Making a case for sprayfoam in the unvented attic

October 17, 2017

Photos ©

By Peter Birkbeck, CTR
Unvented attics have been designed and constructed for some 30 years. They are referred to by various names, including conditioned attics, semi-conditioned attics, indirectly conditioned attics, hot roofs, and compact roofs. Across North America, they have been the subject of numerous technical studies, papers, and—in some cases—building code proposals and changes. However, the strategy has rarely received as much attention as it has in recent years.

For the most part, unvented attics are a residential phenomenon. For residential low-sloped roofs, most building codes require attic space ventilation; Canadian building codes typically call for a net free ventilation area of 1:300, or 0.09 m2 (1 sf) in 29 m2 (300 sf) of insulated floor area.

This is increased to 1:150 for roofs of slope less than 1:6. Other provisions in the National Building Code of Canada (NBC) speak to a high-low distribution of venting, as well as a symmetrical provision of roof venting on either ‘side’ of the roof.

Creating an unvented roof generally involves moving a combination of the thermal, air, and/or vapour control layers to below the roof sheathing without any venting whatsoever. In a traditional design, these layers would be located at the horizontal plane of the ceiling.

In conventional attic construction, two designs are possible. In the first configuration, a sloped roof encloses the entire attic (i.e. roof) space, which is ventilated. Thermal insulation is placed above the uppermost horizontal ceiling finish, and the air and vapour control layers are also located in this plane. Depending on the design and climate, the resulting space may be used to hold mechanical equipment and services (such as space-conditioning equipment and ducting), provide access for repair and maintenance purposes, and possibly allow for storage or future occupancy.

The second option is a cathedralized attic, where all control layers are located at the sloped roof plane, just inside the interior finish. Some designers and owners find this approach more visually pleasing.

From C.J. Schumacher’s 2008 “Hygrothermal Performance of Insulated, Sloped, Wood-framed Roof Assemblies,” Masters Thesis presented to the University of Waterloo, this illustrates the traditional (i.e. unvented) approach to constructing an attic space.
Images courtesy Icynene Inc.
Also from Schumacher’s 2008 thesis, this image illustrates the unvented strategy discussed in this article.










Why construct an unvented roof space?
Other than the practical reasons and visual appeal already mentioned, unvented roof spaces may have additional collateral benefits, such as:

Schumacher’s 2016 “Research Summary – Field Performance of Open-cell SPF Insulated Unvented Roof Assemblies in the Climate of Vancouver, B.C., Canada” illustrates the declining mould index that can apply with open-cell sprayed polyurethane foam.

History of unvented attics and relation to building codes
Some of the first instances of the strategy in actual practice took place in the late 1980s. One Canadian municipality, reporting to an assembly of building officials from large municipalities, noted no early deterioration of roofing materials in its experience with the strategy, spanning some 15 years. (This information is derived from the Toronto Area Building Inspectors Committee [TABIC] Minutes from March 2006.) At this time, it is likely each installation was the subject of a unique building department review and approval, as these would have predated the advent of ‘objective-based’ codes.

The intent of attic space ventilation is to remove unwanted moisture. However, as the Appendix Notes to Section 9.19.1 of the 2015 NBC still point out, for constructions that are sufficiently airtight, the required ventilation may be omitted if it can be shown to be unnecessary. It is interesting to note this language has, with little to no modification, carried on to the present day. To take advantage of the exception now, an application for an alternative solution would typically be presented to the building official.

Even in 1993, research summaries suggested no strong correlation existed between ventilation rates, the location of vents, and moisture content of wood members. (For more, consult “Attic Ventilation and Moisture,” part of Canada Mortgage and Housing Corporation’s [CMHC’s] Technical Series 92-201, Research and Development Highlights.) As part of the same study, high ventilation rates in a marine climate were reported to actually result in higher roof sheathing moisture contents. More building researchers began studying and writing about the topic.

In general, the primary durability concern for unvented attic designs has been the sheathing moisture content and its potential for mould growth throughout the seasons. Approximately 20 per cent moisture content is generally considered a safe level. Contents between 20 and 28 per cent are cautionary, as wood decay can continue following the onset of the process.

Some studies have gone further, using the temperature of a substrate and relative humidity (RH) to calculate a ‘mould index.’ First developed at the VTT Technical Research Centre of Finland and the Fraunhofer Institute, the index is consistent with current consensus standards for moisture design, such as American Society of Heating, Refrigerating, and Air-conditioning Engineers (ASHRAE) 160, Criteria for Moisture-control Design Analysis in Buildings. For wood materials, a rating of no greater than ‘3’ is generally considered acceptable. This would correspond to a coverage of less than 10 per cent visually, and less than 50 per cent under a microscope.

Closed-cell SPF provides a single, monolithic insulation, air, and vapour control layer in this example of a low-slope roof.

More recent studies
Significantly, one study concluded in the Pacific Northwest marine climate, relying solely on attic ventilation actually contributed to elevated moisture content in plywood roof sheathing in sloped wood-framed constructions. (The study in question was G. Finch et al’s “The Problems with and Solutions for Unvented Attics,” from the Proceedings of the 30th RCI International Convention and Trade Show in 2015.) In fact, among other things, the authors concluded the moisture content of the plywood was largely influenced by outdoor climatic conditions, such as exterior ambient temperature and relative humidity (RH). Night-sky cooling of the roof sheathing was a primary mechanism identified for these findings in small, well-constructed, ventilated roof spaces. Further, when the intentional impeding of sources of exterior wetting was investigated by the use of a vapour-impermeable roof membrane, a reduced potential for the plywood roof sheathing to dry was observed.

For more than one year, the University of Waterloo’s Building Enclosure Group (BEG) studied both open- and closed-cell foams in-field in a test-hut environment in Southwestern Ontario. A combination of vented and unvented assemblies was studied, including vented fibreglass. The most notable conclusions were:

In the case of the unvented open-cell assembly, no visible damage or mould was observed on the framed components or wood panel. Both vented and unvented closed-cell SPF assemblies exhibited stable data with little cause for concern.

As is often done in such research, the assemblies were subsequently modelled in an attempt to confirm the in-field results. Most simulations of unvented assemblies can work in all except climate zones in the extreme north (heating degree days [HDD] 18 C  [64 F] > 7000). Specific findings indicated the installation workmanship of the open-cell SPF was important. In all cases, as one might expect, keeping the interior RH level below 35 per cent resulted in better-performing assemblies.

The most important aspects for durability were found to be interior relative humidity and airtightness. These criteria are especially important to open-cell sprayfoam, as well as the provision of adequate vapour control. Although sheet polyethylene yielded good results as a vapour control strategy with open-cell SPF, is was noted it inhibits both drying to the interior and moisture redistribution. Vapour-barrier paint on the drywall could provide some vapour control; however, even more control was suggested in this case.

Vaulted or cathedral ceilings create more usable space. A service space can also be created with a short knee wall.

In the case of the vented fibreglass batt and sheet polyethylene simulation, it was concluded there was no risk of excessive sheathing moisture content in any climate zone, provided there was no air leakage. In traditional vented attic designs with the insulation in the ceiling plane, this would be hard to deliver in practice, especially with the tendency for ceiling penetrations to accommodate such features as recessed light fixtures. Further, the ‘traditional’ fibreglass-insulated assembly with polyethylene was assumed to be airtight; once again, this is likely difficult to deliver in the field.

This work resulted in the Canadian Spray Foam Guide, which concludes SPF installed in unvented attic designs can work in all climate zones, and full closed-cell sprayfoam is recommended for all climate zones. (For more, see the 2013 Canadian Spray Foam Guide from J. Smegal, J. Straube, and A. Grin of Building Science Consulting.) Specific cautions are advised for open-cell SPF, ‘hybrid’ insulation strategies (i.e. foam plus fibre), and insulating sheathing on the exterior of the roof deck. Most of these concerns centre around the provision of adequate levels of vapour control. Care must be taken with a hybrid design to ensure sufficient air-impermeable insulation is provided to control air infiltration. In other words, one should provide a material and thickness that qualifies as an air barrier. In all climate zones, one must also provide a sufficient ratio of air-impermeable to air-permeable insulation, such that the dewpoint does not occur on a surface with condensation potential.

A private study finalized in 2016 used the mould index model to evaluate data from an earlier instrumented actual house, as well as actual roof constructions evaluated in a test-hut scenario. (This study was C. Schumacher’s 2016 publication, “Research Summary–Field Performance of Open-cell SPF Insulated Unvented Roof Assemblies in the Climate of Vancouver, B.C., Canada,” published by Building Science Consulting.) Mould indices no greater than the threshold of ‘3’ were calculated for the test-hut constructions, and indices over multiple seasons were seen to be declining. After four years, they reached an index not unlike that of the traditional ventilated (and fibreglass-insulated) attic.

In terms of the two open-cell SPF insulated attics, a similar declining profile for mould index was observed for both a scenario with no vapour control layer and one with vapour retarder paint applied to the surface of the foam. Interestingly, the performance of the open-cell SPF with no vapour control paint applied was marginally better.

