CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   sprayFoam (http://www.cfd-online.com/Forums/openfoam/112273-sprayfoam.html)

Akuji January 25, 2013 04:34

sprayFoam
 
Hello!

I need a solver to visualized a liquid spray. As I see in tutorials of OF, one of solvers I can use is sprayFoam (I need only first 2 steps: spray breakup and spray tracking).
At 1st I made a tutorial aachenBomb. the results are ok: filed of velocity and pressure seem to be real, but when I try to see particles [foamToVtk and then load sprayCloud], paraview closed with "Segmentation fault".

UPD. On another machine paraview work ok,but I don't see any particles :(.It seems there are no particles.

Does anyone know what do I do wrong?

Maybe someone has a similar task and can help me.

Another question: I found an information about CFDEM and LIGGGHTS. As I understand it is possible to solve my problem using it. But I didn't find any information of their tutorials. Maybe I compiled it not correct, but I didn't see any results after making tutorial cfdemSolverPiso/settlingTestMPI.
Maybe someone can give me more information about it.

mturcios777 January 25, 2013 13:54

Since you mentioned sprayFoam I'm assuming your are using OF 2.1.x. You should be able to just load the defaultCloud in ParaFoam and then extract it using the extractBlock filter. Try that and see what happens.

Akuji January 28, 2013 09:37

2 Attachment(s)
Quote:

Originally Posted by mturcios777 (Post 404056)
Since you mentioned sprayFoam I'm assuming your are using OF 2.1.x. You should be able to just load the defaultCloud in ParaFoam and then extract it using the extractBlock filter. Try that and see what happens.

Thanks for advance, mturcios777. I use OF-2.1.0 you were right.
About paraview: I made as you sad. Load defaultCloud (hope you mean spray.foam), then I used extractBlock filter. After that I choose in pipelan browser interanlMesh, but after it I can load velocity field and pressure field on the same picture together. But I didn't find particles.

I copy from another directory ParticleTrackProperties in constant/ and use particleTracks. After that I have a directory VTK, inside it - particleTracks.vtk
This file I load in paraview, then use Glyph filter - visible spheres. The results of my operation I attach. But I have a doubt - are they right? For my opinion tracks of particles are not right.
Velocity field and pressure field are ok.Attachment 18554
Attachment 18555

mturcios777 January 28, 2013 13:28

Hi Akuji,

What are you scaling the glyph diameter by? Sometimes the distribution can look misleading because of a uniform diameter. Also, what does the temperature look like?

cngz January 28, 2013 15:10

evaporating_drops.ccl
 
Hi,
It is not related to spray foam but could you please send the tutorial file "evaporating_drops.ccl" It is for "Spray Dryer" tutorial? I need it so badly!
alimecngz@gmail.com

mturcios777 January 28, 2013 15:14

I don't have this tutorial case, I'm not sure what version of OpenFOAM it belongs to. Sorry

cngz January 28, 2013 15:17

it belongs to ANSYS CFX 13. Thanks anyway :)

Akuji January 29, 2013 06:40

Quote:

Originally Posted by mturcios777 (Post 404577)
Hi Akuji,

What are you scaling the glyph diameter by? Sometimes the distribution can look misleading because of a uniform diameter. Also, what does the temperature look like?

I use scale mode - scalar. Set scale factor - automatically.
About temperature: at time 0.0015 there is no field of temp. Because I don't need a part of tutorial with combustion, I cut all times where there is no any injection of particles.
The diameter of scaling is 1 (r=0.5). If I choose smaller radius, image is the same, just particles look smaller.

mturcios777 January 29, 2013 14:17

I know you can scale the particles to be proportional to their mean diameter. Is there any evaporation happening?

Akuji February 8, 2013 05:05

1 Attachment(s)
Hello!

About evaporation - really I don't know. The last time when parcels are added is t=0.0125:

Time = 0.00125


Solving cloud sprayCloud

--> Cloud: sprayCloud
Added 56 new parcels

Cloud: sprayCloud
Current number of parcels = 18944
Current mass in system = 4.12019e-07
Linear momentum = (-1.64279e-07 -3.21616e-05 -3.42757e-08)
|Linear momentum| = 3.21621e-05
Linear kinetic energy = 0.0025296
Rotational kinetic energy = 0
Total number of parcels added = 123490
Total mass introduced = 5.99999e-06
Parcel fates:
- escape = 0, 0
- stick = 0, 0
Mass transfer phase change = 5.58798e-06
D32 (mu) = 6.71635
Liquid penetration 95% mass (m) = 0.0193151

As I understand, evaporation exist, because current number of parcels if 18 944, and total number of parcels added is 123 490.
Another step I did, i finally understand how to use vtk file for my time. The spray seems to be real.
Question I have: how to see the distribution of parcels according to their diameter? D32 is 6.7e-6m (am I right?). So, to see the particles I use glyph type sphere, radius = 1, scale mode = 6.7e-6. And my particles have one size and one color.

mturcios777 February 8, 2013 13:36

I would also plot the mass fraction of what is evaporating. If all the particles are constant diameter, then the distribution looks feasible, since parcels break up as we leave the nozzle, increasing the number of parcels while evaporation reduces the number.

I would also plot the mass fraction of the species that is evaporating in the droplets.

Akuji March 5, 2013 07:52

3 Attachment(s)
Well, after some weeks of solving I still have questions.

1. How does sprayFoam calculate D32? According to my calculation I have particles size 1.3mm, and foam calculate as 21581.9 (*10^-6 am I right?).

2. I don't need an evaporation, combustion, chemiacl reactions and other parts. I am interested only in the particle size, path and the particle interaction and particle-to-wall interactions. I turn off evaporation, combustion and etc., change chem.inp file but still have dependence of Temperature:
Code:

DILUPBiCG:  Solving for hs, Initial residual = 0.000893686, Final residual = 1.45528e-10, No Iterations 1
T gas min/max  = 252.514, 320.275

Variation of temperature is too much.


So, example of output in my terminal:
Code:

Cloud: sprayCloud
    Current number of parcels      = 17136
    Current mass in system          = 0.171842
    Linear momentum                = (7.11313e-06 1.24111e-05 -0.218256)
  |Linear momentum|                = 0.218256
    Linear kinetic energy          = 5.49125
    Rotational kinetic energy      = 0
    Total number of parcels added  = 17136
    Total mass introduced          = 0.171842
    Parcel fate (number, mass)
      - escape                      = 0, 0
      - stick                      = 0, 0
    Mass transfer phase change      = 0
    D10, D32, Dmax (mu)            = 2027.54, 21581.9, 99996.5
    Liquid penetration 95% mass (m) = 0.076189

diagonal:  Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG:  Solving for Ux, Initial residual = 0.000416571, Final residual = 5.39093e-11, No Iterations 1
DILUPBiCG:  Solving for Uy, Initial residual = 0.00125617, Final residual = 1.83274e-10, No Iterations 1
DILUPBiCG:  Solving for Uz, Initial residual = 0.000726465, Final residual = 5.80187e-11, No Iterations 1
DILUPBiCG:  Solving for H2O, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG:  Solving for hs, Initial residual = 0.000893686, Final residual = 1.45528e-10, No Iterations 1
T gas min/max  = 252.514, 320.275
DICPCG:  Solving for p, Initial residual = 0.00118179, Final residual = 2.15544e-08, No Iterations 1
diagonal:  Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 4.94038e-11, global = 1.54421e-14, cumulative = 1.204e-08
DICPCG:  Solving for p, Initial residual = 2.68447e-07, Final residual = 2.68447e-07, No Iterations 0
diagonal:  Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 6.18547e-10, global = 4.59799e-15, cumulative = 1.204e-08
DILUPBiCG:  Solving for omega, Initial residual = 3.89693e-07, Final residual = 3.89693e-07, No Iterations 0
DILUPBiCG:  Solving for k, Initial residual = 1.07217e-05, Final residual = 2.39036e-12, No Iterations 1

I have no idea what is wrong (well, maybe I choose wrong solver?:))

If you have an idea or any free time, please, see my attached files: I think there are some mistakes too.

niklas March 5, 2013 09:54

Quote:

Originally Posted by Akuji (Post 411627)
Well, after some weeks of solving I still have questions.

1. How does sprayFoam calculate D32? According to my calculation I have particles size 1.3mm, and foam calculate as 21581.9 (*10^-6 am I right?).

http://foam.sourceforge.net/docs/cpp/a05046_source.html
Check the Dij-function, D32 = Dij(3,2)

The biggest droplets are around 0.1m...which i think probably is a mistake.

Quote:

Originally Posted by Akuji (Post 411627)

2. I don't need an evaporation, combustion, chemiacl reactions and other parts. I am interested only in the particle size, path and the particle interaction and particle-to-wall interactions. I turn off evaporation, combustion and etc., change chem.inp file but still have dependence of Temperature:
Code:

DILUPBiCG:  Solving for hs, Initial residual = 0.000893686, Final residual = 1.45528e-10, No Iterations 1
T gas min/max  = 252.514, 320.275

Variation of temperature is too much.


So, example of output in my terminal:
Code:

Cloud: sprayCloud
    Current number of parcels      = 17136
    Current mass in system          = 0.171842
    Linear momentum                = (7.11313e-06 1.24111e-05 -0.218256)
  |Linear momentum|                = 0.218256
    Linear kinetic energy          = 5.49125
    Rotational kinetic energy      = 0
    Total number of parcels added  = 17136
    Total mass introduced          = 0.171842
    Parcel fate (number, mass)
      - escape                      = 0, 0
      - stick                      = 0, 0
    Mass transfer phase change      = 0
    D10, D32, Dmax (mu)            = 2027.54, 21581.9, 99996.5
    Liquid penetration 95% mass (m) = 0.076189

diagonal:  Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG:  Solving for Ux, Initial residual = 0.000416571, Final residual = 5.39093e-11, No Iterations 1
DILUPBiCG:  Solving for Uy, Initial residual = 0.00125617, Final residual = 1.83274e-10, No Iterations 1
DILUPBiCG:  Solving for Uz, Initial residual = 0.000726465, Final residual = 5.80187e-11, No Iterations 1
DILUPBiCG:  Solving for H2O, Initial residual = 0, Final residual = 0, No Iterations 0
DILUPBiCG:  Solving for hs, Initial residual = 0.000893686, Final residual = 1.45528e-10, No Iterations 1
T gas min/max  = 252.514, 320.275
DICPCG:  Solving for p, Initial residual = 0.00118179, Final residual = 2.15544e-08, No Iterations 1
diagonal:  Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 4.94038e-11, global = 1.54421e-14, cumulative = 1.204e-08
DICPCG:  Solving for p, Initial residual = 2.68447e-07, Final residual = 2.68447e-07, No Iterations 0
diagonal:  Solving for rho, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 6.18547e-10, global = 4.59799e-15, cumulative = 1.204e-08
DILUPBiCG:  Solving for omega, Initial residual = 3.89693e-07, Final residual = 3.89693e-07, No Iterations 0
DILUPBiCG:  Solving for k, Initial residual = 1.07217e-05, Final residual = 2.39036e-12, No Iterations 1

I have no idea what is wrong (well, maybe I choose wrong solver?:))

If you have an idea or any free time, please, see my attached files: I think there are some mistakes too.


How much time have elapsed and how much water are you supposed to inject, because that looks like quite alot. (remember that injected mass is in kg)

niklas March 5, 2013 09:58

remember also that length is in meters, not millimeters

Akuji March 9, 2013 06:30

niklas,

thank you for your help. I find Dji in KinematicCloudI.H, but didn't understand how it calculates (my c++ knowledge is very bad). But I see then more I have mass flow then less diametr my particles have.
About Dmax: I don't know exactly how solver works, but in my work I have drops of water. They interract between themselves: can divide or couple for more smaller or larger particle. Dmax=0.1m is probably puddle on the floor or on the wall (it is my explanation why it is so large).

I reviewed my b.c. and my mass flow is ok (I plan to inject 10 kg/s from my nozzle). My mesh is in meters.

Akuji March 18, 2013 03:58

Hello again!

A added a Blob Atomization model for primary break-up, but have no difference with my previous effort.

Now I have a problem: when my particles reach disk they have absolutely elastic impact. They fly to the patch which from they fall. And for them it is not important that it is injection still work. In fluent I have very nice results for my room, but in OF there is no any spray dissemination :(

Any ideas?..
http://www.ljplus.ru/img4/f/u/fucking_blondy/age.jpg

Akuji April 4, 2013 09:54

Really sad, but particles just bounce between out-patch and disk-patch :(

http://www.ljplus.ru/img4/f/u/fucking_blondy/100.jpg

Vito31388 April 22, 2013 11:59

Hi Akuji, I don't understand how visualize in paraview particles in the spray cloud for aachen bomb tutorial; I've used foamToVTK to create VTK files but when I select spraycloud.vtk and I choose glyph,I don't see the particles.Can you explain me the procedure?

Akuji April 22, 2013 14:45

Quote:

Originally Posted by Vito31388 (Post 422347)
Hi Akuji, I don't understand how visualize in paraview particles in the spray cloud for aachen bomb tutorial; I've used foamToVTK to create VTK files but when I select spraycloud.vtk and I choose glyph,I don't see the particles.Can you explain me the procedure?

Hi, Vito!

In the example I load spraycloud.vtk, choose glyph: type - sphere and scale mode - scalar, then apply. If you still don't see particles, try to edit scalar factor: for example, use not 0.002 but 0.02 - you will see difference. And use for example time=11, not 0 or 1.
Good luck!

ruamojica July 4, 2016 17:44

sprayCloudProperties
 
Hello friends... in this moment I'm studying the code of sprayFoam and I don't know what means de "youngsModulus" and "poissonRatio" at the file sprayCloudProperties, this properties belong to what substance? the liquid or the ambient gas?

Thanks .


All times are GMT -4. The time now is 19:48.