diagnostics
Duct Leakage: Testing and What It Costs
A duct leakage test measures the air leaking out of and into the duct system. Here is how to run the test and what it costs.
The short version
A duct leakage test measures the air leaking out of and into the duct system, supply side, return side, and air handler cabinet. You seal the registers, connect a calibrated fan, pressurize the ducts to 25 pascals, and read how many cubic feet per minute the fan pushes to hold that pressure. That number is the leakage rate, expressed as CFM25.
ENERGY STAR reports that in a typical house, about 20 to 30 percent of the air moving through the duct system is lost to leaks, holes, and poorly connected ducts. The customer pays for that air every month in higher utility bills and rooms that will not hold temperature, and a technician who does not test for leakage chases charge or equipment faults instead, adding diagnostic time to the bill without reaching the root cause.
What the test measures
CFM25 is the airflow, in cubic feet per minute, needed to create a 25 pascal pressure difference in a sealed duct system. You seal every register and return grille, connect a calibrated fan to one opening, and bring the ducts to 25 Pa. The fan’s airflow at that pressure is the leakage rate. Twenty-five pascals is close to the operating pressure of a typical residential duct system, so CFM25 gives a reasonable estimate of how much air leaks during normal operation.
The number alone does not show you where the air goes. The building envelope is the boundary that separates conditioned space from outside. A duct that leaks into a conditioned basement wastes less energy than one that leaks into an unconditioned attic, because the conditioned air stays inside the envelope either way. Energy Vanguard distinguishes benign leakage, which stays inside conditioned space, from malignant leakage, which connects the ducts to unconditioned spaces like attics, crawlspaces, and garages. A house can test high on total leakage and still perform well if most of that leakage stays inside the envelope.
That is why there are two versions of the quantitative test. Total leakage covers every hole in the duct system. Leakage to outside covers only the holes that connect the ducts to unconditioned spaces, whether supply leaks push air out or return leaks pull it in. Both use the same 25 Pa test pressure. They differ in how the house is handled during the test.
Three common ways to test
Total leakage (duct blaster only). Seal every supply register and return grille with tape, cardboard, or reusable covers. Connect a calibrated fan to one opening, typically a return grille or the air handler cabinet. The Minneapolis Duct Blaster, made by the Energy Conservatory, and the Retrotec DucTester are the two instruments in common use for residential work. Pressurize the system to 25 Pa and read the fan’s airflow. That reading is total CFM25. Every hole in the system, supply side and return side, contributes to it. Code thresholds are usually written as total leakage limits.
Leakage to outside (duct blaster plus blower door). This version zeros out the benign leaks so only the malignant ones show up in the reading. Energy Vanguard describes the procedure: run the blower door to pressurize the house to 25 Pa at the same time the duct blaster pressurizes the ducts to 25 Pa. With equal pressure on both sides of any duct that passes through conditioned space, no air crosses those leaks. Everything the duct blaster pushes into the system escapes through leaks to unconditioned space, so the fan’s airflow reading is CFM25 to outside. This test requires two instruments running simultaneously and takes more setup, but it separates the leakage that costs energy from the leakage that does not.
Pressure pan (blower door, pan, and manometer). This test is qualitative rather than quantitative, and it starts with a safety step. A blower door depressurizes the house, and that suction can reverse the flow in the flue of a furnace, water heater, or fireplace that vents with indoor air, pulling combustion gases, including carbon monoxide, into the house instead of up the flue. That reversal is called backdrafting. Set combustion appliances the way your blower door protocol requires before you depressurize the house, so nothing can backdraft during the test. The Department of Energy’s Building America Solution Center calls for a worst-case depressurization test where appliances vent with indoor air, and for removing or covering fireplace ashes, which the same suction would otherwise pull into the room.
Depressurize the house to 50 Pa with a blower door, then place a pressure pan over each register one at a time. A pressure pan is a gasketed register cover with a pressure tap connected to a gauge. The gauge reads the pressure difference between the duct behind that register and the room. A reading near zero means that run has no significant connection to the outside. Higher readings mean a stronger connection to unconditioned space, and the cutoff varies by program: one weatherization guide flags a median reading of 4 Pa or more for sealing. The pressure pan does not give you a leakage rate in CFM. It shows which runs are most connected to unconditioned space, so you know where to focus sealing work.
| Test | Equipment | What it measures | Result type |
|---|---|---|---|
| Total leakage | Duct blaster, sealing materials | All leakage in the duct system | CFM25 (quantitative) |
| Leakage to outside | Duct blaster and blower door | Leakage to unconditioned spaces only | CFM25 to outside (quantitative) |
| Pressure pan | Blower door, pressure pan, manometer | Which runs connect to outside | Pa per register (qualitative) |
Running a total leakage test
The total leakage test requires only the duct blaster. On an existing system, it shows how much air the ducts lose overall, which is the starting point for deciding what to seal and whether the sealing made a difference.
Before you start. Turn off the HVAC system so the air handler fan does not run during the test. Open one door or window so the fan does not change house pressure. Remove the filters from the ducts and the air handler cabinet, and set the panels and dampers the way your test standard requires.
Seal the registers. Seal every supply register and return grille at the face and perimeter. If you connect at a return grille, that grille is the only one you leave open, and you seal around the connection. Use tape and cardboard, or reusable covers made for testing. A missed register vents directly to the room, so the fan reads higher than the real leakage rate and the result is not valid.
Connect the fan. ANSI/RESNET/ICC 380 permits connection at the blower compartment on any system. The largest return grille is allowed only in limited cases: three or fewer return grilles, total leakage under 50 CFM25, or a blower access panel you cannot legally or physically reach. Attach the duct blaster at the connection point you are using, and seal it so all the air the fan moves enters the duct system.
Pressurize and read. Increase fan speed slowly until the duct pressure gauge reads 25 Pa. Read the fan’s airflow at that pressure. That airflow is total CFM25. The flow ring is the insert that sets the fan’s measurement range. If the system is large or particularly tight, adjust it to get an accurate reading.
Normalize the result. Divide the raw CFM25 result by the conditioned floor area in hundreds of square feet to compare against code thresholds. A 2,000 square foot house that tests at 80 CFM25 gives 80 divided by 20, or 4.0 CFM25 per 100 square feet. You can also express leakage as a percentage of air handler flow by dividing CFM25 by the system’s rated airflow. Energy Vanguard notes that these two normalizations produce very different numbers for the same house, so always confirm which one a code or program threshold is using before you compare.
What the code requires
The 2021 International Energy Conservation Code (IECC) sets the model duct leakage thresholds. The limits are expressed as CFM25 per 100 square feet of conditioned floor area, and testing is done to ANSI/RESNET/ICC 380 or ASTM E1554 at 25 Pa.
| Condition | Limit (CFM25 per 100 sq ft) |
|---|---|
| Post-construction, air handler installed | 4.0 |
| Rough-in, air handler not installed | 3.0 |
| All ducts and air handler inside the thermal envelope | 8.0 |
The 2021 edition removed an exemption from earlier versions. Before 2021, duct systems located entirely inside the building thermal envelope did not require a leakage test at all. That exemption is gone. Heating and cooling ducts now require testing wherever they sit, and standalone ventilation ducts are excepted. The higher 8.0 threshold for ducts inside the envelope reflects the lower energy penalty when leakage stays in conditioned space.
The IECC applies to new construction, and to additions, alterations, repairs and changes of occupancy in existing buildings. Existing homes are not required to test unless a renovation triggers the energy code, though energy audit programs and utility rebate programs often include duct testing. State and local adoption varies, and not every jurisdiction enforces the 2021 edition. Some have amended the thresholds. Check what your jurisdiction has adopted before quoting a number to a customer.
What leakage costs the system
ENERGY STAR estimates that 20 to 30 percent of the air moving through a typical duct system is lost to leaks. Field research from the Lawrence Berkeley National Laboratory (LBNL) has measured similar leakage. A 2002 study by Walker and Sherman found combined supply-and-return leakage averaging 22 percent of air handler flow across the California homes they tested.
The loss is not symmetric. Supply-side leaks send conditioned air into unconditioned spaces, so the house receives less of what the system produces. Return-side leaks pull unconditioned air into the system, so the air handler spends capacity conditioning air that was never in the house. When both occur at once, the penalty compounds rather than adds. Energy Vanguard reports, citing Walker’s 2001 LBNL research, that a system with 10 percent supply leakage and 10 percent return leakage ran at roughly 70 percent duct system efficiency when the ducts sat in a hot attic. That is a 30 percent loss, not the 20 percent the two numbers suggest. Reducing both sides to 5 percent improved efficiency by about 10 percentage points but still left the system well short of 100 percent.
Duct leakage also increases the house’s infiltration rate. Research from the Florida Solar Energy Center (FSEC) found that when the air handler was running, the infiltration rate was four times higher than when the air handler was off. Supply-side leaks depressurize the house, so replacement air is drawn in through envelope leaks and from the attic or crawlspace. Return-side leaks tend to pressurize the house. Both effects add to the conditioning load.
For the homeowner, the result is a higher utility bill every month the system runs, rooms that stay warmer or cooler than the thermostat calls for, and a system that cycles longer to hold its setpoint. The standard fix is sealing the joints and connections with mastic sealant or metal-backed (foil) tape, which ENERGY STAR recommends over cloth-backed duct tape. LBNL accelerated-aging testing found that cloth-backed tape failed consistently under thermal cycling, which is why the name is misleading: cloth-backed duct tape is not a durable duct sealant.
