GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X 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

double-stud wall assembly: dense-packed outer, rockwool batts gap?

buildzilla | Posted in Energy Efficiency and Durability on

talking with an insulation contractor this morning, he was warning about labor costs associated with dense-pack-cellulose especially in double-stud walls because of the extra labor required to place webbing between cavities.

he even suggested just using rockwool batts for entire wall.

i’m thinking that cellulose will be better at dispersing the inevitable moisture that finds its way into a double-stud assembly during the cold season, but wondering if dense-pack-cellulose in an outer (load-bearing) 2×6 wall in combination with rockwool batts in a 3″ gap would be a viable strategy in that it skirts the need for webbing between cavities, and perhaps even address fire concerns at the top which i’m assuming would normally require sheathing gusset blocking the gap at the top of the assembly.

just throwing it out there for thoughts.

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


    That's a very interesting question. We know double walls insulated with cellulose do well. We know they also benefit from good air-sealing, and using a rain-screen gap behind the cladding. Which of those you can remove, or how much of each is necessary is something I'll be interested to hear some discussion on.

    My own intuition is that what you propose will work fine - but that's based on no data at all.

  2. Tim_O | | #2

    Is it common to use web between every stud bay when dense packing? I thought some did that, but I guess I was thinking it was less common.

  3. maine_tyler | | #3

    Is this contractor the one that would be doing the dense packed? Because, like Tim, I'm not sure every installer needs the webbing between walls.

    I have a double stuff wall packed without webbing. The lack of webbing didn't seem to bother the installer-- the much bigger problem with my wall was the use of intello (exterior of interior wall). They had to slit the intello a bunch to get proper air flow and it stretch WAY more than I would have ever imagined under the pressure, making the subsequent installation of batts on the interior a huge pain. The installer said insulweb would have been much stiffer and easier to blow into. But without dividers, it still will 'fetch up.'

    1. Tim_O | | #4

      Out of curiosity, would you do the Intello again as your dense pack support or would you do the webbing and add a different vapor/air barrier later?

      1. maine_tyler | | #9

        I don't think I would build it the same. But I imagine it depends on familiarity of all trades to the build, and perhaps some dense packers would find this set up completely standard.

        My intello did sit around on the wall for awhile so I don't know if that played a role in the stretch (it was quite tight when first installled, but I noticed it loosen in the heat and tighten again in cold.)

        I also had to go around cleaning dust off all the injection holes and slits and tape them. Not a huge deal but another step. I would be curious what those who build these walls all the time do, specifically in my area. Someone like Dan Kolbert or Emerald builders for Maine. I could be wrong, but I don't think they use a mid-wall membrane/barrier, which would allow the application of intello after the fact (my preference if done again).

    2. buildzilla | | #15

      true, i've heard of some installers wanting webbing per cavity and others not, i'm pretty sure i've even read of some webbing every other cavity.

      i was thinking of using intello as well behind my inner wall, but if i was gonna dense pack the gap too, i was thinking that it would be reinforced by webbing as i have no data on intello ability to handle dense-pack.

      i guess in the scenario with dense-pack in the outer and rockwool in the gap, it would be webbing over the 2x6, then batts, then intello on the outside of the inner 2x4 wall.

    3. buildzilla | | #20

      fwiw, this article recommends staples every 4"-6" on the intello to keep it taut:

      i can see the airflow of intello being very different from insulweb, i'm imagining slits cut in the membrane, but then cellulose blowing out of the slits, sounds tricky.

      perhaps another upside of filling an insulweb-inside outer-wall with dense-pack, and then stacking rockwool batts (horizontally?) for the gap, then throwing an intello-outside inner-wall on top of that...

  4. kbentley57 | | #5

    For what it's worth, by the time you start talking about mineral wool batts, you're probably going to be better off looking at an alternative wall system. The double stud wall works because cellulose is cheap. The fact that it buffers moisture is almost inconsequential to its use here.

    You didn't state how thick of a wall you wanted, but I'm assuming it's more than what a standard 2x6 wall + 2" of exterior insulation could provide. If I had to go down the batt route, I'd be looking at a 2x8 wall + external or something like that. Fiberglass on the inside and rigid or mineral wool on the outside. That gives you the possibilities of up to an R40 wall, which ought to be good enough for pretty much anywhere. It's also easy to build, easy to source, and wont prove any more difficult than a normal stud wall, except for the weight increase.

    In some sense, it's your double stud wall, smooshed back together.

    1. buildzilla | | #6

      i was planning on 2x6 + 3" + 2x4, but i was gonna apply something like intello to the outside of the inner 2x4 and use it for a service-cavity.

      even without insulating the service-cavity, i would be at code which is 20+12c.

      looks like r13 mineral-wool as about 1.20 sf, so roughly comparable to what i was quoted for dense-pack.

      my other issue with exterior-insulation is in addition to direct labor and material, it gets into in-direct labor and material associated with window-bucks, long-screws, rain-screen, etc.

      in my particular case, i'm looking at everlast siding which has inherent rain-screen when applied directly to sheathing, but if applied to something like comfortboard, would require strapping for rain-screen.

      so, for my case, i'm not convinced that trading 3" of dense-pack for 3.5" of rockwool batts tips the scales to the external-insulation assembly.

      1. kbentley57 | | #8

        Food for more thought:

        You can buy R-30C fiberglass batts that are 8 1/4” thick or so, which would give you about R27 when compressed into the 7.25” space. That cost is about $1.25 per sq ft.

        That would give you an effective R24 or so wall. Add 1” of polyiso and you could direct attach the siding through it and be around R30 effective with a pretty simple setup.

        1. buildzilla | | #13

          that could work as well, polyiso is dense enough to provide a good plane for the siding. my concern would be dew-point in the fiberglass and resultant moisture dissipation in the spring. i don't want to be the guinea-pig, but would be interested in the results of a study. i feel like there are many untapped innovative ideas like this out there that could present positive value-engineering outcomes.

      2. maine_tyler | | #10

        "i was gonna apply something like intello to the outside of the inner 2x4 and use it for a service-cavity."

        I'm not sure how you would divide the bays with this set up. It's why I had no choice but to rely on my installer letting the cellulose fetch up. Best to talk to the specific installer.

        1. buildzilla | | #16

          good point, no way to web across cavities with membrane on outside of inner wall!

          i guess that's gonna be an issue with the service-cavity strategy with or without intello, in order to web between the walls, they need to web between, then web on inside of inner wall, so no service-cavity possible.

          did your installer have an issue with intello being less transparent than web so they missed out on that visual aspect?

          1. maine_tyler | | #18

            They didn't mention an issue with visibility. It seemed like they did a really good job with what they were working with. Knew what they were doing. But as stated earlier, they did need to slit the corners to encourage airflow. The intello is definitely rated for dense packing, but in my particular situation it did stretch a good deal making fitting batts a pain. I probably wouldn't do it the same again.

            I also really like/liked the idea of a servi e cavity, but it's worth considering how many penetrations you would end up actually having. With a little planning around the electrical, it doesn't seem like a total game changer. But do plan ahead on the electrical if not using a service cavity (airtight boxes, etc).

  5. Jeremiah_Sommer | | #7

    What about a hybrid approach of sorts. We are in pre-con on a project where we have specified a staggered stud double wall. The assembly goes like this from exterior to interior...

    1x4 furring rainscreen
    WRB system Solitex Mento 1000
    1/2 ply sheathing
    2x8 @ 24"oc load bearing wall
    Rockwool batts or dense pack fibreglass BIBS
    Intello vapour retarder
    *2x4 @ 24"oc service cavity wall with studs offset from structural wall
    Rockwool or BIBS

    What we like about this wall is that we can eliminate the labour associated with exterior insulation and window bucks etc. We are currently leaning toward using RW batts in the 2x8 wall since there will little to no services in it and hand batting will be efficient. Intello will be a very simple install over the int face of the 2x8 wall and then we simply build our service wall in front of that. BIBS may be the preferred choice for the SC wall since we are fortunate to have excellent local installer and the system is great for getting around wiring etc.

    *We may nerd out a little more and use 2x3 for our top and bottom plating and place a 1" strip of EPS or polyiso between our 2x8 and 2x3 plates. Not sure it will make a noticeable difference with thermal bridging but simple and worth a try.

    1. Expert Member
      MALCOLM TAYLOR | | #11


      Which brings us back to the original question of how much these double-wall assemblies rely on the moisture buffering of cellulose for their safety?

      1. Jeremiah_Sommer | | #12

        I agree that cellulose can be an important component in a deep wall assembly as a moisture buffer. However I do think a well thought out assembly that includes a rainscreen and a stout air sealing strategy may rank slightly higher than the insulation material especially if fibreglass is used. One of the reasons I like the idea of placing the air/vapour control layer roughly 1/3 of way into wall is to promote a little more drying to the interior which should lower the moisture burden of the deepest part of the assembly a little.

        1. Expert Member
          MALCOLM TAYLOR | | #19


          As I said in my first post, I suspect you are right - but that's based on nothing concrete at all. We have a three legged stool of things we know help make the wall safe. That we don't know is if cutting one off makes the stool tip over.

  6. maine_tyler | | #14

    How does having the air/vapor control 1/3 of the way in (from the interior) promote more inward drying? Do you just mean that the inner third of the wall would essentially be in open communication with the interior so would have no throttle? I don't see how it would improve drying of the outer two-thirds of the wall though. And in what season are you thinking; summer?

    I believe most if not all smart retarders open up in higher RH conditions, so wouldn't it do sort of the opposite? The inner most surface of the wall assembly would have lowest RH in winter (most throttling) and highest RH during air conditioning (most open).

    Don't get me wrong, I like that location as well, but i thought of it as being mostly for physical protection.

    1. Jeremiah_Sommer | | #17

      Yes, I was suggesting that inner 1/3 of the wall would be essentially a direct connection to the conditioned space which would allow it to dry inward without much resistance. This wouldn’t necessarily increase inward drying of the exterior (although a smart VR would keep the option open) wall but could reduce the amount moisture that the outer 2/3 has to deal with.

      However as you mentioned, the primary benefit of a mid-wall VR is for its protection which is key to a stout air/vapour control strategy.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |