Poiseuille Flow

[Lucas]

Hi,

I removed the cylinder from the example lbm_cylinder.py so that it can simulate a Poiseuille flow.

I want to simulate entry effects either, so I set velocity boundaries on both beginning and end sides of the slit (left and right).

My question is, from the generated files by this simulation, choosing a particular iteration, I have the same graph for all x-positions of the grid.

So I can only see the development of the velocity profile if I use different iterations, not different positions of a single time step. Is this right?

Thanks!

[Michal Januszewski]

Hi,

Are you aware that we already have a Poiseuille simulation (lbm_poiseuille.py)?

Are you setting a specific velocity profile at the inlet and outlet,
or are you just using a constant value? If you're doing the latter,
the profile should be different at different X positions.

Michal

--
Michal Januszewski
http://people.gentoo.org/spock

[Michal Januszewski]

LBMGeo2D and FluidLBMSim are generic classes -- the first specifies an
empty geometry (only fluid nodes) and the second is used for single
phase fluid simulations.

LBMGeoPoiseuille adds walls for the channel and pressure nodes at the
inlet and outlet if the flow is to be driven by a pressure difference.
LPoiSim just adds a bunch of command line parameters that make it
easier to set up the simulation. Both are defined in
lbm_poiseuille.py, which is the example you should probably be looking
at for a Poiseuille simulation.

Hope that helps,
Michal

On Fri, Mar 2, 2012 at 15:26, Lucas <fabiol...@gmail.com> wrote:
> I´ve seen the specific Poiseuille simulation, but couldn´t understand
> everything on it yet..
>
> I was setting a constante value at the inlet and outlet, using
> NODE_VELOCITY in set_geo, but it doesn't work.
>
> My professor tried to simulate Poiseuille flow from the example
> lbm_ldc.py, and it worked right. The differences between his example
> and mine is that I used the classes LBMGeoPoiseuille and LPoiSim and
> he used the general class geo.LBMGeo2D and FluidLBMSim. Can you tell
> me the differences between these classes or where can I see them?
>
> Thank you

>
>
>
> On 16 fev, 21:25, Michal Januszewski <mich...@gmail.com> wrote:
>> Hi,
>>
>> Are you aware that we already have a Poiseuille simulation (lbm_poiseuille.py)?
>>
>> Are you setting a specific velocity profile at the inlet and outlet,
>> or are you just using a constant value?  If you're doing the latter,
>> the profile should be different at different X positions.
>>
>> Michal
>>

[Fabíola]

Hi,

I was able to run poiseuille flow well, thank you very much!

I would like to know how can I show data in prompt line like some
examples do. Does anybody know how to do that?

Fabíola

[Michal Januszewski]

Hi,

Do you mean this:

[0001.54] Simulation parameters:

  Re: 4800

  bc_pressure: equilibrium

  bc_slip: slipbb

  bc_velocity: equilibrium

  bc_wall: fullbb

  dt: 0.000434027777778

  dx: 0.0208333333333

  grid: D2Q9

  incompressible: False

  model: bgk

  precision: single

  size: (192, 48)

  visc: 0.001

?  This is displayed when --verbose is passed on the command line. In order to add your own entries, you need to override the sim_info property in your simulation class.  See e.g. this example from sailfish/lbm.py;

class FreeSurfaceLBMSim(LBMSim):

    @property

    def sim_info(self):

        ret = LBMSim.sim_info.fget(self)

        ret['gravity'] = self.gravity

        return ret

Hope that helps,

Michal