[annu]

Hi friends,
My work is on 3D, but due to shortage of time i shifted to 2D. I wanna know is there is more variation in the results, and is 2D results are acceptable?
Is this possible in 3D, gas is dispersing in both sides from the release source and in 2D, it remains in one direction with same boundary conditions? am i doing wrong somewhere.

Any response is highly appreciated..

[venkateshaero]

We cant do simulate 2D for all cases. Please explain about your Geomerty and BC details

[annu]

its a tunnel of 1000 m.

[venkateshaero]

WInd Tunnel?

[annu]

transportation tunnel (road)

basically gas is released from a hole which is at centre of the tunnel means at 500 m.
b.c. for release source is mass flow inlet
outlet are pressure outlet

[FMDenaro]

The key is that 2d assumption has a fundamental constraint on the vorticity evolution, no stretching exists. This is fundamental in DNS/LES where 3D is mandatory (apart some geophysical flows...)
You can simulate 2d turbulent flows in the statistical meaning (URANS), provided that your problem is consistent to such assumption.

[annu]

FM denaro: I have done some work on 3D its is dispersing on both sides, I used standard k-epsilon model in both cases, but in 2D it is flowing in one side..its unsteady state problem.

.
And you mean that it is not possible to
simulate in 2D if i want to check concentration in longitudinal directions.??

Thankyou for replying

[FMDenaro]

Originally Posted by annu

FM denaro: I have done some work on 3D its is dispersing on both sides, I used standard k-epsilon model in both cases, but in 2D it is flowing in one side..its unsteady state problem.

.
And you mean that it is not possible to
simulate in 2D if i want to check concentration in longitudinal directions.??

Thankyou for replying

without details is not possible to give you an answer ... is your flow problem incompressible and turbulent? you have a geometry that has a 2d geometry elonged in the spanwise direction? are you sure that the statistical fields are unsteady? in any case you can switch to URANS.

what do you mean for "to check concentration"?

[marek_ac]

Hi,
It is better to do it in 3D.
Here you have equation (description is in polish)

http://feazone.org/readarticle.php?article_id=11

and sample results:
http://commons.wikimedia.org/wiki/Fi...is_a1_test.gif
http://commons.wikimedia.org/wiki/Fi...is_a2_test.gif

animation at the bottom of the page (with description):
http://pl.wikipedia.org/wiki/Modelow...odorant%C3%B3w

Regards,
Marek

[annu]

FM Denero : I am using species transport without reaction. I am checking mass fraction profiles of gas in longitudinal and transversal direction of the tunnel.

Marek: U mean 2 D results are not acceptable? I dont have much time..

Thankyou

[marek_ac]

Hi Annu,
If it is Wind Tunnel with simple geometry - like a pipe - 2D is OK
but if you use transportation tunnel - like a urban canyon - 3D would be much better.
Regards,
Marek