Learning Forum

[luna6362]

Hello everyone,

I am currently simulating a compressor which includes a rotating rotor using the MRF method. Previously, I used version 17.2 for the simulation and the flow field appeared normal. However, with the same settings, the simulation results in version 24 are very strange. There is a significant high-speed flow on the walls, which I believe is not a normal simulation result. Below are my settings. Are there significant differences between version 17 and version 24 for these settings?

Mesh:

• Meshing Element size: 0.8mm
• Face sizing to MRF area: 0.6mm

Fluent setting:

• Steady
• Reynolds stress model
• Frame motion Y = 1, -4850 rev/min

Material:

• R32
• Density: 61.5295 kg/m³
• Viscosity: 1.5923e-05 kg/ms

Boundary conditions:

• Mass Flow inlet: 0.01353 kg/s
• Pressure outlet: 2824315 Pa

Method:

• Coupled

Thank you!

[Rob]

Are you reading in an old case or starting from scratch? I'm not aware of anything that will stop the latter from working. For the former it's about a 8 major release jump so check all of the settings, especially the interface between the rotor & stator. 

[luna6362]

Hi Rob!

I resolved the issue by changing the boundary condition of the rotor wall.

I set it to a rotational wall with no velocity.

However, since I was following a video tutorial online, I'm not entirely sure of the purpose behind this change, as it didn't involve setting a velocity.

Thank you!

 

[Rob]

You need to check whether walls are moving with a cell zone or otherwise. So, depending on what is/isn't moving you may have set something moving at twice the speed it should have. The moving reference frames are very simple once you see what's happening, and very hard to explain!