You are on page 1of 19

SETTLEMENT OF SHALLOW FOUNDATIONS ON

GRANULAR SOILS
By G. A. Leonards," Fellow, ASCE,
and J. D. Frost,2 Student Member, ASCE
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

ABSTRACT: A conceptual framework for understanding the effects of


overconsolidation in reducing the compressibility of all types of soil is
presented. A generally applicable method for estimating the settlement
of footings on granular soils is proposed. The procedure uses a combi-
nation of dilatometer and cone-penetration test results to identify the
preconsolidation pressure, while soil moduli—either Young's modulus
or constrained modulus, depending on the boundary conditions—
are obtained from the dilatometer test results. Calibration chamber test
results are used to adjust the dilatometer moduli for the effects of stress
path and for disturbance due to insertion of the instrument. Detailed
examples are given to illustrate the use of the method and to compare
the results obtained with those calculated using currently accepted
methods for estimating settlements.

INTRODUCTION
It has long been recognized (Leonards 1975; Rowe 1975; Lambrechts
and Leonards 1978) and fully confirmed by more recent studies (Jami-
olkowski et al. 1985; Bellotti et al. 1986) that penetration resistance tests of
whatever nature are inherently incapable of sensing the effects of pre-
stress, or overconsolidation (OC), on the compressibility of granular soils.
For example, a typical ratio of the cone-penetration resistance qc in the OC
condition to that in the normally consolidated condition (NC) is 1.1 to 1.2,
while the corresponding ratio of the secant Young's modulus at 25% of the
failure stress ( E y is 6 to 10, with the higher value corresponding to lower
relative densities. Accordingly, procedures for calculating the settlement
of footings on granular soils based on correlations between penetration
resistance and soil modulus [e.g., Terzaghi and Peck (1967) and Schmert-
mann (1970)] will seriously overestimate the amount of settlement if the
deposit has been prestressed. Recognizing that some natural deposits and
all compacted granular fills, including in-situ deposits that have been
subjected to various ground improvement procedures, are prestressed, it
must be accepted that the state of conventional practice is unsatisfactory.
It is clear that in-situ tests to measure soil modulus must in some way
sense compressibility directly. Among the tools available for this purpose
at present are plate load tests, the Iowa stepped blade (Handy et al. 1982),
the Marchetti dilatometer (Marchetti 1975), the screw plate (Kummeneje
1956) and the self-boring pressuremeter (Wroth and Hughes 1973; Baguelin
et al. 1974). It has been argued (Leonards 1985) that, at the present time,
the Marchetti dilatometer is the most generally applicable practical tool for
^rof., School of Civ. Engrg., Purdue Univ., West Lafayette, IN 47907.
?Res. Asst., School of Civ. Engrg., Purdue Univ., West Lafayette, IN 47907.
Note. Discussion open until December 1, 1988. To extend the closing date one
month, a written request must be filed with the ASCE Manager of Journals. The
manuscript for this paper was submitted for review and possible publication on July
9, 1987. This paper is part of the Journal of Geotechnical Engineering, Vol. 114, No.
7, July, 1988. ©ASCE, ISSN 0733-9410/88/0007-0791/$1.00 + $.15 per page. Paper
No. 22591.

791

J. Geotech. Engrg., 1988, 114(7): 791-809


sensing soil compressibility directly, at least for those deposits loose
enough so that compressibility is a significant practical problem. The
purpose of this paper is to review, in principle, the advantages and
limitations of this tool and assess its potential for improving the current
state-of-the-art in regard to settlement prediction. In this connection, a
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

new method is outlined for calculating settlements of shallow foundations


supported on granular soils; also, a suitable framework is provided for
incorporating the results of future research on the interpretation of
dilatometer parameters to obtain soil moduli in situ, thereby enhancing the
accuracy of settlement predictions.

MARCHETTI DILATOMETER
The construction, calibration, and operation of the Marchetti dilatome-
ter and the subsequent determination of parameters from the test results
have been described in detail elsewhere (Marchetti 1980; Schmertmann
1981, 1984, 1986a). Briefly, the dilatometer consists of a thin steel blade
with a flexible sensing membrane mounted on one side. Tests are per-
formed by pushing the blade vertically into the ground and measuring the
lift-off pressure (A-pressure) and the pressure required to expand the
membrane 1.1 mm (B-pressure) at the test depths. Tests are typically
performed at 20-cm intervals to avoid possible disturbances due to
previous tests. The rate of penetration of the blade to test depths is of
minor importance in sands and varies between 1-10 cm/s (Schmertmann
1986a); typically, a penetration of 2 cm/s is used. In dense granular soils
the dilatometer may be driven into place, which will likely influence the
results obtained, although when driving is necessary to achieve penetra-
tion, settlements will usually not be a problem. From the measured lift-off
and expansion pressures, several index parameters are determined as
follows:

dilatometer modulus:
ED = 34.7 {Px - P0) (1)
horizontal stress index:

^ = "b-2 (2)
and dilatometer index:
Pi-Po

where P0 and />, = A- and 5-pressures, respectively, corrected for offset in


the measuring gage and membrane stiffness; u0 = in-situ hydrostatic pore
pressure; and a'vo = in-situ effective vertical overburden pressure. Using
these dilatometer test indices, correlations are available for interpreting:
(1) The coefficient of lateral earth pressure K from KD and either the thrust
on the dilatometer rods (Schmertmann 1983) or the cone resistance
normalized with respect to the effective overburden pressure, qJu'vo,
(Marchetti 1985; Baldi et al. 1986a); and (2) the constrained modulus MD
from KD,ID, and ED (Marchetti 1980).
792

J. Geotech. Engrg., 1988, 114(7): 791-809


EFFECTS OF OVERCONSOLIDATION

In order to provide a fuller and more general understanding of the effects


of stress path and overconsolidation, the physical aspects involved will
first be examined. Consider an element of soil in an ideal NC deposit. The
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

initial state of effective stress is given by condition A in Fig. 1. If the state


of stress is increased along the KD stress path (condition B) and then
unloaded to condition C, allowing lateral strain to maintain the K„ stress
path, then the OCR with respect to either vertical (OCRV) or horizontal
(OCRH) stress is the same (Fig. 1 and Table 1). If, instead, the stress state
at B is unloaded by removing ACT„ under conditions of zero lateral strain
(condition D), then OCRV is the same as for condition C but OCRH =

v (effective) + 4a

K (a'+io)

Condition A Condition B

o'

K a'
KITK (a'+Aa)
H o v

Condition C Condition D

FIG. 1. Definition of Stress States: (A) Ideal NC Soil; (B) Loaded along K0 Line; (C)
Unloaded along K„ Line; (D) Unloaded, Zero Lateral Strain

TABLE 1. OCR Parameters


Case OCR formulation OCR
d) (2) (3)
C OCRV a (crt', + ACTVCT,'

c OCRH" (cr,', + Aa)/o-,',


D OCRV « , + Aa)/CT,',
D OCRH b \IKH
"With respect to vertical stresses.
b
With respect to horizontal stresses.
Note: Where KH = ah(D)l<r„(B).

793

J. Geotech. Engrg., 1988, 114(7): 791-809


Critical State Line
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

(large strain envelope)


V
K Line

Yield locus for


case A

/ Yield locus for


cases B, C and D

p ' - (oj + a'3)/Z

FIG. 2. Stress Paths for Cases Shown in Fig. 1

IIKH, where KH = ratio of <rH(D)/crH(B). As the response to subsequent


loading is strongly influenced by the value of KH, it is recommended that
a data bank be compiled from which KH can be estimated with reasonable
reliability given the soil description, the void ratio (or relative density Dr),
and the OCR. For reconstituted samples of silica sand comprising bulky
grains of relatively uniform particle size distribution and low fines content,
Dr = 40-70%, and OCRV = 2-3, a representative value of KH = 0.80. For
this case, the stress path corresponding to points A, B, C, and D and for
stress increments in the vertical (V) and horizontal (H) directions are
shown in Fig. 2.
It is seen that for loading in the horizontal Hc and vertical Vc directions
from point C, and in the vertical direction VD from point D, the sand
remains in the overconsolidated ("elastic") range for substantial stress
increments, while loading in the horizontal direction HD for case D reaches
the yield locus at a relatively smaller stress increment. Thus, measurement
of moduli at operational stress levels using instruments such as pres-
suremeters, dilatometers, or Glotzl cells will grossly underestimate the
modulus if subsequent loading in situ is in the vertical direction, even if soil
disturbance is kept to a minimum. In the case of the dilatometer or Glotzl
cell, mere insertion of the instrument in OC sands may bring the stress
state close to yield. It is therefore recommended that, in the future, the
effects of overconsolidation be assessed in terms of proximity of the
existing state of stress to the yield locus along the expected stress path.
Otherwise the potential exists for grossly misinterpreting the effects of
prestressing on a particular stress path when loading is along a different path.

CORRELATION OF DILATOMETER AND CALIBRATION CHAMBER TESTS RESULTS


Although calibration chamber (CC) tests do not account for the impor-
tant effects of aging and, thus far, test data are limited to granular soils with
a small range in particle characteristics deposited in a specific manner,
794

J. Geotech. Engrg., 1988, 114(7): 791-809


they are the best means currently available to assess the ability of in-situ
measurements to evaluate soil moduli. Good results are obtained from
tests performed in chambers that are large enough so that boundary effects
are not a serious problem and in which provision has been made to control
boundary stresses and displacements with high precision. In this manner,
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

parameters such as the coefficient of earth pressure at rest K0 , constrained


modulus Mcc, and Young's modulus Ecc can be reliably measured. The
latter two parameters are influenced by soil type, fabric, normal and shear
stress levels, overconsolidation ratio (OCR), relative density Dr, and
stress path. As all of these effects have not yet been appropriately
accounted for in CC tests, available correlations must be tempered by
judgment and local field experience. Nevertheless, sufficient information is
available to arrive at useful general conclusions. Emphasis is placed here
on the effects of prestress and of stress path, particularly, on the fact that
the compressibility due to loads applied in a vertical direction is of interest
for settlement calculations, whereas the dilatometer applies stress incre-
ments to the soil in a horizontal direction.
A review of the best data currently available relating the dilatometer and
CC test results for sands in NC and OC conditions (Bellotti et al. 1985,
1986; Baldi et al. 1986a, 1986b; Robertson et al. 1983; Marchetti, personal
communication, 1986) indicates that the ratio RE = E2S/ED for NC sands is
typically of the order of 0.9, where E25 = Young's modulus measured at a
deviatoric stress level equal to 25% of the failure stress. The corresponding
ratio for OC sands shows appreciable scatter and appears to be more
dependent on the type of sand. For example, for Ticino sand, RE ranges
from 4-10, with the higher ratios being generally associated with lower
relative densities, while for Hokksund sand the typical range is from 2-3.5.
Thus, advantage can be taken of the CC test results to adjust the
dilatometer modulus for the effects of stress path and disturbance due to
inserting the instrument in order to obtain practical estimates of settle-
ment. While further research is needed to obtain more generally applicable
values of RE for OC sands, it is emphasized that the use of an approximate
value of RE is superior to making no allowance whatsoever to account for
the effects of disturbance and stress path on the modulus interpreted from
the dilatometer.
In this paper, a general framework for calculating settlement of founda-
tions supported on both NC and OC granular deposits is presented. The
only other general method for estimating such settlements known to the
writers was proposed recently by Schmertmann (1986b). Schmertmann's
procedure makes no adjustment to the dilatometer modulus for the effects
of stress path and disturbance due to insertion and involves use of a
constrained modulus obtained from correlations with the dilatometer
parameters. As the settlement of isolated footings hardly occurs under
constrained conditions, a method is proposed herein that is both more
direct and more generally applicable to different types of foundations and
layer thicknesses. Comparisons will be made between the settlements
predicted by the two methods to highlight their basic differences.

SETTLEMENT CALCULATIONS
The basis for the proposed method for estimating settlements of footings
founded on granular soils is the procedure previously recommended by
795

J. Geotech. Engrg., 1988, 114(7): 791-809


Schmertmann (1970). The rationale of this procedure is well suited to deal
with soil moduli that vary erratically with depth, with granular strata of
limited thickness, with fluctuations of the groundwater level that may
occur subsequent to the site investigation, and other variables commonly
encountered in practice, e.g., the effects of variable depths of embedment
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

of the foundation. If desired, a modification to account for the effects of


rectangular-shaped footings can be used (Schmertmann et al. 1978).
Schmertmann's expression for settlement S is generalized in the following
form:
D Rz(OC) RZ(NQ
S= Ciqne,JjIzA. (4)
Ez(OC) EZ(NC)
o
where cx = the correction to take into account the effect of embedment
(Schmertmann's c2 factor for time effects is not included because it is
considered to equal one for clean granular soils); qnet = the net increase in
contact pressure; D = the thickness of the granular stratum (<2 x width
of footing); Iz = the strain influence factor; Az = the height of the sublayer;
Rz(OC) — the ratio of the stress increment corresponding to the OC portion
in a given layer to the total increment of stress in that layer; i?z(NC) = the
ratio of the stress increment corresponding to the NC portion for a given
layer to the total increment of stress in that layer; and EZ(QC) and EZ(NC)
= the appropriate £-values corresponding to the OC and NC portions of
the stress increment, respectively, in a given layer.
It is noted that if the deposit remains in the NC range for the total
increment of stress in all layers then alli?z(OC) = 0 and alli?z(NC) = 1, and
thus Eq. 4 reduces to

° /A
S
~ Ciq"e' | E^NC) (5)

which is the standard form of the Schmertmann (1970) expression. For the
NC condition, it is recommended that 2sz(NC) = 0.9 ED .
Similarly, if the deposit remains in the OC range for the total increment
of stress in all layers, then all Rz(OC) = 1 and all i?z(NC) = 0, and thus Eq.
4 becomes

5 - cxqMt ^ zJoc} (6)

which again is the standard form of the Schmertmann expression. For


sands with different material characteristics than Hokksund and Ticino
sands, the use of a value of iiz(OC) = 3.5 ED will generally be conservative
(i.e., low), especially in loose sand deposits. It is recognized that isz(OC)
depends on the reloading stress path and the magnitude of the stress
increment, and that these values vary beneath the footing. Nevertheless, it
is believed that the suggested approximation is acceptable for practical
settlement calculations.
These two conditions are limiting cases. In general, the soil in a given
layer of a deposit will be in the OC range for a certain portion of the stress
increment and then in the NC portion for the remainder of the increment.
796

J. Geotech. Engrg., 1988, 114(7): 791-809


In this case, Eq. 4 is valid where Rz(OC) and RZ(NC) have values >0 and
< 1 , and ^(OC) and EZ(NQ are selected appropriately, i.e., £z(OC) « 3.5
ED and £Z(NC) = 0.7 ED . The reduced coefficient of 0.7 used for the NC
portion results from the fact that the measured ED is an OC modulus, and
calibration chamber tests indicate this to be about 1.3 times the NC
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

modulus. Thus the normal coefficient of 0.9 must be reduced by a factor of


1.3.
To determine the relative contributions of the OC and NC portions and
thus the values of Rz(OC) and RZ(NC), the writers adopted the following
procedure. The stress increase with depth is determined based on an
assumed 2 vertical: 1 horizontal stress distribution, although any suitable
method to calculate the stress distribution may be used. It is necessary first
to estimate K (OC) and § so that OCR and thus p'c can be determined. Two
alternative methods are available to do this, depending on whether CPT
results as well as DMT results are available. If CPT data are available, then
the K (OC)- and (^-values can be determined from correlations with KD and
qch'v, as suggested by Marchetti (1985) [see also Baldi et al. (1986a)] using
the following correlation:

3000
Assumed Cone Roughness

2000
plane strain

(after Marchetti, 1985)


1000 -

Ratio 500
fc
0' 300
V

200

100 -

0.2 0.4 0.6 0.8 1

Coefficient of Earth Pressure K

FIG. 3. Chart for Determining (j> as Function of K and qju[,

797

J. Geotech. Engrg., 1988, 114(7): 791-809


K(OC) = c, + c2KD + c3(qc/<r$ (7)
where c , , c2, and c3 are experimentally determined coefficients whose
values are (Baldi et al. 1986a): cx = 0.376; c 2 = 0.095; and c 3 = -0.0017.
It has been suggested (Baldi et al. 1986a; Marchetti personal communi-
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

cation, 1986) that some modification to Eq. 7 may be required in natural


sands. Thus selection of appropriate values of the coefficients should be
tempered by local correlations and experience. With this value of K(OC)
and the qch'v ratio, it is possible to determine §ps (plane-strain) using the
D&M equations (Durgunoglu and Mitchell 1975). These equations can
readily be programmed. A chart prepared by Marchetti (1985) is shown in
Fig. 3, from which the interrelationship between fyps, qjv'v, and K{OC)
can readily be visualized. The chart was prepared using 8/<(> = 0.5. For a

J I L.

Lade and Lee (1976)

.5 4 0 -

Equation (5)

-1-
—r- —I I 1 - T-
32 34 40 42 44 46
36
(Ji triaxial (degrees)

FIG. 4. Comparison of <(> (Plane-Strain) and $ (Triaxial)

798

J. Geotech. Engrg., 1988, 114(7): 791-809


steel cone in contact with typical sands, the angle of wall friction 8 ranges
between 18°-34°, depending on the angularity, hardness, and roughness of
the grains, and how "polished" the steel is (Brumund and Leonards 1973;
Uesugi and Kishida 1986). For example, for a hard-grained, loose, angular,
fine sand penetrated by a used cone, 8/cJ> would approach 1.0.
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

The cj>pi.-value is converted to 4>uv (axial) using the approximate relation-


ship (Schmertmann 1983)
(<tv " 32)
<l>a.v - <bns - (8)

for fyps > 32°. If <$>ps < 32°, then assume §ax = §ps.
Eq. 8 is in good agreement with results of laboratory tests presented by
Marachi et al. (1981), Lade et al. (1976), and Rowe (1969), as shown in Fig.
4.
Alternatively, the procedure suggested by Schmertmann (1983) that
involves DMT data only can be used. In this case <j>pi. is determined from
the D&M (1975) equations using the dilatometer penetration resistance qD
estimated from the total thrust required to advance the DMT blade. The
c^-value is converted to <j>uv using Eq. 8. The K (OC) value is then
determined from the KD- and <|>aJ.-values using the following expression
suggested by Schmertmann (1983), based on calibration chamber tests:
40 + 2-JKp - 86ffD(l - sin 4>„.t) + 152(1 - sin <KJ - 717(1 - sin V - ) 2
K(OC) =
192 - 717(1 - sin c|>„.r)
(9)

Mayne-Kulhawy (((1=43)

Mayne-Kulhawy ($=35)

Modified Mayfie -Kulhawy (+-43)

^ Modifled Mayne-Kulhawy ($"»35)

Wabash River Sand


Range of $ values
(Hendrou, 1963)

II \ i |

10 12 H 16 18 20

rcbnsolidation Ratio OCR

FIG. 5. Comparison of Measured and Calculated K-OCR Relationships


799

J. Geotech. Engrg., 1988, 114(7): 791-809


Regardless of which method is used to calculate K (OC) and $ax, the
OCR is determined using the Mayne-Kulhawy (1982) procedure but
modified as suggested by Schmertmann (1983) to
(1/0.8 sin <b)
K(OC)
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

OCR = (10)
1 - sin §
Eq. 10 is further confirmed by the plot in Fig. 5, which shows good
agreement with the results of Hendron's (1963) one-dimensional compres-
sion tests. The Schmertmann (1983) method to obtain OCR calculates qD
from the total thrust required to advance the DMT blade. This relies on the
assumption of negligible friction on the rods and DMT blade. The
Marchetti (1985) procedure can be in error due to spatial variations in the
soil properties, since CPT and DMT data cannot be obtained from
precisely the same location. It is noted, however, that evidence to date
indicates good agreement between the K (OC) values as determined using
the two methods (Baldi et al. 1986a). Further support of this agreement is

Overconsolidation Ratio
5 6 7 8 9 10
_J • i I I I—

4.0 .

