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

Cellulose Hot roof application?

user-887642 | Posted in Green Building Techniques on

I am trying to build a super or well insulated workshop with upstairs room. The main story are walls made from ICF, the upper are trusses with “room in attic” setup. It has a steel roof on purlins. with the top chord of the trusses being 2by8 and 2by4 purlins (stood up) i think i have created a decent amount of space to insulate. i was hoping that this would mitigate issues with the sloped part of the wall ( this area in my house, a cape cod, seems to be problematic for heat loss)

my original plan was to have a sprayfoam hotroof approach, but the more i learn about sprayfoam, the less i like it… i like the idea of using cellulose. i then thought about doing a flash and bat system, using cellulose over 1-2 inch of sprayfoam (to seal things up) . i wouldn’t be opposed to cellulose only method of insulation (dense pack). but the problem seems to me is: can cellulose be used for a hot-roof approach?. i have never heard of this being done, but i’m sure there are countless vaulted ceilings out there that are full of cellulose (although that doesn’t make it right). also, if this is an approved method, will the pressure of densepack make my metal roofing bulge? its 28 gauge screwed on 24″ centers

i thought i had it all figured out, but it seems to me the more i research (as i build) the more decisions i have to make.

details are:
building is 25 by 32, i live in wisconsin, simple gable roof, 12 foot ceilings on first floor, and 8′ on second (attic) floor room.


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. Riversong | | #1

    ICF walls are hardly super-insulated for a Wisconsin climate, and they are the second worst way to build with concrete and foam. Interior foam is a bit worse, but ICI offers little dynamic thermal mass benefit and leaves both sides with a fire hazard.

    You may run into racking problems with your 2x4 purlins on edge under a heavy snow load which will tend to make them roll.

    And a metal roof should not have any type of insulation in direct contact since the metal has a large coefficient of expansion and is a near-perfect condensation plane. There should be an air gap, if not ventilation, under the metal and some membrane to isolate the thermal layer from the roofing.

    Other than all that, cellulose is used in hot roofs, though the advantages of a vented roof are significant enough that I would never consider anything but. If it's used in an unvented roof, there must be a near perfect air/vapor barrier underneath, continuous with the air/vapor barrier of the wall assembly.

  2. GBA Editor
    Martin Holladay | | #2

    I agree with Robert. No insulation should touch your metal roofing, and it's best to have a ventilation gap above the cellulose.

    Ideally, you should have an air barrier on the top side of your insulation. Obviously, metal roofing is not an air barrier.

  3. jklingel | | #3

    Martin: Could you elaborate on this a tad? "Ideally, you should have an air barrier on the top side of your insulation. " I'm missing something. If you want ventilation, then why do you need an air barrier above the ventilation, which is above the insulation? Or were you referring to a hot roof there? thanks. john

  4. user-887642 | | #4

    thanks for the answers.... i was lured into using ICF mostly by the promise of speed and ease of construction... i havent heated the structure yet, so i cannot attest to how well it works yet.. i'm sure it will do better than a conventional framed 2x4 house... other than that, i wouldn't wager anything. as, i didn't originally plan to vent the roof, i am unsure now how to go about venting it.. the purlins run horizontal to the roof plane, and and i worry that they may block any airflow. i am also unsure on how to get any sort of barriers up into the gridwork that is my rafters/purlins. The insulation contractors i have spoken with all suggested shooting sprayfoam onto the bottom of the roof.

    would you reccomend a way to go about insulating this roof? will the ribs in the roofing serve as proper ventilation (to get around the purlins) that way i can use the purlin space for insulation also. would putting XPS or EPS with standoffs up against the bottom of the roof be a proper air barrier?

    i'm just so frustrated, because every step of the way i have consulted with manufacturers, and contractors and builders and have followed what i thought was the right way to do things, and it seems like the information that i now gather is that they didn't know as much as they said they did.


  5. user-869687 | | #5

    Tyler, have you already installed the trusses, 2x4 purlins and the roof? If you haven't then the best answer would be a somewhat different assembly. For example you could add a panel of some type over the purlins before the roof is installed (could be plywood, exterior gypsum sheathing or homasote) and use that as an upper boundary for cellulose. Then add a layer (or two) of asphalt paper / roofing felt to help carry out any condensation from under the metal.

  6. user-869687 | | #6

    If you do have the roof on already, then here's a recommendation for that: First place a vapor permeable panel against the underside of the purlins in between the 2x8 rafters. This could be as suggested above, gyp sheathing or homasote. Plywood is really less permeable than would be ideal. Then furr out a dropped ceiling using 2x3s parallel to the rafters, attached with gusset plates. The ceiling will be low obviously, and the lower you take it the more room there will be for insulation. Insulate the 2x8 bays and the extended space below with dense packed cellulose.

    Add vents in the soffit to ventilate between the purlins (not between the rafters), and you'll have a cold roof. This is a safer strategy than an unvented roof, and especially one with insulation touching metal directly below the sky. The downside would be that condensation below the metal would drip onto the panel below and would have to evaporate dry, because it can't run out. But so long as the rate of drying exceeds the rate of wetting there's not a problem, and a 3.5" air space should allow a decent rate of drying for any drips.

