|
You are not authorized to post a reply.
|
 Prev Next
|
| Author |
Messages |
|
|
|
wildblue
Registered Users
New Member
Posts:37
 |
| 06/16/2009 11:55 PM |
|
Interesting stuff but it’s all theoretical. Knowing this stuff is good but ultimately the specific situation should rule.
As far as the radiant barrier, solar gain in winter is negligible around here. In any event I’ll gladly trade what there is for more comfort in summer. Even if my cost savings for winter heating is a little less, we would much rather have the windows opened in summer than closed with the AC on. Additionally we have a very high efficiency dual fuel heating system. It’s not too expensive so I’ll be hanging it from the rafters.
I will install fiber glass insulation because cellulose burns and it absorbs water. It rains a bit here and there is much electrical wiring in my attic. Even if cellulose would perform a little better the few bucks a year are not worth the risk to me.
The bottom line is when I am done with all this I WILL be saving big on my current energy costs. |
|
|
|
|
Dana1
Registered Users
Advanced Member
Posts:702
|
| 06/17/2009 10:43 AM |
|
With a foot of fiberglass in the attic, the radiant barrier will be a total waste of money, and the effect on heat-gain will be next to nothing, which is what the ORNL table points out (they DO measure that stuff, and their assumptions are conservative.) RB only makes a noticeable difference if the insulation layer is thin to VERY thin (even in cooling dominated climates.) If it's worth it in AC power savings at R19, but not R30 insulation levels in Phoenix, with 10x the cooling season, it's surely not worth it by any measure in Seattle at R30+. See: http://www.ornl.gov/sci/roofs+walls/radiant/rb_tables.html#table4
The number in the table is the max you should pay per square foot to get a 25 year payback. Figure if you get a really good deal, and get it at $0.10/square foot, it's still not going to ever pay off, and the effect at R30+ will be academic (measurable with instrumentation, but not a noticable difference in comfort level.)
Canadian (and other) studies have shown that cellulose in the attic improves it's fire-resistance (it'll only light up with extended exposure to heat/flame), and fiberglass REDUCES fire resistance (since it passes some IR and allows superheated gas convection), but if you're not comfortable with it I'm not gonna push you on it. (Roof leaks can be somewhat messier to deal with using cellulose.)
But I WILL push you on this:
Rather than installing the radiant barrier, take that money and put it into another inch or three of fiber insulation (take it to 13-15" instead of sticking to the 12").
The net reduction in heat flux through the ceiling during the summer will be similar (or better), and it'll improve the wintertime performance as well. It's simply money better spent to achieve a comparable-to-better summertime result, with a more favorable annual result. Whatever the "...not very expensive..." cost of RB you come up with, just add it to the fiberglass installation. Unlike RB, the payback will be there.
|
|
|
|
|
Dana1
Registered Users
Advanced Member
Posts:702
|
| 06/17/2009 11:05 AM |
|
FYI: With 3-4" of FG you're at ~R11, in which case the (very conservative) ORNL cost-effectiveness insulation value estimator recommends bumping it up another R30 to final of R41:
http://www.ornl.gov/~roofs/Zip/tmp/results21497.html
Try it yourself: http://www.ornl.gov/~roofs/Zip/ZipHome.html
(insert your actual fuel & fuel costs & zip. I used 980__ as the first 3 digits of the zip for climate correlation.) |
|
|
|
|
wildblue
Registered Users
New Member
Posts:37
|
| 06/18/2009 4:56 AM |
|
| Thanks. The money is not a problem. I'm interested in comfort not pay back. It's not like I'm going to remove all the widows in the house. The difference with or without RB will be minimal. There is already 3-4 inches of fiber glass so the added 12 will make 15 to 16 inches.
There is no way to tell if RB will be effective or not in this particular situation. Studies are averages and bell shaped curves. I believe this situation is an outlier. However once I blow it the insulation I CAN'T go back in and add the RB.
Additionally, cellulose is nothing but treated recycled news paper. Implying fiberglass is not more fire safe makes me go hmmmm.
Nuff said on this thread... |
|
|
|
|
Dana1
Registered Users
Advanced Member
Posts:702
|
| 06/18/2009 9:18 AM |
|
Posted By wildblue on 06/18/2009 4:56 AM
Thanks. The money is not a problem. I'm interested in comfort not pay back. It's not like I'm going to remove all the widows in the house. The difference with or without RB will be minimal. There is already 3-4 inches of fiber glass so the added 12 will make 15 to 16 inches.
There is no way to tell if RB will be effective or not in this particular situation. Studies are averages and bell shaped curves. I believe this situation is an outlier. However once I blow it the insulation I CAN'T go back in and add the RB.
Additionally, cellulose is nothing but treated recycled news paper. Implying fiberglass is not more fire safe makes me go hmmmm.
Nuff said on this thread...
Actually there, is a way, and it's a done-deal: The Nat'l Labs folks who actually test & measure stuff (in the field as well as labs) and have pretty-good models say definitively: "NOT!!"
The effectiveness of RB diminishes with increasing R-value- right now you've got next to nuthin', and if all you were to do was add RB the effect would be substantial. But at ~R40 the biggest effect you would get would be be like going from R40 to R45, best case. Sure, the effective R45 would be a cut in heat gain from R40, but only a 9% difference (on an already75-80%% reduced heat gain.) Going from R10 (what you have now) to and effective R15 is more like a 33% difference.
There's a nearly encylopedic synopsis of the radiant barrier testing that was chased quite intensively in the '80s here:
http://www.fsec.ucf.edu/en/publications/pdf/FSEC-CR-211-88.pdf
DO read the section on the effect of attic insulation level (starting
on P21 of the PDF pagination). (Note as an asided, that in their
analysis they presume that without RB the surface of the FG is ~15F
above the attic ambient, based on reasearch data not cited. They don't address the differences in emissivity between fiberglass & cellulose, since most of the research was done on the then-more-common R19 fiberglass batting.)
