CIEM 5740

COMPUTER METHODS FOR SLOPE ENGINEERING

1 – Introduction & Soil Strength

Jidong Zhao
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 2

Lecture & Format

• Instructor:
• Jidong Zhao, Room 4606, Email: jzhao@ust.hk

• Lecture schedule
• Friday, 7:00 pm – 10:00 pm
• Venue: 3207 (Lift 21)
• Tutorials uses the same venue and time

• Lecture format
• Lecture slides
• Demonstration Q&As
• Note taking and Self study
• Extra reading material after class

• Consultation
• During class breaks or make an appointment with me by email
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 3

Assessments & Weightings

• Assignments (50%)
• Five assignments
• Each takes a weighting of 10%

• Group term projects (40%)

• Free grouping to around 8 groups
• Grouping in CANVAS
• To be announced in CANVAS

• Term project presentation (10%)

• Group presentation
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 4

Course Description
• This course aims to teach students to apply the fundamental
knowledge of soil mechanics and numerical methods for
analysis and design of slope stability.

• The course covers mechanics of slope failures, shear strength

of soils, transient seepage analysis, major methodologies on
slope stability analysis, and use of popular computer software
programs for slope stability analysis, including SLOPE/W,
SEEP/W & PLAXIS 2D.

• The course include tutorials on hand-on experience using

computer software on slope stability analysis.

• Preferred background (not compulsory)

• Soil mechanics, Geotechnical Analysis and Design, Numerical Methods
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 5

Course Outline & Schedule

1. Introduction & Soil Strength
2. Methods of Slope Stability Analysis
3. Use of SLOPE/W for Slope Stability Analysis
4. Steady Seepage Analysis
• Basics & Use of SEEP/W & SLOPE/W for Slope Stability Analysis
5. Transient Seepage Analysis:
• Basics & Use of SEEP/W & SLOPE/W for Stability Analysis of Slopes subject to
rainfall infiltrations and Embankments under rapid drawdown
6. Design of Slope Stabilization Measures:
• Use of SLOPE/W for Slope Reinforcement
7. 2D Finite Element Method & PLAXIS 2D
8. PLAXIS 2D for Stability Analysis of Embankment
9. PLAXIS 2D for Tunnelling Analysis and others
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 6

Textbooks and References

• No Textbook
• Lecture notes will be provided
• All course materials (Syllabus, lecture notes, assignments, supporting files
etc.) will be provided in CANVAS, if not stated otherwise
• Logon http://canvas.ust.hk/ using your student ID and password
• Choose CIEM5740 from the course list
• Check the announcements from time to time

• References
• J.M. Duncan, S.G. Wright, Soil Strength and Slope Stability. John Wiley & Sons
(Imprint: Hoboken, N.J.), 2005. (Available online at UST library)
• E.N. Bromhead, The Stability of Slopes. Blackie Academic & Professional,
1992.
• GEO-SLOPE International Ltd. SEEP/W and SLOPE/W Manuals. www.geo-
slope.com
• PLAXIS 2D Manual. www.plaxis.nl
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 7

This Lecture – Introduction & Soil Strength

• 1 - Introduction
• The stability and instability of slopes

• Past experience on landslides

• Definition/classifications

• Typical configuration/velocity classes

• Causes/triggering mechanisms of slope failures

• Historical development of slope analysis

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 8

Landslide
• Definition
• The movement of a mass of rock
debris or earth down a slope
• Mass movements take place in
many different materials and
have widely varying shapes

• Classification
• Mass movements can be defined in
terms of
• Material involved
• Rock, debris or earth
• Debris: assembly of fragments of soft
rocks or hard soil
• Type of movement taking place
(kinematics)
• Falls, topples, slides, spreads or
flows
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 9

Landslide Velocity Scale

(Cruden & Varnes, 1996, Landslide types and processes, TRB, 247)

Class 1: Movements are not noticeable without instruments; construction possible

with precautions
Class 7: Movements are catastrophe of major violence; buildings destroyed by impact
of displaced materials; many deaths; escape unlikely.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 10

Introduction - Past landslide Events

• Rossberg Landslide near Goldau (Switzerland, 1806)
• Rapid rock fall/slide caused by heavy rainfall, killed 457 people
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 11

Introduction - Past landslide Events

• Po Shan Road Landslide (Hong Kong, 1972)
• Soil slope failure/debris flow, killed 76 people,
• Leading to the establishment of Hong Kong GEO in 1977
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 12

Introduction - Past landslide Events

• Po Shan Road Landslide (Hong Kong, 1972)
• Soil slope failure/debris flow, killed 76 people,
• Leading to the establishment of Hong Kong GEO in 1977
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 13

Introduction - Past landslide Events

• Vaiont Landslide (Italy 1964) (Rock slide)
• Rock slide rushed into Vaiont reservoir and caused over-spilling
• Water all over 60 m high, sweeping all nearly villages, killed 2600 people
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 14

Introduction - Past landslide Events

• La Conchita Landslide, CA, USA
• A slump was formed in 1995
• Major slide occurred in 2005 after heavy rainfall, killed 10 people
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 15

Introduction - Past landslide Events

• La Conchita Landslide, CA, USA
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 16

Introduction - Past landslide Events

• La Conchita Landslide, CA, USA, 2005
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 17

Sources of Landslides - illustration

 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 18

Causes of Slope Failure

tf
FS S =
td
ttFSfs = Factor of safety with respect to strength
FS = f = Shear strength of the soil/rock
FS S = t = Shear stress developed along the potential failure surface
S
td d

FSs = 1.0: Impending failure

= 1.5: acceptable for the design of a stable slope

Terzaghi divided landslide causes into two:

(1) Internal causes: Reduction in the shear strength of the soil
(2) External causes: increase in the shear stress
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 19

Internal causes – Reduction of shear strength

1. Increased pore pressure (reduced
effective stress)
• Change of groundwater levels due to
• Intense rainfall
• Rapid snowmelt

t f = c¢ + s ¢ tan f ¢
 20

Internal causes – Reduction of shear strength

2. Swelling of soil
• Leads to dilation and increase in void ratio
• Frequently happens in heavily over-consolidated clays subjected to
long time of access to water under low confining pressures

3. Cracking of soil
• Loss of tensile strength

t f = c¢ + s ¢ tan f ¢
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 21

Internal causes – Reduction of shear strength

4. Development of slickensides
• Parallel striations on rock surfaces
produced by relative motion across
opposite sides of fault planes

5. Decomposition of clayey rock fills

6. Creep under sustained loads

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 22

Internal causes – Reduction of shear strength

7. Leaching

8. Weathering
• Mechanical weathering
• Thermal/frost/hydraulic/biological
• Chemical weathering
• Dissolution/hydration/hydrolysis
/Oxidation/Biological/Carbonation

9. Cyclic loading
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 23

External causes – Increase in shear stress

1. Loads at the top of the slope

2. Water pressure in cracks at the top of the slope

3. Increase in soil weight due to increased water

4. Excavation at the bottom of the slope

5. Drop in water level at the base of a slope

6. Earthquake shaking
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 24

External causes – Increase in shear stress

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 25

2 – Soil Strength
• Mohr-Coulomb failure criterion
• Frictional resistance
• Angle of shear resistance /Angle of internal friction
• Apparent cohesion / True cohesion
• Total stress and effective stress
• Drained and un-drained conditions/strength
• Tri-axial Compression Test
• Mohr Circle representation of stress state at failure
 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 26

Mohr-Coulomb Failure Criterion

• Mohr-Coulomb (MC) failure criterion
t = c + s tan f
t = c + s tan f compressive stress
Normal
t = c +Cohesion
s tan f (apparent)
= c + s tan f Friction Angle
• Angle of shearing resistance according to Terzaghi

t = c + s tan f
t = c + s tan f
 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 27

Mohr-Coulomb Failure Criterion

t = c + s tan f
t = c +s tan f tand
• Both + sNOT
= c are f constants. They depend on the initial state
tansoil
of the soil and the type of loading.
= c +s • f - Clay: up to 20 degrees; loose sand: up to 35 degree.
tan

t
t = c + s tan f
t = f (s )

s
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 28

Mohr-Coulomb Failure Criterion

• The combined normal and shear stresses on a plane in a soil mass
are such that they can be plot as
• Point A: shear failure will not occur along the plane
• Point B: shear failure will occur along the plane
• Point C: this stress state cannot exist
• Shear failure would have occurred before this condition is reached
t
C
Failure envelope
B

s
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 29

Cohesion-less Soil
• Sand/Granular material: cohesion-less
t = s tan f
• The shear resistance depends only on the normal stress on
the potential surface of sliding and is provided solely by
friction
• Frictional resistance

• f : Angle of internal friction

• Can be regarded constant which is independent on the stress
rates/level.
• Its value depends on the initial density of the sand: 30-50
degrees.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 30

Friction Angle

Dilation and shear strength

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 31

True Cohesion
• True cohesion
• Results from adherence between particles in the absence of any
externally applied or self-weight forces.
• Involves the breakup of
• short-range and cementation bonds (up to 500kPa),
•
• May be lost through solution.
• And/or electrostatic and electromagnetic attractions (e.g., stiff OC clay 5-
24kPa)
• May be lost through weathering.

• And/or root cohesion

• May be lost by logging or fire.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 32

Apparent Cohesion
• Involve capillary stresses due to the surface tension
properties of water and apparent mechanical forces
due to interlocking rough surfaces on the failure plane.

• Unsaturated soils
• As soil dries out, water menisci form at grain contacts. These are
under negative capillary pressure. When capillary forces are netted
over a unit area, we refer to “matric suction stress”, which creates
inter-particle forces to be the apparent cohesion.
• May be lost during wetting.

• Due to pore pressure during fast (undrained) loading.

• The sources of apparent cohesion cannot be isolated in the current
testing methods, and its magnitude can only be inferred from the
experimental results.

• Though there is no direct relation between the two cohesions other

than the name, Terzaghi argued that the true cohesion is never smaller
than the apparent cohesion. This, however, does not apply to dense
sand.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 33

Total Stress & Effective Stress

• Saturated soil
• The total amount of compressive stress applied to the sample does not
produce a measurable compression of the soil layer.

• Total stress
• The sum of all forces, including those transmitted through inter-particle
contacts and those transmitted through water pressures, divided by the total
area (both the area of voids and the area of solid matrix)

• Effective stress
• Only includes the forces that are transmitted through particle contact
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 34

Total Stress & Effective Stress

• The total compressive stress in a saturated soil consists of two parts
with different mechanical effects:

• Pore-water pressure
• Effective stress: the part of stress producing effects such as compaction or
an increase of the shearing resistance, grain-to-grain

s = s ¢ + uw
• Influence of pore-water pressure on the stress-strain relation and shear
resistance can be investigated by triaxial compression tests.
• Direct measurement of total stress and pore-water pressure
 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 35

Effect of Water Water

buoyant effect

Negative pore pressure in unsaturated soils/rocks

Positive pore pressure in saturated soils/rocks
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 36

Drained/undrained Conditions
• Drained condition
• Under which water is able to flow into/out of soil in the duration that
the soil is subjected to change of load.
• Changes in the load do NOT cause changes in the water pressure

• Un-drained condition
• Under which there is no flow of water into/out of the soil mass in the
duration of the soil is subjected to change of load.
• Changes in the loads DO cause changes in the water pressure in the
voids.
• The volume of the sample can be regarded constant.

• Major difference in practice

• Time of load change with respect to that of water flow in/out of the
soil.
• The soil strength differs significantly under the two conditions.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 37

Soil Strength
• Soils are essentially frictional materials

• The strength depends on the applied stress

• Strength is controlled by effective stresses

• Pore-water pressures need to be considered

• Soil strength depends on drainage

• Different strengths will be measured for a given soil that

(a) deforms at constant volume (undrained) and
(b) deforms without developing excess pore pressures (drained)
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 38

Drained Strength
• Drained strength is the strength of the soil when it is loaded
slowly enough so that no excess pore pressures are induced by
applied loads.

• In the field drained conditions result when loads are applied

slowly to a mass of soil, or where they persist for a long enough
time so that the soil can drain.

• In the lab, drained conditions are achieved by loading test

specimens slowly so that excess pore pressure do not develop
as the soil is loaded.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 39

Volume Changes During Drained Shear

• Contraction/Dilation depends on the density and effective stress that
confines the soil
• In loose soils the void tends to be filled up be particles such that the
sample appears to be contractive

• In dense soils the particles are packed tightly together, which results in a
great deal of interference between particles when they move relative to
one another

• In very dense soils, particles cannot move relative to each other unless
they ride up over each other, which cause dilation.

• Higher effective stresses tend to prevent dilation, because work is

required to cause the soil to expand against the effective confining
pressure.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 40

Pore Pressure Response

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 41

Effective Stress Failure Criterion

• t = c¢ + s ¢ tan f ¢
c¢ and f ¢ are effective (drained) strength parameters

• Soil behaviour is controlled by effective stresses, and the

effective strength parameters are the fundamental strength
parameters. But they are not necessarily soil constants.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 42

Undrained Strength
• The strength of the soil when loaded to failure under un-
drained conditions.

• In the field, un-drained conditions result when loads are

applied to a mass of soil faster than the soil can drain.

• In the lab, un-drained conditions are achieved by loading test

specimens so rapidly that they cannot drain, or by sealing them
in impermeable membrances.
 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 43

Total Stress Failure Criterion

• If the soil is taken to failure at constant volume (undrained)
then the failure criterion can be written in terms of total stress
as
t = cu + s tan fu
• cu and fu are un-drained strength parameters. They are not
constant either and depend strongly on water content of the
soil.

• The undrained strength is only relevant in practice to clayey

soils that in the short term remain undrained. Note that as the
pore pressures are unknown for undrained loading the effective
stress failure criterion cannot be used.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 44

Tests to measure Soil Strength

Direct Shear Test
Top platen Normal load

Load cell to
Motor measure Shear
drive Force

Soil

Porous plates

Rollers

Measure relative horizontal displacement, dx

vertical displacement of top platen, dy
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 45

Direct Shear Test

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 46

Direct Shear Test

• Usually only relatively slow drained tests are performed in
shear box apparatus. For clays rate of shearing must be chosen
to prevent excess pore pressures building up. For sands and
gravels tests can be performed quickly
• Tests on sands and gravels are usually performed dry. Water
does not significantly affect the (drained) strength.
• If there are no excess pore pressures and as the pore pressure
is approximately zero the total and effective stresses will be
identical.
• The failure stresses thus define an effective stress failure
envelope from which the effective (drained) strength
parameters c’, f’ can be determined.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 47

Typical Direct Shear Results

t
tp
Dense tp
Medium tr
Ultimate
Loose
Exp

Shear displacement
Volume change

Dense
Medium

Shear displacement
Loose
Comp
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 48

Direct Shear Test - Advantages

• Easy and quick test for sands and gravels

• Large deformations can be achieved by reversing shear

direction. This is useful for determining the residual strength of
a soil

• Large samples may be tested in large shear boxes. Small

samples may give misleading results due to imperfections
(fractures and fissures) or the lack of them.

• Samples may be sheared along predetermined planes. This is

useful when the shear strengths along fissures or other
selected planes are required.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 49

Direct Shear Test - Disadvantages

• Non-uniform deformations and stresses in the specimen. The
stress-strain behaviour cannot be determined. The estimated
stresses may not be those acting on the shear plane.

• There is no means of estimating pore pressures so effective

stresses cannot be determined from undrained tests

• Undrained strengths are unreliable because it is impossible to

prevent localised drainage without high shearing rates

• In practice shear box tests are used to get quick and crude
estimates of failure parameters.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 50

Tests to Measure Soil Strength

Triaxial Compression Test
Deviator load
Confining cylinder

Cell water
Rubber membrane

O-ring seals

Soil
Porous filter disc

Cell pressure
Pore pressure and
volume change
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 51

Tri-axial Compression Test

CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 52

Stresses in Tri-axial Tests

F = Deviator load
sr

sr sr = Radial stress (cell

pressure)

sa = Axial stress

F
From equilibrium we have sa = sr +
A
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 53

Stresses in Tri-axial Tests

• F/A is known as the deviator stress, and is given the symbol: q
q = ( s a - s r ) = ( s1 - s 3 )
• The axial and radial stresses are principal stresses
• If q = 0 increasing cell pressure will result in
• Volumetric compression if the soil is free to drain. The effective stresses
will increase and so will the strength
• Increasing pore water pressure if soil volume is constant(that is,
undrained). As the effective stresses cannot change it follows that Du =
Dsr
• Increasing q is required to cause failure

Can a soil fail by increasing p but keeping q = 0?

 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 54

Strain in Tri-axial Tests

From the measurements of change in height, dh, and change in volume dV we
can determine
dh
Axial strain ea = -
h0
dV
Volume strain eV = -
V0

where h0 is the initial height and V0 is the initial volume

It is assumed that the specimens deform as right circular cylinders. The cross-
sectional area, A, can then be determined from
æ dV ö
ç 1 + ÷
V æ 1 - evö
Ao ç ÷
0
A = = Ao ç ÷
ç 1 + dh ÷ è 1 - ea ø
ç ÷
è h0 ø
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 55

Types of Tri-axial Tests

• UU (unconsolidated undrained) test
• Cell pressure applied without allowing drainage. Then keeping cell
pressure constant increase deviator load to failure without drainage.

• CIU (isotropically consolidated undrained) test

• Drainage allowed during cell pressure application. Then without allowing
further drainage increase q keeping sr constant as for UU test.

• CID (isotropically consolidated drained) test

• Similar to CIU except that as deviator stress is increased drainage is
permitted
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 56

Tri-axial Tests - Advantages

• Specimens are subjected to (approximately) uniform stresses
and strains

• The complete stress-strain-strength behaviour can be

investigated

• Drained and undrained tests can be performed

• Pore water pressures can be measured in undrained tests,

allowing effective stresses to be determined

• Different combinations of cell pressure and axial stress can be

applied
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 57

Typical Tri-axial Test Results

q

Increasing cell
pressure

ea
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 58

Mohr Circle
• To relate strengths from different tests we need to use some
results from the Mohr circle transformation of stress.

t
t = c + s tan f

c s
s3 s1

• The Mohr-Coulomb failure locus is tangent to the Mohr circles

at failure
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 59

Mohr Circle
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 60

Mohr Circle
t
æp fö
(ta, sa) a = ç - ÷
è4 2ø

(s 1 + s 3 ) ( s1 - s 3 )
sa = - cos 2a
2 2
f 2a s

s3 (s 1 + s 3 ) ( s 1s1- s 3 )
sa = - cos 2a
2 2

(s 1 + s 3 ) ( s1 - s 3 )
sa = - cos 2a
2 2
( s1 - s 3 )
ta = sin 2a
2
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 61

Mohr Circle
• The Mohr circle construction enables the stresses acting in different
directions at a point on a plane to be determined, provided that the
stress acting normal to the plane is a principal stress.

• The construction is useful in Soil Mechanics because many practical

situations may be approximated as plane strain.

• The sign convention is different to that used in Structural analysis

because it is conventional to take compressive stresses positive

• Sign convention:
• Compressive normal stresses positive
• Anti-clockwise shear stresses positive (from inside element)
• Angles measured clockwise positive
 CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 62

Mohr-Coulomb Failure Criterion

t

f c s
s3 s1

c cot f p

Failure occurs if a Mohr circle touches the failure criterion. Then

R = sin f ( p + c cot f )

s 1 + c cot f =
1 + sin f
=
ép2
+
fù
=
tan ê ú Nf
s 3 + c cot f 1 - sin f ë4 2û

s1 = Nf s 3 + 2 c Nf
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 63

Effective Stress Criterion

• Effective strength parameters are the fundamental parameters.
The Mohr-Coulomb criterion must be expressed in terms of
effective stresses

t = c' + s ¢n tan f' 1 + sin f ¢

Nf =
s 1¢
1 - sin f ¢
= N f s ¢3 + 2 c ¢ Nf

s ¢n = s n - u
s1¢ = s1 - u

s ¢3 = s 3 - u
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 64

Effective and Total Mohr Circle

t

s
s ¢3 s 1¢ s3 s1

u
u

For any point in the soil a total and an effective stress Mohr
circle can be drawn. These are the same size with
s 1¢ - s ¢3 = s 1 - s 3
The two circles are displaced horizontally by the pore
pressure, u.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 65

Interpretation of Laboratory Tests

• Drained shear loading
• In laboratory tests the loading rate is chosen so that no excess water
pressures will be generated, and the specimens are free to drain.

• Effective stresses can be determined from the applied total stresses and
the known pore water pressure.

• Only the effective strength parameters c’ and f’ have any relevance to

drained tests.

• It is possible to construct a series of total stress Mohr circles but the

inferred total stress (undrained) strength parameters are meaningless.

At the same confining pressure, can we obtain the

strength parameters by drained triaxial tests?
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 66

Interpretation of Laboratory Tests

• Effective strength parameters are generally used to check the
long term stability (that is when all excess pore pressures have
dissipated) of soil constructions.

• For sands and gravels pore pressures dissipate rapidly and the
effective strength parameters can also be used to check the
short term stability.

• In principle the effective strength parameters can be used to

check the stability at any time for any soil type. However, to do
this the pore pressures in the ground must be known and in
general they are only known in the long term.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 67

Interpretation of Laboratory Tests

• Undrained loading

• In undrained laboratory tests no drainage from the sample must occur,

nor should there be moisture redistribution within the sample.

• In the shear box this requires fast shear rates. In triaxial tests slower
loading rates are possible because conditions are uniform and drainage
from the sample is easily prevented.

• In a triaxial test with pore pressure measurement the effective stresses

can be determined and the effective strength parameters c’, f’ evaluated.
These can be used as discussed previously to evaluate long term stability.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 68

Interpretation of Laboratory Tests

• The undrained tests can also be used to determine the total (or
undrained) strength parameters cu, fu. If these parameters are to be
relevant to the ground the moisture content must be the same. This
can be achieved either by performing UU tests or by using CIU tests
and consolidating to the in-situ stresses.

• The total (undrained) strength parameters are used to assess the

short term stability of soil constructions. It is important that no
drainage should occur if this approach is to be valid. For example, a
total stress analysis would not be appropriate for sands and gravels.

• For clayey soils a total stress analysis is the only simple way to assess
stability.

• Note that undrained strengths can be determined for any soil, but
they may not be relevant in practice.
CIEM 5740 – Computer Methods for Slope Engineering – Hong Kong University of Science and Technology 69

Significance of un-drained strength parameters

• It is often found that a series of undrained tests from a
particular site give a value of fu that is not zero (cu not
constant). If this happens either
• The samples are not saturated, or
• The samples have different moisture contents

• If the samples are not saturated analyses based on undrained

behavior will not be correct

• The undrained strength cu is not a fundamental soil property. If

the moisture content changes so will the undrained strength.