  7. GBA Editor
    Martin Holladay | | #7

    John Klingel,
    You asked, "If you want ventilation, then why do you need an air barrier above the ventilation, which is above the insulation?" The answer is, to prevent wind-washing.

    The problem of wind-washing is greater with fiberglass batts than with cellulose, but it's a problem with either insulation. If you are building an insulated cathedral ceiling, with an air space with insulation below, you don't want cold outdoor air flowing over the top of your insulation fibers. If you have that, the air flow is robbing heat from the air between the fibers.

    Ideally, you want to install a rigid panel as an air barrier between each pair of rafters. Many builders install 1x1 or 1.5x1.5 inch sticks in the corners of each rafter bay, followed by Thermoply, cardboard, thin plywood, or even rigid foam. (High-permeance materials are better than low-permeance materials.) These panels should be installed in an airtight manner, with caulk or canned foam.

    The fibrous insulation can then be installed below the panels.

  8. Tyler H | | #8

    Thanks for all the info guys.. i need to digest it a little... its so nice to have access to a community with this kind of information.

    The roof is already on, to answer your question. i guess i am more in damage control mode now =/

  9. jklingel | | #9

    "John Klingel,
    You asked, "If you want ventilation, then why do you need an air barrier above the ventilation, which is above the insulation?" The answer is, to prevent wind-washing." Thanks. I did not see exactly where you were stopping the air. That makes perfect sense (or, cents, as the case may be). If I may ask, is there any point in preventing wind washing with a cold roof and 16" of cellulose blown in on a flat lid? I don't see any practical way to cover the cellulose w/out compressing it, so I guess the answer is "no; rely on the soffits and gables keeping the wind to a minimum." Thanks. john

  10. GBA Editor
    Martin Holladay | | #10

    If you have deep cellulose on an attic floor, you definitely need wind-washing protection at the perimeter, where air enters at the soffits. Ideally there is solid blocking between the rafters, to block air movement between the perimeter wall top plate and the underside of the ventilation chute.

    People have experimented with the use of Tyvek as an air barrier on top of attic insulation, but it usually makes more sense just to install a deeper layer of cellulose to address any performance degradation attributable to wind-washing.

  11. jklingel | | #11

    Martin: Got it. Thanks. j

  12. J Chesnut | | #12

    Tyler I am wondering what exactly the use of this building is. Since it is not your home you might not necessarily have to condition all the spaces at all times during the year. What will the attic space be used for? If your thermal envelope were defined at the ceiling of the main story workshop you could simplify things.
    Do you have to maintain certain interior conditions to store certain materials? A workshop implies physical activity and you can go a long way with minimal heat with layers of clothing and still be quit comfortable.
    If you're in Wisconsin and are building a workshop you are probably an ice fisherperson are your not?
    Go Packers : )

  13. Tyler H | | #13

    The upper area was going to be like a rec room/workout room for kids and they wife.. the downstairs will be more an autobody type area. my work may eventually spill over into the upstairs though, and it could be turned into workspace. that is good food for thought though. i guess i could only heat it when i plan to use it, but it floors walls and equipment take a while to heat up, and that may prove to be annoying... would airtite drywall help me at all? also would there be enough airflow if i vented from rake eaves as suggested above? am i looking for air flow or just air space? Yes to ice fishing =)

  14. J Chesnut | | #14

    When designing highly insulated air tight homes the strategy is heat retention. In a cold climate this makes sense for a home because the interior spaces are being conditioned 24/7. It sounds like the workshop does not have to always maintain a certain temp and relative humidity because there is neither plumbing or sensitive materials.
    Next question then is how often you would use the workshop. This will determine whether it is worth retaining heat which is what a highly insulated and air tight assembly does. A weekend cabin for instance will not retain heat over the course of the week unless it approaches Passivhaus standard assemblies that leverage passive solar heat gains.
    The amount of time it takes to heat up the space depends on the temperature the space starts at and the kind of heating equipment. A propane heater can provide heat almost immediately. Electric resistance will take awhile.
    It is still a question in my mind how much effort should go into air tightness and high insulation value. It is a large structure and it would probably be more energy efficient in this scenario to keep the people warm with heating equipment rather than keeping the whole space conditioned.
    An exercise room about a workspace with potential off gassing chemicals could be an air quality concern. If your trying to achieve a tight envelope you'll need to include mechanical ventilation.
    Ventilation below the roofing serves two purposes: extending the life of the roofing and allowing any moisture accumulation in the roof assembly to dry out. Keeping the attic cooler could be considered a third concern but this is better accomplished with insulation. In this case the insulation value above your second floor space will be as important to keeping the space cool in the summer as it will be to retain heat in the winter.
    So you have no roof sheathing? Your metal roof just sits on purlins and the purlins are attached directly to the trusses? In this case as long as the insulation doesn't touch the metal roofing as has been recommended you have sufficient drying potential. The only concern would be with the air tightness of your ceiling assembly. Warm humid air could leak through gaps and condensation could occur on the steel roof. But I think this being a periodically used workshop this is a low risk scenario because you are not talking about people cooking and showering in the space during the winter 3 times a day.
    In any case it is worth air sealing between the ICFs and the roof assembly and the airtight drywall approach on you second floor. As a precaution I would include a vapor retarder (poly or latex paint) on the ceiling/ slant wall interior plane if you are only going to insulate will cellulose or another highly permeabe insulation.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |