<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=ISO-8859-1">
</head>
<body bgcolor="#ffffff" text="#000000">
Hi all,<br>
<br>
I am wondering if there is any reference on the pressure boundary
condition for Pn-Pn case? I know Dr. Fisher has a 1997 JCP paper
discussing the Pn-Pn-2 setting. <br>
<br>
Also what subroutines/functions handle with the boundary conditions
for both velocity and pressure? Thank you,<br>
<br>
Zhu Wang<br>
Virginia Tech<br>
<br>
<br>
----------------------------<br>
<small>Dear Fred,<br>
<br>
The pressure bcs are chosen to ensure div.U=0.<br>
<br>
For the Pn-Pn case, this involves addition of an appropriate
inhomogeneity<br>
that cancels the b.l. divergence that arises from std. splitting <br>
formulations.<br>
<br>
For the Pn-Pn-2 case, there are no pressre bcs, only velocity bcs.<br>
<br>
Paul<br>
<br>
----------------------------------<br>
On Thu, 24 Jun 2010, nek5000-users at lists.mcs.anl.gov wrote:<br>
<br>
> Dear NEKs,<br>
><br>
> I wanted to ask which boundary condition is used for the
pressure in nek?? <br>
> are all pressure gradients normal to wall set tto zero??
because in my <br>
> current simulation i have a pressure gradient normal to a
wall and i am <br>
> wondering why!?<br>
><br>
> best, fred<br>
</small>---------------------------<br>
</body>
</html>