Replies

1. 
   mr_reference_Hugh | Nov 02, 2022 05:14am | #1

   DJK447,

   You do in fact need to be very careful when modifying a roof assembly that has performed well for a very long time. Many of the older assemblies perform well because they are drafty and allow for lots of air circulation. The air movement has typically removed any humidity or dried any water infiltration from occasional strong winds during a rain storm.

   Your question

   i) would there be any benefit / tradeoff to going with mineral wool batts in the sloped section of the roof?
   Reply: I like both. Mineral wool does not absorb water to the same degree as cellulose. On the other hand, dense packed cellulose can reduce the air movement from the living space to the roof deck. I prefer other solution for air-sealing (see below) so the advantage of cellulose is not as important if you have a different air-sealing strategy. Cellulose captures carbon while mineral wool batts require lots of carbon to be produced. Otherwise, I don't think there is much difference. If you think that mineral is better to detect leaks, go with that.

   ii) can have the roofer install a couple of roof vents at the same time,
   Reply: I talk about ventilation below, and my view is that things need to dry. If you have not air-sealed the bottom of the "triangle", then you will get moisture coming from the living space below.
   CAVEAT: I mention below n article with at link that explain that you need to air-seal your ceilings if you add ventilation or else you will suck more air from the inside of your house/basement and that causes other issues.

   iii) questions / answers where people recommend installing baffles between the roof and insulation to allow airflow along the roof line even though it’s an unvented roof
   Reply: This would be a really good idea. The article I mention from Martin H. also recommends this for slate roofs. This ventilation would still be important even if you don't add ventilation at the roof peak (as discussed below). The importance of that gap is discussed in different ways in this (rather long) article from Building Science Corp.
   https://www.buildingscience.com/documents/insights/bsi-038-mind-the-gap-eh

   ********Additional analysis below*********************

   I found this article from Martin Holladay from 2011
   https://www.greenbuildingadvisor.com/question/proper-way-to-insulate-an-unvented-slate-roof

   1 - Spray foam is like tar - it is a "no-no".
   > "If you want to insulate the sloping roof, you definitely don't want to install spray foam against the back side of the slates."

   2. Venting channels under roof deck
   Both articles recommend a venting channel.
   > "create a vent channel under the slates"

   3. Adding venting to increase air movement to the outside - there are different opinions:

   There is a general rule that if something cannot dry, it is going to rot. I believe that the Building Science Corp guru is quoted as saying "If it cannot dry, it is going to die."

   In modifying the amount of air movement in the assembly with the changes contemplated, it would make a lot of sense to me to add venting. BUT - all this costs money and we all have a budget. First, look at two different professional opinions, then what I think.

   Michael Maines says: "Elsewhere in the attic provide venting equivalent to at least 1/150 of the floor area. It can be a combination of ridge and gable vents. If your house is 24' x 36', you should have at least 5.76 ft² in vent area."
   https://www.greenbuildingadvisor.com/question/attic-insulation-slate-roof-double-brick-walls-no-soffit-vents (the article you referenced).

   Martin H. indicates that you don't need to add venting. He says "Slate roofing is far from airtight. Such roofs are ventilated to a high degree, even without soffit vents or ridge vents, because of the cracks between the slates. The driving forces are sunlight (which creates temperature differences) and wind."

   I don't know how much air actually moves between the slate, through the paper, and through the wood decking. There is some, but is it enough to keep the roof in the same condition you found it. If you have a budget for it, adding venting could be a good investment. The article from Building Science Corp above appears to suggest that the most important thing is the gap behind the sheeting (in this case roof deck). (I am contradicting Martin H??? and Building Science Corp???)

   - Venting at the peak
   I don't know whether you have a "triangle" at the top. If you do, you have the option of adding gable vents in that upper "triangle" or a ridge vent. If you don't have a "triangle", then you would only have the option of adding a ridge vent.

