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

Open-cell foam on roof deck

mark h | Posted in General Questions on

My house is spray foamed with open-cell Icynene on the roof deck as well as all walls. I have an HRVv for fresh air.

Everything’s been good but I’m noticing now that the humidity in the attic starts out around 36% in the a.m. but gets up to 60-70% during the day then back down to 40-50% at night. The temp is always within 5% of the house temp. I’m not sure what to do.

There isn’t any moisture sources in the attic. To make it more confusing is, there are two separate attics and both are doing the same thing.

Any help would be greatly appreciated. I don’t know what to do.
Thank you,

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. Blakeobeans | | #1

    Do you mean relative humidity? How are you monitoring the humidity?

  2. mark h | | #2

    im monitoring with a humidistat.i also have an hrv system to bring in fresh air.the thermastat for the hrv unit also has a reading for rh.

  3. GBA Editor
    Martin Holladay | | #3

    Please describe what kind of roofing you have, what kind of roofing underlayment, and what kind of roof sheathing. It's possible that the moisture is being driven inwards through your roof assembly due to inward solar vapor drive. However, this only occurs if all of the layers above your insulation are vapor-permeable.

    What is your climate zone? In cold climates, you can't install Icynene on the underside of roof sheathing unless the cured foam is protected on the interior with a vapor retarder.

    Several other GBA readers have had similar problems. Here are links to two threads on the issue:

    Icynene Moisture Problem in Cathedral Ceiling

    Icynene & moisture

    Moisture problem between roof decking and open cell insulation

  4. mark h | | #4


  5. GBA Editor
    Martin Holladay | | #5

    In Massachusetts, you need to install gypsum wallboard painted with vapor retarder paint on the interior side of a roof assembly insulated with Icynene.

  6. mark h | | #6

    oh boy... if i were to put a dehumidifier in the attic would that take care of it? i cant hang sheetrock over it. the rafters are encapsulated with foam and the roof is less then 2yrs old so stripping and putting a different vapor barrier instead of tar paper due to finances isnt possible either.

  7. GBA Editor
    Martin Holladay | | #7

    First of all, the daily swings in indoor RH may not be a problem. Are you seeing any signs of moisture problems in your attic?

    Second, the reason that you need the interior vapor retarder has nothing to do with the elevated RH in your attic. You need the vapor retarder to protect your roof sheathing, which is at risk of rotting due to accumulating moisture in the roof sheathing during the winter.

  8. mark h | | #8

    thank you for all the information. thers no moisture build up in the attic. how would i be able to tell if moisture is accumulatiing on the roof deck?

  9. GBA Editor
    Martin Holladay | | #9

    The standard method is to remove a small amount of Icynene, exposing the underside of the roof sheathing. Insert moisture-meter probes, with the wires leading into your house. Then fill up the hole with canned spray foam.

    Monitor the moisture content of the sheathing over the course of a year. You'll usually see the moisture level climb in late winter. This is what happens: indoor moisture diffuses through the Icynene (which is vapor-permeable) and accumulates the in the roof sheathing, which is cold.

  10. mark h | | #10

    wow! this is a nightmare! thanks for all the info martin

  11. davidmeiland | | #11

    What about a vapor retarder sprayed onto the underside of the Icynene?

    Frankly, I am surprised that we continue to read stories like this. You'd think that the insulation installers would understand the issues and know how to furnish a complete installation, rather than a time bomb.

  12. GBA Editor
    Martin Holladay | | #12

    Joe Lstiburek used to advise builders to spray vapor-retarder paint on the interior side of the cured spray foam. Then he tested the paint in a lab, and found out that when the paint is installed on a bumpy, porous surface like cured foam, it is worthless. The paint only works as a vapor retarder if it is sprayed on a smooth surface like gypsum drywall.

    More information here: Joe Lstiburek Discusses Basement Insulation and Vapor Retarders.

  13. mark h | | #13

    if the vapor barrier is on the inside would this trap moisture on the sheathing from the outside due to inward solar vapor drive?. i think my only option to correct this is to scrape the foam to the rafters and cover them with sheet rock & the vapor retarding paint.
    thanks for everyones help

  14. davidmeiland | | #14

    Or... could you apply a thin coat of closed-cell to the existing open-cell and get the VB/VR you need that way?

  15. GBA Editor
    Martin Holladay | | #15

    If you have asphalt shingles, I doubt that inward solar vapor drive is occurring. Armin Rudd, a researcher at Building Science Corp., calls this phenomenon "ping-pong moisture" -- moisture that comes and goes on a daily schedule. It is still somewhat mysterious, and deserves more research.

  16. mark h | | #16

    thats would be great if i could do that. im so confused about this . some say no vapor barrier some say i need one. i def have high humidity during the day in the attic but not in the morn or at night so its gotta be inward solar vapor drive. i bet closed cell sprayed overtop would be the best solution

  17. Expert Member
    Dana Dorsett | | #17

    With plank roof decking you would have quite a bit of time before this is a real problem but with OSB it's pretty urgent, and with plywood it's something in-between.

    Convert your relative humidity numbers to dew point or wet-bulb temp (measures of absolute humidity) which will give a more relevant picture, since the temperature is probably not constant in those attic spaces. (Use either a psychrometric chart, or an online psychrometric calculator.) If the afternoon dew point in the attic space air is at or below the night time lows, you can bet a substantial amount of that moisture is cycling back into the roof decking.

    Solar drives baking the moisture out of the plywood to be re-adsorbed at night through the highly permeable open cell foam is almost certainly what is going on. Give it at least until mid-July before closing it in with a vapor retarder though, since those solar drives are what's purging the moisture out of the plywood, and you DO want to get as much out as possible before putting a vapor-retardent element on the interior. You may see the humidity swing moderate as the nighttime outdoor temps moderate, which would be an indication that it is finally drying out. Closing it in with air-tight gypsum and a half-perm vapor retarder like vapor-barrier latex before October will keep it from accumulating much wintertime moisture next year, and but the drying rate will also be slow (but still fast enough- well under 6 months.)

    If you were planning to re-roof in short years, the alternative solution would be to make THIS the year, and add at least R20 in rigid insulation above the roof deck. That raises the average mid-winter temperature of the roof deck substantially, reducing the adsorption rate as well as the peak moisture content. If you're in National Grid's service territory for either electricity or natural gas, you could even get a substantial subsidy for doing that under their Deep Energy Retrofit program:

    For pictures & analysis of the foam-over roof deck solution, see:

    BTW: On clear dry nights like many recent days, the roof deck temperature radiation-cools to below the outdoor ambient, which makes the re-adsorption issue somewhat larger than it otherwise would be.

  18. mark h | | #18

    this keeps turning into a worse nightmare. where would i find a roofer that has expertise in this situation? my existing roof is only 2yrs old. i dont have the money to re-roof but i dont want my roof deck rotting out. i wish i never had sprayfoam installed. the whole purpose of the foam was to save money and have a comfortable home for my family and neither has happened.
    thanks for the info. i appreciate everyones help

  19. GBA Editor
    Martin Holladay | | #19

    The advantage of enclosing your Icynene with gypsum wallboard (painted, of course, with vapor-retarder paint) is that the drywall protects the spray foam from ignition. That's always a good idea from a fire safety point of view, and many building inspectors insist that spray foam be covered with a thermal barrier or an ignition barrier like drywall.

    However, if your local building inspector doesn't require protection with drywall, and you don't have a vapor barrier, you might consider covering the inside of the Icynene with a smart vapor retarder like MemBrain. That would be cheaper than installing a new roof.

    You might also contact the contractor who installed the Icynene. The contractor should have known that you need an interior vapor retarder in your climate.

    The Icynene website is still under the mistaken impression that vapor-retarder paint works on cured foam. They evidently haven't heard about the tests performed at the Buidling Science Corp. that show that vapor retarder paint doesn't work on cured foam.

    More info:
    "When open-cell spray foam insulation is used, a separate vapor retarder may or may not be necessary. In U.S. climate zones 1 through 4, a vapor retarder is not required by code, and Icynene’s building science engineers do not recommend it. Exceptions include places with high interior relative humidity, such as interior swimming pools. When a separate vapor retarder is needed, it should be applied to the warm side of the wall."
    "Icynene is an effective AIR BARRIER, but NOT a vapor barrier."
    "While Icynene LD-C-50™ is an excellent air barrier, it is not a vapor retarder. According to the state of Wisconsin building codes, a separate vapor retarder may still be required. A low vapor retarding paint/primer (perm rating below 1 perm) is recommended to be applied to the finished surface of the drywall to fulfill these requirements."
    "If you’re using an open-cell spray foam insulation such as Icynene LD-C-50 in a cold climate, you’re usually required to have a vapor barrier or vapor diffusion retarder on the interior (warm in the winter) side of the insulation. This blocks moisture movement from inside your house to the outside, where it may condense and freeze."

