Wow, Dana! I was uncomfortable just reading about how uncomfortable that house was. And horrified at how they tried to solve the building enclosure flaws with mechanical systems.

Yes, ducts and air handlers need to be inside the building enclosure. Absolutely! Thanks for making the point about that here in the comments since I glossed over it in the article.

(OK, time to pet the peeves again I guess… 🙂 )

ANY ducts (even insulated ducts) in exterior walls is just plain stupid- both an energy waste and an invitation for future problems. In a sane world gaps in the building insulation like that would not exist, punching holes in the building insulation to route HVAC ducting (or even install air handlers) outside the thermal & pressure boundary of the house just wouldn’t happen, but they do.

Ducts & air handlers above the floor insulation in vented attics, anyone? 😉

How about ducts & air handlers in basements or crawlspaces under the floor insulation?

Why not just put it all outside, up in the air? They’re out there.

A project I advised on in central MA a handful of years ago had barely a insulated flat roof /cathedralized ceiling (no attic, thus) and a finished walk out basement that housed a tiny mechanical room with a 200KBTU/hr hydronic boiler and a 50 gallon water heater. Cooling was provided by 9 tons of commercial building style AC package units (a 4 ton and a 5 ton) mounted above the roof, with a duct sculpture (insulated with R6) running to ceiling diffusers and a few ceiling return grilles. It had enough heat & coolth to cover the loads, but it was MISERABLE to live in! The AC was deafening when running (but oversized to the point of not adequately controlling latent loads), and when the radiant ceiling heat was running the house reeked of bitumen roofing indoors. After every snowfall (in a city that will sometimes get 10+ feet of total snowfall over a winter) the snow melt formed ice dams & icicles, sometimes big enough to a hazard to cars or people below when they fell.

It took some convincing (the owners were ready to give up and sell), but building a very low slope gabled roof above the existing roof with 6″ of polyiso and new membrane on the new roof deck, creating a cramped crawl-attic to house RIGHT SIZED (per Manual-J) mechanicals, which consisted of a 60KBTU/hr 2 stage furnace and either a 3 or 4 ton (just one) AC unit. Combined with some window upgrades and spot insulation fixes that great looking but downright miserable uncomfortable energy hog Mid Century Modern (circa 1958) transformed from being one of the least comfortable houses ever, to something VERY comfortable in all seasons, using less than half the heating & cooling energy that it had previously. (This was not a project for the faint of heart or light of wallet, but it was later sold at a net gain despite only staying 3-4 years after fixing it.)

The point being- bring ALL of the mechanicals COMPLETELY inside the thermal and pressure boundary of the house, and right size the equipment. That makes it possible to actually control heat & moisture flows to properly condition the living space, and avoids a whole range of problems for the house itself. The practice of installing ducts in vented attics or exterior wall stud bays needs to just end!

Great question, Gordon. I agree with Curt’s recommendation for code-minimum fiberglass duct insulation with a vapor barrier. Bubble wrap by itself is about R-1, so it wouldn’t meet the code. If it’s installed with proper air spaces, it might work, but those gaps are hard to achieve and even harder to maintain.

The bigger issue is that ducts shouldn’t be in exterior walls unless you have enough insulation outboard of the duct to meet the code requirement for that wall. It also needs to be air sealed to keep humid outdoor air from getting anywhere near the duct.

Duct insulation should then be covered with a continuous vapor impermeable barrier. This serves as a secondary defense against air leaks, but, much more importantly, prevents the insulation from being wetted by condensation. Typical outer layers include mylar, metal foil, reinforced scrim tape, or fairly thick polyethylene (similar to sub-slab vapor barrier film)

In theory, an assembly consisting of “batts and bubble wrap”, properly installed, might meet requirements, but I’m skeptical what you describe performs properly.

What do you think the minimum r-value should be for these ducts? Should “bubble wrap” be considered insulation?

Curt, that’s a great way to put it. Humidity problems are all about keeping humid air away from cool materials, with “humidity” and “cool materials” defined in terms of dew point. Of course, bad flashing, roof leaks, flooded basements, and other liquid water problems cause more damage than humidity problems, but water vapor still claims its share.

“Manage the dewpoint, or it will manage you!”

Matt, the leakage I was talking about was outdoor air getting into wall cavities, where it can condense on the uninsulated sheet metal. Duct leakage certainly can make things worse, but it’s the outdoor air with all its humidity near the cold sheathing and siding that causes this problem. If that outdoor air gets into the wall cavity—and it usually does in old houses with ducts in exterior walls—then the condensation happens inside the all, too.

You wrote: “…someone is going to grab a can of foam insulation and inject it between the two. I can see issues with that. What about you?” Yes, I can. That’s not what I meant by that statement, but yes, if someone interprets it that way and does what you suggest, they probably won’t solve the problem. At least here in Georgia, the boot fills up the whole wall cavity so they wouldn’t really be able to put foam in that space. At best, they’d get foam around the edge. I’ll see what I can do to clean that recommendation up. Thanks for pointing that out, Matt.