gas dispersion and ignition

[Andrew]

I am in the process of modelling natural gas being expelled into the
atmosphere and then ignited however i have a few questions:

Has anyone used FDS for this before and did it give reliable outputs?

How do you specify a wind consisting of air blowing into the domain to
simulate a wind and have natural gas being expelled from a simulated
leak? When i attempted it the wind either consisted of natural gas or
when i specified it as air it gave a numerical instability error after
1.5s

How do you ignite the gas? I have isosurfaces dictating the
flammability limits however when the gas ignites it does not engulf
the rest of the gas but burns in an isolated area without spreading.

Any advice would be much appreciated.

[Kevin]

I have done this, but I have to warn you that it is a difficult
calculation to set up, and we have no validation data other than
Heskestad's flame height correlations. I suggest you first just create
the wind blowing through your domain. Make sure everything makes
sense. Then create a small VENT at the ground that ejects the gas. Set
HRRPUA so that HRRPUA*(VENT area) = your desired HRR. This will be the
default mixture fraction model. There is no ignition. The fuel and
oxygen will burn on contact.

[Andrew]

Thanks for the information, ive now got a sub model running.

However the gas is not burning very well unless the vent with the
HRRPUA is accross the whole floor of the domain and the gas is
expelled at the floor.

Therefore how do i define a vent/obst that will set fire to the gas
realisticly? The gas is being expelled from an vent attached to an
obstacle in the middle of a domain to represent a structure.

If i put the obst with a HRRPUA in the same place as the vent emitting
the gas nothing happens.
If i extend it so that it is a larger area the gas ignites but does
disperse properly as the obst is obstructing it.
Is there a way to define HRRPUA on the same line as the vent which is
expelling the gas? (instead of having two lines an obst and a vent,
one with the gas and another igniting it) I have tried but cant get it
to work.

Thanks for your help

On Jul 3, 5:45 pm, Kevin <mcgra...@gmail.com> wrote:
> I have done this, but I have to warn you that it is a difficult
> calculation to set up, and we have no validation data other than
> Heskestad's flame height correlations. I suggest you first just create
> the wind blowing through your domain. Make sure everything makes

> sense. Then create a small VENT at the ground that ejects thegas. Set

> HRRPUA so that HRRPUA*(VENT area) = your desired HRR. This will be the
> default mixture fraction model. There is no ignition. The fuel and
> oxygen will burn on contact.
>
> On Jul 3, 12:28 pm, Andrew <ajbasf...@googlemail.com> wrote:
>
>
>

> > I am in the process of modelling naturalgasbeing expelled into the

> > atmosphere and then ignited however i have a few questions:
>
> > Has anyone used FDS for this before and did it give reliable outputs?
>
> > How do you specify a wind consisting of air blowing into the domain to

> > simulate a wind and have naturalgasbeing expelled from a simulated

> > leak?  When i attempted it the wind either consisted of naturalgasor
> > when i specified it as air it gave a numerical instability error after
> > 1.5s
>

> > How do you ignite thegas? I have isosurfaces dictating the
> > flammability limits however when thegasignites it does not engulf
> > the rest of thegasbut burns in an isolated area without spreading.
>
> > Any advice would be much appreciated.- Hide quoted text -
>
> - Show quoted text -

[Kevin]

Please read the FDS User's Guide. From your description of the
problem, it seems to me that you do not understand the different modes
of operation of FDS in regard to combustion. When you specify HRRPUA
anywhere in the domain, FDS automatically assumes the mixture fraction
combustion approach. In this mode, additional gases will not ignite. I
cannot explain all the details here. Tech Guide and User's Guide.

> > - Show quoted text -- Hide quoted text -

[Andrew]

Ive fixed the problem, it was not igniting on entering the domain as
the MASS_FRACTION(1) was set wrongly causing the gas not to ignite. I
was using this to ensure that the wind consisted of air. I now have
two working models set up, one showing gas dispersion and the other
showing gas combustion both with a wind blowing into the domain.

Thanks for your assistance