Need help for boundary condition

lehoanganh07 · June 3, 2014, 01:36

Hi everyone,

I am doing my numerical simulation of cavitating flow in rectangular Nozzle. I am having trouble with boundary condition.

Firstly, I use mixed Neumman and Drichlet boundary condition at inlet and outlet section. Inlet is subsonic flow so that I fixed total pressure, flow direction and void fraction. Outlet also assume that subsonic flow, so just static pressure fixed. But the solution divergent due to very strong pressure wave propagation.

I read in book that, for this situation, I need use non-reflective BC or Convective BC.

example, convective BC at oulet: d(phi)/dt + U d(phi)/dx = 0 with phi: velocity, pressure, density, mass fraction,...

But my question, if use CBC, how to fix the outlet pressure?
And which boundary is suitable for my case?

harishg · June 5, 2014, 13:45

Check this paper

Poinsot T. et Lele S. (1992). Boundary conditions for direct simulations of compressible reacting flows. Journal of Computational Physics. 101, 1, 104-129.

CBC will perform a decent job in absorbing the waves but may not be sufficient depending on the problem. An easy option is to have a sponger layer/PML to absorb the outgoing waves.

Depending on your numerical scheme(if it is high-order), you can also use the energy transfer and annihilation method of Visbal.

June 3, 2014, 01:36	Need help for boundary condition	#1
lehoanganh07 Member le hoang anh Join Date: Oct 2012 Posts: 96 Rep Power: 13	Hi everyone, I am doing my numerical simulation of cavitating flow in rectangular Nozzle. I am having trouble with boundary condition. Firstly, I use mixed Neumman and Drichlet boundary condition at inlet and outlet section. Inlet is subsonic flow so that I fixed total pressure, flow direction and void fraction. Outlet also assume that subsonic flow, so just static pressure fixed. But the solution divergent due to very strong pressure wave propagation. I read in book that, for this situation, I need use non-reflective BC or Convective BC. example, convective BC at oulet: d(phi)/dt + U d(phi)/dx = 0 with phi: velocity, pressure, density, mass fraction,... But my question, if use CBC, how to fix the outlet pressure? And which boundary is suitable for my case?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
UDF for degassing boundary condition	peaker007	Fluent UDF and Scheme Programming	5	November 23, 2015 12:55
pressure oulet condition	nam su	FLUENT	2	January 24, 2012 04:48
İnlet boundary condition	aggie	FLUENT	10	December 23, 2011 04:28
How to resolve boundary condition problem?	sam	FLUENT	2	July 20, 2003 02:19
The Boundary Condition about the Flat Plate	boing	Main CFD Forum	1	January 6, 2002 16:53

June 5, 2014, 13:45		#2
harishg Senior Member N/A Join Date: Mar 2009 Posts: 189 Rep Power: 17	Check this paper Poinsot T. et Lele S. (1992). Boundary conditions for direct simulations of compressible reacting flows. Journal of Computational Physics. 101, 1, 104-129. CBC will perform a decent job in absorbing the waves but may not be sufficient depending on the problem. An easy option is to have a sponger layer/PML to absorb the outgoing waves. Depending on your numerical scheme(if it is high-order), you can also use the energy transfer and annihilation method of Visbal.