Convergence issue in CHT simulation

funDevil · May 8, 2015, 14:33

Hi All,

For my first CFD simulation, I was required a simulate a CHT system for NATURAL CONVECTION. The system has 2 components - plastic tube (almost U-shaped)and a heat sink. The heat sink snugly fits into the open face (along the length), such that the both together form a hollow tube with fins of the heat-sink in the air passage. The was to study the temperature distribution over the system at steady-stage.

When I start the simulation, for some reason all the residuals don't seem to decrease (see attachment). I have attached the images of geometry, mesh, and the residuals plot. I am not sure where I am going wrong. I need help!

mesh.jpg

residuals.PNG

setup.jpg

geometry.jpg

Deepak

alg · May 8, 2015, 21:56

I might not be the best person to give advice on how to converge a CHT problem (see my other post), but you could try refining your mesh in the fluid domain (and adding more inflation layers). I'm assuming that your opening boundary conditions are asserting a pressure drop along the channel (and hence a flow)?

funDevil · May 8, 2015, 22:47

Alg,

Its a natural convection problem. The ends of the tube are open to atmospheric pressure.

Like you suggested, I increased the number of inflation layers (to 5), and it did not help. Thanks for the inputs.

alg · May 8, 2015, 23:11

Ok, no worries.

How are your opening boundary conditions set then? and which way does your gravity vector point? I've run a similar simulation recently but modelled a fuild domain 'around' the channel, linking the inlet and outlet, with openings on the outer surfaces of this larger domain. This avoided the situation where you have a pressure set at each end of the channel, which will dictate the mean flow. There may be an easier way to do this though...

ghorrocks · May 9, 2015, 07:48

Before we dive into details on this simulation - have you read the FAQ on this issue? http://www.cfd-online.com/Wiki/Ansys...gence_criteria

May 8, 2015, 14:33	Convergence issue in CHT simulation	#1
funDevil New Member Deepak Join Date: May 2015 Location: Arizona Posts: 2 Rep Power: 0	Hi All, For my first CFD simulation, I was required a simulate a CHT system for NATURAL CONVECTION. The system has 2 components - plastic tube (almost U-shaped)and a heat sink. The heat sink snugly fits into the open face (along the length), such that the both together form a hollow tube with fins of the heat-sink in the air passage. The was to study the temperature distribution over the system at steady-stage. When I start the simulation, for some reason all the residuals don't seem to decrease (see attachment). I have attached the images of geometry, mesh, and the residuals plot. I am not sure where I am going wrong. I need help! mesh.jpg residuals.PNG setup.jpg geometry.jpg Deepak

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Wind Tunnel Flow Simulation Mass Flow BC Issue	ledelman	SU2	4	August 3, 2014 19:38
Convergence issue on Solar chimney simulation	Mani.juve	FLUENT	0	November 19, 2012 18:00
DPM simulation convergence problem	gemini	FLUENT	2	May 22, 2012 02:41
[CFX] Simulation of Flow Separation in a rectangular diffuser - Convergence Problem	anon_b	CFX	11	May 6, 2012 21:16
Convergence problem with axial pump simulation!	cnchilavert	CFX	2	October 5, 2011 05:23

May 8, 2015, 21:56		#2
alg New Member Alan Join Date: Apr 2015 Posts: 13 Rep Power: 11	I might not be the best person to give advice on how to converge a CHT problem (see my other post), but you could try refining your mesh in the fluid domain (and adding more inflation layers). I'm assuming that your opening boundary conditions are asserting a pressure drop along the channel (and hence a flow)?

May 8, 2015, 22:47		#3
funDevil New Member Deepak Join Date: May 2015 Location: Arizona Posts: 2 Rep Power: 0	Alg, Its a natural convection problem. The ends of the tube are open to atmospheric pressure. Like you suggested, I increased the number of inflation layers (to 5), and it did not help. Thanks for the inputs.

May 8, 2015, 23:11		#4
alg New Member Alan Join Date: Apr 2015 Posts: 13 Rep Power: 11	Ok, no worries. How are your opening boundary conditions set then? and which way does your gravity vector point? I've run a similar simulation recently but modelled a fuild domain 'around' the channel, linking the inlet and outlet, with openings on the outer surfaces of this larger domain. This avoided the situation where you have a pressure set at each end of the channel, which will dictate the mean flow. There may be an easier way to do this though...

May 9, 2015, 07:48		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Before we dive into details on this simulation - have you read the FAQ on this issue? http://www.cfd-online.com/Wiki/Ansys...gence_criteria