Why do double-paned windows have a 2cm gap?

March 6, 2009 8:46 PM

Why is a 2cm gap between window panes optimal?

I came across this 2cm/0.75in figure while researching solar oven design. It appears to also apply to regular housing windows, and walled insulation in general. Nowhere have I found a justification.

Is it based on some engineering principle, or law of thermodynamics?

posted by unmake to Science & Nature (7 answers total)

expansion and contraction
posted by pianomover at 9:07 PM on March 6, 2009

If the panes are too close together, then heat can easily conduct from one pane to the other. If they are too far apart, then the air inside can circulate to make convection currents that transfer heat from one pane to the other.
posted by JackFlash at 9:08 PM on March 6, 2009 [1 favorite]

Best answer: In engineering terms, the spacing is a compromise between two heat transfer mechanisms, conduction and convection. Conduction is reduced when you move them farther apart. Convection is reduced when you move them closer together. There is an optimal spacing that reduces the combination of the two competing effects to a minimum. The optimal spacing depends somewhat on the temperature difference between the inside and outside of the window. The optimal spacing also depends on the gas used to fill the space between the panes. If you use a heavier gas than air, like argon, krypton or xenon, then conduction is reduced so you can space the panes closer together to reduce convection. Panes can also be coated with materials that reduce transmission or reflects thermal radiation, which is the third mechanism of heat transfer.
posted by JackFlash at 9:32 PM on March 6, 2009 [2 favorites]

Best answer: The glass itself doesn't provide much insulation.
Most of insulation is provided by the air against the glass.
have a look at the diagram on this page:

http://webserver.dmt.upm.es/~isidoro/bk3/c11/Exercise1.htm

it shows the temperature across a double pane window.

of particular interest, have look just inside the inside window.
The warmest spot isn't just inside the glass it actually some distance away from the glass.
This is because a still layer of air against the glass acts as a insulating blanket.

You can test this yourself by holding your fingertip about a cm away from a window inside a warm house on a cold day, as you move your finger tip towards the glass you can feel it get colder.

This still layer isn't very thick, this diagram seems to indicate that most of the benefit is in the first 0.4 cm. I suspect it might be a little different if you use different internal and external temperatures.

so if the blanket of air provides good insulation for thickness x, it makes sense that you'd want to make the gap between the panes at least 2x.
posted by compound eye at 10:47 PM on March 6, 2009

Others have it: two inches (-ish) is too small to form a stable convection cell, and still air is a lousy heat conductor. Finding that it's two inches would make a nice thermodynamics final exam problem.
posted by fantabulous timewaster at 10:50 PM on March 6, 2009

Irrespective of insulation characteristics, a window thicker than 2 cm wouldn't sell. No one wants windows that are ridiculously thick.
posted by Chocolate Pickle at 11:06 PM on March 6, 2009

That is not true. Historically the first double pane windows were created with storm windows and double single pane windows. The double singles on my house are a good 3" thick overall. They sold plenty of double single aluminum sliders mounted in a wood jamb in the 60s and 70s.

Also people in colder areas routinely buy triple pane windows when they can afford them which are twice as thick (give or take) as a double pane window.
posted by Mitheral at 12:16 AM on March 7, 2009