Closed cell insulation on interior of old stone home – Firestop required?
We are renovating an old stone home, and on the advice of this group hired a building scientist who confirmed what many offered on this forum: best way to insulate was to use closed cell insulation applied directly to the interior stone in a continuous fashion from basement up to the roof (and in the roof rafters as well) as exterior application was not possible.
In order to facilitate that we reframed the interior walls off of the stone by 1-2” (varies given the stone undulations). Some of the LVL joists and steel beams for the floors still live in and on the stone at certain points but broadly the goal was to create a continuous barrier to the stone, keeping the interior air out of the cavity.
Here’s the issue: we are now getting guidance that a firestop may be required between the floors. This would require installing 5/8” plywood or other between the interior wall studs and the stone at each floor (basement to first; first to second; second to attic). If we do so, I’m concerned we are going to break that continuous closed cell barrier against the stone and allow air into the cavity during the winter (we are zone 6), hitting the cold stone (12-18” thick but still no R-value) and condensing, creating moisture, rot, etc. Or at a minimum have thermal bridging on that plywood layer from interior to exterior.
I’ve read some of the articles on the forum as well as other locations and it seems that there is some guidance that the firestop between floors is meant to be defined inside the studs if the cavity is filled, but I now can’t find that for confirmation. I was hoping to get some help/guidance that I might take to our building department for discussion. This type of construction is relatively unusual in our area.
It’s possible, as well, that I’m overthinking it and we shouldn’t be that worried about the 5/8” layer as long as we insulate above and below it inside the cavity.
Thanks for any help or guidance.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
This is the relevant section of the IRC: https://codes.iccsafe.org/content/IRC2021P1/chapter-3-building-planning#IRC2021P1_Pt03_Ch03_SecR302.11. It's not a crystal-clear situation and not one I've faced but you could just stuff those spaces with mineral wool and use a variable permeance membrane to control air and moisture.
R302.11.1.7 specifically allows mineral wool batts:
"Fireblocking shall consist of the follow materials.
...
7. Batts or blankets of mineral wool or glass fiber or other approved materials installed in such a manner as to be securely retained in place."
I’ve struggled as well to understand whether the closed cell insulation is fine to have continuous and meets the code. I’m wondering if there have been clarifications. The way the home is built we either have the ability to maintain that continuous closed cell against the stone (and it becomes the draft stop/firestop behind the studs or we have to have a break with a layer of plywood/other behind the studs. Not sure if that break is a big deal or not for condensation, etc.
I'd think that you'd want the floor to floor fire-stops installed, code required or not.
you could parge the face of stone flat with a type O mortar which would give you a consistent joint width which simplifies the issue.
if you have a building scientist onboard, then you might consider having them contact a fire protection manufacturer such as STI. https://www.stifirestop.com/solutions/wood-frame you might be looking for an Engineering Judgement to see whether their product would address the intent of the code regarding floor-to-floor fire protection in wood frame construction, [the manufacturer might have already run into this]
some combination of mineral wool and water-resistant smoke seal may be all you need, though the fire creates a pressure differential that the joint has to resist. in commercial construction, there is a metal plate on top of the fire-protected joint. your building scientist would need to weigh in whether a horizontal metal closer piece fastened to the top of the joist, set 1/4" from the face of stone and caulked to your stone would have any effect on condensation of within the assembly.
another alternative is to parge the masonry to a consistent width and then install a horizontal fire-rated expansion joint such as Willseal FR-H or EMSEAL DFR-2. these commercial products are designed for use with non-combustible substrates like concrete or masonry, but may satisfy the code official, particularly if they have ASTM E-84 testing or an Engineering Judgement. the install work would be trivial for a commercial caulking company.
How are the floor joists supported? Presumably they're already penetrating the foam? I don't see the fireblock being any different.
Also, I don't see 5/8" plywood as an approved material under R302.11.1.
The floor joists do go directly into the stone, but unlike the plywood fire block (I think it is 3/4” now that you flag), it penetrates the stone vs. resting close to it so is embedded. My worry is all the other articles about stone wall condensation in the winter if you don’t have a continuous closed cell barrier as I will likely end up with 3/4” or so that isn’t covered and thus exposed to the internal temps. Maybe that doesn’t matter that much, but the postings on this topic have me nervous so I thought I’d ask about just cutting back the area so the closed cell can be continuous on the stone wall.
What I would do is put a rectangle of pressure treated 3/4" plywood between the studs, one side of the rectangle is flush with the interior side of the studs, the other side it touching the stone, but since the stone isn't regular, only at the closest point. Fill between the plywood and wall. Technically this should be mortar. Fireblock foam would be easier, but it's only approved for use around "vents, pipes, ducts, cables and wires." You may want to ask your inspector if he would approve the use of foam, it's seems like a distinction without a difference in this case.
Then spray foam the whole thing. I don't see why the 3/4" discontinuity in the foam makes any difference, it's still going to be well sealed.
Agree with this. I'd even try to eliminate the discrete fire foam step if the inspector doesn't particularly care, since it's not magically better at this function than closed cell. (If anything, two part closed cell is denser and more robust than single part fire foam in a can.)
This whole exercise is somewhat theoretical in the first place, because if you have live fire in one of these bays, it means that all of that foam is actively burning (actually hard to do, because you would need a lot of heat and oxygen for it to get to this point) and the structure would have probably been very compromised by then anyway.
(Interesting video on trying to burn closed cell: https://www.youtube.com/watch?v=2ZBa1ijiHwY)
Generally speaking, I like 3/4" blocking in stud bay in old houses because you often are filling old stud bays that are different depth than today's nominal lumber sizes. And you can cut a lot of ~ 14 1/4" x 4" blocks out of random scraps of floor sheathing you may have around.
The fires I've read about with foam and fire spread don't have the fire spreading *through* the foam, they have the fire spreading in a vertical air channel *alongside* the foam. The textbook case of this was in the Grenfell Tower fire in London in the UK. The air gap acts like a chimney, helping the fire to rapidly rise through the assembly. The fire is on the surface of the foam here, not within the layer of foam. The purpose of fire blocking in vertical cavities is to limit the rise of hot gases and slow the spread of fire, it's not so much to be fireproof barrier.
Note that I'm talking about the theory here, not the specific code requirements. If your inspector specifically wants a fire block, you might try using "non com" (non combustible) 3/4" plywood here, which is treated with a fire retardant and will be dyed a reddish color. You usually have to get this at lumber yards, not box stores, and you typically have to buy it in full sheets. I could see embedding this into the foam a few inches, but not going all the way through to the masonry, which would act as a fire block to hot gasses. I don't know if that would necassarily satisfy your local inspector though. You may try talking to your local fire marshall (not the building department), and ask what they think. They may have some ideas for you that you could then run past the building department people. Ultimately, whatever you come up with will have to pass muster with your local inspectors.
Bill
R302.11.1.7 specifies "one thickness of 23/32-inch wood structural panel." "Wood structural panel" is plywood that is rated for structural use.
DC, that would include most 3/4" plywood too, and the non-com stuff is also one of those. It's just more fire resistant, which is why it might be a good option if inspectors are being extra picky. I don't normally see anything other than "regular" wood used for fire blocking in studbays though. The only oddball requirement I've seen for non-com plywood was for a roof. In the commercial world, it's most commonly used as a backboard for electrical stuff to mount on.
Bill
He's being extra-extra-picky if he doesn't accept what the code specifies.
I'd argue too that it's what the rest of the house is made of. It's already what's keeping the fire from spreading from floor to floor in most of the house.
> The air gap acts like a chimney, helping the fire to rapidly rise through the assembly. The fire is on the surface of the foam here, not within the layer of foam. The purpose of fire blocking in vertical cavities is to limit the rise of hot gases and slow the spread of fire, it's not so much to be fireproof barrier.
This is my point, though. :-) If the entire cavity is filled with closed cell spray foam, the condition for the closed cell burning uncontrolled doesn't exist, absent all the interior drywall having been destroyed (bad) or the exterior stone disappearing (more bad).
To wit, the foam *is* the functional fire blocking, even if the code doesn't prescriptively acknowledge it as such.