And why doubling a fan's speed doesn't just double the airflow.
Every duct system carries two different kinds of pressure at once, and a single gauge reading "0.8 in. w.g." on a system means nothing until you know which one it's measuring. One component does the actual work of pushing air through resistance. The other exists purely because the air is moving. Confusing the two — or assuming that speeding up a fan a little only costs a little more energy — is one of the most common and most expensive misunderstandings in duct design and fan selection.
Static pressure (SP) is the potential-energy component of air pressure in a duct — it pushes outward equally in all directions, the way pressure in an inflated balloon does, whether or not the air inside is moving. It's the component that does the actual work of overcoming the resistance of ductwork, fittings, filters, and coils, and it's measured with a gauge or manometer tap held perpendicularto the direction of airflow, so the moving airstream doesn't contribute anything to the reading.
Velocity pressure (VP) is the kinetic-energy component — it exists specifically because the air has velocity, and it vanishes if the air stops moving. It's measured facing directly into the airflow (a pitot tube is the classic instrument), and for standard air it relates to velocity by a simple formula: VP = (V / 4005)², with VP in inches of water gauge and V in feet per minute. Add the two together at any point in the system and you get total pressure (TP = SP + VP) — the full mechanical energy the air is carrying at that point.
Notice what happens at the constriction: the duct narrows, the same volume of air has to move through a smaller area, so velocity — and therefore velocity pressure — spikes locally. At the same instant, static pressure drops. That's not a coincidence or a measurement error; it's energy converting from one form (potential/static) to another (kinetic/velocity), the same trade-off that shows up in Bernoulli's equation for any flowing fluid. As the duct widens back out downstream, velocity falls again and some of that velocity pressure converts back into static pressure — though never fully, because the transition itself costs a bit of total pressure to turbulence. Meanwhile, total pressure — static plus velocity, added together at any single point — only ever decreases as air travels through the system, consumed continuously by friction against duct walls and losses through fittings.
A centrifugal or axial fan running on a fixed duct system follows three simple scaling relationships — the fan laws (or affinity laws) — whenever its speed changes and everything else about the system stays the same. Airflow (CFM) is directly proportional to fan speed (RPM). Pressure — static, velocity, or total — is proportional to the square of fan speed. And the power required to drive the fan is proportional to the cube of fan speed. Those three exponents (1, 2, 3) look like a minor technicality until you plot them side by side.
Airflow at a given point in a rotating fan is proportional to the blade tip speed, which is proportional to RPM — hence CFM scales linearly. Velocity pressure itself is proportional to the square of velocity (that's exactly what VP = (V/4005)² says), and on a fixed duct system, faster fan speed means proportionally faster air, so pressure scales with RPM squared. Power is force times velocity, and since the "force" here (pressure) already scales with RPM² while flow rate scales with RPM, multiplying the two gives power scaling with RPM³. None of this is a coincidence — it's the same relationship showing up three times at three different exponents, which is exactly why a "small" speed increase produces a disproportionately large jump in energy cost. Going from 100% to 120% RPM only sounds like a 20% bump — but power required climbs by a factor of 1.2³ ≈ 1.73, a 73% increase in energy consumption for a 20% airflow gain.
Half right. Doubling fan speed does double the airflow (CFM) — that part scales linearly, exactly as intuition suggests. But the pressure the fan develops against the duct system quadruples (RPM squared), and the power the motor has to deliver increases eightfold(RPM cubed) — not double. This is a critical distinction for anyone bumping up a fan's speed (via a VFD, pulley change, or drive adjustment) to chase more airflow: the airflow benefit scales gently, but the motor sizing, drive train stress, and energy-cost implications scale far more steeply. It's also exactly why the fan laws work just as powerfully in reverse — slowing a fan down even modestly (as VAV and demand-controlled ventilation systems do) saves disproportionately more energy than the airflow reduction alone would suggest.
Explains the two components of air pressure in a duct system — static pressure (potential energy, does the work of overcoming resistance) and velocity pressure (kinetic energy, exists because the air is moving) — and the fan laws that describe how airflow, pressure, and power scale with fan speed, including why doubling fan speed quadruples developed pressure and requires eight times the power.
Most gauge readings and duct calculations quote a single pressure number, which makes it easy to forget that duct air pressure has two distinct components with different physical meanings and different measurement methods. Static pressure is measured perpendicular to flow and represents the potential energy doing the work of pushing air through resistance; velocity pressure is measured facing directly into the airflow (a pitot tube) and represents kinetic energy that exists only because the air is moving. The two convert back and forth as air speeds up or slows down through the system (a constriction raises velocity pressure and lowers static pressure locally), while their sum — total pressure — only ever decreases downstream of the fan due to friction and fitting losses. Separately, the fan laws (airflow scales linearly with RPM, pressure with RPM squared, power with RPM cubed) are frequently underestimated: a modest speed increase intuitively feels like it should cost a modest amount more energy, but the cubic power relationship makes the actual cost far steeper.
Total pressure (TP) equals static pressure (SP) plus velocity pressure (VP) at any point in a duct: TP = SP + VP. Velocity pressure for standard air relates to velocity by VP = (V / 4005)², with V in feet per minute and VP in inches of water gauge. As air accelerates through a constriction, continuity requires velocity to increase, so VP increases and — since total pressure can only stay level or fall — SP correspondingly drops; as the duct widens back out and velocity falls, some VP converts back to SP, though not completely, since the transition itself dissipates some total pressure as turbulence (analogous to Bernoulli's equation for an incompressible fluid, with friction losses added).
The fan laws (affinity laws) apply when a fan's speed changes while it continues operating on the same system curve, at a dynamically similar operating point, with unchanged air density: CFM₂/CFM₁ = RPM₂/RPM₁ (airflow scales linearly with speed); SP₂/SP₁ = VP₂/VP₁ = TP₂/TP₁ = (RPM₂/RPM₁)² (all pressures scale with the square of speed); and BHP₂/BHP₁ = (RPM₂/RPM₁)³ (power scales with the cube of speed). Doubling RPM therefore doubles CFM, quadruples pressure, and multiplies required power by eight.
Static vs. velocity pressure matters directly in duct design and testing/balancing: total external static pressure (TESP) across an air handler is what selection software and manufacturer fan curves are built around, while velocity pressure and pitot-tube traverses are how field technicians actually measure airflow (CFM) in a duct when a direct flow meter isn't practical. The fan laws matter every time a fan or blower speed is adjusted — via a VFD, sheave/pulley change, or damper — to hit a different airflow target: they explain why variable-speed drives and demand-controlled ventilation save disproportionately more energy at partial speed than the airflow reduction alone suggests, and conversely why "just speeding up the fan a bit" to fix an airflow shortfall can quietly blow past a motor's rated horsepower.
Static pressure is the potential-energy component of air pressure — it pushes equally in all directions and does the work of overcoming resistance from ductwork, fittings, filters, and coils. Velocity pressure is the kinetic-energy component that exists only because the air is moving, measured facing into the airflow, and it converts to and from static pressure as the airstream speeds up or slows down through the system.
For standard air (0.075 lb/ft³), VP = (V / 4005)², where V is velocity in feet per minute and VP comes out in inches of water gauge. This relationship, plus a pitot-tube traverse, is the basis for measuring airflow in a duct when a dedicated flow station isn't available: measure VP, solve for V, multiply by duct cross-sectional area to get CFM.
Because the same volume of air has to move through a smaller cross-sectional area, its velocity must increase (continuity of mass flow). Velocity pressure rises with the square of velocity, and since total pressure can't increase on its own, static pressure has to give up energy to velocity pressure at that point — a direct application of the Bernoulli-type energy trade-off, with total pressure still trending downward overall due to friction and fitting losses.
Per the fan laws (on the same duct system): airflow (CFM) doubles, since it scales linearly with RPM. Pressure developed (static, velocity, or total) quadruples, since it scales with RPM squared. Power required increases eightfold, since it scales with RPM cubed. The airflow gain is modest compared to the power penalty, which is why speeding up a fan significantly to chase more airflow can require a much larger motor and drive than expected.
No — the fan laws as stated here specifically describe how a single fan's performance scales with its own speed on an unchanged system curve. Changing the system (adding resistance, opening/closing dampers) instead moves the operating point along the fan's existing performance curve at a fixed speed, which is a different calculation (reading the fan curve at the new system curve intersection), not a fan-law speed scaling.
Try our HVAC Studio
More calculators, simulators, and guides for this discipline.