Originally Posted by Square Eye
Part of the reason I'm confused is that, the radiant barrier type of bubble wrap used here, has foil on both sides.
Radiant Barrier (bubble wrap) does not have an R-value because it's intended purpose is to slow radiant heat and is not intended to reduce conducted heat. Although it may actually have a low R-value, R-values apply to conducted heat and would be misleading when comparing radiant barriers. The air gap that is mentioned in your link is intended to separate the the radiant barrier from the regular insulation. The radiant barrier is still a vapor barrier, so it needs to be directly against the metal. Then an air gap, then the fiberglass insulation or foam panels can be used...
Actually the bubble wrap does
have an R-Value. It's the radiant barrier by itself that doesn't. R Values are tested using what is called a Hot Box Test. Basically, an insulating material is put in the hot box with a heat lamp above it, and the time it takes to resist the heat is quantified as the R value. A plain radiant barrier (without bubble insulation) is an R 0, because, even though it blocks 97% of the heat, the 3% gets through instantly. That's why radiant barrier companies came out with adding insulating material. The bubble material in 5/32", for example, is about an R 8. Then foil is wrapped around it and it's put in the Hot Box. The hot box registers it as an R 14.9, because the bubble insulation was only under 3% of the heat load. Thus, the radiant barrier bubble insulation has a certified 14.9 R-Value in 5/32". This R-Value per dimentional thickness is even better than aviation grade foam.
To put this under a roof and over the decking, you could use this as a vapor or moisture barrier instead of felt or tar paper. Then, if you wanted to, you could stop there and add the metal roof with about an R 8. However, if you just add furring strips before putting on the metal roof, you'll also have all the benifits of the radiant barrier.
I hope this helps.http://www.radiantbarrier.com/tempsh...le_double_.htm