Please help me decide between 2 attic insulation options.....

Facts
* I live in Dallas, Texas.  My objective is to make my house comfortable in the summer.
* I live in a 2-story, 2600 sq foot house built in 1920, which is essentially a perfect rectangle. 
* The attic is 1128 square feet with a 6/12 pitch.
* The attic has 2 windows, one on the south and one on the north.
* A recent energy audit confirmed that my house leaks a lot of air (windows, doors, etc.), but especially in the attic.
* The attic is vented and gets very, very hot in the summer.
* I have two HVAC units, one of which is in the attic.
* I also have an electric hot water heater in my attic.

Option #1 (unvented solution)
* Remove all old batt insulation from floor of attic and vacuum.
* Air seal entire attic floor using 3/4" close cell foam.
* Fill entire floor with blown-in fiberglass to achieve R-38 (filled on top of foam).
* The floor joists are mostly 2X6, but some 2X8.
* Test and seal HVAC ducts in attic.


Option #2 (vented solution)
* Remove all old batt insulation from floor of attic and vacuum.
* Seal the entire attic ceiling with open as much open cell foam between rafters as possible.
* Rafters are 2X6, so I can get approximately R21 using the open cell foam (i.e., 6” X 3.5).
* Test and seal HVAC ducts in attic.


**** Which option will give me the best energy performance, minimize my bills, and maximize my comfort in the summer????????

Thanks.

I think you got the vented & unvented descriptors swapped.

#2 wouldn't even meet CURRENT code-min (R30) for R value, but would likely perform better than what you currently have. To meet the (not currently in force) IRC 2012 you would need at least R5 in rigid insulation on the exterior of the roof decking, and a total R of at least R38.

#1 would be the likely performance winner, provided the air-tightness of both the ducts and attic are verified, and the ducts are insulated as well as sealed.

But even better would be to go with a hybrid:

Insulate and air seal the roof deck & gable ends as the primary air barrier, and blower-door verify air tightness.

Then blow cellulose (not fiberglass) over the batts, (rather than removing the old stuff) to a depth where it covers the joists by 3" (or fully filling the joist bays up to the floor boards.) With 2x10 joists filled to the top that would be ~R32-ish, but R40+ if the joist tops are covered by 3" (which is an R10 thermal break over the bridging joists, cutting the heat transfer at the joists in half.) Low density fiberglass isn't air-retardent enough to perform well a the winter lows, since it allows convection loops through the fiber with the cooler attic air, and in summer the partial translucency of fiberglass to radiated heat coming down from the roof deck (mostly rafter-edges now) make it underperform it's rated-R at the summertime peak temps. Cellulose doesn't have these issues, and has a more stable R over temp.

This puts the ducts & air handler at least partway inside the thermal envelope, fully within the pressure envelope, and brings the total R up to something substantial for higher overall comfort. The attic will still run hotter than in conditioned space, but it's humidity will track that of conditioned space, and in a Dallas climate would not have any condensation issues. While this would be a violation of code under IRC2012 on a couple of counts (it still needs R5 outside the roof deck, and the attic space would not be fully within the thermal envelope, a requiremnt for an unvented attics), it's performance would be good.

Caveat with either the hybrid approach or #1: If the HVAC unit in the attic has a gas-burner there would be combustion air supply and other issues that would need some attention.

cooper 24000;
From the Building Science Corporation:
"Buildings should be suited to their environment. It is irrational to expect to construct the same manner of building in Montreal, Memphis, Mojave and Miami. It's cold in Montreal, it's humid in Memphis, it's hot and dry in Mojave and it's hot and wet in Miami. And that's just the outside environment. It is equally irrational to expect to construct the same manner of building to enclose a warehouse, a house or a health club with a swimming pool. The interior environment also clearly matters."
They lean towards the unvented attic in hot humid climates and so do I

"They lean towards the unvented attic in hot humid climates and so do I"

Me too, but not if it's only R20 center-cavity on the rafters in a location with a 98F 1% design temp. The hybrid approach gives you higher R between the roof deck and the conditioned space ceiling lowering the overall load and increasing comfort. It doesn't meet the letter of IRC 2012, but it sure works better.

And any of the above would beat the leaky-ceiling with ducts & air handler above sub-code batts.

