Embed Size (px)
Citation preview
eScholarship provides open access, scholarly publishingservices to the University of California and delivers a dynamicresearch platform to scholars worldwide.
Lawrence Berkeley National Laboratory
Title:The Measurement of Soil Properties in-situ -- Present Methods -- Their Applicability and Potential
Author:Mitchell, J.K.Guzikowski, FrankVillet, Willem C.B.
Publication Date:03-01-1978
Publication Info:Lawrence Berkeley National Laboratory
Permalink:http://escholarship.org/uc/item/0375f55m
THE MEASUREMENT OF SOIL PROPERTIES IN-SITU
Present Methods - Their Applicability and Potential
by
James K. Mitchell, Frank Guzikowski and Willem C.B. Villet
Department of civil EngineeringUniversity of California, Berkeley
ABSTRACT
The measurement.of soil properties in-situ offers the advantages of minimal disturbance, retention ofthe in-situ state of stress, temperature, chemical and biological environments, and cost effectiveness
relative to many types of laboratory tests for evaluation of undisturbed soil properties.
This report is concerned with techniques for in-situ measurement of permeability, strength, stress-deformation properties, and volume change properties; property classes which are of interest in most geo-technical ,engineeringproblems. Emphasis is on test concepts, data analysis and interpretation, andadvantages and limitations of methods, as opposed to details of apparatus and procedure.
Permeability (hydraulic conductivity) is often measured in-situ by means of inflow or outflow bore-hole pumping tests employing either,constant or falling heads. Large scale pumping tests provide themost accurate results, but their high cost generally restricts their use to large projects.
Tests for determining shear strength include the standard penetration test (SPT), static cone pene-tration, vane shear, pressuremeter, and the Iowa borehole s~ear tests. The SPT has been the most widelyused, but is least accurate. The static cone and pressuremeter provide more reliable data and are expect-ed to be increasingly employed. The vane shear test, previously considered to be very reliable in softclays, is now known to frequently overestimate soil strength. The borehole shear test, a rapid and lowcost technique, is limited to soils with some cohesion suitable for stage testing.
In-situ stresses may be determined by pressure cells, hydraulic fracturing and the pressuremeter;deformation characteristics by.the pressuremeter, plate load tests, seismic methods and back analysisof completed projects. The pressuremeter is seen as being of great promise. Hydraulic fracturing tests,while suitable for use in rock, provide results which are often extremely difficult to interpret insoils. Seismic methods of determining elastic moduli involve very small strains, so results need tobe corrected before application in most cases. Plate load tests can yield accurate estimates ofproperties, but costs may be prohibitively high.
Volume change parameters have not often been measured by in-situ methods. However, tne following
techniques may be successfully employed: borehole permeability tests, penetration resistance (both dynamicand static), plate bearing tests, screw plate tests, and the pressuremeter. In-situ permeability testsare particularly well suited for the evaluation of the consolidation rate of fine grained soils. Correla-
tions of penetration resistance with volume change characteristics are empirical and may yield misleadingresults. Both the load bearing and screw plate tests can provide reliable volume change parameters.However, because of practical time limitations, their use is mostly restricted to sands and slightly co-hesive soils.
State of the art equipment, testing techniques and evaluation methods, as reviewed in this report,can provide many geotechnical design parameters with a degree of accuracy which compares favorably withconventional laboratory testing, often with substantial savings in cost and time. It is anticipated thatexisting methods for the in-situ measurement of soil properties will be even more widely accepted, fur-ther developed and supplemented by the introduction of new techniques in the foreseeable future.
I. INTRODUCTION
The determination of soil properties by in-situ measurement has assumed greatly increased im-portance in Geotechnical Engineering in recentyears. Improvements in apparatus, instrumentation,measurement techniques and analysis procedureshave been significant. It is probable that formany projects the determination of properties byin-situ measurement will assume an importanceequal to, or greater than laboratory testing.
There are several reasons for in-situ testing,including:
1) To determine propertiestinental shelf and sea floor
that can't be easily sampledstate.
of soils, such as con-sediments and sands,in the undisturbed
2) To avoid some of the difficulties of labora-tory testin~ such as sample disturbance and the
proper simulation of in-situ stresses, temperature,and chemical and biological environments.
3) To test a volume of soil larger than can con-veniently be tested in the laboratory.
4) To increase the cost effectiveness of an ex-
ploration and testing program.
In-situ tests cannot be considered a panacea,
however, for the following reasons:
1) Some of the tests may not be cost effectivein all cases.
2) Uncertain empirical correlations between mea-sured quantities and properties are often used.
3) Flow (in permeability tests) and stress direc-
tions cannot ,be independently varied in most cases.Principal stress directions in the test may differfrom those in real problems.
4) The possible effects of future changes in en-vironmental conditions cannot be readily deter-mined.
This report considers the evaluation of soilproperties that are needed for engineering analy-ses from the results of in-situ tests. Detailsof test apparatus and measurement methods are notconsidered except as they influence the valuesobtained for the properties of interest.
There are five property classes, one or moreof which may be important in most geotechnicalproblems:
1) Permeability (hydraulic conductivity)
2) Strength
3) In-situ stress and deformation characteristics
4) Volume change characteristics
5) DUrability cor susceptibility to changes inproperties due to time and changes in environmen-tal conditions.
This report is concerned mainly with the firstfour of these property classes. In-situ tests arenot generally suitable for prediction of the fifthproperty class, but they are very appropriate formonitoring changes with time. Some of the in-situtest types, such as pressuremeter and penetrationtest~ can be used to deduce information about morethan one property. Similarly, some of the princi-ples and considerations in testing relate to allthe properties.
This report draws heavily on material present-ed at, and contained in, the proceedings of tworecent conferences; namely the European Symposiumon Penetration Testing (ESOPT), held in Stockholm,
June 1974, and the ASCE Geotechnical EngineeringDivision Speciality Conference on In-Situ Measure-ment of Soil Properties, held at North CarolinaState University, June 1975. It is organized interms of property class rather than by test type.A list of pertinent references summarizing the
current state-of-the-art is appended to thisreport.
2
II. PERMEABILITY
A. Introduction
Many difficult construction problems are di-rectly related to the presence of ground waterflow. .Groundwater seepage has been responsiblefor slope and base instability in excavations,for face and roof instability in tunnels and forpiping and erosion in earth/rock dams. Seepagelosses in dam foundations or from reservoirs di-
rectly influence the safety and economics of waterconservation projects. Large local variations ofpermeability are a common condition and lead totlow concentrations which may cause dewateringdifficulties such as local piping. The success ofmany projects therefore depends upon a knowledgeof in-situ permeability and its degree of varia-tion.
All soil and rock deposits are permeable tosome degree. The coefficient of permeability, k,probably exhibits the largest range of magnitudeof any parameter indicative of geotechnical prop-erties. The general range of in-situ permeabilityvalues for different soil and rock types is pre-sented in Figure II~l.
102
k-cm/sec.
10-2 10.3 10.4 10.5 10-6 10.7 10.810 10-1
4-- Gravels
Silts
Sands
XBL 776-9396
Fig. 11-1 Approximate range of permeability (kl.
in soil and r~ck (from Milligan, 1975)
An analysis of dozens of dam foundations has
shown that a conceptually correct, simplified ap-proach based on reliable geologic data will yieldequally good results as the most involved mathema-tical simulations (Milligan, 1975). In many cases,if the coefficient of permeability can be evaluatedto within an order of magnitude a safe, economicaldesign will be possible.
For many applications, the object of a perme-ability measurement is simply to determine if aproblem exists. If so, a defensive design consist-
ing of four steps may be appropriate (Gordon, 1975):
1) Assess the problem.
2) Design to accomodate a chosen range of perme-ability.
3) Monitor subsequent behavior.
4) Take appropriate action.
Defensive design may be applied to seepagecontrol for dams and reservoirs, storage of conta-minated water, disposal of solid wastes and pre-vention or correction of landslides.
In-situ permeability values may be utilized
in fluid flow calculations or in conjunctio~ withlaboratory measurements of stress-strain behaviorin order to deduce rates of consolidation.
~
B. Fluid Flow in Soils and Rock
In soils and rocks, the flow of water throughpore space, voids, discontinuities or cracks isusually assumed to be laminar and to obey Darcy'sLaw, i.e.,
v = k . i (II-I)where:
v = flow velocity
i = hydraulic gradient, head lost per
unit length of flow path
k = coefficient of permeability (hy-
draulic conductivity), normallyexpressed in velocity units
Experience has shown that this relationshipis valid for a wide range of soil conditions. Itis possible, however, that the presence of pre-ferred flow paths in an otherwise essentially im-pervious material may decrease the significanceof an "average" permeability value.
In addition to being the governing equationfor fluid flow in geotechnical problems, Darcy'slaw also provides the basis for the measurementof permeability. The procedure common to all di-rect methods of permeability measurement is to im-pose a hydraulic gradient and measure the resultingflow across a known cross sectional area. Darcy'slaw is then employed to calculate a value of per-meability. Deviations from direct proportionalitybetween flow velocity and gradient may sometimesdevelop when changes in soil structure occurduring flow (Mitchell, 1976).
c. Variation of In-Situ Permeability
The in-situ permeability of soil and rock isinfluenced by both microscopic and macroscopicfeatures. Microscopic properties such as voidratio and particle size, shape and orientation may,in some cases, be retained through undisturbedsampling.
The in-situ permeability of clean sandy soils,which lose their in~situ structure during evencareful sampling, may be estimated in the labora-tory by testing samples which have been carefullyrecompacted to the appropriate relative density.At best, such a procedure may provide an estimate
of in-situ permeability, since laboratory compac-tion cannot be expected to reproduce a complicatedin-situ soil structure which is the product ofmany physical and chemical processes.
The effects of important macroscopic features,such as sand lenses, fissures and clay seams, can-not usually be duplicated in laboratory testing.However, it is these macroscopic features whichwill most likely govern field behavior.
A comparison of corresponding field and lab-
oratory permeability test results will usuallyshow the field permeability to be considerably,but unpredictably, higher than the values measuredin the laboratory. A reliable in-situ determina-
3
tion of permeability is therefore a necessity formany projects.
D. Direct Measurement of In-Situ Permeability
Methods which are commonly used for the di-rect measurement of permeability have been summar-ized by Milligan (1975) and are reproduced inTable 11-1 of this report. The procedural detailsof these tests are thoroughly discussed in thereferences cited in Table II-I.
It has been noted that the approach which iscommon to all these direct methods is to impose aknown hydraulic gradient and measure the result-ing flow. If a constant hydraulic gradient ismaintained during testing the test may be classi-fied as a "constant head" test. If the imposed
hydraulic gradient decreases or varies with timeduring testing, the test is classified as a "fall-ing head" or "variable head" permeability test.
If the induced flow during testing proceedsfrom the measuring device into the soil to betested, the test is described as being an "in-
flow" test. Similarly, if the imposed gradientinduces flow from the soil mass into the measuringdevice, the test is classified as an "out-flow"test.
In-flow tests tend to clog the soil voidswith dislodged particles, resulting in the measure-ment of a coefficient of permeability, kin' thatmay be lower than the true in-situ value. Out-flow tests tend to erode particles from the soilskeleton, increasing soil porosity and resultingin the measurement of a coefficient of permeabi-
lity, kout, which may be higher than the true in-situpermeability. .
The value of the ratio, (kout/kin) may be aslarge as 500 in extreme cases (Milligan, 1975).According to Milligan, the true in-situ permeabi-lity, k, can be estimated by:
k = f(k. ). (k.)V tKout l.n (11-2)
Use of this equation implies a judgment that
the true in-situ permeability lies between kin andkout, but is closer to kin than kout. Therefore,the implicit assumption is that the erosion effectof outflow tests is more severe than the cloggingeffect of in-flow tests.
In the remainder of this section, the methodsof in-situ permeability measurement are examinedin more detail, together with their applicabilityto engineering practice.
The Borehole Permeability Test
Borehole permeability tests are performed bypumping water either ou~ of a borehole (drawdowntest) or into a borehole (infiltration test). Thetest may be performed as a constant head test in
which the rate of pumping which is necessary tomaintain a constant water level in the borehole is
measured. Alternatively, a variable head test may
be performed by observing the change in water lev-el in the borehole after pumping has stopped.
A fully or partially cased borehole is usual-
METHOD
..
Augerhole
A
Test Pit
B Cased borehole
(no inserts)
C Cased borehole
(inserts used)
i) Sand filter plug
ii) Perforated/slotted
casing in lowest
section
iii) Well point placed
in hole, casing
drawn back
D Piezometers/Permeameters
(with OR without casing)
E Well pumping test
F Test excavation
pumping testes)
TABLE II-1-
TECHNIQUE
Shallow uncased hole in
unsaturated material
above G.W.L.
Square OR rectangulartest pit (equivalent to
circular hole above)
i) Falling/rising head,
~ in casing measured
VS time
ii) Constant head main-
tained in casing, out-
flow, Q VS time
i) Generally falling head
~ measured VS tiIOO only
ii) Variable heads possible
iii) As for (ii) above
i) Suction Bellows appara-
tus (independent of
boring) inflow ONLY
measured VS time
ii) Short Cell (Cementation)(independent of boring)
Outflow ONLY measured
VS .time
iii) Piezometer tip pushed
into soft deposits/
placed in boring,
sealed, casing with-
drawn/pushed ahead of
boring.Constant head( outflowmeasured VS t~me. Vari-able heads also possible
Drawdown in central wellmonitored in observationwells on, at least, two 900radial directions
Monitoring more extensive
than E , during excava-
tion dewatering
(Initial construction stage)
DIRECT TESTING OF IN SITU PERMEABILITY IN SOILS
(from Milligan (1975»
GRAVEL
APPLICATION TO:
.;
Onlywherek>1O-3cm/sec.;
.;
.;
.;
.;
.;
-
.;
-
.;
.;
SAND
.;
.;
.;
.;
.;
?
.;
-
.;
.;
I
!1
I
+II
I
SILT
1
?
?
?
?
?
.;
.;
-
IY
.;
(?)
.;
(1)
PROBLEMSCLAY
- Difficult to maintai~
water levels in coarse
gravels
-
- Borehole must be flush-
ed. Possible lines clog
base (falling ~).
Pumping (rising ~)
where WL lowered
excessively.
-
-- Single tests only. .
Cannot be used as
boring is advanced.
-
-Restricted to fine
I sands, coarse silts,variable bellows
required 'k' range
lO-~ to 10-7 cm/sec.
- Carried out in adit
OR tunne 1
.;
Possible tip' smea:!:'
when pushed. ~u set upin pushing tip.
Danger of hydraulic
fracture
- Screened portion
should cover complete
stratum tested
-Expensive, but of
direct benefit to
contractual costing
METHODRATING
Poor
Poor
Fair
Fair
Fair toGood
Good(local zones)
Excellent
(Masspermeability
1
0f foundationmaterial)
I
REFERENCE
U5BR, (1974)
Lacroix (1960)
Hvorslev (1951)
USBR, (1974)
Hvorslev (1951)
~
Golder, Gass (1963}
Golder, Gass (1963)
Gibson (1966)Wilkinson (1968)Hvorslev (1951)
Bjerrum et aI, (1972)
Todd (1959)
--
ly employed in this method. The permeability of alocalized ?one of $oil surrounding the base anduncased portions of the borehole is computed bythe application of theoretically derived equations.The equations which are appropriate for a numberof test configurations are reproduced in Figure'II-2, from Hvorslev (1951).
The borehole permeability test is usuallyperformed as an in-flow test, partially becauseinexpensive pumps are ~imit~d in their ability toaffect drawdown ~t a cqn$tant pumping rate, andpartially because it is usually difficult toaccurately monitor the drawdown water level in theborehole. Con$tant head tests are preferred to
variable he~d tests, b~cau$e they are easier toperform properly and have been found to providemore reliable and consistent data, as discussed
by Schmidt et ale (1976).
Permeability testing ~n a single borehole islikely to yield misleaging ~esults. Twq simpli-fied geologic profiles ar~'presenteCiln F~gureII-3. It can be visualized that the results of
p~rmeabilitytesting in these profiles will varyg~eatly with borehole location and testing depth.Permeab~lity tests in Borehole #1 will consisten-ly yield the fairly low value of k exhibited bythe silt stratum, but the important flow pathsafforded by the sandy layers will remain undetect~ed. Test results from Borehole #2 may, or may not,
reflect the presence of the pervious jointing pat-terns. The probability of detecting a joint dur,..
ing testing will depend upon the relative dimen-sions of the test section and the joint spacings.
BHiI
(a)
--" ~
I
.'
I
<,""''j-'''~':-.'_.L..' I
1~~S~{i;:,
r,:"~LL.=...~iJ slt'~I''''~~1~~~~~
J FineSanqJJ:.:':~""'1*'-
)'"==~-J ' L..J. .L.-I:-:-.J~
- .J. k'
l
:-:r::~
,r
' -'
.
,,1,
';-:""""'7
,
'.
,
:
,
:::-
,
., ,,' : '
,
" ,.:;:, 'i;;":" "
,
::~
,
:'cr
'..,,'.'.'""'f. . ," :'",,':::'Medsaiid,Giavei\:".1 L
,
sible ~Li~~(sJi~r r
~'
1
:'d.:'
l
':'~~
I
'::{)
r
:-\~!
r
":~
l
:'::~
of samples, In SItu ~I
Jtests
.1 "
BH2
(b)
Fig. II"..3 Limitations of sampling/in-situ testsin a single borehole: (a) in soils
(b) in rock (from Milligan, 1975)
5
If such geologic discontinuities are encoun-tered in a borehole permeability test, their pres-encemay dominate the testing flow conditions andbias the measured value 9f permeability. However,the effect of geologiCicliscontinuitieson fullscale performance cannot be directly predictedfrom their effect on small scale tests. The in-
vestigator must refer to a geologic site model,and, if necessary, perform more small scale testsor resort to full scale testing.
The primary advantage of the borehole permea-bility test is its simplicitY. The 'results may beviewed as general indicators of the order of mag-nitude of in-situ permeability appropriate for therelatively small zone of tested soil. Theuncer-tainties involved in a single test are signifi-cant, but their importance can be reduced by per-forming a large number o~ tests.
Large Scale Pumping Tests
Large sca~e pumping tests are performed bypumping water into or out of a screened well em-bedded below the natural groundwater table. As
pumping proceeds, the resulting change of ground-water leve+., is monitored.~:n surrounding, observa-
tion wells. Generally, .fouror more of these ob-servation wells are.employed. Pumping tests maybe categorized as equilibrium (steady state) ornon-equilibrium (transient flow) tests.
In an equilibrium pumping test, the ground-water level measurements are recorded once a con-stant water level has been attained in the obser-
vation wells. The time required for this equili-brium to occur may be very great, especially whenthe test is performed in soils of low permeabili-ty. An average value of permeability for the de-posit under study can be computed from the testdata by relationships such as the Thiem Formula,which a~e described by Lang (1967) and others.
In the non-equi~ibrium test, the rate ofpumping, and the consequent rate of water levelchange in the observation wells are recorded. Thetest data can be analyzed by a procedure such asthe one proposed by Theis (1935). The advantagesof transient flow testing are that testing timemay be reduced as compared with equi~ibrium testing,and groundWater drawdown may be controlled, orhalted if necessary.
Large scale pumping tests with observationwells provide the most reliable, but most expen-sive, permeability data for relatively perviousdeposits. Such a testing program was employed byAhmad et ale (1975) in order to estimate seepagelosses from an artificial lake. The soil depositin question was an unconfined sand aquifer, sam-ples of which exhibited a uniformity coefficientranging from 2 to over 20. Surprisingly consis-tent values of permeability coefficients were ob"..
tained, ranging from 0.10 to 0.15 cm/sec.
Large scale pumping tests are currently thepreferred method for determining the in-situ per-meability.for large construction projects whichcan justify the.costs involved. The test con-ditions closely simulate full scale site dewater-
ing. In:f~ct'aJu:nct~qning site dewatering sys-tem may be analyzed'as a pumping test, and the
c
~0u
"a
~II
~~
t2
He
laboratorypermeameter
Flush bottomat impervious
boundaryBA
Fig. II-2.
6
Flush bottomin uniform
soilC
Soil in casingat impervious
boundaryD
Soil in casingin uniform
soilE
Well point-filterat impervious
boundaryF
Well point-filterin uniform'
soilG
Configurations and appropriate equations for borehole permeability tests (after Hvors1ev, 1951)
I~Constant Head
4'q'Lk. = ". D' . H.
A
--
B 'Ikm = 2 . D . H,
C 'Ikm= 2.75. D'.H,
D4. (
" k: D
k: == 'I 8"' k." .;;:; + L)". D"H,
E (" k: D)4''1' -'-'- +L11 k. m
,,'D"~k:
Variable Head
d"L HIk.= D'. (t, - 'I) InIi.
L HIk.= -In - for d= D" - I) H,
,,'d' H)km= 8, D . (t, - ,) InIi.
".D H)k = -In- for d=Dm 8, (I, - '1) H,
,,'d' HIkm = 11 . D . (t, - '1) In Ii.
".D H)km=-In- for d=D
II . (t, - '1) H,
(" k: D
)dO. -'-'-+Lk ' 8 k. m In !!2. D''(/,- ,) H,
" D-'- +L8 m H)
(k: =k.k.=-ln H- for d - D" - I) 2 -
(" k: D )dO. -'-'-+L
k: = 11 k. m In~D' .(I.- ,) H," D-'-+L11 m H) (k: =k.k.=-In n- for d - D'I - I) I -
[2mL ;
- d"ln [2mL ~ (F IkA=q'ln D+vl+C~L )
'JI kA= 8~+yl+\~n H
2. " .L .H L. (/, - 'I) In-.!, H,
-
GkA q'ln [~+)I + (";,L)'J
2,,,, L 'He
Basic Time Lag
d' 'L
k.= D"T
Lk. = T for d = D
"d'km = 8 . D . T
,,'Dkm = g-:-r for d = D
" .d'km= 11. D. T
,,'Dkm = ~ for d = D
(" k: .!!.)+ Ld"' 8"'~ m
k: = D. . T"D
L
{k' =k.-'- + .
~ for d=Dk. = T
("k;D)d". -'-'- +L
11 k. mk: = D' . T
" D, IT.;;:;+ L (k: =k.
k. =-r- for d=D
Notation
D = Diam, intake. sample(em)
d = Diameter, standpipe(em)
L = Length, intake, sample(em)
H. = Constant piez. head(em)
H) = Pia. head for I = I)
(em)H. = Pia. head. for I = I.
(em)'I = Flow of water (cm"/s«)1 = Time (see)
T ,,;.Basic time lag (see)k; = VerI. perm. casing
(cm/see)
k. = Vert. perm. ground(em/see)
kA= Harz. perm. ground(cm/see)
km = Mean coeff. perm.(cm/see)
m = Transformationratio
k,.;= vkA'k. m = vkAlk.In = log. = 2.3 log)o
[2mL ~ (2mL
)~k - d'.ln D+yl + \D)JA- '8'L'T
d2 . In (4mL
)2mL Dfor -
D >41
kA =- 8 'L. TkA
(4mL
)d"ln -D H)In-
8 . L . (I, - ,) H,
2mLfor D > 4
d"ln[~+jt + (~nkA= 8 . L . T
d' . In(2mL
)kA= D~mL
for Ii > 4Delerminalion basic time
lag T
ASSUMPTIONS
Soil at intake, infinite depth. and directional isotropy (k. and kA constant). No disturbance, segregation, swelling, or consolidation of soil. No sedimentation orleakage. Nq air or gas in soil, well point. or pipe. Hydraulic losses in pipes, well point, or filter negligible.
d"ln[~+~J H)kA 8'L'(/,-II) InII.
d' . In(2mL
)D HI mLkA 8 . L . (I. - ,) In II. for Ii > 4
results used to refine the system as work proceeds.
Potential Errors in In-Situ PermeabilityMeasurement
The sources of error in a typical in-situpermeability test include:
E.
I} Inaccurate water quantity measurement
2} Inaccurate head measurement
3} Inaccurate test length
4} Inaccurate measurement of test sectiondimensions
5} Plugging or smearing of fractures or pores
6} Use of excessive pressures
7} Unknown flow resistance of measuring system
The effects of 1, 2, 3, and 6 may be mitigatedby the use of careful procedures, but in general,the contributions of the other sources of error are
difficult to isolate or interpret.
F. Conclusions
Most fine grained soils are relatively imper-vious, and fluid flow problems in these soils occuras the result of geologic discontinuities, such asfissures or sand lenses. A painstaking soils in-vestigation is reqpired to detect the presence andextent of these discontinuities. Coarser soils
which exhibit a coefficient of permeability greaterthan about 10-5 cm/sec can be expected to causeproblemsiri seepagecontrol.
Conventional soil investigations, ~d manyof the commonly used borehole permeability methods,can yield misleading.permeability information.Consequently, large scale field.observations arean essential supplement to any important permea-bility testing program.
If existing methods are applied with discre-tion and engineering judgment, the coefficient ofpermeability can often be predicted to within oneorder of magnitude, which is usually sufficientto enable a saf~, economical design.
III. SHEAR STRENGTH
A. Introduction
Almost all geotechnical projects involve someconsideration of.soil shear strength. Soil strengthis most commonly described as a peak shear str~ngthin terms of the Mohr failure envelope parameters,friction angle ~, and cohesion c, but the residual,low strain and yield strength values may also beimportant in many projects. This discussion doesnot concern the many different ways of describingsoil strength, other than to make a distinction
between drained and undrained loading conditions.For clay soils, the peak undrained shear strength,
Su (half the unconfined compressive strength), istaken as the measure of strength. For non-clays,the angle of internal friction, ~', in terms ofeffective stresses is considered as the governingstrength parameter.
7
Once an in-situ strength parameter has beenobtained, it may be substituted in a limit equili-brium design analysis, used to classify the soilstratigraphically, employed as an index of anothergeotechnical property, or used as a quantitativebasis for decisions regarding further testing.
In-Situ Strength Testing
The advantages of strength measurement in-situare threefold: the large effects of sampling dis-turbance on soil strength can be minimized, thecosts of undisturbed sampling and testing can bereduced, and in-situ measurement eliminates the
need to reproduce complex chemical, biological,thermal and stress environments.
A disadvantage of in-situ testing is that thesoil strength parameters must be measured "as is."Soil strength can be affected by the imposition ofpost-construction environmental conditions. Post-construction changes in the stress, thermal andchemical environments may often be simulated inlaboratory testing. It is, however, not usuallypossible to measure the resulting "future" beha-vioral properties in-situ.
The most commonly used in~situ strengthtests are listed in Table III-I, after Schmert-mann (1975). The following sections describethe use of these tests in engineering practice.
TABLE 111-1
Commonly Used In-Situ Strength Tests(after Schmertmann, 1975)
B. The Standard Penetration Test (SPT)
A simple method of obtaining at least someinformation concerning the degree of compactnessof soil in-situ consists of counting the number
of blows of a weight dropping a given distancerequired to drive a sampling spoon for a distanceof one foot (30 cm). The resulting blow count, N,is at the present time probably the most widelyused index of subsurface soil conditions in theU.S.A.
AlthoughSPT, as it iswidely known,on blow count
the details and advantages of thedefined in ASTM Standard 1586, arethe effects of different factors
are a subject of continuing study.
Advantages of the SPT include economy of use,simplicity of procedure and widespread acceptanceand familiarity among practicing civil engineers.Disadvantages of the SPT include the uncertain
Test Abbreviation
Standard Penetration SPT
Quasi-Static Cone Penetra-tion Q-CPT
Vane Shear VST
Pressuremeter PMT
Borehole Shear BST
effects of a large number of influencing factors,
and the poor reproducibility and large variabilityof test results.
Estimating Strength of Sand from the SPT
A correlation between blow count and ~' forsands has been established by DeMello (1971) andis reproduced as Figure III-I. Use of this chartyields conservative values of ~', which are notapplicable to shallow soil depths.
60
C:J0(J~0CD 20
I-a..U')
<U.20~ 40z-
00 I 2
Overburden Pressure- kgf/cm2
Fig. 111-1 Method for estimating effective fric-tion angle (~') from SPT blowcount (N)
(based on DeMello's 1971 analysis,
USBR data)
Another common practice is to estimate ~' ,
,using relative density, Dr' as an intermediateparameter. A correlation of relative density,Gibbs and Holtz (1957), is often employed, al-though at high relative densities the relation-although at high relative densities the relation-ship suggested by Bazaraa (1967) may be more ap-propriate. The original Gibbs and Holtz relationis reproduced in Figure 111-2. In Figure 111-3,the Bazaraa, Gibbs and Holtz, and a third rela-tionship suggested by Schultze and Melzer (1965)are compared. An inspection of this compositechart shows that an estimate of Dr from the SPTmay easily involve a large uncertainty, even ifthe overburden pressure and "true" blow count areknown with certainty.
Once the relative density estimate and splitspoon sample identification have been obtained, acorrelation such as that recommended for quartzsands by Schmertmann (1975) may be used to obtain~'. This correlation is reproduced in Figure 111-4. Because estimating Dr by the SPT can easilyinvolve an error of I20%, the error in ~' may beas large as I5° (Schmertmann, 1975). Analyses,such as bearing capacity calculations, which are
sensitive to variations in ~', must thereforeinclude large safety factors if this approachis used.
8
3
N
~~ 00QlNE
~ 0,5l
N~ 1.0mz
~g 1.5wa::I-m
w2: 2.0
t;w
I::w
<l 2.5u
i=a::w>
3.00 60 70 8010 20 30 40 50
STANDARD PENETRATION RESISTANCE N, bl/f! XBL 776.939B
Fig. III-2 Correlation between relative densityand standard penetration resistanceaccording to Gibbs and Holtz (1957)
STANDARD PENETRATION RESISTANCE, BLOWS/ft.
60 70 80a
-- GIBBSANDHOLTZ(1957)
BAZARAA (1967)
-'-'-'- SCHULTZE AND MELZER (1965)
gf~ 1.0Ii;w>t 1.5wu.u.w
<l 2.00i=a::w>
~6!
2.5
3.0
XBL 776.9395
Fig. III-3 Three correlations between relative
density and standard penetrationresistance
45
CII
~ 43CJ'I
~ 41I
~39
.~ '5l0
~ 35EE 33)(
0:2:
31
290 20 40 60 80
Relative Density. Dr - percent
100
Fig. 111-4 Approximate correlation between effec-tive friction angle (~') and relative
density (Dr) in quartz sands (fromSchmertmann, 1975)
Estimating Strength of Clay from the SPT
SPT results are occasionally employed to es-timate the undrained strength of clays. A number
of published correlations between Su and blowcount for insensitive clays are presented in
Figure 111-5. There is a wide degree of scatterin the correlations, and corrections for overbur-den pressure and overconsolidation ratio are notavailable. Clay sensitivity may decrease the blowcount for a given undisturbed strength because ofstrength loss during penetration, in the mannerdepicted in Figure 111-6.
One "rule of thumb" is that su' in tsf., isat least as great as NilS (Schmertmann, 1975).
30
zi20::)
00
~0CD 10
.-a..en
Fig. 111-5 Some correlationscount (N) and unconfined
(qu) in clays (from DM-7(from Schmertmann, 1975)
between SPT blow
compressive strengthunlessnoted *)
1.0
0.8
EO.6(J1
~ 0.4z
0.2 Richmond Cloy(After Casagrande, /966)
aa 2 4 6 8 10
Partial Sensitivity Along Inside and Outside of SPT Sampler
(assumed: 0.50' remolded sensitivity=0.50' St)v.tIere N =observed blow count
NSt =correctedblow count for undisturbedstrength
Fig. 111-6 Estimated decrease in standard blow
count with increasing clay sensitivity
at constant undrained strength
(from Schmertmann, 1975)
9
Factors Which Influence SPT Blow Count
The methods used for the prediction of ~'from the SPT allow variously for the effects ofsoil type, overburden pressure and relative densi-ty. SPT results are, in addition, affected to asignificant degree by a number of other influencingfactors:
1) State of Stress. Blow count is affected not
simply by overburden pressure, but by the entirein-situ stress state, as well as stress history.
The in-situ horizontal effective stress, 0h' mayactually have twice the proportional effect ofvertical stress on blow count (Schmertmann, 1975).The 1ifficulties of estimating horizontal stressesare discussed in Section IV of this report. At
present, the possible influence of 0h' on measuredN values may be considered in only a very quali~tative sense.
2) Pore Pressures. The generation and dissipa-tion of pore pressure during the SPT may have amajor influence on the resulting blow count. Soils,such as fine,loose sands, which generate signifi-cant positive pore pressures during driving, willyield decreased values of N, as a result of thedecreased effective stress levels during driving.Dilatant soils will generate negative pore pres-sures during driving and increase the recordedblow count.
3) Sampler Side Friction. There is evidencethat side friction between soil and the samplingspoon may corttributea significant fraction of themeasured penetration resistance. The results ob-tained by Schmertmann (1975), reproduced in Table111-2 show that the contribution of soil-samplerside friction to penetration resistance increasesdramatically with increasing soil friction ratio,FR. The friction ratio, in turn, will generallyincrease with increasing soil cohesion. For theinsensitive clay in Table 111-2, 86% of the mea-sured penetration resistance originated from sidefriction.
Table 111-2 CPT-Based N-Resistance Distributions
(Schmertmann, 1975)
FR = Friction ratio = unit side friction f unitend bearing stress
St = Sensitivity
Part of Blow CountDue to
FR Typical Soil TypesSideEnd
Bearing Friction
1% Sand, above and 56% 44%below groundwater
2% Silty marl, silt, 34% 66%clayey sand
4% Sandy clay, clay 25% 75%with St - 4
8% Insensitive clay, 14% 86%peaty soils
10
TABLE III-3. MEASURED INCREASES IN "N" WITH IMPEDANCE
OF HAMMER FREE FALL (Schmertmann, 1975)
GWT = ground water table
TABLE 111-4. SOME FINDINGS FROM WAVE EQUATION STUDIES OF SPT(from Schmertmann, 1975)
r- . . .. . -~'ar
I
~_._------
1. Energythruh-----
2. Totalon sam
---.---.---
3. Distri
R alon
4. Rods:
4) Testing Technique. One of the most signifi-cant sources of uncertainty in the SPT relates to
above ground apparatus and test conditions. Afull discussion of these considerations may befound in DeMello (1971), Fletcher (1965) andSanglerat (1972).
The usual cathead-slackened rope procedureof releasing the drive weight may result in a sig-
nificantly larger blow count than that obtained ifa "free fall" weight release mechanism is used. Asummary of data on this effect is presented inTable 111-3, from Schmertmann (1975). The unanswer-
ed question is: "How much drop friction is built
into existing correlations between N-values andsoil properties?"
Application of the one-dimensional wave
equation to the SPT by McLean et al. (1975), hasprovided some important information concerning therelative importance of different details in SPTtechnique. Some findings of this study have beensummarized by Schmertmann (1975), and are present-ed in Table 111-4.
Conclusions
The SPT is, and will remain, an important
--
Investigators Soil Depth GWT N-range N/NFFNotes
Frydman (70) Natural ? ? 2 - 100+ 1.4 In Israel, 2I turns sliprope
I
over 10-12"cathead
Zolkov, (71, Dune sands I-12m 'above 7 - 60 1. 8 @2m do
72) SP, SP-SM 1. 5 @llm w = 1-6%
Serota & Dry sand At sur- Dry 10 - 20 1.06 In England,
Lowther (73) compacted charged Dr=95%, 1
in a drum surface rope turn oncathead
1.21 do., 2 ropeturnsdo. .but cat-
1.4head hammer
system weighedless
i:-:lati: rtance
IComments
nput=E . N l/Et Very 1mportantmer sys em
sistanceVery important
N 8R (kips) ifer =R E=70%. (350 ft.lb)
for N .2:.15
but ion of N increases c. 40%
g samplerImportant
going from 100 -+0%end bearing
length Minor % change important
at very low N
type (A,N) Minor NN slightly larger
loose joints i Minor than NA1buckling
I
Minor Maybe Impt. in 3D
geotechnical tool. The precision of the test isadmittedly low, but many practicing engineers areable to apply judgment and local experience to de-rive satisfactory designs on the basis of personalor published correlations. The SPT is most prop-erly applied to the "day to day" design projectswhich do not justify more sophisticated testingand in areas where the soil conditions are reason-
ably well known. Standards for equipment and pro-cedure would make the SPT results more quantita-tively meaningful. The general effects of manyinfluencing factors in the SPT are becoming betterunderstood; it is important that they be consider-ed in the interpretation and application of testresults.
C. Cone Penetration Tests
Many types of cone-tipped penetration devicesare used in Europe and are receiving increasingacceptance in the United States because of thesimplicity of testing, reproducibility of results
and the greater amenability of the test data torational analysis. The general types of cone pene-tration devices are summarized in Table 111-5.
The quasi-static, Q-CPT or "Dutch Cone" methodis the most commonly used in engineering practice.
Dynamic cone penetration tests are subject to the
same disadvantages as the SPT with an additional
drawback of not. pr~viding simultaneous sampling. Acombination dynamic-quasi-static system facilitates
penetration through stiff layers which resist thestatic cone.
Methods for Estimating Strength
In the widely used Dutch cone test, a pene-trometer of 10 cm2 base area and 60° apex angle isadvanced vertically at a constant rate (2 cm/sec)into the soil to be tested. The friction jacketadvances simultaneously (electrical cones) oralternately (mechanical cones) with the tip. The
resulting tip resistance, qc, and side friction,
TABLE III-5.
11
fs, are measured separately.
A conservative method has been developed
de Beer for determining~' fromqc on thebasibearing capacity theory (Sanglerat, 1972, ESO!1974). A less conservative, semi-empirical c(
lationbetween~' and qc has been used in theand is presented in Figure 111-7, after ESOPT
NE~ 0~ 0e::J
:a 0.2ea..c:
~ 0.4~
B~
~ 0.6
~3-
Fig. 111-7 Method for estimating effective an
of friction (~') from static cone:bearing resistance (qc) reported i:in USSR (ESOPT, 1974, p. 151)
Large scale model footing tests on sand (Muhs .
Weiss, 1971) have shown that:
qc(kg2
)= 0.80N
cm y(I:
where Ny is, Terzaghi's bearing capacity factor,dimensionless function of ~I. This correlatiorapplies most directly to the bearing capacity (
GENERAL TYPES OF CONE PENETRATION TESTS
(from Schmertmann, 1975)
r-------'-Tip Advance! t"lhere
Type ---.-.-......-- ,'-''',,,,,, -.-,.. .-.-. ---.-.----c------.----. NotesMethod Rate Used
--'-------- -'
1. Static During increments of 0 Research Too slow forconstant load general field use
2. Quasi- Hydraulic or mechanical 1-2 World- Usually 10 cm2static jacking cm/see wide 60° cone point
3. Dynamic Impact of drive weight variable world- Great.varietyofwide sizes, weights, etc.
4. Qu.asi- Combines 2. & 3. using France Uses specialstatic & dynamic when Q-CPT can- switzer- penetrometerdynamic not penetrate further land tips
5. Screw Rotation of a weighted, variable Sweden
helical cone Norway
6. Inertial Dropped or propelled variable Offshore, Useful for soils
into soil/rock surface during Military in inaccessiblemeasured areas
deaccelera-Ition
---
0'{\'
U\'\\\
41-, '
y'\ ~
6 \ 1CD \ III
0 \~ 8 \!- \ Ii ,~ \ ..~ ,'1ii10. ,1'ii -a::
12
4000
NI'q
6000 eooo 12QX)r--COE~ =90.cJ> =105.BIcJ>=1.0
I)()()()
Biarn and Gr.sillon ,1972; Equation (17)parabolic variation( DI B )cr! 20
I " LinNI:""
<.. variationM.~rhof
"-..(19E1a)""
0..toto't!
I
L14
16
18
Fig. 111-8 Effect of depth on resistance factor
Nyq according to different theories(Durgunoglu-Mitchell, 1975)
putations for shallow footings. In reality, qc isnot uniquely correlated with ~'. The effects ofother factors, including soil compressibility,stress stat~ and penetrometer base roughness havebeen recbgnized by Schmertmann (1975) and.Durguno-glu and Mitchell (1975).
A procedure has been developed for estimating
~' from qc on the basis of a rigid-plasticwedgedisplacement bearing capacity theory with empiricalcorrections for a qircular cone shape by Durgunogluand Mitchell (1975):
qc = c N S + y B N Sc c s yq yq (111-2)
For cohesionless materials, c = 0; and
N = F (~', ~, o/~', D/B)y~
where:~ = penetrometer base semi~apex angle
B = width of penetrometer base
D = depth of penetrometer base
Sc,Syq = shape factors
0 = friction angle between penetrometerbase and soil
Ys = unit weight of soil
Bearing capacity factors, Nyq' for the caseof a Dutch cone and ~'= 45°, are presented in Fig-ure 111-8 as a function of soil depth, together
with N as calculated using other methods proposedyq
12
NE
Z 100.0~uCT
L&JUZ 50.0<{
tninL&Ja::
L&Jz0uuj::
;5If)
L&J
~ 10.0~5::>
Fig. 111-9
500.0
IjIpr.dict.d ultir'nat. valu.s. .,(: 30. 8 let>:0.4 -0.5
{
e Muhs Weissdata obtained t. Kahl et at.from Meyer-hof A Kerisel(1974) 0 Melzer
5.0
50
. - -~._-I
.--I
I
-r--I
1.035 40 45
ANGLE OF INTERNAL FRICTION' _deg
Ultimate cone resistance as a function
of friction angle for several sands(from Durgunoglu and Mitchell, 1975)
by Berezantzev (1961), Biarez and Gresillon (1972)and Meyerhof (1961). Using equation (111-2), andthe concept of a critical depth (Durgunoglu and
Mitchell, 1975), ~' has been correlated with qcas presented in Figure 111-9. The data pointsshowthe qc ~ ~' correlationpresentedby Meyerhof(1974) for actual field cases.
Sand profiles which possess a constant ~' and
exhibit a linear variation of qc with depth may beanalyzed using a method suggested by Janbu and
Senneset (1973):
qc + a = Nq (pi + a)(111-3)
where:
a = penetration resistance intercept
parameter, as indicated in Fig-ure 111-10
p' = effective overburden pressure
Np = slope of qc - p' profile
Nq = F (tan ~') = Np + 1, as indicatedin Figure III-II
Net Cone Resistance
N
~ I~ £1u.a'-b. T
~
.. \
&i \
~ 100 \.0 \...Q)>
0Q)>
"";:
~ 200\I-\I-
I.rJ
Cloy or Silt~u>a
Fig. 111-10 Principle of static cone interpreta-tion, Janbu and Senneset (1973) method
00Q.0
U
01c
"i:0Q)m
:; 300,0.200"Z,OO..:Q)
.0 50E:JZ
.~ 201
I0 0.2 0.4 0.6 0.8 1.0
Friction. toncp'
Fig. 111-11 Values of Ng versus tan ~ (after Janbu& Senneset, I973)
An example of the use of this method is pre-sented in Figure 111-10. In practice, the inves-
tigator estimates the slope, Np' and intercept, a,of the average linear qc - p' profile. The rela-
tionship of Nq = Np + 1 with ~I, as presented inFigure 111-11, is then used to estimate ~'. Thismethod has the advantage of including overburdeneffects, and appears to provide reasonable resultsfor appropriate profile conditions.
As with the SPT, relative density, Dr' may beemployed as an intermediate parameter for estimat-ing ~' frompenetrometerdata. A correlationofqc' Dr' and overburden pressure for normally con-solidated, uncemented, primarily quartz, saturatedfine sands is presented in Figure 111-12, afterSchmertmann (1976)*. Use of this relationship
*The correlation applies when using the Fugro-
type electrical cone, 10 cm2, 60°, cylindricaltip, advanced continuously at 2 cm/sec.
13
300
200
100
gQ!L!.Q.!!.%
00 1.0 2.0 3.0Vertical Effective Stress- kgf /cm2
4.0
Fig. 111-12 Q-CPT bearing capacity to estimaterelative density in normally consoli-dated, silty fine to uniform mediumsands (after Schmertmann, 1976)
requires a correction for overconsolidated sands,which Schmertmann (1974a),based on the results ofchambertests,suggestedbe taken as:
qc
(Ko'
)()= 1 + 0.75 ()- 1qc NC Ko NC(111-4)
where:q = measured penetrometer tipc resistance
(qC)NC = penetrometer tip resistance forthe normally consolidated case
Ko' = in-situ lateral stress coefficient
(Kd~c = lateral stress coefficient for thenormally consolidated case
Before applying the relationship presented inFigure 111-12, the investigator must estimate Ka'/
(Ka')NC and then employ Equation 111-4 in order toconvert the measured tip resistance, qc' to the'equivalent"normally consolidated" tip resistance,
(qc)NC. If the overconsolidation ratio (OCR) isknown, then Ko'/(Ko')NC can be estimated usingKo'/(Ko')NC = (OCR)O.42 and (Ka)NC = 1 - sin~'.
Strength of Clays from the CPT
The shear strength of clays, su' is mostoften estimated from penetrometer resistance byemploying a relationship of the form:
qc - ps =-u Nc
(111-5)
where:
Su = undrained shear strength
p = overburden pressure
NC - Bearing Capacity Factor appro-priate for deep, circularfoun-dations
The bearing capacity factor, Nc' is, in fact,not a simple constant, but is affected by a numberof factors as summarized in Table 111-6, after
Schmertmann (1975). Values of Nc ranging from 5to 70 have been back-calculated by different in-
vestigators from measured values of qc and valuesof Su determined from other types of strength
tests. Some values of Nc appropriate for differ-ent clay types are presented in Table 111-7, afterBrand et ale (1974). The best approach for design
purposes is to experimentally determine Nc for agiven clay type, penetrometer apparatus, testingprocedure and reference suo If a friction-conetip is employed, the measured soil-steel friction,
fs' is suggested by Schmertmann (1975) as a lowerbound for suo
14
TABLE 111-6. SOME OF THE VARIABLES THAT INFLUENCE NcIN EQUATION (111-5) (Schmertmann, 1975)
D. The Vane Shear Test
Introduction. In contrast with methods which
derive strength parameters from intermediate var-iables, such as penetration resistance, the VaneShear Test, VST, attempts to measure undrainedshear strength directly. The test procedure isto advance a vane configuration to a desired soildepth and measure the applied torque as the vaneis rotated at a constant rate. Shearing resis-tance is considered to be mobilized on a cylin-drical failure surface of rotation, corresponding
to the top, bottom and sides of the vane assembly.
Three common methods for installing the vane
apparatus are illustrated in Figure 111-13. Asindicated in the figure, the vane may be install-ed at the bottom of a predrilled borehole, or
pushed into the ground by means of an extension
Approx. NcVariable factor Direction Notes
potential
1. Changing the test 2 to 3 Better sampling,
method for obtain- thinner vanes, use
ing reference Su of sUPMT alldecrease Nc ..
2. Clay stiffness 3 Increases with Vesic (1972)
ratio = G/suincreasing stiff-ness
-
3. Ratio increasing/ 3 Decreases with Ladanyi (1967)
decreasing modulus decreasing ratio(E+/E-) at peak Su
4. Effective friction, 2 to 3IncreQses with Janbu (1974)
tan CP' increasing cpr
5. K, or OCR 3 Increases with Janbu (1974)increasing Kor OCR
6. Shape of pene- 2 Clay adhesion on Example in
trometer tip mantle of mechani- Amar et al.,
cal tips increases (1975, Fig. 2)
Nc
1.5 Reduced diameter Schmertmann
above cone can (1972)
decrease Nc invery sensitiveclays
7. Rate of pene- 1.2 Increasing rate Viscous, no
tration increases Nc pore pressureeffects
8. Method of 1.2 Continuous (electrical tips)
penetration penetration decreases Nc compared toincremental (mechanical tips) because
of higher pore pressures.
15
TABLE III-7. CONE FACTORS DETE'RMINED FOR CLAYS (after Brand et a1., 1974)
I
C Clay PropertiesReference Clay one
I
Factor /2 ' , ,
w, % wL' % Ip' % su'ton m Sensl.tl.vl.ty
Thomas (1965) London Clay 18 20-30 80-85 55 5-29+ -
Ward et a1.
(1965) London Clay 15.5 22-26 60-71 36-43 21-52+ -
Meigh & Corbett ,(1969) Arabl.anGulf Soft Clay 16 30-47 38-62 20-35 0.5-4* 5
Ladanyi & Eden(1969) Leda Clay (Gloucester) 7.5 50-70 50 23 2.5* 30-50
Ladanyi & Eden(1969) I Leda Clay (Ottawa) 5.5 72-84 40 20 5.7* 10-35
I
Pham (1972)
I
I Soft Bangkok Clay (City) 16 60-70 70-80 40-50 1.3-2.9* 5-7
Anagnostopou1os(1974) PatrasClay 17 30 35 18 3-7+ 1.5-3
I
Brandet. a1. '
I
Soft Bangkok Clay 19 60-130 60-130 60-120 1.3-3.8* 5-7(1974) ,(Bangp1i)
jBrandet. a1. ! WeatheredBangkokClay 14 100-130 100-135 60-80 1.3-3.2* 6-8(1974) i (Bangp1i)
i
IGroun
tSurface
PossiblyDrillingFluid
ExtenSl
~on
Rod
6Ic-;f
!2Friction
.'l Rod-ClayVane
Uncased Cased
(0) Drillholemade by augering,washboringor rotarydrilling
Fig. III-13
tGrOUnd
1Surface
ExtensionPipe
E ,tens;on E xlens;on
£P:'::t;on RodShoe for
VaneWithdrawal
50c;T" . No Friction t Frict!On':1. Rod-Cloy Couphng
Vane Vane
(b) Equipment pushedinto the ground bythe extension pipe
(c) Vane pushedinto the groundby the extensionrod
Methods for installing shear vaneS
'3I~r-6.5.,em
~oc~
,
IOem
~)
(I) (2)
Some shear vane configurations in common use
(3) (4) (5)
Fig. 111-14
rod, with or without the.use of a protection shoe.A number of different vane configurations are incurrent use, as depicted in Figure 111-14, but thepreferred design, as specified in ASTM StandardD2573, is a four bladed vane with a height/diameterratio of 2.
Determination of Strength from the VST
Undrained shear strength may be calculated
from measured torque in the VST, provided thathorizontal and vertical shear strengths are as-
sumed equal, by employing the following equation:
suv2'r
'IT D3 (H/D + a/2)(III-6a)
where:
suv = the undrained shear strength, fromthe VST
T = maximum applied torque
H = vane height
D = vane diameter
a = factor which depends on the assumedshear distribution along the top
and bottom of the failure cylinder
= 2/3 if uniform shear is assumed
= 1/2 if triangular distribution isassumed (i.e., shear strength mobi-
lized is proportional to strain)
= 3/5 if parabolic distribution isassumed
For a uniform shear strength distribution, ,
and ASTM D2573 vane configuration,Equation1II-6a
reducesto: 6 Ts =- -uv 7 D3
(III-6b)
Under these conditions, shearing resistanceon the side (vertical) face of the failure cylin-der contributes some 85% of the measured.resist-ance to rotation, T. What is actually measured inthe VST is therefore a weighted average of the
shear strengths Sv on vertical and sh on horizon-tal planes. It is possible to determine both sh
and Sv separately by repeating a test using a vaneof a different shape or H/D ratio. This methodwas employed by Richardson et al. (1975) in orderto determine the relative values of sh and Sv inBangpli Clay. The results of their investigation
are reproduced in Figure 111-15. The ratio Sh/Svis seen to be fairly constant with depth, with anaverage value of 0.6.
16
rn~m1-13em-l
Measurements of suv using different vane con-figurations will be influenced by different pro~portions of sh and sv,as noted in Figure 111-14.It has been found that, in general, the ratio
sh/sv is less than unity (Duncan and Seed, 1966;and Lemasson, 1974). It therefore seems that anaccurately determined.value of suv may be used as
a conservative estimate of sv.
Accuracy of the VST
Until recently, the VST was consider~d to bea most reliable means of measuring undrainedshear strength in soft to medium clays, possess-ing the advantages of economy of use and reducedsoil disturbance, compared to sampling and lab-oratory testing. A number of cases have been en-countered, however, in which the use of suv leadsto unconservative results in undrained stability
analyses (Bjerrum, 1972; Pilot, 1972). Accord-ingly, a correction procedure was developed(Bjerrum, 1973) which attributes the discrepancyin field behavior mainly to strain rate effects.The true undrained strength is related to themeasured shear strength as follows:
s = ].l su (field) uv
(111-8)
The Bjerrum correction factor,].l , is correla-
.tedwith soil plasticity index, Ip' as shown inFigure III-16~ An inspection of Figure 111-16shows considerable scatter in the data obtained
subsequent to dev~lopment of the correlation. Thiscorrection<procedure is subject to the additional
uncertainty inherent in the determination of Ipand suv' and in the formulation of stability ana-lysis assumptions. Use of the Bjerrum correctionfactor may 'actually yield occasional unconserva-
tive results as noted by LaRochelle et al. (1974)and Ladd (1973). Schmertmann (1975) notes also
that as Ip is determined using disturbed clay itcannot account for differences among clays havingdiffering undisturbed structures.
Factors \ihich Affect the VST
It is now recognized that the VST is subjectto the uncertain effects of a large number of
influencing factors:
I} Disturbance. It is evident that the vanecannot be installed in the ground without causingsome degree of disturbance in the soil around it.If predrilling is employed (see Figure III-13a), azone of a certain width around the bottom of theborehole must be considered to be disturbed. In
practice, it is commonly assumed that the disturbedzone does not extend more than 60 cm from the bot-
0
2
~ 4~(I.)EI 6
.s::a.(I.)