   - Venting from the lower part of the roof
   If you have a vent at the top, you would need to have air coming up from the bottom of the roof. This could come from a gable vent in each lower "triangle". You could also look at adding soffit vents, but I don't know the exact construction of you roof.

   4 - Air Sealing to prevent air moving from the interior

   This is a good brief article.
   https://www.greenbuildingadvisor.com/article/air-leaks-from-your-home-to-your-attic-need-to-be-sealed
   The article makes it clear that you need to air seal between living space, more-so if you are adding ventilation to the attic. So adding ventilation, as you might choose to do, means you would also need to deal with air sealing.

   A) Opinions
   As explained in the article linked above, the infiltration of air from the living space carries lots of moisture/water vapours. You want to avoid this because the moisture can get trapped in the assembly and/or create condensation under the roof deck. This gets to be critical with an unvented assembly, but still standard practice with vented roofs. In my opinion, this was less important when the roof was not insulated and naturally
   vented with air freely moving through the assembly - and you can tell because your roof is still in good condition.

   Michael Maines suggests "While you're up there, air-seal any penetrations between the living space and the attic."

   Martin Holladay does not mention air sealing in his 2011 reply but he does suggest "Once the rafter bays are filled, you can install rigid foam under the rafters to address thermal bridging." In effect, that rigid insulation, as long as it is well taped, will be a very good means of air sealing. I would interpret Martin's suggestion to deal with air movement from living spaces and not just to address thermal bridging.

   B) - Rigid foam for air sealing

   Martin Holladay suggested using rigid foam under the rafters. That makes sense if the underside of the rafters are open.

   Michael Maines suggest "air-seal any penetrations".

   If you read GBA articles, you will often see a recommendation to use a rigid material like rigid foam taped/sealed at the seams and edges to air seal. Air-sealing any penetrations is important but it is difficult to properly air-seal the larger surfaces of the ceiling without using a rigid air-tight material.

   This video explains how to install rigid foam to air seal the some parts of the lower "triangle":
   https://www.greenbuildingadvisor.com/article/how-to-air-seal-an-attic-kneewall
   This video does not mention air sealing the bottom of the air triangle, which is really puzzling to me. Ideally, the bottom of the triangle (the ceiling of the living space below) would need the same air sealing for the same reasons the other parts need air sealing.

   The video does mention why it is ideal to use a rigid insulation board instead of flexible material like house wrap.

   Hope this helps.

2. 
   [email protected] | Nov 02, 2022 09:13am | #2

   Thank you for the detailed response!! Some very helpful stuff there.

   > I found this article from Martin Holladay from 2011
   https://www.greenbuildingadvisor.com/question/proper-way-to-insulate-an-unvented-slate-roof

   I found that one too, unfortunately, he seems to be discussing a different type of slate roof where there are big gaps between the wood slats and the backside of the slates is open to the attic? (Or at least I've seen him discuss that in other places and given his concern about the spray foam sticking to the back side of the slate).

   The venting we are considering is a couple of box vents near the ridge line, a ridge vent would be too expensive, but a couple box vents aren't crazy.

   I've been getting conflicting opinions about the utility of adding the air channel in the rafter bays of the roof. At first I thought it would be necessary, but then, watching this: https://www.greenbuildingadvisor.com/article/the-bs-beer-show-unvented-roofs-and-fluffy-insulation it seems like the air gap at the back of an unvented roof can be really *harmful* and like you're creating the situation where the insulation isn't up against the roof and potentially causing problems. But I'm not sure how exactly to evaluate.

   And the insulation folks mostly seem to think that it wouldn't help anything without a real soffit vent at the bottom of the roof, which is impossible for us because the bottom of the roof is the neighbor's roof because we're a townhouse.

   But I agree, I'm not sure how much the tar paper and wood is actually keeping air movement from happening, with that said, with something like mineral wool, which allows air and vapor movement, would the gap be all that different? It seems like with most unvented roofs the damage is at the very top, so the fact that that's open should help?

3. 
   mr_reference_Hugh | Nov 02, 2022 10:03am | #3

   Do you have a “triangle” at the top (I.e. at the peak or only at the bottom?

   Can you attach a hand drawn sketch of the roof profile?

1. 
   [email protected] | Nov 02, 2022 10:18am | #4

   It looks like this screenshot from the video, and we're doing the warm storage option where we insulate in the sloped sides of the closets.

   (so yes, we do have the triangle at the top as well!)

   
1. 
   mr_reference_Hugh | Nov 03, 2022 02:32am | #7

   Djk447,

   The solution in the video to fill the rafters with cellulose also assumes that you have soffit venting, plus ridge venting. You have neither but I will share what I would do. The Building Science Corp guy in the video said the assembly he spoke about has not tested and that the occupants would need to keep the interior humidity levels "low" for the life of the building.

   In the initial posted question, the warm storage was not mentioned so was not accounted for. It was mentioned that there was no soffit ventilation but was not clear that none could be added. So now I take that into account.

   You will see in the attached sketch what I would do myself and I can explain.

   Air sealing
   1. Air seal and seal well. Using a rigid material like rigid foam (e.g. foil faced polyiso) would be my choice. That would be covered with an appropriate drywall - likely type x mold resistant green board as a personal preference. This would be installed on the inside. The type x is likely not required by code but it offers protection against fire and I like that myself.

   New triangle at eaves
   2. You see that I would add a smaller triangle near the eaves. I would need a space to introduce outside air like what I explain in point number 5.

   I would definitely add some ventilation.
   3. I would add the box vents at the top (as you are thinking) near the ridge in the top triangle. This would allow moisture to escape.

   4. I would have a gap between the insulation and the roof deck that could be created with manufactured baffles or using plywood. This would connect the new very small triangle (see sketch) at the eaves to the existing triangle at the top, which is vented through box vents.

   5. If I don't have soffit vents then maybe the only solution is to "pipe-in" air by installing insulated duct from the outside into a "new" very small triangle at the eaves. That vent could be place almost anywhere around the house where there is access to fresh air. It would even be possible to have a fan (continuous run/humid space rated) to move air from the outside. I would not want too much air because the pressure in the attic would likely pus air into the living space. I would box the ducts in with a small narrow bulkhead. I would place this right near an interior wall to have it blend in more.

   Vapor retarder
   6. I am not certain whether you need a vapor retarder but your building code would tell you. Regardless I would use a rigid material to create the air barrier (see item 1 above). This material would create a vapor retarded - whether it is needed or not.

   Insulation
   7. All things considered, I would use dense packed cellulose under as the insulation under the baffles and fluffy cellulose in the top triangle, but not so much to block the vents or the baffles.

   One advantage of cellulose is that it can act as a sponge to hold water vapour and then release it. Mineral wool does not have that property. This was mentioned in the video. It is what I have in my own walls and for that exact reason. In one study I read, I was surprised to see that the wood framing was much more "wet" in an assembly that had mineral wool than the one that had cellulose. That is because the cellulose was absorbing some of the water vapors - basically sharing the water load with the wood frame. The mineral wool was not absorbing any water vapors so the wood frame had to take all the water vapors alone - more risk of mold.

   That is what I would do. Happy to answer questions if you wanted clarification on any points.

   
4. 
   walta100 | Nov 02, 2022 12:07pm | #5

   Half story homes like yours tend to be very leaky before you insulate you need to air seal your home in order to keep the conditioned air inside.

   I hate spray foam but one of the few times it is worth the cost and risk is a house like yours that has a half story of living space in the attic.

   I say find and read the articles about conditioned attics make peace with the idea that you are going to spend the money on equipment and fuel to heat and cool the attic.

   Walta

5. 
   mr_reference_Hugh | Nov 02, 2022 12:15pm | #6

   I will look at it this evening.