Depth
(in)
Schmertmann (1983)
M a r c h e t t i (1985)

Data for p r e - d e n s i f i c a t i o n p r o f i l e s from:


MIT 31-1655
CPT 11-1655

Data for p o s t - d e n s i f i c a t i o n p r o f i l e s from:


DMT 32-4943
CPT 12-4943

_, ,—_, _,. ! r—_, ,——!—_r. , _, , _,__,.


FIG. 6. Comparison of OCR Calculated by Different Methods

800

J. Geotech. Engrg., 1988, 114(7): 791-809


shown in Fig. 6, where OCR profiles for predominantly NC and predomi-
nantly OC deposits are plotted based on the two methods.
Once the preconsolidation pressure p[, has been determined from the
initial vertical effective stress o^ and the OCR, then the ratios Rz{OC) and
i?z(NC) can readily be obtained for each layer using the following relations:
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

0C
^ )=(f^)-- <">
N C
^ ) = ( ^ ) - - V ^
where <x'f = the final stress at the center of the layer.

EXAMPLES

In order to evaluate the proposed method of settlement calculation, data


for a soil deposit (Table 2) which has been dynamically compacted and
additional data (Table 3) from the same location but obtained prior to
dynamic compaction were considered. Expected settlements were calcu-
lated using three different approaches:

1. The conventional Schmertmann (1970) method (E = 2qt), which is


based solely on cone penetration resistance tests.
2. The procedure outlined herein, which utilizes the Schmertmann
(1970) procedure with a modified E = 3.5 ED in the OC portion of the stress
increment and E = 0.7 ED in the NC portion of the stress increment [if
Rz(OC) = 0, use E = 0.9 ED].
3. The Schmertmann (1986b) method.
The three methods considered a square footing 4.0 m x 4.0 m founded
at a depth of 2.0 m below ground level with a net contact pressure of 500
kPa. The soil profile was divided into 0.5-m thick layers for settlement
calculations.
The settlements calculated using the procedures described are summa-
rized in Table 4. If only CPT data qc and Schmertmann's (1970) procedure
were used with E = 2qc, it is likely that the ground improvement
procedure would be rejected (in error) due to the excessive deformations
that are estimated since the effects of prestress are not accounted for.
Using the procedure suggested by the writers, a settlement of 34.9 mm is
calculated precompaction and 11.2 mm postcompaction. A detailed, step-
by-step presentation of settlement calculations using this procedure is
given in Appendix I.
Settlements of 22.3 mm precompaction and 11.7 mm postcompaction
were calculated using the Schmertmann (1986b) method. This method uses
a constrained modulus derived from ED-values that are uncorrected for
stress path and disturbance effects. In a predominantly NC profile, the
effect of using a constrained modulus in the case of an isolated footing
dominates, and smaller settlements than the writers' proposed method are
predicted. In a predominantly OC profile, the use of a constrained modulus
compensates for not correcting the moduli values, and the Schmertmann
(1986b) method predicts settlements of comparable value, although the

801

J. Geotech. Engrg., 1988, 114(7): 791-809


TABLE 2. Summary of Test Results—Post-Compaction Profile
z «c // u
v A B ED
(m) (MPa) (kPa) (kPa) (MPa) (MPa) (MPa) KD h qjo,',
(D (2) (3) (4) (5) (6) (7) (8) (9) (10)
2.00 16.84 0.0 36.86 0.60 1.98 48.0 14.83 2.53 456.99
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

2.20 17.84 0.0 40.82 0.72 2.07 46.9 16.33 2.03 436.89
2.40 17.54 0.0 44.69 0.54 1.89 46.9 10.93 2.77 392.46
2.60 16.15 0.0 48.55 0.54 1.84 45.5 9.90 2.73 332.65
2.80 16.15 0.97 51.52 0.46 1.65 41.5 7.88 2.95 313.46
3.00 16.15 2.92 53.41 0.43 1.62 41.2 7.01 3.17 302.41
3.20 15.16 4.86 55.39 0.46 1.76 45.5 7.16 3.31 273.70
3.40 14.66 6.80 57.37 0.45 1.68 43.0 6.76 3.19 255.61 .
3.60 12.78 8.75 59.25 0.51 1.88 48.0 7.40 3.15 215.72
3.80 14.27 10.69 61.24 0.56 1.99 49.8 8.06 2.91 233.01
4.00 16.15 12.63 63.12 0.70 2.31 56.3 9.85 2.61 255.89
4.20 15.16 14.58 65.10 0.70 2.38 58.8 9.46 2.75 232.88
4.40 17.04 16.52 67.08 0.80 2.66 65.3 10.50 2.67 254.06
4.60 18.03 18.46 68.96 0.82 2.57 61.4 10.55 2.43 261.49
4.80 19.92 20.41 70.95 0.76 2.67 67.1 9.28 2.94 280.73
5.00 21.80 22.35 72.93 0.86 2.51 57.8 10.54 2.17 298.91
5.20 24.67 24.29 74.81 0.84 2.55 59.9 9.94 2.32 329.80
5.40 19.92 26.24 76.79 0.74 2.55 63.5 8.30 2.87 259.35
5.60 19.02 28.18 78.67 0.90 2.84 68.6 10.01 2.51 241.81
5.80 14.66 30.12 80.95 0.93 3.22 81.2 9.85 2.93 181.15
6.00 13.77 32.07 83.23 0.96 3.22 80.1 9.94 2.79 165.48
6.20 13.28 34.01 85.41 0.97 2.79 64.3 10.03 2.16 155.45
6.40 12.78 35.95 87.39 0.82 2.31 52.0 8.27 2.07 146.26
6.60 13.28 37.90 89.28 0.90 2.43 53.4 8.94 1.93 148.72
6.80 10.40 39.84 91.56 0.99 2.95 69.7 9.37 2.34 113.64
7.00 3.77 41.79 93.54 0.81 2.38 55.3 7.46 2.28 40.25
7.20 3.77 43.73 95.72 0.99 2.58 56.0 9.12 1.85 39.34
7.40 3.77 45.67 97.70 0.65 1.76 38.7 5.71 2.00 38.54
7.60 10.90 47.62 99.38 0.34 0.97 21.3 2.65 2.34 109.67
7.80 11.89 49.56 101.17 0.38 1.01 21.3 2.97 2.05 117.53
8.00 15.16 51.50 102.95 0.44 1.57 39.4 3.24 3.40 147.26
8.20 15.66 53.45 104.83 0.67 1.98 45.9 5.35 2.36 149.34
8.40 15.66 55.39 106.81 0.52 1.82 45.9 3.74 3.31 146.57
8.60 13.28 57.33 108.60 0.41 1.59 41.2 2.72 4.02 122.26
8.80 15.66 59.28 110.48 0.50 1.77 44.8 3.42 3.42 141.70
9.00 20.91 61.22 112.46 0.64 2.18 54.2 4.55 3.05 185.90
9.20 18.03 63.16 114.44 0.66 2.49 64.7 4.50 3.62 157.58
9.40 17.04 65.11 116.33 0.72 2.72 71.2 4.84 3.64 146.51
9.60 17.04 67.05 118.31 0.76 3.20 87.0 4.90 4.33 144.05
9.80 20.91 68.99 120.49 0.96 3.55 92.5 6.38 3.47 173.52
10.00 21.80 70.94 122.77 0.94 3.08 76.2 6.26 2.86 177.56
Note: Dilatometer calibration data;
DA = 17.84 kPa; DB = 41.62 kPa (above 6.7 m); DA = 12.88 kPa; DB = 35.67 kPa (below 6.7 m); ZM = 9.91
kPa; water table at 2.70 m (data from J. H. Schmertmann, personal communication, 1987).

closeness of the agreement will depend on the estimated preconsolidation


pressures. Since in an OC deposit the settlements are generally small,
differences in predictions by the two methods are unimportant. However,
in an NC deposit, the difference is significant and may affect the design
decisions.
It is clear from the data presented in Table 4 and from the results of
numerous tests conducted in the laboratory under controlled conditions
that assessment of the effects of ground improvement using only penetra-
tion resistance data grossly underestimates the benefits achieved. In this
case, as in naturally OC deposits, it is essential to account for the effects
of prestressing. The dilatometer is a generally applicable, practical tool to
accomplish this purpose.
802

J. Geotech. Engrg., 1988, 114(7): 791-809


•o
o
o
ss

o
3
-o
o

3
J=*
±P-
s|=

»i>
F

*!>
31-

**-
sf
3$
S

3
2.00 4.26 35.77 0.16 0.91 26.2 3.62 5.83 119.11

O
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

2.20 5.25 39.44 0.23 1.3, 38.1 4.63 6.02 133.17

o
2.40 • 8.52 43.10 0.39 1.78 49.7 7.54 4.40 197.70
p
2.60 10.01 47.07 0.47 2.06 56.9 212.63

Oo
4.16
p

2.80 10.90 50.93 0.57 2.15 56.2 9.91 3.21 214.01


o

3.00 11.39 54.79 0.56 1.97 50.0 9.18 2.87 207.96


p

3.20 9.51 58.66 0.56 2.13 55.8 8.44 3.25 162.16


p

*-.
3.40 9.51 62.42 0.37 38.8 5.13 3.50 152.38
O

3.60 8.52 1.94 64.11 0.34 37.4 4.53 3.71 132.92

o
65.89 0.27

A.
3.80 8.03 3.89 30.5 3.47 3.85 121.80
4.00 6.14 5.83 67.68 0.24 34.5 2.83 5.19 90.78

IO
4.20 5.65 7.77 69.16 0.19 29.1 2.13 5.69 81.66

b
4.40 9.72 70.95 0.23 34.9 2.50 5.67 86.59

to
6.14
4.60 5.65 11.66 72.73 0.25 31.6 2.74 4.57 77.66
Ul
4.80 6.14 13.60 74.22 0.20 o 29.8 2.03 5.71 82.78
.5.55 76.00 0.25 34.5 2.52 5.19 105.61
UJ

5.00 8.03
5.40 9.51 19.43 79.47 0.35 42.7 3.48 4.45- 119.70
00
Ui

81.25 0.36 134.15


S3 NONJ

5.60 10.90 21.38 43.8 3.48 4.46


5.80 10.40 23.32 83.03 0.35 40.6 3.32 4.24 125.30
6.00 9.02 25.27 84.72 0.38 46.4 3.49 4.52 106.43
o

86.50 0.33
NO

6.20 6.64 27.21 33.0 3.03 3.62 76.75


29.15 88.29 0.28 34.1 2.37 4.69 59.48
~~i

6.40 5.25
6.60 5.25 31.10 89.97 0.26 32.3 2.11 4.89 58.37
= O

6.80 5.65 33.04 91.75 0.26 31.2 2.07 4.74 61.56


93.54 0.24
ON
ON

7.00 7.63 34.98 40.2 7.45 81.57


jo
O

95.22 0.31 41.3 2.33 5.37 109.26


C?

7.20 10.40 36.93


7.40 4.76 38.87 96.81 0.14 0.62 15.7 0.89 5.23 49.13
7.60 2.87 40.81 98.39 0.15 0.77 20.7 0.89 6.84 29.20
7.80 2.38 42.76 99.98 0.14 0.84 23.6 0.72 9.52 23.79
101.66 0.32 42.0 2.19 5.44 65.30
UJ

8.00 6.64 44.70


Ul

103.45 0.33
o\

8.20 7.63 46.64 42.7 2.22 5.36 73.75


ON

105.13 0.33 45.6 2.13 5.88 98.96


UJ

8.40 10.40 48.59


CO

106.91 0.38 45.6 2.54 4.85 97.31


ON

8.60 10.40 50.53


NO
bo

108.70 0.30 82.95


OO

UJ

9.02 52.47 37.7 1.85 5.40


o

ON

0.32
NO

110.38 45.3 6.26 103.23


NJ

9.00 11.39 54.42


00
~~J

9.20 10.90 56.36 112.17 0.36 49.6 5.95 97.17


£

0.41
NO

9.40 113.95 53.6 2.48 5.47 107.83


UJ

12.29 58.30
bo

9.60 10.01 60.25 115.63 0.42 50.7 2.54 4.96 86.55


-fc*

0.36
a*

9.80 10.40 62.19 117.42 44.9 2.05 5.37 88.61


S

be

119.20 0.44 48.5 87.28


NJ

10.00 10.40 64.14 2.63 4.46


ill
Z

III

IS'
J D S

" >
3 O
-

II
3

3
30
0}

CO
nr-
m

-n
m

T?
(0
>
3

H
H
z

o £•'

£0 ft"
O 3"

ci cr.

P'
CL
=r 3-

5. a SS Cft L

en r^.

° 2

i- ? i. § 1 .
«-. ^

CL 2
p ft
s- *g2-ft 23 3

C
S"S
S
an

O <*> ft <
£> o ft

1 TO

PS X !

