September 17, 2013

Abstract: 

Unvented roof strategies with open-cell and closed-cell spray polyurethane foam insulation sprayed to the underside of roof sheathing have been used since the mid-1990's to provide durable and efficient building enclosures. There have been isolated moisture related incidents that raise potential concerns about the overall hygrothermal performance of these systems. This project involved hygrothermal modeling of a range of rainwater leakage and field evaluations of in-service residential roofs using spray foam insulation. The exploration involved taking a sample of spray foam from the underside of the roof sheathing, exposing the sheathing, then taking a moisture content reading. All locations had moisture contents well within the safe range for wood-based sheathing.

Executive Summary

Unvented roof strategies with open cell and closed cell spray polyurethane foam (ocSPF and ccSPF) insulation sprayed to the underside of roof sheathing have been used since the mid-1990s to provide durable and efficient building enclosures. However, there have been isolated moisture-related incidents reported anecdotally that raise potential concerns about the overall hygrothermal performance of these systems. The incidents related to rainwater leakage and condensation concerns. Condensation concerns have been extensively studied by others and are not further discussed in this report (Straube et al. 2010).

This project involved hygrothermal modeling of a range of rainwater leakage and field evaluations of in-service residential roofs using spray polyurethane foam (SPF) insulation. All of the roof assemblies modeled exhibited drying capacity to handle minor rainwater leakage. All field evaluation locations of in-service residential roofs had moisture contents (MCs) well within the safe range for wood-based sheathing.

The quantity of water passing through a roof system is difficult to quantify, but hygrothermal modeling is possible using ASHRAE 160, Minneapolis Typical Meteorologial Year 2 and U.S. Climate Normals weather data, and WUFI weather data. WUFI 5 was used to determine the effect of 0.01% to 1.00% of the rainfall entering the unvented roof system as a leak and coming in contact with the wood-based structural roof sheathing. The 2012 International Residential Code-compliant roofing system using ccSPF on plywood sheathing with cellulose insulation on the interior has the capability according to WUFI to safely dry a leak up to 0.6% of the rainfall in Minneapolis. In Seattle the roof systems modeled were able to accommodate up to 0.6% for ocSPF and 1.0% for ccSPF and in Miami up to 1.5% could be dried out when using ocSPF. Assuming the recommended fully adhered membrane is properly designed, detailed, and installed water should have very little likelihood of ever entering the system through leaks. ocSPF dries more readily than ccSPF, but ocSPF allows more wetting of the sheathing during the winter months and accordingly requires a Class II vapor retarder coating directly applied to its interior surface as specified by International Residential Code. Interior relative humidity can directly affect the sheathing MC in all scenarios and hence wintertime relative humidity in a climate zone 6 home should be < 40%. Orientation and sheathing materials created variations within the system, but these variations were relatively small compared to the type of SPF and vapor permeance of coatings used. Oriented strand board (OSB) sheathing, ocSPF, and roofs facing north maintain the highest MCs, but all systems modeled were within the safe range for wood-based sheathing. Damage could occur where the volume of the leak, frequency of the leak(s), or quantity of interior moisture driven into the system are more than modeled in this study.

Explorations of 11 in-service roof systems were completed. The exploration involved taking a sample of SPF from the underside of the roof sheathing, exposing the sheathing, then taking a moisture content reading. All locations had MCs well within the safe range for wood-based sheathing. One full-roof failure was reviewed, as an industry partner was involved with replacing structurally failed roof sheathing. In this case the manufacturer's investigation report concluded that the SPF was installed on wet OSB, based on the observation that the SPF did not adhere well to the substrate and the pore structure of the ccSPF at the ccSPF/OSB interface was indicative of a wet substrate.

1 Background and Significance for Building America

1.1. Introduction

Open cell spray polyurethane foam (ocSPF) and closed cell spray polyurethane foam (ccSPF) insulation sprayed to the underside of roof sheathing is a popular strategy for increasing roof insulation levels in all climate zones. Unvented roof strategies with spray polyurethane foam (SPF) have been used since the mid-1990s to provide durable and efficient building enclosures. However, there have been isolated incidents of failures (either sheathing rot or SPF delamination) reported anecdotally that raise some potential concerns about the hygrothermal performance and durability of these systems. The incidents were related to rainwater leakage and condensation concerns.

