Published in TAB Journal Summer 2017 and written by Denny Whitzel, TBE, CxA Pacific Coast Air Balancing. View the original article here. 

As required by Washington State Code, all commercial multi-unit residential structures are required to have air barrier testing performed to ensure they meet the state requirements for infiltration levels. During the construction process special procedures, materials and methods are utilized to ensure the airtight integrity of the inside and outside of the building are maintained. Yet even when the previous three are followed diligently, achieving acceptable results can still be very challenging. This study hopes to bring to light a number of conditions, some “inherent” and some “created”, that when recognized can make the difference in testing multi-story wood-built structures to meet state requirements.

To better understand the study, the type of building tested is a multi-story/unit wood-built apartment structure; these units vary in size from 3 to 4 stories, and 12 to 30 apartments per unit. Individual apartments are 1-3 bedrooms. The square footage ranges from 750 to 1350 square feet, with a 9-foot ceiling height. The buildings are wood-framed with standard siding over a barrier wrap on the outside. The interior finish is sheetrock-taped, and finished ceiling to floor. The lower floor units have an overhead area consisting of a 24” sub floor for the next level, which is used to route utilities. The top floor units have an insulated open vented attic above the finished ceiling, and connecting walls. The individual apartments were tested using the blower door air infiltration testing system. The standard for the test is no more than 5 air changes per hour under an induced pressure of 50 Pascals(Pa). Each type of apartment is field measured to determine total volume in cubic feet, which is then used to calculate the maximum allowable leakage amount. A preliminary short manual test is performed to test infiltration level at 50 Pa. When the results are good the full automated test is performed; if not, additional sealing work is done to ensure compliance before the final test.

Conditions that cause leakage due to penetrating the interior or exterior envelope are what is considered “created” conditions. These include electrical, plumbing and venting just to name a few. Failure to seal these types of penetrations will have a negative impact on test results. Failure to adhere to protocol throughout the construction process for the interior /exterior envelope can create leaks that are not easily located, or able to be fixed. Regardless, with attention to detail most created conditions can be avoided.

Leakage conditions that exist due to building design and/or routing of utilities are what is considered to be “inherent” type. These types of conditions cannot always be completely rectified; however, most can be minimized to reduce the impact. One condition encountered is the apartment size versus the number of utility penetrations. Mechanical exhaust systems for bathrooms, laundry, range hoods etc. are intended air flow paths. These systems are equipped with backdraft dampers which are intended to stop air infiltration when the fans are off. These dampers are lightweight and are forced open by fan discharge pressure when they are operating. These dampers when operating properly will be forced open under a 50 Pa positive induced pressure. This will result in a significant increase in leakage rates during the pressurization mode. This is also why testing is performed in both directions with the pressurization test followed by a depressurization cycle. The depressurization test is used to determine compliance, and as an added bonus it reveals whether or not the backdraft dampers are operational. In some cases the actual leakage rate at 50 Pa during the depressurization cycle dropped as much as 300 CFM. This is a significant amount considering the maximum allowable leakage rate varies from 545 CFM to just over 1000 CFM at 50 Pa depending on the apartment size. The number of these systems along with the total number of penetrations to the interior envelope i.e. unit heaters, service outlets/panels, plumbing, dryer vents, lighting fixtures, etc. is not proportional to the size/volume of the apartment. Essentially a small 1 bedroom apartment has 3 exhaust systems whereas a large 3 bedroom unit has 4. Although not quite as high proportionally the same applies to the numbers of other penetrations mentioned. What this really means is that on smaller apartments there are as many penetrations to seal, with a much lower allowable leakage requirement, which will make obtaining acceptable results more challenging.

Another inherent condition encountered with this type of construction is the vented attic. Unlike the lower floors which have a sealed sub floor space above, the top floor has a large vented attic overhead. The vented attic is a low resistance path where air is encouraged to vent. Basically this creates a ducted path via wall penetration, and a direct path via ceiling penetration to atmosphere for leakage to occur. The lower floor overhead areas are sealed at the outer envelope. By sealing the outer envelope the holes in the interior envelope have nowhere to leak to. This makes the top floor apartments much more prone to leakage. Initial result  from preliminary testing showed higher than allowable amounts of leakage for the top floor apartments. Just as the blower door system was used to identify this condition it was then utilized to rectify the problem. By putting the system in decompression mode to the rated 50 Pa pressure most leaks become evident, and can be felt by the rush of air inward to the apartment at their location. When inspected in this mode literally every penetration to the wall or ceiling had air movement that could be felt by the human hand. Again the capability of the blower door system would be put to use.

Being post-construction, sealing the attic side of the inner envelope was not an option. This left sealing the inner envelope from the inside as the remedy of choice. To prove that sealing all the penetrations would bring the apartments within acceptable limits, the blower door system was again set up in decompression mode at 50 Pa. With a constant negative pressure maintained, all penetrations to the walls and ceilings were inspected and temporarily sealed one by one. A baseline leakage amount was established, and a new leakage reading was taken after each point was sealed. By doing so the difference in total leakage equaled the amount of each individual leak. Depending on the type and size of leak sealed, the individual leak amounts varied from inconsequential 2-3 CFM on light switches to measureable 20-30 CFM on larger gaps around panels or vents. By eventually sealing all accessible penetrations, the leakage levels were brought below the maximum allowable levels. The contractor subsequently sealed all the wall and ceiling penetrations for the top floor. The apartment was re-tested and passed the minimum requirement for infiltration. The extra effort sealing of all top floor penetrations was adopted as a standard to ensure compliance for the upper apartments. Due to the nature of the sealed exterior envelope on the lower floors, the standard sealing of larger wall penetrations achieved acceptable results without the extra effort of sealing every plug, switch or light fixture.

During the process of identifying the different dynamics between top versus lower floors while measuring individual leakage per penetration, another point came to light; next to the fan wall air was felt coming from the fan wall itself. Upon further investigation the air was primarily coming from the interior slots where the fan wall adjusts to fit the doors. Using the same method, leakage was recorded, and then the fan wall was sealed using tape. The system was returned to negative 50 Pa, and another reading was taken with the difference being the leakage at the fan wall. This test was performed multiple times and depending on apartment size, the fan wall leakage was 25 to 45 CFM. Although 45 CFM leakage is less than 5% of the total allowable leakage 1330 CFM, it is still enough that taping these seams was adopted during all testing to improve accuracy.

By understanding the dynamics with the particular construction covered in this study, one can take the lesson learned to identify areas that will need extra effort to be sealed. This will allow you to inform the contractor ahead of time, ensuring they get addressed at the proper point during construction. By doing this the general contractor can save hours instead of going back and sealing after the fact. Sometimes the toughest part is explaining to the general contractor the dynamics involved so they understand why so much more effort is required. Including the general contractor during the test process of proving the upper floor units could be sealed by temporarily taping was found to be very effective. Afterwards there was no argument or lack of understanding about the need to seal. Essentially feeling is believing when it comes to leaks, even if they don’t understand the entire concept. In closing the hope is by knowing the created and inherent conditions that require diligent sealing, difficulty achieving acceptable results can be avoided in these types of structures.