2 way or 1 way coupled

Pled · March 5, 2018, 08:40

I have a problem that needs solution. I want to simulate a microfluidic pressure driven fluid flow in a square section microchannel in STEADY STATE and watch the tube's deformation. Do I need to make a 1 or 2 way coupling between the cfx and structural?

Gert-Jan · March 5, 2018, 13:49

Strange question in steady state.
If the deformation is large, do it 2 way.
If small you might get away with 1 way. But if it is small, then way do you want to study it anyway?

ghorrocks · March 5, 2018, 16:44

The choice between 1 way or 2 way depends on how the fluid flow is coupled to the motion. If the motion is small enough that the fluid flow is not going to change significantly then 1 way is OK. If the motion is large enough that the fluid flow will change because of the motion then you need 2 way coupling.

Gert-Jan · March 5, 2018, 17:40

@Glenn, You mention motion. If I assume you really mean the motion of the wall, then the simulation is not steady state, but transient.
If I assume you don't mean motion but the final deformation of the solid, then it is possible to apply a steady state calcultion.

@Pled: If deformation is significant, the final shape of the solid at the end of a (in fact transient) deformation process, can be determined in a 2-way steady state calculation. Then the deformation is calculated in structural and the new shape is transferred back to CFX which will determine the effect of the deformation on the flow and vice versa several times.

In a 1-way calculation the steady state pressure distribution on the wall is calculated in CFX and is tranferred to the structural part to determine the deformation. The deformation is not transferred back to CFX. This is only valid if the deformation is small.

Bottomline, the choice for 2 or 1 way depends on if the deformation is significant or not.

ghorrocks · March 5, 2018, 17:45

Thanks for the clarification - I should have said deformation, not motion.

March 5, 2018, 08:40	2 way or 1 way coupled	#1
Pled Member Dimitrios S. P. Join Date: Jan 2018 Posts: 63 Rep Power: 8	I have a problem that needs solution. I want to simulate a microfluidic pressure driven fluid flow in a square section microchannel in STEADY STATE and watch the tube's deformation. Do I need to make a 1 or 2 way coupling between the cfx and structural?

March 5, 2018, 13:49		#2
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,854 Rep Power: 27	Strange question in steady state. If the deformation is large, do it 2 way. If small you might get away with 1 way. But if it is small, then way do you want to study it anyway? Last edited by Gert-Jan; March 5, 2018 at 16:48.

March 5, 2018, 16:44		#3
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,781 Rep Power: 143	The choice between 1 way or 2 way depends on how the fluid flow is coupled to the motion. If the motion is small enough that the fluid flow is not going to change significantly then 1 way is OK. If the motion is large enough that the fluid flow will change because of the motion then you need 2 way coupling. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

March 5, 2018, 17:45		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,781 Rep Power: 143	Thanks for the clarification - I should have said deformation, not motion. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ANSYS Meshing] Help with element size	sandri_92	ANSYS Meshing & Geometry	14	November 14, 2018 07:54
How to give Two-way coupled and one-Way coupled at a time in fluent?	krishna.n	FLUENT	0	June 27, 2016 07:09
Divergence detected in AMG solver: pressure coupled	n.jithin.1991	FLUENT	6	January 5, 2014 11:43
Some confusion about coupled solver for incompressible flow	bearcat	Main CFD Forum	0	February 14, 2010 20:40
Coupled 1D/3D STAR-CD Training	CD adapco Group Marketing	Siemens	1	November 13, 2002 15:48

March 5, 2018, 17:40		#4
Gert-Jan Senior Member Gert-Jan Join Date: Oct 2012 Location: Europe Posts: 1,854 Rep Power: 27	@Glenn, You mention motion. If I assume you really mean the motion of the wall, then the simulation is not steady state, but transient. If I assume you don't mean motion but the final deformation of the solid, then it is possible to apply a steady state calcultion. @Pled: If deformation is significant, the final shape of the solid at the end of a (in fact transient) deformation process, can be determined in a 2-way steady state calculation. Then the deformation is calculated in structural and the new shape is transferred back to CFX which will determine the effect of the deformation on the flow and vice versa several times. In a 1-way calculation the steady state pressure distribution on the wall is calculated in CFX and is tranferred to the structural part to determine the deformation. The deformation is not transferred back to CFX. This is only valid if the deformation is small. Bottomline, the choice for 2 or 1 way depends on if the deformation is significant or not.