GGU-CONSOLIDATE: Multi-layered system
When you investigate a multi-layered system with differing permeabilities or - strictly theoretically - differing consolidation coefficients CV, the program must always orient itself around the larger permeability value with regard to defining the necessary time steps. As rapid changes in pore water pressure can take place in layers with high permeability, a very small time step must be selected in order to achieve sufficiently accurate results. If both low-permeability and high-permeability soils are present, long consolidation times are required. In extreme situations this can lead to modelling times which, depending on the power of the computer used, can easily require several days (!).
High-permeability soils require only small time steps.
Low-permeability soils require longer consolidation times.
For example, extreme modelling times may occur if you arrange a permeable sand layer (e.g. k = 10-4 m/s) between two cohesive layers of low permeability (e.g. k = 10-9 m/s). In such systems the sand layer can generally be neglected, unless it can drain externally to the system. In this case it is simpler not to model this soil as sand, but instead to define boundary conditions for the sand region with, e.g., a pore water pressure u = 0.
Before each computation commences, the program will estimate the expected modelling time; this helps to visualise the problem of long modelling times. You can then attempt to reduce the modelling time by making sensible increases in the iteration steps (depth). But also remember that this decreases the quality of the solution. You should also investigate the system with a view to possible simplifications. A permeable layer at the upper or lower system boundary can generally be ignored. A permeable sand layer between two layers of low permeability also plays an important role only if it can drain externally to the system under consideration (see above).