Building Enclosure Testing (BET): Field Air and Water Leakage Verification
What this domain covers
BET tests whether you can turn the envelope into a measured, defensible number. Questions cluster around the ASTM methods that govern field practice: E779 and E1827 for fan pressurization, E3158 for large-building leakage, E783 for installed fenestration air leakage, E1105 and AAMA 501.1/501.2 for water penetration, and E1186 for locating leaks by smoke, chamber, or infrared. Reporting, baseline correction, and normalization carry as many marks as the test itself.
Core rules and concepts
- Report envelope tightness at the 50 Pa reference pressure as CFM50 and ACH50, where ACH50 = CFM50 x 60 / conditioned volume in ft3.
- Normalize for air barrier comparison by gross envelope area, not floor area: CFM50 / ft2 of envelope.
- Take a zero-flow baseline before and after the test, correct induced pressures for it, and record wind speed and indoor/outdoor temperature.
- Build a multipoint regression over a wide range, commonly 10-60 Pa with at least five points spanning a 5x range, and confirm R2 of about 0.99 or better.
- E3158 large-building work uses multiple fans and building pressure control; average readings over several minutes so wind gusts wash out.
- AAMA 501.2 hose test: 30-35 psi nozzle pressure, 5 minutes minimum, nozzle held about 12 in. from the assembly, working from the bottom upward.
- E1105 spray rack duration and pressure differential follow the project specification, typically 15 minutes at a stated fraction of design pressure.
Worked example
A 10,000 ft2 single-story building with a 10 ft ceiling has a volume of 100,000 ft3. The envelope is the 10,000 ft2 roof plus 400 ft of perimeter wall x 10 ft = 4,000 ft2, for 14,000 ft2 gross. The fans deliver 2,000 CFM at 50 Pa. ACH50 = 2,000 x 60 / 100,000 = 1.2. Normalized leakage = 2,000 / 14,000 = 0.143 CFM50/ft2, so a 0.15 CFM50/ft2 specification is met.
Traps that cost marks
Using floor area instead of envelope area; skipping the zero-flow baseline; testing with HVAC running or intentional openings unsealed and undocumented; confusing ACH50 with natural air changes; accepting a regression from too few points; and running water tests at the wrong nozzle pressure or duration.