Duct Heat Loss and Gain Calculator

A supply run through a 130 degree attic in August or a 25 degree crawl space in January is a heat exchanger nobody designed. It has an area, it has an R-value, and the air spends a fixed number of seconds inside it — that is enough to work out what arrives at the register rather than what left the plenum. The answer is usually smaller than people fear and larger than the drawing assumed, and it is very sensitive to one thing that has nothing to do with insulation: how fast the air is going. Halve the airflow in the same duct and the air spends twice as long losing heat.

The flow through this length. This matters more than most people expect — the air in a slow duct has longer to give heat away.
Only the part outside the thermal envelope. Duct buried in insulation or inside a conditioned soffit is a different situation and is not this length.
The value the product data gives at the thickness it is actually installed at. Duct wrap compressed by a strap or lying on a joist is thinner than its label where it is compressed.
The still-air film on the outside of the jacket. Used only to work back to a jacket surface temperature. A windy or fan-blown attic reads lower.
Measured in the plenum or at the start of this length. A furnace supply and a heat pump supply are not the same temperature.
Attic, crawl space, garage or unheated basement, at the condition you care about. A summer attic and a winter attic are two separate runs of this page.
Optional, and only interesting in cooling. Compared against the jacket surface temperature and reported as a margin in degrees, with no verdict attached.
Optional. Enter another R and the page prints what the run does at that instead. Set to 0 to leave it out.
Optional, for the energy figure. Blower run hours, not hours in the season.
Optional. All-in delivered rate off your own bill, divided by the efficiency of whatever makes the heat if you want the figure at the meter rather than at the duct.
Duct Heat Loss Calculator — Attic and Crawl Space RunsBuildFigure

Why the answer is an exponential and not a subtraction

Air moving down a duct in a colder space loses heat at a rate set by how far it is above that space. As it cools, the difference shrinks and the loss slows, so the temperature approaches the surrounding air along a curve rather than dropping at a fixed number of degrees per foot. The whole run collapses to one exponent: the conductance of the duct divided by the heat capacity flow rate of the air.

Work the default through. Thirty feet of 8 in round has 62.8 square feet of outer surface. R-8 of wrap plus 0.68 of outside film is 8.68 of resistance, so the conductance is 7.24 Btu per hour per degree. Four hundred CFM of standard air carries 432 Btu per hour per degree. The ratio is 0.0168, the exponential of minus that is 0.9834, and the air closes 1.66 percent of the 80 degree gap — it arrives at 118.7 F having entered at 120, and 574 Btu per hour went into the attic.

Airflow is the input people forget

Insulation is where attention goes and airflow is where the sensitivity is. Drop that same run from 400 CFM to 150 — a branch with the damper most of the way shut, or a blower on a low tap — and the exponent nearly triples. The air now closes 4.4 percent of the gap and arrives at 116.5 F. The heat crossing the wall barely changes at 567 Btu per hour, because the duct is the same duct in the same attic, but the drop at the register goes from 1.3 degrees to 3.5 — two and a half times as far.

This is the mechanism behind a familiar complaint: the far room is cold, someone chokes the near branches to push more air to it, and the far room gets marginally more air at a noticeably lower temperature. The two effects partly cancel and the occupant is unimpressed.

The outside film is a bigger share than it looks

Bare sheet metal in still attic air has essentially only the film, around 0.68. Wrapping it with R-4 takes the total to 4.68 and cuts the loss from 6,655 Btu per hour to 1,058 — to about a sixth. Going from R-4 to R-8 takes the total from 4.68 to 8.68 and cuts it to 574, by 46 percent rather than by half. The first layer is doing an enormous amount of work and each subsequent layer does less, which is the reason the page prints both R-values side by side rather than a rule about which is right.

The same arithmetic explains why compressed wrap is worse than it seems. Insulation strapped tight over a joist or squashed where the duct passes a truss is not at its label thickness at those points, and heat crosses at the thin places preferentially.

What it does not know

Leakage. A duct that loses 574 Btu per hour through R-8 wrap and dumps five percent of its air into the attic is losing roughly 1,730 Btu per hour out of the joints as well, and the joints are not in this arithmetic at all. Conduction is the tractable part; leaks want a pressure test.

It also does not know your attic. Attic temperature is entered once and held for the whole run, and a real attic swings sixty degrees through a summer day, gets warmer near the ridge, and is warmed locally by the duct itself. Treat the answer as the run at that one condition, and run the page twice — once for a design winter morning and once for a design summer afternoon — rather than looking for one number that covers both.

Questions people ask

How much heat does an attic duct lose?

For thirty feet of 8 in round at R-8 carrying 400 CFM of 120 F air through a 40 F attic, about 574 Btu per hour, and the air arrives 1.3 degrees cooler than it left. The two numbers move differently: the heat depends on the duct and the attic, while the temperature drop depends on how fast the air is going through it. Slow the same duct to 150 CFM and the drop goes from 1.3 degrees to 3.5 while the heat stays near 567.

Why does the temperature drop follow a curve rather than a straight line?

Because the loss at any point depends on how far the air still is above the surrounding space, and that gap shrinks as the air cools. The result is an exponential approach to the temperature of the attic or crawl space, set by the duct conductance divided by the heat capacity flow of the air. Doubling the length does not double the loss, and halving the airflow doubles the exponent.

Does going from R-4 to R-8 halve the duct loss?

No, and the reason is the outside air film. There is roughly 0.68 of resistance on the outside of the jacket before any insulation, so R-4 of wrap is 4.68 of total resistance and R-8 is 8.68. The loss falls by about 46 percent, not 50. Going from bare metal to R-4, by contrast, cuts it by 84 percent, which is why the first layer is worth so much more than the last.

Will an insulated duct sweat in an attic?

That is not a question arithmetic can settle, and this page does not try. It reports the jacket surface temperature the resistance split implies and puts it beside the dew point you enter, as a margin in degrees. Whether condensation actually forms depends on the humidity through the whole day, on whether the vapour barrier is continuous and sealed at every joint and boot, and on what happens at the fittings where the wrap is thinnest.

Should I count duct that is buried in attic insulation?

Not as part of this length. Duct buried under blown insulation is in a different thermal situation, with a resistance that depends on the depth and the coverage over it and a moisture situation of its own. Enter only the length that is exposed to the unconditioned air, and treat buried runs separately.

Related