9.3 Duct Testing Protocols & Leakage Limits (R403.3.5–R403.3.6)

9.3 Duct Testing Protocols & Leakage Limits (R403.3.5–R403.3.6)

Quick Answer: Pressure-test ducts to ANSI/RESNET/ICC 380 or ASTM E1554. Measure total leakage at 0.1 inch w.g. (25 Pa). Tape or otherwise seal registers. Include the manufacturer's air handler enclosure when it is installed. File a signed written report with the code official. Limits per 100 square feet of CFA are 4.0 (rough-in with air handler), 3.0 (rough-in without air handler), 4.0 (postconstruction), and 8.0 (all ducts and air handlers entirely within the building thermal envelope). The 8.0 path is still a test. The 2018 idea that ducts in conditioned space are exempt from testing is false in 2021.

Independent OpenExamPrep material for ICC Exam 79 treats R403.3.5 as the protocol and R403.3.6 as the scoreboard. Confirm the live exam bulletin still lists 2021 IECC residential provisions. Do not grade a 2021 house with a 2018 highlighter still parked on an exemption that the first printing removed.

R403.3.5 — how the test is run

Ducts shall be pressure tested in accordance with ANSI/RESNET/ICC 380 or ASTM E1554 to determine air leakage by one of the following methods. You pick a method. You do not skip both.

Rough-in test

Total leakage shall be measured with a pressure differential of 0.1 inch w.g. (25 Pa) across the system, including the manufacturer's air handler enclosure if installed at the time of the test. Registers shall be taped or otherwise sealed during the test.

If the air handler is already set, its cabinet is in the tested volume. If the air handler is not yet set, you are testing ducts (and whatever fittings are installed) without that cabinet. That is why R403.3.6 then gives two rough-in limits. A rough-in with open boots, a missing filter-box door, or a flex run still coiled in the attic is not the test the section describes.

Postconstruction test

Total leakage shall be measured with a pressure differential of 0.1 inch w.g. (25 Pa) across the entire system, including the manufacturer's air handler enclosure. Registers shall be taped or otherwise sealed during the test.

Postconstruction always includes the cabinet. Finishes may be on. The measured number is the system the occupant will live with, not a partial trunk. Boots must still be sealed for the test. Testing with registers open so the house can help is not R403.3.5.

Report

A written report of the results of the test shall be signed by the party conducting the test and provided to the code official. An unsigned printout in the permit packet is not the report. A text message with a photo of a gauge is not the report. The code official receives a signed writing. Where the jurisdiction requires an approved third party, that is a local administration overlay on top of this sentence — the IECC still wants the signed report.

Ventilation-only exception

Exception: A duct air-leakage test shall not be required for ducts serving ventilation systems that are not integrated with ducts serving heating or cooling systems. An HRV with its own dedicated ducts can use this exception. An HRV that dumps into the return of the heating and cooling duct system is integrated. Kitchen and bath exhaust that never joins the supply/return network is not this exception's homework — those are exhaust ducts, not the heating/cooling duct system R403.3 is testing. Do not expand the exception into every duct that happens to move air.

Pressure is 25 Pa, which the code also writes as 0.1 inch water gauge. Blower-door envelope testing in R402.4 is 50 Pa. Do not mix the two manometer targets. Envelope ACH50 is not a duct-leakage substitute.

R403.3.6 — the numbers that pass

The total leakage of the ducts, where measured in accordance with R403.3.5, shall be as follows.

Read the table as four rows, not as permission to average them. A rough-in without the air handler uses 3.0, not 4.0. The ducts are being scored before cabinet leakage is in the measurement, so the ductwork itself must be tighter. After the cabinet is installed, rough-in with air handler and postconstruction both use 4.0 unless the entire system qualifies for the envelope row.

Field scenario — postconstruction 5.2 cfm per 100 square feet. After finishes, a dwelling is tested postconstruction at 25 Pa. Some ducts run in a vented attic. Total leakage is 5.2 cfm per 100 square feet of CFA. The report is signed. 5.2 exceeds 4.0. The test fails R403.3.6 item 2. It does not pass because it is under 8.0. The 8.0 row is only for systems whose all ducts and air handlers are located entirely within the building thermal envelope. Attic runs are not that row. Arithmetic on a 2,400 square-foot CFA house: 4.0 cfm/100 sf is 96 cfm at 25 Pa; 5.2 cfm/100 sf is about 125 cfm. The extra 29 cfm is a fail, not a rounding story.

Field scenario — ducts in conditioned space, test skipped. All ducts and the air handler are in a basement and in interior soffits, entirely within the building thermal envelope. The contractor skips duct-leakage testing and cites a 2018 rule that ducts in conditioned space are exempt. That exemption is not in 2021 R403.3.5. Testing is still required. When the location sentence in R403.3.6 item 3 is truly met, the limit relaxes to 8.0 cfm per 100 square feet of CFA. Relaxed limit means you still run ANSI/RESNET/ICC 380 or ASTM E1554, you still seal registers, you still include the air handler, and you still hand the code official a signed report. A letter that says inside conditioned space, no test is a 2021 fail.

8.0 is not a nickname for R403.3.2

R403.3.2 can consider some buried attic, floor-cavity, or exterior-wall ducts to be inside conditioned space for wrap and classification. R403.3.6 item 3 uses different words: ducts and air handlers located entirely within the building thermal envelope. A buried trunk in a ventilated attic is still in that attic. Do not give that system the 8.0 allowance just because option 2 was used to call it inside for insulation. Option 2 even demands a 1.5 cfm per 100 square feet to-outside tightness to earn the classification — a stricter to-outside number than 8.0 total leakage. Keep three families of numbers separate:

Protocol details that fail otherwise-good tests

• Wrong pressure. Scoring at 50 Pa because the blower door is already set up is not R403.3.5. Convert or retest at 25 Pa per 380 or E1554 as those standards direct; do not treat ACH50 as cfm per 100 sf.
• Registers not sealed. The section twice says registers shall be taped or otherwise sealed. A test with living-room grilles open measures rooms, not ducts.
• Air handler left off the postconstruction test. Postconstruction includes the manufacturer's enclosure. Capping the supply and return at the plenum to hide a leaky cabinet is not the entire system.
• Wrong floor area. Limits are per 100 square feet of conditioned floor area. Using garage area, porch area, or a single-story footprint on a two-story house changes the score. For the 1.5 classification number, use CFA served by the duct system.
• No signature. R403.3.5's last sentence is as mandatory as the manometer.

Plans review and inspection sequence

1. Identify whether the project will test at rough-in or postconstruction. Both are legal. Neither is optional except for the ventilation-only exception.
2. Identify whether all ducts and air handlers are entirely within the building thermal envelope. If any supply or return is in a vented attic, vented crawl, or garage, you are on the 4.0 / 3.0 rows, not 8.0.
3. At rough-in, ask whether the air handler is installed. Yes → 4.0. No → 3.0.
4. Collect the signed report. Compare cfm per 100 sf of CFA at 25 Pa to the matching row.
5. Remember Section 9.1 and 9.2 still apply on a passing test: R-8/R-6 wrap, sealed joints, no panned joists. A passing 3.8 cfm/100 sf does not legalize cloth tape or a cavity return.

Keep the wording independent: this is OpenExamPrep teaching of 2021 IECC residential duct-testing rules for Exam 79 study, not an ICC, RESNET, or ASTM endorsement.