0 8
10
17
Moisture Content -% Sensitivity0 50 100 150 2 4 6
12
Properties of the Bangpli clay at the vane boring site
00sh/s v
05 1.0
~(I.)QjEI
.s::a.(I.)0
12
Ratios of vane strengthson horizontal and verticalplanes
Fig. III-IS Ratio of strengths on horizontal (sh) to vertical (sv)planes for Bangpli clay (from Richardson et al., 1975)
0.4a
Fig. 111-16
Su(Field)= fLx Su(Vane)
Reference
0 .* 8jerrum (1972)
/:1A* Milligan(972)0 Ladd and Foott(1974)
V Flaateand Preber(r374)<> LaRochelle et 01.(1974)
* Layered and ,vorved clays
I
A
0
./:1
t-O-t
20 40 60 80Plasticity Index, P.I.-%
100 120
Field vane correction factor vs plasticity indexderived from embankment failures (from Ladd, 1975)
---- v .-J". ':1 - 1-..Weathered (., - l
Clay'-'
§- "\
-- ., \...... -- ?-Ij..... ....
'-r . ...-Soft
..... .... "-..
Clay r'-" ...:::"-......:::' J'"1"'"
-I- ,;:;'-
I'-'
I I
-IU-.J
I_...
I
0 Plastic Limit0 Liquid Limit. Natural MoistureContent. .
I1\.
\!
)
(
I
1.4V
1.2
:::L.: 1.0E(.)0IL.
c0
. 0.8
0u
0.6
tom of the borehole. It has been noted (Flaate,
1966), that this assumption can be questioned ifthe borehole is wide and the drilling procedureunfavorable.
If the vane is installed by means of a pro-tective shoe forced into the ground without pre-drilling (see Figure III-13b), a plastic zone willbe formed around the vane head with a varying de-gree of disturbance. Studies by Cadling and Oden-stad (1950) and Andresen and Bjerrum (1965) indi-cate no measurable disturbance caused by this pro-cedure when the vane test is made at least 50 cm
from the protective shoe.
When the vane is installed without any pre-drilling or protective shoe (see Figure III-13c),the disturbance is due'only to the vane rod andthe vane itself. This component of disturbance isalso present if predrilling or a proteptive shoeis employed. In this regard, it is considered ad-vantageous to minimize the "area ratio" of thevane. The area ratio is defined as the cross sec-tional area of the vane cross and stem as a per-centage of the cross sectional area of the failurecylinder. It is recommended that the 'arearationot be higher than about 15% (Flaate, 1966).Laboratory tests performed by Vey (1955) show thatcohesive soils can stick to,the vane surface, and,
in effect, increase the area ratio of the appara-tus. It has been suggested (Flaate, 1966), thatthis effect will probably depend upon the sticki-ness of the soil and be of less importance insensitive clays.
It is recognized that the use of a damagedvane apparatus may greatly increase the magnitudeof soil disturbance.
2} Mode of Failure. The stress distribution andstress strain behavior in the VST are little known.
The soil is generally assumed to fail along thesides and ends of a circumscribed cylinder. It isalso assumed that the shear strength is fully mo-bilized all along the surface at the ... same time,i.e., no progressive failure takes place. In theidealized case, the application of torque willproduce large stress concentrations at the end ofthe vane blades (Flaate, 1966). The actual stressdistribution will depend upon the degree of dis-.turbancearound the vane as well as the stress-
strain properties of the undisturbed soil. Thepotential for progressive failure will depend uponthe type of soil, its sensitivity and its stress-strain characteristics.
A laboratory program was carried out byLeBlanc (1975) in order to examine the actualfailure pattern and stress-strain behavior duringthe VST. It was observed that the peak strengthwas obtained at very low angular deformations.Only for rotations in excess of 45° was a cylin-drical failure surface observed. On the basis ofthese results, Roy (1975) concluded that it isquestionable to interpret VST results in the clas-sical manner which assumes a cylindrical shearfailure surface at peak strength.
The failure mode in the VST may not corres-pond to the failure mode in a prototype situation.The effect of preferred failure plane orientations
18
may govern full scale behavior, yet remain unde-tected in the VST. Because of the uncertainties
involved in describing the failure mode in a givenVST, it is desirable that VST results be usedin conjunction with other strength tests.
3} Dimensions of Failure Cylinder. The actualdiameter of the failure cylinder in a soft clay,and hence the moment arm of mobilized shear
strength, has been observed to be some 5% largerthan the diameter of the vane blades (Arman et al.,
1975). An uncorrected discrepancy of this magni-tude will result in an over-estimate of 16% in
calculated shear strength. At present, there isnot enough data available to enable a quantitativecorrection for this effect in different soil types.
E. The Pressuremeter Test (PMT)
A borehole pressuremeter, of the type devel-oped by L.Menard, is depicted in Figure III-17.These devices are widely used in Europe and arereceiving increasing acceptance in the U.S. The
test proceeds once the apparatus is lowered to adesired borehole depth. Increments of hydraulicpressure are applied to the center cell, and the.resulting deformation of the borehole wall, atset time intervals, is d~termined from fluid vol-ume change measurements of the cell chamber. Thetwo pressurized guard cells serve to stabilize thedevice within the borehole and to insure essen-
tially axial plane strain conditions at mid-height
CO2Gas
GasTubesto
Guard CellsVolume ChangeMeasured by Changein Water Level
Pressure-Volumeter
I Protective Sheath
Guard Cell Under Gas Pressure
Measuring Cell. Filled withWater Under Gas Pressure
Guard Cell Under Gas Pressure
Fig. III-17 Classic Menard pressuremeter
Pressure
Time
, VolumetricExpansion
Fig. 111-18 Pressure-volume change history
during a pressuremeter test
Pressure Gage
Probe
Chopping Tool
Cutting Edge
Fig. III-19a Sketch of IIAutoforeuse" probe
(from Baguelin et al., 1972)
19
of the apparatus. The relationships between pres-sure increments, volumetric expansion, and timeare presented in Figure 111-18 (Menard, 1975).
Important recent improvements of the PMT arethe "Autoforeuse" and the "Camkometer" self boring
pressuremeters, developed respectively by Baguelinet al. (1972) and Wroth and Hughes (1973). A selfboring pressuremeter is schematically presented inFigure III-19a, and the Camkometer is depicted inFigure III-19b.
The PMT is seen to possess at least three
distinct advantages:
1) The test models the axisymmetric expansion ofan infinite cylindrical cavity -- a problem withwell developed elastic and elasto-plastic solu-tions which are suited for application to soilmechanics.
2) The conduct of the test permits an estimateof the in-situ lateral stresses with no limits onK.0
3) The test results provide not only strengthdata, but stress-strain soil properties applicableto the direction perpendicular to the axis of theborehole cavity.
RushingWater
Pressure LineforExpanding RubberMembrane
Rubber Membrane
Spring Fine Thread
Diameter of Membranebefore Expansion is thesame as the CuttingHead
ClompJ
~'@
Section X-X
Cutting~
Head
Fig. III-19b Main details of Cambridge in-situinstrument showing stress gages
(fromWroth and Hughes,1972)
~E~ 600~cuC\c:0.cUcu 400E::J
gcu.c0
It 200
0
Fig. III-2O
Po Pf PIPressure - kPa
Theoretical pressuremeter curve
tNVo
Plastic
Elastic
Recompression of Disturbed Soil
Pf P
Typical result of a pressuremetertest in soft clay
Obtaining Strength and Stress-Strain Data from thePMT
PMT results are usually presented as a plotof chamber volumevs. applied pressure. Appliedpressure should be corrected to account for mem-brane stiffness and the hydrostatic head of thefluid in the connecting tubes. An idealized pres-sure-volume curve is presented in Figure III-2O.The results in Fig. 111-21 for soft clay reflectthree phases of the test: (1) recompression ofdisturbed soil, (2) linear elastic compression ofthe soil and (3) plastic deformation of the sur-rounding soil mass. The in-situ lateral pressure,
Po' is estimated as the pressure corresponding tothe kink in the curve between the recompressionand elastic compression phases. By definition, the
the limit pressure, PI' is the abscissa value ofthe vertical asymptote to the pressuremeter curve.The limit pressure may be determined directly fromthe curve, but more conventionally it is taken asthe pressure corresponding to a volume increase ~V
equal to the initial volume of the borehole Vi(i.e.,P=Pl @ ~V=Vi)as volume increasesfor pres-sure increments at this point are normally rela-tively large.
20
Three independent theories for extracting acomplete stress-strain curve from the results ofpressuremeter tests in incompressible, nondilatingsoil were simultaneously presented by Ladanyi(1972),Palmer (1972) and Baguelin et al. (1972).Each investigator assumed ~ = 0, Poisson's ratio~ = 0.50 and showed the validity of the followingequations:
d (p - P )
T = E 0ps 0 d E0
(1 + Eo) (1 + Eo/2) (111-9)
For small strains:
d (p - P )
T ~ E 0ps 0 d E0
~VE =1 C(V. + ~V)
J.
where:E = radial deformation of the probe,
0 ~a/a, where a is the cell radius;computed from ~V
T = equivalent plane-strain shear stressps
El = equivalent axial compressive strain
p, p = corrected applied pressure and in-0 situ lateral pressure, respectively
V., ~V = initial and differential chamberJ. volumes, respectively
A graphical procedure developed by Amar eta~. (1975), may be employed to deduce the com-plete stress-strain curve.
Shear strength is frequency determined fromthe Menard PMT through the semi-empirical rela-tionship:
su
PI - PoN
(111-10)
where PI and Po are as indicated in Figure 111-20,and N is a correlation factor generally taken to
equal 5.5.
Until recently, <j)'was determined from PMT
resultsthroughIIin-houseII correlations of ~' with
th~ pseudo-elastic modUlus EpMT (see section IV E)and the limit pressure Pl' The reliability ofsuch correlations may be questionable when theyare generalized to different soil types.
Cavity expansion theo;-y has been applied to
the determination of ~'f:r6m the PMT through thework of Gibson and Anderson (1961), Vesic (1972)
and Ladanyi (1963). Using,Vesic's theory, close
agreement between ~'PMT and~' from triaxial testshas been shown by winter and Rodrigues (1975),but investigations by Laier (1973) and AI-Awkati(cited byJ. Schmertmann, 1975), showed a poorabilityof the abovetheoriesto predict~' fromthe results of pressurementei tests in largetriaxial chambers.
The application of these theories to the de-termination of ~' is hindered by four factors:
1) Pressuremeters with ordinary length to dia-meter ratios require major correction factors toconvert real PMT limit pressures to the limitpressure corresponding to infinite pressuremeterlength (Laier et al., 1975).
2) At or near the limit pressure, the bulge inthe pressuremeter cell seriously violates the planestrain assumption (Schmertmann, 1975).
3) Any evaluation of ~' requires a knowledge ofin-situ stress conditions which may only be crude-
ly determined in the conventional PMT (Laier, 1975;Hartmann and Schmertmann, 1975).
4) The initial soil modulus used in these theo-ries may be underestimated when the conventionalpressuremeter test is employed (Schmertmann, 1975).
A new method for determining ~' from pres-suremeter tests in sand has been developed byAl-Awkati and reported by Schmertmann (1975). Themethod employs only the low strain portion of thePMT curve and thus avoids the uncertainty associ-
ated with evaluating Pl'
Al-Awkati's method proceeds as follows:
1) Plot PMT data as In (Eo = ~a/a) vs. In (cor-rected pressure). A sample curve is presented in
Figure 111-22.
2) Measure e as defined in Figure 111-22.
3) Enter chart in Figure 111-22 with tan e andmove vertically to measured or estimated ~d where~d representsthatpart of ~' not due to dilatan-cy.
4) Read predicted triaxial ~' on horizontalscale.
The predicted triaxial ~' applies to the mean
normal stress, ad ' where:
pa ' = ~'0 1 + sin~'
(III-II)
and
Pf = chamber pressure when sand entersfailure state.
One may determine ~o' from a triaxial test inwhich the volumetric strain rate becomes zero at
peak strength or by correcting for volumetric workin a test where this rate "differs from zero. Al-
ternately, Al-Awkati presents a correlation be-
tween--<p-~-and the volumetric strain between 0 and100% relative density for quartz sands which isreproduced in Figure 111-23.
This method has been applied to correspond-
ing PMT and laboratory triaxial test results,
yielding a close agreement between predicted andmeasured values of ~', as summarized in Figure111-24 after Al-Awkati.
21
45
(0)
40
CI)CIICII
C.CII",I-9- 35"0';(0
~
2'5
0.2 0.3
Fig. 111-22 Al-Awkati's theoretical correlationbetween PMT data and triaxial maximum
~', using ~o' (from Schmertmann, 1975)
Factors Which Affect the PMT
The standard Menard type PMT is highly sensi-tive to variations in test procedure.
A method of boring, and inserting the appara-tus must be chosen which minimizes soil disturbance.
Many methods will produce approximately the same
40
~0,c: 0 30'(50-=,(/)0
E '~ 20,=00>
Quartz Sand
---~~-- -(""'~11 --I~
~----~~--
1025 30 40
Fig. 111-23
35Triaxia,l 4>~-degrees
Correlation for estimating ~o' using
volume strain from relative density
tests (suggested by Al-Awkati inSchmertmann, 1975)
45II)Q,)
~0'1Q,)1:1
...,
~ 400.;(0.i:I-1:1
~ 35::III)0Q,)~
35 40 45Predi.cted Triaxial cp'-degrees
Fig. 111-24 A1-Awkati's comparisons betweenmeasured and predicted ~'(Schmertmann, 1975)
'limit pressure, but an accurate evaluation of soildeformation requires the best possible boreholepreparation, especially in sensitive clay deposits.
Duringa standard PMT, the variation of porepressure is not measured. As a result, the rateof expansion to assure essentially drained or un-drained test conditions is not known. A common
practice is , to' set an arbitrary time interval (1 -2 minutes) .' for each pressureincrement. Since themagnitude of soil volume change will increase aspressure increases in successive increments, a
fixed time interval proceduremay result in nearlydrained conditions for the initial pressure incre-ments, but essentially undrained conditions forthe final pressure increments. It is likely thatno part of the test will,be fully drained or un-drained.
This uncertainty in pore pressure behaviormay be resolved by measuring pore pressure with asuitable transducer affixed to the face of the de-
vice. At present, very few pressuremeters are soequipped. Versions of the Cankometer, depicted in
Figure 1II-19b, contain such a pore pressuretransducer. An alternateapproach,tothe porepressure problem is to experiment with differentrates of cell expansion to determine at what rates(fast = undrained, slow = drained) the test results
become unaffected. Such an approach is generally
impractical' in engineering applications, because
of the large number of tests required.
F. The Iowa Borehole Shear Test
The Iowa Borehole Shear Test is performed bylowering a shear head, consisting of two opposinghorizontally grooved shear plates to an uncasedsectionof ,a borehole(Figures 111-25. and 111-26).At the required test positionthe two shearplatesare expanded until skated in the ,boreholewalls ata preselectedpressure. Some time is allowedfor
22
Hydrauli cPressure~
Upward Pull '
Creates ShearFarce
UncaseclBorehole
PullingYoke
ExpandableShear Head
Fig. 111-25 Iowa borehole shear test apparatus
consolidation to occur. When consolidation is com-
plete the shear head is either pulled upwards, orpusheddownwardsat a steadyrate of 2 mm/min.
The required forces for shearing are measured,and the shearing stress plotted against the normal
press~re. At this point the shear plates may becontracted, the shear head lowered to its originalposition, rotated through 90° and the test re-peated.
The shear head is then re~urned to the ori-
ginal position, another seating pressure selectedand the test repeated.
By performing a number of.tests at differentseating pressures, a Mohr-Coulomb failure line andsubsequently c and ~may be determined. In prac-
tice the number of tests that can be reliably re-peated at the same position are influenced by theinherent friction of the soil. (In clays, thenumber may be as low as 2, whereas in sand thelimit may be determined by how far the shear heads
can be'expanded.) c and ~values determined bythis method show good corr,elationwith those deter-mined in triaxial.and direct shear tests.
Major Limitations of the Method
i) Drainage conditions are not known, and, there-fore, it is not possible to determine whether the
strength parameters are those for drained, partial-ly drained or undrained conditions.
Pistons
Knife Edge
. ,}
Sheor Planes-2 .' . .. .:-LVOT
- Contact Plate
...' Borehole
Fig. III-26. Schematic of borehole shear device
(A) circumferential shearing resis-tance must be minimized for uniform
distribution for contact pressure nl'and (B) knife edges minimize andresistance
ii) The shear strength of unconsolidated portionsof the borehole wall, not under pressure from theshear head plates but which extend along the endsof the plates, may add to the resisting shear forcebut is not accounted for in the calculations.
Advantages of Borehole Shear Test
i) The Mohr-Coulomb strength parameters may bedetermined in less than one hour for some soils.
ii) Erroneous shear values resulting from poorseating or remolding are usually apparent and maybe systematically eliminated.
iii) Tests may'be located at selected strata, asthe tests are only performed after a hole is boredand logged.
G. Conclusions
Soil strength may be expressed in a number ofways, and the particular strength parameters to
use in design will depend on the nature of a spe-cific problem. In-situ strength testing methodsand programs should be selected on this basis.
23
The Standard Penetration Test and DynamicCone Penetration Test suffer from a lack of theor-etical understanding and controlled use, as wellas uncertainties in interpretation. Use of thesemethods will continue, however, in the absence ofeconomical replacements and because of the greatnumber of available empirical correlations, espe-cially for the SPT. In many cases SPT data arethe only information available.
The quasi-static cone penetration test pos-sesses many practical advantages and prospects forthe development of a reasonable theoretical analy-sis of the test in the future are good. At presen~best use of the method for strength evaluationrequires the establishment of a cone factor for
a'particular soil deposit by means of a fewseparate strength tests by another method.
In spite of the several difficulties citedrelative to the interpretation of vane shear testresults, it can be anticipated that its use willcontinue for evaluation of the in-situ strength ofsoft clays.
The pressuremeter test has great potentialfor use in measuring soil strength, as well asstress-strain behavior. The potential accuracy ofthe self-boring PMT is promising, but the highcosts of reliable results indicate that this ap-proach may be most efficient if it is used in con-junction with simpler strength index measurements.
The borehole shear test is expected to increasein popularity as a result of its relative simplicityand the direct design value of the raw test data.Its use is limited to testing those soils which aresuitable for stage testing. The applicability ofstage testing to different soil types is a subjectwhich needs to be studied further.
The different methods are not equally appli-cable in all soil types. The SPT and dynamic conetests are most suitable for strength estimates insands and stiffer soils. The static cone and vane
shear devices cannot penetrate very stiff deposits.Pressuremeter testing, while suitable in principlefor most soil types except gravels and rockfills,is not suitable in deposits containing interspersedcobbles or rock fragments. The borehole shear de-vice has had its greatest applicability thus far instiffer deposits. The vane shear test is bestsuited for soft clays; whereas greatest applicabi-lity of the static cone has been to sands, althoughit appears well-suited to soft and medium clays aswell.
IV. IN-SITU STRESSES AND DEFORMATION CHARAC-
TERISTICS
A. Introduction
The response of soil to the application ofload stresses depends greatly upon the initialstate of stress and the stress path experienced bythe soil. Deformations and local yielding are con-sequently affected to a marked degree by the geo-logic stress history, the magnitude of the in-situvertical and lateral effective stresses and the
nature of the stress changes imposed during con-struction. The importance of stress path consid-erations have been giscussed by Lambe and Whitman(1969) and Ladd and Foott (1974), among others.
Most limit analyses, as performed by geotech-nical engineers, are independent of soil deforma-
tion characterIstics and the magnitude of initiallateral stresses. On the other hand, the predic-tion of settlements depends upon the accurate de-termination of deformation parameters, which inturn depend upon the in-situ effective stresslevels. Historically, deformation parameters havebeen evaluated from the results of laboratory ex-
periments on nominally undisturbed samples andthen substituted in a simple elastic analysis toyield values of stresses and displacement for usein analysis and design.
Some inadequacies of this approach are evi-dent. Significant soil disturbance is inevitableas a result of the sampling and laboratory testingprocesses. A consequence of this disturbance isthat predictions based-upon laboratory test re-sults may grossly overestimate the movements,resulting in overconservative designs (Marsland,1973). Proper simulation of field conditions inlaboratory tests requires knowledge of in-situstress conditions, quantities-that can best bedetermined by in-situ tests. In fact, Wroth (1975)suggests that correct evaluation of the in-situlateral stress may not be possible from the resultsof laboratory tests.
B. The Estimation of In-Situ Stresses
The simplest concept of an :in-situ stress regime
is based upon several simplifying assumptions:
1) The ground surface is reasonably level, so that
the vertical stress is a principal stress which
may be calculated from the weight of overlyingmaterial.
2) Horizontal stresses are equal in alldirections.
3) The pore pressure is known.
4) The coefficient of lateral earth pressure,
Ko' is known.
In reality, the in-situ stress state may de-viate significantly from these assumed conditionsas a result of tectonic or geologic processes.Even if these assumptions hold true for a particu-lar deposit, uncertainty will arise in the deter-
mination of the vertical stress, avo' and porepressure, uo'
The calculation of avo may be influenced by
24
several sources of uncertainty, as considered byMassarsch et al. (1975b) and Tavenas et al. (1975):
1) Determination of total unit weight, Yt, issubject to laboratory measurement errors.
2) The presence of a dry, fissured crust will
produce horizontal variations in avo'
3) The assignment of an average unit weight toan interval of depth may not reflect the true ave-rage unit weight of that depth interval.
4) The determination of depthnesses is subject to error. Ininvestigations, borehole depthsonly to the nearest 0.1 meter.
and layer thick-conventional soil
are probably known
The measurement of pore water pressure isprobably only accurate to within IO.l m of waterhead (Il.O kPa), even with the best piezometerscurrently available. Sources of error in the mea-surement of Uo include:
1) Temperature effects
2) Fluctuations in atmospheric pressure
3) Calibration errors
4) Zero shift in electrical transducers
5} Uncertain depth of transducer placement
6) Seasonal fluctuations of the groundwater table
7) Failure to achieve complete equilibrium be-tween the groundwater and the measuring system.
C. In-Situ Lateral Stresses
The nature of in-situ lateral stresses may beunderstood by examining the results of laboratoryoedometer tests which model the one-dimensional
consolidation experienced by an idealized soil de-posit as the overburden pressure.is varied (Wroth,1975). The observed stress paths corresponding toone-dimensional laboratory consolidation and subse-quent unloading take the form presented in FigureIV-I, where:
q = (al - a3)/2(IV-I)
p' = (aI' + a2' + a3')/3(IV-2)
T) = q/p'and
aI' ,a2', a3' = major, intermediate and minorprincipal effective stresses,
respectively
For the usual case where the in-situ horizon-tal stress is less than the overburden stress,
a' = a'v 1
a~ = a~ = a~
In Figure IV-I, the vector AB represents thestress path of a soil under a one-dimensional load-ing which models the increasing overburden pressureand normal consolidation during sedimentation. A
constant ratio, Ke, is observed between ah'and a;.Therefore, for a given soil, under these loading
a:'v. B
a;'h(0)
YJ
A B
0
Inp'
(c)
q
B
(b)
e
--E~\ I
~
(d)
Fig. IV-l Typical effective stress paths for soilconsolidated one-dimensionally (fromWroth, 1975)
conditions:
K =C5h'/C5' (a constant less than 1.0)o. v(IV-4a)
For normally consolidated soils, a theoreticalrelationship may be derived (Jaky, 1944):
KO = (1 + 2/3 sin cp')
1 - sin <P'1 + sin cpt
(IV-4b)
where <P' = soil friction angle, in terms ofeffective stresses.
A convenient approximation for this expres-sion is often used that has been found to be ac-
curate enough for most engineering applications(Wroth, 1972):
Ko ~ 1 - sin cpt(IV-4c)
The derivation of these expressions is pre-sented in Huck et ale (1974).
If a normally consolidated soil, at stressstate "B" in Figure IV-I, is subjected to one-dimensional unloading, the stress path will followa vector similar to BC. At point "C", the horizon-
tal and vertical stresses are equal, and Ko isequal to 1.0. For many soils (Ladd, 1975), thisstate corresponds to an overconsolidation ratio,OCR, ranging from 4 to 5.
If the stress path BC is assumed to be linear,
the theory of elasticity may be applied to show
that, for any point on the stress path BC:
~C5 ' =~ ~C5 'h 1-11' v
(IV-5)
25
and, ,K' = (OCR) (K' ) - ~ (OCR - 1)0 nc 1 - ll'
(IV-6)
where:K' = coefficient of lateral earth pres-nc sure for a normally consolidated
state
ll' = Poisson's ratio, in terms of effec-tive stresses
pi
By applying Equation IV-5 to the results oflaboratory tests on 2 sands and 6 remolded clays,representative values of ll'have been obtained and
correlated with Plasticity Index, Ip' by Wroth(1975). The test data appear in Table IV-I, and
the resulting plot of ll'vs Ip is reproduced inFigure IV-2. It is emphasized that this correla-tion only applies to lightly overconsolidatedsoils. Heavily overconsolidated soils may exhibitmuch lower values of ll' as reported by Wroth (1975).
pi ':lOA.~
~
0CL
D-° 20
Fig. IV-2
Plasticity Index-%Valuesof Poisson'sratio for lightlyoverconsolidatedsoils (fromWroth,1975)
An important consequence of these findings is
that for a lightly overconsolidated clay, K6 may
be readily estimated from I~, cpt,and OCR. Avalue of cpt is substituted ln Equation IV-4 to
calculate K~c' and ll'is obtained from Ip usingthe plot in Figure IV-2. The best estimate ofthe OCR is made, based upon geologic informationand/or laboratory test results. Equation IV-6 is
then employed to determine Kb.
Heavily Overconsolidated Soil
If a soil at stress state "c" in Figure IV-lis subjected to additional one-dimensional unload-ing, the stress path takes on a curved shape simi-lar to the arc segment CD. Along the stress path
CD, the value of Ko'is greater than unity, andsteadily increases with increasing OCR. The max-imum value of Ko'occurs at point "D", where thesoil is in a state of'passive failure.
Along the "heavily overconsolidated" portions
of the stress path between C and D, the horizontal
stress, C5h ' is greater than the vertical stress,
C5v', and therefore becomes the major principal
stress, C5l' For clarity in Figure IV-I, C5h is
still identified as (J3 in this range, however.
26
TABLE IV-I. Values of Soil Parameters Plotted in Figure IV-2
Wroth (1975)
It has been shown by Wroth (1972) that thestress path segment BCD in Figure IV-l may betransformed into a log-linear plot of the stressratio, n, and the natural logarithm of the stresspoint, In (p'). The results of Brooker's (1964)experiments on Bearpaw shale have been presentedin this manner, and are reproduced in Figure IV-3,after Wroth (1975).
7]
0.4 B
0
P'/Pe
-0.4
Fig. IV-3 One-dimensional unloading of Bearpawshale (from Wroth, 1975)
The unloading stress path, BCD, may thereforebe represented by the equation:
m (n - nB) = In (p' /p~ )(IV - 7 )
where:
p ~, nB = stress point, stress ratio at"B" in Figure IV-l
m = inverse of the gradient of BCD
in Figure IV-l
Equation IV-7
l~ 3 (l-K' )
ncm 1+2K'
nc
may be rewritten as:
- 3 (l-K~)
]
-
[
OCR (1+2KilC)
]
-1+2K' - ln 1+2K' (IV8)a a
The values of the OCR corresponding to selec-
ted values of K~ may thus be calculated for a
given soil, if the parameters m and ~c are known.The parameter m has been obtained for the soilslisted in Table IV-I, and correlated with I aspresented in Figure IV-4, after Wroth (19751. The
valueof ~c may be calculatedfrom ~' usingEquation IV-4.
The K~ - OCR relationship predicted in thismanner is compared in Figure IV-5 with the values
of Kd and OCR measuredduringone-dimensionalun-
m
3
2
00 20 40
PlasticityVariation of the rebound gradient, m,
with plasticity (from Wroth, 1975)
Fig. IV-4
2
Ka
01 2 4 8 16 32
Overconsol idotionRoti 0
Fig. IV-5 Comparison of computed and experimentaldat~ for one-dimensional unloading ofBoston blue clay (from Wroth, 1975)
No. I SoilI %
]11 AuthorP
m
1 Pennsylvania sand- 0.281 1.54 Hendron
2 Wabash river sand - 0.267 1.60 Hendron
3 Chicago clay 10 0.254 1.36 Brooker
4 Goose Lake flour 16 0.282 1.60 Brooker
5 Weald Clay 20 0.287 1.58 Brooker
6 Kaolin 32 0.325 1.81 Nadarajah
7 London clay 38 0.346 2.26 Brooker
8 Bearpaw shale 78 0.371 2.92 Brooker
loading of Boston blue clay, after Wroth (1975).The predicted and observed values are seen to be inclose agreement. The method described previously,applicable to lightly overconsolidated soils, is
also seen to produce~reliable predictions of ~ inBoston blue clay, for low values of OCR.
Reconsolidated Soil
If the soil at stress state "D" in Figure IV-lis reloaded one-dimensionally the effective stresswill follow the approximately linear segment DE.The value of Kq' will decrease rapidly, as a large
increase in aJ is accompanied by little change in
ari .
Stress states liEII and "C" correspond to the
same vertical stress, aJ ' and preconsolidationpressure, (a~ )B, and hence possess identical over-consolidation ratios. The values of Kd will dif-fer, however, because stress state "C" is located
on a path of primary unloading, while stress state"E" is located on a reloading path.
Therefore, the current stress state is un-
predictable if a soil deposit has been subjectedto more than one cycle of loading (deposition) andunloading (erosion).
Distribution of Kd in the Ground
The distribution of Ke' in a natural soil de-
posit may be the result 6f a complex sequence ofstress variations caused by deposition, erosion,tectonic movement, water table fluctuations andprevious construction history. The effects ofsecondary compression, vegetation, capillary pres-sures and heterogeneous soil distributions willserve to complicate this situation even further.
Knc0
z(0) (h)
Fig. IV-6 Typical vertical profile of Kd. for an
overconsolidated deposit
An idealized soil deposit in the normallyconsolidated state, with constant unit weight andhorizontal ground surface, will exhibit a uniform
distribution of KQ with depth, as sketched inFigure IV-6(a). If a soil layer of thickness, zo'were to be removed through erosion, the value ofthe overconsolidation ratio at any depth, z, wouldbecome equal to (z + zo)/z. If the soil is onlylightly overconsolidated, the methods presented
previously could be employed to show that Kq'should aecrease with depth as sketched in FigureIV-6(b). At a point near the ground surface,where the overconsolidation ratio is sufficiently
27
high, the value of ~ will be limited by a passivefailure condition. An equivalent distribution of~ would also result from the effective stress in-crease and decrease caused by a fall and subsequentrise of the water table.
q
CurrentYield
Envelope
pi(0)
e
pi(b)
Fig. IV-7 Change of state caused by delayedconsolidation
Ko Delayed, or secondary compression, as des-cribed by Bjerrum (1967), is a common cause ofapparent overconsolidation. If a soil element at
stress state "G" and void ratio eG in Figure IV-7undergoes secondary consolidation, the void ratio
will decrease to a value, eH' while the stressstate, H, remains unchanged from pointG. At thisstate, the soil will behave as if it were lightlyoverconsolidated. If additional vertical stresses
are imposed, the void ratio will decrease along
the recompression curve "eB" - eln, Figure IV-7.The vertical stress at state "1" would be inter-
preted as the preconsolidation pressure, and avalue of Kq'would be estimated accordingly. Infact, Ka'is that of a normally consolidated soil,since the effective stresses have not changedduring secondary consolidation. This phenomenonwill produce an effect of apparent OCR which isindependent of depth.
Laboratory Measurement of ~
The most widely used assumptions in the lab-oratory evaluation of in-situ lateral stresses areof questionable validity (Wroth, 1975). Even themost advanced techniques, such as those developedby Poulos and Davis (1972) or Tavenas et all (1975),may be expected to produce accurate results onlywhen applied to normally consolidated soils.
FlowmeterOutflow
_From 2-To I
(a) Total earth pressure
cell and read-out unit
28
1ft
Cell Pressure
(b) Schematic diagram of the earth
pressure cell (Glotzl) method
Fig. IV~8 Measurement of the in-situ lateral pressure using ~lofiU,cells
A thorough consideration of the factors in-fluencing,the determination of Ko has promptedWroth (1975) to conclude that "...there is no sat-
isfactory method of determining Ko for a naturalsoil by laboratory testing. II
D. In-Situ Measurement of Lateral Stress
The measurement of .in-situstresses is clear-
ly an area which holds great potential for thesuccessful application of in-situ testing methods.
The direct measurement of stresses in earth
fills may be achieved through the installation ofpressure cells during construction. Accurate re-suIts are difficult to obtain and verify, but thetechniques of installation and measurement arefairly well developed.
Strictly speaking, it is not possible to di-rectly measure stresses in undisturbed soil, sincethe insertion of a measuring device creates dis-turbance and modifies the local state of stress.
In practice, in-situ measurement may be very suc-cessful if soil disturpance can be minimized,andadequate time allotted for the re-establishment ofequilibrium stress conditions. A number of imagi-native techniques have been developed which attemptto realize these goals. These techniques are onlysummarized in the following paragraphs, and thereader is referred to the references for more de-tails.
Stress Measurement in Soft Clays
Pressure Cells: Total pressure cells, whichare pushed into undisturbed soil or placed in un-cased boreholes, may provide reliable stress infor-mation if disturbance is kept to a minimum and ade-
quate time is,allowed for the dissipation of porepressures. The Glatzl cell, depicted in FigureIV-S, has been found to be suitable, and is dis-cussed in detail by Tavenas et ale (1975) andMassarsch et ale (1975b).
Hydraulic Fracturing
The hydraulic fractu~ing technique is fre-
quently used.to determine stresses in rock strata,and is experiencing increasing use in soil studies.A schematic of the operation is presented in ~ig-ure "IV-9. Increasing wa,terpre~sure is applied toan isolated section of a borehole. Soil fracture,
In principle along the minor principal plane, isaccompanied by a marked increase in water in-flowwhich is monitored at the surface. The applied
pressure is gradually reduced, and when the waterpressure is equal to the stress which is normalto the cracks, the,cracks close and water flowdrops significantly. The "closure pressure IIis
MercuryManometer
Screw Cant rols
PlasticTubing
Piezometer
Fig. IV-9 Schematic diagram of the hydraulicfracture method
recorded and is an estimate of ah'
In theory; hydraulic fracture can measure ahonly in soils in which Ko is less than unity. Inpractice, the closure pressure often represents aweighted average of ah and av, since the preferredcrack orientation may be along horizontal stratifi-cations. The closure pressure is not always clear-ly defined, and is subject to the uncertain effectsof soil strength and deformation characteristics.
The theory of the fracture mechanism has beenpresented by Bjerrum et ale (1972). The detailsof the testing procedure are considered by Tavenaset ale (1975) and Massarsch et ale (1975Pl, who pre-
fer the use of total pressure cells to the hydrau-lic fracturing method.
Pressuremeter Tests
The idealized results of a pressuremetertest in soft clay are presented in Figure III-21and the details of the pressure-volume curve arediscussed in Section III E. The pressure at pointB (in Figure III-21) is generally assumed to
represent ah' but in practice, point B is oftenpoorly defined: Borehole advancement producesdisturbance and stress relief in the adjacent soil,and small details of test technique have beenfound to introduce large uncertainties in testresults. Tavenas et ale (1975) conclude that the
conventional pressuremeter test is unsuitable fordetermining lateral stresses in soft clay becauseof this.
The s~lf boring pressuremeter, depicted inFigure 111-19, minimizes soil disturbance by lim-iting lateral soil movements to small values. Eventhis small disturbance produces excess pore pres-sures which must be allowed to dissipate beforetesting. The equilibrium values of lateralstresses are not greatly affected, however, oncethis dissipation has occurred.
Measurement of Ko in Stiff Soils
The measurement of Ko in stiff clays, sands,and soft rocks poses particular problems. Pushed-in-place total pressure cells generally cannot beused because of the high resistance of these soilsto penetration. Hydraulic fracturing is effective
only if Ko is less than one, and is not possiblein previous sand deposits. Self boring instru-ments with load cells have been used, but may pre-sent problems of compliance, in that the stiffnessof these soils may be significant when compared tothe stiffness of the load cell itself.
Thesuitablestresses(1975) .
pressuremeter device may be the mostmeans of directly measuring lateralin these materials according to Wroth
A great deal of development work needs to bedone in this area of in-situ testing. It is anti-cipated that "secondary" methods, which deducestress conditions from other parameters, will re-ceive increasing attention~
29
In-Situ Measurement of DeformationCharacteristics
It has been emphasized that soil deformationcharacteristics must be measured under the appro-
priate combination of effective stresses, sincestress-strain behavior is stress path dependent.The estimation of these in-situ stresses may be asource of error in conventional laboratory test-ing. In-situ testing has the potential of "by-passing" this source of uncertainty.
E.
Most current methods for predicting soildeformations model the soil skeleton as a linear-elastic material. Nonlinear stress-strain res-
ponse is treated in apiece-wise linear manner.Newer methods, which predict ground response onthe basis of elastic-plastic behavior, are re-ceiving increased attention, but for this dis-cussion in-situ measurement procedures will beconsidered in the framework of the theory ofelasticity.
Application of the theory of elasticitynecessitates a clear distinction between totaland effective stresses. Terms which apply to to-tal stress will be denoted with the subscript "u"while terms relat~d to effective stress will be
identified with a prime, '.
The shear modulus, G, is a convenient defor-mation parameter, because it is defined in termsof a stress difference which may be calculatedfrom either effective or total stress measure-
-ments. For an isotropic elastic material:
al - a3 cri - a;2G = =
El - E3 El - E3(IV-9)
where:G = shear modulus
aI' a3 = major and minor principal stresses
El' E3 = major and minor principal strains
Equation IV-9 may be transformed by applyingrelationships between the elastic constants, yield-ing a convenient relationship between the drainedand undrained modulus of compression:
E' Eu2Gu= 2G'= 1+""'U" = 1+U
where: uE', E = Young'smodulusfor drained
u and undrained loading, respec-tively
(IV-IO)
U', Uu = Poisson's ratio for drained andundrained loading, respectively
It is re-emphasized that these moduli aredependent upon stress level, and should be mea-sured at stress levels appropriate for the proto-type situation.
Methods of Measuring Deformation Parameters
The methods currently used for measuring de-formation parameters in~situ have been reviewed byWroth (1975), and may be catalogued as follows:
1) Pressuremeter Tests have been found to yieldthe most appropriate values of parameters for theprediction of ground deformation caused by founda-tion and earthwork 'construction.
2) Plate Loading Tests are frequently employed,and yield reliable results if performed with care.However, the expense and complexity ofth"e proce-dure may make use of this method impractical inmany cases.
3) Seismic Methods possess unique advantages, butyield deformation parameters which correspond tolevels of strain an order of magnitude or moresmaller than the strains which occur in other test-
ing methods. A correction must therefore be madebefore the resultsare applied to the predidtion ofground deformations other than those caused by vi-bration. "
4) Back Analysis of well documented case histor-
ies is an important means of understanding any as-
pect of soil behavior. This approach is not prac-
tical for "everyday" ,design problems, but does con-
stitute the means of, evaluating the "everyday" de-
sign methods.
In the following sections, each of thesemethods is discussed, with special emphasis placedupon the Pressuremeter Test.
F. Analysis of the Pressuremeter Test (Wroth, 1975)
The results of a two-cycle PMT in sand arepresented in Figure IV-10 as a plot of appliedpressure vs. radial deformation. The two majorcycles of loading are identified by the paths be-tween points ~ through ~ and ~ through ~.
K
£!
60
N 40c-S
,~
O.H0 6
Fig. IV-lO Results of Pressuremeter test in sand
at Landvetter, Sweden (from Wroth, 1975). 1st cycle of loading
0 2nd cycle of loading
30
The secondary unloading and reloading path,through points C, D and E in Fig. IV-10, illus-trates that for small strains, the unloadingbehav-i 0 r is linear and recoverable. When the sand is
reloaded from point D, it responds elastically un-til the previous maximum pressure is reached atpoint E. The sand then yields, undergoing plasticstrain as it rejoins the virgin compression curveBCF.
Upon unloading from point E to g, the sandonce again exhibits elastic rebound. As the ap-plied pressure is decreased from point Q to ~,the membrane of the pressuremeter deflates to itsoriginal size, and a loose annulus of soil col-lapses around the pressuremeter.
The reloading curve, HIJK, exhibits an ini-tial gradient which is an order of magnitude lessthan that of the original loading curve, l[8C, thedifference being due to the extremely disturbedstate of the soil at point H. The behavior ofthe disturbed sand, illustrated in segment HIJ,serves as an eXample of the effects of disturb-ance on the measurement of deformation parametersin conventional pressuremeter tests. This pheno-menon might well occur as the result of perform-ing a PMT in an oversize borehole; a loose annulusof soil may collapse around the apparatus, andsmall pressure increments will induce inordinatelylarge deformations in the disturbed material.
Interpretation of a Drained PMT
The interpretation of an undrained PMT isdiscussed in section III-E. It was explained thatthe complete undrained stress-strain curve couldbe obtained from PMT results in an incompressiblesoil. That analysis has been extended by Wrothand Windle (1975) to take account of volumechanges. The drained stress-strain curve may,thus be derived from a PMT in permeable soil.
The soil surrounding the expanding cylindri-cal membrane is assumed to be deformed in axial
symmetry and plane strain, all displacementstherefore being radial in direction. These as-sumptions have been shown to hold true by Wrothand HUghes (1973) and Hartman and Schmertmann(1975) . As a result" an analysis of the PMT needonly consider one horizontal plane of soil at mid-height of the apparatus and consider the variationof stresses, strains, and displacements along atypical radius ABC shown in Figure IV-ll(a).
A cylindrical coordinate system (r,z,S) isintroduced and the coordinate, r, is chosen torefer to the displaced position B' of the generalpoint B. The displacement, B B', of a general
point is indicated as ~B' Compressive strains aregiven positive values.
The circumferential strain, ES' is everywheretensile, so:
8 ~-E = ; > 0S r - ~ (IV-II)
PMT results have been plotted in Figure IV-lO
in terms of the dimensionless parameter, E = ~A/aO'which is equal to the strain, -Ee, at the wall ofthe cavity in the soil.
(e-j) £ (-EO)£
>~<I
(d) (e-ii)
Fig. IV-II Stress-strain behavior of dense sand
during pressuremeter test (from Wroth,1975)
As the test proceeds, the membrane expands,and an element of soil which was originally locat-ed at point A in Figure IV-ll(a) will be displaced
a distance ~A' and will now be located at point A'.Similarly, soil at point B will move some distance
~B to the point B'.
The radial strain in the soil, Er' is every-where compressive, so that:
E = - d~r dr > 0 (IV-12)
If some drainage is assumed to occur, a soil
element of initial volume Vo will exhibit a volumechange ~V in response to the increased effective
stresses. The volume change may be negative orpositive.
The stress-strain and volume change behavior
are illustrated in Figure IV-II. At any instantduring the test, it may be said that, for the soilin contact with the membrane, the volumetric
strain, Ev, is equal to the tangential strain, Ee,times a factor, -to
.That is to say:
lw = -tEe 'E =v Vo
(IV-13)
where ~V represents an increase in soil volume,
31
and the factor t is positive for soil expansion(dilation), and negative for soil contraction. InFigure IV-ll(c), the volumetric strain is seen tovary linearly with shear strain for a large por-tion of the test. For that portion of the test,the factor t is therefore a constant. The value
of t need not be constant, however, since themethod is applied to the test data at specificpoints in time, and different values of t may besubstituted in each iteration.
The Mohr circle of strain for the soil ele-
ment which has been distorted from point A to A'is presented in Figure IV-ll(b). The angle of di-lation, V, is indicated in the figure, and is de-fined as:
d!:.V/VO
sin V = --a:y-
The parameter t may be expressed in terms ofthe angle of dilation. The plane strain assump-tions imply that the vertical strain, Ez, is zero,and that for small deformations:
!:.V- E + E = +tEe- - - eVo r(IV-14)
therefore: Er
t-l+'E- ecry,.. 15 )
1 - sinvt = 1 - 1 + sinv
2sinV1 + sinv (IV-16)
Alternatively, t may be expressed in terms ofthe stress dilatancy parameter, D, introduced by
Rowe (1962): !:.V/V0 1D=l+-=-
E 1 - tr
The detailed derivation by Wroth and Windle(1975) leads to the following expression for
shearstress,T = 1/2(ar- ae) in terms of theappliedpressurep, and correspondingstrainE:
(IV-17)
E (l+E) (2+E-t)T = 2 (1+tE)
dpdE (IV-IS)
The case of no soil volume change (i.e., a com-pletely undrained test in a saturated soil)corresponds to a value of 0 for the factor t,and equation (IV-IS) becomes
T = E(l+E) (2+E)2
dpdE (!V-19)
Equation (IV-19) is identical to equation (111-9)presented earlier.
The important consequence of this analysisis that if the volume change in the annulus ofsoil immediately surrounding the membrane canbe measured, t can be caiculated and the completestress-strain curve may therefore be derived fromequation (IV-IS). It should be noted that t neednot remain constant during the pressuremeter test,as the analysis applies to each small incrementof the expansion of the membrane, and t may becalculated for each such increment.
Electrical resistivity methods have been
employed to measure this volume change by Windle
-e-..a 8' c c'
8£! r
(b)(0)
fNIv I% 8
/",I a'
I
6I
1t4r- /h
I I ' .10 20 y%
(e)
and Wroth (1975). In ,thelaboratory, soil porositywas correlated with soil resistivity. Soil resis-tivity was measured dur.ingfield pressuremetertests by means of electrodes mounted on the mem-brane face. The field measurements indicated
porosity changes which were consistent and reason-able. The approach suffers, however, from the factthat the measured field resistivity is actually theaverage resistivity of a fairly lar.gevolume ofsoil. The method of analysis calls for considera~tion of volume changes only with a narrow annulusof soil.
In practice, it is not yet feasible to measuresoil volume changes during a conventional PMT. The"drained" analysis is useful nOI1etheless,in esti-mating the probable errors involved when the "un-drained" analysis is applied to a test in whichsome drainag~ occurs.
Laboratory tests performed by Windle and Wroth(1975) indicate that a value of -0.3 for the volume
change factor, R",is appropriate for a normallyconsolidated clay'in a fully drained PMT. If thisvalue of R, is substituted in the "drained analysis,
the computed values of shear stress will be some-what higher than the values of shear stress cal-culated according to the "undrained" analysis.The "undrained" analysis is seen to underestimateshear stress by some 25% for R, = -0.3 and E = .06.
Use of the undrained analysis for a partiallydrained PMT in clay will therefore result in anunderestimate of the peak shear stress. Thiserror is partially compensated, however, by thefact that an undrained PMT,measures a shearstrength which is not,the true in-situ undrainedstrength of the soil, but rather the strength ofthe soil after some degree of consolidation hasoccurred. The actual degree of this compensationis not known.
Analysis of a Pressuremeter Test in Sand
The original analysis of a PMT in sand, pre-sented by Gibson and Anderson (1961), assumed thatthl~sand failed at a constant stress ratio, with no
volume change. ~~ese assumptions are reasonableol1lyfor the case,of ~ very l..)csesand. The stress-strain behavior of a dense sand in plcme strain isillustra'tedin Figure IV-II d and,e which indicatethat for a substantial range of strain the sanddeforms at a nearly constant stress ratio,
0'1'-0'3"/0'1'+0'3',and rate of dilation. The beha-vior may be described by the stress dilatancyrelationship:
a ' ,I
(fN/V
)
2
(
'IT O'f
)-= 1+- tan -+-'a3' E 4 2
(IV-33)
For plane strain:
CP'=CP'f c.v.
where cP pertains to shear at constantvolume. c.v.
The angle of dilation, V, has been introducedin the analysis of a 'drained PMT. The definitionof V is recalled, and through triginometric sub-stitution Equation (IV-20) can be transformed to:
32
1 + sincp'
1 - sincp'
cpt
1 + sin c.v. (IV-21)
1 - sincp'c.v.
1 + sinV1 - sinV
which relates cptat any point to the angle ofdilation, V Hughes, Wroth and Windle (1975) haveshown that it is possible to obtain a completesolution to the stress and displacement fieldswithin the annular soil failure zone. The mem-
brane pressure, p, is related to the strain, E,
by the equation:
In E- -Pf -
a In ...£.
Ef(IV-221
where:
sin.cp' (1 + sin V)
1 + sin CP'
and subscript f denotes the onset of failure.
a (IV-,231
If the stress dilatancy relationship inEquation (IV-20) is employed, the parameter, a,may be expressed in terms of either V or cpt:
sin cpt + sin Vc.v.
1 + sin cptc.v.a (IV- 241.
a(1 - sin cpt ) sin cptc.v.
(1 - sin cpt sin cpt)c.v.(IV-251
Equation (IV-22) indicates that if PMT resultsare plotted as lnp vs. lnE, a straight line with aslope, a, should result. The virgin loading seg-ment of the PMT curve in Figure IV-IO has beenreplotted in this manner in Figure IV-12. Anaverage gradient, a = 0.455, is seen to fit the
data closely. If the value of cp'c.v.can beestimated, equations (IV-24) and (IV-2fi) may then
be employed to determine ~' and v.
80
K
60
PIbf;'n~
40
This method is not sensitive to small vari-
ations in CP' c ' and so for most purposes it issufficiently aXcurate to estimate ~'c.v. fromdrained triaxial tests on loose samples of the
sand in question.
20-I 2 4 6 8 10
€%
Fig. IV-12 Results of part of pressuremeter teston sand (after Wroth, 1975)
The calculation of the complete stress-strainfrom a PMT in sand proceeds according to thismethod as follows:
1) While the sand is failing, (along curveBCDEFG in Figure IV-lO), the principal stress ratio,
cr'r/cr'e'is given by the value of~' deduced fromthe experimentally observed value of a. For sand
in contact with the membrane, the major principal
stress, cr'r ,is equal to the ,applied pressure, p,
so the minor principal stress, cr'e, and the devia-
tor stress, (cr'r - cr'e)' may be calculated fromthe stress ratio.
2) For the "pre-failure" stage of the test,corresponding to segmentAB in Figure IV-lO, shearstress may be calculated by means of Equation(IV-IS), if the rate of volume change, i, is known.Application of this equation requires the selectionof a value of i for each data PQint. Since soilvolume changes are small during this phase of thePMT, it has been found convenient, and not inaccu-
rate, to set t = o. when applying Equation (IV-19)to the "pre-failure" stage.
This method has been applied to the PMT datain Figure IV-lO. The resulting stress-straincurves are presented in Figure IV-13. It is seen
that the calculated value of cr'eremains nearlyconstant throughout the test, providing a confir-
mation of the in-situ effective stress, po.
60 a:'r
40
StressIbf/in.2
'/2 (ar'- 0"8)
a;'8
00
L6
~2 4
f%
Fig. IV-13 Stress-strain curves derived from the
pressuremeter test on sand (data inFigure IV-lO) (from Wroth, 1975)
The importance of this method of analysis isevident. The complete stress-strain behavior of asand may be derived from a PMT with a minimum ofcomputation effort. Furthermore, the method does
not require an estimate of the limit pressure, PI'and thus permits the termination of a PMT after
a small strain, of the order of 10%. The importantparameters ~' and V may be accurately evaluated,and deformation parameters may be obtained fromthe stress-strain curve.
33
The gradient of the derived shear-stress curvein Figure IV-13 may be related to the shearmodulus G'. Recalling Equations (IV-13) and(IV-14)
Er - Ee (t - 2)E8= (2 - t)E
therefore:
d(cr' - cr' )1 r e2 dE
d(cr' - cr' )r e
deEr - Ee)
(2 - t)2
= (2 - t) G' (IV- 26)
Pressuremeter Tests in Sand
A number of considerations should be kept inmind when evaluating a PMT in sand, including thefollowing.
The elastic modulus deduced from a PMT is
appropriate for stresses and strains in the hori-
zontal direction only. This may be important when
testing anisotropic materials.
Finite Element studies performed by Hartmanand Schmertmann (1975) have shown that non-linear
soil behavior is required to produce ~V vs. pcurve of the usual type presented in Figure 111-20.
The pressuremeter modulus, EpMT' suggested byMenard (1975) is frequently used in practice asopposed to values deduced theoretically accordingto the methods just described and is defined asfollows:
EpMT 2 (1 + 11) V /jp/~V0
tI:V~27)
where: V0 = volume of the cavity at the
instant when ~p/~V is measured,provided ~p/~V is measured inthe pseudo elastic phase ofthe test'indicated in Figure111-20.
11 = Poisson's ratio
Non-linear soil behavior limits the validityof the assumption of a constant pressuremeter
modulus, and hence EpMT must be considered as afictitious property.
The effect of a smear zone of disturbed soil
on the value of the pressuremeter modulus isillustrated in Figure IV-14 after Hartman and
Schmertmann (1975). In pre-bored holes, soil dis-turbance results in the measurement of a decreasedmodulus. On the other hand, the hysteresis ofborehole unloading and reloading results in themeasurement of an increased modulus. In practice,the two effects may compensate, resulting in themeasurement of a more nearly accurate modulusvalue. The presence of a smear zone does notalter the general shape of the PMT curve.
An investigation by Laier et. ale (1975) has
shown that the length/diameter ratio, L/D, of the
pressuremeter apparatus will affect the limitpressure as indicated in Figure IV-IS. A correc-tion for the limit pressure is therefore required
E2=50
45
UndisturbedMaterial
40
35
30
EI=250 rl =0.1
Widthof SmearZone-ft2rl =0.2
Fig. IV-14 Moduluscalculations as a function ofwidth of smear zone in linear plasticmaterial with V = 0.3 (from Hartmanand Schmertman, 1975)
3DR, 0/0
o~0 97
PLj{PL)~ =Gt)2
OJI0 20 L40
3JM = d60
Fig. IV"'15 Influence of PM length/diameter onmeasured limit pressure (~Vmeasuredalong middle1/3 of pressuremeter)(from Laier et al.,1975)
if the test interpretation is based upon an assump-tion of infinite pressuremeter length. As long asthe ratio, LID, is greater than about 4, measure~ments of elastic modulus do not need to be corrected
for the infinite length assumption.
Pressuremeter Tests in Clay
An investigationby Roy et a1. (1975)hasshown that soil disturbance has a dramatic effecton the observed lateral stress and modulus for
a PMT in clay. The effects of disturbance on the
observed limit pressure were found to be minor.
The effects of different borehole preparation
techniques on observed lateral stress, modulus andlimit pressure are summarized in Figures IV-16, 17,and 18.
Probe Auger DensitySize.mm Size,mm cycm3 Q----
"V 6 AX:44 51 1.00 >114~d 7 AX:44 51 1.04 >1
~ 9,10 AX:44 44 1.04 10 II. 12 BX;60 5. 1.04 <1
0 ~. H.t,He.AX:44 51 1.09 >1HI2
BoreholeNo., Pr-
Fig. IV-16 Horizontal pressure measured at the
Saint-Alban site (from Roy et al.,1975)
Deformation Modulus, Ep -100 KN/m2
00 20 40 60 80
2
4
6EI
:: 8Q.Q)
0
10
12
.
Borehole ~robe Auger DensityRatio'NO'"Pt Slze.mm Size,mm,g/Cm! ~
v 6 AX:44 51 1.00)114~ 0 7 AX:44 51 1.04 >1
~ 9. 10 AX:44 44 1.04 I0 11,12 BX:60 51 1.04 <I0 ~.H... H. AX:44 51 1.09 >1
HI2. AX:44DrivinqandRelaxation
Fig. IV-17 Deformation modulus measured at the
Saint-Alban site (from Roy et al.,1975)
At present, the most effective way of mini-mizing soil disturbance in thePMT is through theuse of self boring devices, such as those sketchedin Figure 111-19.
34
HorizontalPressure, Fb-IOOKN/m2
00 0.8 1.6 2.4 3.2
2
4
6E,
t!Q)0 i
10 J-
35
Limit Pressure, PL-100 KN/m245. a I 2
002 3
2A
E 4..r::.-~ 6a
8
10
3 4 5
8
Fig. IV-18 Limit pressure measured at the Saint-Alban site: A-withdifferent techniques. B-with different probes
(from Roy et al., 1975)
In order to insure that a pressuremeter testis truly undrained, it must be carried out at aspeed such that the rate of strain experienced bythe soil is considerably higher than would normallybe the case in good laboratory practice. Values ofsoil moduli determined in the laboratory may beexpected to be considerably less than the corre-sponding values measured in an undrained PMT. Boththe difference in strain rate between the two
methods and the disturbance of laboratory samplescontribute to this discrepancy.
G. Plate Bearing Tests
Plate bearing tests have been a traditional
in-situ method .for estimating the soil modulus, E,for the purpos~s of' estimating the immediate settle-
ment of spread footings. The vertical modulus ofdeformation is,deduced from the load settlementbehavior of a suitable bearing plate placed at adesired depth. A variation of the method is tomeasure the horizontal modulus using one side ofa trench for the bearing test and the other sidefor the reaction. Experience has proven the re-liability of plate tests, but the large costs in-volved have stimulated increasing interest inalternate methods.
Recent tests at the Division of BuildingResearch Station in the U.K., as reported byMarsland (1973, 1975), have shown that an order ofmagnitude difference may exist between field andlaboratory values of soil moduli. This discrepancymay be minimized by leaving the excavated surfaceexposed as short a time as possible, and by re-establishing the in-situ effective stresses beforetesting.
Shields and Bauer (1975) have undertaken a
comparative study of full scale footing tests,plate bearing tests, pressuremeter tests and lab-
oratory triaxial tests on sensitive clay in Ottawa,Canada. The soil profile is reproduced in FigureIV-19 and a schematic of the footing test arrange-ment is presented in Figure IV-20. The modulidetermined from these tests are summarized in
Figure IV-21.
Two series of pressuremeter tests were per-formed. The first series yielded low modulusvalues as the result of a large degree of soildisturbance. The more painstaking second seriesyielded higher modulus values. The calculationsassumed a value of 0.33 for Poisson's ratio. Useof the correct undrained Poisson's ratio of 0.5
produces an increase of 13% in the calculatedmoduli. The PMT moduli in Figure IV-2l have been
increased by this amount
The plate loading tests are seen to providemodulus values which most closely coincide withthose determined from the model footing test.However, the plate load test is considerably moreexpensive than the other in-situ methods, par-ticularly when it is performed with the degree ofcare which is required for accurate results.
H. Vibro-Seismic Tests
The J:;>asisof all seismic exploration techniquesfor evaluation of deformation characteristics is
an interpretation of the time required for par-ticular types of elastic waves to travel known orinferred distances thro~gh the subsurface. Avariety of information of interest to the geo-technical engineer may be obtained from vibro-seismic test results, and a more complete review
of this method is planned for a subsequentreport.
A state of the art review of vibroseismicmethods for the determination of soil moduli
a orehole No. Probe Size Auger Siz e Ratiop - mm mm ar -
V 6, 7 AX: 44 51 >1A 8,9,10 AX: 44 44 I<> II, 12 ax: 60 51 <I. 17 ax:60 Sampler(57mm). 18 Mini: 22 Drivino0 I, 2,3, 14 Drivino. 23,24 I I Dr_ivino
36
has been presented by Ballard and McLean (1975).The following five approaches were discussed:
Refraction
Rayleigh wave - vibratoryRayleigh wave dispersionCrosshole methods
UpholejDownhole methods
source
s:
~E0
Natural Water Content I
and Atterbero Limits20 40 60%
SoilType
Clay-Brown
1.5 IFissuredCiQy:-Grey-
. Brown3.0
Clay-Grey-
.0. Clay-Grey,Silty,SiltJ-nt.s-SomeGrovelGrovelOverRoc
tt t
~
0
~
to
I 0
10 '36. 1.7
100
::::',:imi',:' :. ::. ".:::~.::".11111,: '. " :' '.:' -.E1857m
ReinforcedConcrete' BottomOfFootinQTest Footing
3.1mx3.lmxo.66m 7.6cm I.D.x2m LongHydra Rigid TendonSheath
Soil Sleeve. 2. em I.D.Pipe
Tendon Casino 16cm 1.0.Pipeto Rock
.Tendon 12/0.6 Extra HighStrength Conenco ITest Load of 2.7MN isO.85h
-EJ. 80.2 mBottom of Soil Sleeve
Deep SettlementAnchors
E
t4Q)a
Casing wos !7tdEI.68.3m RockSurface
Reamed to 1.2 m. . .
Below Rock Surface 11( l~m'Rock Socket 15.2cmD.~ E'.61.0m Bottom of
Rock Socket
Fig. IV-20 Test arrangement for large scale plateload tests (from Shields and Bauer,1975)
Fig. IV-2l Comparison of modulus values as deter-mined by different techniques (fromShields and Bauer, 1975)
Fig. IV-19
1.8' .25
1.8.24
00
2
6
Eu.UU Cycled
-Triaxial Tests
Es. Secant Modulus.Static KoTriaxial Tests
81
Subsoil profile at
site of large scale
plate load tests(from Shields and
Bauer, 1975)
Ee.CUCycledTriaxial Tests
The study showed the crosshole method to bethe most versatile and the most adaptable to tech-
niques of signal enhancement. It should be noted,however, that under some conditions crosshole
tests may not be possible, economical, or evenreliable because of uncertainties in hole loca-
tion and alignment. Seafloor testing is onesituation wherein uphole-downhole tests may be
more appropriate. It was also found by B~llardand McLean (1975) that the most common errors inthe execution of vibroseismic m~thods lead tounconservative conclusions: i.e., velocities and
moduli which are inaccurately high.
Corresponding field crosshole and laboratoryresonant column test results have been compared byAnderson and Woods (1975). A schematic diagramof their crosshole set-up and triggering systemis presented in Figure IV-22. A time dependent
increase of the shear wave velocity, vs' wasobserved in the laboratory and attributed to
secondary consolidation. The measured variation
of Vs with time is presented in Figure IV-23. Forthe deposits studied, 20 years corresponded to thetime elapsed since the last major stress change
due to excavation or construction. Values of Vsafter 20 years of secondary consolidation wereextrapolated from the lab data, and were found tobe in agreement with the measured field values.
CapacitiveCircuit
Hammer
Tronsd~ BodyWave
- ( ( ( 11I1 I ,~
Impulse Rod
Fig. IV-22 (a) , Sc~ematic of equipment used duringcross-hole tests
::=:: 12.6V 1..10mf
t::--IcopocOOr1000 .n. Resistor
ToScope
ToImpulse Rod To Hammer
Fig. IV-22 (b) Triggering system for cross-holetests (from Anderson and Woods,1975)
37
600
1/1~
eTo =20 psi
~400?::''u0
jCII>03=~ <g200.c ,«)
10psi
5psi-
~Detroit Cloy
Sample R2-1
Primary Response' Secondary Response
10'Z
Time, t -min.
103 10410
Fig. IV-23 Comparison of Vs with time of con-solidation for Detroit clay (fromAnderson and Woods, 1975)
The Hardin-Black (1969) empirical correlation
of v with void ratio, OCR and Po was also appliedto tge soil deposits, but yielded velocities whichwere generally higher than the field and labora-tory velocities.
The in-situ impulse test shown schematicallyin Figure IV-24 (Miller et al., 1975) allows for
the determination of shear modulus at varyingshear strain levels (10-5 to 10-1%). A plot ofarrival time vs. distance of each sensor from theenergy source allows evaluation of the shear wave
velocity, vs' vs. distance relationship. Thetransducer records give vertical particl~ velo~city v vs. time. Shear strain y can be calculated
from:
vy :::Vs
(IV- 2B)
and the corresponding shear modulus G is given by
2G = pvs (IV-29)
where p is the soil density.
1. Conclusions
The in-situ measurement of initial stresses and
deformation characteristics is especially sensitiveto the degree of soil disturbance caused by appa-ratus installation and testing procedure. Distur-bance alters soil behavior by destroying naturalstructure or fabric and by modifying the systemof stresses experienced by the soil.
To some extent, the effects of disturbance maybe alleviated by the re-establishment of in-situeffective stresses, provided they are accuratelyknown. Due to hysteresis effects, however, areversible stress change will not result in a
reversible change in initial ,deformation behavior.In many cases, the effects of changes in soilstructure or fabric are irrecoverable. The most
productive approach towards dealing with soildisturbance is therefore to try to avoid it inthe first place.
RECORDS
~ AnChor~:= Sensor I'f.2 Sensor 2f. ~ Sensor 3
Time GuideRod/ AnchorHole
/ IOin.Dia, I,\ I I/
~6 ISensorHole ~-z; I I
I IL-L--II T I
,I I II I'I WAVE I I" GENERATING I ,
II STATION I III I II, I . IJI I.: I" ImpactHammerl!!,I, - . I ~ ,II BellevilleSPring I' III Load Cell~ .-'I /'I / "",'"-I,
~I."AnchorII. o.,e.II ~ 0 ::Recording II /f?tOrpq --~
M .//
l 'Slolion2 IV~, "':"- -" / :1/ /, . / ,I
,,---~ ' I '/' Vertical II\ / t- -r TransducerII""-' , I ,II ,
II, I I
: I L_Jesting- , I 11 PlaneI,I I
A
r
ilI "
/ ,III
I , JL_- 'Vertical
Displacements
38
Sensor Hole I
GroundSurf a ce
Fig. IV-24 Schematic representation of in-situ i~ulse test(from Miller, Troncoso and Brown, 1975)
with any test method, soil disturbance can beminimized only through the careful use of appro-
priate test procedures. Selection of a test pro-cedure should involv~ a consideration of the par-ticular soil type to be studied. The value ofcompetent and experienced testing personnel isevident.
The self-boring method of advancing instru-ments into the ground is seen as being a verypromising approach for minimizing soil disturbance.It is expected that increased experience with self
boring devices in different soil types will helpto define the effects of details of technique,such as cutting rate, instrument advance rate anddrilling fluid pressure.
The accurate measurement of initial stresses
in the ground is very difficult. Laboratory testsmay provide useful bounds to the likely values ofin-situ stresses, but the number of assumptionsrequired precludes the accurate determination ofin-situ stresses in the laboratory.
Some progress has been made in the measurement
of in-situ stresses in clay soils using pressure
cells, but there is a conspicuous lack of infor-mation on the Deasurement of stresses in sands,stiff clays and soft rock. The determination ofin-situ lateral stress by means of the hydraulicfracturing method is not very reliable in manycases.
The pressuremeter test is seen as being amethod of great promise, and its increased useis predicted. The advantages of the methodinclude:
1) A potential minimization of soil distur-
bance, especially with the use of a self-boringdevice.
2} The possible derivation of the completestress-strain behavior of the tested soil, in-cluding the effects of soil volume change.
3) The fairly accurate estimat~on of in-situ lateral stresses.
Vibroseismic methods of measuring shear and
compression moduli are also of great value. Themethod avoids many of the effects of soil distur-
bance, and measures parameters which are repre-sentative of large volumes of soil. A disadvan-
tage of the method is that the levels of inducedstrain are very small, and thus tpst results mustbe corrected before they are used in most predic-tions of ground deformation. Unlike the PMT,seismic methods do not yield estimates of soilstrength.
The most accurate predictions of deforma-tion behavior may be obtained from large scalefield tests which closely simulate the proto-type loadir.gconditions. The expense of thisapproach, however, precludes its general use.
V. VOLUME CHANGE CHARACTERISTICS
A. Introduction
Of the four geotechnical property classesdiscussed in this report,soil volume change isprobably the one least studied by means of in-situmeasurements. The direct determination of in-situ
soil volume change is difficult, and may requirelong time periods in the case of fine grainedsoils. In-situ measurement of volume change pro-
perties remains desirable, however, since diffi-culties of laboratory testing, such as sample dis-turbance and simulation of the in-situ environment,are partially avoided.
A number of empirical and analytical proce-dures have been developed to predict the rate andamount of soil volume changes in response to stresschanges. These methods are summarized in thefollowing section which is based on portions of thestate of the art paper prepared by Mitchell andGardner (1975).
B. Properties and Procedures
1. Properties
The specific properties that may be used tocharacterize the amount of soil compression andexpansion which may be computed or estimated fromthe results of 'in-situtests include the following:
Cc - One-dimensional compression index =
de
d(loglOPl)
where e is void ratio and pi is effectiveconsolidation pressure
Cs - One dimensional swelling index =
deI
, d'
) on un oa ~ngoglOP ,
Cr - One-dimensional recompression index =
ded(lo~ on reloading10
ff"f '
b'l' de
a - Coe ~c~ent 0 compress~ ~ ~ty = --d Iv p
a
mv - Compressibility = 1 +ve0
39
E - One-dimensional deformation (constrained)s
1modulus m
v
!J.u- C;hange in pore water pressure
y - Unit weight
w - Water content
DR RelativedensityE' - Deformation modulus under drained conditions
~' - Poisson's ratio under drained cionditions.
pi - Effective 9reconsolidation pressure
c (maximum 9ast effective pressure)
- Overconsolidation ratio = 9'~Ip~where p' is the 9resent overburden effectivestress.o
The specific properties needed to describe theconsolidation behavior (rate of volume change) that
may be computed or estimated from the results ofin-situ tests .are the following:
k - Hydraulic conductivity (permeability) in theh horizontal direction
OCR
k - Hydraulic conductivity in the verticalv direction
c - Coefficient of consolidation for vertical
vv compression ~nd vertical flow
c h- Coefficient of consolidation for verticalc compression and horizontal flow.
2. Procedures
A variety of field tests are available that.can yield data from-which volume change propertiescan be deduced. These include:
Plate' load
Screw platePressuremeter
PermeabilityPenetration resistance (dynamic and static)
Monitoring of fills and excavations
In-situ density and water content
Special tests, including blasting, electro-osmosis, dielectricreponse properties,collapse measurements.
C. Consolidation Properties from Borehole orPiezometer Permeability Tests
Theory
If a hydrostatic head difference is appliedbetween the water in a piezometer or borehole probeand the water in the surrounding compressible soil,consolidation or swelling will occur. If the'applied pressure and rate of water flow into orout of the soil are measured, the permeability(hydraulic conductivity) and coefficients of con....solidation or swelling may be calculated.
Spherical and cylindrical piezometer con-figurations are sketched in Figure V-I. A con....
stant head difference, !:m, is maintained through....
out the test. Variable headitests are also
feasible, but their interpretation is moredifficult.
rTwin leads to
piezometer tip
Manometer orpressuregauge.
connected to lead
k = constant
(a)
40
,~o
4uU(r.l)
,,-F Constant' pressure
l;>O UO='Yw"
rh >"?a
Spherical Piezometer
0
;IS<I r:;:;
H
A
k = constant
.a~
(b)
t =0
,All
u (r.o. t)UO-'Yw'"
r
h >>a,b
Cylindrical Piezometer
Fig. V-I Schematic representation of in-situ test for determinationof consolidation properties from permeability measurement
For the spherical piezometer, the governingequation for consolidation or swell is:
2c (a u + ~ au) = au
ar2 r ar at(r>a) (V-I)
where c is the coefficient of consolidation or
swelling, u is hydrostatic excess pressure, r isradial distance and t is time.
The solution for the flow rate, Q, as a func-tion of time is (Gibson, 1963)
k~u
[
l
]Q(t) = 41Ta- 1+-,
Yw \11fT
where: k = permeability
. fct
T = t~me actor = ~a
(V-2)
The values of k and c may therefore be de-duced from the intercept and slope of a plot ofQ(t) vs. l/~, in the manner illustrated inFigure V-2.
This solution has been extended by Wilkinson(1967, 1968) to consider the case of a cylindricalpiezometer in anisotropic soil. The effects of asmear zone and the variation of k and ffivduring
testing are also considered. Large smear zoneswere shown to render the test invalid, but the
influence of variation of k and ffivwas found to
be minor. The effect of soil anistropy is toimparta curvedshape to the plot of Q(t) vs.It\;t.
TESTED NOVEMBER 13-15,1964; 6U APPLIED AT TIP-5-8PS.I.
-f--I- TESTED NOVEMBER 29-30,1965; 6U APPLIED AT TIP-3-7 PS.I.
~ TESTED NOVEMBER 1-2,1966; 6U APPLIED ATTIP-5-9P.S.I.
2.5
0.5
21T16u)bkh.-Yw ~O
0 0.1 0.2 0.3 0.4
YIT Imin.-V2)
0.5 0.6
Fig. V-2 Constant head test results for foundationclay at fiddler's ferry ash lagoon em-bankment foundation illustrating basisfor evaluation of permeability and co-efficient of consolidation (data fromAI-Dhahir et al., IQ69)
For the cylindrical case:
a2u 1 au a2u auc. (- + - -) + c - =-h ar2 r2 ar v az2 at
(V- 3)
The solution for T < 0.44 is
8u [Q(t) = ka - 2~ (N - 0.5)Yw
+ 4Y~(N - 0.5) + 12.6 + 7.0JrT IT
(V-4)
where: N = length = 2bdiameter 2a(V-5)
For T < 0.44, water flow through the central
section in Figu~e V-I is purely radial in direction.
For a long permeameter, with a ratio, N > 2,water flow through the ends of the cylinder may beneglected, and the solution for the flow rate be-comes:
2- 2~8u b k (1 + -)
Q(t) - Y h IWTw(V-6 )
T = cht2a
(V-7 )
If a short permeameter is used, the verticalpermeability, k , may be computed according to ame~hod prescrib~d by Jezequel and Mieussens (1975):
1. Obtain ch and kh using a long permeameterand Equations (V-6), (V-7)
Normalize short permeameter data, using:2.
41
Yw
QR = Q 2~ (8u) b ~(V-8)
3. Enter Figure V-3 with QR and T to obtainRK
kv = ~/RK (V-9)4.
12
..c
~ 20:;,~I<J
~ ~ '16I::,(\/
0
a:::0
8
4
02 3 4 5 7 9 I/Vf6 8
Fig. V-3 Normalized flow rate as a function oftime factor for a short permeameter(b/a = 0.2) (from Jezequel and
Mieussens, 1975)
Application of Methods
The methods described above have been appliedto field tests in embankment soils as reported byBishop and Al-Dhahir (1969). The field perme-ability values were in agreement with both thefull scale field performance and the laboratory
values. Field tests predicted values of Cv whichwere substantially lessor greater than fieldperformance values in some cases. The discrep-
ancies in Cv were attributed to probable stresshistory effects.
Additional descriptions of the application ofthese methods are provided by Al-Dhahir and Mor-
genstern (1969), Al-Dhahir et al. (1969) and
Jezequel and Mieussens (1975).
Discussion
The availableevidenceindicatesthat thevalues of k and c determined in the field aregenerally greater than the values predicted fromlaboratory tests.
The testing time required for a pressureincrement in field testing is comparable to thatof laboratory testing. The field testing time may,potentially~be greatlyreducedthroughthe use ofa piezometer probe such as describedby Wissa etal. (1975)and shown in FigureV-4 of this report.
Differences in the values of c calculated
from field; laboratory and full scale performancemeasurements are almost unavoidable as the result
of differences in flow direction, stress systemand disturbance.
Borehole and piezometer permeability tests do
not give soil compressibility, mv, directly but itis possible to estimate IDv using the values of k
and c: k
mv = cYw (V-IO)
2.0
c:E.... 1.5uu
0
1.0
SquareA-Rod Thread
PratectivePolyethyleneTubing
3/8' 0.D.
Ferrule
TransducerElectric Cable
O-ringSeals
'pressureTronsducerLocknut
Stainless SteelPorous Tip
Fig. V-4 Schematic of the piezometer probe (fromWissa et al., ~975)
An approach for the evaluation of c that hasgiven good results is to compute it by means of
equation (V-IO) using mv determined from laboratorYtests and k from field measurements.
D. Penetration Resistance as a Measure of SoilCompressibility
General Considerations
As noted in an earlier section the static or
dynamic penetration resistance of a soil is a com-plex function of soil deformation, strength andcompressibility characteristics. Analyticalsolutions for penetration resistance as a functionof compressibility are not yet available, but anumber of empirical correlations have been proposed.
Factors such as pore pressure behavior andpossible grain crushing must be considered whencomparing compressibility and penetration resis-tance. Contact stresses greater than about 100kgf/cm2 may cause crushing of soil grains, withsusceptibility to crushing dependent on mineralogy,gradation, particle size and shape, void ratio andin-situ stress.
A wide variety of penetrometers are in currentuse,butthis discussionwill be limitedto the .8PTand Dutch Cone methods. The methods to be des-cribed apply only to dry or completely saturatedsoils.
Penetration resistance usually decreases whenthe ground water table is crossed, but the effectsof pore pressure generation must be considered forindividual'tests. Soil dilation will induce neg-
42
ative pore pressures and increase penetration re-sistance in a saturated soil.
Penetration tests do.riCit provide information
on the preconsolidation pressure in sands. Thisis a serious limitation, since the compressibilityof sand is significantlyinfluencedby its OCR.
Compressibility as.Indicated by the StandardPenetration Test
Compressibility based on the value of StandardPenetration Resistance, SPR, is usually expressed
as an equivalent constrained modulus, Es' that is,
1 1 + eE =-=-sma v v
(V-II)
Proposed correlations between Es and SPR fordifferent sands are summarized in Table V-I and
Figure V-5. A wide degree of variation is evident.The correlations are of 'little use unless it can
be. establiShed which correlation, if any, holds
true for a particular sand in question.
Correlations betweenSPR and the compressi-
bility of cohesive soils have not been well-defined.
800
700
0Q."""02i
C\I
E0
.......-01"""
600
500
400-111+-
InWCJ):>...J:>00~
300
200
100
O. .0 10 20 30 40 50
STANDARD PENETRATION RESISTANCE - BLOWS/30cm
Fig. V-5 Relationships for cqmpressibilitymodulus as a function of standard
penetration resistance (from Mitchelland Gardner, 1975)
43
Table V-I
Compressibility as Indicated by Standard Penetration Resistance (Mitchell and Gardner, 1975)
Reference Relationship
Schultze andMelzer (1965)
Es = vaO.S22 kg/cm2
v = 246.2 10gN- 263.4 Po + 375.6:!: 57.6
0< Po < 1.2 kg/cm2
Po = effective overburdenpressure
Webb (1969) tons/ft2
tons/ft2Es = 5 (N + 15)
Es = 10/3(N+ 5)
Farrent(1963)
Es = 7.5 (1- jJ2)N tons/ft2
jJ = Poisson's ratio
Begemann(1974)
ES = 40 + C(N-6) kg/cm2 N> 15
Es = C(N+6) kg/cm2 N < 15
C = 3(silt with sand) to
l2(gravel with sand)
Trofimenkov
(1974)E = (350 to 500) logN kg/cm2
Meyerhof
(1974)S = P¥'B /2N inches
S = pvB /N inches
p in tons/ft2, B in inches
Soil Types
Dry sand
Sand
C:Jayey sand
Sand
Silt with
sand to .
gravelwith sand
Sand
Sand and
gravelSilty sand
Note: N is penetration resistancein blows per 30 cm. (blows/ft.)
Compressibility as Indicated by Static ConeResistance
Many correlations between static cone resis-
tance, qc' and compressibility have been proposed.A number of them are listed in Table V-2. Corre-lations have been developed using both the con-stant of compressibility, C, and the compression
index, Cc' For one dimensional compression:,
llh 1 . llcr
- = - - In (1 + ---,)h C cr
(V~12)
and,,
llcrlle = -C log (1 + )
c cr(V-13)
thus, C and Cc are related:
C = 2.3 (1 +e)Cc
(v-14)
The parameter, a, may be used to relate the
cone resistance, qc' with Cc:
a = 2.3 (1 +e)Cc
cr'~
qc(V-IS)
where
cr' = effective overburden stress.0
Basis
Penetration tests infield and in testshaft. Compressibilitybased on e, e andmax.emin. (SchultzeandMoussa, 1961)
Screw Plate Tests
Terzaghi and Peckloading settlementcurves
Analysis of field dataof Schultze and Sherif(1973)
Remarks
Correlationcoefficient= 0.730 for 77 tests
Below water table
Used in Greece
USSR practice
Conservativeestimateof maximum settlement
of shallow foundations
The compressibility, IDv' may be expressed as:
1E s = m = aqc
v(V-16)
Recommended values of a range from 1.5 to 8
for normally consolidated soils, with higher valuesassociated with softer more compressible soils. Aconsistent value of a may be exhibited by a given
deposit, as illustrated in Figure V-6.
A general correlationet al. (1972) is presented
for a given sand, a is nottionship
Dr . 2a = 2 (1 + 100) ]
proposed by Sangleratin Figure V-7. In fact,constant. The rela-
(V-17)
has been proposed by Vesic (1970) for OgeecheeRiver sand.
In general, it is not clear whether a will
increase or decrease withqc' Conflicting re-sults exist and have not yet been explained(Mitchell and Gardner, 1975).
It is evident that since a may vary by tlOO%,
predictions of volume change based on q areroughestimatesat best. C
Table V-2
Compressibility as Indicated by Static Cone Resistance (Mitchell al1d.Gardner, 1975)
Reference
Buisman (1940)
Trofimenkov (1964)
De Beer (1967)
Schultze and Melzer(1965).
Bachelier and Parez
(1965)
De Beer (1967)
Thomas (1968)
Webb (1969)
Relationship
Es = 1.5 qc
Es = 2.5 qc
Es = 100 + 5 qc
Es = 1.5qc
E =.J:.\laD. 522s mv
\I= 301.1 log qc - 382. 3 Po + 60.3 :!:50.3
Es = Cl.qc
CI. = 0.8.,.0.9
CI. = 1.3-1.9
CI. = 3.8-5.7
CI. = 7.7
A = C AoedCoed
ES =Cl.qc
CI. = 3 - 12
5Es = Z(qc + 30) tsf
5Es = 3(qc + 15) psf
Soil Types
Sands
Sand
Sand
Dry sand
Pure sand
Silty sand
Clayey sand
Soft clay
Overconsolidated
sand
3 sands
Sand below watertable
Clayey sand belowwater table
Remarks
Overpredictssettlementsby a factorofabout two
Lower limit
Average
Overpredictssettlementsby a factoroftwo
Based on " field and lab penetration tests
compressibilitybased on e,emaxamd emnCorrelation coefficient = 0.778 for
90 tests valid for Po = 0 to 0.8 kg/cm2
~~
C from field tests
Aoed and Coed from lab oedometertests
Coed = 2.3 (l+e)Cc
Aoed = 2.3 (1+ e)Cs
Based on penetrationand compressiontests in large chambersLower valnes of CI.at higher values. of
qc; attributedto grain crushing
Based on screw plate testsCorrelatedwell with settlementofoil tanks
Reference
Heigh and Corbett(1969)
Vesic (1970)
Schmertmann (1970)
Gielly et al. (1969)
}Sanglerat et al. (1972)
Table V-2 (continued)
E = -1. = aqs m cv
Relationship
Es = 2(1+DR2)qc
DR = relative density
Es = 2qc
Es = aqc
q < 7 barsc
7 < qc < 20 barsq > 20 barscq > 20 barscq < 20 barscq < 20 barscq < 12 barscq < 7 bars:c
50'< w < 100
100 < W < 200
w > 200
20 < qc < 30 barsq > 30 barscq < 50 barscq > 100 barsc
3 < a < 8
}2 < a < 5
1 < a < 2.5
3 < Ct.< 6
}1 < Ct.< 3
2 < Ct.< 6
2 < a < 8
1.5 < Ct.< 4
}1 < Ct.< 1.5
0.4 < Ct.< 1
}
}
2 < a < 4
1.5 < a < 3
a = 2
Ct.= 1.5
Soil Types
Soft silty clay
Sand
Sand
Clays of lowplasticity (CL)
Silts of low
plasticity (ML)
Highly plastic siltsand Clays (MH,CH)
Organic silts (OL)
Peat and organicclay (Pt, OH)
Gravel
Sand
Remarks
See Fig. 2 and text
Based on pile load testsand assumptionsconcerningstate of stress
Based on screw plate tests ~cr= 2 tsf
Basedon 600 comparisonsbetweenfieldpenetration and lab oedometer tests
,j:>.111
See Fig. 9
qc > 12 bars, w < 30% C < 0.2c
qc< 12 bars, w < 25%' C < 0.2c
25 < w < 40% 0.2 < Cc < 0.340 < w <100% 0.3 < C < 0.7c
qc<
7 bars, 100 < W < 130% 0.7 < Cc < 1w > 130 C > 1c
Reference
Bogdanovic (1973)
Schmertmann (1974a)
De Beer (1974b)
Trofimenkov (1974)
Meyerhof (1974)
AlpersteinandLeifer (1975)
Dahlberg (1974)
Table V-2 (continued)
Relationship
ES = a.qc
qc > 40 kg/cm2
20 < qc <40
10 < qc < 20
5 < qc < 10
a.= 1.5
a.= 1.5 - 1.B
a.= 1.B - 2.5
}a. = 2.5 - 3.0
Es = 2.5 qc
Es = 3.5 qc
C > .2qc2cr0
3 qc
A > E: "2 0'0
ES=1.6qc-B
Es = 1.5 qc ' qc > 30 kg/cm2
}Es = 3 qc ' qc < 30 kg/cm2
Es > 3/2qc or Es = 2 qc
Es = 1. 9 qc
Es = f (qc + 3200) kN/m2
Es = t (qc + 1600) kN/m2
Es = a.qc' 1.5 < a. < 2
Es = 3 qc
Es = 7 qc
S = pB/2 qc in consistent 1,1nitsS = settlement
E = (11 - 22) qc
E = a.qc1 < a.< 4
Bulgarianpractice
} Greek practiceItalianpractice
Sand.
)
Fine to medium eand South Africanpractice
Clayeysands,PI< 15%
Sands U.K. practice
Soil Types
Sands, sandy gravels
Silty saturated sands
Clayey silts withsilty sand and siltysaturated sands with
silt
NC sands
NC sands
NC sands
OC sands
Sand
Sand
Sand
Sands
Clays
Cohesionless soil
Overconsolidatedsand
NC and OC sand
Remarks
Based on analysisof silo settlementsover a period of 10 years
L/B = 1 to 2
L/B ~ 10
axisymmetric
plane strain
Belgianpractice
3 < E:< 10, Belgian practice
~0'\
} U.S.S.R. practice
Conservativeestimate,based on analysisof verticalstrain
ES determinedby lab tests on recon-stitutedsamplesof sand
Es back-calculatedfrom screw platesettlementusing Buisman-DeBeerandSchmertmannmethods;a.increaseswith
increasingqc; see text
Fig. V-6
modulus
00
t'..._--'---10
20
...
.-"30.>
.!
§I.Oe01
~500
...
.J:j
~60a..'a70
-...
----.-.
80
90 . otdomthr testD dissipation test
47
sq.tt./ ton0.15
GtVL
Crust
Very soft silty
clay becominqfirmer with dept:h
ISanglerat) Some laminations
of silty finesand and
occasional pocketsof shells
Comparison between compressibility of a soft silty clay
as measured by oedometer tests and based on cone resistance(from Meigh and Corbett, 1969)
Overconsolidation Ratio and Maximum Past Pressurefrom Static Cone Resistance
A knowledge of the maximum past pressure atpoints in a soil deposit is needed for many analy-ses. A determination of this quantity by meansof the static cone penetration test alone is not
possible in cohesionless soils. An estimate ofthe OCR in clays may be obtained through thefollowing method suggested by Schmertmann (1974a):
1) estimate Su from qc
from profile information2) estimate p~
compute su/p~
4) refer to the relationships between su/p;and OCR for different clays, as published for
example by Ladd and Foott (1974).
3)
Alternatively, ~ plot of qc vs. depth may beextrapolated above the ground surface to where
qc = O. The distance between this point and theexisting ground surface corresponds to the depthof soil removed, assUming a constant unit w~ightthroughout the profile.
Load Bearing Tests for th~ Determination ofCompressibility Properties
The load bearing test, LBT, represents thefirst know~ application of in-situ testing toinvestigate soil compressibility. Factors ofcost and testing time have increased the relative
advantages of other in-situ methods, but the LBT
E.
remains a valuable test method, suited particularlyto the testing of uncontrolled fills and stoneysoils.
Basis of Load Bearing Test Interpretation
The load vs. settlement curve derived from
the static LBT is usually interpreted to deducethe allowable bearing pressure, or more funda-mentally, the moduli applicable to volume changeprediction. Only the derivation of Es will beconsidered in this report.
The approaches used for the evaluation of Esfrom the LBT generally fall into three classifi-cations:
1) Elastic Solutions are based on settlementof test plate, soil deformation at differentdepths, and deformations at surface points locatednearby.
2) Statistical Methods use relationships be-tween the .settlementof bearing plates and largescale footings.
3) Finite Element Analyses consider bothelastic and inelastic material response, employingthe same measurements.cited in 1).
Interpretation from.Elastic Solutions
Foundation materials may be idealized as an
isottopic, elastic half space. For a rigid LBTplate:
t)u
>:OJ"tII::H
I::0'r-!UJ
~ I
~J °'I0 !
U :I
48
Cc
6
" "" n
C'
..- .. ..-.
~I
w < 2~. 25 < W < ,30. 30 < W < I.'J- 100< W < 100. 100 < W < 130
W > 130
... ..... .
. ..i
0 J I
.t ' . .. . '. ~.. ~~.-
: ". -"". -'-. "". .
. .I
0zI
I
0 I
..
~
.;,:
. .
,,
" ::.. 'i'. . " .I . . t. . .,;:':':.;. "~--:~ -
. t
.. ,. .I
'j'
ooL0 '°.~ 30 !~I~ 1010
Static Cone Resistance - qc in bars
Fig. V-7 Correlation between compression index and cone resistance(from Sanglerat et al., 1972)
E = I II k (1 - ~2) D
s 0
where:
D
(V-18) given by Mitchell and Gardner (1975).
= diameter of test plate
Interpretation from Statistical Relationships
The best known ,statistical relationship be-tween settlement and footing size for a given
bearing pressure is that'proposed,by TerzaghiandP~ck (1948) for sands. Based on a statisticalanaiysis of LBT results on 'footingsand testplates, the settlement of a footing or plate ofdiameter D has been relatedtd that of a smaller
footing or plate of diameter Dl by the followingequation:
11,1 = influence factors for surface0 loading and loading at depth
k
~
= coefficient of subgrade reaction
= Poisson's ratio
The simplest case is that of a rigid LBT
plate located'at the ground surface. The appro-priate influence factors and the solution forthis case are presented in Figure V-8. The in-fluence factors for a number of other cases are
P(
2D
)
2
PI =, D + Dl
(V-19)
Q
Po --=r-
Fig. V-8 Influence factors for LBT on surface
By recalling Equation (V-18), the ratio ofthe "equivalent elastic moduli," Es/Esl' may beexpressed as follows:
E
(
D+D
)
2~ - ---1. D- - 2D DESI 1
(V-20)
where: elastic modulus det~rmined fromLBT, (through an equation ofthe form of Equation (V-18»
ESI
Es elastic modulus appropriate for
a footing bf breadth or dia-meter D
Dl' D = breadth or diameter of LET or
footing respectively.
For a LBT with Dl equal to unity, Equation(V-20) may be simplified:
-
(
D + 1
)
2 -E = D Es 2D sl
(V-21)
An independent analysis of LBT data by
Bjerrum and Eggestad (1963) has indicated that asingle relationship, such as Equation (V-21), isnot appropriate for sands at various relativedensities. The wide range of observed settlement
ratios, pIPl' for different relative densities andfooting sizes is illustrated in Figure V-9~ TheTerzaghi-Peck correlation is also sketched in thefigure, and is seen to underestimate prediqtionsof settlement in a number of cases.
An amended version of Equation (V-21) hasbeen proposed by Bazaraa (1967):
Es
D(D + 1.5\ 2 D2.5D I sl
(V-22)
where: D is measured in feet.
A plot of settlement ratio vs. diameter whichcorresponds to that predicted by Equation (V-22)has been superimposed on the data in Figure V-9.
<19
The correlation is seen to be somewhat better than
Equation (V-19), but one is persuaded to concurw~th Bjerrum and Eggestad in concluding that aunique relationship independent of relative density,or other parameters, does not exist.
Lambe and Whitman (1969) have noted that aLBT will be most successful if soil disturbance
below the plate is minimized, and if the soil inquestion is homogeneous for a depth which is largerelative to the size of the actual footing. Asoil profile which will yield misleading LBT re-sults is depicted in Figute V-IO. The settlement,ofthe test plate will be due primarily to strains
ih sq~l A, while the footing settlement will bedlie.piim~rilyto strains in soil B. If the twosoils have.different volume change characteristics,the settlement behavior of the plate and footing
may be considerably different.
Interpretation ofLBT from Finite Element Analyses
The finiteelementmethod (FEM) of analysisis readily adaptable to the interpretation ofvolume change parameters in the LBT. Complex test
geometries, as well as non-homogeneous linearlyelastic or non-elastic behavior may be accommo-dated.
Carrier and Christian (1973) have presented
comprehensiveFEM solutions for the settlement ofa rigid circular plate on an isotropic linearlyelastic and on a non-homogeneous linearly elastichalf space. The solutions were based upon thethree following variations of deformation modulus,
Es' with depth, z:
The closed form isotropic elastic solutionsof Case #1 and the non-homogeneous elastic solu-
tion for case #2 as'./presentedby Gibson (1967)are in agreement with the FEM solutions. Nosolutions are avafHible for comparison with theFEM solutions for Case #3.
Compressibility Characteristics from ScrewPlate Tests
A flat pitch.auger device of the type de-picted in Figure V~ll is used in the screw platetest--termed a field 'compres someter by Janbu and
Senneset (1973)-to observe the load-settlementbehavior of soil at depth. The auger is screwedto the desired depth in the soil, increments ofpressure are applied to the hydraulic piston, andthe settlement of the screw plate is observed. Thehorizontally projected area over the single 3600auger flight is taken as the loading plate area.
F.
Screw plate tests are well suited for use insandy soils where undisturbed sampling is diffi-
cult. They are less suitedfor in-situtests onclays except for estimating immediate deformations,because of the long consolidation time that would
be required to define drained deformations.
Q°0=
E = 10 °0 (1-J.l2)D00 s
Po
RIGID PLATESHAPE INFLUENCEFACTORCircular 0.785Square 0.815
Case #1 E = Es 0
Case #2 E = n zs v
Case #3 E = E + n z0 v
50
100I I I. - - I
_8 Dense
]' 0
+ Medium Raft and sin'" footinp
- ~~O::k:
J I"}
' I y
- '" Spre8dfootinp .....~ ",,"'"
:; ~ "..j<" r-
,,' ~" L.&. BAztr..II967). "'i.' ,tJ'" ~; .,'" /. 1M ~ T~lhi-Peck
~cy~t::::--- --~-'_.--/~ _.A--
~,. ~ ..., "L: 100"""
y~.
~ + Da. O.36m(drcuI8r)Da,- O.32m (square),
10WidthratioDIDo-
100 1000
i'Q:10
I
I
Fig. V-g Comparison between settlement and dimension of loaded area(from Bjerrum and Eggestad, 1963)
Test Plotec::J
Footino
Soil A
Soil B
Fig. V-lO A soil profile which will yield misleading load bearingtest results
SCREW PLATE
_.. Hydraulic hose
HYDRAULIC HOSEINNER PIPE 5clliement gouges
OUTER PIPE
PISTON
Anchorfrom e
Earl\,anchor
~ Field cOlTf)ressometer
0Scole
50 100 cm
SCALE
0~cm
FiS. V-ll Screw plate test as developed in Norway
(from Janbu and Senneset, 1973).
The modulus for one-dimensional compressibil-ity, Es' may be calculated from screw plate testresults by employing a relationship proposed bySchmertmann (1970):
p = Cl C2 I:.p
2B I
)L (E: I:.z0 .
(::) (embedment correction)
(V-23)
where:Cl = 1 - 0.5
C = 1 + 0.2 log (t years)2 0.1 (creep
correction)B = least dimension of r~ctangular
footing or diameter ~f bearingplate.
I = settlement influence factorz
'1:. = increment in stressp
The summation form of Equation (V-23) enables
soil layering to be accounted for in the analysis.
Another procedure for determining Es is des-cribed by Janbu and Senneset (1973). The modulusis related to a derived modulus number, m, asfollows:
I-a
E = m p(
~)s a Pa
(V-24)
= 1 for NC fine grained soils2
p = reference stress = 10 ton/m
a 1 atmosphere
Corresponding values of screw plate pressure andsettlement are used as a basis for evaluation of
m, following procedures given by Janbu andSenneset (1973) and summarized by Mitchell andGardner (1975).
The coefficient of consolidation may also bededuced from the time settlement record obtained
during a screw plate test. The method is illu-strated in Figure V-12. Drainage during the screwplate test is essentially radial so the parameter
measured is cvh. The method has yielded resul~sthat compare well with results of oedometer testsas illustrated in Figure V-13.
The pre consolidation pressure may also be
obtained from the screw plate test in the mannerillustrated in Figure V-14.
The Evaluation of Volume Change Propertieswith the PMT
The procedure and apparatus of the pressure-meter test have been described in earlier sec-
tions. For an infinitely thick elastic cylindercorresponding to the tested soil in a PMT, it maybe shown that:
G.
Es
2V (1 + t:.O6.t}J0
---xv (V-25)
51
00
TIME, t, 5 (V-SCALE)25 100 225 1.00 625
0
EE 0.10
~
0.10R1
C, .0.335 t;1-'ZIJJ1::IJJ
JI-
~Q20(/)
IMPERVIOUS
~Q30
0 225 40025 100
EXAMPLE TEST J 3 + 16.95LOAD STEP (['. 392 kPo(SEE ALSO FIG.18)
Fig. V-12 Determination of cvh by screw platetest. Data shown are for a field test
in a sand near Stockholm,SwedenI:.cr' = 392 kPa (from Dahlberg, 1974)
The terms of Equation (V-25) have been de-fined in Section (IV). The significance of the
computed Es depends upon the point of the testcurve being considered. An idealized pressure-volume curve from a PMT is presented .inFigureV-15. The moduli.which are relevant to each phaseof the test are indicated in the figure.
Time limitations and creep effects limit theusefulness of this method in clays for the measure-ment of volume change properties. The method hasproved successful in the interpretation o~ pres-suremeter tests in soft rock.
windle and Wroth (1975) have suggested anelectrical resistivity method for measuring soilvolume change during a partially drained PMT. Themethodhas been describedin Section (IV-E). Thebasis of the test interpretation is a correlationwhich has been observed between soil resistivityand porosity:
resistivity, p = ap n-mw (V...26t
where: a =1.0
n = porosity
m = 1.5 to 2.0 for marine soils
Pw = resistivity of pore wat~r
where: a = 1 for DC clays and undrained loadingof soft saturated clays
= 1/2 for many sands and silts
52
0Cr and Cv in m2/ year
0 100 2000
2 --: -- 1 - ...j_. ....
(a) Silt depositin Stjf6rdal
0 0crein m2/year
200 400
4 4 I
-'1- ---T
~
~
16 --r--
.: 20 -t-..s:
1-=t: 24-10 !
28 ..-t-. --
8(b) Sand deposit
in Steinkjer
Fig. V-13 Comparisons between modulus numbers and coefficients ofconsolidation determined by screw plate tests and byoedometer tests (from Janbu and Senneset, 1973)
PRESSURE (f', kPa
0117'..,'0'
J.. 0 nat'";; ,
10':1,00 800
...:ZW 8~-W-.J': 12w(/)w~ 16~......
wa: 20
Fig. V-14 Evaluation of the preconsolidation pressure, crt, fromthe results of a screw plate test c
(from Dalberg, 1974)
tEd = Deformation modulusEo = Alternated or rebound modulusE+ = Recompression modulusPo - Horizontal earth pressure at restPv = Creep limit~ = Limit pressure
CJtJICC
~U
I
J/// '
v/ II»E::J~
Pseudo-Elastic Zone Plastic Zone
~ Py ~0
0 pressure--XBL 776-9429
Fig. V-15 Idealized pressuremeter test curveindicating soil moduli which areapplicable to each phase of testing
Monitoring of Earth and Earth SupportedStructures
Monitoring during construction and after con-struction should be considered as much a part ofin-situ measurement as is testing strictly forinitial design purposes. Data gathered from aninstrumentation program both shapes the design offuture facilities and influences the construction
rate and sequence of the facility under study. Ina defensive design format, monitoring becomes anintegral part of the design process.
H.
Good quality field measurements are the firstrequirements of a "back calculation" procedure.The most commonly measured quantities are verticalor horizontal deformation, porewater pressure, and
less frequently, soil pressure or load. A com-prehensive description of field instrumentation isgiven by Cooling (1962) and more recent innovationsare described by Dunnicliff (1971) and Selig (1974).
The proper evaluation of effective stressesand the influences of soil creep behavior con-
stitute the largest obstacles to the implementationof .back analyses. Nonetheless, if the appropriatedata are available, a number of theories may be
applied to deduce values of cv' Es, ~, Cs' Cr andCc from the behavior of constructed facilities.The utilization of performance records from largescale instrumented earth loads and excavations
has particular appeal because of the close approx-imation of conditions which will control the be-havior of the as-constructed facility. There are,however, serious problems concerned with this
approach, which include the following:
1) Unknown boundary conditions
2) Isolation of the volumetric component ofmeasured deformation
3)stress
Measurement and prediction of effective
53
4)effects
Evaluation of loading and unloading rate
5) Cost considerations
In-Situ Density, Relative Density and Porosityas Indicators of Volume Change Properties
A knowledge of the in-situ density or relativedensity and water content provides an indicationof the potential for, or magnitude of, severalvolume change phenomena including col!apse, ex-pansion, settlement under cyclic loading and lique-faction.
1.
Undisturbed sampling, the Washington densom-eter, the sand cone and nuclear probes providereliable means for evaluating in-situ water con-tent and density. Penetration resistance, electri-cal resistivity and plate load tests have also beensuccessfully employed to measure density.
Penetration Resistance as a Measure of Relative
Density
Some correlations between relative
density and blow count have been presented inFigures 111-2 and 111-3. Field data indicate thatthe effective overburden stress, a' , should be
mul tiplied.. by KID ~4 before entering v these charts.
This correction may alter the derived relativedensity considerably, since compacted sand fillsmay exhibit a coefficient of lateral pressure asgreat as 1. 5.
1.
CPT
Some suggested correlations between relative
density and qc are presented in Figure V-l6. Therelationships were developed from chamber testresults and statistical correlations. Correla-
tions of this type are subject to a number of un-certainties, including the determination of thereference relative densities themselves.
2. Collapse and Expansion Behavior
In~situ density is a useful indicator ofpotential expansion or collapse in partly satu-rated soils. Figure V-17 presents a general guidefor assessing this behavior on the basis of natural
dry density and liquid limit. Soils with combi-nations of liquid limit and natural dry densitythat plot below and to the right of the solidcurved lines are susceptible to expansion; thoseplotting above and to the left are susceptible tocollapseon wetting. .
J. Volume Change Properties and Expansive Soils
Expansive soils pose very serious problems inmany areas of the world, yet methods for in-situmeasurement of their volume change properties are
in a very early stage of development.
Satisfactory prediction of the amount andrate of heave in any case depends at least on thefollowing factors:
1)
2)
3)
Thickness of expansive soi11ayer
Depth to water table
Extent of soil dessication
4)
5)
6)swelling
""
",
'" /",,,,'" ?
",,,,'" '~/ /' /
/ /' ///' /
"', //"',/ /
/~/ //./ /
// / VERTICALEFFECTIVE. STRESS.4.0 I19/cm!
~ 100z
~90a::w 80a.
I,,II,
~ 70
, t
l
,
a: IC ,'60 I
~ ,!:: 50 ,(/)Z 40wc 30w
~ 201/« 10...JWa::
- SCHULTZE-MELZER(1965SCHMERTMANN(1970
VERTICAL EFFECTIvE STRESS-OI I I I I I I
0 25 50 75 100 125 150 175 200
CONE RESISTANCE Qc (kg/cm2)
~ 100 "~ ,90 ,/ ....---u '/ffi 80 //./~ 70 //./a: // /"
C~ 60 /1 ,/'
>- 50,
/1/.- h~ 40 //1~ 30 // /~ 20 '/1i= 10-/~ /,Wa:: 0 25 50 75 100 125 150 175 200
CONE RESISTANCE, Qc (kg/cm2)
;::,00Z~ 90a::w 80a.
SCHMElnMANN (1971)
- - THOMAS(1969)-,- TURNBULL., 01(1960)
~ 70
cf60~.- 50U)Z 40wc 30w> 20i=« 10...JWa:: 0 25 50 15 100 125 150 115200 225 250 215 300
CONE RESISTANCE, Qc (kQ/cm2)
54
~IOOw~90w0.80
I /
1// /",/
/( ./'I' /
a:70 /,' /'': 60 /,' /'>- I"!:: 50 I ~/~ 40 .!c 1/' -SCHULTZE-MELZER(1965w 30 /' ==~~~:~~T(~::~ (1971)~ 20 , -.- TURNBULL., 01(1960).- /« 10 VERTICALEFFECTIVEG1 / STRESS.O,!5kG/cm!
a:: 0 25 50 75 100 125 150 175 200
CONERESISTANCE Qc (kg/cm2)
;:: 100Z /"w 90 ,/
~ /' /"wao ,," /./"a. / / '~ 70 ,,/ ~,./"
c 60 ,/ ./'- , ~~ 50 /'/' /- i/ /(/) 40 ,,/'~ " '/ /C 30 /? / SCHMERTMANN (1971)
W I' / --THOMAS (1969)> 20 ,,' -.-TURNBULLOf 01 (1960)
i= 10 I / VERTICAL EFFECTIVE
:3 '/ STRESS' 2,0 kG/cm!
~ 0 25 50 75 100 125 150 175 200 225 250
CONE RESISTANCE, Qc(kg/cm2)
Fig. V-l6 Correlations betweenstaticcone resistance,verticaleffective stress, and relative density
In-situ stress
Applied stresses
Soil characteristics which determine theindex and coefficient of swelling
7)
8)
9)
Depth of'seasdnal moisture variation
Chemical and biological environment
Temperature distributions.
In-situ methods are capable of providing in-formation on each of these factors, but the tech-
niques are not yet commonplace.
A distribution of soil swell which varies
parabolically with distance above the water table
has been considered by vijayvergiya and Sullivan(1974). The distribution is sKetched in FigureV-18. The surface heave; H, corresponds to thesummation of soil swell over the depth, h, andis given by the equation
1(
~h)H='3h h 0
(V-27)
where:(~h)0
is the unit swell at the groundsurface
55
10 20
L I QUID LI M I T- .,.40 50 60 7030
60
&LU
Il. 70I>-.... ,VERY HIGH(/)Z~ 80c .V
V>-a:c 90...Jcra:
::) 100I-crZ
V'1/
110M
, ,'" "Specific gravity = 2.70 1
""'Speclflc gravity = 2.60! Lines of 100.,. saturation
S~bolHYMXL
Estimated Degree of EKpansi venessvery highHighMedi um highMediumMedium lowLow
- Fig. V-17 Guide to collapsibility, compressibility, and expansionbased on in-situ dry density and liquid limit
(adapted from Gibbs, 1969)
h
(~h/h )0
SURFACE HEAVE =SHADED AREA
GROUND WATER TABLE
~
Fig. V-18 variation of swell with depth according to Vijayvergiyaand Sullivan (1974)
90
v
Indirect methods which measure "suchparame-
ters as pH, salt content, or electrical propertiesare seen to hold some promise fo~ the in-situmeasurement of volume change properties in expan-sive soils.
K. Conclusions
Several types of in-situ tests ,are available
for determination of volume change properties of
soils and soft rocks, including(l) borehole or
piezometer permeability tests, (2) penetrationresistance measurements, (3) :j.oadbearihgtests,
including screw plate tests" (4) presstiremetertests, (5) and test fills or excavations. Inaddition in-situ tests that provide a measure ofdensity or porosity yield an indicatiqnof volumechange properties. Each:of these,types of testhas advantages and disadvantages, and none can beconsidered suitable for all soils under all con-ditions. Nonetheless, in-situtestsi if:'properlydone and interpreted, woUld seem to overcome.'someof the problems associated with the use of lab-oratory tests for evaluation of volume change'properties, such as sampling (particularly incohesionless soils) and simulation of in7si~ustressconditions. '
The borehole or piezometer permeability testis particularly well::-suited +.or evaluation of .the
rate of consolidation properties of fine-grainedsoils, giving results th,ataccord with full,scaleperformance at least as well as, if not betterthan, predictions based on the results of labora-
tory tests. Whether this type of test can beused for evaluation of compressibility as well
(see equation V-lO) remains to be seen. Problemsbecause of disturbance, rapidly Varyirigpropertieswith distance from the probe and changing stressfields adjacent to the test device may be impor-tant, however.
Penetration resistance measurements, becauseof their simplicity, speed, and low cost areattractive. Correlations with volume change pro-perties are empirical, and the test results incohesionless soils are sensitive to in-situ stress
conditions. The static cone penet~ationtest canbe well suited for evaluation of the compress-
ibilityand relative density of normally consol-idated sands, but can give misleading results ifoverconsolidated sands are misinterpreted asnormally consolidated.
The screw plate test offers a method ,forevaluation of the preconsolidation pressure ofmost soil types, although time limitations willprobably restrict its use to cohesionless toslightly cohesive soils. It provides data fromwhich the compressibility and coefficient ofconsolidation can be calculated.
Analysis of pressuremeter tests can be madein terms of undrained or drained deformation para-meters depending on the time allowed for consoli-
dation under each applied stress. Most appli-cations to date in cohesive soils have been for
determination of uhdrained properties. Analysisof the data in terms of effective stresses or the
56
allowal1.ce .ofconsq:j.idation under each applied
stress could provide a basis for determination ofvolume change properties, however.
Load:bearing tests are usually interpretedusing elastic theory to give a modulus of de-formation for undrained or drained conditions,
depending on soil type and time of loading. Therequired testing time for a fully drained loadb~aring test on a'c6hesive soil may be prohibitive.with recent developments in finite difference andfinite element techniques for analysis of con-solidation, one could conceive a load bearing testwith pore pressure measuremE:mt:sand/or artificialdrains from which both compressibility and con-
solid~tion properties might be determined in ashorttime. '
Large scale test fills and excavations pro-bablY,bffer-the best means for.assessing potentialsettlements 'or volume change. High cost and longtesting time preclude their,use in most cases,however." Significant advances"should be possiblewere more"of the available data for full scalestructures used to back-calculate properties.
This would permit/both the establishment of moredata points on correlation curves and predictionof behavior for other cases on the same soil.
The use of in-situ tests for evaluation of
the swell properties of expansive soils is in itsearly'stages, and'there would appear to be needfor considerable'further work on tests for use inthese materials.
The most suitable currently available in-situtests for evaluation of the volume change pro-
perties of differen-tsoil types are as follows:
Normallyconsolidatedsands - staticpenetra-tion tests, screw plate tests, pressure-meter tests
Overconsolidatedsands - screwplate tests,loadbearingtestspressuremetertests
Soft clay - permeabilitytests, static penetration tests, self~boring pressuremeter tests
Stiff clay, shale - load bearingtests, pressuremeter tests
In the case of cohesionless soils, a know-
ledge of the relative density may be of primaryimportance because of concern over liquefactionpotential in seismic areas. Although the standardpenetration test has been widely used for thispurpose, the available correlations are tenuous,particularly if the in-situ stress conditions areunknown, with substantial variations in relativedensity indicated by a given penetration resis-tance according to different correlations. Thestatic cone resistance may offer more reliablemeans for assessment of relative density, althoughsuitable correlations are still in early stage of
development. To sort out the combined influenceof relative density, in-situ stress and fabric willprobably necessitate the combined use of more thanone test type; e~g. static cone and screw plate.
VI. SU~1ARY AND CONCLUSIONS
This report describes many of the methodscurrently used for the in-situ measurement of soilproperties and indicates procedures for deductionof specific property values from the acquireddata. The references cited provide more completedetails of the procedures for each test.
An assessment of the applicability of themethods discussed is presented in Table VI-I.Each method is listed and its suitability for de-
termining various different geotechnical para-meters is indicatedb~T a "grade" of A, B or C,with A indicating high app1icabi1ity,C indicatinglimited applicability, and no mark indicating .hOapplicability for determination of the property.On the basis of i~formation obtained for prepar-ationof this report, the potential for futuredevelopment of the different methods for improveddetermination of different parameters has beenassessed. These conclusions are summarized inTableVI-2 It is recognized that.such an assess-
ment is very subjective in nature.
It is believed that a number of techniques,such as the SPT, which enjoy widespread use 'at
present, offer relatively little potential forfuture development. This is not to say that thesemethods will not see continued use and refinement.
In many cases, what is necessary is not a drasticimprovement, but rather a standardization of tech-nique.
On the other hand, some methods which are in-
frequently used at present are expected to becomeincreasingly popular, as their potential isrecog-nized and exploited. The PMT is a prime exampleof a method which is expected to see increasinguse. Similarly the static cone penetration testis likely to see greatly increased use in the U.S.
Sophisticated analytical methods for predic-ting soil behavior can now be used to evaluatecomplex boundary conditions and the effects ofnon-homogeneous, anisotropic and non-linear soilproperties. In~situ testing constitutes a poten-tial means of defining such conditions and char-acterizing such behavior, but at.the present timethe accuracyof analyticalmethoas.usually'"sur-
passes the accuracy attainable in measuring inputparameters. In this context, the goal of in-situtesting research is two-fold: to improve ourunderstanding of the behavior of undisturbed soildeposits, and to provide accuracy and relevant de-sign parameters for specific engineering problems.
The potential advantages of in-situ measure-ment of soil properties are both fundamental and
practical. The fundamental advantages include:
57
1) Decreased soil disturbance as comparedto that incurred during sampling and laboratory
testing.
21 Retention of the in-situ stress, temper-
ature, chemical and biological environments.
The practical advantages include:
1) Increased cost-effectiveness of testing.
2) The potential ability to increase theintensity of testing for a given project and thusdevelop a data base which is suitable for statis-tical evaluation.
3) Decreased number of error sources.
The disadvantages.of'in-situ testing lie inthe difficulty of attaining the realization of
these potential advantages, as well as the lack ofa sample (in many instances) for verification ofthe actual soil type being tested. The problem oftest "loading" is a fundamental consideration ofproperty measurement, i.e., '''How much does the
measurement of a given quantity affect the mag-nitude of that quantity?"
Future Developments
possible improvements in in-situ testing maybe categorized as follows:
1. Refinement of existing procedures and
.me~hods of interpretation.
2. Introduction of new methods.
Synthesis of existing methods.'3.
The refinement of existing procedures and
methods of interpretation is an on-going process.Some significant contributions such as the drainedanalysis of the PMT have been described in thisreport, and some important methods, such.as aninterpretation of the CPT in terms of cavity ex-pansion, are expected to be hvai1ab1e ih the nearfuture. A second report is planned con-cerning potential new approaches and methods forevaluation of soil geotechnical properties in-situ. Of particular interest in this regard aregeophysical and remote sensing approaches.
Acknowledgements
Financial support for studies of in-situtesting and analysis procedures and for preparationof this report was provided by the GeoscienceResearch Program of the Lawrence Berkeley Lab-oratory and by the Professional Development Programof Woodward-Clyde Consultants, San Francisco,California. Work supported in part by the U. S.Department of Energy.
TABLE VI-I.
CURRENT SUITABILITYOF METHODS
Dynamic Penetration
Static Penetration
Vane Shear
Falling/ConstantHead, BoreholePermeability
Large Scale Pumping
Hydraulic Fracture
Piezometer Probe
Pressuremeter
Borehole Shear
Direct Shear
Plate Load
Screw Plate
Piezometer
Q)
~r-I
-.-4
0U)
B I
B
A
C
C
TableVI-l.
Q)>-.-4
+JUQ)\1-1lI-/r:r:I
r-I
It! II)
U II)
-.-4 Q)
+J ~~ +JQ)U)>
Q)>-.-4
+JUQ)lI-/
\1-1
r:r:I
Ir-I
It!+Js:: II)
0 II)to:!Q)-.-4~~+J0U):I:
C
C
B
A
C
58
CUrreAt suitability of methods
Q)~::III)II)Q)~III
Q)Io-!
0Ilt
A
A
A
>t+J-.-4II)s::Q)Q
B
B
>t+J-.-4II)0~0III
.Js::Q)+Js::0U
~Q)+JIt!
~
B
C
s::
0-.-4+JIt!~::I+JIt!U)
:>t+J
\1-1 -.-4
Or-l-.-4
+J,Qs:: -.-4
Q) II)
-.-4 II)
U Q)-.-4~\1-1~\1-1SQ) 0out)
C
C
C
C
B
C
.
+JIo-!
~~-.-4
r-II-.-4
~ NQ) -.-4
~ ~Q):2III
A
A
C
A
A
i
Q)t1I
s:: It!0 s::
-.-40.-4
+J It!It!~'tjQ-.-4
'd N'II)0.-4
s:: ~0 0Ute
lI-/ Io-!
0 0
lI-/+J\1-1~Q) OJ0>u.......
B
C
B
C
B
II)
::Ir-I
::I'tj
~~It!
~U)
t.(S
II)
t1I
§:8
C
C
A
C
A
A
s::00.-4+JU0.-4
~~r-I
It!s::~Q)+Js::H
lI-/
0
Q)r-It1I
~
B
B
B
B
A
A
A
.c::+Jt1Is::Q)~+JU)
~It!Q).c::U)
'tj
Q)s::-.-4It!
~
'B0
C
B
A
C
B
B
A
A
A
.c::+Jt1I
§~+JU)
~It!Q).c::U)
'tjQ)s::-.-4
It!~t:)
C
B
B
~+JII)-.-4SQ)
.c::
U
'tj-.-4
::IH~III)
It!
t!1
Soil
Profile
r-IIt!
U-.-4+J~Q)>
r-IIt!+Js::0to:!
0.-4
~
:2
II)s::0-.-4+JU::I
Io-!+JII)
~
C
A
A A
A A
B
C
B
B
A = high applicability; B = moderate applicability; C = limited applicability; D = No applicability.
TABLE VI-2.
POTENTIAL OF METHODS
FOR FURTHER FUTURE
DEVELOPMENT AND
IMPROVEMENT
Dynamic Penetration
Static Penetration
Vane Shear
Falling/Constant
Head, Borehole
Permeability
Large Scale Pumping
Hydraulic Fracture
Piezometer Probe
Pressuremeter
Borehole Shear
Direct Shear
Plate Load
Screw Plate
Piezometer
59
Table VI-2. Potential methods for further future
development and improvement
r--'-r '-r-" r--rl ~! I! ,I I: ......
I
I J : ! i I; Q), I , ; 01i : I
I
i;:>. I I c: rtli ! I I!+J . 10C:;! I i j'M I I 'M'MI I <1J':' UJ I I +J rtl'I :> I! 0 I
! rtl 1-1
iI Q)'M i I 1-1 ! '"dQ
! , :> +J ,I 0 .
I
I
'M
I : 'M 0 I p., :>. .-I .I I +J <1J +J . 0 N
II 0 ~ I I.. 4-f'M +J fJj'M. 1\1)
I
~ (]), I+J 0.-1 1-11:1-1
I' I 4-f ~ 1-1 I
I
c:
.
'M
I
:>.(]) 0 0
i
I
~ ::J. Q) +J.o +J:> UII:~ r-I UJ +J C C:'M 'M
(]) rtl fJj I>: 0 \1)UJ.-I1 I!-! 1-1
I~ 1.-1 I +J ell I ! 0 oM OM fJj'M 0 0~ I ~ ~ ' 8~ !:! t' I U i;j .~. ~ I ~.N
I
I OM Q)
I
N (l) I OM 1-1 1-1 ~ P.,I
elI'M
I
~ +J,..; +JI-I ''''''1-1 ill UJ (]) ::J ~E 1-1 ~I-I'M
I1-1 +> 1-1+J 1-1 I c: I +J +J (]j 0 "
e 0 (l) <1J'
I
O,Q)OO ooo o,(]jjcO rtl OU QJII: 0:>
oo ! :> ::r: p.,! Q IS:(/) U p., U---
C B C
A B B A B B B
B C
B C
C B B
C C
e e B A B e A
e B A A A
C
I
I e1
C
eI
jI ei
A
e e
II c:i .~
UJ I +J::J I 0.-I I'M::J I 1-1
'8 I.r..~
I
r;;j1-1 c:
m I ~..c I+>oo c
~""
~UJ
I
~01 (]jc: .-I::J
I
010 c::>i ..x:
B
B
B
B
B A
e
A
C
A
B
e
e
AI!i
..c+JtJ"1c:())1-1+>oo
'1-1cO(l)..c:oo
'"dQ)c:
'Mrtl1-1
"dc:D
IeA
B
A
B
e
e
A
!
I
I I
~I
~ Soil i
~ P.
rofile 1.
1
OM '.I
E --,.---.....
c5 I UJ
I
c:
'"d .-I 0OM rtl 'M
.
::J .-I +J .j.J.-Ilrtl c: U~ o~
1
2 SI .j.J OM +JUJ 1-1 1-1 fJjrtl Q) 0 ..0(.!) :> II: 0
I
,\t
..c+JtJ"1c:Q)1-1+J(/)
1-1rtl(]j
..coo'"dQ)C
'McO1-1Q
B
B
I
tI
I
I
I
i
I
I
I
I
, eI
I C
!
A = high applicability; B = moderate applicability; e limited applicability; D = No applicability.
C
C
C
A
e
REFERENCES
Ahmad, S., Lacroix, Y. and Steinbach, J. (1975)
"pumping Tests in an Unconfined Aquifer,"
Proceedings of the Conference on In-Situ Meas-
urement of Soil Properties, Specialty Confer-ence of the Geotechnical Division,
North Carolina State University, Raleigh, Vol. I,
pp. 1-21.
Al Dhahir, Z.A. ,Kennard, M.F. andMorgenstern,N.R. (1969) "Observations on Pore Pressures
Beneath the Ash Lagoon Embankments at Fiddler's
Ferry Power Station," Conference on the In-SituBehavior of Soil and Rock, InSt. of Civil
Engrs., pp. 175-186.
Al Dhahir, Z.A.. and Morgenstern, N.R. (1969) "In-
take Factors for Cylindrical Piezometer Tips,"Soil Science, Vol. 107, January.
Alperstein, R. and Leifer, S.A. (1975) "Site In-vestigation with Static Cone Penetrometer,"Journal, Geotechnical Engineering Division,ASCE, Vol. 102, No. GT5, pp. 539-555.
Amar, S., Baguelin, F., Jezequel, J.F. and LeMehaute, A. (1975) "In Situ Shear Resistance of
Clays,".Proceedings of the Conference on In-Situ Measurement of Soil Properties, SpecialtyConference of the Geotechnical Division, NorthCarolina State University, Raleigh, Vol. I,pp. 22-45.
Anagnostopoulos, A.G. (1974) "Evaluation of theUndrained Shear Strength from Static ConePenetration Test in a Soft Silty Clay in patras,Greece," Proc. European Symposium on Penetra-
tion Testing, Stockholm, June 5~7, 1974, Vol.2.2, pp. 13-14.
Anderson, D.G. and Woods, R.D. (1975) "Comparisonof Field and Laboratory Shear Moduli," Pro-
ceedings of the Conference on In-Situ Measure-ment of Soil Properties, Specialty Conferenceof the Geotechnical Division, North Carolina
State University, Raleigh, Vol. I, pp. 69-92.
Andresen, A. and Bjerrum, L. (1965) "Vane Testingin Norway; Contributions from Scandinavia,"Symposium, Vane Shear Testing of Soils, ASTMSTP 193, p. 54.
Arman, A., Poplin, J.K. and Ahmad, N. (1975) "AStudy of Vane Shear," Proceedings of the Con-ference on In-Situ Measurement of Soil Proper-ties, Specialty Conference of the GeotechnicalDivision, ASCE, North Carolina State Universit~
Raleigh, Vol. I, pp. 93-120.
Bachelier, M. and Parez, L. (1965) "Contributionto the Study of Soil Compressibility by Meansof a Cone Penetrometer," Proceedings, 6th In-ternational Conference on Soil Mechanics and
Foundation Engineering, Montreal, Vol. 2,pp. 3-7.
Baguelin, F., Jezequil, J-F., LeMee, E. and Le-
Mehaute, A. (1972) "Expansion of Cylindrical
Probes in Cohesive Soils," Journal of the Soil
Mechanics and Foundations Division, ASCE,
60
Vol. 98, No. SMll, .pp. 1129~1142.
Ballard, R.G. and McLean, F.G. (1975) "SeismicField Methods for In-Situ Moduli," proceedingsof the Conference on In-Situ Measurement ofSoil Properties, Specialty Conference of theGeotechnical Division, North Carolina State
University, Raleigh, Vol. I, pp. 121-150.
Bazaraa, A. (1967) "Use of the Standard Penetration
Test for Estimating Settlements of Shallow
Foundations on Sand," Ph.D. Thesis, University
of Illinois, Urbana.
Begemann,H. (1974) "GeneralReportfor CentralandWestern Europe," Proceedings of the European
Symposium on Penetration Testing, Stockholm,June 5-7, 1974, published by National SwedishBuilding Research, 1975, Vol. 2.1, pp. 29-39.
Bemben, S.M. and Myers, D.A. (1974) "The Influenceof Rate of Penetration on Static Cone Resis-tance Values in Connecticut River valley Varved
Clay," Proceedings of the European symposium onPenetration Testing, Stockhblm, June 5-7, 1974,
published by National Swedish Building Research,1975, Vol. 2.2, pp. 33-34.
Berezantzev, V.G. (1961) "Load Bearing Capacityand Deformation of piled Foundations," Proceed-
ings, 5th International Conference on SoilMechanics and Foundation Engineering, Paris,Vol. II, p. 4.
Biarez, J. and Gresillon, J.M. (1972) "Tests andProposed Methods for the Calculation of theBearing Capacity of Piles in a Granular Medium,"Geotechnique, Vol. 22, No.3, pp. 433-450.
Bishop, A.W. and Al Dhahir, Z.A. (l969) "Some Com-parisons Between Laboratory Tests, In-SituTests and Full Scale Performance, with SpecialReference to Permeability and Coefficient ofConsolidation," Conference on In-Situ Behaviorof Soil and Rock, Inst. of Civil Engineers,
pp. 251-264.
Bjerrum, L. (1967) "Engineering Geology of Norweg-ian Normally Consolidated Marine Clays asRelated to Settlement of Buildings," Geotech-
nique, Vol. 17, pp. 81-118.
Bjerrum, L. (1972) "Embankments on Soft Ground,"Proceedings, ASCE Specialty Conference on Per-formance of Earth and Earth Supported Struc-tures, Vol. 2, Purdue University, pp. 1-54.
Bjerrum, L. (1973) "Problems of Soil Mechanics andConstruction on Soft Clays," Proceedings 8thInternational Conf. on Soil Mech. and Found.
Engrg., Vol. 3, Moscow, pp. 111-159.
Bjerrum, L. and Eggestad, A. (1963) "Interpretationof Loading Tests on Sand," Proceedings, Euro~anConf. on Soil Mech. and Found. Engrg., Vol. I.
Bjerrum, L., Nash, J.K.T.L., Kennard, R.M., Gibson,
R.E. (1972) "Hydraulic Fracturing in Field
permeabili ty Testing," Geotechnique, Vol. 22,No.2.
Bogdanovic, L. (1973) "Settlement of Stiff Struc-tures (Silos) Founded on Soft Soil with LowPenetration Resistance," Transactions of the SRInstitute for Testing Materials, Publ. No. 34,Belgrade.
Brand, E.W., Moh, Z-C. and Wirojanagud, P. (1974)"Interpretation of Dutch Cone Tests in SoftBangkok Clay," Proceedings of the EuropeanSymposium on Penetration Testing, Stockholm,June 5-7, 1974, published by National SwedishBuilding Research, 1975, Vol. 2.2, pp. 51-58.
Brooker, E.W. (1964) "The Influence of StressHistory on Certain Properties of RemoldedCohesive Soils," Ph.D. Thesis, University ofIllinois.
Brown, R.E. (1974) "Estimating Relative Densityfrom Standard Penetration Test Resistances,"
Proc. of the European Symposium on PenetrationTesting, Stockholm, June 5-7, 1974, Vol. 2:1,p. 161.
Brown, R. E. (1975) "Cost Effectiveness of In-Situ
Testing," Proc. ASCE Specialty Conference on
In-Situ Measurement of Soil Properties, Vol. II,pp. 346-348.
Buisman, A.S.K. (1940) Groundmechania, Waltman,Delft.
Cadling, L. and Odenstad, S. (1950) "The VaneBorer," Proceedings, Swedish GeotechnicalInstitute, No.2, Stockholm, 1950.
Carrier, W.D. and Christian, J.T. (1973) "RigidCircular Plate Resting on a Non-HomogeneousElastic Half-Space," Geotechnique, Vol. 23,No.1.
Cooling, L.F. (1962) "Field Measurements in SoilMechanics," Second Rankine Lecture, Geotechni-
que, Vol. 12, No.2, pp. 77-104.
Dahlberg, R. (1974) "Penetration, Pressuremeter
and Screw Plate Tests in a Pre loaded Natural
Sank Deposi,t," -Proc., European Symposium onPenetration Testing, Stockholm, Vol. 2.1,
National Swedish Building Research, 1975,pp. 69-87.
DeBeer, E.E. (1967) "Bearing Capacity and Settle-ment of Shallow Foundations on Sand," Sympos-ium on Bearing Capacity and Settlement ofFoundations, Duke University, Durham, N.C.,April 5/6, 1965, pp. 15-33.
DeBeer, E.E. (1974a) "Decrease in PenetrationResistance Due to Excavation in an Over-
consolidated Glauconitic Sand," Proc. EuropeanSymposium on Penetration Testing, Stockholm,Vol. 2.2, National Swedish Building Research,1975, pp. 103-104.
DeBeer, E. (1974b) Discussion on Interpretation ofResults of Static Penetration Tests, Proceed-
ings of the European Symposium on PenetrationTesting, Stockholm, June 5-7, 1974, publishedby National Swedish Building Research, 1975.
61
de Mello, V.F.B. (1971) "The Standard PenetrationTest," Proceedings, 4th panamerican Conferenceon Soil Mechanics and Foundation Engineering,Vol. 1, pp. 1-86.
De Ruiter, J. (1971) "Electric Penetrometer for
Site Investigations," Journal of the Soil
Mechanics and Foundations Division, ASCE, Vol.
97, No. SM2, pp. 457-472.
DM-7 (1971) "DESIGN MANUAL - Soil Mechanics Founda-
tions, and Earth Structures," NAVFAC-DM7, Naval
Facilities Command, Washington, D.C.
Duncan,J.M. and Seed, H.B. (1966) "Anisotropy and
Stress Reorient<3;tionin Clay," Journal of theSoil Mechanics and Foundations Division, ASCE,
Vol. 92, No. SM5, Sept., pp. 21-50.
Dunnicliff, J.C. (1971) "Equipment for FieldDeformation Measurements," Proc., Fourth Pan-american Conference on Soil Mechanics,and
Foundation Engineering, Vol. II, pp. 319-332.
Durgunoglu, H.T. and Mitchell, J.K. (1975) "StaticPenetration Resistance of Soils, I - Analysis,
II - Evaluation of Theory and Implications forPractice," Proceedings of the Conference onIn-Situ Measurement of Soil Properties, Speci-alty Conference of the Geotechnical Division,ASCE, North Carolina State University, Raleigh,Vol. I, pp. 151-189.
ESOPT (1974) Proceedings of the European Symposiumon Penetration Testing, Stockholm, published byNational Swedish Building Research, 1975.
Farrent, T.A. (1963) "The Prediction and FieldVerification of Settlements in Cohesionless
Soils," Proc. 4th Australia - New Zealand Conf.
on Soil Mech. and Found. Engrg., pp. 11-17.
Flaate, K. (1966) "Factors Influencing the Resultsof Vane Tests," Canadian Geotechnical Journal,Vol. III, No.1.
Fletcher, G.F.A. (1965) "Standard Penetration
Test: Its Uses and Abuses," Journal of the Soil
Mechanics and Foundations Division, ASCE, Vol.
93, No. SM4, p. 67.
Gibbs, H.J. (1969) "Discussion on Expansive Clays,Specialty Session No.3, "Expansive Soils andMoisture Movement in Partly Saturated Soils,"7th Int. Conf. on Soil Mech. and Found. Engrg.,Mexico City.
Gibbs, H. J. and Holtz, W.G. (1957) "Research onDetermining the Density of Sands by Spoon Pene-tration Testing,".Proceedings, 4th Int. Conf.on Soil Mech. and Found. Engrg., Vol. I, p. 35.
Gibson, R.E. (1963) "An Analysis of System Flexi-bility and its Effect on Time Lag in Pore WaterPressure Measurements," Geotechnique, Vol. 13,No.1, pp. 1-11.
Gibson, R.E. (1966) "A Note on the Constant HeadTest to Measure Soil Permeability In-Situ,"Geotechnique, Vol. XVI, No.3, pp. 256-259.
Gibson, R.E. (1967) "Some Results Concerning Dis-placements and stresses in a Non-HomogeneousElastic Half-Space," Geotechnique, Vol. 17,No.1, pp. 58-67.
Gibson, R.E. and Anderson, W.F. (1961) "In-SituMeasurement of Soil properties with the Pres-
suremeter," Civil Engineering and Public Works
Review, Vol. 56, pp. 615-618.
Gielly, S., Lareal, P. and Sanglerat, G. (1969)"Correlation Between In-Situ Penetrometer
Tests and the Compressibility Characteristicsof Soils," Proc. Conf. on In-Situ Investiga-
tions in Soils and Rocks, Inst. of civil
Engineers, London, pp. 167-179.
Golder, H.Q. ,and Gass, A.A. (1963) "Field Tests.for Determining Permeability of Soil Strata," .
ASTM Sp. Pub 1. 322 (Fie.ld Testing of Soils) .
Gordon, B.B. (1975) Panel Discussion: Field Meas-urements of Permeability in Soil and Rock,
Proceedings of the Conference on In-Situ Meas-urement of Soil Properties, Specialty Confer-ence of the Geotechnical Division, ASCE, NorthCarolina State University, Raleigh, Vol. II,
pp. 37-38.
Hardin, B.O. and Blac:k, W.L. (1969) Closure to
"Vibration Modulus of Normally Consolidated
Clay," Journal of the Soil Mechanics andFoundations Division, ASCE, Vol. 95, No. SM6,
November, pp. 1531-1539.
Hartman, J.P. and Schmertmann, J.H.(1975) "FEMStudy of Elastic Phase of Pressuremeter Test,"Proceedings of the Conference on In-Situ Meas-urement of Soil Properties, Specialty Conferenceof the Geotechnical Div., ASCE, North CarolinaState University, Raleigh, Vol. I, pp. 190-207.
Hirst, T.J., Richards, A.F. and Inderbitzen, A~L.(1972) "A Static Cone Penetrometer for OceanSediments," ASTM, STP 501, pp. 69-80.
Huck, P.J., Pincus, H~J., Singh, M.M. and Chugh,Y.P. (1974) "Determination of the In situState of StresS in Soil Masses," Report No.FHWA-RD-74-68, prepared for Federal HighwayAdministration, Available through Natl. Techni-cal Information Service, Springfield, VA 22151.
Hvorslev, M.J. (1951) "Time Lag and Soil Perme-
ability in Ground Water Measurements," Corpsof Engineers, Waterways Experiment Station,vicksburg, Miss., Bulletin 36, 50 pp.
Jaky, J. (1944) "The Coefficient of Earth Pressureat Rest," (In Hungarian), Magyar Mernok esEpitesz Egylet Kozlonye.
Janbu, N. (1974) Discussion to ESOPT "Penetra-tions in Clay at Undrained Conditions," Vol.2.1, pp. 129-132.
Janbu, N. (1975) Panel Discussion, Session V, Pro-ceedings of the Conference on In-Situ Measure-ment of Soil Properties, Specialty Conferenceof the Geotechnical Div., ASCE, North Carolina
State University, Raleigh, Vol. II, pp. 349-351.
62
Janbu, N. and Senneset, K. (1973) "Field Compres-someter--Principlesand Applications," Proc.8th Int. Conf. on Soil Mech. and Found. Engrg.,Moscow, Vol. l-l,pp. 191-198.
Jezequel, J.F. and Mieussens, C. (1975) "The In-Situ Measurement of Coefficients of permeabilityand Consolidation in Fine Soils," proceedingsof the Conf. on In~Situ Measurement of Soil
Properties, Specialty Conf. of the GeotechnicalDiv., ASCE, North Carolina State University,Raleigh, Vol. I, pp. 208-224.
Lacroix, Y. (1960) "Notes on the Determination ofCoefficients of Permeability in the Laboratoryand In-Situ," polytechnic School Zurich(Unpublished) .
Ladanyi, B. (1963) "Evaluation of PressuremeterTests in Granular Soils," Proceedings, 2nd Pan-american Conference on Soil Mechanics and Foun-
dation Engineering,Sao Paulo, Vol. I, pp. 3-20.
Ladanyi, B. (1967) "Deep Punching of SensitiveClays," Proc. 3rd panamerican Conference on SoilMechanics and Foundation Engineering, Caracas,Vol. I, pp. 533-546.
Ladanyi, B. (1972) "In-Situ Determination ofUndrained Stress-Strain Behavior of Sensitive
Clays with the pressuremeter," Canadian Geo-technical Journal, Vol. 9, No.3, pp. 313-321.
Ladanyi, B. and Eden, W.J. (1969) "Use of DeepSounding Penetration Tests in Sensitive Clay,"Proc. 7th Int. Conf. Soil Mech. and Found. Eng.,
Mexico City, Vol. 1, pp. 225-230.
Ladd, C.C. (1973) Invited Discussion of Bjerrum.(1973), Presented at 8th International Conf.
on Soil Mechanics and Foundation Engineering,Moscow.
Ladd, C.C. (1975) Discussion of "The Measurement ofIn-situ Shear Strength," by J.H. Schmertmann,Proceedings of the Conf. on In-Situ Measurementof Soil Properties, Specialty Conference of theGeotechnical Division, North Carolina State
University, Raleigh, Vol. II, pp~i 153-160.
Ladd, C.C. and Foott, R. (1974) "New Design Pro-cedure for Stability of Soft Clay," Journal,Geotechnical Engineering Division, ASCE, Vol.100, No. GT7, pp. 764-786.
Laier, J.E. (1973) "Effects of Pressure meter Probe
Length/Diameter Ratio and Bore Hole Disturbance
on PMT Results in Dry Sand," Ph.D. Thesis,
University of Florida.
Laier, J.E., Schmertmann, J.H. and Schaub, J.H.(1975) "Effect of Finite Pressuremeter Lengthin Dry Sand," Proceedings of the Conference onIn-Situ Measurement of Soil Properties, Speci-alty Conference of the Geotechnical Division,ASCE, North Carolina State University, Raleigh,
Vol. I, pp. 241-259.
Lambe, T. W. and Whitman, R.V. (1969) Soil Mechan-ics, John Wiley, New York, 553 pp.
Lang, S.M. (1967) "pumping Tests for DeterminingAquifer Characteristics," ASTM, STP 417, Perme-ability and Capillarity of Soils, pp. 35-55.
LaRochelle, P., Trak, B. , Tavenas, F., Roy, M.
(1974) "Failure of a Test' Embankment on a Sen-
sitive Champlain Clay Deposit," Canadian Geo-technical Journal, Vol. XI, No.1, pp. 142-164.
Leblanc, A. (1975) "Analyse De La Resistahce Non
Draine des Argiles Mesuree au Scissometre,"M.Sc. Thesis, Universite Laval.
Lemasson, H. (1974) "Un Exemple D'Utilisation DuScissometre de Cahntier," Bulletin de Liaisondes Laboratoires de Ponts de Chaussees, No. 20,
July.
Marsland, 'A. (1973) "Laboratory and In-Situ Meas-urements of the Deformation Moduli of London
Clay," Building Research Establishment Current
Paper 24.
Marsland, A. (1974) "Comparisons of the Resultsfrom Static Penetration Tests and Large In SituPlate Tests in London Clay," Proceedings of the
European Symposium on Penetration Testing,Stockholm, June 5-7, 1974, published by NationalSwedish Building Research,1975, Vol. 2:2,pp. 245-252.
Marsland, A. (1975)
on Glacial Clays
Symposium on theBirmingham.
"In-Situ and Laboratory Testsat Redcar," Proceedings,Behavior of Glacial Materials,
Massarsch, K.R., Broms, B.B. and Sundqui, O. (1975a)"Pore Pressure Determination with MultiplePiezometer," Proceedings of the-Conf. on In-situ Measurement of Soil Properties, SpecialtyConference of the Geotechnical Division, ASCE,North Carolina State University, Raleigh, Vol.
I, pp. 260-265.
Massarsch, K.R., Holtz, R.D., Holm, B.G. andFredriksson, A. (1975b)"Measurement of Hori-zontal In-Situ Stresses," Proceedings of theConference on In~Situ Measurement of Soil
Properties, Specialty Conf. of the GeotechnicalDiv., ASCE, North Carolina State University,Raleigh, Vol. I, pp. 266-286.
McLean, F.G., Franklin, A.G. and Dahlstrand, T.K.(1975) "Influence of Mechanical Variables onthe SPT," Proceedings of the Conference onIn-Situ Measurement of Soil Froperties, Spec-ialty Conf. of the Geotechnical Div., ASCE,North Carolina State University, Raleigh, Vol.I, pp. 287-318.
Meigh, A.C. and Corbett, B.O. (1969) "A Comparisonof In~Situ Measurements in a Soft Clay withLaboratory Tests and the Settlement of oilTanks," Proc. Conf. on In-Situ Behavior of Soil
and Rock, Inst. of Civil Engineers, pp. 173-179.
Menard, L. (1975) "The Interpretation of Pressure-meter Test Results," Sols Soils, Vol. VII, No.
26, p. 7.
63
Meyerhof, G.G. (1961) "The Ultimate Bearing Capac-ity of Wedge Shaped Foundations," Proceedings,Fifth International Conf. on Soil Mechanics andFoundation Engrg., Paris, Vol. 2, p. 105.
Meyerhof, G.G. (1974) General Report for OutsideEurope, Proceedings of the European Symposiumon Penetration Testing, Stockholm, June 5-7,1974, published by the National Swedish Build-ing Research, 1975, Vol. 2.1, pp. 40-48.
Miller, R.P., Troncosco, J.H. and Brown, F.R., Jr.(1975) "In-situ Impulse Test for Dynamic ShearModulus of Soils," Proceedings of the Conferenceon In-Situ Measurement of Soil Properties,
Specialty Conference of the Geotechnical Div.,ASCE, North Carolina State University, Raleigh,Vol. I, pp. 319-335.
Milligan, V. (1972) "Pan~lDi~cussion, Session 1,Embankments on Soft Ground," Proc. ASCE
Specialty Conference on Performance of Earthand Earth~Supported Structures, Vol. 1, PurdueUniversity, pp. 4l~48.
Miliigan, V. (1975):"Field Measurement of Perme-ability in soil and Rock," State of the ArtPaper, Session I, Proceedings of the Conf. onIn-Situ M~asurement of Soil Properties, Spec-
ialty Conf. of the Geotechnical Div., ASCE,North Carolina State university, Raleigh, Vol.
II, pp. 3-36.
Mitchell, J.K. (1976) Fundamentals of Soil Behavior,
John Wiley and Sons, Inc., New York, 422 p.
Mitchell, J.K. and Gardner; W.S. (1975) "In SituMeasurement of Volume Change Characteristics,"State-of-the-Art Report, Proceedings of theConference on In-Situ Measurement of SoilProperties, Specialty Conf. of the GeotechnicalDiv., North Carolina State University, Raleigh,Vol. II, pp. 279-345.
Muhs, H. and Weiss, K. (1971) "Untersuchung vonGreritztragfahigkeitun'Setzungverhaltenflachgegrundeter Einzelfundamente imUngleichformigen nichtbindigen Boden," Deg~boHeft 26.
Palmer, A.C. (1972) "Undrained Expansion of aCylindrical Cavity in Clay: A Simple Inter-pretation of the Pressuremeter Test," Geo-technique, Vol. 22, No.3, pp. 451-457.
Phami T.N. (1972) "Application of the Dutch Cone in
Bangkok Area," M.Eng. Thesis No. 377,. AsianInst. of Technology, Bangkok.
Pilot, G. (1972) "Study of Five Embankment Failureson Soft Clay," ASCE Specialty Conf. on Perfor-mance of Earth and Earth Supported Structures,Lafayette, Purdue University, Vol. I, pp. 81-100.
Poulos, H.G. and Davis, E.H. (1972) "LaboratoryDetermination of In-Situ Horizontal Stress in
Soil Masses," Geotechriique,VoL 22, pp. 177-182.
Prevost, J.H. and Hoeg, K. (1975) "Analysis ofPressuremeter in Strain Softening Soil,"Journal of the Geotechnical Engineering Div-ision, ASCE, Vol. 101, No. GT8, pp. 717-732.
Richardson, A.M., Brand, E.W.and Memon, A. (1975)"In-Situ Determination of Anisotropy of a Soft
Clay," Proceedings of the Conf. on In-SituMeasurement of Soil Properties, Specialty Conf.of the Geotechnical Div., ASCE, North Carolina
State University, Raleigh, Vol. I, pp. 336-349.
Rowe, P.W. (1962) "The Stress-Dilatancy Relationfor Static Equilibrium of an Assembly of Par-ticles in Contact," Proceedings, Royal Societyof London, A, Vol. 269, pp. 500-527.
Roy, M. (1975) "Measurement of In-Situ ShearStrength, II'Proceedings, ASCE Specialty Con-ference on In-Situ Measurement of Soil
Properties, Vol. II, pp. 164-167.
Roy, M., Juneau, R., LaRochelle, P. and Tavenas, F.(1975) "In Situ Measurement of the Propertiesof Sensitive Clays by Pressuremeter Tests,"Proceedings of the Conf. on In-Situ Measure-ment of Soil Properties, Specialty Conf. of theGeotechnical Divo, ASCE, North Carolina StateUniversity, Raleigh, Vol. I, pp. 350-372.
Sanglerat, G. (1972) The Penetrometer and SoilExploration, Developments in GeotechnicalEngineering, Vol. 1, Elsevier, Amsterdam.
Sanglerat, G., Gielly, J., Lareal, P. and Chapeau,C. (1972) "Le penetromitre Statique et la Com-
pressibilite des Sols," Annales de L'InstitutTechnique du Baitment et.Des Travaux Publics,Supplement No. 298, October.
Schmertmann, J.H. (1970) "Static Cone to Compute.Static Settlement Over Sand," Journal of theSoil Mechanics and Foundations Division, ASCE,
Vol. 96, No. SM3, p. 1011.
Schmertmann, J.H. (1971) Discussion to de Mello
(1971) Proc. 4th panamerican Conf. on Soil
Mech. and Foun~. ~ng., Puerto Rico, Vol. III,pp. 90-98.
Schmertmann, J.H. (1972) "Static Cone Penetration
Tests," Canadian National Research Council,
Division of Building Research, Ottawa, ProjectG-l, Internal Report 1.
Schmertmann, J.H. (1974a) "Guidelines for DesignUsing CPT Data," Prepared for Fugro-Cesco,Leidschendam, The Netherlands, July.
Schmertmann, J.H. (1974b) "Penetration Pore Pres-
sure Effects on Quasi-Static Cone Bearing,.qc'"Proceedings, European Symposium on Penetration
Testing, Stockholm, June 5-7, 1974, published
by National Swedish Building Research, 1975,Vol. 2.2, p. 345.
Schmertmann, J.H. (1975) "In-Situ Measurement of
Shear Strength," State-of-the-Art Paper,Session III, Proceedings of the Conf. on In-Situ Measurement of Soil Properties, SpecialtyConf. of the Geotechnical Div., ASCE, North
64
Carolina State University, Raleigh, Vol. I,pp. 57-138.
Schmertmann, J.H. (1976) "pilot Field and Labora-
tory Study of the Feasibility of Using theWissa-Type Piezometer Probe to Identify the
Liquefaction Potential of Saturated Fine Sands,"
Appendix III - An Updated Correlation Between
Relative Density Dr and Fugro-Type Electric
Cone Bearing, ~, Waterways Experiment Station,DACW-39-76-M-6646.
Schmidt, F., Matarazzi, B., Dunnicliff, C.J. and
Alsup, S. (1976) "Subsurface Exploration Methodsfor Soft Ground Rapid Transit Tunnels," ReportDOT-TSC-UMTA-76-3.II, U.S. Dept. of Transporta-tion, Urban Mass Transportation Administration,Office of Research and Development, Washington,D.C.
Schultze, E. and Melzer, K-J. (1965) "The Deter-mination of the Modulus of Compressibility ofNon-Cohesive Soils by Soundings," Proceedings,6th International Conf. on Soil Mechanics and
Foundation Engineering, Montreal, Vol. 1, pp.354-358.
Selig, E.T. (1974) "Soil Strain Measurements UsingInductance Coil Method," ASTM Symposium onPerformance Criteria and Monitoring for Geo-technical Construction.
Shields, D.H. and Bauer, G.E. (1975) "Determinationof the Modulus Deformation of .aSensitive Clayusing Laboratory and In-Situ Tests," Proceed-ings of the Conf. on In-Situ Measurement of SoilProperties, Specialty Conference of the Geo-technical Div., ASCE, North Carolina State
University, Raleigh, Vol. I, pp. 395-449.
Stokoe, K.H. and Abdel-razzak, K.G. (1975) "ShearModuli of Two Compacted Fills," Proceedings ofthe Conference on In-Situ Measurement of Soil
Properties, Specialty Conference of the Geo-technical Div., ASCE, North Carolina State
University, Raleigh, Vol. I, pp. 422-449.
Tabor, G.F., Jr., "In Situ Te?ting: Permeability,"Woodward-Clyde Consultants Field MeasurementsSubcommittee, Information Bulletin.
Tavenas, F.A. (1975) "The Interpretation and Use ofthe Pressuremeter Test for Settlement Predic-
tions," Discussion, Proceedings of the Confer-ence on In-Situ Measurement of Soil Properties,
Specialty Conference of the Geotechnical Div.,ASCE, North Carolina State University, Raleigh,
Vol. II, pp. 352-355.
Tavenas, F., Blanchette, G., Leroueil, S., Roy, M.and LaRochelle, P. (1975) "Difficulties in the
In-situ Determination of Ko in Soft SensitiveClay," Proceedings of the Conference on In-SituMeasurement of Soil Properties, Specialty Con-ference of the Geotechnical Division, ASCE,North Carolina State University, Raleigh, Vol.I,
pp. 450-476.
Terzaghi, K. and Peck, R. (1948) Soil Mechanics inEngineering Practice, 1st Edition, John Wiley,New York.
Theis; C.V. (1935) "Relation between Lowering ofPiezometric Surface and Duration of Dischargeof a Well Using Groundwater Storage," AmericanGeophysical Union Transcripts, pp. 519-524.
Thomas, D. (1965) "static Cone Penetration Testsin London Clay," Geotechnique, Vol. 15, pp. 175-179.
Thomas, D. (1968) "Deepsounding Test Results andthe Settlement of Spread Footings on NormallyConsolidated Sands," Geotechnique, Vol. 18,pp. 472-488.
Todd, D.K. (1959) Groundwater Hydrolog~ John Wile~New York.
Trofimenkov, J.G. (1964) "Field Methods for Test-
ing the Structural Properties of Soils,"Moscow: Building Literature Publishing House.
Trofimenkov,J.G. (1974) "PenetrationTestinginEastern Europe," Proceedings of the EuropeanSymposium on Penetration Testing, Stockholm,June 5-7, 1974, Published by National SwedishBuilding Research, 1975, Vol. 2.1, pp. 24-28.
USBR (1974) (United States Department of theInterior, Bureau of Reclamation), "EarthManual," Denver, Colorado, 2nd Edition.
Vesic, A.S. . (1970) "Tests on Instrumented Piles,
Ogeehee River Site," Journal of the Soil Mech-anics and Foundations Division, ASCE, Vol. 96,No. SM2, pp. 561-584.
vesic, A.S. (1972) "Expansion of Cavities in In-finite Soil Mass," Journal of the Soil Mech-anics and Foundations Division, ASCE, Vol. 98,No. SM3.
Vey, E. (1955) Discussion of paper "Field VaneS.hearTests of Sensitive CohesiveSoils," by
Hamilton.Gray, Proc. ASCE, 81, Paper 843, pp. .17-20, 1955.
vijayvergiyi, V.N. and Sullivan, R.A. (1974) "Sim-ple Technique for Identifying Heave Potential,"Bulletin of the Association of EngineeringGeologists, Vol. XI, No.4, pp. 277-291.
Ward, W.H., Marsland, A. and Samuels, S.G. (1965)"Properties of the London Clay at the AshfordCommon Shaft; In-Situ and Undrained StrengthTests,~'Geotechnique, Vol. 15, pp. 321-344.
Webb, D.L. (1969) "Settlement of Structures onDeep Alluvial Sandy Sediments in Durban, SouthAfrica," Conference on In-Situ Behavior of Soil
and Rock, Inst. of civil Engineers, London,pp. 181-188.
Wilkinson, W.B. (1967) "A Note on the Constant
Head Test to Measure Soil Permeability In-Situ,"Geotechnique, Vol. 17, No.1, pp. 68-71. .
Wilkinson, W.B. (1968) "Constant Head ~n-SituPermeability Tests in Clay Strata," Geotech-nique, Vol. 18, No.2, pp. 177-194.
65
Windle, D. and Wroth, C.P. (1975) "Electrical Re-sistivity Method for Determining VblumeChangesthat Occur During a PressuremeterTest," Pro-ceedings of the Conference on In-Situ Measure-ment of Soil Properties, Sp~cialty Conferenceof the Geotechnical Div., ASCE, North CarolinaState University, Raleigh, Vol. I, pp. 487-510.
Win~land, J.D. (1975)j"Borehol~ Shear Device,"
Proceedings of the Conference on In-SituMeasurement of Soil Properties, Specialty Con-ference of the G~otechnical Div., North CarolinaState University, Raleigh, Vol. I,. pp. 511-522.
Winter, E. aridRodriguez,A. (1975) "Evaluation ofPreconsolidation andFriction.Angle in GranularSoils Using th~ Pressuremeter," Proceedings ofthe Conferenc~ cmln-Situ Measurement of Soil
Prop~rties, Speci~lty Conference.of the Geo-technical Div., ASCE, North Carolina State Uni-
versity,Raleigh, Vol. I, pp. 523-535.
Wissa, A.E.Z., Martin, R.T. and Garlang~r, J.E.(1975) "The Piezometer Probe," Proceedings ofthe Conference on In-Situ Measur~ment of Soil
Properties, Spepialty Conference of the Geo-technical Div., North Carolina State University,
Raleigh, Vol. I, pp. 536-545.
Wroth, C.P. (1972) "General Theories bf EarthPressure and Deformations," Proceedings, 5th
European Conference on Soil Mechanics andFoundation Engineering, Madrid, Vol. 2, pp.33-52.
Wroth, C.P. (1975) "In situ M~asurement of InitialStresses and Deformation Character.istics,"
State-of-the-Art Paper, Proceedings of the Con-Ference on In-Situ Measurement of Soil Proper-ties, Specialty Conference of the GeotechnicalDivision, North Carolina State University,Raleigh, Vbl. II, pp. 181--230.
Wroth, C.P.and l{ughes,J.M.O. (1973) "An Instru-ment for the In-Situ Measurement of the Proper~ties of Soft Clays;" Proceedings, 8th Int. Conf.on Soil Mech. and Found. Engrg., Moscow, Vol.1.2, pp. 487-494.
Wroth, C:P. and'Wihdle, D. (1975) "Analysis of thePressuremeter"Te,stAllowing for Volume Change,"Technical Note, Geotechnique, .Vol. XXV, No.3,pp. 598-604.
a
LIST OF SYMBOLS ANDABBREvIATIO:NS
a
a
. a
a.vB
b
BST
CcCrCs
Cl'C2
.c
cch
cvv
D
D
D
D
DR,Drd
.E
E
Ea
Ed
EpMT
ES
Es,Est
Eu
E+
E'
eFR
fsG
GWT
H
Hc
HlH2
Cell radius of the presstiremeter
Diameter of borehole.for permeabilitytests
Penetration resistance interceptfadtor
Shear stress distribution factor, per-.tainingto~hevane.shear test
Coefficienb of compressibility
Base width dimension
Length pf borehole£or-permeability tests
Iowa.borehole shear test
One7dimensional compression index
One-dimensional recomPress~onindex
One- dimensional swelling ..index
Embedment and. creep correction factorsfor the screw plate 'test :
Cohesion
Coefficient of cohsolidationfor vertical
compression and horizontal flow
Coefficient of consolidation for vertical
compression and vertical flow
bepthof penetrometer .base
Diameter
Stress dilatancy parameter
Vane diameter for the vane. shear test
Relative 'density .
,Diameter
E:lastic modtiltis
Energy input.throughthestandardpene...tration test
Alternated loading br rebound modulus
Deformation modulus,
MOdulus obtained .by.the pressuremetertest .~ .
One dimensional (constrained} modulus
determined by thedloadElastic modulus
bearing test
Young's modulusconditions
Recompression modulus
under undrained loading
Deformation modulus under drainedconditions
Void ratio
Friction ratio on the static penetrometer""<";,
Side fri9.tiO'n"or-.. the static penetrometer
Shear mti~lUS~~i'\::""
Ground water table
shear vane height
Constant Piezometric head
Piezometrichead at time tl' t2
m
my
N
N
NA,NN
Depth increment
;Plasticity index
Settlement influehce factor
Influence factors for surface loading and
. loading at. depth for the' load bearing test
Hydraulic gradient
Lateral earth pressure coefficient for
normally consolidated soils.
In~situ.lateralearth pressure coeftic~ent
In-situ lateral earth pressure coefficient
Maximum passive earth pressure coefficient
Coefficient of subgrade reaction
Coefficient of permeability (~ydraulic
conductivity)
Coefficient.of..permeability.in the hori-zontal direction
Mean coefficient of permeability
.(k = lk .'k. )m v h
Coefficientofp~rmeability in.thevertical direction
Vertical coefficient of permeability forbasing
Length
Factorutilizedihpressuremetertesta~aiysisrelatingvolumetricto tangentialstrain. . .
Transformation ratio, pertaining to
.horizontal and. vertical permeability
(/k7k )h v
Modulus number
Compressibility
Blow.count for standard penetration test
Correlation factor for strength in thepressuremeter test analysis
Standard penetration test blow counts forA, N rods
. Nc' Nyg Bearing capacity -factors
NFF .Standard penetration test blow count for
,ideally free falling hammer
N.P
Nst
n
NC
OCR
Po
PyP
,1-Rf.'
Pf
The slope of the point resistance versuseffective overburden pressure profilefor the static penetration test
Standard penetration test blow counts in
. serisitiveclays
Porosity
Normally conSQlidated
Overconsolidation ratio
Horizontal earth pressure at rest
Creep ,limit
Total overburden pressure
Applied pressure during screw plate test
'\F~~~rpressure when soil fails duringpfes~uremeter test
66
h
Ip
Iz
II' IO
i'
c
K.0
1<0'
.KpK
k
km..,.
k v
k'v
L
Jl,
m
p
Po
p'
p'
p'cPI
PMT
p. s.
Q
q
q
qc
qu
QR
Q - CPT
R
RK
sh
Su
.sUV
Sv
SPR
SPT
T
T
T
t
V
V.~
v
Vs
V
VST
W
WL
Z
a
a.
Y
Yt
Limit pressure during pressuremetertest
In-situ lateral pressure in pressure-meter test
Effective overburden pressure
Stress point, average of the three prin-ciple effective stresses
Effective preconsolidation pressure
Plasticity index
Pressuremeter test
Plane strain
Volumetric flow rate
Deviator stress
Volumetric flow rate
Static cone point resistance
Unconfined compressive strength
Normalized flow rate
Quasi-static penetration test
Total resistance on standard penetrationtest samples
Ratio of horizontal to verticalpermeabilityHorizontal shear strengthin vane sheartest
Undrained shear strength
Undrained shear strength from the VaneShear Test
Vertical shear strength
Standard Penetration Resistance
Standard penetration test
Maximum applied torque in the vane sheartest
Time factor
Time lag in the borehole permeabilitytest
Time or elapsed time
Chamber volume of the pressuremeter
Initial chamber volume of the pressure-meter
Flow velocity
Shear wave velocity
Particle velocity
Vane shear test
Water content
Liquid limit
Layer thickness
Parameter relating cone resistance withconstrained modulus
Penetrometer base semi-apex angle
Shear strain during pressuremeter test
Total unit weight
67
Ys
Yw<5
Er
Ev
Ez
Eo
Saturated unit weight of soil
Unit weight of water
Friction angle between penetrometer andsoil
Radial strain during pressuremeter test
Volumetric strain
Vertical strain
Radial strain of the pressuremeter cell
Equivalent axial compressive strain duringthe pressuremeter test
El,E2' E3 Majo::, intermediateand minor principalstra~ns.
El
Ee
on
11
11
v
~A' ~B
~c'~qPp
P
P
Pw
PI
crh
cr'h
crr
crvcr'v
Circumferential strain during pressure-meter test
Stress ratio between deviator stress and
average principal effective stresses
Bjerrum correction factor during vaneshear tests
poisson's ratio
Angle of dilation
Radial displacement of point A, B duringpressuremeter test
Shape factors
Bulk resistivity during pressuremeter test
Resistivity
Settlement
Soil density
Pore water resistivity
Settlement of one foot diameter bearingplate
Total horizontal stress
Effective horizontal stress
Radial stress during the pressuremetertest
Total vertical stress
Effective vertical stress
crl,cr2,cr3 Major, intermediate and minor principalstresses
T
Tps
<P
<P'
<Pcv
<Po
<t>~MT
Shear stress
Shear stress, plane strain
Angle of internal friction in terms oftotal stress
Angle of internal friction in terms ofeffective stress
Angle of internal friction pertaining toshear at constant volume
Component of angle of internal friction,not due to dilatancy, in terms ofeffective stress
Angle of internal friction, in terms ofeffective stress, derived by thepressuremeter test
