Thermal Break In Stud Walls
Hi,
Just found this site 2 months into a renovation of one part of our 45 year old home. We have taken the main floor of the house back to stud/rafters and are now at the insulation stage. The original plan was Roxul with a vapour barrier in the walls with blown in ceiling, but we decided to go with 2lb closed cell foam in the walls to try and close the gaps and make the space more airtight and comfortable.
The plan now is to put 4″ of 2lb closed cell in the 6″ wall cavities and 5″ of 2lb closed cell on the first two feet of our roof(over applied soffit venting) to help us attain the R60 we are going for there but simply cannot attain due to a lack of adequate space to blow it in there.
The question I have is there a way to create a thermal break between my drywall and my wooden stud walls? I have added at 2×2 on top of the existing 2×4 to give me a cavity large enough to spray in R24 or 4″ of 2lb closed cell foam. The exterior is brick so I simply have not way to do this externally.
I was wondering could I use something like Sill Gasket? cut into appropriate sized strips attached between the drywall and the stud? or should I have done this between the 2×4 and the 2×2 I added and there is no sense in doing it where I was considering?
Thanks for you help in advance. Looks like a great site, I’m off to do lots of reading as I have two more back to stud renos still to do in this house in the next two years.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
I should also note were upgrading the 4 windows in this area to triple pane awning style windows.
Did you already install the 2x2s?
Putting the 2x2s at 90 degrees to the studs would make some difference, although I don't know how significant. The difference would not show up with a simple calculation of whole wall R value because it is a three dimensional effect.
Using something other than wood to extend the studs, like rigid foam, would be more effective.
Yes, the 2x2's are already installed on top of the 2x4's. I know its a little late to the game, just wondering if the sill gasket would work between them and the drywall, or if the difference would be negligible.
Using closed cell foam where it is thermally bridged by framing is usually a waste of good foam.
If you're putting 2x2s on to 2x4s that's 1.5" for a milled 2x2, and 3.5" for milled 2x4 for a total depth of 5", not 6". Assuming a 5" total stud depth:
When you put 4" of closed cell foam in a 5" wall cavity you have a nice center-cavity R, but the 25% framing fraction has shrunk to about R4.8 from about R6, and the framing is conducting more heat than the insulation. If you filled the whole thing up with open cell foam you'd have comparable (slightly higher) performance, but at a much lower net cost. The "whole wall" R of the 4" closed cell solution comes in at about R13.9, after thermal bridging, whereas the open cell solution comes in at R14 whole-wall.
If you take Reid Baldwin's advice and peel off the 2x2s and re-installed them perpendicular to the studs and filled the whole thing with open cell foam you'd be getting a ~R5.5 thermal break over much of the 2x4s and you'd have a whole-wall R north of R16.
If you're going to leave the 2x2s as-is, the R0.25 you get out of a sill gasket isn't worth the trouble. But scrapping the closed cell foam concept and going with a full fill of open cell foam is well worth it. It's higher performance, and unlike 4" of closed cell, it can be installed in one pass. (Closed cell foam needs to be installed in 2" lifts with a cooling period between, or it's a fire hazard in the 24 hours following as it cures, and it has more problems with adhesion, cracking and sinking.)
In colder climates depending on sheathing & siding type it's usually necessary to use a smart vapor retarder (MemBrain , Intello Plus) between the wallboard and the open cell foam. MemBrain is about 12-13 cents per square foot, Intello more like 50 cents, but either will work. It's more effective than the typically 4 or 6 mil polyethylene vapor barrier, since it becomes vapor open if moisture ever accumulates in the assembly, enhancing the ability to dry toward the interior. It's a cost adder that wouldn't be necessary with 4" of closed cell foam (which is highly vapor retardent), but it's still a lot cheaper than the cost difference between open cell foam and close cell foam.
Cheaper still would be to compress R23 rock wool batts designed for 5.5" deep 2x6 framing, which will perform at about R21 center cavity. Or, you could add edge strips of 1/2" rigid polyiso and get the full R23 out of the batts with R3 of thermal break on the framing. To do batts correctly you'd need to air seal the framing to the sheathing inside every stud bay, and install the batts to near-perfection, but with the R3 polyiso edge strips it would end up at about R16 whole-wall. (You'd still need the smart vapor retarder, of course.)
To explain why the sill gasket won't work, the concept of a "thermal break" as a thin strip of an insulating material is only really valid in one specific context, thin pieces of metal (particularly aluminum) with significant exposed surface area on both sides.
With a flat plane, pretty much the worst possible assembly R-Value is roughly R-1.4, thanks to the conductivity of air & air films on either side (the standard U factor for single pane windows is worse than this, but it assumes a 35mph wind on one side). Add your R-0.25 sill gasket to that, and now you're up to R-1.6. Even the worst possible wall (with 2-dimensional heat flow) only gets a 15% performance improvement from this so-called "thermal break".
But materials with high thermal conductivity can "cheat" the simple 2D heat flow system by exposing more surface area to the air, reducing the effect of the the air film. The aluminum passing through an assembly might be only 1/20th of an inch thick, but with 5 inches surface area exposed on either side; that 0.05" section acts like it only has an ~R-0.014 air film, with the aluminum itself only adding another R-0.008 per inch. Adding a tiny strip of R-0.25 in the middle of an effective R-0.025 now gives us a miraculous 1000% improvement!
But wood framing doesn't have high enough conductivity to use the surface area "cheat" to any significant extent, so we can't cheat with a tiny "thermal break" either. When we talk about a thermal break in a stud wall, we just mean that we want to make sure that heat has to travel through a bunch of insulation no matter what path it takes, and there's no way to do that without a bunch of insulation.
Very nice explanation Tim!
OK, thanks Tim and Dana, I appreciate the detailed responses. The closed cell is ordered and a deposit already made, I called, the don't do open cell. I guess I am all out of options other than disassemble the wall and move the 2x2's to run horizontal to the 2x4's.
Google "Mooney Wall"
How much do they ding you on the deposit if you cancel the order on the closed cell foam?
It might still be cheaper (and a lot nicer to the planet) to go with open cell foam + smart vapor retarder.
You might find this document a useful read:
http://apps1.eere.energy.gov/buildings/publications/pdfs/building_america/high-r_value_walls_case_study_2011.pdf
Case 3 is a 2x6 wall with 2x3s horizontal to the framing. Yours being a 2x4 + 2x2 it'll come in lower performance than that, but Table 3 lays out the whole-wall R of a number of different wall assembies & types.
No useful information to add but I'm adding my encouragement to at least ask how much of your deposit you can get back if you cancel the order. You may find that they are willing to return most or even all of it.
Since you have two more renos to do, finding a better system is important whether you can recover your deposit or not.
When you are trying to minimize flow through a system (like a wall) that has good parts (cavity insulation) in parallel with bad parts (studs), the performance of the systems is driven by how bad the bad parts are. Improving the good parts helps little. It is better to spend your money making the bad parts less bad. In terms of air tightness, the bad parts are the joints, so you started down the right path - sealing the bad parts. However, you could have accomplished that with only a couple inches of closed cell foam or with other techniques altogether. The remaining inches of closed cell foam only served to make the R-value of the good part better which doesn't help the system much. On your remaining renos, if you are willing to have the wall be wider, make sure you use that width to improve the R-value of the bad parts (the studs).
The deposit was 25% of the total of the job. Not an amount I would like to give up. The Mooney wall looks like a great solution, if I was coming into this at a different stage of the project, perhaps when we do the next floor as we have a second story and a basement still to do. I see there is not the easy fix for thermal bridging I had hoped for. We will just have to live with what we have planned, and take comfort in the improvements we have made over a space that had no insulation in about half the cavities in the wall when we took down the plaster.