These findings were confirmed with purposely made inspection openings in the test-hut scenario, as well as in actual in-service installations in the same climate zone. Roof sheathing moisture content observed was no higher than 16 per cent, with the majority being well below that. No visible mould was observed.

Where suitable vapour control is provided (not shown), open-cell SPF has been shown to be an effective performer in some climate zones.

Other considerations
Many factors come into play when ensuring the performance of an unvented attic.

Vapour control with open-cell SPF
Vapour control can be achieved by installing a material with a vapour permeance of 1 perm or less. Naturally, closed-cell SPF, at thicknesses that would be typically installed in a roof space, would meet this requirement without the addition of any other materials. However, the alternative strategy of applying a vapour retarder paint over the interior surface of open-cell SPF first appeared in the literature around 2003. (More information is available in 2004’s “Unvented Cathedralized Conditioned Attics: A Comprehensive Update,” by A. Rudd of Building Science Corporation, and in 2005’s “Understanding Attic Ventilation,” by J. Lstiburek of Building Science Corporation.)

It may be helpful to the designer or regulator if the sprayfoam manufacturer is able to supply the assembly testing that was performed to qualify the particular performance of the foam plus paint combination.

Shingle life
The first and second greatest influences on roof temperatures are the roof’s geographic location and the direction it is facing. (This is derived from “What’s the Value of Ventilation” by C.G. Cash and E. Lyon, published in Professional Roofing Magazine in March 2002.) A rise of about 2 C (2 to 3 F) in asphalt shingle temperature and a corresponding rise of 6 C (10 F) in sheathing temperature were reported by researchers as early as the late 1990s. However, since the effect of ventilation decreases closer to the exterior surface of the shingle, an absolute temperature change of less than two per cent was suggested. Reviewers of published work examining this issue concluded it is unlikely the service life of roof shingles was extended by the use of attic ventilation, and that their colour played a more significant role. (Read the 1999 publication, “Issues Related to Venting of Attics and Cathedral Ceilings,” by A. TenWolde and W. Rose in ASHRAE Transactions V.105, Pt.1, for more information.) Shingle durability and all commonly cited benefits of attic ventilation were more influenced by other strategies.

Recent developments
There is a long history of unvented attic performance in the United States, beginning with Building America demonstration houses in the early 2000s, which resulted in published design guidance and building code prescriptive language. Nonetheless, it is now the position of some U.S. researchers that reported problems with unvented attic designs in all climate zones are related to installations of open-cell sprayfoam where moisture is accumulating in the upper portions of the attic space. The proposed solution involves the introduction of air (via various possible strategies) or dehumidification to the space. (For further reading, see “Building Science Insight [BSI] 016: Ping Pong Water and the Chemical Engineer” by J. Lstiburek, published by Building Science Corporation in October 2016.) This radical departure from earlier guidance needs more substantiation.

When combined with strategies responding to the cautions building enclosure researchers have identified, all types of sprayfoam can, for the most part, work in all climate zones. As with any construction product, installation and workmanship are important. Combined with the proper detailing, sprayfoam is an effective solution for providing air impermeability (i.e. air control layer) and, in the case of closed-cell SPF, an integral vapour control layer.

Guidance to inform homeowners, such as was published by Canada Mortgage and Housing Corporation (CMHC) in the “About Your House” series at least as far back as the early 2000s, suggested that, especially as a standalone strategy, attic ventilation is overrated. A well-sealed ceiling and the maintenance of lower interior humidity levels were cited as the key influencers of roof assembly durability. The unvented attic design generally has a good in-service track record and is deserving of a more explicit treatment in Canadian building codes as a viable design option.

Peter Birkbeck, CTR, LEED Green Associate, is a codes and standards specialist with the engineering division of Icynene Inc., a manufacturer of sprayed polyurethane foam (SPF) systems. He has nearly 20 years of experience in polyurethane formulation, manufacturing, and product technical support. Birkbeck is active on numerous industry committees, and is a member of CSC, ASTM, American Society of Heating, Refrigerating, and Air-conditioning Engineers (ASHRAE), Canadian Home Builders’ Association (CHBA) Technical Research Committee, and the Spray Foam Coalition–Canadian Work Group. He can be reached via e-mail at[8].


 Further Discussion…
 After we published this article in the October 2017 issue of Construction Canada, a reader had concerns regarding ice damming. To read the letter and see the author’s response, click here.
  1. [Image]:
  2. [Image]:
  3. [Image]:
  4. [Image]:
  5. [Image]:
  6. [Image]:
  7. [Image]:

Source URL: