What’s the range of window air leakage values for good-quality US windows vs. garden-variety vs. Passive House-quality?
It’s very hard to find published window air leakage rates (in cfm/ft2), harder still to get independent test reports with this information. So far I’ve found the following data points:
* a maximum of 0.3 for all of one US mfr’s product line (in an internal manual);
* .02 for a tilt-turn unit by same mfr (from independent test report);
* .003 for a Canadian mfr’s tilt-turn unit (from website literature).
I’ve also been told that the test apparatus doesn’t yield a result smaller than .01.
What’s the real story here? What is a typical value for a US good-quality (i.e., double-pane, low-e, wood or fiberglass) residential window? What is a typical value for an older, single-pane window? What is a typical value for a Passive House-worthy window?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Ann,
To get you started, I found this document from Wasco Windows to be helpful:
http://www.wascowindows.com/Documents/U-factor.pdf
Here's the relevant section:
"How is air infiltration measured?
Air leakage is measured by physical testing of a standard sized window. The test window is installed on a large wall. On one side, a vacuum is of 75 pascals (1.57 psf) is applied. This corresponds to approximately a 25 mph wind blowing perpendicularly to the window. Flow meters measure the rate of air leakage. This is divided by the total window area to get a reading (in the U.S) in cfm/ft2 to the nearest tenth (i.e., a test value of 0.17 would be given a rating of 0.2). The best possible rating by the NFRC is 0.1, as they will not round to 0. ...
Why doesn't manufacturer “X” have air leakage ratings on their windows?
The NFRC does not require manufacturers to test for air leakage. ...
Does air leakage have a large influence on energy bills?
If a manufacturer elects to have its product tested, it cannot have more than 0.3 cfm/ft2 of air leakage. If this maximum is met, the energy loss through air leakage is likely considerably less than that through heat loss (ASHRAE 2005 Fundamentals, P. 31.55). Nonetheless, air infiltration has a major effect on perceived performance of the window. Even at the passing value of 0.3, disturbing drafts may be felt. These drafts are often the very reason a customer is replacing a window!
Are some window designs better than others for air infiltration?
Yes. Casement and awning windows offer excellent air infiltration performance because pressure from the wind tightens up the seals. Tilt-and-turn windows, with their dual compression seals and multiple locking points around the perimeter offer equal or better performance. Traditional double-hung windows (sometimes called “vertical sliders”) cannot do as well due to their sliding surfaces where compression seals are not possible. And, horizontal sliders generally have the worst air infiltration performance of all window types. Beyond the sliding surfaces of the double hung, these windows have frames designed for sash removal. This feature, along with necessary provisions for drainage, make it impossible to seal these windows as tightly as a double hung, much less a casement."
Thanks, Martin, this is very helpful!
HA- HA. The gentleman who wrote this is definately one who any window replacement costumer should hear! I almost made the mistake of buying a double pane window from a company. They had all the low U-Factor ratings, and SHGC numbers, but when I operated their sample, you could easily see how when opening the upper, or lower window up & down, air could seep through the sides where they slide on each other !!!
Any single hung will provide for less dusting of your home. Always check how tight, and the micro- fibers that seal the moveable parts, and make up your own decision.
Personally, I don't want outside air comming in (drafts)... that's why I replaced my windows in the first place !!!
Don't get ripped off... make your own assessment of a window with your own eyes, not just words in a brochure, or numbers. AIR LEAKAGE between moveable parts is something to LOOK AT - EVEN IF not too many people have figured out it's importance regarding window replacements.
I've been considering this same issue as I look to replace my windows. After reading Martin's article on blower door basics on this site, I learned that a reasonable target for air leakage at 50 pascals is around 1000 -2000 cfm with older leaky homes measuring up to 5000 cfm. For my house, I have roughly 250 ft2 or window area, so a .3 cfm/ft2 window that meets the AAMA/NFRC standard tested at 75 pascals would likely contribute around 75 cfm of air leakage for double hung windows. Given that this is only 7.5% of the leakage at a 1000 cfm target, there are probably many other larger contributors to overall air infiltration than the windows. Hopefully this performance is maintained over the life of the window so durability for the weather stripping something to consider.
http://www.nfrc.org/documents/ReplacementWindows.pdf states that air leakage numbers typically range from .1 to .3 and this is consistent with values I found on the Andersen, Pella, and Paradigm sites. The casement number are down to .01.
