Duct Leakage Testing and Sealing
A leaky duct system quietly throws away a fifth of the air you paid a furnace or condenser to move. Here is how a duct blaster test actually works, what the numbers mean, and where the biggest leaks almost always hide.
Chasing delivered airflow?
Convert velocity and duct dimensions into real CFM to see what the leaks are costing you.
In This Guide
Why Duct Leakage Matters
Duct leakage is the single most common performance defect in forced-air systems, and it is almost always invisible until you test for it. A typical existing home loses roughly 20–30% of its conditioned air to leaks — air that escapes into attics, crawl spaces, wall cavities, and framing chases instead of reaching the rooms it was meant to condition.
The damage is not just wasted energy. Leaks distort the whole system:
- Supply leaks in an attic dump paid-for heating or cooling into unconditioned space and starve the far rooms.
- Return leaks pull hot attic air or dusty crawl-space air straight into the blower, hurting capacity and indoor air quality.
- Pressure imbalance from a leaky return can depressurize the house enough to backdraft a water heater or gas furnace — a genuine combustion-safety hazard.
- Comfort complaints and short cycling follow, because delivered airflow no longer matches the equipment's rated CFM.
If you have ever measured a low temperature split or a static pressure that made no sense, leakage is frequently the culprit. Testing turns a guess into a number.
Total Leakage vs. Leakage to Outside
Every duct blaster test measures one of two things, and the distinction matters for both code and diagnostics.
Total Leakage
All air escaping the sealed duct system at 25 Pa, whether it leaks into conditioned or unconditioned space. This is the number used for most code compliance tests. The house itself stays at rest; only the ducts are pressurized.
Leakage to Outside
Only the air escaping to unconditioned space. Measured by running a blower door and duct blaster together at the same pressure so leaks between duct and living space cancel out. This is the number that actually drives energy loss.
A duct system buried entirely inside the conditioned envelope can have high total leakage but near-zero leakage to outside — the escaped air is still inside the house. For an attic or crawl-space duct run, though, total leakage and leakage to outside are nearly the same thing, and every CFM is money leaving the building.
How the Duct Blaster Test Works
A duct blaster is a small calibrated fan with a digital manometer. You seal the duct system, pressurize it to a known reference, and read the airflow the fan needs to hold that pressure — more leakage means the fan has to move more air.
The 25-Pascal Standard
Duct testing is standardized at 25 Pa (0.1 inch w.c.) relative to the surrounding space. Reporting all leakage at the same reference pressure is what makes results comparable between houses and against code — a raw "it leaks a lot" means nothing without the pressure it was measured at.
- Seal every register and grille. Mask off all supplies and returns with tape or foam plugs so the only intentional opening is the fan.
- Connect the fan. Mount the duct blaster at the return grille or air-handler cabinet and route the reference tube into the duct to sense system pressure.
- Pressurize (or depressurize) to 25 Pa. Ramp the fan until the manometer reads 25 Pa between the duct and the room.
- Read the fan flow. The airflow through the fan at that moment is your total leakage, expressed as CFM25.
- Normalize and compare. Divide by conditioned floor area and scale to 100 sq ft, then check against the applicable limit.
Field tip: A rough-in test (done before drywall, with the air handler installed) catches leaks while they are still cheap to fix. A postconstruction test on a finished system is what most inspectors want to see. Whenever possible, test at rough-in so you are not chasing sealant behind finished walls.
Ventora — Your AI HVAC Companion
Duct blaster targets, static pressure references, and airflow math in your pocket. Free on iOS.
Code Limits and Pass/Fail Numbers
Under the IECC (Section R403.3.5 in the 2018 and 2021 editions), duct testing is required whenever any portion of the duct system is outside the conditioned space. Results are always normalized to CFM25 per 100 sq ft of conditioned floor area. The common thresholds:
| Test Type | Limit (CFM25 / 100 sq ft) | Notes |
|---|---|---|
| Postconstruction, total | ≤ 4.0 | Finished system, air handler installed |
| Rough-in, with air handler | ≤ 4.0 | Tested before drywall |
| Rough-in, no air handler | ≤ 3.0 | Handler not yet set |
| Ducts fully inside envelope | Exempt | No test required if all ducts + handler are inside |
Watch the edition. Older codes (IECC 2009) allowed 8–12 CFM25 per 100 sq ft, and some jurisdictions still reference legacy limits or amend the state amendments. Always confirm the exact adopted code and limit with the local AHJ before you certify a result.
Worked Example: CFM25 per 100 sq ft
Problem:
You test a 2,400 sq ft home with ducts in a vented attic. The duct blaster reads 132 CFM to hold the system at 25 Pa. Does it pass a 4.0 CFM25 per 100 sq ft limit?
Step 1: Record the measured leakage
Total leakage = 132 CFM25
Step 2: Divide by conditioned floor area
132 / 2,400 = 0.055 CFM25 per sq ft
Step 3: Scale to 100 sq ft
0.055 × 100 = 5.5 CFM25 per 100 sq ft
Step 4: Compare to the limit
5.5 > 4.0 → over the threshold
Result: FAILS
At 5.5 CFM25 per 100 sq ft, this system exceeds the 4.0 limit. To pass, total leakage has to drop below 96 CFM25 (4.0 × 2,400 ÷ 100). That is roughly a 27% reduction — very achievable with mastic on the plenum seams and boot-to-drywall connections.
Where the Biggest Leaks Hide
Leakage is rarely spread evenly. A handful of locations account for the majority of most systems' total leakage. Check these first:
- Air-handler and furnace cabinet seams. Factory cabinets and field-cut openings leak badly. This is often the single largest source.
- Supply and return plenum connections. Where sheet metal meets the equipment is a chronic gap.
- Boot-to-subfloor / boot-to-drywall joints. The register boot rarely seals to the finished opening; each one leaks a little and there are many.
- Building-cavity returns. Panned joist bays and wall-stud returns are almost never airtight and pull in unconditioned air.
- Flex-duct collar connections. Inner liner not fully seated over the collar, or a zip tie without mastic underneath.
- Longitudinal seams and takeoff collars on trunk lines.
- Disconnected or crushed sections — a flex run knocked off its collar in the attic can single-handedly blow a test.
Finding leaks fast
Pressurize the ducts and use a smoke pencil, or feel for airflow with a wetted hand along seams. On a rough-in, a light haze of theatrical fog in the ducts makes escaping air visible at every gap. Mark leaks with tape as you find them, then seal in one pass.
Sealing Ducts the Right Way
The rule that trips up new techs: despite the name, standard cloth-backed duct tape is not approved for sealing ducts. It dries out and fails within a few years. Use the right materials:
Use These
- Water-based duct mastic on seams and joints
- Mastic with fiberglass mesh tape on gaps over 1/8 in.
- UL 181 foil tape for flex-duct collars
- Mechanical fasteners (screws) before sealing metal
- Aeroseal-style aerosol sealing for buried or inaccessible runs
Avoid These
- Cloth-backed "duct tape" (not code-approved)
- Sealing over a joint with no mechanical support
- Skipping the return side — returns leak too
- Mastic applied too thin (aim for nickel thickness, ~1/16 in.)
Work from the equipment outward: cabinet and plenum first (biggest wins), then trunk seams, then every boot and takeoff. Fasten metal joints mechanically, then brush or glove mastic over the seam. On flex, seal the inner liner to the collar with mastic and a draw band before pulling the insulation and outer jacket over and banding that too.
Don't seal a system too tight without checking combustion safety. If you tighten a return that was feeding makeup air, re-verify that atmospheric appliances still draft properly and that the house isn't being over-depressurized. Tightening ducts is a system change, not just an energy upgrade.
After sealing, retest. The number tells you whether you found the leaks or just moved them around — and gives you the documentation the inspector and the homeowner both want.
Duct Diagnostics, Powered by AI
Ventora — the AI HVAC assistant in your pocket. Get instant airflow math, code limits, static pressure references, and AI troubleshooting on every call. Start with a 7-day free trial.