Update on Vented, Vaulted, Closed-Cell Roof
A few months ago I posted about the insulation in my home addition.
Original post here:
https://www.greenbuildingadvisor.com/question/vaulted-hip-roof-venting-strategy
After some discussion and reading, I opted to follow the plan of installing site-built baffles (2 inch depth), followed by approximately 6 inches of CCF (Zone 5). This is a vaulted hip roof, so I opted for Cobra hip vents along the hip and ridge vent at the peak.
The project just had drywall finished about a week ago and the furnace was turned on yesterday. The heat in the project rose quickly and the inside was very comfortable.
This morning, I noticed a lot of water on the underside of the gutters and soffit. It froze overnight and this was during the morning thaw. My initial thought was that maybe some improper drip edge/flashing was leading to the issue, but my heart sank when I saw water visibly dripping out of a soffit vent hole nearest the house. A scope camera confirmed there are water drips from the roof sheathing across the soffit, and beads of water accumulating on the sheathing near and upward from the eave. I can see water tracks running down the baffles, which I made with 1/4″ Hardboard.
This feels like a bit of a disaster. I imagine, despite the best of my efforts, the ventilation is inadequate. I did this entirely to avoid a hot roof situation, but may have created a larger problem for myself. I’m now extremely stressed and wondering how I can retroactively completely insulate the rafter bays without a complete roof and/or drywall CCF removal… I’ve turned the furnace off, and I’m prepared to keep the project at ambient temps until I have a solution.
Any ideas/thoughts are beyond welcomed at this moment.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Is the addition still closed off from the occupied part of the house?
If your local weather has been consistently cold enough, it's possible that some frost had formed on the underside of the roof sheathing. When you turned on the furnace, enough heat entered the vent space to melt it, or maybe you just had a day of warmer weather. If you installed the baffles yourself, do you recall seeing any frost? I think you should keep the heat on and just continue to monitor the roof for the next few days. The heat should help dry it out.
I installed the baffles well before frost set in. Insulation went on about a month ago, and then temporary heat (electric and propane) was used for about 2-3 weeks during drywall.
Understand if your propane heater is unvented, it will release a lot of water vapor into your building about 7 pounds per gallon so almost gallon per gallon.
Walta
Do you think it's worth running the furnace for a few more days, even though it seems to be the heat in the home causing the condensation in the baffles space? I scoped it again this morning and ice all along the deck.
There should be no difference between your assembly and a more conventional vented roof using fluffy insulation. What have the weather conditions been like there over the last few days and nights? Do you mean that your vent spaces are 2"? In your other post, it sounds like some of the bays don't extend to the mini-attic; do the areas that are dripping have any correlation to those bays? Have you been heating the interior with an unvented propane or diesel heater, and/or did you pour new concrete or use a lot of wet framing? Or did it rain a lot during construction?
Temps have been colder. 18 yesterday morning and today. You can visibly see on the soffits and gutters where the condensation is the worst, but not clear why it's worse in some locations.
There's not a mini attic, unless you're referring to the baffles themselves?
I did heat the interior with an unvented propane heater several times over the past few weeks. Now the furnace is up and running (off now). Even with the propane, how would that make it's way to the roof deck through 6 inches of CCF? Thanks for weighing in.
Edit: Two of the more frosted areas are right below plumping vent (plumbing drain and furnace exhaust) penetrations. Other heavily frosted areas have no penetrations nearby.
Deleted
I have a feeling you have some big air leaks somewhere. Usually with additions the connection between the old and new is a very hard detail especially if you are talking about cathedral ceilings.
A quick first check would be to get a box fan plus some cardboard and mount it into a window. If you can heat the place, set the fan to blow out thus depressurizing the space. In cold weather, if there are major air leaks, you will feel it right away. If there are major air leaks, you need to fix those first as no amount of venting will get you out of trouble.
With new construction especially after propane heater, you'll have very high interior moisture. You'll probably need to run a dehumidifier even during the winter to remove that.
The source of condensation is warm, humid interior air getting against the sheathing. The source of the humidity is the occupants of the house. If the house is unoccupied there is no source of humidity.
So I'm willing to hold out hope that moisture got under the roof during construction and now it's being driven out. That seems more consistent with what is being observed than insufficient venting.
If that's the case I would expect it to clear up within a few days.
The addition is still unoccupied, just initial heating.
How many days would you give the dry out while watching troublesome water leak from the soffit each morning. What worries me is that's the only water I'm able to see, no telling where other condensation runoff is going... I was out of town the past 24 hours and had the heat off. Got back around noon today and the soffits were dry, but I'm guessing obviously because the temperature differential in the vent cavity is now neutral and far from dew point.
Are you using central heating or a space heater?
Central is available now that the furnace is active. Currently turned off. I was looking at some photos of the spray foam and it looks like it was done much shallower than agreed upon. There was 7.25" of rafter showing due to the baffles and we agreed upon six inches of foam. Could the foam be too thin, leading to due point issues?
With the vented design the underside of the roof was always going to be close to outdoor temperature. The purpose of the foam is to act as a vapor barrier, as long as it doesn't leak it doesn't really matter how thick it is.
This is why I'm leaning toward some other source of moisture, what you're describing doesn't seem like what you'd see if interior air was reaching the underside of the sheathing. And there's no source of humidity on the interior -- although newly built buildings are often quite damp from being exposed to the weather.
I'd give it a week with the heat on, it's not going to hurt anything. The reason I asked what kind of heat it was is that propane heaters pump a lot of moisture into the air.
Is the spray foam continuous over that interior wall with the duct (ie does the spray foam stop at the wall and continues on the other side)? How is that duct getting to this new space?
The thickness of the spray foam won't matter with what you are seeing. My gut feel is still you have a large air leak into your rafter space somewhere.
Akos, the spray is continuous. That's just a jumper/transfer duct for the returns in the bedrooms. All heat comes from floor registers.
So am I understanding right that the issue cannot be too thin of foam, leading to temperature mixing in the vent channel? There seems to only be 3 inches of foam (maybe less) in some areas. I know that vapor likely isn't getting through, but if the topside of the foam is warm wouldn't that create condensation when the cold exterior air mixes with it?
Got a moisture meter and the drywall is between 4.5 and 6.5%. and humidity is at 30%. Doesn't seem like there's an abundance of moisture still working it's way out.
[response to #16]
If I understand correctly, you're getting condensation in the vent channel? IE, between the insulation and the roof. If that's the case then the mystery is how the moisture that is being condensed is getting in there. It's not coming in on the ventilating air that's coming through the vent, because it wouldn't condense because it's at the same temperature. It's either coming from the living space, or it's already under the roof.
In order for it to be coming from the living space there would have to be a leak in the spray foam. From the pictures it looks like it's a pretty continuous layer, but there may be a leak somewhere, that's Akos' hypothesis.
The other possibility is that the moisture is in the underside of the roof. When the roof heats up in the sun it's driven out, and when it hits the cold vent air it condenses. If that's the case, and presuming there isn't a continuous source of wetting but it got wet during construction, that water should all get driven out pretty quickly, think a towel on a clothesline.
The thickness of the foam really doesn't matter so long as it is continuous enough to keep moist interior air out of the vent. Regardless of the thickness of the foam the vent is going to be at the temperature of the outdoor air coming in.
Thanks, DC. If that's what's happening (moisture on underside of roof), it would be a temporary occurrence and stop once the towel on the clothesline is dry, so to speak? I may run the furnace for 3-4 days this week and if it remains a daily occurrence, I likely have another problem to solve.
Another thought is that some of the vents in the hip/ridge vent may be insufficient, and or blocked. Sorry for another Q, but if air is only getting in from the eave, could that be contributing to this?
It appears the culprit is not enough insulation. Based on photos, we may only have R-10 - R-20 in some areas of the rafters, and the top plates are significantly lacking.
We also took Flir images today and there are major hotspots in the eaves where the water is leaking through the soffit.
The plan is to have a blower door test done this week, and next week remove all the drywall and do a thorough inspection of insulation depth. This obviously isn't ideal, but seems necessary. At this point I'm left with these big glaring questions:
1. Is it safe to spray directly against the existing foam. Any reason to avoid that?
2. Even if the proper amount of insulation is added, will this baffle system be prone to condensate if any hip or ridge vent is blocked (fully or partially)? That's my biggest fear is that we go to the lengths of removing drywall and correcting the insulation, but I may (no way of really knowing) have an issue with ventilation as well. My thought is that R-38+ would keep that from happing, but trapped cold air in the winter, or warm air in the summer, may lead to condensation. If so, I'm inclined to rip it all out and spray directly to the deck.
Major hotspots at the eaves could indicate significant air leaks at the top plates. If the lumber was still wet when the drywall went on, it will be drying into the stud cavities. The warm humid air will rise up through any gaps into the eaves area, where the moisture will quickly condense out.
There is no issue with adding more foam to the existing. Thin areas should certainly be topped up, but the bigger issue is addressing the air leakage.
Check the moisture content of the lumber before putting the drywall back up. If I remember right, it should be no more than 19% MC.
Thanks, Frank. Lumber was up for three months before drywall, but I'll check. Any insight on that second question?
Insulation baffles are never prone to condensation, which is why your problem is vexing. Did you say that some of the vent channels are blocked at the top?
What has your weather been like? You answered me above that it's been 18° in the morning. What's it like during the day? Have you had rain? Has the sky been clear at night?
If the soffits are lighting up on IR, the issue is air leaks. Most likely wall top plates.
The fact that the roof is vented doesn't matter, this would have created the same condensation issue but probably a bit further down into the soffits.
Short rafters with only soffit vents are fine as long as there are no major air leaks. I'm in the land of 2.5 story structures and properly vented roofs are very rare yet most still hold up.
Find the air leaks and plug them. This will solve the condensation issue.
As for the spray foam thickness, see if you can have somebody dense pack the space between the drywall and foam. This will save you having to demo the ceiling and would get you pretty much the same assembly R value. Even if you do demo the ceiling, I would not put more spray foam but just fill the rafter space gap with cheap batts.
Akos, you are most likely correct, but the main reason for venting roofs is for the vents to whisk away moisture. The small amount of moisture that would leak through air gaps after drywall is finished is usually not enough to move a large amount of moisture; if it were, a huge percentage of new housing would have the same problem.
"Lighting up" on an IR camera is usually entirely relative. If another area had a bigger leak, the one that shows now would look small.
Warm, humid air from outside getting into the vent space but not being able to escape at the ridge, and night sky radiation cooling nails that penetrate the roof sheathing below the air temperature, could easily cause condensation. Whether it would cause more than an air leak from the interior is debatable.
The baffles are not hurting anything, even if the vent area at the top is less than ideal. Without the baffles, moisture would still be getting to the roof sheathing, except you wouldn't have noticed the problem right away. If you have a good air barrier continuous from the ceiling to the walls, very little moisture should reach the vent space. What actually is your air barrier on the ceiling and walls?
The air barrier is Certainteed Membrain on the walls (plus latex paint on the drywall later), and then the closed cell on the ceiling.
I'll plan to post another update once the drywall is down, we've done the blower door, and additional thermal scans. I'm still leaning toward removing all the closed cell and baffles, because it's just one more thing that could lead to a problem. There's been no shortage of them with this project!
I can't see the vent space causing any problem. But if you want to remove it, I suggest leaving the existing foam as is and using a pour-in-place foam. You'd have to block off the vent spaces at the bottom and then the installer would inject the foam in through the ridge.
Deleted
Is it safe to say you may have not inspected the foam depth prior to adding sheetrock?
Correct. That was overlooked by the contractor.
We would use a 1.5" foil face polyiso board as the vent (held 1.5" off of roof), but we would always use it as a "baffle" to keep spray foam out of the soffit & connect the vent to the outside of the top plate - then shoot foam to seal vent to top plate.
~formerly worked for a spray/injection foam contractor
An update for everyone. The drywall was removed today. And in the "hot spots", there was very little (if any) insulation on the top plate and underside of the baffles. In one location, I could see straight to the 1.5" foamular that I used for eave blocking.
It's looking like inadequate insulation and air sealing is the culprit for my soffit ice and moisture. The image is looking across the top plate to the foamular.
I'll keep providing updates.