Remolded shear

[ZTGR]

Hello,
I have a simple question about the remolded shear testing, all the materials are not meeting the specs required (C of 590, Phi of 28 degrees) the materials are silty sands most of the time. i am performing the remolded shear on the material passing sieve #4, the ASTM don't mention anything about remolded shear passing a certain sieve.
your comments are appreciated
Sincerely,

 

[Focht3]

What results are you actually getting?

Are you using a direct shear box? Or triaxial setup?

How long are you allowing the samples to sit ("cure&quot before you are testing them?

 

[ZTGR]

Hello Focht3,
all results are not meeting the specs, i think that is because of the materials, not enough cohesion, to meet their standards i know the materials should be clayey sands (Standards call for c:23.9 KN/m2 , Phi 28 degrees and 25% passing #200). some of the materials meets the passing 200 and the phi is close to 28, but i am getting low cohesion, the samples are being consolidated a minimum of half hour to 45 minutes in the direct shear box and at least 24 hours consoled in water. my wonders are maybe i shouldn't use the -#4 materials for remolded shear.
Sincerely,

 

[Focht3]

You will increase the phi value, but it probably won't help cohesion. But run one test that way - see what happens. Perhaps you'll get lucky...

 

[MRM]

ZTGR
Is there any way you could modify the specs and use effective stress parameters instead of parameters involving "cohesion?" The effective stress approach would yield a phi angle (without a cohesion intercept) that represents the conditions you are modeling (drained or undrained shear). This approach is fundamentally more correct in terms of soil behavior. The triaxial test would probably be a better set-up to monitor pore pressures during testing to allow you to use effective stress parameters. The cohesion intercept is, in many ways, one of the great "untruths" of the soil mechanics world. In my opinion, solid rock is really the only geo-material that has cohesion. Any cohesion you're observing in the test as you have it now is probably the result of pore pressure build-up because I don't believe that silty sand, by the description alone, would have any "cohesion", as anyone knows it, anyway. How much silt is present in the specimen? How does the remainder of the grain size distribution look?

On a side note, you could probably meet the "cohesion" specs by over consolidating the silty sand like crazy and running a very quick test (saturated, undrained). The pore pressure created in the highly oc silty specimen would cause a larger "cohesion" intercept. Of course I'm only joking now...this test would be a trick. It would not represent the actual loading conditions. Also, the "cohesion" intercept would dissipate with time.

 

[Focht3]

MRM is right, of course. "Cohesion" is really a remnant of the curved failure envelope of the soil.

And that brings me to an important point for your testing program: you could probably meet the test result requirements by increasing the normal (vertical) force. This will have the effect of reducing the phi value somewhat, but will increase the apparent value of cohesion.

But that isn't a good strategy, after all - it isn't ethical to choose only those test results that allow the material to pass the criteria. You would be better served to perform enough tests to define the curved failure envelope, then present the results to the engineer(s) who specified the criteria. Plot the criteria on the same graph, and show them that the soil meets their criteria under certain circumstances. This should help you resolve your problem -

 

[BigH]

Good points by all above - but how are you using the test results? For what purpose. The c,phi in most cases are only a mathmatical model of the stress behaviour. What you look for is what is the shear stresses computed in the range of shear normal stresses I encounter. If your c,phi at a normal stress, say of 100kN is X, then if you have material with c=0,phi= 31, say, also giving X, then I believe you could make a case that you have met the "spec". It would be interesting to know what the "curves" of the original investigation show - on which the designer probably just "copied". It is like the case of fill behind retaining walls. I've seen far too many designers specify: "Fill to meet following criteria - c=0, phi = 31deg, delta = 20deg, gamma = 1.85g/cc". Why is he specifying the fill this way? Because that is the parameters in which he got his ka = 0.33! Now, translate that into a real soil - he may also request compaction to 95% modified and hence, you'd be getting, say phi = 35deg; then you are outside, eh????

 

[Focht3]

BigH: This is one of the reasons that separating the design and implementation functions of geotechnical engineering is such a dangerous mistake -

I know that you understood the implications of 95% compaction behind your hypothetical retaining wall, but for the novice:

Specifying a high degree of compaction behind a retaining wall will cause the soils to be more dense, giving a larger effective value of phi. But it will also induce lateral stresses within the soil mass that can exceed the capacity of the wall, causing the wall to crack. The high stresses will also require more horizontal movement to occur before the design active pressures are achieved -

Just a little clarification.

 

[BigH]

Yeah - had a masonry wall crack horizontally in China once. It was the abutment for a small span bridge. Contractor had the deck in place and then backfilled behind. Used a big roller right up to the abutment. So you can imagine that the horizontal stresses induced during rolling caused a bit of tension on the inside of the abutment - then a nice line crack developed about 750mm down from the seating beam.

The point I was making about the backfill parameters is that many times, the designer just puts this in as he doesn't know what the "fuzziness" of his design can take. Then, he can always blame the constructors.

Still, too, it is hard to fathom how experienced contractors keep on piling on the big rollers in the abutment backfill area!!

Best to you all