The use of the Modified Cam-Clay model in practical applications is not recommended?

Dear Jiayao Wang,

The use of MCC for practical applications is not recommended for 2 reasons:

1) You already mentioned, it can't handle overconsolidated clays. I wouldn't even say "very well", because quite often it's "not at all" because of mesh dependency issues.

2) The MCC model has a coupling between the strength parameter M and the ratio between horizontal and vertical stresses in primary loading K0nc. If we choose the correct value of M the model overestimates K0nc, which means that the horizontal stresses are overestimated and therefore also the mean effective stress p'. And overestimating the mean effective stress p' then causes the model to overestimate the (stress-dependent) stiffness as well. The overestimating of the horizontal stresses also influences the prediction of horizontal displacements.
If on the other hand we would like to have the K0nc correct (for instance because we are interested in horizontal displacement at let's say the toe of a dam or so) then we can only do so by overestimating the strength. We would have to choose often an M >> 1, which gives an equivalent friction angle of significantly over 30 degrees.

Both problems have been overcome in the Soft Soil model. In the Soft Soil model we choose M in such a way that K0nc is correct. To avoid that we would now overestimate the strength, the model has an additional Mohr-Coulomb failure surface with the correct desired strength. This means that actually the CSL is outside the Mohr-Coulomb failure criterion, hence the Soft Soil model cannot calculate overconsolidated soil and not even soil at critical volume.

With kind regards,

Dennis Waterman