Rock wool insulation advice?
I’m building a house for my wife and myself and I’m about to start insulation using mineral wool. Exterior walls are 2 X 4 studs w/ tyvek house wrap, 2″ polyiso, 1 X 4 vertical firring strip and d 1 X 8 T&G clear cedar installed horizontally. Roof is 1/2″ on 12″ epdm with 4 1/2″ polyiso on 2/4″ plywood over 2″ X 12″ ceiling joists/rafters. I’m using R30 7″ Roxual bats in ceiling cavity. My question is should I hold the bats tight to the roof sheathing or down tight to sheetrock? Walls are getting R15. we are in zone 6. Also, another question: what is the best paint and primer to use on sheetrock for breathability (walls are breathing to the inside). Thanks, Ben
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
By the way 3/4" plywood, not 2/4". That was a typo.
Ben,
With 4.5 inches of polyiso the batts should be tight to the roof sheathing because there should be no risk of condensation there. That is, assuming that the roof plane is airtight - taped sheathing, (then felt?), taped polyiso. Was the polyiso one layer? If not, were the seams staggered and taped as well?
Why not fill the entire ceiling joist bay with Roxul? Or fill that extra space with fiberglass insulation. It's insulation space that you need to take advantage of especially in zone 6. Now is the best, least expensive time to do it.
I think Dana and Martin always talk about latex paint on the sheetrock.
Hope that helps.
I didn't tape plywood but i used tyvek on roof sheething and thay's taped, polyiso is 1 layer of 4" 1 layer of 1/2" taped and caulked then 1/4" dense deck then epdm. epdm is wrapped around curb and caulked. I was told i should not fill entire cavity with insulation because of dew point. I'm zone 6 but on border of 5. thanks L
ucy.
location is just above kingston n.y.
Ben,
Ordinary latex paint is fine on the interior side of your drywall.
Building codes require (and building scientists recommend) that the air-permeable insulation in the type of roof assembly that you describe be installed in direct contact with the underside of the roof sheathing. For more information, see How to Install Rigid Foam On Top of Roof Sheathing.
thanks martin, seems like i used wrong insulation for my zone and i'm stuck with it. i think i'm going to stay with plan of r 30 roxul in ceiling/ and r 15 in walls unless anybody has better idea,thanks
In zone 6 you need the exterior foam layer to be at least 50% of the total roof-R for dew point control. The fiber insulation has to be tight to the roof deck, in part to avoid convective by passing of moisture laden air directly to the roof deck. The IRC Chapter 8 prescriptive for R25 exterior foam is based on a total center-cavity value of R49. R25/R49= 51%
http://publicecodes.cyberregs.com/icod/irc/2012/icod_irc_2012_8_sec006.htm
With 4.5" even at the labeled R6/inch that's R27.6, with R30 under the roof deck you'd only have a ratio of 48%. But polyiso loses performance when the average temp through the foam layer is below 55F or so falling even more rapidly when the average temp through the foam falls below 40F.
If you de-rate the foam to R5/inch (optimistic, in that climate and stack-up) that 4.5" is really only R22.5 during the winter months a ratio of only 43%, well below the amount necessary to be protective. During colder years the average winter performance may even fall under R20, which would make the exterior R less than 40% of the total.
You would be able to get away with R30 rock wool in the cavities at the warm edge of zone 6 if you install an air-tight smart vapor retarder on the interior side as cheap insurance- after all, in zone 5 a ratio of 41% ratio works. Some box stores in MA are starting to carry Certainteed MemBrain- not sure about your neighborhood, but if you can't find it through local distribution channels you can get MemBrain or Intello Plus via internet sales.
It might make even more sense to use a full cavity fill of cellulose blown in netting, in combination with an interior side smart vapor retarder, which would guarantee no large air-currents could bypass the fiber insulation or smart vapor retarder with only a partially filled cavity. Cellulose would also wick and redistribute any wintertime moisture accumulations without losing function, protecting the structural wood.
The wall stackup is fine. Even derated to R5/inch the 2" polyiso would be R10, and for walls you'd only need ~35% of the total R to be exterior for dew point control. (The average wintertime temperatures of siding is higher than that of roofing, since it's not radiating heat to the night sky.) You'd be fine even at R7.5, which is the IRC prescriptive for 2x4 framing in zone 6:
http://publicecodes.cyberregs.com/icod/irc/2012/icod_irc_2012_7_sec002_par025.htm
Disclaimers: I'm an amateur not a pro. And I love Roxul and am no big fan of foam inside the building envelope.
First, who is installing the insulation? If you are hiring a pro, perhaps cellulose is a good plan.
But if you are doing it yourself, speaking as someone who has installed the common insulation types as an amateur, you'll get a lot better install with Roxul. In a 2x12 joist space you can fit two R23 Batts or an R15 and R30. I like two R23s for installation ease (they stay up on their own; R15 doesn't).
Are you strapping down for an utility cavity or just burying the wire in the insulation? You should be aware you can't pull new wire in Roxul. The other reason I ask is whether you will be able to have a penetration less air barrier or whether you'll have a bunch of difficult to detail electrical penetrations.
I can't really speak to whether your assembly will be vapor safe but if you have your doubts you could have it modeled in WUFI.
Keith: Neither an R15 plus an R30, or a pair of R23s would snugly fill an 11.25" nominal depth of a 2 x 12 rafter. It needs to be a compression fit to limit the potential for bypass gaps. The R15 + R30 is only 10.75" (a half-inch shy) and a pair of R23 is only 11" (a quarter-inch shy.)
I'm i little confused, if i fill rafters with 11' of roxul won't that be making the dew point problem worse by increasing the % inside? better R but still the dew point prob. thanks, ben
Yes, more fiber-R requires more foam-R for dew point control.
Using cellulose as a full cavity fill at 11-1/4" would be about R40-R42 in the cavity, which is modestly better than the R45-ish rock wool cavity fill options (that don't quite actually fill it up), but would be more protective than rock wool due to it's much higher moisture buffering capacity.
An alternative would be to install 2" of closed cell spray polyurethane directly on the underside of the roof deck, reducing the cavity depth to 9.25". The foam would run R12 and ~0.5-0.7 perms (sufficiently vapor-tight to protect the roof deck on it's ow and) the interior side of the spray foam becomes the first condensing surface. If you figure the polyiso is good for R22-ish and the spray foam is good for R12, that means you can safely install up to (R22+ R12= ) R35 in the remaining 9.25". An R38-C high density fiberglass batt manufactured at 10.25" nominal loft will perform at R35 when compressed to 9.25", which is probably your safest-best choice here rather than jamming not quite a full fill of rock wool and relying on smart vapor retarders.
For compressed batt performance see:
https://www.greenbuildingadvisor.com/sites/default/files/Compressing%20fiberglass.JPG
As much as I hate to use closed cell polyurethane for it's disproportionate environmental impact, this is the safest-best way forward short of ripping up the roof and adding more exterior foam. The closed cell foam is going to cost about two bucks per square foot, but it will reliably protect the roof deck, and provide just enough R outside the condensing surface to keep the fiber insulation layer from accumulating moisture.
While an R70-75ish unvented roof may seem excessive, that's about what it takes to design for Net Zero Energy with a solar array that fits on the house in a zone 6 climate.
thanks dana, i really don't like the idea of using spray foam inside and my wife would divorce me anyway, i thought rock wool was excellent with moisture and was hoping r 30 roxul would squeak by but i guess polyiso really is crap in cold climate. So if i do my original idea what is going to happen in your opinions and any more ideas? thanks a lot everyone. ben
Spray foam is a divorceable offense?
only if it's sprayed inside
i know it's not ideal but what if i use 5.5' of rockwool in the rafters at R 21 i think?
Ben,
I vote for R-30 mineral wool, tight to the sheathing, along with a smart vapor retarder like MemBrain.
Ben, I sympathize with your domestic treaty regarding spray foam. Many of my solutions are oriented toward domestic detente under a similar treaty.
As said before I'm not a pro so I'm not at all dismissing Danas expertise but ... I've retrofitted 2 R23 roxul batts into 2x12 roof. Perhaps my batch of R23 was just a little thick but it was a brilliant fit. The intello had just a little overpressure from the roxul. You might not do as well with R30/R15 As the stack up is a little thinner.
Again my original comment was not to denigrate cellulose and its hygroscopic properties but to say a DIY install might go better with roxul batts.
If the 2" ccSPF is a complete non-starter, Martin's recommendation for R30 + air-tight smart vapor retarder almost meets code-min for labeled R value (probably makes for real on U-factor) , and would not be very risky.
Rock wool isn't particularly great for moisture, compared to other fiber insulation. It'll wick some, but has very little capacity for storing moisture. The microscopic structure of cellulose fibers is hollow tubes, which can take on more than 2x the amount of moisture as adsorb than other fiber insulation before liquid forms, and without damage or losing function.
Keith: The 2x R23 is still a bit THIN- in terms of actually filling up the space it's not quite there. It's only a bit thick in terms of optimal R-ratio.
thanks again for your time, just a little unsure of a membrain type product trapping moisture in cavity and not letting it dry to the inside.
Ben,
MemBrain is a brand name for a so-called "smart" retarder with variable vapor permeance. It will allow the assembly to dry to the interior when necessary.
I'm just not so confident in" smart" products, sometimes they don't think so well.
MemBrain is 2-mil nylon, which has decades of history in food-wrap applications. It's moisture regulating properties are long lived and well understood. It simply cannot trap moisture, but will limit the rated moisture transfer quite well if the air next to it is sufficiently dry. So if the cavity becomes humid for any reason, the moisture gets out fairly freely (the limiting factor will be the latex paint on the gypsum.) But in the winter the cool roof deck adsorbs the moisture in the cavity, making the proximate air quite dry.
I've heard a good bread knife is the best way to cut roxul. I think I'm going with R30 roxul in the roof cavity and possibly membrain or similar product. Really appreciate you guys' help. Any other suggestions and ideas would be greatly appreciated. Thanks, Ben
For batts as thick as R30 you might want to spring for a saw-handle type insulation knife. Most bread knives will work OK for up to R23s, but for batts thicker than 5.5" even a 12" bread knive isn't really all that great. Bread knives are great for sculpting out around electrical outlets, etc, but for making clean even cuts on thick batts, there's nothing quite like the purpose-made tools.
http://www.finehomebuilding.com/toolguide/product-finder/cepco-tool-insul-knife-insulation-knife.aspx
https://www.youtube.com/watch?v=b_jRFDBpthk
http://toolmonger.com/2007/07/13/dealmonger-cepcos-insulation-knife-for-25/
thanks dana, that looks great
Ben, your local Lowes may sell Roxul and an inexpensive (<$10) insulation knife that looks like a bread knife (mine does). Yes, a cheap bread knife will work but I personally found the insulating knife well worth it. If you have a lot of odd size cutting to do, you may want to make a piece of plywood with a 1" grid on it to use as a cut table. Unnecessary if you nice clean layouts but essential if you have to rip batts to 19.25" +/'1".
thanks keith, good idea
Ben,
You're right -- mineral wool insulation is surprisingly fragile. Do your best. The traditional way to hold up batts in a case like this is with insulation support wires. Here are some links:
http://www.homedepot.com/p/Simpson-Strong-Tie-16-in-Insulation-Support-100-Qty-IS16-R100/100375163
http://www.lowes.com/pd_12324-41075-12324_0__?productId=1007637
so i installed some r 30 today, stuff is petty fragile and tough to get tight to roof deck without compressing the sides to much. do i over compress then try to pull back out or just leave over compressed ? also should i use wires to hold up ? ( 2x12 rafters 8'' rockwool ) seems to hold fine but wonder in 25 years it still will. and i just assume stamped side out? thanks
when i go to roxul's site for installing in ceiling it says for flat and low pitch roofs allow at least 2'' gap between top of insulation an roof sheathing yet everyone says hold tight to sheathing and leave the 4'' space i have with 2x12 rafters between bottom of insulation and sheetrock, pushing tight to roof sheathing makes more sense but please confirm i'm not making a mistake, thanks
@Ben
IIRC Roxul is assuming that the roof is vented so the 2" gap would allow air to flow from the soffits under the roof decking and out the ridge cap at the apex of the roof.
BTW remember the Roxul batts are considered to be "friction fit" so it's expected to have some compression along at the sides of the batts as they're fitted into place. Without the compression/friction the batts would fall out like fiberglass.
THANKS CHRIS, ACTUALLY I MEANT COMPRESSED UP AT EDGES TO GET TIGHT TO ROOF SHEATHING, IT TAKES SOME PUSHING OR IT WON'T CONTACT
Ben
Top gap: you likely can't get 'pressed' against the roof deck due to roofing nails of excess length etc. While a large air gap could create convective looping, I personally wouldn't lose any sleep over <0.5". I'd put all your air space on warm side. Plus it makes room for wiring etc.
Friction fit: I have two tests pieces looking at friction fit in my home. One is in an trussed floor R-23, no wire, two years, no movement. The other is in a garage cathedral ceiling, R-23, no wire, two years, no movement. So I don't worry about it. But I still use wires. BTW the 2x4 stuff doesn't stay and is a hassle to install overhead.
Compression: I find tip in works much better than straight in. Hold at 45 angle, set the top corner, compress the bottom edge a little to get it started. Another trick is to run that bread knife (assuming no wiring) or a big putty knife along the joist or truss edge to help slide if it is catching on a plate or something.
Are you custom ripping or are your joists stock 24" o.c.?
Last don't worry about a little bit of compression or flaking. It will still work.
Hope my DIYer experience helps.
thanks keith, that's what i did but also added wire just in case