Buried Ducts in Humid Climates WITHOUT Condensation
![Various Ways To Install Ducts Overhead, Including Buried Ducts [Image Credit: IBACOS]](https://www.energyvanguard.com/wp-content/uploads/2025/12/attic-duct-system-configurations.jpg)
Think about moving cold air through a hot attic or hot air through a cold attic. Building codes require only a measly R-8 insulation on the ducts. It may seem natural to bury those ducts beneath the attic insulation to improve energy efficiency. My contractor friends out west do this all the time and have no problems. But if you try to do buried ducts in humid climates, you run the risk of condensation.
That’s what the 2018 IRC tried to address, but that first attempt was a dud. The code requirement that was nearly impossible to meet was to use R-13 duct insulation in IECC climate zones 1A, 2A, and 3A. None of the duct insulation manufacturers made R-13 duct wrap, and the code change didn’t create much demand for it either. Now we have a much better protocol for doing it without the condensation risk, and it’s based on solid research recently completed in Florida.
What kind of buried ducts?
Before jumping in, let me define what I’m talking about—as well as what’s not included—when I use the term “buried ducts.” The research did not address ducts that are only partially buried. Nor did they look at encapsulated ducts, which are first covered with an inch of closed-cell spray foam before being buried in blown insulation.
The two categories of buried ducts in the study are “completely buried” and “deeply buried.” They’re both buried but they differ only in the amount of insulation over the top of them. Completely buried ducts need to have 3.5 inches of loose-fill insulation on top, and deeply buried ducts need to have R-30 on top.
They call these out separately because you can model them differently. You treat completely buried buried ducts as in an attic with R-25 duct insulation. When you do deeply buried ducts, you get to treat them as being in conditioned space.
More research led to a better protocol
Last month we had Nelson Conarroe from Owens Corning (OC) as the speaker at our Atlanta BS & Beer meeting. (BS stands for building science, of course.) He gave a fascinating presentation on the research that OC has been doing the Florida Solar Energy Center (FSEC). (Download the slides here.)
I had doubts about how well just going to R-13 duct insulation would protect against condensation. In my 2017 article on the 2018 IRC requirements, I wrote this:
There are pitfalls here, though. Here are a few:
- Buried ducts that are poorly insulated, even in one small area
- Too much duct leakage
- Improperly sealed vapor barrier on the duct insulation
One problem was that it was relying on duct installers to get the details right. If you understand ducts, you probably know that a high percentage of duct systems are installed poorly. And the other glaring omission was leaving the attic vented with outdoor air…and all its humidity.
Adding the missing ingredients
The new research that Owens Corning did at FSEC came up with a much safer way to do buried ducts in humid climates. They even made it more doable by reducing the duct insulation to R-8, which is the code required level for vented attics. The key to making it work was also addressing the humidity in the attic. Here are the main factors they found that make buried ducts in humid climates work:
- R-8 duct insulation
- Unvented attic
- Small amount of supply air from the air conditioner (50 cfm per 1,000 square feet of attic floor area)
- Vapor diffusion port (See this article by Dr. Joseph Lstiburek to learn about this feature.)
They put the various attics in the FSEC test hut under what they thought might be the best range of conditions to stress the system. The varied the color of the roof, since light-colored roofs reflect more heat and could make condensation more likely. They also tested with the thermostat set at 68 °F (20 °C), since that would make the ducts colder and more likely to have condensation.
When they did those four things above, they found: “Attic air relative humidity and roof wood moisture in unvented attics with buried ducts and insulation on the ceiling was acceptable during summer and winter conditions.”
What about using a dehumidifier?
Since they’re sealing up the attic, you might wonder if they tested how well using a dehumidifier might work. And yes, they did test that. Here’s what the full FSEC report says in the executive summary:
Using a dehumidifier in unvented attic to control duct surface moisture is not reliable enough and requires a high amount of energy and long runtimes.
- The RH [relative humidity] sensor of a dehumidifier is under very different conditions than the buried duct surface. The lowest RH setting of control was needed to get just marginal moisture control during hot and humid weather.
- Based on measured data and estimates from attic test cells, dehumidifier energy use is expected to be at least an order of magnitude higher than using conditioned supply air into the attic.
So just stick with the items in the previous section if you want to do buried ducts in humid climates.
Code requirements for buried ducts in humid climates
Even though they just finished up this research, some of what they recommend made it into the 2024 International Residential Code (IRC). Here’s the relevant section:
N1103.3.5 (R403.3.5) Ductwork buried within ceiling insulation
3. In Climate Zones 0A, 1A, 2A and 3A, the supply ductwork shall be completely buried within ceiling insulation, insulated to an R-value of not less than R-13 and in compliance with the vapor retarder requirements of Section M1601.4.6.
4. In Climate Zones 0A, 1A, 2A and 3A where installed in an unvented attic with vapor diffusion ports, the supply ductwork shall be completely buried within the insulation in the ceiling assembly at the floor of the attic, insulated to an R-value of not less than R-8 and in compliance with the vapor retarder requirements of Section M1601.4.6.
Item 3 above is basically the provision from the 2018 IRC, which requires R-13 duct insulation but not the other stuff. Item 4 shows the new requirements. (Items 1 and 2, which I didn’t show, cover duct insulation.)
The 2024 IRC doesn’t say anything about putting a little supply air into the attic, but they don’t say you can’t do it either.
Don’t sleep on duct condensation
If you’re wondering how important all these precautions are, go up into an attic in a humid climate sometime. Condensation on ducts is common there. Usually it happens where two ducts are touching. That causes the space between the ducts to get colder, but it’s still in contact with the humid attic air.
![Microbial growth on ceiling of New Orleans home [Photo courtesy of Bill Robinson]](https://www.energyvanguard.com/wp-content/uploads/2025/12/ceiling-mold-attic-duct-condensation-humid-climate-1000.jpg)
Allison A. Bailes III, PhD is a speaker, writer, building science consultant, and the founder of Energy Vanguard in Decatur, Georgia. He has a doctorate in physics and is the author of a bestselling book on building science. He also writes the Energy Vanguard Blog. For more updates, you can follow Allison on LinkedIn and subscribe to Energy Vanguard’s weekly newsletter and YouTube channel.
Lead image by IBACOS, used with permission.
Related Articles
Buried Ducts Risk Condensation in Humid Climates
Buried Ducts Allowed in 2018 Building Code
Is Venting an Unconditioned Attic Necessary?
Comments are welcome and moderated. Your comment will appear below after approval. To control spam, we close comments after one year.
Thank you for this article, what an important topic!
I’m slightly confused with the design though; they put a supply in an unvented attic above the thermal boundary? This seems like this would result in that boot sweating on top of wasted energy. Or was it buried within the insulation?
Another question, wouldn’t by installing a supply in an unvented attic put the conditioned space under a negative pressure? I thought this was bad because you don’t want unconditioned, humid air being drawn into homes in the South.
Thank you again for the article. Any clarity or additional reasoning here, would be greatly appreciated.
Or is this saying that the roof deck is also insulated, bringing the entire ducts within the thermal boundary? If that is the case, I would struggle to understand the purpose of burying the ducts if they are already within the building pressure boundary/ thermal boundary.
Mitch, no, the insulation is on the attic floor, not at the roof deck.
If you have an unvented attic with an air barrier at the roof and insulation in the ceiling with no vapor barrier there, what happens in the winter. In the north, there would be a lot of water condensing or freezing under the roof deck. Does it never get that cool in the south?
RoyC: Yes, that could be a problem if the weather is cold for a long enough period. Since they did the study in Florida, they didn’t have much cold weather. They wrote:
“The generally colder conditions did result in higher roof wood sheathing WMC compared to the very dry summer and fall periods, however, they were not high enough to elicit concern of building degradation under the test conditions reported here.”
In climate zone 3A, though, there might be times when this could be a problem.
Mitch: You’re right to be confused about those things. Yes, the thermal boundary is at the attic floor, but the air barrier is at the roofline. If the air barrier gets sealed completely, the supply air doesn’t get lost to outdoors. It just lowers the temperature a bit and helps to dry the air. And yes, adding supply air could cause pressure imbalances. Since the air barrier is at the roof deck, a small transfer grille to the conditioned space below could solve that problem, but I couldn’t find any mention of that in the report.
I just can’t see the benefit of an unvented attic with ceiling insulation instead of roof insulation. Is it a cost issue? If so, there are many other benefits of insulating the roof instead of the ceiling. Duct insulation and condensation is not a big issue, equipment maintenance in the attic is much simpler, the attic can be used for storage, etc.
Hi Roy, I had much the same gut reaction — but the first cost of SPF is huge, AND it happens out of the builder’s usual construction sequencing. In contrast, an unconditioned unvented attic allows the builder to skip having soffit vents and soffit baffles, and to use the same amount and type of insulation s/he is used to using for the attic, while making minor process tweaks to build the ridge vent as a vapor diffusion port and to add a small supply duct into the attic.
Allison, wouldn’t a transfer grill from an unfinished attic introduce additional pollutants into conditioned space? Just revisited the Ashrae 62.2 discussion from way back in 2013… and one thing I absolutely agree with Dr. Lstiburek is that reducing contaminant sources is a better strategy than increasing ventilation rates, but that’s a whole other topic…
Brian, I still cannot imagine an air handler in an uninsulated attic in FL, even with a little supply air helper (ironically coming from the said air handler!). I always thought that a career as a builder is dealing with constant process tweaks. If a builder cannot deal with or pay for roof insulation, tweak the conventional attic, or unvented, but you will have to be ready for more tweaks coming at you in the form of conditioned space ducts and equipment.
Equipment not working with you? Change it. A ducted mini split air handler enclosure needs about 18 vertical inches or less out of a 9-foot ceiling in a closet.
Benefits come with tight homes. Ducts can get smaller and shorter, they don’t have to reach into every nook and cranny, throwing air at trouble spots becomes a non-issue, even dedicated return ducts (OK, I better zip it…).
I know they’re doing this so the industry can continue to place HVAC systems in the attic but there’s no way productions builders will get this right.
“3.5 inches of loose-fill insulation on top” of an object or a line (a round duct) in practice is a very tricky situation. Not only is the imaginary border plane tricky (where blown-in insulation ends and open air begins), but the blowing method throws a curve ball as well. For fiberglass, was it blown to 0.3lbs/cu ft? Was it 0.4? 3.5 inches is such a short distance in loose-fill fluff.
If you are asked to do 3.5 inches, do 7 and report back “yes, I did 3.5”.
Allison, the drawing at the top, I could not find it in the Florida report. Could you explain what ICSP stands for? In my opinion, the last method (encapsulating in foam) is assuming that flex ducts last for ever, or that they are so magnificent as to deserve a foam blanket.
In practice, if you’re going after deeply burying the ducts, you need to be aiming for somewhere between R-49 and R-60 overall cover to meet the R-30 on top. And the ducts better lay flat.
Paul: Yes, ICSP stands for in conditioned space.
Allison, thank you! Is that drawing from the Florida study?
No, I saw it in the Owens Corning presentation slides, which included a link to another set of slides from IBACOS. I’m not sure if it was used somewhere else also, but IBACOS did make the diagram. Here’s the link to their slides:
https://www.phrc.psu.edu/assets/docs/Webinars/Ductwork_in_Attics_Handouts.pdf
Thank you, Allison, for the link. As one of their slides mentioned, “More research needed to fully characterize the phenomena”. More empirical studies such as the one from FL that you cited. These phenomena are hard to simulate, and get even harder when human sloppiness (installation, materials, or both) is added.
Allison,
Gotcha, thank you for the response. I understan4d this better now, just not something I’ve seen before, so talking through it is helpful.
I guess my only other question would be: If the attic is unvented with an air barrier and no insulation (spray foam) wouldn’t that allow for convection/ conduction heat transfer and significantly raidse temperatures inside of that sealed attic space, creating a hot stagnant micro environment?
I understand that there would be a supply to help mitigate that, but would it be able to keep up? Maybe so? Maybe a cool roof would help too?
There are homes here in Charleston SC that have very similar challenges with the design, and again this is a great article. I appreciate your expertise on building science.
Mitch: The attic doesn’t need to be kept cool. It just needs to be dry, and that’s the only purpose of the supply air. Yes, a reflective or light-colored cool roof would help keep the attic cooler. So would a radiant barrier. But with the ducts deeply buried in the insulation and R-38 insulation would prevent most of that heat from getting into the living space below.
Thank you Allison, this makes sense. The Goal is to learn something new every day and grow/ refine knowledge. Happy Holidays.
Thanks for sharing this research, Allison! I’ve been eagerly awaiting the details of Nelson Conarroe’s presentation. Definitely some surprises in here!
Allison,
I’ve been thinking about this a bit more, and I am trying to visualize/ understand how the supply air would dry the air as intended. From thinking of the HVAC side of things specifically, that supply register is just an extension of the rest of the ductwork system serving the conditioned space, and the system would only be providing cool air when the house calls for it, not the attic.
So how would the attic receive enough cool to dry the space out? Is the thought that the air barrier is sealed so well, that negligant levels of humid outside air will infiltrate the space? If so, I guess an advanced blower door test would be recommended on these spaces to verify the seal.
This aligns with doing blower door tests on sealed attics with open cell spray foam applied to the roof deck; I find a lot of unintentional pathways at the ridge, rake and eaves.
I have this issue in one of our bathrooms for the HRV vent. On extremely cold days, like today where it is sub 10F, we get some drips out of the vent. Running the fan extra usually clears this up, but it is annoying.
Scott: That’s condensation on ducts, but it’s a different kind from what I wrote about here. What you’re experiencing is humid air going through a cold duct rather than a cold duct surrounded by humid air. The duct temperature is below the dew point of the air going through, so condensation happens inside and then drips down to the vent in the bathroom. Does your HRV extract vent duct run through unconditioned space? It shouldn’t but some installers don’t know that. See this recent article I wrote about H/ERV problems:
11 Ways to Screw Up an ERV Installation
https://www.energyvanguard.com/blog/11-ways-to-screw-up-an-erv-installation/
Yes, I understand the warm moist air is condensing inside the hrv exhaust. I don’t blame the installation, no other way to do it (2nd floor, vented attic). And this was installed almost 10 years ago before the good idea fairies have had time to figure a lot of this out.
I’ve been in the attic and these are buried pretty deep- realistically 10-15” under the blown in cellulose considering the pipes run across the lower roof trusses. About 18-24” total depth of insulation.
Allison, in an existing home, would a practical first step to assessing the risk of duct condensation be comparing supply air temperature at the air handler vs. the farthest supply register? Of course, this kind of testing would need to be done under specific exterior conditions.