Building Air Barrier Testing and Veriﬁcation Using Smoke Flow Testing and Infrared Thermography in the Canadian Arctic: A Case Study - 2002-2003

By: Richard Ogle & Bill Wyness

Introduction

This paper summarizes the techniques used to verify and correct air tightness of the building envelope in the new Inuvik Regional Health and Social Services Centre, a recently commissioned $40 million dollar facility in the high arctic.

Since the 1980’s the requirement for air-barrier continuity of building envelopes has become increasingly recognized. Part 5 of the National Building Code of Canada (NBCC 1995) Environmental Separation requires that wherever a building assembly separates an interior conditioned space from an exterior space that it contain an air barrier system. The air-barrier system is required to be continuous, be durable through mechanical support, and conform to stipulated air leakage rates based on the use of low permeability materials and joints between those materials. Uncontrolled air leakage can cause condensation problems, increase energy consumption and adversely affect occupant comfort.

The need for an effective air barrier system is particularly acute in Canada’s Arctic where environmental loads are harsh and maintenance and operating costs are high. Inuvik has a 2-1/2% January design temperature of –46 oC and averages 10,050 Degree Days Below 18 oC. For comparison, Calgary has a 2-1/2% January design temperature of –31 oC and averages 5200 Degree Days Below 18 oC.

Case Study Building

The Inuvik Regional Health and Social Services Centre was constructed in the years 2001 through 2003. Figures 1 and 2

Figure 1: Inuvik Regional Health and Social Services Centre.

Figure 2: The exterior wall system consists of a structural insulated panel system.

The project delivery method was a negotiated, fast-tracked design-build model, where the Project Manager, Ninety North Partners, consisting of Ninety North Construction and Stuart Olsen Construction retained the project design consultants, Stantec. Technical performance requirements stipulated for the facility by the Owner included the NBCC 1995, applicable technical standards for Canadian health care facilities, and recommendations from Good Building Practice for Northern Facilities, a recommended best practices document developed by the GNWT after ten years of technical performance evaluation of buildings throughout the NWT. Since this is a health care facility, the minimum interior relative humidity was required to be 30% and the building was required to be positively pressurized by the air handling system. These are considered rigorous environmental loads for a building envelope in such a cold climate.

The plan area of the facility is approximately 8000 m2 with an irregular shaped ﬂoor plan. The roofs are low-slope with a 2-ply SBS modiﬁed bitumen water shedding membrane in a conventional conﬁguration. The walls are constructed of 100 mm thick structural insulated panels (SIPS). Each panel consists of a pre-ﬁn-ished metal facer bonded to a urethane insulation core. The panels interlock and seal on the vertical edges by means of a tongue & groove type of proﬁle, and ﬁeld applied sealant in the joint between the panels, with a modiﬁed bituminous membrane (MBM) facing strip on the interior panel face. Figure 3. The panels rely for the air-barrier seal at the top and bottom on correctly installed low-permeance materials bridging from the panel face to the roof air-barrier system and the crawl-space air barrier system at the panel base.

Figure 3 The SIPS panels consist of a 100 mm thick urethane foam core with metal facers and interlocking edges.

The SIPS wall system was selected because of its light shipping weight, high insulating value, rapid deployment to quickly enclose the building during the short arctic construction season, and lateral load resistance. However there were some concerns with the use of this system with respect to continuity of the air barrier, particularly at locations such as wall-to-roof junctions, window connections, panel-to-panel joints and wall-to-base support junctions. A typical wall section and details of the parapet and grade level panel joints are presented in Figures A1 through A3 in the Appendix.

Smoke Testing

As wall panel installation proceeded, air barrier continuity was checked using smoke testing. The technique involves creating a pressure difference across the air barrier system and then using a portable smoke generator to identify breaks. Since the building was not completely closed in and the mechanical system was not operational, the pressure difference had to be created by means of a variable speed portable fan. Figure

4. Small areas of the exterior walls were enclosed with gypsum board and polyethylene sheets and then pressurized to approximately 25 Pa with the fan. Smoke was sprayed at critical joints and details on the interior while observations were made on the exterior. Walkietalkies facilitated communication.

Figure 4 Air pressure differences were achieved in sections of the partially completed building using a blower door fan (right). A smoke generator (above) was then used to identify typical air leakage paths.

An effort was made to have representatives of all interested parties present during the testing. This included representatives of the owner, project manager, subcontractors, design consultant and building envelope consultant.

Although air tightness requirements are often speciﬁed in terms maximum air ﬂow rates per unit area of wall at a given pressure difference, it is seldom practical to conduct quantitative tests in the ﬁeld to verify these ﬁgures. It is therefore important to understand that on this project the smoke testing was strictly qualitative in nature. Once a discontinuity was identiﬁed using smoke, the leakage path was conﬁrmed by visual inspection. No ﬂow rate measurements were made.

As a result of the smoke testing the following major air leakage paths were identiﬁed: a) Roof to panel junction; b) The base of the panels near grade level; c) The roof to wall curb on the second ﬂoor; d) some vertical panel to panel joints.

Following the smoke testing the general contractor and the sub-contractors began a program of air barrier remediation. This mainly consisted of using sprayed urethane foam and caulking to seal various joints throughout the building.

Infrared Thermography – General Concepts

Although smoke testing readily demonstrates typical air barrier discontinuities, it is difﬁcult to verify the air tightness of the entire building envelope of a large facility using this technique. Positive building pressure cannot be achieved until the building’s mechanical system is operational. By that time, the inside surface of the building envelope is usually covered by various interior ﬁnishes. Furthermore, the sheer surface area involved makes complete smoke testing impracticable.

Infrared thermography has been used to identify building envelope deﬁciencies for over thirty-ﬁve years. Thermography relies on the physics principle that objects emit a frequency of electromagnetic radiation in proportion to their absolute temperature. In the case of objects which are at terrestrial temperatures, this radiation turns out to be beyond the visible spectrum, and is called infrared i.e. “below” red in the visible spectrum. Thermographic cameras use sensors to measure the infrared light and create a visible image based on an electronic mapping of the infrared light to the visible spectrum.

Modern thermographic cameras can display infrared images in a variety of chromatic schemes. Besides producing a visible image, the absolute temperature at any point in the image can be determined. A so-called ‘spot temperature’ can be identiﬁed by the camera for the reference cross hairs in Figure 5.

Figure 5

The thermographic image (thermogram) on the right shows air leakage at the base of the SIPS panels at grade level. Note the snow on the ﬂashing is melted where the air leakage is most intense.

One important feature of thermography is that large areas of the building can be surveyed in a relatively short period of time. Usually one session under optimal conditions (the absence of solar spectral radiation and the absence of excessive exterior winds) is sufﬁcient to complete the entire survey and identify spots where the usual temperature of the surface is substantially varied. These variations are typically referred to as thermal anomalies, and are caused by conductive or convective heat migration through the building envelope. A second important feature is that the survey can be recorded both with still images and on videotape or DVD.

Since the infrared images are read from the surface temperatures of the object, deﬁciencies other than air leakage are identiﬁed. These thermal anomalies include missing or improperly installed insulation and low thermal resistance pathways through the envelope, usually called thermal bridges. While infrared thermography can locate and measure temperature anomalies, it is not capable of determining the cause of a particular anomaly in a wall or roof assembly. The cause must usually be conﬁrmed by physical veriﬁcation, involving review of the construction detail, smoke ﬂow testing to locate air leakage paths through the assembly, and opening up the assembly to precisely locate the failed air-barrier component.

Initial Thermographic Survey

The ﬁrst thermographic survey was carried out in November 2002. Prior to carrying out the survey a proposed set of procedures and conditions was circulated for comment among the stakeholders. A meeting was then held at the site with representatives of the owner, general contractor, building envelope consultant and thermographer to conﬁrm the survey procedure. Among the more important points that were arrived at were:

• The thermographic survey would be carefully documented in terms of environmental conditions, viewing angles and anomaly locations. This would establish a benchmark, or reference, for future year thermographic surveys, a requirement of the owner.

The survey would be recorded on documentation sheets with thermographic images (‘thermograms’) and corresponding daylight images. Figures 5 through 9. A videotape of the entire thermographic survey would also be made.

Spot temperature measurements would be taken of apparent thermal anomalies and these temperatures would be compared to the ambient viewframe temperature. Based on previous experience, it was suggested that a temperature difference on the building surface of greater than 10oC was indicative of a very signiﬁcant anomaly.

It was agreed that the deﬁnition of what constituted a deﬁciency would likely require some physical investigation after the survey was complete.

The outside air temperature at the time of the 2002 survey was approximately –17 oC, the inside temperature was 21 oC and wind was calm. These conditions are considered suitable for producing good thermal images. The building was pressurized to a target pressure of approximately 35 Pa using the mechanical system. Positive pressurization was essential for causing exﬁltration of warm interior air through the envelope during the whole period of thermographic surveying.

Over ninety-ﬁve documentation sheets were prepared. The documentation sheets and the videotape of the survey were reviewed and physical investigation of the as-built construction was carried out. A typical documentation sheet is presented in Figure A4 of the Appendix.

It was concluded that the following types of thermal anomalies due to air leakage were signiﬁcant deﬁciencies:

Some locations at the panel base at grade, Figure 5.

Some locations at the roof to wall junction, Figure 6.

Some locations at the roof to wall curb on the second ﬂoor level, Figure 7.

Some locations at panel joints, Figure 8.

Miscellaneous penetrations, Figure 9.

The sealant bead between insulating glass units and the window frame.

Figure 6

Signiﬁcant air leaks are apparent at two locations at the parapet. A minor leak can be seen in a panel seam on the right.

Figure 7

Air leakage also occurred at the base of the SIPS panels on the second ﬂoor level.

Figure 8

Air leakage was identiﬁed at the joints between some panels however the extent of this problem was not as severe as the leakage at the parapets or base of the

Figure 9

Air leaks were also identiﬁed at various penetrations through the walls (rectangles). The bright spot above the two rectangles is heat radiating from a surveillance camera housing and is not considered a deﬁciency.

In order to conﬁrm that thermal anomalies were caused by air leakage, the air pressure in the building was reversed using the mechanical system, so that air was inﬁltrating into the building rather than exﬁltrating. The reversal of air pressure across the air barrier conﬁrmed that the thermal anomalies were due to air leakage and not from conductive thermal bridging. Figure 10.

In order to conﬁrm that the large anomaly on the left was due to air leakage, the pressure in the building was reversed from positive to negative, and the anomaly was surveyed again. Note that the anomaly completely disappeared. The same result proved true for other major anomalies when the air-pressure was reversed.

After conclusion of the survey a comprehensive report was compiled and distributed to various stakeholders.

Remedial Repairs

After review of the ﬁrst thermographic survey report the general contractor and the sub-trades began a program of remediation. Remediation eventually proceeded on the basis that anomalies showing temperature differences greater than 10 oC with respect to ambient would be repaired.

An action plan was prepared identifying over one hundred and ﬁfty repair locations referenced to the thermographic report thermograms. (This did not include the windows, all of which required remedial sealant repairs). Copies of the report were distributed to the sub-trades to clearly identify the locations for remedial repairs. Presentation of unambiguous visual information (thermograms) clearly showing the air leakage locations was instrumental in getting the sub-trades to buy into the remedial repair process, and allowed a precise remediation work plan to be developed.

Where accessible, repairs to the anomalies were undertaken with sprayed urethane foam insulation and with silicone sealants. Unfortunately, the large anomalies associated with the roof – wall junction could not be accessed from the interior. This required cutting open the roof at over thirty-ﬁve locations to expose the air leakage paths. Repairs were undertaken with self-adhering modiﬁed bitumen membrane and caulking, and then the roof membrane was made good. Figure 11.

Figure 11 In order to seal air leakage paths at the parapet the roof water prooﬁng membrane was opened up at leakage locations and self-adhering membrane was used to seal defects.

