GUI pitfalls - how does solver treat duplicate SIF entries

[landersohn]

I ran the Elmer GUI with the intent to do a bunch of flow simulations (Navier Stokes) with different viscosities.
Since I was lazy, I started from a library material and thought to just edit viscosity in the Navier-stokes tab on the material properties. The results were a bit unexpected so I looked at the SIF file more closely and I found 2 "viscosity =" lines in my solver block, one with my number the other some other value. After looking around a bit, I found that the second number came from a viscosity entry in the "Reynolds" tab in the materials dialog. Apparently it gets populated from the material library with a viscosity and even though this solver was not exlicitly activated, the GUI wrote this viscosity also into the SIF file.

Questions:
1) if there are duplicate lines in a block for the same parameter, which one does the solver pick? the first or last one?
2) Is it possible to fix this in the GUI so that the SIF generator only writes parameters from solvers that are actually active?
3) Even better, if I change a parameter in the GUI for one solver and this parameter is also used in other solvers, it should automatically be updated globally OR these parameters should not be split up by solvers but be in the General tab in the GUI.

[raback]

Hi,

I fully agree that if this is true then it is clearly a bug. I don' know which one it chooses. As you say it would be optimal if just one instance value needed to be changed.

There is a mode "scanning" that you could use for this kind of cases, at least if the problem is steady state.

-Peter