Boundary Limit : Velocity Ramp Vs Smoke

[ChrlBzt]

Hi Everyone,

I'm starting to work with FDS. It's very interesting. I have an issue that I do not have solved for the moment. I'm simulating a tunnel with a specific velocity ramp at one head. My problem is that the smoke is going through both head. As I am applying this velocity boundary limit at one head, smoke is not going through like it was on "open" defined limit. The smoke is like facing a wall. Do you have any advices for me ? I can not used the "DYNAMIC PRESSURE" function because the obtained velocity inside the tunnel is too different from the one I want to simulate with my velocity ramp. I cannot apply the boundary condition to the opposite end either, because the smoke also reaches that end.

I was thinking about creating a kind of box / reservoir at the extremity at which I am applying the velocity. Like this the smoke can go into this box and applying my boundary limit on the face of this box in front of the entrance of my tube. I have attached a picture and the code.txt with / without this box.

Thank you for your help,
Charles

Tun-Box-indA.txt
Tun-indA.txt

[mcgratta]

When you specify a velocity at a boundary, you are essentially creating a solid boundary that blows air in (or sucks air out). Is that what you want?

[ChrlBzt]

Thank you for your help. The problem with this modelization is that the smoke can not got through this boundary. In some cases, it could work because the smoke is directed only in one direction and so I can choose on which side to apply my boundary condition.
Here, it's not what I want because I need the smoke to exit my tunnel through both extremities. In the same time, I still need to represent this velocity at the boundary.
This velocity condition is calculated thanks to the wind and meteorological data around the tunnel.
I run a new simulation using the function "WIND SPEED" but I am not absolutely sure of how it works. I compared the velocity in different sections of my tunnel for the two different cases : using the "wind speed" condition and using a velocity ramp at one boundary.

My concern with this "wind speed" condition is to know exactly how it works : is the wind speed applied as a boundary conditions or in all the domain ? I want to be sure that my fire modelisation will still be valid and physically coherent. For example, I want to be sure that the thermal draft effect is well represented and
not “crushed” by this imposed speed.
I have attached the new code. Thank you :)

Charles

RUN_Sc1.txt

[mcgratta]

If you want to allow smoke to naturally vent from each end of the tunnel, then you cannot expect to be able to set the velocity at the ends exactly. I suggest you experiment with different values of DYNAMIC_PRESSURE for the tunnel with no fire until you achieve a desired flow velocity. In the meantime, I will run your case to see if there is anything else that could be done.

[ChrlBzt]

Thank you for your help. I'm quite satisfied with the wind speed function. For the moment, the velocity inside the tunnel is closer of the physical reality rather than using the function dynamic pressure. The problem with the dynamic pressure is, I think, that the friction and pressure losses are not well modelized.
Thank you very much for your reactivity.

[drjfloyd]

If you want to get a specific velocity in the tunnel with no fire and dynamic pressure, then you need to tune the pressure to account for the friction loss in the tunnel.