eddy length scale

[HEROISCOMING1]

Hi~ Happy new year everyone

I'm studying fds, I have a question about eddy.

In the fds user guide : eddies are injected into the flow at random positions on the boundary and advect with the mean flow over a short distance near the boundary equivalent to the maximum eddy length scale.

And, I want to know what the maximum eddy length scale means.
Does that mean the maximum value of diameter(eddy length scale) at which the eddy can be maintained?
Does the eddy length scale gradually increase and advent the mean flow when it reaches a maximum eddy length scale?
What is the maximum eddy length scale value set in fds?

This may be a bit of a difficult question, but I need someone to answer it, please...

[mcgratta]

Have you looked at the FDS Technical Reference Guide. That contains more of the theoretical explanation for what is in the User's Guide.

[mcgratta]

Actually, details of the synthetic eddy method are here

[12] N. Jarrin. Synthetic Inflow Boundary Conditions for the Numerical Simulation of Turbulence. PhD
thesis, The University of Manchester, Manchester M60 1QD, United Kingdom, 2008.

[mcgratta]

I am not, because I am not familiar with this feature. I suggest that you look at Jarrin's thesis to confirm the meaning of this parameter. In general, a length scale is usually not referred to a as "maximum" value. It is usually taken as "characteristic" or "on the order of" some length in the physical domain. Empirical sub-models typically have time and length scales over which the relations are appropriate. Occasionally they take on an actual length, like the height of the ceiling, width of a burner, etc. More often, they are just a value with which the basic flow equations have been non-dimensionalized or scaled.

[rmcdermo]

@HEROISCOMING1 I think there is a confusion here. Jarrin's Synthetic Eddy Method (SEM) is not applied by default. It is only invoked at a VENT if N_EDDY is set to some value > 0. From there, you can try to control the turbulent statistics, as discussed in the thesis. If you are new to FDS, this is one of the last things you need to be worrying about.

[HEROISCOMING1]

Thank you for your helpful replies.
When using the LES model, if the value of N_EDDY is 0 for the vent(the boundary where the monin-obukhov condition is applied and the wind speed and wind direction are set), is the turbulence not considered in that vent?

[rmcdermo]

You are being too cavalier with the word "turbulence". LES uses a subgrid stress to account for turbulent transport of momentum. Also, an eddy viscosity together with turbulent Schmidt and Prandtl numbers are used to account for turbulent transport of mass and heat. At a vent, the mean profile you specify is usually stationary (does not fluctuate in time) but the mean profile would not be that shape if not for turbulence. So, it is more precise to say that if N_EDDY=0 the vent uses a stationary profile. It will not inject coherent turbulent structures (spatially and temporally varying) into the domain. Hence, large-scale (larger than grid scale) hydrodynamic instability will not occur just downstream of the vent (it may occur later, depending on the flow scenario).

If I were you, I would put a slice of VISCOSITY across your vent and normal to the vent. And DEVC measuring VELOCITY where you have experimental measurements to compare. And I also recommend you spend some time working with simple verification tests until you understand the code and LES better. Outdoor flow scenarios are a research topic in fire dynamics.

[HEROISCOMING1]

Thank your for your reply
It was a great help in understanding LES