FDS 6, stack_effect example

[Dave McGill]

Hi,

I have a few questions regarding the stack_effect.fds example in the FDS 6 release.

1.  Is there a reason to use HVAC features instead of zones and leakage paths? (The latter seems more understandable to me for this particular example.)

2. Is there a reason for defining a ZONE in the file?  I can't figure out its role. The results appear to be the same without it.

Thanks

Dave

[dr_jfloyd]

1.  Leakage paths use the zone pressure without correction for hydrostatic and are evaluated as an average flow over all surfaces with leakage.  So if we defined a leakage area and assigned top and bottom there would be no pressure difference and no flow.  Even if we did do leakage with hydrostatic pressure, the pressure differences across the neutral plane would average out to be essentially zero (as noted leakage flows are averaged over all leaking surfaces in the zone).  By using HVAC we avoid those problems.
2. In this case there is no difference between the background pressure in the building and outside; however, if you had a fire in the building, you would raise the building background pressure as the gas heated and would want the ZONE to capture that effect.

[Dave McGill]

Jason,

I want to do a multi-storey building with air infiltration through the outer walls and windows into the apartments and then  infiltration around the doors into the hallways and then from the hallways into the elevator shafts. I'll use HVAC for the leakage paths. Would I still have just have one internal zone even though there are floor separations?

Thanks

Dave

[dr_jfloyd]

The HVAC routines and their use for leakage are new, so not a lot of experience to draw on in answering your question.  It might be worth doing some low resolution and/or simplified geometry simulations looking at different ZONE approaches to see what seems to work best: one internal zone, shaft+each floor, shaft+each hallway+each apartment. My gut says you might be better off with more than one internal zone.

[Dave McGill]

OK, I'll give it a play around with it and report back.

Thanks

[Dave McGill]

Jason,

You're right; it works if each floor and the elevator shaft has its own zone. The attached file is of a 10 floor building with zones at the 1st, 5th and 10th floors.

As a follow-up to my earlier question, I noticed the door_crack example also uses HVAC. Is HVAC better than zone/leakage in all circumstances?

Dave

[dr_jfloyd]

When using an HVAC vent, the vent surface temperature is the upstream air temperature.   When using leakage, the surface temperature is given by the heat transfer solution (e.g. based on the &SURF definition).  So if you were to apply an HVAC VENT to the entire wall of a compartment and give the duct a leakage area, that entire wall would then have the temperature of the air flowing into the VENT.  You probably wouldn't want that to happen as you would no longer be computing the correct heat transfer to that wall.  In the door crack case, we are defining flow through a small region.  Not doing the heat transfer to the door over those few cells isn't likely to have a significant impact on the solution, so we can take advantage of the additional capabilities of the HVAC rather than using leakage.

[Dave McGill]

Hmmm........ so what is the best solution for a multi-storey building with a different zone for each floor and a single zone for the elevator shaft? Using HVAC with a relatively small vent to represent the leakage over the entire wall?

I am guessing that having a zone for each floor would allow use of leakage, to represent horizontal leakage. The problem would be using leakage in a single zone representing the tall elevator shaft. Is that correct?

[dr_jfloyd]

With a tall building, the leakage from room to hallway, hallway to shaft, should be OK with the normal leakage.  The problem is the shaft leakage to ambient.  Leakage is just using the background zone pressure which for the shaft would be at the bottom.  To capture the change in pressure over the height of a tall shaft, you need to use HVAC.  This is an area where there still some development/refinement needed to make this easier for the User.

[Dave McGill]

I modified my test case to have the leakage from outside to the room by leakage. The path from the room to the shaft remains by HVAC. The results are the same as using all HVAC.

[dr_jfloyd]

Thanks for the update, it is good to know that expectations and reality have coincided in this case.

[Dave McGill]

Jason,

You wrote,

...In this case there is no difference between the background pressure in the building and outside; however, if you had a fire in the building, you would raise the building background pressure as the gas heated and would want the ZONE to capture that effect.

I want to apply what we've been discussing to a real building, and am confused as the need for zones. I understand that they are required whenever leakage is used. It will be a difficult task to add zones to a floor plan where each apartment is not a simple rectangle (See attached), which is a good argument for using just HVAC and not leakage.   Is a zone required in a stack-effect case because the building is a sealed box containing warm air in a cold domain? 

Thanks

Dave

[dr_jfloyd]

Without ZONEs, every grid cell uses the same background pressure as a function of height and there is no rise in background pressure due to bulk heat or mass transfer in or out of the ZONE.  There can be of course be changes in the perturbation pressure that drive flow. 

In a region defined with a ZONE, the initial background pressure as a function of height is defined using the initial conditions in the ZONE (in the stack_effect case we use INIT to change the inside temperature which means there is a different decay in pressure as a function of height).  That background pressure can increase or decrease due to bulk heat or mass transfer, this is in addition to any perturbation pressure required for flow within the ZONE to occur.

Odd shapes should be OK as long as they are enclosed in walls.  Pressure zones are defined by first setting everything within the zone XB, then we look at neighboring cells to see if they are connected.  If so they are added to the pressure zone.  This is discussed in 9.3.1 of the FDS 6 User's Guide.