Best Placement of an In-Attic Furnace, Among Roof Trusses

The title of this post is revised, deleting the furnace adjective, gas. Information will apply to a gas furnace with or without air conditioning, and to HVAC heat pump installations having the same footprint and air duct arrangements.

Customer Meyer:

I have called this a good installation.

Here is the Meyer installation as I found it. Badly installed ducts are in the way of harm, everywhere. Ducking ducts, one could at best get at the furnace access panels. The untrained "professional" installer ruined the side of the furnace, hacking on an oversize return box with no possibility of the imperative return-air high-efficiency filter. 

See outlet Medusa of fully unguided flow losing all kinetic head.

See accommodation of two return air ducts at the narrow inlet D-box.

I will demolish all of this awful ducting less than two years old and blessed as conforming with required methods. Leaky and unstable duct connections to a D-box are hacked-in unsupervised at the job site. I will spend several hours dangerously stripping off goop and screws that might have achieved an airtight assembly. Worse to me, the assembly is hydraulic foolishness. Dampers full-open only add to the wasteful blower head. There is a better way.

It is October 30, 2017 in mild-maritime weather of Portland, Oregon, and the home will lack HVAC for several days. Here the furnace is reset on a raised floor with space below that will pass furnace-output flex ducts, under flooring, buried in the floor insulation. Is there a better way to place ducts in "conditioned space?"

Here trial-fit the new, lined 180° turning plenum. Behind the furnace see bags of fiberlass insulation employed to fully stuff space under flooring, to R60. The new floor supports are up 20 inches from ceiling drywall. There is good return in filling all available space with insulation.

The improved furnace has a  insulated 180° turning plenum on top, directing conditioned air downward with ducts to be buried under a raised attic floor. A couple of vanes in  the second 90° turn might do some good in reducing flow turbulence.

Look down through the progression of wye fittings below the turning plenum, that keep air moving steadily at ideally a constant velocity. All of this guiding steel is with some cost of efficiency. Steel mass that responds to elevated temperature of each furnace cycle, has heat capacity. Heat capacity cooling or heating output furnace air, acts just like leakage of the conditioned air. Do what I can to reduce responding thermal mass, by palm-applying many layers of flexible grout. Flexible grout is insulating ceramic microballoons more than half by volume. I allege that the coating reduces responding thermal mass by more than half, a really good investment of my time with material cost-free.

Framework that supports flooring, and passes buried ducts, can be complicated. There is a good R25 base layer of insulation under this ducting pit. It gets more difficult now, to tightly place insulation.

This is all that will remain visible, of HVAC flex ducts. With planning, excess R8 duct liner pulls up to cover almost all area of the non-lined steel fittings. Full wrapping is of about equal importance, to the internal lining of ceramic microballoons; neither perfect insulation. An in-attic furnace must have been thought necessary. I have perhaps minimized the penalties, while making the attic a well-lighted, useful space. 

Let's look around this now-pretty attic, reviewed in other blog post about its new access hatch cover and R30 insulation of a skylight.




Finished attic, access beyond the skylight, two shelves nestled in trusses. Storage space was needed and affordable. no one may command that attics be left dark and dangerous, with precariously-hung, energy-losing tangles of flex ducts and pretend-only hatch and skylight insulation. Extensive flooring was essential to placing the R60 floor insulation, now protected for sixty years or more. I think flex ducts sheltered from attic heat and yanking in passage should survive far beyond warranted life of not more than 25 years. The creation of useful space was a very good investment and is perhaps affordable for anyone.

An Opportunity Not Yet Resolved, Customer Coleman:

(Someday, maybe with conversion to heat pump elimination of a gas furnace.)

With the deep dimension of the furnace parallel to trusses, the return air at the bottom via a service box efficiently collects two 12" ducts in a wye, to 16" round inlet. The output at top is then blocked by truss elements at sides, and a  return must not obstruct front service doors and air supply and exhaust ducts. Output duct compromises at top are not acceptable. This furnace is only one truss bay distant from an end wall.Set closer to attic centerline and turned 90°, service doors then facing to left, could then be reached with best-headroom access along the attic centerline.

An in-attic furnace adversely placed parallel to trusses and handily near an end wall. This time, the home owner has ample garage storage, and desire to cope with the bad furnace installation. I need to convince a reader that this installation is wrong. A builder should keep control over how the necessary in-attic work, is done. Perhaps the architect could have provided an interior furnace location. A builder must cope. So here we are. What could have been done differently, within the control of the builder?


An alternative attic with furnace perpendicular to trusses, has excellent access. The found hatch then interferes with duct paths. A better access is possible. 

Here are found conditions of the badly-oriented furnace and badly-conceived HVAC ducts. I have pinioned my back and legs to move through the trusses in the alley at RHS. Appreciate the needless hostility of this attic, and the consequences to decrepit conditions.

This one of the too-few soffit vents moderating temperature in the attic. The house was built in 2001, and I am the first person to see that the cardboard baffles are already falling-down. Two others are fully detached We must ban all that is not durable. Who now will fix this? How? The HVAC flex ducts will disintegrate within the next twenty years. Perhaps they will last much longer, buried, temperature moderated, not under tension. Shouldn't something be done here, now? 

I hope I will be allowed to fix this attic.I hope then that lessons shared, will cause an end of cheap foolishness. I am up here early in the failure cycle, only because there lately has been a loud clicking in the walls at each furnace startup. A 12" steel pipe connecting the attic to the crawl space and carrying conditioned air to floor registers of the first floor, was pinned against ability of thermal expansion. I was able to hack away the wood interference,.


Look for web consensus that an in-attic furnace should be perpendicular to trusses:
An attic furnace should be oriented perpendicular to trusses for best deployment of ducts. 
Hope that this post will be detected, and will lead a change of direction.

Search results include: https://www.greenbuildingadvisor.com/article/dealing-with-ductwork-in-an-unconditioned-attic  (I have joined this conversation.)

https://www.energyvanguard.com/blog/73245/Go-Up-Go-Small-and-Get-Those-Ducts-Inside-Conditioned-Space    (I don't like that duct Medusa.  I think the USDOE reverse bulkhead idea has no traction. How could you get to ducts for service?