The equation itself is exactly right, always — for the narrow set of conditions it was actually derived under. Almost all of the confusion is about applying it outside that set.
P + ½ρv² + ρgh = constantis a statement of conservation of energy per unit volume, tracked along a single streamline of a moving fluid. It says the sum of pressure energy, kinetic energy, and potential energy stays fixed as that fluid moves — which is a genuinely powerful idea. But "conserved along a streamline" is doing a lot of quiet work in that sentence, and the equation only holds when a specific set of assumptions about the flow are actually true.
Bernoulli's equation, in its classic form, assumes: steady flow (conditions at any point don't change with time), incompressible fluid (constant density — fine for liquids and low-speed gas flow, wrong for high-speed compressible gas), inviscid flow (zero viscosity, so no friction losses), and that you're comparing two points along the same streamline. Drop any one of those and the simple equation stops being an exact description of what's actually happening — most commonly, it's the "inviscid" assumption that fails, because every real fluid has some viscosity, and every real pipe has some friction against its walls.
The extended form engineers actually use in pipe-system design adds exactly one thing to Bernoulli: P₁/ρg + v₁²/2g + z₁ = P₂/ρg + v₂²/2g + z₂ + hL, where hL is the head lost to friction and fittings between the two points. That term is calculated from the Darcy-Weisbach equation using a friction factor pulled from a Moody chart (which depends on Reynolds number and pipe roughness) — it isn't something Bernoulli's idealized form can predict on its own, because Bernoulli's derivation explicitly assumes zero viscosity. This is exactly why pump and fan selection isn't "solve Bernoulli and done" — it's reading a manufacturer's pump curve against a system curve built from friction losses, so the pump supplies enough head to overcome both elevation change and friction, not just elevation and velocity change.
Incomplete, and it causes real errors. That pressure-velocity trade-off is only guaranteed along the same streamline, under steady, incompressible, frictionless flow — the exact assumptions behind the equation. It says nothing about what happens when you move along the pipe in a constant-diameter, constant-elevation section, where velocity isn't changing at all — yet real pressure still drops there, continuously, because of friction against the pipe wall. A person who thinks "Bernoulli explains all pressure changes in a pipe" will be confused the first time they measure pressure dropping in a straight run with no area change. Bernoulli explains the velocity-driven part of the trade-off. Head loss explains the friction-driven part.Real systems need both — which is exactly why friction factors, Moody charts, and pump/fan curves exist alongside Bernoulli's equation instead of being replaced by it.
Explains exactly which assumptions Bernoulli's equation depends on — steady, incompressible, inviscid flow along a single streamline — and why real pipe systems need an added head-loss term (friction factors, Moody charts, pump/fan curves) to account for what the idealized equation alone cannot predict.
Bernoulli's equation is often taught as a single memorable statement — faster flow means lower pressure — without equal emphasis on the conditions required for that statement to hold exactly: steady flow, constant density, zero viscosity, and comparison points along the same streamline. Students and early-career engineers then try to apply it to situations where friction dominates, such as long straight pipe runs, and are surprised when measured pressure drop doesn't match the simple equation's prediction of zero change.
Bernoulli's equation is derived from Euler's equation of motion for an inviscid fluid, integrated along a streamline under steady, incompressible conditions: P + ½ρv² + ρgh = constant. Real pipe flow adds a viscous shear stress term at the pipe wall that the inviscid derivation excludes entirely. Engineers account for it separately using the extended energy equation, P₁/ρg + v₁²/2g + z₁ = P₂/ρg + v₂²/2g + z₂ + hL, where hL (head loss) is computed from the Darcy-Weisbach equation, hL = f(L/D)(v²/2g), with the friction factor f found from the Moody chart as a function of Reynolds number and relative pipe roughness.
This distinction is central to pipe network design, pump and fan sizing, HVAC duct design, and venturi/orifice flow metering — anywhere a system combines a Bernoulli-driven pressure-velocity trade-off with friction-driven head loss. It's also a frequent trip-up point on the PE exam's fluid mechanics section, where problems are specifically designed to test whether a candidate applies the idealized equation where an extended, loss-inclusive form is actually required.
Not in its classic textbook form — that version explicitly assumes an inviscid (frictionless) fluid. The version used in real piping and HVAC design is the extended energy equation, which adds a head-loss term (hL) computed separately from the Darcy-Weisbach equation and a friction factor.
Because that constant-along-a-streamline relationship only holds for a frictionless (inviscid) fluid. Real fluids have viscosity, which produces shear stress against the pipe wall and a continuous loss of mechanical energy as heat — a loss the basic Bernoulli equation doesn't include at all, which is why the extended equation adds the hL term.
The basic derivation is strictly along a single streamline. It can be extended across streamlines only under additional conditions — irrotational flow with uniform total energy (Bernoulli's constant the same everywhere) — which hold in many practical situations like flow from a large reservoir, but aren't guaranteed in general.
No — the standard form assumes constant fluid density. For gas flows at higher speeds (commonly cited around Mach 0.3 and above), density changes become significant and a compressible flow energy equation must be used instead.
Because real systems lose mechanical energy to friction (pipe walls, fittings, valves) that the idealized Bernoulli equation assumes away entirely. A pump or fan has to supply enough head to overcome both elevation change and velocity change (the Bernoulli part) and all the friction losses along the way (the head-loss part) — which is exactly what plotting a pump curve against a friction-based system curve is for.
Try our Mechanical Studio
More calculators, simulators, and guides for this discipline.