Swell model in effective stress


As I understand, constitutive models in FLAC3D operate on effective stress. Why does the “Swell model” use use total stress ? :


Is there a way to switch this formulation to effective stress instead of total stress ?


The swelling effect in this specific constitutive model is currently based on total stress. However, if there is evidence, either from experimental data or literature, suggesting that effective stress is more influential, we can introduce an option for users to select which type of stress to use. This would involve adding a flag to the model, allowing users to choose between total stress and effective stress.

I think people use, abusively, total stress instead of effective stress when talking about swelling pressure because swelling pressure is always determined from laboratory oedometer tests, where the pore pressure is almost nul, so total stress = effective stress. But to be rigorous, swelling pressure is an effective stress. Only in the particular case of oedometer test without pore-pressure that it is equal to the total stress.

Just imagine 2 samples under the same total stress loading at sig_v = 1 MPa. One is wetted under a water pressure p ~0, and the other is wetted, with a pump for exemple, with a pressure of p = 0.5 MPa. At equilibrium, the effective stress in the 2 samples will be sig’_v = 1 and 0.5 MPa. Assuming a swelling pressure of 2 MPa, will they have the same swelling strains ?

Thus, it would be great if we can have the possibility to use effective stress instead of total stress in this model.


While we typically rely on solid experimental evidence for such changes, adding a flag property to allow users to select the stress type for the swelling effect is feasible. However, considering our reduced frequency of updates for FLAC3D version 7, this feature will be implemented in version 9 instead. You can expect this update to be available in the upcoming update release, version v9.166.