Ductulator — online duct size calculator
An online ductulator that solves duct size the way the standard actually does: Darcy-Weisbach friction with the Altshul-Tsal friction factor, air density computed from your temperature and site elevation, and the Huebscher equation for rectangular and flat oval sections. Equal friction, velocity reduction and static regain are all built in, and every intermediate value is printed so the answer can be checked by hand.
Selected duct size
16.0 in ø
Equivalent diameter 16.0 in · calculated requirement 14.1 in · governed by the friction rate
| Actual velocity in the section | 3.64 m/s · 716 fpm |
| Friction rate | 0.354 Pa/m · 0.0433 in.wg/100 ft |
| Total pressure drop over the run | 7.0 Pa · 0.028 in.wg |
| Cross-sectional area | 0.1297 m² · 1.396 ft² |
| Hydraulic diameter / perimeter | 16.0 in / 50.3 in |
| Velocity pressure | 7.4 Pa |
| Aspect ratio | n/a — round |
| SMACNA gauge (indicative) | 24 ga · 0.70 mm |
| Sheet metal weight | 7.73 kg/m · 153.1 kg for the run |
| Reynolds number / friction factor | 91,305 / 0.01939 |
| Air density used | 1.1198 kg/m³ |
Velocity is just under the 5–8 m/s band for commercial — general work — quiet and low in pressure drop, at the cost of a slightly larger duct. Fine to build.
Why a ductulator asks for temperature and altitude
The cardboard ductulator wheel that HVAC engineers carried for forty years read a single printed friction chart, and that chart was plotted for standard air — 20 °C at sea level, 1.204 kg per cubic metre. It had no way to ask where the building was. Pressure loss is directly proportional to air density, so on a job at 2000 metres the real loss is about 22 per cent below what the wheel reads, and in a 200 °C kitchen extract it is lower still. The wheel was never wrong; it was answering a different question from the one being asked.
That is the whole reason this tool takes temperature and elevation as inputs and computes density from the standard atmosphere rather than assuming it. The duct size barely moves — velocity is a volume-flow question, not a density one — but the pressure the fan has to overcome moves a great deal, and so does the mass of air actually delivered at a given CFM. On a Gulf or highland project, sizing on sea-level density quietly oversizes the fan and undersizes the cooling.
How to use the ductulator
- Enter the design airflow. The flow carried by this section, not the fan total. Sizing a branch on the main duct flow is the single most common way a duct run ends up oversized.
- Choose the shape. Round for the lowest friction and cost per unit of air moved, rectangular where the ceiling void is shallow, flat oval where the void is tight but round friction is still wanted.
- Pick the sizing method. Equal friction for normal low-velocity work, velocity limit where noise governs. Fill both boxes and whichever needs the larger duct governs — the result says which one did.
- Set the air conditions and the material. Temperature and site elevation set the air density, which scales pressure loss directly. The material sets the absolute roughness, which is the difference between spiral steel and flexible duct.
- Read the size and the velocity together. The selected size comes with the actual velocity in the real section. Check that velocity against the band for the application before accepting the size.
- Open the calculation panel to verify. Every intermediate value is printed with its equation — Reynolds number, friction factor, the branch of Altshul-Tsal taken, and the density used. An answer you cannot check is an answer you cannot defend.
The equations used
- Darcy-Weisbach: Δp = f × (L / D) × (ρV² / 2)
- Altshul-Tsal friction factor: f′ = 0.11 × (ε/D + 68/Re)0.25; f = f′ when f′ ≥ 0.018, otherwise f = 0.85 f′ + 0.0028
- Reynolds number: Re = ρVD / μ
- Huebscher equivalent diameter: De = 1.30 × (ab)0.625 ÷ (a+b)0.250
- Flat oval: De = 1.55 × A0.625 ÷ P0.250, A = πa²/4 + a(W−a), P = πa + 2(W−a)
- Standard atmosphere: p = 101.325 × (1 − 2.25577×10−5 z)5.2559 kPa
- Density: ρ = p ÷ (0.287 T), T in kelvin
- Sutherland viscosity: μ = 1.458×10−6 T1.5 ÷ (T + 110.4)
Worked example
0.5 m³/s through a 300 mm round galvanised spiral duct (ε = 0.09 mm) at 20 °C and sea level. Area = π × 0.3² ÷ 4 = 0.070686 m², so velocity = 0.5 ÷ 0.070686 = 7.074 m/s. Density = 1.2043 kg/m³ and viscosity = 1.8134×10−5 Pa·s, giving Re = 1.2043 × 7.074 × 0.3 ÷ 1.8134×10−5 = 140,932. Then f′ = 0.11 × (0.0003 + 68/140932)0.25 = 0.018398, which is above 0.018, so f = f′. Friction rate = 0.018398 × (1.2043 × 7.074² ÷ 2) ÷ 0.3 = 1.848 Pa/m. Over a 20 m run that is 37.0 Pa.
Chart-read ductulators typically show about 1.7 Pa/m for this case. The 8 per cent difference is the chart, not the maths — reading a logarithmic nomogram by eye carries that much error, which is precisely why the working is printed here.
Recommended duct velocities
| Application | m/s | fpm |
|---|---|---|
| Residential | 3–5 | 591–984 |
| Commercial — low noise | 4–6 | 787–1181 |
| Commercial — general | 5–8 | 984–1575 |
| Industrial | 8–12 | 1575–2362 |
| Branch / runout | 2–4 | 394–787 |
Main duct figures. Branches and runouts sit lower; final runouts to a diffuser are usually held to 2–4 m/s so the terminal, not the duct, sets the sound level. See the duct velocity chart for the full breakdown by position in the system.
Absolute roughness of duct materials
| Material | ε (mm) | ε (ft) |
|---|---|---|
| Galvanized steel, spiral seam | 0.09 | 0.00030 |
| Galvanized steel, longitudinal seam | 0.15 | 0.00049 |
| Aluminium | 0.05 | 0.00016 |
| Stainless steel | 0.05 | 0.00016 |
| PVC / plastic | 0.01 | 0.00003 |
| Fibrous glass duct board (rigid) | 0.90 | 0.00295 |
| Fibrous glass liner, air side | 1.50 | 0.00492 |
| Flexible duct, fully extended | 3.00 | 0.00984 |
| Flexible duct, 70% extended | 6.00 | 0.01969 |
| Concrete | 3.00 | 0.00984 |
| Brick | 3.00 | 0.00984 |
Values per ASHRAE Fundamentals Chapter 21. The jump from 0.09 mm for spiral steel to 3.0 mm for fully extended flexible duct is a factor of 33, and it is the single largest lever on the answer this tool gives.
Things to keep in mind
- This sizes straight duct only. Elbows, tees, transitions and dampers carry their own losses, and on a congested run they routinely exceed the straight-duct friction. Add fitting losses before selecting a fan.
- Equal friction does not self-balance. Velocity falls as flow drops along the run, so downstream branches see more static pressure than upstream ones. Long systems need dampers, or static regain sizing instead.
- Use the real velocity for noise. The equivalent round velocity belongs to the friction calculation. Acousticians want Q divided by the actual section area.
- Gauge output is indicative. Reinforcement spacing, joint class and duct support all feed the real selection. Confirm against the SMACNA construction tables before issuing for fabrication.
- Standard sizes differ by market. The metric list follows EN 1506 and the imperial list follows common North American spiral sizes. They do not map onto each other, so a metric and an imperial answer for the same duty will land on genuinely different sizes.
Frequently asked questions
What friction rate should I design a duct to?
For commercial low-velocity supply, 0.8 to 1.0 Pa per metre — 0.08 to 0.10 inches of water gauge per 100 feet — is the normal band. Use a lower rate on long runs where fan energy over the life of the plant outweighs the sheet metal saving, and lower again where a low sound level is specified. Above roughly 1.5 Pa per metre you are usually buying noise and fan power in exchange for a smaller duct.
Is 0.8 Pa/m the same as 0.08 in.wg per 100 ft?
No, and the difference matters. Converted exactly, 0.8 Pa per metre is 0.0979 inches of water gauge per 100 feet, while 0.08 in.wg per 100 ft is 0.654 Pa per metre — about 22 per cent apart. They are two separate rules of thumb that happen to look like the same number, and design guides in each unit system quote their own. This ductulator converts honestly rather than snapping one to the other, so a metric and an imperial user get the same physics, not the same digits.
Why is the velocity in my rectangular duct different from the equivalent round duct?
Because the Huebscher equivalent diameter is defined on an equal-friction, equal-airflow basis, not equal area. The equivalent round duct loses the same pressure per metre as your rectangular duct carrying the same air, but its cross-sectional area is smaller, so its velocity is higher. Use the equivalent diameter for friction, and the real rectangular area for velocity, noise and terminal selection.
Does duct aspect ratio matter if the area is the same?
Yes. A 1000 by 250 duct and a 500 by 500 duct have the same area, but the flat one has more perimeter, so more friction, more sheet metal, more insulation and more hanger and reinforcement work. Keep the aspect ratio at or below 4 to 1 wherever the ceiling void allows. Above 6 to 1 the cost and pressure penalty is severe enough that splitting the run into two ducts is usually cheaper.
How much does flexible duct really cost in pressure?
A great deal. Fully extended flexible duct has an absolute roughness around 3.0 mm against 0.09 mm for spiral steel, and published data shows that compressing it to about 70 per cent of its extended length multiplies the pressure drop several times over. Either size flexible duct on the compressed figure, or keep runs short and pull them genuinely tight on site.
Why does site elevation change the answer?
Pressure loss is proportional to air density, and density falls with altitude. At 2000 metres the density is roughly 22 per cent below sea level, so the same airflow through the same duct produces proportionally less pressure drop. The duct size barely moves, but the fan sees a lower resistance — and, importantly, the mass of air delivered at the same volume flow is lower too, so the cooling capacity at a given CFM drops with it.
Why not use Hazen-Williams for duct sizing?
Because it is a water correlation. Hazen-Williams is an empirical fit calibrated for water near ambient temperature in full pressurised pipes, and it carries no term for fluid density or viscosity. Applied to air it produces confident nonsense. Duct friction is solved with Darcy-Weisbach, which is physically grounded and takes density and viscosity as inputs — which is exactly why this tool asks for the air temperature and the site elevation.
Is a ductulator the same thing as a duct calculator?
Yes — the name comes from the cardboard slide-rule wheel that HVAC engineers carried for decades, and it is often written ductilator or duct-u-lator. The wheel read a friction chart plotted for standard air at sea level. This version solves the same equations directly, at your actual air density, and shows the working the wheel never could.
Last updated: 26 July 2026