Duct Size & Air Velocity Calculator
Duct diameter against air velocity — an inverse-square law, so doubling the duct quarters the speed and one size up fixes a noisy run.
Velocity is only half of duct sizing — the other half is friction loss along the run, which depends on length, fittings and material and is not calculated here. A duct can be quiet and still starve a register at the far end. Nothing is uploaded.
Velocity goes as one over the diameter squared
Area is π r², so doubling a duct quadruples its area and quarters the velocity — exactly, not roughly. 400 CFM through four inches is 4584 feet a minute, a noise you can hear two rooms away. The same air through twelve is 509, which is inaudible.
| Diameter | Area | Velocity at 400 CFM | Against 6 in |
|---|---|---|---|
| 4 in | 0.087 sq ft | 4584 fpm | 2.25× |
| 5 in | 0.136 sq ft | 2934 fpm | 1.44× |
| 6 in | 0.196 sq ft | 2037 fpm | 1.00× |
| 8 in | 0.349 sq ft | 1146 fpm | 0.56× |
| 10 in | 0.545 sq ft | 733 fpm | 0.36× |
| 12 in | 0.785 sq ft | 509 fpm | 0.25× |
| 16 in | 1.396 sq ft | 286 fpm | 0.14× |
Which is why one nominal size is never a small compromise. Dropping from six inches to five raises the velocity by 44% — the sort of change that turns a quiet run into an audible one, from a decision that looks like rounding.
Which cuts the other way too, and rectangles are worse than their area
Because the relationship is squared, going the other way is a square root: halving the velocity needs only 1.41 times the diameter. That is why going one size up fixes a noisy run so effectively — a small change in the duct is a large change in the air.
| Velocity target | Diameter needed at 400 CFM |
|---|---|
| Quiet — a bedroom branch — 500 fpm | 12.1 in |
| Normal residential branch — 700 fpm | 10.2 in |
| A main trunk — 900 fpm | 9.0 in |
| Commercial trunk — 1200 fpm | 7.8 in |
| Loud, and losing pressure — 1800 fpm | 6.4 in |
And a rectangular duct is not worth its area. A 10 by 10 has the same 100 square inches as a 20 by 5, but their friction-equivalent round diameters are 10.9 and 10.3 inches — the flatter one behaves like a smaller pipe, because more of the air is close to a wall. Aspect ratio costs you something, every time.
How to use
- Enter the airflow in CFM.
- Enter the duct diameter in inches.
- Read the velocity in feet per minute.
- Or set a target velocity and read the diameter it needs.
Frequently asked questions
How do I work out duct velocity?
Airflow divided by cross-sectional area. For a round duct the area is pi times the radius squared, so 400 CFM through an eight-inch duct is about 1,150 feet a minute.
Why does duct size matter so much?
Because area goes as the square of the diameter, so velocity goes as one over the diameter squared. Doubling a duct does not halve the velocity, it quarters it — exactly.
Is one nominal size a small change?
Never. Dropping from six inches to five raises the velocity by about 44%, which is the sort of change that turns a quiet run into an audible one from a decision that looks like rounding.
What velocity should I aim for?
Roughly 500 feet a minute for a quiet bedroom branch, 700 for a normal residential branch, 900 or so for a main trunk. Past about 1,200 it starts to be audible, and past 1,800 it is both loud and losing pressure.
How much bigger to halve the velocity?
Only 1.41 times — the square root of two. That is why going one size up fixes a noisy run so effectively: a small change in the duct is a large change in the air.
Is a rectangular duct the same as a round one of equal area?
No, it is worse. A 10 by 10 and a 20 by 5 both have 100 square inches, but their friction-equivalent round diameters are about 10.9 and 9.7 inches — the flatter one behaves like a smaller pipe, because more of the air is close to a wall.
Does velocity tell me everything about a duct?
No — the other half is friction loss along the run, which depends on length, fittings and material and is not calculated here. A duct can be quiet and still starve a register at the far end.
Does this send anything anywhere?
No. Every figure is computed in your browser, and nothing is uploaded or stored.
🔒 This tool runs entirely in your browser. Nothing you enter is uploaded, logged, or stored.