JakobCHR.com
 
Quick Navigation:
 
Personal:
 Go to Home
 MS Research
 PhD Research
 Curriculum Vitae

General:
 Linux

Soon to come:
 Matlab
 On-line Stores
 Cycling
 Medicine & Health
 LaTeX
 OOP & C++
 Sony PCM-R500 DAT


next up previous contents index
Next: Core flow tests Up: Grid generation Previous: Grid points not influenced   Contents   Index

Grid points influenced by a spacer

The second requirement treats grid points which are in the vicinity of a spacer, ie a grid point, zK, which satisfies

\begin{displaymath}
(z_K-z_{\mbox{\protect\scriptsize sp}} < \Delta z_{\mbox{\p...
...criptsize sp}}) \wedge (z_K>z_{\mbox{\protect\scriptsize sp}})
\end{displaymath} (8.22)

where $z_{\mbox{\protect\scriptsize sp}}$ is the position of the spacer closest to point i and $\Delta
z_{\mbox{\protect\scriptsize sp}}$ is the width of the spacer pressure variation (see section 6.7.2.3).

If (8.22) is fulfilled we proceed as follows

.
The first grid point which is influenced by the spacer, zS, is is placed on the location of the spacer, ie $z_S = z_{\mbox{\protect\scriptsize sp}}$.
.
The next N grid points are distributed uniformly over the width, $\Delta z_{\mbox{\protect\scriptsize sp}}$, ie

\begin{displaymath}z_{i+S} = \frac{\Delta z_{\mbox{\protect\scriptsize sp}}}{N} ...
...\scriptsize sp}} \qquad
{\mbox{for}} \quad i=\{1,2,\ldots,N\} \end{displaymath}

where N and $\Delta z_{\mbox{\protect\scriptsize sp}}$ are user specified quantities.

In summary the grid is uniquely defined when the user has specified the following quantities

>
average steplength in the computational grid [m].
>
number of steplengths which follow a spacer [--].
>
width of a spacer pressure variation [m].
Note that in order to keep the truncation error associated with the pressure, p(z), reasonably low is advisable to choose $N \ge 5$.

A typical computational grid is depicted in Figure 8.8.

\begin{figure}
% latex2html id marker 31162\rule{\textwidth}{0.2mm}
\rule{0cm}...
... $h_{\mbox{\protect\scriptsize av}} = 0.1{\mbox{ m}}$\ and $N = 5$.}\end{figure}

One could question the grid generation procedure since it only considers the variation of q' and not, for instance, the fact that the frictional pressure drop increases along the channel due to increased flow quality. The reason for not considering the pressure variation is that in practice the curvature of the pressure as a function of z is small when we compare to curves of $\mbox{$<\!{\alpha}\!>$}$ and $\mbox{$<\!{x}\!>$}$. This means that it is absolutely no problem to control the truncation error in regard to the pressure.


next up previous contents index
Next: Core flow tests Up: Grid generation Previous: Grid points not influenced   Contents   Index  
 
 
 
Revision 2.0, Copyright © 1999-2004 Jakob Christensen
http://www.JakobCHR.com
E-Mail: webmaster@JakobCHR.com
Top Quality
Developed with

Danish
Brain Power
Linux Powered!