Condensation concerns have been extensively studied by others (Straube et al. 2010) and are not further discussed in this report.

It is unclear whether the rainwater leakage issues are a material issue, an application issue, both, or neither—or even whether issues actually exist in sufficient numbers to be of concern. The 2011 Standing Technical Committee on Enclosures has identified this as an important topic for additional research work (Lstiburek 2011).

  • The primary risks for roof systems are:
  • Rainwater leaks
  • Condensation from diffusion and air leakage
  • Built-in construction moisture.

This report deals directly with rain and indirectly with built-in construction moisture. The technical approach used in this project combined hygrothermal modeling of a range of rainwater leakage scenarios, and field evaluations of residential roofs using SPF insulation.

1.2 Project Background

Spray foams have advantages over alternative methods with respect to providing air sealing in complex assemblies—particularly roofs. SPF can provide the thermal, air, and vapor control layers in both new and retrofit construction. In cases where mechanical systems are located in attics, moving the air control layer and thermal control layer to the underside of the roof deck has particularly large advantages compared to sealing and insulating attic ceilings and ductwork. In addition, it might not be desirable (in hurricane or wildfire areas) or practical (in retrofits) to add roof vents at soffit locations. Accordingly, there may not be any practical alternative to moving the air control layer and thermal control layer to the underside of the roof deck.

1.3 Relevance to Building America’s Goals

The energy savings goals set by the U.S. Department of Energy’s Building America program for both new and existing homes are 30%–50% relative to a home built based on the 2009 International Energy Conservation Code.

The insulation methods used to achieve the energy use reduction goals set out by the U.S. Department of Energy should not result in moisture-related durability risks or failures. Research nto the performance of current recommended assemblies that use SPF should be completed, through theoretical modeling and field review of installed roof insulation methods, to determine if the recommended assemblies have long-term associated durability risks.

1.4 Cost Effectiveness

Durable assemblies exist for a long time and make the best use of construction resources. Because the assemblies last for a long time, their energy resource use should be considered over the life of the assembly, and should therefore be substantially reduced during the initial design. Improving the moisture tolerance and durability of an assembly is also necessary to determine the whole-house life cost.

The roof system’s lifespan may be significantly reduced from durability flaws in the design. Replacement of a roof system due to a design flaw is costly and could be avoided with a thorough understanding of the system interactions. The cost effectiveness of this investigation is in the savings of never having to replace a roof system (framing, insulation, and sheathing) over the intended life of the house. According to Building Science Corporation’s (BSC) experience it is estimated that the cost associated with replacing the roof framing, insulation, and sheathing for a typical home is on the order of $30,000–$50,000. Even in the case of localized bulk water leakage failures (approximately 2–10 ft2) the costs to repair the roofing system (and the damage below it) can be very significant. These costs should be avoided with a long-term, durable, energy-efficient design and installation of the original roofing system.

1.5 Tradeoffs and Other Benefits

The increased R-value and airtightness of SPF roof systems improve energy efficiency and occupant comfort by reducing drafts and improving surface temperatures. The durability of these systems and their maintenance requirements and tolerance to the possible operating conditions within the home should be investigated.

1.6 Integration Opportunities

The information developed from this research will help enable the safe implementation of high R-value, airtight SPF roof systems on both prototype homes and homes in production-built communities.

1.7 Contact Information

The following are the BSC Industry Team members involved in this project:

Table 1. Industry Team Member Contact Information

Company
Name
Team
Member
DowGary Parsons
BASFPaul Campbell
HonewellXuaco Pascual
IcynenePaul Duffy

2 Experiment

2.1 Research Questions

The following research questions are answered by this project:  • Are there risks associated with installing SPF under plywood and oriented strand board (OSB) roof decks, specifically moisture and durability issues? • Are roof leaks a serious problem with SPF roof assemblies?  • What is happening in these systems with high measured sheathing moisture content (MC) (i.e., 20+) but with no evidence of damage to the sheathing? • Are there moisture durability risks associated with installing SPF under OSB roof decks in climates with high rainfall?

