A SIMULATION INVESTIGATION OF STUCCO CLADDING WALL SYSTEM
VAPOR TRANSPORT PERFORMANCE IN A COLD CLIMATE
Louise F. Goldberg, Ph.D (Eng)
Building Physics and Foundations Research Programs
College of Design
University of Minnesota.
MN Lath and Plaster Bureau
St. Paul, MN
Date: July, 2006
ACKNOWLEDGEMENT AND CERTIFICATION
The research described herein has been performed with funding provided by the Minnesota Lath and Plaster Bureau. While this support is gratefully acknowledged, the Principal Investigator assumes complete responsibility for the contents herein.
Hygrothermal: Combined moisture and heat.
Bulk Water: Liquid water.
Advection: Movement of a fluid produced by a difference in pressure.
Hydraulic transport: Movement of material produced by the flow of water.
Parametric: A measure that is changed independently to investigate the response of a system.
ASHRAE: American Society of Heating, Refrigerating and Air-Conditioning Engineers. Recognized authority that publishes a handbook that contains basic principles and essential fluid flow and heat transfer engineering data.
NIST: National Institute of Standards and Technology. A National test laboratory that provides measurements and standards for U.S. industry.
NRC: National Research Council, Canada
Permeance: A measurement of the ability of a material to retard the diffusion of water vapor at 73.4°F (23°C) in response to an applied vapor pressure gradient. A perm (a permeance measurement unit) is the number of grains of water vapor that pass through a square foot of material per hour at a differential vapor pressure equal to one inch of mercury.
WRB: Water-Resistive Barrier. Any in the variety of housewraps such as Tyvek, Building Paper or No. 15 felt used to wrap the external sheathing of a building prior to the installation of the cladding.
Fluid Dynamics: The study of the motion of fluids (liquids and gases) in relation to the forces imposed upon them.
The Minnesota Lath and Plaster Bureau entered into an agreement with the University of Minnesota on April 7, 2003 for the Building Physics and Foundations Research Programs to perform research on the moisture performance of residential building envelope walls covered with stucco cladding. The impetus for this research was an underlying and persistent theme laid out in symposiums by credible scientists and engineers that “reservoir” claddings (stucco in particular) are prone to condensation-related failures produced by solar radiation induced water vapor transport. At the time of this agreement, anecdotal speculation was the basis for the inquiry, as there was no quantifiable research substantiating these claims.
Hence this report examines the simulated hygrothermal performance of stucco-clad wall systems in response to interior and ambient relative humidity and temperature; solar radiation on a south-facing exposure; and, wind-driven exterior rain. The simulations specifically exclude systemic bulk or liquid water effects. Examples of these effects include drainage from flange-mounted windows onto the water resistive barrier between the cladding and the sheathing and, leakage into the wall cavity from rough openings. Thus the simulations focus on the impact of stucco cladding on the water vapor transport performance of the wall system, not on the response of the wall system to flashing-related and other bulk water failures.
The simulations were carried out using the Fraunhofer Institut Bauphysik WUFI-2D (version 2.1) 2-dimensional transient wall system hygrothermal simulation program because, despite its well-known limitations, this was the only commercially available program in common usage at the time. These limitations include the absence of advective (pressure-driven) moist air transport; no continuous or free-surface hydraulic transport; an excessively critical dependence on empirically derived material properties; the absence of rigorous phase change modeling; and, limitations in dealing with wind driven rain. As a result of these limitations, a model was adopted for the simulations in which the hygrothermal performance relative to the baseline is significant as a function of specific input parameter changes. Hence the absolute performance in any particular case may not be in agreement with experiment which is to be expected in terms of the above limitations.
A total of 6 interior and exterior boundary condition combinations were evaluated in combination with three sets of material properties in order to quantitatively assess the impact of climate as well as the critical dependence of the WUFI program on material properties. The simulations were carried out in 2 phases. The first phase included parametric variations of 4 climates and 2 sets of published material properties (ASHRAE and NIST); 6 variations of the baseline composite stucco configuration as defined by the project sponsor; and, 20 variations of wall system design which included alternative insulation, water resistive barriers and interior vapor retarders. The second phase investigated 2 additional climate variations; a third set of published material properties (NRC); 3 different double layer water resistive barriers in common use with stucco claddings; and, 4 alternative claddings (clay brick, cedar, fiber cement board and vinyl siding).
Taken as a whole the phase I and phase II simulations investigate two approaches to achieving unconditional stucco-clad wall system hygrothermal durability in response to temperature, relative humidity and exterior-only water impingement:
by engineering the properties of the relevant materials to permit the use of standard or traditional, code compliant stucco-clad wall systems, or
by designing fault-tolerant, non-standard wall systems allowing the use of stucco and other wall system components with a broad range of material properties.
The primary conclusion that can be drawn from the phase I simulations is that the hygrothermal performance of stucco-clad wall systems with exterior surface wetting only is complex and success or failure is extremely sensitive to material properties and details of the wall configuration. The importance of treating the wall assembly as a system whose overall moisture performance is synergistically dependent on the composite performance of its constituent layers is apparent. If installed materials have published properties that fall outside the design limits, the simulations show that the wall system as a whole will fail. For example, if the installed wood fiberboard has ASHRAE material properties, it generally performs better than if it has NIST material properties or worse than if it had some other material properties. This highlights the importance of having a wall system design that is intrinsically fault tolerant and accommodates a broad range of material properties and, in particular, stucco material properties. The phase I simulations demonstrated the following two such fault-tolerant stucco wall systems designs (in exterior to interior order):
Standard stucco cladding with generic material properties / 2 layers 60-minute grade-D building paper / 25/32” wood fiberboard / closed cell spray polyurethane insulation in a 2 x 6 wood stud frame / 0.5” gypsum board with an acrylic paint finish.
Standard stucco cladding with generic material properties / 2 layers 60-minute grade-D building paper / 25/32” wood fiberboard / 2” extruded polystyrene insulation / unfaced fiberglass batts in 2 x 4 wood stud frame / 2-mil. PA-6 membrane / 0.5” gypsum board with an acrylic paint finish.
The phase II results indicate that with exterior bulk water sources only (wind-driven rain), vapor transport through a “standard” stucco-clad wall system does not produce condensation-related failures with “correctly” engineered stucco material properties corresponding roughly to those used in the phase II simulations (NRC). This finding includes the current practice of installing a warm-side polyethylene vapor retarder.
Unfortunately, current building practice predominantly uses an interior drainage plane (that is, between the cladding and the WRB) as the means by which liquid is drained away from rough openings, most notably, by the use of flange-mounted windows. In stucco systems, this introduces a moisture source interior to the system between the stucco and the WRB, a moisture source that the “standard” stucco system with a warm-side polyethylene vapor retarder is not designed to accommodate. This changes the boundary conditions on the wall system very substantially and thus does not permit the results of the phase I and phase II simulations to be extrapolated to the interior moisture source case with much reliability. The impacts of such moisture sources on the overall wall system performance will be investigated in the next phase of the research.
Finally, the simulations undertaken have revealed the significant complexity of the physics governing the hygrothermal performance of reservoir cladding wall systems, even within the simplifying assumptions of the WUFI analysis. This illustrates how ill-advised anecdotal forensic analyses of stucco system failures can be and why popular articles, fact sheets and the exhortations of experts are so inconsistent and confusing. Upon removing the simplifying assumptions on which WUFI is based, the physics become even more complex and less amenable to simplistic explanations of observed failures. In this regard, there is a real need for a rigorous, public domain computational fluid dynamics simulation that can provide reliable design guidance for the hygrothermal performance of building envelopes in general and reservoir cladding wall systems in particular. Such a simulation should minimize the use of first-order empirical material properties and include specific test standards for experimentally quantifying the empirical material properties that are necessary.
This report has been peer-reviewed by 2 reviewers so far – their comments are included in Appendix 2.
Download the full report: LP-Stucco-Final-C.pdf
return to contents