GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Musings of an Energy Nerd

Rainscreen Gaps and Igloos

A report from the 2013 Westford Symposium on Building Science

Achilles Karagiozis and John Straube watching a funny cat video on Achilles' laptop.
Image Credit: Martin Holladay
View Gallery 7 images
Achilles Karagiozis and John Straube watching a funny cat video on Achilles' laptop.
Image Credit: Martin Holladay
Armando Cobo's equations have John Straube stumped.
Image Credit: Martin Holladay
Some of the conference attendees hang out near the beer coolers in Joe and Betsy's backyard.
Image Credit: Karyn Patno
I stayed at the Westford Regency Hotel. The bathroom exhaust fan made a lot of noise, but its duct wasn't connected to the outdoors; the fan failed the toilet-paper test. Instead of fixing the broken bath exhaust fan, the hotel management just put up this sign.
Image Credit: Martin Holladay
Every evening, the Building Science Boogie Band played in Joe and Betsy's barn. Steve Sauer is at the keyboard; Collin Olson plays bass guitar; and Randy Martin is on drums.
Image Credit: Martin Holladay
Steve Sauer and Mike Lubliner entertaining the crowd in Joe and Betsy's barn.
Image Credit: Martin Holladay
GBA blogger Rob Moody plays the sax.
Image Credit: Martin Holladay

For the past 17 years, Joe Lstiburek and Betsy Pettit have hosted an annual conference, the Westford Symposium on Building Science, near their home in Massachusetts. Informally known as “summer camp,” the invitation-only gathering attracts hundreds of builders, engineers, architects, professors, and building science researchers.

The attendees listen to presentations at a conference center during the day and relax in Joe and Betsy’s backyard during the evening.

This year’s conference was held on August 5-7. Speakers included Michael Blasnik, Mike Steffan, Gary Klein, Kent Browning, Achilles Karagiozis, Chris Schumacher, Dave Ober, Joerg Birkelback, Marcy Tyler, Ed Retzbach, Andreas Holm, and Joe Lstiburek.

Covering familiar territory

If there were any devoted readers of GBA at the conference, they probably recognized the information provided at several of the summer camp presentations this year. At the Westford conference, Blasnik discussed the limitations of energy modeling software, a topic covered in a GBA blog published in March 2012. Blasnik also discussed bad energy retrofit advice, a topic which GBA has covered in at least three articles: A Plague of Bad Energy-Saving Tips, More Energy Myths, and How Not to Save Energy.

(The PowerPoint slides that Blasnik used in his presentation can be viewed by clicking one of the links at this Web page. The page also includes links to some of the other presentations from this year’s Symposium.)

Gary Klein, a renowned expert on domestic hot water systems, discussed several ways to reduce the time required for hot water to travel from a water heater to a faucet. GBA has reported on Klein’s calculations and advice in several articles, including Inefficient Hot Water Piping Layouts Waste Hot Water and All About Water Heaters.

Joe Lstiburek’s presentation focused on his proposal that ASHRAE 62.2 be replaced with a new residential ventilation standard. His arguments were reported in a June 2013 GBA article, How Much Fresh Air Does Your Home Need? GBA’s follow-up articles on this topic include Resistance May NOT Be Futile in the Residential Ventilation Wars and An Interview with Dr. Joseph Lstiburek.

Lstiburek’s proposed new ventilation standard was published this week, and is available online: Ventilation for New Low-­Rise Residential Buildings.

R-mageddon: The thermal metric project

Chris Schumacher and Dave Ober reported on research that is leading to the development of a new metric to replace R-value. (This year, Schumacher told the crowd that he wanted to call his presentation “R-mageddon.”) Much of the material presented by Schumacher and Ober was a reprise of presentations made at previous conferences. (To read GBA’s reports on these earlier presentations, see A Bold Attempt to Slay R-Value and Air Leakage Degrades the Thermal Performance of Walls.)

As he did last year, Schumacher shared some bar graphs showing the effects of high temperatures and infiltration on the thermal performance of wall insulation. Schumacher’s colleague, Dave Ober, proposed several fairly complicated ways to replace R-value with a new metric that takes these factors into account. One proposal called for using a line graph to report the new metric. Another proposal called for reporting the new metric in a very large table listing different R-values for different cities; according to this proposal, the R-value (or perhaps the “Joe-value” or “Schumacher-value”) for a fiberglass batt would be different in Omaha than in Boston.

A new use for WUFI

Achilles Karagiozis is the director of building science at Owens Corning in Ohio. His presentation was a delight. He started out with an amusing scenario for an imaginary TV reality show, involving a helicopter crash in Alaska with only one survivor, Joe Lstiburek. The only tool that Joe had to help keep him alive at -51°F was a laptop loaded with WUFI software. Karagiozis explained how Joe could use WUFI to design an igloo, assuring his survival — and incidentally proving the usefulness of the oft-derided hygrothermal modeling software that Karagiozis helped develop.

Once Karagiozis had completed his Alaskan fantasy, he discussed his efforts to model air flows through rainscreen gaps on walls. If there were any math-averse listeners present, they were probably having an anxiety attack. Karagiozis explained why it is difficult to model turbulent 3-D air flows; why air flow equations need a friction factor to account for roughness in laminar flow; how to calculate buoyancy pressures; when you need to consult Darcy-Weisbach equations; and when you need to perform a multivariate regression analysis.

If (like most of us) you have no interest in studying the Darcy-Weisbach equations, here’s all you need to know:

  • The main driving factor for air movement through ventilated rainscreen gaps in walls is solar gain.
  • Rainscreen gaps in walls are very effective at drying damp siding and sheathing.
  • Rainscreen gaps provide an important safety factor to handle the higher moisture loads in today’s high-R walls.
  • Karagiozis recommends that all walls should include a ventilated rainscreen gap.

Summing up, Karagiozis said, “Increasing levels of insulation are making us rethink moisture transport in walls. We need to think harder. We need to design. To withstand the moisture loads, we need to passively ventilate walls to increase the drying potential.”

Placebo tune-ups

Michael Blasnik is an energy consultant from Boston, Mass., who is renowned for his skill at analyzing large datasets. Here are some quotes from his presentation.

Michael Blasnik: “I went to the Home Energy Saver website and entered information on my house. The model said that I will save 24% of my heating energy by insulating my basement ducts. But studies show that measured savings are on the order of 1% to 3%. The model told me that if I seal my basement ducts, insulate the ducts, and insulate the floor above the basement, I will save 41% of my heating bill!”

“A SEER 13 air conditioner uses less energy than a SEER 10 air conditioner. But a SEER 17, 19, or 20 air conditioner doesn’t really perform as well as the SEER numbers would lead you to believe.”

“Cooling load is not driven by delta T. Cooling load is driven by solar gain through windows, solar gain through the attic, duct efficiency, and internal gains.”

“Here is a list of things that are hard to model well: Foundation heat loss, infiltration, wall and attic heat loss, window loss and gain, and HVAC performance. Everything else we’ve got nailed down.”

“If you have a vented attic, most of the ventilation air isn’t coming from the soffit vents. Most of the ventilation air comes from the house to the attic, and then out.”

After Blasnik explained that annual tune-ups of gas furnaces don’t save any energy, Henry Gifford commented, “You can’t tune up a gas furnace. There is nothing to adjust.” Blasnik responded, “Right. It’s a placebo measure. There is nothing to do — but pay the guy $180 for the visit.”

Tankless water heaters are not instantaneous

Gary Klein is an energy specialist who used to work for the California Energy Commission. He is now leading efforts to change plumbing codes to legalize hot water piping systems that minimize the time it takes for hot water to reach a distant tap. (Here are the two most important principles to remember: small diameter pipes work better than large diameter pipes, and short pipes work better than long pipes.)

Gary Klein: “It does not make sense to discuss efficiency if you are not providing the service people want. Most people want hot water in 3 to 4 seconds. For that to happen, you need a source of hot water that is close to the tap, and it has to be hot before you turn on the tap. A tankless water is not instantaneous. It takes 4 seconds of water flow before the igniter turns on, and it takes 30 to 60 seconds until the water is hot.”

“This study shows that 92% of the time, hot water draws are within one hour of each other. That’s why pipe insulation makes sense.”

Talking to contractors about duct leakage testing

Kent Browning is an HVAC engineer from Austin, Texas, who commissions homes, condominium buildings, and museums. Browning said, “If a contractor can’t show duct leakage testing experience, then the contractor is not qualified.”

He also said, “If you are working in any area where contractors aren’t used to duct leakage testing, you will encounter six predictable stages in your conversations with contractors:

1. ‘Our ducts are so good and they don’t need to be tested.’

2. ‘Ducts can’t be leak tested.’

3. ‘This test is wrong.’

4. ‘Ducts can’t possibly be sealed as specified so the current leakage should be accepted as is.’

5. ‘This testing and sealing wasn’t included in our bid; it will cost extra.’

6. ‘See, I told you we would take care of you.’ ”

Electric blues band

Each evening, the Building Science Boogie Band provided live entertainment in Joe and Betsy’s barn. On Sunday night, Steve Sauer was on keyboard, Mike Lubliner was on guitar, Collin Olson was on bass, Rob Moody was on sax, and Randy Martin was on drums. The band was mesmerizing the crowd with an electric blues riff that was slowly building in intensity, eliciting cheers and hoots as the musicians’ power cranked up.

Right behind my chair, I was distracted by some shuffling and scraping. There was Betsy Pettit, bending down to move the musicians’ equipment cases that were on the floor under the stairs. “They’re blocking the HVAC return,” I heard her say.

In a crowd of energy nerds, most of whom are either leaning on the bar or captivated by the band, it’s a pretty safe bet that there will be at least one architect in the room who (thank goodness) is worried about return-system airflow.

Martin Holladay’s previous blog: “If Only Green Homes Could Be Sold Like Breakfast Cereal.”

Click here to follow Martin Holladay on Twitter.


  1. Michael Blasnik | | #1

    quick clairifcation
    I didn't actually mean that you should adjust the insulation R value by 20% (i.e. multiply by 0.8) -- I meant that you should reduce your estimated energy impact from infiltration by 20%. You should leave the conduction estimate alone.

    I completely disagree with describing this interaction as air leakage reducing the performance of insulation. The actual impact is the opposite -- air leakage improves the performance of insulation. It's just that the air leakage itself adds more to your loads than it reduces the conduction losses. Put another way, if you calculate the conduction losses and infiltration losses separately using standard methods, you end up over-stating the combined heat loss that would occur if the air flows through insulation or building cavities.

  2. User avater GBA Editor
    Martin Holladay | | #2

    Response to Michael Blasnik
    Thanks for your comments; I have edited my article to reflect your useful feedback. I'm sorry that I misinterpreted your "multiply by .8" comment.

    I understand that air infiltration reduces the conduction losses though air-permeable insulation, and have discussed the phenomenon in several recent articles, including Air Leakage Degrades the Thermal Performance of Walls and Energy Modeling Isn’t Very Accurate. You're quite right, of course, that my phrasing was sloppy when I wrote that "air leakage reduces the performance of insulation."

    Thanks for keeping me on my toes.

  3. Ted Kidd | | #3

    Why Are Savings Overestimated? REALLY?!
    When my Quickbooks balance and my bank balance don't match, I don't blame the software. If I don't journal a deposit or withdrawal, the inaccuracy is my fault. Blaming Intuit seems a bit absurd.

    To use a golf analogy for modeling, I think everyone would agree accuracy is landing on the green. Landing in the hole is not required, particularly when current shots are landing on the fairway of the adjoining hole. The fault does not lie with the club, it lies with the player and the game.

    That models overstate consumption initially should be no more surprising than the fact a reasonably ACCURATE Manual J oversizes equipment by as much as 30%. It's the fudge, Kaptain Krunch!

  4. Ted Kidd | | #4

    Why Are Savings Overestimated? REALLY?!
    Looking over Michael Blasnik's presentation I kept hoping to see these possibly controversial proposed answers in the "Why Are Savings Overestimated" slides:

    "Because there is no incentive for accuracy and a lot of disincentive" or
    "Because there is no reconciliation step" or
    "Because bigger savings are easier to sell than smaller savings" or
    "Because EE programs require COST EFFECTIVENESS at project level, and the less you exaggerate savings, the fewer jobs you will get approved" or
    "Because savings are never tracked, reported to the consumer or contractor, or otherwise made public." or
    "Because you can not expect accuracy without measurement, feedback, and incentive."

    (Not that there was NO beef for me in the slides, I did enjoy much of the content. I loved the debunk of "takeback" and studies indicating that people don't suddenly keep improved homes at 80°, and the "free energy advice Overpriced?" I do think the fact "Movers" energy consumption went up vs "Stayers" might have correlation to drop in energy prices.)

    OK, I understand that such direct call out against bad program design could lead to unpleasant repercussions, and possible uncertainty of future employment. As one friend recently told me when discussing publicly calling out issues some might be embarrassed by - "I've got to WORK with these people". And I get a lot of "I'm glad you are doing this, but I've got a family to support..."

    Which prompts this concluding question - Is there anybody in a position of influence who is willing to put truth in front of self-preservation?

  5. User avater GBA Editor
    Martin Holladay | | #5

    Response to Ted Kidd
    I can assure you that Blasnik considered the factors you proposed. Looking back on my notes, I found that Blasnik noted in Westford, "There are problems with data collection and auditing problems. They look at your attic. The form asks, 'What is the R-value of this attic insulation?' There is a bias toward assuming the existing insulation is bad."

    He mentioned that there are many utility-sponsored programs where auditors miraculously find a great many houses that are just bad enough to meet the threshold required to participate in the program.

    The last time I heard Michael give this presentation (in 2012), I wrote an article called Energy Modeling Isn’t Very Accurate. In that article, I wrote, "Many models ask for inputs which are open to interpretation. Blasnik asked, 'How do you decide if a basement is conditioned or unconditioned? Perhaps it’s semi-conditioned? Or unintentionally conditioned? Or maybe unintentionally semi-unconditioned?' When making these types of assessments, it’s hard for technicians to avoid unintentional bias. Technicians entering pre-retrofit information on an older home often come up with pessimistic R-value estimates for existing insulation levels, leading to overestimated savings projections. Because of these problems with input accuracy, default assumptions are often more accurate than data collection."

    If these types of data-entry errors can be shown to be consistent, then they need to be considered by the developers of the model. If default values are more accurate than inputs by auditors, then software developers who fail to change their models are promoting bad models.

  6. User avater
    Albert Rooks | | #6

    Rain screen gaps

    Great re-cap on the symposium. I'm sure it was great being around the "brethren" complete with music and a bar!

    I did not see where Achilles slides were available. I'm really interested in hearing more on his presentation on rainscreens. It makes great sense that solar drive is the chief mechanism that creates airflow. With today's thicker walls and higher moisture loads did he present any specifics on assmbally values of R36 and above?

    I keep encountering the old idea of "all you need is a gap wider than the surface tension of water" idea. This usually means something in the range of 1/8 - 1/4". This seems like a poor definition of a layer that is to relieve moisture load by evaporation enabled by air flow.

    It would be great if you we're able to recall if he cited a gap value or thickness based on a wall type or thickness, and if that was tied to a particular climate. It's a really interesting subject to me since I still like the "high drying potential to the exterior" assemblies that use high perm sheathing like the Agepan that we use, or something similar like Doug McEvers favorite North American fibre board that he's sourced.

    It becomes interesting to consider that from Achilles's "WUFI perspective", perhaps a larger gap is less useful over plywood at around 1 to 1.5 perms since it may not allow enough diffusion to feed airflow created. And... at what point does the wall potentially diffuse more that the airflow can evaporate?

    This is a round about way of asking: is one way of "sizing" the rain screen gap based on how permeable the exterior layers are? It sure makes sense to me that these values are interrelated.

  7. Kevin Dickson, MSME | | #7

    Domestic Hot Water and Gary Klein
    Gary's Structured Plumbing strategy is the most efficient way to plumb a new house. It reduces the wait time for hot water, and saves a lot of energy without adding cost.

    I don't see many builders promoting the concept, maybe because it takes a few minutes to understand, and some of the concepts are tricky (like longer runs of smaller pipe).

    Michael Chandler's sketch makes it easier to grasp for a residence, so I wrote this article:

  8. Mark Johnson | | #8

    About that attic ventilation
    Blasnik said among other things, “If you have a vented attic, most of the ventilation air isn’t coming from the soffit vents. Most of the ventilation air comes from the house to the attic, and then out.”

    Somewhere there probably is a study which documents this. I would like to see that study if possible. My expectation is the houses in the study, have some qualities where I can pursue the opposite.

  9. User avater GBA Editor
    Martin Holladay | | #9

    Response to Albert Rooks (Comment #6)
    You wrote, "I did not see where Achilles' slides were available." You're right -- they haven't been posted (yet).

    Q. "With today's thicker walls and higher moisture loads, did he present any specifics on assembly values of R-36 and above?"

    A. Not specifically. However, Karagiozis mentioned that today's high-R walls are subjected to higher moisture loads and have less drying potential than older low-R walls. Therefore, he said, it's important to provide designs that maximize drying -- if only to provide an extra safety factor to cover workmanship defects.

    Q. "I keep encountering the old idea of 'all you need is a gap wider than the surface tension of water' idea. This usually means something in the range of 1/8 - 1/4 inch. This seems like a poor definition of a layer that is to relieve moisture load by evaporation enabled by air flow."

    A. As I wrote in my article All About Rainscreens, a rainscreen gap provides many functions. Even when the gap is not optimized for ventilation drying, it is still valuable. However, increasing the gap size from 1/8 inch (a depth which, frankly, is hard to maintain consistently on a job site using wood-based materials) to 3/4 inch, and making sure that the gap has ventilation openings at the top as well as the bottom, greatly increases the drying rate. So if you care about ventilation drying rate, you want the gap to be big enough, and you want openings at top and bottom.

    Q. "At what point does the wall potentially diffuse more that the airflow can evaporate?"

    A. If you have a decent air barrier, you're stopping the most important moisture transport mechanism. Diffusion is a minor concern, unless you have an indoor swimming pool or a very humid greenhouse filled with tropical plants.

    Q. "This is a roundabout way of asking: is one way of 'sizing' the rainscreen gap based on how permeable the exterior layers are?"

    A. Using vapor-permeable sheathing is a double-edged sword, because of the problem of inward solar vapor drive. In many cases, walls with vapor-impermeable sheathing (especially rigid foam of an adequate thickness) stay dryer than walls with vapor-permeable sheathing. Interestingly, Karagiozis is a big fan of Delta-Dry, an impermeable plastic product with egg-carton dimples. Delta-Dry creates two distinct air gaps: one between the siding and the Delta-Dry, and one between the Delta-Dry and the sheathing. Karagiozis says that Delta-Dry performs phenomenally well -- partly because it takes advantage of ventilation drying, and partly because it is vapor-impermeable (and therefore stops inward solar vapor drive).

    I will post a photo of Delta-Dry below. For more information on Delta-Dry, see All About Water-Resistive Barriers.


  10. User avater GBA Editor
    Martin Holladay | | #10

    Response to Kevin Dickson (Comment #7)
    Thanks for the link to your blog; I'm glad to hear that you found Michael Chandler's sketch to be helpful.

    For more information on Gary Klein's approach to piping hot water systems, see my article, All About Water Heaters.

    For still more information, here is a link to a Web page with links to four articles by Gary Klein. Klein's articles address his approach in depth.

  11. User avater GBA Editor
    Martin Holladay | | #11

    Response to Mark Johnson (Comment #8)
    Michael Blasnik is a smart guy, with a background in weatherization work as well as number-crunching. While I'm sure that he is right about air leakage through ceilings in existing houses, his description does not apply to a new home that is built with attention to airtightness.

    I'm not sure whether he was thinking of a specific study when he made the comment; but any weatherization contractor with blower-door experience can testify that the ceilings in old homes tend to be incredibly leaky.

    If you want to build an unconditioned vented attic that works well, your task is not too complicated. Pay attention to airtightness when building your house. Don't install any recessed can fixtures in an insulated ceiling. Seal the cracks between partition top plates and partition drywall. Seal all penetrations through the ceiling, including electrical penetrations and plumbing vents. Don't install any ductwork in your attic. Seal all chases and chimneys. Carefully weatherstrip your attic hatch, and make sure the hatch has a tight latch.

    Finally, use a blower door and a smoke pencil to verify that your ceiling is tight.

  12. Michael Blasnik | | #12

    Response to Mark Johnson
    Mark- I don't think that quote is completely accurate or perhaps I misspoke. I meant to say that in many homes the stack effect pushes more air out through an attic than the vents let in. This is what I found studying old leaky flat roof rowhouses in Philadelphia (admittedly an unusual housing type) and it's also consistent with work elsewhere when you look at tracer gas measurements and driving forces in older leaky homes.

    In many homes, the ceiling plane is leaky and so the stack effect is pushing lots of air 24x7 while wind induced ventilation is often intermittent and low in typically well shielded homes. But a tight ceiling completely changes this conclusion -- as would a very windy site or high levels of venting with a fairly windy site.

    In "good" homes the attic ventilation will be dominated by outside air coming through the vents.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |