RD 2011

CFRD 17

REVENTAZN DAM GEOTECHNICAL DESIGN

Roy Ruiz Vzquez1, Eduardo Avils Madrigal2
1 Instituto Costarricense de Electricidad, [email protected],
2 Instituto Costarricense de Electricidad, [email protected],

ABSTRACT
This paper presents the geotechnical design of Reventazn 130 m high CFRD dam, located in Reventazn River
in Costa Rica, Central America. The dam foundation consists of conglomerates and breccias (weak rocks). The
plinth width was designed using the concept of internal plinth with a permissible gradient of 6 in breccias and 10
in conglomerates. Rockfill design has adopted the modern concept with thinner layer thickness and a high
modulus of deformability of upstream zone of the dam. Rockfill comes from the alluvial banks near the dam and
from the required excavations of the dam site. The use of alluvial materials compacted in thin layers and with
careful monitoring will result in very small settlements, as it could be verified with the results of fill test. Two
things make different the design of the dam: The presence of a weak layer below the dam also serves as a seal
layer of an aquifer connected to the reservoir. For this, it was designed a drain curtain and relief wells. The high
seismicity of the area. For this, lower angles were used in the slopes, crest width is bigger and zone 3B was
extended to the downstream slope in the upper part of the dam. A dynamic analysis of the dam is beginning. The
diversion works and the dam excavation is just beginning and the final drawing is almost finished.

Key Word: CFRD, plinth, rockfill, deformation modulus, permeability, shear strength.

1. INTRODUCTION
Reventazn hydroelectric project is located in the middle basin of the same name, about 8 km
southwest of the city of Siquirres, Costa Rica. The project has a CFRD dam with a height of 130 m
and a fill volume of 7,89 hm3, creating a regulation reservoir 120 hm3 bimonthly useful. Normal
operating levels minimum and maximum dimensions are 245 m and 265 m, respectively. Figure 1
shows the plan of the dam.

Fig.1 Dam plant

The Second International Symposium on Rockfill Dams

The crest has a width of 12,0 m, a long of 535 m and is located to 270 msnm of elevation. The
upstream slope is 1,5 H:1V, the downstream slope is 1,6 H:1V. The volume of excavation of the
foundation of the dam is 2,29 hm3.
The dam face area (A) is 76.024 m2 and the maximum hydraulic load (H) is 127,5 m. Then, the
relationship A/H2 (called the shape of the valley) is 4,68. In this case this value is greater than 4, so it
is a wide valley. A plant of the dam concrete face is shown in the fig. 2.

Fig. 2 Dam concrete face plant

The diversion works are designed to allow a peak flow of 3.700 m3/s, corresponding to a return period
of 200 years. These consist of two tunnels located on the left bank of the river, with 14 m diameter
excavation. The longest tunnel is 793 m length, and the other is 667 m length. The cofferdam is
rockfill dam with a geomembrane as an impervious element. The crest is at 186 m elevation.
The dam excavation is currently under construction and river diversion is planned for March 2012. It
is considered a construction period of 36 months at the dam is expected to be completed by January
2015.

2. GEOLOGY
In the dam site there are three lithologies: Fm. Doan, Fm. Suretka and Fm. Guayacn. The top is
generated by a volcanic breccia associated with the Fm. Doan, whose thickness varies between 50 and
200 m. The middle part consists of a conglomerate of Fm. Suretka (alluvial fan deposit) with a
thickness varying between 65 and 130 m. The bottom is formed by several lava flows (Fm. Guayacn)
and paleosols. These basaltic lavas are alkaline with a variable thickness between 30 and 190 m. The
basaltic lava flows underlying the conglomerates in the dam site by a contact of altered lava which is a
weak and weathered zone.
The faults that could have some influence on the dam site are strike-slip faults and thrust-slip faults.
Of these faults, the closest to the dam site is Lomas fault, which has a left-handed strike-slip fault.
The stratigraphy of lava flow present fault system that could be related with some tectonic stress
influence North - South compression associated with this region.
Reventazn hydroelectric project is located in a region of complex geology and tectonic activity,
resulting in complex fault systems, many of which have not been studied in detail. From a
deterministic point of view, the most adverse condition of occurrence of a seismic event with respect
to the dam project site shows a magnitude of 7,0 MW, with a peak horizontal acceleration of 0,41 g.
From the hydro geological point of view, there are four aquifers in the dam site: two lavas confined
aquifers (each one separated by layers of altered lava and paleosol, respectively), a free aquifer in
The Second International Symposium on Rockfill Dams

conglomerate, and an aquifer hung in breccias. The aquifers of the dam site are continental, formed in
volcanic and sedimentary fractured materials.

Fig. 3 Geological section of the plinth

Fig. 4 Maximum section of the dam that shows altered lava layer

3. FOUNDATION
Both Doan breccia and Suretka conglomerate are soft rocks formed with hard clasts embedded in a
matrix of variable consistency. The average compressive strength of Doan breccia is 5,0 MPa and
Suretka conglomerate is 7,0 MPa. Since these rocks do not present a systematic jointing, their
behavior tends to be that of a massive rock, but of high deformability and low resistance. However, the
conglomerate has an anisotropy related to their sedimentary origin and has resulted in planes of
weakness parallel to bedding.
Deformability modules of breccias Doan are between 1000 and 1.200 MPa. Deformability modules of
conglomerates are between 1.500 and 2.000 MPa. However, in both materials decompression depths
are between 30 and 40 m where deformability modules do not exceed 300 MPa.
The Guayacn lavas flow has the typical behavior of the lava flows: anisotropic behavior in both
strength and deformability by the presence of breccias and paleosols in contacts between flows. The
lavas have uniaxial compressive strength between 40 and 70 MPa. Breccias have compressive
strengths simple average of 2,5 MPa.
The contact between conglomerates and lavas dip downstream and consists of a layer of volcaniclastic
material with low mechanical qualities know as altered lava. This contact has an unfavorable dip
The Second International Symposium on Rockfill Dams

and it works as a hydrogeological seal that serves as a confined aquifer that is fed to the filling of the
reservoir.
Therefore, we need to analyze what will happen with this weak layer when the reservoir is full. Direct
shear tests on site and in the laboratory were carried out. This allowed estimating their strength
properties. The results of stability analysis (pseudo static conditions) have demonstrated the need to
bring down the uplift pressure existing under this weak layer (see figure 5).
To bring down this uplift, a grout curtain was designed under the plinth and a drainage curtain was
designed below the dam. Also, there are ten relief wells located near the downstream toe of the dam.

Fig. 5 Examples of failure surface of pseudoestatic analysis

The results of pseudo static stability analysis should be verified by a dynamic analysis of the dam,
which should also include the effect of the presence of this weak layer in the foundation.

4. BORROW MATERIALS
Reventazn dam has a total volume filled with about 7.9 million cubic meters. Materials close to the
dam available for construction are the following:
Sites

Type of material

Available
Volume (m3)

Dam area to be used

Reventazn River
alluvial banks
Conglomerates from
the excavations.

Sandy Gravel, healthy and

7.168.408
2A, 2B, 3A, 3B, 3C, 4
clean.
Old gravel slightly lithified
1.073.871
Cofferdam, T, 3C
sandy matrix.
Angular blocks in a sandyBreccias from the
silty matrix with low-grade
307.354
Cofferdam, T
excavations.
welding.
Healthy fractured lavas mix
Reservoir Right Bank
with altered fractured lavas
Undefined
T, 3C
Quarry
and breccias.
Table 1. Materials available for Reventazn dam
Rockfill material to get of excavation of the spillway and diversion works is a relatively small,
because a lot of this material will be soils and poor quality breccias. Conglomerate and breccias of
good quality will be available towards the end of the excavation of the spillway and are a small
volume.

The Second International Symposium on Rockfill Dams

5. PLINTH GEOTECHNICAL DESIGN

The geotechnical design of the plinth was performed according to the criterion of internal plinth where
the plinth width requirements are met in part by extending downstream of the plinth, under the
upstream slope of the dam. The width of the plinth is dimensioned according to the terms of erodibility
of the rock mass on which will be founded (ICOLD, 2004). The part of the plinth upstream of the
perimeter joint was designed with a width of 4,5 m to facilitate the work of the foundation grouting.
Boreholes give little information on the erodibility and the geophysics could not be performed because
the topography is very steep. For the final design, it was expected to have additional drilling
information and information from a seismic refraction to get a better characterization the different
areas on which the plinth would be founded.
Materials in contact with the foundation of the plinth are shown in the figure 3 and are the following:
On the left side: above 207 m elevation breccia appears healthy, and below 207 m elevation
conglomerate appears healthy.
On the river: at 140 m level shows healthy conglomerate
On the right side: above 251 m elevation breccia appears healthy, and below 251 m elevation
conglomerate appears healthy.

Segment
Elevations
Hydraulic
Load H
(m)

Segment

Both materials in healthy condition show a P wave speed of 1.600 to 1.900 m/s. That is, the
mechanical properties of both materials are similar.
Width of the Plinth (m)
Foundation
(ICOLD
2004)

Maximun
acceptable
gradientd

140 al
126
189

Moderately
Erodible

10

12,6

4,50

8,10

8,50

PI1A- 189 al
77
PI2
227
PI1- 227 al
39
PI1A
266
PI3 al 140 al
126
PD3
140

Moderately
Erodible

10

7,7

4,50

3,20

3,50

Erodible

6,5

4,50

2,00

2,00

Moderately
Erodible

10

12,6

4,50

8,10

8,50

PD3- 140 al
126
PD3A 185

Moderately
Erodible

10

12,6

4,50

8,10

8,50

PD3APD2
PD2PD1

Moderately
Erodible

10

8,1

4,50

3,60

4,00

Erodible

5,75

4,50

1,25

2,00

PI2PI3

185 al
81
231,5
231,5
34,5
al 266

Minimun External Minimun

Total
Width
Internal
Width (m)
(m)
Width (m)

Proposed
Internal
Width
(m)

