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

Spray foam from crawlspace onto bottom of plank subfloor in Pacific NW?

frasca | Posted in Energy Efficiency and Durability on

Hi there – I’m in a 1955 house in Seattle, WA, over a vented, unconditioned crawlspace, with hardwood flooring on top of plank subfloor with no insulation under it.

I have read “Building an Unvented Crawl Space” and several related threads, and the advice to seal one’s crawlspace all seem to come with the caveat that in drier or Western climates the benefits are less. I even read the WSU study ( that showed several sealed crawlspaces had higher winter energy consumption than vented ones, even when there was ductwork down there.

Given that I am starting from scratch, the easiest solution seems to be to have 2″ of closed cell sprayfoam sprayed directly onto my plank subfloor. It won’t nearly prevent thermal bridging from my 2×10 joists, but will be a perfect airseal and insulate a heck of a lot better than what I have.

The next easiest seems to be to install rigid foam boards against the bottoms of the joists and tape the seams. I’d also have to insulate the rim joists and I’d be sentencing myself to several weekends in the shallow (2-4′) crawlspace, but maybe the method is superior (and cheaper?) enough to justify the time.

Any advantages or disadvantages I might am missing? Is spraying foam against the bottom of plank subfloor a no-no for future repairability?


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
    Martin Holladay | | #1

    Either approach will work. Before you commit to the spray foam option, consult a spray foam contractor to make sure that access is adequate. A 2-foot-high crawlspace is rather tight -- you may need to lower the floor for better access.

    For more information, see How to Insulate a Cold Floor.

  2. Expert Member
    Dana Dorsett | | #2

    The notion that there is a way to "...prevent thermal bridging from my 2x10 joists..." is misguided. The only way to prevent thermal bridging is to not use the framing cavities for insulation at all.

    For the same amount of money as 2" of closed cell foam (R12-R14) you could install ~5.5-6" of half pound open cell foam (R20-ish), and it would cut the amount of heat moving through the framing by about 2/3. The air tightness of 5.5" of open cell foam would be somewhat better than 2" of closed cell foam.

    At 6" the total amount of polymer in half pound open cell foam is about the same as a mere 1.5" of most closed cell foam, and it's blown with water, which has very low environmental impact compared to the industry standard HFC245fa ( and extremely powerful green house gas) used for closed cell foam. There are low impact HFO blown closed cell foams now too, but they command a premium. Applying that premium to simply more open cell foam would get you north of R25.

    Summertime outdoor dew points in the Puget Sound region average WELL below the indoor temperatures of even the chilliest air conditioned house, and there is effectively zero risk of moisture accumulating in open cell foam or the plank subfloor, a risk that could at least theoretically occur in the tropics with air conditioned houses insulated with open cell foam.

    Rigid foam board on the underside of the joists can be pretty cheap too, especially if done as a DIY using reclaimed roofing foam. It takes at least 2.5" of continuous polyiso or 3.5" of EPS to hit the thermal equivalent of 5.5" of open cell between floor joists. But rigid foam under the joists doesn't air seal the band joist or subfloor anywhere near as reliably as 5.5" of open cell foam would.

    Even with a vented crawlspace and it's still worth putting down a ground vapor barrier. With insulation at the subfloor the exposed portion of the joists will run cooler and higher moisture content year-round than it has been. A ground vapor barrier guarantees that the crawlspace air's moisture content will track pretty closely to the outdoor air's dew point which in your climate would be sufficient to control the mold hazard.

  3. frasca | | #3

    Thanks to both of you.

    Dana - re: "For the same amount of money as 2" of closed cell foam (R12-R14) you could install ~5.5-6" of half pound open cell foam (R20-ish), and it would cut the amount of heat moving through the framing by about 2/3."

    Is this true even though the 2x10 joists will still not be 'buried' in foam? My thought was that any depth of sprayed foam less than 9 1/4 + would still let those joists transfer heat conductively, at R 1.2 per inch or whatever framing lumber does, right through the insulation...

  4. mackstann | | #4

    The joist is a low-R-value short circuit. The length of the short circuit is the depth of the insulation the heat is going around.

    If you have 2" of closed cell foam, then the short circuit is only 2" long. That's how far heat needs to travel through wood to get around the foam. So about a quarter of the floor area is R-2.4.

    If you have 6 inches of open cell foam, then the short circuit is 6" long and that same ~quarter of the floor area is R-7.2.

  5. Expert Member
    Dana Dorsett | | #5

    What Nick said: With 2" of closed cell foam you have a 2" short circuit through the framing, with 6" of open cell foam it's a 6" path. That's 3x the R value, and thus 1/3 the amount of heat traveling through the thermal bridge.

    With a full cavity fill it's a 9.25" path and about a 4/5 reduction in heat transfer through the framing, but it's still an R10-ish thermal bridge through insulation that has 3x or higher R.

    If you were considering encapsulating the sides of joists in 2" of closed cell foam you'd end up at about the same performance as a full cavity fill of open cell foam or a combination of foam + fiber, but at a much higher cost.

    If air sealing and getting close to code-min R performance is the goal, 3.5" of open cell foam does about as well at air sealing as 2" of closed cell, leaving ~5.75" of depth to fill with something cheaper than open cell foam. Cheap low density R22 fiberglass get you the remaining R needed. R22s are manufactured at nominal 6.75" thickness, and would compress easily into 5.75", performing at about R20 at that thickness. Combined with the R13 open cell foam that would be about R33, which exceeds the IRC 2015 code minimum of R30 by a small amount. That delivers the 4/5 reduction of heat transfer through the thermal bridge rather than just 2/3 you'd get with just 5.5-6" of open cell foam.

    Alternatively, installing 5.5- 6" of open cell and contractor rolls of R13 fiberglass delivers about the same performance.

  6. frasca | | #6

    Ok I'm sold on having my insulation contractor quote me an open-cell foam job at that depth, which I hadn't even been considering before posting here. I've been mulling over this statement by Dana though:

    "Summertime outdoor dew points in the Puget Sound region average WELL below the indoor temperatures of even the chilliest air conditioned house, and there is effectively zero risk of moisture accumulating in open cell foam or the plank subfloor..."

    I totally agree; to see a dew point of 60°F is rare here, even in the summer. But what about the converse, when I'm heating in the winter?

    Quick searches of this site and others suggest that a comfortable indoor air dew point in the winter might be around 45°F, - 40% RH at 70°F, (higher than I keep it but roll with me for purposes of explanation) - which is certainly higher than the OAT.

    Is there a reason I'm missing why we don't worry about moisture accumulating in the open cell foam in the winter?

  7. GBA Editor
    Martin Holladay | | #7

    Q. "Why don't we worry about moisture accumulating in the open cell foam in the winter?"

    A. Moisture doesn't really "accumulate in the open-cell spray foam" as long as the open-cell spray foam can dry in at least one direction. During the winter, when the vapor drive is from the warm, humid interior toward your cooler crawlspace, the open-cell spray foam will dry downward. If there were a condensing surface in the assembly -- for example, if someone stupidly installed a sheet of polyethylene sheeting on the underside of the joists, and the polyethylene got cold -- then moisture might accumulate on the interior side of that condensing surface during the winter. But you won't be installing polyethylene.

  8. Expert Member
    Dana Dorsett | | #8

    The outdoor dew points are even lower in winter, and the vented crawlspace air's dew point will track that of the outdoors. In winter the temperature in the crawlspace is also warmer than the outdoors by at least a few degrees, so even when it's 100% relative humidity saturate pea soup fog outdoors, the relative humidity in the crawlspace will be at a lower RH.

    Three quarter inch thick ship-lap is also a vapor retarder, as are most finish floor materials. Open cell foam isn't even Class-III vapor retarder until it's approaching a foot thick, and will readily pass any moisture that diffuses through the flooring into the vented crawlspace. You can think of the floor and crawlspaces as something like a horizontal wall with rainscreened siding (in this case the siding is the dirt) that has an interior side vapor retarder (the flooring), and NO exterior sheathing- just open cell foam facing the rainscreen gap. The moisture content of the open cell foam and joists are then almost entirely a function of the outdoor air's humidity, and the temperature of the crawlspace floor.

  9. frasca | | #9

    Makes sense. I’m assuming that the same logic in about of open cell foam does NOT apply to sprayfoam applied from my attic onto the underside of my roof - technically onto a mixture of polypropylene smartbaffles and the 2x4 top sections of my trusses - because in that case moisture from the warm air inside could condense against those two materials and accumulate in the open cell foam?

  10. GBA Editor
    Martin Holladay | | #10

    The situation with open-cell spray foam adjacent to roof sheathing is more complicated than for crawlspaces. Some of the reasons for these differences in attics are known, and some are unknown.

    Suffice it to say that if your roof assembly is vented -- as is implied by your mention of ventilation baffles -- your roof assembly should be unlikely to experience moisture problems.

    Unvented roof assemblies are another matter. For more information, see these articles:

    Open-Cell Spray Foam and Damp Roof Sheathing

    High Humidity in Unvented Conditioned Attics

  11. Expert Member
    Dana Dorsett | | #11

    In a vented roof in Seattle you can safely go with open cell foam too. Even at R49 open cell foam is more vapor open than standard latex ceiling paint, and it would dry into the vent space faster than moisture can find it's way in from the interior. It's only a problem if applied directly to the underside of the roof deck.

    If the open cell foam would encapsulate the interior half of the top chords of the trusses you don't even need R49, you'd be able to hit code min with about 10" of half-pound foam, as long as it's still 10" deep over the top plates of the structural walls. (That's usually only possible with "energy heel" trusses.)

  12. frasca | | #12

    Thanks to both of you for the advice on this.

  13. 1900DisasterHome | | #13

    I have a crawl space 3 feet high with about 5" open cell from previous owner. I encapsulated with good quality materials to keep pipes from freezing, but radon in that room went up from about 3-5 to about 7-14 depending on weather. What is best solution? I have an unused HRV and thought about using that, but it seems to be more popular to put a fan under vapor barrier. I'm in Columbia MD, we get warm humid summer and cold dry winter and everything in between.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |