Best way to superinsulate?

Started by jimgranite, January 22, 2007, 11:12:14 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.


No interior insulation, but 4 inches of foam insulation on the outside.  The inside studs are open leaving room for plumbing and wiring and air movement if needed.  I haven't had a chance to go look at the house under construction but I guess he has built several that way.  Some friends currently own one and LOVE it.  

I guess I need to explain that the studs are not exposed, he does have interior drywall and a finished wall.


Doesn't make any sense to me to spend money on expensive exterior foam and then not put the inexpensive insulation in the cavities you will be closing up anyway.


I'll talk to the builder, but remember, this is a real cold climate.  Not as bad as Fairbanks, but it was -39 a couple weeks ago.

The complete foam wrap eliminates the problem with wood transferring heat and also the problems that are associated with settling or moving batts over time, Just a little bit of incomplete insulation along the edge of a stud will cause a lot of thermal transfer.

I notice in my little cabin that I am building that there was a huge amount of heat loss at the top and bottom plate.  I neglected to put in a foam insulating pad at that point.  Of course I caulked that up.  But foam on the outside would have extended above and below the plates.

You are right about the expense however, don't know if it's worth it.


Quotethis is a real cold climate.  
All the more reason for more insulation.   :-/   Leaving the cavities empty is a puzzler to me.   :-?


One of the goals of these foam-on-the-outside schemes is to eliminate thermal bridging by the wood studs. But I wonder if there ends up being almost as much or more thermal bridging due to all the nails through the foam. Steel conducts heat 400 times better than wood, so a single large nail (large enough to go through 4" of foam) conducts about as much heat as a twelve inch length of 2x lumber. If the nail spacing through the strapping and foam is closer than that, I don't think you're gaining much.

One possible reason for not filling the cavities behind the foam is to avoid condensation inside the wall. In cold humid climates, the added batt insulation might cool the inner foam surface enough to cause condensation there. And a thick layer of foam would act as a vapor barrier keeping the wall from drying out. I think this would definitely be a concern in Juneau. I'm not familiar with the climate in the Fairbanks area though.


Excellent point Bartholemew.  If there were batts in the cavity, there would almost surely be a soaking wet layer right on the sheathing.  

That is a little bit of what I fear with the 1 inch foam all around the exterior.  I worry about it with the 4 inches of foam as well.  Where exactly does the condensed moisture end up, and how does it dissipate after condensing?

With the 6 inch fiberglass batt, I know that it does condense somewhere in there, but at least it has a chance to dry out a little after


Quote#1... eliminate thermal bridging by the wood studs. to avoid condensation inside the wall.

Bartholomew, both points are very well made. Wish I had thought of the first, and I've forgotten about the second. I've been in  a warm/hot low humidity climatetoo long. (OK winters are cool to cold, but it's still pretty dry.

Re #2: How about applying the foam board to the interior side of the studs and covering over with drywall?  :-/ Install fiberglass in the framing cavities as usual. That would place the super vapor barrier characteristics of foam on the warm side. (Could be a problem with the pipes in the framed wall?) If that works, then...

Re #1: Run 1x furring strips horizontally on the inside every 16 inches and securing the foam to them. Drywall on top. Of course that would really puzzle someone coming along later trying to find studs to secure something to.  ;D

I think I still prefer more or less conventional framing and sealing the bejeebers out of it before installing fiberglass in the cavities. I especially like the staggered interior/exterior 2x framing method if one is concerned about thermal breaking the studs.

If going with conventional framing, it might be possible to work out something with foam covering the top and bottom plate areas (thermal break) on the exterior and using the added thickness as some sort of an architectural detail.    :-?  Depends on the exterior materials. Easy to do with stucco. Around here the builders use a lot of foam like that (covered with housewrap or felt and chicken wire) for raised borders around windows, doors or wherever to help break up the monotony of flat stucco areas.


Yes, an exterior (or interior) layer of foam is a wonderful conductive break in the framing. Such foam sheathing is doing its best work where it covers the solid wood (you can see this in the calcs on my diagram above where 1" of foam doubles the R-value of the framing.)

The cost effectiveness of the foam does go down pretty fast as you add more inches. And since it is so much less expensive, the cost effectiveness of fluffiness such as BIBS or fiberglass stays quite high - at least until the cavities are fully filled and you have to build new ones  ;).

You always want to max out the most cost effective measures first. This makes both economic and resource utilization sense.

PS - 1. Any wall with 1" + of foam (plus batt) will be too well insulated to ever reach the condensation (dew) point on the moist interior. 2. The surface area that nails contribute to conduction is negligible. They are anchored into wood with an R-value of 1/inch. 3. Where you really see conduction take off is when people build with steel framing. There you get some real bridging and the nails or screws are part of that inside to outside heatflow freeway. Steel buidings (if heated) have lots of problems with condensation and corrossion unless they are fully foam sheathed.


True John, the remaining wood does provide a bit of a thermal break. However, the nail is drawing heat from not just the small cross-sectional area of it's tip but also from the sides (think of a cone on its side with its base on the inner wall surface and the nail sticking into the point). An analogy is a bridge across a river. The people living right next to the bridge might use it the most but it also serves the surrounding area. Likewise, the nail acts as a heat conductance bridge across the foam for a fairly large area of inner wall surface. How large is the question. In this thermal image, you can see the largish blobs where heat is conducting away through the drywall nails.

Simplifying now, suppose each nail provides a thermal bridge for a rectangular area of inner wall surface which is the width of the stud and 3" up and down from the nail. Well, if the siding is horizontal lap siding with 6" exposure then there is a nail every 6" and virtually all the heat transfering through the studs can bypass the foam.

The added foam does provide more insulation between the studs. However, it seems to be sold/justified mainly on the basis that it reduces thermal bridging. I believe that the thermal bridging effect is much less than most people assume. As a result, it might make more sense to spend the money elsewhere... upgrade from 2x4 to 2x6, BIBS or cellulose in place of batts, icynene, etc. What I'd really like to see is actual whole wall test results from Oak Ridge National Labs or similar impartial organization.


On the condensation issue, I'm talking about the surface of the foam, not the inside wall surface. In martyv's example, suppose that R-19 of fiberglass is added behind the R-20 foam. The inner surface of the foam will then be roughly halfway between the outside and the inside temps. Given high enough indoor humidity and cold temps outside, there can easily be condensation inside the wall. cold and very cold regions the thickness of the foam sheathing should be determined by hygro-thermal analysis so that the interior surface of the foam sheathing remains above the dew point temperature of the interior air ...
The hygro-thermal analysis is done using software such as WUFI from ORNL (free to download and fairly easy to use). Probably worth doing if you are building in an extreme climate or doing something beyond the tried-and-true for your area....


Coastal Contractor, November 2006, did an article on foam sheathing which said that 1" xps would likely work for every U.S. area south of Long Island.  I guess they were just thinking of the Eastern seaboard.  The article is posted at the Journal of Light Construction website ( ) but I'm not sure if you need a JLC subscription to access it.  I subscribe to the online version and have access to their 20 year archive of JLC articles--a good deal.

As far as thermally-leaky fasteners, I think a weighted average of material conductivity based on contribution to wall area might be appropriate--at least that's what they use for computing the R loss from framing members.  When I calculated for 10d nails (8x16 pattern like for fastening strapping), I'm getting about 4.4 compared with R5 for non-perforated 1" xps.  Using stainless-steel 10d nails, I'm getting about 4.6.  Don't use aluminum nails: it drops to R3.   I'm not an engineer and pretty new to this so I could be doing it wrong; however, it seem like the building science experts are probably aware of fastener conductivity and still use external foam in many of their model wall designs. ;)


If it's 70 degrees inside and 20 below zero outside, and the moisture is moving from the warmth to the cold, it would seem that it needs to condense somewhere.  It certainly isn't 70 degees only 1/4 inch into the foam from the -20 degree exterior air.  

Since the foam is a vapor barrier it would seem to trap the moisture just inside the foam, would it ever be cool enough to codense right there?  I would think so.

In fact the place it might be most likely to condense would be along the studs because they would be cooler.

And after it condenses it would run down the inside of the wall I suppose.

I am not a thermal engineer, but I am certainly interested in this question for practical reasons.


Hopefully the interior vapor barrier keeps the RH in the insulated area low enough that the moisture never reaches it's dew point or that the dew point occurs somewhere in the foam where there is no humidity or problem.  Not a scientist either - learned some of this in pilot training though.

If the moisture stays in the heated area and doesn't reach it's dew point there will be no condensation.


QuoteAs far as thermally-leaky fasteners, I think a weighted average of material conductivity based on contribution to wall area might be appropriate
I'm not entirely sure. Doesn't that approach basically make the simplifying assumption that heat will only flow in a straight line from inside to out (or vice versa) instead of following the path of least resistance? That assumption might not affect the results too much when all the materials are of roughly similar conductivity. But when they differ by three orders of magnitude, the result could be overestimating the actual r-value by quite a bit.

Nonetheless, if there are 17 10d nails per stud (i.e. for 6" exposure siding) and the studs are 16" OC, I calculate the combined r-value as 3.62. If instead I assume the foam is fitted in between 2x lumber, I get 3.64 for the combination. That suggests that upsizing the studs and fitting the foam between them would be just as effective. You'd also get additional benefit from having an extra 2" of wood providing a thermal break for the nails.

Another possible option might be to use plastic nails. But I don't know if they are approved for siding use.


QuoteHopefully the interior vapor barrier...
You do have to make sure you don't effectively end up with vapor barriers on both sides, since some foam has fairly low permeability and there is 4 inches of it. But that's what the WUFI software is for.


I guess that is where some of the new methods run into problems -- things happen in places they never used to.


OK, first on condensation.

You have to have airflow to have problems with condensation. If the dew point is reached inside your wall insulation, foam sheathing or wood framing it will condense moisture only if air leaks are bringing fresh moisture there. While there can be "vapor pressure" through a material such as framing, such condensation is trivial. Over 95% of the moisture transport in real houses is through the air leaks in the structure itself. Airflows from a warm moist interior reaching a cold surface (most often the inside of the wall sheathing or siding) will drop their moisture there. This only becomes a problem when the transport of the airflow is sufficient to overwhelm the ability of the materials themselves to absorb that moisture. Wood framing, sheathing and siding can hold lots of water before it becomes a problem.

Framing lumber, for instance, might rise in moisture content seasonally from 15% to 45% absorbing scores of gallons of water transported by small air leaks. Then this will dry out in the summer. There will be no surface moisture at any time during this entire yearly cycle.  Steel framing and vinyl siding holds much less water and manifest moisture (real water) problems much faster. These buildings are much more likely to have mold issues.

On conduction through fasteners. It is based on an area weighted calculation of total assembly R-value. Steve044 is doing the right calculation. The R-value calculation is a representation of the "line of least resistance". Lower R-values do pump heat faster than higher R-values. Nothing more magic than that is going on. It is why moisture problems show up first on the surface of the windows where the R-value is 1 to 3 rather than the 14 to 21 you have in the wall.

Themography can show small differences in heat conduction and can thus make these differences visible. The windows in such a photo are off the chart until the sensitivity is dialed way down.


Is Fibre Brace sheathing better than wood in regard to increasing insulation?


Here is a very good technical article on building with insulated sheathing. Included many details such as how to handle windows and doors, housewrap, structural bracing, etc. These are folks who do their homework (Building Science Corp.)


Thanks John! I guess that's why you're the architect and we're...  :-? well, we're not.   ;D  So one should take extra time and care to stop the air in/exfiltration above all else.


Absolutely! Airsealing is the big key to controlling moisture.

RE: On the question (youngins) of composite (fiber) sheathing. I don't know that particular product (there are many brands made locally from waste wood products). Most fiberboard sheathings are only slightly better than OSB or plywood at insulation (check the R-value. Solid wood is about R-1 per inch) and far worse at structural integrity for bracing and worse at nail holding for siding. They are cheaper and may be code approved if you are in low wind areas. These are sometimes sold as an "insulation" product but this is used as a marketing term and the supplier hopes you don't understand R-values.

I call out structural sheathing (OSB or plywood) in my plans but lots of builders use fiberboard ("punkboard" :-?) sheathings too.


Did you know:

John wrote a book "Superinsulated Design and Construction."  1987 I think.

ISBN 0-442-26051-2  It's out of print - available on Amazon from serveral used book sellers.  This book is an investment.


Is the composite sheathing what we used to call black board?

Because, strangely enough, it was black.  

Either nobody's using it any more or it gets covered up by tyvek.


QuoteHere is a very good technical article on building with insulated sheathing. Included many details such as how to handle windows and doors, housewrap, structural bracing, etc. These are folks who do their homework (Building Science Corp.)
That is an excellent, very informative article. Thanks a lot John!   :)


QuoteIs the composite sheathing what we used to call black board?
Because, strangely enough, it was black.  
Either nobody's using it any more or it gets covered up by tyvek.
If what you're talking about is a kind of fibrous material it's still in use around here by some builders. They may use OSB at the corners and fill in the spaces in between with the black stuff. I can't for the life of me remember the name, but there's a stack of it at Home Depot. The houses in my neighborhood have a lot of it in them. (22 years old) Currently it costs more than OSB so I don't understand why anyone would use it.


Fibre Brace -

Rob LeMay used it on his -

Its cheaper than OSB and claims to be better at sealing moisture. - Is this true?