Table 2. Design width of the plinth as the geotechnical conditions of foundation level adopted
It is considered that the tops of the abutments, the depth of cut for the plinth is lower and thus will be
founded in breccias or altered and decompressed conglomerates (erodible). Therefore, it was decided
to assign a maximum acceptable gradient of 6 (ICOLD, 2004). At lower levels, the conglomerates are
generally cemented soft rock, but may have discontinuous sandy lenses, so they are considered as
moderately erodible. Then a maximum acceptable gradient is 10 (ICOLD, 2004).
Table 2 shows the total width of the plinth based on the geotechnical conditions present at the dam site
Reventazn hydroelectric project. The plinth has a minimum width of 6,50 m at the top of the

The Second International Symposium on Rockfill Dams

abutments of the dam, a width varying between 8 and 8,50 m in the middle of the slope and total width
of 13 m in the riverbed.
In addition to this, it is necessary to indicate that the fault zones (if present) should apply the usual
cleaning procedures and application of dental concrete to a depth of 3 times the width of the fault.
Moreover, both conglomerates and breccias have unconsolidated lenses, therefore, the layers of
material 2A (filters), 2B (supporting slab) and 3A (transition) are extended over the internal plinth up
to H/2, where H is the hydraulic head.

6. DAM EXCAVATION
The usual practice for CFRD dams is that only the excavation of the plinth should keep a formal
design, because the slopes will be permanent and the geometry of the excavation in the vicinity of the
plinth should encourage good behavior of the slab. The rest of the excavation of the dam is done
according to specifications of the deformation modulus E presents at the foundation level.
Additionally should be checked that the resulting slopes are stable in a temporary condition during the
excavation of the dam.
The foundation level of the different zones of the dam should follow the criteria shown in table 3.
Zone of the Dam

Type of the Rock Specified to Founded

Plinth

According plane bedrock excavation: vp breccia with at least

1.600 m/s and vp conglomerate least 1.900 m/s.

2A (filter) , 2B (Support the

Concrete Slab , 3A (Transition)

According plane bedrock excavation: vp breccia with at least

1.600 m/s and vp conglomerate least 1.900 m/s.

3B (Upstream Rockfill)

Rock decompressed: breccias, conglomerate (vp between 800 m/s

and 1.600 m/s) or compact alluvium.

T (Transition), 3C (Downstream
Rockfill), 4 (Drain).

Uncompressed or altered rock: breccias and conglomerate (vp

between 800 m/s and 1.300 m/s), compact alluvium, hard soils.

Table 3. Criteria adopted to determine the foundation level of the zones of dam
The rock types were defined as follows:
1. Hard soil: Residual hard soil with N greater than 25 blows.
2. Compact alluvial: Alluvium of the river well graded with a percentage of fines less than 15%
and a relative density equal to or greater rockfill placed on top.
3. Decompressed Rock: Rock with compressional wave velocity (vp) of 800 m/s - 1.600 m/s,
whose resistance is such that it cannot be removed by a 200 HP tractor with universal blade.
4. Bedrock: Rock with compressional wave velocity (vp) of at least 1.600 m/s, whose resistance
is such that it cannot be removed by a heavy ripper.

7. DESIGN OF THE ZONING OF THE DAM

The design of the zoning of the dam was carried out according to the availability of materials in the
area (see Table 4) and the recommendations of the last dams built worldwide CFRD. It also has used
the results of a test fillings made by ICE to refine the grain size curves and values of density of the
materials associated with a compaction energy for each type of material.
The Second International Symposium on Rockfill Dams

Zone

Function

Description

Origin

Maximun
Size
(mm)

Layer
Thickness
(m)

1A

Slab Protection

Cohesionless Silty
Sand

Underground
Excavation Works
conduction

2,5

0,25

1B

1A Zone
Protection and
Containment

Fill Unselected

Excavation Works

300

0,60

2A

Perimeter Joint
Protection

Fine Filter

Alluvial Deposits
Processed

19

0,20

2B

Support the
Concrete Slab

Semi Permeable Filter

Alluvial Deposits
Processed

100

0,30

3A

Transition Zone
between 2B and
3B

Alluvial Graves

Alluvial Deposits

250

0,30

3B

Upstream Rockfill
Dam Axis

Alluvial Graves

Alluvial Deposits

550

0,60

Transition Zone
downstream of the
Dam Axis

Conglomerates,
Alluvial with Breccia
Mixture or Lavas

Underground
Excavation,
Mixtures of
Materials

550

0,60

3C

Downstream
Rockfill Dam Axis

Alluvial Graves or
Lavas

Alluvial Deposits
or Lavas

700

0,80

Horizontal Drain
4

0,90
Alluvial Graves

Alluvial Deposits

300

Vertical Drain or
Chimney

Rip Rap

Protection of
Downstream the
slope face

0,40

Boulders, Lavas

About Size from

the River, Lavas
Blocks

> 700

---

Table 4. Description of materials to be used by landfill area of the dam

For upstream areas (upstream the axis of dam) it will use alluvial materials obtained from the alluvial
banks of the river Reventazn. For downstream areas it will use lower quality materials and mixtures
of alluvial materials produced by the excavations. By using materials with higher fines content as
conglomerates in the downstream area, it must enter a chimney drain in the dam. In the Figure 5
presents the maximum section of the dam. Below is the description and characteristics of each of the
zones.

The Second International Symposium on Rockfill Dams

Fig. 5 Maximum cross section of the dam

Zone 1A
The zone 1A material is used as a protective element of the concrete face of the dam, located just
above the slab. This zone consists of a cohesionless silty sand with a plasticity index between 5% and
7% to be obtained from the matrix of conglomerates from the excavations. This zone is located on the
lower third of the face of concrete below the elevation 200 m and up to bed in the elevation 137 m.
Zone 1B
The zone 1B materials should be used to confine the area 1A. This material will not be selected. Be
obtained from the excavation of the spillway, consider the breccias mainly.
Zone 2A
The 2A zone material should be used for the construction of the filter on the zone adjacent to the
perimeter joint of the concrete face. Processed materials will be used, obtained from the crushing of
relevant materials extracted from alluvial river terraces.
The zone 2A should not have fines content greater than 10%, the gravel should be around 30% and
sand content of approximately 65%. In the Figure 6 shows the grading band defined for this material.
In addition, the materials for zone 2A should meet the requirements shown in Table 3.
Testing
Absoption (ASTM C-127)
Abrasion Los Angeles Machine
(ASTM C-535)
Sodium sulfate soundness
(ASTM C-88)

Requirement
8% maximun
50% maximun after 1000 revolutions
15% maximun after 5 cycles

Table 5. Additional requirements for the material of the Zone 2A of the dam filled
Zone 2B
The material in zone 2B should be used to build a semi-impermeable concrete face. A mixture of low
permeability processed, from the best materials from the alluvial terraces that comply with the grading
of Sherard (Sherard, 1987).
This zone has a width of 4,0 m. This material must have between 35% and 55% of sand (passing the
No. 4 sieve) and a maximum of 8% of silt or clay material passing the No. 200 sieve. With this
The Second International Symposium on Rockfill Dams

gradation is obtained permeability between 10-3 cm/s and 10-4 cm/s. In the Figure 6 shows the
grading bands proposed for this material. These bands are taken from the report's recommendation
first visit of the consultant Ing. Bayardo Matern (Matern, 2009).
The material in zone 2B also is used as a transition between zone 4 and zone T and 3C, the horizontal
and vertical drain. The zone 2B of the dam must meet the same requirements specified for zone 2A
(see Table 5).
Zone 3A
The material of the zone 3A will be used as a transition between the material of zone 2B and zone 3B
gravels of the backstop upstream of the dam. In the Figure 6 shows the grading bands of this material
were specified taking into account the grading evaluated in the exploration that took place in the
feasibility stage and during investigations in the fills test that the ICE performed.
In order to prevent erosion of the foundation of the dam leaks in the vicinity of the joint perimeter the
zones 2A, 2B and 3A extend over the surface of the abutments behind the perimeter joint width varies
from 5 m downstream of the axis of the fillers (at the top of the dam) to 0,50 H from the axis of the
plinth at the base of it, where H is the hydraulic head.
The materials used for zone 3A of the dam must meet the same requirements specified for Zone 2A
(see Table 5).
GRANULOMETRIC BANDS

REVENTAZN DAM HYDROELECTRIC PROJECT

TYPE: CFRD
HYDROMETER

24 h

100

4h
10 m in

1h

30 15
min m in

5
min

PRESENTED ZONES: 1A, 2A, 2B, 3A.

U. S. GRID STANDAR

2
m in

200

100

5040

30

16

GRID IN INCHES

10

3/8" 1/2" 3/4" 1

1 1/2"

SIZE IN INCHES

3"

6"

12"

24"

90

PASSING PERCENTAGE

80
70
60
2A LS
50
2A LI
40

2B LS

30

2B LI

20

3A LS
3A LI

10
1A LS
0
0.001

0.010

0.100

CLAY TO SILT

1.000
APERTURE OF SIEVE (mm)
FINE

MEDIUM
SAND

GROSS

10.000

100.000

FINE

GROSS

PEBBLES

1000.000

BOULDER

GRAVEL

Fig. 6 Granulometric bands for the materials 1A, 2A, 2B and 3A

Zone 3B
The upstream zone of the dam (3B zone) will be built with gravel fill. The grading bands defined for
zone 3B were established from the envelopes of grading tests to samples taken from the terraces of
Reventazn river during the feasibility stage. Addition, more grading test were completed during the
filled test using materials from alluvial terraces downstream of dam axis. The terraces located
upstream of the dam site, are the main sources of project materials.
In the Figure 7 shows the grading bands for this material. As a criterion of acceptance of the gravel
will be considered a value of the abrasion test according to the Los Angeles Machine less than or equal
to 50% and an absorption value less than or equal to 5%.

The Second International Symposium on Rockfill Dams

GRANULOMETRIC BANDS

REVENTAZN DAM HYDROELECTRIC PROJECT

TYPE: CFRD

PRESENTED ZONES: 3B, DRAINS T, 3C.

