3ddp-solver

use 3DDP solver....

parameter u.r of momentum and pressure by default...

results:
after iteration 9 AMG Solver still serro...

according to the cfdfluent.com (discussion) says that need to use u.r one by one .... still don't have pattern or formula how to set the discretization methods....

parameter u.r of momentum and pressure by momentum=0.39 and pressure=0.31...
after iteration 4139 error AMG solver...

parameter u.r of momentum and pressure by momentum=0.39 and pressure=0.61...
after iteration 1 error AMG solver...

parameter u.r of momentum and pressure by momentum=0.61 and pressure=0.39...
after iteration 937 error AMG solver...

so try about BC's of the Simulation.....

equisize-skew->-0.97

Meshing success? but EquiSIZE SKEW >0.97 (11 or 16 cells)?

May have to re-fine meshing of the geometry...by:
*. coarse mesh then if need more fine mesh have to try methods below...

Because Meshing in GAMBIT automatically generated mesh (especially in 3-D) in most cases, can not provide uniformly high quality in un-structured mesh (hexahedral or tetrahedral) for complex geometry.

On the surface, the unstructured mesh does have advantages to cover large areas of complex geometry, but, in reality, a high quality mesh is always hard to obtain.

One says that need to go back to the blocking of the geometry (meshing process): to divide the geometry into smaller pieces and re-create the surface geometry so that can avoid the highly skewed cells.

**. have to try long iteration until 10^-6 criterion meet.... (already done, but still not convergence)....

have to try one by one....

And .... just an information from CFD source: it must take more time in Meshing Process and Convergence Process......

Under-relaxation-momentum

About:....

Error: AMG Solver ...temperature

Solution:
* by Modification of Under-relaxation of Momentum become less than default (0.7) have to try become 0.6....or: 0.5, 0.4, 0.3, 0.35, 0.39....
for 0.5: 500iteration success but after iteration 701 error AMG Solver again...
for 0.4: error after iteration 1185...
for 0.3: error afater iteration 359...
for 0.35: error after iteration 763...
for 0.39: error after iteration 1750...
for 0.396 or 0.389: have to simulate beside check about Meshing or Coupled Solver and not Segregated...

for 0.396: error after iteration 197...
for 0.389: error after iteration 857...
for 0.37: error after iteration 456...
So, want to check it up of coupled...

conclution for u.r. of momentum is about=0.39 and for pressure is about=0.3

but ok before fixed it...need to set the under-relaxation of Pressure....the default is 0.3 but i have to try the 0.28, 0.29 and 0.32 to know influence of this value...

for this use the same convergence criteria that is 1.0e-06 for all graphics...
for 0.28: error after iteration 1086...
for 0.29: error after iteration 264...
for 0.3: error after iteration 1750...already done
for 0.31: error after iteration 2229...
for 0.32: error after iteration 838...

So, the best configuration based-on under-relaxation of pressure and momentum is 0.31 and 0.39 for this case eventhough it's still has error in AMG solver: temperature.

Now, before try coupled solver....need to try to change the the BC's becouse one says that AMG solver and Reversed flow come from settings in BC's....

AMG-Solver-Temperature

Message in the screen.....when simulate in FLUENT after GAMBIT process....

Error: divergence detected in AMG solver: temperature
Error Object: ()

Is it about computer?

One suggest me to:
*Make sure about Under-Relaxation:.....I have to try to do it....
*the message indicates that doing something wrong in settign up the case (boundary conditions)...
*Make up Coupled Solver.
*check about mesh.

I have to do it....hope fully it become succeesfull


About the reversed flow pressure outlet.....please to make sure about the BCs and ICs...Because: In fact (real conditions), there should be no backflow at the pressure outlet in the converged solution. But, backflow occurs during the iteration. If so what should I do?

8-possibilities-meshing-process

elements type smoother spacing sources Results
HEX MAP None 0.6 NA ERROR: Entity V.5 can not be mesh on lower entity f.79
0.15 NA f.77
0.1; 1.0; 1.0 NA f.75; f68; f67
Hex
Sub map
NA 1.0; 0.6 NA F67; F62
hex Tet primitive NA 0.6 NA Connectivity for v.5 does not allow meshing using the tetrahedral primitive scheme
hex cooper NA 0.6; 1.0 F.63; 67; 68; 69 = all from 4th ventilating lids F38, 37, 57, 5, 42, 53, 24, 22 is Not appropriate for use as a face to project along. Either it is not sub mapped or the choice of source faces is incorrect
HEX STAIRSTEP NA 1.0; 0.6; 2.0 NA Successfully meshed V.5. Created faceted volume (s). F_volume.6 with mesh volumes = 5564
HEX/wedge cooper NA 0.6 NA F38, 37, 57, 5, 42, 53, 24, 22 is Not appropriate for use as a face to project along. Either it is not sub mapped or the choice of source faces is incorrect

Tet/hybrid hexcore NA 0.6 NA Mesh generated for v.5: mesh volumes = 47457. Contains 16 highly skewed element (EQUISIZE SKEW > 0.97)
Tet/hybrid tgrid na 0.6 NA Mesh generated for v.5: mesh volumes = 51780. Contains 11 highly skewed element (EQUISIZE SKEW > 0.97)

There are three meshing process that success for the geometry....
[1]. Stairstep =mesh volume=5,564.
[2]. Tgrid =mesh volume=51,780.
[3]. Hexcore =mesh volume=47,457.

Meshing-Process

Not 'convergence' yet? Is is caused by meshing still displayed "equisize skew > 0.97" (although already done about three of eight meshing 'success')?

reversed-flow

...
reversed flow in 8 faces on pressure-outlet 9.

reversed flow in 9 faces on pressure-outlet 13.
6009 3.3860e-04 1.7957e-01 3.4764e-01 1.9047e-01 5.7751e-04 0:00:00 0

why not displayed word "CONVERGENCE" in the monitor (althought the simulation already done for 1000 iteration = 13minutes and 33seconds --> 81 minutes and 3 seconds)?

[Check the CONVERGENCE status……….by:]
1. Compute from: no need to set… (because already set from geometry by GAMBIT)
2. Solve-Control-Solution—Under-relaxation
Solve-Control-Solution—Pressure-Velocity Coupling
Solve-Control-Solution—Discretization; or
3. Solve-Monitor-Residual--Convergence Criteria; or
4. Meshing process about Equisize SKEW > 0.97 (11, 16 or … by Tet/Hybrid- TGrid and Hexcore

:-)smile

divergence-in-AMG-solver

after running the simulation for airflows I see only an error:

...
reversed flow in 23 faces on pressure-outlet 4.

reversed flow in 31 faces on pressure-outlet 5.

reversed flow in 35 faces on pressure-outlet 6.

reversed flow in 38 faces on pressure-outlet 7.

reversed flow in 52 faces on pressure-outlet 8.

reversed flow in 8 faces on pressure-outlet 9.

reversed flow in 8 faces on pressure-outlet 13.

Error: divergence detected in AMG solver: temperature
Error Object: ()

why?.....how to fixed it?.....

solution for that error: in windows of FLUENT click SOLVE then click Initialize-Initialize-set Velocity Magnitude = 1 or other values then click Reset-Apply-Close.

Another solution is to click Initialize-Initialize- then click Init...

...continuousimprovement of lasmanp's thesis...

Holman1992-Natural-Convection-System

Natural, or free, convection is observed as a result of the motion of the fluid due to density changes arising from the heating process.

The movement of the fluid in free convection, wheter it is a gas or liquid, results from the bouyancy forces imposed on the fluid when its density in the proximity of the heat-transfer surface is decreased as a result of the heating process.

The bouyancy forces would not be present if the fluid were not acted upon by some external force field such as gravity, although gravity is not only type of force field which can produce the free-convection currents; a fluid enclosed in a rotating machine is acted upon by a centrifugasl force field, and thus could experince free-convection currents if one or more of the surfaces in contact with the fluid were heated.

The bouyancy forces which give rise to the free-convection currents are called body forces.

heat-flux

Heat flux is the rate of energy transfer through a given surface. This quantity can be measured using a heat flux sensor. The measurement of heat flux is of importance to many sciences. Most common applications are in building physics, where the heat flow through walls is one of the factors determining the indoor climate, in agricultural meteorology, where the heat flux into the soil is a parameter in the study of evaporation of water, and biology to measure heat flux from humans or animals. The accurate measurement of heat flux can lead to energy saving in buildings and to more efficient use of water in irrigated agricultural area's [http://www.hukseflux.com/heat%20flux/flux.htm]

Leonhard-Euler-Wikiedia

In fluid dynamics, the Euler equations govern the motion of a compressible, inviscid fluid. They correspond to the Navier-Stokes equations with zero viscosity and heat conduction terms, although they are usually written in the form shown here because this emphasises the fact that they directly represent conservation of mass, momentum, and energy. The equations are named after Leonhard Euler. This page assumes that classical mechanics applies; see relativistic Euler equations for a discussion of compressible fluid flow when velocities approach the speed of light. [http://wikipedia.com]

Madsen,2006-use-eulerian-model

In the Eulerian multi-fluid model, the gas and particle phases are treated considered as the primary considered as dispersed or secondary phases. The gas and particle phases as interpenetrating continua in an Eulerian framework. The gas phase is considered as primary phase whereas the particle phases are secondary phases. The gas and particle phase are characterized by volume fractions, and by definition, the volume fractions of all phases must sum to unity: (lasmanp thesis: 0.9 for gas, 0.1 for particle).

The governing equations of the multi-fluid model can be derived by conditionally ensemble averaging of the local instant conservation equation of single-phase flow (Drew, 1983; Drew and Passman, 1999). In the Madsen, J., (2006) thesis the flow assumed to be isothermal; hence, energy balances are not needed. Furthermore, there is no interfacial mass transfer between the gas and particle phases.

turbulence-model-for-each-phase

+*solves a set of k and ε transport equations for each phase.
+*is appropriate choice when the turbulence transfer among the phases plays a dominant role.

+*Turbulence predictions are obtained from two long equations which both of that equations have two terms that must be approximated like these:
* Clq = 2,
* Cql = 2ηlq/(1+ηlq).
where: ηlq is the ratio of the characteristic particle relaxation time and the Lagrangian integral time scale of the phases q and l.

+*The turbulent viscosity defined as a equations.

Recomended for the thesis (1st priority), and use LAUNDER and SPALDING (1974), because so famous used by researchers like Lu and Howatrh (1996a, 1996b). Although we can use in simulation as 2nd priority,dispersed t. m. or mixture t. m. as the 3rd priority for these works.

dispersed-turbulence-model

This model is appropriate when the concentrations of the secondary phases are dilute. In this case, interparticle collisions are negligible and the dominant process in the random motion of the secondary phases is the influence of the primaryphase turbulence. Fluctuating quantities of the secondary phases can therefore be given in terms of the mean characteristics of the primary phase and the ratio of the particle relaxation time and eddy-particle interaction time.

Turbulence in the continuous phase is described by:
- a modified k-ε model,
- the influence of the dispersed phase on the continuous phase q, and
- the production of turbulent kinetic energy
- For the turbulence quantities of the dispersed phase user can use Mr. 'X' approach (for e.g. Muehlbauer, P 2004 use Simonin, Viollet (1990) approach).

Think that 1st phase=gas, air, and 2nd phase=particles, water,

mixture-turbulence-model

Mixture turbulence model (default in FLUENT), applicable for stratified (or nearly stratified) multiphase flows, and when the density ratio between phases is close to 1. In these cases,
using mixture properties and mixture velocities is sufficient to capture important features
of the turbulent flow. There are 5th equation for modeling turbulence in this mixture turbulence model, i.e.:
- ske-2 eqn
- mixture density
- mixture velocity
- the turbulent viscosity
- the production of turbulence kinetic energy

Sohn, et. al. 2004-using-CFD

CFD is a mathematical modeling procedure whereby the fluid parameters of velocity,
temperature, pressure, turbulence, and contaminant concentrations are calculated
by solving the governing partial differential equations for fluid flow, heat
transfer, and conservation of species. These differential equations describe a threedimensional
viscous fluid flow field. Due to the non-linearity of these equations,
they cannot be solved analytically. The CFD approach is to transform these differential
equations into a set of discrete algebraic equations and solve the algebraic
equations by an iterative procedure.
Researchers have used CFD since the early 1970s. Its use has increased dramatically
in the last decade as a result of advances in the computing power. In the
1980s, Cray supercomputers typically were used to process CFD simulations, with
solution times taking days (which followed a long wait to even gain access to use the
supercomputer). The same simulations may now be run on a personal laptop computer
in a matter of hours. CFD modeling has been continually validated since its
inception against many known fluid phenomena. It is considered a very useful tool
for engineers and scientists working on fluid flow problems across many disciplines.

