Principal Investigator: Louise F. Goldberg
Building Physics and Foundations Research Programs
University of Minnesota
Project Manager: Stan Gatland
CertainTeed Corporation
Date: January 06, 2006
Revision B

The research described herein has been performed with funding provided by the CertainTeed Corporation.  While this support is gratefully acknowledged, the Principal Investigator assumes complete responsibility for the contents herein.


This experiment was conducted at the University of Minnesota's Foundation Test Facility (FTF) from December 2003 through February 2005.  The principal purpose of the experiment was to quantitatively evaluate the hygrothermal performance of 2-mil. polyamide-6 membrane (PA-6) as a warm-side vapor retarder in a masonry-block, full basement application with an interior fiberglass batt / stud frame insulation system and compare that performance to standard, Kraft-facing.  The following insulation configurations were evaluated (interior to wall sequence):

  1. PA-6 | unfaced fiberglass batt
  2. 0.5" gypsum | PA-6 | unfaced fiberglass batt
  3. 0.5" gypsum | PA-6 | unfaced fiberglass batt | PA-6
  4. PA-6 | unfaced fiberglass batt | PA-6
  5. 0.5" gypsum | PA-6 | unfaced fiberglass batt | adhered elastomeric waterproofing membrane
  6. PA-6 | unfaced fiberglass batt | adhered elastomeric waterproofing membrane
  7. Kraft-faced fiberglass batt

With the exception of the waterproofing membrane systems, all the systems were evaluated on both northeast and southeast quadrant exposures to determine the influence of solar effects.  As a qualitative control (that is, no numerical data were collected, only weekly observations), a 0.5 lb/ft3 free-rise, open-cell spray-applied foam insulation system from a previous experiment was used to reflect the hygrothermal performance of a no-vapor retarder system as well as a conventional 6-mil. polyethylene interior vapor retarder system.  While not as high as fiberglass batts, the permeance of the open-cell foam is high enough to represent the hygrothermal performance of fiberglass batts on a qualitative basis.

Data collected included the following:

The hygrothermal data are presented in terms of relative humidity, vapor pressure, and for the systems with removable sections,  wetting/dying profiles.


        1.    Boundary conditions
        2.    Relative humidity temporal profiles
        3.    Vapor pressure temporal profiles
        4.    Wetting/drying profiles



All the conclusions apply to full-height, hollow core masonry block walls with an 18" above-grade exposure located in a well-draining group I soil (medium sand) with a very low water table.  The exterior wall surfaces are bare and are in direct contact with the soil.

  1. Full-height, bare masonry block walls with no more than 18" of the wall above grade that are insulated with unfaced fiberglass batts and covered with an interior, air-sealed PA-6 vapor retarder meet the performance requirements of the MN Energy Code Building Foundation Rule Proposal Final Report, specifically, they conservativelya satisfy the following criteria:

    On the interior side of the water separation plane, the system has:
  2. Kraft-facing without an interior air barrier appears to yield better hygrothermal performance than air-sealed PA-6 on the equivalent insulation system, however the facing paper became wet in places yielding the potential for biotic activity, thus these experimental data do not assure that this system will meet the "biotic activity" performance criterion over the long term.  Additional, longer term experimentation will be necessary to provide this assurance, particularly in the presence of a well-sealed air barrier and an exterior water separation plane with a permeance of 0.1 perm or less.
  3. When the water separation plane does not have a permeance of 0.1 perm or less, there is considerable vapor gain from the surrounding earth into the masonry block wall cores that condenses onto the surface of the above-grade block core faces and thence into the interior insulation system .  With the inclusion of such an adhered interior water separation plane, the hygrothermal performance of an air-sealed interior PA-6 vapor retarder system is considerably improved over systems without such a water separation plane.  However, there was significant condensate rundown on the interior side of the adhered water separation plane to a level 30.5" below the wall top so that these experimental data do not offer the assurance that the "accumulation of free water" performance criterion would be met over the long term.  Further experimentation with more detailed instrumentation is needed to provide this assurance.
  4. PA-6 encapsulated fiberglass batt interior insulation systems (that is, PA-6 membranes on both the interior and wall sides of the insulation) do not offer any improvement in hygrothermal performance over an interior only PA-6 membrane system.  These systems can produce additional condensate rundown on the interior side of the wall side membrane and impede the moisture sequestration performance of the masonry block wall surface.
  5. Unpainted and non-air-sealed interior gypsum does not measurably affect the hygrothermal performance of any of the PA-6 systems tested.
  6. In order to accurately capture the continuous wetting/drying performance of PA-6 membrane insulation systems at the wall surface, it is necessary to use relative humidity sensors that operate in condensing conditions with an accuracy of 1% or better in the range of 90 to 100% RH.


  1. The results are conservative because the exterior water separation plane consisting of free-draining medium sand does not provide any resistance to soil-sourced water vapor ingression.  As shown in conclusion 3, with the addition of a 0.1 perm water separation plane, the hygrothermal performance of the insulation system increases significantly.  Thus, were the exterior water separation plane also to have a permeance of 0.1 perms, then the same hygrothermal performance improvement would occur while the bare masonry block surface would still absorb the heating season interior wall surface condensate.


  1. The experiment should be repeated with sensors that satisfy the requirements of conclusion 6.
  2. The experiment should be repeated on poured concrete and on permanent wood foundation walls.
  3. Technologies for improving the moisture sequestration performance of the wall surface should be developed and evaluated under conditions where the above-grade wall exposure is increased to at least 50%.
  4. The experimental results should be used to guide the development and validation of a 3-dimensional, below-grade hygrothermal simulation code.  Current commercially available simulation codes do not capture enough of the heat and moisture transport physics occurring in building foundations to enable their validation by the experimental data.

1.    Weekly manual data readings
2.    References

return to contents