HYDROMETER

24 h

100

4h
10 min

30 15
min min

1h

5
min

U. S. GRID STANDAR

2
min

200

100

5040

30

16

GRID IN INCHES

10

3/8" 1/2" 3/4" 1

1 1/2"

SIZE IN INCHES

3"

6"

12"

24"

90

PORCENTAJE PASANDO

80
70
60
3B LS
50
40
30

3B LI
Dren LS
Dren LI
T LS

20
10

T LI
3C LS
3C LI

0
0.001

0.010

0.100

1.000
ABERTURA DE MALLA (mm)
FINE

MEDIUM

GROSS

10.000

100.000

FINO

GRUESO

CLAY TO SILT

PEBBLES
SAND

1000.000

BOULDER

GRAVEL

Fig. 7 Grading bands obtained for the materials 3B, Drains, T and 3C
Zone T
The material of the zone T is used to build the central backstop downstream of the dam, as shown in
Figure 5. The Figure 7 shows the grading bands specified for the use in this zone.
As acceptance criteria for zone T material, furthermore of the defined grading, may be considered an
absorption value less than or equal to 8%, a maximum value on the abrasion test in Los Angeles
machine less than or equal to 50% and a maximum loss of 15% in the sanity test.
Zone 3C
The material in Zone 3C will be used to build the downstream body of the dam, as shown in Figure 5.
The material should be free of clay, organic matter, material susceptible to decay and any other
inappropriate material. The grading bands specified for the zone 3C of the dam are shown in Figure 7.
In addition to meeting the grading bands shown in Figure 7, the material must meet the same
requirements for Zone 3B of the dam.
Zone 4
The zone 4 consists of a horizontal drain placed under the zones T and 3C and a vertical chimney
drain between the zones 3B and T. The horizontal drain is placed over the entire bed of the river from
the axis of the dam and downstream to have a drainage blanket to evacuate in a controlled manner and
with sufficient capacity, seepage from the reservoir through the concrete face. This ensures that the
downstream body of the dam (zones T and 3C) are kept dry and not affecting its stability.
It is possible that the zone T material present a lot of fines, because such material is derived from the
conglomerates of the excavation of the spillway. The same could happen with the lava in the altered
nature of these deposits. The chimney drain would be 4 m thick, and will be protected upstream and
downstream by a layer of 2 m thickness from material 2B. The vertical drain is an average height of
100 m.
Rockfill protection

The Second International Symposium on Rockfill Dams

10

The slope protection material downstream of the dam will be a rockfill layer consisting of fragments
clean rock sizes from alluvial beaches, excavation of the spillway or underground excavations. Its size
will be 0,70 m and the width of the buffer zone is 2 m measured horizontally. The quality of the rock
fragments should ensure that there is no long-term decline.
The following table shows the characteristics necessary for the placement and compaction of materials
for each zone.
Zone
1A
1B
2A
2B
3A
3B
3C
T
4 Chimney
Drain
4 Horizontal
Drain

Equipment
Compaction
Tractor D-8L or similar
Tractor D-8L or similar
Manual or Vibratory 10 Ton
Vibratory 15 Ton
Vibratory 15 Ton
Vibratory 15 Ton
Vibratory 15 Ton
Vibratory 15 Ton

Number Past
2
2
4
4
4
4
6
6

Tractor D-8L or similar

Tractor D-8L or similar

Table 6. Characteristics of placement the zones of the dam

8. GEOMECHANICAL PROPERTIES OF THE MATERIALS OF THE DAM

8.1 Hydraulic Properties
8.1.1 Revision of the laws of filters between materials
It is very important to check that compliance with the laws of filters on the materials placed
downstream of the concrete face or to a chimney drain, because the filter system provides a second
line of defense against leaks caused by cracks in the slab or perimeter joint failures. Moreover, the
filter should be much more permeable than the interface sealant or material trapped on the perimeter
joint.
The following criteria condense the laws that material with basic filter function must meet (ICOLD,
2004).
1. Function of the internal erosion resistance given by the following relation (Marsal and
Resndiz, 1975):
D15F < 5* D85S
2. Function of permeability given by the following relationship (Marsal and Resndiz, 1975):
D15F 5* D15S
Where:
D15: Size of the particles in the filter (protective agent) which constitute 15% by weight with a
diameter less than that.
d85: Size of the particles to protect which constitute the 85% by weight with a diameter less than that.
d15: Size of the particles to protect which constitute the 15% by weight with a diameter less than that.
The Second International Symposium on Rockfill Dams

11

Compact Material

D15F

d15s

d85S

5*d15s

5*d85S

1A - 2A Average
1A - 2B Average
2A - 2B Average
2B - 3A Average
3A - 3B Average
3B - 2B Average
2B - Drain Average
T - 2B Average

0,19
0,34
0,33
2,30
8,40
0,33
40,00
0,33

0,08
0,08
0,19
0,33
2,30
8,25
0,33
3,20

1,82
1,82
10,07
45,00
150,00
320,00
45,00
377,78

0,38
0,38
0,95
1,65
11,50
41,25
1,65
16,00

9,10
9,10
50,35
225,00
750,00
1.600,00
225,00
1.888,90

1st Law of
Filters
Complies
Complies
Complies
Complies
Complies
Complies
Complies
Complies

