Residuals are constant after 5 Iterations

  • Last Post 4 weeks ago
  • Topic Is Solved
hessma posted this 02 June 2020



I'm using Ansys Fluent 17.0 and trying to calculate a pressure surge in a shock tube. 

The tube is 10 mm in diameter, the driver is 1000 mm and the driven is 200 mm long. I'm using the desinty-based-solver and the Material is oxygen as ideal gas.


During the calculation I have the problem that in later time steps the residuals only fall during the first 5 iterations and remain almost constant afterwards.


can someone tell me what my mistake is?

Order By: Standard | Newest | Votes
rahkumar posted this 02 June 2020


Can you please insert image of pressure and velocity contour, so that it would give us a better idea of what is going on? Thanks. 

Regards, Rahul

hessma posted this 02 June 2020

the pressure after 1.8e-4 s:

(at the beginning on the right site was a pressure of 1e6 Pa and on the left there was a pressure of 1e5 Pa)


and the mach-number

rwoolhou posted this 02 June 2020

How well refined is the mesh around the shock? How long does it take the material in the high speed flow region to cross one cell?

hessma posted this 03 June 2020


I tried diffrent Meshes (all of them Tetrahedrons) and time steps.

In the moment use this Mesh (42,742 Nodes and 203,188 Elements):


I have chosen the following time steps: 1e-6 s, 1e-7 s and 1e-8 s:
So in the worst case it takes around 5 time steps to cross one cell.

rwoolhou posted this 03 June 2020

Try 5e-9s and also plot the Mach Number on the surface with node values off. You may need a higher resolution mesh to capture the shock: read up on adaption. 

hessma posted this 4 weeks ago

The problem is solved for the moment, but I do not know what the problem was.

It must have been either the time steps or the mesh, but I have not yet found a clear connection to the problem.