2.2 Technical Approach

This project hygrothermally modeled a range of rainwater leaks in code-compliant residential unvented roofs that used SPF insulation. Three climate zones were used in this analysis. In addition, field explorations were conducted of in-service residential roofs using SPF insulation. WUFI 5 hygrothermal modeling software was used for the analysis.

The hygrothermal modeling was used to determine drying potentials and assess the relative risks of rainwater leakage in SPF roof systems. Past field work and published work (Straube et al. 2010) has already addressed the condensation risks of various systems in various climate zones.

Hygrothermal modeling was conducted on roof assemblies located in the principal climate zones of interest defining building performance in the lower 48 states—a hot climate with significant rain (Miami), a cold climate (Minneapolis), and a marine climate with significant rain (Seattle). These locations “bracket” the expected in-service conditions of concern for unvented roof assemblies. A number of roofs were reviewed to visually inspect the sheathing of in-service residential roofs using SPF against the underside of the roof sheathing. This enabled the correlation of modeled low sheathing MCs to in-service roofs with measured low sheathing MCs.

The field explorations were designed to provide information about the actual performance of roofs using SPF insulation. The isolated failures of SPF roofs have led some practitioners to believe that these systems trap moisture in the sheathing. The intent of the explorations of in- service roofs was to measure the sheathing MC and verify that it is within a safe level. Field explorations of installed ocSPF and ccSPF roof assemblies were conducted via visual inspection and core samples. Industry partners were approached to source specific installations of SPF under roof sheathing that could be investigated. The evaluations of the assemblies were based on visual examination of the materials, supported by quantitative moisture meter readings, and product sampling where necessary.

The field evaluation locations were selected based on availability and timing. All locations that were made available by the industry partners were evaluated.

3 Analysis Background

3.1 Unvented Roof Systems

A successful roof (or roof-ceiling assembly) will perform the following tasks:

  • Provide a water management system to keep precipitation out.
  • Provide an air barrier system between the indoors and outdoors.
  • Provide a thermal control system to keep the heat out during the summer and retain heat during the winter.
  • Provide a vapor control system to maintain a durable environment that does not allow condensation and does not promote mold growth.

BSC experience suggests that when failures occur in wood-frame roofs insulated with SPF at the deck, it is typically due to leakage of bulk water (precipitation), or vapor diffusion condensation. Vapor diffusion condensation can occur as an outward drive or as an inward drive. Proper roof enclosure system design can avoid the majority of failures. SPF insulated roofs are common in retrofit work. In retrofit work, the order of work to be considered is important. Health and safety issues must be addressed first and are more important than durability issues. Durability issues are in turn more important than saving energy. Lstiburek (2010) provides the background and approach for the preparatory work necessary prior to insulating an attic. The guide focuses on combustion safety, ventilation for indoor air quality, and attic ventilation for durability. The guide provides a scope of work and specification for the air sealing of many points of air leakage in common attic spaces.

Unvented attic assemblies, or cathedralized attics that move the insulation and airtightness planes to the slope, have been developed to overcome two major problems with vented attics (Figure 1). These problems are:

Locating ducts/air handling units in the attic space causes major air leaks of conditioned air (and thus forced infiltration/exfiltration), and heat/loss gain through the ductwork.

Designs with complex coffered ceiling planes, numerous penetrations by lights, speakers, vents, etc. make it practically difficult to achieve the airtightness required just below the insulation layer.

Figure 1. Cathedralized or unvented attics

All unvented attic and cathedral ceiling designs must provide for either a very high degree of airtightness or avoidance of condensation by warming sensitive surfaces. To meet durability goals in most applications, the airtightness must be provided by a continuous membrane— preferably adhered to the top surface of the structural roof deck and under rigid insulation that provides condensation control. In designs where the airtightness is provided between framing elements, SPF has been found to be a practical solution. However, all wood-to-wood joints in the framing must still be sealed. Figure 2 shows the application of SPF to form an air barrier in a hybrid roof system.

Figure 2. Example high-R hybrid unvented cathedralized ceiling/attic

3.2 Code Requirements for Roofs

The 2012 International Residential Code (IRC) defines vapor retarder class information. A vapor retarder is defined as: A measure of the ability of a material or assembly to limit the amount of moisture that passes through that material or assembly. . .

Download the complete report here.