[FLASH-USERS] Adding a Collision term to the momentum equation in the USM solver

Christian Fischer fischer at geo.uni-koeln.de
Tue Oct 20 05:32:27 EDT 2015


Dear FLASH-Users,

I am currently working on an MHD-problem where I want to simulate a 
cloud of neutral gas which is embedded in a plasma.
The Setup is quite simple:
The cloud is located in the middle of a cartesian 3D domain as an 
obstacle for the plasma. The interactions are described by a Collision 
frequency nu.
The simplest way to realise this would be by adding the term - nu * v * 
rho into the Equation of Motion.

Currently I am struggling with the issue that FLASH does not run this 
problem.
As soon as everything is compiled and I want to run the simulation the 
Error of a negative dt occurs in the Hydro-Unit.

I have defined the initial setup in my Simulation_initBlock.F90 with a 
Scratch-Variable for the Collision frequency.
Then I have added the Collision-term in the file 
hy_uhd_getRiemannState.F90 in a similar fashion as the Gravity is 
implemented.

I don't know what could be wrong with it, although I guess that my 
changes in the USM-Solver are not right.

I would be very happy if anyone is able to help me!

Best regards
Christian



More information about the flash-users mailing list