Environmental Engineering Reference
In-Depth Information
unconfined compressive strength of the rock substance (MPa).
This function resulted in a variance explained of 61.7%. The addition of the uniformity
coefficient, c
u
and the maximum and minimum particle sizes, d
max
and d
min
, did not result
in a better fit to the data.
Douglas (2003) also carried out an analysis of the database using the principal stresses
UCS
1
. This showed an equation of the form shown in Equation 6.33 was found to be
the most effective for relating
and
3
1
to
3
.
RFI
(6.33)
′
′
1
3
The analysis resulted in the following equations (based on 869 data sets and with a
variance explained
98.8%):
0.8726
(6.34)
RFI
6.3491RFI
0.48763RFI
0.0027RFI
1.1568RFI
e
ANG
d
max
FINES
0.30598RFI
(6.35)
UCS
1
1e
(6.36)
RFI
e
i
RFI
1 if angular, otherwise
0
(6.37)
ANG
RFI
d
(mm)
(6.38)
d
max
max
e
1e
(Fines 20)
(Fines
(6.39)
RFI
where fin
es is in %
FINES
20)
e
1e
(UCS 110)
(UCS
(6.40)
RFI
where UCS is
in MPa
UCS
110)
For many analysis of stability it will be sufficient to use the data in
Figure 6.37
and
equations 6.29 to 6.32 to select the strength of well compacted rockfills. A bi-linear
strength envelop should be used to better represent the curved Mohr-Coulomb envelope.
This can be plotted by calculating
sec
from equations 6.29 to 6.32 at varying
n
applica-
ble to the stability analysis.
Loose dumped rockfill at low
n
will have
sec
about 4 degress lower than calculated
using this approach. At high
n
the difference will be negligible. For more important proj-
ects, equations 6.33 to 6.40 should be used, to give the relationship between
1
and
3
. To
convert this to the more familiar c
,
, use the method described in Section 16.3.2.5,
Equations 16.16 to 16.18.
If the rockfill has high percentages of finer particles (say greater than 25-30%, passing
2 mm), it would be wise to carry out triaxial tests, using scaled coarse fraction particle size
distribution, but retaining the same percentage passing 2 mm.
For very large dams, it may be necessary to carry out triaxial tests on scaled gradings of
the rockfill to confirm the strength, but only if the strength is critical (which in reality it
seldom is).
