[wali]

Dear CFD Experts,

I am performing a natural convection problem, where the objective of the analysis is to determine the temperature distribution inside the box due to solar heat.
The reason for determining the temperature inside the box is that the client wants to put a box in an open environment and the box contains some fire suppression agent therefore there are some limitations that the surrounding air should not exceed than the certain temperature.

I also explained the problem graphically in the attached document. In order to perform the above analysis following steps are performed in ANSYS fluent 2020R2
a) Ground = define thermal properties of soil, soil thickness, temperature bc for soil and define absorptivity = 0.8
b) 5 sides of outer domain = pressure outlet, solar transmissivity =1
c) box wall = define shell conduction with absorptivity and participate in solar ray tracing
d) box bottom = define shell conduction, with absorptivity, and participate in solar ray tracing.
e) cylinder = define thermal properties of the cylinder wall, thickness and define adiabatic bc
f) Turbulence = k-epsilon realizable
g) air = ideal gas law
h) Operating density =0 (as per FLuent manual for ideal gas law)
i) gravity on = negative y-direction
j) Dimension of box = 0.5 x 0.5 x 1(m)
k) Dimension of outer domain = 5 x5 x5 (m)


Queries: When I plot the velocity vectors, i found that the flow is travelled from top to bottom ideally as per my understanding the flow should enter from the bottom faces of the pressure outlet and takes heat and then exit from the top faces of the pressure outlet. See attached image.
I understand the same problem is discussed earlier but none of the discussion reached to the final conclusion, therefore I would like to request to all CFD experts please provide your valuable thoughts about this physical behaviour of the velocity vectors. What conditions are wrong in my CFD simulation

Thanks in advance
Regards
Wali

[Alfinmp]

hey do you find your answer yet? i am facing the same problem here, i want to do some simulation of a rotating cylinder that being heated from bottom, but can not be able to do that.

[lei]

Hi WALI,

If you want to simulate natural convection in this problem, you need to correctly define the pressure boundary conditions at the five boundaries of your box (simulation domain).

As you mentioned in item b), you used pressure outlet b.c., which by default uses constant back pressure. This causes problems in calculating the right flow directions. Since you include gravity and reference air density of zero in operating conditions, you should put the correct boundary pressure profiles in the four vertical outlets, as a function of your elevation. If you have reference static pressure of 1 atm on the ground, then, p=1atm-rho*g*h, in those four boundary surfaces, and correspondingly the top surface. You can use profiles or expressions to define these pressure profiles in FLUENT.

You may also need to consider modeling correct heat source on ground due to solar radiation reception, etc.

By doing so, your simulation should show correct flow directions. Please let us know if this works for you.