Geotechnical Engineering I: Soil Classification
Geotechnical Engineering I: Soil Classification
Soil Classification
- Moisture
- Voids Ratio, e = void volume to solid volume
- Specific Volume = v = volume to solid volume = 1 + e
- Saturation Ration Sr = water volume to void volume
- Water Content = w = water mass to solid mass
- Relative Density ID = fraction of way from emax to emin
- Colour
- Consistency
- Structure
- Soil Type
- Particle Size by sieving or water column (size ~ terminal velocity)
- % smaller than: ~void size = d10, ~floater size = d60
- Uniformity Coefficient UC = d60/d10 (uniform<4, well graded>6)
- Plastic Fines electrostatically stick to water, can mix with lots of water and be stable
- Non Plastic Fines held by surface tension, turn to slurry with lots of water
- Liquid Limit LL = w when soil transitions plastic to ~liquid
- Plastic Limit PL = w when soil transitions brittle to plastic
- Plasticity Index PI = LL - PL
-Other Features
- Origin
- Mineralogy
- Density, normalised by water density = Gs = 2.65 for silica so good guess
- Groundwater
Effective Stress
Loads in soil carried by fluid and grains, effective stress represents stress at particle contacts, which
governs behaviour- strength/stiffness. Strength of soil originates from both friction and interlocking.
σ = σ’ + u
Seepage
The pressure head at a point in fluid is the height above that point the fluid will reach if a standpipe is
inserted at that location.
The potential head is the height of the top of the same standpipe wrt a fixed datum- differences in this
govern flow.
Deriving from pressure drop in pipes, and likening to capillaries in soil, we get Darcy’s Law:
v = -k (γw/μ) dh
/ds = -K dh
/ds
K = 0.01 d102
Darcy’s Law can be applied in 3D vector form by separating velocity and K (may be anisotropic) into
components. If K is constant/isotropic, seepage can be calculated by flownets.
Artificial Soil
Graph:
- Start at given w , e
- Compaction causes vertical increase (constant w)
-Adding/Removing Water causes horizontal
movement (adding is easier)
If soil is compacted to state with Sr < 75%, it may collapse on wetting, if it is utilising its menisci surface
tension strength. Can be seen as reducing compaction (distance to after compaction curve).
Soil Compression
Consolidation
If pore pressures not equillibrium, groundwater will flow. If flow in and out of soil element aren;t equal,
element undergoes volumetric strain. If a saturated soil undergoes instantaeous increase in total stress, it
cannot be carried by effective stress (this would lead to volumetric strain, but water is incompressible)-
must be carried by pore pressure.
Oedometers can measure consolidation rate, to assess what would happen in field:
- Soil disc with porous top and bottom plates loaded with total stress = σv’ equivalent to initial
field value
- Left to settle before adding load increment Δσv = expected Δσv’ in the field at t=0
- Compression measured at intervals until finished
Isochrones show change of excess pore pressure over time for primary consolidation (due to excess pore
pressures, secondary due to creep):
By
considering a control volume with pore pressure (therfore potential head) gradient, using Darcy’s Law for
in and outflow, then relating this volumetric strain to effective stress rise (pore pressure drop), we get
Terghazi 1D consolidation equation:
There are two stages of primary consolidation, shown below. Both c be approximated as a parabola to
simplify, some rules for parabolas are:
Both cases can be solved by forming two equations and applying parabola rules:
- Integrating change in height for small element
- Darcy’s Law at surface
Cv scales with E0 which scales with λ. Therefore, swelling and non-virgin compression (κ instead of λ) is
much faster. Smaller strains mean less water to be expelled.
In layered ground, each layer can be analysed separately- total settlement at given time = sum of all
individual settlements at that time. If Tv for one is much lower than the other, full consolidation occurs for
one much faster- computation can be minimised.
Removing water also causes compression- it lowers the water table and hence increases effective stresses.
Soil Strength
Coulomb’s law details strength arising due to friction: F = μ R ~~ τ = σ’ tanφ’
Failiure occurs when τ exceeds shear strength (function of σ’ ) i.e. when Mohr’s circle crosses line. The
angle of failiure plane can be found by plotting a pole on the Mohr’s circle, tends to be:
- 45 + φ/2 for ‘active’ (vertical stress larger)
- 45 - φ/2 for ‘passive’ (horizontal stress larger)
Triaxial Tests:
- Apply confining pressure everywhere to cylinder, increase axial (deviatoric) stress until failiure,
measuring axial force, radial and axial strain
For Triaxial tests, drainagae may be open or closed while increasing cell pressure, and while apply axial
stress (Consolidated/Unconsolidated), or while increasing axial stress (Drained/Undrained)
CD on sands:
- Cell (confinnig) pressure increased to desired level with valve open to consolidate
-Axial load (deviatoric stress = 2 * max shear) applied, axial and volumetric strains recorded at each load
-Following profile obtained:
For loose sands, as strain increases, force increases as more
grains become mobilised and need to be sheared for higher
strains. Also as strain increases, particles are pushed into gaps
(those in gaps can be pushed out also), so volume decreases.
Rate of pushing into gaps decreases until it = rate of pushing
out of gaps (dilation), all grains are mobilised so force is
constant. This in known as critical state, and is considered
failiure.
For dense sands, initially last few particles are pushed into gaps,
causing volume decrease. Again, higer strains mean more
mobilised grains so stress increases. Eventually, particles begin
to dilate, and when net dilation-gapfilling becomes positive,
volume increases. Amount of dilation starts to increase, and at
its max rate, force is at its highest. This is known as peak
strength and considered failiure. After this, it becomes easier to
shear, and when rate of dilation decreases to = rate of filling,
critical state is reached.
At high confining pressure, dilation is less favourable, critical state is denser, so even dense sands become
looser than critical., hence the following failiure criterion for a particular dense sand:
CD on clay:
For sand, due to low compressibility, initial density changes little with confining stress, so has a big
influence on strength, as well as confining pressure. For clays, density is less useful, as it changes a lot
during compression, so OCR is the more relevant paramenter.
For a clay sample, if confining stress is reduced and it becomes more OC, it will not expand very much,
so will be denser than critical at that stress. If it is loaded and becomes more NC, critical state will be
denser, but it will not compress very much, so it will be ‘looser’ than critical.
CU on clay:
Volume can’t
change so excess u
scale swith how v
wants to change.
As can be seen from the graph above, increasing confining stress (more NC) increases Su. The Su –
confining stress curve takes a similar shape to the Mohr-Coulomb failiure surface above (curved then
linear for NC).
UU on clay:
For a given clay, shear strength always constant, as confining pressure always taken by water, so effective
stress always the same. Shear strength found is Su.
Undrained Footings
In drained analysis, strength is considered φcrit’ is used everywhere as a safe design parameter. In
undrained analysis, Su can considered to be constant everywhere. In reality, it may vary with depth, but a
design decision must be made to assess significance (e.g. If clay has high OCR, slightly higher confining
stress lower down will not affect the overall OCR that much, so strength will not change significantly).
Upper bound = if compatible mechanism found where work = dissipation, collapse must occur and loads
are upper bound to true collapse loads
Lower bound = if internal stress distribuion found which satisfies external loads, collapse can’t occur and
loads are lower bound to true collapse loads
Stress disconinuity is sudden change in stress parallel to discontinuity. Mohr’s cirlce geometry can be
used to relate rotation of principle stresses to mean stress difference and angle of discontinuity stress (data
book). Perpendicular equillibrium maintained, so (like refraction) if major prinicple stress (velocity) is
increasing, bends away from normal.
Solved as in IB, displacement diagrams then equate WD to ED for Nc = 5.77 and 5.52 respectively. Also
Prandtl solution – slip fan:
WD = qf B δv
ED = sliding at blocks E, B and H
+ sliding along circumference of
each fan
+ sliding on internal radii
Combined Loading:
Under V-H loading, the two slip fans are of different sizes. An inclined load means inclined principal
stresses under the footing.More slip fans are therefore required in one direction (larger principal stress
rotation).
- Therefore soil on right must be mobilising less than su, su,mob = su (π/2 – ψ)/( π/2 + ψ) from above
equation
- Other failiure surface due to sliding, when shear ecseeds su
V/Vult = 0.5 +0.5 sqrt(1-H/Hult ) (stress fan - V/Vult > 0.5 – simple approximation to derived form)
Under V-M loading, M is converted into eccentricity (e=M/V), and conservatively, only symmetrical
footing about eccentric load is considered (B becomes B-2e). Prantl solution can be used on effective
area:
Normally, V,M and H must be considered together. If multiple foundations, M can be ignored and assume
differences in V/H produce required couple.
Shape:
Shape factor Sc is applied to Nc = 5.14 to increase Vult. For circular foundations, shape factors for smooth
and rough (more common) are 1.11 and 1.18 respectively, so are stronger than strips. Square foundations
are treated as circles with same area, so Sc = 1.18. Rectangles use relationship Sc = 1 + 0.18 B/L to scale
between strip and square.
Embedment:
Conservative approach treats soil above base as surcharge only. In reality, mobilising strength of
embedding soil leads to embedment factor dc, which varies with h/D (experiments have been done which
suggest dc = 1 + 0.33 arctan (h/D), but may be conservative).
Strength Heterogeneity:
Su may increase with depth, in which case capacity will e higher. To analyse, use dimensionless gradient
κ = kB/su0 where k = rate of increase of su with depth, and su0 is at base. . Amplification factor F for Nc as
function of κ. Then use
In heterogeneuos conditions, shape factors also decrease with F for circular foundations- mechanism is
shallow so soil is wekaer on average.
Drained Footings
Problem parameters- surchage from embedment or added material.
Evaluate effective stress in soil due to self weights, then ‘augment’ with foundation load. Strength due to
soil self weight = Nγ and surcharge = Nq :
Lower bound approach using discontinuities used again, though more complicated:
In elastic deformation, soil stays on kappa line. No dilation/gap-filling. Grains are gradually mobilised up
to limiting friction but don’t slide past eachother (that would cause dissipation). Increase in force because
more grains are mobilised adding more friction as shear strain increases. When it turns to elastic-plastic
the grains start to slide past eachother, moves off kappa line, dissipation occurs and dilation/gap-filling
can take place.
For loose soil, in plastic, number of gaps filled increases, so force increases, until critical amount where
number is constant. This is hardening.
For dense soil, in plastic, number of gaps filled decreases because of dilation, so force decreases until
critical point. This is softening.
Note this is a different model to the one proposed in 3D1, and is NOT meant for detailed curve fitting of
data.