Their conclusion (at the bottom of p25) is,
"All other things being equal, the percentage reduction in ceiling heat transfer for attic radiant barrier systems as compared to standard attics should remain constant regardless of ceiling insulation level."
So let's say your RB cuts the heat gain by 50% (way more effective than most installations- that would be an upper-bound). At your current ~R10 level that's a significant amount of heat rejection.
But at R45 (about where you'll end up with 15" of FG) you've already cut the gain by 78%, so adding the RB is going to cut the remaining 22% of gain in half, only 11% of the heat gain you were getting before the increase in insulation level.
Putting it in perspective (on page 26): "...50% of a large number may be substantial, whereas 50% of nothing is still nothing."
As the fiberglass gets ever deeper, the absolute amount of heat flux reduction is reduced, because it it only rejects a percentage function of what the insulation passes. That IS the way it works.
Broadly tested in the field by the ORNL, it correlates well with the FSEC synopsis. The ORNL tables could be denominated in BTUs/hr as well as dollars- it'll scale linearly:
http://www.ornl.gov/sci/roofs+walls/radiant/rb_tables.html#table4
The net difference in heat gain of adding RB to an R38 attic in Seattle is ~1/5 of what you'd get out of adding RB to an R11 attic. (And you'll be around R45.)
At R45 in the attic you'll be seeing far bigger gains from your glazing than heat flux through the attic. Going to low-e storm windows (or adding low-e window-films) will reject more heat than adding RB to the attic.
|
|
|
|
|
jonr
Registered Users
Basic Member
Posts:335
|
| 06/18/2009 8:59 PM |
|
All insulation suffers from diminishing returns on a $ saved per inch basis. And savings are even less than one would think (double the insulation is not 1/2 the heat loss) when one considers the entire building and other sources of heat loss.
Evidently radiant barriers being effective without an air gap is a myth that only people selling RBs promote.
Far better than blocking radiant heat in the attic is blocking it on the external surface. Aluminum shingles for example.
IMO, the best system for hot climates is Dupont AtticWrap (or similar systems with a sealed attic and ventilation under the roofing).
|
|
|
|
|
Dana1
Registered Users
Advanced Member
Posts:702
|
| 06/19/2009 10:18 AM |
|
Posted By jonr on 06/18/2009 8:59 PM
All insulation suffers from diminishing returns on a $ saved per inch basis. And savings are even less than one would think (double the insulation is not 1/2 the heat loss) when one considers the entire building and other sources of heat loss.
Evidently radiant barriers being effective without an air gap is a myth that only people selling RBs promote.
Far better than blocking radiant heat in the attic is blocking it on the external surface. Aluminum shingles for example.
IMO, the best system for hot climates is Dupont AtticWrap (or similar systems with a sealed attic and ventilation under the roofing).
What makes the diminishing returns curve of RB different from other insulation is that it's a fixed ratio of the total heat flux, independent of the total R value. (Whereas additional R-value's reduction is a ratio of the R-values.) If you plot it out it's a steeper asymptote to zero with additional R, but it has a fixed cost independent of R.
That makes it a great deal when the R-values are low since the same incremental cost of more insulation to achieve the same result is higher. But it's lousier deal than additional insulation when R-values are high, buying you only a fraction of the heat rejection that even more R-value would give, per dollar. The crossover point depends on your climate, since it's usefulness is far higher for cooling. The financial (not heat-rejection) crossover occurs at much lower R-values in heating dominated climates since the heating-season benefits of addtional R-value far exceed the heating-season benefits of RB. According to the ORNL calc, the crossover occurs below R40 in any climate (even Phoenix.)
It's true that rejection of the radiated heat at the exterior of the structure is indeed more effective (by shade or reflectivity) than between the roof & attic floor, also much-studied by FSEC & Texas academics. But even there it's relative value can be trumped by higher R-value. In most heating dominated climates the R-value is already well beyond where paying extra for a reflective roof is worthwhile, even in the cooling season. But this is another situation where the annual energy costs are
net-negative in heating dominated climates, since it also reduces attic
temps in winter, not just summer. (Even if it had ZERO addtional cost up front , it's not the right solution in Seattle.)
Even where reflective roofs make sense, SOME emissivity of the exterior finish is important for optimum performance. Very low-E surfaces like unfinished aluminum (with emissivites under 0.15) run at higher temps, since they can't re-radiate much heat, and their cooling relies on conduction to the roof decking (increasing the attic temp), and convecting heat to the exterior air, which isn't very efficient in low-turbulence low-wind conditions. See:
http://www.fsec.ucf.edu/en/publications/pdf/FSEC-CR-1411-03.pdf
http://www.fsec.ucf.edu/en/publications/html/FSEC-CR-1220-00-es/roofing.pdf
With a surface emissivity of 0.38, Galvalume is good stuff! I'd be curious to see how it's emissivity & performance changes over 10 years though- grit, algae, anything on the surface will likely raise it's emissivity considerably over time (just as dust on RB changes it's characteristics.) Whether 0.38 represents the optimal emissivity/reflectivity balance point for roofing in hot climates seems an open question.
|
|
|
|
|
jonr
Registered Users
Basic Member
Posts:335
|
| 06/19/2009 3:22 PM |
|
Yes, aluminum should be anodized for use as a reflector.
I believe white paint will both preserve galvalume and increase weathered performance.
|
|
|
|
|
|
| You are not authorized to post a reply. |
|
|
|
ActiveForums 3.6
|
Professionals Serving
Your Location:
|