N2 discharge simulation

[HunterXHunter1997]

Hello team,
I want to check if FDS can simulate a N2 gas discharge to see the gas distribution within a room and it's concentration. I thought about eject particles of N2 through the space using a nozzle, but you know that the K-factor is high in gas nozzles (around 726 in my case). and the pressure is 3 bar
So when I launched my test, a numerical instability message showed up at the beginning. Although I used a fine mesh
Also, I thought about using a Vent to flux the species will not be realistic because I will ignore the discharging angels.

Please let me know if there is a better method to that.
Thanks for support!

[rmcdermo]

FDS cannot model choked flow near the nozzle exit. But it can do a good job of modeling the transport and dispersion once the flow decelerates, which happens very close to the nozzle. Your first step needs to be finding a near field solution to your jet mixing problem. There are similarity solutions for this, it is a canonical fluid mechanics problem. (Try your favorite fluid mechanics book or search the internet for similarity solutions for compressible jets).

Once you have a profile of velocity and concentration at a cross-section of the jet (axially downstream) you can use this as a VENT in FDS.

[drjfloyd]

For @rmcdermo's solution this should be done as two VENTs. One that injects the subsonic N2+Air mixture, and one that removes the Air portion, so that the net mass is just the N2.

[rmcdermo]

That's a good point, I had not considered. Another thought might be to use a velocity patch to specify the momentum profile and then inject the N2 in an INIT region. That way you'd have control over the velocity and N2 mass without adding extra Air.

[HunterXHunter1997]

Thanks for answer
So you mean I can define the ejected n2 species from one vent with angles, right?