CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Help with CFX-Solver Manager (http://www.cfd-online.com/Forums/cfx/72870-help-cfx-solver-manager.html)

DavidA February 19, 2010 23:26

Help with CFX-Solver Manager
 
Hey People,

I am a student trying to work on my Disso using CFX. I have a problem. Every time I have tried to run the solution in CFX Solver Manager, a windows message shows up saying

"solver-hpmpi.exe has encountered a problem and needs to close."

So I send the report, and CFX Solver Manager says,

" An error has occurred in cfx5solve:
The ANSYS CFX solver could not be started, or exited with return
code 255: . No results file has been created. "

This has happened everytime I have tried to run this simulation and its always on outer loop iteration = 2.

Please could you help me since time is getting really short and I need results.

The simulation I am trying to run is a rotor stator arrangement which is very similar to the one in the ansys tutorials.

HELP!!!

sans February 20, 2010 01:35

Hi,
Have you been able to run the tutorial problem?
I have got this error in the past for a variety of reasosn, could be mesh, setup etc. Try to debug one at a time. I would start with a remesh.

ghorrocks February 20, 2010 06:22

If it occurs on outer loop iteration 2 then it is a convergence problem. You need to improve the stability of your simulation - are your physics correct? Is the mesh of high enough quality? Is your time step small enough?

DavidA February 20, 2010 16:35

Thank you.

I tried to change a few of the details for this run, but now, I won't even start up.

As soon as I try to start the run, the same message is displayed and this time, it does no iterations....

Help, this is getting worse

zandi February 20, 2010 16:59

did you get help from the tutorial you've mentioned. as the friends say more than one exact reason can have role here. so because of time the tutorial is a good start. chang the time scale gradually and little by little. test all the ways one by one to find the best model, don't take all of them in one step
good luck

feizaghaee February 21, 2010 04:32

are you running a solver with parallel mode?

DavidA February 21, 2010 09:26

No, I am using Serial mode

feizaghaee February 21, 2010 17:04

check the effective factors as glenn said.

ghorrocks February 21, 2010 18:39

Quote:

I tried to change a few of the details for this run, but now, I won't even start up.
Then you have made it even less numerically stable. You are heading in the wrong direction. Whatever you changed, try going the other way.

I repeat my previous post:
Quote:

are your physics correct? Is the mesh of high enough quality? Is your time step small enough?

DavidA February 23, 2010 18:27

Thank you
 
People, I don't believe you can imagine how grateful I am for you help.

I have my first set of results, and I'm on my way to getting more.

It was the time steps.

made them 10-4 and it worked

Thank you again


All times are GMT -4. The time now is 14:30.