CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Simulation of a Silo (http://www.cfd-online.com/Forums/cfx/80926-simulation-silo.html)

Attesz October 11, 2010 10:55

Simulation of a Silo
 
Hi CFX users,

I'm going to simulate a silo full of corn, with hot air blowing in. The main problem is the material properties of the corn :) My first idea is model the corn as a porous domain (i've never done such a thing). The main goal is to determine the temperature distribution in the silo, so actually the temp.dist of that 6 tonns of corn.

Thanks in advance for any suggestions!

Attila

ghorrocks October 11, 2010 18:07

Does the corn move? Then maybe a fluidised bed. If the corn is still then a porous domain approach makes sense. In the preview version of CFX V13 I think you can couple the porous material temperature with the air temperature and that is probably important for this - so talk to CFX support to get the latest preview version.

Attesz October 12, 2010 03:03

Hello Glenn,
the corn is still, and it fills up the silo.

Thank you,
Attila

michael_owen October 12, 2010 15:40

Glenn is correct, to do the CHT in the porous domain, you will either have to get R13 Preview 3, or you will have to do is yourself using CEL, sources, and an Additional Variable.

Attesz October 12, 2010 15:49

Hi Michael,
we have research license at the university, can we download the R13 in this case? If not, doing it myself is difficult? Is there good documentations in the topic? Thanks.

michael_owen October 12, 2010 16:01

Quote:

Originally Posted by Attesz (Post 278873)
Hi Michael,
we have research license at the university, can we download the R13 in this case?

I'm not sure. You should contact your account rep and ask them. Also, it's in beta in R13, so you probably shouldn't be relying on it for production work anyway.

Quote:

If not, doing it myself is difficult? Is there good documentations in the topic? Thanks.
Define "difficult". To my knowledge there is no documentation on that specific topic. You will need to create a diffusive transport Additional Variable that represents the temperature of the porous solid and create an Energy Source in the Porous domain to remove/add energy to model the heat transfer, and an associated Additional Variable source. You will need to create CEL expressions to model the heat transfer. You will need a volumetric surface area density (m^2 m^-3) and a heat transfer coefficient. You'll need the specific heat of the corn, the thermal conductivity, which would have to be converted to a diffusivity for the additional variable, etc. It might take me a day or two to develop such a model from scratch.

Attesz October 12, 2010 16:22

Okay, thanks for your reply. I will need documentations in simulation of porous domains, and heat transfer(i think the tutorials will be not enough to learn these topics). The parameters of the corn is more problematic. Anyway, it is a little bit difficult for me, but that's the challange of CFD.

Thanks again,
Attila

michael_owen October 12, 2010 16:43

Also, FYI R13 Preview 3 officially ended back in August, so you are probably out of luck trying to get a copy.

michael_owen October 12, 2010 16:48

You may be in luck. It turns out from the world's largest coincidence that I now have to implement this in R12.1 by tomorrow for a client. If I get it working, I'll tell you how I did it.

michael_owen October 12, 2010 17:35

Ok, it only took me 45 minutes. :p

michael_owen October 12, 2010 17:53

Here's the CCL for the model that does CHT for a porous domain (the mesh is just a tube with an inlet and an outlet):

The domain:

FLOW: Flow Analysis 1
&replace DOMAIN: Porous Domain
Coord Frame = Coord 0
Domain Type = Porous
Location = B6
BOUNDARY: Inlet
Boundary Type = INLET
Interface Boundary = Off
Location = F8.6
BOUNDARY CONDITIONS:
ADDITIONAL VARIABLE: Solid Temperature
Option = Zero Flux
END
FLOW REGIME:
Option = Subsonic
END
HEAT TRANSFER:
Option = Static Temperature
Static Temperature = 25 [C]
END
MASS AND MOMENTUM:
Normal Speed = 10 [m s^-1]
Option = Normal Speed
END
TURBULENCE:
Option = Medium Intensity and Eddy Viscosity Ratio
END
END
END
BOUNDARY: Outlet
Boundary Type = OUTLET
Interface Boundary = Off
Location = F9.6
BOUNDARY CONDITIONS:
ADDITIONAL VARIABLE: Solid Temperature
Option = Zero Flux
END
FLOW REGIME:
Option = Subsonic
END
MASS AND MOMENTUM:
Option = Average Static Pressure
Pressure Profile Blend = 0.05
Relative Pressure = 0.0 [Pa]
END
PRESSURE AVERAGING:
Option = Average Over Whole Outlet
END
END
END
BOUNDARY: Porous Domain Default
Boundary Type = WALL
Create Other Side = Off
Interface Boundary = Off
Location = F7.6
BOUNDARY CONDITIONS:
ADDITIONAL VARIABLE: Solid Temperature
Additional Variable Value = 0 [C]
Option = Transfer Coefficient
Transfer Coefficient = 10 [m s^-1]
END
HEAT TRANSFER:
Option = Adiabatic
END
MASS AND MOMENTUM:
Option = No Slip Wall
END
WALL ROUGHNESS:
Option = Smooth Wall
END
END
END
DOMAIN MODELS:
AREA POROSITY:
Option = Isotropic
END
BUOYANCY MODEL:
Option = Non Buoyant
END
DOMAIN MOTION:
Option = Stationary
END
MESH DEFORMATION:
Option = None
END
REFERENCE PRESSURE:
Reference Pressure = 1 [atm]
END
VOLUME POROSITY:
Option = Value
Volume Porosity = 0.7
END
END
FLUID DEFINITION: Fluid 1
Material = Air at 25 C
Option = Material Library
MORPHOLOGY:
Option = Continuous Fluid
END
END
FLUID MODELS:
ADDITIONAL VARIABLE: Solid Temperature
Kinematic Diffusivity = Thermal Diffusivity
Option = Diffusive Transport Equation
END
COMBUSTION MODEL:
Option = None
END
HEAT TRANSFER MODEL:
Option = Thermal Energy
END
THERMAL RADIATION MODEL:
Option = None
END
TURBULENCE MODEL:
Option = k epsilon
END
TURBULENT WALL FUNCTIONS:
Option = Scalable
END
END
INITIALISATION:
Option = Automatic
INITIAL CONDITIONS:
Velocity Type = Cartesian
ADDITIONAL VARIABLE: Solid Temperature
Additional Variable Value = 25 [C]
Option = Automatic with Value
END
CARTESIAN VELOCITY COMPONENTS:
Option = Automatic with Value
U = 0 [m s^-1]
V = 0 [m s^-1]
W = 10 [m s^-1]
END
STATIC PRESSURE:
Option = Automatic with Value
Relative Pressure = 0.0 [Pa]
END
TEMPERATURE:
Option = Automatic with Value
Temperature = 25 [C]
END
TURBULENCE INITIAL CONDITIONS:
Option = Medium Intensity and Eddy Viscosity Ratio
END
END
END
POROSITY MODELS:
LOSS MODEL:
Loss Velocity Type = Superficial
Option = Isotropic Loss
ISOTROPIC LOSS MODEL:
Option = Linear and Quadratic Resistance Coefficients
Quadratic Resistance Coefficient = 650 [kg m^-4]
END
END
END
SUBDOMAIN: Sources
Coord Frame = Coord 0
Location = B6
SOURCES:
EQUATION SOURCE: Solid Temperature
Multiply by Porosity = No
Option = Source
Source = Solid Temperature Source
Source Coefficient = Solid Temperature Source Coefficient
END
EQUATION SOURCE: energy
Multiply by Porosity = No
Option = Source
Source = Energy Source
Source Coefficient = Energy Source Coefficient
END
END
END
END
END



The Additional Variable:

LIBRARY:
&replace ADDITIONAL VARIABLE: Solid Temperature
Option = Definition
Tensor Type = SCALAR
Units = [K]
Variable Type = Volumetric
END
END



The Expressions:

LIBRARY:
CEL:
&replace EXPRESSIONS:
Energy Source = -Heat Transfer Coefficient*Volumetric Surface Area Density*(Temperature - Solid Temperature)
Energy Source Coefficient = -Heat Transfer Coefficient*Volumetric Surface Area Density
Heat Transfer Coefficient = 100 [W m^-2 K^-1]
Solid Density = 1000 [kg m^-3]
Solid Specific Heat Capacity = 100 [m^2 s^-2 K^-1]
Solid Temperature Source = -Energy Source/(Solid Specific Heat Capacity * Solid Density)
Solid Temperature Source Coefficient = -Energy Source Coefficient/(Solid Specific Heat Capacity * Solid Density)
Solid Thermal Conductivity = 10 [W m^-1 K^-1]
Thermal Diffusivity = Solid Thermal Conductivity/Solid Density/Solid Specific Heat Capacity
Volumetric Surface Area Density = 100[m^2 m^-3]
END
END
END



Solver Control:

FLOW: Flow Analysis 1
&replace SOLVER CONTROL:
Turbulence Numerics = First Order
ADVECTION SCHEME:
Option = High Resolution
END
CONVERGENCE CONTROL:
Length Scale Option = Conservative
Maximum Number of Iterations = 100
Minimum Number of Iterations = 1
Timescale Control = Auto Timescale
Timescale Factor = 100
END
CONVERGENCE CRITERIA:
Conservation Target = 0.01
Residual Target = 1e-6
Residual Type = RMS
END
DYNAMIC MODEL CONTROL:
Global Dynamic Model Control = On
END
END
END


Be careful of the Solid Density, thermal conductivity, specific heat capacity; they are for the porous matrix itself, not the material it is made of.

Attesz October 13, 2010 11:30

Dear Michael,

thank you , it was very kind from you to help me so much! Additional variables, expressions are known for me, i was in trouble only with the theoretical background, but it seems to be getting clear now.

Do you have any ideas how to set the thermal parameters for the porous domain if I would have the parameters of the corn? If i will lucky, there will be measures too, so I can validate the material properties.

Thanks again,
Attila

Attesz October 13, 2010 11:34

Quote:

You may be in luck. It turns out from the world's largest coincidence that I now have to implement this in R12.1 by tomorrow for a client.
That client was my built-in man, but it's a secret :)

Attesz October 13, 2010 11:51

Quote:

Volumetric Surface Area Density = 100[m^2 m^-3]
And it means that for example the gross surface area of the maize grains is 100 m^2 in 1 m^3 volume?

These parameters will be hard to found out later...but the thermal setup is clear. Thanks again.

michael_owen October 13, 2010 13:01

Quote:

Originally Posted by Attesz (Post 279013)
Dear Michael,

thank you , it was very kind from you to help me so much! Additional variables, expressions are known for me, i was in trouble only with the theoretical background, but it seems to be getting clear now.

No problem. Was a fun nut to crack, and I had to do it anyway.

Quote:

Do you have any ideas how to set the thermal parameters for the porous domain if I would have the parameters of the corn? If i will lucky, there will be measures too, so I can validate the material properties.
The density is just the mass of a volume of packed corn divided by that volume. Get a bucket and some corn and measure it. The specific heat capacity is already per unit volume, so you should be able to just convert the specific heat capacity of an individual kernel using the volume porosity. The thermal conductivity I'm unsure of; I think you will have to directly measure it (think about this, what's the thermal conductivity of a matrix of packed ideal spheres? Zero!).

Quote:

Originally Posted by Attesz (Post 279017)
And it means that for example the gross surface area of the maize grains is 100 m^2 in 1 m^3 volume?

Yes, although at the granular level are are geometric effects that are being neglected.

Quote:

These parameters will be hard to found out later...but the thermal setup is clear. Thanks again.
That one should be fairly easy to estimate. You should be able to calculate the surface area of a kernel of corn, and how many kernals there are per unit packed volume. Voila.

michael_owen October 13, 2010 13:02

By the way, what is this for? Research paper? Dissertation? Class project? Work?

Attesz October 14, 2010 06:00

Quote:

You should be able to calculate the surface area of a kernel of corn, and how many kernals there are per unit packed volume. Voila.
It's clear, but how to measure a the surface area of a kernel?

Quote:

By the way, what is this for? Research paper? Dissertation? Class project? Work?
We are going to modernize the heating system of a corn pre-heating silo and its parts. The current gas heater will be replaced by a biomass based heater, and so the ventilation system of the silo will be modified. The main problem could be that the corn will not reach the necessary temperature, or maybe it could be overheated. The temperature distribution could be also important for the client. Otherwise, I'm going to be a PhD student hopefully from february, so it can be a research project as well.

Thanks for your suggestions again, Michael!

Attila

michael_owen October 14, 2010 21:14

Assume the kernel is a sphere. Rectilinear. It's somewhere in between. Cad model it and measure it.

Attesz October 15, 2010 07:01

Okay, it's a good idea.

Thanks!

Best regards,
Attila

Josh October 15, 2010 07:52

If the heat input is very high, can we expect popcorn to form?


All times are GMT -4. The time now is 19:49.