Muehlbauer, et. al., 2001-future-needs-on-CFD

The document which written by Muehlbauer, P., (2001) describe the present capabilities of more than three CFD-type computer codes to simulate two-phase flows and the future needs on the two-phase CFD computer codes. Principal approaches to description of two-phase flows, their advantages, and disadvantages are discussed. Also for each of the codes two-phase flow and heat transfer are already described in some detail. From all of CFD-type computer codes that described, there are three CFD codes i.e., FLUENT, STAR-CD, and CFX which have more capabilities especially for “3D codes for open medium” to simulate two-phase flows and there is a short description of the two-phase flow capabilities of several other CFD-type computer codes i.e., TRIO-U, NEPTUNE, and PHOENICS.

CFD-Packages-in-the-market

[1]. CBNP Transport & Fate Team, http://www.lanl.gov/orgs/d/d4/pdf/air/model_review.pdf accesed May 2007. (Chemical & Biological Nonproliferation Program) had been classified become five-types of computer codes that already used in the world to modeling transport and fate phenomenons, i.e.,: CFD Models for Flow around Building, Building Interior Flow and Transport Models, Mesoscale Atmospheric Models, Plume Dispersion Models, and Subway System & Interior Flow Models.

[Examples of CFD Models for flow around buildings are: Dynaflow, FEM3C, GASFLOW, HIGRAD, and TEMPEST.
Examples of building Interior Flow and Transport Models are: COMIS, MIAQ4, CONVECT9, GASFLOW.
Examples of Mesoscale Atmospheric Models are: COAMPS, HOTMAC, NORAPS, RAMS.
Examples of Plume Dispersion Models are: ADPIC, HYPACT, LODI, RAPTAD, SALB.
Examples of Subway system & Interior flow models: GASFLOW, SES.]

[2]. Sohn, et. al. (2004) already reviewed many existing technologies that could protect the supply air systems of buildings from chemical/ biological (chem/bio) contaminants such as FEM3MP, HPOC, CONTAM, dan FLOVENT. FEM3MP and HPOC for external dispersion modeling, CONTAM and FLOVENT for internal dispersion modeling. The results provide a modeling approach that can be used to simulate the dispersion of contaminants inside buildings. Such a numerical simulation will help designers and managers evaluate possible ways to reduce occupant exposures to chem/bio contaminants.

[M]odeling Package Name: FLOVENT
[A]uthor: Flomerics Incorporated)
[N]umerical Simulation Description: CFD - Specific to the built environment(internal/external flows)

[M] Airpack
[A] FLUENT Incorporated
[N] CFD - Specific to the built environment (internal/external flows)

[M] STAR-CFD
[A] Adapco Group
[N] CFD - Multi-purpose code

[M] CFX
[A] ANSYS
[N] CFD - Multi-purpose code

CONTAM 2.1
National Institute of Science and Technology (NIST)
Nodal Analysis - Multizone indoor air quality and contaminant transport analysis.

IAQX (MIAQ4) (Nazaroff and Cass, 1989)
EPA
Add on to RISK for particulate, sources, spills

COMIS (Festel, 1999)
LBNL
Conjunction of Multizone Infiltration Specialists

Integration of CFD and COMIS (Sohn, et. al., 2006)
LBNL
Current research area for LBNL Airflow and Pollutant Transport Group

[3]. CFD is a mathematical modeling procedure whereby the fluid parameters of velocity, temperature, pressure, turbulence, and contaminant concentrations are calculated by solving the governing partial differential equations for fluid flow, heat transfer, and conservation of species. These differential equations describe a threedimensional viscous fluid flow field. Due to the non-linearity of these equations, they cannot be solved analytically. The CFD approach is to transform these differential equations into a set of discrete algebraic equations and solve the algebraic equations by an iterative procedure.

[4]. This thesis is about internal dispersion,... and have two fluid flow i.e.: particle and gas.

[5]. STAR-CFD, CFX dan FLUENT in two phase modeling.

[6] Sippola and Nazaroff (2002), describe four broad methods of predicting particle deposition rates that found in the literature: empirical equations, Eulerian models, sublayer models and Lagrangian simulations. These methods usually require information about the particle size and density, as well as the air speed and dimensions of the duct containing the flow. Deposition rates are most commonly reported in the form of the dimensionless deposition velocity, Vd+, versus the dimensionless relaxation time, τ+, a measure of particle inertia.

...[to be continued]...

Hinds1999-characteristic-of-particles

Real particles are not spherical in form, but have different shapes. That is why the aerodynamic particle diameter is introduced for simulations. The aerodynamic diameter used in the simulations, dp, is the diameter of the unit density ( ρp = 1 g/cm3) sphere that has the same settling velocity as the particle. Particles can appear as human hair, parts of human skin and parts of different building or other materials (Hinds, 1999).