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

Heat Loss from Ducts in Unconditioned Crawlspace

jonathanb | Posted in Energy Efficiency and Durability on

We bought a beautiful home in sunny California (Zone 3C) that has long runs of R-8 ducts in an unconditioned crawlspace, with a very recent 24,000 btu heat pump that can’t quite keep up.

If I’m calculating things right, the heat loss from just the ducts is roughly this:

(10″ diameter flex duct * 1 ft / 12 in) * pi * (100 ft duct length for supply and return) * (0.125 U-value for R-8) * (28 degrees temperature difference at the design condition of 40 degrees outside and 68 degrees inside)
= 920 btu

According to CoolCalc, the whole area served by this heat pump only needs 18,000 btu.

Assuming this is accurate, our 24,000 btu heat pump should be more than adequate.

It gets a little worse. It’s likely that the ducts are losing some additional heat from any crimps from the supports hanging them from the floor joists, and there might be a little air leakage anywhere the joins aren’t perfect. Even if it’s double the loss, that’s still under 2,000 btu. That’s not enough to explain what’s going on.

The house must be leaking more than 4,000 btu more than what CoolCalc says it should be. Where is the heat going?

Some ideas:

1. ($) Insulate the ducts better. Unfortunately, I don’t think there’s much we can do. We have a tiny crawlspace, only about 20″ high, and 10″ R8 duct takes up most of that.  Next idea:

2. ($$) Encapsulate the crawlspace. If we could reduce the temperature differential between the crawlspace and the outdoors, that could help. There’s just not enough heat loss to explain the problem though. Even if duct loss was zero, that’s small compared to the performance gap we’re seeing.

3. ($$$) Replace the very new new 24,000 btu air handler with a new 36,000 btu unit. This would work, and it’s what any HVAC contractor might recommend. But it’s not addressing the root of the problem, which is air infiltration or loss.

4. (-) Wear extra layers of clothes on cold days. Get a space heater. This would work, but again, it doesn’t address what’s wrong.

5. ($$) Air seal the attic and crawl space. Maybe we have a chimney effect that’s causing much greater loss than what the Manual J / CoolCalc predicts.

Anyone know if poor air sealing can cause more than 4,000 btu of heat loss? With what we’re seeing, it would have to be 6k or 8k btu.

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. jonathanb | | #1

    Hmm, maybe it's even worse in the crawl space than I calculated?

    The *supply* ducts are sending warm air. That means they have an even greater temperature differential. The return ducts are sending room temperature air, which is what I was using to estimate.

    So about 30 ft of duct are for the return air. About 70 ft is for supply. It's a little bit of a simplification to call it all 10" diameter, since it's mostly 10", but the branches for the registers are 6" and in one case 4". It does add up to about the equivalent of 70 ft of 10" flex duct though.

    Updated calculation:

    Return duct heat loss = (10″ diameter flex duct * 1 ft / 12 in ) * pi * (30 ft duct length for return) * (0.125 U-value for R-8) * (28 degrees temperature difference at the design condition of 40 degrees outside and 68 degrees inside)
    = 275 btu

    Supply duct heat loss = (10″ diameter flex duct * 1 ft / 12 in) * pi * (70 ft duct length for return) * (0.125 U-value for R-8) * (55 degrees lift in temperature)
    = 1,250 btu

    TOTAL LOSS = 1,525 btu

    That's worse than my first calculation, but still not even close to enough to explain the problem.

    Let me try plugging in different values to CoolCalc and see if there's any setting that could cause the heating load to be many thousands of btu underestimated.

  2. Deleted | | #2


  3. this_page_left_blank | | #3

    You're multiplying inches and feet. Divide your results by 12 for the correct numbers.

    1. jonathanb | | #4

      Ack, you're right!! I'll update the previous comments. Thanks, @Trevor Lambert!

  4. charlie_sullivan | | #5

    Trevor correctly spotted the error. Still, you'll want to make some kind of envelope improvement to help it keep up, and encapsulating the crawl is an option. Have you had a blower door test?

    1. jonathanb | | #6

      I've got a quote to get that blower door test. Sounds like I should do it!

  5. joshdurston | | #7

    Get your ducts sealed. I bet your supply or return is leaking to the unconditioned crawl space. Then your envelope.
    Maybe aero barrier for the Ducts?

    1. jonathanb | | #9

      Could be! The HVAC contractor didn't find a leak, but I should get a second opinion.

      AeroBarrier looks really interesting. Can they use it just for ducts, not for the whole house? We have a house full of furniture and stuff, and it seems like they target new construction.

      1. mathiasx | | #14

        They did say on Matt Risinger’s recent video that Aerobarrier can do a finished house, but it’s a lot more prep work and time on site, which adds up $$$. The caulk shouldn’t stick to vertical surfaces but it will stick to horizontal. So everything needs to be under plastic and well-sealed/taped.

        I’m debating Aerobarrier but going to time it for when we can remove all the furniture from our upstairs for a remodeling project (and then sealing the stairwell and ducts beforehand.)

        1. MVinColorado | | #15

          We did Aerobarrier on my occupied house two weeks ago as an experiment. Moving out and in again was more expensive than the process, and we had some bad luck. The door to the basement accidentally blew shut and then the sprayers down there didn't work. While there were measurable gains they were not big enough to justify the effort/expense.

  6. Expert Member
    Akos | | #8

    Your house looses much more heat from the unconditioned crawlspace than what is lost through well installed R8 ducts. Encapsulating and insulating your crawlspace will make the house much more comfortable and also reduce your energy use significantly.

    Unless you are in an area with a lot of radon or flooding, this is worth doing.

    1. jonathanb | | #10

      Thanks, @1at. Crawlspace encapsulation is on our list. We'd been thinking of doing it after air sealing and increasing the attic insulation, which is currently R-19 fiberglass bats. Would you recommend a different order, like doing the crawlspace before adding more attic insulation?

      1. Expert Member
        BILL WICHERS | | #19

        Crawlspaces are usually pretty leaky, especially if they're made that way intentionally with "vents" :-) If you have to split your projects, I'd do this order:

        1- Do a good check of all your crawl space ductwork, seal any seams/gaps/leaks you find with mastic. I like the fiber reinforced duct mastic. You'll want some fiberglass mesh reinforcement tape for any larger gaps (this stuff is similar to the mesh-type drywall tape). This will at least cut down on air leaks from the ducts themselves, and it's relatively quick and easy to do.

        2- Do your air sealing project, but BE SURE to include the rim joist area in this work. You'll be a doing a bit of "double work" this way, because you're going to come back to this area again when you encapsulate your crawl space to tie in the perimeter membrane/insulation with the rim joist insulation, so ideally you'd do the rim joist and crawl space together at the same time. The reason I'm recommending you do the rim joist area first is because you're particularly concerned with air leaks. In my own home, I did rim joist air sealing and insulating (using EPS and canned foam) prior to encapsulating the crawl space for exactly this reason.

        3- Add attic insulation. R19 is awfully thin here, so it should be a high priority insulation upgrade project.

        4- Encapsulate the crawl space, and insulate the perimeter walls. Make sure to tie this into the rim joist area in terms of keeping a continuous air barrier.


        1. jonathanb | | #20

          Thanks, Bill! This is really valuable advice, especially about the rim joist air sealing.

          There's so much great expertise on GBA. Thanks to everyone.

  7. bluesolar | | #11

    I'm a bit confused – how do you know that there's a problem? I must be missing it in your post. So you ran the math on how your system should be performing, but I don't see anything about how it is actually performing, or what the problem is exactly. Is it not heating the house?

    Relatedly, to know if there's a system-level problem wouldn't you need a thermal model of the home? I don't know how you're computing leakage or energy loss without a fuller model, but I've not done it before. When you say that 24,000 BTUs should be more than adequate, the implication is that the system is not in fact adequate, but I don't see any math related to that or any measure of what's happening. The only math you've presented is theoretical losses, but there's nothing about the actual behavior or performance of your system.

    1. jonathanb | | #12

      Fair comment, @BlueSolar. I didn't explain why we believe the system isn't adequate. Here's the data:

      It's a Carrier ducted mini split. It maintains the set temperature within half a degree during normal operation in mild weather, both heating and cooling.

      On cold nights that get down close to the design conditions of 40 degrees, the indoor temperature drops by several degrees and in the extreme case can't heat the house back up to the set temperature even running 24 hours.

      We have made sure the air filters are replaced and clean, the ducts have proper air flow, the air coming out of the registers is warm, all windows and doors are closed and sealed.

      We had the HVAC company come and check out the system and look for any issues. They went into the crawl space and checked all of the ducts for kinks or leaks and were unable to find a problem.

      Given all of that, our best hypothesis is that we have some kind of heat loss. We've used CoolCalc to explore what might cause a big enough heat loss to explain the problem. We know that infiltration, attic insulation, crawl space encapsulation, and replacing windows that aren't already low E ought to help. But we can't completely rule out some kind of equipment issue.

      Does that help explain? I left those details out. Any ideas?


  8. MVinColorado | | #13

    You mentioned AeroBarrier but in this case you'll want AeroSeal. Same process but franchised into two different companies. Cost ranges between $1800-2500. That said, if I read this right the HVAC company didn't really find anything wrong with the ducts, so the effect may be limited.

    1. jonathanb | | #17

      Thanks @MVinColorado! I'm not sure how the HVAC company checked, so I might also want to get a second opinion before I rule out duct leaks.

  9. user-2310254 | | #16

    Ducts in a ventilated crawl space are problematic. (There are multiple posts and articles on GBA about why they should be avoided.) If it were my house, sealing that space would be at the top of my to-do list. (If the "floor" is dirt, you might want the contractor dig it out to increase headroom and make the work a bit easier.)

    That step and air sealing where appropriate might solve your problem (or not). If the problem persists, you might want to order a new Manual J and a new Manual D from an independent engineer.

    1. jonathanb | | #18

      Thanks, @Steve Knapp!

  10. jonathanb | | #21

    Thanks to all who replied to this thread!

    I wanted to follow up and let you know what we ended up concluding. We did get a Manual J and Manual D done. And we realized many things: the system was oversized, especially for the bedroom. It was located at one end of the house instead of centrally. The ducts weren't leaking, but they were badly designed, had inadequate return air, fittings that reduced airflow, and overall the static pressure was too high.

    It turned out that the problems were so great that we replaced the entire system, ducts and mini splits both. We got a Fujitsu that was half the size, and it heats and cools better, produces higher airflow, and cut our electric bill in half.

    We used a different contractor. The original that installed the Carrier had hundreds of 5 star reviews on Yelp. The new contractor was not more highly rated, but clearly had MUCH more experience with mini splits and new ductwork. This was a very, very expensive learning experience.

    Thanks to everyone on GBA for pointing us in the right direction!

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |