic,direct, simple-shear tests.

7.5.3 State of Stress of Soils

Assessment of the vertical σ′νo and horizontal σ′hoeffective stresses within a soil deposit and the max-imum effective stresses imposed on the depositsince deposition σ′νm is a general requirement forcharacterization of soil behavior. The ratio σ′νm /σ′νo istermed the overconsolidation ratio (OCR). Anotheruseful parameter is the ratio of σ′ho /σ′νo, which iscalled the coefficient of earth pressure at rest (Ko).

For a simple gravitation piezometric profile, theeffective overburden stress σ′νo is directly related tothe depth of groundwater below the surface andthe effective unit weight of the soil strata. Ground-water conditions, however, may be characterizedby irregular piezometric profiles that cannot bemodeled by a simple gravitational system. Forthese conditions, sealed piezometer measure-

ments are required to assess σ′νo.

Maximum Past Consolidation Stress n

The maximum past consolidation stress σ′νm of asoil deposit may reflect stresses imposed prior togeologic erosion or during periods of significantlylower groundwater, as well as desiccation effectsand effects of human activity (excavations). Themaximum past consolidation stress is convention-ally interpreted from consolidation (oedometer)tests on undisturbed samples.


Normalized-shear-strength concepts providean alternate method for estimating OCR fromgood-quality UU compression tests. In the absenceof site-specific data relating cu /σ′νo and OCR, a formof Eq. (7.5) may be applied to estimate OCR. In thisinterpretation, σ′νo represents the effective overbur-den pressure at the depth of the UU-test sample. Avery approximate estimate of σ′νm can also beobtained for cohesive soils from relationships pro-posed between liquidity index and effective verti-cal stress (“Design Manual—Soil Mechanics, Foun-dations, and Earth Structures,” NAVDOCKSDM-7, U.S. Navy). For coarse-grained soil deposits,it is difficult to characterize σ′νm reliably from eitherin situ or laboratory tests because of an extremesensitivity to disturbance.

The coefficient of earth pressure at rest Ko canbe determined in the laboratory from “no-lateralstrain” TC tests on undisturbed soil samples orfrom consolidation tests conducted in speciallyconstructed oedometers. Interpretation of Ko fromin situ CPT, PMT, and dilatometer tests has alsobeen proposed. In view of the significant impact ofsample disturbance on laboratory results and theempirical nature of in situ test interpretations, thefollowing correlations of Ko with friction angle φ′and OCR are useful. For both coarse- and fine-grained soils:

(7.8)

A value for m of 0.5 has been proposed for over-consolidated cohesionless soils, whereas for cohe-sive soils it is proposed that m be estimated interms of the plasticity index Ip as 0.581Ip

-0.12.

7.5.4 Deformability of Fine-GrainedSoils

Deformations of fine-grained soils can be classifiedas those that result from volume change, (elastic) dis-tortion without volume change, or a combination ofthese causes. Volume change may be a one-dimen-sional or, in the presence of imposed shear stresses, athree-dimensional mechanism and may occurimmediately or be time-dependent. Immediate

deformations are realized without volume changeduring undrained loading of saturated soils and as areduction of air voids (volume change) within unsat-urated soils.



The rate of volume change of saturated, fine-grained soils during loading or unloading is con-trolled by the rate of pore-fluid drainage from orinto the stressed soil zone. The compression phaseof delayed volume change associated with pore-pressure dissipation under a constant load istermed primary consolidation. Upon completionof primary consolidation, some soils (particularlythose with a significant organic content) continueto decrease in volume at a decreasing rate. Thisresponse is usually approximated as a straight linefor a plot of log time vs. compression and is termedsecondary compression.

As the imposed shear stresses become a sub-stantial fraction of the undrained shear strength ofthe soil, time-dependent deformations may occurunder constant load and volume conditions. Thisphenomenon is termed creep deformation. Failureby creep may occur if safety factors are insufficientto maintain imposed shear stresses below thecreep threshold of the soil. (Also see Art. 7.10.)

Fig. 7.2 Typical curves plotted from


One-dimensional volume-change parameters

are conveniently interpreted from consolidation(oedometer) tests. A typical curve for log consoli-dation pressure vs. volumetric strain εν (Fig 7.2)demonstrates interpretation of the strain-refer-enced compression index C′c , recompression indexC′r , and swelling index C′s. The secondary com-pression index C′α represents the slope of the near-linear portion of the volumetric strain vs. log-timecurve following primary consolidation (Fig. 7.2b).The parameters C′c , C′r , and C′α be roughly esti-mated from soil-index properties.

Deformation moduli representing three-dimensional deformation can be interpreted fromthe stress-strain curves of laboratory shear tests forapplication to either volume change or elasticdeformation problems.

(“Design Manual—Soil Mechanics, Foundations,and Earth Structures,” NAVDOCKS DM-7, U.S.Navy; T. W. Lambe and R. V. Whitman, “SoilMechanics,” John Wiley & Sons, Inc., New York.)

data obtained in consolidation tests.



7.5.5 Deformability of Coarse-Grained Soils

Deformation of most coarse-grained soils occursalmost exclusively by volume change at a rateessentially equivalent to the rate of stress change.Deformation moduli are markedly nonlinear withrespect to stress change and dependent on the ini-tial state of soil stress. Some coarse-grained soilsexhibit a delayed volume-change phenomenonknown as friction lag. This response is analogousto the secondary compression of fine-grained soilsand can account for a significant amount of thecompression of coarse-grained soils composed ofweak or sharp-grained particles.

The laboratory approach previously describedfor derivation of drained deformation parametersfor fine-grained soils has a limited application forcoarse-grained soils because of the difficulty inobtaining reasonably undisturbed samples. Testsmay be carried out on reinstituted samples butshould be used with caution since the soil fabric,aging, and stress history cannot be adequately sim-ulated in the laboratory. As a consequence, in situtesting techniques currently appear to be a morepromising approach to the characterization ofcohesionless soil properties (see Art. 7.6.3) and aregaining increasing acceptance.

The Quasi-Static Cone PenetrationTest (CPT) n The CPT is one of the most useful insitu tests for investigating the deformability ofcohesionless soils. The secant modulus E′s , tons/ft2,of sands has been related to cone resistance qc bycorrelations of small-scale plate load tests and loadtests on footings. The relationship is given by Eq.(7.9a). The empirical correlation coefficient α in Eq.(7.9a) is influenced by the relative density, graincharacteristics, and stress history of the soil (seeArt. 7.6.3).

(7.9a)

The α parameter has been reported to rangebetween 1.5 and 3 for sands and can be expressedin terms of relative density Dr as 2(1 + Dr

2 ). α mayalso be derived from correlations between qc andstandard penetration resistance N by assumingthat qc /N for mechanical (Delft) cones or qc /N + 1for electronic (Fugro)-type cone tips is about 6 forsandy gravel, 5 for gravelly sand, 4 for clean sand,


and 3 for sandy silt. Bear in mind, however, that E′scharacterizations from qc or N are strictly empiricaland can provide erroneous characterizationsunder some circumstances. (See also Art. 7.13)

Load-Bearing Test n One of the earliestmethods for evaluating the in situ deformability ofcoarse-grained soils is the small-scale load-bearingtest. Data developed from these tests have beenused to provide a scaling factor to express the set-tlement ρ of a full-size footing from the settlementρ1 of a 1-ft2 plate. This factor ρ/ρ1 is given as a func-tion of the width B of the full-size bearing plate as:

(7.10)

From an elastic half-space solution, E′s can beexpressed from results of a plate load test in terms ofthe ratio of bearing pressure to plate settlement kν as:

(7.9b)

µ represents Poisson's ratio, usually considered torange between 0.30 and 0.40. Equation (7.9b)assumes that ρ1 is derived from a rigid, 1-ft-diametercircular plate and that B is the equivalent diameter ofthe bearing area of a full-scale footing. Empirical for-mulations such as Eq. (7.10) may be significantly inerror because of the limited footing-size range usedand the large scatter of the data base. Furthermore,consideration is not given to variations in the char-acteristics and stress history of the bearing soils.

Pressuremeter tests (PMTs) in soils and softrocks have been used to characterize E′s from radi-al pressure vs. volume-change data developed byexpanding a cylindrical probe in a drill hole (seeArt. 7.6.3). Because cohesionless soils are sensitiveto comparatively small degrees of soil disturbance,proper access-hole preparation is critical.

(K. Terzaghi and R. B. Peck, “Soil Mechanicsand Engineering Practice,” John Wiley & Sons,Inc., New York; H. Y. Fang, “Foundation Engineer-ing Handbook,” 2nd ed., Van Nostrand Reinhold,New York.)

7.5.6 California Bearing Ratio (CBR)

This ratio is often used as a measure of the qualityor strength of a soil that will underlie a pavement,



for determining the thickness of the pavement, itsbase, and other layers.

(7.11)

where F = force per unit area required to pene-trate a soil mass with a 3-in2 circularpiston (about 2 in in diameter) at therate of 0.05 in/min

F0 = force per unit area required for cor-responding penetration of a stan-dard material

Typically, the ratio is determined at 0.10 in pene-tration, although other penetrations sometimes areused. An excellent base course has a CBR of 100%.A compacted soil may have a CBR of 50%, where-as a weaker soil may have a CBR of 10.

Tests to determine CBR may be performed inthe laboratory or the field. ASTM standard tests areavailable for each case: “Standard Test Method forCBR (California Bearing Ratio) for LaboratoryCompacted Soils,” D1883, and “Standard TestMethod for CBR (California Bearing Ratio) of Soilsin Place,” D4429.

One criticism of the method is that it does notsimulate the shearing forces that develop in sup-porting materials underlying a flexible pavement.

7.5.7 Soil Permeability

The coefficient of permeability k is a measure of therate of flow of water through saturated soil undera given hydraulic gradient i, cm/ cm, and is definedin accordance with Darcy’s law as:

(7.12)

where V = rate of flow, cm3/s.

A = cross-sectional area of soil convey-ing flow, cm2

k is dependent on the grain-size distribution, voidratio, and soil fabric and typically may vary from asmuch as 10 cm/s for gravel to less than 10-7 cm/s forclays. For typical soil deposits, k for horizontal flowis greater than k for vertical flow, often by an orderof magnitude.

Soil-permeability measurements can be con-ducted in tests under falling or constant head, eitherin the laboratory or the field. Large-scale pumping


(drawdown) tests also may be conducted in thefield to provide a significantly larger scale measure-ment of formation permeability. Correlations of kwith soil gradation and relative density or void ratiohave been developed for a variety of coarse-grainedmaterials. General correlations of k with soil indexand physical properties are less reliable for fine-grained soils because other factors than porositymay control.

(T. W. Lambe and R. V. Whitman, “Soil Mechan-ics,” John Wiley & Sons, Inc., New York.)

7.6 Site InvestigationsThe objective of most geotechnical site investiga-tions is to obtain information on the site and sub-surface conditions that is required for design andconstruction of engineered facilities and for evalu-ation and mitigation of geologic hazards, such aslandslides, subsidence, and liquefaction. The siteinvestigation is part of a fully integrated processthat includes:

1. Synthesis of available data

2. Field and laboratory investigations

3. Characterization of site stratigraphy and soilproperties

4. Engineering analyses

5. Formulation of design and construction criteriaor engineering evaluations

7.6.1 Planning and Scope

In the planning stage of a site investigation, allpertinent topographical, geologic, and geotechni-cal information available should be reviewed andassessed. In urban areas, the development historyof the site should be studied and evaluated. It isparticularly important to provide or require that aqualified engineer direct and witness all fieldoperations.

The scope of the geotechnical site investigationvaries with the type of project but typicallyincludes topographic and location surveys,exploratory borings, and groundwater monitoring.Frequently, the borings are supplemented bysoundings and test pits. Occasionally, air photointerpretations, in situ testing, and geophysicalinvestigations are conducted.



7.6.2 Exploratory Borings

Typical boring methods employed for geotechnicalexploration consist of rotary drilling, augerdrilling, percussion drilling, or any combination ofthese. Deep soil borings (greater than about 100 ft)are usually conducted by rotary-drilling tech-niques recirculating a weighted drilling fluid tomaintain borehole stability. Auger drilling, withhollow-stem augers to facilitate sampling, is awidely used and economical method for conduct-ing short- to intermediate-length borings. Most ofthe drill rigs are truck-mounted and have a rock-coring capability.

With percussion drilling, a casing is usuallydriven to advance the boring. Water circulation ordriven, clean-out spoons are often used to removethe soil (cuttings) in the casing. This method isemployed for difficult-access locations where rel-atively light and portable drilling equipment isrequired. A rotary drill designed for rock coring isoften included.

Soil Samples n These are usually obtained bydriving a split-barrel sampler or by hydraulically ormechanically advancing a thin-wall (Shelby) tubesampler. Driven samplers, usually 2 in outsidediameter (OD), are advanced 18 in by a 140-lb ham-mer dropped 30 in (ASTM D1586). The number ofblows required to drive the last 12 in of penetrationconstitutes the standard penetration resistance

(SPT) value. The Shelby tube sampler, used forundisturbed sampling, is typically a 12- to 16-gageseamless steel tube and is nominally 3 in OD (ASTMD1587). In soils that are soft or otherwise difficult tosample, a stationary piston sampler is used toadvance a Shelby tube either hydraulically (pumppressure) or by the down-crowd system of the drill.

Rotary core drilling is typically used to obtaincore samples of rock and hard, cohesive soils thatcannot be penetrated by conventional samplingtechniques. Typically, rock cores are obtained withdiamond bits that yield core-sample diametersfrom 7/8 in (AX) to 21/8 in (NX). For hard clays andsoft rocks, a 3- to 6-in OD undisturbed sample canalso be obtained by rotary drilling with a Denni-

son or Pitcher sampler.

Test Boring Records (Logs) n These typi-cally identify the depths and material classificationof the various strata encountered, the sample loca-


tion and penetration resistance, rock-core intervaland recovery, groundwater levels encounteredduring and after drilling. Special subsurface condi-tions should be noted on the log, for example,changes in drilling resistance, hole caving, voids,and obstructions. General information requiredincludes the location of the boring, surface eleva-tion, drilling procedures, sampler and core barreltypes, and other information relevant to interpre-tation of the boring log.

Groundwater Monitoring n Monitoringgroundwater levels is an integral part of boringand sampling operations. Groundwater measure-ments during and at least 12 h after drilling areusually required. Standpipes are often installed intest borings to provide longer-term observations;they are typically small-diameter pipes perforatedin the bottom few feet of casing.

If irregular piezometric profiles are suspected,piezometers may be set and sealed so as to measurehydrostatic heads within selected strata. Piezome-ters may consist of watertight ½- to ¾-in OD stand-pipes or plastic tubing attached to porous ceramic orplastic tips. Piezometers with electronic or pneumat-ic pressure sensors have the advantage of quickresponse and automated data acquisition. However,it is not possible to conduct in situ permeability testswith these closed-system piezometers.

7.6.3 In Situ Testing Soils

In situ tests can be used under a variety of circum-stances to enhance profile definition, to providedata on soil properties, and to obtain parametersfor empirical analysis and design applications.

Quasi-static and dynamic cone penetration

tests (CPTs) quite effectively enhance profile defi-nition by providing a continuous record of pene-tration resistance. Quasi-static cone penetrationresistance is also correlated with the relative densi-ty, OCR, friction angle, and compressibility ofcoarse-grained soils and the undrained shearstrength of cohesive soils. Empirical foundationdesign parameters are also provided by the CPT.

The standard CPT in the United States consistsof advancing a 10-cm2, 60° cone at a rate between1.5 and 2.5 cm/s and recording the resistance tocone penetration (ASTM D3441). A friction sleevemay also be incorporated to measure frictionalresistance during penetration. The cone may be



incrementally (mechanical penetrometer) or con-tinuously (electronic penetrometer) advanced.

Dynamic cones are available in a variety of sizes,but in the United States, they typically have a 2-inupset diameter with a 60° apex. They are driven byblows of a 140-lb hammer dropped 30 in. Auto-matically driven cone penetrometers are widelyused in western Europe and are portable and easyto operate. The self-contained ram-soundingmethod has the broadest application and hasbeen standardized.

Pressuremeter tests (PMTs) provide an in situinterpretation of soil compressibility and undrainedshear strength. Pressuremeters have also been usedto provide parameters for foundation design.

The PMT is conducted by inserting a probe con-taining an expandable membrane into a drill holeand then applying a hydraulic pressure to radiallyexpand the membrane against the soil, to measureits volume change under pressure. The resultingcurve for volume change vs. pressure is the basisfor interpretation of soil properties.

Vane shear tests provide in situ measurementsof the undrained shear strength of soft to firm clays,usually by rotating a four-bladed vane and measur-ing the torsional resistance T. Undrained shearstrength is then calculated by dividing T by thecylindrical side and end areas inscribing the vane.Account must be taken of torque rod friction (ifunsleeved), which can be determined by calibrationtests (ASTM D2573). Vane tests are typically run inconjunction with borings, but in soft clays the vanemay be advanced without a predrilled hole.

Other in situ tests occasionally used to providesoil-property data include plate load tests (PLTs),borehole shear (BHS) tests, and dilatometer tests.The PLT technique may be useful for providingdata on the compressibility of soils and rocks. TheBHS may be useful for characterizing effective-shear-strength parameters for relatively free-drain-ing soils as well as total-stress (undrained) shear-strength parameters for fine-grained soils.Dilatometer tests provide a technique for investi-gating the horizontal effective stress σ′ho and soilcompressibility. Some tests use small-diameterprobes to measure pore-pressure response,acoustical emissions, bulk density, and moisturecontent during penetration.

Prototype load testing as part of the geotechni-cal investigation represents a variation of in situtesting. It may include pile load tests, earth load


tests to investigate settlement and stability, andtests on small-scale or full-size shallow foundationelements. Feasibility of construction can also beevaluated at this time by test excavations, indicatorpile driving, drilled shaft excavation, rock rippabil-ity trials, dewatering tests, and so on.


7.6.4 Geophysical Investigations

Geophysical measurements are often valuable inevaluation of continuity of soil and rock stratabetween boring locations. Under some circum-stances, such data can reduce the number of bor-ings required. Certain of these measurements canalso provide data for interpreting soil and rockproperties. The techniques often used in engineer-ing applications are as follows:

Seismic-wave-propagation techniques includeseismic refraction, seismic rejection, and directwave-transmission measurements. Refractiontechniques measure the travel time of seismicwaves generated from a single-pulse energysource to detectors (geophones) located at variousdistances from the source. The principle of seismicrefraction surveying is based on refraction of theseismic waves at boundaries of layers with differ-ent acoustical impedances. This technique is illus-trated in Fig. 7.3.

Compression P wave velocities are interpretedto define velocity profiles that may be correlatedwith stratigraphy and the depth to rock. The P

Fig. 7.3 Illustration of seismic refraction concept.



wave velocity may also help identify type of soil.However, in saturated soils the velocity measuredrepresents wave transmission through water-filledvoids. This velocity is about 4800 ft/s regardless ofthe soil type. Low-cost single- and dual-channelseismographs are available for routine engineeringapplications.

Seismic reflection involves measuring thetimes required for a seismic wave induced at thesurface to return to the surface after reflectionfrom the interfaces of strata that have differentacoustical impedances. Unlike refraction tech-niques, which usually record only the first arrivalsof the seismic waves, wave trains are concurrentlyrecorded by several detectors at different positionsso as to provide a pictorial representation of for-mation structure. This type of survey can be con-ducted in both marine and terrestrial environ-ments and usually incorporates comparativelyexpensive multiple-channel recording systems.

Direct seismic-wave-transmission techniques

include measurements of the arrival times of Pwaves and shear S waves after they have traveledbetween a seismic source and geophones placed atsimilar elevations in adjacent drill holes. By mea-suring the precise distances between source anddetectors, both S and P wave velocities can be mea-sured for a given soil or rock interval if the holespacing is chosen to ensure a direct wave-trans-mission path.

Alternatively, geophones can be placed at dif-ferent depths in a drill hole to measure seismicwaves propagated down from a surface sourcenear the drill hole. The detectors and source loca-tions can also be reversed to provide up-holeinstead of down-hole wave propagation. Althoughthis method does not provide as precise a measureof interval velocity as the cross-hole technique, it issubstantially less costly.

Direct wave-transmission techniques are usual-ly conducted so as to maximize S wave energygeneration and recognition by polarization of theenergy input. S wave interpretations allow calcula-tion of the small-strain shear modulus Gmaxrequired for dynamic response analysis. Poisson’sratio can also be determined if both P and S wavevelocities can be recorded.

Resistivity and conductance investigationtechniques relate to the proposition that strati-graphic details can be derived from differences inthe electrical resistance or conductivity of individ-


ual strata. Resistivity techniques for engineeringpurposes usually apply the Wenner method ofinvestigation, which involves four equally spacedsteel electrodes (pins). The current is introducedthrough the two end pins, and the associatedpotential drop is measured across the two centerpins. The apparent resistivity ρ is then calculatedas a function of current I, potential difference V,and pin spacing a as:

(7.13)

To investigate stratigraphic changes, tests are run atsuccessively greater pin spacings. Interpretationsare made by analyzing accumulative or discrete-interval resistivity profiles or by theoretical curve-matching procedures.

A conductivity technique for identifying sub-surface anomalies and stratigraphy involves mea-suring the transient decay of a magnetic field withthe source (dipole transmitter) in contact with thesurface. The depth of apparent conductivity mea-surement depends on the spacing and orientationof the transmitter and receiver loops.

Both resistivity and conductivity interpreta-tions are influenced by groundwater chemistry.This characteristic has been utilized to map theextent of some groundwater pollutant plumes byconductivity techniques.

Other geophysical methods with more limitedengineering applications include gravity and mag-netic field measurements. Surveys using these tech-niques can be airborne, shipborne, or ground-based.Microgravity surveys have been useful in detectingsubsurface solution features in carbonate rocks.

Aerial surveys are appropriate where large areasare to be explored. Analyses of conventional aerialstereoscopic photographs; thermal and false-color,infrared imagery; multispectral satellite imagery; orside-looking aerial radar can disclose the surfacetopography and drainage, linear features that reflectgeologic structure, type of surface soil and often thetype of underlying rock. These techniques are par-ticularly useful in locating filled-in sinkholes in karstregions, which are often characterized by closelyspaced, slight surface depressions.

(M. B. Dobrin, “Introduction to GeophysicalProspecting,” McGraw-Hill Book Company, NewYork.)



7.7 Hazardous Site andFoundation Conditions

There are a variety of natural hazards of potentialconcern in site development and foundationdesign. Frequently, these hazards are overlookedor not given proper attention, particularly in areaswhere associated failures have been infrequent.

7.7.1 Solution-Prone Formations

Significant areas in the eastern and midwesternUnited States are underlain by formations (carbon-ate and evaporate rocks) susceptible to dissolution.Subsurface voids created by dissolution rangefrom open jointing to huge caverns. These featureshave caused catastrophic failures and detrimentalsettlements of structures as a result of ground lossor surface subsidence.

Special investigations designed to identifyrock-solution hazards include geologic reconnais-sance, air photo interpretation, and geophysical(resistivity, microgravity, and so on) surveys. Tomitigate these hazards, careful attention should begiven to:

1. Site drainage to minimize infiltration of surfacewaters near structures

2. Limitation of excavations to maximize the thick-ness of soil overburden

3. Continuous foundation systems designed toaccommodate a partial loss of support beneaththe foundation system

4. Deep foundations socketed into rock anddesigned solely for socket bond resistance

It is prudent to conduct special proof testing ofthe bearing materials during construction in solu-tion-prone formations. Proof testing often consistsof soundings continuously recording the penetra-tion resistance through the overburden and therate of percussion drilling in the rock. Suspectzones are thus identified and can be improved byexcavation and replacement or by in situ grouting.

7.7.2 Expansive Soils

Soils with a medium to high potential for causingstructural damage on expansion or shrinkage arefound primarily throughout the Great Plains and


Gulf Coastal Plain Physiographic Provinces. Heaveor settlement of active soils occurs because of a changein soil moisture in response to climatic changes, con-struction conditions, changes in surface cover, andother conditions that influence the groundwater andevapotransportation regimes. Differential founda-tion movements are brought about by differentialmoisture changes in the bearing soils. Figure 7.4 pre-sents a method for classifying the volume-changepotential of clay as a function of activity.

Investigations in areas containing potentiallyexpansive soils typically include laboratory swelltests. Infrequently, soil suction measurements aremade to provide quantitative evaluations of vol-ume-change potential. Special attention during thefield investigation should be given to evaluation ofthe groundwater regime and to delineation of thedepth of active moisture changes.

Common design procedures for preventingstructural damage include mitigation of moisturechanges, removal or modification of expansivematerial and deep foundation support. Horizontaland vertical moisture barriers have been utilized tominimize moisture losses due to evaporation orinfiltration and to cut off subsurface groundwaterflow into the area of construction. Excavation ofpotentially active materials and replacement withinert material or with excavated soil modified bythe addition of lime have proved feasible whereexcavation quantities are not excessive.

Fig. 7.4 Chart for rating volume-change poten-tial of expansive soils.



Deep foundations (typically drilled shafts) havebeen used to bypass the active zone and to resist orminimize uplift forces that may develop on theshaft. Associated grade beams are usually con-structed to prevent development of uplift forces.

(“Engineering and Design of Foundations onExpansive Soils,” U.S. Department of the Army, 1981.

L. D. Johnson, “Predicting Potential Heave andHeave with Time and Swelling Foundation Soils,”Technical Report S-78-7, U.S. Army EngineersWaterways Experiment Station, Vicksburg, Miss.,1978.)

7.7.3 Landslide Hazards

Landslides are usually associated with areas of sig-nificant topographic relief that are characterizedby relatively weak sedimentary rocks (shales, silt-stones, and so forth) or by relatively impervioussoil deposits containing interbedded water-bearingstrata. Under these circumstances, slides that haveoccurred in the geologic past, whether or not cur-rently active, represent a significant risk for hillsidesite development. In general hillside developmentin potential landslide areas is a most hazardousundertaking. If there are alternatives, one of thoseshould be adopted.

Detailed geologic studies are required to evalu-ate slide potential and should emphasize detectionof old slide areas. Procedures that tend to stabilizean active slide or to provide for the continued sta-bility of an old slide zone include:

1. Excavation at the head of the sliding mass toreduce the driving force

2. Subsurface drainage to depress piezometric lev-els along potential sliding surface

3. Buttressing at the toe of the potential slidingmass to provide a force-resisting slide movement

Within the realm of economic feasibility, the relia-bility of these or any other procedures to stabilizeactive or old slides involving a significant slidingmass are generally of a relatively low order.

On hillsides where prior slides have not beenidentified, care should be taken to reduce the slid-ing potential of superimposed fills by removingweak or potentially unstable surficial materials,benching and keying the fill into competent materi-als, and (most importantly) installing effective sub-


surface drainage systems. Excavations that result insteepening of existing slopes are potentially detri-mental and should not be employed. Direction andcollection of surface runoff so as to prevent slopeerosion and infiltration are recommended.

7.7.4 Liquefaction of Soils

Relatively loose saturated cohesionless soils maybecome unstable under cyclic shear loading suchas that imposed by earthquake motions. A simpli-fied method of analysis of the liquefaction potentialof cohesionless soils has been proposed for pre-dicting the ratio of the horizontal shear stress τaν tothe effective overburden pressure σ′νo imposed byan earthquake. (τaν represents a uniform cyclic-stress representation of the irregular time historyof shear stress induced by the design earthquake.)This field stress ratio is a function of the maximumhorizontal ground-surface acceleration amax, theacceleration of gravity g, a stress-reduction factorrd , and total vertical stress σνo and approximated as

(7.14)

rd varies from 1.0 at the ground surface to 0.9 for adepth of 30 ft. (H. B. Seed and I. M. Idriss, “A Sim-plified Procedure for Evaluating Soil LiquefactionPotential,” Report EERC 70-9, Earthquake Engi-neering Research Center, University of California,Berkeley, 1970.)

Stress ratios that produce liquefaction may becharacterized from correlations with field observa-tions (Fig. 7.5). The relevant soil properties are rep-resented by their corrected penetration resistance

(7.15)

where σ′νo is in units of tons/ft2. The stress ratio caus-ing liquefaction should be increased about 25% forearthquakes with Richter magnitude 6 or lower. (H.B. Seed, “Evaluation of Soil Liquefaction Effects onLevel Ground during Earthquakes,” Symposium onLiquefaction Problems and Geotechnical Engineering,ASCE National Convention, Philadelphia, Pa., 1976.)

Significantly, more elaborate, dynamic finite-element procedures have been proposed to evalu-ate soil liquefaction and degradation of undrainedshear strength as well as generation and dissipa-tion of pore-water pressure in soils as a result of



Fig. 7.5 Chart correlates cyclic-stress ratios that produce soil liquefaction with standard penetration resis-tance. (After H. B. Seed.)

cyclic loading. Since stress increases accompanydissipation of pore-water pressures, settlementsdue to cyclic loading can also be predicted. Suchresidual settlements can be important even thoughliquefaction has not been induced.

(P. B. Schnabel, J. Lysmer, and H. B. Seed, “AComputer Program for Earthquake ResponseAnalysis of Horizontally Layered Sites,” ReportEERC 72-12, Earthquake Engineering ResearchCenter, University of California, Berkeley, 1972; H.B. Seed, P. P. Martin, and J. Lysmer, “Pore-WaterPressure Changes During Soil Liquefaction,” ASCEJournal of Geotechnical Engineering Division, vol. 102,no. GT4, 1975; K. L. Lee and A. Albaisa, “Earth-quake-Induced Settlements in Saturated Sands,”ASCE Journal of Geotechnical Engineering Division,vol. 100, no. GT4, 1974.)

Shallow FoundationsShallow foundation systems can be classified asspread footings, wall and continuous (strip) foot-ings, and mat (raft) foundations. Variations arecombined footings, cantilevered (strapped) foot-ings, two-way strip (grid) footings, and discontin-uous (punched) mat foundations.


7.8 Types of FootingsSpread (individual) footings (Fig. 7.6) are the mosteconomical shallow foundation types but are moresusceptible to differential settlement. They usuallysupport single concentrated loads, such as thoseimposed by columns.

Fig. 7.6 Spread footing.



Combined footings (Fig. 7.7) are used wherethe bearing areas of closely spaced columns over-lap. Cantilever footings (Fig. 7.8) are designed toaccommodate eccentric loads.

Continuous wall and strip footings (Fig. 7.9) canbe designed to redistribute bearing-stress concen-trations and associated differential settlements inthe event of variable bearing conditions or local-ized ground loss beneath footings.

Mat foundations have the greatest facility forload distribution and for redistribution of sub-grade stress concentrations caused by localizedanomalous bearing conditions. Mats may be con-stant section, ribbed, waffled, or arched. Buoyancymats are used on compressible soil sites in combi-nation with basements or subbasements, to createa permanent unloading effect, thereby reducingthe net stress change in the foundation soils.

(M. J. Tomlinson, “Foundation Design and Con-struction,” John Wiley & Sons, Inc., New York; J. E.Bowles, “Foundation Analysis and Design,”McGraw-Hill Book Company, New York.)

7.9 Approach to FoundationAnalysis

Shallow-foundation analysis and formulation ofgeotechnical design provisions are generallyapproached in the following steps:

l. Establish project objectives and design or eval-uation conditions.

2. Characterize site stratigraphy and soil properties.

Fig. 7.7 Combined footing.


3. Evaluate load-bearing fill support or subsoil-improvement techniques, if applicable.

4. Identify bearing levels; select and proportioncandidate foundation systems.

5. Conduct performance, constructibility, and eco-nomic feasibility analyses.

6. Repeat steps 3 through 5 as required to satisfythe design objectives and conditions.

The scope and detail of the analyses vary accord-ing to the project objectives.

Project objectives to be quantified are essential-ly the intent of the project assignment and the spe-cific scope of associated work. The conditions thatcontrol geotechnical evaluation or design workinclude criteria for loads and grades, facility oper-ating requirements and tolerances, constructionschedules, and economic and environmental con-straints. Failure to provide a clear definition of rel-evant objectives and design conditions can resultin significant delays, extra costs, and, under somecircumstances, unsafe designs.

During development of design conditions forstructural foundations, tolerances for total and dif-ferential settlements are commonly established asa function of the ability of a structure to toleratemovement. Suggested structure tolerances interms of angular distortion are in Table 7.5. Angu-lar distortion represents the differential verticalmovement between two points divided by the hor-izontal distance between the points.

Fig. 7.8 Cantilever footing.



Fig. 7.9 Continuous footings for (a) a wall, (b) several columns.

AngularStructural Response Distortion

Cracking of panel and brick walls 1/100

Structural damage to columns andbeams 1/150

Impaired overhead crane operation 1/300

First cracking of panel walls 1/300

Limit for reinforced concrete frame 1/400

Limit for wall cracking 1/500

Limit for diagonally braced frames 1/600

Limit for settlement-sensitive machines 1/750

*Limits represent the maximum distortions that can be safelyaccommodated.

Source: After L. Bjerrum, European Conference on Soil Mechanicsand Foundation Engineering, Wiesbaden, Germany, vol. 2, 1963.

Table 7.5 Limiting Angular Distortions*


Development of design profiles for foundationanalysis ideally involves a synthesis of geologicand geotechnical data relevant to site stratigraphyand soil and rock properties. This usually requiressite investigations (see Arts. 7.6.1 to 7.6.4) and insitu or laboratory testing, or both, of representativesoil and rock samples (see Arts. 7.3 to 7.5.6).

To establish and proportion candidate founda-tion systems, consideration must first be given toidentification of feasible bearing levels. The depth ofthe foundation must also be sufficient to protectexposed elements against frost heave and to pro-vide sufficient confinement to produce a factor ofsafety not less than 2.5 (preferably 3.0) againstshear failure of the bearing soils. Frost penetrationhas been correlated with a freezing index, whichequals the number of days with temperaturebelow 32° F multiplied by T – 32, where T = aver-age daily temperature. Such correlations can beapplied in the absence of local codes or experience.Generally, footing depths below final grade shouldbe a minimum of 2.0 to 2.5 ft.

For marginal bearing conditions, considerationshould be given to improvement of the quality ofpotential bearing strata. Soil-improvement tech-



niques include excavation and replacement oroverlaying of unsuitable subsoils by load-bearingfills, preloading of compressible subsoils, soil densifi-cation, and grout injection. Densification methodsinclude high-energy surface impact (dynamic con-solidation), on-grade vibratory compaction, andsubsurface vibratory compaction by Vibroflotationor Terra-Probe techniques.

Another approach to improvement of bearingconditions is to incorporate reinforcement. Associ-ated techniques are stone columns, lime columns, geo-fabric reinforcing, and reinforced earth. The choice ofan appropriate soil-improvement technique ishighly dependent on the settlement tolerance ofthe structure as well as the magnitude and natureof the applied loads.

Assessment of the suitability of candidate foun-dation systems requires evaluation of the safety fac-tor against both catastrophic failure and excessivedeformation under sustained and transient designloads. Catastrophic-failure assessment must consid-er overstress and creep of the bearing soils as well aslateral displacement of the foundation. Evaluationof the probable settlement behavior requires analy-sis of the stresses imposed within the soil and, withthe use of appropriate soil parameters, prediction offoundation settlements. Typically, settlement analysesprovide estimates of total and differential settle-ment at strategic locations within the foundationarea and may include time-rate predictions of set-

Fig. 7.10 Bearing capacity factors for use in Eq.


tlement. Usually, the suitability of shallow founda-tion systems is governed by the systems’ load-set-tlement response rather than bearing capacity.

7.10 Foundation-StabilityAnalysis

The maximum load that can be sustained by shal-low foundation elements at incipient failure (bear-ing capacity) is a function of the cohesion and fric-tion angle of bearing soils as well as the width Band shape of the foundation. The net bearing

capacity per unit area qu of a long footing is con-ventionally expressed as:

(7.16)

where αf = 1.0 for strip footings and 1.3 for cir-cular and square footings

cu = undrained shear strength of soil

σ′νo = effective vertical shear stress in soilat level of bottom of footing

βf = 0.5 for strip footings, 0.4 for squarefootings, and 0.6 for circular footings

γ = unit weight of soil

B = width of footing for square and rec-tangular footings and radius of foot-ing for circular footings

(7.16) as determined by Terzaghi and Meyerhof.



Nc , Nq , Nγ = bearing-capacity factors, functions ofangle of internal friction φ (Fig. 7.10)

For undrained (rapid ) loading of cohesive soils,φ = 0 and Eq. (7.16) reduces to

(7.17)

where N′c = αf Nc. For drained (slow) loading ofcohesive soils, φ and cu are defined in terms ofeffective friction angle φ′ and effective stress c′u.

Modifications of Eq. (7.16) are also available topredict the bearing capacity of layered soil and foreccentric loading.

Rarely, however, does qu control foundationdesign when the safety factor is within the range of2.5 to 3. (Should creep or local yield be induced,excessive settlements may occur. This considera-tion is particularly important when selecting asafety factor for foundations on soft to firm clayswith medium to high plasticity.)

Equation (7.16) is based on an infinitely longstrip footing and should be corrected for othershapes. Correction factors by which the bearing-capacity factors should be multiplied are given inTable 7.6, in which L = footing length.

The derivation of Eq. (7.16) presumes the soilsto be homogeneous throughout the stressed zone,which is seldom the case. Consequently, adjust-ments may be required for departures from homo-geneity. In sands, if there is a moderate variation instrength, it is safe to use Eq. (7.16), but with bear-ing-capacity factors representing a weighted aver-age strength.

For strongly varied soil profiles or interlayeredsands and clays, the bearing capacity of each layer

Nc

Rectangle† 1 + (B / L) (Nq /Nc)

Circle and square 1 + (Nq /Nc)

*After E. E. De Beer, as modified by A. S. Vesic. See H. Y. Fang, “FouNew York.

† No correction factor is needed for long strip foundations.

Table 7.6 Shape Corrections for Bearing-Capacity

Shape ofFoundation


should be determined. This should be done byassuming the foundation bears on each layer suc-cessively but at the contact pressure for the depthbelow the bottom of the foundation of the top ofthe layer.

Eccentric loading can have a significant impacton selection of the bearing value for foundationdesign. The conventional approach is to propor-tion the foundation to maintain the resultant forcewithin its middle third. The footing is assumed tobe rigid and the bearing pressure is assumed tovary linearly as shown by Fig. 7.11b. If the resultantlies outside the middle third of the footing, it isassumed that there is bearing over only a portionof the footing, as shown in Fig. 7.11d. For the con-ventional case, the maximum and minimum bear-ing pressures are:

(7.18)

where B = width of rectangular footing

L = length of rectangular footing

e = eccentricity of loading

For the other case (Fig. 7.11c), the soil pressureranges from 0 to a maximum of:

(7.19)

For square or rectangular footings subject tooverturning about two principal axes and forunsymmetrical footings, the loading eccentricitiese1 and e2 are determined about the two principal

Correction Factor

Nq Nγ

1 + (B / L) tan φ 1 – 0.4(B / L)

1 + tan φ 0.60

ndation Engineering Handbook,” Van Nostrand Reinhold, 2d ed.,

Factors of Shallow Foundations*



Fig. 7.11 Footings subjected to overturning.

axes. For the case where the full bearing area of thefootings is engaged, qm is given in terms of the dis-tances from the principal axes c1 and c2 , the radiusof gyration of the footing area about the principalaxes r1 and r2 , and the area of the footing A as:

(7.20)

For the case where only a portion of the footing isbearing, the maximum pressure may be approxi-mated by trial and error.

For all cases of sustained eccentric loading, themaximum (edge) pressures should not exceed theshear strength of the soil and also the factor of safe-ty should be at least 1.5 (preferably 2.0) againstoverturning.

The foregoing analyses, except for completelyrigid foundation elements, are a very conservativeapproximation. Because most mat foundations andlarge footings are not completely rigid, their defor-mation during eccentric loading acts to produce amore uniform distribution of bearing pressuresthan would occur under a rigid footing and toreduce maximum contact stresses.


In the event of transient eccentric loading, experi-ence has shown that footings can sustain maxi-mum edge pressures significantly greater than theshear strength of the soil. Consequently, somebuilding codes conservatively allow increases insustained-load bearing values of 30% for transientloads. Reduced safety factors have also been usedin conjunction with transient loading. For caseswhere significant cost savings can be realized,finite-element analyses that model soil-structureinteraction can provide a more realistic evaluationof an eccentrically loaded foundation.

Allowable Bearing Pressures n Approx-imate allowable soil bearing pressures, without tests,for various soil and rocks are given in Table 7.7 fornormal conditions. These basic bearing pressuresmay be increased when the base of the footing isembedded beyond normal depth. Rock values maybe increased by 10% for each foot of embedmentbeyond 4 ft in fully confined conditions, but the val-ues may not exceed twice the basic values.

In any case, bearing pressures should be limitedto values such that the proposed construction will besafe against failure of the soil under 100% overload.



Resistance to Horizontal Forces n Thehorizontal resistance of shallow foundations ismobilized by a combination of the passive soilresistance on the vertical projection of theembedded foundation and the friction betweenthe foundation base and the subgrade. The soilpressure mobilized at full passive resistance,however, requires lateral movements greaterthan can be sustained by some foundations. Con-sequently, a soil resistance between the at-restand passive-pressure cases should be determinedon the basis of the allowable lateral deformationsof the foundation.

The frictional resistance f to horizontal transla-tion is conventionally estimated as a function ofthe sustained, real, load-bearing stresses qd from

Soil Material Pressure,

Unweathered sound rock 6

Medium rock 4

Intermediate rock 2

Weathered, seamy, or porous rock 2 to

Hardpan 1

Hardpan

Gravel soils 1

Gravel soils

Gravel soils

Gravel soils

Sand soils 3 to

Fine sand 2 to

Clay soils

Clay soils

Silt soils

Silt soils 1

Compacted fills

Fills and soft soils 2 to

Table 7.7 Allowable Bearing Pressures for Soils


(7.21)

where δ is the friction angle between the founda-tion and bearing soils. δ may be taken as equiva-lent to the internal-friction angle φ′ of the sub-grade soils. In the case of cohesive soils, f = cu.Again, some relative movement must be realizedto develop f, but this movement is less than thatrequired for passive-pressure development.

If a factor of safety against translation of at least1.5 is not realized with friction and passive soilpressure, footings should be keyed to increase soilresistance or tied to engage additional resistance.Building basement and shear walls are also com-monly used to sustain horizontal loading.

tons/ft2 Notes

0 No adverse seam structure

0

0

8

2 Well cemented

8 Poorly cemented

0 Compact, well graded

8 Compact with more than 10% gravel

6 Loose, poorly graded

4 Loose, mostly sand

6 Dense

4 Dense

5 Hard

2 Medium stiff

3 Dense1/2 Medium dense

Compacted to 90% to 95% of maxi-mum density (ASTM D1557)

4 By field or laboratory test only



7.11 Stress Distribution UnderFootings

Stress changes imposed in bearing soils by earthand foundation loads or by excavations are con-ventionally predicted from elastic half-space theo-ry as a function of the foundation shape and theposition of the desired stress profile. Elastic solu-tions available may take into account foundationrigidity, depth of the compressible zone, superpo-sition of stress from adjacent loads, layered pro-files, and moduli that increase linearly with depth.

For most applications, stresses may be comput-ed by the pressure-bulb concept with the methodsof either Boussinesq or Westergaard. For thickdeposits, use the Boussinesq distribution shown inFig. 7.12a; for thinly stratified soils, use the Wester-gaard approach shown in Fig. 7.12b. These charts

Fig. 7.12 Stress distribution under a square footinwidth B, as determined by equations of (a) Boussinesq


indicate the stresses q beneath a single foundationunit that applies a pressure at its base of qo.

Most facilities, however, involve not only multi-ple foundation units of different sizes, but alsofloor slabs, perhaps fills, and other elements thatcontribute to the induced stresses. The stressesused for settlement calculation should include theoverlapping and contributory stresses that mayarise from these multiple loads.

7.12 Settlement Analyses ofCohesive Soils

Settlement of foundations supported on cohesivesoils is usually represented as the sum of the pri-mary one-dimensional consolidation ρc , immedi-ate ρi , and secondary ρs settlement components.

g with side B and under a continuous footing with and (b) Westergaard.



Settlement due to primary consolidation is con-ventionally predicted for n soil layers by Eq. (7.22)and (7.23). For normally consolidated soils,

(7.22)

where Hi = depth below grade of ith soil layer

C′c = strain referenced compression indexfor ith soil layer (Art. 7.5.4)

σν = sum of average σ′νο and averageimposed vertical-stress change ∆σν

in ith soil layerσ′νο = initial effective overburden pressure

at middle of ith layer (Art 7.5.3)

For overconsolidated soils with σν > σ′νm ,

(7.23)

where C′r = strain referenced recompressionindex of ith soil layer (Art. 7.5.4)

σ′νm= preconsolidation (maximum pastconsolidation) pressure at middle ofith layer (Art. 7.5.3)

The maximum thickness of the compressible soilzone contributing significant settlement can betaken to be equivalent to the depth where ∆σν =0.1σ′νo.

Equation (7.22) can also be applied to overcon-solidated soils if σν is less than σ′νm and C′r is substi-tuted for C′c.

Inasmuch as Eqs. (7.22) and (7.23) representone-dimensional compression, they may providerather poor predictions for cases of three-dimen-sional loading. Consequently, corrections to ρchave been derived for cases of three-dimensionalloading. These corrections are approximate butrepresent an improved approach when loadingconditions deviate significantly from the one-dimensional case. (A. W. Skempton and L. Bjerrum,“A Contribution to Settlement Analysis of Founda-tions on Clay,” Geotechnique, vol. 7, 1957.)

The stress-path method of settlement analysisattempts to simulate field loading conditions byconducting triaxial tests so as to track the sequen-tial stress changes of an average point or pointsbeneath the foundation. The strains associatedwith each drained and undrained load increment


are summed and directly applied to the settlementcalculation. Deformation moduli can also bederived from stress-path tests and used in three-dimensional deformation analysis.

Three-dimensional settlement analyses usingelastic solutions have been applied to both drainedand undrained conditions. Immediate (elastic) foun-dation settlements ρi , representing the undraineddeformation of saturated cohesive soils, can be cal-culated by discrete analysis [Eq. (7.25)].

(7.24)

where σ1 – σ3 = change in average deviator stresswithin each layer influenced by applied load. Notethat Eq. (7.24) is strictly applicable only for axisym-metrical loading. Drained three-dimensional defor-mation can be estimated from Eq. (7.24) by substi-tuting the secant modulus E′s for E (see Art. 7.5.5).

The rate of one-dimensional consolidation canbe evaluated with Eq. (7.26) in terms of the degreeof consolidation U and the nondimensional timefactor Tν. U is defined by

(7.25)

where ρt = settlement at time t after instanta-neous loading

ρc = ultimate consolidation settlement

ut = excess pore-water pressure at time t

ui = initial pore-water pressure (t = 0)

To correct approximately for the assumed instanta-neous load application, ρt at the end of the loadingperiod can be taken as the settlement calculated forone-half of the load application time. The time trequired to achieve U is evaluated as a function ofthe shortest drainage path within the compressiblezone h, the coefficient of consolidation Cν , and thedimensionless time factor Tν from

(7.26)

Closed-form solutions Tν vs. U are available for avariety of initial pore-pressure distributions. (H. Y.Fang, “Foundation Engineering Handbook,” 2nded., Van Nostrand Reinhold Company, New York.)



Solutions for constant and linearly increasing uiare shown in Fig. 7.13. Equation (7.27) presents anapproximate solution that can be applied to theconstant initial ui distribution case for Tν > 0.2.

(7.27)

where e = 2.71828. Numerical solutions for any uiconfiguration in a single compressible layer as wellas solutions for contiguous clay layers may bederived with finite-difference techniques.

(R. F. Scott, “Principles of Soil Mechanics,” Addison-Wesley Publishing Company, Inc., Reading, Mass.)

The coefficient of consolidation Cν is usuallyderived from conventional consolidation tests byfitting the curve for time vs. deformation (for anappropriate load increment) to the theoreticalsolution for constant ui. For tests of samplesdrained at top and bottom, Cν may be interpretedfrom the curve for log time or square root of timevs. strain (or dial reading) as

(7.28)

Fig. 7.13 Curves relate degree of


where H = height of sample, in

t = time for 90% consolidation (√–t curve)or 50% consolidation (log t curve),days

Tν = 0.197 for 90% consolidation or 0.848for 50% consolidation

(See T. W. Lambe and R. V. Whitman, “SoilMechanics,” John Wiley & Sons, Inc., New York, forcurve-fitting procedures.) Larger values of Cν areusually obtained with the √–t method and appearto be more representative of field conditions.

Secondary compression settlement ρs isassumed, for simplicity, to begin on completion ofprimary consolidation, at time t100 correspondingto 100% primary consolidation. ρs is then calculat-ed from Eq. (7.29) for a given period t after t100.

(7.29)

Hi represents the thickness of compressible layersand Cα is the coefficient of secondary compressiongiven in terms of volumetric strain (Art. 7.5.4).

consolidation and time factor Tν.



The ratio Cα to compression index Cc is nearlyconstant for a given soil type and is generally with-in the range of 0.045 + 0.015. Cα , as determinedfrom consolidation tests (Fig. 7.2), is extremely sen-sitive to pressure-increment ratios of less thanabout 0.5 (standard is 1.0). The effect of overcon-solidation, either from natural or construction pre-load sources, is to significantly reduce Cα. This is animportant consideration in the application of pre-loading for soil improvement.

The rate of consolidation due to radial drainage isimportant for the design of vertical sand or wickdrains. As a rule, drains are installed in compress-ible soils to reduce the time required for consolida-tion and to accelerate the associated gain in soilstrength. Vertical drains are typically used in con-junction with preloading as a means of improvingthe supporting ability and stability of the subsoils.

(S. J. Johnson, “Precompression for ImprovingFoundation Soils,” ASCE Journal of Soil Mechanicsand Foundation Engineering Division, vol. 96, no.SM1, 1970; R. D. Holtz and W. D. Kovacs, “An Intro-duction to Geotechnical Engineering,” Prentice-Hall, Inc., Englewood Cliffs, N.J.)

7.13 Settlement Analysis ofSands

The methods most frequently used to estimate thesettlement of foundations supported by relativelyfree-draining cohesionless soils generally employempirical correlations between field observationsand in situ tests. The primary correlative tests areplate bearing (PLT), cone penetration resistance(CPT), and standard penetration resistance (SPT)(see Art 7.6.3). These methods, however, are devel-oped from data bases that contain a number ofvariables not considered in the correlations and,therefore, should be applied with caution.

Plate Bearing Tests n The most commonapproach is to scale the results of PLTs to full-sizefootings in accordance with Eq. (7.10). A less con-servative modification of this equation proposedby A. R. S. S. Barazaa is

(7.30)


where B = footing width, ft

ρ = settlement of full-size bearing plate

ρl = settlement of 1-ft square bearingplate

These equations are not sensitive to the relativedensity, gradation, and OCR of the soil or to theeffects of depth and shape of the footing.

Use of large-scale load tests or, ideally, full-scaleload tests mitigates many of the difficulties of thepreceding approach but is often precluded by costsand schedule considerations. Unless relatively uni-form soil deposits are encountered, this approachalso requires a number of tests, significantlyincreasing the cost and time requirements. (See J. K.Mitchell and W. S. Gardner, “In-Situ Measurementof Volume-Change Characteristics,” ASCE SpecialtyConference on In-Situ Measurement of Soil Proper-ties, Raleigh, N.C., 1975.)

Cone Penetrometer Methods n Correla-tions between quasi-static penetration resistance qcand observation of the settlement of bearing platesand small footings form the basis of foundationsettlement estimates using CPT data. The Buis-man-DeBeer method utilizes a one-dimensionalcompression formulation. A recommended modi-fication of this approach that considers the influ-ence of the relative density of the soil Dr andincreased secant modulus E′s is

(7.31)

where ρ = estimated footing settlement. The σ′νoand ∆σν parameters represent the average effectiveoverburden pressure and vertical stress change foreach layer considered below the base of the foun-dation (see Art 7.12). Equation (7.31) has limitationsbecause no consideration is given to: (1) soil stresshistory, (2) soil gradation, and (3) three-dimension-al compression. Also, Eq. (7.31) incorporates anempirical representation of E′s , given by Eq. (7.32),and has all the limitations thereof (see Art. 7.5.5).

(7.32)

The foregoing procedures are not applicable tolarge footings and foundation mats. From fieldobservations relating foundation width B, meters,



to ρ/B, the upper limit for ρ/B, for B > 13.5 m, isgiven, in percent, approximately by

(7.33)

For the same data base, the best fit of the averageρ/B measurements ranges from about 0.09% (B =20 m) to 0.06% (B = 80 m).

Standard Penetration ResistanceMethods n A variety of methods have been pro-posed to relate foundation settlement to standardpenetration resistance N. An approach proposedby I. Alpan and G. G. Meyerhof appears reasonableand has the advantage of simplicity. Settlement S,in, is computed for B < 4 ft from

(7.34a)

and for B > 4 ft from

(7.34b)

where q = bearing capacity of soil, tons/ft2

B = footing width, ft

Fig. 7.14 Curves relate relative density to standard


N′ is given approximately by Eq. (7.34c) for σ′νo < 40psi.

(7.34c)

and represents N (blows per foot) normalized forσ′νo = 40 psi (see Fig. 7.14).

(G.G Meyerhof, “Shallow Foundations,” ASCEJournal of Soil Mechanics and Foundation EngineeringDivision, vol. 91, no. SM92, 1965; W. G. Holtz and H.J. Gibbs, “Shear Strength of Pervious GravellySoils,” Proceedings ASCE, paper 867, 1956; R. B. Peck,W. E. Hanson, and T. H. Thornburn, “FoundationEngineering,” John Wiley & Sons, Inc., New York.)

Laboratory Test Methods n The limitationsin developing representative deformation parame-ters from reconstituted samples were described inArt. 7.5.5. A possible exception may be for the set-tlement analyses of foundations supported bycompacted fill. Under these circumstances, consol-idation tests and stress-path, triaxial shear tests onthe fill materials may be appropriate for providingthe parameters for application of settlement analy-ses described for cohesive soils.

(D. J. D’Appolonia, E. D’Appolonia, and R. F.Brisette, “Settlement of Spread Footings on Sand,”ASCE Journal of Soil Mechanics and Foundation Engi-neering Division, vol. 94, no. SM3, 1968.)

penetration resistance and effective vertical stress.



Deep FoundationsSubsurface conditions, structural requirements,site location and features, and economics general-ly dictate the type of foundation to be employedfor a given structure.

Deep foundations, such as piles, drilled shafts,and caissons, should be considered when:

Shallow foundations are inadequate and structur-al loads need to be transmitted to deeper, morecompetent soil or rock

Loads exert uplift or lateral forces on the foundations

Structures are required to be supported over water

Functionality of the structure does not allow fordifferential settlements

Future adjacent excavations are expected

7.14 Application of PilesPile foundations are commonly installed forbridges, buildings, towers, tanks, and offshorestructures. Piles are of two major types: prefabricat-ed and installed with a pile-driving hammer, or cast-in-place. In some cases, a pile may incorporate bothprefabricated and cast-in-place elements. Drivenpiles may be made of wood, concrete, steel, or acombination of these materials. Cast-in-place pilesare made of concrete that is placed into an auger-drilled hole in the ground. When the diameter of adrilled or augured cast-in-place pile exceeds about24 in, it is then generally classified as a drilled shaft,bored pile, or caisson (Arts. 7.15.2, 7.21, and 7.22).

The load-carrying capacity and behavior of asingle pile is governed by the lesser of the struc-tural strength of the pile shaft and the strength anddeformation properties of the supporting soils.When the latter governs, piles derive their capaci-ty from soil resistance along their shaft and undertheir toe. The contribution of each of these twocomponents is largely dependent on subsurfaceconditions and pile type, shape, and method ofinstallation. Piles in sand or clay deposits withshaft resistance predominant are commonlyknown as friction piles. Piles with toe resistanceprimary are known as end-bearing piles. In reali-ty, however, most piles have both shaft and toeresistance, albeit to varying degrees. The sum ofthe ultimate resistance values of both shaft and toeis termed the pile capacity, which when divided by


an appropriate safety factor yields the allowable

load at the pile head.The capacity of a laterally loaded pile is usually

defined in terms of a limiting lateral deflection ofthe pile head. The ratio of the ultimate lateral loaddefining structural or soil failure to the associatedlateral design load represents the safety factor ofthe pile under lateral load.

Piles are rarely utilized singly, but are typicallyinstalled in groups. The behavior of a pile in agroup differs from that of the single pile. Often,the group effect dictates the overall behavior of thepile foundation system.

The following articles provide a general knowl-edge of pile design, analysis, construction, and test-ing methods. For major projects, it is advisable thatthe expertise of a geotechnical engineer with sub-stantial experience with deep-foundations design,construction, and verification methods be employed.

7.15 Pile TypesPiles that cause a significant displacement of soilduring installation are termed displacement piles.

For example, closed-end steel pipes and precastconcrete piles are displacement piles, whereasopen-end pipes and H piles are generally limited-

displacement piles. They may plug during drivingand cause significant soil displacement. Auger-castpiles are generally considered nondisplacement

piles since the soil is removed and replaced withconcrete during pile installation.

Piles are usually classified according to theirmethod of installation and type of material. Pre-formed driven piles may be made of concrete,steel, timber, or a combination of these materials.

7.15.1 Precast Concrete Piles

Reinforced or prestressed to resist handling and pile-driving stresses, precast concrete piles are usuallyconstructed in a casting yard and transported to thejobsite. Pretensioned piles (commonly known as pre-

stressed piles) are formed in very long casting beds,with dividers inserted to produce individual pile sec-tions. Precast piles come in a variety of cross sections;for example, square, round, octagonal. They may bemanufactured full length or in sections that arespliced during installation. They are suitable for use asfriction piles for driving in sand or clay or as end-bear-ing piles for driving through soft soils to firm strata.



Prestressed concrete piles usually have solidsections between 10 and 30 in square. Frequently,piles larger than 24 in square and more than 100 ftlong are cast with a hollow core to reduce pileweight and facilitate handling.

Splicing of precast concrete piles should gener-ally be avoided. When it is necessary to extend pilelength, however, any of several splicing methodsmay be used. Splicing can be accomplished, forinstance, by installing dowel bars of sufficientlength and then injecting grout or epoxy to bondthem and the upper and lower pile sections. Over-size grouted sleeves may also be used. Alternativesto these bonding processes include welding ofsteel plates or pipes cast at pile ends. Some special-ized systems employ mechanical jointing tech-niques using pins to make the connection. Thesemechanical splices reduce field splice time, but theconnector must be incorporated in the pile sectionsat the time of casting.

All of the preceding methods transfer some ten-sion through the splice. There are, however, sys-tems, usually involving external sleeves (or cans),that do not transfer tensile forces; this is a possibleadvantage for long piles in which tension stresseswould not be high, but these systems are notapplicable to piles subject to uplift loading. Forprestressed piles, since the tendons require bond-development length, the jointed ends of the pilesections should also be reinforced with steel bars totransfer the tensile forces across the spliced area.

Prestressed piles also may be posttensioned.Such piles are mostly cylindrical (typically up to66-in diameter and 6-in wall thickness) and arecentrifugally cast in sections and assembled toform the required length before driving. Stressingis achieved with the pile sections placed end to endby threading steel cables through precast ductsand then applying tension to the cables withhydraulic devices. Piles up to 200 fit long havebeen thus assembled and driven.

Advantages of precast concrete piles includetheir ability to carry high axial and inclined loadsand to resist large bending moments. Also, concretepiles can be used as structural columns whenextended above ground level. Disadvantagesinclude the extra care required during handling andinstallation, difficulties in extending and cutting offpiles to required lengths, and possible transportationdifficulties. Special machines, however, are availablefor pile cutting, such as saws and hydraulic crushing


systems. Care, however, is necessary during allstages of pile casting, handling, transportation, andinstallation to avoid damaging the piles.

Precast concrete piles are generally installedwith pile-driving hammers. For this purpose, pileheads should always be protected with cushioningmaterial. Usually, sheets of plywood are used.Other precautions should also be taken to protectpiles during and after driving. When driving isexpected to be through hard soil layers or intorock, pile toes should generally be fitted with steelshoes for reinforcement and protection from dam-age. When piles are driven into soils and ground-water containing destructive chemicals, specialcement additives or coatings should be used toprotect concrete piles. Seawater may also causedamage to concrete piles by chemical reactions ormechanical forces.

(“Recommended Practice for Design, Manufac-ture, and Installation of Prestressed Concrete Pil-ing,” Prestressed Concrete Institute, 175 W. JacksonBlvd., Chicago, IL 60604; “Recommendations forDesign, Manufacture and Installation of ConcretePiles,” American Concrete Institute, P.O. Box 19150,Detroit, MI 48219.)

7.15.2 Cast-in-Place Concrete Piles

These are produced by forming holes in the groundand then filling them with concrete. A steel cagemay be used for reinforcement. There are manymethods for forming the holes, such as driving of aclosed-end steel pipe, with or without a mandrel.Alternatively, holes may be formed with drills orcontinuous-flight augers. Two common methods ofconstruction are (1) a hole is excavated by drillingbefore placement of concrete to form a bored pile,

and (2) a hole is formed with a continuous-flightauger (CFA) and grout is injected into the holeunder pressure through the toe of the hollow augerstem during auger withdrawal. A modification ofthe CFA method is used to create a mixed-in-placeconcrete pile in clean granular sand. There arenumerous other procedures used in constructingcast-in-place concrete piles, most of which are pro-prietary systems.

Advantages of cast-in-place concrete pilesinclude: relatively low cost, fast execution, ease ofadaptation to different lengths, capability for soilsampling during construction at each pile location,possibility of penetrating undesirable hard layers,



high load-carrying capacity of large-size piles, andlow vibration and noise levels during installation.Construction time is less than that needed for pre-cast piles inasmuch as cast-in-place piles can beformed in place to required lengths and withouthaving to wait for curing time before installation.

Pile foundations are normally employed wheresubsurface conditions are likely to be unfavorablefor spread footings or mats. If cast-in-place con-crete piles are used, such conditions may createconcerns about the structural integrity, bearingcapacity, and general performance of the pile foun-dation. The reason for this is that the constructedshape and structural integrity of such piles dependon subsurface conditions, concrete quality andmethod of placement, quality of work, and designand construction practices, all of which requiretight control. Structural deficiencies may resultfrom degraded or debonded concrete, necking, orinclusions or voids. Unlike pile driving, where theinstallation process itself constitutes a crude quali-tative pile-capacity test and hammer-pile-soilbehavior may be evaluated from measurementsmade during driving, methods for evaluating cast-in-place piles during construction are generallynot available. Good installation procedures andinspection are critical to the success of uncasedaugured or drilled piles.

(“Drilled Shafts: Construction Procedures andDesign Methods,” Federal Highway Administration;various publications of The International Associationof Foundation Drilling (ADSC), P.O. Box 280379, Dal-las, TX 75228.)

7.15.3 Steel Piles

Structural steel H and pipe sections are often usedas piles. Pipe piles may be driven open- or closed-end. After being driven, they may be filled withconcrete. Common sizes of pipe piles range from 8to 48 in in diameter. A special type of pipe pile is theMonotube, which has a longitudinally fluted wall,may be of constant section or tapered, and may befilled with concrete after being driven. Closed-endpipes have the advantage that they can be visuallyinspected after driving. Open-end pipes have theadvantage that penetration of hard layers can beassisted by drilling through the open end.

H-piles may be rolled or built-up steel sectionswith wide flanges. Pile toes may be reinforced withspecial shoes for driving through soils with


obstructions, such as boulders, or for driving torock. If splicing is necessary, steel pile lengths maybe connected with complete-penetration welds orcommercially available special fittings. H piles,being low-displacement piles, are advantageous insituations where ground heave and lateral move-ment must be kept to a minimum.

Steel piles have the advantages of being rugged,strong, and easy to handle. They can be driventhrough hard layers. They can carry high compres-sive loads and withstand tensile loading. Because ofthe relative ease of splicing and cutting to length,steel piles are advantageous for use in sites wherethe depth of the bearing layer varies. Disadvantagesof steel piles include small cross-sectional area andsusceptibility to corrosion, which can cause a signif-icant reduction in load-carrying capacity. Measuresthat may be taken when pile corrosion is anticipat-ed include the use of larger pile sections than other-wise needed, use of surface-coating materials, orcathodic protection. In these cases, the pipes areusually encased in or filled with concrete.

Specifications pertaining to steel pipe piles aregiven in “Specification for Welded and SeamlessSteel Pipe Piles,” ASTM A252. For dimensions andsection properties of H piles, see “HP Shapes” in“Manual of Steel Construction,” American Instituteof Steel Construction, 400 N. Michigan Ave., Chica-go, IL 60011.

7.15.4 Timber Piles

Any of a variety of wood species but usually south-ern pine or douglas fir, and occasionally red orwhite oak, can be used as piles. Kept below thegroundwater table, timber piles can serve in a pre-served state for a long time. Untreated piles thatextend above the water table, however, may beexposed to damaging marine organisms anddecay. Such damage may be prevented or delayedand service life prolonged by treating timber pileswith preservatives. Preservative treatment shouldmatch the type of wood.

Timber piles are commonly available in lengthsof up to 75 ft. They should be as straight as possi-ble and should have a relatively uniform taper.Timber piles are usually used to carry light to mod-erate loads or in marine construction as dolphinsand fender systems.

Advantages of timber piles include their rela-tively low cost, high strength-to-weight ratio, and



ease of handling. They can be cut to length afterdriving relatively easily. Their naturally taperedshape (about 1 in in diameter per 10 ft of length) isadvantageous in situations where pile capacitiesderive mostly from shaft resistance. Disadvantagesinclude their susceptibility to damage during harddriving and difficulty in splicing.

Timber piles should be driven with care toavoid damage. Hammers with high impact veloci-ties should not be used. Protective accessoriesshould be utilized, when hard driving is expected,especially at the head and toe of the pile.

Specifications relevant to timber piles are con-tained in “Standard Specifications for Round Tim-ber Piles,” ASTM D25; “Establishing Design Stress-es for Round Timber Piles,” ASTM D2899; and“Preservative Treatment by Pressure Processes,”AWPA C3, American Wood Preservers Association.Information on timber piles also may be obtainedfrom the National Timber Piling Council, Inc., 446Park Ave., Rye, NY 10580.

7.15.5 Composite Piles

This type of pile includes those made of more thanone major material or pile type, such as thick-walled, concrete-filled, steel pipe piles, precastconcrete piles with steel (pipe or H section) exten-sions, and timber piles with cast-in-place concreteextensions.

Fig. 7.15 Approximate ranges of design loadsfor vertical piles in axial compression.

*For shaft diameters not exceeding 18 in.† Primary end bearing.‡ Permanent shells only.§ Uncased only.


7.15.6 Selection of Pile Type

The choice of an appropriate pile type for a partic-ular application is essential for satisfactory founda-tion functioning. Factors that must be consideredin the selection process include subsurface condi-tions, nature and magnitude of loads, local experi-ence, availability of materials and experiencedlabor, applicable codes, and cost. Pile drivability,strength, and serviceability should also be takeninto account. Figure 7.15 presents general guide-lines and approximate ranges of design loads forvertical piles in axial compression. Actual loadsthat can be carried by a given pile in a particularsituation should be assessed in accordance withthe general methods and procedures presented inthe preceding and those described in more special-ized geotechnical engineering books.

7.16 Pile-Driving EquipmentInstallation of piles by driving is a specialized fieldof construction usually performed by experiencedcontractors with dedicated equipment. The basiccomponents of a pile-driving system are shown inFigure 7.16 and described in the following. All

Fig. 7.16 Basic components of a pile-drivingrig. (From “The Performance of Pile Driving Systems—Inspection Manual,” FHWA/RD-86/160, Federal High-way Administration.)



components of the driving system have someeffect on the pile-driving process. The overall sta-bility and capacity of the pile-driving crane shouldbe assessed for all stages of loading conditions,including pile pickup and driving.

Lead n The functions of the lead (also knownas leader or guide) are to guide the hammer, main-tain pile alignment, and preserve axial alignmentbetween the hammer and pile. For proper func-tioning, leads should have sufficient strength andbe straight and well greased to allow free hammertravel.

There are four main types of leads: swinging,fixed, semifixed, and offshore. Depending on therelative positions of the crane and the pile, pile size,and other factors, a specific type of lead may haveto be employed. Swinging leads are the simplest,lightest in weight, and most versatile. They do not,however, provide much fixity for prevention of lat-eral pile movement during driving. Fixed leadsmaintain the position of the pile during driving andfacilitate driving a pile at an inclined angle. How-ever, they are the most expensive type of lead.Semifixed leads have some of the advantages anddisadvantages of the swinging and fixed leads. Off-shore leads are used mostly in offshore construc-tion to drive large-size steel piles and on land ornear shore when a template is used to hold the pilein place. Their use for inclined piles is limited by thepile flexural strength.

Pile Cap (Helmet) n The pile cap (alsoreferred to as helmet) is a boxlike steel elementinserted in the lead between the hammer and pile(Fig. 7.17). The function of the cap is to house bothhammer and pile cushions and maintain axialalignment between hammer and pile. The size ofthe cap needed depends on the pile size and thejaw-opening size of the lead. In some cases, anadaptor is inserted under the cap to accommodatevarious pile sizes, assuring that the hammer andpile are concentrically aligned. A poor seating ofthe pile in the cap can cause pile damage and buck-ling due to localized stresses and eccentric loadingat the pile top.

Cushions n Hammers, except for somehydraulic hammers, include a cushion in the ham-mer (Fig. 7.17). The function of the hammer cush-


ion is to attenuate the hammer impact forces andprotect both the pile and hammer from damagingdriving stresses. Normally, a steel striker plate, typ-ically 3 in thick, is placed on top of the cushion toinsure uniform cushion compression. Most cush-ions are produced by specialized manufacturersand consist of materials such as phenolic or nylonlaminate sheets.

For driving precast concrete piles, a pile cush-ion is also placed at the pile top (Fig. 7.17). Themost common material is plywood. It is placed inlayers with total thickness between 4 and 12 in. Insome cases, hardwood boards may be used (withthe grain perpendicular to the pile axis) as pilecushions. Specifications often require that a freshpile cushion be used at the start of the driving of apile. The wood used should be dry. The pile cush-ion should be changed when significantly com-pressed or when signs of burning are evident.Cushions (hammer and pile) should be durableand reasonably able to maintain their properties.When a change is necessary during driving, thedriving log should record this. The measured resis-tance to driving immediately thereafter should bediscounted, especially if the pile is being drivenclose to its capacity, inasmuch as a fresh cushionwill compress significantly more from a hammerblow than would an already compressed one.Hence, measurements of pile movement per blowwill be different.

Fig. 7.17 Pile helmet and adjoining parts.



With the aid of a computer analytical program,such as one based on the wave equation, it is pos-sible to design a cushion system for a particularhammer and pile that allows maximum energytransfer with minimum risk of pile damage.

Hammer n This provides the energy neededfor pile installation. Basically, an impact pile-drivinghammer consists of a striking part, called the ram,and a means of imposing impacts in rapid succes-sion to the pile.

Hammers are commonly rated by the amountof potential energy per blow. This energy basicallyis the product of ram weight and drop height(stroke). To a contractor, a hammer is a mass-pro-duction machine; hammers with higher efficiencyare generally more productive and can achievehigher pile capacity. To an engineer, a hammer isan instrument that is used to measure the qualityof the end product, the driven pile. Implicitassumptions regarding hammer performance areincluded in common pile evaluation procedures.Hammers with low energy transfer are the sourceof poor installation. Hence, pile designers, con-structors, and inspectors should be familiar withoperating principles and performance characteris-tics of the various hammer types. Following arebrief discussions of the major types of impact pile-driving hammers.

Impact pile-driving hammers rely on a fallingmass to create forces much greater than their weight.Usually, strokes range between 3 and 10 ft.

These hammers are classified by the mode usedin operating the hammer; that is, the means used toraise the ram after impact for a new blow. There aretwo major modes: external combustion and internalcombustion. Hammers of each type may be single ordouble acting. For single-acting hammers (Fig. 7.18),power is only needed to raise the ram, whereas thefall is entirely by gravity. Double-acting hammersalso apply power to assist the ram during down-ward travel. Thus these hammers deliver moreblows per minute than single-acting hammers; how-ever, their efficiency may be lower, since the powersource supplies part of the impact energy.

External-combustion hammers (ECHs) rely ona power source external to the hammer for theiroperation. One type is the drop hammer, which israised by a hoist line from the crane supporting thepile and leads and then dropped to fall under theaction of gravity to impact the pile. Main advan-


tages of drop hammers are relatively low cost andmaintenance and the ability to vary the stroke eas-ily. Disadvantages include reduction of the effec-tiveness of the drop due to the cable and winchassembly required for the operation, slow opera-tion, and hammer efficiency dependence on oper-ator’s skills. (The operator must allow the cable togo slack when the hammer is raised to dropheight.) Consequently, use of drop hammers isgenerally limited to small projects involving light-ly loaded piles or sheetpiles.

For some pile drivers, hydraulic pressure is usedto raise the ram. The hammers, known as air/steamor hydraulic hammers, may be single or double act-ing. Action starts with introduction of the motivefluid (steam, compressed air, or hydraulic fluid)under the piston in the hammer chamber to lift theram. When the ram attains a prescribed height,flow of the motive fluid is discontinued and theram “coasts” against gravity up to the full stroke. Attop of stroke, the pressure is vented and the ram

Fig. 7.18 Single-acting, external-combustionhammer driving a precast concrete pile.



falls under gravity. For double-acting hammers, thepressure is redirected to act on top of the pistonand push the ram downward during its fall. Manyhydraulic hammers are equipped with two strokeheights for more flexibility.

The next cycle starts after impact, and the startshould be carefully controlled. If pressure is intro-duced against the ram too early, it will slow downthe ram excessively and reduce the energy avail-able to the pile. Known as preadmission, this is notdesirable due to its adverse effect on energy trans-fer. For some hammers, the ram, immediately pre-ceding impact, activates a valve to allow themotive fluid to enter the cylinder to start the nextcycle. For most hydraulic hammers, the ram posi-tion is detected by proximity switches and the nextcycle is electronically controlled.

Main advantages of external combustion ham-mers include their higher rate of operation thandrop hammers, long track record of performanceand reliability, and their relatively simple design.Disadvantages include the need to have addition-al equipment on site, such as boilers and compres-sors, that would not be needed with another typeof hammer. Also disadvantageous is their relative-ly high weight, which requires equipment withlarge lifting capacity.

Diesel hammers are internal-combustion ham-mers (ICHs). The power needed for hammer oper-ation comes from fuel combustion inside the ham-mer, therefore eliminating the need for an outsidepower source. Basic components of a diesel ham-mer include the ram, cylinder, impact block, andfuel distribution system. Hammer operation isstarted by lifting the ram with one of the hoist linesfrom the crane or a hydraulic jack to a presetheight. A tripping mechanism then releases theram, allowing it to fall under gravity. During itsdescent, the ram closes cylinder exhaust ports, as aresult of which gases in the combustion chamberare compressed. At some point before impact, theram activates a fuel pump to introduce into thecombustion chamber a prescribed amount of fuelin either liquid or atomized form. The amount offuel depends on the fuel pump setting.

For liquid-injection hammers, the impact of theram on the impact block atomizes the fuel. Underthe high pressure, ignition and combustion result.For atomized-fuel-injection hammers, ignitionoccurs when the pressure reaches a certain thresh-old before impact. The ram impact and the explo-


sive force of the fuel drive the pile into the groundwhile the explosion and pile reaction throw the ramupward past the exhaust ports, exhausting the com-bustion gases and drawing in fresh air for the nextcycle. With an open-end diesel (OED), shown in Fig.7.19, the ram continues to travel upward untilarrested by gravity. Then the next cycle starts. Thedistance that the ram travels upward (stroke)depends on the amount of fuel introduced into thechamber (fuel-pump setting), cushions, pile stiff-ness, and soil resistance. In the case of closed-enddiesel hammers (CEDs), the top of the cylinder isclosed, creating an air-pressure, or bounce, chamber.Upward movement of the ram compresses the air inthe bounce chamber and thus stores energy. Thepressure shortens ram stroke and the stored energyaccelerates the ram downward.

The energy rating of diesel hammers is com-monly appraised by observing the ram stroke (orbounce-chamber pressure for closed-end ham-mers). This is an important indication but can bemisleading; for example, when the hammer

Fig. 7.19 Open-end, single-acting diesel ham-mer driving a pile.



becomes very hot during prolonged driving.Because the fuel then ignites too early, the ramexpends more energy to compress the gases andless energy is available for transmission into thepile. The high pressure still causes a relatively highstroke. This condition is commonly referred to aspreignition. In contrast, short ram strokes may becaused by lack of fuel or improper fuel type, lack ofcompression in the chamber due to worn pistonrings, excessive ram friction, pile stiffness, or lackof soil resistance.

Internal-combustion hammers are advanta-geous because they are entirely self-contained.They are relatively lightweight and thus permituse of smaller cranes than those required for exter-nal-combustion hammers. Also, stroke adjustmentto soil resistance with internal-combustion ham-mers is advantageous in controlling dynamicstresses during driving of concrete piles. Amongdisadvantages is stroke dependence on the ham-mer-pile-soil system, relatively low blow rate, andpotential cessation of operation when easy drivingis encountered.

Table 7.8 presents the characteristics of impactpile-driving hammers. The hammers are listed byrated energy in ascending order. The table indi-cates for each model the type of hammer: ECH,external combustion hammer, or OED, open-enddiesel hammer; manufacturer; model number; ramweight; and equivalent stroke. Note, however, thatnew models of hammers become available at fre-quent intervals.

Vibratory hammers drive or extract piles byapplying rapidly alternating forces to the pile. Theforces are created by eccentric weights (eccentrics)rotating around horizontal axes. The weights areplaced in pairs so that horizontal centrifugal forcescancel each other, leaving only vertical-force com-ponents. These vertical forces shake piles up anddown and cause vertical pile penetration underthe weight of the hammer. The vibration may beeither low frequency (less than 50 Hz) or high fre-quency (more than 100 Hz).

The main parameters that define the character-istics of a vibratory hammer are amplitude pro-duced, power consumption, frequency (vibrationsper minute), and driving force (resultant verticalforce of the rotating eccentrics). Vibratory ham-mers offer the advantages of fast penetration, lim-


ited noise, minimal shock waves induced in theground, and usually high penetration efficiency incohesionless soils. A disadvantage is limited pene-tration capacity under hard driving conditions andin clay soils. Also, there is limited experience in cor-relating pile capacity with driving energy and pen-etration rate. This type of hammer is often used toinstall non-load-bearing piles, such as retainingsheetpiles.

(“Vibratory Pile Driving,” J. D. Smart, Ph.D. the-sis, University of Illinois, Urbana, 1969; variouspublications of the Deep Foundations Institute, 120Charolette Place, Englewood Cliffs, NJ 07632).

Other Pile Driving Accessories n In addi-tion to the basic equipment discussed in the pre-ceding, some pile driving requires employment ofspecial accessories, such as an adaptor, follower,mandrel, auger, or water jet.

An adaptor is inserted between a pile cap andpile head to make it possible for one cap to accom-modate different pile sizes.

A follower is usually a steel member used toextend a pile temporarily in cases where it is nec-essary to drive the pile when the top is belowground level or under water. For efficiency intransmitting hammer energy to the pile, stiffnessof the follower should be nearly equal to that of thepile. The follower should be integrated into thedriving system so that it maintains axial alignmentbetween hammer and pile.

Mandrels are typically used to drive steel shellsor thin-wall pipes that are later filled with con-crete. A mandrel is a uniform or tapered, roundsteel device that is inserted into a hollow pile toserve as a rigid core during pile driving.

Water jets or augers are sometimes needed toadvance a pile tip through some intermediate soillayers. Jet pipes may be integrated into the pileshaft or may be external to the pile. Although pos-sibly advantageous in assisting in pile penetration,jetting may have undesirable effects on pile capac-ity (compression and particularly uplift) thatshould be considered by the engineer.

(Department of Transportation Federal High-way Administration, “The Performance of Pile Dri-ving Systems: Inspection Manual,” FHWA ReportNo. FHWA/RD-86/160, National Technical Infor-mation Service, Springfield, VA 22161.)


Table 7.8 Characteristics of Impact Pile-Driving Hammers

Rated Ram EquivalentEnergy, Hammer Weight, Stroke, Hammerkip-ft Manufacturer Model kips ft Type

1.00 MKT No. 5 0.20 5.00 ECH

2.50 MKT No. 6 0.40 6.25 ECH

4.15 MKT No. 7 0.80 5.19 ECH

7.26 VULCAN VUL 30C 3.00 2.42 ECH

7.26 VULCAN VUL 02 3.00 2.42 ECH

8.10 LINKBELT LB 180 1.73 4.68 CED

8.13 ICE 180 1.73 4.70 CED

8.23 DELMAG D 5 1.10 7.48 OED

8.65 DAWSON HPH 1200 2.29 3.77 ECH

8.75 MKT 9B3 1.60 5.47 ECH

10.50 DELMAG D 6-32 1.32 7.94 OED

13.11 MKT 10B3 3.00 4.37 ECH

14.20 MKT C5-Air 5.00 2.84 ECH

15.00 CONMACO C 50 5.00 3.00 ECH

15.00 RAYMOND R 1 5.00 3.00 ECH

15.00 VULCAN VUL 01 5.00 3.00 ECH

15.02 LINKBELT LB 312 3.86 3.89 CED

15.10 VULCAN VUL 50C 5.00 3.02 ECH

16.00 MKT DE 20 2.00 8.00 OED

16.20 MKT C5-Steam 5.00 3.24 ECH

16.25 MKT S-5 5.00 3.25 ECH

17.32 DAWSON HPH 2400 4.19 4.13 ECH

17.34 BANUT 3 Tonnes 6.61 2.62 ECH

17.60 DELMAG D 8-22 1.76 10.00 OED

18.00 BERMINGH B200 2.00 9.00 OED

18.20 LINKBELT LB 440 4.00 4.55 CED

18.56 ICE 440 4.00 4.64 CED

19.15 MKT 11B3 5.00 3.83 ECH

19.18 VULCAN VUL 65C 6.50 2.95 ECH

19.50 CONMACO C 65 6.50 3.00 ECH

19.50 RAYMOND R 65C 6.50 3.00 ECH

19.50 RAYMOND R 1S 6.50 3.00 ECH

19.50 RAYMOND R 65CH 6.50 3.00 ECH

19.50 VULCAN VUL 06 6.50 3.00 ECH

19.57 VULCAN VUL 65CA 6.50 3.01 ECH

20.00 MKT 20 DE333020 2.00 10.00 OED

21.00 MKT DA 35B 2.80 7.50 CED

21.20 MKT C826 Air 8.00 2.65 ECH

22.13 IHC Hydh SC 30 3.64 6.08 ECH

22.40 MKT DE 30 2.80 8.00 OED

22.50 FEC FEC 1200 2.75 8.18 OED

22.50 ICE 30 S 3.00 7.50 OED

22.99 BERMINGH B23 2.80 8.21 CED

22.99 BERMINGH B23 5 2.80 8.21 CED

23.12 ICE 422 4.00 5.78 CED


23.59 DELMAG D 12 2.75 8.58 OED

23.80 MKT DA35B SA 2.80 8.50 OED

23.80 MKT DE30B 2.80 8.50 OED

24.38 RAYMOND R 0 7.50 3.25 ECH

24.40 MKT C826 Stm 8.00 3.05 ECH

24.48 RAYMOND R 80CH 8.00 3.06 ECH

24.48 RAYMOND R 80C 8.00 3.06 ECH

24.48 VULCAN VUL 80C 8.00 3.06 ECH

24.76 MENCK MHF3-3 7.05 3.51 ECH

24.88 UDDCOMB H3H 6.60 3.77 ECH

28.14 MITSUB. MH 15 3.31 8.50 OED

28.31 DELMAG D 15 3.30 8.58 OED


29.25 BERMINGH B225 3.00 9.75 OED

29.48 IHC Hydh SC 40 5.51 5.35 ECH

30.36 IHC Hydh S 40 5.51 5.51 ECH

30.37 ICE 520 5.07 5.99 CED

30.41 HERA 1500 3.37 9.02 OED

30.72 MKT DA 45 4.00 7.68 CED

30.80 MKT MS-350 7.72 3.99 ECH

30.96 MENCK MHF3-4 8.82 3.51 ECH

31.33 DELMAG D 12-32 2.82 11.11 OED

32.00 MKT DE 40 4.00 8.00 OED

32.50 CONMACO C 100 10.00 3.25 ECH

32.50 CONMACO C565 6.50 5.00 ECH

32.50 MKT S 10 10.00 3.25 ECH

32.50 RAYMOND R 2/0 10.00 3.25 ECH

32.50 VULCAN VUL 506 6.50 5.00 ECH

32.50 VULCAN VUL 010 10.00 3.25 ECH

32.55 FAIRCHLD F-32 10.85 3.00 ECH

32.90 VULCAN VUL 100C 10.00 3.29 ECH

33.00 MKT 33 DE333020 3.30 10.00 OED

33.18 UDDCOMB H4H 8.80 3.77 ECH


35.37 JUNTTAN HHK 4 8.82 4.01 ECH

35.40 BERMINGH B250 5 3.00 11.80 OED

35.98 VULCAN VUL 140C 14.00 2.57 ECH

37.38 CONMACO C 115 11.50 3.25 ECH

37.52 MKT S 14 14.00 2.68 ECH

37.72 ICE 110-SH 11.50 3.28 ECH

38.20 MKT DA 55B 5.00 7.64 CED

38.69 MENCK MHF3-5 11.02 3.51 ECH

38.69 MENCK MHF5-5 11.02 3.51 ECH

39.00 VULCAN VUL 012 12.00 3.25 ECH

39.25 DELMAG D 16-32 3.52 11.15 OED

40.00 CONMACO C 80E5 8.00 5.00 ECH

40.00 ICE 40-S 4.00 10.00 OED

40.00 MKT DA55B SA 5.00 8.00 OED

40.00 MKT 40 DE333020 4.00 10.00 OED

40.00 VULCAN VUL 508 8.00 5.00 ECH

40.31 BERMINGH B300 3.75 10.75 OED

40.31 BERMINGH B300 M 3.75 10.75 OED


40.61 DELMAG D 22 4.91 8.27 OED

40.62 ICE 640 6.00 6.77 CED

40.63 RAYMOND R 3/0 12.50 3.25 ECH

41.47 UDDCOMB H5H 11.00 3.77 ECH

42.00 CONMACO C 140 14.00 3.00 ECH

42.00 ICE 42-S 4.09 10.27 OED

42.00 VULCAN VUL 014 14.00 3.00 ECH

42.40 DELMAG D 19-32 4.00 10.60 OED

42.50 MKT DE 50B 5.00 8.50 OED

43.01 MITSUB. M 23 5.06 8.50 OED

43.20 BERMINGH B400 4.8 4.80 9.00 OED

43.37 BSP HH 5 11.02 3.94 ECH



7.41


Table 7.8 Characteristics of Impact Pile-Driving Hammers (Continued)


44.00 HPSI 110 11.00 4.00 ECH

40.31 BERMINGH B300 M 3.75 10.75 OED


40.61 DELMAG D 22 4.91 8.27 OED

40.62 ICE 640 6.00 6.77 CED

40.63 RAYMOND R 3/0 12.50 3.25 ECH

41.47 UDDCOMB H5H 11.00 3.77 ECH

42.00 CONMACO C 140 14.00 3.00 ECH

42.00 ICE 42-S 4.09 10.27 OED

42.00 VULCAN WL 014 14.00 3.00 ECH

42.40 DELMAG D 19-32 4.00 10.60 OED

42.50 MKT DE 50B 5.00 8.50 OED

43.01 MITSUB. M 23 5.06 8.50 OED

43.20 BERMINGH B400 4.8 4.80 9.00 OED

43.37 BSP HH 5 11.02 3.94 ECH

44.00 HPSI 110 11.00 4.00 ECH

44.00 MKT MS 500 11.00 4.00 ECH

44.23 JUNTTAN HHK 5 11.03 4.01 ECH

44.24 IHC Hydh SC60 7.72 5.73 ECH

44.26 IHC Hydh S. 60 13.23 3.35 ECH

45.00 BERMINGH B400 5.0 5.00 9.00 OED

45.00 FAIRCHLD F-45 15.00 3.00 ECH

45.07 MENCK MRBS 500 11.02 4.09 ECH

45.35 KOBE K22-Est 4.85 9.35 OED

46.43 MENCK MHF5-6 13.23 3.51 ECH

46.43 MENCK MHF3-6 13.23 3.51 ECH

46.84 MITSUB. MH 25 5.51 8.50 OED

47.20 BERMINGH B350 5 4.00 11.80 OED

48.50 DELMAG D 22-13 4.85 10.00 OED

48.50 DELMAG D 22-02 4.85 10.00 OED

48.75 CONMACO C 160 16.25 3.00 ECH

48.75 RAYMOND R 4/0 15.00 3.25 ECH

48.75 RAYMOND R 150C 15.00 3.25 ECH

48.75 VULCAN VUL 016 16.25 3.00 ECH

49.18 MENCK MH 68 7.72 6.37 ECH

49.76 UDDCOMB H6H 13.20 3.77 ECH

50.00 CONMACO C 100E5 10.00 5.00 ECH

50.00 FEC FEC 2500 5.50 9.09 OED

50.00 MKT 50 DE70/50B 5.00 10.00 OED

50.00 VULCAN VUL 510 10.00 5.00 ECH

50.20 VULCAN VUL 200C 20.00 2.51 ECH

50.69 HERA 2500 5.62 9.02 OED

51.26 DELMAG D 22-23 4.85 10.57 OED

51.52 KOBE K 25 5.51 9.35 OED

51.63 ICE 660 7.57 6.82 CED

51.63 LINKBELT LB 660 7.57 6.82 CED

51.65 IHC Hydh S70 7.72 6.69 ECH

51.78 CONMACO 160 ** 17.26 3.00 ECH

53.05 JUNTTAN HHK 6 13.23 4.01 ECH

53.75 BERMINGH B400 5.00 10.75 OED

53.75 BERMiNGH B400 M 5.00 10.75 OED

54.17 MENCK MHF3-7 15.43 3.51 ECH

54.17 MENCK MHF5-7 15.43 3.51 ECH

55.99 FEC FEC 2800 6.16 9.09 OED

56.77 HERA 2800 6.29 9.02 OED

56.88 RAYMOND R 5/0 17.50 3.25 ECH

57.50 CONMACO C 115E5 11.50 5.00 ECH

58.00 ICE 60 S 6.96 8.33 OED

58.90 IHC Hydh SC 80 11.24 5.24 ECH

59.00 BERMINGH B400 5 5.00 11.80 OED

59.50 MKT DE 70B 7.00 8.50 OED

59.60 DELMAG D 30 6.60 9.03 OED

60.00 CONMACO C 200 20.00 3.00 ECH

60.00 HPSI 150 15.00 4.00 ECH

60.00 MKT S 20 20.00 3.00 ECH

60.00 VULCAN VUL 320 20.00 3.00 ECH

60.00 VULCAN VUL 512 12.00 5.00 ECH

60.00 VULCAN VUL 020 20.00 3.00 ECH

60.78 BSP HH 7 15.43 3.94 ECH

61.49 DELMAG D 25-32 5.51 11.16 OED

61.71 MITSUB. M 33 7.26 8.50 OED

61.91 JUNTTAN HHK 7 15.44 4.01 ECH

61.91 MENCK MHF5-8 17.64 3.51 ECH

62.50 CONMACO C 125E5 12.50 5.00 ECH

63.03 FEC FEC 3000 6.60 9.55 OED

65.62 MITSUB. MH 35 7.72 8.50 OED

66.00 DELMAG D 30-02 6.60 10.00 OED

66.00 DELMAG D 30-13 6.60 10.00 OED

66.36 IHC Hydh S90 9.92 6.69 ECH

67.77 MENCK MRBS 750 16.53 4.10 ECH

69.34 UDDCOMB H8H 17.60 3.94 ECH

69.43 MENCK MH 96 11.02 6.30 ECH

69.50 BSP HH 8 17.64 3.94 ECH

69.65 MENCK MHF5-9 19.84 3.51 ECH

70.00 ICE 70-S 7.00 10.00 OED

70.00 MKT 70 DE70/50B 7.00 10.00 OED

70.96 HERA 3500 7.87 9.02 OED

72.00 ICE 80-S 8.00 9.00 OED

72.18 KOBE K 35 7.72 9.35 OED

72.60 ICE 1070 10.00 7.26 CED

73.00 FEC FEC 3400 7.48 9.76 OED

73.66 DELMAG D 30-32 6.60 11.16 OED

73.66 DELMAG D 30-23 6.60 11.16 OED

75.00 RAYMOND R 30X 30.00 2.50 ECH

77.39 MENCK MHF5-10 22.04 3.51 ECH

77.42 IHC Hydh SC 110 15.21 5.09 ECH

77.88 BERMINGH B450 5 6.60 11.80 OED

78.17 BSP HH 9 19.84 3.94 ECH

80.00 HPSI 200 20.00 4.00 ECH

80.41 MITSU8. M 43 9.46 8.50 OED

81.25 RAYMOND R 8/0 25.00 3.25 ECH

83.82 DELMAG D 36-13 7.93 10.57 OED

83.82 DELMAG D 36-02 7.93 10.57 OED

83.82 DELMAG D 36 7.93 10.57 OED

85.13 MENCK MHF5-11 24.25 3.51 ECH

85.43 MITSUB. MH 45 10.05 8.50 OED

86.88 UDDCOMB H10H 22.05 3.94 ECH

88.00 BERMINGH B550 C 11.00 8.00 OED

88.42 JUNTTAN HHK 10 22.05 4.01 ECH

88.50 DELMAG D 36-32 7.93 11.16 OED

88.50 DELMAG D 36-23 7.93 11.16 OED



7.42


Table 7.8 Characteristics of Impact Pile-Driving Hammers (Continued)


90.00 HPSI 225 22.50 4.00 ECH

90.00 ICE 90-S 9.00 10.00 OED

90.00 VULCAN VUL 330 30.00 3.00 ECH

90.00 VULCAN VUL 030 30.00 3.00 ECH

90.44 DELMAG D 44 9.50 9.52 OED

92.04 BERMINGH B500 5 7.80 11.80 OED

92.75 KOBE K 45 9.92 9.35 OED

92.87 MENCK MHF5-12 26.45 3.51 ECH

93.28 MENCK MRBS 850 18.96 4.92 ECH

93.28 MENCK MRBS 800 18.98 4.92 ECH

95.54 BSP HH 11 24.25 3.94 ECH

96.53 DELMAG D 46-13 10.14 9.52 OED

100.00 CONMACO C 5200 20.00 5.00 ECH

100.00 ICE 200-S 20.00 5.00 OED

100.00 RAYMOND R 40X 40.00 2.50 ECH

100.00 VULCAN VUL 520 20.00 5.00 ECH

101.37 HERA 5000 11.24 9.02 OED

103.31 IHC Hydh SC 150 24.25 4.26 ECH

104.80 MENCK MH 145 16.53 6.34 ECH

106.10 JUNTTAN HHK 12 26.46 4.01 ECH

106.20 BERMINGH B550 5 9.00 11.80 OED

107.18 DELMAG D 46-02 10.14 10.57 OED

107.18 DELMAG D 46 10.14 10.57 OED

107.18 DELMAG D 46-23 10.14 10.57 OED

110.00 MKT 110 DE110150 11.00 10.00 OED

113.16 DELMAG D 46-32 10.14 11.16 OED

113.60 VULCAN VUL 400C 40.00 2.84 ECH

115.57 HERA 5700 12.81 9.02 OED

116.04 MENCK MHF10-15 33.06 3.51 ECH

120.00 VULCAN VUL 340 40.00 3.00 ECH

120.00 VULCAN VUL 040 40.00 3.00 ECH

121.59 BSP HH 14 30.86 3.94 ECH

123.43 MENCK MRBS110 24.25 5.09 ECH

123.79 JUNTTAN HHK 14 30.87 4.01 ECH

124.53 DELMAG D 55 11.86 10.50 OED

125.70 HERA 6200 13.93 9.02 OED

130.18 KOBE KB 60 13.23 9.84 OED

135.15 MITSUB. MH 72B 15.90 8.50 OED

135.59 MENCK MRBS150 33.07 4.10 ECH

138.87 BSP HH 16 35.27 3.94 ECH

141.12 MENCK MH 195 22.05 6.40 ECH

147.18 IHC Hydh S 200 22.00 6.69 ECH

147.38 MENCK MHUT 200 26.46 5.57 ECH

149.60 MITSUB. MH 80B 17.60 8.50 OED

150.00 CONMACO C 5300 30.00 5.00 ECH

150.00 MKT 150 DE110150 15.00 10.00 OED

150.00 RAYMOND R 60X 60.00 2.50 ECH

150.00 VULCAN VUL 530 30.00 5.00 ECH

151.00 IHC Hydh SC 200 30.20 5.00 ECH

152.06 HERA 7500 16.85 9.02 OED

152.45 DELMAG D 62-02 13.66 11.16 OED

152.45 DELMAG D 62-22 13.66 11.16 OED

152.45 DELMAG D 62-12 13.66 11.16 OED

154.69 MENCK MHF10-20 44.07 3.51 ECH

173.58 BSP HH 20S 44.09 3.94 ECH

173.58 KOBE KB 80 17.64 9.84 OED

176.97 IHC Hydh SC 250 39.24 4.51 ECH

178.42 HERA 8800 19.78 9.02 OED

179.16 VULCAN VUL 600C 60.00 2.99 ECH

180.00 VULCAN VUL 360 60.00 3.00 ECH

180.00 VULCAN VUL 060 60.00 3.00 ECH

184.64 IHC Hydh S 250 27.60 6.69 ECH

186.24 DELMAG D 80-12 17.62 10.57 OED

189.81 MENCK MRBS180 38.58 4.92 ECH

196.64 DELMAG D 80-23 17.62 11.16 OED

200.00 VULCAN VUL 540 40.90 4.89 ECH

206.51 IHC Hydh S 280 29.80 6.93 ECH

225.00 CONMACO C 5450 45.00 5.00 ECH

225.95 MENCK MRBS250 55.11 4.10 ECH

225.95 MENCK MRBS250 55.11 4.10 ECH

245.85 DELMAG D100-13 22.03 11.16 OED

260.37 BSP HA 30 66.13 3.94 ECH

262.11 MENCK MRBS250 63.93 4.10 ECH

289.55 MENCK MHU 400 50.71 5.71 ECH

294.60 IHC Hydh S 400 44.30 6.65 ECH

295.12 MENCK MHUT 400 52.70 5.60 ECH

300.00 VULCAN VUL 3100 100.00 3.00 ECH

300.00 VULCAN VUL 560 62.50 4.80 ECH

325.36 MENCK MRBS300 66.13 4.92 ECH

347.16 BSP HA 40 88.18 3.94 ECH

350.00 CONMACO C 5700 70.00 5.00 ECH

368.30 IHC Hydh S 500 55.30 6.66 ECH

368.55 MENCK MHUT 500 59.54 6.19 ECH

433.64 MENCK MHU 600 77.16 5.62 ECH

498.94 MENCK MRBS460 101.41 4.92 ECH

500.00 VULCAN VUL 5100 100.00 5.00 ECH

510.00 CONMACO C 6850 85.00 6.00 ECH

513.34 MENCK MRBS390 86.86 5.91 ECH

516.13 MENCK MHUT700 92.83 5.56 ECH

542.33 MENCK MRBS500 110.23 4.92 ECH

589.85 IHC Hydh S 800 81.57 7.23 ECH

619.18 MENCK MHUT700 92.83 6.67 ECH

631.40 MENCK MRBS700 154.00 4.10 ECH

736.91 MENCK MHUT100 132.30 5.57 ECH

737.26 IHCHydh S 1000 101.41 7.27 ECH

737.70 MENCK MHU 1000 126.97 5.81 ECH

750.00 VULCAN VUL 5150 150.00 5.00 ECH

759.23 MENCK MRBS600 132.27 5.74 ECH

867.74 MENCK MRBS800 176.37 4.92 ECH

954.53 MENCK MRBS880 194.01 4.92 ECH

1177.80 IHC Hydh S 1600 156.00 7.55 ECH

1228.87 MENCK MHU 1700 207.23 5.93 ECH

1547.59 MENCK MHU 2100 255.80 6.05 ECH

1581.83 MENCK MBS12500 275.58 5.74 ECH

1694.97 IHC Hydh S 2300 226.60 7.48 ECH

1800.00 VULCAN VUL 6300 300.00 6.00 ECH

2171.65 MENCK MHU 3000 363.76 5.97 ECH

2210.12 IHC Hydh S 3000 332.00 6.66 ECH



7.43



7.17 Pile-Design Concepts

Methods for evaluating load-carrying capacity andgeneral behavior of piles in a foundation rangefrom simple empirical to techniques that incorpo-rate state-of-the-art analytical and field verificationmethods. Approaches to pile engineering include(1) precedence, (2) static-load analysis (3) static-load testing, and (4) dynamic-load analytical andtesting methods. Regardless of the method select-ed, the foundation designer should possess fullknowledge of the site subsurface conditions. Thisrequires consultations with a geotechnical engi-neer and possibly a geologist familiar with the localarea to ensure that a sufficient number of boringsand relevant soil and rock tests are performed.

Design by precedent includes application ofbuilding code criteria, relevant published data, per-formance of similar nearby structures, and experi-ence with pile design and construction. Undersome circumstances this approach may be accept-able, but it is not highly recommended. Favorablesituations include those involving minor and tem-porary structures where failure would not result inappreciable loss of property or any loss of life andconstruction sites where long-term experience hasbeen accumulated and documented for a well-defined set of subsurface and loading conditions

Static-load analysis for design and prediction ofpile behavior is widely used by designers who arepractitioners of geotechnical engineering. Thisapproach is based on soil-mechanics principles,geotechnical engineering theories, pile characteris-tics, and assumptions regarding pile-soil interaction.Analysis generally involves evaluations of the load-carrying capacity of a single pile, of pile-groupbehavior, and of foundation settlement under ser-vice conditions. Designs based on this approachalone usually incorporate relatively large factors ofsafety in determination of allowable working loads.Safety factors are based on the engineer’s confi-dence in parameters obtained from soil explorationand their representation of the whole site, anticipat-ed loads, importance of the structure, and thedesigner’s experience and subjective preferences.Static-load analysis methods are used in preliminarydesign to calculate required pile lengths for cost esti-mation and bidding purposes. Final pile design andacceptability are based on additional methods of ver-ification. Pile-group behavior and settlement predic-tions are, however, usually based entirely on static


analyses due to the lack of economical and efficientroutine field verification techniques.

Field testing should be performed on a suffi-cient number of piles to confirm or revise initialdesign assumptions, verify adequacy of installa-tion equipment and procedures, evaluate the effectof subsurface profile variations, and form the basisof final acceptance. Traditionally, piles were testedwith a static-load test (by loading in axial com-pression, uplift, or laterally). The number of suchtests that will be performed on a site is limited,however, due to the expense and time requiredwhen large number of piles are to be installed.

See Art. 7.18 for a description of static-loadanalysis and pile testing.

Dynamic-load pile testing and analysis is oftenused in conjunction with or as an alternative to statictesting. Analytical methods utilizing computers andnumerical modeling and based on one-dimensionalelastic-wave-propagation theories are helpful inselecting driving equipment, assessing pile drivability,estimating pile bearing capacity, and determiningrequired driving criteria, that is, blow count. Dynam-ic analysis is commonly known as wave equation analy-sis of pile driving. Field dynamic testing yields infor-mation on driving-system performance of piles: staticaxial capacity, driving stresses, structural integrity,pile-soil interaction, and load-movement behavior.

See Art. 7.19 for a description of dynamic analy-sis and pile testing.

7.18 Static-Analysis and PileTesting

Static analysis of piles and pile design based on itcommonly employ global factors of safety. Use of theload-and-resistance-factors approach is growing,however. Steps involved in static analysis include cal-culation of the static load-carrying capacity of singlepiles, evaluation of group behavior, and assessmentof foundation settlement. Normally, capacity and set-tlement are treated separately from each other andeither may control the design. Pile drivability is usu-ally treated as a separate item and is not consideredin static-load analysis. See also Art. 7.17.

7.18.1 Axial-Load Capacity ofSingle Piles

Pile capacity Qu may be taken as the sum of theshaft and toe resistances, Qsu and Qbu respectively.



The allowable load Qa may then be determinedfrom either Eq. (7.35) or (7.36)

(7.35)

(7.36)

where F, F1 , F2 , are safety factors. Typically F forpermanent structures is between 2 and 3 but maybe larger, depending on the perceived reliability ofthe analysis and construction as well as the conse-quences of failure. Equation (7.36) recognizes thatthe deformations required to fully mobilize Qsu andQbu are not compatible. For example, Qsu may bedeveloped at displacements less than 0.25 in,whereas Qbu may be realized at a toe displacementequivalent to 5% to 10% of the pile diameter. Con-sequently, F1 may be taken as 1.5 and F2 as 3.0, if theequivalent single safety factor equals F or larger. (IfQsu /Qbu < 1.0, F is less than the 2.0 usually consid-ered as a major safety factor for permanent struc-tures.)

Fig. 7.20 Variation of shear-strength (adhesion) r(After “Recommended Practice for Planning, Designing, aPetroleum Institute, Dallas.)

αα=


7.18.2 Shaft Resistance in CohesiveSoils

The ultimate stress f–s of axially loaded piles in cohe-

sive soils under compressive loads is conventional-ly evaluated from the ultimate frictional resistance

Qsu = As f–s = Asαc–u (7.37)

where cu = average undrained shear strength ofsoil in contact with shaft surface

As = shaft surface area

α = shear-strength (adhesion) reductionfactor

One relationship for selection of α is shown inFig. 7.20. This and similar relationships are empiri-cal and are derived from correlations of load-testdata with the cu of soil samples tested in the labo-ratory. Some engineers suggest that f

–s is influenced

by pile length and that a limiting value of 1 ton/ft2

be set for displacement piles less than 50 ft longand reduced 15% for each 50 ft of additionallength. This suggestion is rejected by other engi-neers on the presumption that it neglects the

eduction factor α with undrained shear strength.nd Constructing Fixed Off-Shore Platforms,” American



effects of pile residual stresses in evaluation of theresults of static-load tests on piles.

The shaft resistance stress f–s for cohesive soils

may be evaluated from effective-stress concepts:

f–s = βσ′νο (7.38)

where σ′νο = effective overburden pressure of soil

β = function of effective friction angle,stress history, length of pile, andamount of soil displacement inducedby pile installation

β usually ranges between 0.22 and 0.35 for inter-mediate-length displacement piles driven in nor-mally consolidated soils, whereas for piles signif-icantly longer than 100 ft, β may be as small as0.15. Derivations of β are given by G. G. Meyer-hof, “Bearing Capacity and Settlement of PileFoundations,” ASCE Journal of Geotechnical Engi-neering Division, vol. 102, no. GT3, 1976; J. B. Bur-land, “Shaft Friction of Piles in Clay,” GroundEngineering, vol. 6, 1973; “Soil Capacity for Sup-porting Deep Foundation Members in Clay,” STP670, ASTM.

Both the α and β methods have been applied inanalysis of n discrete soil layers:

(7.39)

The capacity of friction piles driven in cohesivesoils may be significantly influenced by theelapsed time after pile driving and the rate of loadapplication. The frictional capacity Qs of displace-ment piles driven in cohesive soils increases withtime after driving. For example, the pile capacityafter substantial dissipation of pore pressuresinduced during driving (a typical design assump-tion) may be three times the capacity measuredsoon after driving. This behavior must be consid-ered if piles are to be rapidly loaded shortly afterdriving and when load tests are interpreted.

Some research indicates that frictional capacityfor tensile load Qut may be less than the shaft fric-tion under compression loading Qsu. In the absenceof load-test data, it is therefore appropriate to takeQut , as 0.80Qsu and ignore the weight of the pile.Also, Qut is fully developed at average pile defor-mations of about 0.10 to 0.15 in, about one-halfthose developed in compression. Expanded-base


piles develop additional base resistance and can beused to substantially increase uplift resistances.

(V. A. Sowa, “Cast-In-Situ Bored Piles,” CanadianGeotechnical Journal, vol. 7, 1970; G. G. Meyerhofand J. I. Adams, “The Ultimate Uplift Capacity ofFoundations,” Canadian Geotechnical Journal, vol. 5,no. 4, 1968.)

7.18.3 Shaft Resistance inCohesionless Soils

The shaft resistance stress f–s is a function of the soil-

shaft friction angle δ, deg, and an empirical lateralearth-pressure coefficient K:

f–s = Kσ–′νοtan δ < fl (7.40)

At displacement-pile penetrations of 10 to 20pile diameters (loose to dense sand), the averageskin friction reaches a limiting value fl. Primarilydepending on the relative density and texture ofthe soil, fl has been approximated conservativelyby using Eq. (7.40) to calculate f

–s. This approach

employs the same principles and involves thesame limitations discussed in Art. 7.18.2.

For relatively long piles in sand, K is typicallytaken in the range of 0.7 to 1.0 and δ is taken to beabout φ – 5, where φ′ is the angle of internal fric-tion, deg. For piles less than 50 ft long, K is morelikely to be in the range of 1.0 to 2.0 but can begreater than 3.0 for tapered piles.

Empirical procedures have also been used toevaluate f

–s from in situ tests, such as cone penetra-

tion, standard penetration, and relative densitytests. Equation (7.41), based on standard penetra-tion tests, as proposed by Meyerhof, is generallyconservative and has the advantage of simplicity.

(7.41)

where N—

= average standard penetration resis-tance within the embedded length of pile and f

–s is

given in tons/ft2. (G. G. Meyerhof, “Bearing Capac-ity and Settlement of Pile Foundations,” ASCEJournal of Geotechnical Engineering Division, vol. 102,no. GT3, 1976.)

7.18.4 Toe Capacity Load

For piles installed in cohesive soils, the ultimate tipload may be computed from



(7.42)

where Ab = end-bearing area of pile

q = bearing capacity of soil

Nc = bearing-capacity factor

cu = undrained shear strength of soilwithin zone 1 pile diameter aboveand 2 diameters below pile tip

Although theoretical conditions suggest that Ncmay vary between about 8 and 12, Nc is usuallytaken as 9.

For cohesionless soils, the toe resistance stress qis conventionally expressed by Eq. (7.43) in termsof a bearing-capacity factor Nq and the effectiveoverburden pressure at the pile tip σ′νo.

(7.43)

Some research indicates that, for piles in sands, q,like f

–s , reaches a quasi-constant value ql after pene-

trations of the bearing stratum in the range of 10 to20 pile diameters. Approximately,

(7.44)

where φ is the friction angle of the bearing soilsbelow the critical depth. Values of Nq applicable topiles are given in Fig. 7.21. Empirical correlationsof CPT data with q and ql have also been appliedto predict successfully end-bearing capacity ofpiles in sand. (G. G. Meyerhof, “Bearing Capacityand Settlement of Pile Foundations,” ASCE Journalof Geotechnical Engineering Division, vol. 102, no.GT3, 1976.)

7.18.5 Pile Settlement

Prediction of pile settlement to confirm allowableloads requires separation of the pile load into shaftfriction and end-bearing components. Since q andf–s at working loads and at ultimate loads are differ-

ent, this separation can only be qualitatively evalu-ated from ultimate-load analyses. A variety ofmethods for settlement analysis of single pileshave been proposed, many of which are empiricalor semiempirical and incorporate elements of elas-tic solutions.



7.18.6 Groups of Piles

A pile group may consist of a cluster of piles or sev-eral piles in a row. The group behavior is dictatedby group geometry and direction and location ofload as well as by subsurface conditions. Design ofpile groups for axial load has usually beenapproached by considering the safety of the pilegroup against failure by overstressing of the sub-soils and by analyzing group settlement underworking loads. Ultimate-load considerations areusually expressed in terms of a group efficiency fac-tor, which is used to reduce the capacity of each pilein the group. The efficiency factor Eg is defined asthe ratio of the ultimate group capacity to the sumof the ultimate capacity of each pile in the group.

Eg is conventionally evaluated as the sum of theultimate peripheral friction resistance and end-bearing capacities of a block of soil with breadth B,width W, and length L approximately that of thepile group. For a given pile spacing S and numberof piles n:

(7.45)

where f–s = average peripheral friction stress of

block

Qu = single-pile capacity

Fig. 7.21 Bearing-capacity factor for granularsoils related to angle of internal friction.



The limited number of pile-group tests and modeltests available suggest that for cohesive soils Eg > 1if S is more than 2.5 pile diameters D and for cohe-sionless soils Eg > 1 for the smallest practical spac-ing. A possible exception might be for very short,heavily tapered piles driven in very loose sands.

In practice, the minimum pile spacing for con-ventional piles is in the range of 2.5 to 3.0D. A largerspacing is typically applied for expanded-base piles.

Field tests have demonstrated that not all pilesin a group carry the same load and that progres-sive failures are produced, often starting at a cor-ner pile. Consequently, the simple-block failuremodel should be viewed as a very crude and total-ly empirical representation of group behavior. Sim-ilarly, other empirical group-capacity reductionformulas based solely on pile-group geometry areof limited value.

A more rational evaluation of group behaviorcan be made from settlement considerations byapplying elastic or elastoplastic analyses. A readilyapplied approach is based on superposition of anelastic solution for a single pile. This methodrelates the settlement of a single pile to that of thegroup. Single-pile settlement can be evaluatedfrom load tests as well as from static analyses.

Another very approximate analysis of groupsettlement, applicable to friction piles, models thepile group as a raft of equivalent plan dimensionssituated at a depth below the surface equal to two-thirds the pile length. Subsequently, conventionalsettlement analyses are employed (see Arts. 7.12and 7.13).

Downdrag loads are imposed on piles driventhrough soils that subsequently move downward(consolidate) with respect to the piles. (Consolida-tion is typically induced by site fills placed overcompressible subsoils or by lowering of the watertable.) Time-dependent, these downdrag forcesdepend on the amount of relative downwardmovement of the soil, the amount and distributionof positive friction (shaft resistance) initially devel-oped by the pile, the fill thickness, pile-group size,and the stiffness of the end-bearing materials. Themaximum negative skin friction that can be devel-oped on a single pile can be calculated with Eq.(7.38) with β factors for clay of 0.20 to 0.25, for silt of0.25 to 0.35, and for sand of 0.35 to 0.50.

A very approximate method of pile-groupanalysis calculates the upper limit of group dragload Qgd from


(7.46)

HF, γF, and AF represent the thickness, unit weight,and area of fill contained within the group. P, H,and cu are the circumference of the group, thethickness of the consolidating soil layers penetrat-ed by the piles, and their undrained shearstrength, respectively. Such forces as Qgd couldonly be approached for the case of piles driven torock through heavily surcharged, highly com-pressible subsoils.

(H. G. Poulos and E. H. Davis, “Elastic Solutionsfor Soil and Rock Mechanics,” and K. Terzaghi andR. B. Peck, “Soil Mechanics and Engineering Prac-tice,” John Wiley & Sons, Inc., New York; J. E. Gar-langer and W. T. Lambe, Symposium on Downdrag ofPiles, Research Report 73-56, Soils Publication no.331, Massachusetts Institute of Technology, Cam-bridge, 1973.)

7.18.7 Design of Piles for LateralLoads

Piles and pile groups are typically designed to sus-tain lateral loads by the resistance of vertical piles,by inclined, or batter, piles, or by a combination.Tieback systems employing ground anchor or dead-men reactions are used in conjunction with lateral-ly loaded sheetpiles (rarely with foundation piles).

Lateral loads or eccentric loading produce over-turning moments and uplift forces on a group ofpiles. Under these circumstances, a pile may haveto be designed for a combination of both lateraland tensile load.

Inclined Piles n Depending on the degree ofinclination, piles driven at an angle with the verti-cal can have a much higher lateral-load capacitythan vertical piles since a large part of the lateralload can be carried in axial compression. To mini-mize construction problems, however, pile batters(rake) should be less than l horizontal to 2 vertical.

Evaluation of the load distribution in a pilegroup consisting of inclined piles or combined ver-tical and batter piles is extremely complex becauseof the three-dimensional nature and indeterminan-cy of the system. A variety of computer solutionshave become available and allow a rational evalua-tion of the load distribution to inclined group piles.The same methods of axial-capacity evaluation



developed for vertical piles are applied to inclined-pile design, although higher driving energy lossesduring construction suggest that inclined pileswould have a somewhat reduced axial-load capac-ity for the same terminal resistance. (A. Hrennikoff,“Analysis of Pile Foundations with Batter Piles,”ASCE Transactions, vol. 115, 1950.)

Laterally Loaded Vertical Piles n Vertical-pile resistance to lateral loads is a function of boththe flexural stiffness of the shaft, the stiffness of thebearing soil in the upper 4D to 6D length of shaft,where D = pile diameter, and the degree of pile-head fixity. The lateral-load design capacity is alsorelated to the amount of lateral deflection permit-ted and, except under very exceptional circum-stances, the tolerable-lateral-deflection criteria willcontrol the lateral-load design capacity.

Design loads for laterally loaded piles are usu-ally evaluated by beam theory for both an elasticand nonlinear soil reaction, although elastic andelastoplastic continuum solutions are available.Nonlinear solutions require characterization of thesoil reaction p versus lateral deflection y along theshaft. In obtaining these solutions, degradation ofthe soil stiffness by cyclic loading is an importantconsideration.

The lateral-load vs. pile-head deflection rela-tionship is readily developed from charted nondi-mensional solutions of Reese and Matlock. Thesolution assumes the soil modulus K to increaselinearly with depth z; that is, K = nhz, where nh =coefficient of horizontal subgrade reaction. A char-acteristic pile length T is calculated from

(7.47)

zmax Ay By Aθ

2 4.70 3.39 –3.403 2.65 1.77 –1.754 2.44 1.63 –1.65

>5 2.43 1.62 –1.62

*Coefficients for maximum positive moment are located at about Source: L. C. Reese and H. Matlock, “Non-Dimensional Solutions fo

al to Depth,” 8th Texas Conference of Soil Mechanics and Foundation Eng

Table 7.9 Deflection, Moment, and Slope Coefficie


where EI = pile stiffness. The lateral deflection y ofa pile with head free to move and subject to a lat-eral load Pt and moment Mt applied at the groundline is given by

(7.48)

where Ay and By are nondimensional coefficients.Nondimensional coefficients are also available forevaluation of pile slope, moment, shear, and soilreaction along the shaft.

For positive moment,

(7.49)

Positive Mt and Pt values are represented by clock-wise moment and loads directed to the right on thepile head at the ground line. The coefficientsapplicable to evaluation of pile-head deflection andto the maximum positive moment and its approxi-mate position on the shaft z/ T, where z = distancebelow the ground line, are listed in Table 7.9.

The negative moment imposed at the pile headby pile-cap or other structural restraint can be eval-uated as a function of the head slope (rotation) from

(7.50)

where θs, rad, represents the counterclockwise (+)rotation of the pile head and Aθ and Bθ are coeffi-cients (see Table 7.9). The influence of the degreesof fixity of the pile head on y and M can be evalu-ated by substituting the value of –Mt from Eq.(7.50) in Eqs. (7.48) and (7.49). Note that for thefixed-head case:

Bθ Am* Bm* z/T*

–3.21 0.51 0.84 0.85–1.85 0.71 0.60 1.49–1.78 0.78 0.70 1.32–1.75 0.77 0.69 1.32

the values given in the table for z/T.r Laterally Loaded Piles with Soil Modulus Assumed Proportion-

ineering, University of Texas, 1956.

nts



(7.51)

Improvement of Lateral Resistance n

The lateral-load capacity of a specific pile type canbe most effectively increased by increasing thediameter, i.e., the stiffness and lateral-bearing area.Other steps are to improve the quality of the surfi-cial bearing soils by excavation and replacement orin-place densification, to add reinforcement, andto increase the pile-head fixity condition.

Typical lateral-load design criteria for buildingslimit lateral pile-head deformations to about ¼ in.Associated design loads for foundation piles drivenin medium dense sands or medium clays are typi-cally in the range of 2 to 4 tons, although signifi-cantly higher values have been justified by loadtesting or detailed analyses or a combination.

Resistance of pile groups to lateral loads is notwell-documented by field observations. Results ofmodel testing and elastic analysis, however, indicatethat pile spacings less than about 8 pile diameters Din the direction of loading reduce the soil modulusK. The reduction factors are assumed to vary linear-ly from 1.0 at 8D to 0.25 at a 3D spacing if the num-ber of piles in the group is 5 or more and the passiveresistance of the pile cap is ignored. The effect of thisreduction is to “soften” the soil reaction and producesmaller lateral resistance for a given group deflec-tion. Elastic analyses also confirm the long-heldjudgment that batter piles in the center of a pilegroup are largely ineffective in resisting lateral load.

(B. B. Broms, “Design of Laterally LoadedPiles,” ASCE Journal of Soil Mechanics and FoundationEngineering Division, vol. 91, no. SM3, 1965. H. Y.Fang, “Foundation Engineering Handbook,” VanNostrand Reinhold, New York. B. H. Fellenius,“Guidelines for Static Pile Design,” Deep Founda-tion Institute, 120 Charolette Place, EnglewoodCliffs, NJ 07632. H. G. Poulos and E. H. Davis,“Elastic Solutions for Soil and Rock Mechanics,”John Wiley & Sons, Inc., New York. L. C. Reese andR. C. Welch, “Lateral Loading of Deep Foundationsin Stiff Clay,” ASCE Journal of Geotechnical Engineer-ing, vol. 101, no. GT7, 1975.)

7.18.8 Static-Load Pile Testing

Because of the inherent uncertainty in static pile-design methods and the influence of construction


procedures on the behavior of piles, static-loadtests are desirable or may be required. Static-loadtests are almost always conducted on single piles;testing of pile groups is very rare.

Engineers use static-load tests to determine theresponse of a pile under applied loads. Axial com-pression testing is the most common, althoughwhen other design considerations control, uplift orlateral loading tests are also performed. In somespecial cases, testing is performed with cyclic load-ings or with combined loads; for example, axial andlateral loading. Pile testing may be performed dur-ing the design or construction phase of a project sothat foundation design data and installation criteriacan be developed or verified, or to prove the ade-quacy of a pile to carry design load.

Use of static-load pile testing is limited byexpense and time required for the tests and analy-ses. For small projects, when testing costs add sig-nificantly to the foundation cost, the increasedcost often results in elimination of pile testing. Forprojects involving a large number of piles, static-load pile tests usually are performed, but only afew piles are tested. (A typical recommendation isthat of the total number of piles to be installed innormal practice 1% be tested, but the percentageof piles tested in actual practice may be muchlower.) The number and location of test pilesshould be determined by the foundation designengineer after evaluation of the variability of sub-surface conditions, pile loadings, type of pile, andinstallation techniques. Waiting time between pileinstallation and testing generally ranges from sev-eral days to several weeks, depending on pile typeand soil conditions.

The foundation contractor generally is respon-sible for providing the physical setup for conduct-ing a static-load test. The foundation designershould supervise the testing.

Standards detailing procedures on how toarrange and conduct static-load pile tests include“Standard Test Method for Piles under Static AxialCompression Load,” ASTM D1143; “StandardMethod of Testing Individual Piles under Static AxialTension Load,” ASTM D3689; and “StandardMethod of Testing Piles under Lateral Loads,” ASTMD3966. See also “Static Testing of Deep Foundations,”U. S. Federal Highway Administration, Report No.FHWA-SA-91-042, 1992; “Axial Pile Loading Test—Part 1: Static Loading,” International Society for SoilMechanics and Foundation Engineering, 1985; and



“Canadian Foundation Engineering Manual,” 2nded., Canadian Geotechnical Society, 1985.

Load Application n In a static-load pile test,a hydraulic jack, acting against a reaction, appliesload at the pile head. The reaction may be provid-ed by a kentledge, or platform loaded with weights(Fig. 7.22), or by a steel frame supported by reac-tion piles (Fig. 7.23), or by ground anchors. The dis-tance to be used between the test pile and reaction-system supports depends on the soil conditionsand the level of loading but is generally three pilediameters or 8 ft, whichever is greater. It may benecessary to have the test configuration evaluatedby a structural engineer.

Hydraulic jacks including their operationshould conform to “Safety Code for Jacks,” ANSIB30. 1, American National Standards Institute. Thejacking system should be calibrated (with loadcells, gages, or machines having an accuracy of atleast 2%) within a 6-month period prior to pile test-ing. The available jack extension should be at least6 in. The jack should apply the load at the center ofthe pile (Fig. 7.24). When more than one jack isneeded for the test, all jacks should be pressurizedby a common device.

Loads should be measured by a calibrated pres-sure gage and also by a load cell placed betweenthe jack and the pile. Internal forces in the pile maybe measured with strain gages installed along thepile. Two types of test-loading procedures are

Fig. 7.22 Static-load test on a pile with deadweight as the reaction load.


used: the maintained load (ML) and the constant rateof penetration (CRP) methods.

In the ML method, load is applied in incre-ments of 25% of the anticipated pile capacity untilfailure occurs or the load totals 200% of the designload. Each increment is maintained until pilemovement is less than 0.01 in/h or for 2 h, whichev-er occurs first. The final load is maintained for 24 h.Then, the test load is removed in decrements of25% of the total test load, with 1 h between decre-ments. This procedure may require from 1 to 3days to complete. According to some practices, theML method is changed to the CRP procedure assoon as the rate exceeds 0.8 in/h.

Tests that consist of numerous load increments(25 to 40 increments) applied at constant timeintervals (5 to 15 min) are termed quick tests.

In the CRP procedure, the pile is continuouslyloaded so as to maintain a constant rate of pene-tration into the ground (typically between 0.01 and0.10 in/min for granular soils and 0.01 to 0.05in/min for cohesive soils). Loading is continueduntil no further increase is necessary for continu-ous pile penetration at the specified rate. As longas pile penetration continues, the load inducingthe specified penetration rate is maintained untilthe total pile penetration is at least 15% of the aver-age pile diameter or diagonal dimension, at whichtime the load is released. Also, if, under the maxi-mum applied load, penetration ceases, the load isreleased.

Alternatively, for axial-compression static-loadtests, sacrificial jacks or other equipment, such asthe Osterberg Cell, may be placed at the bottom ofthe pile to load it (J. O. Osterberg, “New Load Cell

Fig. 7.23 Reaction piles used in static-load teston a pile.



Fig. 7.24 Typical arrangement of loading equipment and instrumentation at pile head for a compressionstatic-load test. (From “Static Testing of Deep Foundations,” FHWA SA-91-042, Federal Highway Administration.)

Testing Device,” Deep Foundations Institute). Oneadvantage is automatic separation of data on shaftand toe resistance. Another is elimination of theexpense and time required for constructing a reac-tion system, inasmuch as soil resistance serves as areaction. A disadvantage is that random pile test-ing is not possible since the loading apparatus andpile installations must be concurrent.

Penetration Measurements n The axialmovement of the pile head under applied load maybe measured by mechanical dial gages or electro-mechanical devices mounted on an independentlysupported (and protected) reference beam. Figure7.24 shows a typical arrangement of equipment andinstruments at the pile head. The gages should haveat least 2 in of travel (extendable to 6 in) and typi-cally a precision of at least 0.001 in. For redundancy,measurements may also be taken with a surveyor’srod and precise level and referenced to fixed bench-marks. Another alternative is a tightly stretchedpiano wire positioned against a mirror and scalethat are attached to the side of the pile. Movementsat locations along the pile length and at the pile toemay be determined with the use of telltales.


For the ML or quick-test procedures, pile move-ments are recorded before and after the applica-tion of each load increment. For the CRP method,readings of pile movement should be taken at leastevery 30 s.

Pile-head transverse displacements should bemonitored and controlled during the test. For safe-ty and proper evaluation of test results, move-ments of the reaction supports should also be mon-itored during the test.

Interpretation of Test Results n A consid-erable amount of data is generated during a static-load test, particularly with instrumented piles. Themost widely used procedure for presenting testresults is the plot of pile-head load vs. movement.Other results that may be plotted include pile-headtime vs. movement and load transfer (from instru-mentation along the pile shaft). Shapes of load vs.movement plots vary considerably; so do the pro-cedures for evaluating them for calculation of limitload (often mistakenly referred to as failure load).

Problems in data interpretation arise from thelack of a universally recognized definition of fail-ure. For a pile that has a load-carrying capacity



greater than that of the soil, failure may be consid-ered to occur when pile movement continuesunder sustained or slightly increasing load (pileplunging). In general, the term failure load shouldbe replaced with interpreted failure load for evalua-tions from plots of pile load vs. movement. Thedefinition of interpreted failure load should bebased on mathematical rules that produce repeat-able results without being influenced by the sub-jective interpretation of the engineer. In the offset

limit method, interpreted failure load is defined asthe value of the load ordinate of the load vs. move-ment curve at ρ + 0.15 + D/120, where ρ is themovement, in, at the termination of elastic com-pression and D is the nominal pile diameter, in.One advantage of this technique is the ability totake pile stiffness into consideration. Anotheradvantage is that maximum allowable pile move-ment for a specific allowable load can be calculatedprior to proof testing of a pile. Interpretation meth-ods that rely on extrapolation of the load-move-ment curve should be avoided.

The test report should include the following aswell as other relevant data:

1. Information on general site subsurface condi-tions, emphasizing soil data obtained fromexploration near the test pile

2. Descriptions of the pile and pile installationprocedure

3. Dates and times of pile installation and statictesting

4. Descriptions of testing apparatus and testingprocedure

5. Calibration certificates

6. Photographs of test setup

7. Plots of test results

8. Description of interpretation methods

9. Name of testing supervisor

The cost, time, and effort required for a static-load test should be carefully weighed against themany potential benefits. A static-load test on a sin-gle pile, however, does not account for the effectsof long-term settlement, downdrag loads, time-dependent soil behavior, or pile group action, nordoes the test eliminate the need for an adequatefoundation design.


7.19 Dynamic Pile Testingand Analysis

Simple observations made during impact pile dri-ving are an important and integral part of the pileinstallation process. In its most basic form, dynamic-load pile testing encompasses visual observations ofhammer operation and pile penetration during piledriving. Some engineers apply equations based onthe Newtonian physics of rigid bodies to the pilemovements recorded during pile driving to esti-mate the load-carrying capacity of the pile. Thebasic premise is that the harder it is to drive the pileinto the ground, the more load it will be able tocarry. The equations, generally known as energyformulas, typically relate hammer energy and workdone on the pile to soil resistance. More than 400formulas have been proposed, including the widelyused and simple Engineering News formula.

This method of estimating load capacity, how-ever, has several shortcomings. These includeincomplete, crude, and oversimplified representa-tion of pile driving, pile and soil properties, andpile-soil interaction. Often, the method has beenfound to be grossly inaccurate and unreliable tothe extent that many engineers believe that itshould be eliminated from contemporary practice.(See, for example, “Manual on Design and Con-struction of Driven Pile Foundations,” FederalHighway Administration.)

7.19.1 Wave Equation

In contrast to the deficiencies of the energy formu-las, analysis of pile-driving blow count or penetra-tion per blow yields more accurate estimates of theload-carrying capacity of a pile, if based on accu-rate modeling and rational principles. One suchtype of analysis employs the wave equation basedon a concept developed by E. A. Smith (ASCE Jour-nal of Geotechnical Engineering Division, August1960). Analysis is facilitated by use of computerprograms such as GRLWEAP (Goble RauscheLikins and Associates, Inc., Cleveland, Ohio) thatsimulate and analyze impact pile driving. Sophisti-cated numerical modeling, advanced analyticaltechniques, and one-dimensional elastic-wave-propagation principles are required. Computa-tions can be performed with personal computers.A substantial improvement that the wave equationoffers over the energy approach is the ability to



model all hammer, cushion, pile-cap, and pile andsoil components realistically.

Figure 7.25 illustrates the lumped-mass modelused in wave equation analyses. All componentsthat generate, transmit, or dissipate energy arerepresented by a spring, mass, or dashpot. Thesepermit representation of mass, stiffness, and vis-cosity.

Fig. 7.25 Lumped-mass model of a pile used in wspring represents mass and stiffness; dashpot, the loadblock with spring, the soil resistance forces along the ppile velocity. (c) Variation of static soil resistance with p


A series of masses and springs represent themass and stiffness of the pile. Elastic springs andlinear-viscous dashpots model soil-resistance forcesalong the pile shaft and under the toe. The springsrepresent the displacement-dependent static-loaded components, and the dashpots the loading-dependent dynamic components. Springs modelstiffness and coefficient of restitution (to account

ave equation analysis. (a) Rectangular block withing-dependent dynamic components; small square

ile shaft. (b) Variation of dynamic soil resistance withile displacement.



for energy dissipation) of hammer and pile cush-ions. A single mass represents the pile-cap. Forexternal-combustion hammers, the representationis straightforward: a stocky ram, by a single mass;the hammer assembly (cylinder, columns, etc.), bymasses and springs. For internal-combustion ham-mers, the modeling is more involved. The slenderram is divided into several segments. The gas pres-sure of the diesel combustion cycle is calculatedaccording to the thermodynamic gas law for eitherliquid or atomized fuel injection.

The parameters needed for execution of a waveequation analysis with the GRLWEAP computerprogram are:

Hammer: Model and efficiency

Hammer and Pile Cushions: Area, thickness, elas-tic modulus, and coefficient of restitution

Pile Cap: Weight, including all cushions and anyinserts

Pile: Area, elastic modulus, and density, all as afunction of length

Soil: Total static capacity, percent shaft resistanceand its distribution, quake and damping constantsalong the shaft and under the toe

In practice, wave equation analysis is employedto deal with the following questions:

1. If the input to the computer program provides acomplete description of hammer, cushions, pilecap, pile, and soil, can the pile be driven safelyand economically to the required static capacity?

2. If the input provides measurements of pile pene-tration during pile driving or restriking blowcount, what is the static-load capacity of the pile?

For case 1, pile design and proper selection of ham-mer and driving system can be verified to ensure thatexpected pile-driving stresses are below allowablelimits and reasonable blow count is attainable beforeactual field work starts. For case 2, given field obser-vations made during pile driving, the analysis is usedas a quality-control tool to evaluate pile capacity.

Generally, wave equation analysis is applied to apile for the cases of several static-load resistancescovering a wide range of values (at a constant pilepenetration corresponding to the expected final pile-toe depth). Analysis results are then plotted as a bear-


ing graph relating static pile capacity and drivingstresses to blow counts.

Figure 7.26 presents a bearing graph from ananalysis of a single-acting external-combustionhammer (Vulcan 012) and a precast concrete pile(18 in square, 95 ft long).

For a diesel hammer, the stroke or bounce-chamber pressure is also included in the plot.Alternatively, for an open-end diesel hammer (orany hammer with variable stroke), the analysismay be performed with a constant pile staticcapacity and various strokes. In this way, therequired blow count can be obtained as a functionof the actual stroke.

Wave equation analysis may also be based onpile penetration (commonly termed pile drivabili-ty). In this way, variations of soil resistance withdepth can be taken into account. Analysis resultsare obtained as a function of pile penetration.

Pile specifications prescribe use of wave equationanalysis to determine suitability of a pile-driving sys-tem. Although it is an excellent tool for analysis ofimpact pile driving, the wave equation approach hassome limitations. These are mainly due to uncertain-ties in quantifying some of the required inputs, suchas hammer performance and soil parameters. Thehammer efficiency value needed in the analysis isusually taken as the average value observed in manysimilar situations. Also, soil damping and quake val-ues (maximum elastic soil deformation) needed inmodeling soil behavior cannot be readily obtainedfrom standard field or laboratory soil tests or relatedto other conventional engineering soil properties.

Dynamic-load pile testing and data analysisyield information regarding the hammer, drivingsystem, and pile and soil behavior that can be usedto confirm the assumptions of wave equationanalysis. Dynamic-load pile tests are routinely per-formed on projects around the world for the pur-poses of monitoring and improving pile installa-tion and as construction control procedures. Manyprofessional organizations have established stan-dards and guidelines for the performance and useof this type of testing; for example, ASTM (D4945),Federal Highway Administration (“Manual onDesign and Construction of Driven Pile Founda-tion”). Dynamic-load testing methods are alsoeffectively employed for evaluating cast-in-placepiles (“Dynamic Load Testing of Drilled Shaft—Final Report,” Department of Civil Engineering,University of Florida, Gainesville 1991).



Main objectives of dynamic-load testinginclude evaluation of driving resistance and static-load capacity; determination of pile axial stressesduring driving; assessment of pile structuralintegrity; and investigation of hammer and drivingsystem performance.

7.19.2 Case Method

A procedure developed at Case Institute of Tech-nology (now Case Western Reserve University),Cleveland, Ohio, by a research team headed by G.G. Goble, enables calculation of pile static capacityfrom measurements of pile force and accelerationunder hammer impacts during pile driving. The

Fig. 7.26 Bearing graph derived from a wave equpressive stresses with blows per foot. (b) Ultimate capation distribution along the tested pile. Driving was dociency. Helmet weighed 2.22 kips (k). Stiffness of hamk/in. Pile was 95 ft long and had a top area of 324 in2. elastic deformation), 0.100 in for shaft resistance and s/ft for shaft and toe resistance.


necessary equipment and analytical methodsdeveloped have been expanded to evaluate otheraspects of the pile-driving process. These proce-dures are routinely applied in the field using adevice called the Pile Driving Analyzer (PDA). Asan extension of the original work the researchersdeveloped a computer program known as theCAse Pile Wave Analysis Program (CAPWAP),which is described later.

Measurements of pile force and velocityrecords under hammer impacts are the basis formodern dynamic pile testing. Data are obtainedwith the use of reusable strain transducers andaccelerometers. Strain gages are bolted on the pileshaft, usually at a distance of about two pile diam-

ation analysis. (a) Variation of pile tensile and com-city of pile indicated by blows per foot. (c) Skin fric-

ne with a Vulcan hammer, model 012, with 67% effi-mer cushion was 5765 k/in and of pile cushion, 1620Other input parameters were quake (soil maximum0.150 in for toe resistance; soil damping factor, 0.150



eters below the pile head. The PDA serves as a dataacquisition system and field computer that pro-vides signal conditioning, processing, and calibra-tion of measurement signals. It converts measure-ments of pile strains and acceleration to pile forceand velocity records. Dynamic records and testingresults are available in real time following eachhammer impact and are permanently stored indigital form. Using wave propagation theory andsome assumptions regarding pile and soil, the PDAapplies Case method equations and computes in aclosed-form solution some 40 variables that fullydescribe the condition of the hammer-pile-soil sys-tem in real time following each hammer impact.

When a hammer or drop weight strikes the pilehead, a compressive-stress wave travels down thepile shaft at a speed c, which is a function of thepile elastic modulus and mass density (Art. 6.82.1).The impact induces at the pile head a force F and aparticle velocity v. As long as the wave travels inone direction, force and velocity are proportional;that is, F = Zv, where Z is the pile impedance, andZ = EA/c, where A is the cross-sectional area of thepile and E is its elastic modulus. Changes inimpedance in the pile shaft and pile toe, and soil-resistance forces, produce wave reflections. Thereflected waves arrive at the pile head after impactat a time proportional to the distance of their loca-tion from the toe. Soil-resistance forces or increasein pile impedance cause compressive-wave reflec-tions that increase pile force and decrease velocity.Decrease in pile impedance has the opposite effect.

For a pile of length L, impedance Z, and stress-wave velocity c, the PDA computes total soil resis-tance from measured force and velocity recordsduring the first stress-wave cycle; that is, when 0 <t < 2L/c, where t is time measured from start ofhammer impact. This soil resistance includes bothstatic and viscous components. In the computationof pile bearing capacity under static load RS at thetime of testing, effects of soil damping must beconsidered. Damping is associated with velocity.By definition, the Case method damping force isequal to ZJcvb, where Jc is the dimensionless Casedamping factor, and vb the pile toe velocity, whichcan be computed from measured data at the pilehead by applying wave mechanics principles. Thestatic capacity of a pile can be calculated from:

(7.51a)


where t2 = t1+ 2L/c and t1 is normally the time ofthe first relative velocity peak. The damping con-stant Jc is related to soil grain size and may be takenfor clean sands as 0.10 to 0.15, for silty sands as 0.15to 0.25, for silts as 0.25 to 0.40, for silty clays as 0.4to 0.7, and for clays as 0.7 to 1.0.

The computed RS value is the pile static capac-ity at the time of testing. Time-dependent effectscan be evaluated by testing during pile restrikes.For this purpose, the pile must have sufficient pen-etration under the hammer impact to achieve fullmobilization of soil-resistance forces. (F. Rausche,G. Goble, and G. Likins, “Dynamic Determinationof Pile Capacity,” ASCE Journal of Geotechnical Engi-neering Division, vol. 111, no. 3, 1985.)

The impact of a hammer subjects piles to a com-plex combination of compression, tension, torsion-al, and bending forces. Maximum pile compressivestress at the transducers’ location is directlyobtained from the measured data as the maximumrecorded force divided by the pile area. For pileswith mainly toe soil resistance, the compressiveforce at the pile toe is calculated from pile-headmeasurements and one-dimensional wave propa-gation considerations. Maximum tension force inthe pile shaft can be computed from measure-ments near the pile head by considering the mag-nitude of both upward- and downward-travelingforce components. Pile damage occurs if drivingstresses exceed the strength of the pile material.

For a pile with a uniform cross-sectional areainitially, damage after driving can be indicated bya change in area. Since pile impedance is propor-tional to pile area, a change in impedance wouldindicate pile damage. Hence, a driven pile can betested for underground damage by measuringchanges in pile impedance. Changes in pile imped-ance cause wave reflections and changes in theupward-traveling wave measured at the pile head.From the magnitude and time after impact of therelative wave changes, the extent and location ofimpedance change and hence of pile damage canbe determined. Determination of pile damage canbe assisted with the use of the PDA, which com-putes a relative integrity factor (unity for uniformpiles and zero for a pile end) based on measureddata near the pile head. (F. Rausche and G. G.Goble, “Determination of Pile Damage by TopMeasurements,” ASTM STP-670.)

The PDA also is helpful in determining theenergy actually received by a pile from a hammer



blow. Whereas hammers are assigned an energy rat-ing by manufacturers, only the energy reaching thepile is of significance in effecting pile penetration.Due to many factors related to the hammer mechan-ical condition, driving-system behavior, and generaldynamic hammer-cushions-pile-soil incompatibility,the percentage of potential hammer energy thatactually reaches the pile is quite variable and oftenless than 50%. (“The Performance of Pile DrivingSystems—Main Report,” vol. 1–4, FHWADTFH 61-82-1-00059, Federal Highway Administration.) Fig-ure 7.27 presents a summary of data obtained onhundreds of sites to indicate the percentage of allhammers of a specific type with an energy-transferefficiency less than a specific percentage. Givenrecords of pile force and velocity, the PDA calculatesthe transferred energy as the time integral of the

Fig. 7.27 Comparison of performance of two typeThe percentile indicates the percentage of all hammethan a specific percentage.


product of force and velocity. The maximum trans-ferred energy value for each blow represents the sin-gle most important parameter for an overall evalua-tion of driving-system performance.

7.19.3 CAPWAP Method

The CAse Pile Wave Analysis Program (CAPWAP)combines field-measured dynamic-load data andwave-equation-type analytical procedures to pre-dict pile static-load capacity, soil-resistance distrib-ution, soil-damping and quake values, pile load vs.movement plots, and pile-soil load-transfer charac-teristics. CAPWAP is a signal-matching or systemidentification method; that is, its results are basedon a best possible match between a computed vari-able and its measured equivalent.

s of hammers when driving steel or concrete piles.rs in each case with a rated transfer efficiency less



The pile is modeled with segments about 3 ft longwith linearly elastic properties. Piles with nonuni-form cross sections or composite construction can beaccurately modeled. Static and dynamic forces alongthe pile shaft and under its toe represent soil resis-tance. Generally, the soil model follows that of theSmith approach (Art. 7.19.1) with modifications toaccount for full pile-penetration and rebound effects,including radiation damping. At the start of theanalysis, an accurate pile model (incorporatingsplices, if present) is established and a complete set ofsoil constants is assumed. The hammer model usedfor the wave equation method is replaced by themeasured velocity imposed as a boundary condition.The program calculates the force necessary to inducethe imposed velocity. Measured and calculated forcesare compared. If they do not agree, the soil model isadjusted and the analysis repeated. This iterativeprocess is continued until no further improvement inthe match can be obtained. The total number ofunknowns to be evaluated during the analysis isNs + 18, where Ns is the number of soil elements.Typically, one soil element is placed at every 6 ft ofpile penetration plus an additional one under the toe.

Results that can be obtained from a CAPWAPanalysis include the following:

Comparisons of measured values with corre-sponding computed values

Soil-resistance forces and their distribution for staticloads

Soil-stiffness and soil-damping parameters alongthe pile shaft and under its toe

Forces, velocities, displacements, and energies as afunction of time for all pile segments

Simulation of the relationship between static loadsand movements of pile head and pile toe

Pile forces at ultimate soil resistance

Correlations between CAPWAP predicted valuesand results from static-load tests indicate verygood agreement. (ASCE Geotechnical Special Pub-lication No. 40, 1994.)

7.19.4 Low-Strain DynamicIntegrity Testing

The structural integrity of driven or cast-in-placeconcrete piles may be compromised during installa-


tion. Piles may also be damaged after installation bylarge lateral movements from impacts of heavyequipment or from slope or retaining-wall failures.Procedures such as excavation around a suspect pileor drilling and coring through its shaft are crudemethods for investigating possible pile damage. Sev-eral testing techniques are available, however, forevaluation of the structural integrity of deep foun-dation elements in a more sophisticated manner (W.G. Fleming, A. J. Weltmen, M. F. Randolph, and W. K.Elson, “Piling Engineering,” Surrey University Press,London.) Some of these tests, though, require thatthe pile be prepared or instrumented before or dur-ing installation. These requirements make randomapplication prohibitively expensive, if not impossi-ble. A convenient and economical method is thelow-strain pulse-echo technique, which requires rel-atively little instrumentation and testing effort andwhich is employed in low-strain, dynamic-loadintegrity testing. This method is based on one-dimensional wave mechanics principles and themeasurement of dynamic-loading effects at the pilehead under impacts of a small hand-held hammer.

The following principle is utilized: Whenimpacted at the top, a compressive-stress wavetravels down the pile shaft at a constant speed cand is reflected back to the pile head from the toe.Changes in pile impedance Z change wave charac-teristics and indicate changes in pile cross-section-al size and quality and thus possible pile damage(Art. 7.19.2). Low-strain integrity testing is basedon the premise that changes in pile impedance andsoil-resistance forces produce predictable wavereflections at the pile head. The time after impactthat the reflected wave is recorded at the pile headcan be used to calculate the location on the pile ofchanges in area or soil resistance.

Field equipment consists of an accelerometer, ahand-held hammer (instrumented or withoutinstrumentation), dedicated software, and a PileIntegrity Tester (Fig. 7.28), a data acquisition sys-tem capable of converting analog signals to digitalform, data processing, and data storage. Pilepreparation involves smoothing and leveling of asmall area of the pile top. The accelerometer isaffixed to the pile top with a jell-type material, andhammer blows are applied to the pile head. Typi-cally, pile-head data resulting from several ham-mer blows are averaged and analyzed.

Data interpretation may be based on records ofpile-top velocity (integral of measured accelera-



tion), data in time or frequency domains, or morerigorous dynamic analysis. For a specific stress-wave speed (typically 13,000 ft/s), records of veloc-ity at the pile head can be interpreted for pilenonuniformities and length. As an example, Fig.7.29 shows a plot in which the abscissa is time,measured starting from impact, and the ordinate isdepth below the pile top. The times that changes inwave characteristics due to pile impedance or soilresistance are recorded at the pile head are repre-sented along the time axis by small rectangles. Theline from the origin extending downward to theright presents the position of the wave travelingwith velocity c after impact. Where a change in pileimpedance Z occurs, at depth a and time a/c, a line(I) extends diagonally upward to the right andindicates that the wave reaches the pile top at time2a/c. Hence, with the time and wave velocityknown, the distance a can be calculated. Similarly,from time 2b/c, as indicated by line II, the distanceb from the pile top of the change in soil resistanceR can be computed. Line III indicates that the wavefrom the toe at distance L from the pile head reach-es the head at time 2L/c.

Dynamic analysis may be done in a signal-matching process or by a method that generates apile impedance profile from the measured pile-topdata. (F. Rausche et al., “A Formalized Procedure forQuality Assessment of Cast-in-Place Shafts UsingSonic Pulse Echo Methods,” TransportationResearch Board, Washington, D.C. 1994.)

The low-strain integrity method is applicable toconcrete (cast-in-place and driven) and wood piles.Usually, piles are tested shortly after installation so

Fig. 7.28 Pile Integrity Tester. (Courtesy of PileDynamics, Inc., Cleveland, Ohio.)


that deficiencies may be detected early and correc-tive measures taken during foundation construc-tion and before erection of the superstructure. Asfor other nondestructive testing methods, theresults of measurements recorded may be dividedinto four main categories: (1) clear indication of asound pile, (2) clear indication of a serious defect,(3) indication of a somewhat defective pile, and (4)records do not support any conclusions. The foun-dation engineer, taking into consideration structur-al, geotechnical, and other relevant factors, shoulddecide between pile acceptability or rejection.

The low-strain integrity method may be used todetermine the length and condition of piles underexisting structures. (M. Hussein, G. Likins, and G.Goble, “Determination of Pile Lengths under ExistingStructures,” Deep Foundations Institute, 1992.)

The method has some limitations. For example,wave reflections coming from locations greaterthan about 35 pile diameters may be too weak to bedetected at the pile head with instruments current-

Fig. 7.29 Graph relates distance from a pilehead to a depth where a change in the pile cross sec-tion or soil resistance occurs to the time it takes animpact pulse applied at the pile head and travelingat a velocity c to reach and then be reflected fromthe change back to the pile head. Line I indicates thereflection due to impedance, II the reflection due topassive resistance R (modeled velocity proportion-al), and III the reflection from the pile toe.



Position . . . within 6 in of plan location (3 in for pile groups with less than 5 piles)

Plumbness . . . deviation from vertical shall not exceed 2% in any interval (4% from axisfor batter piles)

Pile-hammer assembly Verification of the suitability of the proposed hammer assembly to drivethe designated piles shall be provided by wave equation or equiva-lent analysis subject to the engineer's approval.

Pile-driving leaders All piles shall be driven with fixed leaders sufficiently rigid to maintain pileposition and axial alignment during driving.

Driving criteria . . . to an elevation of at least ____ and/or to a terminal driving resistanceof__ blows/__ in

Indicator piles Before start of production driving, indicator piles should be driven at loca-tions determined by the engineer. Continuous driving resistancerecords shall be maintained for each indicator pile.

Preboring Preboring immediately preceding pile installation shall extend to elevation____. The bore diameter for friction piles shall not be less than 1 ormore than 2 in smaller than the pile diameter.

Heave* The elevation of pile butts or tips of CIPC pile shells shall be establishedimmediately after driving and shall be resurveyed on completion of thepile group. Should heave in excess of 1/4 in be detected, the piles shallbe redriven to their initial elevation or as directed by the engineer.

Relaxation or setup† Terminal driving resistance of piles in-place for at least 24 h shall be redriv-en as directed by the engineer.

*Can be initially conducted on a limited number of pile groups and subsequently extended to all piles if required.† May be specified as part of initial driving operations.

Table 7.10 Guide to Selected Specification Provisions

ly available. Also, gradual changes in pile imped-ance may escape detection. Furthermore, themethod may not yield reliable results for steel piles.

Concrete-filled steel pipe piles may be evaluat-ed with this method.

7.20 Pile Specification NotesSpecifications for pile installation should providerealistic criteria for pile location, alignment, andminimum penetration or terminal driving resis-tance. Particular attention should be given to pro-visions for identification of pile heave and relax-ation and for associated remedial measures.Corrective actions for damaged or out-of-positionpiles should also be identified. Material quality andquality control should be addressed, especially forcast-in-place concrete piles. Tip protection of pilesis an important consideration for some types of


high-capacity, end-bearing piles or piles driventhrough obstructions. Other items that may beimportant are criteria for driving sequence in pilegroups, preexcavation procedures, protectionagainst corrosive subsoils, and control of pile driv-ing in proximity to open or recently concreted pileshells. Guidelines for selected specification itemsare in Table 7.10.

7.21 Drilled ShaftsDrilled shafts are commonly used to transfer largeaxial and lateral loads to competent bearing mate-rials by shaft or base resistance or both. Alsoknown as drilled piers, drilled-in caissons, or large-diameter bored piles, drilled shafts are cylindrical,cast-in-place concrete shafts installed by large-diameter, auger drilling equipment. Shaft diame-ters commonly range from 2.5 to 10 ft and lengths



from 10 to 150 ft, although shafts with dimensionswell outside these ranges can be installed. Shaftsmay be of constant diameter (straight shafts, Fig.7.30) or may be underreamed (belled. Fig. 7.30) orsocketed into rock (Fig. 7.30). Depending on loadrequirements, shafts may be concreted with orwithout steel reinforcement.

Under appropriate foundation conditions, asingle drilled shaft is well-suited for support ofvery heavy concentrated loads; 2000 tons withrock bearing is not unusual.

Subsurface conditions favoring drilled shaftsare characterized by materials and groundwaterconditions that do not induce caving or squeezingof subsoils during drilling and concrete placement.High-capacity bearing levels at moderate depthsand the absence of drilling obstructions, such asboulders or rubble, are also favorable conditions.Current construction techniques allow drilledshafts to be installed in almost any subsurface con-dition, although the cost-effectiveness or reliabilityof the system will vary significantly.

7.21.1 Construction Methods forDrilled Shafts

In stable soil deposits, such as stiff clays, concretewith or without reinforcement may be placed inuncased shafts. Temporary casing, however, maybe employed during inspection of bearing condi-tions. Temporary casing may also be installed inthe shaft during or immediately after drilling toprevent soil intrusion into the concrete duringplacement. During this process, the height of con-

Fig. 7.30 Types of drilled shafts.


crete in the casing should at all times be sufficientso that the weight more than counterbalanceshydrostatic heads imposed by groundwater or byfluid trapped in the annular space between the soiland the casing. The lack of attention to thisrequirement is perhaps the greatest contributor todrilled-shaft failures.

Unstable soil conditions encountered within alimited interval of shaft penetration can be han-dled by advancing a casing into stable subsoilsbelow the caving zone via vibratory driving or byscrewing down the casing with a torque-barattachment. The hole is continued by augeringthrough the casing, which may subsequently beretracted during concrete placement or left inplace. A shaft through unstable soils may also beadvanced without a casing if a weighted drillingfluid (slurry) is used to prevent caving.

For a limited unstable zone, underlain by rela-tively impervious soils, a casing can be screwedinto these soils so as to form a water seal. Thisallows the drilling slurry to be removed and theshaft continued through the casing and completedby normal concreting techniques.

The shaft may also be advanced entirely by slur-ry drilling techniques, although the procedure isnot commonly used. With this method, concrete istremied in place so as to completely displace theslurry. One disadvantage is that in-place inspectionof the shaft before concrete is placed is precluded.

Reinforcing steel must be carefully designed tobe stable under the downward force exerted by theconcrete during placement. Utilization of rein-forcement that is not full length is not generallyrecommended where temporary casing is used tofacilitate concrete placement.

Concrete can be placed in shafts containing notmore than about 4 in of water (less for belledshafts). Free-fall placement can be used if an unob-structed flow is achieved. Bottom-discharge hop-pers centered on the shaft facilitate unrestrictedflow, whereas flexible conduits (elephant trunks)attached to the hopper can be used to guide con-crete fall in heavily reinforced shafts. Rigid tremiepipes are employed to place concrete in water or inslurry-filled shafts.

Equipment and Tools n Large-diameterauger drills are crane- and truck-mounted,depending on their size and weight. The capacityof the drill is rated by its maximum continuous



torque, ft-lb, and the force exerted on the auger.This force is the weight of the Kelly bar (drill stem)plus the force applied with some drills by theirKelly-bar down-crown mechanism.

Downward force on the auger is a function ofthe Kelly-bar length and cross section. TelescopingKelly bars with cross sections up to 12 in squarehave been used to drill 10-ft-diameter shafts inearth to depths over 220 ft. Solid pin-connectedKelly sections up to 8 in square have also beeneffectively used for drilling deep holes. Additionaldown-crowd forces exerted by some drills are onthe order of 20 to 30 kips (crane-mounts) and 15 to50 kips (truck-mounts).

Drilling tools consisting of open helix (single-flight) and bucket augers are typically used forearth drilling and may be interchanged duringconstruction operations. To drill hard soil and softand weathered rock more efficiently, flight augersare fitted with hard-surfaced teeth. This type ofauger can significantly increase the rate of advancein some materials and provides a more equitabledefinition of “rock excavation” when compared tothe refusal of conventional earth augers. Flightaugers allow a somewhat faster operation and insome circumstances have a superior penetrationcapability. Bucket augers are usually more efficientfor excavating soft soils or running sands and pro-vide a superior bottom cleanout.

Belled shafts in soils and soft rocks are con-structed with special underreaming tools. Theseare usually limited in size to a diameter three timesthe diameter of the shaft. Hand mining techniquesmay be required where hard seams or otherobstructions limit machine belling.

Cutting tools consisting of roller bits or corebarrels are typically used to extend shafts into rockand to form rock sockets. Multiroller bits are oftenused with a reverse circulation type of rotary drillingrig. This technique, together with percussion bitsactivated by pneumatically powered drills, usuallyprovide the most rapid advance in rock but havethe disadvantage of requiring special drill typesthat may not be efficient in earth drilling.

7.21.2 Construction Precaution forDrilled Shafts

During the preparation of drilled-shaft design andconstruction specifications, special attentionshould be given to construction-related design fea-


tures, including shaft types, shaft-diameter varia-tions, site trafficability, ground-loss potential, andprotection of adjacent facilities. Proper technicalspecifications and contract provisions for suchitems as payment for rock excavation and drillingobstructions are instrumental in preventing signif-icant cost overruns and associated claims.

Some cost-reduction or quality-related precau-tions in preparation of drilled-shaft designs andspecifications are:

1. Minimize the number of different shaft sizes;extra concrete quantities for diameters largerthan actually needed are usually far less costlythan use of a multiplicity of drilling tools andcasings.

2. Use larger-diameter, plain-concrete shafts inlieu of reinforcing, whenever feasible.

3. Delete a requirement for concrete vibration anduse concrete slumps not less than 6 + 1 in.

4. Do not leave a casing in place in oversized holesunless pressure grouting is used to preventground loss.

5. Shaft diameters should be at least 2.5 ft, prefer-ably 3 ft, to facilitate in-place inspection.

6. Limit drilled shafts with diameters less than 2.5ft to shafts with a ratio of length to diameter lessthan about 15.

7. Avoid uncased shafts less than 1.5 ft in diameter.Such shafts have a relatively high probability offailure because of potential discontinuities inthe concrete.

Tolerances for location of drilled shafts shouldnot exceed 3 in or 1/24 of the shaft diameter, whichev-er is less. Vertical deviation should not be more than2% of the shaft length or 12.5% of the diameter,whichever controls, except for special conditions.

Provisions for proof testing are extremely impor-tant for shafts designed for high-capacity end bear-ing. This is particularly true for bearing materialsthat may contain discontinuities or have randomvariations in quality. Small percussion drills (jack-hammers) are often used for proof testing and maybe supplemented by diamond coring, if appropriate.

Because many drilled-shaft projects involvevariations in bearing levels that cannot be quanti-fied during the design stage, the limitations of thebearing level and shaft quantities estimated for



bidding purposes must be clearly identified. Varia-tions in bearing level and quality are best accom-modated by specifications and contract provisionsthat facilitate field changes in shaft type and con-struction technique. The continuous presence of aqualified engineer, experienced in drilled-shaftconstruction, is required to ensure the quality andcost-effectiveness of the construction.

7.21.3 Drilled-Shaft Design

Much of the design methodology for drilled shaftsis similar to that applied to pile foundations andusually differs only in the manner in which thedesign parameters are characterized. Consequently,drilled-shaft design may be based on precedent(experience), load testing, or static analyses. Forexample, most shafts drilled to rock bearing areproportioned in accordance with local precedent(including building codes), whereas shaftsdrilled in soil deposits are more often designedwith the use of static analyses and occasionallyfrom load-test results. The ultimate-load designapproach is currently the most common form ofstatic analysis applied, although load-deforma-tion compatibility methods are being increasing-ly used (see Art. 7.17).

7.21.4 Skin Friction in Cohesive Soils

Ultimate skin friction for axially loaded shaftsdrilled into cohesive soils is usually evaluated byapplication of an empirically derived reduction(adhesion) factor to the undrained shear strengthof the soil in contact with the shaft [see Eq. (7.37)].For conventionally drilled shafts in stiff clays (cu >0.50 tons/ft2), the adhesion factor α has beenobserved to range usually between 0.3 and 0.6.

Based on analysis of high-quality load-testresults, primarily in stiff, fissured Beaumont andLondon clays, α factors of 0.5 and 0.45 have beenrecommended. Unlike the criteria applied to piledesign, these factors are independent of cu. Reesehas recommended that the shaft length assumedto be effective in transferring load should bereduced by 5 ft to account for base interactioneffects. and that a similar reduction be applied toaccount for surface effects such as soil shrinkage.

Table 7.11 lists recommended α factors forstraight shafts as a function of the normalized


shear strength cu /σ′νo (Art. 7.5.1) and plasticity indexIp (Art. 7.4). These factors reflect conventional dry-hole construction methods and the influence ofthe stress history and plasticity of the soil in con-tact with the shaft. The α factors in Table 7.11 maybe linearly interpreted for specific values of cu /σ′νoand Ip. To account for tip effects, the part of theshaft located 1 diameter above the base should beignored in evaluation of Qsu (see Art. 7.17).

Where shafts are drilled to bearing on relative-ly incompressible materials, the amount of relativemovement between the soil and the shaft may beinsufficient to develop a substantial portion of theultimate skin friction, particularly for short, verystiff shafts. Under these circumstances, Qs shouldbe ignored in design or analyzed with load-dis-placement compatibility procedures.

Because belled shafts usually require largerdeformations than straight shafts to developdesign loads, Qsu may be reduced for such shafts asa result of a progressive degradation at relativedeformations greater than that required to developpeak values. Limited data on belled vs. straightshafts suggest a reduction in the α factor on theorder of 15% to account for the reduced shaft fric-tion of the belled shafts. It is also conservative toassume that there is no significant load transfer byfriction in that portion of the shaft located about 1diameter above the top of the bell.

There is some evidence that when shaft drillingis facilitated by the use of a weighted drilling fluid(mud slurry), there may be a substantial reductionin Qsu, presumably as a result of entrapment of theslurry between the soil and shaft concrete. Wherethis potential exists, it has been suggested that theα factor be reduced about 40%. This reduction,however, does not apply to intervals of the shaftdrilled dry.

Normalized Shear Strength cu /σ′νo*

0.3 or Less 1.0 2.5 or More

20 0.33 0.44 0.5530 0.36 0.48 0.6060 0.40 0.52 0.65

* Based on UU tests on good-quality samples selected so that cu

is not significantly influenced by the presence of fissures.

Table 7.11 Adhesion Factors α for Drilled Shafts

PlasticityIndex



(A. W. Skempton, “Summation, Symposium onLarge Bored Piles,” Institute of Civil Engineers, Lon-don; L. C. Reese, F. T. Toma, and M. W. O’Neill,“Behavior of Drilled Piers under Axial Loading,” ASCEJournal of Geotechnical Engineering Division, vol. 102, no.GT5, 1976; W. S. Gardner, “Investigation of the Effectsof Skin Friction on the Performance of Drilled Shafts inCohesive Soils,” Report to U.S. Army EngineersWaterways Experiment Station (Contract no. DACA39-80-C-0001), vol. 3, Vicksburg, Miss., 1981.)

7.21.5 Skin Friction in CohesionlessSoils

Ultimate skin friction in cohesionless soils can beevaluated approximately with Eq. (7.40). In theabsence of more definitive data, K in Eq. (7.40) maybe taken as 0.6 for loose sands and 0.7 for mediumdense to dense sands, on the assumption that thesoil-shaft interface friction angle is taken as φ′–5°. Asfor piles, limited test data indicate that the averagefriction stress fsu is independent of overburden pres-sure for shafts drilled below a critical depth zc offrom 10 (loose sand) to 20 (dense sand) shaft diam-eters. The limiting skin friction fi for shafts with ratioof length to diameter L/D < 25 should appreciablynot exceed 1.0 ton/ft2. Average fsu may be less than1.0 ton/ft2 for shafts longer than about 80 ft.

Equation (7.52) approximately represents a cor-relation between fsu and the average standard pen-etration test blow count N

–within the embedded

pile length recommended for shafts in sand witheffective L/D < 10. The stress fsu so computed, how-ever, is less conservative than the foregoing designapproach, particularly for N

–> 30 blows per foot.

(7.52)

(L. C. Reese and S. J. Wright, “Drilled Shaft Designand Construction Guidelines Manual,” Office ofResearch and Development, Federal HighwayAdministration, U.S. Department of Transportation,Washington, D.C., 1977.)

Currently there are no indications that the useof drilling mud to penetrate potentially unstablecohesionless soil zones will significantly affect theskin friction of such shafts.

7.21.6 End Bearing on Soils

End bearing of drilled shafts in cohesive soils istypically evaluated as described for driven piles


[Eq. (7.42)]. The shear-strength term in this equa-tion represents the average cu within a zone of 2diameters below the shaft space. A 75% reductionin c–ufor fissured clay has been suggested for shaftswith diameters greater than 3 ft [see Eq. (7.37)]. Forsmaller shafts, the suggested reduction factor is 0.8.

End bearing in cohesionless soils can be estimat-ed in accordance with Eq. (7.43) with the same criti-cal-depth limitations described for pile foundations.Nq for drilled shafts, however, has been observed tobe significantly smaller than that applied to piles (seeFig. 7.19). Meyerhof has suggested that Nq should bereduced by 50%. Alternatively, qu can be expressed interms of the average SPT blow count N

–as:

(7.53)

where qu = ultimate base resistance at a settlementequivalent to 5% of the base diameter. (G. G. Mey-erhof, “Bearing Capacity and Settlement of PileFoundations,” ASCE Journal of Geotechnical Engi-neering Division, vol. 102, no. GT3, 1976.)

7.21.7 Shaft Settlement

Drilled-shaft settlements can be estimated byempirical correlations or by load-deformationcompatibility analyses (see Art. 7.17). Other meth-ods used to estimate settlement of drilled shafts,singly or in groups, are identical to those used forpiles (Art. 7.18.5). These include elastic, semiempir-ical elastic, and load-transfer solutions for singleshafts drilled in cohesive or cohesionless soils. (H.G. Poulos and E. H. Davis, “Elastic Solutions forSoil and Rock Mechanics,” John Wiley & Sons, Inc.,New York; A. S. Vesic, “Principles of Pile Founda-tion Design,” Soil Mechanics Series no. 38, DukeUniversity, Durham, N.C., 1975; H. Y. Fang, “Foun-dation Engineering Handbook,” 2nd ed., Van Nos-trand Reinhold, New York.)

Resistance to tensile and lateral loads bystraight-shaft drilled shafts should be evaluated asdescribed for pile foundations (see Art. 7.18). For rel-atively rigid shafts with characteristic length Tgreater than 3, there is evidence that bells willincrease the lateral resistance. The added ultimateresistance to uplift of a belled shaft Qut can beapproximately evaluated for cohesive soils modelsfor bearing capacity [Eq. (7.54)] and friction cylinder[Eq. (7.55)] as a function of the shaft diameter D andbell diameter Db. (G. G. Meyerhof and J. I. Adams,



“The Ultimate Uplift Capacity of Foundations,”Canadian Geotechnical Journal, vol. 5, no. 4, 1968.)

For the bearing-capacity solution,

(7.54)

The shear-strength reduction factor ω in Eq. (7.54)considers disturbance effects and ranges from ½(slurry construction) to ¾ (dry construction). curepresents the undrained shear strength of the soiljust above the bell surface, and NC is a bearing-capacity factor [see Eq. (7.16)] .

The failure surface of the friction cylinder modelis conservatively assumed to be vertical, startingfrom the base of the bell. Qut can then be determinedfor both cohesive and cohesionless soils from

(7.55)

where fut is the average ultimate skin-friction stressin tension developed on the failure plane; that is,fut = 0.8 c–u for clays or Kσ–′νo tan φ for sands [see Eqs.(7.37) and (7.40)]. Ws and Wp represent the weightof soil contained within the failure plane and theshaft weight, respectively.

7.21.8 Rock-Supported Shafts

Drilled shafts may be designed to be supported onrock or to be socketed into rock. Except for long,relatively small-diameter (comparatively com-pressible) shafts, conventional design ignores theskin friction of belled or straight shafts founded onrelatively incompressible materials. Where shaftsare socketed in rock, the design capacity is consid-ered a combination of the sidewall shearing resis-tance (bond) and the end bearing of the socket. Inpractice, both end-bearing and rock-socket designsare usually based on local experience (precedence)or presumptive bearing values contained in build-ing codes.

Bearing values on rock given in building codestypically range from 50 to 100 tons/ft2 for massivecrystalline rock, 20 to 50 tons/ft2 for sound foliatedrock, 15 to 25 tons/ft2 for sound sedimentary rock,8 to 10 tons/ft2 for soft and broken rock, and 4 to 8tons/ft2 for soft shales.

The supporting ability of a specific rock type isprimarily dependent on the frequency, orienta-tion, and size of the discontinuities within the rockmass and the degree of weathering of the rock


minerals. Consequently, application of presump-tive bearing values is not recommended withoutspecific local performance correlations. (R. W.Woodward, W. S. Gardner, and D. M. Greer,“Drilled Pier Foundations,” McGraw-Hill BookCompany, New York.)

Some analyses relate the bearing values qu injointed rock to the uniaxial compressive (UC)strength of representative rock cores. These analy-ses indicate that qu should not be significantly lessthan UC, possibly excluding weak sedimentaryrocks such as compacted shales and siltstones.With a safety factor of 3, the maximum allowablebearing value qa can be taken as: qa < 0.3UC. Inmost instances, however, the compressibility ofthe rock mass rather than rock strength governs.Elastic solutions can be used to evaluate the settle-ment of shafts bearing on rock if appropriatedeformation moduli of the rock mass Er can bedetermined. (H. G. Poulos and E. H. Davis, “Elas-tic Solutions for Soil and Rock Mechanics,” JohnWiley & Sons, Inc., New York; D. U. Deere, A. J.Hendron, F. D. Patton, and E. J. Cording, “Break-age of Rock,” Eighth Symposium on Rock Mechanics,American Institute of Mining and MetallurgicalEngineers, Minneapolis, Minn., 1967; F. H. Kul-hawy, “Geotechnical Model for Rock FoundationSettlement,” ASCE Journal of Geotechnical Engineer-ing Division, vol. 104, no. GT2, 1978.)

Concrete-rock bond stresses fR used for thedesign of rock sockets have been empiricallyestablished from a limited number of load tests.Typical values range from 70 to 200 psi, increasingwith rock quality. For good-quality rock, fR may berelated to the 28-day concrete strength f′c and tothe uniaxial compressive (UC) strength of rockcores. For rock with RQD > 50% (Table 7.3), fR canbe estimated as 0.05f′c or 0.05UC strength,whichever is smaller, except that fR should notexceed 250 psi. As shown by Fig. 7.31, the ultimaterock-concrete bond fRu is significantly higher thanfR except for very high UC values. (P. Rosenbergand N. L. Journeaux, “Friction and End-BearingTests on Bedrock for High-Capacity SocketDesign,” Canadian Geotechnical Journal, vol. 13, no.3, 1976.)

Design of rock sockets is conventionally basedon

(7.56)



Fig. 7.31 Chart relates the bond of a rock socket to the unconfined compressive strength of cores.

where Qd = allowable design load on rocksocket

ds = socket diameter

Ls = socket length

fR = allowable concrete-rock bond stress

qa = allowable bearing pressure on rock

Load-distribution measurements show, however,that much less of the load goes to the base than isindicated by Eq. (7.56). This behavior is demon-strated by the data in Table 7.12, where Ls /ds is theratio of the shaft length to shaft diameter and Er /Epis the ratio of rock modulus to shaft modulus. The


finite-element solution summarized in Table 7.12probably reflects a realistic trend if the averagesocket-wall shearing resistance does not exceed theultimate fR value; that is, slip along the socket side-wall does not occur.

A simplified design approach, taking intoaccount approximately the compatibility of thesocket and base resistance, is applied as follows:

1. Proportion the rock socket for design load Qdwith Eq. (7.56) on the assumption that the end-bearing stress is less than qa [say qa /4, which isequivalent to assuming that the base load Qb =(π/4) ds

2qa /4].



2. Calculate Qb = RQd, where R is the base-loadratio interpreted from Table 7.12.

3. If RQd does not equal the assumed Qb, repeat theprocedure with a new qa value until an approxi-mate convergence is achieved and q < qa.

The final design should be checked against theestablished settlement tolerance of the drilledshaft. (B. Ladanyi, discussion of “Friction and End-Bearing Tests on Bedrock,” Canadian GeotechnicalJournal, vol. 14, no. 1, 1977; H. G. Poulos and E. H.Davis, “Elastic Solutions for Rock and Soil Mechan-ics,” John Wiley & Sons, Inc., New York.)

Following the recommendations of Rosenbergand Journeaux, a more realistic solution by theabove method is obtained if fRu is substituted for fR.Ideally, fRu should be determined from load tests. Ifthis parameter is selected from Fig. 7.31 or fromother data that are not site-specific, a safety factorof at least 1.5 should be applied to fRu in recognitionof the uncertainties associated with the UCstrength correlations. (P. Rosenberg and N. L.Journeaux, “Friction and End-Bearing Tests onBedrock for High-Capacity Socket Design,” Canadi-an Geotechnical Journal, vol. 13, no. 3, 1976.)

7.21.9 Testing of Drilled Shafts

Static-load capacity of drilled shafts may be veri-fied by either static-load or dynamic-load testing(Arts. 7.18 and 7.19). Testing by applying staticloads on the shaft head (conventional static-loadtest) or against the toe (Osterberg cell) providesinformation on shaft capacity and general behav-ior. Dynamic-load testing in which pile-head forceand velocity under the impact of a falling weight

Er /Ep

0.25 1.0 4.0

0.5 54* 48 441.0 31 23 181.5 17* 12 8*2.0 13* 8 4

* Estimated by interpretation of finite-element solution; forPoisson’s ratio = 0.26.

Table 7.12 Percent of Base Load Transmitted toRock Socket

Ls /ds


are measured with a Pile Driving Analyzer andsubsequent analysis with the CAPWAP method(Art. 7.19.3) provide information on the static-loadcapacity and shaft-movement and shaft-soil load-transfer relationships of the shaft.

Structural integrity of a drilled shaft may beassessed after excavation or coring through the shaft.Low-strain dynamic-load testing with a Pile Integri-ty Tester (Art. 7.19.4), offers many advantages, how-ever. Alternative integrity evaluation methods areparallel seismic or cross-hole sonic logging.

For parallel seismic testing, a small casing isinserted into the ground near the tested shaft and toa greater depth than the shaft length. A hydrophoneis lowered into the casing to pick up the signalsresulting from blows on the shaft head from a smallhand-held hammer. Inasmuch as wave velocity inthe soil and shaft are different, the unknown lengthof the pile can be discerned from a series of mea-surements. One limitation of this method is the needto bore a hole adjacent to the shaft to be tested.

Cross-hole testing requires two full-length longi-tudinal access tubes in the shaft. A transmitter is low-ered through one of the tubes to send a signal to areceiver lowered into the other tube. The arrival timeand magnitude of the received signal are interpretedto assess the integrity of the shaft between the twotubes. For large-diameter shafts, more than two tubesmay be needed for thorough shaft evaluation. A dis-advantage of this method is the need to form two ormore access tubes in the shaft during construction.Furthermore, random testing or evaluations of exist-ing shafts may not be possible with this method.

(C. L. Crowther, “Load Testing of Deep Founda-tions,” John Wiley & Sons, Inc., New York.)

Retaining Methods ForExcavationThe simplest method of retaining the sides of anexcavation in soil is to permit the soil to form a nat-ural slope that will be stable even in the presenceof water. When there is insufficient space for thisinside the excavation or when the excavation sidesmust be vertical, construction such as thatdescribed in the following must be used.

7.22 CaissonsLoad-bearing enclosures known as caissons areformed in the ground, usually to protect excavation



for a foundation, aid construction of the substruc-ture, and serve as part of the permanent structure.Sometimes, a caisson is used to enclose a subsurfacespace to be used for such purposes as a pump well,machinery pit, or access to a deeper shaft or tunnel.Several caissons may be aligned to form a bridgepier, bulkhead, seawall, foundation wall for a build-ing, or impervious core wall for an earth dam.

For foundations, caissons are used to facilitateconstruction of shafts or piers extending from nearthe surface of land or water to a bearing stratum.This type of construction can carry heavy loads togreat depths. Built of common structural materials,they may have any shape in cross section. Theyrange in size from about that of a pile to over 100 ftin length and width. Some small ones are consid-ered bored, or caisson, piles (Art. 7.15.2). For someof the smaller caissons used for foundation shafts,the casing generally is not assigned any load-car-rying capacity, or it may be withdrawn as concretefills the hole.

Caissons often are installed by sinking themunder their own weight or with a surcharge. Theoperation is assisted by jacking, jetting, excavating,and undercutting. Care must be taken during thisoperation to maintain alignment. The caissons maybe built up as they sink, to permit construction tobe carried out at the surface, or they may be com-pletely prefabricated. Types of caissons used forfoundation work are as follows:

Chicago caissons are used for constructingfoundation shafts through a thick layer of clay tohardpan or rock. The method is useful where thesoil is sufficiently stiff to permit excavation forshort distances without caving. A circular pit about5 ft deep is dug and lined with wood staves. Thisvertical lagging is braced with two rings made withsteel channels. Then, 5 ft of soil is removed, andthe operation is repeated. If the ground is poor,shorter lengths are dug until the bearing stratum isreached. If necessary, the caissons can be belled atthe bottom to carry large loads. Finally, the hole isfilled with concrete. Minimum economical diame-ter for hand digging is 4 ft.

Sheeted piers or caissons are similarly con-structed, but the vertical lagging of wood or steel isdriven down during or before excavation. This sys-tem usually is used for shallow depths in wetground.

In dry ground, horizontal wood sheeting maybe used. This is economical and necessary where


there is inadequate vertical clearance. Louveredconstruction should be used to provide drainageand to permit packing behind the wood sheetingwhere soil will not maintain a vertical face longenough to permit insertion of the next sheet. Thistype of construction requires over-excavating sothat the wood sheets can be placed. Openingsmust be wide enough between sheets to allowbackfilling and tamping, to correct the excavationirregularities and equalize pressure on all sides.Small blocks may be inserted between successivesheets to leave packing gaps. If the excavation islarge, soldier beams, vertical cantilevers, can bedriven to break up the long sheeting spans.

Benoto caissons up to 39 in in diameter may besunk through water-bearing sands, hardpan, andboulders to depths of 150 ft. Excavation is donewith a hammer grab, a single-line orange-peelbucket, inside a temporary, cylindrical steel casing.The hammer grab is dropped to cut into or breakup the soil. After impact, the blades close aroundthe soil. Then, the bucket is lifted out and dis-charged. Boulders are broken up with heavy, per-cussion-type drills. Rock is drilled out by churndrills. To line the excavation, a casing is boltedtogether in 20-ft-deep sections, starting with a cut-ting edge. A hydraulic attachment oscillates thecasing continuously to ease sinking and withdraw-al, while jacks force the casing into the ground. Asconcrete is placed, the jacks withdraw the casing ina way that allows concreting of the caisson. Beno-to caissons are slower to place and more expensivethan drilled shafts, except in wet granular materialand where soil conditions are too tough for augersor rotating-bucket diggers.

Open caissons (Fig. 7.32) are enclosures with-out top and bottom during the lowering process.When used for pump wells and shafts, they oftenare cylindrical. For bridge piers, these caissons usu-ally are rectangular and compartmented. The com-partments serve as dredging wells, pipe passages,and access shafts. Dredging wells usually have 12-to 16-ft clear openings to facilitate excavation withclamshell or orange-peel buckets.

An open caisson may be a braced steel shell thatis filled with concrete, except for the wells, as it issunk into place. Or a caisson may be constructedentirely of concrete.

Friction along the caisson sides may range from300 to over 1000 lb/ft2. So despite steel cuttingedges at the wall bottoms, the caisson may not



Fig. 7.32 Construction with an open concrete caisson.

sink. Water and compressed-air jets may be used tolubricate the soil to decrease the friction. For thatpurpose, vertical jetting pipes should be embed-ded in the outer walls.

If the caisson does not sink under its ownweight with the aid of jets when soil within hasbeen removed down to the cutting edge, the cais-son must be weighted. One way is to build it high-er, to its final height, if necessary. Otherwise, aplatform may have to be built on top and weightspiled on it, a measure that can be expensive.

Care must be taken to undercut the edges even-ly, or the caisson will tip. Obstructions and varia-tions in the soil also can cause uneven sinking.

When the caisson reaches the bearing strata,the bottom is plugged with concrete (Fig. 7.40b).The plug may be placed by tremie or made byinjecting grout into the voids of coarse aggregate.


When a caisson must be placed through water,marine work sometimes may be converted to aland job by construction of a sand island. Fill isplaced until it projects above the water surface.Then, the caisson is constructed and sunk as usualon land.

Pneumatic caissons contain at the base a work-ing chamber with compressed air at a pressureequal to the hydrostatic pressure of the water inthe soil. Without the balancing pressure, the waterwould force soil from below up into a caisson. Aworking chamber clear of water also permits handwork to remove obstructions that buckets, air lifts,jets, and divers cannot. Thus, the downwardcourse of the caisson can be better controlled. Butsinking may be slower and more expensive, andcompressed-air work requires precautions againstsafety and health hazards.



Access to the working chamber for workers,materials, and equipment is through air locks, usu-ally placed at the top of the caisson (Fig. 7.33). Steelaccess cylinders 3 ft in diameter connect the airlocks with the working chamber in large caissons.

Entrance to the working chamber requires onlya short stay for a worker in an air lock. But thereturn stop may be lengthy, depending on thepressure in the chamber, to avoid the bends, orcaisson disease, which is caused by air bubbles inmuscles, joints, and the blood. Slow decompres-sion gives the body time to eliminate the excess air.In addition to slow decompression, it is necessaryto restrict the hours worked at various pressuresand limit the maximum pressure to 50 psi aboveatmospheric or less. The restriction on pressurelimits the maximum depth at which compressed-air work can be done to about 115 ft. A medical, orrecompression, lock is also required on the site fortreatment of workers attacked by the bends.

Floating caissons are used when it is desirableto fabricate caissons on land, tow them into posi-tion, and sink them through water. They are con-structed much like open or pneumatic caissons butwith a “false” bottom, “false” top, or buoyant cells.When floated into position, a caisson must be keptin alignment as it is lowered. A number of meansmay be used for the purpose, including anchors,templates supported on temporary piles, anchoredbarges, and cofferdams. Sinking generally is accom-plished by adding concrete to the walls. When thecutting edges reach the bottom, the temporarybulkheads at the base, or false bottoms, areremoved since buoyancy no longer is necessary.With false tops, buoyancy is controlled with com-pressed air, which can be released when the caissonsits on the bottom. With buoyant cells, buoyancy isgradually lost as the cells are filled with concrete.

Closed-box caissons are similar to floating cais-sons, except the top and bottom are permanent.Constructed on land, of steel or reinforced concrete,they are towed into position. Sometimes, the site canbe dredged in advance to expose soil that can safelysupport the caisson and loads that will be imposedon it. Where loads are heavy, however, this may notbe practicable; then, the box caisson may have to besupported on piles, but allowance can be made forits buoyancy. This type of caisson has been used forbreakwaters, seawalls, and bridge-pier foundations.

Potomac caissons have been used in wide tidalrivers with deep water underlain by deep, soft


deposits of sand and silt. Large timber mats areplaced on the river bottom, to serve as a templatefor piles and to retain tremie concrete. Long, steelpipe or H piles are driven in clusters, vertical andbattered, as required. Prefabricated steel or con-crete caissons are set on the mat over the pile clus-ters, to serve as permanent forms for concreteshafts to be supported on the piles. Then, concreteis tremied into the caissons. Since the caissons areused only as forms, construction need not be soheavy as for conventional construction, wherethey must withstand launching and sinking stress-es, and cutting edges are not required.

(H. Y. Fang, “Foundation Engineering Hand-book,” 2nd ed., Van Nostrand Reinhold Company,New York.)

7.23 Dikes and CribsEarth dikes, when fill is available, are likely to bethe cheapest means for keeping water out of anexcavation. If impervious material is not easilyobtained, however, a steel sheetpile cutoff wallmay have to be driven along the dike, to permit

Fig. 7.33 Pneumatic caisson. Pressure in work-ing chamber is above atmospheric.



pumps to handle the leakage. With an imperviouscore in the dike, wellpoints, deep-well pumps, orsumps and ditches may be able to keep the excava-tion unwatered.

Timber cribs are relatively cheap excavationenclosures. Built on shore, they can be floated to thesite and sunk by filling with rock. The water sidemay be faced with wood boards for watertightness(Fig. 7.34). For greater watertightness, two lines ofcribs may be used to support two lines of woodsheeting between which clay is tamped to form a“puddle” wall. Design of timber cribs should pro-vide ample safety against overturning and sliding.

7.24 CofferdamsTemporary walls or enclosures for protecting anexcavation are called cofferdams. Generally, one ofthe most important functions is to permit work tobe carried out on a nearly dry site.

Cofferdams should be planned so that they canbe easily dismantled for reuse. Since they are tem-porary, safety factors can be small, 1.25 to 1.5, whenall probable loads are accounted for in the design.But design stresses should be kept low whenstresses, unit pressure, and bracing reactions are

Fig. 7.34 Timber cr


uncertain. Design should allow for constructionloads and the possibility of damage from construc-tion equipment. For cofferdams in water, thedesign should provide for dynamic effect of flow-ing water and impact of waves. The height of thecofferdam should be adequate to keep out floodsthat occur frequently.

7.24.1 Double-Wall Cofferdams

These may be erected in water to enclose largeareas. Double-wall cofferdams consist of two linesof sheetpiles tied to each other; the space betweenis filled with sand (Fig. 7.35). For sheetpiles drivento irregular rock, or gravel, or onto boulders, thebottom of the space between walls may be pluggedwith a thick layer of tremie concrete to seal gapsbelow the tips of the sheeting. Double-wall coffer-dams are likely to be more watertight than single-wall ones and can be used to greater depths.

A berm may be placed against the outside faceof a cofferdam for stability. If so, it should be pro-tected against erosion. For this purpose, riprap,woven mattresses, streamline fins or jetties, orgroins may be used. If the cofferdam rests on rock,a berm needs to be placed on the inside only ifrequired to resist sliding, overturning, or shearing.

ib with stone filling.



Fig. 7.35 Double-wall cofferdam.

On sand, an ample berm must be provided so thatwater has a long path to travel to enter the coffer-dam (Fig. 7.35). (The amount of percolation is pro-portional to the length of path and the head.) Oth-erwise, the inside face of the cofferdam may settle,and the cofferdam may overturn as water perco-lates under the cofferdam and causes a quick, orboiling, excavation bottom. An alternative to awide berm is wider spacing of the cofferdam walls.This is more expensive but has the added advan-tage that the top of the fill can be used by con-struction equipment and for construction plant.

Fig. 7.36 Cellular s


7.24.2 Cellular Cofferdams

Used in construction of dams, locks, wharves, andbridge piers, cellular cofferdams are suitable forenclosing large areas in deep water. These enclo-sures are composed of relatively wide units. Aver-age width of a cellular cofferdam on rock shouldbe 0.70 to 0.85 times the head of water against theoutside. When constructed on sand, a cellular cof-ferdam should have an ample berm on the insideto prevent the excavation bottom from becomingquick (Fig. 7.36d).

heetpile cofferdam.



Fig. 7.37 Soldier beams and wood lagging retain the sides of an excavation.

Steel sheetpiles interlocked form the cells. Onetype of cell consists of circular arcs connected bystraight diaphragms (Fig. 7.36a). Another typecomprises circular cells connected by circular arcs(Fig. 7.36b). Still another type is the cloverleaf, com-posed of large circular cells subdivided by straightdiaphragms (Fig. 7.36c). The cells are filled withsand. The internal shearing resistance of the sandcontributes substantially to the strength of the cof-ferdam. For this reason, it is unwise to fill a coffer-dam with clay or silt. Weepholes on the insidesheetpiles drain the fill, thus relieving the hydro-static pressure on those sheets and increasing theshear strength of the fill.

In circular cells, lateral pressure of the fill caus-es only ring tension in the sheetpiles. Maximumstress in the pile interlocks usually is limited to8000 lb/lin in. This in turn limits the maximumdiameter of the circular cells. Because of numerousuncertainties, this maximum generally is set at 60ft. When larger-size cells are needed, the cloverleaftype may be used.

Circular cells are preferred to the diaphragmtype because each circular cell is a self-supporting


unit. It may be filled completely to the top beforeconstruction of the next cell starts. (Unbalancedfills in a cell may distort straight diaphragms.)When a circular cell has been filled, the top may beused as a platform for construction of the next cell.Also, circular cells require less steel per linear footof cofferdam. The diaphragm type, however, maybe made as wide as desired.

When the sheetpiles are being driven, care mustbe taken to avoid breaking the interlocks. The sheet-piles should be accurately set and plumbed against astructurally sound template. They should be drivenin short increments, so that when uneven bedrockor boulders are encountered, driving can be stoppedbefore the cells or interlocks are damaged. Also, allthe piles in a cell should be started until the cell isringed. This can reduce jamming troubles with thelast piles to be installed for the cell.

7.24.3 Single-Wall Cofferdams

These form an enclosure with only one line ofsheeting. If there will be no water pressure on thesheeting, they may be built with soldier beams



(piles extended to the top of the enclosure) andhorizontal wood lagging (Fig. 7.37). If there will bewater pressure, the cofferdam may be constructedof sheetpiles. Although they require less wallmaterial than double-wall or cellular cofferdams,single-wall cofferdams generally require bracingon the inside. Also, unless the bottom is driven intoa thick, impervious layer, they may leak excessive-ly at the bottom. There may also be leakage atinterlocks. Furthermore, there is danger of flood-ing and collapse due to hydrostatic forces whenthese cofferdams are unwatered.

For marine applications, therefore, it is advan-tageous to excavate, drive piles, and place a seal oftremie concrete without unwatering single-wallsheetpile cofferdams. Often, it is advisable to pre-dredge the area before the cofferdam is construct-ed, to facilitate placing of bracing and to removeobstructions to pile driving. Also, if blasting is nec-essary, it would severely stress the sheeting andbracing if done after they were installed.

For buildings, single-wall cofferdams must becarefully installed. Small movements and conse-quent loss of ground usually must be prevented toavoid damaging neighboring structures, streets,and utilities. Therefore, the cofferdams must beamply braced. Sheeting close to an existing struc-ture should not be a substitute for underpinning.

Bracing n Cantilevered sheetpiles may beused for shallow single-wall cofferdams in water oron land where small lateral movement will not betroublesome. Embedment of the piles in the bot-tom must be deep enough to insure stability.Design usually is based on the assumptions thatlateral passive resistance varies linearly with depthand the point of inflection is about two-thirds theembedded length below the surface. In general,however, cofferdams require bracing.

Cofferdams may be braced in many ways. Fig-ure 7.38 shows some commonly used methods.Circular cofferdams may be braced with horizontalrings (Fig. 7.38a). For small rectangular cofferdams,horizontal braces, or wales, along sidewalls andend walls may be connected to serve only as struts.For larger cofferdams, diagonal bracing (Fig. 7.38b)or cross-lot bracing (Fig. 7.38d and e) is necessary.When space is available at the top of an excavation,pile tops can be anchored with concrete dead men(Fig. 7.38c). Where rock is close, the wall can be tiedback with tensioned wires or bars that are


anchored in grouted sockets in the rock (Fig. 7.39).See also Art. 7.39.4.

Horizontal cross braces should be spaced to min-imize interference with excavation, form construc-tion, concreting, and pile driving. Spacing of 12 and18 ft is common. Piles and wales selected should bestrong enough as beams to permit such spacing. Inmarine applications, divers often have to install thewales and braces underwater. To reduce the amountof such work, tiers of bracing may be prefabricatedand lowered into the cofferdam from falsework orfrom the top set of wales and braces, which isinstalled above the water surface. In some cases, itmay be advantageous to prefabricate and erect thewhole cage of bracing before the sheetpiles are dri-ven. Then, the cage, supported on piles, can servealso as a template for driving the sheetpiles.

All wales and braces should be forced into bear-ing with the sheeting by wedges and jacks.

When pumping cannot control leakage into acofferdam, excavation may have to be carried outin compressed air. This requires a sealed workingchamber, access shafts, and air locks, as for pneu-matic caissons (Art. 7.22). Other techniques, suchas use of a tremie concrete seal or chemical solidifi-cation or freezing of the soil, if practicable, howev-er, will be more economical.

Braced sheetpiles may be designed as continu-ous beams subjected to uniform loading for earthand to loading varying linearly with depth forwater (Art. 7.26). (Actually, earth pressure dependson the flexibility of the sheeting and relative stiff-ness of supports.) Wales may be designed for uni-form loading. Allowable unit stresses in the wales,struts, and ties may be taken at half the elastic limitfor the materials because the construction is tem-porary and the members are exposed to view. Dis-tress in a member can easily be detected and reme-dial steps taken quickly.

Soldier beams and horizontal wood sheetingare a variation of single-wall cofferdams often usedwhere impermeability is not required. The soldierbeams, or piles, are driven vertically into theground to below the bottom of the proposed exca-vation. Spacing usually ranges from 5 to 10 ft (Table7.13). (The wood lagging can be used in the thick-nesses shown in Table 7.13 because of arching ofthe earth between successive soldier beams.)

As excavation proceeds, the wood boards areplaced horizontally between the soldiers (Fig.7.37). Louvers or packing spaces, 1 to 2 in high, are



left between the boards so that earth can betamped behind them to hold them in place. Hay

Nominal In Well- In CohesiveThickness of Drained Soils with LowSheeting, in Soils Shear Resistance

2 5 4.53 8.5 64 10 8

Table 7.13 Usual Maximum Spans of HorizontalSheeting with Soldier Piles, ft

Fig. 7.38 Types of cofferdam bracing include compwales, and tiebacks.


may also be stuffed behind the boards to keep theground from running through the gaps. The lou-vers permit drainage of water, to relieve hydrosta-tic pressure on the sheeting and thus allow use ofa lighter bracing system. The soldiers may bebraced directly with horizontal or inclined struts;or wales and braces may be used.

Advantages of soldier-beam constructioninclude fewer piles; the sheeting does not have toextend below the excavation bottom, as do sheet-piles; and the soldiers can be driven more easily inhard ground than can sheetpiles. Varying the

ression rings; bracing, diagonal (rakers) or cross lot;



Fig. 7.39 Vertical section shows prestressed tiebacks for soldier beams.

spacing of the soldiers permits avoidance ofunderground utilities. Use of heavy sections forthe piles allows wide spacing of wales and braces.But the soldiers and lagging, as well as sheetpiles,are no substitute for underpinning; it is necessaryto support and underpin even light adjoiningstructures.

Liner-plate cofferdams may be used for exca-vating circular shafts. The plates are placed in hor-izontal rings as excavation proceeds. Stampedfrom steel plate, usually about 16 in high and 3 ftlong, light enough to be carried by one person,liner plates have inward-turned flanges along alledges. Top and bottom flanges provide a seat forsuccessive rings. End flanges permit easy boltingof adjoining plates in a ring. The plates also arecorrugated for added stiffness. Large-diameter cof-


ferdams may be constructed by bracing the linerplates with steel beam rings.

Vertical-lagging cofferdams, with horizontal-ring bracing, also may be used for excavating cir-cular shafts. The method is similar to that used forChicago caissons (Art. 7.22). It is similarly restrictedto soils that can stand without support in depths of3 to 5 ft for a short time.

Slurry trenches may be used for constructingconcrete walls. The method permits building a wallin a trench without the earth sides collapsing. Whileexcavation proceeds for a 24- to 36-in-wide trench,the hole is filled with a bentonite slurry with a spe-cific gravity of 1.05 to 1.10 (Fig. 7.40a). The fluid pres-sure against the sides and caking of bentonite on thesides prevent the earth walls of the trench from col-lapsing. Excavation is carried out a section at a time.



Fig. 7.40 Slurry-trench method for constructing a continuous concrete wall: (a) Excavating one section;(b) concreting one section while another is being excavated.



A section may be 20 ft long and as much as 100 ftdeep. When the bottom of the wall is reached in asection, reinforcing is placed in that section. (Testshave shown that the bond of the reinforcing to con-crete is not materially reduced by the bentonite.)Then, concrete is tremied into the trench, replacingthe slurry, which may flow into the next section to beexcavated or be pumped into tanks for reuse in thenext section (Fig. 7.40b). The method has been usedto construct cutoffs for dams, cofferdams, founda-tions, walls of buildings, and shafts.

7.25 Soil SolidificationChemical grouting may be used to solidify soils.Although it is effective in water-bearing soils, themethod is not suitable in silts and clays, which areimpervious to the chemicals. One method, the Joost-en process, calls for successive injection of sodiumsilicate and calcium chloride, which react to form ahard, insoluble, calcium silicate gel; In sand, theresulting product resembles sandstone. Anothermethod is based on oxidation of lignin with achromium salt; a premixed single solution is inject-ed, and time for gel formation may be controlled.This solution offers the advantage of low viscosity,which enables the liquid to penetrate less permeablesoils. It is, however, expensive and has low strength.

Freezing is another means of solidifying water-bearing soils where obstructions or depth precludepile driving. It can be used for deep shaft excava-tions and requires little material for temporaryconstruction; the refrigeration plant has high sal-vage value. But freezing the soil may take a verylong time. Also, holes have to be drilled below thebottom of the proposed excavation for insertion ofrefrigeration pipes.

(L. White and E. A. Prentis, “Cofferdams,”Columbia University Press, New York; H. Y. Fang,“Foundation Engineering Handbook,” 2nd ed.,Van Nostrand Reinhold Company, New York.)

7.26 Lateral Active Pressureson Retaining Walls

Water exerts against a vertical surface a horizontalpressure equal to the vertical pressure. At anylevel, the vertical pressure equals the weight of a 1-ft2 column of water above that level. Hence, thehorizontal pressure p, lb/ft3 at any level is


(7.57)

where w = unit weight of water, lb/ft3

h = depth of water, ft

The pressure diagram is triangular (Fig. 7.41).Equation (7.57) also can be written

(7.58)

where K = pressure coefficient = 1.00.The resultant, or total, pressure, lb/lin ft, repre-

sented by the area of the hydrostatic-pressure dia-gram, is

(7.59)

It acts at a distance h/3 above the base of the trian-gle.

Soil also exerts lateral pressure. But the amountof this pressure depends on the type of soil, itscompaction or consistency, and its degree of satu-ration, and on the resistance of the structure to thepressure. Also, the magnitude of passive pressurediffers from that of active pressure.

Fig. 7.41 Pressure diagram for water.



Active pressure tends to move a structure inthe direction in which the pressure acts. Passive

pressure opposes motion of a structure.Free-standing walls retaining cuts in sand

tend to rotate slightly around the base. Behindsuch a wall, a wedge of sand ABC (Fig. 7.42a) tendsto shear along plane AC. C. A. Coulomb deter-mined that the ratio of sliding resistance to slidingforce is a minimum when AC makes an angle of 45°+ φ/2 with the horizontal, where φ is the angle ofinternal friction of the soil, deg.

For triangular pressure distribution (Fig. 7.42b),the active lateral pressure of a cohesionless soil at adepth h, ft, is

(7.60)

where Ka = coefficient of active earth pressure

w = unit weight of soil, lb/ ft3

The total active pressure, lb/lin ft, is

(7.61)

Because of frictional resistance to sliding at theface of the wall, Ea is inclined at an angle δ with thenormal to the wall, where δ is the angle of wall fric-

Fig. 7.42 Free-standing wall with sand backfill (a


tion, deg (Fig. 7.42a). If the face of the wall is vertical,the horizontal active pressure equals Ea cos δ. If theface makes an angle β with the vertical (Fig. 7.42a),the pressure equals Ea cos (δ + β ). The resultant actsat a distance of h/3 above the base of the wall.

If the ground slopes upward from the top of thewall at an angle α, deg, with the horizontal, thenfor cohesionless soils

(7.62)

The effect of wall friction on Ka is small and usual-ly is neglected. For δ = 0,

(7.63)

Table 7.14 lists values of Ka determined from Eq.(7.63). Approximate values of φ and unit weightsfor various soils are given in Table 7.15.

) is subjected to (b) triangular pressure distribution.



φ = 10° 15° 20° 25° 30° 35° 40°

α = 0 0.70 0.59 0.49 0.41 0.33 0.27 0.22α = 10° 0.97 0.70 0.57 0.47 0.37 0.30 0.24

β = 0 α = 20° — — 0.88 0.57 0.44 0.34 0.27α = 30° — — — — 0.75 0.43 0.32α = φ 0.97 0.93 0.88 0.82 0.75 0.67 0.59

α = 0 0.76 0.65 0.55 0.48 0.41 0.43 0.29α = 10° 1.05 0.78 0.64 0.55 0.47 0.38 0.32

β = 10° α = 20° — — 1.02 0.69 0.55 0.45 0.36α = 30° — — — — 0.92 0.56 0.43α = φ 1.05 1.04 1.02 0.98 0.92 0.86 0.79

α = 0 0.83 0.74 0.65 0.57 0.50 0.43 0.38α = 10° 1.17 0.90 0.77 0.66 0.57 0.49 0.43

β = 20° α = 20° — — 1.21 0.83 0.69 0.57 0.49α = 30° — — — — 1.17 0.73 0.59α = φ 1.17 1.20 1.21 1.20 1.17 1.12 1.06

α = 0 0.94 0.86 0.78 0.70 0.62 0.56 0.49α = 10° 1.37 1.06 0.94 0.83 0.74 0.65 0.56

β = 30° α = 20° — — 1.51 1.06 0.89 0.77 0.66α = 30° — — — — 1.55 0.99 0.79α = φ 1.37 1.45 1.51 1.54 1.55 1.54 1.51

Table 7.14 Active-Lateral-Pressure Coefficients Ka

Density or Angle of Internal Unit Weight w,Consistency Friction φ, deg lb/ft3

Coarse sand or sand and gravel Compact 40 140Loose 35 90

Medium sand Compact 40 130Loose 30 90

Fine silty sand or sandy silt Compact 30 130Loose 25 85

Uniform silt Compact 30 135Loose 25 85

Clay-silt Soft to medium 20 90–120

Silty clay Soft to medium 15 90–120

Clay Soft to medium 0–10 90–120

Table 7.15 Angles of Internal Friction and Unit Weights of Soils

Type of Soil



For level ground at the top of the wall (α = 0):

(7.64)

If in addition, the back face of the wall is vertical (β= 0), Rankine’s equation is obtained:

(7.65)

Coulomb derived the trigonometric equivalent:

(7.66)

When information on the value of the angle of wallfriction is not available, δ may be taken equal to φ/2for determining the horizontal component of Ea.

Note: Even light compaction may permanentlyincrease the earth pressure into the passive range.This may be compensated for in wall design by useof a safety factor of at least 2.5.

Fig. 7.43 Braced wall retaining sand (a) may have (b). (c) Uniform pressure distribution may be assumed


Unyielding walls retaining cuts in sand, such asthe abutment walls of a rigid-frame concrete bridgeor foundation walls braced by floors, do not allowshearing resistance to develop in the sand alongplanes that can be determined analytically. For suchwalls, triangular pressure diagrams may be assumed,and Ka may be taken equal to 0.5. But only sand orgravel should be permitted for the backfill, and com-paction should be light within 5 to 10 ft of the walls.See preceding Note for free-standing walls.

Braced walls retaining cuts in sand (Fig. 7.43a)are subjected to earth pressure gradually anddevelop resistance in increments as excavation pro-ceeds and braces are installed. Such walls tend torotate about a point in the upper portion. Hence,the active pressures do not vary linearly withdepth. Field measurements have yielded a varietyof curves for the pressure diagram, of which twotypes are shown in Fig. 7.43b. Consequently, someauthorities have recommended a trapezoidal pres-sure diagram, with a maximum ordinate

(7.67)

to resist pressure distributions of the type shown infor design.



Fig. 7.44 Assumed trapezoidal diagrams for a braced wall in soils described by boring logs in (a) and (d).

Ka may be obtained from Table 7.14. The total pres-sure exceeds that for a triangular distribution.

Figure 7.44 shows earth-pressure diagramsdeveloped for a sandy soil and a clayey soil. Inboth cases, the braced wall is subjected to a 3-ft-deep surcharge, and height of wall is 34 ft. For thesandy soil (Fig. 7.44a), Fig. 7.44b shows the pressurediagram assumed. The maximum pressure can beobtained from Eq. (7.67), with h = 34 + 3 = 37 ftand Ka assumed as 0.30 and w as 110 lb/ft3.

p1 = 0.8 × 0.3 × 110 × 37 = 975 lb/ft2

The total pressure is estimated as

P = 0.8 × 975 × 37 = 28,900 lb/lin ft

The equivalent maximum pressure for a trapezoidaldiagram for the 34-ft height of the wall then is


Assumption of a uniform distribution (Fig. 7.43c),however, simplifies the calculations and has little orno effect on the design of the sheeting and braces,which should be substantial to withstand construc-tion abuses. Furthermore, trapezoidal loading termi-nating at the level of the excavation may not apply ifpiles are driven inside the completed excavation.The shocks may temporarily decrease the passiveresistance of the sand in which the wall is embeddedand lower the inflection point. This would increasethe span between the inflection point and the lowestbrace and increase the pressure on that brace.Hence, uniform pressure distribution may be moreapplicable than trapezoidal for such conditions.

See Note for free-standing walls.



Flexible bulkheads retaining sand cuts are sub-jected to active pressures that depend on the fixityof the anchorage. If the anchor moves sufficientlyor the tie from the anchor to the upper portion ofthe bulkhead stretches enough, the bulkhead mayrotate slightly about a point near the bottom. Inthat case, the sliding-wedge theory may apply. Thepressure distribution may be taken as triangular,and Eqs. (7.60) to (7.66) may be used. But if theanchor does not yield, then pressure distributionsmuch like those in Fig. 7.43b for a braced cut mayoccur. Either a trapezoidal or uniform pressure dis-tribution may be assumed, with maximum pres-sure given by Eq. (7.67). Stresses in the tie should bekept low because it may have to resist unanticipat-ed pressures, especially those resulting from aredistribution of forces from soil arching. The safe-ty factor for design of ties and anchorages shouldbe at least twice that used in conventional design.

Free-standing walls retaining plastic-clay cuts

(Fig. 7.45a) may have to resist two types of activelateral pressure, both with triangular distribution.If the shearing resistance is due to cohesion only, aclay bank may be expected to stand with a verticalface without support for a height, ft, of

(7.68)

Fig. 7.45 Free-standing wall retaining clay (a) mayor (d). For mixed soils, the distribution may approxima


where 2c = unconfined compressive strength ofclay, lb/ft2

w = unit weight of clay, lb/ft3

So if there is a slight rotation of the wall about itsbase, the upper portion of the clay cut will standvertically without support for a depth h′. Belowthat, the pressure will increase linearly with depthas if the clay were a heavy liquid (Fig. 7.45b):

p = wh – 2c

The total pressure, lb/lin ft, then is

(7.69)

It acts at a distance (h – 2c /w) / 3 above the base ofthe wall. These equations assume wall friction iszero, the back face of the wall is vertical, and theground is level.

This condition is likely to be temporary. In time,the clay will consolidate. The pressure distributionmay become approximately triangular (Fig. 7.45d)from the top of the wall to the base. The pressuresthen may be calculated from Eqs. (7.60) to (7.66)with an apparent angle of internal friction for thesoil (for example, see the values of φ in Table 7.15).The wall should be designed for the pressures pro-

have to resist the pressure distribution shown in (b)te that shown in (c).



Fig. 7.46 Braced wall retaining clay (a) may have to resist pressures approximated by the pressure dis-tribution in (b) or (d). Uniform distribution (c) may be assumed in design.

ducing the highest stresses and overturningmoments.

Note: The finer the backfill material, the morelikely it is that pressures greater than active willdevelop, because of plastic deformations, water-levelfluctuation, temperature changes, and other effects.As a result, it would be advisable to use in design atleast the coefficient for earth pressure at rest:

(7.70)

The safety factor should be at least 2.5.Clay should not be used behind retaining walls,

where other economical alternatives are available.The swelling type especially should be avoidedbecause it can cause high pressures and progres-sive shifting or rotation of the wall.

For a mixture of cohesive and cohesionless

soils, the pressure distribution may temporarily beas shown in Fig. 7.45c. The height, ft, of the unsup-ported vertical face of the clay is

(7.71)

The pressure at the base is

(7.72)


The total pressure, lb/lin ft, is

(7.73)

It acts at a distance (h – h′′) /3 above the base of thewall.

Braced walls retaining clay cuts (Fig. 7.46a) alsomay have to resist two types of active lateral pres-sure. As for sand, the pressure distribution may tem-porarily be approximated by a trapezoidal diagram(Fig. 7.46b). On the basis of field observations, R. B.Peck has recommended a maximum pressure of

(7.74)

and a total pressure, lb/lin ft, of

(7.75)

[R. B. Peck, “Earth Pressure Measurements in OpenCuts, Chicago (Ill.) Subway,” Transactions, AmericanSociety of Civil Engineers, 1943, pp. 1008-1036.]

Figure 7.46c shows a trapezoidal earth-pressurediagram determined for the clayey-soil conditionof Fig. 7.46d. The weight of the soil is taken as 120lb/ft3; c is assumed as zero and the active-lateral-pressure coefficient as 0.3. Height of the wall is 34ft, surcharge 3 ft. Then, the maximum pressure,



obtained from Eq. (7.60) since the soil is clayey, notpure clay, is

From Eq. (7.75) with the above assumptions, thetotal pressure is

The equivalent maximum pressure for a trape-zoidal diagram for the 34-ft height of wall is

To simplify calculations, a uniform pressure dis-tribution may be used instead (Fig. 7.46c).

If after a time the clay should attain a consoli-dated equilibrium state, the pressure distributionmay be better represented by a triangular diagramABC (Fig. 7.46d), as suggested by G. P. Tschebotari-off. The peak pressure may be assumed at a dis-tance of kh = 0.4h above the excavation level for astiff clay; that is, k = 0.4. For a medium clay, k maybe taken as 0.25, and for a soft clay, as 0. For com-puting the pressures, Ka may be estimated fromTable 7.14 with an apparent angle of frictionobtained from laboratory tests or approximatedfrom Table 7.15. The wall should be designed forthe pressures producing the highest stresses andoverturning moments.

See also Note for free-standing walls.Flexible bulkheads retaining clay cuts and

anchored near the top similarly should be checkedfor two types of pressures. When the anchor is like-ly to yield slightly or the tie to stretch, the pressuredistribution in Fig. 7.46d with k = 0 may be applic-able. For an unyielding anchor, any of the pressuredistributions in Fig. 7.46 may be assumed, as for abraced wall. The safety factor for design of ties andanchorages should be at least twice that used inconventional design. See also Note for free-stand-ing walls.

Backfill placed against a retaining wall shouldpreferably be gravel to facilitate drainage. Also,weepholes should be provided through the wallnear the bottom and a drain installed along the


footing, to conduct water from the back of the walland prevent buildup of hydrostatic pressures.

Saturated or submerged soil imposes substan-tially greater pressure on a retaining wall than dryor moist soil. The active lateral pressure for a soil-fluid backfill is the sum of the hydrostatic pressureand the lateral soil pressure based on the buoyedunit weight of the soil. This weight roughly may be60% of the dry weight.

Surcharge, or loading imposed on a backfill,increases the active lateral pressure on a wall andraises the line of action of the total, or resultant,pressure. A surcharge ws , lb/ ft2, uniformly distrib-uted over the entire ground surface may be takenas equivalent to a layer of soil of the same unitweight w as the backfill and with a thickness ofws /w. The active lateral pressure, lb/ft2, due to thesurcharge, from the backfill surface down, thenwill be Kaws. This should be added to the lateralpressures that would exist without the surcharge.Ka may be obtained from Table 7.14.

(A. Caquot and J. Kérisel, “Tables for Calcula-tion of Passive Pressure, Active Pressure, and Bear-ing Capacity of Foundations,” Gauthier-Villars,Paris.)

7.27 Passive Lateral Pressureon Retaining Walls andAnchors

As defined in Art. 7.26, active pressure tends tomove a structure in the direction in which pressureacts, whereas passive pressure opposes motion of astructure.

Passive pressures of cohesionless soils, resistingmovement of a wall or anchor, develop because ofinternal friction in the soils. Because of frictionbetween soil and wall, the failure surface is curved,not plane as assumed in the Coulomb sliding-wedge theory (Art. 7.26). Use of the Coulomb the-ory yields unsafe values of passive pressure whenthe effects of wall friction are included.

Total passive pressure, lb/lin ft, on a wall oranchor extending to the ground surface (Fig. 7.47a)may be expressed for sand in the form

(7.76)



Fig. 7.47 Passive pressure on a wall (a) may vary as shown in (b) for sand or as shown in (c) for clay.

where Kp = coefficient of passive lateral pressure

w = unit weight of soil, lb/ft3

h = height of wall or anchor to groundsurface, ft

The pressure distribution usually assumed forsand is shown in Fig. 7.47b. Table 7.16 lists values ofKp for a vertical wall face (β = 0) and horizontalground surface (α = 0), for curved surfaces of fail-ure. (Many tables and diagrams for determiningpassive pressures are given in A. Caquot and J.Kérisel, “Tables for Calculation of Passive Pressure,Active Pressure, and Bearing Capacity of Founda-tions,” Gauthier-Villars, Paris.)

Since a wall usually transmits a downwardshearing force to the soil, the angle of wall frictionδ correspondingly is negative (Fig. 7.47a). Forembedded portions of structures, such asanchored sheetpile bulkheads, δ and the angle ofinternal friction φ of the soil reach their peak val-ues simultaneously in dense sand. For those condi-

φ = 10° 15° 20°

δ = 0 1.42 1.70 2.04δ = – φ/2 1.56 1.98 2.59δ = – φ 1.65 2.19 3.01

*For vertical wall face (β = 0) and horizontal ground surface (α =

Table 7.16 Passive Lateral-Pressure Coefficients Kp


tions, if specific information is not available, δ maybe assumed as – 2/3φ (for φ > 30°). For such struc-tures as a heavy anchor block subjected to a hori-zontal pull or thrust, δ may be taken as – φ/2 fordense sand. For those cases, the wall friction devel-ops as the sand is pushed upward by the anchorand is unlikely to reach its maximum value beforethe internal resistance of the sand is exceeded.

When wall friction is zero (δ = 0), the failuresurface is a plane inclined at an angle of 45° – φ/2with the horizontal. The sliding-wedge theorythen yields

(7.77)

When the ground is horizontal (α = 0):

(7.78)

25° 30° 35° 40°

2.56 3.00 3.70 4.603.46 4.78 6.88 10.384.29 6.42 10.20 17.50

0).

*



If, in addition, the back face of the wall is vertical(β = 0):

(7.79)

The first line of Table 7.16 lists values obtainedfrom Eq. (7.79).

Continuous anchors in sand (φ = 33°), whensubjected to horizontal pull or thrust, develop pas-sive pressures, lb/lin ft, of about

(7.80)

where h = distance from bottom of anchor tothe surface, ft.

This relationship holds for ratios of h to height d, ft,of anchor of 1.5 to 5.5, and assumes a horizontalground surface and vertical anchor face.

Square anchors within the same range of h/ddevelop about

(7.81)

where P = passive lateral pressure, lb

d = length and height of anchor, ft

Passive pressures of cohesive soils, resistingmovement of a wall or anchor extending to theground surface, depend on the unit weight of thesoil w and its unconfined compressive strength 2c,psf. At a distance h, ft, below the surface, the pas-sive lateral pressure, psf, is

(7.82)

The total pressure, lb/lin ft, is

(7.83)

and acts at a distance, ft, above the bottom of thewall or anchor of

The pressure distribution for plastic clay is shownin Fig. 7.47c.

Continuous anchors in plastic clay, when sub-jected to horizontal pull or thrust, develop passivepressures, lb/lin ft, of about


(7.84)

where h = distance from bottom of anchor tosurface, ft

d = height of anchor, ft

Equation (7.84) is based on tests made with hori-zontal ground surface and vertical anchor face.

Safety factors should be applied to the passivepressures computed from Eqs. (7.76) to (7.84) fordesign use. Experience indicates that a safety fac-tor of 2 is satisfactory for clean sands and gravels.For clay, a safety factor of 3 may be desirablebecause of uncertainties as to effective shearingstrength.

(G. P. Tschebotarioff, “Soil Mechanics, Founda-tions, and Earth Structures,” Mc-Graw-Hill BookCompany, New York; K. Terzaghi and R. B. Peck,“Soil Mechanics in Engineering Practice,” John Wiley& Sons, Inc., New York; Leo Casagrande, “Commentson Conventional Design of Retaining Structures,”ASCE Journal of Soil Mechanics and Foundations Engi-neering Division, 1973, pp. 181–198; H. Y. Fang, “Foun-dation Engineering Handbook,” 2nd ed., Van Nos-trand Reinhold Company, New York.)

7.28 Vertical Earth Pressureon Conduit

The vertical load on an underground conduitdepends principally on the weight of the prism ofsoil directly above it. But the load also is affectedby vertical shearing forces along the sides of thisprism. Caused by differential settlement of theprism and adjoining soil, the shearing forces maybe directed up or down. Hence, the load on theconduit may be less or greater than the weight ofthe soil prism directly above it.

Conduits are classified as ditch or projecting,depending on installation conditions that affectthe shears. A ditch conduit is a pipe set in a rela-tively narrow trench dug in undisturbed soil (Fig.7.48). Backfill then is placed in the trench up to theoriginal ground surface. A projecting conduit is apipe over which an embankment is placed.

A projecting conduit may be positive or nega-tive, depending on the extent of the embankmentvertically. A positive projecting conduit is installedin a shallow bed with the pipe top above the sur-



face of the ground. Then, the embankment isplaced over the pipe (Fig. 7.49a). A negative pro-jecting conduit is set in a narrow, shallow trenchwith the pipe top below the original ground sur-face (Fig. 7.49b). Then, the ditch is backfilled, afterwhich the embankment is placed. The load on theconduit is less when the backfill is not compacted.

Load on underground pipe also may bereduced by the imperfect-ditch method of con-struction. This starts out as for a positive projectingconduit, with the pipe at the original ground sur-face. The embankment is placed and compactedfor a few feet above the pipe. But then, a trench aswide as the conduit is dug down to it through thecompacted soil. The trench is backfilled with aloose, compressible soil (Fig. 7.49c). After that, theembankment is completed.

Fig. 7.48 Ditch conduit.

Fig. 7.49 Type of projecting condu


The load, lb/lin ft, on a rigid ditch conduit maybe computed from

(7.85)

and on a flexible ditch conduit from

(7.86)

where CD = load coefficient for ditch conduit

w = unit weight of fill, lb/ft3

h = height of fill above top of conduit, ft

b = width of ditch at top of conduit, ft

D = outside diameter of conduit, ft

From the equilibrium of vertical forces, includingshears, acting on the backfill above the conduit, CDmay be determined:

(7.87)

where e = 2.718

k = 2Ka tan θ

Ka = coefficient of active earth pressure[Eq. (7.66) and Table 7.14]

θ = angle of friction between fill andadjacent soil (θ < φ, angle of inter-nal friction of fill)

it depends on method of backfilling.



Table 7.17 gives values of CD for k = 0.33 for cohe-sionless soils, k = 0.30 for saturated topsoil, and k= 0.26 and 0.22 for clay (usual maximum and sat-urated).

Vertical load, lb/lin ft, on conduit installed bytunneling may be estimated from

(7.88)

where c = cohesion of the soil, or half the uncon-fined compressive strength of the soil, psf. Theload coefficient CD may be computed from Eq.(7.87) or obtained from Table 7.17 with b = maxi-mum width of tunnel excavation, ft, and h = dis-tance from tunnel top to ground surface, ft.

For a ditch conduit, shearing forces extend fromthe pipe top to the ground surface. For a projectingconduit, however, if the embankment is sufficient-ly high, the shear may become zero at a horizontalplane below grade, the plane of equal settlement.Load on a projecting conduit is affected by thelocation of this plane.

Vertical load, lb/lin ft, on a positive projectingconduit may be computed from

(7.89)

where CP = load coefficient for positive projectingconduit. Formulas have been derived for CP and thedepth of the plane of equal settlement. These formu-

h/b Cohesionless Soils Saturated

1 0.85 0.82 0.75 0.73 0.63 0.64 0.55 0.55 0.50 0.56 0.44 0.47 0.39 0.48 0.35 0.39 0.32 0.3

10 0.30 0.311 0.27 0.212 0.25 0.2

Over 12 3.0b/h 3.3

Table 7.17 Load Coefficients CD for Ditch Conduit


las, however, are too lengthy for practical application,and the computation does not appear to be justifiedby the uncertainties in actual relative settlement ofthe soil above the conduit. Tests may be made in thefield to determine CP. If so, the possibility of anincrease in earth pressure with time should be con-sidered. For a rough estimate, CP may be assumed as1 for flexible conduit and 1.5 for rigid conduit.

The vertical load, lb/lin ft, on negative project-ing conduit may be computed from

(7.90)

where CN = load coefficient for negative project-ing conduit

h = height of fill above top of conduit, ft

b = horizontal width of trench at top ofconduit, ft

The load on an imperfect ditch conduit may beobtained from

(7.91)

where D = outside diameter of conduit, ft.Formulas have been derived for CN , but they

are complex, and insufficient values are availablefor the parameters involved. As a rough guide, CNmay be taken as 0.9 when depth of cover exceedsconduit diameter. (Sec also Art. 10.31.)

Clay

Topsoil k = 0.26 k = 0.22

6 0.88 0.895 0.78 0.807 0.69 0.738 0.62 0.672 0.56 0.607 0.51 0.552 0.46 0.518 0.42 0.474 0.39 0.432 0.36 0.409 0.33 0.377 0.31 0.35b/h 3.9b/h 4.5b/h



Superimposed surface loads increase the load onan underground conduit. The magnitude of theincrease depends on the depth of the pipe belowgrade and the type of soil. For moving loads, animpact factor of about 2 should be applied. A super-imposed uniform load w′, lb/ft2, of large extent maybe treated for projecting conduit as an equivalentlayer of embankment with a thickness, ft, of w′/w.For ditch conduit, the load due to the soil should beincreased by bw′e–kh/b, where k = 2Katan θ, as in Eq.(7.87). The increase due to concentrated loads can beestimated by assuming the loads to spread out lin-early with depth, at an angle of about 30° with thevertical. (See also Art. 7.11).

(M. G. Spangler, “Soil Engineering,” Interna-tional Textbook Company, Scranton, Pa.; “Hand-book of Steel Drainage and Highway ConstructionProducts,” American Iron and Steel Institute,Washington, D. C.)

7.29 Dewatering Methods forExcavations

The main purpose of dewatering is to enable con-struction to be carried out under relatively dryconditions. Good drainage stabilizes excavatedslopes, reduces lateral loads on sheeting and brac-ing, and reduces required air pressure in tunnel-ing. Dewatering makes excavated material lighterand easier to handle. It also prevents loss of soilbelow slopes or from the bottom of the excavation,and prevents a “quick” or “boiling” bottom. Inaddition, permanent lowering of the groundwatertable or relief of artesian pressure may allow a lessexpensive design for the structure, especiallywhen the soil consolidates or becomes compact. Iflowering of the water level or pressure relief istemporary, however, the improvement of the soilshould not be considered in foundation design.Increases in strength and bearing capacity may belost when the soil again becomes saturated.

To keep an excavation reasonably dry, thegroundwater table should be kept at least 2 ft, andpreferably 5 ft, below the bottom in most soils.

Site investigations should yield informationuseful for deciding on the most suitable and eco-nomical dewatering method. Important is aknowledge of the types of soil in and below thesite, probable groundwater levels during construc-tion, permeability of the soils, and quantities of


water to be handled. A pumping test may be desir-able for estimating capacity of pumps needed anddrainage characteristics of the ground.

Many methods have been used for dewateringexcavations. Those used most often are listed inTable 7.18 with conditions for which they general-ly are most suitable. (See also Art. 7.36.)

In many small excavations, or where there aredense or cemented soils, water may be collected inditches or sumps at the bottom and pumped out.This is the most economical method of dewatering,and the sumps do not interfere with future con-struction as does a comprehensive wellpoint sys-tem. But the seepage may slough the slopes, unlessthey are stabilized with gravel, and may hold upexcavation while the soil drains. Also, springs maydevelop in fine sand or silt and cause undergrounderosion and subsidence of the ground surface.

For sheetpile-enclosed excavations in pervioussoils, it is advisable to intercept water before itenters the enclosure. Otherwise, the water will puthigh pressures on the sheeting. Seepage also cancause the excavation bottom to become quick,overloading the bottom bracing, or create piping,undermining the sheeting. Furthermore, pumpingfrom the inside of the cofferdam is likely to leavethe soil to be excavated wet and tough to handle.

Wellpoints often are used for lowering thewater table in pervious soils. They are not suitable,however, in soils that are so fine that they will flowwith the water or in soils with low permeability.Also, other methods may be more economical fordeep excavations, very heavy flows, or consider-able lowering of the water table (Table 7.18).

Wellpoints are metal well screens about 2 to 3 in indiameter and up to about 4 ft long. A pipe connectseach wellpoint to a header, from which water ispumped to discharge (Fig. 7.50). Each pump usually isa combined vacuum and centrifugal pump. Spacingof wellpoints generally ranges from 3 to 12 ft c to c.

A wellpoint may be jetted into position or set ina hole made with a hole puncher or heavy steelcasing. Accordingly, wellpoints may be self-jettingor plain-tip. To insure good drainage in fine anddirty sands or in layers of silt or clay, the wellpointand riser should be surrounded by sand to justbelow the water table. The space above the filtershould be sealed with silt or clay to keep air fromgetting into the wellpoint through the filter.

Wellpoints generally are relied on to lower thewater table 15 to 20 ft. Deep excavations may be



dewatered with multistage wellpoints, with onerow of wellpoints for every 15 ft of depth. Or whenthe flow is less than about 15 gal/min per well-point, a single-stage system of wellpoints may beinstalled above the water table and operated withjet-eductor pumps atop the wellpoints. Thesepumps can lower the water table up to about 100ft, but they have an efficiency of only about 30%.

Deep wells may be used in pervious soils fordeep excavations, large lowering of the watertable, and heavy water flows. They may be placedalong the top of an excavation to drain it, to inter-cept seepage before pressure makes slopes unsta-ble, and to relieve artesian pressure before itheaves the excavation bottom.

Saturated-Soil Conditions

Surface water

Gravel

Sand (except very fine sand)

Waterbearing strata near surface; water tabledoes not have to be lowered more than 15 ft

Waterbearing strata near surface; water table tobe lowered more than 15 ft, low pumping rate

Excavations 30 ft or more below water table;artesian pressure; high pumping rate; largelowering of water table—all where adequatedepth of pervious soil is available forsubmergence of well screen and pump

Sand underlain by rock near excavation bottom

Sand underlain by clay

Silt; very find sand (permeability coefficientbetween 0.01 and 0.0001 mm/s)

Silt or silty sand underlain by pervious soil

Clay-silts, silts

Clay underlain by pervious soil

Dense or cemented soils: small excavations

Table 7.18 Methods for Dewatering Excavations


Usual spacing of wells ranges from 20 to 250 ft.Diameter generally ranges from 6 to 20 in. Wellscreens may be 20 to 75 ft long, and they are sur-rounded with a sand-gravel filter. Generally, pump-ing is done with a submersible or vertical turbinepump installed near the bottom of each well.

Figure 7.51 shows a deep-well installation usedfor a 300-ft-wide by 600-ft-long excavation for abuilding of the Smithsonian Institution, Washington,D.C. Two deep-well pumps lowered the generalwater level in the excavation 20 ft. The well instal-lation proceeded as follows: (1) Excavation towater level (elevation 0.0). (2) Driving of sheetpilesaround the well area (Fig. 7.51a). (3) Excavationunderwater inside the sheetpile enclosure to ele-

Dewatering Method Probably Suitable

Ditches; dikes; sheetpiles and pumps or underwa-ter excavation and concrete tremie seal

Underwater excavation, grout curtain; gravitydrainage with large sumps with gravel filters

Gravity drainage

Wellpoints with vacuum and centrifugal pumps

Wellpoints with jet-eductor pumps

Deep wells, plus, if necessary, wellpoints

Wellpoints to rock, plus ditches, drains, automatic“mops”

Wellpoints in holes 3 or 4 into the clay, backfilledwith sand

For lifts up to 15 ft, wellpoints with vacuum; forgreater lifts, wells with vacuum; sumps

At top of excavation, and extending to the pervioussoil, vertical sand drains plus well points or wells

Electro-osmosis

At top of excavation, wellpoints or deep wellsextending to pervious soil

Ditches and sumps



Fig. 7.50 Wellpoint system for dewatering an excavation.

vation –37.0 ft (Fig. 7.51b). Bracing installed as dig-ging progressed. (4) Installation of a wire-mesh-wrapped timber frame, extending from elevation0.0 to – 37.0 (Fig. 7.51c). Weights added to sink theframe. (5) Backfilling of space between sheetpilesand mesh with 3/16- to 3/8-in gravel. (6) Removal ofsheetpiles. (7) Installation of pump and start ofpumping.

Vacuum well or wellpoint systems may be usedto drain silts with low permeability (coefficientbetween 0.01 and 0.0001 mm/s). In these systems,wells or wellpoints are closely spaced, and a vacu-um is held with vacuum pumps in the well screensand sand filters. At the top, the filter, well, and risersshould be sealed to a depth of 5 ft with bentonite oran impervious soil to prevent loss of the vacuum.Water drawn to the well screens is pumped out withsubmersible or centrifugal pumps.


Where a pervious soil underlies silts or siltysands, vertical sand drains and deep wells canteam up to dewater an excavation. Installed at thetop, and extending to the pervious soil, the sandpiles intercept seepage and allow it to drain downto the pervious soil. Pumping from the deep wellsrelieves the pressure in that deep soil layer.

For some silts and clay-silts, electrical drainagewith wells or wellpoints may work, whereas grav-ity methods may not (Art. 7.36). In saturated clays,thermal or chemical stabilization may be necessary(Arts. 7.37 and 7.38).

Small amounts of surface water may beremoved from excavations with “mops.” Sur-rounded with gravel to prevent clogging, thesedrains are connected to a header with suction hoseor pipe. For automatic operation, each mop shouldbe opened and closed by a float and float valve.



Fig. 7.51 Deep-well installation used at Smithsonian Institution, Washington, D.C. (Spencer, White &Prentis, Inc.)

When structures on silt or soft material arelocated near an excavation to be dewatered, careshould be taken that lowering of the water tabledoes not cause them to settle. It may be necessaryto underpin the structures or to pump dischargewater into recharge wells near the structures tomaintain the water table around them.

(L. Zeevaert, “Foundation Engineering for Dif-ficult Subsoil Conditions,” H. Y. Fang, “FoundationEngineering Handbook,” 2nd ed., Van NostrandReinhold Company, New York.)

UnderpinningThe general methods and main materials used togive additional support at or below grade to struc-tures are called underpinning. Usually, the addedsupport is applied at or near the footings.

7.30 Underpinning ProceduresUnderpinning may be remedial or precautionary.Remedial underpinning adds foundation capacity to


an inadequately supported structure. Precautionaryunderpinning is provided to obtain adequate foun-dation capacity to sustain higher loads, as a safe-guard against possible settlement during adjacentexcavating, or to accommodate changes in groundconditions. Usually, this type of underpinning isrequired for the foundations of a structure whendeeper foundations are to be constructed nearby foran addition or another structure. Loss of ground,even though small, into an adjoining excavation maycause excessive settlement of existing foundations.

Presumably, an excavation influences an exist-ing substructure when a plane through the outer-most foundations, on a 1-on-1 slope for sand or a1-on-2 slope for unconsolidated silt or soft clay,penetrates the excavation. For a cohesionless soil,underpinning exterior walls within a 1-on-1 slopeusually suffices; interior columns are not likely tobe affected if farther from the edge of the excava-tion than half the depth of the cut.

The commonly accepted procedures of struc-tural and foundation design should be used forunderpinning. Data for computing dead loads



may be obtained from plans of the structure or afield survey. Since underpinning is applied toexisting structures, some of which may be old,engineers in charge of underpinning design andconstruction should be familiar with older types ofconstruction as well as the most modern.

Before underpinning starts, the engineersshould investigate and record existing defects inthe structure. Preferably, the engineers should beaccompanied in this investigation by a representa-tive of the owner. The structure should be thor-oughly inspected, from top to bottom, inside (ifpossible) and out. The report should includenames of inspectors, dates of inspection, anddescription and location of defects. Photographsare useful in verifying written descriptions of dam-aged areas. The engineers should mark existingcracks in such a way that future observationswould indicate whether they are continuing toopen or spread.

Underpinning generally is accompanied bysome settlement. If design and field work aregood, the settlement may be limited to about ¼ to3/8 in. But as long as settlement is uniform in astructure, damage is unlikely. Differential settle-ment should be avoided. To check on settlement,elevations of critical points, especially columns andwalls, should be measured frequently duringunderpinning. Since movement may also occur lat-erally, the plumbness of walls and columns alsoshould be checked.

One of the first steps in underpinning usually isdigging under a foundation, which decreases itsload-carrying capacity temporarily. Hence, prelimi-nary support may be necessary until underpinningis installed. This support may be provided by shores,needles, grillages, and piles. Sometimes, it is desir-able to leave them in place as permanent supports.

Generally, it is advisable to keep preliminary sup-ports at a minimum, for economy and to avoid inter-ference with other operations. For the purpose,advantage may be taken of arching action and of theability of a structure to withstand moderate over-loads. Also, columns centrally supported on largespread footings need not be shored when digging isalong an edge and involves only a small percent ofthe total footing area. A large part of the column loadis supported by the soil directly under the column.

When necessary, weak portions of a structure,especially masonry, should be repaired orstrengthened before underpinning starts.


7.31 ShoresInstalled vertically or on a slight incline, shores areused to support walls or piers while underpinningpits are dug (Fig. 7.52a). Good bearing should beprovided at top and bottom of the shores. Oneway of providing bearing at the top is to cut aniche and mortar a steel bearing plate against theupper face. An alternate to the plate is a Z shape,made by removing diagonally opposite halfflanges from an H beam. When the top of the shoreis cut to fit between flange and web of the Z, move-ment of the shore is restrained. For a weak mason-ry wall, the load may have to be distributed over alarger area. One way of doing this is to insert a fewlintel angles about 12 in apart vertically and boltthem to a vertical, heavy timber or steel distribut-ing beam. The horizontal leg of an angle on thebeam then transmits the load to a shore.

Inclined shores on only one side of a wallrequire support at the base for horizontal as well asvertical forces. One way is to brace the shoresagainst an opposite wall at the floor. Preferably, thebase of each shore should sit on a footing perpen-dicular to the axis of the shore. Sized to providesufficient bearing on the soil, the footing may bemade of heavy timbers, steel beams, or reinforcedconcrete, depending on the load on the shore.

Loads may be transferred to a shore by wedgesor jacks. Oak wedges are suitable for light loads;forged steel wedges and bearing plates are desir-able for heavy loads. Jacks, however, offer greaterflexibility in length adjustments and allow correc-tions during underpinning for settlement of shorefootings.

(H. A. Prentis and L. White, “Underpinning,”Columbia University Press, New York; M. J. Tomlin-son, “Foundation Design and Construction,” Halst-ed Press, New York.)

7.32 Needles and GrillagesNeedles are beams installed horizontally to trans-fer the load of a wall or column to either or bothsides of its foundation, to permit digging of under-pinning pits (Fig. 7.52b). These beams are moreexpensive than shores, which transmit the loaddirectly into the ground. Needles usually are steelwide-flange beams, sometimes plate girders, usedin pairs, with bolts and pipe spreaders between thebeams. This arrangement provides resistance to



Fig. 7.52 Temporary supports used in underpinning: (a) shores; (b) needle beams; (c) grillage.

lateral buckling and torsion. The needles may beprestressed with jacks to eliminate settlementwhen the load is applied.

The load from steel columns may be transmit-ted through brackets to the needles. For masonrywalls, the needles may be inserted through niches.The load should be transferred from the masonryto the needles through thin wood fillers that crushwhen the needles deflect and maintain nearly uni-form bearing.

Wedges may be placed under the ends of theneedles to shift the load from the member to besupported to those beams. The beam ends may becarried on timber pads, which distribute the loadover the soil.

Grillages, which have considerably more bear-ing on the ground than needles, often are used asan alternative to needles and shores for closelyspaced columns. A grillage may be installed hori-zontally on soil at foundation level to support and


tie together two or more column footings, or it mayrest on a cellar floor (Fig. 7.52c). These preliminarysupports may consist of two or more steel beams,tied together with bolts and pipe spreaders, or of asteel-concrete composite. Also, grillages sometimesare used to strengthen or repair existing footings byreinforcing them and increasing their bearing area.The grillages may take the form of dowels or ofencircling concrete or steel-concrete beams. Theyshould be adequately cross-braced against bucklingand torsion. Holes should be made in steel beamsto be embedded in concrete, to improve bond.

7.33 Pit UnderpinningAfter preliminary supports have been installedand weak construction strengthened or repaired,underpinning may start. The most commonmethod of underpinning a foundation is to con-struct concrete piers down to deeper levels with



adequate bearing capacity and to transfer the loadto the piers by wedging up with dry packing. Tobuild the piers, pits must be dug under the foun-dation. Because of the danger of loss of groundand consequent settlement where soils are saturat-ed, the method usually is restricted to dry subsoil.

When piers have to be placed close together, acontinuous wall may be constructed instead. Butthe underpinning wall should be built in short sec-tions, usually about 5 ft long, to avoid undermin-ing the existing foundation. Alternate sections arebuilt first, and then the gaps are filled in.

Fig. 7.53 Vertical section through White House, Wning was used for the walls. (Spencer, White & Prentis,


Underpinning pits rarely are larger than about5 ft square in cross section. Minimum size for ade-quate working room is 3 × 4 ft. Access to the pit isprovided by an approach pit started alongside thefoundation and extending down about 6 ft. Thepits must be carefully sheeted and adequatelybraced to prevent loss of ground, which can causesettlement of the structure.

In soils other than soft clay, 2-in-thick woodplanks installed horizontally may be used to sheetpits up to 5 ft square, regardless of depth. Sides ofthe pits should be trimmed back no more than

ashington, D.C., during restoration. Pit underpin- Inc.)



absolutely necessary. The boards, usually 2 × 8’s,are installed one at a time with 2-in spaces betweenthem vertically. Soil is repacked through these lou-vers to fill the voids behind the boards. In runningsands, hay may be stuffed behind the boards toblock the flow. Corners of the sheeting often arenailed with vertical 2 × 4 in wood cleats.

In soft clay, sheeting must be tight and bracedagainst earth pressure. Chicago caissons, or a simi-lar type, may be used (Art. 7.22).

In water-bearing soil with a depth not exceed-ing about 5 ft, vertical sheeting can sometimes bedriven to cut off the water. For the purpose, lightsteel or tongue-and-groove wood sheeting may beused. The sheeting should be driven below thebottom of the pit a sufficient distance to preventboiling of the bottom due to hydrostatic pressure.With water cut off, the pit can be pumped dry andexcavation continued.

After a pit has been dug to the desired level, itis filled with concrete to within 3 in of the founda-tion to be supported. The gap is dry-packed, usu-ally by ramming stiff mortar in with a 2 × 4 pound-ed by an 8-lb hammer. The completed piers shouldbe laterally braced if soil is excavated on one side toa depth of more than about 6 ft. An example of pitunderpinning is the work done in the restorationof the White House, in which a cellar and subcellarwere created (Fig. 7.53).

7.34 Pile UnderpinningIf water-bearing soil more than about 5 ft deepunderlies a foundation, the structure may have tobe underpinned with piles. Driven piles generallyare preferred to jacked piles because of lower cost.The feasibility of driven piles, however, dependson availability of at least 12 ft of headroom andspace alongside the foundations. Thus, drivenpiles often can be used to underpin interior build-ing columns when headroom is available. But theyare hard to install for exterior walls unless there isample space alongside the walls. For very lightlyloaded structures, brackets may be attached tounderpinning piles to support the structure. Butsuch construction puts bending into the piles,reducing their load-carrying capacity.

Driven piles usually are 12- to 14-in diametersteel pipe, 3/8 in thick. They are driven open-ended,to reduce vibration, and in lengths determined byavailable headroom. Joints may be made with cast-


steel sleeves. After soil has been removed from thepipe interior, it is filled with concrete.

Jacked piles require less headroom and may beplaced under a footing. Also made of steel pipeinstalled open-ended, these piles are forced downby hydraulic jacks reacting against the footing. Theoperation requires an approach pit under the foot-ing to obtain about 6 ft of headroom.

Pretest piles, originally patented by Spencer,White & Prentis, New York City, are used to pre-vent the rebound of piles when jacking stops andsubsequent settlement when the load of the struc-ture is transferred to the piles. A pipe pile is jackeddown, in 4-ft lengths, to the desired depth. Thehydraulic jack reacts against a steel plate mortaredto the underside of the footing to be supported.After the pile has been driven to the required depthand cleaned out, it is filled with concrete andcapped with a steel bearing plate. Then twohydraulic jacks atop the pile overload it 50%. As theload is applied, a bulb of pressure builds up in thesoil at the pile bottom. This pressure stops down-ward movement of the pile. While the jacks main-tain the load, a short length of beam is wedgedbetween the pile top and the steel plate under thefooting. Then, the jacks are unloaded and removed.The load, thus, is transferred without further settle-ment. Later the space under the footing is concret-ed. Figure 7.54 shows how pretest piles were usedfor underpinning existing structures during con-struction of a subway in New York City.

7.35 MiscellaneousUnderpinning Methods

Spread footings may be pretested in much the sameway as piles. The weight of the structure is used tojack down the footings, which then are wedged inplace, and the gap is concreted. The method may beresorted to for unconsolidated soils where a highwater table makes digging under a footing unsafe orwhere a firm stratum is deep down.

A form of underpinning may be used for slabson ground. When concrete slabs settle, they maybe restored to the proper elevation by mud jack-ing. In this method, which will not prevent futuresettlement, a fluid grout is pumped under the slabthrough holes in it, raising it. Pressure is main-tained until the grout sets. The method also maybe used to fill voids under a slab.



Fig. 7.54 Pretest piles support existing buildings, old elevated railway, and a tunnel during construc-tion of a subway. (Spencer, White & Prentis, Inc.)

In loose sandy soils, Vibroflotation (Art. 7.36.5)may be used for underpinning. One difficulty withthis method is that the structure has to be shoredbefore underpinning starts, and the structure andshoring must be isolated from the vibrating equip-ment and soil being compacted.

Chemical or thermal stabilization (Arts. 7.37 and7.38) sometimes may be used to assist underpinning.

Soil ImprovementSoil for foundations can be altered to conform todesired characteristics. Whether this should bedone depends on the cost of alternatives.

Investigations of soil and groundwater conditionson a site should indicate whether soil improvement,or stabilization, is needed. Tests may be necessary todetermine which of several applicable techniques


may be feasible and economical. Table 7.19 lists someconditions for which soil improvement should beconsidered and the methods that may be used.

As indicated in the table, soil stabilization mayincrease strength, increase or decrease permeabili-ty, reduce compressibility, improve stability, ordecrease heave due to frost or swelling. The maintechniques used are constructed fills, replacementof unsuitable soils, surcharges, reinforcement,mechanical stabilization, thermal stabilization, andchemical stabilization.

7.36 Mechanical Stabilizationof Soils

This comprises a variety of techniques for rearrang-ing, adding, or removing soil particles. The objec-



tive usually is to increase soil density, decreasewater content, or improve gradation. Particles maybe rearranged by blending the layers of a stratifiedsoil, remolding an undisturbed soil, or densifying asoil. Sometimes, the desired improvement can beobtained by drainage alone. Often, however, com-pactive effort plus water control is needed.

Soil Probable Type ProbaDeficiency of Failure Cau

Slope instability Slides on slope Pore-wapress

Loose grsoil

Weak so

Mud flow Excessiv

Slides—movement Toe instaat toe

Low bearing Excessive settlement Saturatecapacity

Loose grsoil

Weak so

Heave Excessive rise Frost

Expansiclay

Excessive Seepage PerviousPermeability fissur

“Quick” bottom Loss of strength Flow un

Table 7.19 Where Soil Improvement May Be Econ


7.36.1 Embankments

Earth often has to be placed over the existingground surface to level or raise it. Such construct-ed fills may create undesirable conditions becauseof improper compaction, volume changes, andunexpected settlement under the weight of the fill.

ble Possiblese Remedies

ter Drain; flatten slope; freezeure

anular Compact

il Mix or replace with select material

e water Exclude water

bility Place toe fill, and drain

d clay Consolidate with surcharge, and drain

anular Compact; drain; increase footing depth;mix with chemicals

il Superimpose thick fill; mix or replacewith select material; inject or mixwith chemicals; freeze (if saturated);fuse with heat (if unsaturated)

For buildings: place foundations belowfrost line; insulate refrigeration-roomfloors; refrigerate to keep groundfrozen

For roads: Remove fines from gravel;replace with nonsusceptible soil

on of Exclude water; replace with granular soil

soil or Mix or replace soil with select material;ed rock inject or mix soil with chemicals; con-

struct cutoff wall with grout; enclosewith sheetpiles and drain

der Add berm against cofferdam inner face;increase width of cofferdam betweenlines of sheeting; drain with well-points outside the cofferdam

omical

ompanies, Inc. All rights reserved. Use of License Agreement. Click here to view.


To prevent such conditions, fill materials and theirgradation, placement, degree of compaction, andthickness should be suitable for properly support-ing the expected loads.

Fills may be either placed dry with convention-al earthmoving equipment and techniques or wetby hydraulic dredges. Wet fills are used mainly forfilling behind bulkheads or for large fills.

A variety of soils and grain sizes are suitable fortopping fills for most purposes. Inclusion of organ-ic matter or refuse should, however, be prohibited.Economics usually require that the source of fillmaterial be as close as possible to the site. For mostfills, soil particles in the 18 in below foundations,slabs, or the ground surface should not be largerthan 3 in in any dimension.

For determining the suitability of a soil as filland for providing a standard for compaction, themoisture-density relationship test, or Proctor test(ASTM D698 and D1557), is often used. Several ofthese laboratory tests should be performed on theborrowed material, to establish moisture-densitycurves. The peak of a curve indicates the maxi-mum density achievable in the laboratory by thetest method as well as the optimum moisture con-tent. ASTM D1557 should be used as the standardwhen high bearing capacity and low compressibil-ity are required; ASTM D698 should be used whenrequirements are lower, for example, for fills underparking lots.

The two ASTM tests represent different levelsof compactive effort. But much higher compactiveeffort may be employed in the field than that usedin the laboratory. Thus, a different moisture-densi-ty relationship may be produced on the site. Proc-tor test results, therefore, should not be consideredan inherent property of the soil. Nevertheless, thetest results indicate the proposed fill material’s sen-sitivity to moisture content and the degree of fieldcontrol that may be required to obtain the specifieddensity.

See also Art. 7.39.

7.36.2 Fill Compaction

The degree of compaction required for a fill is usu-ally specified as a minimum percentage of themaximum dry density obtained in the laboratorytests. This compaction is required to be accom-plished within a specific moisture range. Minimumdensities of 90 to 95% of the maximum density are


suitable for most fills. Under roadways, footings, orother highly loaded areas, however, 100% com-paction is often required. In addition, moisturecontent within 2 to 4% of the optimum moisturecontent usually is specified.

Field densities can be greater than 100% of themaximum density obtained in the laboratory test.Also, with greater compactive effort, such densitiescan be achieved with moisture contents that do notlie on the curves plotted from laboratory results.(Fine-grained soils should not be overcompactedon the dry side of optimum because when they getwet, they may swell and soften significantly.)

For most projects, lift thickness should berestricted to 8 to 12 in, each lift being compactedbefore the next lift is placed. On large projectswhere heavy compaction equipment is used, a liftthickness of 18 to 24 in is appropriate.

Compaction achieved in the field should bedetermined by performing field density tests oneach lift. For that purpose, wet density and mois-ture content should be measured and the dry den-sity computed. Field densities may be ascertainedby the sand-cone (ASTM D1556) or balloon vol-ume-meter (ASTM D2167) method, from an undis-turbed sample, or with a nuclear moisture-densitymeter. Generally, one field density test for each4000 to 10,000 ft2 of lift surface is adequate.

Hydraulically placed fills composed of dredgedsoils normally need not be compacted during place-ment. Although segregation of the silt and clay frac-tions of the soils may occur, it usually is not harmful.But accumulation of the fine-grained material inpockets at bulkheads or under structures should beprevented. For the purpose, internal dikes, weirs, ordecanting techniques may be used.

7.36.3 Soil Replacement or Blending

When materials at or near grade are unsuitable, itmay be economical to remove them and substitute afill of suitable soil, as described in Art. 7.36.1. Whenthis is not economical, consideration should be givento improving the soil by other methods, such as den-sification or addition or removal of soil particles.

Mixing an existing soil with select materials orremoving selected sizes of particles from an existingsoil can change its properties considerably. Addingclay to a cohesionless soil in a nonfrost region, forexample, may make the soil suitable as a base coursefor a road (if drainage is not too greatly impaired).



Adding clay to a pervious soil may reduce its per-meability sufficiently to permit its use as a reservoirbottom. Washing particles finer than 0.02 mm fromgravel makes the soil less susceptible to frost heave(desirable upper limit for this fraction is 3%).

7.36.4 Surcharges

Where good soils are underlain by soft, compress-ible clays that would permit unacceptable settle-ment, the site often can be made usable by sur-charging, or preloading, the surface. The objectiveis to use the weight of the surcharge to consolidatethe underlying clays. This offsets the settlement ofthe completed structure that would otherwiseoccur. A concurrent objective may be to increasethe strength of the underlying clays.

If the soft clay is overlain by soils with adequatebearing capacity, the area to be improved may beloaded with loose, dumped earth, until the weightof the surcharge is equivalent to the load that laterwill be imposed by the completed structure. (Ifhighly plastic clays or thick layers with little inter-nal drainage are present, it may be necessary toinsert sand drains to achieve consolidation withina reasonable time.) During and after placement ofthe surcharge, settlement of the original groundsurface and the clay layer should be closely moni-tored. The surcharge may be removed after little orno settlement is observed. If surcharging has beenproperly executed, the completed structure shouldexperience no further settlement due to primaryconsolidation. Potential settlement due to sec-ondary compression, however, should be evaluat-ed, especially if the soft soils are highly organic.

7.36.5 DensificationAny of a variety of techniques, most involving someform of vibration, may be used for soil densification.The density achieved with a specific technique,however, depends on the grain size of the soil. Con-sequently, grain-size distribution must be taken intoaccount when selecting a densification method.

Compaction of clean sands to depths of about 6ft usually can be achieved by rolling the surfacewith a heavy, vibratory, steel-drum roller. Althoughthe vibration frequency is to some extent adjustable,the frequencies most effective are in the range of 25to 30 Hz. Bear in mind, however, that little densifi-cation will be achieved below a depth of 6 ft, andthe soil within about 1 ft of the surface may actually


be loosened. Compactive effort in the field may bemeasured by the number of passes made with aspecific machine of given weight and at a givenspeed. For a given compactive effort, density varieswith moisture content. For a given moisture con-tent, increasing the compactive effort increases thesoil density and reduces the permeability.

Compaction piles also may be used to densifysands. For that purpose, the piles usually are madeof wood or are a sand replacement (sand pile). Tocreate a sand pile, a wood pile or a steel shell isdriven and the resulting hole is filled with sand.Densification of the surrounding soils results fromsoil displacement during driving of the pile orshell and from the vibration produced during piledriving. The foundations to be constructed neednot bear directly on the compaction piles but maybe seated anywhere on the densified mass.

Vibroflotation and Terra-Probe are alternativemethods that increase sand density by multipleinsertions of vibrating probes. These form cylindri-cal voids, which are then filled with offsite sand,stone, or blast-furnace slag. The probes usually areinserted in clusters, with typical spacing of about4½ ft, where footings will be placed. Relative den-sities of 85% or more can be achieved throughoutthe depth of insertion, which may exceed 40 ft. Useof vibrating probes may not be effective, however,if the fines content of the soil exceeds about 15% orif organic matter is present in colloidal form inquantities exceeding about 5% by weight.

Another technique for densification is dynamiccompaction, which in effect subjects the site tonumerous mini-earthquakes. In saturated soils,densification by this method also results from par-tial liquefaction, and the elevated pore pressuresproduced must be dissipated between each appli-cation of compactive energy if the following appli-cation is to be effective. As developed by Tech-niques Louis Menard, dynamic compaction isachieved by dropping weights ranging from 10 to40 tons from heights up to 100 ft onto the groundsurface. Impact spacings range up to 60 ft. Multipledrops are made at each location to be densified.This technique is applicable to densification of largeareas and a wide range of grain sizes and materials.

7.36.6 Drainage

This is effective in soil stabilization becausestrength of a soil generally decreases with an



increase in amount and pressure of pore water.Drainage may be accomplished by gravity, pump-ing, compression with an external load on the soil,electro-osmosis, heating, or freezing.

Pumping often is used for draining the bottomof excavations (Art. 7.29). For slopes, however,advantage must be taken of gravity flow to attainpermanent stabilization. Vertical wells may be usedto relieve artesian pressures. Usually, interceptingdrains, laid approximately along contours, suffice.

Where mud flows may occur, water must beexcluded from the area. Surface and subsurfaceflow must be intercepted and conducted away atthe top of the area. Also, cover, such as heavymulching and planting, should be placed over theentire surface to prevent water from percolatingdownward into the soil. See also Art. 7.39.

Electrical drainage adapts the principle thatwater flows to the cathode when a direct currentpasses through saturated soil. The water may bepumped out at the cathode. Electro-osmosis is rela-tively expensive and therefore usually is limited tospecial conditions, such as drainage of silts, whichordinarily are hard to drain by other methods.

Vertical sand drains, or piles, may be used tocompact loose, saturated cohesionless soils or toconsolidate saturated cohesive soils. They providean escape channel for water squeezed out of thesoil by an external load. A surcharge of perviousmaterial placed over the ground surface alsoserves as part of the drainage system as well as partof the fill, or external load. Usually, the surcharge isplaced before the sand piles are formed, to supportequipment, such as pile drivers, over the soft soil.Fill should be placed in thin layers to avoid forma-tion of mud flows, which might shear the sanddrains and cause mud waves. Analyses should bemade of embankment stability at various stages ofconstruction.

7.37 Thermal Stabilization ofSoils

Thermal stabilization generally is costly and isrestricted to conditions for which other methodsare not suitable. Heat may be used to strengthennonsaturated loess and to decrease the compress-ibility of cohesive soils. One technique is to burnliquid or gas fuel in a borehole. Another is to injectinto the soil through pipes spaced about 10 ft apart


a mixture of liquid fuel and air under pressure.Then, the mixture is burned for about 10 days, pro-ducing a solidified soil.

Freezing a wet soil converts it into a rigid mate-rial with considerable strength, but it must be keptfrozen. The method is excellent for a limited exca-vation area, for example, freezing the ground tosink a shaft. For the purpose, a network of pipes isplaced in the ground and a liquid, usually brine, atlow temperature is circulated through the pipes.Care must be taken that the freezing does notspread beyond the area to be stabilized and causeheaving damage.

7.38 Chemical Stabilization ofSoils

Utilizing, portland cement, bitumens, or othercementitous materials, chemical stabilizationmeets many needs. In surface treatments, it sup-plements mechanical stabilization to make theeffects more lasting. In subsurface treatments,chemicals may be used to improve bearing capaci-ty or decrease permeability.

