CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   On the Courant number . (http://www.cfd-online.com/Forums/fluent/31396-courant-number.html)

raintung May 4, 2003 22:10

On the Courant number .
 
Hello everyone, When I use fluent to simulate 3D flow field, I set the Courant number 0.5 at the beginning of the iterating. Later, once I increase the Courant number, the following information appears: divergence detected - temporarily reducing Courant number to 0.2 and trying again... What should I do? If I doní»t increase the Courant number, convergence become slow, and I doní»t know whether this will affect the computation results. Thank you!

Holidays May 5, 2003 04:25

Re: On the Courant number .
 
That the simulation slows down is pretty normal as you are relaxing the solution more, not allowing it to changeso much at each iteration. Cr number criterion are usually good to get the simulation going; later on you may want to change to a less stringent condition.

raintung May 5, 2003 08:11

Re: On the Courant number .
 
Thank you! But what does the stringent condition mean? I know I set the low succssive ratio(spanwise), because I want to obtain the streamline distribution on the endwall(there is a thin fence on it). How can I modify this? Thanks.

Holidays May 7, 2003 06:03

Re: On the Courant number .
 
What I meant is that later on, when you start converging, you should change for a "physical timescale" in place of the Cr number approach so that the solution can continue going, but at a faster pace.


All times are GMT -4. The time now is 09:43.