  20. homedesign | | #20

    Martin, You suggested : " covering the inside of the Icynene with a smart vapor retarder like MemBrain"

    I live in a house with open cell foam under the roof deck (note: would not do it again if I had a do-over)...and the roof rafters are encapsulated (like Mark's roof)... I can not imagine or visualize how your suggestion could possibly work.
    How could anyone create a COMPLETE air barrier against the lumpy foam with a sheet membrane?
    Not to mention all of the places where rafters meet ceiling joists and plates.

  21. homedesign | | #21

    Mark, I agree "this is a nightmare"....I do not not have any useful suggestions.
    However, I am curious... what season (date) was it when you took the humidity readings? and what were the approx. corresponding temperatures for the humidity readings? For example in the morning when the attic was 36% RH what was the approx attic temp? etc.
    The reason I ask is that the RH in my attic does not fluctuate wildy(daily) like yours seems to.

    It makes me wonder if you have a "bad" installation....for instance if the foam was off-ratio when installed...or the roof deck was not at the proper temperature or moisture content at the time of installation.
    I have read of a similar case to yours where there were air pockets between the foam and the roof deck.
    The foam looked fine...but when some excavation was done...the voids were revealed.

    Makes me wonder how many "time bombs" are "festering" out there!!

  22. mark h | | #22

    the tempature dosent fluctuate. it stays consistant with the downstairs. the morning the rh was 36% the temp outside was in the mid 30's and the temp in the attic was 70 degrees. its only the rh that fluctuates. iam going to have a blower door test done to check for air leakage to rule that out.
    martin would i try and adhere the the men brain right to the under side of the foam?

  23. GBA Editor
    Martin Holladay | | #23

    I understand what you are saying. Trying to attach MemBrain to lumpy foam and encapsulated rafters would be difficult and would look goofy. That said, it's not impossible -- and it's cheaper than a new roof. I was trying to think outside the box in an effort to be helpful.

  24. mark h | | #24

    martin do you think i could adhere it directly to the foam with some type of glue? if i could get it on somehow this would solve the problem of vapor passing through the foam in the winter and condisating on the roof deck? if so this seems like the most cost effective way to fix the problem and give me piece of mind at the same time.

  25. homedesign | | #25

    Even if you could somehow create a "tent" of MemBrain ....The part that I see as virtually impossible would be "connecting" the "tent" of fabric to the wall Air Barrier.....
    Without a continuous Air Barrier (A 3 dimensional "Red-Line") .... the house and attic air WILL find a way around and enter the unavoidable space between the membrane and the foam..

    I am not trying to be unhelpful here... I just can not visualize how a retrofitted membrane air barrier could work.

  26. GBA Editor
    Martin Holladay | | #26

    The vapor transport mechanism here is diffusion, not air movement. No air is moving through the roof assembly. Unless there is a leak at the ridge, there is no reason for air to be pulled into the area.

    The MemBrain could be attached at the perimeter of the attic with 1x3s attached with screws. The same method could be used elsewhere -- using 1x3s screwed through the foam to the rafters.

    If 90% of the area is covered with a vapor retarder, then 90% of the vapor diffusion has been addressed.

  27. homedesign | | #27

    Martin, Air is NOT "pulled" is "pushed" ;--)
    I believe that "vapor" is also transported by variations in fluid density..."Buoyancy"
    In addition to thermal buoyancy there is also moisture buoyancy.
    Paraphrasing Julius Sumner Miller.....What happens is the less dense air (the warm moist air)is pushed up by the more dense air.

  28. mark h | | #28

    matin do you think i should attach it to the gable ends as well?

  29. GBA Editor
    Martin Holladay | | #29

    I'm not sure what your point is. I still think that MemBrain, attached with 1x3s and long screws, would address the diffusion issue.