After completion of each repair, the building mechanical system was used to create a positive air pressure and a smoke generator was used to conﬁrm that the air leak had been sealed. (Some locations, such as the roof to wall curb on the second ﬂoor, could not be reached for smoke testing due to interior ﬁnishes, and were not repaired. Once repairs have been made at those locations, a smoke pencil or smoke puffer on the exterior will be useful to test for airﬂow.)

Follow-up Thermographic Survey

A second thermographic survey was conducted in December of 2003 in order to verify whether remedial repairs done in the summer had been effective. Exterior air temperatures averaged about minus 27o C, which was lower than the previous survey. Other environmental factors such as interior air temperature, wind speed and building air pressurization were similar to the ﬁrst survey conditions.

The procedure for the follow-up survey was similar to the ﬁrst survey. Wherever possible the same vantage points were utilized. The results of the follow-up survey were as follows:

The large anomalies repaired at the roof to wall junction were eliminated, Figure 12.

The majority of anomalies along the panel bases were either eliminated or signiﬁcantly reduced in magnitude, Figure 13.

The anomalies located at the roof to wall curb on the second ﬂoor level were reduced, however some signiﬁcant anomalies remained.

There were no signiﬁcant anomalies detected with the windows.

The anomalies at joints between panels were largely eliminated.

Figure 12

The Figure on the left was taken before remedial air sealing repairs were undertaken. The Figure on the right was taken during the follow-up survey and shows that the remedial repairs were successful.

Figure 13

Large volume air leakage at the base of the panels was also successfully eliminated in most locations.

The remedial repair program was determined to be largely successful. One remaining concern was the anomalies located at the roof to wall curb on the second ﬂoor level. It is considered that repairs at these locations were not effective. No smoke testing was undertaken as they were completed. Thus the sub-contractor responsible was not aware that many air leakage paths had been missed. It was therefore recommended that this detail be re-worked and veriﬁed with smoke testing at a future time.

Although the remedial repairs signiﬁcantly reduced the number and magnitude of anomalies due to air leakage, the follow-up thermographic survey showed that the building was still not perfectly airtight. It is suggested that once the anomalies located at the roof to wall curb on the second ﬂoor level are repaired, the building will have achieved a workable level of air tightness, and that further efforts of remedial air sealing may not be cost effective. It was recommended to the owner that the exterior walls of building be monitored periodically for signs of air leakage such condensation, icicles or frost formation. It was also recommended that building temperature, relative humidity and pressure be recorded by the building mechanical control system so that environmental loads would be known should air leakage problems arise in the future. Finally, it was recommended that another thermographic survey be conducted within ﬁve years to determine if the airtightness of the building envelope changes over time with settlement and other structural movement.

Summary and Conclusions

Air-tightness veriﬁcation of the building envelope was considered critical for the satisfactory performance of the Inuvik regional health care facility, situated in the harsh environment of the Canadian arctic, particularly on a fast-tracked project with an innovative panelized wall system. In order to achieve a known level of air tightness, testing and veriﬁcation of the air barrier was undertaken during and after construction. Smoke testing early in the project was valuable in identifying and demonstrating systematic air-barrier deﬁciencies to the project team in speciﬁc test locations. Infrared thermography proved to be an essential tool for identifying, locating and recording the overall condition of the air barrier. The use of temperature differences derived from the thermograms for guidance as to what constituted a signiﬁcant thermal anomaly streamlined the repair process. A follow-up thermographic survey conducted with the same protocols as the initial survey veriﬁed that the largest signiﬁcant air-leakage locations had been sealed.

It is the authors’ view that the process described in this paper was instrumental in achieving a veriﬁable level of air-tightness of the building envelope. Smoke testing, and in particular thermographic surveying, was essential for demonstrating envelope air-barrier deﬁciencies and their locations to the project manager and sub-trades. The use of these techniques during the construction and commissioning process encouraged a cooperative team approach, which led to the successful remediation of the largest air barrier deﬁciencies.

Acknowledgements

The authors wish to acknowledge the Government of Northwest Territories, Stuart Olsen Project Management Ltd. and Magna IV Engineering Ltd. for their assistance in preparing this paper.

Building air barrier testing continued on page 10

APPENDIX

Figure A2 Typical parapet detail from architectural design drawings.