< Previous20 Spring 2019 • Ontario Building Envelope Councilwith the aerogel blanket thermal bridge correction.It was observed that the annual heating and cooling energy increased in the model with the thermal bridges in the building. From the study, concrete buildings saw up to a 32 per cent increment, whereas steel buildings saw up to a 35 per cent increment when thermal bridges were accounted to the model. This was expected, as the thermal bridges increase the amount of heat loss, thereby increasing the energy demand of the building.The use of an aerogel-enhanced blanket in the steel and concrete building signifi-cantly reduced the amount of space heating energy. Figure 2 (on page 18) shows the improvement that can be achieved annually per square-metre of the conditioned area for concrete and steel buildings, respective-ly, for Toronto’s climate.The equivalent U-value method under-estimates the space heating load and over-estimates the cooling load, which is similar to the previous studies.1,6 Figure 3 (on page 19) illustrates the comparison in percentage difference of the energy demand calculated using the equivalent U-value method and the 3D dynamic method for concrete and steel buildings, respectively.The difference of heating load between the 3D dynamic method and the equiva-lent U-value method was found to be 21 per cent in the concrete building and 22 per cent in the steel building. The cooling load difference was found to be one per cent in the concrete building and six per cent in the steel building. With the use of an aerogel-enhanced blanket to correct the thermal bridges, the discrepancy in annual heating load decreases. It was found that the difference between the 3D dynamic and the equivalent U-value method is reduced by between nine per cent and four per cent, respectively, for the concrete building and steel building. The discrepancy in annual cooling load between the 3D dynamic and equivalent U-value method seems higher, as the annual cooling load is comparative-ly much less than the annual heating load. The reduction of the discrepancy is due to the improvement of the thermal bridges. As a result, it also reduces the thermal in-ertia effect in the building.From the literature studies, the thermal inertial effect not only changes the resist-ance but also changes the dynamic behav-iour of the wall, which is consistent with the findings from the comparison of the 3D dy-namic method and equivalent U-value meth-od. The decrease in difference suggests the equivalent U-value method can also give ac-curate results if the details are improved. The improved details not only save the energy demand but also avoid the complex and time-consuming 3D dynamic method. In this research, the aerogel-enhanced blanket of 10 millimetres / 20 millimetres was proven to improve the construction de-tails in two different building models. A sig-nificant reduction of the heating load was observed with the application of an aero-gel-enhanced blanket in the building. This study shows that the aerogel-enhanced blanket can provide a better opportunity for the thermal bridging correction, as it lowers the building heating by six to eight per cent in the concrete construction and by 11 to 18 per cent in the steel construc-tion. This shows that an aerogel-enhanced blanket with a thickness of 10 millimetres / 20 millimetres can be used as a thermal break in both new and retrofit buildings. A higher improvement can be achieved with an increased thickness. Due to high ther-mal performance, low density, weight, and thickness of an aerogel enhanced blanket, it can be applied in air spaces and areas with both space and weight constraints. The two different modeling approach-es in the building suggest the equivalent U-value underestimates the heating energy and overestimates the cooling load. The 3D dynamic method is more accurate than the equivalent U-value method for proper calculation of the energy demand. The difference between the 3D dynamic method and the equivalent U-value meth-od decreases when the details are im-proved with the aerogel-enhanced blan-ket, as the dynamic effect of the thermal bridging will be reduced. Nonetheless, for standard details, the 3D dynamic method is required for the most accurate results, but if details are improved, the equivalent U-value method can also predict an accur-ate energy demand. nRashmi Sharma, B.Arch., CPHC, has an undergraduate degree in architecture from Tribhuwan University and a master’s degree in building science from Ryerson Universi-ty. In addition, Sharma is a certified Passive House consultant. For extended results from this study, please contact the authors.Dr. Umberto Berardi, M.Sc., Ph.D., P.Eng., is an associate professor in the Faculty of En-gineering and Architectural Science at Ryer-son University in Toronto. His main research interests are related to the study of new inno-vative materials for energy saving applications. Berardi directs the BeTOP Ryerson lab for new building system and technology develop-ment and testing.REFERENCES1. Baba, F., & Ge, H., 2017, Dynamic Effect of Thermal Bridges on the Energy Demand of Residential Buildings in British Columbia, 15th Canadian Confer-ence on Building Science & Technology2. Cuce, E., & Cuce, P., 2016, The Impact of Internal Aerogel Retrofitting on the Thermal Bridges of Residential Buildings: An Experimental & Statistical Research3. Berardi, U., 2018, Aerogel-Enhanced Solutions for Building Energy Retrofits: Insights from a Case Study4. Berardi, U., 2018, Aerogel-Enhanced In-sulation for Building Applications, Nano-technology in Eco-Efficient