Our homes include insulation to reduce heat flow through floors, walls, and ceilings. Some parts of our homes’ thermal envelopes (for example, insulated ceilings) are well insulated and have a high R-value; other parts (like windows) have a much lower R-value. But during the winter, as long as we have an adequate heating system that keeps the indoor air temperature at 72°F, we should be comfortable — right?
Not quite. Even when the air temperature is held to a steady 72°F, occupants can be cold during the winter — especially if they are standing or sitting next to a large window.
Why windows give us a chill
Someone who sits next to a window on a cold night might say, “I feel a chill. I think this window is drafty.”
That person might be right. If the window has bad weatherstripping, outdoor air might be leaking around the sash, creating a cold draft.
There’s another possibility, however: the cold window pane is probably cooling the interior air near the window glass. As that air gets colder, it also gets denser, so it falls to the floor. The falling cold air pulls warmer air near the ceiling toward the glass — air which is cooled in turn. A convective loop is set up, and people sitting nearby might accurately note that they feel a slight breeze — even if the window isn’t letting any outdoor air leak in.
There’s a third way that windows can give us a chill, and it has to do with radiative heat transfer. The surfaces in a room emit longwave radiation: that is to say, they radiate heat. People also emit longwave radiation. If I am sitting near a hot wood stove, the stove is emitting more radiation in my direction than I’m emitting to the wood stove.…
Get building science and energy efficiency advice, plus special offers, in your inbox.
This article is only available to GBA Prime Members
Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.Start Free Trial
Already a member? Log in
Over thinking things
I need more science. So if I have a large expanse of glass and pull a blind, I feel warmer. But is it because I've reduced the radiant heat loss to the glass...
1. because the room side of the blind is likely warmer than the room side of the glass by a few degrees (and a quilted blind would be warmer yet)? Or,
2. your radiant heat loss is technically to the outdoor temperature through the transparent glass (depending on coatings), and the non-transparent blind inhibits the radiant heat waves?
Or, in other words, is it simply and only a matter of your body temp relative to the glass surface, or does material transparency of the glass and blind play a part too?
Response to Elden Lindamood
Q. "If I have a large expanse of glass and pull a blind, I feel warmer. Is it because I've reduced the radiant heat loss to the glass because the room side of the blind is likely warmer than the room side of the glass by a few degrees (and a quilted blind would be warmer yet)?"
Q. "If I have a large expanse of glass and pull a blind, I feel warmer. Is it because I've reduced the radiant heat loss to the glass because your radiant heat loss is technically to the outdoor temperature through the transparent glass (depending on coatings), and the non-transparent blind inhibits the radiant heat waves?"
A. No. It's the temperature of the innermost pane of glass that matters; the transparency of the glass is irrelevant, and the outdoor temperature matters only indirectly (because it affects the temperature of the glass).
When you pull a blind, radiational effects continue. Your warm skin is still radiating heat to the surface of the blinds.
You can use the tool to show
You can use the tool to show that the emissivity of the interior side also matters. Eg, a room-side low-e coating reduces the glass temperature (increasing draft discomfort) but slightly increases the radiant comfort (ie, feels warmer).
Response to Jon R
Thanks for your comments, and for reminding me that emissivity matters -- not just the temperature of the surface.
One can also use the tool to
One can also use the tool to verify that triple glazing isn't need to maintain comfort in zones 7 & 8.
Response to Jon R
Everybody has a different definition of "comfort" (although of course ASHRAE 55 has attempted to standardize the definition).
The level of discomfort will depend on the area of glass and the outdoor temperature.
In their 2006 paper, "Window Performance for Human Thermal Comfort," the authors (Charlie Huizenga, Hui Zhang, Pieter Mattelaer, Tiefeng Yu, and Edward Arens) published the graph that I reproduced as Image #2, above. Using criteria that the authors explain in the paper, they note that the lowest outdoor temperature for comfort is about -6 degrees C (about 21 degrees F) for double glazed low-e windows.
I'm not particularly uncomfortable near a double-glazed window until the temperature drops to about 0 F, but everybody's different. Larger windows are more likely to lead to comfort complaints.
Forgive me if I'm going "too far into the weeds" with this, but it has piqued my interest.
