automatic time step "error"

hello,

i´m working on an unsaturated transient 2D problem. First tried it as steady, and there the simulation went well, after adjusting the problem settings.
Now I tried it with transient information, so i put my time series in the modell.
I choose "automatic time step control" with an inital time step 0.1d , numerical parameters with 100 iterations and the SAMG solver changed to 1e-04 termination criterion.

BUT if I start the simulation, the time steps are very very small and don´t get bigger over time, so even after an hour of computing, the simulation was not finished with even the first of more than 100 days. (see picture[PDF] )

I checked the observation points hydraulic heads- they were stable.

What could be the problem? I really tried to changed SAMG solver or numerical parameters in different ways, but never got the simulation over 1 day of time.

Check out the numerous white papers available for download from feflow.info site under documentation.  Probably your mesh is too coarse for the pressure-saturation-relative conductivity specifications you've chosen.  Its likely you either have to reduce the nonlinearity of those equations (make them more linear) or you have to increase the resolution of the mesh.