I'll pose a purely academic question for something I'm pondering.,... How much does infiltration in a house increase as wind speed increases? If a house tests out at, say, 5 ACH5, I have read various "rules of thumb" that say to find the "natural" infiltration rate divide the ACH50 by about 20. So in this case the ACHnat would be 0.25 ACH. Since 50 Pascals represents the pressure of wind at about 16 mph can I assume the infiltration rate in any house will be 20 times as much in a 16mph wind as it is in still air? Is it only 8 times as much because with a blower door it's all coming in and in a wind half is in and half is out? At what wind speed does the magic R-value of "air film" disappear? Another way to phrase my question might be this - I know that air sealing a building is important anywhere, and that it's even more critical where it's typically windy - but can anyone tell me HOW MUCH MORE important it gets as the wind increases?? Happy New Year to all!

Too many factors to consider- there IS no rule of thumb that's reliable across any residential structure. Even the shape of your roofline changes the venturi-effect on actual & parasitic flue structures that might exist. The infiltration rates may be quite different depending on the direction of the wind as well. The delta-T between indoor & outdoor air drives the whole-structure stack effect even in still air, and even with identical ACH50 numbers the would be quite a range from that factor as well, independent of wind speed.

The magic air-film is just that- magic. Consider it toast at sub 1 mph flow rates or high delta-Ts. It really only applies to dead-calm air, where the delta-T is low enough that the induced convection currents from the delta-T are laminar, at a ants-pace. At big delta-Ts the convection currents even within fiberglass batts kills off R value measurably. With not even fiber to slow the flow, any presumed-R from the air films at an relevant delta-T are barely more than a fantasy even before the turbulence of a 0.1mph "wind" breaks up the laminar flow.

According the US military MIL-HDBK-1003/1 spec, a theoretical air film in dead-calm air, low delta-T is good for ~ R0.68, but that drops to ~ R0.25 in 7.5mph wind, R0.17 in a 15mph wind. (I'm not sure about the test conditions by which that spec was verified.) That's still a difference of only R0.5 between dead-still and 15mph. (You'll see a bigger difference than that in 2" of EPS or fiberglass batts between 75F vs. 25F outdoor conditions- albeit in opposite directions.)

Dead-air spaces between layers in a wall stackup with highly reflective (read "low emissivity") surfaces facing each other different story- you can sometimes achieve R1.5 or better. It works better on horizontal surfaces with the hot side up though. Vertically, not so much.

Dana,
Thanks for the cogent reply. Intuitively I have always thought the "air film" was magic and never could bring myself to figure it into any calcs.

I'm trying to develop some recommendations/incentives for building "beyond the code" in an area where the building code rates the wind gusts at 120 mph. The IRC (dealing mostly with houses) and I think (but still need to research) the IBC also ignores the added effects of high wind. In essence, I'd like to be able to say something like - "Given an average wind speed during the heating season in this area of 18 mph the infiltration portion of the heating load could be "x" to "y" times the infiltration load for an identical house located in a sheltered area with the same temperature ranges." I suppose I could make up some numbers and no one could prove me wrong - but that seems a bit unethical - actually more than a bit. Maybe I just shouldn't go there, but my gut tells me air sealing is critical and WAY MORE critical when the wind blows 20 - 25 and I'd love to find a credible way to make others believe.

One interesting bit I found in searching this was a study done on wind breaks and an actual test done in central Pennsylvania - infiltration was measured in test structures with and without windbreaks of white pine trees and found the infiltration was reduced by 12% or more. Most effective was a break locate a distance from the structure equal to the tree height. More effective than I would've thought.....

Birdman-

I think that your questions about wind effects and shielding are directly addressed by Max Sherman and Nance Matson in the "Appendix: Modeling Tools" of their report at http://epb.lbl.gov/Publications/lbl-39036.pdf. If you have further questions beyond what is discussed there, I would start out doing a literature search on Max Sherman, as he has a number of publications that are pertinant.

Lee

If I do this right, there will be a PDF chart attrached with windspeed converted to windpressure in a variety of units with the corresponding Beufort scale thrown in for good measure.

This chart is used for structural calculations of commercial building facades and is of absolutely NO use for residential energy calcualtions.

As the others have pointed out, for any given wind speed and direction, different portions of your building will be under varying negative and positive wind pressure. 

If you had pressure readings from a wind tunnel test (which we normally do for tallish buildings) calculating actual infiltration/exfiltration could probably be modeled.

Bruce