Divergence (pressure correction) for axysymmetric density driven flow

Sxq · November 29, 2013, 18:33

Hi!

I'm trying to simulate an unsteady density driven/negative buoyancy event where I have a volume of low density ambient fluid, and a smaller cylindrical volume of more dense fluid which is initialized inside the ambient fluid and released. There is no forcing other than the density difference, and the idea is that the volume of dense fluid will fall downward and hit the bottom wall of the region.

I made a 2D axisymmetric domain with one of the side walls having an axis BC, the other side wall being a pressure outlet, the top being a pressure outlet, and the bottom being a wall.
I used the Eularian 2 phase model to define both liquids.
I'm using the SST k-omega turbulence model.

However the solution diverges as soon as I start trying to solve. I think it has something to do with my BCs... It doesn't seem to like the way I have set up the axisymmetric condition perhaps? I am able to run a similar flow condition when I approximate it as a 2D planar flow instead.
Any help on why it diverges, or confirmation that the models I'm using are valid would be great... Here is a schematic of the domain:

http://imgur.com/Wqsc2l1

flotus1 · November 30, 2013, 07:28

Does your axis boundary condition coincide with the x-axis?
Fluent can only handle axisymmetric flows with this domain orientation.

Sxq · November 30, 2013, 16:56

My axis boundary condition is coincident with the y-axis. Will this actually cause it to not work?

Sxq · November 30, 2013, 18:39

Yep... it only works along the x-axis... Thanks. Now I just need to figure out the smaller details, but it seems to be running well.

November 29, 2013, 18:33	Divergence (pressure correction) for axysymmetric density driven flow	#1
Sxq New Member Join Date: Nov 2013 Posts: 4 Rep Power: 12	Hi! I'm trying to simulate an unsteady density driven/negative buoyancy event where I have a volume of low density ambient fluid, and a smaller cylindrical volume of more dense fluid which is initialized inside the ambient fluid and released. There is no forcing other than the density difference, and the idea is that the volume of dense fluid will fall downward and hit the bottom wall of the region. I made a 2D axisymmetric domain with one of the side walls having an axis BC, the other side wall being a pressure outlet, the top being a pressure outlet, and the bottom being a wall. I used the Eularian 2 phase model to define both liquids. I'm using the SST k-omega turbulence model. However the solution diverges as soon as I start trying to solve. I think it has something to do with my BCs... It doesn't seem to like the way I have set up the axisymmetric condition perhaps? I am able to run a similar flow condition when I approximate it as a 2D planar flow instead. Any help on why it diverges, or confirmation that the models I'm using are valid would be great... Here is a schematic of the domain: http://imgur.com/Wqsc2l1

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Divergence detected in AMG solver: pressure correction	thisisit	FLUENT	10	November 9, 2022 06:20
Pressure Driven Supersonic Flow in Converging-Diverging Nozzle	cdm	OpenFOAM Running, Solving & CFD	7	August 28, 2019 01:37
Divergence detected in AMG solver: pressure correction	xinquanzhoucn	FLUENT	5	July 21, 2014 04:49
Help! Divergence detected in AMG solver: pressure correction	user0314	FLUENT	3	September 2, 2011 16:12
Hydrostatic pressure in 2-phase flow modeling (CFX4.2)	HB &DS	CFX	0	January 9, 2000 13:19

November 30, 2013, 07:28		#2
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,399 Rep Power: 46	Does your axis boundary condition coincide with the x-axis? Fluent can only handle axisymmetric flows with this domain orientation.

November 30, 2013, 16:56		#3
Sxq New Member Join Date: Nov 2013 Posts: 4 Rep Power: 12	My axis boundary condition is coincident with the y-axis. Will this actually cause it to not work?

November 30, 2013, 18:39		#4
Sxq New Member Join Date: Nov 2013 Posts: 4 Rep Power: 12	Yep... it only works along the x-axis... Thanks. Now I just need to figure out the smaller details, but it seems to be running well.