Wednesday, May 9, 2018

Declaring Thin Insulation Batts - Defective

Unfaced R15 insulation recommended by my professional supplier as most cost-effective is:

Owens Corning R15 23x93 "Unfaced" Batts 103.98 sf per bag  See that despite forcing a Google search result of unfaced batts, only big-box store offerings show, mainly with kraft facing. A push of kraft facing upon consumers, is adverse. I can buy unfaced batts at Home Depot as a special order, minimum eight bags on a pallet, store arrival in 7 to 10 days, bag price $73.82.

At my professional supplier, I bought a 4-bag pod, at $77.97 per bag, ahead of a declared 9% price increase on 5/1/2018. Imminently then, the price to a professional installer of OC R15 UNF, is 1.09*$77.97/104 sf =  $817.34 per 1000 sq ft.

I brought the four bags of OC R15 UNF to my job site, a detached garage of a new home.
Determined to be trusting, I pushed all four bags up this very excellent Fakro LWP 22x54  newly-installed attic ladder. A 20.5" clearance is tight for the large bags, but there is nothing to snag a bag, with intelligent springs-on-door durable ladder hardware.

I have reinforced the 10' 2x4 roof truss bottom elements of a 20' wide two-car garage, by stitching on 6" rips of 5/8" or 1/2" CDX plywood, then capped by stitched-on new 2x4s. Apply American Wood Council Span Calculator to judge that 2x4 spans longer than 5' 9" can't otherwise bear storage loads. Truss bottom elements strengthened to 2x8 can carry prescribed 40 psf design live loads with span up to 11' 11".

I want to insulate the ceiling to R15, which one expects to be 3.5" thick. In fact, thickness must be 3.5" to have that value. I am beyond disappointment, although I am not surprised. I have known that unfaced R15 fiberglass batts from Johns Manville and Owens Corning, if not from all manufacturers, are dependably thinner than promised. The same is true of R21 batts not measuring up to 5.5" thickness (2x6), and R30 Cathedral batts not measuring up to 7.25" (2x8). High Density R15, R21 and R30c are premium materials, more expensive by more than ratio of density, than ordinary batts.

Building Science Corporation: Thermal Metric Summary Report (June 2015 Update) , reports at page iii:   When walls are constructed with the same installed R-value in the stud space, and are air sealed both inside and outside (i.e. there is effectively zero air leakage through the assembly), they  exhibit essentially the same thermal performance regardless of the type of insulation material used.
This, to me, is definitive proof from elaborate experiments, that no material has magic, to have declared R-value, even if it does not fill the space. Conversely this is credited by makers of loose-fill insulation, that uniform added and sustained fluff, makes added insulation from thin air. If makers of insulation, both as batts and as loose-fill, would defend the magic in thin batts, they can not have it both ways.

Rejecting magic then, I pulled up the placed batts, some hacked around lighting junction boxes, and returned them with bags not opened, to be credited back to me, and to be returned to the manufacturer, as defective, not to be passed on to a gullible other buyer. I am probably the first buyer to do this, and I hope I will not be the last. I imagine though that my distributor will not follow my instructions, except for the opened bag. I am compelled then, to be a constructive bother, in other ways. There will be first, this blog post that is within my control.

I have an associated action in progress at Insulation Institute Blog of NAIMA, North American Insulation Manufacturers Association . I perhaps abuse a post of April 5, 2018, in posted replies, accusing product design and quality as not-mentioned elements of achieving RESNET Grade 1 standing. Other readers in the few regions where training is offered, only care about the free instruction for their workers.
Free Grade I Training Coming to Four States 
I think I do not need to experience the training, to know its content. In service to RESNET, Insulation Institute has published a handout checklist guide: GRADE I INSULATION INSTALLATION . Nothing in the checklist tests product quality in the matter of reliably filling the space promised. RESNET Grade 1 is not defined in the Checklist. RESNET Grade 1 is not defined anywhere at the RESNET web site . RESNET is in the business of a Home Energy Rating System, HERS.

