Geoscience Reference
In-Depth Information
for individual plots on Figure 8.17 are adjusted based on the slopes of regression curves
to evaluate
R
L
20
at the target relative densities
D
r
=
20%-90%. The stress ratio
R
L
20
increases with increasing
Cu
for
D
r
50% and 90% while it stays almost constant
for other
D
r
. Hence, it may be said that the undrained cyclic strength is not so much
dependent on
Cu
or soil particle gradation in contrast to its large dependency on
D
r
.
Data points by Tanaka et al. (1987) in Figure8.18 indicate that the trend is essentially
the same although there exists one distinct separation. Thus, undrained cyclic strength
defined by the stress ratio attaining 5% DA strain is strongly dependent on the relative
density despite large difference in absolute density due to the difference in particle gra-
dation. In other words, liquefaction strength (normally defined by 5% double amplitude
strainornearly100%pore-pressurebuildup)isnotsomuchsensitivetotheabsoluteden-
sity but more dependent on the relative density. Combining thisfinding with the fact that
N
-values of well-graded gravels are considerably larger than that of sands of the same
D
r
for
D
r
>
=
50%asshowninFigure8.11,thereasoncanbeexplainedwhytherelation-
ship between liquefaction strength versus SPT
N
1
-value of well-graded gravels largely
deviates fromthat of poorly-graded sands for
N
1
>
25-30.
3.4. EFFECT OF FINES CONTENT
As indicated from previous case histories, liquefied gravelly soils sometimes contained
measurable fine soil particles smaller than 0.075mm. It was demonstrated for poorly-
graded sands by quite a few laboratory tests that liquefaction strength clearly decreases
withincreasing content of lowplasticity fines under aconstant relative density (e.g. Sato
et al., 1997). In order to examine the effect of fines content on undrained cyclic strength
of well-graded soils and compare it with poorly-graded sands, a series of tests have been
conducted bymixing fines with RS1 and RS3 as depicted inFigure 8.13.
Figure8.19exemplifiestypicalrelationshipsbetweenthecyclicstressratio,
R
L
forattain-
ing 5% DA strain, and the number of loading cycles
N
L
for RS1 and RS3 of relative
density
D
r
50%withfinescontent
Fc
changingstepwisefrom0%to30%.Thoughthe
datapointsshowlargescattersforsmaller
Fc
forRS3inparticular,thestrengthobviously
tends to decrease with increasing
Fc
. It can be pointed out that the slopes of
R
L
versus
N
L
curves are apparently steeper for well-graded RS3 than for poorly-graded RS1 with
various fines content. Also noted is that the decrease in the strength is drastic for initial
small increase of fines content from 0% to 10% particularly in RS3 while the change
occurs more gradually inRS1.
≈
InFigure8.20,thestressratiosfor5%DAstraincorrespondingto
N
L
=
20
,
R
L
20
(
DA
=
5%
,areplottedversus
Fc
forRS1(smallersymbols)andRS3(largersymbols)ofrelative
densities
D
r
)
0% correspond to the
data shown in Figure 8.17. Despite some data dispersions, liquefaction strength defined
by
R
L
20
(
≈
30%, 50% and 70%. Solid symbols at
Fc
=
obviously decreases with increasing
Fc
particularly for larger
D
r
not only for RS1 (poorly-graded sand) but also for RS3 (well-graded soil). The strength
reduction may be somehow attributable to the role of fines as lubricators to decrease
positive dilatancy, though the exact mechanism isnot yet clarified.
DA
=
5%
)