  30. GBA Editor
    Martin Holladay | | #30

    Q. "Do you think i should attach it to the gable ends as well?"

    A. Yes.

  31. mark h | | #31

    thank you martin

  32. davidmeiland | | #32

    I'm having a hard time with the MemBrain idea. Vapor diffusion is the issue, but unless this non-rigid material can be fashioned into an effective air barrier, won't humid indoor air simply get behind it and diffuse into the foam? I would be more inclined to shave the foam flush to the rafters and then install either drywall or maybe a layer of rigid foam that's taped (to make up for lost insulation value, provide an air barrier and a vapor barrier. Maybe you get an ignition barrier that way too. The MemBrain installation would depend on perfect taped seams and inside corners, etc.

  33. GBA Editor
    Martin Holladay | | #33

    Like you, I think that the best solution is to install gypsum drywall. That was my first recommendation. The drywall will provide a thermal barrier, and will provide a smooth surface for vapor retarder paint.

    Mark is reluctant to install drywall, so I tried to suggest another approach. Admittedly, it's not as good.

    Concerning the risks of vapor diffusion problems if Mark installs MemBrain without drywall: I don't think there will be any air flow through the roof assembly with my suggestion, and the amount of air exchange due to "pumping" from temperature changes will be slight. Remember, even with your suggested approach -- using drywall -- there will still be air between the drywall and the bumpy foam.

  34. homedesign | | #34

    Mark reported the outside temperature was in the Mid 30's...
    Mid 30's is not-so-cold.
    We often get Winter temperatures in the mid 30's in North Texas.
    And we experience "night sky/sky" cooling....and solar heating.
    My house has rafters encapsulated with open cell foam.
    The foam has no vapor retarder coatings, no gypsum.
    The temperature and RH in my attic are always very similar to the Living Zone.

    I have NEVER noticed the WILD RH fluctuations that Mark is reporting.

    It makes me think that there is more going on here than a missing vapor retarder.

    The other cases I have read about with wild daily Humidity swings involved rafters that were not encapsulated or else some type of hidden foam failure(where the foam separated from the roof deck).

  35. GBA Editor
    Martin Holladay | | #35

    Maybe it's a bad hygrometer? Perhaps it's time to verify the readings with another meter.

  36. mark h | | #36

    when the temp starts out in the mid 30's the rh is good. as the day gets warmer the rh statrs to rise. then as evening comes the rh starts to fall back to normal. the temp in the attic is always the same as the house within a few degrees. if it was bad foam or not adhering wouldnt the temp flucuate not just the rh? martin i also think drywall is the best fix. i need to find a reputable contractor to do it that has experience with tight homes which as of now seems impossible

  37. homedesign | | #37

    Hi Mark,
    Thanks for sharing your "problem".
    You have given a limited description of your house & attic.
    How about posting a few pictures of the exterior and the attic?

    You say your home is "tight" do you know it is tight?
    Is the house 1 or 2 story? basement or crawlspace or other?
    What size are the roof rafters? 2x6?
    How thick is the open cell foam between rafters?
    How thick is the foam that covers the bottom of the rafters?

  38. homedesign | | #38

    I'm trying to noodle on why the RH in my attic does not "Yo-Yo" like Mark's Attic...
    Is the difference in the R-Value?
    The open cell under my OSB roof deck is at least 8 inches thick and covers the bottom of each 2x8 rafter by at least 1 inch.

    Is the difference in the properties of the foam?
    I think every open cell foam job is different...
    not only are there variations in formula by company...
    there are variables at the job site....
    it depends on micro climate, surface conditions and who is "cooking" the batch(Walter or Jesse).
    How fussy is the equipment? clean are the nozzles and hoses?
    How "Crusty" is the outside Skin of the foam?
    (the foam in my attic has a definite crust/skin)
    Was it laid up in one pass or two?

    Maybe I got lucky and Walter was "cooking" that day.

    What if "Jesse" was the "cook" for Mark's foam?
    Did Jesse miss a few hard to get to spots?
    Did Jesse have a hangover that day?

  39. mark h | | #39

    the house is a raised ranch. 10inch of foam. the rafters are totatly encapsulated. the basement is walk out with no moisture issues. if there were voids in the foam i would think i would have temp swings but as i said before the temp in the attic is the same as house. the foam covers the rafters by at least an inch. & ive had a blower door test done so i know how tight the house is. the foam does have a crust on it

  40. homedesign | | #40

    Thanks Mark
    do you remember the ACH-50 ?
    you mentioned earlier that you were "going" to have a blower door test done... so I assumed you had not had one.

  41. davidmeiland | | #41

    If cost were no object, you could have someone evaluate the foam installation with a blower door and IR camera. If there are voids, it should be possible to detect them this way, if the inspection is done correctly.

    At this point, I would probably track down the humidity swings a bit more carefully. The easiest way is to use simple USB dataloggers, and you would probably need 3 of them if not more. Getting an accurate picture of the temp/RH movements over a period of days might prove useful. Maybe there is a workable solution other than encapsulating the foam. Finding out where the moisture originates is important. Specifically, I would want to look at whether there is any relationship between RH inside the house, RH outdoors, and RH in the attic.

    Mark, who is in the household? Any chance you have a teenager who doesn't like to use the bath fan, or some other indoor source of humidity?

  42. mark h | | #42

    i have 3 kids & a wife in the house. during the afternoon when the humidity spikes theres no showers going. sometimes the kids take showers at night and this is when the humidity drops whether they take showers or not. i have booked a blower door test to be done to see if there is any air leakage from the attic. vapor retarder is very confusing. i spoke ti an icynene rep who said boston shoudnt need one. the insulation co. said i didnt need one but here and some other sites online recomend spraying vapor retardent paint but now i read that dosent work. i could try and sheetrock the attic but its a raised ranch so the attic is very tight and the roof line is low so i dont even know how i could do that. thanks for everyones help

  43. Expert Member
    Dana Dorsett | | #43

    "I'm trying to noodle on why the RH in my attic does not "Yo-Yo" like Mark's Attic..."

    The answer lies in:

    "We often get Winter temperatures in the mid 30's in North Texas."

    Mid-30s peak is not the same as mid-30s average, and it takes time for moisture to accumulate in the wood. Lubbock's January mean temp is a bit north of 40F, which is about the dew point of comfortable healthy 35%RH/70F air, so it wont' have a sustained period of moisture adsorption unless you were actively adding a LOT of humidity to the air.!dashboard;a=USA/TX/Lubbock

    In Mark H's southeastern MA the January mean temp is ~28F, well below the 40F dew point of health conditioned air, and it has many weeks of moisture accumulating temperatures:!dashboard;a=USA/MA/Taunton

    Once accumulated, it takes time to bake that moisture back out. So during the daytime under the solar heating of the sun the wood is outgassing water vapor, but when it cools off at night it stops, and if cold enough, even re-adsorbs a bit. With warmer weather it'll be purging moisture more continuously, and the yo-yo will eventually stabilize. That's when it's time to close it in with something more vapor-tight, but not before.

    If MemBrain is the solution it doesn't matter when you close it in. When the humidity rises behind Membrain during the day it becomes more vapor-open, and allows it to pass, but when the humidity drops at night it becomes more vapor tight, slowing any moisture drives from the conditioned space air.

  44. mark h | | #44

    dana do think membrain would work? do think i would be able to attach it to the foam 100% around the edges? what do you think the best way to attatch it is
    thanks mark

  45. Expert Member
    Dana Dorsett | | #45

    You can (carefully) staple it in place, and foam-seal the edges with can-foam, tape over any micro-tears at the staples.

    Even though it's a PITA, I'd still recommend putting half-inch gypsum up over it, painted or not. With the MemBrain as an air & vapor barrier the air-seal detailng on the gypsum needn't be perfect, and you wouldn't need to paint it. The advantage of a smart-vapor retarder rather than vapor barrier paint is that the assembly dry at 10x faster through unpainted gypsum, even though it's allowing moisture in at roughly 2x faster than v..b. paint. If you do your best to make both the MemBrain and gypsum air tight, and you should be fine. Even if you get SOME air leakage behind it, the total amount of mid-winter moisture getting to the roof deck will have been reduced by at least 2 or more orders of magnitude, yet the drying rate is only reduce 1 order of magnitude in the parts where the air leaked. It doesn't need to be absolutely perfect to do a world of good.

  46. mark h | | #46

    thank you for all your help dana

  47. mark h | | #47

    also thanks to everyone for all the info

  48. mark h | | #48

    dana do you know of any roofers that have done the foam board on the outside of roof? as you said in an earlier post thats the best solution. i was in the attic last night and due to all the hvac in the attic it dosent seem possible to apply the membrain and sheet rock.

  49. Expert Member
  50. davidmeiland | | #50

    I've heard that name before. A guy that I consulted a bit for hired them to solve a gnarly issue with his house.

  51. mark h | | #51

    i tried to get a hold of him yesterday no awnser & no machine. its the number for synergy construction.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |