Concrete face rockfill dams - Geotechnical Engineering of Dams - page 611

It should be noted that E rf is a pseudo modulus, only to be used in estimates of face slab
deformation using the simplified method shown in Figure 15.10 but allowing for the
stress distribution using the Poulos and Davis (1974) method. For finite element analyses
the modulus should be obtained as detailed in Section 15.2.4.1.
The reasons for a number of the outliers on Figure 15.12 are detailed in Hunter and Fell
(2002, 2003c). They mostly relate to complications arising from complex zoning (Ita,
Crotty, Scotts Peak) and localised narrow section of the valley causing arching (Khao
Laem). The apparently higher modulus on first filling implied by high E rf /E rcc values is
due to the actual stress paths in the rockfill giving an initial reduction in deviator stress,
as filling begins, and an overall relatively small increase in deviator stress on filling, as
well as the simplifying assumptions made in the analysis.
Numerical analyses were carried out to assess the hypothesis put forward by Cooke
(1984) that the layer of rockfill would have a higher horizontal modulus than vertical due
to the high degree of compaction in the upper part of each layer, but these showed this
was not the case.
The steps to estimate E rf are:
(a) Estimate the ratio E rf /E rcc based on the embankment height and upstream slope angle
using Figure 15.12. Trendlines are given for upstream slopes of 1.3-1.4 H to 1 V, and
1.5 H to 1 V, which cover most CFRD designs.
E rcc is the E rc value estimated from Figure 15.11 , adjusted for vertical stress such
that it is representative of the average vertical stress in the lower 50% of the rockfill in
the central region of the embankment.
(b) Estimate E rc , the representative secant moduli at end of construction, of the Zone 3A
rockfill from the methods outlined above. The E rc value should be adjusted for verti-
cal stress such that it is representative of the lower 50% of the rockfill in the central
region of the embankment.
(c) The E rcc values used in the derivation of Figure 15.12 were not corrected for arching
effects due to valley shape because valley shape is potentially likely to affect both E rc
and E rf , and would therefore be taken into consideration in the E rf /E rcc ratio. Hence
E rc estimates derived from Figure 15.11 must be adjusted by dividing by the stress cor-
rection factors in Table 15.7 to give E rcc .
(d) E rf can then be estimated by multiplication of the value with the E rf /E rcc ratio.
The method applies to compacted rockfill. There is insufficient data on gravel fill
dams to draw a trendline, but it appears that E rf /E rcc may be smaller for gravel fill than
rockfill.
As will be apparent from the scatter of data in Figure 15.12, this method is approxi-
mate. It is most likely to be in error where the dam has complex zoning, with varying
moduli in the zones, and in situations where valley shape effects are significant. The trend-
line for the upstream slope of 1.5 H:IV is particularly uncertain, being based on a limited
number of cases.
15.2.4.3 Effect of valley shape
Several authors including Pinto and Filho Marques (1998), and Giudici et al. (2000) have
concluded that the valley shape can have a significant effect on the settlements, because of
3D effects shedding vertical stresses to the valley sides. This has been assessed empirically,
aided (for Guidici et al., 2000) with 3D numerical analyses.
Hunter (2003), Hunter and Fell (2002) assessed this along with the other factors, and
used simple 2D finite difference models to assess the likely effects of valley shape on cal-
culated moduli. This showed that the effects were probably not as great as assumed by