So a particular low-e coating will reflect the radiant heat waves back at the source (person), making you feel warmer but preventing the heat waves from warming the glass surface, correct? In a sense, making the glass surface "appear" the same temperature as your body, less the effective modifier of the coating?
And this is what you see when you look at your heat reflection in a window with an infrared camera?
I may be venturing into the world of meaningless BS terms like "effective R value", but I'm just curious if you could do some math and arrive at a legitimate claim such as "this low-e coating has the same comfort effect as adding a PVC window shade", or some other vaguely worded statement about undefined values. Not that anyone would ever do that...
Tangentially, I have a wall of glass in my zone 7 house, with Cardinal tri-pane 180/180 IGUs and I do not feel cold sitting next to them when it is 0 degrees out. When it is really cold out (-10 or less), I do close the curtains at night. The curtains are about 12" inboard of the glass surface. They rest on the floor at the bottom, and have about a 1" gap at the top. Am I actually saving energy by pulling the curtain, or potentially just increasing comfort, or maybe a bit of both?
Also, when I get up in the morning, the bottom of the blinds has pulled against the glass (unless one of the cats gets in there). This would indicate to me that the warm air is leaving the 12" cavity at the top through the 1" gap, and can't be replaced at the bottom, thus pulling in the bottom of the curtain. Can I assume from this that I have arrested the cold air washing off the glass and onto the floor and slowed the convective loop?
Response to Elden Lindamood
Q. "So a particular low-e coating will reflect the radiant heat waves back at the source (person), making you feel warmer but preventing the heat waves from warming the glass surface, correct? In a sense, making the glass surface "appear" the same temperature as your body, less the effective modifier of the coating?"
A. Yes. Here's what I wrote about surface #4 low-e coatings a while back: "A relatively new type of double glazing includes a hard-coat (pyrolitic) low-e coating on surface #4 (that is, the innermost surface of glass facing the interior of the house). The use of a surface 4 coating allows double-glazing to have two low-e coatings, thereby lowering the U-factor of the IGU. Somewhat paradoxically (counterintuitively?), a surface 4 coating reflects radiant heat energy back into the room, and IGUs with surface 4 coatings have a colder interior surface than conventional low-e double-glazing without the surface 4 coating. The glazing performs better, but the surface is colder. However, if you are sitting near the window stark naked, the glass will feel warmer (even though it is colder)."
Image #2 (and their "Winter
Image #2 (and their "Winter comfort rating") is for some specific conditions and should not be over generalized to imply the need for triple glazing under other conditions. For example, use the Payette tool, reduce window height and bump the interior temperature up a few degrees. Or add perimeter heat. Comfortable even at -30 with higher U factor/value windows.
To continue the questions asked by Elden:
1. I'd be interested in a description of the how perimeter heat affects the radiant transfer between occupants and windows. Is it simply a matter of it increasing the surface temperature of the inner-pane?
2. How does perimeter heat, as opposed to heat from other locations, affect energy-loss through the windows?
Another response to Elden Lindamood (Comment #7)
Q. "I'm just curious if you could do some math and arrive at a legitimate claim such as, 'this low-e coating has the same comfort effect as adding a PVC window shade,' or some other vaguely worded statement about undefined values."
A. Let's try to avoid vaguely worded statements about undefined values. Here's what we can state: Adding a low-e coating always changes the U-factor (and therefore the R-value) of a window. A triple-glazed window with two low-e coatings will have a lower U-factor than a triple-glazed window with one low-e coating. So the comfort effects of the low-e coating is reflected in the lower U-factor. (I know that surface 4 low-e coatings seem to confuse the issue, but let's stick with U-factors -- that simplifies the concepts.)
Q. "The curtains are about 12 inches inboard of the glass surface. They rest on the floor at the bottom, and have about a 1 inch gap at the top. Am I actually saving energy by pulling the curtain, or potentially just increasing comfort, or maybe a bit of both?"
A. Probably a bit of both. In the worst-case scenario, there is so much airflow between the curtain and the window that no energy savings occur (or conceivably, there are increased energy losses due to the convective loop that has been set up). Everything you can do to interrupt the convective loop improves the energy performance of the window + curtain assembly. Extending the curtain to the floor is obviously better than leaving a 2-inch gap between the bottom of the curtain and the floor.
Even better: You could throw away the curtain and install a window quilt with weatherstripping at the bottom, tracks on the sides, and an airtight valance at the head. These window quilts aren't cheap, but they are designed to prevent convective looping.
Q. "When I get up in the morning, the bottom of the blinds has pulled against the glass (unless one of the cats gets in there). This would indicate to me that the warm air is leaving the 12 inch cavity at the top through the 1 inch gap, and can't be replaced at the bottom, thus pulling in the bottom of the curtain. Can I assume from this that I have arrested the cold air washing off the glass and onto the floor and slowed the convective loop?"
A. I'm going to play armchair detective and guess that (a) you sleep late, and (b) the sun is shining through this window when you observe the phenomenon.
On a cold night, the typical convective loop would involve cold air falling off the glass onto the floor (like a waterfall), pulling warm air in at the top.
Once the sun comes up, the convective loop reverses, and warm air leaves from the top of the assembly, pulling the bottom of the curtain toward the window.
Response to Malcolm Taylor (Comment #10)
Q. "I'd be interested in a description of the how perimeter heat affects the radiant transfer between occupants and windows. Is it simply a matter of it increasing the surface temperature of the inner-pane?"
A. Good question. According to my understanding, there are two ways that comfort is improved when a forced-air register or hydronic baseboard unit is installed directly under the window: (a) a curtain of warm air is established between the occupants of the house and the window, raising the air temperature on that side of the room (compared to the situation that would occur of the heat source were on an interior wall), and (b) the curtain of warm air raises the temperature of the innermost window pane (compared to the situation that would occur of the heat source were on an interior wall).
Q. "How does perimeter heat, as opposed to heat from other locations, affect energy-loss through the windows?"
A. To the extent that perimeter heat raises the temperature of the window glass, heat loss through the window will inevitably increase. There's no free lunch. Scenario one (heat sources on an interior wall) result in discomfort and lower energy bills. Scenario two (heat sources under the windows) result in greater comfort and higher energy bills.
Of course, switching from double-glazed windows to triple-glazed windows can improve comfort while lowering energy bills.
IMO, the primary benefit of
IMO, the primary benefit of below the window perimeter heat is to counteract downdrafts. Which apparently is usually the limiting factor in window related comfort.
In theory, it might be perfectly balanced, eliminating convective airflow over the window and reducing energy loss.
A secondary effect of a radiant heat source near a window would be to raise the mean radiant temperature and improve the distribution. Even if it doesn't raise the glass temperature.
Search for "Impact_position_radiator_Gary_Gong.pdf"
Response to Jon R
The real solution to this problem, as I'm sure you know, is to specify good windows. If the windows have a low enough U-factor, all of the heat distribution can occur near the center of the house, reducing the cost of the duct system. Comfort will be improved compared to high U-factor windows.
Heat distribution systems rarely run continuously, since peak load conditions happen rarely. Most equipment turns on and off in response to a wall-mounted thermostat. With high U-factor windows, here's what happens: the windows cool the indoor air, setting up a convective loop (a "waterfall" of cold air falling to the floor). Then the heating system comes on, and the heat source under the window reverses the direction of the convective loop -- at least for a little while. Then the thermostat is satisfied, and the heat source turns off.
At that point, the cold waterfall resumes. During most of these cycles, the occupants aren't particularly comfortable.
As usual Elden tries to blame the cat.
I'd say that window perimeter
I'd say that window perimeter heat may only be viable if one has already decided to install hydronic radiant with outdoor reset (can run continuously). Or perhaps there are some variable output electric heaters suitable for this purpose.
One can use the Payette tool to show that with cold temperatures, even good windows can cause draft comfort problems. And, for example, that a 2F bump in interior temperature can supposedly make double panes as comfortable as triple.
Response to Jon R
Your statement, "a 2°F bump in interior temperature can supposedly make double panes as comfortable as triple," is true, as far as it goes.
Another way to express this: When people feel a chill, they crank up the thermostat until they feel more comfortable. When this happens, the homeowner pays higher energy bills, for two reasons: (a) The double-pane windows leak more heat at night than triple-pane windows, and (b) Setting the thermostat at 74°F requires more energy than setting the thermostat at 72°F.
All of the comfort and thermal distribution problems resolve themselves with the installation of the InFlector see-through radiant barrier window insulator. Drapes and the like, films, etc. eventually absorb enough heat to re-establish the original problem.
Log in or become a member to post a comment.Sign up Log in