|
[Sponsors] |
March 6, 2015, 14:09 |
IATE problem
|
#1 |
New Member
Join Date: Nov 2014
Posts: 11
Rep Power: 11 |
Hi,
I changed the dispersed phase (air) of "bubbleColumnIATE" to "perfectFluid" and the code can still run. Does this physically make sense? However when I change the density of dispersed phase (modeled as "perfectFluid" now) to be larger than the density of continuum phase, the code will give errors. Can anyone explain these two problem? Thank you. Mingzhao |
|
March 7, 2015, 05:50 |
|
#2 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 339
Rep Power: 28 |
Have a look at the file IATEsource.C and the definition of Ur. The calculation of Ur involves taking the difference of densities and taking the power of a term containing this difference.
As the density difference changes signs, you will get (probably) a floating point error. pow025() returns the 4th root of its argument (i.e. square root of the square root). Since, roots are not defined for negative numbers, you will get an error. If you want to use the IATE model for e.g. droplets dispersed in air, you will have to create a modified version of the IATE model with all density differences reversed. Code:
Foam::tmp<Foam::volScalarField> Foam::diameterModels::IATEsource::Ur() const { const uniformDimensionedVectorField& g = phase().U().db().lookupObject<uniformDimensionedVectorField>("g"); return sqrt(2.0) *pow025 ( fluid().sigma()*mag(g) *(otherPhase().rho() - phase().rho()) /sqr(otherPhase().rho()) ) *pow(max(1 - phase(), scalar(0)), 1.75); } |
|
March 9, 2015, 09:28 |
|
#3 | |
New Member
Join Date: Nov 2014
Posts: 11
Rep Power: 11 |
Quote:
Thank you very much for explaining and it is helpful. I also found there are high concentration of kappa in the area where we get pure primary phase. Does this make sense? I am expecting 0 interfacial area if alpha of dispersed phase is 0. Would you please also help me with this? Thank you. |
||
March 9, 2015, 09:43 |
|
#4 |
Senior Member
Gerhard Holzinger
Join Date: Feb 2012
Location: Austria
Posts: 339
Rep Power: 28 |
The interfacial area is the product of kappa and alpha. The transport equation for kappa is independent of the dispersed phase's volume fraction field. Having non-zero values of kappa everywhere in your domain is natural for this model. Nevertheless, this is a good thing since the Sauter diameter is defined as dSM = 6/kappa.
In regions with no dispersed phase present, you multiply a non-zero kappa with zero dispersed volume fraction. So, you get zero interfacial area in this region just as you expect. Chapter 43 of [1,2] is a summary of my findings. [1] https://github.com/OSCCAR-PFM/OSCCAR...Manual_PFM.pdf [2] http://www.cfd-online.com/Forums/ope...-openfoam.html |
|
March 9, 2015, 10:06 |
|
#5 |
New Member
Join Date: Nov 2014
Posts: 11
Rep Power: 11 |
Hi GerhardHolzinger,
Thank you very much for explaining and I understand now. The documents are also very helpful and it is very nice of you to offer these. Mingzhao |
|
April 6, 2015, 08:09 |
|
#6 |
Member
Join Date: May 2014
Location: Germany
Posts: 31
Rep Power: 11 |
Hello,
why is Ur modelled this way? In the originl paper by Wu [1] Ur is calculated with . Why is the implementation so different? Regards, hester [1] Wu, Q., Kim, S. and Ishii, M. One-group interfacial area transport in vertical bubbly flow. International Journal of Heat and Mass Transfer 41 (1998), p. 1103–1112. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
UDF compiling problem | Wouter | Fluent UDF and Scheme Programming | 6 | June 6, 2012 04:43 |
Gambit - meshing over airfoil wrapping (?) problem | JFDC | FLUENT | 1 | July 11, 2011 05:59 |
natural convection problem for a CHT problem | Se-Hee | CFX | 2 | June 10, 2007 06:29 |
Adiabatic and Rotating wall (Convection problem) | ParodDav | CFX | 5 | April 29, 2007 19:13 |
Is this problem well posed? | Thomas P. Abraham | Main CFD Forum | 5 | September 8, 1999 14:52 |