GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

ICF Cape Cod design worries: the roof deck nightmare

ejkessler | Posted in Energy Efficiency and Durability on

Last summer/fall my wife and I started designing a retirement home and after posting my worries on moisture to this blog, I received some great advice that gave much “cause for pause” to study more and also to provide the more detail members needed to better understand the design and what I was trying to accomplish. Since then, and as members of both and, I’ve studied everything I could get my hands for 9-months from these two websites to help me with my Zone 4 Potomac River/Chesapeake Bay region home. Last week we broke ground and started footers so I’m trying to get my mind around some finer details.

I rely especially on the numerous Martin Holladay and Dr. Joe Lstiburek articles (hockey puck humor and all, I love his humor and examples), along with gleaning elements from the Fine Home Building (FHB) “The Passive House Build” 4-part series, the GBA Energy Star drawing details, the GBA “Pretty-good-house” (R-10, 20, 40, 60 formula), RESNET, HERS and such. Having said that, my wife and I chose to break one of Martin Holladay’s top rules for beginners and others, especially the simple roof line rule … because we loved the Cape Cod look … so we do have those nasty dormers (2 on front and 1 on rear), though we are trying hard to mitigate this poor choice by close attention to sprayfoam and air sealing. Our Insulated Concrete Form (IntegraSpec ICF) Cape has about 1700-1800 sq-ft on both 1st floor with MBR and a walkout basement floors. The 900 sq-ft 2nd floor includes 2 BR & 1 Bath which also has an adjacent room for the 2nd stair case via an exterior-grade firedoor directly to the Garage; it also has access to the room truss room above the 2-car garage as a storage room at the top of those same stairs.

Our home & lot is such that we have a long south facing front of house, 7degrees-45minutes from perfect south and a roof pitch of 36 degrees in hopes that we find a pot of gold to purchase some amount of PV array (a major financial stretch for us). Our plan is to use the VRF ducted and non-ducted splits from the likes of Dakin, Mitsubishi, LG, Fujitsu, or… along with a Venmar or someone’s ERV to ventilate and filter air, then finally perhaps an Ultra-Aire XT105 dehumidifier to help control our humid season. We’ve maintained our basic ICF approach for the full basement and main floor and now have specified a very detail Air and Vapor barriers in our drawing package. So now I figure I have arrived at a point where I am becoming dangerous and expensive and a bit confused about the finer details; … correction, I’m starting to get wicked confused in some of these details.

Here is the first of my worries, for fear of scaring you with all my nightmares at once and overwhelming you with living in my head, haha. I’ll post a couple more following this.

The Roof Deck Nightmare:
While we have changed our insulation approach to one where we use Spray Foam against the bottom of the roof deck and encapsulating the top cord of the truss so that all Fire Code accesses and any ventilation equipment are contained inside this barrier; I am still confused if I should run one of my ERV ventilation intake and exhaust tube up there to capture and remove moisture in this tight house give the OCSF layer ability to retain moisture? For fear of the Goldilocks principle and “believing” we need this barrier to be both an air barrier and vapor barrier, I believe I learned from you correctly that I needed “about” 3.0-inches of Closed-Cell Spray Foam (CCSF) as the air/vapor barrier. We plan to apply an additional balance of only 6” of Open-Cell Spray Foam (OCSF), because R-60 cost is making me question the Pretty-good-house R-60 “spec” for my Roof; the combo of 3” CCSF and 6” OCSF only gets me to about R-40, aarg? I am not sure how to cost effectively get to R-60 out of my roof, should I spec 5.7 more inches of OCSF or maybe Rigid foam above my roof deck? Do I really need 12” of OCSF in addition to the 3” of CCSF… Gee, some say after certain point spray foam does not really provide the specified R-Value, is this true? Also, for this semi-conditioned attic which will NOT be used to store anything except insulation, wiring and plumbing … should I run one of my ERV dirty air intakes (& maybe a fresh air supply)? Do I need to apply fire retard coating; I recall reading that I must NOT run ventilation to such unoccupied attic space?

Part of my roof concern is how I design to drive out heat gain BEFORE it ever hits my insulation; but also … is this even cost effective? Am I chasing pennies with cubic dollars? I suspect everyone worries about this, but a clear strategy and cost effective approach seems missing? I seriously worry about this whole cool roof concept and articles because when I examine the so-called cool roof shingles all I see is bragging over a tiny 10-degree attic temp drop and no reference to roof deck temp reductions … I was expecting much more. Where do I see the specific strategies for 100degree roof deck temp reductions some articles mention and what should that translate into as attic changes? So am I wrong, that these so-called energy star shingles are not the answer. I purchased and read everything in the FHB “Siding, Roofing, and Trim” book for pros by pros book but no strategy emerged there either. Oh yeah I am so confused by now. This had me looking at Standing Seam metal (ouch major expense) some venting ideas like the “Cool-Vent” vented nailbase insulated panels … wow is that expensive too ($57+ per 4×8 panel) which is in addition to the basic roof deck. Another idea was from Martin Holladay’s May 30, 2014 Musings, the Smart Vent which when used with their 16” OC baffle created a 2” airflow space under the roofdeck that Spray foam can be over and encapsulate along with their Smart Vent and Smart Ridge system.

Or … Maybe I should just build a roof over a roof with a 1-2” air-gap and a waterproof, but highly vapor permeable, underlayment from DuPont or someone on the first under roof that had the sprayfoam applied to its bottomside; and then some kind of radiant foil facing the sun on the 2nd upper roof that has the shingles applied. It might allow any moisture drying of the topside of the roof with the air flow between the layers and also get rid of some of the radiant heat trying to get to my sprayfoam insulation. Maybe Zip System in combination has a solution?

Aaarg, you see I am lost at this point with my new and dangerous “knowledge/information” because I’m trying to find and understand what is economically wise. Help 🙂

I have provided attachments however the final drawings because the .pdf file was 2.6MB and larger than 2MB limit. Below you will find the kessler31314am.pdf that shows the general layout of home and some detail toward the end of the file showing sprayfoam detail and ICF Foam alignment while the 2nd file show some other sprayfoam and alignment effort to create an air sealed and vapor sealed Gable end of Cape that may not be part of the dwg package.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. Expert Member
    Dana Dorsett | | #1

    "Gee, some say after certain point spray foam does not really provide the specified R-Value, is this true? "

    No. The R-value is there.

    What they are really saying is that when you are using insulation as expensive as closed cell foam the FINANCIAL value of the energy savings for additional foam isn't as high as the first-cost of that foam on any reasonable lifecycle basis:

    Hitting R60 with cellulose at 3 cents per R per square foot has a very different financial crossover point than closed cell foam at 17 cents/R-foot. Close cell foam is 5-6x as expensive per unit-R(!)

    Open cell foam is about 10-12 cents/R-foot, and does a great job of air sealing, but is too vapor-permeable to be used alone, without taking into account the other layers in the stackup. But that's still 3-4x the price/R of open-blown cellulose.

    Rigid EPS or polyiso above the roof deck run about 10 cents/R-foot at about 3x that of cellulose.

    R60 may have no financial rationale in a cathedralized ceiling or conditioned attic, even if R100 might if insulated with open-blown low density cellulose.

    Radiant barriers and "cool roof" shingles are a waste of money in US climate zone 4 in any code-min stackup, and are likely to increase the moisture content of the roof deck if implemented poorly. Foils are true vapor retarders, and if you're putting it on the exterior of the roof deck it's difficult to avoid creating a moisture trap at the roof deck. Permeable shingle underlayments can't appreciably raise the vapor permeance of the shingles, but even #30 felt + shingles almost an order of magnitude higher vapor permeance than 3-mil aluminum.

    Some design considerations are:

    * In a zone 4 climate, as long as 30% or more of the R-value is on the exterior of the roof sheathing you can get by with class-III vapor retardency on the interior side. (per IRC 2012 chapter 8: When code-min is R49, R15 is sufficient for dew point control with class-III interior vapor retarders in zone 4A, which is you.)

    *Spraying open cell with "vapor barrier" latex paint delivers about 5-perm performance. While that is an order of magnitude higher than it's ~0.5 perm performance when applied to wallboard, it's "good enough" to meet the conditions spelled out in chapter-8

    So, an R50 roof would take 4" of EPS above the roof deck with 10" of o.c. foam sprayed with v.b. latex, at a cost of about $5-5.50 per square foot.

    In the R60 roof would take 2" EPS + 2" polyiso (with the EPS on the exterior side, the polyiso next to the roof deck) and 11" of o.c. foam at about $6-6.50 per square foot. (Using the combination of foams keeps the thickness at 4", which is comparatively easier to build than 5" of EPS would be. Putting the EPS on the exterior keeps the polyiso layer warm enough that you don't have to discount it's performance for low outdoor temp operation.)

    Using 2x4 furring though-screwed into to the truss rafter elements 24" o.c. with pancake head timber screws (eg FastenMaster HeadLok ) to hold the foam in place gives you something on which to mount a (now ventilated) nailer deck for the shingles. If you like, you can use foil faced roof decking as the nailer (with the foil facing the ventilation gap, but the boost in thermal performance isn't really enough to matter. A generic half-inch (or 7/16") OSB is enough of a nailer in most applications.

  2. GBA Editor
    Martin Holladay | | #2

    Dana has given you some good advice.

    In general, insulating at the roofline (creating an insulated roof assembly) is more expensive (per R) than insulating the floor of an attic.

    You don't have to build a house with an unvented conditioned attic (or an insulated cathedral ceiling), but if you choose that style of house, you have to be prepared to spend more for your insulation.

    If you want less expensive insulation (cellulose), then design your house to have a vented, unconditioned attic, and put all the insulation on your attic floor.

  3. GBA Editor
    Martin Holladay | | #3

    In case my last response left any uncertainties, let me be explicit: a Cape Cod design is one of the hardest to air seal and insulate. If you are building a Cape Cod, the best way to handle the problematic roof design is to put all of the insulation above the roof sheathing. I would recommend that you consider a SIP roof or the installation of thick nailbase above your roof sheathing.

  4. ejkessler | | #4

    Dana and Martin,

    Thank you so much for the super quick and thoughtful advice! I do have a few follow-up questions to better understand and sort out these options. I so appreciate the pricing info, which really helps sort out ideas in a very practical way – just awesome!

    I chose the conditioned attic/insulated roof assembly approach to help mitigate the challenge of having a Cape Cod style, partly because the Fire Code required access to attic and under roof areas outside of the rooms and I did not want air & moisture leaky access openings and other leak concerns like wall or ceiling fixtures; but also because it provided access for ventilation pathways.

    I am very intrigued by the idea and possible simplicity of the SIP idea for the roof, but am concerned about losing my “Air, Vapor & Insulation” continuous uninterrupted connections of my Foam-to-Foam connections at every boundary possible in my design from the Slab-to-ICF Wall and then the ICF Wall-to-Roof spray foam. Basically the my drawings detail my Foam-to-Foam seal from the Foam roof insulation through the Closed-Cell Spray Foam Raised heal interface at the ICF Wall, down to the ICF Wall to the Basement R-20 Foam Slab. You can see part of this in my 1 sheet attachment titled "kessler30714_Roof&Wall_Note_TheBottomRightDetail.pdf" and also on page 15 & 21 of the “kessler31314_am.pdf” titled file.
    • If I use a SIP on the roof, then I suppose I must still apply enough CCSF in the ICF Wall to Roof interface and then up the underside of the SIP and between the top-cord of the Trusses so that I maintain the tight air-seal with the Walls all the way to the roof? If this is correct then if I wanted say an R-60 roof then I should consider an R-50ish SIP and then CCSF another 2-3” inside the roof deck and also fill the Roof to Wall cavities with R-40 or so?
    • Say, will the SIP be as strong against wind and storms as Spray-foam encapsulated roof trusses?
    • With a SIP would you still install a 2” air gap and other nail base for the shingles in order to provide ventilation so some of the radiant energy is removed before it conducts through the SIP?
    • Do the 2” air gap approaches create a risk of condensation or other problems? Maybe my desire for removing some of the heat gain before it hits the insulation is risky in some way???

    The other idea of your I so far seem to favor is that combo of the 2” Poly-Iso and 2” EPS on the outside with CCSF on the inside encapsulating the roof Trusses.
    • Is installing such layers dangerous for roofers? I imagine that stuff is slippery … or do they do it all the time?
    • As with my above concern about the 2” air gap, I like the idea of removing some heat gain before it gets to the insulation but in my Zone 4 area am I either chasing pennies with dollars or creating a moisture/condensation issue between the shingle layer and the 2” of EPS? And won’t the EPS absorb moisture and get soggy?

    Sorry if these are dumb questions but I am really trying to understand these details.

    Eric Kessler

  5. GBA Editor
    Martin Holladay | | #5

    Since you have so many questions, it may be time for you to consider paying for the services of an architect and an engineer.

  6. jackofalltrades777 | | #6

    I have an ICF wall and SIP roof design and I had to hire an engineer to get the conditions resolved. Most counties will require an engineers stamp on the drawings when you use unconventional building methods (ICF & SIPs).

    As far as SIP roof strength goes. SIPs are basically an "I-Beam" construction so they can resist some serious loads. A roof truss vs a SIP is a no contest in terms of strength, the SIP wins. They can up to a Seismic "C" design and "D" is allowable with engineering. They can take wind loads up to 150mph without issues.

  7. ejkessler | | #7

    Yes, I understand Martin. I wanted to better understand your good ideas. Of course I did hire "expert" services and they produced the drawings you see attached, and we had planed to use the combo of the spray foams shown but after reading GBA articles about driving out the temperature I wanted to improve on that plan. The final plans with the engineers stamp were to 2.6MB and too large to upload which is why you don't see the engineers stamp.

    Of course I have learned rather expensively so far that there is much miss-information and general lacking knowledge by many in the industry out there; especially when I compare advice with what i read in GBA and FHB. So between "trust with verification" and joining and reading everything I could on Fine Homebuilding I learned that cooler roofing was possible but could not find the expertise that said what I had been reading about.

    I've also hired a RESNET reviewer who has my plans (we meet again on Tuesday) and he is reviewing the plan for refinement and will also conduct the certification and blower door testing portions.

    That said, I want to learn! That's why am trying to study as much as I can on my own to both personally understand the Building Science (I also signed up for the GBA/HeatSpring Mastering Building Science course starting in October) but also to be able to build trust and verification of what I am told by various competing expert opinions.

    Anyway Martin that's why I wanted to better understand Dana and your suggestions since you folks at GBA and FHB seem to be the Center-Hub of all the leading experts. I certainly did not mean to over burden anyone trying to more clearly understand. I've been enjoying the leaning process and the blog dialog and perhaps got too enthusiastic. I hope this helps clarify better why I asked the follow up questions.


  8. GBA Editor
    Martin Holladay | | #8

    Q. "If I use a SIP on the roof, then I suppose I must still apply enough closed-cell spray foam in the ICF wall-to-roof interface and then up the underside of the SIP and between the top-cord of the trusses so that I maintain the tight air-seal with the walls all the way to the roof?"

    A. No. There are many products that can be used for air sealing other than closed-cell spray foam. Most energy-conscious builders would use European tapes for this purpose; these tapes are available from 475 Building Products or the Small Planet Workshop.

    Q. "Will the SIP be as strong against wind and storms as spray-foam encapsulated roof trusses?"

    A. SIP roofs are routinely engineered to meet all wind-load requirements. If you have any doubts on this issue, talk to the engineers at your SIP supplier or consult with your own engineer.

    Q. "With a SIP would you still install a 2 inch air gap and other nail base for the shingles in order to provide ventilation so some of the radiant energy is removed before it conducts through the SIP?"

    A. I think that it is a good idea to provide a 1.5-inch-deep or 2-inch-deep air gap (ventilation channel) above a SIP roof, but it has nothing to do with radiant energy. The ventiltation channel is there to assure good moisture management -- not to reduce radiant heat flow. The ventilation channel does two things: (a) it lowers the temperature of the top layer of roof sheathing during the winter, reducing the chance of ice dams, and (b) it helps remove any moisture that accumulates in the exterior OSB facing of the SIP. (This moisture accumulation occurs when there are air leaks at the SIP seams.)

  9. GBA Editor
    Martin Holladay | | #9

    Q. "Do the 2-inch air gap approaches create a risk of condensation or other problems?"

    A. No.

    Q. "Is the installation of such layers dangerous for roofers? I imagine that stuff is slippery … or do they do it all the time?"

    A. Roofing is one of the most dangerous jobs in the U.S., according to published statistics. More roofers are injured on the job than any other type of construction worker. It's always a good idea to make sure that your roofing contractor follows OSHA fall protection guidelines. If you case about this issue, ask your contractor about fall protection measures.

  10. GBA Editor
    Martin Holladay | | #10

    Q. "I like the idea of removing some heat gain before it gets to the insulation."

    A. There is no need to do that. The insulation is there to limit heat flow.

    Q. "In my Zone 4 area, am I chasing pennies with dollars?"

    A. If you are spending money to reduce the amount of radiant heat that reaches the top layer of your insulation, the answer is "maybe." When it comes to this strategy, only low-cost measures make sense.

    Q. "In my Zone 4 area, am I creating a moisture/condensation issue between the shingle layer and the 2 inches of EPS?"

    A. As I wrote earlier, the air gap won't cause moisture problems; it will prevent them.

    Q. "Won’t the EPS absorb moisture and get soggy?"

    A. SIP failures aren't caused by soggy EPS. They are caused by soggy OSB.

  11. Expert Member
    Dana Dorsett | | #11

    EPS at better than 0.5lb density doesn't get soggy- it can't. EPS is a closed cell foam, with many uses such as dock/raft floats, surfboard cores, lobster pot buoys, etc. Type-II EPS runs about 1.5lbs per cubic foot nominal density, and will withstand extended submersion with at about a 7% reduction in thermal performance while submerged for months, but will dry and regain full performance when the tide goes out. The performance hit is from water getting into the interstitial spaces between the macroscopic beads (which fill up and drain relatively rapidly under full immersion), but the microscopic cell structure responsible for most of the thermal performance is still dry & intact.

    In climate zone 4 the use of radiant barrier or "cool roof" roofing on anything but very low pitch roofs (under 2:12) results in an annualized net increase in energy use, trading off modest cooling season reductions for slightly less modest heating energy use increases. In zone 3 it get's into the "maybe yes, maybe no" gray zone where the particulars matter.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |