Attic Insulation with a new roof

Users who are viewing this thread

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
No idea if anyone here is an attic insulation/ roofing expert... but I figure it's worth a shot :D

Ok, here's the current situation:

1920s house with a steep slate roof. The attic (3rd floor) is finished with knee walls. There is no roof venting. The knee walls are uninsulated. There is paper-backed batt insulation in the rafter bays directly under the decking. The attic floors are uninsulated. The ceiling is insulated with some sort of fiberglass.

Here's what's changing:

The roof has done its time and is going to be replaced with asphalt shingles. The plan is also to also cut in a ridge vent. What I am thinking is that it makes sense to remove the insulation from the rafter bays if we're attempting to vent the space, and to also add in shingle over intake vents.

I figure I'll then go back and insulate the kneewalls and the attic floors behind those walls.

Does this make sense?

A variety of other options I've seen:
-Remove the insulation, do not vent the roof, spray in foam insulation
-Do not remove the insulation, just cut in the ridge vent
-Do not remove the insulation, do not vent the roof

I should mention that this is in Pittsburgh, PA -- so we do see all temperatures and levels of humidity.

At the moment, I'm not overly concerned with energy efficiency, as I plan on leaving access to behind the kneewalls pretty open and can always make changes, but I am mostly concerned with the longevity of the roof and preserving the roof boards.

I greatly appreciate the assistance and please ask questions about the existing conditions if anything isn't clear!
 
Last edited:

Reach4

Well-Known Member
Messages
38,749
Reaction score
4,400
Points
113
Location
IL
I like the first option, presuming they do a good job. This is the most expensive, but it gives you the most space and seals against air leaks too.
 

hj

Master Plumber
Messages
33,599
Reaction score
1,037
Points
113
Location
Cave Creek, Arizona
Website
www.terrylove.com
ideally, when you insulate under a shingle roof, you want the insulation to have an air space between it and the roof for air circulation, which your existing insulation probably has already. IF you spray insulation directly under the shingles, heat will build up and probably destroy the roof fairly quickly. That is one of the reasons for venting the eaves AND the ridge vent.
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
ideally, when you insulate under a shingle roof, you want the insulation to have an air space between it and the roof for air circulation, which your existing insulation probably has already. IF you spray insulation directly under the shingles, heat will build up and probably destroy the roof fairly quickly. That is one of the reasons for venting the eaves AND the ridge vent.

The current insulation is right up against the decking. That said, a slate roof isn't anything close to airtight, no tar paper / ice-water shield, no sealing between the shingles, so I don't think it's been a huge problem as the decking, from what I can tell, seems to be in decent condition.

Are you voting for removing the existing insulation, venting eaves and ridges, and then insulating kneewalls and floors?

Thanks.
 

Jadnashua

Retired Defense Industry Engineer xxx
Messages
32,771
Reaction score
1,191
Points
113
Location
New England
If it's in the budget, you might consider something like an aluminum shake look alike roofing materials...they'll be the last ones you ever need to replace. One place in TX where they had been installed in 1927 at a golf course clubhouse had them tested somewhere in the 1980's (over 50-years later) had the lab say, at the current degradation, they'd last another 800+ years!

A co-worker installed some on his home part way up a mountain that sees lots of snow. The first day, he said it changed the feel of the whole house. Those have about 1/2" or so of air gap, and the aluminum acts like a great radiant barrier so the roof deck doesn't get hot, and home heat gets radiated back into the home. WIth those, I think you could use the spray foam and not worry about anything.

FWIW, slate can last centuries, but the fasteners may wear out before then!
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
If it's in the budget, you might consider something like an aluminum shake look alike roofing materials...they'll be the last ones you ever need to replace. One place in TX where they had been installed in 1927 at a golf course clubhouse had them tested somewhere in the 1980's (over 50-years later) had the lab say, at the current degradation, they'd last another 800+ years!

A co-worker installed some on his home part way up a mountain that sees lots of snow. The first day, he said it changed the feel of the whole house. Those have about 1/2" or so of air gap, and the aluminum acts like a great radiant barrier so the roof deck doesn't get hot, and home heat gets radiated back into the home. WIth those, I think you could use the spray foam and not worry about anything.

FWIW, slate can last centuries, but the fasteners may wear out before then!

Unfortunately, nope, not in the budget. A) It's a big roof B) Anything but asphalt or slate is very rare around here. The slate has lasted 90+ years which, as I understand it, is the anticipated lifetime of this particular variety. I'd love to replace it with new slate, but that's not going to happen.
 

hj

Master Plumber
Messages
33,599
Reaction score
1,037
Points
113
Location
Cave Creek, Arizona
Website
www.terrylove.com
quote; That said, a slate roof isn't anything close to airtight, no tar paper / ice-water shield, no sealing between the shingles, so I don't think it's been a huge problem as the decking, from what I can tell, seems to be in decent condition.

That may be, but it has absolutely NOTHING to do with the new roof which is going to be asphalt shingles.
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
quote; That said, a slate roof isn't anything close to airtight, no tar paper / ice-water shield, no sealing between the shingles, so I don't think it's been a huge problem as the decking, from what I can tell, seems to be in decent condition.

That may be, but it has absolutely NOTHING to do with the new roof which is going to be asphalt shingles.

I agree -- which is why I'm looking to remove the insulation from the rafter bays when the new roof goes on. I think the insulation in the bays was vaguely acceptable with the slate as air was moving through the roof pretty readily, thereby preserving the decking.
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
Slate roofs are inherently ventilated due to the air spaces between slate. Shingle roofs are not.

The difference in shingle life between vented & unvented roofs aren't worth even thinking about- the color makes a bigger difference in peak & average shingle temps and shingle life than vented/unvented, even in hot sunny places like FL or AZ. The PRIMARY purpose for venting the roof is to keep the roof deck sufficiently dry- the shingle cooling effects are negligible even in hot climates and even less in PA.

But in a PA climate an unvented roof assembly without exterior insulation will leave the roof deck below the dew point of the of the conditioned space air for most of the winter and it will accumulate quite a bit of moisture over the winter, leading to mold/rot risks. But you can't just put up sheet poly on the interior to block the interior moisture drives or you will create a moisture trap between the shingles & felt and the poly.

There are a couple of ways to deal with it in an unvented roof assembly: Put rigid insulation above the roof deck at the IRC prescriptive minimum R for your climate zone, then fill out the R with fiber between the rafters (which you already have). Or, you can temporarily pull the batts and air-seal & insulate the underside of the roof deck with 1-2" of closed cell spray polyurethane. Closed cell polyurethane is only semi-impermeable at about 0.6-1.2 perms (1.2perms @ 1"), compared to 6-mil poly at 0.05 perms. The cc foam is vapor tight enough to limit wintertime moisture accumulation to acceptible levels, but not so tight that the roof can't dry in a season. Spray foam that thin doesn't meet the letter of the IRC from an R-value point of view, but it works, and works well. (The IRC prescriptives don't factor in the low vapor permeance of the foam, only the air-permenance & R-value.)

The best approach is to spring for the exterior rigid foam and apply it over #15 felt, ( not #30), held in place with half-inch OSB (seams staggered with those of the foam for air-tightness), through-screwed to the rafters 24" o.c. with pancake head timber screws, then do a standard felt & shingle layup over that. EPS runs about 10cents/R per square foot, and doesn't have wintertime loss of performance the way polyiso does, nor does it lose performance over time the way XPS does. It takes 4" to beat R15 (needed the climate zone 4 eastern-PA region), 5" for R20 (for most of PA. Find yourself on this map to figure out what applies for you.

It's damned-near impossible to adequately air seal kneewall attic spaces- you'll get a lot better thermal performance (both winter and summer) and far fewer ice-damming issues if you do the foam-over approach on the roof deck. But if that's too expensive or awkward, the closed cell flash & batt between the rafters is still pretty good.
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
Dana strikes again :)

Thank you for the insight into the complete reason behind roof ventilation and why it's now a concern once the slate is removed. I like the EPS foam option, namely that it keeps me out of the crawlspaces, but 5" is quite a bit and I'm wondering how complicated that install will turn into with the many angles on my roof. The more I consider the logistics of the spray foam, without really tearing up the walls to get real, proper access, it's going to be a chore and a half.

More to think about.
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
If you go less than IRC 2012 code on the total roof-R you can get away with less exterior foam. The R20 exterior insulation presumes at total R of about R50: Fully fiber-filled 2x10 rafters for about R30, plus R20 exterior comes in at ~R50 center-cavity (meeting the R49 prescriptive code-min), which is a foam/fiber ratio of 2/3.

If yours is 2x6 rafters with R19s snugged up against the decking you'd be good with 3" (R12.6) of EPS, since the foam to fiber fiber ratio would be about the same which makes the average temp of the roof deck at the same point. While you'd only be at about R32 center-cavity it would actually outperform R38 between joists, due to the breaking of the ~R7 thermal bridging of the rafters with ~R13 of exterior foam.
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
If you go less than IRC 2012 code on the total roof-R you can get away with less exterior foam. The R20 exterior insulation presumes at total R of about R50: Fully fiber-filled 2x10 rafters for about R30, plus R20 exterior comes in at ~R50 center-cavity (meeting the R49 prescriptive code-min), which is a foam/fiber ratio of 2/3.

If yours is 2x6 rafters with R19s snugged up against the decking you'd be good with 3" (R12.6) of EPS, since the foam to fiber fiber ratio would be about the same which makes the average temp of the roof deck at the same point. While you'd only be at about R32 center-cavity it would actually outperform R38 between joists, due to the breaking of the ~R7 thermal bridging of the rafters with ~R13 of exterior foam.

Ok, 3" seems a bit more reasonable, I'm not entirely sure what's in the bays at the moment, I'll check over the weekend.

A question though, how does this change the fascia profile? Seems like it's going to involve a fair bit of construction to make it work.

Something like this?
http://www.buildingscience.com/documents/insights/bsi-063-over-roofing

Though I suppose on a smaller scale since he's got 6"+ of materials.
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
Another thought I've had, something like this:

http://www.raycore.com/images/vented-sip-panel.png

(Ignore that the picture is a SIP panel)

In this case it would be, existing insulation remains, seal over current decking, furring to create an air gap, new ply or osb over the whole thing. Ridge vents up top and then some sort of behind the drip edge intake at the bottom. Does this actually properly vent the space though? I guess it would keep the shingles cooler (which we've concluded isn't much of a deciding factor), and it would properly vent the new decking, but does it address then air leaving the house and hitting the existing decking? (Obviously this method does not improve insulation)

edit: Apparently this is know as a "cold roof"

Here's the idea merged with adding EPS foam to the mix:
http://www.danperkinsroof.com/pdf/metal-roof-venting-and-insulation.pdf

Page 3 shows a nice detail, my question would be how does this then interface with the existing gutters. They're K style and I guess this would just sort of shift the drip edge out away from the existing fascia and more into the middle of the gutter?
 
Last edited:

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
I think you have the picture!

From a "how green is it" point of view steer away from XPS (the pink/blue/green board) and closed cell polyurethane (like the RayCore SIPs) until/unless the manufacturers stipulate they are usin blowing agents with a greenhouse gas potential less than 10x CO2. In the US XPS is blown with a mix of HFCs, the primary component of which is HFC134a (~1400x CO2), most of which bleeds out in the first 25-30 years, and when it does it loses performance. At age 50 it's barely higher per inch than EPS of the same density. Almost all 2lb and 3lb polyurethane in the US is blown with HFC245fa at about 1000x CO2. It's aged performance hit isn't quite as severe as with XPS, but it's there.

EPS and polyiso are usually blown with pentane (7x CO2 ), and for unfaced goods most of it is gone before it's up on your roof, and it has a stable performance over time. Polyiso takes a huge hit with cold temperature in this type of application. From a dew-point control point of view with it mounted outside the sheathing of the assembly you should treat it as if it were R5/inch in your climate, even though it's labeled performance (tested at 75F center-foam temp per ASTM C518) is usually R6-R6.5/inch.

Venting the nailer deck with furring or purlins definitely works, and adds a lot to the overall longevity of the nailer deck and maybe a ~5% life extension for the shingles (varies). Without venting the nailer deck minor leaks may require replacing a few sheets the next time it's re-roofed. But even without venting the nailer deck most do just fine even over a 2-3 decades. (There isn't a lot of history on those stackups older than 3 decades or so.)
 

Indigo

New Member
Messages
21
Reaction score
0
Points
1
Location
Pittsburgh
Do you by any chance know of a resource where I might find detail drawings for things like dormers with the additional roofline thickness?
 

Dana

In the trades
Messages
7,889
Reaction score
509
Points
113
Location
01609
You might need a subscription do download the detail drawings from the site operator's library, but there's a blog where foam-over details oft discussed. Building Science Corp has many example drawings, some of which may address particular issue that you may run into as well.
 
Top
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.

If any particular ad is your REASON for blocking ads, please let us know. We might be able to do something about it. Thanks.
I've Disabled AdBlock    No Thanks