Duct sizing is the relationship between three quantities: airflow, duct size and pressure drop. Fix any two and the third follows.
The starting figures
Rough capacities for smooth rigid round metal duct at a friction rate around 0.08–0.10 in. w.c. per 100 feet:
| Diameter | Approx. CFM |
|---|---|
| 4" | 25 |
| 5" | 50 |
| 6" | 100 |
| 7" | 150 |
| 8" | 220 |
| 9" | 300 |
| 10" | 400 |
| 12" | 700 |
| 14" | 1,100 |
| 16" | 1,700 |
These are the numbers most quick charts give. They are a starting point rather than an answer, for the reason below.
Friction rate is chosen, not assumed
Friction rate is the pressure drop the design allows per 100 feet of duct.
It is calculated from the equipment's available static pressure — the total external static pressure it is rated for, minus the losses in the coil, the filter, the registers and the grilles — divided by the total effective length of the longest run, including the equivalent length of every fitting.
friction rate = (available static pressure × 100) ÷ total effective length
A house with a short, simple duct system can be designed at a high friction rate and use smaller ducts. A house with a long run and many fittings must use a lower friction rate and therefore larger ducts to carry the same air.
Using a fixed 0.10 for every job, which is what a chart implicitly does, undersizes long systems and oversizes short ones.
Equivalent length is where systems fail
Fittings behave like additional straight duct. An elbow, a takeoff, a boot, a register — each has an equivalent length published in ACCA Manual D, sometimes 20 to 60 feet for a single fitting.
A run of 30 feet of duct with six fittings can have a total effective length of 200 feet. Sizing it as 30 feet produces ducts far too small, and this is the single most common cause of a system that measures high static pressure.
Velocity limits
Even where friction allows a smaller duct, velocity has to stay within limits or the system will be noisy:
Supply trunk, roughly 700–900 feet per minute in a residence. Supply branches, around 600. Return trunk, 600–700. At the grilles, lower still.
Noise is the usual first complaint about an undersized duct, well before anyone measures anything.
Flexible duct is not equivalent
Flex duct has considerably more friction than smooth metal of the same diameter — the corrugated inner liner disrupts the flow.
Design tables treat it separately, and fully stretched flex carries far more than the same duct compressed. Flex left slack in an attic, sagging between supports, can lose a large fraction of its rated capacity.
A common field failure is flex sized from a rigid-duct chart and then installed loose.
Rectangular duct
Rectangular ducts are sized by equivalent round diameter, and an aspect ratio far from square carries less air than its area suggests because there is more wall surface per unit of flow.
Keeping the aspect ratio below about 4:1 avoids most of that penalty.
The right sequence
Room-by-room load calculation, giving the CFM each room needs. Then the total effective length of the longest run. Then the friction rate. Then the sizes.
That is Manual D, and it is a different exercise from reading a chart — which is why two duct systems carrying identical airflow in two different houses are correctly built with different duct sizes.