|
[Sponsors] |
August 12, 2014, 20:26 |
Unsaturated Van Genuchten
|
#1 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
Hello,
Imīm trying to model a simple model of unsaturated flow using the Van Genuchten model. However, after setting all the inputs the prepricessor gives me de following error: *************************************** ** fatal error ** ** capillary pressure curves not ** ** properly defined ** *************************************** Any guess? Thank you! |
|
August 13, 2014, 17:49 |
|
#2 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
These are my porous component inputs:
remark='!! Component 1 properties', opor(1)=0.7, oadrg(1)=6.38e-12, obdrg(1)=0.55337, ofcmn(1)=0.029, ofcmx(1)=0.366, iopcfd(1)=1, iodfit(1)=3, alphwobs(1)=3503.57, xnwobs(1)=2.239, xmwobs(1)=0.366, itpobs(1)=0, iaqsrb(1)=0, Is there anything else I should enter? |
|
August 14, 2014, 11:32 |
|
#3 |
Senior Member
Jeff Burnham
Join Date: Apr 2010
Posts: 204
Rep Power: 17 |
If you have an older version, try adding the following namelist:
&pcap fpcmx=0.12, remark=' capillarity parameters ', pcmx=2.62e+4, pexp=0.5, / These parameters are global and should not be used by the solver (since the component parameters you've provided over-write them), but in older versions of FLOW-3D the namelist might still be called and these placeholders will let you run the case better. Try it and let me know if it works. If not, the variable list you need to define for the Mualem drag model + van Genuchten model, and it looks like you might be missing the drying curve coefficients: ALPHWOBS = wetting curve alpha coefficient in units of pressure XNWOBS = van Genuchten wetting coefficient, fit from data XMWOBS = 1 - 1/n = 1 - 1/XNWOBS ALPHDOBS = drying curve alpha coefficient = ALPHWOBS or fit from data XNDOBS = XNWOBS or fit from data XMDOBS = 1 - 1/XNDOBS or fit from data OPOR = Vpores/Vbulk = 1 - ρ_bulk/ρ_solids OADRG = OPOR|g|/K_HC = OPOR μ/(k_IP ρ) OBDRG = m = XMWOBS, or find from linear regression of the data where OPOR = volumetric porosity, in FLOW-3D (units of m^3 open/m^3 solid = dimensionless). OADRG = drag coefficient 'a' of Mualem drag model, a function of saturated flow in the media (units of s^-1). OBDRG = drag coefficient 'b' of Mualem drag model, a function of unsaturated capillary pressure in the media (dimensionless). XMWOBS = van Genuchten model wetting coefficient 'n' (see User Manual for equations). ρ_bulk = bulk density of the media (units of kg media/m^3 media, including solid and pores). ρ_solids = micro density of the solid component used in the media (units of kg solids/m^3 solids). |g| = magnitude of gravity (units of m^2/s). K_HC = coefficient of saturated hydraulic conductivity in units of length (units of m/s), k_IP = coefficient of saturated intrinisic permeability (units of m^2). μ = fluid viscosity (units of kg/m/s). ρ = fluid density (units of kg/m^3) m = coefficient of Van Genuchten capillary pressure model (dimensionless) |
|
August 14, 2014, 19:31 |
|
#4 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
Thank you Mr. Burnham! It's running know!
However I'm not sure I understood which was the problem... Could you explain me what these comands you told me to add are exactly doing? Again, thank you! |
|
August 14, 2014, 19:47 |
|
#5 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
Should I paste these commands whenever I want to rerun?
|
|
August 14, 2014, 20:47 |
|
#6 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
My prepin has only:
&PCAP ivg=1, / |
|
August 15, 2014, 15:24 |
|
#7 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
Iīm confused about how flow 3d defines the capillarity pressure.
For example, if I put a water level as initial condition, does it recognize the capillarity pressure on every point of the domain based on the water level? And what if I want to put a capillarity pressure as boundary condition, to simulate a wetting condition, how do I do that? |
|
August 15, 2014, 15:54 |
|
#8 |
Senior Member
Jeff Burnham
Join Date: Apr 2010
Posts: 204
Rep Power: 17 |
There's an unsaturated porous media in the included examples: File > Add Example > Porous Medium Flow. That shows how to use pressure boundaries to make a wet boundary condition. As for the capillary pressure, it's computed in all cells in the porous component. It's based on the fluid fraction in the cell, the history of the cell (wetting or drying) and the capillary pressure curves you defined with the coefficients. Please refer to the User Manual for more detail.
|
|
August 17, 2014, 10:33 |
|
#9 |
New Member
Victor Dobrochinski Barbosa
Join Date: Aug 2014
Posts: 7
Rep Power: 12 |
Thank you! You helped a lot!
|
|
Tags |
flow 3d, porous media input, unsaturated, van genuchten |
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
van Driest 1956 , classic paper | Ferramis | Main CFD Forum | 3 | January 18, 2017 09:16 |
Incomp. LES in pisoFOAM, how to set up Smagorinsky model with van Driest dampin | Jan.Östh | OpenFOAM Running, Solving & CFD | 21 | August 10, 2016 12:03 |
Particle Interaction is CFX; Adding Interparticle Attraction such as Van Der Waals | FrankS | CFX | 11 | November 12, 2013 19:53 |
van leer second order method | k_tafazoli | Main CFD Forum | 0 | March 12, 2013 11:20 |
Adding Van der Waals force by coding UDF | Long Sang | FLUENT | 1 | June 23, 2008 07:44 |