Common mistakes
Do
- Seal every supply register and return grille at the face and perimeter, since a missed one vents to the room and overstates the leakage rate
- Confirm the air handler is off, because the blower running during the test corrupts the pressure reading
- Remove the filters from the ducts and the cabinet, because the standard test runs with the filters out
- Normalize the CFM25 result to conditioned floor area before comparing it to a code threshold
- Use the pressure pan to locate problem runs before sealing, because total CFM25 shows the size of the problem but not where it sits
- Check what your jurisdiction has adopted before quoting the 2021 IECC threshold to a customer
Don’t
- Run the test with the air handler fan on, which pressurizes the system independently and produces a meaningless reading
- Assume a low total leakage number means low energy loss, since total leakage includes benign leaks into conditioned space that cost little
- Use cloth-backed duct tape for permanent sealing, because it degrades under the temperature swings that ducts in attics and crawlspaces see
- Apply mastic over dirty or oily surfaces, which prevents adhesion and allows the seal to separate
- Skip the test because the ducts are inside the building envelope, since the 2021 IECC removed that exemption
- Quote a code threshold without checking your jurisdiction’s adopted version of the energy code
Frequently asked questions
What is CFM25?
CFM25 is the airflow in cubic feet per minute needed to hold a 25 pascal pressure difference in a sealed duct system. It is the standard measure of duct airtightness. Twenty-five pascals is close to the operating pressure of a typical residential duct system, so CFM25 approximates how much air leaks during normal operation. The higher the number, the more air the system loses.
How much does a duct leakage test cost?
A residential duct leakage test typically runs between $200 and $450 for a single system, depending on the size of the house and the tester’s rate in your area. The test takes about an hour. Some utility rebate programs and energy audit programs cover part or all of the cost.
How much air does a typical duct system lose?
ENERGY STAR reports that about 20 to 30 percent of the air moving through a typical home’s duct system is lost to leaks, holes, and poorly connected ducts. LBNL field research measured combined supply-and-return leakage averaging 22 percent of air handler flow in California homes. Systems with ducts in unconditioned attics and crawlspaces tend to lose more than those with ducts inside conditioned space.
What is the difference between total leakage and leakage to outside?
Total leakage counts every hole in the duct system, whether it leaks into conditioned space or unconditioned space. Leakage to outside counts only the leaks connecting the ducts to unconditioned spaces like attics, crawlspaces, and garages, in either direction. Leakage to outside is the number that directly costs energy. A system with high total leakage but low leakage to outside may perform better than the total number suggests.
Does the pressure pan test replace the duct blaster test?
The pressure pan test does not replace the duct blaster test. A pressure pan shows which duct runs connect to unconditioned space, but it does not measure how much air is leaking. Code compliance requires a quantitative CFM25 result, which takes a calibrated duct leakage tester. The pressure pan is a diagnostic tool for locating the worst runs before and after sealing.
What should I use to seal duct leaks?
ENERGY STAR recommends mastic sealant or metal-backed (foil) tape at joints and connections. Mastic fills irregular gaps up to about a quarter inch and hardens into a permanent seal. Metal-backed tape works well on flat seams and smaller connections. Cloth-backed duct tape is not a durable seal for duct joints. For seams wider than an eighth of an inch, the Department of Energy’s Building America Solution Center calls for fiberglass mesh tape under the mastic.
Glossary
- Benign leakage: duct leakage into conditioned space, which does not waste energy because the air stays inside the building envelope.
- Blower door: a calibrated fan mounted in an exterior doorway, used to pressurize or depressurize a house for envelope and duct testing.
- Building envelope: the boundary that separates conditioned space from unconditioned space and outside; the energy code calls it the building thermal envelope.
- CFM25: cubic feet per minute at 25 pascals, the standard measure of duct leakage rate.
- Duct blaster: a calibrated fan designed to pressurize a duct system for leakage testing. The Minneapolis Duct Blaster from the Energy Conservatory and the Retrotec DucTester are the residential models in common use.
- IECC: International Energy Conservation Code, the model energy code adopted in modified form by most U.S. states.
- LBNL: Lawrence Berkeley National Laboratory, a federal research lab whose duct leakage field studies are cited here.
- Malignant leakage: duct leakage between the ducts and unconditioned spaces like attics, crawlspaces, and garages, which wastes energy.
- Mastic: a wet sealant applied to duct joints that hardens into a permanent seal.
- Pa: pascal, the metric unit of pressure used in duct and envelope testing. 25 Pa equals 0.1 inches of water column.
- Pressure pan: a gasketed register cover with a pressure tap, used with a blower door to identify which duct runs leak to the outside.
- RESNET: Residential Energy Services Network, the organization that publishes duct testing protocols and accredits the providers that certify energy raters.
Drafted with AI assistance and reviewed by the author.