SimScale CAE Forum

Temperature Field diverging

Hi guys,
I’m having trouble with my simulation. I am currently completing an assignment in which I need to run a simulation on a ramjet engine and prove how they work. I have my Ramjet design in which my current problem is appearing. Whenever I run a simulation with Compressible air, a few minutes in the Temperature field diverging error appears, this is the exact message:

“Temperature field started diverging. Please check the mesh quality near the reported location and try refining the mesh. If the problem occurred near a boundary, please check the boundary conditions.” with: “Temperature = -3.10656e+07 at position: (0.01725 m, -0.04716 m, -0.02411 m).”

I am unsure why this is happening as my mesh appears to be fine, and I haven’t changed any temperature settings. This error is happening in Compressible 3 and does not happen in incompressible simulations.

Any help would be greatly appreciated! I am pretty new to simscale as this is my first real project. My project can be found here:
Thanks in advance.

Hey @jbathie!

Ramjet’s have a supersonic inlet and outlet (!?), but the combustion happens at subsonic speeds.

My colleague Vuko suggests to use “special” boundary conditions.

  1. Fixed velocity and pressure at the inlet via custom BC,
  2. Zero gradient on everything via custom BC.



Hi @jousefm
Yes, that is correct, supersonic inlet, the compressor section then slows the air to subsonic speeds for combustion which the exits the exhaust nozzle at over supersonic speeds (higher speed than which it came in of course).

Could you explain how I would do that, so I can understand what is happening by doing those things?

P.s. I tried running a sim with those different boundary conditions that you (I believe) put in, and it didn’t work, it came up with the same error.

1 Like

Hi @jbathie

This type of flow is quite challenging to model. In an ordinary subsonic flow, a common combination is a velocity inlet & a fixed value pressure outlet.

For supersonic flow, this approach wouldn’t work very well, because static pressure can’t propagate upstream, thus it has to be defined at the inlet. This is what Jousef is referring to in point 1.

The second point addresses the outlet, trying to achieve a stable configuration/avoiding return flow. It would definitely be a challenging project.

This project from @Get_Barried comes to mind: