Peter Schätzl Grundwassermodellierer

1) Yes, this is intended. By hitting stop, the simulation-related data are released from memory. The best option to keep the values is writing a DAC file.
2) I don't know what the reason for this is. In general, the export always worked fine for me.

In case that there are many time steps in the dac file, loading can take long (possibly hours for very large files). The reason for this is that FEFLOW will reconstruct the damaged diagram data from the results of all the time steps as far as possible. Thus each time step has to be loaded (one after another) in order to extract the corresponding data.

General results should be fine, just the content of the diagrams may be corrupt. Possible reasons from my experience: Saving interrupted (due to full disk, etc.), version differences between writing (simulation) and reading DAC.

The grouping can be done in an elemental user data distribution. It does not support strings ('rock1'), but you can encode your rock types in float numbers (1.0, 2.0, ...) and apply the 'if' statement to those.

Sounds like a license issue... Pls. check whether you run with a valid license or in DEMO mode (in the latter case with a small model only).

Sounds like no license is found. Could you check in Tools - License whether a license is available (or demo mode is set in case you want to do a small mesh within the bounds of the demo version)?

Had not seen that your mesh is unstructured - so forget what I said about slices...

1) You can do exactly as you wrote. Use an expression with the formula "Elev - 1"
2) Yes, the Cauchy BC needs to cover an area (either horizontal or vertical) - thus you need two rows of nodes on one slice, or you use nodes on two slices (then only a line on each). The value is the reference water level, you need to put a value for the transfer rate in addition (in material properties).

There's no way in FEFLOW to do this. For FEFLOW one file is one case.

It's hard to find out what the difference is between the results without playing with the models. I'd recommend to modify parameters one by one to see whether the two results can be brought closer together. I've never directly compared EED and FEFLOW, but at the time the BHE were introduced into FEFLOW, the results were thoroughly cross-checked with CFD simulations. The rapid development towards the steady state points towards one of the storage parameters, and I'd think it might be in the aquifer rather than in the borehole (as I'd expect the borehole to have shorter influence only). Could it be different assumptions for fluid/solid heat capacity in the aquifer (and porosity, of course)?