[Greenbuilding] windows - flush vs recessed

[John Straube]

I have heard numerous sources of "window must be in middle", but page 5 of your reference is an excellent example. Thanks for pointing it to me John.
The presentation is full of interesting and useful information, but also some obviously incorrect stuff.

The difference is U=0.85 vs 0.83 W/m2K for the whole window. This is R6.68 vs R6.84 or R0.16.
This is beleivable, although the THERM model is not realistic for the first two models (no space around window) and dangerous (how will water drain out at head? how will it drain at sill?)

On page 6 the author leaps to a 22% difference.  Clearly not possible with the kind of numbers she presents on page 5. Essentially unbeleivable unless there is a large window area and a massive thermal bridge. Or calculation errors.

On page 7 she shows another therm analysis. Again, one that traps water, but likely buildable.  No numbers given. Likely the window detail she planned on using

On page 8 there is a bunch of gobbly gook.  I can see the she is using  292 meter (3000 feet!) of window edge. This is about ten times or more what a house of the size she is looking at would be.  The addition seems odd. U is a measure of heat flow per unit area, you cant multiply by length, L!  You can add psi (linear heat loss) to U (area heat loss), and this is normal.  Of course, the answer is a simple UA(wall) + UA(window) +psi (length of frame-wall intersection). Very worrisome.

pp 9-11 have good examples which one rarely buildings (covering the frame with insulation on top and not the bottom makes the window look odd to most folks, and is challenging to build and maintain) but notwithstanding some subtlety about details, all those curves show a typical and expected result:
-the thermal bridging impact is small and not far from optimal if you avoid installing window in uninsulated part of wall.
- locating window near the middle of the insulation is usually best location and may reduce heat flow by as much as 2-3% versus near the out edge or inner edge of the insulation.

p 12. Finally a real window with flashing.  Good warning: using a chunk of steel that penetrates is bad.  Use plastic or tar-based products when you penetrate the insulation layer. Oh, note there is no insulation attached to the face of the window frame.

Equation on page 16 is ridiculous. If only it were so easy. It could be correct for a certain type of building with a specific window area in a specific climate, but almost certainly wrong for most of the US and Canada.
Pg 18 ignores the overheating during warm spring and fall days that are sunny.  SHGC=0.6 can work in many cold to cool climate homes, but only if the window area is tightly constrained or operable exterior shade is deployed. In Zones 4A or less, choosing this SHGC will result in very small windows or cooking in the swing seasons.


So, this is further evidence that middle of wall is not critical, but may be best by a couple percent. Windows installed outside the insulation layer can be significant thermal shorts, cause comfort and condensation and must be avoided.

While all the details matter, the big heat losses matter more, and designing for the last 100 Watts of heat loss is somewhat misplaced when there are so many other important building details that need concern (health, safety, durability affordability, maintenance etc) Even heat loss up the plumbing drain stack is likely more significant than a "decent" window install.


On 12-01-30 10:48 AM, John O'Brien wrote:
> On Mon, Jan 30, 2012 at 8:47 AM, John Straube<jfstraube at gmail.com>  wrote:
>> Hiya George. Of course I know it is per unit length, as I quoted the units
>> of W/mK
>> So if I have a 5' by 4' window (1.5 x 1.2m) the additional heat loss of 0.01
>> W/mK is 0.054 W/mK or in 0 F weather, 2 Watts. Lets say we allow us 0.03
>> W/mK difference between the ideal location and the location preferred by the
>> owner, or builder, or the window durability.  Now we are up to 6
>> Watts/window. In a house with 12 windows, this is 72 Watts, at DESIGN
>> CONDITIONS for Zone 5, and we have 2000 W flowing through the windows
>> themselves.  The prescence of absence of overhangs and neighboring buildings
>> will likely have this much ipact on heat loss.
>> The real reason to worry about this stuff are the examples of when it is
>> done horribly, and you have a 0.2 W/mK (which works out to 41W/window or
>> 500W for 12 windows at design) detail and you start to risk condensation on
>> the window edge.
>>
>
> http://passivehouse.us/passiveHouse/2010_Passive_House_Conference_Presentations,_November_5_files/2010%20Conference-Windows%20Roundtable-Bronwyn%20Barry.pdf
>
> Where most of the 'window should be in the middle' has been coming
> from. Anyone want to take a stab at her numbers?
>
> She claims "By moving the windows to the center of the wall the Annual Heat
> Demand was reduced by 22%."
>
> Seems somewhat not insignificant.
>
> J
>
> _______________________________________________
> Greenbuilding mailing list
> to Send a Message to the list, use the email address
> Greenbuilding at bioenergylists.org
>
> to UNSUBSCRIBE or Change your List Settings use the web page
> http://lists.bioenergylists.org/mailman/listinfo/greenbuilding_lists.bioenergylists.org

-- 
Prof. John Straube, Ph.D., P.Eng.
Dept of Civil Engineering / School of Architecture
www.buildingscience.com