CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS (http://www.cfd-online.com/Forums/ansys/)
-   -   Spontaneous failure of moving mesh interface (http://www.cfd-online.com/Forums/ansys/78207-spontaneous-failure-moving-mesh-interface.html)

gareth__it_power July 15, 2010 07:12

Spontaneous failure of moving mesh interface
 
hello,
My problem is that interface between a moving and stationary part of my mesh seems to spontaneously start behaving like a wall, i.e. not allowing fluid across it. This may happen after the simulation has been running successfully for a relatively long time. I don't know what causes it.

A bit more background:

I am a Phd student and I am modelling a vertical axis wind turbine in 2D using uRANS. My domain is a large square, with a rotating circle in the middle which contains the blades.

The mesh is quads, and has been refined along the blade surfaces to give y+ around 1. I have also refined the mesh along interface (the boundary of the circle) with about 2000 cells along the interface. The timestep is such that one revolution takes around 1000 timesteps.

The mesh was built in Gambit, where I specified two zones; the moving and stationary areas. In Fluent I specified the rotational velocity of the inner circle, and the Mesh Interfaces (which I had to change from default Wall.)

When the simulation first begins it is fine, and it takes a while for the expected velocity fields to settle down and periodic behaviour to begin. At some point, it is impossible to predict when, the periodic behaviour ceases because fluid is no longer entering the circle, but is flowing around it.

I'd be very grateful for any suggestions, or to hear from anyone else who may have experienced this fault.

Kind Regards,

Gareth Uglow

gareth__it_power August 26, 2010 07:50

I seem to have solved the problem.

For some reason the diameter of the rotor or stator was changing slightly, causing a gap between the interfaces. This gap was interpreted as a very small wall.

I have now allowed a small, 1mm, overlap between the interfaces. The cell size in this region is about 1cm, so the overlap is small relative to the cells.

I'd be interested to hear from anyone else who has encountered this phenomenon.

Regards,

Gareth


All times are GMT -4. The time now is 15:07.