Posted By ICFHybrid on 15 Apr 2011 12:45 PM
Crummy R19 batts...
ROTFL
Am I detecting some functional bias in Dana1's normally "'bot-like" exterior? :-0
Mayhaps- I've been known to refer to R19 batts as "junk" on this, and other public forums.
My take on R19 batts:
R19 batts are the same weight as R13 batts- it's exactly the same amount of material spun slightly differently for higher-loft/lower-density. (That's why they're price the same, right?
)
It get's better: They're only R19 at the spun thickness of 6.25". When installed per manufacturers instructions they're only R18 in a 5.5" deep (!), a 5% hit on the labeled performance the instant they're installed. See:
http://www.owenscorning.com/around/...nChart.xlsBut it doesn't stop there. The low density has consequences. The reduction in density when fluffed out to R19 makes them poor performers at high delta-T due to the relatively high rate of convection within the fiberglass layer itself. Installed perfectly they perform pretty much at spec in a 70F house when it's 45F outdoors, but well-under spec when it's 0F outdoors. High density R21 batts in the same cavity lose some effectiveness at high delta-T, but not nearly to the same extent.
Note, R22 batts actually underperform R21 batts and are only delivering R19 in the very 2x6 cavity they were designed for- see spreadsheet in the Owens Corning link above. What's with that? More low-density junk, with R-labeling bordering designed to deceive. R19 is a SUBSTANTIAL hit from the labeled R22- 16% lower performance than the label implies, even when installed perfectly to the manufacter's spec. (And would seriously underperform an R22 ICF or SIP, eh?) R21 batts deliver least the nominal center cavity performance as-labeled when installed as-intended. R22 is one better than R21, right? Riiiiiiight... (caveat emptor)
Cellulose at any standard density doesn't have the loss of R at high delta-T issue, nor does dense-packed (1.8lbs or higher) new-school fiberglass blowing wools such as Spider or Optima. These are also sufficiently air-retardent that minor faults in the air-barrier don't matter too much- orders of magnitude less air can move through the leak than with the low density R19 batt.
There are similar issues with low density batts with radiated heat transfer across the cavity. Being somewhat translucent to infra-red the temperature gradient across the batt isn't linear- the outer inch is soaking up heat from the hot side in a non-conducted manner that becomes apparent when the outer wall is hot. On a hot sun-drenched wall on a hot day it performs little better than if it were compressed down to the mid-density R13 batt it really wants to be. (Higher density fiber does much better here too.)
So, yeah, crummy 'tis.
But the more R there is in exterior foam, the lower the delta-T across the batt, and the less s
ome of my objections really matter. Flash & batt with R19 + 1" of closed cell foam in a 2x6 cavity can be a reasonable option, since it raises the density and K-value of the batt under compression, and the flash-foam air-seals the cavity pretty well. Compressed to 4.5" they yield an honest R16, to which you can add the R6 of the foam, for an R22 center-cavity R, but in a cavity thats more air tight than you'd get with R21-R22 batts. (The foam also protects the sheathing from interior condensation issues in cold climates, but that's another subject.) You get about the same theoretical moderate delta-T performance out of flash'n'fill with spray cellulose, but the spray cellulose has no gaps or compressions, doesn't lose R at higher delta-Ts, and adds about as much thermal mass as increasing the wallboard thickness by 1/4". It's sometimes cheaper, but even at a modest premium, going with wet spray cellulose is usually "worth it" over a compressed R19 batt flash'n'fiber approach to studwall cavity insulation.