The usefulness of a radiant barrier sandwich – an exercise in futility?
I’ve been throwing around a few ideas, mostly to myself, that I wanted to open up for discussion. One of them has to do with radiant barriers, and more importantly, a double radiant barrier.
For example, suppose you built a cathedral ceiling, with the layers as follows, inside to outside. Asking for a friend, it completely doesn’t already exist this way on a project in my back yard.
1/2″ drywall, 2×6 rafters, R21 fluffy, 19/32 OSB (taped seams), #15 felt paper, 1.5″ foil faced polyiso (foil facing out, taped seams), 2×4’s on the flat, 7/16 LP TechSheild (taped seams), #15 felt paper, arch. shingles.
In climate zone 4A, or mixed humid, my thoughts were that having a single radiant barrier is good, so two must be better, eh? the foil facing on the TechShield guards against some of the IR transmission from the (160 F+) shingles in the summer, and the foil face on the polyiso reflects another portion of that back up into the TechShield, (also making for one of those optical Fresnel nightmare problems), and so on. In the winter, any IR loss from the outer surface of the polyiso will (should, might) be mitigated by the low(?) emissivity of the foil face on the polyiso.
In the worst case, it seems that it’s only a small waste of money, in the difference in cost between regular 7/16″ OSB and the TechShield. In the best case it helps to any degree, and since the building is small, and mostly unconditioned, the extra $50 will pay for itself in the long run.
Anyone care to offer up a thought? It’s not a permanently habitable living space, and the effective R value seems to be around R27. It’s not code in this area, but it’s probably not far off by U value.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
The polyiso is probably enough. My view has always been that the primary benefit to radiant barriers is to help with solar gain in the summer, so "facing out" is the way to go. Inward facing radiant barriers, when setup correctly, are usually said to perform somewhere around R1 or so, which isn't very much. With the solar gain issue, there is a lot more to gain, so it makes more sense to face the radiant barrier out, especially in cooling-dominated climate zones.
I wouldn't bother with the second radiant barrier. My usual sneaky way to add a bit of radiant barrier is to build insulation baffles with foil faced polyiso, a sort of win-win here. I've never bothered to add a radiant barrier solely for the purpose of adding a radiant barrier, although there are situations where doing so can be a benefit. I don't think the effort/cost of more than one radiant barrier is really justified though.
Bill
There are a lot of misconceptions about how radiant barriers work. A lot of this seems to arise from false conclusions drawn from the association between the fact they work well for combating heat from the summer sun hitting a roof, and that fact that that summer sun arrives at the roof in the form of radiation. Those are both true, but there's no causal relationship. A radiant barrier on the underside of your roof deck retards heat that is coming through that deck equally well whether that heat originated from solar radiation, a hot wind, or from a roofing crew spreading hot tar on the roof.
When the sun hits the roof, the radiation all gets absorbed or reflected. None of it goes through the roof. What gets absorbed can only proceed through the decking by conduction. No thermal or solar radiation goes through plywood or OSB. After it has conducted through, if there's an air gap, it has to jump that air gap to proceed further and heat your building. It can jump that gap by radiating off the bottom of the deck, or by heating the air in the gap (covnection). If the bottom of the deck has a low-e coating (e.g. foil), that cuts the radiation from that surface to only 5% of what it would have been. That's pretty negligible, so the heat has to cross by convection. Convection doesn't work as well to move heat down as it does to move heat up, and if the gap is vented, a bunch of that heat gets swept out rather than proceeding across that gap. So the assembly can work pretty well: heat has a hard time getting across the gap.
Putting the foil on the top of the layer under the gap theoretically works just as well. The radiation that comes off the decking goes across the gap, but doesn't get absorbed--it gets reflected back. In practice, the upward facing foil doesn't work as well longer term, because it gets dirty faster. The dirt absorbs the radiation rather than reflecting it, and the foil does nothing to stop the heat conducting from the dirt into the layers below.
Note that which surface you want low-e does not depend on the direction of heat flow you want to retard. The only reason it matters for performance is where the dirt accumulates. In windows, where it's sealed inside a cavity and kept clean regardless of whether surface 2 or 3 is coated, you get the same U-factor with the coating on surface 2 or 3. The only reason it sometimes matters which you coat is when solar heat gain comes in, because the coatings can do different things for solar radiation wavelengths. But in a roof, the materials are 100% opaque to solar radiation, so there's no such effect.
With some of those misconceptions out of the way, what happens when you have foil on more than one surface? You can reduce the radiation heat transfer a little bit more. If both are shiny foil with emissivity of 0.05, the combination has an effective emissivity of 0.025, and that cuts the tiny bit of remaining radiation heat transfer in half. But even with 0.05, most of the heat transfer across that gap is by convection. That factor of two in radiation is negligible in the total heat transfer amount.
But there's no harm in having two foil layers, and if the one facing up was the one you got for free, adding the one on the underside of the decking is worthwhile, since the one facing up will soon be dirty.