Two Views of Double-Stud Walls

musingsheader image

Two Views of Double-Stud Walls

John Straube and Jesse Thompson talk about double-stud walls and risk

Posted on Apr 7 2017 by Martin Holladay

At the recent BuildingEnergy 17 conference in Boston, there were at least two presentations that touched on double-stud walls. John Straube, a professor of building envelopeExterior components of a house that provide protection from colder (and warmer) outdoor temperatures and precipitation; includes the house foundation, framed exterior walls, roof or ceiling, and insulation, and air sealing materials. science at the University of Waterloo in Ontario, used his presentation to raise a warning flag, noting that “these walls will work if everything works — if there aren’t any defects — but they don’t work if there is something wrong.”

Jesse Thompson, an architect from Portland, Maine, was one of several presenters at a session called “Evolving Assemblies.” Thompson is clearly more of a double-stud fan than Straube. On many of his projects, Thompson said, discussions “keep coming back to the double-stud wall. We tell builders, ‘Let’s build a house, and then we’ll put an extra wall on the inside.’ It’s easy to explain.”

So who’s right? Is the double-stud wall risky or robust?

Measuring the moisture content of OSB sheathing

John Straube’s comments were made in a presentation called “Moisture Safe? The Writing on the Wall.” His co-presenter was Kohta Ueno, an engineer at Building Science Corporation.

Straube told the audience, “There has been a rise in interest in double-stud walls in the last decade. If labor is cheap, this is an inexpensive wall.”

Over the years, Straube has been involved in many research projects that show that the sheathing on a double-stud wall has a higher moisture content in late winter than the sheathing on a wall with continuous exterior rigid foam.

At the Boston conference, Straube described one of these research studies. (A report on the study cited by Straube can be found on pages 55 through 117 of The Hygrothermal Performance of Exterior Insulated Wall Systems by Trevor Trainor.)

Straube explained that at the University of Waterloo, “We built a test hut with seven different wall types. We had a 2x6 wall insulated with fiberglass batts; we called it the ‘datum’ wall or the ‘base case’ wall. We had a 2x8 wall filled with spray foam — that one was sponsored by the spray foam industry. One wall had 2.5 inches of exterior XPSExtruded polystyrene. Highly insulating, water-resistant rigid foam insulation that is widely used above and below grade, such as on exterior walls and underneath concrete floor slabs. In North America, XPS is made with ozone-depleting HCFC-142b. XPS has higher density and R-value and lower vapor permeability than EPS rigid insulation., and one wall had 2 inches of exterior polyisoPolyisocyanurate foam is usually sold with aluminum foil facings. With an R-value of 6 to 6.5 per inch, it is the best insulator and most expensive of the three types of rigid foam. Foil-faced polyisocyanurate is almost impermeable to water vapor; a 1-in.-thick foil-faced board has a permeance of 0.05 perm. While polyisocyanurate was formerly manufactured using HCFCs as blowing agents, U.S. manufacturers have now switched to pentane. Pentane does not damage the earth’s ozone layer, although it may contribute to smog. . We has a wall framed with I-joists. We had a double-stud wall that was 11 1/2 inches thick. We monitored the OSB sheathing temperature at multiple locations per wall.”

He continued, “When it comes to OSB temperature, the walls with exterior mineral wool and XPS were the best. The wall with polyiso was not quite as good, because polyiso doesn’t perform well in cold temperatures. It was operating as if it had an R-valueMeasure of resistance to heat flow; the higher the R-value, the lower the heat loss. The inverse of U-factor. of R-5 per inch.”

John Straube at the NESEA conference in Boston

[Photo credit: Martin Holladay]

What about the I-joist wall? “Some people think that a wall framed with 9 1/4-inch I-joists won’t have as much thermal bridgingHeat flow that occurs across more conductive components in an otherwise well-insulated material, resulting in disproportionately significant heat loss. For example, steel studs in an insulated wall dramatically reduce the overall energy performance of the wall, because of thermal bridging through the steel. as a wall framed with 2x10s,” said Straube. “We thought maybe — but not. The OSB has a higher density than a 2x10, so the thermal bridging through the I-joist is about the same as the thermal bridging through the 2x10s. But the I-joists cost twice as much.” (After the presentation, John Straube sent me an email in which he cited data from a paper authored by seven European building scientists to support this point; see Image #2, below, for a table from that paper.)

The OSB on the double-stud wall was colder and wetter than the other walls (see the graph immediately below). “The double-stud wall and I-joist wall had the coldest sheathing. All of the walls had OSB with a moisture content below 17% — but the OSB on the double-stud wall and the I-joist wall was the most wet. The driest OSB was on the walls with exterior mineral wool.”

Moisture content of the OSB on the north-facing wall

During the second phase of the study, researchers introduced a deliberate air leak through the test walls. “The double-stud wall and I-joist wall both picked up moisture when we introduced an air leak — more moisture than a simple 2x6 wall. The moisture content of the sheathing in the double-stud wall rose to 27% or even 30%.”

In his conclusions, Straube noted, “Double-stud walls have an increased risk of condensation wetting but the rain drying rate is only slightly reduced.”

At the end of his presentation, someone in the audience asked Straube, “Do you have any recommendations on what you should use for exterior insulation?”

Straube answered, “Yes. Use more of it.”

A time-tested wall

At his presentation, Jesse Thompson noted that energy-conscious builders have been building double-stud walls for at least 40 years. The wall assembly has been used for so long that Thompson’s first slide announced his topic as “Ye Olde Double-Stud Wall.”

Thompson noted that a lot of builders are worried about cold, damp sheathing problems. “People talk about the ‘cold sheathing epidemic,’ ” said Thompson. “Why is there all this talk about a deadly, dangerous wall that will kill you?”

As an architect, Thompson has designed many homes with double-stud walls. “We’re often working with builders who have never done a low-energy house before,” he said. “So why do we use this wall?”

“It’s easy to explain to builders. We can use this wall on a conventional foundation. We don’t often get to build on flat easy lots. We get rocky lots on a steep hill — not the kind of site where you would choose a raft slab. If it was easy, they wouldn’t have called you. With a double-stud wall, the exterior wall lands on a cold concrete foundation, the way the builders have always done it.”

On these jobs, the basement wall insulation goes on the interior. “If you install insulation on the exterior of a foundation, it’s difficult to prevent damage from weed-whackers.”

Double-stud walls don’t require exotic components. “We prefer to use readily available materials,” said Thompson. “We like the idea of an exterior air barrierBuilding assembly components that work as a system to restrict air flow through the building envelope. Air barriers may or may not act as a vapor barrier. The air barrier can be on the exterior, the interior of the assembly, or both.. You can tape a piece of sheathing you can get at the lumberyard. Or you can use Zip sheathing. You don’t need products form Europe or from shipping containers. You can use non-proprietary generic materials — materials that don’t have a registered trade mark symbol beside it.”

While double-stud walls require extra labor, that’s usually not a problem. “We work in Maine,” said Thompson. “Maine has embarrassingly low labor costs. We can get framers for under $15 an hour. I know these are terrible wages, and the framers can’t afford health care. I’m not saying this is a good thing. But if we want to avoid expensive materials, we can afford to throw more labor at the problem.”

The double-stud wall is time-tested. “This is not a revolutionary system,” said Thompson. “We’re not reinventing the wheel. We don’t want to do research. This is a battle-tested system.”

The rainscreen gap is essential

Ultimately, said Thompson, builders may be focusing on the wrong issues. “We spend a lot of time talking about walls,” said Thompson. “But we want to change the process, not the product. What matters is getting air barriers right. The good thing about exterior air barriers is that you can inspect them. It’s a process change.”

Summing up, Thompson explained that double-stud walls are simple. “It’s strapping plus Zip sheathing plus cellulose plus gypsum wallboard, without a vapor retarder,” he said. “The key is to have the rainscreen strapping. Without that, we would probably be in a lot of trouble. The rainscreen is the savior of the double-stud wall.”

[Credit for two above images: Kaplan Thompson Architects]

The system can always be tweaked. “You can use taped CDX instead of Zip sheathing. Plywood is tougher than OSB. If you want, you can specify a modern WRB — maybe a product from 475. We now have lots of smart vapor retarders to choose from: Intello, Siga Majpell 5, CertainTeed MemBrain, or kraft paper. So you can do it this way: Strapping plus WRB plus taped CDX plus cellulose plus a smart vapor retarder plus GWB. Or you can use a fluid-applied WRB instead of regular housewrap.”

Thompson provided some warnings. “Don’t do fiberglass batts,” he said. “Don’t use cheap OSB; use CDX, Advantech, or Zip. And remember, you need a rainscreen gap. And you need to use a blower door to test your assembly.”

Looking for the yeti

At the end of his presentation, Thompson returned to the question of risk.

“When [Maine builder] Dan Kolbert did some renovations on a house with 10-year-old Zip-sheathed double-stud walls, he found no problems,” Thompson said. “I think that double-stud wall sheathing failure is the yeti of building science. Everyone talks about it, but nobody has found one. If they found one, Martin would have written about it.”

The best candidate for the yeti award is probably the failing double-stud walls described by Jim Bradley in a JLC article titled “Major Surgery for a Failing Fat Wall.” The article describes repairs at a 1980s house where the cellulose insulationThermal insulation made from recycled newspaper or other wastepaper; often treated with borates for fire and insect protection. had been installed at such a low density that the insulation slumped.

Point / counterpoint

I’ve chosen to describe these two conference presentations in reverse order. In fact, Jesse Thompson made his presentation before John Straube. A few conference attendees were at both sessions. At Straube’s presentation, an audience member raised his hand and said, “At another session, I heard that a double-stud wall is the best way to go.”

Straube responded, “I don’t know why they said that. I bet they didn’t measure anything. These walls are more prone to failure.” When I emailed Straube after the conference, he noted that he had an addendum to his “prone to failure” statement: “But double-stud walls can be made low risk with small changes in design, good materials, and care during construction.”

Martin Holladay’s previous blog: “What’s the R-value of Cedar Shingle Siding?”

Click here to follow Martin Holladay on Twitter.

Tags: , , , , , , ,

Image Credits:

  1. Image #1: Peter Amerongen / Riverdale Net Zero
  2. Image #2: Thomas Haavi, Bjørn Petter Jelle, Arild Gustavsen, Steinar Grynning, Sivert Uvsløkk, Ruben Baetens, and Roland Caps
  3. Image #3: Eco Village, Ithaca, New York
  4. Image #4: Paul Biebel, Prudent Living Homes
  5. Image #5: Carter Scott

Apr 7, 2017 8:37 AM ET

John is right
by Kaplan Thompson Architects

Just to be clear (never hurts on the internet...), I agree with John Straube.

The double stud wall certainly isn't for all project types and John sees a far wider range of climate zones and building types than we do. And to top it all off, he's an actual real live scientist doing monitoring and research, which we certainly are not.

Double stud has a great history in New England custom single family home building so far. 1 - 2 story buildings with small stack effect at work, careful builders who can quickly learn to install their air barrier as carefully as they do their finish work, and a wide spread network of skilled dense-pack cellulose installers seem to be a good mix for this construction system to be built successfully.

Remove any of these characteristics and your risk certainly rises. If I was John and had to give recommendations to Code officials and builders all across the country, I would certainly say different things from what I said at the Building Energy Conference.

Jesse Thompson
Kaplan Thompson Architects

Apr 7, 2017 8:54 AM ET

Reply to Jesse Thompson
by Martin Holladay

Thanks for your comments. And thanks for your contributions, here in New England and elsewhere up and down the Atlantic coast, to improved details for superinsulated homes.

-- Martin Holladay

Apr 7, 2017 10:44 AM ET

So, what ARE the most forgiving optimally-insulated systems?
by Skip Harris

I'd love to see a consensus (or even a discussion) for roof and wall systems, for each climate.

Apr 7, 2017 11:27 AM ET

I-studs with OSB are better than the ones in the test
by Charlie Sullivan

To me, the most surprising thing in this article was the analysis that seemed to indicate that I-studs have about as much thermal bridging as regular studs.

The paper ( available at ) includes more details: They compared 36-mm thick wood studs (1.42") to I-studs with 8-mm thick (5/16") fiberboard webs. The thermal conductivity they measured in the fiberboard is 0.38 W/(mK). And they estimated 0.1 W/(mK) in the studs, based on moisture content. Based on those numbers, the thermal bridging is about 18% worse in the wood studs than in the fiberboard webbing, and the difference gets a little smaller when you consider the 3-dimsional multi-material I-stud.

However, the typical value most manufacturers publish for OSB thermal conductivity is 0.13 W/(mK). So it appears that this problem with I-studs is only really true of fiberboard I-studs and not ones you can buy in North America made with OSB. I'm not sure whether you can buy I-studs in Europe made with fiberboard; the ones used in the paper were assembled by the authors. But I'm not sure how carefully the thermal conductivity of OSB has been measured--it would be good to know that better for an assembly where it matters significantly.

Apr 7, 2017 11:56 AM ET

In a cold climate, it's going
by Jon R

In a cold climate, it's going to be warm sheathing and breathable to the exterior (eg, rock wool).

Apr 7, 2017 11:56 AM ET

Fiberboard sheathing for double-studwalls in cold climates,
by Dana Dorsett

With 16" o.c. framing 3/4" asphalted fiberboard is sufficiently structural for most wind loading, and strapped by 1x4 girts 16" o.c. to provide the back-ventilation for siding it should stand up pretty well to dense packing too. (Don't try this with half-inch fiberboard, wider stud spacing, or without girts.) The insulating properties of fiberboard add another R1.5 to the whole-wall performance too, but the real reason to use it is the high vapor permeance and high moisture tolerance.

In Europe 2" low density fiberboard sheathing is available good for about R5-R5.5, but it's not adequate as structural sheathing (the walls would need let-in bracing or shear panels), and it's pretty pricey stuff even in Europe, let alone the US. One version is described in this GBA blog:

also here:

I've seen one example of a UK level 6 (= ultra-sustainable) house covered in popular media using double-studwall /cellulose with 2" fiberboard sheathing, but in the video of the Grand Designs television episode covering that project it was clearly not dense-packed. (That episode currently available on Netflix, but frustratingly thin on some details.)

On this side of the puddle 3/4" asphalted fiberboard isn't necessarily more expensive than better grades of OSB, and can be structural if the fastener schedule is followed (though you might choose to still use shear panels or let-in bracing, and fewer nails on the fiberboard), eg:

Apr 7, 2017 1:22 PM ET

Edited Apr 7, 2017 1:23 PM ET.

should I be worried?
by Brian Croston

I've been lurking here for a long time, but still no clear answers regarding double stud walls. I am an owner-builder, starting my project in a few months.

I am in southern Montana (7000 HDD, -10*F & +91*F design temps) and had planned a wall construction similar to Jesse Thompson's design (1/2" air gap + Tyvek Drainwrap + taped CDX + cellulose + Intello Plus + 1.5" utility chase w/Rockwool infill + GWB), but want to make sure it will work in my climate. Most of the discussion and testing seems to have occurred in New England, with significantly different levels of humidity.

The simplicity of my proposed wall design and significantly reduced construction cost are why I'm wanting to go in this direction. Here in Montana, items like Rockwool, asphalt fiberboard, and gypsum board are still considered exotic and expensive and I want to steer away from foam where possible.

Apr 7, 2017 2:17 PM ET

Alternative to ext foam
by Jim Baerg

We built double stud houses in the 80's. Based on many years of discussions at GBA and BSC, I thought that the consensus was to stay away from fat walls with exterior sheathing. I'm surprised at the continued popularity of traditional double stud walls. Ease of construction seems to be one argument.
I'm also wondering why there is no mention of using exterior D.P. cellulose. If exterior mineral wool works so well, why wouldn't cellulose? We've done a number of houses this way in Montana and builders adapt well to it.


Apr 7, 2017 2:59 PM ET

The drier air of Montana makes it easier @ Brian
by Dana Dorsett

The drying capacity toward the exterior during the late winter / early spring is going to be more favorable in high-dry MT climate than New England. As temps warm up in MT the outdoor dew points and outdoor relative humidity remain low, whereas in New England the outdoor dew points & RH track a bit higher in the early part of the drying season, which slows the drying rate of the cold sheathing. f you're still nervous, run a WUFI simulation for your location but it should be fine to clone a Jesse Thompson double-studwall stackup.

Intello works, but it has to be made air tight. An OSB (the cheap stuff is fine here) or CDX air barrier at that location would be more robust, at a similar varying vapor retardency. OSB will be somewhat more vapor-tight than CDX, and may be the better choice for the interior side air-barrier & vapor retarder if you go that route. (An OSB air barrier/vapor retarder in a layer near the interior is found in a number of PassiveHouse designs.)

Fiberboard can be ordered through box stores (HD, Menards, etc) , if nobody else carries it, but if you're dense packing you'll need at least 1x girts (horizontal furring, not vertical, which limits some siding options) for a 3/4" rather than 1/2" air gap. IIRC half-inch fiberboard has better structural specs than 3/4" stuff (better fastener retention?) and it's cheap- under $10/sheet (half the price of ZIP OSB.)

Apr 7, 2017 3:03 PM ET

response to Jim
by Brian Croston

Jim, I've also considered the wall construction you propose, using OSB/plywood on the interior stud wall and reinforced exterior WRB such as Mento Plus on the outside. Based on the few houses I've built, I'd think it's more difficult (dense pack from exterior, additional strapping, installing both insulweb + housewrap, etc.) to build.

I'll let others with more experience comment on other reasons why builders continue to choose exterior sheathing.

Apr 7, 2017 3:49 PM ET

Edited Apr 7, 2017 3:50 PM ET.

Appropriate timing for this
by David Powers

Appropriate timing for this article. Just yesterday I did some moisture testing in a deep energy retrofit/large addition my company did in Sutton Vt in 2013-2014.

Wall system is GWB, 12" double stud, dense pack cellulose, Zip sheathing, 6mm home slicker and cedar clapboards. Sheathing MC was in the high 20s on the north and south sides. Higher up the wall the higher the MC. Unvented 18" roof assembly with dense pack cellulose was surprisingly in the low teens MC. I am tempted to open up a wall for a more thorough inspection.

Since this project, we have built two other double stud homes but used an interior smart vapor retarder and service cavity on each. Plan to test these homes in the near future. Sure hope to see dryer results.

Apr 7, 2017 5:26 PM ET

Martin So if....
by Kye Ford

So if cold sheathing is in fact ok for double stud walls as many contend, where does this leave us for recommendations for continuos exterior insulation values for other wall assemblies?

Apr 7, 2017 6:34 PM ET

Response to Kye Ford
by Martin Holladay

The recommendations for minimum R-values for exterior rigid foam apply to walls that can't dry to the exterior.

Double-stud walls can dry to the exterior.

-- Martin Holladay

Apr 7, 2017 10:08 PM ET

Edited Apr 7, 2017 10:09 PM ET.

Free double stud wall.

We are building double stud walls, floor and roof like a truss, all as one frame, no plates. Essentially taking the labor equation out of the double stud, reducing cost to almost zero and the part I like: no hairy eyeball from contractor. I order the frames from truss company and say here build this. Its patent pending. WE have sheathing both sides of wall for strength then a finish of gyp or just use nice plywood. We spray foam from inside of outer sheathing layer to approx 2 inches from inner sheathing leaving a zero cost electrical chase, electric boxes are inside the heat envelope. Exterior wall has slight pitch to help break the box look.

Wall is 12 inches thick at bottom and 8 or 9 at top but can be wider. . We are building small modular s with them. With walls, floor and roof framed at once its very fast.

Tim McCarthy Utility Patent Legal Zoom Order 47527295.pdf 167.3 KB

Apr 8, 2017 5:16 AM ET

Response to Tim McCarthy
by Martin Holladay

That's a truss-framed wall, not a double-stud wall.

It looks like you give your clients two choices: either have an exterior wall that isn't plumb, or an interior wall that isn't plumb. If I were offered that choice, I'd say, "No, thanks."

-- Martin Holladay

Apr 8, 2017 6:24 AM ET

Edited Apr 8, 2017 6:26 AM ET.

Response to Skip Harris (Comment #3)
by Martin Holladay

Q. "So, what ARE the most forgiving optimally-insulated systems? I'd love to see a consensus (or even a discussion) for roof and wall systems, for each climate."

A. When it comes to walls, my advice (tailored to different climate zones) is provided in this article: How to Design a Wall.

For ceilings, I'm a firm believer in vented unconditioned attics with a thick layer of cellulose on the attic floor. This approach requires raised-heel trusses or rafters installed above the joists.

For sloped roof assemblies, the best assemblies require an adequate layer of rigid foam above the sheathing, as explained in this article: How to Install Rigid Foam On Top of Roof Sheathing.

-- Martin Holladay

Apr 8, 2017 6:31 AM ET

Response to Charlie Sullivan (Comment #4)
by Martin Holladay

I have emailed your comment on the U-factor of I-studs to John Straube. I hope he responds, because this is certainly a question that interests me as well.

-- Martin Holladay

Apr 8, 2017 6:37 AM ET

Response to Dana Dorsett (Comment #6)
by Martin Holladay

GBA has published several articles about walls with fiberboard sheathing that have been insulated with dense-packed cellulose or dense-packed fiberglass. Many of the builders who have tried this technique have had problems with bulging fiberboard, so builders interested in fiberboard sheathing should (a) choose their product with care, and (b) research the bulging issue, and solutions to the bulging issue, before proceeding.

For more information on this topic, see Wall Sheathing Options.

-- Martin Holladay

Apr 8, 2017 6:42 AM ET

Edited Apr 8, 2017 6:43 AM ET.

Response to Brian Croston (Comment #7)
by Martin Holladay

Q. "Should I be worried?"

A. In my opinion, no. I agree with Jesse Thompson's statement: "The rainscreen is the savior of the double-stud wall.”

Other important details that help keep these walls out of trouble are:

1. The use of plywood sheathing instead of OSB sheathing.

2. Airtight construction techniques, verified with a blower door.

3. The use of a smart vapor retarder on the interior side of the wall.

-- Martin Holladay

Apr 8, 2017 6:47 AM ET

Response to Jim Baerg (Comment #8)
by Martin Holladay

Q. "I'm also wondering why there is no mention of using exterior dense-packed cellulose. If exterior mineral wool works so well, why wouldn't cellulose?"

A. GBA has published lots of articles on walls that include dense-packed cellulose on the exterior side of the wall sheathing. Here are two examples:

All About Larsen Trusses

The Klingenberg Wall

The second article includes many links to more GBA articles on this topic.

-- Martin Holladay

Apr 8, 2017 6:53 AM ET

Edited Apr 8, 2017 6:54 AM ET.

Response to Dana Dorsett (Comment #9)
by Martin Holladay

You wrote, "Intello works, but it has to be made airtight."

That statement is only true if the builder is depending on the Intello to be an air barrier. If the builder chooses to install Intello as a smart vapor retarder, it will perform very well in that capacity even if it isn't installed in an airtight manner.

Of course, every wall needs at least one (and sometimes two) air barriers, but a variety of materials can be used for this purpose, including the exterior sheathing and the interior drywall.

(I suspect that you know all these facts, Dana, but it's important to clarify them for GBA readers.)

-- Martin Holladay

Apr 8, 2017 7:01 AM ET

Response to David Powers (Comment #11)
by Martin Holladay

First of all -- hi, neighbor. (It's always good to hear from GBA readers from the Northeast Kingdom).

Thanks for sharing your moisture content data. The readings are consistent with the findings of academic researchers. In general, sheathing with a high moisture content in February or March tends to start drying out in April or May. This type of annual moisture cycling is probably more concerning in OSB sheathing than plywood sheathing.

While the long-term effects of this type of annual moisture cycling are unknown, double stud walls that experience this type of cycling appear to be doing OK, as long as the drying that occurs in April and May can occur rapidly. A ventilated rainscreen gap (with openings at the bottom of the wall and the top of the wall) really help.

-- Martin Holladay

Apr 8, 2017 3:28 PM ET

new systems defy categorization

Martin The sloped wall is an option. The current model Im working on neither wall is sloped or what you call "not plumb". Art does play a part in successful buildings. New systems transcend definition or categorization. My system has full thermal break between the inner frame member and the outer frame member, the most important aspect of the double stud wall and its done WITHOUT THE COST. High Performance housing has to think out of the box, reduce cost, otherwise it will remain as it is, a very fringe percentage whats being built.

Apr 8, 2017 3:43 PM ET

Batts rather than Cellulose
by Malcolm Taylor

John Straub raises concerns about substituting batts for the more common cellulose in double walls. Does anyone have a feeling for how much more vulnerability this represents if the other precautions (rain screen gap, plywood sheathing, good air-sealing) are taken?

Apr 8, 2017 3:49 PM ET

Basic Question
by Malcolm Taylor

Does eliminating or moving the sheathing inwards on a double-stud wall completely remove the risk? Is the remaining exterior wall framing a condensing surface subject to smaller but still possible damage, or at that point am I asking how many angels can fit on the head of a pin?

Apr 8, 2017 4:48 PM ET

Response to Malcolm Taylor (Comment #24)
by Martin Holladay

It was Jesse Thompson, not John Straube, who emphasized that cellulose is preferred to fiberglass batts for a double-stud wall. I imagine that the main reason to use cellulose is that it cuts down on air leakage, making the wall less susceptible to moisture damage.

That said, there are other ways to cut down on air leakage, and fiberglass batts can work well if perfectly installed.

-- Martin Holladay

Apr 8, 2017 4:54 PM ET

These sort of systems are discouraged through taxes.
by user-6717844

Double stud walls take large amounts of floor space, and building footprints grow along with wall thickness. When the building footprint grows, taxes grow along side. Cities are discouraging sustainable construction by punishing owners with higher taxes.

Apr 8, 2017 5:20 PM ET

by Dan Kolbert

I remember Jesse saying that at the conference, but I don't remember compared to what.

We stick pretty much entirely to double stud walls. For me at least (and we keep framing in-house, and I'm paying a lot more than $15/hr), double stud is cheaper than the alternatives. Exterior foam is expensive and fussy to install. It requires additional labor and material to prepare for siding. And you still need to insulate the walls. The biggest expense with dense pack cellulose is the prep - filling a 12" thick wall is a lot cheaper per unit than a 6" wall.

Exterior mineral wool has the same problems, in spades.

I've priced out exterior i-joists but never done it. Between the price of the material, and the difficulty of working vertically on the exterior of a building, my budget was much higher than for double stud. Perhaps in a factory setting like EcoCor's, the numbers are better, but not for a site built house, in my opinion.

As for the cold sheathing problem, I just haven't seen it. As Jesse pointed out, the very first double stud house we built (with his firm, back in 2008) was recently renovated. We turned a relatively unfinished 3rd floor into a studio, and in the process replaced a window with a door. It was the north side of the house, facing the ocean. We saw zero evidence of any wetting or deterioration of the sheathing.

Of course the stakes are higher with well-insulated buildings. Caution and careful detailing is critical. But even Straube seems to admit (and I wasn't at his session) that the dangers aren't as grave as BSC's documents would lead you to believe.

Martin & I have discussed the perils of the "expert-ization" of green building. One way to resist it is precisely thru sessions like the one Chris Briley moderated at NESEA - to compare real-world experiences of various wall sections.

Apr 8, 2017 5:30 PM ET

Air barrier in double stud wall
by stephen sheehy

Our exterior wall was sheathed with Advantech, with all joints taped. But our primary air barrier was Siga Majpel, applied to the outside of the inner stud wall. By running the membrane a foot or so over the top plate and outside the ends of the framed walls, it was easy to lap the ceiling membrane (stapled to the bottom truss chord) over the wall membrane and to overlap at the wall corners.

Our contractor had not done a double stud wall before, but found it pretty easy to execute. By using a membrane on the outer side of the interior wall, we were able to run plumbing and wiring inside the inner stud space with minimal penetration of the membrane. Insulating the space between the sheathing and membrane, about 8 1 /2" , was simple. We just cut holes in the membrane, blew the cellulose in and taped the holes.
Jesse and his colleague, Jamie Broadbent, did the design.

Apr 8, 2017 6:18 PM ET

by Malcolm Taylor

That argument has come up before and I'm not sure the impact is large enough to have any effect. Increasing the width of the exterior walls from 6" to 1'-0" would add about 2.5% to the floor area of a 1600sf house. And given that property taxes are based on the assessed value of a house, not just its floor area, I can't see there being much of a noticeable change.

Apr 8, 2017 6:22 PM ET

by Malcolm Taylor

Sorry, you are quite right. Is it just the ability of the cellulose to reduce air leakage, or does it's ability to absorb moisture help too?

Apr 8, 2017 6:33 PM ET

Dan Kolbert
by Malcolm Taylor

The sheathing is dry. That is very reassuring to hear - and it's what you had been predicting all along.

What do you see as a viable alternative to the "expert-ization"? The counteracting force is I suppose market-driven short-cuts which here in the PNW lead to widespread building failures. Do you think more robust building codes and the use of a few trusted assemblies for each climate are the answer?

Apr 8, 2017 7:20 PM ET

Response to Charlie Sullivan (Comment #4)
by John Straube

Hi Charlie,
I think there is just a confusion about language. They use the term "fibreboard" for the web. Fibreboard in the US would often be low-density fiberboard (an slightly insulating product) that could never be used as a web because it is too weak.

OSB is what we would use in the US, and that is also very common for I-joist in Sweden today, although tempered hardboard (what I assume they made for the paper, as that is more what the photo seemed to show) is also very common, more in the past.

See for the largest I-joist manufacturer in Sweden (who use 10 mm OSB webs whereas 9.5 mm is more common here) For solid wood, thermal conductivity across the grain (what is normally measured) is in the 0.1 W/mKrange but routinely varies from 0.08 to 0.12 : this is why people who do hotbox testing routinely test the materials used. That is what our lab does. Along the grain, the thermal conductivity is higher.

For OSB, across the panel, a value of 0.13 is reasonable (we have measured this in our lab), because the OSB is compressed and hence higher density. But higher density products have higher thermal conductivity.
The thermal conductivity of wood along the grain is about twice that across the grain (a value of 0.22 is often used) The thermal conductivity produced by the manufacturer is in the direction perpendicular to pressing direction: the web in an I-joist is parallel to the direction of pressing. This makes a big difference to the conductivity.

Hence, while initially surprised by the results, when I dug into their results I found them both plausible and relevant to North American practice.

There is a reason researchers need to measure real walls: we often don't know as much about this stuff as people think, and there are a lot of variables.

It would be great if we could get funding to measure this stuff… but really it does not change much, because both physics and measurements show that I-joist walls have other limitations, and they are not cheap!
The quote from the paper, describing the “fibreboard plates” may useful for others.

"It is common to measure the thermal conductivity through the thickness of fibreboard plates, but the thermal conductivity in the longitudinal direction is not well known. The thermal resistance in the longitudinal direction, i.e. the direction of the heat flow through the I-stud and the U-stud, was therefore measured in a heat flow meter apparatus according to the governing standard (NS-EN 12667 2001). The corresponding thermal conductivity λfb║ = 0.38 W/(mK) were used in the numerical simulations.”

-- John Straube

Apr 9, 2017 4:49 AM ET

Edited Apr 9, 2017 4:50 AM ET.

Response to Malcolm Taylor (Comment #31)
by Martin Holladay

Q. "Is it [the advantage of cellulose over fiberglass] just the ability of the cellulose to reduce air leakage, or does its ability to absorb moisture help too?"

A. That's a complicated question. After researching the purported value of hygric buffering and hygric resistribution, I'm somewhat of a skeptic when it comes to the purported benefits of hygric buffering. If there is a mechanism that allows the moisture content of a building assembly to ratchet up, the building assembly will eventually fail unless there is a periodic opportunity for the building assembly to dry. On an annual basis, it is always essential for the drying rate to exceed the wetting rate.

Here is a link to my article on the topic: Hygric Buffering and Hygric Redistribution.

The paragraphs I quote below came from that article.

Most building scientists agree that some wall assemblies that seem risky are made safer by hygric redistribution. The classic example is a double-stud wall insulated with dense-packed cellulose. In a cold climate, moisture tends to accumulate on the cold side of this type of wall during the winter months, leading to damp wall sheathing. Yet when these walls are disassembled and inspected, the wall sheathing is almost always in good shape. The lack of mold or rot is often attributed to two factors: hygric redistribution by the cellulose (which pulls moisture from the sheathing and redistributes it toward the center of the wall) and outward drying in April and May.

That said, it’s not as if builders can cut corners with water-management details, and glibly announce, “We’re fine — the cellulose provides hygric redistribution, and that will keep us out of trouble.” So hygric redistribution may be one of those phenomena which is interesting and worth studying, but is so hard to model that it is useless as a design principle.

Moreover, there is no easy way to compare the advantages and disadvantages of blown-in fiberglass insulation with the advantages and disadvantages of cellulose insulation. Water is more likely to drain quickly through fiberglass than cellulose, and that’s good — right? And damp fiberglass dries more quickly than cellulose — also good, right?

But on the other hand, cellulose provides both a hygric buffer and hygric redistribution — characteristics that are absent from fiberglass insulation — so the cellulose must be preferable, right?

Assessing the effects of these pluses and minuses is difficult even for building scientists. To weigh all the relevant factors, field observations may be just as valuable as, or more valuable than, hygrothermal modeling.

In our recent phone conversation, [Joseph] Lstiburek noted that cellulose insulation can't perform miracles. “A mass wall built out of several wythes of brick is a spectacular example of hygric distribution,” Lsitubrek told me. “The rainwater penetrates, is wicked away, stored, and redistributed in a material that is not water-sensitive. But hygric redistribution doesn’t work that way with cellulose. Yes, the cellulose wicks and it stores moisture, but the moisture all ends up on one side of the wall because of the thermal gradient. The hygric redistribution buys you something, but it doesn’t buy you enough to keep you out of trouble. It you have a window that leaks, cellulose will not save you compared to fiberglass insulation.”

-- Martin Holladay

Apr 9, 2017 7:54 AM ET

Edited Apr 9, 2017 7:54 AM ET.

"Water is more likely to drain quickly through fiberglass"
by Skip Harris

Really? This might be more akin to saying a bigger hole in the bottom of my boat will allow the water to drain out faster when I bring it ashore after it sinks. If my wall ever needs to actually drain water, something is SERIOUSLY wrong.

Apr 9, 2017 11:16 AM ET

John Straube (I-joist webs and anisotropic thermal condutivity)
by Charlie Sullivan

Re: comment 33 by John Straube: Thanks for sorting out some more of those details. Putting the numbers together I would estimate that the advantage of a typical commercial I-joist over a regular stud is probably significantly more the 15% found in the paper, and would be worthwhile if those were the only two options on the table, but as you say, those are rarely the two best options.

The higher thermal conductivity along the grain is a useful thing to keep in mind, in considering, for example, why rafter tails can be such severe thermal bridges.

Apr 9, 2017 11:22 AM ET

Wrong Question?
by Andy Kosick

Really appreciate hearing about the research and recommendations here but I can't help but think when reading about wall assemblies these days that what we're really talking about is OSB. If it's wet, were it's at, etc. Is the elephant in the room actually an industry wide obsession with an inferior building material, something that's too risky to be in modern high R wall and and unvented roof assemblies. Many solutions seam to involve CDX, or Advantech, or eliminating OSB form the assembly all together. Despite that last one being a personal favorite, I don't mean getting rid of structural sheathing. My question is does the standard for what constitutes the minimum properties for structural sheathing in a residential wall assembly need to change. Are market prices made possible by using an inferior material making it overcomplicated to move forward with high R assemblies. I'm just thinking out loud here, any thoughts?

Apr 9, 2017 2:16 PM ET

by Malcolm Taylor

It would be interesting to have some idea of how much more resilient to continued high moisture cycling plywood is than OSB. We know it's better, but from my experiences with rotten sheathing it isn't immune to the same problems when wetted repeatedly. High moisture levels in exterior walls have other effects too. Fasteners, electrical connections, framing and other components also suffer. Is a wet wall where the sheathing maintains its structural properties a good idea? I'm also just thinking out loud.

Apr 9, 2017 5:15 PM ET

Edited Apr 9, 2017 7:48 PM ET.

ZIP and vapor retarder
by Ethan T ; Climate Zone 5A ; ~6000HDD

Quick question... The diagrams suggest that ZIP sheathing eliminates the need for a vapor retarder. Am I reading this correctly?

Apr 9, 2017 5:50 PM ET

If one is going to add
by Jon R

If one is going to add furring strips, is there a reason that Densglass (or maybe DensElement) isn't the standard for double walls?

Apr 10, 2017 12:31 AM ET

by Malcolm Taylor

Once you move from sheathing with plywood or OSB that can provide shear resistance and a solid backing for trim and flashing, is there any real advantage to having sheathing at all? Does fibreboard or something like Densglass offer anything that house wrap alone wouldn't?

Apr 10, 2017 6:55 AM ET

Edited Apr 10, 2017 6:57 AM ET.

Response to Andy Kosick (Comment #37)
by Martin Holladay

You wrote, "What we're really talking about is OSB."

Every building material has appropriate uses and inappropriate uses. Sometimes, asphalt felt makes sense as a roofing underlayment; other times, you really need Grace Ice & Water Shield. That doesn't mean that asphalt felt is a bad product.

Designers have to specify the right product for the assembly they are building.

OSB performs well as long as it stays dry. GBA does not recommend the use of OSB on the exterior side of double-stud walls. For example, you can find this GBA recommendation in my article, How to Design a Wall.

For more information on this topic, see Wall Sheathing Options.

-- Martin Holladay

Apr 10, 2017 7:04 AM ET

Response to Ethan T (Comment #39)
by Martin Holladay

Q. "The diagrams suggest that Zip sheathing eliminates the need for a vapor retarder. Am I reading this correctly?"

A. The diagrams show wall assemblies used by Kaplan Thompson Architects (the company founded by Jesse Thompson and Phil Kaplan). GBA isn't endorsing these drawings, although they look fine to me.

An interior vapor retarder is required by most building codes. You can satisfy this requirement with a variety of approaches, including vapor retarder paint or an interior "smart" membrane like Intello or CertainTeed MemBrain. In my opinion, including a smart vapor retarder (one with variable vapor permeance) makes a double-stud wall more robust.

In my article on walls (How to Design a Wall), I recommended the use of a smart vapor retarder on the interior side of a double-stud wall.

-- Martin Holladay

Apr 10, 2017 7:13 AM ET

Edited Apr 10, 2017 7:14 AM ET.

Response to Jon R (Comment #40)
by Martin Holladay

Q. "If one is going to add furring strips, is there a reason that Densglass (or maybe DensElement) isn't the standard for double walls?"

A. As far as I know, there is no standard for double-stud walls. But GBA recommendations for sheathing choices for double stud walls have always included DensGlass or other brands of fiberglass-faced gypsum panels.

For example, in my article on walls (How to Design a Wall), I recommended that double-stud walls should have "any type of exterior sheathing material other than OSB — for example, plywood, diagonal boards, fiberboard, or fiberglass-faced gypsum panels."

In my article titled How Risky Is Cold OSB Wall Sheathing?, I wrote, "OSB is more susceptible to rot than plywood. So if you’re worried about the durability of your sheathing, choose plywood, DensGlass Gold, or diagonal board sheathing over OSB. One other possible (vapor-permeable) sheathing choice is structural fiberboard sheathing."

-- Martin Holladay

Apr 10, 2017 9:48 AM ET

> Does fibreboard or
by Jon R

> Does fibreboard or something like Densglass offer anything that house wrap alone wouldn't?

My understanding is that Densglass provides structural support for racking and is stiff enough for dense pack cellulose (unlike fiberboard). And can't rot like OSB/Zip or even plywood. I see 5/8" sold for $18/sheet.

So when I see "use CDX, Advantech, or Zip", I think "but which is best" and "where does Densglass fit in".

"These walls are more prone to failure... but double-stud walls can be made low risk” - this comment makes me wonder "can it be made lower than a insulation-outside-the-sheathing wall".

Apr 10, 2017 10:47 AM ET

Response to Malcolm Taylor (Comment #41)
by Martin Holladay

Q. "Does fibreboard or something like Densglass offer anything that housewrap alone wouldn't?"

A. Yes. Structural fiberboard and DensGlass can both be used as structural wall bracing. For more information, see Wall Sheathing Options.

-- Martin Holladay

Apr 10, 2017 11:48 AM ET

by Malcolm Taylor

Looking more closely at the spec sheets for various brands of fibreboard and Densglass they can't be used as a fastening surface for siding or trim, and the shear strength values they use to call them "structural" are too low to meet our code.

They are used primarily on commercial projects where the cladding systems and structures don't require these attributes. and their shortcomings are why they haven't penetrated the wood frame residential market.

Apr 10, 2017 2:46 PM ET

Response to Malcolm Taylor
by Martin Holladay

As the article I linked to (Wall Sheathing Options) clearly noted, neither fiberboard nor DensGlass is designed or intended to hold siding fasteners. All I said in my comment is that these sheathing products can be used to brace walls. Housewrap can't.

While structural fiberboard and DensGlass are routinely used to brace residential walls in the U.S., any bracing plan should be checked by an engineer. In high wind areas, earthquake zones, or for use on a multi-story building, a different type of sheathing might well be called for.

If these products can't be used to brace residential walls in Canada, that is news to me. Thanks for letting us know.

-- Martin Holladay

Apr 10, 2017 8:50 PM ET

by Malcolm Taylor

No, there are regions where both insulated fibreboard and gypsum based sheathing can be used to resist shear (although on where I build). My point was a bit different.

I see switching to these materials with marginal structural values in much the same way I look at Advanced Framing. You are taking a robust, established way of building and reducing the structure as much as possible - removing the redundancy that makes it resilient - and this is done in the name of efficiency.

This seems like looking at building assemblies as a zero-sum game. I can't see how it makes sense for high performance houses to have less structural integrity simply because they are designed to use less energy. Surely they should have at least as robust a structure as their mass market cousins. If the wall assembly can't safely support plywood or OSB, any alternative needs to perform the roles we traditionally have asked sheathing to do. Otherwise why have it?

Apr 10, 2017 9:37 PM ET

higher perm sheathing
by william murray

We're just wrapping up construction of a PHIUS house here in southern Quebec. We are not keen on OSB. Having seen it in soggy shreds in countless renovations, it doesn't inspire a lot of confidence in our team. "Performs well when it's dry" says it all. We sheathed the house in Eco4, a fibreboard made here in Quebec. At 1-1/2", it was more than rigid enough to resist the bulge of 17" of dense pack cellulose. It is essentially three sheets of 1/2" fibreboard glued together with a wax emulsion. It's a far cry from Agepan — no sexy tongue & groove and the edges don't always match up but with a perm rating of 26 and its little extra R-4, we were more than willing to put up with it. I don't know what kind of racking rating Malcolm needs in BC but it definitely fit the bill here in frosty Quebec.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!