Www.fugro.com © Fugro 2013 Amin Rismanchian Nick Ramsey Feb. 2014 Drained or Undrained: Is That the Question?

www.fugro.com© Fugro 2013

Amin RismanchianNick Ramsey

Feb. 2014

Drained or Undrained:Is That the Question?

www.fugro.comDate

What I am going to talk about

What do I mean by “Drained” and “Undrained”?

Relevant soil parameters, and methods of assessing these parameters

Drained vs. undrained breakout resistance

Flaws of current methods

Conclusions

www.fugro.com

What do I mean by “drained” and “undrained”?

“Fully Drained is the condition under which water is able to flow into or out of a mass of soil in the length of time that the soil is subjected to some change in load.”

“Fully Undrained is the condition under which there is no flow of water into or out of a mass of soil in the length of time that the soil is subjected to some change in load. Changes in load cause changes in pore water pressure, because the water cannot move into or out in response to the tendency of volume change.”

Partially drained is the “twilight zone” between fully drained and fully undrained behaviour.

(Duncan and Wright 2005)

www.fugro.com

Example: Cone Penetration Test, Offshore WA

Is this test indicating undrained/drained or partially drained conditions in the surficial soils?

-0.05 -0.025 0 0.025 0.05Pore pressure (MPa)

Generated pore pressure, u2

Hydrostatic pore pressure, uo

www.fugro.com

Twilight Zone: Partially Drained

Zone Soil Type

1a SILTS and low Ir CLAYS

1b CLAYS

2 Essentially drained SANDS

3 Transitional soils

Refer to Schneider et al. (2008) for the boundary lines.

www.fugro.com

Twilight Zone: Partially Drained

Zone Soil Type

1a SILTS and low Ir CLAYS

1b CLAYS

2 Essentially drained SANDS

3 Transitional soils

Refer to Schneider et al. (2008) for the boundary lines.

www.fugro.com

Example: Cone Penetration Test, Offshore WA

-0.05 -0.025 0 0.025 0.05Pore pressure (MPa)

Generated pore pressure, u2

Hydrostatic pore pressure, uo

Twilight zone (partially drained)

Fully drained

Fully undrained

www.fugro.com

Drainage is a Function of the Normalised Velocity

CPT

Spudcan penetration

Pipeline penetration

Partially drained penetration

Undrained penetration

Pe

netr

atio

n re

sist

an

ceSpool

Drained penetration

Twilight zone

www.fugro.com

Parameters Affecting Drainage Behaviour of Soil

Rate and duration of loading/shearing

– Estimated from installation/operation conditions

Drainage length

– Estimated from the geometry of the problem

Coefficient of consolidation

– Laboratory methods

• Rowe cell

• CRS

• Indirectly from permeability

– Estimated from in-situ tests:

• Dissipation tests

• Twitch tests• Parkable piezoprobe

(Chatterjee et al. 2014)(Randolph and Hope 2004)

Range of uncertainty: Up to 1.5 times

Range of uncertainty: 100 to 1,000 timesBut it can be decreased!

Range of uncertainty: Depends on the application. Sometimes very high.

www.fugro.comDate

What I am going to talk about

What do I mean by “Drained” and “Undrained”?

Relevant soil parameters, and methods of assessing these parameters

Drained vs. undrained breakout resistance

Flaws of current methods

Conclusions

www.fugro.com

Twilight zone Lateral equivalent friction factor, H/W′

Normalised time, T = cvt/D2

Short breakout duration; impermeable soil

Long breakout duration; permeable soil

Drained behaviour

Undrained behaviour

Slow thermally- induced buckling

‘Snap’ buckling

Significant difference in lateral equivalent friction factor dependent on breakout duration

Dilatant soil (e.g. silty SAND/sandy

SILT in NWS)

Why Drainage is the Question

www.fugro.com

Lateral equivalent friction factor, H/W′

Normalised time, T = cvt/D2

Short breakout duration; impermeable soil

Long breakout duration; permeable soil

Drained behaviour

Undrained behaviour

Slow thermally- induced feed-in

‘Snap’ buckling

Dilatant soil (e.g. silty SAND/sandy

SILT in NWS)

Why Drainage is the Question

More permeable soil (or longer breakout duration)

Insignificant difference in lateral equivalent friction factor dependent on breakout duration

www.fugro.com

Lateral equivalent friction factor, H/W′

Normalised time, T = cvt/D2

Short breakout duration; impermeable soil

Long breakout duration; permeable soil

Drained behaviour

Undrained behaviour

Slow thermally- induced feed-in

‘Snap’ buckling

Dilatant soil (e.g. silty SAND/sandy

SILT in NWS)

Why Drainage is the Question

Insignificant difference in lateral equivalent friction factor dependent on breakout duration

Less permeable soil (or shorter breakout duration)

www.fugro.com

Drained behaviour

Undrained behaviour

Best estimate

LB

UB

Possible values

UB

LB

P50

Probability

Uncertainty

UB

LB

BE

H/W′Lateral equivalent friction factor, H/W′

Normalised time, T = cvt/D2

High Uncertainty

www.fugro.com

Best estimate

LB

UB

Possible values

UB

LB

P50

ProbabilityUncertainty

UB

LB

BE

H/W′

Reasonable definition of volume change behaviour and velocities Narrowing of uncertainty

Lateral equivalent friction factor, H/W′

Normalised time, T = cvt/D2

Drained behaviour

Undrained behaviour

Reducing the Uncertainty

www.fugro.com

Conclusions and Discussions

Every soil type can behave drained/partially-drained/undrained depending on:– Rate or duration of loading

– Drainage length

– Coefficient of consolidation (cv)

Site investigations should be specifically targeted to suit the field events and design requirements

In specific soils (e.g. silty sands/sandy silts) both drained and undrained behaviours should be checked

Narrowing down the range of the above parameters, significantly reduces uncertainties (e.g. by in-situ estimation of cv)

There is no means of being conservative or unconservative.

www.fugro.com

Thank You

All materials and content contained within this presentation remain the property of Fugro (© Fugro 2013) and may not be copied, reproduced, distributed or displayed without Fugro's express written permission. All third party information featured within the presentation remains the intellectual property of their respective originators.

First author:

– Prof. David White, Dr. Fraser Bransby and other colleagues at Fugro.

Acknowledgments