NAIMA then, steps in to do the quality description in words and illustration focused entirely on RESNET, despite a confusion of rival quality rating-by-number schemes. I think this is the most-complete expression of NAIMA teaching:


Advanced Energy is a non-profit energy consulting group in North Carolina. Advanced Energy separately publishes its' advice as:
prepplusinsulation.pdf, SUCCESS WITH HOME ENERGY UPGRADES What every contractor needs to know

NAIMA publications including those by Advanced Energy, take care to not conflict with a thin-batts problem  in manufacturing, that defies achieving claimed insulation value. Nothing may compel an installer to fully fill insulation space, to achieve best rating. Here is deliberately-foggy instruction, in words:

Summary of RESNET Grading Criteria:
Grade I – Minor Defects

Occasional very small gaps are allowed. Voids can’t extend from the interior to the exterior (i.e. the full width of a wall cavity). The product is installed according to manufacturer’s specification and cut to fit around electrical junction boxes and is split around wires and pipes. Compression or incomplete fill amounting to 2% or less, if the empty spaces are less than 30% of the intended fill thickness, are acceptable for Grade I.

If a batt fills the insulation space from side to side and is on-the-bag called the correct R-value, this is good enough for Grade I.


Know that the installer has carefully avoided pushing batts to contact with the exterior sheathing, to then unavoidably, very difficult of retraction, lack Ithe appearance of a full fill.

I think the NAIMA/RESNET Grade I definition in words, allows under-fill on thickness by up to 30%. Maybe Grade I, not Grade 1, is abbreviation of Grade Incomplete. Grade II is really-incomplete.

The full-appearance-only is quite difficult with R13 batts in 2x4 framing, where batts are never more than 3" thick. Watch the technique, in this NAIMA training video .

Consequences of thin batts are especially awful, where batts are fully out of contact with exterior sheathing, then not stilling the convection of air where temperature gradients are most steep. Find proof of this consequence in  training videos produced by Owens Corning, one showing installation techniques good and bad, and another reviewing installation details alongside infrared scans upon covering drywall.

Part 1, Installations 

Part 2, Include Infrared Scans 

These are R13 kraft-faced batts about 3" thick yet bag-labeled as code-compliant in 2x4 framing.The center joist cavity in the photo above purports to have a thin batt. It is "thinner" than other batts only because it is "compressed," gently pushed into good contact with exterior sheathing, then with kraft facing nominally 1/2" out from the drywall.

Here are the filled cavities including that on the right where there are voids outboard, challenged by parting and sloppy fit around a badly-foamed outlet from some long-ago other demonstration and around romex wires.

The "thin" batt has warmest drywall surface temperature. Carelessly placed batts left and right, especially that at left under a window, have lots of cold green and blue infrared indication. The outlet badly bleeds heat despite the foam, where evidently the foam has no penetration behind the outlet wall box. Acceptable insulation has generally-pink or red infrared indication on interior drywall, governed much by indoor and outdoor air temperatures. None of these three cavities is filled to Grade 1. Actual insulation value can only be known with full fill.

(An ice cube in a glass of hot water, I imagine.)

Know that the consensus quality labeling attributed to RESNET is in Arabic , not Roman numerals.

Should I be concerned that I am threatening to fiberglass insulation manufacturers and their lobbyists, NAIMA? If so, this is not the first instance published. At this thought, I have added Label: Thin Insulation Batts in this blog. Find associated writing at that Label, and at Label: Building Codes . I have been engaged in this for at least ten years, and addressed incomplete insulation fill of walls in a proposal for the 2018 revision of the International Energy Conservation Code.

My so-far-best summary of the 2018 building codes quest is in this post:
Math Assessment Of Poorly-Placed Insulation . Copied from that post is this statement of the proposal to deal with thin batts:

My proposal 11881 offered:
R402.2.1 Over-fill of insulation. Wherever possible, insulation shall be confined on all six sides. Insulation fill shall be complete and without voids that could allow air bypass of the insulation. Overfill of insulation shall be required to achieve complete insulation fill. In addition to installation methods that prevent voids within the insulation, the methods shall prevent the presence of air spaces between a barrier film such as the kraft facing of a batt or the containment mesh for blown cellulose, and hard wall coverings such as drywall.

I was an inexperienced individual, investing about $5,000 in preparation and attendance at hearings in the Louisville, KY Convention Center. I stood up for ten proposals and in each, in less than five minutes, accepted unanimous disapproval. Things move so fast, that each Judge surely has somehow noted his vote in advance, I accuse, with small regard for technical merit. If there is weight in standing up to defend or oppose, it is with those honored for repeat engagement, the lobbyists. My sure opponent each time was the National Association Of Home Builders, NAHB . NAHB guards the short-term interests of major home builders, to pocket as much as possible in each home sale, not much caring about interests of home buyers. For the Overfill of insulation proposal, the overwhelming single opponent, was NAIMA.

I have remained engaged with NAIMA, to no effect. I have been willing to try again in Louisville, but know it will be futile if NAIMA has not seen that the proposal is in fact good for its represented manufacturers. How could it be harmful to manufacturers if better code allows them to sell more insulation, I figured. Where in this conversation I state my observation that placement technique for photos and instructional videos, thin batts are delicately placed proud of joist front faces, for appearance of fullness, I am threatened: I must not question the honor and good intentions of NAIMA. Else, I will be shut out of conversation. I have been in defiance of this threat for more than a year, still thinking I might have NAIMA as an ally in Louisville. This post is perhaps the ending of that hope.

Get on then to what I have done on a job, after returning Owens Corning batts R15 unfaced, as defective.

Here is Rockwool R15 in nominal 22 1/2" space. No fussing needed to have uniform 3 1/2" thickness.

There is no buckling in accommodating 5/8" reduction of available space where I have used plywood webbing in growing the truss bottom elements to load-bearing 7" thickness. In a five-star review at the Rockwool web site, I credit edge compression without buckling,  to material friability . Crushing without buckling is virtue, but in trade, the batts are subject to falling apart when carried. The densely-packed insulation bags could not be pushed through 24" oc framing reduced to 20.5" clearance in the attic ladder frame. I carried insulation up one batt at a time, with occasional fragmenting in my grasp.

I am quite happy with 3 1/2" fill of R15 Rockwool batts (always unfaced). The honest-value Rockwool batts cost less too. Unit cost is $39/60.1*1000 = $649 per 1000 sf. Compare to an unsupportable, price-fixed $817 per 1000 sf cost of Owens Corning R15 unfaced fiberglass batts (see above).

Addition to this post will detail layup of batt insulation in layers, rather than thinking batts are readily parted to sandwich wall plumbing and wiring. Knowing that underfill shall not be allowed, unfaced batts about R11 shall be readily available, for sandwiching, and for fill-in at any low spot at an internal or outer layer. Layering/ fill-in batts must be visibly on offer for novice installers at big box stores too. Where fill-in might slow a pro, batts thinner than promised might remain unacceptable.

I will have may examples of layering with Rockwool batts. Here 2x10@16" framing is extended to 2x14. All wiring is laid atop the 2x10s, and is revealed by lifting plywood and top-layer R15. A trampled 12" depth of loose-fill fiberglass, with many voids and bare spots, was effectively not better than R10. Loose-fill has been redistributed to 4" depth now under R23 Rockwool. The serviceable new total of insulation is nearly R49.

Unhappily, I have buried can lights, with much effort in the placement of Rockwool batts. I have several thicknesses of Rockwool AFB batts useful in this, especially 1 1/2" thickness.At my professional insulation supplier, IDI, I may choose between Rockwool, and Owens Corning Thermafiber. I stick with Rockwool because there are fewer thickness choices with Thermafiber.