phi angle for silty or clayey sands

[kcall2]

Local jurisdictions require that safety factor calculations for static slope stability and infinite slope stability be based on residual, drained shear strength parameters obtained using a direct shear machine. I am consistently getting values for cohesion around 100 to 200 psf, and values for phi around 22 to 26 degrees, for well compacted fill materials (92% modified proctor) with fines contents ranging from 20 to 45 percent. Test samples are both in-situ (obtained from a hand-driven california drive), and remolded (to avoid potential disturbance). Use of these values will result in unsatisfactory safety factors for infinite stability and static stability of high 2:1 (h:v) slopes. Competitors seem to be reporting values up to 50% greater. Any suggestions on how to obtain more suitable (and economical) shear strength values (i.e. shear rate, normal load, sampling method, etc)?

 

[GeoPaveTraffic]

For silty to clayey sands with 20 to 45% fines your values appear about right. For peak values the cohesion looks a little low and the phi angle mayby 2 or 3 degrees low. However, for residual strenths; I know a lot of folks who would set the cohesion to zero.

You don't say how may cycles you are running to determine the residual phi angle. What are you getting for peak strengths? Is there any chance that your competitors are reporting peak instaed of residual strengths?

At the lower end of the fines content you could have significant grain to grain contact in the sand thereby getting a higher phi angle. Are the competitors reporting higher strengths or higher factors of safety? There are a lot of ways to get any answer you want out of stabitlity software...

 

[kcall2]

Thanks for the response geopave. Peak strengths seem to be just fine. I agree that as the fines content approaches zero, so should the residual cohesion. In theory, every soil will have zero residual cohesion if sheared at a slow enough rate.

We are running one cycle on the direct shear machine, unless it is on a sample of landslide plane or a clay bed in bedrock to model adverse bedding, then it is three cycles. Our competitors should be reporting residual values as directed by local guidelines.

I am aware of how the stability software can be manipulated, but I am truly interested in logically finding the most critical failure surface. As for infinite slope, it is so basic that if you don't have at least 32 deg and 200 psf (or various other combinations), the safety factor will be less than 1.50. Obviously fill materials with the above index parameters will have some cohesion and typically do not have surficial stability problems, so if local jurisdictions require residual strengths for analysis, then it is impractical to use anything less than 200 psf. I guess it comes down to engineering judgement.

 

[fattdad]

You know if you make your normal loads high enough, there just won't be a peak strength. I say this as this type of thing frustrates me! If you have a cut slope and see blocky fracture, then you NEED to use residual strength. These things are determined by engineering experience, not by mindless regulation.

Sorry for the morning rant.

f-d

¡papá gordo ain’t no madre flaca!

 

[BigH]

One cycle of direct shear will not be enough to obtain a true "residual" friction angle of the clay. You go through peak, then fully softened and only at great strain, will you get the residual value. I think that it was Skempton who wrote the "grandfather" paper on residual strength (a clayey slope in Britain failed at 10 deg slope when they made a 1 m vertical cut into the slope with a retaining wall (not immediately). I wonder why the local jurisdictions are requiring the residual friction angles. Normally this, I would think, would only apply to HOC clayey soils. Terzaghi Peck and Mesri (1995) offers a lot of good pages of discussion on strengths of clayey soils. I won't comment on what appears to be the "competitions" take on analysis.

 

[theCorkster]

Just some thoughts....

"...residual, drained shear strength parameters using direct shear tests." That implies to me that a lot of engineering judgement has to be brought to bear.

While it is feasible to obtain "residual" shear strength from a direct shear test system, it takes a lot of diligence in running the test, especially when you have 55 to 80 percent sand. Improper spacing between the upper and lower box can result in excess friction.

Another thought is the degradation of the shearing surface at the edges of the specimen. Exactly how much area are you really measuring the shear force on?

Are your competitors running the tests too quickly, and are they "wetting" the specimens. Also, are you running the tests at the appropriate stresses for the stability analyses (infinite versus deeper seated failure)?

 

[SmokeyBear]

Some late thoughts...
Probably the best method of assessing residual strength is to use a ring shear device. I would guess around 22-25 deg with c'=0. I don't think there is anyway to get "better" values. Residual strength is residual strength.

However, the method of testing will certainly influence the strength results, as theCorkster says. Under rapid loading-higher strength, at low strains-higher strength.

I would have thought that the amount of stain necessary to develop residual stregt would have resulted in failure along the way, or at least, very significan movement.

If residual values were adopted a Factor of Safety slightly above 1.0 would seem more appropriate than the 1.5 inferred in your post.

 

[epongra2]

Perhaps the strength values you, and your jurisdiction, are looking for are "ultimate", rather than residual.

Unless your fill slope has a pre-existing failure, why would you use residual strenghts?