***< 3 «
->

o X
0 00

>-* o
CL ft 2- c

CL p
E.3
CL en

-
i-b ft

• r t en ft

rt

ft
c

? £ c-*f T '-5"
en S»

°>en
en
ST « T3
;<»
3

ft

h 3

3
^j
<T 3 ft'

ft ££•
ft ^ .
ft 4 , ft <-*•

*hi
Pt-
pre. o <*

S. 3
a. ^
ET pj

X O
ps
rf

*-w
o
ft

en
o
to

<£.

5
es
3"

en

" 5
l 1

en
en '-I

ft
~ ft

o
tt &• o
p

ft

ft
ft

O
3
i-i»

~> CL
3- S-

°*£
3 W

C
en' p
•< 3
TO O

TO O
en O

ft ft
en

pj ps
•— O
ft

<'§

3
&

05
ft

3
ft
rt

3
>
<•)

oo
00
o

J. Geotech. Engrg., 1988, 114(7): 791-809


TABLE 4. Summary of Calculated Settlements
Pre-compaction Post-compaction
settlement settlement
(predominantly (predominantly
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

NC profile) OC profile)
Method (mm) (mm)
(1) (2) (3)
In situ modulus interpreted solely from
penetration resistance, i.e., E - 2qc for entire
stress increment and using Schmertmann
(1970) method 77.6 43.8

Deposit partly in OC and partly in NC range as


determined by Leonards/Frost method; E =
3.5 ED in OC portion; and E = 0.7 ED in NC
portion 34.9 11.2

Schmertmann (1986b) method using constrained


dilatometer modulus (ordinary method) 22.3 11.7

where ACT,, = change in vertical stress; Az = thickness of the layer; and M


= constrained modulus.
Calibration chamber test results suggest that the use of M= 1.3 MD for
the NC case and M = 2.4 MD for the OC cases (Baldi et al. 1986a) is
reasonable; however, the importance of additional research to develop
more appropriate factors is stressed. The constrained dilatometer modulus
MD can be obtained as suggested by Marchetti (1980) as follows:
MD = RMED (14)
where RM is obtained as a function of ID and KD according to the following
simplified expression, for ID > 1.8:
RM = 0.5 + 2.0 log KD (15)
RM is always > 0.85. If ID < 1.8 (clayey soils), the procedure discussed
herein is not applicable. While the scatter in these correlations is consid-
erable, they are still considered to be of practical value in cases were
settlements occur approximately under I-D compression.

CONCLUSIONS

1. A conceptual framework for understanding the effects of overconsol-


idation in reducing the compressibility of all types of soils has been
presented. The stress range within which the soil remains overconsoli-
dated is dependent on the stress path, both during unloading and reloading,
with respect to the location of the yield locus. This concept can be applied
usefully to the interpretation of all types of field and laboratory tests.
2. The advantage of including dilatometer tests as a routine part of site
investigations in granular soils has been outlined. Using correlations with
calibration chamber test results, the parameters obtained from the dilatom-
eter can be corrected for the effects of stress path and of disturbance due
804

J. Geotech. Engrg., 1988, 114(7): 791-809


to inserting the instrument to obtain in-situ values of compressibility of
both normally consolidated and overconsolidated sands. Whether the
preconsolidation pressure p{. is determined by evaluating ^T(OC) using the
Marchetti (1985) or Schmertmann (1983) procedures, the dilatometer is the
most practical tool presently available to evaluate p'v at depth in granular
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

soils. Moreover, as dictated by the boundary conditions, either Young's


modulus or the constrained modulus can be evaluated.
3. A method for calculating the settlement of shallow foundations
supported on granular soils has been proposed. The procedure is applica-
ble both to normally consolidated or overconsolidated sands and can deal
with isolated footings or rafts, deposits of limited thickness, variability in
the soil profile or depth of embedment of the footing, and fluctuations in
the water table that may occur subsequent to the time the site investigation
was carried out. Thus, it is believed that the new method is sounder in
principle than any others that have been proposed to date.
4. Although the correlations used to account for stress path and
disturbance effects are at present limited in scope, it is better to use
approximate correction factors than to ignore these effects entirely,
particularly in the case of overconsolidated sands at lower relative
densities. Moreover, the results of future research can readily be incorpo-
rated into the procedure to improve the precision of the calculated
settlements.
5. It is common practice to assess the effects of ground improvement
techniques, e.g., rolling in layers, dropping heavy weights, or detonating
explosives, using only "before" and "after" penetration resistance data.
This practice, although valid from the standpoint of drained shear strength,
can grossly underestimate the benefits achieved from the standpoint of
controlling settlements.

ACKNOWLEDGMENTS

The writers wish to express their gratitude to J. H. Schmertmann, M.


Jamiolkowski, R. Lancellotta, and the ASCE journal reviewers whose
constructive criticisms challenged the writers to improve the clarity of the
paper. Special thanks are due to J. H. Schmertmann who provided the
unpublished data listed in Tables 2 and 3.

APPENDIX I. STEP-BY-STEP SETTLEMENT CALCULATION

1. Perform DMT and CPT soundings at appropriate locations through


soil layers of interest.
2. Divide the soil profile into layers with similar characteristics for
settlement calculations. In this sample calculation, layer 2 (2.5-3.0 m) is
selected. The test data for this layer is shown set off in Table 3.
3. Determine the average qch'v ratio and the KD value for each layer.
For layer 2, average qja'v = 211.5 and average KD = 9.16.
4. Use Eq. 7 to determine K(OQ. For layer 2, A"(OC) = 0.376 +
0.095(9.16) - 0.0017(211.5) = 0.89.
5. Using (for convenience here) the chart prepared by Marchetti (1985)
from the Durgunoglu and Mitchell (1975) equations, determine §ps (Fig. 3).
For layer 2, K(OC) = 0.89; qcla'v = 211.5 = 41.5°.
805

J. Geotech. Engrg., 1988, 114(7): 791-809


6. Calculate the value of <j>nv from Eq. 8. For layer 2, $ax = 41.5 - (41.5
- 32)/3 = 38.3°.
7. Use Eq. 10 to determine OCR for the layer. For layer 2, A:(OC) =
0.89; <|>„ = 38.3°
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

