Insulating ceiling with a vented cathedral ceiling
Hey all, I’ve read a few GBA articles on the topic, and end with just a few questions that I was hoping for some advice on!
I’m converting a shed with a vented cathedral ceiling to a fulltime house. This upstairs will be the bedroom and bathroom, finished with TG pine ceiling.
There is an underlayment below the roofing material as you can see in the photos, and vented at the ridge from the soffits.
I was planning to install:
– Plastic air channel baffles, and to make sure that is sealed airtight leaving about a 1″ air gap
– About R30 of mineral wool*
– Polyiso boards attached to the bottom of the rafters and tape sealed
– TG pine ceiling
There are three main questions I have:
1. The mineral wool bats fit in between the rafters, but won’t there be about 1″ between each bat where the rafter space is open, how can I get that filled if I need to?
2. The depth of the rafters changes from about 10″ at the wall to over 16″ at the ridge. I was planning to layer 3.5″ thick R15 mineral wool batts so that where there is sufficient depth, I can just add another batt, which would mean the ridge will be more insulated than the sides. I was guessing that because heat should want to find the highest ceiling point that wouldn’t be a mistake to do?
3. There are some random 2x4s nailed down as you can see, those aren’t structural are they, in that, I could remove them to make sure everything fits better?
Any other reactions? Thank you all!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
For sealed baffles you need them to be permeable so standard plastic ones don't work.
High density batts such as mineral wool don't suffer from wind washing, so you can skip the baffles except at the eaves intakes.
Trusses are best insulated with loose fill. If this won't work, you can get one layer of batts the same thickness as the bottom chord.
For the layer above, get batts for steel studs which are wider and will sit over the framing. You'll still have to do some trimming around the webs.
You can also cut and cobble rigid between the truss webs and insulate with a single layer of batts.
The random 2x4s are for bracing the trusses. Since your ceiling will not be directly applied to the bottom of the trusses, you'll have to come up some other form of bracing if you want to remove those.
I agree with what Akos says. I'd fill the truss web spaces first with pieces of foam or mineral wool board cut to fit. Focus on making the ceiling airtight rather than worrying about the baffles.
You could move the truss braces from the top side of the bottom truss chords to the bottom side. It will be easier to cut the polyiso to fit between them than trying to cut and fit batts around them.
On your question 2, putting more insulation where you have the space for it is a good idea.
Ok thank you both for the information. I'll look into some blown-in options and see what the pricing is there, in addition to thinking about trying to fill the voids with more material if I go with batts.
I was planning on there being two air/moisture barriers, one directly above the ceiling material, and then one above the batts, basically to create the bottom or the air channel. Akos, you're saying the top barrier isn't as important/necessary?
Thank you!
Correct. A house only needs one solid air barrier. That would be the taped rigid on the inside. Outside of that your only worry is wind washing which is not an issue with high density batts.
The 2nd air barrier doesn't hurt but it needs to be something permeable (house wrap, fiberboard, permeable rigid).
Akos, is the notion that high density batts would be effectively an air barrier at that length and density? If so, then is there a risk from condensation at the underside of the roof deck with its vapor barrier (foil)?
Or is the taped rigid Insulation going to be enough of an air and vapor barrier to prevent any vapor from getting to the roof surface?
High density batts are still very much air permeable, so you must have an air barrier elsewhere. The taped rigid under the rafters is the air and vapor barrier and as long as it is detailed properly, it is very effective.
The OP is referencing a vented roof, so as long as air and moisture transferee is limited, the venting will take care of any condensation that makes its way above the fluffy. The fact there is a foil vapor barrier under the roof deck doesn't matter as no moisture needs to move through it.
Ah... So high density batts still maintain their entire r-value when air moves through them. That's what I understand of your claim that wind washing won't have any effect on them. Did I get this right?
I would expect significant air movement from eave to ridge due to solar gain on roof. Wind will also induce air movement but in ways that are less scrutable to me. It would be good to know these effects will have no impact on the batts while still maintaining the vapor/moisture transport out of the assembly.
Malcolm posted a great study which demonstrates this with data!!! Thanks Malcolm... Very interesting...
In a vented roof you want the airflow to move over the batts not through it. This is why you still want baffles by the eaves.
If the airflow actually moved through the batts (this is often the case of very leaky walls with low density fiberglass batts), the effective R value of the batts is pretty much zero.
Air velocity in a vented roof is not as much as you might think, it is low enough that it is hard to measure (at least on a still day).
The extra bit of fluffy in the high density batts prevents this bit of air flow from getting inside the top of the fibers and reducing the R value of the batts. This is why the baffles are not really needed.
I'm squinting at the study that Malcolm posted, and at least in the wind washed wall scenario that they mocked up, there was significant reduction in r-value for low density fiberglass batts, but much less (negligible) for mineral wool batts, which are also much higher density. This speaks to a more generalized resilience to wind washing though the paper doesn't speak to the difference between low density and higher density insulation materials.
Regardless of the building science, this assembly is not something that I could ever use or recommend in my projects since the building departments are pretty diligent about observing code required venting. There was another thread on this recently so I won't get into that again here
Luke,
Our code up here in Canada acknowledges you might not be using baffles by making the minimum vent gap 2 1/2"
I realize now that I had judged the rafter depth in the pictures to be a bit less than the OP describes...
(Facepalm)
If my questions seemed a bit daft, well... There's the reason
(Sigh)
Alex,
Not exactly the same situation, but tables #8 and #9 in this link give you an idea of what reduction in R value wind-washing causes with various types of insulation.
https://www.rdh.com/wp-content/uploads/2017/10/Van-Straaten-Windwashing.pdf
This is helpful to see you all discuss. @Akos, wanting to make sure I'm following right:
The idea is that we need the strong vapor/air barrier under the batts (sealed polyiso boards fastened under the rafters) which should stop moisture from coming from the interior.
Then from the soffits, I would just need to direct the air upward towards the ridge vent with something like this: https://www.homedepot.com/p/Amerimax-Home-Products-Accuvent-22-5-in-x-1-5-in-Black-Soffit-Insulation-Baffle-ACCUVENT/202962730. Because the mineral wool is dense enough, a strict barrier isn't required on top of it to keep the air from running through it? Or I could do additional baffles that keep the air moving upward but those don't or shouldn't be airtight?