GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

Exterior sheathing and insulation

cdgatti | Posted in Energy Efficiency and Durability on

Hello new member here. Plan on building a home in TN, climate zone 4A. Want to make this home very efficient and comfortable. Planning on geothermal ground exchange HVAC as no natural gas is available in the area. My question is around exterior wall insulation.

My builder’s standard is 2X6 walls with OSB and tyvek style house wrap with standard fiberglass bat insulation. I am interested in 2X6 walls with rockwool bat insulation and a thermal bridge preventing exterior insulation like the rockwool over the tyvek OR an insulated ZIP sheathing. Is this extra exterior insulation overkill for this application and if NOT what would people recommend for this zone? Thank You!

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

    There is no such thing as "...thermal bridge preventing ...". If there is insulation in the framing cavities rather than continuous insulation on the exterior, there IS thermal bridging. A continuous layer of insulating sheathing is a thermal break, reducing the amount of heat transfer through the thermally bridging framing, but it's not "preventing" the heat transfer.

    A 2x6/R23 16" o.c. wall with typical siding & interior options comes in at about R15-R16 "whole wall", after factoring in the thermal bridging of the framing and the R-values of the other layers. Adding 1.5" of rock wool or an inch of polyiso (or 1.5" ZIP-R) improves that to the low-mid R20s.

    In a zone 4 climate at typical material & labor pricing up to R25 is not overkill, still usually financially rational on a lifecycle basis, according to Table 2 ,p10 of this now classic document from Building Science Corp.

    Unless your builders already have experience with mounting exterior insulating sheathing over the structural sheathing they're probably going to be charging you extra to cover their learning curve risk. Even though the material costs are higher, going with 1.5" ZIP-R may be cheaper, since the learning curve is a lot shorter, with less detailing, and fewer ways to screw it up.

    Geothermal (ground source heat pump, or "GSHP") may not be the most financially rational solution. Before going down that path very far start with a preliminary load calculation and adjust it as changes to the window sizes/types and wall & roof insulation & stackups change. With a reasonably elegant footprint and house design a 2500' house in the TN hills with U0.20-ish double low-E double panes and an R23+ R6 c.i. wall can come in under 25,000 BTU/hr @ +15F (a typical hill-country 99% outside design temp in TN: ) At loads that low there are several high efficiency modulating air source heat pumps (ducted & ductless) to consider that can come close to GSHP efficiency at a fraction of the installed cost. In my area (southern New England) GSHP is more expensive than some regions, and of the difference in up front cost can pay for enough rooftop solar to cover ALL of the space heating & cooling electricity use, not just the difference related to the somewhat lower efficiency.

    A code-min window is typically an air-filled single low-E double pane running U-0.35 & an SHGC of 0.27-0.33. Going with an argon-fill can bring that down to U0.28-U0.32, with a similar SHGC. But with glass that has a second coating of hard-coat low-E on the surface in contact with the room air the center glass U-factor can drop to the U0.20 range, roughly the same as a triple pane, but at much smaller up-charge. The low-E coating selected for surface #2 (the interior surface of the exterior pane) can set the SHGC, and for west facing windows you'd want something less than 0.20 to keep the cooling loads well bounded, but on south facing windows something north of U0.50 allows you to take in wintertime sun, and shade those windows in the summer from roof overhangs. Better tuned higher performance glass like that would be a better expenditure of funds in most locations than going with GSHP for reducing electricity use.

    But start running the load numbers. If you know how to use standard computer spreadsheet tools start an I=B=R style room-by-room spreadsheet and up-date it as you lock down some of the design conditioned. See:

  2. cdgatti | | #2

    Wow that is alot of detailed info you provided, Thank You Dana. The house is a 3800 ft^2 basement ranch, 2300 up and 1500 finished basement. It is facing south, but the view and most of the windows are in the rear where 2 levels are exposed. Only one level exposed in southern exposure front. I don't think my builder has much experience with external insulation as the climate is considered temperate. Geo is big in that area, however. My thought was if I could payback the premium in ~10 years I would go with the geothermal. Also like the idea of radiant floor heat and reducing domestic hot water costs with the geo. I may just go with the 24" OC 2x6 with rockwook (r23 I believe) forgo the exterior CI and use the geo. What do you think of that plan?

  3. GBA Editor
    Martin Holladay | | #3

    First of all, can you tell us your name?

    The high cost of a ground-source heat pump (what some contractors call a "geothermal" system) is rarely justified by the small energy savings. It usually makes more sense to install an air-source heat pump (one or two minisplits) and to take the money you saved on the cheaper equipment and use the money to buy a PV system. For more information, see Are Affordable Ground-Source Heat Pumps On the Horizon?

    You mention in-floor radiant heat. Remember that the decision on whether or not to install a hydonic (water-based) heating system has nothing to do with the decision on whether to install a ground-source heat pump. You can have in-floor radiant heat with a variety of heat sources, and you can have a forced-air (ducted) system with a variety of heat sources (including a ground-source heat pump).

    In Tennessee, you don't really need an in-floor radiant heating system. For more information, see All About Radiant Floors.

    When it comes to your wall assembly, it's hard to know whether the investment in continuous exterior insulation will make sense. A lot depends on the skill of your contractor and the upcharge. Here's an important point to remember: it's far more important to pay attention to airtightness (when building your walls, as well as when building your floor and ceiling) than to worry about R-value. Make sure that you have a goal for airtightness, and verify the results with a blower door.

  4. cdgatti | | #4

    Thank You Martin. Yes that all makes sense to me. Sorry about the abbreviated screen name. Name is Chris Gatti.

    The front of my house faces south - don't think I would want PV cells on the front roof of the house for aesthetic purposes. I am very surprised by the relatively negative stance on geothermal in this forum about energy savings. Especially for new construction when the upcharge over a state of the art conventional system makes the most sense. How do you beat $3-$5 of energy for $1 of electricity? I do understand that the payback period is 10 years or so. One of my main drivers here is every air sourced heat pump house I have been in in the winter in this zone feels cold in the winter due to the temperature of the 'heated' air being so far less than a good old gas fired hot air system. I want to be able to raise my thermostat with impunity in the winter time. Making the house as airtight as possible is definitely in the plan - besides a good house wrap what else needs to be pursued here beyond just sound construction?

  5. GBA Editor
    Martin Holladay | | #5

    Q. "How do you beat $3-$5 of energy for $1 of electricity?"

    A. Lots of ways. You are comparing the theoretical COP of a ground-source heat pump with electric resistance heat. But electric resistance heat is not a realistic option, so the comparison is silly. A better comparison might be to natural gas heat.

    When it comes to air-conditioning, the energy comparison you cite is meaningless.

    It makes more sense to compare the COP of a ground-source heat pump (generally in the range of 3 to 4) with the COP of a good air-source heat pump (generally in the range of 2.7 to 3 or more). True, the COP of the ground-source heat pump will be a little better -- but the equipment costs so much more that it is unlikely that the upcharge can be justified. Moreover, the complications inherent in a site-built system (the ground-source heat pump) -- the possibility of design errors, installation errors, and commissioning errors -- are serious. These problems are much more likely than with an air-source heat pump.

    Finally, your report of comfort problems with an air-source heat pump are based on old technology. Homeowners with newer versions of ductless minisplits or ducted minisplits don't report comfort problems.

    A PV array can be roof-mounted, but it doesn't have to be. You can also install a ground-mounted PV array in a side yard or your back yard (if your property is big enough).

  6. STEPHEN SHEEHY | | #6

    Chris: My house is in Maine, zone 6, which is a lot colder than TN. My air source heat pumps keep us as warm as we want, even when temperature is way below zero. As Martin says, key is good air sealing, as well as good insulation. Geothermal is complicated and expensive.

  7. GBA Editor
    Martin Holladay | | #7

    Q. "Making the house as airtight as possible is definitely in the plan. Besides a good housewrap, what else needs to be pursued here beyond just sound construction?"

    A. The housewrap has nothing to do with airtightness. The purpose of the housewrap is to act as a water-resistive barrier (WRB) -- that is, a secondary barrier to stop any wind-driven rain that gets past the siding. For more information on WRBs, see All About Water-Resistive Barriers.

    There are lots of articles on GBA about airtightness, including lots of articles by owner-builders who have reported on their air-sealing efforts. Use the "search" box at the top of every GBA page to learn more. You can start by reading this article: Questions and Answers About Air Barriers.

  8. Jon_R | | #8

    Anyone considering residential PV needs to consider what net metering agreements are available and the chances of them continuing. Don't assume that the utility will act as a free battery.

    Consider running BEoptâ„¢ (Building Energy Optimization) software.

  9. Expert Member
    Dana Dorsett | | #9

    With GSHP vs. ASHP (air source heat pump) it's a matter of upfront cost/lifecycle cost and design risk. Air is pretty much the same everywhere, and weather data sets can tell you with reasonable precision what sort of average seasonal efficiency to expect if the ASHP is right-sized for the load. In TN a right sized modulating heat pump should deliver a seasonal average all-in coefficient of performance (COP) of about 3.5-4.

    GSHP has a higher up front cost and higher design risk, since there are more variables at play. A best in class GSHP might hit a seasonal all-in COP of 5, but the average real world performance is less than that, often no better than 3. A lot depends on the ground loop design and the amount of pumping energy necessary to deliver that ground heat to the heat pump. A sloppy design or poor pump specification can bring even the most efficient GSHP to it's knees.

    A 2300' above grade floor are with 1500' of mostly below grade finished basement should have a design heat load less than 30,000 BTU/hr @ +15F even for a tight 2x4 framed house with U 0.50 double-panes. A tight 2x6 /R23 + R6c.i. house with an R12-R15 c.i.basement and even U0.35 windows would be in the 20-24,000 BTU/hr @ +15F range. With U0.20 windows it might even come in under 20K. Even 2 tons of cold climate ductless or ducted modulating mini-split can cover a 30K of heat load, and would likely more than cover your cooling load as well. As a general rule the output temperature of mini-splits is higher than larger 'merican style ducted ASHP, which is in turn higher than the temperature of air-output GSHP system (though with the latter that is also a matter of design- efficiency can be traded for higher air temperatures). Two tons of ductless cold climate mini-split would cost about $7.5-8K (all in, fully installed, no subsidy) in my neighborhood in competive bidding. Two tons of mini-ducted mini-split would run about $10-11K, sometimes less if the house framing design was optimized for that sort of system. In my area 2 tons of GSHP would be well north of $20K, often north of $30K. With the 30% federal income tax subsidy that gets knocked back to ~$15-21K, still twice the cost of a mini-split solution.

    Would the difference in efficiency between a COP of 3.5 vs. 5 pay back the cost difference in less than 10 years? In my area, even with 20 cent electricity (roughly the regional average) it wouldn't even pay back in 25 years. Even if in your case it's only a $3K cost adder it's unlikely that it would pay back in ten years, even if your electricity became as expensive as mine rather than the ~10 cent/kwh average currently paid in TN.

    The biggest problem with most ducted old-school type ASHP is still grotesque oversize factors. Even with an HSPF 8/SEER 14 single speed comfort and as-used efficiency is all about sizing it right. Those "cold" houses would feel a lot warmer if the systems were sized properly and not delivering intermittent blasts of tepid air at higher than needed cfm, but were instead running a steady but lower cfm run at a much higher duty cycle. So many of them have been sized for air conditioning with a "ton per 500 square feet" or "ton per 700 square feet" rule of thumb, with oversizes it by 2-3x for the cooling, and more than 1.5x for heating. Typical cooling loads for houses with 2300' of above grade space in the southeast would be less than 2 tons, and the basement will have negative or zero sensible load, only latent load. A better tuned better designed house that size could have a cooling load of about 1-1.5 tons. Allison Bailes ( a frequent contributor here) in the Atlanta area ran this plot of actual Manual-J calculated square feet per ton against house size that his firm did the math on:

    (Clipped from this blog piece he did here: )

    The average house in the 2300' range comes in a ratio of about a ton per 1300', or about 1.75 tons, but most houses that size would be outfitted with 3- 4 tons, some even 5 tons by the rules of thumb folks. A dumb 1 speed 3 ton heat pump has an AHRI output of over 30,000 BTU/hr @ +17F, a 3-ton cold climate heat pump would have more than 40,000 BTU/hr of capacity @ +17F.

    It's actually better for average efficiency with the dumb 1-speed to undersize it slightly, and use properly sized heat strips to make up the difference. The lower efficiency of having to suppplement with heat strip for 2-5% is paid back several times by the higher as-used efficiency that comes with higher duty cycle the rest of the season. But with modulating mini-splits (or modulating large air handler heat pumps) they can usually be sized to where they will cover the 99% heat load with decent margin, yet still modulate at a VERY high duty cycle most of the season.

    But with any heat pump (including GSHP) you simply HAVE to nail down the actual load numbers to get both the efficiency and comfort you're paying for.

  10. cdgatti | | #10

    Thank You very much to all the experts that have chimed in here on my design questions. What I am taking away is that I would probably be better off focusing on:

    1. better air tightness
    2. R23 2x6 with CI
    3. ASHP with mini-splits

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |