Simulation the free convection in fluent

ashkan_1984 · May 25, 2013, 16:16

Dear all,
I want to simulate natural free convection in fluent but I cant.I do this thing up to now and I cant reach to answer:
1-In operation condition I give -9.81 in the y part.
2-I give the density of water(my channel is full of water) in businesqe
please help me

ashkan_1984 · May 26, 2013, 06:16

I cant understand any thing

alireza9359300 · May 26, 2013, 15:50

hi. its steady or unsteady simulation.

ashkan_1984 · May 27, 2013, 02:38

Unsteady 3D simulation

alireza9359300 · May 27, 2013, 11:14

3d+natural+unsteady simulation is out of the power of normal computer.
change your problem into 2d
if you use temperature bc increase its value to see the result sooner
reduce the mesh in your problem to use less memory of your computer
your time must be large to see anything.

ashkan_1984 · May 28, 2013, 07:50

I don't have problem about memory only I can't simulate natural convection.

alireza9359300 · May 28, 2013, 16:21

what about time of your simulation?
by small time you cant see any changes in domain of natural sim.

so you must use large time:
by: little time step + large number of time steps + large iteration number in each time step + 3d + natural + small mesh + high order of solving = be sure you will have memory problem of using normal computers.

if you don't have memory problem just you must insert large time for natural simulation to see changes in domain another way is to increase the value of the bc ( if you are using temperature bc ) to have more influence on your fluid.

if I were you
first I simulate it steady after it try to solve it unsteady , cause of for being sure it is true.

if you like . can you send me case & data file of your problem.(little size .
before solving ) alireza9359300@yahoo.com

JimKnopf · May 30, 2013, 03:48

And don't under relax the enegry equation

alireza9359300 · May 31, 2013, 17:33

I think your problem is density (change it to compressible fluid not boussings)
I sent it for you.

JimKnopf · June 1, 2013, 03:52

He already said he's using Boussinesque ansatz, that should work.

As he is writing about a channel, it sounds to me as if he is trying to solve a Rayleigh-Benard convection type of problem. That always takes a while to start.
To force starting the convective motion one could initialize the field with a smal amount of velocity in the assumed direction of motion.

Anyways, more info is needed.

Greetz
Jim

May 25, 2013, 16:16	Simulation the free convection in fluent	#1
ashkan_1984 New Member ashkan Join Date: May 2013 Posts: 12 Rep Power: 12	Dear all, I want to simulate natural free convection in fluent but I cant.I do this thing up to now and I cant reach to answer: 1-In operation condition I give -9.81 in the y part. 2-I give the density of water(my channel is full of water) in businesqe please help me

May 26, 2013, 06:16	free	#2
ashkan_1984 New Member ashkan Join Date: May 2013 Posts: 12 Rep Power: 12	I cant understand any thing macfly likes this.

May 27, 2013, 02:38	free	#4
ashkan_1984 New Member ashkan Join Date: May 2013 Posts: 12 Rep Power: 12	Unsteady 3D simulation

May 28, 2013, 07:50	free	#6
ashkan_1984 New Member ashkan Join Date: May 2013 Posts: 12 Rep Power: 12	I don't have problem about memory only I can't simulate natural convection.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Chemical Equilibrium Simulation using Fluent	enigma	ANSYS	2	October 27, 2011 08:02
Free surface simulation	Moti	ANSYS	0	August 13, 2009 13:41
Fluent simulation over airships	Aragon	FLUENT	8	March 30, 2009 12:12
Fluent simulation over airships	Aragon	FLUENT	0	March 27, 2009 12:38
Can Fluent solve free interface problem?	xueying	FLUENT	3	July 18, 2002 15:29

May 26, 2013, 15:50		#3
alireza9359300 New Member alireza Join Date: May 2013 Location: shiraz Posts: 11 Rep Power: 12	hi. its steady or unsteady simulation.

May 27, 2013, 11:14		#5
alireza9359300 New Member alireza Join Date: May 2013 Location: shiraz Posts: 11 Rep Power: 12	3d+natural+unsteady simulation is out of the power of normal computer. change your problem into 2d if you use temperature bc increase its value to see the result sooner reduce the mesh in your problem to use less memory of your computer your time must be large to see anything.

May 28, 2013, 16:21		#7
alireza9359300 New Member alireza Join Date: May 2013 Location: shiraz Posts: 11 Rep Power: 12	what about time of your simulation? by small time you cant see any changes in domain of natural sim. so you must use large time: by: little time step + large number of time steps + large iteration number in each time step + 3d + natural + small mesh + high order of solving = be sure you will have memory problem of using normal computers. if you don't have memory problem just you must insert large time for natural simulation to see changes in domain another way is to increase the value of the bc ( if you are using temperature bc ) to have more influence on your fluid. if I were you first I simulate it steady after it try to solve it unsteady , cause of for being sure it is true. if you like . can you send me case & data file of your problem.(little size . before solving ) alireza9359300@yahoo.com

May 30, 2013, 03:48		#8
JimKnopf Member Jim Knopf Join Date: Dec 2010 Posts: 60 Rep Power: 15	And don't under relax the enegry equation

May 31, 2013, 17:33		#9
alireza9359300 New Member alireza Join Date: May 2013 Location: shiraz Posts: 11 Rep Power: 12	I think your problem is density (change it to compressible fluid not boussings) I sent it for you.

June 1, 2013, 03:52		#10
JimKnopf Member Jim Knopf Join Date: Dec 2010 Posts: 60 Rep Power: 15	He already said he's using Boussinesque ansatz, that should work. As he is writing about a channel, it sounds to me as if he is trying to solve a Rayleigh-Benard convection type of problem. That always takes a while to start. To force starting the convective motion one could initialize the field with a smal amount of velocity in the assumed direction of motion. Anyways, more info is needed. Greetz Jim