PRESSURE ZONEs

[giacomo]

Hi All,
I am exploiting pressure zones to handle leakage and pressurization
effects between two adjacent zones (A and B). I have Zone A which is
positively pressurized by a fan with flow leaking into Zone B, which
leaks to the outside (Zone B leak >> zone a leak as the latter is to
be the dominant flow resistance).

MESH IJK=25,50,27, XB=0.0,5.0,0.0,10.0,0.0,2.7 /

&ZONE XB= 0.0, 5.0, 0.0, 5.0, 0.0, 2.7 LEAK_AREA(0) = 1.00 /
&ZONE XB= 0.0, 5.0, 5.0,10.0, 0.0, 2.7 LEAK_AREA(1) = 0.03 /

The two zones are separated by a wall:
&OBST XB= 0.0, 5.0, 5.0,5.2, 0.0, 2.7 /

The leak between the two zones is assigned to the a "common" door
(which is represented by a VENT on a solid obstruction not
representing a boundary between the two zones)

&OBST XB= 2.0, 3.0, 4.8, 5.0, 0.0, 2.2 /
&VENT XB= 2.0, 3.0, 4.8, 4.8, 0.0, 2.2 SURF_ID='LEAK_DOOR'/

&SURF ID='LEAK_DOOR' , LEAK_PATH=1,2/

Zone A is pressurized by a fan. If I set VOLUME_FLUX = 0.5 then I have
a resulting pressure of 167 Pa (which is the result that can be
predicted using the simple leakage formula in the Users Guide). If I
reduce VOLUME_FLUX then the agreement between calculated and simulated
pressures is not mantained (simulated pressures are greater than
calculated ones).
Any indication about the possible cause of the lack of agreement will
be appreciated,
Thanks in advance,

Giacomo

[Kevin]

Submit a simple case to the Issue Tracker.

[dr_jfloyd]

What version of FDS are you using?  If you are using version 5, then it is not stable for large leakage areas.  If you are using version 6, make sure you are using the latest source and if you still have errors then open an issue with simple input file that replicates your problem.

[Cian Davis]

Hi Dr. Floyd,
Any advice on where you'd draw the line for large leakage areas? I use
leakage areas between 0.001m2 and 0.36m2 and wondering if I should
change approach for larger holes.

Regards,
Cian

[Kevin]

If the leak area is comparable to the size of a grid cell, you might
want to consider using HOLEs and OBSTs directly.

> > then open an issue with simple input file that replicates your problem- Hide quoted text -
>
> - Show quoted text -

[dr_jfloyd]

The model was intended for leakage areas associated with typical construction which Klote and Milke report as being area ratios from 1E-3 to 1E-4 (area leak / area wall).  If you get larger than these, then you might have problems with the leakage model in FDS 5.  If your cases are running for you, and you don't see odd pressure behavior, then you are probably OK.

[Cian Davis]

Hi Kevin,
Would the hole need to be greater than one grid cell to obtain
sufficient resolution? I always try and ensure any directly modelled
gaps are at least 3 grid cells by 3 grid cells

Regards,
Cian

[Kevin]

How good do you think that leakage model is? Best to do some V&V to
determine what is most accurate.

> >> Cian- Hide quoted text -

[Cian Davis]

The comment was meant in a more general sense. A fluid path with only a
single cell would provide only a single node for solution. I was
wondering if anyone had an opinion on whether fluid paths should have a
greater number of cells.

Regards,
Cian

[Cian Davis]

Hi Dr. Flloyd,
Do you mean the Klote and Milke book "Principles of Smoke Management" or
some other report? I've been unable to find the reference to the area
ratio in the book.

Regards,
Cian

[dr_jfloyd]

Yes.
Table 4.3: Typical Leakage Areas....
Column 3: area ratio

[Cian Davis]

Ah, so the recommendation is based more of typical observed areas rather than some investigated limit on the numerical model?

I haven't been able to identify the exact table - I must have a different version of the book but I think there is a similar table in BS EN 12101-6 (the British Standard used to calculate and specify pressure differential systems for smoke and heat control)

[dr_jfloyd]

We've not done a detailed study on the leakage model that is in FDS 5, and would have no plans to do so since it no longer exists in its current form in FDS 6.

The general observation has been that when leakage areas in inputs are commensurate with leakage fractions of 1E-4 to 1E-3 that the model in FDS 5 tends to be stable.  When the leakage areas are much larger, the model tends to not be stable.

[Cian Davis]

Sorry if I came across as suggesting someone else should do the work! Not my intention at all! I'm just wondering if people had some guidance/experience.

The reason I'm looking at this is because I'm seeing some weird effects when I have actual mesh-resolvable holes - it seems to act as a flow barrier for about 700 sec and then acts as expected. I'm still running some test models to nail down the problem and ensure it's not something stupid I'm doing.

It's also related to my other questions about OPEN vents in pressure zones. I was wondering if specifiying a large LEAK_AREA(0) would be a better way to do things. Various test models running trying to do some investigation.

Thanks for your help!

Cian