Critique for gut-job re-do — insulation
Hi,
I’m working on a gut job rehab of a property I bought in Houston, TX. (Hot Humid, AC 9 months of the year). It is 1965 construction 2 story. First story is brick veneer with an air gap and exterior drywall over stud framing. Second story is vinyl siding over vertical wood siding with deteriorated 30# felt over stud framing. Roof is approximately 18 years old, typical redo for Houston… they removed cedar shingles and put down 1/2″ plywood deck with asphalt shingles. It was insulated with 1.5″ rock wool and there was evidence of moisture accumulation on the interior side of the exterior drywall.
Slab has a vapor barrier, but is gaining moisture from the ground somehow…..
My goal is to have a comfortable house that has low energy costs planned on a 15 to 20 year ownership cycle.
Right now I am planning to do closed cell foam in the exterior walls, convert to a closed attic with open cell on the roof deck and use ducted mini splits for zoning purposes.
The vinyl siding is going to go away. I will either keep the wood siding or apply tyvek and fiber cement siding. There are some flashing issues where the upstairs walls meet the first story roof. Minor leak under the right wind driven rain conditions.
I plan to apply the foam myself and I already own the equipment to do it and am certified to handle the materials and have a local distributor lined up to purchase from. I will be applying some open cell in selected areas as a sound deadener (like on the bottom of the upstairs floor).
I chose closed cell in the exterior walls to provide an air and moisture barrier. I would use open cell, but I am concerned with moisture migrating in and staying there. Short of removing the brick I can’t fix the exterior envelope. Either way I stack it, I am in for two sets of material to insulate this house properly, so shifting one set to closed cell lines up nicely with the exterior walls and isn’t that much more expensive as I am applying it myself and own the equipment.
I replaced the windows with Plygem units from Home Depot that meet the energy star requirements for my region on emissivity and solar gain.
Issues and Questions:
– I am trying to calculate the benefit of installing a 5v style metal roof against asphalt shingle. The roof is 20 years old and has some flashing issues and 2 areas with minor leaks. That’s enough for me to replace it as part of the scope of this project. I have installed utility style (corrugated) metal roofs and built a metal building before, so I’m not new to metal working. I’ve also done 3 tab shingle roofs. My working theory is that a metal roof will cost about 25 to 50% more than a good quality 3 tab roof (Owens Corning Duration or similar). That savings will be paid back in insurance discounts and energy savings by going with a silver, Galvalume, or light grey roof. White would be ideal, but I have trees and that will get the white roof dirty.
– I considered a R style corrugated roof on wood stringers on top of the deck and then doing a radiant barrier under it. That is easy and low-risk but I can’t find any references on doing that. It makes sense that the radiant barrier would be highly effective in this style and then I’d be free to put whatever color I want in.
– HVAC equipment has to stay in the attic due to 8 foot ceilings. Therefore I think it makes sense to cathedralize the attic with spray foam.
– Any thoughts on the first story wall assembly open cell vs closed cell? Specifically – will open cell in a hot humid climate with air conditioning on one side suck in and retain moisture from the exterior? The exterior drywall is not taped/sealed.
– I’ve searched for a model assembly to use, but can’t find anything that matches this construction style with a retrofit.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
The slab's moisture gain may be from condensation/adsorption from the room air if there is a lot of air infiltration, but it deserves some investigation.
A flash-foam of 1" nominal closed cell foam on the walls would be sufficient vapor retardency to manage the moisture drives off the brick, and you can then fill the rest of the cavity with compressed batts or sprayed fiber. Installing 2" of closed cell foam to meet the prescriptive code-minimum would be both lower performance and higher cost than a flash & batt (or sprayed) solution, due the the very short 2" /R2.5-ish thermal bridging on the framing fraction compared to the 3.5" R4+ of a full cavity fill. A 3"+ cavity fill of closed cell foam has a much higher environmental cost (both polymer & blowing agent), for little-to no performance gain over a flash & batt/spray solution.
An open cell cavity fill behind brick veneer would air seal just fine, but would be prone to moisture migration into your interior-side finish wall, during the cooling season at Houston's average summertime dew points.
