Design of the Cua Dat CFRD
in Vietnam,
and innovative
river diversion ×
M. Ho Ta Khanh (×) and his
system Vietnamese colleagues at the dam site
Michel Ho Ta Khanh (Paris) Giang Pham Hong (Hanoi)
(Some slides presented at “ASIA ‘2006 Symposium”, Dec. 2006)
Design of the Cua Dat CFRD in Vietnam,
and innovative river diversion system
Michel Ho Ta Khanh Giang Pham Hong
Vietnam:
83 000 000 people
331 000 km2 territory
(75% mountains & hills)
Mean annual rainfall: 2 000 mm
(70% in three months of the rainy
season)
flood drought
flow redistribution
Dam Development
Program
Cua Dat is a multipurpose scheme under construction in Vietnam. The main dam will
be the highest CFRD in Vietnam (H= 119 m, L= 930 m, V = 13 hm3). The Design flood
(1/1 000) is 13 200 m3/s and the Check flood (1/10 000) is 18 900 m3/s. The dam will be
constructed in 5 years (2004-2009). The 20 year-flood, adopted as the Construction
flood, has a peak inflow discharge of 1 230 m3/s during the dry season and 5 050 m3/s
during the wet season.
To minimize the cost of the diversion structures, it was admitted to divert the flow during
the most critical stage of the construction (wet season of 2007) by only one tunnel (L =
802 m, D=9 m) with an overtopping of the upstream and downstream cofferdams and of
the main dam about 20 m higher than the river bed.
Extensive laboratory tests (scale 1/40), taking into account the zoning and materials of
the three structures, have been performed in order to check their stability and to
measure the pressure and velocity of the flows at different parts of the structures.
Various alternatives have been tested : protection by concrete slabs, RCC, gabions or
downstream stepped rockfill for different dam crest and water levels depending on the
construction schedule and probability of floods. These tests have permitted to verify the
resistance of the dams and to define precisely the most suitable protections for their
crest, upstream/downstream faces and toe during the overtopping, with their respective
advantages, drawbacks and costs.
This presentation provides some results of this study. Finally, the tests performed
indicate that this solution leads to significant cost and delay savings compared with
conventional method (high upstream cofferdam with several large diameter tunnels) and
is therefore very promising for the future rockfill dams constructed in Asia, on large rivers
with a marked seasonality regime
(Viewers can see something about the Cua Dat Dam in details in this website (“Dams in
Vietnam”) or directly http://www.vncold.vn/En/Web/Content.aspx?distid=231 )
Chu river in the dam area in dry season
Flood Control
Reservoir 8 m3/s discharge
storage: for industry
1 450 hm3
Dat tributary
Water supply
source for
2.5 mil.people
Chu river
Auxiliary dams
87 000 ha
irrigation
97 MW
hydropower Headworks : Previously
(main CRFD, spillway, suggested
tunnels, dam location
hydropower plant)
Project area plan
Spillway
Reservoir
Power
house
Main CRFD
Bridge
Chu
river
3D Model of Cua Dat Headwork System
Cua Dat CFRD cross section
Some features about the critical phase of the diversion
• Inflow peak discharge ≅
5 050 m3/s
• Tunnel discharge
variable with the MWL
( ≅ 700 m3/s)
• Overflow discharge
variable with the MWL
( ≅ 4 400 m3/s)
• River bed W.L ≅ 28 m
• U/S WL, Dam crest L,
MWL on the dam =
variable according to
the tests.
• D/S WL = 38.1 m
Cros section of the cofferdam &
the CFRD crest is at level 45.0 m and the MWL on the dam is 50.0 m.
Purpose of the Laboratory Tests
(scale 1/40)
• To determine the flow conditions over the cofferdams and the
CFRD under construction for different flood discharges and
dam crest levels.
• To explore water velocities field and depths along various
profiles of the reach between U/S and D/S cofferdams.
• To explore water pulsation at critical locations.
• To work out solutions to protect main dam as well as
cofferdams.
• To examine eventual action of erosion.
• To adjust dimensions of diversion works, if necessary.
(1~2) Alternatives of downstream slope face protection
by concrete slabs or gabion mats when overtopping
View (from upstream) of downstream concrete slabs protection
View (from upstream) of downstream gabion mats protection
Note that in these tests, the required resistances of the concrete slabs or gabion mats
are not representative of the prototype. They are determined only by the values of the
water velocities and pressure fluctuations measured on the models.
Lower part of CFRD under construction
Detail on erosion at the end of the protection layer after the flood. The maximum
water levels and velocities (max = 8.2 m/s) are indicated on the figure.
(3) Alternative of rock fill without protection.
Lower part of CFRD under construction Hydraulic jump
This alternative was contemplated in order to avoid an expensive protection layer.
The steps are 2.4 m high and 24 m large (1.6 m and 16 m respectively for alternative
2A). The maximum size of the rock fill is 0.8 m. This figure shows the rockfill steps
levels (blue line) and the water levels (yellow line) at the beginning of the flood. The
dam crest is at level 50.0 m and the MWL on the dam is 54.0 m.
View (from upstream) of the lower part of dam
Longitudinal profile after the flood. The figure shows the rockfill and water levels and
the maximum water velocities, with the erosion of the upper part of the dam and the
material deposit downstream the toe.
Conclusion: with such high head, important nappe depth and small size of the
rock fill (0.8 m, limited by the low quality of the quarry), the dam cannot resist to
the overflow without a protection layer.
(4) Alternative with stepped gabions entirely on the downstream
slope face of the dam (no slope in the upstream slope part).
Note the thickness of the nappe compared with the height of the steps: the
dissipation of energy is low on the steps and the erosion is still important at the
toe of the dam.
View of the steps showing the disorganization of the compacted rockfill just under
the gabions due to the piping of the small rocks through the gabions.
This type of protection requires the use of steel bars in place of wire mesh, a filter layer
between the rockfill and the gabions and a reliable protection at the toe of the dam.
CONCLUSION
• CFRD can be overtopped during the construction if its crest,
downstream face and toe are adequately protected against erosion.
• If the dam is constructed on a site with low discharge during the dry
season, but with very large floods during some months of the wet
season, it is possible to reduce, by this mean, the delay and cost of
construction.
• As more and more CFRD will be constructed in Asia in such
condition (monsoon), this practice seems very promising.
• It will be therefore interesting to collect in the future the experience
of this method of construction to improve its economy and safety.
