[tahtawy]

I am solving High Rayleigh natural convection problem ( air in square cavity whose vertical sides are heated and horizontal ones are adiabatic )
so do i consider gas an ideal gas or use Boussinesq approximation?
when Boussinesq approximation is used?
thank you

[Carlo_P]

Hey Tahtawy.
In the Boussinesq approximation you consider the density function of the static temperature, also if normally it depends also on the pressure and the velocity.


So, if in your simulation there are small variations in pressure and small velocity, you can use the Boussinesq approximation.


Otherwise, no.


Cheers,
Carlo

[FMDenaro]

Originally Posted by tahtawy

I am solving High Rayleigh natural convection problem ( air in square cavity whose vertical sides are heated and horizontal ones are adiabatic )
so do i consider gas an ideal gas or use Boussinesq approximation?
when Boussinesq approximation is used?
thank you

This model can be applied for small temperature difference

[tahtawy]

Originally Posted by FMDenaro

This model can be applied for small temperature difference

by saying small , it is not specific
my case is 40 degree difference

[tahtawy]

Originally Posted by Carlo_P

Hey Tahtawy.
In the Boussinesq approximation you consider the density function of the static temperature, also if normally it depends also on the pressure and the velocity.


So, if in your simulation there are small variations in pressure and small velocity, you can use the Boussinesq approximation.


Otherwise, no.


Cheers,
Carlo

will bousinessq here will aid in converging the residual ?

[FMDenaro]

Originally Posted by tahtawy

by saying small , it is not specific
my case is 40 degree difference

The approximation is based on a linear expansion. Just valid for few degree of variation, 40 is too much

[Carlo_P]

Uhm...I don't agree to much..CFX Tutorial use this approxiamtion with a 70 degree of difference..

[FMDenaro]

Originally Posted by Carlo_P

Uhm...I don't agree to much..CFX Tutorial use this approxiamtion with a 70 degree of difference..

https://www.sciencedirect.com/scienc...1793101300820X

[Carlo_P]

Hey FMDenaro, thanks to post that article!


Uhm..so maybe the CFX tutorials are not so trustable..


Thanks again,
Cheers,
Carlo

[Moreza7]

Originally Posted by Carlo_P

Uhm...I don't agree to much..CFX Tutorial use this approxiamtion with a 70 degree of difference..

For 15 degrees difference for the air, it generates 1 percent of error. So, maybe for the air, 70 degrees difference may cause a small error and reasonable solution where the accuracy is not critical.
Its only valid for the air, for other fluids, it depends on the expansion factor at constant pressure. For example the error is much bigger for the water.