Verifying that the brick veneer is vented to the exterior both top & bottom can also be important for managing the moisture uptake. It should have periodic weep holes on the bottom course of bricks, with either an open top to the cavity into a vented attic (common), or vent slots in the vertical mortar of comparable size & frequency to the weep holes. If sealing the attic is going to block the top of the masonry cavity, drilling some venting to the exterior is in order.
Open cell foam on the roof deck at the full IRC2012 prescriptive R38 for US zone 2 would work, as long as it's verified to be fully air-tight (blower door testing using theatrical smoke machines makes any leakage at the harder-to-seal soffits dead easy to spot.) That's a fairly pricy proposition though. Depending on your rafter/truss configurations there may be cheaper alternatives (or not.)
If you end up doing the metal roof over wood stringers, you can get the radiant barrier sort of for free if you use bare galvalume for the roof; its emissivity is much lower than painted metal.
Great info Dana. Thank you for confirming what my research has pointed to with the moisture ingress during our cooling season. I agree flash/batt would probably work but I have to buy the foam in sets and it just happens that a set of closed cell will align with filling the wall cavities.
Some updated info on the metal roof. I reached out to my insurer and they confirmed that basically there is no savings for a metal roof. The metal roof raises the rebuild cost and while they do offer a discount for an impact resistant metal roof, it actually raises the premium against an asphalt roof.
So the question becomes this... with R38 Open cell on the roof deck and a sealed attic, does a metal roof offer a substantial heat rejection over asphalt? I don't think it does. I'm leaning towards a 5v type profile and the manufacturer documentation that I can find indicates a requirement to be in contact with the roof deck, thus ruling out a radiant barrier.
I'm not clear what's meant by "a set of closed cell"???
In my neighborhood any closed cell foam jobs from 500 board-feet or larger are usually cheaper to have done by a pro than with DIY foam kits, and many installers would still do a 200 board foot job (at a very high price per board foot.)
You could paint the interior side of the exterior gypsum sheathing with vapor-barrier paint (rated at about 0.5 perms, about the same as 2" of closed cell foam) to manage the high exterior moisture drives, then use open cell foam in the cavities.
Using a "cool roof" asphalt shingle can be every bit as heat rejecting as a non-white finish on a metal roof. To compare apples to apples you can search the CRRC by product type, and compare the 3 year aged solar reflective index (SRI).
http://coolroofs.org/
http://coolroofs.org/products/results
The highest SRI products tend to be a variation on titanium-white, but there are a broad range of colored finishes that still reflect very well in the near-infrared spectrum (about half of the incident solar energy) with good deep infra-red emittance (radiating absorbed heat back toward cold radiant-temperature sky), while having a decent amount of visible spectrum color for aesthetic purposes.
Somewhat dated, but with many links to various details on cool roof materials development can be found here:
http://coolcolors.lbl.gov/
Dana,
The coolroofs.org site is great, thank you.
As for the foam. I own a small machine that is capable of applying SPF. It's not a froth-pak or DIY foam kit style unit. It's made by a small company in Oregon called MixSyn. http://www.sprayfoammachines.com/lilpuplowpressure.html . Foam is sold by suppliers in 2 drum sets that are commonly referred to as simply "sets". It's an A (isocyanate) and B (resin) set that are typically matched chemistry. It has a shelf life of up to a year if stored properly, but this varies by manufacturer.
I've done quite a bit of research and the yield from a set of Closed Cell foam should be about the same as what I need for my walls. My walls are 1372 s/f ignoring doors and windows. At 3.5" depth this works out to 4872 board feet. The reality is that the walls have 2x4's, pipes, outlet boxes, doors, windows etc, so I probably won't use 4872 b/f. A set should produce in the vicinity of 4500 to 5500 board feet depending on factors like temperature and the machine being adjusted properly.
When I contacted various SPF contractors I found two camps. One camp wanted nothing to do with a remodel. They were too busy. The other camp was interested, but expensive and the way they described wanting to install the product didn't mesh with how it ought to be installed. They wanted to install 2 inches of open cell in the walls. That's lovely for the contractor because it avoids waste and trimming labor. It's a #FAIL for the insulation project as it requires coming back with fiberglass or leaving open space in the wall.
There are a few advantages for me in doing it myself.
1) The going rate I found to be in the area of 50 cents a board foot. Material cost is actually around 14 cents a board foot using a name brand product like BaySeal or Quardrant. Closed cell has about 1/4 the yield so it's probably closer to 60 cents a board foot. Now in all fairness people need to make a living and a trailer mounted rig is expensive. So first, I can save money by doing it myself. I have the proper safety gear and understand what I'm working with. My rig is smaller and has limitations that have to be worked around. I won't be able to apply 30 pounds of material a minute in a snowstorm.... I need to work in warm weather and it will take me longer to apply material.
2) By doing it myself I know exactly how it's applied and can take the time to detail it properly. I'll end up insulating areas that most people wouldn't insulate... one example being putting a couple inches on the bottom of the upstairs floor to act as sound attenuation. Another example being insulating between the living room and the master and half bathroom to stop any noise. It wouldn't be worth bringing a contractor back when it's time to insulate those areas, but with me having the equipment and material it's an easy add-on. Another example is insulating the half wall next to my in-wall aquarium to muffle any water noise from the equipment under it.
3) with a contractor I need to get the whole house ready to spray at once.... with my equipment I can work on it in sections which is works better for me.
Doing it yourself isn't for everyone, especially with spray foam. I've run the calculations and it works in my case. I can also add that I tested cartridge applied systems and found it to be unacceptable. I won't name names but folks can find that if they are interested in it. The DIY kits are really only good for very small projects.
If anyone is interested in the technical side of calculating yield Mason Knowles has a great article here on it: http://masonknowles.com/docs/calculating_yield_in_sprayfoam.pdf
I've narrowed my options for roofing down to a 5V profile galvalume roof vs architectural asphalt shingles. I'm leaning towards the metal roof right now, but having some trouble with some of the details the fabricator is specifying.
The roof pitch is a 5/12 which is a decent pitch roof.
MBCI is the fabricator I'm talking to and I've used their products before, plus they are local in Houston. They are saying that the metal needs a waterproof layer under it because the 1/2" high seams can leak under the right circumstances. When I asked what the layer should be the sales person told me Ice and Water on the whole roof. That seems very drastic. I looked at their Miami Dade Notice of Acceptance (NOA) to see what the specification was. In the NOA It's a ASTM 226D 30# felt, which appears to primarily have about twice the asphalt as regular 30# felt that Home Depot sells. I would have preferred to use something like Grace Tri-Flex, but Grace is mysteriously silent on using that product under a metal roof. Instead they want to sell their Ice and Water HT, aka the most expensive product they make.
The 5V is spec'd to be installed directly on a barrier on top of the sheathing.
The assembly is 2x6 rafters 24" OC with 1x3 stringers about every 8 to 10 inches. On top of this is what appears to be 1/2" plywood, covered with what looks like 15# felt and 18 year old architectural shingles. I will be using 7 to 8 inches of Open Cell on the underside of the roof deck and converting to a non-vented attic.
The shingles are in good condition and have reasonable glue down between layers. I estimate that they have 5 to 10 years or remaining serviceable life. I am replacing the roof because there is a leak in one of the valleys and there are flashing assembly mistakes where the first story roof and the second story walls meet. I don't think it would work well to replace just the shingles at the intersection. The flashing was reused from the original cedar shake roof. there are gaps in the original flashing which is under vertical wood siding. There is no step flashing. I don't think I could retrofit step flashing without destroying the existing shingles. They will break if bent.
I considered a click-down system, but it's about 25% more expensive and they insist that it needs Ice and Water under it... which adds $40/square to the cost... or about $1200 to the project. 26 Ga metal is very competitive with shingles if you ignore the trim being more expensive. My expectation is that it will go faster and require less labor which will balance out the trim costs.
I have a few questions:
1) What do most people put under their metal roof? Does it really need Ice and Water on the whole roof?
2) If I do put Ice and Water on the roof, will it bond to the metal and cause me hell if I ever need to remove a panel?
3) Could I use the existing roof as a waterproof membrane and put a layer of 30# felt over it to get rid of any abrasion? In theory the existing roof is waterproof. Code in Houston says you can't have more than 2 layers of shingles. It's silent on metal over asphalt.
I want the metal to be done right, but at the same time I need to manage cost and not go hog wild with Ice and Water. I have inspected the sheathing. The building is open to the inside with no sheetrock so I can see all of the sheathing and don't see any evidence of leaks or rot. The leaks I mentioned earlier only occur during torrential downpours with the right wind conditions. I suspect the building has always leaked when exposed to wind-driven rain.
I will be re-siding with fiber cement siding and using the existing siding as sheathing with a layer of Tyvek over it.
Brian,
You are raising a lot of issues here.
Are you the roofer? Or do you have a roofing contractor?
If you are hiring a roofing contractor, it usually makes sense to hire a contractor with good references, and to defer to the contractor if the contractor has a system that he is comfortable with. A reputable contractor will stand behind his work.
If you are the roofer, and you aren't sure how to proceed, that's not a good sign. Maybe you should hire a roofer.
Second: Building materials should be installed according to manufacturers' instructions. If you aren't comfortable with the manufacturer's instructions, maybe you should choose a different product.
Third: Like you, I think that it rarely make sense to install Ice & Water Shield over the entire roof. In many parts of the country, metal roofing is regularly installed over asphalt felt, and these roofs are leak-free. However, a leak-free installation depends on the details -- especially the flashing details -- and local climate. I'm not familiar with hot climate details, because I live in northern Vermont, but I know that metal roofing can get very hot in Texas. This fact affects which products you can use for underlayment.
Many roofers install metal roofing on top of asphalt shingles. The usual method is to install 1x3 strapping over the shingles, parallel to the ridge, every 16 inches or 24 inches, before installing the metal roofing.
I am the roofer. I will probably have low-skill help when appropriate.
My past experience includes the erection of an engineered metal building, designing it's foundation (reviewed by an engineer, quote from the inspector - "way overbuilt, wish everyone did it this way"), and completing the wiring (under the supervision of a master electrician I hired). I also renovated a 1930's bungalow, raised it a foot while leveling it (block and beam foundation), built an additional wing, wired it, plumbed it, roofed it. Both of those projects were inspected by the City of Houston. Generally I aim to exceed the code. I also built a 900 s/f shed and a lean to mechanical area (approximately 900 s/f as well) both of which had corrugated metal roofs.
I am comfortable with the fabricator/manufacturer and I can read and follow their instructions. The underlayment instructions are ambiguous and manufacturers are typically not concerned with green building practices. They are surely not concerned with avoiding 30 yards of asphalt shingle waste in a local landfill. As a side note there is a recycling option in Houston, but there is not a way to get the material to the recycler without exceeding the cost of trash disposal. Sad.
As with most projects, even those I have done in the past, I like to research what others are doing, check the current code, and check the manufacturer's recommendations.
The sales rep is saying Ice and Water..... the NOA from Miami Dade is saying 30# felt. I'll follow up with the manufacturer, but I wanted to find out what others were using that was practical and sustainable. I also looked at the documentation from other manufacturers and they are either ambiguous or say 30# felt.
While I think hiring a roofer would be wise for some folks, it obviously didn't work out the last 2 times this house was roofed. If the job had done right I wouldn't be tearing off the roof to fix the flashing errors.
So the IBC mandates Ice and Water where the average January temp is below 25 degrees. Houston doesn't experience that type of weather. An unvented attic with SPF on the underside of the roof deck changes the factors for ice damming. Houston does not have Ice issues on roofs. So I agree with you that the reasons for using Ice and Water in beautiful Vermont do not apply to sub-tropical Houston.
We do however have leaf litter and wind driven rain. The sales rep was mostly concerned that if water backed up it might leak through the seam of the roof panels as it is only 1/2" high. It's a valid concern. I'm not sure how heavy it would have to rain for water to back up 1/2" on a 5/12 slope roof with the longest run being 18 feet. However, I think the solution to it is to seal the panel seams with tape sealant. It's cheap, that's an easy solution and it provides an extra defense against water intrusion.