Construction5. National Research Council Canada, 2016, National Energy Code of Canada for Buildings6. ASHRAE 1365-RP, 2011, Thermal Per-formance of Building Envelope Details for Mid- and High-Rise Buildings7. ASHRAE, 2016, ANSI/ASHRAE/IES Standard 90.1-2016, Energy Standard for Buildings Except Low- Rise Residential BuildingsFEATURE n n n...the thermal inertial effect not only changes the resistance but also changes the dynamic behaviour of the wall, which is consistent with the findings from the comparison of the 3D dynamic method and the equivalent U-value method.Pushing the Envelope Canada 23A few years ago, the Toronto office of our engineering firm experi-enced what we termed “the year of the crappy curtain wall.” In reality, it was more like 18 months, and during that period, we investigated persistent water penetration in several low- to mid-rise commercial office buildings ranging in age from five to 20 years old. The failures discovered were not unique to curtain wall systems and were entirely avoidable if those responsible had even a basic understanding of how these systems are supposed to function. For some of the buildings we investigated, the failures were in conventional, fully captured curtain wall systems that were not installed correctly from the get-go. There are many excellent, qualified curtain wall installers out there, but apparently, they were off benefitting other jobs when these buildings were constructed.Any fenestration systems can be ad-versely affected by inexperienced or unqualified installers. A new high-rise condominium, specified to be rainscreen, had the misfortune of being fitted with a modified window wall system that started off being face-sealed and ended up being a reluctantly draining rainscreen system. The developer passed off a cheaper solution to the owners, and the manufacturer sold the developer their modified system with emp-ty promises of performance equality. This “lipstick on a pig” project saw multiple hacks into the window wall system that loosely complied with the project specifi-cations but ended up leaving the building with a systemic water penetration problem. The performance of even the best quality systems can be jeopardized if basic rules of engagement are not followed.Our experience in the Greater Toronto market is, unfortunately, not unique, and we have seen similar unnecessary fenestra-tion failures across North America. With the proliferation of fenestration systems in new construction and retrofit (often with increased level of sophistication to meet aesthetic requirements), the surge toward in-plant glazing, and the frequency of performance problems, it is important to understand some of the most significant causes of failure so they can be avoided. It is also important to know the tells that offer clues to where the failures are occur-ring, and how you can significantly reduce the chances that fenestration failures will damage your interior finishes and spoil comfort and the living experience. Quality fenestration systems are en-gineered to manage water and, at least in Canada, have been designed as rainscreen systems for decades using either wet/wet or wet/dry configurations, which assume that only manageable volumes of water pass the outer moisture seals. Today, rainscreen fenestration systems are still the system of choice for building scientists because they provide redundancy, protection for By Peter Adams, P.Eng., Principal & Senior Building Envelope Engineer, Morrison Hershfield n n n FEATUREIt’s All About the Plumbing!Fenestration Systems: This glazing system on the condominium above can be represented by the slightly more chaotic plumbing on the right. Ok, not quite—but you get the idea.24 Spring 2019 • Ontario Building Envelope Councilsensitive elements from UV, and peace of mind. Building owners and operators should embrace them for the same reasons, although they are sometimes deterred and distracted by other, less costly fenestration systems. There are jurisdictions (such as in Ontario) where professional insurers re-quire rainscreen for certain types of fenes-tration systems as a condition of coverage because of the past high volume of claims related to failures. So why do fenestration systems cry? They need more love! The vast majority of rainwater that hits glazing is, in theory, supposed to be stopped by the exterior metal, seals, or glass surfaces and run harmlessly to grade. As with any rain-screen system, it is expected that the out-er layer is not perfectly sealed and that a portion of the water will end up inside the glazing pocket of the framing system. What comes next is the make or break point for any fenestration system, no mat-ter the level of sophistication.Fenestration is like plumbing—trying to put more water in a piping system than it can handle leads to a back-up and an un-happy owner. Most rainscreen (also called drained and vented) fenestration systems are fully capable of managing an expected volume of water. Things quickly fall apart, however, when the volume of water en-tering the wall exceeds the volume of water that can be successfully drained. The sys-tem drainage capability is a combination of the quality of the original design and how the system is fabricated and installed.Blocked internal drainage paths often walk hand-in-hand with leaky fenestra-tion systems. An overzealous installer (or someone from “Others Contracting”) with a caulking gun can quickly undo all the en-gineering, shop drawing review, and testing that went into what could otherwise have been a successful system installation. Much the same as a clogged pipe, blocked drain-age paths in a glazing system will often re-sult in conditions that are detrimental to internal seals and lead to water intrusion even under modest rain events. There is a balance at play here—the more restricted the drainage flow within the fenestration system, the longer it would take for the same volume of water to drain, causing back up and overflow conditions. This back up condition increases the length of time internal seals are exposed to water, FEATURE n n nRetained water trickled from screw holes when backed-off. A little more love and attention to drainage would have prevented these tears.Pushing the Envelope Canada 25decreasing their longevity and resulting in reoccurring leaks.Design changes over the years (some-times implemented to reduce uncertainty on-site) have resulted in other potential problems. The elimination of soft seals and the movement toward more dry/dry systems (in lieu of wet/wet or wet/dry sys-tems) has meant the water tightness of fenestration assemblies rely heavily on product tolerances and the pressure gen-erated by gasketed pressure plates. All gaskets and tapes are not created equal—some of the gasket compounds are prone to compression set over time, hence de-creased pressure onto glass surface, per-mitting increased water entry. Fastener torques are often not measured or even specified to achieve adequate water tight-ness. Have the manufacturer approved and specified products been installed, and is there a history of long-term perform-ance? These important issues need to be identified, and the underlying problems must be solved before they are permitted to turn into costly liabilities.So, how can we keep the fenestration systems airtight, watertight and draining free? The following is recommended to achieve the maximum performance po-tential of fenestration systems from con-cept to construction:• Select fenestration systems that are “in-staller proof” with quality material, “robust” internal seals and drainage mechanisms that cannot be easily botched during installation. Always make sure quality systems are in play from the start. • Visit the shop where wall compon-ents are being fabricated and check the frame’s “birth certificate” for lineage.• Ensure specifications are clear to elim-inate the inclusion of modified face sealed fenestration systems. Only grass-roots and proven rainscreen systems from reputable manufacturers need apply. • Mock-ups are required to establish the standards of installation, and work with the installers so they have an understanding of intent. They provide the opportunity to review contractors’ workmanship and identify potential transition issues. They also increase the comfort level of all parties involved for future installations. They can be n n n FEATUREThis “lipstick on a pig” project saw multiple hacks into the window wall system that loosely complied with the project specifications but ended up leaving the building with a systemic water penetration problem. The performance of even the best quality systems can be jeopardized if basic rules of engagement are not followed.26 Spring 2019 • Ontario Building Envelope Councilconstruction process (go or no-go). Once completed and agreed upon by all parties, the mock-up could be used as a benchmark for level of quality expected during construction. • Finally, and importantly, push hard to have qualified and frequent adult supervision during construction. Many problems we observed could have been caught very early in a project with an experienced site presence, sharp eye for details, and a good understand-ing of the functions of the fenestration systems. nPeter Adams, P.Eng., is a principal and a senior building envelope engineer at Mor-rison Hershfield. He has been specializing in building science since 1992. Peter grad-uated as a mechanical engineer and spent the early part of his career at the National Research Council in Ottawa. He has con-ducted work on hundreds of properties, and his work has included forensic studies on building failures and building component design. He has extensive experience with indoor environment studies and mould risk assessments.His work has included many buildings with challenging building envelopes and operating conditions, including libraries, water treatment plants, and sporting venues. Adams currently teaches building science and related topics at a university level. He is a past-president of OBEC and past-chair of ASHRAE Technical Committee 4.4 on Building Envelope and Materials.REFERENCES • Ontario Architects Association, OAA Window Wall Endorsement, July 28, 2009 • Pro-Demnity Insurance, Pro-Demnity Non-Drained Exterior Wall Exclusion• Kesik, T., and Saleff, I. 2009, Tower Renewal Guidelines for the Compre-hensive Retrofit of Multi- Unit Resi-dential Buildings in Cold Climates, University of Toronto• Hoffman, Stéphane P. 2001, Adap-tation of Rain-Screen Principles to Window-Wall Design, Thermal Per-formance of the Exterior Envelopes of Whole Buildings VIIIWater retained within an insulating glass unit is a near-sure sign of drainage problems. Okay, the fish was added in Photoshop, but wouldn’t that be cool?FEATURE n n nbuilt, examined and tested in a shop, as a standalone mock-up on-site, or as part of the work to remain.