4 Types of R-Value

26 Comments Read/write comments

thermal-bridging-framing-insulation-r-value.jpg

We talk about R-value all the time. "I've got an R-19 wall," or "Code requires R-38 in my ceiling." But what are those numbers? As it turns out, when we talk about R-value we usually give the R-value of the insulation. That's the case with both of those statements above. But what's the real R-value of the wall or the ceiling? Insulation makes up only a part of each. There's also wood and drywall and sheathing and cladding and...

If we use R-value to describe only the insulating properties of the insulation we install, we neglect the insulation value of the other layers in a building assembly. In a wall, those layers include drywall, sheathing, cladding, and air films. They add to total R-value.

Similarly, using only the insulation's R-value ignores the effects of thermal bridging. Wood has a lower R-value than insulation so including it reduces the overall R-value of a wall. And then if you want to factor that thermal bridging in, how do you do it?

As it turns out, building scientists at the Oak Ridge National Lab worked on this back in the 1990s. Jan Kosny and Jeffrey E. Christian wrote a paper in 1995 titled Whole Wall Thermal Performance and introduced three new types of R-value. Here are the names and definitions, straight from their paper:

Center-of-Cavity R-value

R-value estimation at a point in the wall's cross-sectional R-value containing the most insulation.

Clear wall R-value

R-value estimation for the exterior wall area containing only insulation and necessary framing materials for a clear section with no fenestration, corners, or connections between other envelope elements such as roofs, foundations, and other walls.

Whole-wall R-value

R-value estimation for the whole opaque wall including the thermal performance of not only the "clear wall" area, with insulation and structural elements, but also typical envelope interface details, including wall/wall (corners), wall /roof, wall/floor, wall/door, and wall/window connections.

The first one doesn't include any thermal bridging effects, so it's not a good one to use when comparing buildings. Clear-wall R-value is better, but it doesn't include enough of the thermal bridging since it ignores corners, T-walls, and other important thermal bridges. 

new-home-construction-complex-building-enclosure.jpg

Whole-wall R-value is really what you want to know most of the time. In new home construction, complex building enclosures rule the market. Look at all the corners, roof-wall intersections, and other complexities in the photo above. At least on the front of this house, there's not a whole lot of clear wall there.

This is a code-built house in Atlanta, so they used R-13 insulation in the walls. They're putting half inch drywall on the interior, half inch OSB on the exterior with brick and stone cladding. My estimates of the various R-values are:

  • Insulation:  R-13
  • Center-of-cavity:  R-15
  • Clear-wall:  R-14
  • Whole-wall:  R-10.5

I got the center-of-cavity R-value by adding the R-values of the layers. I calculated a framing factor (ratio of framing to insulated cavity area) of about 13% using 9' ceilings to find the clear-wall R-value. And I estimated a framing factor of about 30% to find the whole-wall R-value.  The difference between those two framing factors is that the clear wall framing factor includes only studs and plates whereas the whole wall framing factor includes studs, plates, corners, T-walls, and headers.

A standard home has a framing factor of about 23%, but this home has a lot of corners, intersections, and openings. If I use the 23% framing factor, the whole-wall R-value is 11.8.

Whole-wall R-value doesn't include the windows and doors. If we include those in our calculation, the number is even lower. A typical window here in Georgia has R-value of 3 to 4, and doors are similar. Average that in with the R-13 insulation, and your overall R-value drops to single-digits.

The thing to remember here is that the R-value in your walls, floors, and ceilings isn't the same as the R-value of the insulation you put in them. Oh, and one more thing: The R-value you calculate according to the definitions above isn't a static number. It changes with temperature and other factors.

 

Related Articles

The Layers and Pathways of Heat Flow in Buildings

The Diminishing Returns of Adding More Insulation

Flat or Lumpy - How Would You Like Your Insulation?

Attic Stairs - A Mind-Blowing Hole in Your Building Envelope

Big News: The R-Value of Insulation Is Not a Constant

 

NOTE: Comments are moderated. Your comment will not appear below until approved.

Comments

Jonathan Webber

My last building project was 25 years ago and at that time, window heatloss was about equal to the rest of the house and infiltration accounted for about half of that.
What are current estimates?

Allison Bailes

Jonathan, I don't know what those numbers would be now but I could look at some of the load calculations we've done to get an idea. Windows are definitely better now than they were 25 years ago, though. Instead of R-1 to R-2, they're now R-3 to R-4, at least here in Georgia. I think that in warm climates, the difference is even bigger because there's probably been more improvement in solar heat gain coefficients than in U-values.

Tapani Talo

This 2016...R-4 went out the window before oil embargo 1973. I swear that we are heading to intellectual ice age. MIT developed South Wall glass that delivers R-12 to R-14 is something that you have to experience to understand what a great fenestration does to you. I had it for 20 years before moving to current location. I feel like I have gone to camping in the 60's or earlier. Summer's sun and winter cold - bot unbearable with regular double glazing. So planning a new home ...and back to comfort

Charles

Thanks Allison for coming back to this - one of my favorite topics - and also one that many (most?) simply do not consider. I believe your numbers for clear wall and whole wall are too generous and should be less. ASHRAE noted in 2003 that the average home's framing factory is 25%, not 23%, and 27% for California. And since many building codes actually increased shear wall requirements in 2012 (bringing more in-line with California) we believe that number has increased, not decreased. And when you consider that these walls are probably 20-25% fenestration & doors - and I'm guessing more in the rear of the house, your whole wall number should decrease dramatically.

I would love to see a more detailed study of corners - they are great for racking/sheer strength of an overall structure but they can significantly decrease the thermal performance of a home. I remember growing up in a house with four corners! Now we often bid on homes with 30 corners or more! Crazy crazy.

Allison Bailes

Charles, I don't think my clear-wall number is off. It's pretty easy to calculate the framing factor for a 16" o.c. wall without the effects of openings, corners, and other complexities. The framing factors you give for whole-wall R-value are actually lower than the one I used for the house in the photo. I guesstimated 30% because of all the openings and corners on the front of the house.

But your point is a good one. Assuming a framing factor of 23% (the number used by RESNET for home energy ratings) means you're not calculating the actual framing factor for the house.

Keith Nelson

As always, great topic Allison! I am typically having this discussion in the commercial construction world with steel studs further reducing the "Whole Wall R-value". What I find most interesting and troubling is the disconnect between design professionals, generally architects and their MEP designer or energy modeler, where an "R-19" wall is converted and input as a 0.053 U-Factor wall in lieu of a something closer to 0.109! To the point of your article, we need to understand the meaning of the values behind the labels we use. Same goes for U-Factors of fenestration as in center-of-glass v. assembly. Shockingly, they are not interchangeable!

I hope to see you in Austin next week!

Allison Bailes

Great points, Keith. I wrote this article not only because it's important by itself but also as a lead-in to one on the effect of steel studs on R-value.

Yes, I'll see you in Austin. Are you going to be there for the Humid Climate Conference or just the BEC?

Keith Nelson

I'll be there for both and the BETEC/BEC National meetings on Wednesday.

David Butler

@Allison, I'm surprised you didn't mention how easy it is to improve the typical wall with advanced framing. Even without going to single top plates you can get the whole-wall opaque R-value below 20%, AND pick up an extra R-6 in the cavity without much additional cost. Add a layer of XPS on the outside and the whole-wall R-value more than doubles, compared to your example.

@Keith, good point about impact of steel studs. The framing factor typically goes down with steel but unless most of the insulation is moved to the outside, the difference between, say, a 25% and 15" framing factor is pretty much irrelevant. By moving the insulation to the outside, the FF becomes a non-issue.

Allison Bailes

Well, David, I'll just have to cover that in another article. I try to keep my articles from getting too long, and I like to have them lead into followup pieces.

Andy Wahl

I like the picture, 9 studs. They are a higher R-value than the missing foam.

Allison Bailes

Yeah, that one caught my eye when I visited that house about 5 years ago.

Debbie Coleman

Thanks for the laugh about the photo with 9 studs lined up - one of my pet peeves in a so-called energy-efficient home! And one reason why I like to spec at least a 1/2" of rigid insulation board on the exterior (more or much more) in all but the mildest climates) for single stud wall assemblies.

Charles

Hi Debbie, or others, how does your cost come out per wall square foot when you are applying an extra layer of insulation on the outside, then having to account for attaching siding, extended jambs for windows/doors, etc. etc. I know it's not straight forward so I'm wondering what that does to cost compared to just single stud framing.

Allison Bailes

You're welcome, Debbie. I've been meaning to use that photo for several years now and this was a great article for it.

David Butler

@Charles, cost is going to depend on the particulars of a given home, especially insulation thickness and type of cladding. For example, stucco cladding systems already include 1" EPS and are still is among the least expensive type of cladding. Windows with nailing fins accommodate stucco/EPS cladding without jamb extensions. For thicker insulation, stucco trim bands (stucco returns) are an attractive and inexpensive alternative to extending the window out from the wall.

John Hatfield

Allison great picture showing a poor application of spray foam and the multiple studs in a pony wall type application. To your point about whole wall R or U-Values this is misunderstood by almost 90% of the inspection and design community. Until the 2012 or 2015 IEC Code started to push the values towards whole wall evaluation most building departments would not review provided insulation values. The framing factor for walls was established in Colorado as 18% framing to 82% cavity on a 16 inch on center wall back in 2003 and still nobody paid attention. The downside of speed framing is that wall skin applications are more dependent upon fasteners in the OSB layer than in studs and therefore the insulation materials are incorrectly installed with bigger gaps. I have observed many applications were the heat loss calculations (Manual J) and HERS values are extremely wrong due to the whole wall value being incorrect as your posting explains. Designers and contractors are not willing to change because the craftsmen do not understand energy principals. Great work keep it up.

Allison Bailes

Thanks, John. It's an important topic that, as you say, isn't well understood.

TAPANI TALO

For his reason alone, I frame house standard 2x6, roof joists 2x10, but add continuous insulation to the exterior side. Walls R+ 40 to 50 and roof R=60 to 80, The buildings are as sweet and comfortable as you can think. And hardly some or any costs to run AC or heating. My clients are the best adds after this process.

Allison Bailes

Comfort begins with a great building enclosure.

TAPANI TALO

the cost of adding R-30 to exterior walls and roof is 3 dollars material cost per sf.
walls are 3 to 4 times more comfortable, roof the same. I have clients who's ac does not go on in summer.
I personally changed my house in stages over 20 years, kept a long on energy usage, the main thing that took my breath away was the comfort and silence that my walls and roof gave. I did have Southwall R-14 glazing on most windows, that added to the comfort level exponentially also. Southwall glass was MIT's best invention to building industry that rest of the window and glazing industry has fought tooth and nail against as the glass thickness is non standard. But could have stopped global warming if adopted 1973.

Stephen Christensen

Any thoughts on products that assign "R value" to building materials with low emmissivity? I think this is a fast and loose application of the term "r value" since my understanding if R is resistance to conductive heat transfer, not radiant heat transfer.

Allison Bailes

You're right, Stephen. Foil-faced bubble wrap, insulating paint, and similar products do play a little fast and loose with the definitions. They're giving credit to a product for the R-value of an assembly.

But R-value actually does include all three forms of heat transfer: conduction, convection, and radiation.

Bill Nickerson

Allison I struggle with this. If a pane of glass is R-1 how do we get to R-4 with one more and an air gap?

Allison Bailes

Great question, Bill. The total U-value of a window is calculated by averaging the U-value for three areas: center of glass, edge of glass, and frame. They can give the glass assembly a better U-value with coatings. They can give the frame a better U-value with insulation. You can also get a better U-value with more frame and less glass.

Sometimes 1 + 1 = 2. Sometimes 1 + 1 can equal 3 or even 4.

So here's a tricky one. How would all these calculations pertain to a house made of steel shipping containers? Normal interior wood framing and insulation.