Hi all,

Robert has it pretty well nailed down in his comments; however, some clarifications could be helpful to those following this discussion.

Robert wrote:  I believe that a plastic vapor barrier is an invitation to problems in almost any thermal envelope. It was promulgated by code when most houses were insulated with fiberglass and there was little attention to (even recognition of) the need for air-tightness.

Putting a Vermont Residential Building Energy Standard (RBES/Code) required vapor retarder with a perm rating of 1.0 or less on the warm-in-winter side of nonframed ceilings, walls, and floors can and should be accomplished without the use of poly.  An appropriate vapor retarder can be installed via the use of the airtight drywall approach (mentioned later in Robert's email) coupled with the use of vapor retarder or barrier paints.  This is a better approach as poly almost never serves as an effective air barrier and in fact usually interferes with efforts to achieve an effective air barrier.  As Robert also correctly implies later in his email, diffusion (water vapor transport through permeable and semi-permeable building materials flowing from areas of high vapor pressure to lower vapor pressure) is the minor vapor transport mechanism.  Air leakage following heat transfer (hot to cold) is the major water vapor transporter in most buildings.

Robert wrote:  The same fiberglass-insulated house with 1" of exterior XPS and 0.35 ACH would have an additional embodied energy payback of less than 1 year, in large part because it results in much greater energy savings than the typical sprayed urethane, mostly by eliminating thermal bridging.

If the building assembly is such that "moisture or its freezing will not damage building materials and/or insulation" (to quote RBES), as would be the case with an exterior XPS wrap or some other exterior rigid foam insulation wrap of sufficient R-value, the home would not require a warm-in-winter side vapor retarder.  I point out though that 2" XPS (R-10) or other insulated sheathing materials at R-10 are needed in our climate zone to prevent moisture or its freezing from damaging building materials and/or insulation.  At R-5, 1" XPS will allow condensing surfaces to exist during sufficient portions of the year.  With only an R-5 exterior wrap, should warm moist interior air get into a wall cavity, it will find surfaces below the dew point and condensation could then damage building materials and/or insulation.

Robert wrote:  I have calculated that, compared to a 2000 SF reasonably well-sealed code-standard fiberglass-insulated house (0.5 ACH), a urethane sprayed house (with insulated gable walls and roof assembly rather than ceiling), even accounting for an increase in air-tightness (min. 0.35 ACH), will have an additional embodied energy cost that would require 23 years of operation to pay back.

The ACH building tightness numbers Robert discusses refer to estimated "natural air changes per hour" or ACHNatural (ACHNat).  These numbers are derived by testing the home with a blower door and then comparing the leakage measured (CFM50 - cubic feet per minute at 50 pascals pressure difference) to home's volume to estimate the air change rate that a house leaks under "normal" conditions.  This can only be roughly estimated and it erroneously implies a specific air change rate as a constant.  The air change rate is not constant and varies widely over time because natural ventilation is driven by temperature difference and pressure difference. 

I admit culpability in training many VT builders to look at and gauge their performance via the ACHNat results they achieved on their Home Energy Rating certificates.  However, it would be far better if we looked at our buildings using air changes per hour at 50 pascals pressure difference (ACH50).  This eliminates the estimate that has to be made as to what's "normal" operation and significantly improves the accuracy of this measurement.  To determine ACH50 the following formula is used:

CFM₅₀ x 60 (minutes/hour) / house volume (cu. ft.) = ACH₅₀ (Air Changes per Hour at 50 Pascals). 

The CFM50 and the house volume are outputs on the new Index Home Energy Rating certificates.  Target ACH50 results should be <= 5 ACH50.  Better yet would be to eliminate the house volume and compare our leakage against the amount of exterior surface area but we are not able to present that in our certificates.

On the last roughly 3,600 homes we have blower door tested (all participants in Efficiency Vermont programs), 73% are at or below .35 ACHNat.  For that same sample, 69% are at or below 5 ACH50.

How long do want our houses to last?  I wouldn't sneeze at a 23 year payback if we're looking at an economic life of 100+ years for the house.

Robert wrote: Open-cell sprayed Icynene has almost the same embodied energy liability as the fiberglass, but it would result in a less energy-efficient house.

Icynene brand open-cell foam has a very similar R-value per inch of thickness as fiberglass; however, the improved air tightness seen in most Icynene insulated homes results in greater heating efficiency not less.  We do generally see spray-in-place insulated homes testing tighter than fiberglass batt insulated homes and usually as tight as or tighter than cellulose insulated homes.

Robert wrote: A similar house with 2x8 framing 24 oc (instead of 2x6 16 oc for the others), which uses no more total wood, and made very tight with the air-tight-drywall system (0.25 ACH, which is adequate in a non-toxic breathable house), would have 41% of the insulation embodied energy of the fiberglass (less than 5% of the urethane) and use 39% less heating energy.

My comments here are that we can also build with 2"x 6" 24" oc and get better performance.  If we are to thicken the wall more we farther with a minimum R-10 exterior insulation wrap to eliminate thermal bridging, lower wetting potential, and enable drying to the inside by diffusion as we can now legitimately eliminate the 1.0 perm vapor retarder.  On top of that we can achieve a whole wall R-value in the R-24 range (with fiberglass batts, cellulose, or Icynene full cavity insulation, though wall performance will vary based on the air sealing properties of the insulation, its installation quality, and the effectiveness of the interior and exterior air barriers).  Houses don't breath.  They should be built tight and ventilated with equipment appropriate to maintain a consistent and appropriate air exchange (when occupied) to control moisture levels, provide adequate fresh air, and dilute indoor air contaminants of all sorts.

I don't recommend spray-in-place foams as 1st choice for new construction because we generally have sufficient building assembly thickness available and cellulose is a preferred material for a multitude of reasons.  However, this is a right tool for right job issue.  Spray-in-place foam does a great job in insulating rim-band assemblies.  It can be very useful when a sloped ceiling constructed with scissors trusses exists that defies dense-packing with cellulose.  To my thinking spray-in-place foams are often the best tool with existing homes where wall thicknesses are small and thickening them is not feasible.  I rebuilt much of my old farmhouse from the outside and closed-cell foam was the only material that would effectively provide me with an R-19.5 insulation, an air barrier, and a vapor barrier in a 3" cavity.  It is also the only reasonable solution for insulating and air sealing the old field stone foundation.

Jeff Gephart