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 vs. Spray-Foam Insulation in Attic

Andre0170 | Posted in Energy Efficiency and Durability on
Hello all – hoping for some some answers regarding attic insulation. I’ve read several of the articles here along with commentary which have all been helpful, but still questions linger…

The situation: I am a building owner, not a builder. New Hampshire climate – two story commercial office building circa 1952. Pic enclosed… First floor brick facade, second floor are gable sides with slate roof on longest sides, brick wall on shortest sides to flat membrane rooftop. The membrane portion is about to be replaced with TPO and 2” tapered polyiso just to shed water to the gable sides and off. AC ducts are in the attic servicing the second floor that will soon carry heat as well from retrofitted exterior heat pumps, to replace old exterior ac condensers. Attic space is a crawl space approximately 36” making for difficult access. Appears to be 2×8 or 2×10 rafters/members. There are three steel girder-style I-beams from  brick wall to brick wall within the space for load. Old degraded batting insulation on attic floor between joists.

The question: Given above situation, loose blown cellulose, or spray foam?


I can remove old attic floor bats, add vapor barrier and loose blow cellulose to r-60 on the floor (blown to depth taller than joists and ducts to reduce thermal bridging), but this entails allowing access to the space from above by opening the roof in spots during roof replacement making it easier for the insulation contractors to blow insulation. This option also necessitates the need for addition of roof venting where currently there is none;
Considering ductwork in the attic and the subsequent possible need for a conditioned air space due to moisture and air temps affecting air in the ducts, I can remove old batting, and use spray foam on the underside of roof sheathing and interior attic side walls, also to r-60, sealing any air leaks and creating a conditioned attic. I’m also Also told this avoids the need for additional venting, and avoids the extra cost of opening the roof during roof replacement. Spraying would be done from the inside despite tight quarters, after roof was completed.
(Assuming one would not spray between attic floor joists, which would prevent the attic from being “conditioned space”…?)
Question on longevity and best energy envelope for efficiency. I’ve read about thermal bridging at the rafters, if spray foam, necessitating the need to either encase the roof rafters in spray foam, or spray to the needed r-value between the rafters and use strapping and rigid board over bottom of the rafters and backfill the remaining intra-rafter voids with cellulose. Due to space and retrofit constraints this is labor intensive and cost prohibitive.
So… would love additional insight on advantages / disadvantages of spray foam vs blown cellulose in this particular situation? Cost difference is significant, but not insurmountable. More concerned with energy efficiency and longevity for roof and energy systems.
Thank you very much in advance…

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. Expert Member
    Dana Dorsett | | #1

    Slate is/was usually installed on skip sheathing, which wouldn't be amenable to spraying foam from the underside. Have you taken a good look? Can you share some pictures of the interior of the attic?

  2. Andre0170 | | #2

    Thanks Dana - it looks to be t&g but might just be planks. It looks too tight to be skip sheathing but I might be wrong. Enclosed are a few pics just scroll to see them. One is looking at the gable sheathing from inside a 2nd floor closet, one the same from in the attic space, one a wider shot looking into the attic space, and one shot of particular concern looking down the side of the building from within the attic space where you can see a space between a brick load wall and the sheathing... not sure how to get insulation into that space down the length of the building. Thanks for having a look...

  3. Andre0170 | | #3

    Shot from inside the attic looking at the gable sheathing under the slate

  4. Andre0170 | | #4

    Wider shot of the attic

  5. Andre0170 | | #5

    Space between brick load wall and gable sheathing under the slate that runs the length of the building...

  6. Expert Member
    Dana Dorsett | | #6

    It's probably ship-lap sheathing with #30 felt or similar on top.

    This is one of those rare situations where I'd be inclined to recommend an HFO-blown closed cell foam on the underside of the roof deck. It only takes ~5" to hit code minimum if the rafters are also encapsulated. Going with 9-10" of open cell MIGHT work on the underside of the slate section, but could be risky under the TPO roof.

    I'm not sure quite how to deal with the long skinny triangle space with brick on on one side and the roof deck & rafters on the other. If it's a south facing pitch with slate on top it's reasonably safe to dense-pack it with cellulose (even though that would technically be a code violation) by installing air barriers on the ends and stuffing it full. If north facing that might be too risky.

    1. Expert Member
      Zephyr7 | | #9

      >"I'm not sure quite how to deal with the long skinny triangle space with brick on on one side and the roof deck & rafters on the other."

      If it's not too deep, a spray foam crew would probably be able to get in there from each side and build up a thick enough layer on the underside of the roof sheathing to be OK. From the look of the hole between the brick and the rafters/roof, I don't think there would be enough clearance to get a foam gun in there and make a bend with the hoses attached, but I think it would be possible to get enough coverage from the two sides at a bit of angle that the foam expansion would fill in sufficiently. It might not be the most pretty part of the install, but it should work.

      It's very difficult to judge the depth of that hole though. If it's much more than 3-4 feet from one end to the other, then it might be a problem to get full depth on spray foam in the middle part.


      1. Andre0170 | | #10

        It would appear that the hole runs the length of the building along that support wall - rounghly 50 feet from end to end. Seems weird that they would have let that stay open...

        1. Expert Member
          Zephyr7 | | #11

          That's going to be a very difficult space to insulate well. If the rafter cavities are open from the top side of the brick wall, then it's probably possible to foam down from above. If there is blocking up there, then that's probably not an option. I don't think a dense pack crew is going to be able to fill that entire space either.

          If you'll have polyiso above the roof sheathing, your best option might just be to accept a little less insulation in that relatively small area, and rely on the polyiso to keep the sheathing warm enough to avoid any moisture isssues in "the void".


    2. Andre0170 | | #12

      Dana - just a follow-up. Care to provide a little more detail on why you would choose the HFO-blown solution rather than cellulose? Thanks so much - and to all, for your thoughts!

  7. Charlie Sullivan | | #7

    Dana's advice is good (as always). Just one other possibility I want to ask about: can you up the polyiso thickness above the roof deck and use a little less spray foam on the flat part?

  8. Andre0170 | | #8

    We can... just trying to find the most cost effective way to get the best result. Increasing polyiso on the roof may not make as much difference if we’re blowing foam anyway, due to the increased cost of the poly vs the labor involved in cracklings around in that space blowing foam. The triangular space is on a west facing wall - and the same likely on the east side, just haven’t crawled all the way in to look. If dense pack cellulose in that space, only what would be the risk to the tpo? Unless it’s because there isn’t venting in the space?

  9. Expert Member
    Akos | | #13

    Slate roofs have a lot of permeability which allows the assembly to dry to the outside. This means you can safely insulate with batts/dense pack in most climates. The issue with this is that if anybody ever decides to replace the slate with shingles, it can create moisture issues.

    The safest in this case would be to spray foam. You can completely skip the SPF on the TPO roof if you bump up the rigid insulation there. Usually the upcharge for extra insulation for a flat roof is pretty small since the labour of installing it is the same. Roofing ISO is significantly cheaper than cc SPF per board feet.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |