CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   De Laval nozzle (http://www.cfd-online.com/Forums/main/83440-de-laval-nozzle.html)

fayaazhussain December 29, 2010 01:21

De Laval nozzle
 
Hey Guys I'm currently trying to simulate flow through a de laval nozzle to show under expansion, over expansion and barrel shocks etc. I used GAMBIT for meshing..I am totally new to CFD as I'm currently a undergrad student so forgive my ignorance. I'm able to get the desired results for a mesh with interval size 0.5 with inviscid and laminar flows that too with third order discretization. Results diverge for second order discretization. When I try for a mesh with 0.1 interval the results diverge for all orders of discretization.

Can anyone please help??:(

ignat December 29, 2010 03:27

Quote:

Originally Posted by fayaazhussain (Post 288752)
I'm able to get the desired results for a mesh with interval size 0.5 ... When I try for a mesh with 0.1 interval the results diverge for all orders of discretization.

It seems your numerical algorithm is bad...
But may be the reason is other. If you use explicit time scheme then you should keep stability Courant condition
\Delta t < \frac{\Delta x}{|u|+c}

So when you have reduced dx (keeping dt=const) this condition could be broken.


All times are GMT -4. The time now is 15:06.