Duct leakage test reports, assembled from field sheets for your project manager to review.

The rig goes up at the end of a section. The crew caps the openings, brings the duct to test pressure, reads the orifice and writes the numbers down, sometimes over several days and after a repair or two. Your project manager then sorts those sheets by system, checks that every section the job called for has a record, moves the values onto the owner's form and chases whichever signature is missing.

For the test, AI works through copies of those sheets and builds the draft package: system or section tested, test pressure, surface area, measured and allowed leakage, who performed it and who witnessed it, then a list of every section with no record or a failed entry. Nothing reaches an engineer or an owner. Your reviewer sets each draft beside the report you really sent.

The report answers to whoever asked for the test. SMACNA's presentation to a Connecticut ASHRAE session in December 2024 said its publications don't require leakage tests: the professionals of record decide when testing is needed, and the local authority may require it. Washington State's energy code reaches ducts designed at 3 inches of water gauge or more, worded earlier as in excess of 3 inches, tests them at a leakage class of 4.0 or lower and wants documentation that representative sections were tested. As that presentation reports it, the 2024 Uniform Mechanical Code covers all ductwork, with a sample that widens after a failure, up to the whole system. The University of Maryland, Baltimore tests every duct system to Class 2 and says that goes beyond SMACNA's own requirement. So the draft quotes the requirement each report answers to.

Whole systems or sections, and at what pressure. The University of Houston's master section tests complete systems and allows partial sections only with the engineer's written approval. Maryland tests whole or in segments as the work progresses, and where it uses a flow-based allowance it prorates each segment by surface area. The University of Michigan's master section wants a testing plan before fabrication showing temporary caps, the surface area of each test section, test pressure, leakage class and allowable leakage. Pressure has its own trap: the SMACNA presentation holds test pressure to the rating of the weakest piece in the section, so a 3-inch trunk with a 2-inch branch tests at 2 inches or less. The draft carries each sheet's section limits and names any section whose pressure or class doesn't match the plan.

The form asks for more than the crew wrote down. Houston's minimum list runs from the system tested, leakage classification and square footage through permissible leakage, test pressure, orifice size, measured pressure differential, measured leakage and duct pressure, to who performed it, the date and time, the temperature and weather, and an optional engineer or owner sign-off. Maryland's form adds total system CFM and construction pressure class at the top and leakage factor and test section CFM by duct, and its rig needs a certified performance curve attached and calibration data on hand. SMACNA's presentation says to leave no blanks, writing N/A where a field doesn't apply. The draft moves each value across, shows the allowed-leakage arithmetic beside the measured figure, and lists anything missing. The result line stays open for your tester.

A failed test stays in the file. Maryland wants a summary report for every test, pass or fail, signed by each party who witnessed it, with seven business days' notice, all tests witnessed by university personnel and selected ones by the balancing contractor. After a failure the joints are surveyed for audible leaks and repaired, and the sealant sets 24 to 36 hours before the retest is rescheduled. Michigan asks for two days' notice and, before anything is measured, a visual, audible and hand check 2 inches from the duct; its welded duct must show zero leakage. The draft keeps each failed entry and its retest in date order and lists any witness line left blank.

Duct leakage is not system leakage. The SMACNA presentation counts only straight duct and fittings as duct. Terminal units, volume and fire dampers, access doors and equipment sit outside it, and SMACNA's System Air Leakage Test Standard, published in 2020, brings them in while leaving out grease duct and smoke control. Shape and sign matter as well: Massillon West Elementary in Ohio tests rectangular duct to Class 6 and round to Class 3, Houston's table writes exhaust tests at negative pressures, and ASHRAE 90.1-2019, as that presentation reports it, accepts a positive-pressure test on negative duct. What your reviewer will check first: each required section has a record, values landed on the right section, failed entries weren't dropped, and signature lines are left for the people who sign them. Treat every rule quoted here as an example from a published document; none of it is legal or engineering advice.

New test sheets get the same assembly for about 30 days once your reviewer agrees.

Your tester signs every report, and your project manager approves each package before it goes to anyone.

We keep comparing on live jobs. Whatever comes next is its own decision, priced on its own.