|
[Sponsors] |
February 10, 2003, 04:47 |
Different answers
|
#1 |
Guest
Posts: n/a
|
Hi, As I wanted to solve a jet flow, by different value initializing I get different answers, do you know why?
|
|
February 12, 2003, 02:10 |
Re: Different answers
|
#2 |
Guest
Posts: n/a
|
Hi,
This is my guess, the calculation is not convergence yet. If you judge only the residuals, the residuals are not good to judge whether a calculation is convergence or not. So you should monitor some values at some points and when the values become constant value during iterations,then we can say the calculation is convergence. Regards, Keil |
|
February 15, 2003, 02:31 |
Re: Different answers
|
#3 |
Guest
Posts: n/a
|
Thanks for your reply KEIL, you mentioned the monitoring, What kind of value (x-vel, y-vel, z-vel, pressure, k, e, ...) do you mean? In an unsteady solution, how can this be used?
|
|
February 20, 2003, 08:10 |
Re: Different answers
|
#4 |
Guest
Posts: n/a
|
Hi,
I mean, the monitoring is a certain physical value at an arbitrary location in your calculation domain. How about the following procedure?. First, you define the location and physical value(for example, velocity magnitude or something). Second ,Define the property(monitoring at every iterations or time steps) on the panel Solve-Monitors-Surfaces. And then calculate it. I recommend every iterations monitoring because we are able to check whether the calculation is convergence or not in the time. Regards, Keil |
|
February 25, 2003, 05:04 |
Re: Different answers
|
#5 |
Guest
Posts: n/a
|
thanks for your help,
|
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Steady State Vs Transient answers | Kushagra | CFX | 25 | July 23, 2017 04:20 |
inconsistent answers | Robreid | FLUENT | 0 | May 27, 2011 17:15 |
[TGrid] Simple Question (Simple Answers?): Edge Length adjustment for tet mesh | booz | ANSYS Meshing & Geometry | 0 | August 29, 2010 13:39 |
Coodles | anja | OpenFOAM Pre-Processing | 33 | January 14, 2010 10:05 |
The answers of the case tutorialsstressFemFoamplateHole seems wrong | dandes | OpenFOAM Pre-Processing | 3 | March 24, 2006 04:31 |