Dana1 has provided some excellent suggestions and analysis. If you use a gas fired burner on your HVAC you would want to go with a direct vented high efficiency (95% or greater AFUE) unit anyway. Virtually all that I have installed in recent years provide for a sealed combustion direct vent (2-pipe) arrangement - usually 2" Sched 40 PVC. They are also generally equiped with either two stage or modulating burners and variable speed DC bolwer motors.

this would be a violation of code under IRC2012 on a couple of counts .. the attic space would not be fully within the thermal envelope, a requiremnt for an unvented attics)

Every time I look at these "partial" ideas, I conclude that they are a bad idea. Vent your attic to the interior or the exterior.

don't do open cell or you'll end up with condensation as I did, see my post.....
6" is simply not enough. you'd need at least 6" of closed cell to get the r factor you need AND the moisture break.

also, keep in mind the foam stinks forever. don't believe those professionals or even the company reps themselves. do a little googling and you'll see it's true.
there are a lot of people complaining about it. I had open and closed cell sprayed. if you can do without it, it's better.
they say it's an inert substance. yes maybe in a lab where it was mixed perfectly to specs but not when it's done at the job site in a truck by someone who doesn't know much about chemicals.

i also now tend to say if you can vent the roof do it. the moisture needs to exit somehow. but also consider moisture comes in in the summer through those vents, too
and needs to be dried somehow. it's a science of its own...

Posted By Sav on 01 Jun 2012 10:22 PM
don't do open cell or you'll end up with condensation as I did, see my post.....
6" is simply not enough. you'd need at least 6" of closed cell to get the r factor you need AND the moisture break.

also, keep in mind the foam stinks forever. don't believe those professionals or even the company reps themselves. do a little googling and you'll see it's true.
there are a lot of people complaining about it. I had open and closed cell sprayed. if you can do without it, it's better.
they say it's an inert substance. yes maybe in a lab where it was mixed perfectly to specs but not when it's done at the job site in a truck by someone who doesn't know much about chemicals.

i also now tend to say if you can vent the roof do it. the moisture needs to exit somehow. but also consider moisture comes in in the summer through those vents, too
and needs to be dried somehow. it's a science of its own...

Polyurethane supposedly off-gasses for years and years. The research is sketchy and it depends on who you read the results from. Spray foam and polyurethane SIP companies claim it is safe but others say otherwise.

EPS is a much more stable insulation as is cellulose insulation.

Open cell foam pretty much gives it all up in the first 1-2 years, closed cell foam takes decades (centuries?) to fully outgas.

Using open cell foam does not increase the risk of condensation compared to closed cell foam. At R30 (current code-min in Dallas TX) closed cell foam is vapor-impermeable trapping moisture at the roof deck if the roofing material is also vapor impermeable (like asphalt composition shingles). If insulating below the roof-deck only, use open cell foam, but foam at least 3" over the rafter ends to thermally break the rafters. And don't assume it's air-tight just because it appears to be foamed all the way down to the soffits- verify it with a blower door and remediate any leaks before the foam truck pulls away.

Splitting the R value between the roof deck and attic floor has far fewer issues of concern in warmer climates than in cold climates, where there's some risk of the attic space falling below the dew point of interior air. In Dallas winter outdoor temps are quite temperate (with a mean outdoor temp in January of ~47-48F, with a mean outdoor dew point of ~34F- it's dry) minimizing the condensation potential a the mid-point of the R. Even if you added humidity to raise the indoor RH to 55% (in general a bad idea even from from a health point of view), the dew point of 70F/55%RH conditioned space air is 53F, and the simple-model temp in the attic at 47F outdoors/70F indoors comes in around 58F. At a more realisitic and healthy 45% indoor RH the dew point is about 48F, which is the mean OUTDOOR temp in this location. The fact that the HVAC unit & ducts are in the space lowers the risk even further, since the parasitic losses of the heating system accrue to the attic space raising the temperature. You have tons of margin against winter dew point/condensation issues in the attic here, even with 2/3 of the R at the attic floor.

In summer, with the roof deck sealed, the attic dew point will continue to track that of the conditioned space air, but will be much warmer, and even less likely to condense. The high summertime dew points are more likely to become an issue if the attic is vented to the outdoors, raising the potential of condensation on the ducts & air handler, or even the attic-side of the ceiling gypsum on extremely muggy days.