Community
CFD Forum
Welcome to Autodesk’s CFD Forums. Share your knowledge, ask questions, and explore popular CFD topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Air flow through sand

11 REPLIES 11
Reply
Message 1 of 12
javorpanev
1426 Views, 11 Replies

Air flow through sand

Hi,

I would like to model an airflow through sand. Any ideas of how to model the sand or any irregular, high porosity structure?

11 REPLIES 11
Message 2 of 12

Typically we would model this using a distributed resistance and then assigning some Permeability to that volume. Note we will not show that the sand will move but rather assume a fixed shape

Message 3 of 12

I tried by by modeling a volume of sand with assigned air ratio of 0.05 in all directions and constant conductivity of 0.71 W/m2K. The sand is surrounded from all sides by a fluid (air) while the least fluid thickness is 5mm. Unfortunately the simulation does not converge.

Message 4 of 12
Jon.Wilde
in reply to: javorpanev

Does the resistance have a good uniform mesh applied to it? 4-5 elements as a minimum through the thickness? 

How about the rest of the model, what is the overall setup?

 

Thanks,

Jon

Message 5 of 12
javorpanev
in reply to: Jon.Wilde

I have attached the simulation fille.

 

Setup:

Air is blown through the pipe on the top at 30l/min in a cylindrical container. There is a heater at the bottom of the cylinder with heat release rate of 100W. The cylinder is filled with sand. All surfaces are exposed to the environment with ambient temperature of 20C and convection coefficient of 5W/m2 K.  I am interested in the air flow and thermal distribution through the cylinder.

 

So far I did a successfully simulation without the sand, but with the sand volume modelled as resistance material it does not converge. Ideally I want the sand to be as much close to the cylinder walls as possible, although I understand that the resistance volume has to be surrounded by a fluid from all sides.

Message 6 of 12
Jon.Wilde
in reply to: javorpanev

Please could you share a CFZ file?

Your setup sounds about right though 🙂

 

We do not have to have a fluid on all walls of a resistance, although you do need a fluid or solid.

Message 7 of 12
javorpanev
in reply to: Jon.Wilde

Here is the CFZ file and a few pictures of the setup.

 

 

Message 8 of 12
Jon.Wilde
in reply to: javorpanev

A few comments which I think will leave this running OK:

 

  1. Remove the film coeff from your outlet, I don't think you will need it
  2. You have no inlet temperature, which is needed, how will the solver know how much the air has heated up over the element?
  3. You could just have sand from wall to wall with air above and below it, that might actually run better too as the flow will not try to bypass the sand and recirculate above it but all just exit upwards (recirculation over a boundary condiction will usually lead to solver instability)
  4. Do you need gravity on and air variable? As you have forced airflow, I would say this is unnecessary
  5. Running with auto forced convection on is actually negating buoyancy as it runs flow first and then locks the flow to run thermal
  6. Also, do you need radiation on? 

Hope that helps.

Jon

 

Message 9 of 12
javorpanev
in reply to: Jon.Wilde

Thank you Jon for the suggestions!

 

I tried forced flow without gravity and with sand touching the walls, but the solution still does not converge 😞

 

Yavor

Message 10 of 12
Jon.Wilde
in reply to: javorpanev

Could you share the CFZ again please? I think it might need a longer outlet as there may be recirculation over the boundary condition, happy to test it here.

Message 11 of 12
javorpanev
in reply to: Jon.Wilde

Here you are. So far I have successfully run simulations without the presence of the sand under both forced and natural convection settings. They both give sensible results and suggest possible circulation in the region of the outlet.

 

Thank you!

Message 12 of 12
Jon.Wilde
in reply to: javorpanev

To get around the recirculation issue, can you extend the outlet air? Try 5x diameter in length at least.

 

I would set your resistance flow direction to z, as that is the main flow direction and try again.

Also use Advection Scheme 5.

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report