CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   swirlinjector (https://www.cfd-online.com/Forums/openfoam-solving/65032-swirlinjector.html)

az1362f June 26, 2009 16:58

3D mesh
 
1 Attachment(s)
hi Dear Niklas

what is my mistake?
what is its correction?
how can i describe an axi-symmetric mesh for diesel foam?

I ran a 3D simulation of dieselfoam with last characteristic of n-heptane experiment and add its result to a new plot and attached it.

there are two strange things:
1- I saw some fluctuations in liquid penetration curve in 3D cases.
2- likewise liquid penetration is far from experimental result.
what is the reason of these?

I set the initial drop diameter to nozzle diameter according to injector model.is it correct?

I am confused.
Please help me.

thanks
regard

az1362f June 27, 2009 13:35

axi-symmetric Mesh
 
2 Attachment(s)
Hi

I tried to solve last problem that I discussed in my last posts by axi-symmetric mesh but my results likewise were incorrect and far from experimental result.

I attached axi-symmetric Mesh.
please check my axi-symmetric blockmesh.
is it correct?
I tried to 3D simulation of dieselfoam of n-heptane experiment but results were far from experiment and had fluctuations.
Why are my result far from experiment?

please help me.
Thanks
regard

az1362f June 29, 2009 11:29

axi-symmetric Mesh
 
Hi Dr Niklas

I tried to solve last problem that I discussed in my last posts by axi-symmetric mesh but my results likewise were incorrect and far from experimental result.

I attached axi-symmetric Mesh.
please check my axi-symmetric blockmesh.
is it correct?
I tried to 3D simulation of dieselfoam of n-heptane experiment but results were far from experiment and had fluctuations.
Why are my result far from experiment?

please help me.
Thanks
regard

az1362f June 30, 2009 03:11

mesh dependency
 
1 Attachment(s)
Hi Dr Niklas;

I tried to solve axi-symmetric dieselfoam for n-heptane fuel,and when I compare my result by experimental result I couldn't find a mesh size of dependency.
I attached my axi-symmetric mesh in last posts and new plot of liquid penetration:

Is my axi-symmetric mesh correct?
if yes why liquid penetration increased as mesh size is refined?

Thanks
Regard

niklas June 30, 2009 04:39

Quote:

Originally Posted by az1362f (Post 220925)
Hi Dr Niklas;

I tried to solve axi-symmetric dieselfoam for n-heptane fuel,and when I compare my result by experimental result I couldn't find a mesh size of dependency.
I attached my axi-symmetric mesh in last posts and new plot of liquid penetration:

Is my axi-symmetric mesh correct?
if yes why liquid penetration increased as mesh size is refined?

Thanks
Regard

looks correct to me.

in general liquid penetration increase if you refine the mesh,
thats because you resolve the jet better and hence reduce the relative velocity.
You also have to investigate the influence on initial turbulent length scale, which also is
very important.
reducing the length scale will also increase the liquid penetration.
there are no ways around these problems for lagrangian/eulerian simulations.
you just have to live with the grid dependency.

az1362f July 1, 2009 12:45

Validating spray
 
Hi Dear Niklas;

then what is the way to validate spray simulation problem,when you don't know is your solution independent of mesh?

regard

az1362f July 4, 2009 11:55

interpolation scheme
 
1 Attachment(s)
Hi Dear Niklas;

Which interpolation scheme did you use for Inter-Phase Coupling in dieselfoam code?
Where can I see its implementation in dieselfoam code?
I attached two method of Inter-Phase Coupling,which one did you use in dieselfoam?

Thanks
Regard

niklas July 6, 2009 03:44

none of them

http://www.nequam.se/files/thesis.pdf
page 34

az1362f July 7, 2009 06:22

1 Attachment(s)
Hi Dear NIklas;

Which of interpolation scheme did you used in dieselfoam?

equation 116 or equation 117:

if 117 how is it for Hexahedron cell?

thanks

niklas July 8, 2009 02:08

in foam its always 117.

The cell is always split into a tet containing the parcel.
if you use cellPoint,
it will use the nodes and cell-center to construct the tet.

if you use cellPointFace it will use the nodes, cell center and face center to
construct the tet.

az1362f July 9, 2009 10:44

implementation
 
Hi Dear Niklas;

I am very thankful for your very good and complete answers.

I have a question about where is the implementation of eq.117 in Foam?

also,I studied your article by subject of "Numerical Investigation of Mesh/Turbulence/Spray
Interaction for Diesel applications",you talked in this article about limiting turbulenc lengh scale, is it limited in openfoam 1.5 for dieselfoam?
if no, where can i limit it?

Thanks
Regard

az1362f July 11, 2009 17:44

turbulent length scale
 
Hi Dear Niklas;

I am very thankful for your very good and complete answers.

I have a question about where is the implementation of eq.117 in Foam?

also,I studied your article by subject of "Numerical Investigation of Mesh/Turbulence/Spray Interaction for Diesel applications",you talked in this article about limiting turbulenc length scale, is it limited in openfoam 1.5 for dieselfoam?
if no, where can i limit it in the code according your article?

Thanks
Regard

az1362f July 13, 2009 11:09

turbulence length scale
 
Hi Dear Niklas;

I am very thankful for your very good and complete answers.

I have a question about where is the implementation of eq.117 in Foam?

also,I studied your article by subject of "Numerical Investigation of Mesh/Turbulence/Spray Interaction for Diesel applications",you talked in this article about limiting turbulence length scale, is it limited in openfoam 1.5 for dieselfoam?
if no, where can i limit it in the code according your article?

Thanks
Regard

az1362f July 20, 2009 04:51

finding parcel
 
Hi to all,

any body knows how openfoam check the existence of particle in an eulerian cell?
in the other word which function or class find if there is a parcel in the current cell?

az1362f July 22, 2009 12:51

limiter
 
Hello Dear Dr Nordin;

I saw in RASProperties file that for K-epsilon model there are two parameters for limiter :" muLimiter on;Lsgs 0.0002;

are these parameters activate turbulent length scale limiter in openFoam version 1.5?

please answer me.
thanks

regard

huoyinghw September 25, 2009 22:44

Hi,az1362f and other foamers:)
i also want to know how to caculate the liquid penetration length,and how i can change the liquid

chbenz September 28, 2009 11:32

Hi,

i use octave/matlab to calculate penetration length. the position of the parcel you can find in $FOAM_CASE/timeDirectory/lagrangian/defaultCloude/position.
The liquid is defined in $FOAM_CASE/thermophysicalProperites

Christof

huoyinghw October 1, 2009 06:23

Quote:

Originally Posted by chbenz (Post 230721)
Hi,

i use octave/matlab to calculate penetration length. the position of the parcel you can find in $FOAM_CASE/timeDirectory/lagrangian/defaultCloude/position.
The liquid is defined in $FOAM_CASE/thermophysicalProperites

Christof

Hi,
i have found the "position" file ,but i want to know how to calculate penetration length.Thank you very much,Christof
huoyinghw:)

MMK October 26, 2011 04:12

Quote:

Originally Posted by az1362f (Post 219236)
hi

I implemented these line in dieselfoam solver to save liquid peneteration lenght in every time step,but my result showed an approximately a constant liquid penetration.(for example 10.xxxx)

std :: fstream penet
penet.open("penetration.txt",std::ios::app);
penet << dieselSpray.liquidPenetration(1.0)<<"\t"<<runTime. timeName()<<"\n"<<endl;
penet.close();

what is my mistake?
I used default setting of dieselfoam tutorials.
I attached a penetration result.
please help me.


Hello Amin,

how did you implement this function can you share it, i m having this problem to implement in dieselFoam

conceptone June 12, 2012 22:42

hi dude,I confront a problem similar to yours now, where should I go to for changing the chem.inp and therm.dat files?


All times are GMT -4. The time now is 04:29.