I've been considering this same issue as I look to replace my windows. After reading Martin's article on blower door basics on this site, I learned that a reasonable target for air leakage at 50 pascals is around 1000 -2000 cfm with older leaky homes measuring up to 5000 cfm. For my house, I have roughly 250 ft2 or window area, so a .3 cfm/ft2 window that meets the AAMA/NFRC standard tested at 75 pascals would likely contribute around 75 cfm of air leakage for double hung windows. Given that this is only 7.5% of the leakage at a 1000 cfm target, there are probably many other larger contributors to overall air infiltration than the windows. Hopefully this performance is maintained over the life of the window so durability for the weather stripping something to consider.
http://www.nfrc.org/documents/ReplacementWindows.pdf states that air leakage numbers typically range from .1 to .3 and this is consistent with values I found on the Andersen, Pella, and Paradigm sites.
Robert,
A leakage target must take into account the volume of the heated space. 1000 CFM50 alone is meaningless. In a 2,000 SF house, that would amount to 3.75 air changes per hour (ACH50), but in a 1200 SF house that rate would equate with 6.25 ACH50.
Air changes per hour under 50 pascals pressure is a more useful number. Typical new homes leak about 3.9 ACH50. A tight house will leak 1-2 ACH50. How that relates to a home's natural leakage rate, under normal stack effect and wind pressures, depends on the climate zone and the height of the building.
IECC standards for maximum fenestration air leakage are:
Operable windows, skylights, and sliding glass doors: 0.34 cfm per lineal foot of operable sash crack or 0.30 cfm per square foot of window area
Residential doors, swinging: 0.5 cfm per square foot of door area
Residential doors, sliding: 0.37 cfm per square foot of door area
Also remember that, when assessing the value of window replacement, existing old weight-and-chain double-hung window sash will perform dramatically less well than the worst case new unit rated at 0.3. This is compounded by the leakage at the weight pocket and casing, which were never designed for air sealing. Replacement sash will not be able to perform well either. The only window replacement scenario that succeeds in the goal of air sealing (not the sole goal, we understand) is a new full-frame unit with modern air and water sealing at the building's air barrier. You can easily verify this yourself with a blower door. Enjoy. Also remember that the NFRC test is at 75 pascals, not 50.
So with this being said:
"Does air leakage have a large influence on energy bills?
If a manufacturer elects to have its product tested, it cannot have more than 0.3 cfm/ft2 of air leakage. If this maximum is met, the energy loss through air leakage is likely considerably less than that through heat loss (ASHRAE 2005 Fundamentals, P. 31.55). Nonetheless, air infiltration has a major effect on perceived performance of the window. Even at the passing value of 0.3, disturbing drafts may be felt. These drafts are often the very reason a customer is replacing a window!"
Does that mean that a window rated at .02 from the manufacturer will not actually perform much better than a window rated at .3 cfm? Would this difference account for double the cost if the U and SHGC values are almost the same?
I'm about to purchase some windows and this is the last hurdle. I only have a total of 6 moving windows in the house, the rest are fixed.
Michael,
I'm not sure what you are asking. You phrased the question this way: "Does that mean that a window rated at .02 from the manufacturer will not actually perform much better than a window rated at .3 cfm?"
There are many ways to talk about window performance. Air leakage through the window is certainly part of window performance, so a window with a very low rate of air leakage certainly performs better than a window with a rate of air leakage that is 15 times worse (as in the example you proposed).
Martin I understand that there is indeed a significant difference between .02 and .3. The question is, is that difference enough to justify spending $1000 on a 6' by 5' window or paying $450? From your response to the original question above it seems that below the .3 rating would not cause a huge change since the thermal performance seems to be the dominant factor below the .3
I'm not as worried about feeling the draft as I am feeling my wallet empty. With 6 windows at .3, I'm loosing 1.8 cfm from my windows in a 2100 sq. ft house with cathedral ceilings (more volume). With 6 windows at .02 I'm still less that one of the other windows at .12 cfm.
After seeing how they installed one of the exterior doors in my house, I'm probably loosing 10 air volumes an hour through these alone. My wife and I removed a door from and could see outside, not to mention the entire header is rotted out from piss poor flashing above it.
Thanks for the help!
Michael,
I agree that 1.8 cfm for 6 windows isn't much, especially since the typical mechanical ventilation system for a single-family home moves 50 to 80 cfm.
On a cold night, however, the infiltration might bother some people.
I'm guessing that the upcharge isn't worth it for the energy savings... but it might be desirable for comfort (for some people -- not everyone).
With the vertical location of most windows on single story homes essentially bisecting the neutral pressure plane, the actual leakage from stack effect (infiltration on cold nights) tends to be less significant than these manufacturer numbers alone would suggest. And as such, a discussion on the dominant ventilation method used within the home, and whether central space conditioning systems include pressure provisions for interior door closures, should be included in discussions of window air leakage to frame the perspective.
The great thing about window air leakage is that, in most cases, that's the good place to get your breathing air from. It's not the place to begin any air sealing efforts, unless of course you like to pick your best tomatoes only to throw them into the dirt.
With so much air sealing opportunity within average homes, a suggested loading order before becoming overly concerned about improving on the air leakage value on new windows is only after a balanced ventilation system has been installed and commissioned, and the 60% air leakage opportunity from attic, crawlspace and air distribution system leakage has been realized.
Michael,
The 0.3 number is cfm per ft^2. For a 6x5 window, that would be 30x0.3=9 cfm for each window @75 pascals. For six such windows, it would be 54 cfm - not 1.8 cfm. For windows close to the neutral pressure plane, your conclusion that the infiltration is small relative to a poorly installed door probably still holds.
Thank you all for your answers. Reid thank you for clarifying the value and calculation. As far as the window and door installation, I will be fixing all of these as I add a WRB and some foam to the exterior of my house. With these new values, would this amount of leakage be enough to contribute to noticeable amount of energy savings with the much tighter windows?
The builder cut many corners... no WRB on the house, corners of the house are not sheathed over but dimensional lumber from framing, rotting door and window frames from lack of WRB and flashing details.... I'm sure I'll find more fun things to contend with.
David Paulus, PhD, P.E., stated:
"Air leakage is measured by physical testing of a standard sized window. The test window is installed on a large wall. On one side, a vacuum is of 75 pascals (1.57 psf) is applied. This corresponds to approximately a 25 mph wind blowing perpendicularly to the window. Flow meters measure the rate of air leakage. This is divided by the total window area to get a reading (in the U.S) in cfm/ft2 to the nearest tenth (i.e., a test value of 0.17 would be given a rating of 0.2). The best possible rating by the NFRC is 0.1, as they will not round to 0. (AAMA records to 0.01)
Air infiltration becomes a greater percentage of total energy loss as the window's U-Factor becomes lower, because the energy loss through heat transfer is reduced. Low air infiltration is critical for a passive solar house, where energy loses must be kept to the absolute minimum. For these houses, casements or tilt & turn windows are the only proper choices."
A European triple pane window like Intus Windows uses a triple window seal & a tilt & turn multiple locking point design to provide better air tightness and so they have readings of < 0.03 CFM/ sq.ft. on their windows. This makes them very, very air-tight.
Peter, thank you. I'm just trying to get my house to pretty good given it's already built and pretty terribly at that. The the windows I was looking at were Soft-Lite Brand with a U of 0.28 and air leakage of 0.02 cfm/ft^2. The 3050 Twins were quoted just under $1000 for a double pane w/argon. The other brand are Window Mart, which are lower quality overall, but with the same U value and the air leakage value is not published, but they are Energy Star Certified which is why I assume a value of 0.3 cfm/ft^2. These were about $500. The picture windows were similar in cost difference with the same U factor and no air leakage.I plan to see both in person soon.
Another problem I have is the Soft Lite Brand does not make trapezoidal picture windows that are as small as the ones I have so I have to find a second brand for these. I may end up using one for the sliding windows and one for the picture windows if the cost justifies it. I'm in zone 4 St. Louis Metro, so it's not extremely cold, but gets somewhat warm in the summer.
I have checked with a few brands I've seen on here and can't find any, or don't get responses from most that do business in my area. This has been my biggest challenge and has led me to some local window shops.
By the way Peter, I ended up ordering three rolls of Siga Wigluv and some Ducskin. After some more research and watching a few installation videos and comparing my own results with other tapes, it seemed like a right idea.
Michael,
Have you gotten any bids from Euro window manufacturers like Intus?
SIGA tapes are awesome and as one GBA poster stated, "SIGA tapes make ZIP Sheathing tapes seem like blue painters tape."
Peter, I checked my emails and the Intus site and I emailed and old dealer and never got a reply. I will try to see if I can get a hold of the ones now listed. This has been my biggest problem, getting someone to respond or that will deal with me without an installer.. I'd rather order directly from a manufacturer if possible to save on the middle man costs. If you know of any that may for sure deal with Missouri please let me know.
For the picture windows I really don't see any reason to spend double for a 0.01 difference in U value.
Michael,
With windows, the manufacturing plant will never sell directly to the end user unless you are a big time builder that buys millions of dollars of windows per year. There is ALWAYS a middle man. It's like with Lowe's or HD, when you buy lumber, screws, plumbing, etc., the manufacturer will never sell directly to you. They sell through Lowe's which is the "middle man".
If you don't get anywhere with the dealer rep, I would contact Intus directly and let them know.
Peter,
You're right that when Lowe's sells lumber, it is the middle man. But it's perfectly possible to buy lumber from a sawmill. I just did it this week.
Peter,
Over the last couple of decades I've bought six different brands of windows. Always from the manufacturer, never a middleman.
Peter, I understand, but I'd rather buy the windows from Intus lets say rather than Bob's windows, who will just buy the window and add a mark-up and sell it to me. I have actually been into a glass manufacturer (PPG) and a window and door manufacturer (Pella), so I know that it would be about impossible to buy direct since the glass isn't even made in the same plant that the windows are assembled. I did however get a response from an Intus rep, I'll wait to see prices before I get excited.
I watched this happen when I ordered siding the other day. I had to purchase it through a rep, the manufacturer faxed their cost to the rep, he added 15% on the quote he gave me and called it a day. I could have just as easily picked up the phone and made the order, but they won't sell direct.
Martin,
Depends where you live. In my neck of the woods there are no sawmills. Lumber is trucked in from out of state. The distribution yards out here will tell you to take a hike if you don't have a GC license.
Malcolm,
You must have the term "middleman" confused. Depending on the manufacturer, they either have in-house sales reps or regional distributors. Intus has authorized regional distributors. Intus is located in Washington state hence the reason they have state distributors. It's not like you are getting the window at "manufacturer cost" by going directly to the manufacturer. Doesn't work that way. The commission/markup is done by either the local distributor or the manufacturer.
So I guarantee you the last 6 windows you purchased over the course of the last decade where marked up, they were marked up by the manufacturer who was then acting as the dealer or "middle man". Sorry to burst your bubble but you didn't buy them at "cost", you still paid the middle man which was the manufacturer sales rep you dealt with. They acted as the middle man. It's not like you walked onto the shop floor and bought them off of the assembly line for cost. The manufacturer (who has also then acting like the sales rep/middle man) marked them up and sold them to you.
Manufacturers either have to pay full-time office sales people or they have to use commission based sale people. Call it what you want but there always is a middle man. Someone has to be there to talk to you and make the sales deal. Whether it is a full-time 40 hour a week sales person at the manufacturers office or a commission only based distributor. The mark up is there for both scenarios.
Peter, When I call the manufacturer's sales desk at their plant the situation is in no way analogous to my buying that or a similar brand at a big box store. Calling the rep who answers the phone a middleman suggests a basic misunderstanding as to how their businesses work - as does your idea I'd expect to get them at cost. No one sells anything at cost, regardless of the volume, they wouldn't stay in business long.
Most builders, such as myself, deal directly with the manufacturer if the product is based relatively nearby. So that's what I have done on all my projects over several decades, with six different manufacturers. Why would I deal with a middleman, and why would you discourage the OP from trying to avoid someone who wants to include installation?
Just out of interest, what is your building experience?