[C.C]

Dear all,

I am trying to simulate the natural convection in an horizontal enclosure: adiabatic vertical walls, a top cold wall and a hot bottom wall. I am simulating the natural convection of water using the Boussinesq model. However, I am not able to get correct pressure values, i.e. for 0.5 m of distance between top and bottom wall I was expecting to get 4180 Pa, but the value obtained was 10 Pa. Moreover, the pressure is higher in the top than in the bottom...
Anyone can help me, I was looking through the manuals for an example, but I was not able to find it. Please, anyone have a reference paper which I can use to simulate and validate what I am doing...
Best regards,

C.C.

[Svetlana]

https://uiuc-cse.github.io/me498cm-f...n-natural.html

De Vahl Davis, G. (1983), Natural convection of air in a square cavity: A bench mark numerical solution. Int. J. Numer. Meth. Fluids, 3: 249–264. doi:10.1002/fld.1650030305

[Kushal Puri]

Originally Posted by C.C

Dear all,

I am trying to simulate the natural convection in an horizontal enclosure: adiabatic vertical walls, a top cold wall and a hot bottom wall. I am simulating the natural convection of water using the Boussinesq model. However, I am not able to get correct pressure values, i.e. for 0.5 m of distance between top and bottom wall I was expecting to get 4180 Pa, but the value obtained was 10 Pa. Moreover, the pressure is higher in the top than in the bottom...
Anyone can help me, I was looking through the manuals for an example, but I was not able to find it. Please, anyone have a reference paper which I can use to simulate and validate what I am doing...
Best regards,

C.C.

What is the operating density and pressure you are using.

[C.C]

Originally Posted by Kushal Puri

What is the operating density and pressure you are using.

I am using operating density equal to 0 kg/m^3 and operating pressure of 101325 Pa... Please do you have any idea that can help me to solve this question?

[Kushal Puri]

Originally Posted by C.C

I am using operating density equal to 0 kg/m^3 and operating pressure of 101325 Pa... Please do you have any idea that can help me to solve this question?

Change the operating density to density of water at temperature say 20 degree celcius. And simulate again

[C.C]

Originally Posted by Kushal Puri

Change the operating density to density of water at temperature say 20 degree celcius. And simulate again

Thank you for your reply. I have changed the operating density to the density of water, but still I am obtaining values of pressure that are unexpected to me..

[Kushal Puri]

Originally Posted by C.C

Thank you for your reply. I have changed the operating density to the density of water, but still I am obtaining values of pressure that are unexpected to me..

Can you provide more details of your problem, like boundary conditions

[C.C]

Originally Posted by Kushal Puri

Can you provide more details of your problem, like boundary conditions

As boundary conditions I have:
- top wall, cold wall, T = 273.15 K
- bottom wall, hot wall, T = 293.15 K
- left and right walls are adiabatic walls

Natural convection in a enclosure of 1m x 0.5m....

[Kushal Puri]

Originally Posted by C.C

As boundary conditions I have:
- top wall, cold wall, T = 273.15 K
- bottom wall, hot wall, T = 293.15 K
- left and right walls are adiabatic walls

Natural convection in a enclosure of 1m x 0.5m....

What about gravity, you turned it on ?

[C.C]

Originally Posted by Kushal Puri

What about gravity, you turned it on ?

Indeed, I have turned it on: gravitational acceleration, Y (m/s2) = -9.81...

[reicht4]

I'm not expert but have few questions. Why people using Boussinesq for a very well known temperaute range for air properties especially.

You set your operating temp to 20 C but your range is between 0-90 degrees right? Try to calculate Boussinesq model density output, by using thermal expension, operating temperature density, instant temperature. And we know air density values vs temperature change. This is a narrow band. So compare your findings with experimental density values.

If you choose an op temperature far away from the average in your temperature range (20 C for 0-90 C band) you will see an error in Boussinesq model density vs experimental which is groving non-lineerly while it is closing to top temperature. I believe this could effect results dramatically. If you are using constant thermal expension rate for operating temperature, this would even make worse.

I was going to open a thread to ask my method but after i saw a natural convection problem in the first page (usual) i instead decided to post here. What do i do is, entering tabular data for density, viscosity, thermal conduction values for air vs temperature changes. Of course this works in a narrow band (-50 ~ 100 C usually) but Boussinesq's itself doesnt work for wider!
I'm making a simulation right nnow and converging faster after i did this. And the most important is it is stable.

So please consider and evaluate my method people.
Cheers