Research Perspectives in Embankment Dams at LNEC · © LNEC 2012 © LNEC 2012 Research Perspectives in Embankment Dams at LNEC Laura Caldeira [email protected] Head of the Geotechnique

© LNEC 2012 © LNEC 2012

Research Perspectives in Embankment Dams at LNEC

Laura Caldeira [email protected]

Head of the Geotechnique Department

Dam World Conference 8-11th October 2012 Maceio, Brazil

© LNEC 2012

Summary

>The influence of upstream zones in the limitation of the progression of internal erosion in zoned dams

>Self-hardening slurry walls design and quality control

2

© LNEC 2012

Earthquake

2%

Slope

instability

4%

Internal

erosion

48%

Overtopping

46%

Importance of internal erosion to dam safety

Source: Foster, Fell e Spannagle (2000)

Overall statistics of embankment dams failures

3

© LNEC 2012

Internal erosion process leading to failure

Section A-A'

Concentrated leakagethrough a transverse crack

Section A-A'

Longitudinal view

INITIATION

CONTINUATION OR FILTRATION

PROGRESSION

BREACH FORMATION

A

A'

Absence ormalfunctioning ofdownstream filter

Enlargment of theconcentrated leak

(erodability of soil)

Reservoirempties afterpipe collapse

Cross Sectional view

Breach formationB

B'

Section A-A'

Pipeformation

Flowincreases

Cracks above pipe

e.g. cracking mechanismby arching effect

Section B-B'

Does the upstream zone: - Limits the eroding flow? - fills in the flaw in the core?

4

© LNEC 2012

Section A-A'

Concentrated leakagethrough a transverse crack

Section A-A'

Longitudinal view

INITIATION

CONTINUATION OR FILTRATION

PROGRESSION

BREACH FORMATION

A

A'

Absence ormalfunctioning ofdownstream filter

Enlargment of theconcentrated leak

(erodability of soil)

Reservoirempties afterpipe collapse

Cross Sectional view

Breach formationB

B'

Section A-A'

Pipeformation

Flowincreases

Cracks above pipe

e.g. cracking mechanismby arching effect

Section B-B'

Does the upstream zone: - Limits the eroding flow? - fills in the flaw in the core?

Internal erosion process leading to failure (cont)

5

© LNEC 2012

Tunbridge Dam, Tasmânia, Australia, 11/28/2008

Source: Jeffery Farrar (2008)

Progression of internal erosion to piping

6

© LNEC 2012

Source: Hanson e Hunt (USDA, 2007)

Progression of internal erosion to piping

7

© LNEC 2012

Erodability of soils in concentrated leaks > Hole Erosion Test (HET)

Specimen in standardProctor mould

Downstream chamber

of acrylic glass

Upstream chamber

of acrylic glassDrilled hole

Ø6 mm

l/h

Flowmeter

Control valve

Upstream tank suppliedby large reservoir above

Downstream tank

Purge Purge

Pea gravel20 mm to 30 mm

Scale

Upstream

piezometer

Downstream

piezometer

300 mm to 1100 mm

˜ 200 mm

Drain

Drain

8

© LNEC 2012

Hole Erosion Test (HET) during test

9

© LNEC 2012

Hole Erosion Test

> Axial hole at the end of a test

10

© LNEC 2012

Limitation of progression of piping

>Flow restriction action

Influence of the presence of upstream zones

Flow Limitation Erosion Test (FLET) Crack-Filling Erosion Test (CFET)

>Crack-filling action

Upstream material Core Filter

Upstream material Core 11

© LNEC 2012

Test cell developed at LNEC for FLET and CFET

12

© LNEC 2012

Flow Limitation Erosion Test (FLET)

Acylic glass

cover

l/h

Electromagnetic

flow meter

Upstream

control valve

Eroding fluid from

storage tank above

Downstream tank

Upstream tank

1250 mm to 2250 mm

200 mm

To drain

To drain

Downstream

control valve

Core material Upstream material

U/S

piezometer

INT

piezometer

D/S

piezometer

Valve Valve

Scale

(mm)

Two

concentric

springs

Digital

cameraOutlet chamber

Pre-formed

pipe (Di)

2300

2200

2100

2000

1900

1800

400

300

200

100

50

Inlet chamber

13

© LNEC 2012

Flow Limitation Erosion Test

> Steps for assembly of test cell and specimen preparation

Step 1 Step 5Step 4Step 3

Upstream

material

90º

Step 2

Core

14

© LNEC 2012

Upstream materials tested in the FLET

Grain Size (mm)

Perc

ent

pa

ssin

gb

yw

eig

ht

(%)

0.001 0.01 0.02 0.05 0.1 0.2 0.5 1 2 3 4 5 7 10 20 30 50

2"3/4"3/8"#4#10#20#40#100#200

ASTM Sieve SizesHydrometer Analysis

Silt or ClayGravelSand

CoarseFineCMediumFine

0

20

40

60

80

100

Core

5% limit

15% limit

30% limit

Upstream materials

with non-plastic fines

with plastic fines

Gap-graded

15

© LNEC 2012 Time (minutes)

Flo

wR

ate

(lit

ers/

h)

Pie

zo

me

tric

le

vel

(cm

)

0 5 10 15 201000 0

1200 50

1400 100

1600 150

1800 200

2000 250Flow rateU/S piezINT piezD/S piez

Df =

20 mmDf = 48 mm

CO RE UPSTREAMZONE

Some results of carried out FLET’s at LNEC > Progression of erosion without flow restriction

16

© LNEC 2012

Some results of carried out FLET’s at LNEC > Flow restriction due to non-erodible upstream material

Time (minutes)

Flo

wra

te(l

iters/

h)

Pie

zo

me

tric

le

vel

(cm

)

0 10 20 30 40 50 60 70 801100 0

1200 80

1300 160

1400 240

1500 320

1600 400

Flow rate stabilizesand effluent clears

Flow rateU/S piezINT piezD/S piez

COREUPSTREAM

Z ONE

FLET B1-02

Df =26 m m

Df = 12 m m

17

© LNEC 2012 Time (minutes)

Flo

wra

te(l

iters/

h)

Pie

zo

me

tric

le

ve

l(c

m)

0 20 40 60 80 100 120 1400 0

400 80

800 160

1200 240

1600 320

2000 400

Onset of pipe

Water bleedRe-open valve

Close valve

Clogged pipe


FLET E1-01a

D f =

14 mm

CORE UPSTREAMZONE

blockage

Some results of carried out FLET’s at LNEC > Flow rate stops completely (self-healing ability)

Core specimen Upstream specimen

Detail of the axial hole of specimens after test 18

© LNEC 2012

Some results of carried out FLET’s at LNEC > Erosion process slows down during a period

Time (minutes)

Flo

wrate

(lit

ers/

h)

Pie

zo

me

tric

le

vel

(cm

)

0 10 20 30 40 50 600 0

400 100

800 200

1200 300

1600 400

2000 500

"Blowout" of gravel part.


Df =

27 mmDf = 36 mm

COREUPSTREAM

ZONE

19

© LNEC 2012

> The performed tests showed that the FLET allows the evaluation of the flow restriction action by an upstream material, that is, if the piping process in the core stops, slows down or progresses.

> The flow restriction action is strongly influenced by some characteristics of the upstream materials, including the fines and gravel contents, as well as the plastic nature of the fines.

> The compaction water content of the upstream material affects strongly the progression of piping erosion.

> The non-plastic fines of soils compacted to the dry side tends to erode more rapidly, leaving unbounded the gravel particles with potential to initiate a self-healing mechanism at the interface or inside the core sample.

Major outcomes of carried out FLET’s

20

© LNEC 2012

Material de

montante

Filling in with eroded particles

from the upstream material

Filter does not retains the

particles from the core

Material settles

Clayey core

Filter retains the coarser particles

eroded from the upstream material

Sinkhole

RockfillRockfill

Crack-Filling Erosion Test (CFET)

> Conceptual model of Crack-filling action mechanism

21

© LNEC 2012

WAC Bennett Dam | Canadá Embankment height=186 m | Length= 2 km Electricity production= 13 biliões kWh/ano

Source: Steve Garner, BCHydro (2007)


> Example of sinkhole formation at the embankment crest

22

© LNEC 2012


> Placement of the filter layer > CFET setup ready to test

23

© LNEC 2012

> Crack-filling of the axial hole on the core with an uniform fine sand


24

© LNEC 2012


> Crack-filling of the axial hole on the core with an uniform fine sand

25

© LNEC 2012

> The preliminary tests showed that the CFET is suitable for the evaluation of the crack-filling action by granular upstream materials.

> The filter layer has an important role in the crack filling action, by retaining some of the particles that are washed in from the upstream material.

> The potential benefits of crack filling action arise from the compatibility between the particle sizes of the upstream material and those of the downstream filter.

> Tests are currently underway examining the crack-filling action due to the presence of several types of coarse grained upstream materials (obtained by blending some fines, and sand and gravel particles).

Major outcomes of preliminary CFET’s

26

© LNEC 2012 © LNEC 2012

SELF-HARDENING SLURRY WALLS

DESIGN AND QUALITY CONTROL

© LNEC 2012

INTRODUCTION

> Objectives A comprehensive literature review.

Characterization of the factors involved in

self-hardening slurry behaviour during

construction and in the long term perfor-

mance.

Definition of numerical models for

analysis and interpretation of the slurry

wall behaviour.

Definition of design principles.