Soil-cement, a mixture of portland cement andsoil, is suitable for subgrades, base courses, andpavements of roads not carrying heavy traffic(“Essentials of Soil-Cement Construction,” Port-land Cement Association). Bitumen-soil mixturesare extensively used in road and airfield construc-tion and sometimes as a seal for earth dikes(“Guide Specifications for Highway Construction,”American Association of State Highway and Trans-portation Officials, 444 North Capitol St., N.W.,Washington, DC 20001). Hydrated, or slaked, limemay be used alone as a soil stabilizer, or with flyash, portland cement, or bitumen (“Lime Stabiliza-tion of Roads,” National Lime Association, 925 15thSt., N.W., Washington D.C. 20006). Calcium or sodi-um chloride is used as a dust palliative and anadditive in construction of granular base andwearing courses for roads (“Calcium Chloride forStabilization of Bases and Wearing Courses,” Calci-um Chloride Institute, Ring Building, Washington,DC 20036).

Grouting, with portland cement or other chem-icals, often is used to fill rock fissures, decrease soilpermeability, form underground cutoff walls toeliminate seepage, and stabilize soils at consider-able depth. The chemicals may be used to fill the



voids in the soil, to cement the particles, or to forma rocklike material. The process, however, general-ly is suitable only in pervious soils. Also, rapid set-ting of chemicals may interfere with thoroughinjection of the underground region. Chemicalsused include sodium silicate and salts or acids;chrome-lignin; and low-viscosity organics.

(K. Terzaghi and R. B. Peck, “Soil Mechanics inEngineering Practice,” John Wiley & Sons, Inc.,New York; G. P. Tschebotarioff, “Soil Mechanics,Foundations, and Earth Structures,” McGraw-HillBook Company, New York; H. Y. Fang, “Founda-tion Engineering Handbook,” 2nd ed., Van Nos-trand Reinhold Company, New York.)

7.39 GeosyntheticsIn the past, many different materials have beenused for soil separation or reinforcement, includ-ing grasses, rushes, wood logs, wood boards, metalmats, cotton, and jute. Because they deterioratedin a relatively short time, required maintenancefrequently, or were costly, however, use of moreefficient, more permanent materials was desirable.

Application Fu

Subgrade stabilization Reinforcement,

Railway trackbed stabilization Drainage, separ

Sedimentation-control silt fence Sediment retent

Asphalt overlays Stress-relieving

Soil reinforcement:

Embankments Reinforcement

Steep slopes Reinforcement

Retaining walls Reinforcement

Erosion control:

Reinforcement Reinforcement,

Riprap Filtration and se

Mats Filtration and se

Subsurface drainage filter Filtration

Geomembrane protection Protection and c

Subsurface drainage Fluid transmissi

Table 7.20 Primary Function of Geosynthetics in G


Synthetic fabrics, grids, nets, and other structures.are now used as an alternative.

Types of geosynthetics, polymer compositionsgenerally used, and properties important for spec-ifying materials to achieve desired performanceare described in Art. 5.29. Principal applications ofgeosynthetics, functions of geosynthetics in thoseapplications, recommended structures for eachcase, and design methods are discussed in the fol-lowing. Table 7.20 summarizes the primary func-tions of geosynthetics in applications often usedand indicates the type of geosynthetic generallyrecommended by the manufacturers of thesematerials for the applications.

7.39.1 Design Methods forGeosynthetics

The most commonly used design methods forgeosynthetics in geotechnical applications are theempirical (design by experience), specification, andrational (design by function) methods.

The empirical design process employs a selectionprocess based on the experience of the project geo-

nction Geosynthetic

separation, filtration Geotextile or geogrid

ation, filtration Geotextile

ion, filtration, separation Geotextile

layer and waterproofing Geotextile

Geotextile or geogrid



separation Geocomposite

paration Geotextile

paration Geotextile

Geotextile

ushion Geotextile

on and filtration Prefabricateddrainage composite

eotechnical Applications



technical engineer or of others, such as designers ofprojects reported in engineering literature, manufac-turers of geosynthetics, and professional associations.

Design by specification is often used for routineapplications. Standard specifications for specificapplications may be available from geosyntheticsmanufacturers or developed by an engineeringorganization or a government agency for its ownuse or by an association or group of associations,such as the joint committee established by theAmerican Association of State Highway and Trans-portation Officials, Associated General Contrac-tors, and American Road and TransportationBuilders Association (Art. 5.29).

When using the rational design method,designers evaluate the performance, constructionmethods required, and durability under serviceconditions of geosynthetics that are suitable forthe planned application. This method can be usedfor all site conditions to augment the precedingmethods. It is necessary for applications not cov-ered by standard specifications. It also is requiredfor projects of such nature that large propertydamage or personal injury would result if a failureshould occur. The method requires the following:

A decision as to the primary function of a geosyn-thetic in the application under consideration

Estimates or calculations to establish the requiredproperties (design values) of the material for theprimary function

Determination of the allowable properties, such asminimum tensile or tear strength or permittivity, ofthe material by tests or other reliable means

Calculation of the safety factor as the ratio of allow-able to design values

Assessment of this result to ascertain that it is suf-ficiently high for site conditions

(“A Design Primer: Geotextiles and RelatedMaterials,” Industrial Fabric Association Interna-tional,” 345 Cedar St., Suite 800, St. Paul, MN 55101;B. R. Christopher and R. D. Holtz, “GeotextileEngineering Manual,” HI-89-050, Federal HighwayAdministration, Washington, D. C., J. E. Fluet,“Geotextile Testing and the Design Engineer,” STP952, ASTM; R. M. Koerner, “Designing withGeosynthetics,” 2nd ed., Prentice-Hall, EnglewoodCliffs, N. J.)


7.39.2 Geosynthetics Nomenclature

Following are some of the terms generally used indesign and construction with geosynthetics:

Apparent Opening Size (AOS). A property desig-nated O95 applicable to a specific geotextile thatindicates the approximate diameter of the largestparticle that would pass through the geotextile. Aminimum of 95% of the openings are the same sizeor smaller than that particle, as measured by thedry sieve test specified in ASTM D4751.

Blinding. Blocking by soil particles of openingsin a geotextile, as a result of which its hydraulicconductivity is reduced.

Chemical Stability. Resistance of a geosyntheticto degradation from chemicals and chemical reac-tions, including those catalyzed by light.

Clogging. Retention of soil particles in the voidsof a geotextile, with consequent reduction in thehydraulic conductivity of the fabric.

Cross-Machine Direction. The direction withinthe plane of a fabric perpendicular to the directionof manufacture. Generally, tensile strength of thefabric is lower in this direction than in the machinedirection.

Denier. Mass, g, of a 9000-m length of yarn.

Fabric. Polymer fibers or yarn formed into asheet with thickness so small relative to dimen-sions in the plane of the sheet that it cannot resistcompressive forces acting in the plane. A needle-punched fabric has staple fibers or filamentsmechanically bonded with the use of barbed nee-dles to form a compact structure. A spun-bondedfabric is formed with continuous filaments thathave been spun (extruded), drawn, laid into a web,and bonded together in a continuous process,chemically, mechanically, or thermally. A wovenfabric is produced by interlacing orthogonally twoor more sets of elements, such as yarns, fibers, rov-ings, or filaments, with one set of elements in themachine direction. A monofilament woven fabricis made with single continuous filaments, whereasa multifilament woven fabric is composed of bun-dles of continuous filaments. A split-film wovenfabric is constructed of yarns formed by splittinglongitudinally a polymeric film to form a slit-tapeyarn. A nonwoven fabric is produced by bondingor interlocking of fibers, or both.



Fiber. Basic element of a woven or knitted fabricwith a length-diameter or length-width ratio of atleast 100 and that can be spun into yarn or other-wise made into a fabric.

Filament. Variety of fiber of extreme length, notreadily measured.

Filtration. Removal of particles from a fluid orretention of soil particles in place by a geosynthetic,which allows water or other fluids to pass through.

Geocomposite. Manufactured laminated or com-posite material composed of geotextiles, geomem-branes, or geogrids, and sometimes also naturalmaterials, or a combination.

Geogrid. Orthogonally arranged fibers, strands,or rods connected at intersections, intended for useprimarily as tensile reinforcement of soil or rock.

Geomembrane. Geosynthetic, impermeable ornearly so, intended for use in geotechnical applica-tions.

Geosynthetics. Materials composed of polymersused in geotechnical applications.

Geotextile. Fabric composed of a polymer andused in geotechnical applications.

Grab Tensile Strength. Tensile strength deter-mined in accordance with ASTM D4632 and typi-cally found from a test on a 4-in-wide strip of fab-ric, with the tensile load applied at the midpoint ofthe fabric width through 1-in-wide jaw faces.

Gradient Ratio. As measured in a constant-headpermittivity test on a geotextile, the ratio of theaverage hydraulic gradient across the fabric plus 1in of soil adjoining the fabric to the averagehydraulic gradient of the 2 in of soil between 1 and3 in above the fabric.

Machine Direction. The direction in the plane ofthe fabric parallel to the direction of manufacture.Generally, the tensile strength of the fabric islargest in this direction.

Monofilament. Single filament, usually of adenier higher than 15.

Mullen Burst Strength. Hydraulic burstingstrength of a geotextile as determined in accor-dance with ASTM D3786.

Permeability (Hydraulic Conductivity). A mea-sure of the capacity of a geosynthetic to allow afluid to move through its voids or interstices, as


represented by the amount of fluid that passesthrough the material in a unit time per unit sur-face area under a unit pressure gradient. Accord-ingly, permeability is directly proportional tothickness of the geosynthetic.

Permittivity. Like permeability, a measure of thecapacity of a geosynthetic to allow a fluid to movethrough its voids or interstices, as represented bythe amount of fluid that passes through a unit sur-face area of the material in a unit time per unitthickness under a unit pressure gradient, withlaminar flow in the direction of the thickness of thematerial For evaluation of geotextiles, use of per-mittivity, being independent of thickness, is pre-ferred to permeability.

Puncture Strength. Ability of a geotextile to resistpuncture as measured in accordance with ASTMD3787.

Separation. Function of a geosynthetic to pre-vent mixing of two adjoining materials.

Soil-Fabric Friction. Resistance of soil by frictionto sliding of a fabric embedded in it, exclusive ofresistance from cohesion. It is usually expressed asa friction angle.

Staple Fibers. As usually used in geotextiles,very short fibers, typically 1 to 3 in long.

Survivability. Ability of geosynthetics to performintended functions without impairment.

Tearing Strength. Force required either to startor continue propagation of a tear in a fabric asdetermined in accordance with ASTM D4533.

Tenacity. Fiber strength, grams per denier.

Tex. Denier divided by 9.

Transmissivity. Amount of fluid that passes inunit time under unit pressure gradient with lami-nar flow per unit thickness through a geosyntheticin the in-plane direction.

Yarn. Continuous strand composed of textilefibers, filaments, or material in a form suitable forknitting, weaving, or otherwise intertwining toform a geotextile.

7.39.3 Geosynthetic Reinforcementof Steep Slopes

Geotextiles and geogrids are used to reinforce soilsto permit slopes much steeper than the shearing



resistance of the soils will permit. (Angle of repose,the angle between the horizontal and the maximumslope that a soil assumes through natural processes,is sometimes used as a measure of the limiting slopesfor unconfined or unreinforced cuts and fills, but it isnot always relevant. For dry, cohesionless soils, theeffect of height of slope on this angle is negligible.For cohesive soils, in contrast, the height effect is solarge that angle of repose is meaningless.) Whengeosynthetic reinforcement is used, it is placed in thefill in horizontal layers. Vertical spacing, embedmentlength, and tensile strength of the geosynthetic arecritical in establishing a stable soil mass.

Fig. 7.55 Stabilization of a steep slope with horizmary reinforcement for a circular failure surface. (b) critical failure surfaces into the backfill. (c) Intermedia


For evaluation of slope stability, potential failuresurfaces are assumed, usually circular or wedge-shaped but other shapes also are possible. Figure7.55a shows a slope for which a circular failure sur-face starting at the bottom of the slope and extend-ing to the ground surface at the top is assumed. Anadditional circular failure surface is indicated in Fig.7.55b. Fig. 7.55c shows a wedge-shaped failure sur-face. An infinite number of such failure surfaces arepossible. For design of the reinforcement, the sur-faces are assumed to pass through a layer of rein-forcement at various levels and apply tensile forcesto the reinforcement, which must have sufficient

ontal layers of geosynthetic reinforcement. (a) Pri-Embedment lengths of reinforcement extend fromte reinforcement for shallow failure surfaces.



tensile strength to resist them. Sufficient reinforce-ment embedment lengths extending into stable soilbehind the surfaces must be provided to ensure thatthe geosynthetic will not pull out at design loads.

Pull-out is resisted by geotextiles mainly by fric-tion or adhesion—and by geogrids, which have sig-nificant open areas, also by soil-particle strike-through. The soil-fabric interaction is determined inlaboratory pull-out tests on site-specific soils andthe geosynthetic to be used, but long-term load-transfer effects may have to be estimated. Design ofthe reinforcement requires calculation of theembedment required to develop the reinforcementfully and of the total resisting force (number of lay-ers and design strength) to be provided by the rein-forcement. The design should be based on safetyfactors equal to or greater than those required bylocal design codes. In the absence of local coderequirements, the values given in Table 7.21 may beused. A stability analysis should be performed toinvestigate, at a minimum, circular and wedge-shaped failure surfaces through the toe (Fig. 7.55a),face (Fig. 7.55c), and deep seated below the toe (Fig.7.55b). The total resisting moment for a circular slipsurface can be determined from Fig. 7.55b as

(7.92)

External Stability

Condition K

Sliding 1.5

Deep seated (overall stability) 1.3

Dynamic loading 1.1

*Td at 5% strain should be less than Ta. † Ta = TL /KcKd , where TL is the creep limit strength, Kc is the safetythe absence of creep tests or other pertinent data, the following maysile strength of the geosynthetic..

‡ For a 3-ft minimum embedment.

Table 7.21 Minimum Safety Factors K for Slope Re


where R = radius of failure circle

FR = soil shearing resistance along theslip surface = τf Lsp

τf = soil shear strength

Lsp = length of slip surface

Ri = radius of slip surface at layer i

Ti = strength of the reinforcing requiredfor layer i

The driving moment, or moment of the forcescausing slip, is

(7.93)

where W = weight of soil included in assumedfailure zone (Fig. 7.55a)

r = moment arm of W with respect tothe center of rotation (Fig. 7.55a)

S = surcharge

d = moment arm of S with respect to thecenter of rotation (Fig. 7.55a)

The safety factor for the assumed circular failuresurface then is

(7.94)

Internal Stability

Condition K

Slope stability 1.3

Design tensile strength Td *

Allowable geosynthetic strength Ta†

Creep 4Construction 1.1 to 1.3Durability 1.1 to 1.2Pull-out resistance

Cohesionless soils 1.5‡

Cohesive soils 2

factor for construction, and Kd is the safety factor for durability. In be used: Ta = Tu /10.4 or Tu > 10.4Td , where Tu is the ultimate ten-

inforcement



A safety factor should be computed for each poten-tial failure surface. If a safety factor is less than therequired minimum safety factor for prevention offailure of the soil unreinforced, either a strongerreinforcement is required or the number of layersof reinforcement should be increased. This proce-dure may also be used to determine the reinforce-ment needed at any level to prevent failure abovethat layer.

The next step is calculation of length Le of rein-forcement required for anchorage to prevent pullout.

(7.95)

where FD = required pull-out strength

K = minimum safety factor: 1.5 for cohe-sionless soils; 2 for cohesive soils

σo = overburden pressure above the rein-forcing level = wh

w = density of the soil

h = depth of overburden

φsr = soil-reinforcement interaction angle,determined from pull-out tests

Embedment length Le should be at least 3 ft. Thetotal length of a reinforcement layer then is Le plusthe distance from the face of the slope to the failurecircle (Fig. 7.55b). The total length of the reinforce-

Fig. 7.56 Geosynthetic applications with retaining(b) Retaining wall anchored into backfill.


ment at the toe should be checked to ascertain thatit is sufficient to resist sliding of the soil mass abovethe base of the slope.

Among the family of potential failure surfacesthat should be investigated is the wedge shape,such as the one shown in Fig. 7.55c. To reinforcethe failure zones close to the face of the slope, lay-ers of reinforcement are required in addition tothose provided for the deep failure zones, as indi-cated in Fig. 7.55c. Such face reinforcement shouldhave a maximum vertical spacing of 18 in and aminimum length of 4 ft. Inasmuch as the tension inthis reinforcement is limited by the short embed-ment, a geosynthetic with a smaller design allow-able tension than that required for deep-failurereinforcement may be used. In construction of thereinforced slope, fill materials should be placed sothat at least 4 in of cover will be between thegeosynthetic reinforcement and vehicles or equip-ment operating on a lift. Backfill should not incor-porate particles larger than 3 in. Turning of vehi-cles on the first lift above the geosynthetic shouldnot be permitted. Also, end dumping of fill direct-ly on the geosynthetic should not be allowed.

7.39.4 Geosynthetics in Retaining-Wall Construction

Geotextiles and geogrids are used to form retainingwalls (Fig. 7.56a) or to reinforce the backfill of aretaining wall to create a stable soil mass (Fig.

walls: (a) Reinforced earth forms a retaining wall.



7.56b). In the latter application, the reinforcementreduces the potential for lateral displacement of thewall under the horizontal pressure of the backfill.

As in the reinforcement of steep slopes dis-cussed in Art. 7.39.3, the reinforcement layers mustintersect all critical failure surfaces. For cohesion-less backfills, the failure surface may be assumed tobe wedge shaped, as indicated in Fig. 7.55c, withthe sloping plane of the wedge at an angle of 45° +φ/2 with the horizontal. If the backfill is not homo-geneous, a general stability analysis should be car-ried out as described in Art. 7.39.3.

The design process for cohesionless soils can besimplified by use of a constant vertical spacing svfor the reinforcement layers. This spacing wouldbe approximately

(7.96)

where Ta = allowable tension in the reinforcement

K = safety factor as specified in a localcode or as given in Table 7.21

Ka = coefficient of active earth pressure(Art. 7.26)

w = density of backfill

H = average height of embankment

Fig. 7.57 Retaining wall anchored with geosynthedom-soil backfill, sand backfill, surcharge, and live loatangular and triangular.


If Eq. (7.96) yields a value for sv less than the mini-mum thickness of a lift in placement of the backfill,a stronger geosynthetic should be chosen. Theminimum embedment length Le may be computedfrom Eq. (7.95). Although the total reinforcementlength thus computed may vary from layer tolayer, a constant reinforcement length would beconvenient in construction.

When the earth adjoining the backfill is a ran-dom soil with lower strength than that of the back-fill, the random soil exerts a horizontal pressure onthe backfill that is transmitted to the wall (Fig. 7.57)This may lead to a sliding failure of the reinforcedzone. The reinforcement at the base should be suffi-ciently long to prevent this type of failure. The totalhorizontal sliding force on the base is, from Fig. 7.57,

(7.97)

where Pb =KawbH2/2

wb = density of soil adjoining the rein-forcement zone

Ps = KawshH

wsh = weight of uniform surcharge

Pν = force due to live load V as determinedby the Boussinesq method (Art. 7.11)

tic reinforcement is subjected to pressure from ran-d. Assumed pressure distribution diagrams are rec-



The horizontal resisting force is

(7.98)

where wrH = weight of soil in the reinforcementzone

φsr = soil-reinforcement interaction angle

c = undrained shear strength of thebackfill

L = length of the reinforcement zonebase

The safety factor for sliding resistance then is

(7.99)

and should be 1.5 or larger. A reinforcement lengthabout 0.8H generally will provide base resistancesufficient to provide a safety factor of about 1.5.

The most economical retaining wall is one inwhich the reinforcement is turned upward andbackward at the face of the wall and also serves asthe face (Fig. 7.57a) The backward embedmentshould be at least 4 ft. If desired for esthetic reasonsor to protect the geosynthetic from damage ordeterioration from exposure to ultraviolet light,sprayed concrete may be applied to the wall face.

As an alternative, the wall may be composed ofconcrete block or precast concrete panels that areanchored to soil reinforcement. The reinforcementshould be installed taut to limit lateral movementof the wall during construction.

See Art. 7.39.3 for other precautions to be takenduring construction.

7.39.5 Geosynthetic Reinforcementfor Embankments

Geosynthetics placed in horizontal layers may beused to reinforce embankments in a manner simi-lar to that used to reinforce steep slopes (Art.7.39.3). The reinforcement may permit greaterembankment height and a larger safety factor inembankment design than would an unreinforcedembankment. Also, displacements during con-struction may be smaller, thus reducing fill require-ments. Furthermore, reinforcement properlydesigned and installed can prevent excessive hori-zontal displacements along the base that can causean embankment failure when the underlying soilis weak. Moreover, reinforcement may decrease


horizontal and vertical displacements of theunderlying soil and thus limit differential settle-ment. Reinforcement, however, will reduce nei-ther long-term consolidation of the underlyingweak soil nor secondary settlement.

Either geotextiles or geogrids may be used asreinforcement. If the soils have very low bearingcapacity, it may be necessary to use a geotextileseparator with geogrids for filtration purposes andto prevent the movement of the underlying soilinto the embankment fill.

Figure 7.58 illustrates reinforcement of anembankment completely underlain by a weak soil.Without reinforcement, horizontal earth pressurewithin the embankment would cause it to spreadlaterally and lead to embankment failure, in theabsence of sufficient resistance from the soil. Rein-forcement is usually laid horizontally in the direc-tion of major stress; that is, with strong axis normalto the longitudinal axis of the embankment. Rein-forcement with strong axis placed parallel to thelongitudinal axis of the embankment may also berequired at the ends of the embankment. Seamsshould be avoided in the high-stress direction.Design of the reinforcement is similar to thatrequired for steep slopes (Art. 7.39.3).

For an embankment underlain by locally weakareas of soil or voids, reinforcement may be incorpo-rated at the base of the embankment to bridge them.

7.39.6 Soil Stabilization withGeosynthetics

Woven or nonwoven geotextiles are used toimprove the load-carrying capacity of roads overweak soils and to reduce rutting. Acting primarily

Fig. 7.58 Geosynthetic reinforcement for anembankment on weak soil is placed directly on thesubgrade.



as a separation barrier, the geosynthetic preventsthe subgrade and aggregate base from mixing. Thegeosynthetic may also serve secondary functions.Acting as a filter, it prevents fines from migratinginto the aggregate due to high water pressure.Also, the geotextile may facilitate drainage byallowing pore water to pass through and dissipateinto the underlying soil. In addition, acting as rein-forcement, the geotextile can serve as a membranesupport of wheel loads and provide lateralrestraint of the base and subgrade through frictionbetween the fabric, aggregate, and soil.

Installation techniques to be used depend on theapplication. Usually, geosynthetics are laid directlyon the subgrade (Fig. 7.59a). Aggregate then isplaced on top to desired depth and compacted.

Design of permanent roads and highways con-sists of the following steps: If the CBR < 3, need fora geotextile is indicated. The pavement is designedby usual methods with no allowance for structuralsupport from the geotextile. If a thicker subbase thanthat required for structural support would have tobe specified because of the susceptibility of theunderlying soil to pumping and subbase intrusion,the subbase may be reduced 50% and a geotextileselected for installation at the subbase-subgradeinterface. For stabilization of the subgrade duringconstruction, an additional determination of thick-ness of subbase assisted by a geotextile is made by

Fig. 7.59 Geosynthetic is used (a) to reinfo


conventional methods (bearing capacity Nc about 3.0without geotextiles and about 5.5 with) to limit rut-ting to a maximum of 3 in under construction vehi-cle loads. The thicker subbase thus computed isselected. Then, the geotextile strength requirementsfor survivability and filtration characteristics arechecked. (Details for this are given in B. R. Christo-pher and R. D. Holtz, “Geotextile Design and Con-struction Guidelines,” FHWA DTFH61-86-C-00102,National Highway Institute, Federal HighwayAdministration, Washington, DC 20590.)

Geosynthetics are also used under railwaytracks for separation of subgrade and subballast orsubballast and ballast (Fig. 7.59b). They also areused for roadbed filtration, lateral permeability,and strength and modulus improvement.

7.39.7 Geosynthetics in ErosionControl

For purposes of erosion control, geosynthetics areused as turf reinforcement, as separators and filtersunder riprap, or armor stone, and as an alternativeto riprap. Different types of geosynthetics are usedfor each of these applications.

Turf Control n To establish a reinforced turf inditches and water channels and on slopes, three-dimensional erosion-control mats often are used.

rce a road, (b) to reinforce a railway roadbed.

ompanies, Inc. All rights reserved. Use of License Agreement. Click here to view.


Entangling with the root and stem network of veg-etation, they greatly increase resistance to flow ofwater down slopes and thus retard erosion.

Mats used for turf reinforcement should have astrong, stable structure. They should be capable ofretaining the underlying soil but have sufficientporosity to allow roots and stem to grow throughthem. Installation generally requires pinning themat to the ground and burying mat edges and ends.Topsoil cover may be used to reduce erosion evenmore and promote rapid growth of vegetation.

When a geosynthetic is placed on a slope, itshould be rolled in the direction of the slope. Hor-izontal joints should not be permitted. Verticaljoints should be shingled downstream. Ditch andchannel bottoms should be lined by rolling thegeosynthetic longitudinally. Joints transverse towater flow should have a 3-ft overlap and be shin-gled downstream. Roll edges should be over-lapped 2 to 4 in. They should be staked at intervalsnot exceeding 5 ft to prevent relative movement.Where highly erodible soils are encountered, ageotextile filter should be installed under the turfreinforcement and staked or otherwise bonded tothe mats. For stability and seeding purposes, woodchips may be used to infill the turf reinforcement.

Use of Geosynthetics with Riprap n

Large armor stones are often used to protect soilagainst erosion and wave attack. Graded-aggre-gate filter generally is placed between the soil andthe riprap to prevent erosion of the soil throughthe armoring layer. As a more economical alterna-tive, geotextiles may be used instead of aggregate.They also offer greater control during construc-tion, especially in underwater applications.Geosynthetics typically used are nonwoven fab-rics, monofilament nonwoven geotextiles, andmultifilament or fibrillated woven fabrics.

The geosynthetics should have sufficient per-meability to permit passage of water to relievehydrostatic pressure behind the riprap. Also, thegeosynthetic should be capable of retaining theunderlying soil. Conventional filter criteria can beused for design of.the geosynthetic, although somemodifications may be required to compensate forproperties of the geosynthetic.

Installation precautions that should beobserved include the following: Riprap should beinstalled with care to avoid tearing the geosynthet-ic inasmuch as holes would decrease its strength.


Stone placement, including drop heights, shouldbe tested in field trials to develop techniques thatwill not damage the geosynthetic. As a generalguide, for material protected by a sand cushionand material with properties exceeding thatrequired for unprotected applications, drop heightfor stones weighing less than 250 lb should notexceed 3 ft; without a cushion, 1 ft. Stone weighingmore than 250 lb should be placed without freefall, unless field tests determine a safe drop height.Stone weighing more than 100 lb should not bepermitted to roll along the geosynthetic. Installa-tion of the armor layer should begin at the base ofslopes and at the center of the zone covered by thegeosynthetic. After the stones have been placed,they should not be graded.

Special construction procedures are requiredfor slopes greater than 2.5:1. These includeincrease in overlap, slope benching, elimination ofpins at overlaps, toe berms for reaction againstslippage, and laying of the geosynthetic sufficient-ly loose to allow for downstream movement, butfolds and wrinkles should not be permitted.

The geosynthetic should be rolled out with itsstrong direction (machine direction for geotextiles)up and down the slope. Adjoining rolls should beseamed or shingle overlapped in the downslope ordownstream direction. Joints should be stapled orpinned to the ground. Recommended pin spacingis 2 ft for slopes up to 3:1, 3 ft for slopes between 3:1and 4:1, 5 ft for 4:1 slopes, and 6 ft for slopes steep-er than 4:1. For streambanks and slopes exposed towave action, the geosynthetic should be anchoredat the base of the slope by burial around theperimeter of a stone-filled key trench. It shouldalso be keyed at the top of the slope if the armor-geosynthetic system does not extend several feetabove high water.

Riprap Replacement n Instead of the riprapgenerally used for erosion control, concrete matsmay be used. For this purpose, the concrete con-ventionally has been cast in wood or steel forms.Use of expandable fabric forms, however, may bemore economical. Such forms are made by joiningtwo fabric sheets at discrete points. After the sheetsare placed over the area to be protected, grout ispumped into the space between the sheets to forma mattress that initially will conform to the shapeof the ground and later harden.. Thickness of themattress is controlled by internal spacer threads.



Filter points and bands are formed in the mattressto dissipate pore water from the subsoil. The fabricforms may be grouted underwater, even in flowingwater, and in hazardous-liquid conditions. Thefabric usually used is a woven geotextile.

7.39.8 Uses of Geosynthetics inSubsurface Drainage

Subsurface drainage is required for many con-struction projects and geotextiles find many usesin such applications. Their primary function is toserve, with graded granular filter media, as a per-meable separator to exclude soil from the drainagemedia but to permit water to pass freely. Nonwo-ven geotextiles are usually used for this purposebecause of their high flow capacity and small poresize. Generally, fabric strength is not a primaryconsideration for subsurface drainage applica-tions, except during installation.

Following are brief descriptions of typical appli-cations of geotextiles in subsurface drainage:

Permeable separators wrapped around trench oredge drains

Drains for retaining walls and bridge abutmentswith the geotextile enclosing the backfill

Encirclement of slotted or jointed drains and wallpipes to prevent filter particles from entering thedrains while permitting passage of water

Wraps for interceptor, toe, and surface drains onslopes to assist stabilization by dissipating excesspore-water pressures and retarding erosion

Seepage control with chimney and toe drains forearth dams and levees with the geotextile laidalong the upstream face and anchored by a berm

7.39.9 Geosynthetics as Pond Liners

Geomembranes, being impermeable, appear to bean ideal material for lining the bottom of a pond toretain water or other fluid. Used alone, however,they have some disadvantages. In particular, theyare susceptible to damage from many sources andrequire a protective soil cover of at least 12 in.Also, for several reasons, it is advisable to lay ageotextile under the geomembrane. The geotextileprovides a clean working area for making seams.It adds puncture resistance to the liner. It increas-es friction resistance at the interface with the soil,


thus permitting steeper side slopes. And it permitslateral and upward escape of gases emitted fromthe soil. For this purpose, needle-punched non-woven geotextiles, geonets, or drainage compos-ites with adequate transmissivity for passing thegases are needed. In addition, it is advantageousto cover the top surface of the geomembrane withanother geotextile. Its purpose is to maintain sta-bility of cover soil on side slopes and to preventsharp stones that may be present in the cover soilfrom puncturing the liner. This type of construc-tion is also applicable to secondary containment ofunderground storage tanks for prevention of leak-age into groundwater.

In selection of a geosynthetic for use as a pondliner, consideration should be given to its chemicalresistance with regard to the fluid to be containedand reactive chemicals in the soil. For determina-tion of liner thickness, assumptions have to bemade as to loads from equipment during installa-tion and basin cleaning as well as to pressure fromfluid to be contained.

7.39.10 Geosynthetics as LandfillLiners

Liners are used along the bottom and sides of land-fills to prevent leachate formed by reaction ofmoisture with landfiil materials from contaminat-ing adjacent property and groundwater. Clay lin-ers have been traditional for this purpose (Fig.7.60a). They have the disadvantage of being thick,often in the range of 2 to 6 ft, and being subject topiping under certain circumstances, permittingleakage of leachate. Geomembranes, geotextiles,geonets, and geocomposites offer an alternativethat prevents rather than just minimizes leachatemigration from landfills.

The U. S. Environmental Protection Agency(EPA) requires that all new hazardous-waste land-fills, surface impoundments, and waste piles havetwo or more liners with a leachate-collection sys-tem between the liners. This requirement may besatisfied by installation of a top liner constructed ofmaterials that prevent migration of any constituentinto the liner during the period such facilityremains in operation and a lower liner with thesame properties. In addition, primary leachate-col-lection and leak-detection systems must beinstalled with the double liners to satisfy the fol-lowing criteria:



Fig. 7.60 Landfill liner systems: (a) with filter soil, gravel, drain pipe, and clay liner; (b) with geotextileseparator-filter, geocomposite leachate drain, primary geomembrane and clay liner, geotextile filter,geonet for leak detection, and secondary geomembrane and clay liner.

The primary leachate-collection system should becapable of keeping the leachate head from exceed-ing 12 in.

Collection and leak-detection systems shouldincorporate granular drainage layers at least 12 inthick that are chemically resistant to the waste andleachate. Hydraulic conductivity should be at least0.02 ft/min. An equivalent drainage geosynthetic,such as a geonet, may be used instead of granularlayers. Bottom slope should be at least 2%.

A granular filter or a geotextile filter should beinstalled in the primary system above the drainagelayer to prevent clogging.

When gravel is used as a filter, pipe drains resistantto chemicals should be installed to collect leachateefficiently (Fig. 7.60a).

Figure 7.60b illustrates a lining system that meetsthese criteria. Immediately underlying the wastes is ageotextile that functions as a filter. It overlies the geo-composite primary leachate drain. Below is the pri-mary liner consisting of a geomembrane above a clayblanket. Next comes a geotextile filter and separator,followed underneath by a geonet that functions as a


leak-detection drain. These are underlain by the sec-ondary liner consisting of another geomembraneand clay blanket, which rests on the subsoil.

The EPA requires the thickness of a geomem-brane liner for containment of hazardous materialsto be at least 30 mils (0.75 mm) with timely cover or45 mils (1.2 mm) without such cover. The sec-ondary geomembrane liner should be the samethickness as the primary liner. Actual thicknessrequired depends on pressures from the landfilland loads from construction equipment duringinstallation of the liner system.

Terminals of the geosynthetics atop the sideslopes generally consist of a short runout and adrop into an anchor trench, which, after insertionof the geosynthetics, is backfilled with soil andcompacted. Side-slope stability of liner system andwastes needs special attention in design.

7.39.11 Geosynthetics Bibliography

“A Design Primer: Geotextiles and RelatedMaterials,” Industrial Fabrics Association Interna-tional, 345 Cedar St., Suite 800, St. Paul, MN 55101-1088.



“Construction over Soft Soils,” The Tensar Cor-poration, 1210 Citizens Parkway, Morrow, GA30260.

“Guidelines for Selection and Installation ofExxon Geotextiles” and “Exxon Geotextile DesignManual for Paved and Unpaved Roads,” ExxonChemical Americas, 380 Interstate North, Suite 375,Atlanta, GA 30339.

“Specifiers Guide,” (annual) Geotechnical FabricsReport, Industrial Fabrics Association International,St. Paul, Minn.

“Woven Geotextiles” and “Nonwoven Geotex-tiles,” Amoco Fabrics and Fibers Company, 900 Cir-cle 75 Parkway, Suite 300, Atlanta GA 30339.

Ahlrich, R. C., “Evaluation of Asphalt Rubberand Engineering Fabrics as Pavement Interlayers,”GL-86-34, U.S. Army Corps of Engineers, Vicksburg,Miss.

Berg, R. R., “Guidelines for Design, Specifica-tion, and Contracting of Geosynthetic Mechanical-ly Stabilized Earth Slopes on Firm Foundations,”FHWA-SA-93-025, Federal Highway Administra-tion, Washington, D. C.

Bonaparte, R., and B. Christopher, “Design andConstruction of Reinforced Embankments overWeak Foundations,” Record 1153, TransportationResearch Board, Washington, D. C.


Cedergren, H. R., “Seepage, Drainage, and FlowNets,” 3rd ed., John Wiley & Sons, Inc., New York.

Christopher, B. B., and R. D. Holtz, “GeotextileDesign and Construction Guidelines,” FHWADTFH61-86-C-00102, Federal Highway Admin-istration, Washington, D. C.

Ingold, T. S., and and K. S. Miller, “GeotextilesHandbook,” Thomas Telford, London.

Kays, W. B., “Construction of Linings for Reser-voirs, Tanks, and Pollution Control Facilities,” JohnWiley & Sons, Inc., New York.

Koerner, R. M., “Designing with Geosynthetics,”2nd ed., Prentice-Hall, Englewood Cliffs, N. J.

Matrecon, Inc., “Lining of Waste Impoundmentand Disposal Facilities,” EPA SW-870, U.S. Environ-mental Protection Agency, Washington, D.C.

Mitchell, J. K., et al., “Reinforcement of EarthSlopes and Embankments,” Report 290, Trans-portation Research Board, Washington, D. C.

Richardson, G. N., and R. M. Koerner, “Geosyn-thetic Design Guidance for Hazardous WasteLandfill Cells and Surface Impoundments, “ EPA68-03-3338, U.S. Environmental Protection Agency,Washington, D. C.

Rixner, J. J., et al., “Prefabricated VerticalDrains,” FHWA/RD-86/168, Federal HighwayAdministration, Washington, D. C.


Documents

Geotechnical Engineering