Pressure zones and external boundaries

[Cian Davis]

Hi All,
The FDS manual states a pressure zone should not have an boundary OPEN
vents.

In my case, this causes a difficulty as I would normally have an OPEN
vent to supply air to the fire. I can replace this with a fan blowing
air into the domain, but I would prefer not to dictate an inlet rate.

Is there another arrangement within blowing in that anyone can suggest?
Or is the suggestion in the manual regarding the OPEN vents just there
to avoid numerical instabilities?

Another question with regard to pressure zones. One of the examples
defines the whole mesh as a pressure zone (with '&ZONE /') and then a
2nd smaller pressure zone within. Presumably then pressure zone 1 is the
entire mesh minus the 2nd and pressure zone 2 is just the smaller
defined section?

Regards,
Cian

[Kevin]

Just do not use a pressure zone. Pressure zones are only useful when
you want to use empirical leak or ventilation models.

[Cian Davis]

The reason I'm using pressure zones is that I want to model leakage
through a closed door using a LEAK_AREA and this is the focus of the
work. Modelling the leakages directly isn't an option.

[Kevin]

We do not want to combine natural ventilation using OPEN with an
empirical leakage model that is based on average compartment pressure.
I'm not confident that the results will make sense.

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

[dr_jfloyd]

Not sure what exactly you are modeling:

1) A room with both an opening to the ambient and a door that leaks:

In this case you aren't going to get any significant air movement through the leak as you can't build up pressure. 

2) A series of rooms, some of which are open to the ambient and some of which only have leaks.

In this case, you can define the rooms that only have leaks to be a ZONE.

[Cian Davis]

1. This is what I tried. I got a pressure rise of about 10Pa. Not
massive, but would be enough to transfer some smoke to an adjacent zone.
Is there any other way you can model small gaps in this scenario? The
holes are too small to model directly.

2. This approach might be a possibility for my problem. I'm guessing
though that I could only define leaks between ZONEs (including ZONE 0)
and not model smoke flow to/from rooms that are open ambient?

Regards,
Cian