Proposal of a quality control and perfor-

mance evaluation methodology.

© LNEC 2012

> A self-hardening slurry cut-off wall is a non-

-structural underground wall that serves as a

barrier to the horizontal flow of water and other

fluids.

> It is constructed with the aid of a viscous

stabilizing fluid known as slurry. Usually,

cement-bentonite slurries are used.

> In Europe, self-hardening slurries walls have

been used since 1960, particularly in seepage

control applications.

> In Portugal, the technology was first applied in

1978, in the remedial works of the Roxo Dam.

DESCRIPTION SELF-HARDENING SLURRY CUT-OFF WALL

© LNEC 2012

> Main applications of the technology.

> Construction procedures.

Excavation dewatering.

Reduction of seepage through embank-

ments or water storage structures.

Reduction of seepage of ponds and

lakes.

Subsurface dams or groundwater

reservoir.

Isolation or maintenance of water tables.

Containment of solid and liquid wastes.

Seismic cut-off.

APPLICATIONS

© LNEC 2012

> Roxo Dam

APPLICATIONS

Plan

Cross-section

Cut-off wall characteristics:

Wall length: 190 m

Maximum depth: 16.8 m

Width: 0,6 m

Self-hardening

slurry wall

Self-hardening

slurry wall Crest

Concrete gravity dam

Earth embankment

dam

Stilling

basin

Irrigation channel

Water intake

Grout curtain

Original upstream slope

© LNEC 2012

Jan. 1977

© LNEC 2012

Parede auto-endurecedora - Abril 1978

© LNEC 2012

Maio 1978

© LNEC 2012 Agosto 1980

© LNEC 2012

> Crestuma-Lever Dam

APPLICATIONS

Self-hardening

slurry wall

Self-hardening

slurry wall

Diaphragm wall

Diaphragm wall

Power plant

Crest

Navigation lock Fish

ladder

Douro

Riv

er

Stilling basin

Alluvial soil

Rock formation

Diaphragm cut-off wall

Self-hardening slurry wall

Plan Cross-section


Wall area: 5 600 m2

Maximum depth: 40 m

Width: 0,8 m

© LNEC 2012

> Águas Industriais Dam

APPLICATIONS

Plan

Cross-section


Wall length: 175 m

Maximum depth: 14 m

Width: 0,4 m

Bottom outlet

Original dam

Downstream

slope

Self-hardening

slurry wall

Original dam profile

Crest Heightening

section

Self-hardening

slurry wall

© LNEC 2012



APPLICATIONS

Phase 1

Phase 2

Phase 3

Phase 4

Phase 5

Etapa 6

P – Primary panel

S – Secondary panel

S

P

P

S

S

P

P P

P P

P

P

P

P

P

S P

P

P

P

P

Alternating panel method.

© LNEC 2012



APPLICATIONS


Continuous trenching method.

© LNEC 2012



APPLICATIONS


Continuous trenching method.

Structural diaphragm wall traditional

method.

© LNEC 2012

> Self-hardening slurry features.

> Self-hardening slurry composition.

> Chemical reactions between water, cement

and bentonite.

Water: 1 m3

Cementitious material: 100 to 350 kg

Bentonite: 30 to 60 kg

SELF-HARDENING SLURRY CHARACTERIZATION

Water (100%)

Cement

(100%)

Bentonite

(100%)

Non-setting

slurries

Semi-fluids

Cut-off slurries

Bleeding

slurries

© LNEC 2012


CAH gel

Flocculated bentonite

particles

CSH gel

Ca2+ ions release

Hydrous silica and

alumina

(OH)- ions release

CSH gel

CAH gel

Hydrated lime

Ca(OH)2

Cement Bentonite suspension

Cement hydration

Dissociation of

hydrated lime

pH rise Modification of

the adsorbed cation

population

Dissolution of

silica and alumina of

bentonite

Shrinkage of

bentonite diffuse

double layer

Interparticle bonding

Primary

cementitous

products

Secondary

cementitous

products

Bentonite-cement clusters

Reaction with Ca2+

and (OH)- ions



> Chemical reactions between water, cement and bentonite:

© LNEC 2012



> Chemical reactions between water, cement

and bentonite.


Pores

Smectite

particle

Slag

particle

Nucleus

CSH gel

(high density)

Clinker

particle CSH gel

(low density)

Bentonite-cement cluster

© LNEC 2012

PROCESSES INVOLVED IN THE FORMATION OF THE CUT-OFF WALL

MATERIAL

VISCOUS

FLUID

(Trench walls

support)

VISCOUS

SOLID

(Time-

depending

properties)

Soil contamination

Penetration

Filtration

Penetration and

Filtration

Sedimentation

Cement hydration

Self-weight

consolidation

Cement hydration

Pozzolanic reactions

EXCAVATION

PHASE

FIRST HOURS

POST-EXCAVATION

FOLLOWING

PERIOD

Cement Setting

© LNEC 2012

EXPERIMENTAL WORK

> Objectives. Identify and quantify the influence of the

slurry composition, and mixing proce-

dures upon the rheological behaviour of

the fresh slurry.

Identify and quantify the influence of the

slurry composition, spoil contamination,

curing time and surcharge loads upon the

physical, mechanical and hydraulic

behaviour of the hardened slurry.

© LNEC 2012

EXPERIMENTAL WORK

> Experimental work description:

Rheological characterization of self-

-hardening slurries.

Characterization of the “cake” formed by

filtration.

Bleeding evolution of self-hardening

slurries.

Physical characterization of hardened

slurry samples.

Compressibility and threshold stress of

hardened slurry samples.

Strength and deformability of hardened

slurry samples.

Permeability of hardened slurry samples. Slurry composition

35 kg bent. + 150 kg cement



Marsh viscosity

47 s

49 s

105 s

Marsh funnel and cup

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.

Permeability of hardened slurry samples.

Fann viscometer

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.


Slurry composition




Viscosity Gel strength

8.0 cP 4.1 Pa

9.5 cP 4.6 Pa

12.5 cP 5.1 Pa

0

5

10

15

20

25

30

35

0 200 400 600 800 1000 1200

Shear rate (s-1)

Sh

ear

str

ess (

Pa)

35 kg bentonite + 200 kg cement



© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.


Filter press

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.



0

50

100

150

200

250

0 2 4 6 8 10

Sqrt[time] (min0.5)

Fil

trate

lo

ss (

ml)

“Cake” permeability: 1.7x10-8 m/s

“Cake” unit mass: 1240 kg/m3

“Cake” water content: 122%

“Cake” void ratio: 3.0

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.


Composition Bleeding

35 kg bent. + 150 kg cement: 6%

35 kg bent. + 200 kg cement: 5 to 6%

50 kg bent. + 200 kg cement: 2%

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.


Slurry composition









Unit mass (average)

1145 kg/m3

1155 kg/m3

1165 kg/m3

Water content

395 to 445%

305 to 350%

300 to 325%

wL IP

128% 22%

151% 38%

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.



0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 10 100 1000 10000

Vertical pressure (kPa)

e /

e0

4 weeks curing

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.



0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 10 100 1000 10000


e /

e0

4 weeks curing 8 weeks curing

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.



0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1 10 100 1000 10000


e /

e0

4 weeks curing 8 weeks curing

12 weeks curing

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.


0

50

100

150

200

250

300

350

400

0 2 4 6 8 10 12 14 16 18 20

e (%)

s' 1

- s

' 3 (

kP

a)

0

50

100

150

200

250

300

350

400

DU

(kP

a)

24 kPa 100 kPa 400 kPa

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.


0

100

200

300

400

0 100 200 300 400

p' (kPa)

s' 1

- s

' 3 (

kP

a)

M=2.5

M=

0

20

40

60

80

100

0 20 40 60 80 100

p' (kPa)

s' 1

- s

' 3 (

kP

a)

M=2.5

M=3

M=3

© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.

Permeability of hardened slurry samples. 1.E-09

1.E-08

1.E-07

1.E-06

3 5 7 9

Void ratioC

on

du

cti

vit

y c

oeff

icie

nt

(m/s

)




© LNEC 2012

EXPERIMENTAL WORK





filtration.


slurries.


slurry samples.




slurry samples.

Permeability of hardened slurry samples. 1.E-09

1.E-08

1.E-07

1.E-06

0 200 400 600 800 1000

Effective stress (kPa)C

on

du

cti

vit

y c

oeff

icie

nt

(m/s

)




© LNEC 2012

FUTURE RESEARCH

> Feasibility study regarding the use of

piezocone penetration tests for assessing the

integrity of self-hardening slurry cut-off walls,

but also for determining permeability, strength

and compressibility of the slurry "in situ".

> Feasibility study regarding the use of geophy-

sical tests in assessing the integrity of self-

-hardening slurry cut-off walls and also in the

characterization of its permeability.

> Sedimentation and self-weight consolidation

analysis of self-hardening slurries using a

consolidation column equipped with a gamma

densimeter.

> Detailed study concerning the influence of

slurry setting upon the development of slurry

filtration, penetration and sedimentation

processes.

© LNEC 2012 © LNEC 2012

Research Perspectives in Embankment Dams at LNEC

Laura Caldeira [email protected]

Head of the Geotechnique Department

Dam World Conference 8-11th October 2012 Maceio, Brazil

Documents

Research Perspectives in Embankment Dams at LNEC · © LNEC 2012 © LNEC 2012 Research Perspectives in Embankment Dams at LNEC Laura Caldeira [email protected] Head of the Geotechnique