phi angle of clays

[pdzeng]

I am working on a project where I need to determine the phi angle of an overconsolidated clay for retaining wall design. I have been informed by colleagues that the best way is to infer this data is from correlations with PI and that triaxial testing on clays typically yields phi angles of around 0 representing short term conditions.

Is this correct and if so is there a more direct way to determine a long-term phi angle in a clay?

 

[jayrod12]

I've been told a few times that 10 degrees is a conservative number for clays. But I'm not a geotech.

 

[pdzeng]

Thanks Jayrod12. I too have been directed towards that sort of value with advice to keep it conservative.

 

[GeoPaveTraffic]

pdzeng and jayrod12,

When talking about friction angles and clays you MUST ALWAYS indicate if you are talking about drained or undrained strengths. Undrained friction angles in clays are often taken as 0 degrees. Drained friction angles are often estimated using coorelations between drained friction angle and plasticity index. Drained friction angles are more acurately determined using drained triax tests or undrained triax tests with pore pressure measurements.

Mike Lambert

 

[pdzeng]

Thanks for you help Mike. I understand now.

How can I interpret the drained friction angle using pore pressure measurements from an undrained test?

 

[pdzeng]

After doing some more reading around this subject, I understand that drained friction angle can be inferred from subtracting the measured pore pressure from the normal and confining stress components to establish the effective stresses and shear strength can be subsequently computed.

 

[BigH]

Long term friction angles in clay (heavily overconsolidated) can be quite low (see London Clay where the residual friction angle is in the order of 10 to 12 deg (my UK colleauges - from memory).

 

[pdzeng]

That sounds about right to me BigH. In the past I have used 15deg for London Clay (interpreted from plasticity index).

 

[PEinc]

Terzaghi & Peck's 1968 Soil Mechanics in Engineering Practice, Page 112, Figure 18.1, has a graph of friction angle vs. plasticity index for clays of moderate to low sensitivity under drained conditions. Terzaghi, Peck & Mesri have a similar graph in their book. See Figure 19.7.

http://www.PeirceEngineering.com

 

[fattdad]

peak friction angles (drained) in clay can exceed 20 degrees. Problem is do you want to rely on the peak frictin angle? In many stiff fissured clays, there are latent failure planes (slickensides) that can affect the overall strength of the soil mass, just like discontinuities can affect the overall strength of the rock mass. Also, in the near surface (i.e., within 10 to 15 ft of the exposed ground surface, the repeated wetting and drying cycles can "soften" the overall soil mass. Such softening can result in loss of cohesion and a lower friction angle. Tim Starke, University of Illinois, Urbanna, has published on the determination of fully-softened clay strength. He loves running torsional ring shear testing on reconstituted samples to derive the fully-softened shear strength. He also has made correlations to clay fraction and Atterberg limits.

Typically fully-softened shear strength is less than peak and greater than residual.

f-d

¡papá gordo ain’t no madre flaca!

 

[pdzeng]

Thanks for your responses. I will have to get a copy of Terzaghi and Peck. I have previously adopted critical strength for design purposes fattdad.

 

[Mccoy]

I wonder what in this case (stiff OC clays, probably fissured) would be the best descriptor of the behaviour of the Whole soil volume under failure.
We agree that peak maybe an overestimation because of existing fissures. But phi_residual maybe an underestimation because it describes the soil frictional resistance with very large deformations involved (about 1 meter would be the relevant scale).
So we are left with phi_cv, which can be determined by torsional ring apparatuses like fattdad describes and by correlations with PI as already described. There is one more choice: some generic post-peak friction angle, greater than phi_cv which by the way is Greater than phi_res in clays.
What I stress is that we should not forget to define precisely which beast are we talking about within the managerie of the different friction angles.

 

[GeoPaveTraffic]

I agree with Mccoy that it is of up-most importance to specify what kind of test/strength we are talking about when talking about clays.

As for fully softened strengths, while I certainly understand the concept and agree that they can occur; I believe that they are being over used/applied. I've seen people argue that fully softened strengths should be used when that calculations show factors of safety below 1 for existing slopes with no indications of instability. I've worked in residual soils, most of which are overconsolidated by desication; for over 25 years. Fully softened strengths would indicate that 80 percent or more of slopes that are existing would have factors of safety well below 1.

Unless there is a history of slope movements, it is unlikely that the fishers will line up enough to significantly reduce the soil mass strength below peak strenghts. Slopes designed using peak strengths and appropriate factors of saftey, 1.4 to 1.5, have shown little chance of failure. Failures happen, but in all cases I've looked at the failures have been caused by poor or no analysis or inadaquate investigations that missed shales or clay shales (caes where the weak planes do line up over long distances.

Overall, I have not found a use for fully softened strengths and will be happy when people finaly stop thinking they should be used for all clays for all situations.

Sorry, all; I'll get off my soap box and go back to my drink

Mike Lambert