Determining flow type of natural convection in cylindrical cavity
 

Hello,

I'm just out of college and staring my career in CFD. I have a question regarding natural convection in closed cavities.

How do I determine flow type for my CFD simulation?

I know that we can judge flow type by Reynolds (Re) number in forced convection and Rayleigh (Ra) number, which is a product of Grashof (Gr) and Prandtld (Pr) numbers, in natural convection. However, only critical values of Ra that i've found is 10^9 for vertical plate cooling (found in Y.Cengel "Heat Transfer. A Practical Approach." as well as T.L Bergman et all "Fundamentals of Heat and Mass Transfer") and 10^6 for squared cavity (found here: https://www.researchgate.net/publica...nclosed_cavity).

How do I proceed when my geometry is not one of those two. For example consider "cylinder in cylinder" cavity presented on the picture below: https://www.cfd-online.com/Forums/at...1&d=1625230301 .

I tried to find some resources with similar geometry and H/R ratio but I was not able to find anything. How then can judge character of my flow?

Best Regards,
Wojtek

I don't understand you case fully.

We have the natural convection, which mostly happens in a boundary layer (it is driven there).

Do we have a rotation cylinder in a fixed surroundings or do we have a rotation cell?

It is simply a cylindrical heat source with gas layer of thickness R outside. No rotation, arrows are just to show where fluid flow is happening. I just provided an example - my question is more general: How to judge weather flow is turbulent or laminar in natural convection in cavities.

I assume that you simulate temperature with the flow. Starting turbulence somewhere in the region means that you have a steep gradient of temperature: Laminar means you have mostly head conduction, a diffusion like process. In turbulence all is mixed and the process is convection dominated.

Nevermind, I just re-re-read your question and you have found the article on convection-driven cavities...
This might be a starting point, since they simulate non-square cavities. However, you might have to dig deeper into bibliography to find out about transition itself.

Also, keep in mind that this is a 2D study and not an axisymmetric one.

Thanks for the replies and sources, i'll look into them