0.89
OCR = . . , - . . [1/0.8 sin (38.3)] = 5.56 (16)
1 - sin (38.3)
8. Calculate the initial vertical effective stress at the center of the layer.
For layer 2, u'v = 49.5 kPa.
9. Determine the preconsolidation pressure/?'cat the test elevation. For
layer 2, a'v = 49.5 kPa; OCR = 5.56p'c = 275.5 kPa.
10. Determine the stress increase at the center of the layer due to the
applied load. For layer 2, ACT = 351.8 kPa (based on assumed 2VAH
distribution).
11. Determine the final stress at the center of the layer: a'f = u'v + ACT.
For layer 2, <r} = 49.5 + 351.8 = 401.3 kPa.
12. Determine what portion of the load increment will be in the OC
range [Rz (OC)] and the NC range [Rz (NC)] using Eqs. 11 and 12. For layer
TABLE 5. Summary of Example Settlement Calculation
Depth to
middle
Layer of layer °"i'' "f Pi- A. £, Sj

number (m) (kPa) (kPa) qja[, K (kPa) «,(OC) fl,(NC) (m) (MPa) (mm)
Kn
(10)
/, (12) (13) (14)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (11)
1 2.25 40.3 479.1 5.26 150.0 0.62 104.9 0.15 0.85 0.075 0.5 134.2 (OC) 0.59
26.8 (NC)
2 2.75 49.9 401.2 9.16 211.5 0.89 275.8 0.64 0.36 0.225 0.5 192.0 (OC) 0.69
38.4 (NC)
3 3.25 59.7 347.2 7.58 174.2 0.80 263.6 0.71 0.29 0.375 0.5 170.4 (OC) 1.14
34.1 (NC)
4 3.75 65.5 305.2 3.61 115.2 0.52 115.9 0.21 0.79 0.525 0.5 120.6 (OC) 4.35
24.1 (NC)
5 4.25 69.7 272.5 2.49 86.3 0.47 92.3 0.11 0.89 0.575 0.5 116.0 (OC) 5.42
23.2 (NC)
6 4.75 73.8 247.7 2.43 88.7 0.46 93.6 0.11 0.89 0.525 0.5 113.0 (OC) 5.08
22.6 (NC)
7 5.25 78.2 229.0 3.00 112.7 0.47 110.5 0.21 0.79 0.475 0.5 136.4 (OC) 3.47
27.3 (NC)
8 5.75 82.6 214.6 3.43 122.0 0.49 132.3 0.38 0.62 0.425 0.5 153.9 (OC) 2.31
30.8 (NC)
9 6.25 87.0 203.4 2.96 80.9 0.52 142.4 0.48 0.62 0.375 0.5 133.5 (OC) 2.08
26.7 (NC)
10 6.75 91.4 194.9 1.95 67.2 0.45 103.6 0.12 0.88 0.325 0.5 122.1 (OC) 2.89
24.4 (NC)
88.1 1.00 0.275 0.5 29.4 (NC)
11 7.25 95.6 188.3 1.63 80.0 0.39 — 2.24

12 7.75 99.6 182.9 1.27 39.4 0.43 89.9 - 1.00 0.225 0.5 26.1 (NC) 2.07

13 8.25 103.8 179.3 2.18 79.3 0.45 123.6 0.26 0.74 0.175 0.5 153.5 (OC) 1.08
30.7 (NC)
14 8.75 108.3 176.9 2.09 94.5 0.41 114.6 0.09 0.91 0.125 0.5 151.4 (OC) 0.92
30.3 (NC)
15 9.25 112.7 175.3 2.17 102.7 0.41 117.1 0.07 0.93 0.075 0.5 174.8 (OC) 0.49
35.0 (NC)
16 9.75 117.0 174.4 2.41 87.5 0.46 148.4 0.55 0.45 0.025 0.5 169.7 (OC) 0.10
33.9 (NC)
Note: Total settlement - 34.9 mm.

806

J. Geotech. Engrg., 1988, 114(7): 791-809


2, p'c = 275.5 kPa; u'v = 49.5 kPa; a} = 401.3 kPa; ^ ( O C ) = 0.64; RZ(NC)
= 0.36.
13. Determine the average ED-value for the layer. F o r layer 2, average
ED = 54.9 Mpa.
14. Determine the strain influence factor Iz for the layer from Schmert-
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

mann's 5/2:2/? approximation. F o r layer 2, Iz .= 0.225.


15. Calculate the settlement within the layer using E q . 4. F o r layer 2, ct
= 0.96 (embedment correction); qnel = 500 kPa; Az = 0.5 m (height of
layer); Ez(OC) = 3.5; ED = 192.0 MPa; £ Z (NC) = 0.7; ED = 38.4 MPa; / ,
= 0.225; s2 = (0.96)(500)(0.225)(0.5) [(0.64/192.0) + (0.36/38.4)] = 0.7 mm.

Summarized settlement calculations for all layers are given in Table 5.

APPENDIX II. REFERENCES


Baguelin, F., Jezequel, J. F., and Le Mehaute, A. (1974). "Self-boring placement
method of soil characteristics measurement." Proc. ASCE Specialty Conf. on
Subsurface Exploration for Underground Excavation and Heavy Construction,
Henniker, N.H., 312-332.
Baldi, G., et al. (1986a). "Flat dilatometer tests in calibration chambers." Proc.
ASCE Specialty Conf on Use of In-Situ Tests in Geotechnical Engineering,
Blacksburg, Va., 431-446.
Baldi, G., et al. (1986b). "Interpretation of CPT's and CPTU's: Part II—Drained
penetration of sands." Proc. of IV Int. Geotechnical Seminar, Singapore.
Bellotti, R., et al. (1985). "Laboratory validation of in-situ tests." Geotechnical
engineering in Italy—an overview. Associazione Geotecnice Italiana, Rome,
Italy, 251-270.
Bellotti, R., et al. (1986). "Deformation characteristics of cohesionless soils from
in-situ tests." Proc. ASCE Specialty Conf. on Use of In-Situ Tests in Geotech-
nical Engineering, Blacksburg, Va., 47-73.
Brumund, W. F., and Leonards, G. A. (1973). "Experimental study of static and
dynamic friction between sand and typical construction materials." ASTM J.
Test. Eval., 1(2), 162-165.
Durgunoglu, H. T., and Mitchell, J. K. (1975). "Static penetration resistance of
soils—Parts I and II." Proc, ASCE Specialty Conf. on In-Situ Measurement of
Soil Properties, Raleigh, N.C., 151-189.
Handy, R. L., et al. (1982). "In-situ stress determination by Iowa stepped blade."
Geotech. Engrg. Div., 108(GT11), 1405-1422.
Hendron, A. J. (1963). "The behavior of sand in one-dimensional compression,"
thesis presented to the Dept. of Civil Engineering, at the University of Illinois, at
Urbana, 111., in partial fulfillment of the requirements for the degree of Doctor of
Philosophy.
Jamiolkowski, M., et al. (1985). "New developments in field and laboratory testing
of soils." Theme Lecture No. 2, Proc. XI Int. Conference on Soil Mech. and
Foundation Engrg., San Francisco, Calif., Vol. 1, 57-153.
Kummeneje, O. (1956). "Foundation of an oil tank in Drammen." Norwegian
Geotech. Inst. Publication No. 12, Oslo, Norway.
Lade, P. V., and Lee, K. L. (1976). "Engineering properties of soils." Report
&UCLA-ENG-7652, Univ. of California, Los Angeles, Calif.
Lambrechts, J. R., and Leonards, G. A. (1978). "Effects of stress history on
deformation of sand." / . Geotech. Engrg. Div., 104(GT11), 1371-1378.
Leonards, G. A. (1975). Discussion of "Session I—Granular materials." Proc.
Conf. on Settlement of Structures, Halsted Press, New York, N.Y., 669.
Leonards, G. A. (1985). Discussion of "New Developments in Field and Labora-
tory Testing of Soils," by M. Jamiolkowski, et al. Proceedings, XI Int. Confer-
ence on Soil Mech. and Foundation Engrg., San Francisco, Calif., Vol. 5.
Marachi, N. D., et al. (1981). "Plane-strain testing of sand." Amer. Society for
Testing and Materials STP 740, Baltimore, Md., 294-302.

