[iksweczaro]

Hello,

I have a question concerning uniformly distributed boundary conditions. The thing is that I have a circular stripe identified as one patch (already meshed stl file which is a part of a bigger installation). I want to set the same uniformly distributed boundary conditions so that it will act as 24 air flow inlets.

Simple draft:


I was thinking about GroovyBC, but I never used this tool, so firstly I would like to know what is your opinion.

Thanks in advance.

[tomf]

Hi,

It looks like you could use something like this:

Inletpatch
{
  type  flowRateInletVelocity;
  volumetricFlowRate constant 1;
  value uniform (0 0 0);
}

You would need to know the total flowrate in m3/s from all your nozzles, but this boundary condition will distribute the flow equally and it will get the normal of all patches from the orientation of the face. Please note I wrote this from memory, might be that the actual name is slightly different.

Regards,
Tom

[iksweczaro]

Hi,

thx for yor your response. Though I'm not sure if I described what I need precisely enough. Thing is I don't need uniform distribution of a velocity through the whole stripe. I need separately distributed inlets introduced by one patch. That's why I was wondering if GroovyBC is not the right solution. But as I'm not familiar with groovy, maybe you know how to do it easier?

Maybe this "high-tech" drawing will clear up what I mean:


Regards.

[tomf]

Hi,

Well easiest (for one simulation) would be to use several inlets instead of the entire strip and use the flowRateInlet. Otherwise you could define specific sections based on the location of the nozzle using groovyBC, probably a coded boundary could also work. But as I am not familiar with your way of working, or what tools you have available, I would not know what would be the best. If you need a lot of different simulations, where different nozzles (from different sections) are to be used between them, groovyBC would be flexible enough I think if you do not want to split up your strip.

Personally I think groovy is rather convenient, once you understand the syntax.

Regards,
Tom

[iksweczaro]

Hi,

Thx for swift response.
The thing about splitting the geometry into several patches is obvious but problematic. The strip is small in response to the whole geometry. When I split it in 26 smaller inlets, those get less clear and some of them get deformed. I know that obvious solution to this problem, is to thicken the mesh by adding a cylinder or a ring with higher level of division of cells, but when I do that the number of cells increases and I'm already over the limit. In order not to loose the shape of inlets I could use surfaceFeatureExtract but that's not very efficient, and it also increases the number of cells. That's why I'm searching for different tool. While I was doing my research I encountered the topic of groovyBC but I'm unfamiliar with this tool.
My idea was to somehow set the BC so that the velocity could be introduced for e.g. as a sinus > 0 around the strip. What do you think about that?

[tomf]

Ok I understand that there may be quality issues with splitting the strip. We generally use an external mesher which helps in this case.

Well yes you would need a mathematical expression which is done with groovyBC, or the codedFixedValue, I would prefer groovyBC. There are plenty of examples around that can give you an idea.

I would not know of any other option.

Regards,
Tom