• Pair the mock-up with a progres-sive site testing program to put the system through its paces, either before it gets on a building, or soon into the Pushing the Envelope Canada 27When a designer is considering options during a low-slope roofing project, in most cases, the first decision is the assembly type. The choice is typically limited to two options: conventional roof assemblies, or protect-ed membrane roof assemblies.In conventional assemblies, the roof membrane is found at the top of the as-sembly, above the insulation, and is con-sidered exposed to the elements. If the roof membrane is not made of a UV stable material, it is protected from UV by a sur-facing layer. The surfacing can include granules, pea gravel, pavers, metal, or liquid-applied coatings. The roof assem-bly layers are attached to the immediate layer underneath, or to the deck, through mechanical fastening or adhesives.Protected membrane roof assemblies are also known as inverted roof assem-blies. In a typical inverted configuration, the roof membrane is located at the level of the structural deck or deck sheathing. As such, it is protected from the elements by other layers of the roof assembly, in-cluding thermal insulation. The roof membrane is protected from UV exposure and from thermally induced dimension-al stress. Roof assembly layers above the roof membrane are, by and large, loose laid and resist wind uplift by a weighted ballast.In general, there are a few advan-tages of protected membrane assem-blies when compared with conventional assemblies. If correctly installed and maintained, protected membrane as-semblies can have a longer lifespan compared to conventional assemblies and can also be expected to have a lower incidence of leaks over the entire life-span of the roof system.For a low-slope roof system to operate most effectively, slope to drains is required. The rule of thumb is two per cent slope at minimum for asphaltic-based convention-al roof systems.1 Ponding of water, even if protected from UV by the layers above the membrane, can be detrimental to the durability of certain roof membranes. Protected membrane roof assemblies are often installed on decks that are structur-ally sloped. Conventional roof assemblies can also be installed on structurally sloped decks, but in many instances, they incor-porate the slope within the assembly itself in the form of tapered insulation.During the design review process for a new construction project in the Greater Toronto Area, roofing was discussed, and the design team understood the benefits of a protected membrane roof. However, earlier in the project, a conventional roof assembly with tapered insulation had been specified and a flat structural steel deck was built.At this stage, the question was whether it was still possible to install a roof assem-bly that provided the benefits of a protect-ed membrane roof. The team ruled out switching to a protected membrane sys-tem, as it was not practical to provide the minimum amount of slope in the struc-ture. Positive drainage and good roofing practice were required by the contract documents.The team considered a “hybrid” roof assembly that combined a convention-al roof assembly, including tapered in-sulation, with a layer of insulation (and other layers) on top of the conventional assembly.Could a problem be created by install-ing this hybrid roof assembly in our heating climate? Consideration was given to the hygrothermal performance of the proposed hybrid roof assembly, taking into account that the roof membrane in a hybrid con-figuration would be kept warmer during Comparing Condensation Potential in Conventional and Hybrid Roof Assemblies: A Case Study By David Wach, M.A.Sc., Engineer-in-Training, & Vladimir Maleev, M.Eng., P.Eng., BSSO, Associate, Engineering Link Inc.n n n FEATURELayers of a typical conventional roof.28 Spring 2019 • Ontario Building Envelope Councilheating seasons by the presence of thermal insulation over the roof membrane when compared to a conventional configuration but would be cooler than a protected roof membrane with the same total amount of insulation. Clearly, the heat transfer and moisture characteristics of the hybrid as-sembly would be different than the two trad-itional options for low-slope roofs. CONDENSATION POTENTIAL IN CONVENTIONAL ROOF ASSEMBLIESIn the past, low-slope roofs were made of dark materials, including pitch and bitumen, which readily absorbed long-wave radiation from the sun. On sunny days, during both winter and summer, the temperature of the dark membrane could be up to 80°C as it absorbed solar radiation, increasing the temperature of the roof assembly and any moisture within it. The heating caused a vapour pressure gradient, driving the vapour down to the interior space. The absence of a vapour retarder at the deck was tolerated because of solar vapour control.2The introduction of light coloured, re-flective roofs, often referred to as “cool roofs,” introduced moisture problems in some assemblies where an effective vapour retarder did not exist. The light colour / reflectivity of the roof membrane resulted in a reduced amount of solar heating of the roof assembly. The reduced absorption of solar radiation and resulting lower temperature of the roof assembly was desirable from an energy perspective, but solar control of embedded moisture was reduced. In some cases, moisture problems from condensation developed, leading to deterioration of the roof assem-bly and structure.3Modern, conventional roof assemblies in heating climates that are installed in low-slope roof systems often contain two vapour barriers that sandwich insulation and other roof system layers. Building codes in Canada require that an air bar-rier and vapour retarder (AB/VR) be pro-vided in all building envelope assemblies, including roofs. The roof membrane acts as a second vapour retarder, as it’s usually made of materials that have a low vapour permeability. During heating seasons, the underside of the roof membrane in a conventional roof system can be cool if the exterior air temperature is cool. If warm, moist air exists within the roof assembly between the two vapour retarders, and it may con-dense into liquid water as it contacts ma-terials with a temperature below the dew point of the air that contacts it.If an AB/VR exists, how could mois-ture end up in the roof assembly? The mechanism could be bulk air movement or vapour diffusion despite the AB/VR. Poor detailing or poor installation of the air and vapour control layer tie-ins with adjacent building envelope assemblies and around penetrations can cause discontinuity in the control layers. Of course, moisture can also enter the roof assembly through breaches in the membrane (leaks) or can be embedded in the roof assembly from construction. As the two vapour retarders exist, once the liquid water is present, it can have a difficult time drying, as vapour diffusion potential is greatly reduced.With the hybrid type roof assembly, the membrane is kept at a temperature closer to the interior space than the ex-terior air when compared to the mem-brane in a conventional roof assembly. Theoretically, the AB/VR should keep moisture out and the exterior insulation will keep the membrane relatively warm. The membrane is protected from solar radiation. But if moisture does get into the roof through poor detailing, installa-tion, or leaks over time, how will the roof perform?To confirm that a moisture prob-lem would not be created by the hybrid FEATURE n n nFigure 1. Simulated temperature data.Figure 2. Simulated relative humidity data.Pushing the Envelope Canada 29n n n FEATUREassembly, and to assess its hygrothermal response under load, a study was under-taken to analyze the moisture response of the assembly over time in a southern On-tario climate.HYGROTHERMAL SIMULATIONS AND DISCUSSIONWUFI Pro 6 modelling software is a one-dimensional hygrothermal analysis program that performs simulations of coupled heat and moisture transfer across user-defined building envelope assemblies. Simulations to compare the performance of the hybrid roof assembly to the conven-tional roof assembly were performed. The simulations were carried out over three years in order to evaluate the long-term hygrothermal performance of the roof assemblies. Exterior conditions were as per the built-in Toronto Cold Weather Year climate file. Interior conditions were held constant at 21°C and 40 per cent RH. Rain exposure was as per ASHRAE 160.The typical conventional roof assembly modelled included a two-ply modified bitu-men roof membrane, asphaltic recovery board, 150 mm of rigid polyisocyanurate insulation, a self-adhered membrane act-ing as both an air barrier and vapour re-tarder, exterior grade gypsum board, and a metal deck. The hybrid assembly contained the same layers as the conventional assem-bly, with the addition of 125 mm of extrud-ed polystyrene (XPS) insulation outboard of the roof membrane. Note, to be con-servative for dew point analysis, the same thickness of polyisocyanurate insulation was included within the hybrid assembly as was in the conventional assembly. In prac-tice, the hybrid roof assembly would be expected to have only tapered insulation below the membrane.Figures 1 and 2 (on page 28) show monthly average temperature and relative humidity data over the three-year time per-iod for the conventional roof assembly, the hybrid roof assembly, and the exterior. The relative humidity data within the roof assem-blies was taken just under the membrane.The temperature and relative humid-ity in the conventional roof vary with the exterior temperature. In Figure 1, the exterior temperature is difficult to see be-cause the membrane temperature in the conventional assembly closely matches it. The hybrid roof has a much smaller tem-perature amplitude than the conventional roof, as it is thermally insulated from tem-perature fluctuations of the exterior air by the exterior insulation. As such, the roof membrane in the hybrid roof assembly is exposed to less thermally induced stress. The analysis also showed that the hy-brid assembly has lower peak and average relative humidity than the conventional roof. As a result, the conventional roof assembly retains more moisture over time than the hybrid roof assembly does. This leads to a higher risk of condensation below the conventional roof membrane in the winter when compared to the hybrid roof assembly. The analysis also suggests that the hybrid roof has better drying po-tential since the relative humidity seems to decrease at a faster rate.CONSIDERATIONS AND LIMITATIONSThe hygrothermal analysis does not include for airflow through the assembly and assumes that the AB/VR in the as-sembly and the roof membrane are air-tight. The AB/VR membrane layer over the sheathing and steel deck should be applied in an airtight manner around all penetrations and the roof perimeter. This helps prevent the flow of warm, moist, bulk air into the roof assembly from the interior space during heating seasons due to convection. Initial moisture within the assembly in modelling was as per program defaults for each material. All wetting and drying in the modelled scenario are ac-counted for by vapour diffusion through materials. In the situation presented, the As a result, the conventional roof assembly retains more moisture over time than the hybrid roof assembly does. This leads to a higher risk of condensation below the conventional roof membrane in the winter when compared to the hybrid roof assembly.Next >