CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   please contribute with ideas: fluent and false/numerical diffusion (https://www.cfd-online.com/Forums/fluent/71937-please-contribute-ideas-fluent-false-numerical-diffusion.html)

mario91 January 20, 2010 08:54

please contribute with ideas: fluent and false/numerical diffusion
 
false/numerical diffusion is a quite classical problem in CFD.

However I think that each simulation case should have more important corrections to handle that.I am having extreme numerical diffusion!

Briefly my case is this:

Fully developed flow along x direction. Free Atmospheric flow between ground and 150m high plane. Ground is flat so vy and vz are zero. I use orthogonal hexaedronds with r=1.07.
I am doing dispersion with species and varying schmidt number. Sct=0.7 gives high numerical diffusion, which increases even more with lower schmidt numbers (e.g. 0.4). vx velocity=(0,43m.s-1/0.4)*ln[(z+z0)/z0]. Z0=0.1m.

I am doing a point source emission and there is concentration of the emitted species even at upwind of its release!

Thanks for your ideas on this.

Bernhard January 20, 2010 10:38

Where (which height) is your point source located?

mario91 January 20, 2010 11:42

i am doing different scenarios

point source (~1m3) at 10 m high or 30 m high.

Is that relevant for Numerical diffusion?
Numerical diffusion is higher near ground (e.g 1 meter high) because wind speed is slower?!


Thanks


All times are GMT -4. The time now is 23:39.