807

J. Geotech. Engrg., 1988, 114(7): 791-809


Marchetti, S. (1975). "A new in-situ test for the measurement of horizontal soil
deformability." Proc. ASCE Specialty Conf. on In-Situ Measurement of Soil
Properties, Raleigh, N.C., Vol. II, 255-259.
Marchetti, S. (1980). "In-situ tests by flat dilatometer." J. Geotech. Engrg. Div.,
ASCE, 106(GT3), 299-321.
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

Marchetti, S. (1985). "On the field determination of K0 in sand." Discussion


Session No. 2A, XI Int. Conference on Soil Mech. and Foundation Engrg., San
Francisco, Calif.
Mayne, P. W., and Kulhawy, F. H. (1982). "K0-OCR relationships in soil." J.
Geotech. Engrg. Div., ASCE, 108(GT6), 851-872.
Mitchell, J. K., and Durgunoglu, H. T. (1973). "In-situ strength by static cone
penetration test." Proc. VIII Int. Conference on Soil Mech. and Foundation
Engrg., Moscow, U.S.S.R., Vol. 1.2, 279-286.
Robertson, P. K., and Campanella, R. G. (1983). "Interpretation of cone penetra-
tion tests—Part 1—Sand," Can. Geotech. J., 20(4), 718-733.
Rowe, P. W. (1969). "The relation between the shear strength of sands in triaxial
compression, plane strain and direct shear." Geotechnique, 19(1), 75-86.
Rowe, P. W. (1975). Discussion of "Session I—Granular materials." Proc, Conf.
on Settlement of Structures, Halsted Press, New York, N.Y., 670.
Schmertmann, J. H. (1970). "Static cone to compute settlement over sand." / . Soil
Mech. and Found. Engrg., ASCE, 96(SM3), 1011-1043.
Schmertmann, J. H. (1981). "A method for determining the friction angle in sands
from the Marchetti dilatometer test." Proc. European Symposium on Penetra-
tion Testing II, Amsterdam, Netherlands, 2, 853-861.
Schmertmann, J. H. (1983). "Revised procedure for calculating K0 and OCR from
DMT's with I.D. > 1.2 and which incorporate the penetration force measurement
to permit calculating the plane strain friction angle." DMT Digest # 1 , GPE Inc.,
Gainesville, Fla.
Schmertmann, J. H. (1984). "The new in-situ Marchetti dilatometer test." Geotech.
News, 2(3), 34-35.
Schmertmann, J. H. (1986a). "Suggested method for performing the flat dilatometer
test." Geotech. Testing J., ASTM, 9(2), 93-101.
Schmertmann, J. H. (1986b). "Dilatometer to compute foundation settlement."
Proc. ASCE Specialty Conf. on Use of In-Situ Tests in Geotechnical Engineer-
ing, Blacksbmg,Va., 303-32\.
Schmertmann, J. H., Hartman, J. P., and Brown, P. R. (1978). "Improved strain
influence factor diagrams." J. Geotech. Engrg. Div., ASCE, 104(GT8), 1131-
1135.
Terzaghi, K., and Peck, R. B. (1967). So/7 mechanics in engineering practice, 2nd
Ed., John Wiley and Sons, New York, N.Y.
Uesugi, M., and Kishida, H. (1986). "Influential factors of friction between steel
and dry sands." So/7* Found., 26(2), 33-46.
Wroth, C. P., and Hughes, J. M. O. (1973). "An instrument for the in-situ
measurement of the properties of soft clays." Proc. VIII Int. Conference on Soil
Mech. and Foundation Engrg., Moscow, U.S.S.R., Vol. 1.2, 487-494.

APPENDIX III. NOTATION

The following symbols are used in this paper:


A = dilatometer lift-off pressure;
B = dilatometer expansion pressure;
CC = calibration chamber;
C
\ ; C2 > C
3 = empirical coefficients;
D = thickness of granular stratum;
Dr = relative density;
E = Young's modulus;
ECc = Young's modulus measured in calibration chamber;
808

J. Geotech. Engrg., 1988, 114(7): 791-809


ED = dilatometer modulus;
Ez = Young's modulus at depth z;
E25 = Young's modulus at" 25% of failure stress measured in
axial compression;
h = strain influence factor at depth z;
Downloaded from ascelibrary.org by MVGR College of Engineering on 02/25/19. Copyright ASCE. For personal use only; all rights reserved.

ID = dilatometer material index;


K = ratio of horizontal to vertical effective stress;
= dilatometer horizontal stress index;
= ratio of horizontal effective stresses after unloading in
given stress path to its value prior to unloading;
K0 = coefficient of earth pressure at rest;
M = constrained modulus;
Mcc = constrained modulus measured in calibration chamber;
MD = constrained dilatometer modulus;
NC = normally consolidated;
OC = overconsolidated;
OCR = overconsolidation ratio;
OCRH = overconsolidation ratio with respect to horizontal
stresses;
OCRV = overconsolidation ratio with respect to vertical stresses;
= corrected dilatometer A pressure;
Pi = corrected dilatometer B pressure;
P' = (aJ + CTJ)/2 stress path parameter;
P'c = preconsolidation pressure;
q = (a, - CT3)/2 stress path parameter;
= cone penetration resistance;
<iD = dilatometer penetration resistance;
Inet = net increase in contact pressure due to footing;
RE = ratio E25/ED;
R
M = ratio M\JED;
RJNQ = ratio of stress increment corresponding to NC portion to
total stress increment;
7?j.(OC) = ratio of stress increment corresponding to OC portion to
total stress increment;
S = total settlement;
Si= settlement of layer /;
«o = in situ hydrostatic pore pressure;
A, = height of layer;
= change in vertical stress;
8 = angle of wall friction;
= horizontal effective stress;
< = vertical effective stress;
= initial vertical effective stress;
= angle of shearing resistance;
= <)) measured in axial compression; and
= <>
j measured in plane strain.

809

J. Geotech. Engrg., 1988, 114(7): 791-809