GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Acceptable duct leakage and AeroSeal?

airfix | Posted in General Questions on

New Construction, Climate 6B.

On my drawing I have a requirement that all ducts shall be sealed and leak tested per IRC 2015 N1103.3.3.

The leakage rate specified in N1103.3.4 seems quite large less than or equal to 4 cubic feet per minute per 100 sq ft of conditioned space.  For a 2000 sq foot home that’s 80 cubic feet per minute.

What is an acceptable leakage rate?  Should I target better than code and if so what is a good target?

The next problem is my installer said the ONLY way to get a sealed HVAC system is to use Aeroseal.  He gave me no other options and didn’t discuss the cost. I don’t think he was even aware of the pressure test requirement until he was 50% done with the duct work when I asked him about the sealing and leakage.  As far as I can see he has only used mastic on a few connections to the air handler.

The reading I’ve done on Aeroseal indicates it’s not a very cost effective method of sealing ducts and it’s only good for 10 years.  It seems it is much better to seal the ducts using mastic during the construction.  Considering that most of the duct work is already installed what options do I have?  I’m I now forced into using Aeroseal?


GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Brian Pontolilo | | #1

    Hi Steve,

    As you know, Aeroseal is not the only way to seal ducts. So at the risk of sounding snide, the option you have is to make sure that the duct system get's sealed and tested now, before construction continues and the work gets more difficult to do well. You can demand it from this contractor, find another contractor, or do it yourself. And you can have the work tested. Here are some articles that may be helpful:

  2. Expert Member
    BILL WICHERS | | #2

    >"The next problem is my installer said the ONLY way to get a sealed HVAC system is to use Aeroseal."

    That may be true in a practical sense for sealing existing, inaccessible, ducting in an old house, but for new construction it is most definitely NOT the case. Mastic and tape can both be used to seal ducts. You just need to be able to access the ducts to be able to do the sealing. Do your sealing work while the walls are still open and you have cheaper options available to you than Aeroseal.


  3. airfix | | #3

    The square ducts look like they are built using sheet metal with some kind of 90 degree slip or press fit joint on 2 diagonals of the 4 corners (the other 2 corners being folded) to make the square cross section. Do each of those seems need to be sealed or is the joint good enough without mastic or tape?

    If those joints need to be sealed and the square duct is up recessed in the floor joists there will be no access to seal the corners even if there is no sheetrock on the walls.

    Is the IRC code minimum for duct leakage a good enough standard to shoot for without going to extreme measures or can you get better than code with normal sealing best practices? How much better than code should I be shooting for to not make it a burden for the installer?


    1. jrpritchard | | #4


      You should be able to hit the code min leakage target but simply sealing the easy joints. You can do much better than that by targeting a few often forgotten about areas. The joints you are taking about are generally called S slip and drive cleats. The s slips are tighter but neither will hold once pressurized. I like to use mastic on all s slips and drive connections. The long seams on your duct are probably snap lock joints which are often easier to tape with a good butyl backed duct tape. Even if they are Pittsburgh seams it’s still a good idea to tape them. The gores on any adjustable elbows should also sealed with mastic. Finally I like to use clear silicone around the equipment on areas that are highly visible. Silicone makes for a beater appearance. Make sure to hit areas like the furnace to a coil seam, coil pan and refer line attachment points and the plenum edge. A plenum edge is often decorative and snd leaks bad when under pressure. Metal to flex connections are also very leaking - I find it best to prime the metals with mastic, slide the flex over the metal, secure with a panduit strap, and then tape the joint. An entry level HVAC apprentice should be able to seal a duct system in less than a day without to much trouble. The biggest mistake is not using enough mastic plus it seems it’s almsot impossible to keep it out of your hair.

    2. Expert Member
      Dana Dorsett | | #8

      >"Do each of those seems need to be sealed or is the joint good enough without mastic or tape?"

      Mastic is the right approach. Tape is second best, but still pretty good if the ducts are shiny-clean when the tape is applied. Leaving it unsealed is not an option. Leaving the register boots unsealed to the subfloor/wall gypsum/ceiling gypsum is also not an option.

      An abbreviated spec for the minimums lives here:

      >"Is the IRC code minimum for duct leakage a good enough standard to shoot for without going to extreme measures or can you get better than code with normal sealing best practices? How much better than code should I be shooting for to not make it a burden for the installer?"

      It's better to shoot for under 40 cfm/25 (one of the Energy Star compliance options) independent of the house size (within reason) rather than code max 4 cfm/25 per 100 square feet of conditioned space. For a 1000' house 40 cfm/25 would be the same as 4 cfm/25 per 100', so for a 2000' house that would be half the maximum leakage allowed under the IRC. That's do-able in a 2000' house but don't sweat it or beat up on the contractor if you end up at 50-60 cfm/25.

      The other Energy Star spec worth shooting for is room-to-room pressure differences of less than 3 pascals (= 0.012" water column) under all air handler speeds, with all conditions of room doors open &/or closed. This is a duct design & balance issue.

  4. airfix | | #5


    Thanks so much for the feedback. I was up there today and it looks like they indeed have used mastic on many of the joints but I did see something troubling.

    I'm not 100% certain but it looks like they used the square box section created by the floor joists as the actual duct work for at least 2 or 3 runs. From where I was standing it looks like they just used mastic to seal a piece of sheet metal to the bottom of the floor joist allowing the subfloor and the two floor joists to be the other 3 sides of the duct.

    This can't be legal can it?

    How on earth do they seal the joists to the sub-floor and the seams between sub-floor sections seams? It just doesn't seem right.

    I'll see if I can get some clearer pictures. If they have done it this way should I have them re-do it or is this okay?


    Edited to attach picture.

    1. jrpritchard | | #6


      Unfortunately what you are seeing there is industry standard practice. I wouldn’t worry much about the subfloor but often there are many other things running through joist spaces - electrical plumbing etc which are much harder to seal around. A fully ducted all metal return system is the best of course but panning joist is most common and I am guessing completely acceptable to your bullding official. If you look around upstairs it is possible that they are using wall cavities to pull return down, which then drop into a joist space which then gets panned back to your air handling unit. If they are using a wall space it doesn’t make sense to change the panning unless you are also going to change wall space returns.
      The bigger problem with using joist space is that most have a free area of something like 12x12 or 14x12 which makes for a big duct and lazy air. A wall cavity is 3.5x14 typically. It is possible they are using a joist space for multiple return runs which can help with sizing but it usually ends up with a wall cavity that’s to small and return space that’s to big and then some trunk that is right sized. With competitive marketplace and metal prices increasing almost monthly returns are often the first place to cut. When doing high performance homes my preferred approach is to use floor returns rather than wall returns. From there you can use a register boot and round pipe to fully duct the return system. You might hear that your supply and returns must be in different planes ie one high and one low but I have found that with a proper manual d and more importantly a proper manual T the room still mixes just fine and the grills in the floor are a better look for most designers. It allows for much easier balancing on the return side of things. Your system will likely work just fine as many others do. My current house is 50 years old with panned returns and works just fine. Issues like this get back to one of my biggest complaints with the building industry - if a customer like you Steve approached me to discuss your desire to build a high performance house we could talk about all these things. We could discuss your budget and your goals and what’s possible. From there it’s easy for everyone to move forward on the same page. Instead - generals beat down their subs and create competition in order to get the cheapest price possible. Oftentimes I never get to talk to the owners until the job is awarded Even good subs have to chose to either cut corners or walk away from the jobs. When I am bidding against other hvac contractors that have never even heard of a manual d and I often wonder what’s the point of even trying. Just doing a manual j s d and t would price me out of most houses in my area. Sorry to rant bottom line Steve - check with your building official and if he says it’s ok it’s probably ok to leave it. Make sure the seams on all the adjustable elbows get sealed up well looks like they missed those in the pictures.

      1. Yupster | | #9

        As a note, "lazy air" e.g. low velocity isn't a problem in ducts in conditioned spaces. Lower velocity just means lower friction rate and less noise.

        1. jrpritchard | | #10

          Sorry should have clarified what I meant there While I typically agree with lower velocity is ok - most panned joist spaces are close to 14x14 and are trying to carry back as little as 100 CFM. The velocity is off the charts on the low end probably less than 50 feet per minute. During cooling mode the main purpose of a return is to bring humid air back to the air handling unit and I prefer to not have ‘lazy’ humid air traveling through a chase made of organic material.

          1. Yupster | | #13

            I can appreciate your concern but I don't think it matters. The wood joists are going to be exposed to the same amount of humidity in the air, regardless of what velocity the air is travelling. To the best of my knowledge low velocity doesn't increase absorption.

      2. airfix | | #11


        Thanks again. I appreciate the rant. It has been a struggle at every stage of this building process to get the subs to know that I'm willing to spend more in order to get a robust design and installation. However the subs being used are conditioned to "this is the way we always do it" so when I come along and question why and want to make changes some (not all) give me pushback.

        It's good to know the right way to do it and then I can make decisions about what I'm willing to accept and what I'm willing to pay for.

        At the end of the day it might not be done exactly how I would want it, but if I keep on top of the processes hopefully I can make improvements in some areas and I'll end up with a "pretty good" home, better than most not as good as some.


    2. GBA Editor
      Martin Holladay | | #7

      Note that the 2009 IRC, Section M1602.2, requires that "Joints of duct systems shall be made substantially airtight by means of tapes, mastics, gasketing, or other approved closure system." Note also that the supply and return duct systems must meet duct leakage test requirements -- even when panned joists are used for returns.

      More information here:

      "Open Spaces as Return-Air Options - Code Notes"

      "Perhaps the Worst HVAC Duct Idea Ever — The Panned Joist Return"

      1. airfix | | #12

        Great resources martin. Thanks.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |