### Image Credits:

1. Bonded Logic

1.
Jun 22, 2011 2:25 PM ET

Superior secondary insulation
by Philipp Gross

I personally like Demin insulation for sound improvement in ceilings and walls. Also we are looking to use it to fill part of an installation wall (which I think every house should have). Doing that I do not believe convection plays such a big role, because it is enclosed by the air barrier (taped plywood) and the drywall. This leads to an earlier discussion on this blog ( http://www.greenbuildingadvisor.com/blogs/dept/musings/one-air-barrier-o... ), if two air barriers are necessary. My opinion on that is: Not needed to get the house tight but helps to prevent convection if batt insulation is used.

One thing I think a lot of people are confused about is your statement that completely stagnant air has a R- value of 5. I saw builders actually advertising with this. I know here in the US nobody seems to work with the value lambda describing the thermal conductivity of a material without the effect of it`s thickness and as the name says without the convection influence! I think this is a very useful value and air, still or not, has a lambda of 0.025 W/(m K) which ignorantly translates to R-5/in . With the influence of the thickness and convection still air never exceeds R- 1.

2.
Jun 23, 2011 9:45 AM ET

Edited Jun 23, 2011 9:48 AM ET.

The thermal conductivity of
by Dick Russell

The thermal conductivity of truly still air increases with temperature. At room temperature, the conductivity (lambda) is indeed around 0.025 w/mK, but conductivity at room temperature is of little interest as far as insulation goes, since temperature difference and thus heat transmission therefore are nil. At 0 F, the conductivity is around 0.023. Multiplying either by 6.93347 gives BTU-in/ft2-hr-F, and taking the reciprocal of that gives R in English units. Thus R ranges from around 5.7 to 6.3 for truly still air in the temperature range of interest (for a heating climate).

Saying "With the influence of the thickness and convection still air never exceeds R- 1" is misleading. If there is convection, the air isn't "still" anymore. Granted, keeping air truly still is an elusive goal. Porous insulating materials work by intercepting air at very tiny intervals to retard convective flow that occurs by pressure gradient either across the layer or induced within the layer by air density differences. The relatively high R of the dead air films around the particles is offset by the higher conductivity of the particles of the insulating material itself and by any convective flow occuring within the layer.

3.
Jun 23, 2011 11:51 PM ET

Cotton Insulation
by Andy Wahl

I predict cotton insulation will shrink. If it can be kept perfectly dry then maybe not. Let’s think about this, if we add small amounts of moisture nearly every day and then bake it. Do this for years and see if it is the same size 30, 40 or more years from now.

4.
Jun 23, 2011 11:57 PM ET

pre-shrunk?
by Tristan Roberts

Andy, that's an interesting thought, but I doubt it would shrink very much, if at all, for a couple reasons. One, it is made of recycled denim—which has already been exposed to water quite a lot. Cotton seems to do most of its shrinking in the first wash. Two, the cotton is fiberized at the plant, mixed with polyester and polyolefin, and baked at 350 degrees F. I think that plastic content would help prevent shrinkage, just like it does in clothing -- maybe even more so since the cotton fibers are so small in the insulation.

This would be interesting to test, though. Maybe I should run a sample through the wash? Err... maybe not.

5.
Jun 24, 2011 1:26 AM ET

Agreed or disagreed?
by Philipp Gross

Dick, I doubt nothing of what you commented on and are actually well aware of the excellent insulation value of air and how insulation material makes use out of it. Why I gave the Lambda value at room temperature is because if we actually calculate with it, the air layer is close to the inside with lots of insulation material on the outside. This brings the "insulating" air closer to room temp than 0ºF and there is still heat transmission. But anyhow not really that important 0.023W/(mK) vs. 0.025W/(mK) the truth is somewhere in between.

Sorry for using the misleading word still... I should have put it in quotation marks. So do you agree with me that an air layer never reaches a higher R-value than 1?

R-values have a thickness that goes with them. R/in obviously takes care of that and with the common insulation materials that is no problem. Now with air or any other gas (as we both agree) that is huge deal. I think that saying "R ranges from around 5.7 to 6.3 for truly still air" is easy to misunderstand to many out there (not wrong).