Letter to Code Officials: Duct Design Issues
"Nothing is impossible for the man who doesn't have to do it himself."
- A. H. Weiler"Insanity is doing the same thing over and over again but
expecting different results."
- Albert Einstein, among others
An open letter
Dear Building Official:
We know the new energy code - the 2000 IECC - has landed. It's having an
impact in a growing number of Colorado jurisdictions.
Like all codes, it's a work in progress, adapting over time to new
information. Sometimes that new information is counter-intuitive.
What's described below is some relatively new info about ductwork to factor
into your code requirements and options. Longmont has proposed a related
amendment to its 2000 IRC, which may be in place in 30 to 60 days. They will
allow installation of simplified return-air ductwork. We hope you'll consider it
as well.
The problem
Testing along the Colorado Front Range shows that our ductwork is leaky. Very
leaky.
You can't tell how tight or leaky ductwork is just by looking at it. The only
way to establish tightness is to test it. This isn't normally done, and it's not
being recommended for code.
Testing duct tightness consists of temporarily taping off all supply and
return grilles, then pressurizing the ducts with a small blower called a "duct
blaster" to a specified test pressure (25 pascals of air pressure). A technician
then measures the amount of air flow-all of which is leaking out of the sealed
ductwork. An additional test determines how much of that duct leakage is
connected to the outdoors.
During 1997, the NAHB Research Center published A Builder's Guide to
Residential HVAC Systems. Their assessment of duct tightness is shown in
Table 1, with Front Range comparison data at the bottom.
As the table illustrates, ductwork in Colorado leaks conditioned air to the
outdoors at just about the average rate for houses nationwide. But our ductwork
ranks well into the "loose" category in terms of overall tightness. While no one
agency or organization has collected and analyzed duct-testing data nationwide,
a recent phone survey by this writer of 10 testing experts nationwide indicates
our ductwork is extremely leaky. Sample quotes:
"I'm not aware of any sample that is leakier (than Colorado's)," said Jamie
Lyons, a research engineer with NAHB's Research Center.
Joe Lstiburek, with Building Science Corp. (Westfield, Mass.) and the U.S.
Dept. of Energy's Building America program, said, "I've tested hundreds of
systems, and the only place that comes close to Colorado's duct leakage is New
Jersey. On a bad day, New Jersey's ductwork is as bad as yours."
"Of the data I've seen, Colorado's would be on the extreme end of duct
leakage," said Ian Walker, scientist at Lawrence Berkeley National Laboratory.
Note: this is relatively new information. No one deliberately aimed for this
result. It just happened. Here's how.
Prime suspect
The rationale that got us and other northern-tier states where we are today
goes like this: "if ductwork is located inside the conditioned space-mostly in
basements-then the leaks simply don't matter very much."
There's a specific culprit here: return-air ducts. Field test data shows that
returns tend to be 50 to 100 percent more porous than supplies. In turn, what
causes this leakage disparity is our widespread use of building cavities for
routing return air back to the furnace.
Think of a master suite above a garage. To return air from that room to the
furnace, you enclose one or two wall cavities on a partition wall, drop the air
down through a fully cut-out bottom plate and down into a floor cavity. Then air
drops down through another wall cavity, into a basement floor cavity that is
panned, carrying the air to the return-air plenum. All the way along, any
penetrations of the cavities have to be sealed and the drywall should have
gaskets or adhesive on all six sides. Then the panning should extend all the way
over the top of the supply plenum (almost never happens) before it drops down
into a return plenum that sags were the hole is cut.
True, that's the worst case, but returns from bedrooms over garages hardly
ever draw much air. In fact, testing data shows that roughly two-thirds of the
air returning to the furnace enters through unintended leaks, not through
intentional grilles. Most second floor grilles are either not connected at all
or are barely connected to the furnace.
Please realize that it is damned near impossible to consistently build tight
return-air ductwork the way it is being designed and installed today.
Possible consequences
Return-air leaks in basements normally create negative pressure. Depending on
the amount of leaks, this negative pressure can exceed the positive pressure of
gases being vented up the flue. Data gathered during 1999 for Fort Collins
Utility's housing performance study indicated that negative pressure in
one-third of randomly selected new homes is sufficient to pose a risk of
back-drafting and spillage of flue gases down water heater flues.
Associated comfort problems (greater than 3-degree F. temperature differences
between rooms) are also commonplace. Between 70 and 80 percent of homeowners
interviewed during the study complained of comfort problems in second-floor
living spaces-either too cold in winter or too hot in summer (two-thirds of all
homes in the study were equipped with air conditioners). Temperature data
gathered in the field supported the scope of the problem.
Consider two stories of comfort complaints brought to this writer's attention
last week. In December, purchasers of a new home near Telluride contacted San
Miguel County's chief building official about comfort problems. He suggested
they get their home tested by an energy rater. Tests showed the return-air duct
in the crawl space was extremely leaky. Once located, fixing the problem was
straightforward, but only because the offending ductwork was accessible.
Dave Van Allen, Longmont's chief building official, relates the recent story
of a local builder (we'll call him Bob). After Bob moved into one of his own
homes, he heard a string of complaints from his wife about the cold bathroom
above the garage. After wrestling with this complaint, Bob told Van Allen, "I'm
going to change our company's entire philosophy. We're going to be a leader in
energy efficiency, and especially in the latest technologies for heating
systems. We're going to lead by example."
The emerging alternative
Working with Joe Lstiburek and the Building America program, Engle Homes is
the first large production builder to opt for simplified return-air systems.
Both Oakwood Homes and Centex Homes are using the scheme in a new project.
Here's how it works. For single-furnace applications, hard duct or flex duct is
run to both high and low levels of the living space. Those ducts are tied
directly into the vertical return-air drop beside the centrally located furnace.
That eliminates the entire horizontal return-air plenum normally attached to the
basement ceiling, with floor panning, etc. Proper design is needed to prevent
the potential for increased blower noise, because all return air is dropping
nearly straight down from the grilles to the furnace. A simplified central
return requires some minor floor-plan modification.
How is pressure relieved in each bedroom and air returned to the furnace?
There's a choice. A contractor can screw on a pair of transfer-air grilles
between the room and a hall or open space, offsetting the grilles high and low
to reduce sound transmission. Or the contractor can install short jump ducts at
the ceiling level, across that same partition wall (using flex duct dampens
noise). For upper level bedrooms, the jump ducts can be buried in attic
insulation. From either of these options, air then flows to the pair of central
return-air grilles. Engle tests its ductwork to assure tightness.
What we're asking for
The old system isn't working. We didn't know how bad it was until air
conditioning became a widespread consumer preference. Now, some HVAC equipment
distributors bring in three air conditioners for every furnace. And more and
more consumers are having oversized AC systems installed to try to compensate
for leaky ductwork.
But here's a rub: several code jurisdictions seem to be insisting on the
installation of individual return-air ducts to all bedrooms on all levels. This
is a step backward, not forward. Requiring old-style. "pretzelized" return-air
ducts doesn't mean they'll work as specified. Please allow the design and
installation of simplified ductwork systems. Please let Bob's guys build a
better ductwork system for his buyers.
Steve Andrews consults with builders for E-Star
Colorado and writes on energy issues (sbandrews@att.net).
E-Star (www.e-star.com),
is a nonprofit home energy rating system that works with both new and
existing homes statewide.
|
