Rigid foam vs spray foam in attic conversion
My 1964 house in North Carolina’s zone 4 has its upstairs HVAC in the attic. The attic is flanked by brick chimneys and gable vents on both sides. There is a ridge vent as well, but no soffit vents. The soffits are not very deep.
I’ve been in the house about 11 months and have an excuse to replace the gas furnace and AC in the attic. Based on the info on this website, I should convert the attic to an unvented space using closed cell spray foam and install a high efficiency gas furnace and AC. I’m concerned about a couple things with this approach:
1) not being able to detect a roof leak under spray foam, and
2) off-gassing of the spray foam and harm to the home’s occupants (children).
Is the solution to install rigid foam under the rafters to create an air and thermal barrier between the attic and the roof? Does anyone have experience with such a “super” attic insulation system?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
HVAC replacement is an opporunity moment for right-sizing the equipment. The average gas hot air furnace installed in the US is at least 3x oversized, and while that isn't an efficiency problem, it takes a dramatic toll on comfort. Before opting for a condensing gas furnace + split system AC, run these numbers to get a handle on the as-is-where-is heat load, including all of the parasitic losses to the attic:
https://www.greenbuildingadvisor.com/article/out-with-the-old-in-with-the-new
If there are multiple systems it's not possible to estimate the heating and cooling loads of just one from the gas bill. Running an aggressive Manual-J load calculation will have to be done, but it's still worth data-logging the duty cycle of the AC against outdoor temperatures on warmer days to put a sanity check stake in the ground on cooling sizing before proceeding. Odds are pretty good that the AC is at least 2x oversized , even with the parasitic loads of the ducts & air handler being in the attic. Methods of sizing by duty cycle are explained here:
https://www.greenbuildingadvisor.com/article/how-to-tell-if-your-air-conditioner-is-oversized
For a first-cut on a Manual-J there are some sort-of reasonable freebie online tools such as loadcalc.net or coolcalc.com. (Even though it's arguably clunkier to use and not as slick, I personally have more faith in loadcalc. YMMV) These are fairly un-nuanced load tools with limitations to the inputs. Experience has shown that the numbers will be closer to what a pro-tool delivers if one assumes zero outdoor air infiltration and zero duct leakage, but it'll at least find the right ball park. Pay attention to how much of the load is attributed to the unconditioned attic vs. direct loads from conditioned space.
There are very few gas furnaces out there appropriately sized for the heat load of 800-1500' of upper floor at NC-style outdoor temperatures. It's more likely that a modulating or multi-stage heat pump solution is the better choice, even if the raw energy cost of gas is comparatively low in your area. But run the numbers- Dettson may have something right-sized for your loads:
https://www.greenbuildingadvisor.com/article/finally-a-right-sized-furnace (They also have a modulating mini-split type compressor compatible with that system.
Converting to an unvented conditioned attic is expensive, and depending on where the load numbers work out it can often be cheaper to install a right sized ducted (or mini-ducted) system fully inside the existing conditioned space, even if some of the duct need to be run in soffits underneath the ceiling. In that case the existing system can be ripped out, ceiling penetrations sealed, maxing out the insulation on the attic floor with cellulose, eliminating any of the issues related to foam. Before insulating it's worth meticulously air sealing the ceiling /attic floor plane, which may be easier using some amount of foam, but it can usually be avoided if desired.
With only a ridge vent and no soffit vents the attic is being depressurized relative to the house, which presents a 24/365 air infiltration drive. If the soffits aren't deep enough to add venting equal to or greater than the free area of the ridge vent there are under-shingle roof edge venting systems out there, or the ridge vent can be sealed up from the interior and gable vents added.
The comfort and lower cost aspects of right-sizing rather than oversizing can't be overstressed. Nate Adams in Ohio has made a business out of fixing comfort & efficiency issues with houses, and more often than not it requires DOWNSIZING existing systems (often nearly brand-new systems) to get there. It's worth reviewing some of his blogs and videos:
http://www.natethehousewhisperer.com/home-comfort-101.html
http://www.natethehousewhisperer.com/hvac-101.html
http://www.natethehousewhisperer.com/hvac-102.html
So, first order if business is to run the load numbers on the conditioned-space-only and we can work on it from there. If attic floor insulation is likely, use the "after upgrades" R-values. If you prefer to use spreadsheet tools rather than a "black box" load calculator tool, the older I=B=R style load calculation is still a valid method of designing the system:
https://www.greenbuildingadvisor.com/article/how-to-perform-a-heat-loss-calculation-part-1
https://www.greenbuildingadvisor.com/article/how-to-perform-a-heat-loss-calculation-part-2 (the air leakage load numbers are always squishy- consider this an upper bound.)
https://www.greenbuildingadvisor.com/article/calculating-cooling-loads
Thank you for the reply, Dana. It'll take me a while to comb through all that info and make sense of it. One of the first two links you provided mentioned data loggers. While I don't have a data logger, I do have ecobee4 t-stats. The ecobee logs indoor and outdoor temps and graphs that with cooling/heating run times. In one illustration of how much the high attic temps in the summer affects cooling, one afternoon it was 89-90 outside, the set point was 77-degrees and the indoor temp was 78-79. The AC ran for 4 hours straight and then for 45 mins/hr for 4 more hours until the set point was increased above the room temp at 9 PM. Several hours later at 3 AM, I dropped the t-stat from 80 to 78 degrees and the AC reached 78 just 30 mins later, i.e., with I can presume was a much cooler attic since the outdoor temp was 74. Do I like keeping my house at 78? Oh no, but as this example showed, there seems to be little cooling capacity in my current arrangement to meet at 75-degree setting on a typical summer day. So, the AC just struggles to cool the upstairs with a hot attic. This is why I'm interested in cooling down my attic. It appears to be big energy waster.
I have to get past the idea of an AC running for a long time is a good thing. There was some mention of improved dehumidification. Is that right? Any other benefits?
As an aside, one of the two contractors I asked for estimates from confessed he's not going to do a Manual J and he didn't need to consider how much insulation was in the attic to size the equipment. From what you're saying, it's unlikely I'll get the right size equipment from him.