[usergk]

Hello Foamers,

I am trying to implement swirl input in a cylindrical tube (axis along z dir). I came across the groovyBC code to implement swirl input:

inlet
{
type groovyBC;
variables "rpm=8000.0;Un=55.8;c=sum(pos()*mag(Sf()))/sum(mag(Sf()));n=sum(normal())/mag(sum(normal()));p=pos()-c;r=mag(p)+1.0e-10;R=max(r);
xt=vector(0,n.z,-n.y);xT=xt/mag(xt);yt=vector(n.y*n.y+n.z*n.z,-n.x*n.y,-n.x*n.z);yT=yt/mag(yt);";
valueExpression "-Un*normal() + (rpm*pi/30)*((p & yT)*xT - (p & xT)*yT)";
}

Can someone explain what xt, yt, and (p & yT)*xT - (p & xT)*yT refer to?

Thanks so much,

Regards,
gk

[usergk]

Hello,

Can anyone provide some hints here?

Many thanks,
gk

[Bernhard]

They are defined in variables. If you don't understand the formulation, make a sketch, using the known vectors (pos and normal), in a random orientation. In your sketch you will understand each term immediately. (With the assumption that this is a correct implementation of course )

[tuominen]

Originally Posted by usergk

Hello Foamers,

I am trying to implement swirl input in a cylindrical tube (axis along z dir). I came across the groovyBC code to implement swirl input:

inlet
{
type groovyBC;
variables "rpm=8000.0;Un=55.8;c=sum(pos()*mag(Sf()))/sum(mag(Sf()));n=sum(normal())/mag(sum(normal()));p=pos()-c;r=mag(p)+1.0e-10;R=max(r);
xt=vector(0,n.z,-n.y);xT=xt/mag(xt);yt=vector(n.y*n.y+n.z*n.z,-n.x*n.y,-n.x*n.z);yT=yt/mag(yt);";
valueExpression "-Un*normal() + (rpm*pi/30)*((p & yT)*xT - (p & xT)*yT)";
}

Can someone explain what xt, yt, and (p & yT)*xT - (p & xT)*yT refer to?

Thanks so much,

Regards,
gk

Hi

I have used following code:

type         groovyBC;
variables
(
    "yp=pos().y;"
    "xp=pos().x;"
    "r=sqrt(pow(xp,2)+pow(yp,2));"

    "U_ax=31.35;"
    "U_t=20.97;"
    "theta=asin(xp/r);"
    "U_tx=sqrt(pow(U_t*cos(theta),2));"
    "U_ty=-U_t*sin(theta);"
);
valueExpression "vector ( U_tx, U_ty, U_ax )";

U_ax is axial velocity along z-axis and U_t is tangential velocity. U_tx and U_ty are tangential velocity's x- and y-components. Sqrt and pow in U_tx only make sure that x component is always positive (I have modelled only sector, not full cylinder).

Lasse

[usergk]

Thanks a lot for the inputs! Really appreciate it!

I was able to get it running using Lasse's code.

gk

[FedericaF]

Hello,

I used funkySetBoundary in openfoam 1.7 to generate a swirl in the inlet patch for a sector of a cylinder and it works. I have attached below the file funkySetBoundaryDict used.

/*--------------------------------*- C++ -*----------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 1.7.1 |
| \\ / A nd | Web: www.OpenFOAM.com |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
FoamFile
{
version 2.0;
format ascii;
class dictionary;
location "system";
object controlDict;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

velocityswirl {
 field U;
   expressions
   (
       {    
            target value;
            patchName inlet;
                       
            variables       "rpm=1000.0;Un=10;c=sum(pos()*mag(Sf()))/sum(mag(Sf()));n=sum(normal())/mag(sum(normal()));n=sum(normal())/mag(sum(normal()));p=pos()-c;r=mag(pos()-vector(0,0,1))+1.0e-10;R=max(r);xt=vector(0,n.z,-n.y);xT=xt/mag(xt);yt=vector(n.y*n.y+n.z*n.z,-n.x*n.y,-n.x*n.z);yT=yt/mag(yt);";
                   
                          expression "-Un*normal() + (rpm*pi/30)*(((pos()-c) & yT)*xT-((pos()-c) & xT)*yT)";
       }
);
}  

//!!! in "variables" do not leave any space between the definitions of two variables and in "espression" try to replace the name of few variables with their definitions !!!

//!!! c=sum(pos()*mag(Sf()))/sum(mag(Sf())) works just for full patches; for other cases you have to set manually !!!

[Pravin Kadu]

Originally Posted by tuominen

Hi

I have used following code:

type         groovyBC;
variables
(
    "yp=pos().y;"
    "xp=pos().x;"
    "r=sqrt(pow(xp,2)+pow(yp,2));"

    "U_ax=31.35;"
    "U_t=20.97;"
    "theta=asin(xp/r);"
    "U_tx=sqrt(pow(U_t*cos(theta),2));"
    "U_ty=-U_t*sin(theta);"
);
valueExpression "vector ( U_tx, U_ty, U_ax )";

U_ax is axial velocity along z-axis and U_t is tangential velocity. U_tx and U_ty are tangential velocity's x- and y-components. Sqrt and pow in U_tx only make sure that x component is always positive (I have modelled only sector, not full cylinder).

Lasse

Thank you Lasse.

For complete (360 degree) cylinder-

inlet
{
type groovyBC;
variables
(
"yp2=pos().y;"
"zp2=pos().z;"
"r2=sqrt(pow(zp2,2)+pow(yp2,2));"

"U_ax2=0.5496*pow((1-mag(r2-0.0095)/(1.001*0.0035)),(1/7));"
"U_t=0.15*(1-pow((mag(r2-0.0095)/(1.001*0.0035)),2));"

"U_tz=-U_t*yp2/r2;"
"U_ty=U_t*zp2/r2;"
);
valueExpression "vector ( U_ax2, U_ty, U_tz )";
value uniform (0 0 0);
}


The 'value' might need to be added since without it some warning will be appeared. According to my understanding 'value' will be used only for first time-step.