Switching between open and closed boundary conditions in a transient solution

jtipton2 · October 10, 2013, 15:33

Hello all,

I am looking at modeling an air intake for an automotive application. From what I've come to understand, it is important to use acoustic resonance to enhance the pressure of the flow when the inlet opens.

I would like to use CFX to model this as a transient problem. The domain would be the manifold and intake header up to the inlet valve.

Would it be possible to oscillate between a wall and pressure outlet conditions at the inlet depending on time? How could I achieve this? I've seen some posts by Glenn H. about doing this at an interface, but this is a domain boundary.

Any guidance would be much appreciated.

Best Regards,

Joseph Tipton

ghorrocks · October 10, 2013, 17:31

A better way of modelling the inlet valve is to define it as a mass flow condition and give it the mass flow versus time curve. You generally know the mass flow better than the pressure anyway, as you have to get the total mass flow correct (to get the correct volumetric efficiency) and that is difficult with a pressure boundary. When the valve is shut this mass flow is obviously zero. There is no need to define a wall when the flow is zero, just defining zero mass flow is good enough.

This approach works fine and I have used it many times to model inlet and exhaust manifolds. You cannot model the coupling of the volumetric efficiency with the inlet manifold wave with this approach, that is quite a bit harder to model as you need a full engine model.

October 10, 2013, 15:33	Switching between open and closed boundary conditions in a transient solution	#1
jtipton2 New Member Joseph Tipton Join Date: Jun 2010 Posts: 27 Rep Power: 15	Hello all, I am looking at modeling an air intake for an automotive application. From what I've come to understand, it is important to use acoustic resonance to enhance the pressure of the flow when the inlet opens. I would like to use CFX to model this as a transient problem. The domain would be the manifold and intake header up to the inlet valve. Would it be possible to oscillate between a wall and pressure outlet conditions at the inlet depending on time? How could I achieve this? I've seen some posts by Glenn H. about doing this at an interface, but this is a domain boundary. Any guidance would be much appreciated. Best Regards, Joseph Tipton

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Boundary conditions for water jet into air in a closed chamber	vishal_s	OpenFOAM Running, Solving & CFD	0	August 20, 2013 03:27
Help UDF to Boundary Conditions in Closed Loop Pipe	Blader	Fluent UDF and Scheme Programming	2	May 14, 2012 09:48
Problems with boundary conditions - velocity and pressure not zero	M_Flodin	FLUENT	0	April 1, 2011 09:32
Change boundary conditions during transient simulation	ZenCef	FLUENT	0	February 6, 2011 16:10
Boundary conditions in CFX-Pre	Saikat	CFX	2	February 10, 2005 07:22

October 10, 2013, 17:31		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,701 Rep Power: 143	A better way of modelling the inlet valve is to define it as a mass flow condition and give it the mass flow versus time curve. You generally know the mass flow better than the pressure anyway, as you have to get the total mass flow correct (to get the correct volumetric efficiency) and that is difficult with a pressure boundary. When the valve is shut this mass flow is obviously zero. There is no need to define a wall when the flow is zero, just defining zero mass flow is good enough. This approach works fine and I have used it many times to model inlet and exhaust manifolds. You cannot model the coupling of the volumetric efficiency with the inlet manifold wave with this approach, that is quite a bit harder to model as you need a full engine model.