Boundary condition for free convection

the new one · October 16, 2013, 04:40

Hi
What are the suitable boundary condition for free convection in starccm+. I am trying to simulate natural convection of a sphere in open atmosphere. so my thought was that freestream conditon all around with very low velocity and boundaries at a very large distanec would work but it dooesnt seem to give physical results. Any suggestions.I have also already triend with adiabatic walls at a large distance.
any suggestions or your experiences with such problems will be appreciated.
Thanks

Cobra · June 13, 2017, 11:49

I know this answer is coming pretty late, but I think it still may be useful for someone looking for it.
You should build a spherical domain around your object, with a diameter 3 to 5 times the major diagonal of the object.
Boundary condition should be 'Pressure Outlet'. As pressure you should probably set Field Function 'Pressure' (not 100% sure here, but for me it worked most of the times).
When I tried to use a bigger domain (say 20 times the major diagonal) the solution couldn't converge.
I already posted some question in the forum about this specific problem.

me3840 · June 15, 2017, 14:44

Quote:

Originally Posted by Cobra

Boundary condition should be 'Pressure Outlet'. As pressure you should probably set Field Function 'Pressure' (not 100% sure here, but for me it worked most of the times).

Don't do this. The field function 'pressure' is solved for. The whole point of a pressure condition is to specify the pressure. You can make your own field function named pressure and use that, but do not use the internal field function pressure. I don't think it makes any sense to do that.

Cobra · June 16, 2017, 05:29

I agree, I was wrong in my previous reply. In the Adapco website I found a guide to set the boundary conditions for open air natural convection

https://thesteveportal.plm.automatio...S/FAQ/RD-5-112

From what I understood the Field Function should be written as:

-1.18*9.81*$$Position[2]*((300/$Temperature)-1)

where 1.18 is representing the reference density.

Right now I'm waiting for a renewal of my licence so I can't test its validity.

cwl · June 16, 2017, 09:17

How do you people get access to StevePortal? - Only via purchasing the license?

me3840 · June 18, 2017, 22:57

Remember that function requires you set your z=0 value to be the altitude where the piezometric pressure is the same as the static pressure.

Cobra · July 10, 2017, 14:39

Update: I tried to run some simulations with the new boundary condition for pressure, i.e. -1.18*9.81*$$Position[1]*((300/$Temperature)-1), which I found on the Steve Portal.
Unfortunately I can't make my simulation work with this BC either.
I still get fluid movements driven by pressure differentials at the atmospheric contour.
I tried the condition with both the Boussinesq and the Ideal Gas models.
I don't know what else I could try.

October 16, 2013, 04:40	Boundary condition for free convection	#1
the new one New Member Join Date: May 2013 Posts: 8 Rep Power: 12	Hi What are the suitable boundary condition for free convection in starccm+. I am trying to simulate natural convection of a sphere in open atmosphere. so my thought was that freestream conditon all around with very low velocity and boundaries at a very large distanec would work but it dooesnt seem to give physical results. Any suggestions.I have also already triend with adiabatic walls at a large distance. any suggestions or your experiences with such problems will be appreciated. Thanks Cobra likes this.

June 16, 2017, 05:29		#4
Cobra New Member Nicholas Join Date: Nov 2015 Location: Modena, Italy Posts: 20 Rep Power: 10	I agree, I was wrong in my previous reply. In the Adapco website I found a guide to set the boundary conditions for open air natural convection https://thesteveportal.plm.automatio...S/FAQ/RD-5-112 From what I understood the Field Function should be written as: -1.189.81$$Position[2]((300/$Temperature)-1) where 1.18 is representing the reference density. Right now I'm waiting for a renewal of my licence so I can't test its validity. Last edited by Cobra; June 18, 2017 at 10:40.*

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Time dependant pressure boundary condition	yosuke1984	OpenFOAM Verification & Validation	3	May 6, 2015 06:16
An error has occurred in cfx5solve:	volo87	CFX	5	June 14, 2013 17:44
Convection Boundary condition	tomcatbobby	FLUENT	2	April 30, 2012 13:50
asking for Boundary condition in FLUENT	Destry	FLUENT	0	July 27, 2010 00:55
RPM in Wind Turbine	Pankaj	CFX	9	November 23, 2009 04:05

June 13, 2017, 11:49		#2
Cobra New Member Nicholas Join Date: Nov 2015 Location: Modena, Italy Posts: 20 Rep Power: 10	I know this answer is coming pretty late, but I think it still may be useful for someone looking for it. You should build a spherical domain around your object, with a diameter 3 to 5 times the major diagonal of the object. Boundary condition should be 'Pressure Outlet'. As pressure you should probably set Field Function 'Pressure' (not 100% sure here, but for me it worked most of the times). When I tried to use a bigger domain (say 20 times the major diagonal) the solution couldn't converge. I already posted some question in the forum about this specific problem.

June 16, 2017, 09:17		#5
cwl Senior Member Chaotic Water Join Date: Jul 2012 Location: Elgrin Fau Posts: 435 Rep Power: 17	How do you people get access to StevePortal? - Only via purchasing the license?

June 18, 2017, 22:57		#6
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	Remember that function requires you set your z=0 value to be the altitude where the piezometric pressure is the same as the static pressure.

July 10, 2017, 14:39		#7
Cobra New Member Nicholas Join Date: Nov 2015 Location: Modena, Italy Posts: 20 Rep Power: 10	Update: I tried to run some simulations with the new boundary condition for pressure, i.e. -1.189.81$$Position[1]*((300/$Temperature)-1), which I found on the Steve Portal. Unfortunately I can't make my simulation work with this BC either. I still get fluid movements driven by pressure differentials at the atmospheric contour. I tried the condition with both the Boussinesq and the Ideal Gas models. I don't know what else I could try.