Dear Mr. Zaryab,

do you have a chart of the grain size distribution of the soil? If your soil has a large amount of clay the increase of cohesion in a dry state is pretty common. In the partially saturated state additional capillary forces are acting which lead to an increase of the cohesion (and friction angle).

Best regards

PS: I recommend to read the attached file.

 shear test  in saturated condition

Hi Abdulhalim

Please find the attached. I believe loess is a kind of silty soil and loses strength when saturated. Please refer to Derek Cornforth's book on "Landslides in Practice: Investigation, Analysis and Remedial/Preventative Options in soils" which refers to the properties of loess.

Best regards

Ravi

http://au.wiley.com/WileyCDA/WileyTitle/productCd-0471678163.html

Dear Zaryab,

It is well known that the cohesion of a soil will decrease in saturated conditions. But according to several authors, the friction angle doesn't vary significantly with water content (Ho et Fredlund, 1982? etc.).

So, in saturated conditions, I expect that the friction angle of your loess will remain nearly constant, and the cohesion will tend to zero.

Even in the conditions of your test, the value of cohesion is very low (6 kPa), so, the shear strength of the loess is essentially due to friction which is not very affected by variations in water content.

Regards

Nabil Abou-Bekr

Thank you my colleagues for great comments. 

I see you are studying a specific area and also a specific soil. In my opinion, the most convinient and practical thing to do is that, in case you have an unperturbed soil sample, you could do an other serie of traxial tests (If possible TX CU or TX CD) but saturating the sample (to a Skepmpton B of 97or more).

The results of those tests must be much more realistic than some theoretical aproches, and indeed you may compare it with the told theories to calibrate them.