2nd Law of
Filters
Not Comply
Not Comply
Not Comply
Complies
Not Comply
Not Comply
Complies
Not Comply

Table 7. Results from the application of the filter law of the materials
According to criteria consulted, the first condition is the most important meet, especially among the
materials 2B, 3A and 3B considering a high-permeability materials 3B (Yez, 2010).
8.1.2

Permeability of materials

To estimate a permeability coefficient of materials (k), the Matsuo Akai test was performed on each
materials valued in the execution of the test fill. The results are shown in the following summary table.
Material
2B
3A
3B
Conglomerate
Lavas
3C

Origin
Igneous
Igneous
Igneous
Sedimentary
Igneous
Igneous

k (cm/s)
8,37E-04
1,44E-03
1,52E-02
2,15E-03
1,70E-03
>> 1,52E-02

Table 8. Values obtained for the permeability coefficient for each of the principal materials of
the dam.
For the permeability of drains (zone 4) adopted a permeability of 1x10-1 m/s, which is typical of a
uniform clean gravel, since in these materials was not carried out any such test to check the value
presented but given the behavior experienced by materials 3B and 3C is estimated that the
permeability of the materials used as drains be higher in at least an order of magnitude.
8.2 Shear Strenght
The methodology proposed by Barton and Kjaernsli (1981) allows for the variation of the effective
friction angle of the gravel and rockfill in function of normal stress applied, taking into account the
roughness of the particles, the compressive strength and the angle of friction materials of the particles
that make up the rockfill. The equation involves obtaining both parameters proposed by Barton and
Kjaernsli presented below (Barton and Kjaernsli, 1981).

Where you have to;

'= Effective friction angle of the rockfill.
The Second International Symposium on Rockfill Dams

12

R = Equivalent roughness of the particles obtained according to the porosity and degree of roundness
and softness of the particles.
S = Equivalent resistance of the rockfill in the size of the particles d50 and uniaxial compressive
strength of intact rock.
n = Effective normal stress.
b = Basic friction angle of the smooth, flat surfaces of the bedrock. Usually this value varies between
25 and 35 .
For materials with gravel from alluvial beaches and indexed for each zone of the dam, as well as
materials such conglomerates, lavas and breaches consolidated estimated parameters for the
application of this methodology and are summarized in the table below.
Material
Conglomerate
Breccia
Lavas
2B
3A
3B
3C

Compressive
Strenght
d 50
(Ton/m2)
9.447,05
40,37
897,08
32,42
6.906,60
58,00
8.247,75
8,88
8.247,75
48,64
8.247,75
116,04
8.247,75
70,00

0,13
0,13
0,17
0,13
0,05
0,18
0,17

11,42
11,14
14,57
11,50
4,99
15,25
14,57

Equivalent
Roughness
R
4,16
5,99
4,47
3,28
3,35
3,35
3,26

Equivalent
Resistance S
(Ton/m2)
6.801,57
657,85
4.834,62
7.285,24
5.841,88
5.773,43
5.773,43

b ()
30,00
30,00
35,00
30,00
30,00
30,00
30,00

Table 9. Parameters of the materials for Reventazn dam fillings

8.3 Compressibility of the gravel and rockfill

The compressibility of the materials used in test fill was obtained by load test of rigid plate (425 mm
diameter) on the surface during the performance of test fill. Also, they were made odometer tests at
the laboratory (600 mm diameter). A stress of at least 2 times the maximum vertical stress of the dam
was applied on the materials.
Of the results of both tests, were obtained global elastic modules, which can determine the elastic
behavior of each of the materials to use and make the respective comparison.
The load was applied in increments, measuring each step, the settlement experienced and recorded.
From the global behavior is determined the elastic modulus experienced by the material. The results
for each of the materials are presented below.
Material
2B
2B
3A
3A
3B
3B
Conglomerate
Lavas
3C

Testing
Number
1
2
1
2
1
2
1
1
1

Poisson Ratio
0,25
0,25
0,25
0,25
0,25
0,25
0,25
0,25
0,25

Global Modulus
(MPa)
169,70
155,36
167,50
170,60
303,59
287,92
131,19
141,25
178,15

Table 10. Global modulus materials obtained by rigid plate load test

The Second International Symposium on Rockfill Dams

13

In the case of odometer testing procedure for obtaining the modulus is directly associated with volume
changes experienced by a material under the action of forces (usually compressors) to which you want
to submit the material; must be similar to those experienced on-site material placement within the
filling of the dam. The results of the odometer test are shown in table 11.

Material

Geological
Origin

2B

Igneous

3A

Igneous

3B

Igneous

3C

Igneous

Conglom.

Sediment.

Lavas

Igneous

Alluvial /
Breccia

Igneous
(Major
Portion)

Load Cycle
(% Applied
Strain)

Obtained
Deformation
(cm)

Applied
Stress
(kg/cm2)

1 al 50%
2 al 75%
3 al 100%
1 al 50%
2 al 75%
3 al 100%
1 al 50%
2 al 75%
3 al 100%
1 al 50%
2 al 75%
3 al 100%
1 al 50%
2 al 75%
3 al 100%
1 al 50%
2 al 75%
3 al 100%
1 al 50%
2 al 75%
3 al 100%

0,33
0,61
0,89
1,40
1,89
2,37
1,09
1,67
2,24
0,83
1,44
2,04
5,75
6,50
7,20
7,17
8,16
8,76
2,91
3,73
4,45

23,87
35,80
47,73
26,73
40,59
54,45
25,74
39,60
52,47
24,75
37,62
50,49
22,77
33,66
44,55
22,77
34,65
46,53
24,75
36,63
49,50

Obtained
Odometer
Modulus
(MPa)
383,76
311,37
284,52
101,30
113,94
121,89
125,29
125,80
124,27
158,20
138,60
131,31
21,01
27,47
32,83
16,85
22,53
28,18
45,12
52,10
59,09

Breakage
Grains (Bg)

30,16%

18,26%

2,76%

2,05%

5,00%

5,95%

3,67%

Table 11. Odometer module obtained for each of the materials used
This test was performed within a steel cylinder, providing confinement to the material being subjected
to only one-way stress. The trial includes three cycles of loading and unloading, showing a remaining
deformation between each cycle which is cumulative and can determine the maximum deformation
experienced by the materials, applying 100% of the determined stress.
Obtaining samples for each material to evaluate the odometer test was subject to respecting a
relationship between the diameter of the odometer and the nominal diameter of particles of 6, hence
the set maximum size to be tested within odometer a nominal maximum size of 3" (75 mm). As this
relationship between the diameter of the odometer and maximum particle size is larger, more
representative results become.
Obtaining the particle size for each of the materials to evaluate within the odometer, was performed by
obtaining the homothetic curves according to the average grading envelopes for each of the materials
obtained. Homothetic curves of the materials are obtained when the coefficient of uniformity and
curvature of the original grading is identical to those obtained in the parallel gradation produced.

9. FILLS TEST
The most important results were obtained for the materials tested during the fill test.

The Second International Symposium on Rockfill Dams

14

The thickness of layers was not a parameter to test during the investigation stage. The thickness of
layer was established previously for each one by recommendation of Ing. Bayardo Matern.
Controls were performed dry density obtained on site, to changes in energy compaction by the number
of passes for each material, assessing the moisture content presented, especially those other than the
alluvium to be used in the dam fill.
Number of Past versus Unit Deformation on Alluvial Materials
16,00

14,00

12,00
)
(% 10,00
n
o
tia
rm 8,00
o
f
e
D
ti
n 6,00
U

2B
3A
3B
3C

4,00

2,00

0,00
0

Number of Passes Compactor

Fig. 8 Response deformation experienced by the alluvial materials to changes in the energy
compaction provided
Number of Past versus Unit Deformation on No Alluvial Materials
16,00

14,00

12,00
)
(% 10,00
n
o
tia
m
r 8,00
o
f
e
D
ti
n 6,00
U

Lavas
Conglomerates
Breach
Alluvial/Breccia

4,00

2,00

0,00
0

Number of Passes Compactor

Fig. 9 Response deformation experienced by no alluvial materials to changes in the energy

compaction provided
In this way we determined the behavior of alluvial and not alluvial material, comparing the strain
respect to the energy compaction by the number of passes, also to make the comparison between the
density achieved by the materials in terms of energy compaction.
The compaction equipment used in the fill test is a BOMAG BW 213 DH-4 BVC, which presents a
static load on the roll of 9,40 Ton with a drum width of 2,13 m, providing a static load per meter of
4,40 Ton/m.

The Second International Symposium on Rockfill Dams

15

The results obtained in this stage of investigation are presented in figures 8, 9, 10 and 11.
Number of Passes versus Density of Alluvial Materials
2,600
2,550
2,500
)
3 2,450
/m
kg(
sl 2,400
a
ri
e
ta
M 2,350
e
h
t
f 2,300
o
yt
is
n
e 2,250
D

2B
3A
3B
3C

2,200
2,150
2,100
0

Number of Passes Compactor

Fig. 10 Response of alluvial material density to changes in the energy compaction provided
Number of Passes versus Density of No Alluvial Materials
2600
2500
2400
)
3 2300
/m
kg(
sl 2200
iar
e
at 2100
M
e
h
t
f 2000
o
y
its
n
e 1900
D

Lavas
Breach
Alluvial/Breccia
Conglomerates

1800
1700
1600
0

Number of Passes Compactor

Fig. 11 Response of no alluvial material density to changes in the energy compaction provided

10. GROUT CURTAIN AND DRAIN CURTAIN

The grout curtain will be constructed by injection from the plinth and from galleries with boreholes
inclined 35 from the vertical, upstream. In the plinth are designed two consolidation rows of
boreholes separate each 1,5 m, with depths varying from 10 m to 30 m. In the river bed zone of the
plinth, the external consolidation line is 15 m deep and spaced 1,5 m (Vargas, 2011).
The internal consolidation line have 30 m deep and spaced 1,5 m. The separation between the rows of
consolidation boreholes and central grouting line is 1,5 m. The main sealing line has a maximum depth
of 83 m. The primary holes are spaced 6 m between them, the secondary holes 3 m and 1,5 m the
tertiary holes. The grout curtain from the plinth will be built with a 1.565 injection holes. The grouting
pressure for consolidation between the plinth and the rock is considered in 1,5 kg/cm2. (Vargas, 2011).
The grouting treatment of fractures and cracks in the conglomerates, lavas and breccias in the
abutments of the dam will be carry out from galleries. These galleries have a height and a width of 4
m. A section is parallel to the axis of the dam and another running parallel to the plinth. The length of
the grouting galleries and drainage varies from 14 m to 538 m, being greater on the right side due to
the sealing of a section adjacent to the spillway in the massif. The drainage gallery, located at 125 m
elevation, has a length of 538 m, from which, drawdown holes must be constructed to relief the
The Second International Symposium on Rockfill Dams

16

confined groundwater pressure of the confined aquifers lavas 1 and 2 (see Figure 3); allows deletions
through drainage holes to cut the confined aquifer (Vargas, 2011).
These vertical mini relief wells have a depth of 72 m on average, and will be constructed with slotted
plastic tube. The slotted section is placed on the geological section that extends from the base of the
altered lava until the roof of the Tobita (Vargas, 2011).
The separation of the galleries is about 40 m placed in a vertical plane. The grouting holes are divided
into drilling down at an angle of 35 respect to the vertical and a fan of boreholes with a length of 12
m and 15 m to intersect the holes from the upper gallery. In addition, the galleries will serve for the
construction of drainage holes with a downstream direction with an angle of 20 with respect to the
vertical and have a depth of 30 m whose function will collect the water that seeps through the grout
curtain. The length of the grout curtain on both sides was calculated using the criteria of the critic
hydraulic gradient for the conglomerate in regular condition, estimated with a value of 3, the location
of the diversion tunnels, the background discharge and the spillway (Vargas, 2011).
The 77,5% of the injection perforations will be held in the galleries and the remaining 22,5% in the
plinth. In relation with these values shows that 74,3% of the length drilled for injection is for drilling
in the galleries and occur the largest cement consumption (Vargas, 2011).

Fig. 12 Detail of the grout curtain

11. INSTRUMENTATION
The dam will have three instrumented sections with the following instruments:
50 vibrating wire piezometers in the foundation,
41 vibrating wire settlement cells,
33 total pressure cells,
8 cross arm type USBR,
5 accelerographs.
The concrete face will be instrumented with:
8 embedment strain gauges with 3 sensor each one
12 1D jointmeters
10 3D jointmeters
29 tiltmeters
Reading most of these instruments will be automated using an automated data acquisition.
The Second International Symposium on Rockfill Dams

17

The aquifer under the weak layer is monitored by some of these piezometers and several additional
open piezometers.

12. CONCLUSIONS
Reventazn dam is a CFRD dam design according the modern concepts latest concepts in technical
information and recommendations of international consultants in this type of dam.
For the construction of the dam with alluvial materials that compacted in thin layers and with careful
monitoring will result in very small settlements. This could be verified with the results of test fillings
made.
Two things make different the design of the dam:

The presence of a weak layer below the dam also serves as a seal layer of an aquifer connected
to the reservoir. For this, it was designed a drain curtain and relief wells.

The high seismicity of the area. For this, lower angles were used in the slopes, crest width is
bigger and zone 3B was extended to the downstream slope in the upper part of the dam. A
dynamic analysis of the dam is beginning.

REFERENCES
Barton, N., y Kjaernsli, B. 1981. Shear strength of Rockfill. Journal of the geotechnical engineering
division. ASCE.107, 873-891.
Cruz, P., Matern, B., y Freitas, M. (2009). Concrete Face Rockfill Dams. Sao Paulo. Brasil.
ICOLD. (1994). Concrete Face Rockfill Dams Concepts for Design and Construction. Committee on
Materials for Fill Dams.
Marsal, R, y Resendiz, D. 1975. Presas de Tierra y Enrocamiento. Mxico; Editorial LIMUSA
221:267 p.
Matern, B. (2009). Informe correspondiente a la primera visita al Proyecto Hidroelctrico
Reventazn. Informe Tcnico, 9-41.
Matern, B. (2010). Informe correspondiente a la segunda visita al Proyecto Hidroelctrico
Reventazn. Informe Tcnico, 6-26.
Matern, B. 2009. Consultas tcnicas referentes a resultados obtenidos para el mdulo de
deformabilidad de los materiales. San Jos. Comunicacin personal.
Sherard, J., y Cooke, J.1987. Concrete Face of Dams: 1.Assessment. Geotechnique. ASCE. 113(10),
1096-1112.
Vargas, S. 2011. Consultas tcnicas referentes al diseo de la cortina de inyeccin. Siquirres, Limn.
Comunicacin personal.
Yaez, D. 2010. Ley de filtros entre materiales de una cortina de enrocamiento. Sociedad Mexicana
de Ingeniera Geotcnica, Mxico.

The Second International Symposium on Rockfill Dams

18