Environmental Engineering Reference
In-Depth Information
particularly if the earth core of the dam is narrow and the rockfill compressible giving lit-
tle lateral constraint.
The compaction process typically overconsolidates the soil to at least 200 kPa, but this
depends on the compaction equipment, soil and the compaction water content. Hunter
(2003) and Hunter and Fell (2003d) determined from internal settlement data that for
earthfill cores placed generally 0.5 to 1% dry of standard optimum moisture content, the
preconsolidation pressure was from 200 to 400 kPa for clayey earthfills, and 700 to
1,000 kPa for silty gravel, silty sand and clayey sand.
The theory and practice of partially saturated soil behaviour is rapidly developing and
readers should seek the latest information from the literature. Some useful references
include Alonso et al. (1990), Bardin and Sides (1970), Khalili and Khabbaz (1998) and
Loret and Khalili (2002).
6.2.1.2.2 Filters and sand-gravel zones
These will, if well compacted, be stiffer (have a higher modulus E) than earthfill and are
usually stiffer than rockfill. The modulus will depend on the grading and degree of com-
paction. Because they are stiff, they may attract load by arching of the core. Some data on
the modulus of compacted “gravel” fill in concrete face rockfill dams is given in Section
15.2.4 and this can be used to estimate the modulus of filters.
6.2.1.2.3 Rockfill
The rockfill compressibility is dependent on the grading, degree of compaction (layer
thickness, roller weight and number of passes), rock substance strength and the effect of
wetting on the substance strength. The modulus is best estimated from the performance
of similar dams and is discussed in detail in Section 15.2.4.
The rockfill upstream of the core of the dam is subject to wetting by the reservoir and to
cyclic loading as the reservoir levels fluctuate. Some rockfills are subject to “collapse set-
tlement” due to softening of the rock substance as they are wetted. Rockfill is likely to dis-
play a lower equivalent modulus than on previous drawdowns of the reservoir if the
reservoir is drawn down below the previous lowest level, that is, the rockfill displays both
over and normally consolidated behaviour.
6.2.2
Methods of estimating the compressibility of earthfill, filters and rockfill
6.2.2.1
Using data from the performance of other dams - earthfill
Generally speaking, it will be sufficient to estimate the settlement of the embankment
earthfill during and after construction, using the results of monitoring of other dams. G.
Hunter (2003) and Hunter and Fell (2003d), has gathered data from a large number of
dams, where the core has been generally compacted to a minimum density of 98% of stan-
dard maximum dry density, with a water content in the range 2% dry to 1% wet of stan-
dard optimum water content. Most of the cores are “dry placed”, with a moisture content
0.5 to 1% dry of standard optimum moisture content.
(a)
Settlement during construction
Figure 6.39
shows total settlements of the core during construction of earth and earth
and rockfill embankments. These have been obtained from monitoring of cross arm
settlement gauges embedded in the dam core near the dam centreline.
Tables 6.5
and
6.6
give the equations of best fit for the data.
(b)
Strains and equivalent moduli during construction
Figure 6.40
and
Table 6.7
shows the vertical strains versus effective vertical stress
(ignoring suction effects) and estimated confined secant moduli in dry placed earthfill
cores at the end of construction. The confined secant moduli (M) can be converted to
Youngs moduli using equation 6.30. The Poissons ratio are likely to be in the range
0.25 to 0.35 for dry placed earthfill.
