[Natural Convection] "Infinite" velocity field

Vittorio · September 5, 2015, 18:43

Hello everybody. I am a student from Roma, Italy and I am working on my thesis on natural convection. I am almost completely new to CFD and I don't have a strong knowledge in fluid dynamics.

I have difficulties in giving a physical explanation to the velocity field that I obtain in natural convection simulation when I have an object in a large (tall) domain. For large I mean 10+ times the size of the object.

I am try to simulate natural convection in a 2D case. I am using COMSOL software and I am working on different cases, starting from this tutorial: https://www.comsol.it/model/natural-...uum-flask-1448

As you can see in the documentation, I am simulating natural convection using a volume force caused by density variations, the formula is: -g_const*(ht.rho-rho_ref) where ht.rho is the actual value of the density and rho_ref is the density at infinity (boundary).

I can't understand why the velocity field is the one that you can see in the attached image ProblemaCOMSOL. I would expect the velocity to decrease at some point but this doesn't happen.

I am also attaching the plot of the density.

I tried different boundary conditions (normal stress=0, pressure=patm, velocity=0) but I always get similar results. Excluding the area next to the boundary, the velocity doesn't seem to decrease.

I would like to know if anybody has a model of natural convection with a large domain.

I am working with ANSYS Fluent to reproduce similar cases to understand what is happening.

Please let me know what you think. Thanks!

khaoula · April 11, 2016, 14:47

Hello,
Have you got any sollution with COMSOL?

I have nearly the same problem in modeling natural convection on COMSOL.
Let me know if you have sollutions.
Best regards.

pivalse · April 16, 2016, 08:56

Dear Vittorio,
The hot air accelerates vertically due to buoyancy. As long as the density difference is present, the flow keeps accelerating or at least maintains the vertical speed.
The flow slows down when the density difference is reduced, but since there is no turbulence this effect is very slow.
Additionally, the flow describes a vertical jet that has no spread. The momentum of the fluid particles moving vertically is not transferred to the surrounding ones. What viscosity are you using?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Foam::error::PrintStack	almir	OpenFOAM Running, Solving & CFD	91	December 21, 2022 04:50
A Question About Setting Whole Velocity Field Using "Proflie" in FLUENT	adsl17754	FLUENT	5	July 31, 2018 04:29
VELOCITY vs VELOCITY IN STN FRAME vs RELATIVE VELOCITY	everest20	FLUENT	1	July 13, 2015 08:35
velocity of massless particles and velocity of flow field in same coordinate	payam_IUST	FLUENT	0	October 18, 2009 23:24
Extract velocity field in certain time step to MATLAB	spatialtime	CFX	3	September 9, 2009 18:30

April 11, 2016, 14:47		#2
khaoula New Member Kalaa kebira Join Date: Mar 2016 Posts: 1 Rep Power: 0	Hello, Have you got any sollution with COMSOL? I have nearly the same problem in modeling natural convection on COMSOL. Let me know if you have sollutions. Best regards.

April 16, 2016, 08:56		#3
pivalse New Member Join Date: Apr 2014 Posts: 11 Rep Power: 11	Dear Vittorio, The hot air accelerates vertically due to buoyancy. As long as the density difference is present, the flow keeps accelerating or at least maintains the vertical speed. The flow slows down when the density difference is reduced, but since there is no turbulence this effect is very slow. Additionally, the flow describes a vertical jet that has no spread. The momentum of the fluid particles moving vertically is not transferred to the surrounding ones. What viscosity are you using?