RE: Setting the initial velocity field

[Arulnambi Palani]

Dear Everyone,

I would like to set up an initial velocity field of 0m/s. I was checking if this was possible using &INIT. But, I suppose we cannot directly specify a value for initial velocity.

My interest is to study only the convective heat transfer within a room (one side of the wall is hot). I have used an option "FREEZE_VELOCITY=.TRUE.". This also does not seem to work.

Then, I found out that the magnitude is velocity is not huge and was approximately equivalent to the noise value (given in the manual). So, I tried using the command: "&MISC NOISE=.FALSE./". But still, I get a velocity magnitude of order E-3 to E-4.

Is there any way that we can set the initial velocity to 0 or just to switch off the existing flow field.

Thank & Regards,

Arul

[fde]

FREEZE_VELOCITY might disable convective heat transfer because convection equation (user guide Eq. 11.2) has velocity component. 

You might try changing the velocity tolerance to low levels (0.0001 m/s).  See >> VELOCITY_TOLERANCE  PRESSURE_TOLERANCE 

[Arulnambi Palani]

Hi,

Thanks for your reply. Ya, that is true. Then, in which case could we use FREEZE_VELOCITY? Because, this option is described briefly in the manual.

I tried using VELOCITY_TOLERANCE. The magnitude is indeed reducing (from E-3 to E-4). But isn't this option used to set error tolerance limit for pressure solver? And also, even with this option, the initial velocity could not be set to 0.

Thanks & Regards,

Arul

[dr_jfloyd]

NOISE=F sets the initial velocity field to zero. However, if T_BEGIN you have any boundary condition present that is something other than the ambient condition, then in that first time step you will see velocities change everywhere in response to that boundary condition. 

[Arulnambi Palani]

Hi,

yes sir. I have other boundary condition: temperature applied on wall (TMP_FRONT = 200). But the gravity is set to 0 in all directions, so that we avoid buoyancy thereby prohibiting the flow.

And in our case, during the start of the simulation (T_BEGIN), the velocity is of the order E-3 and later as time progresses the velocity drops to E-4.

Could you please explain more about this scenario?

Thanks & Regards,

Arul

[Kevin McGrattan]

If you heat the wall, you will still induce a velocity, regardless of gravity. The gases in the vicinity of the wall will expand.