wind vs building air exchange rate

[riskengr]

I'm looking for a ball park figure on the relationship between wind speed and a building's natural air exchange rate(i.e, no mechanical ventilation). A couple sources seem to indicate it is linear, although less than linear as wind speed increases, while others indicate it varies with wind pressure (1/2 of vel. squared).

One of my software packages calculates it (toxic vapor cloud dispersion), using the linear method, but provides no documentation as to where the relationship comes from.

Thanks,

Stephen Haines, PE, CFEI
Haines Fire & Risk Consulting

http://www.hainesfire-risk.com

 

[Zesti]

I would say the velocity-squared relation (wind pressure) seems more logical.

But the actual amount of air exchange for a building due to natural effects (wind, draft)... there are so many factors that are of influence I don't think there is a good way of calculating this.

 

[willard3]

This is not a simple problem and I can think of many variables other than wind velocity just to start:
Delta T
Building height
Age of construction, window types, etc
Type of construction

This list is not exhaustive.

 

[marcoh]

The theory is simple, in practice it is very hard!

Wind pressure on the building is P=½ x ? x Vw^2 where
P = wind pressure
? = air density = 1.2kg/m3
Vw = wind velocity

Pressure is +ve upwind, neutral inside, and -ve downwind

Pressure loss through the building = A x K x V^2 where
A = Total area of openings
K = pressure loss coefficient
V = velocity through openings

Total infiltration = VA

Also need to do the above for both upwind and downwind facades, and what about facades not perpendicular to wind?

The difficulty is determining A and K for which you really need to pressure test the building to establish a pressure vs leakage rate factor or know the maximum leakage rate the building will be designed for (eg 10m3/hour/m2 @ 50Pa)

I just assume the following:
0.25 ACH if no windows or doors
0.5 ACH if windows and no doors
1.0 ACH if there are doors

or run a CFD model

In reality you are totally dependant on how well the building is constructed which is a total guess (eg did the guy have a big night on the booze the night before?)and rules of thumb work as well as any other

 

[riskengr]

Actually we did a prssure test on the building and know the ACH under test conditions (i.e., wind direction and speed, temperature, etc...). We were just looking for a Rule of Thumb or simplified formula to show the client the approximate change in ACH over different wind speeds so that they can get a better feel for the range of ACH that the building might see.

Stephen Haines, PE, CFEI
Haines Fire & Risk Consulting

http://www.hainesfire-risk.com

 

[IRstuff]

But how can you have a rule of thumb if you don't even have information about the direction of the wind?

TTFN

FAQ731-376

 

[riskengr]

The idea was to assume all conditions (temperatures,wind direction, etc...) being the same except for wind speed.

Stephen Haines, PE, CFEI
Haines Fire & Risk Consulting

http://www.hainesfire-risk.com

 

[IRstuff]

But wind blowing into an intake has a different effect than wind blowing in the opposite direction, doesn't it?

TTFN

FAQ731-376

 

[ParadigmEnergy]

The ASHRAE Fundementals Handbook (27.20 in the 2005 edition) has a section on residential infiltration that is helpul. Also the Energy Management Handbook has a section on infiltration (Sec 9.2). All approaches are very rough estimates. I've attached a spreadsheet that I made (also a rough affair) with excepts from these texts. Maybe it will help you.

 

[tys90]

Paradigm, I don't think your attachment worked correctly.

Just curious but why are you trying to find out ACHs without mechanical ventilation?

 

[KiwiMace]

Try reading this from the LBNL. It is more descriptive than ASHRAE, although the equations are hard to discern, so you might want to go back to the fundamentals handbook for these.

Note the Infiltration model section lower down. The wind factor is clearly a non-linear function.

Being that you've run a blower door, you should have good data to run through this model.