Sudden Pipe Expansion: Keep the Loss Coefficient and Reference Velocity Together

AI-generated illustrative scene, not a product screenshot or a completed-project photograph. The worked calculation below uses explicit example inputs.
A pipe-loss schedule can look internally consistent while carrying the wrong reference velocity into a fitting calculation. This is particularly easy at an abrupt enlargement: the same water flow passes through two different bores, so there are two different mean velocities. A coefficient copied from one convention cannot be combined with the velocity from the other convention.
This worked example addresses a sudden, concentric enlargement in a full water pipe. It is not a catalog loss value for a gradual reducer, a valve, a partly full drain or a particular manufactured fitting. The objective is to make the calculation basis auditable before the fitting losses are added to a pump-head schedule.
Start with continuity, not the nominal connection labels
Assume steady incompressible flow of 12 L/s through an actual upstream internal diameter of 80 mm and a downstream internal diameter of 125 mm. Both sections are horizontal. Use gravitational acceleration g = 9.81 m/s². These dimensions are deliberately specified as internal bores; nominal pipe labels alone would not establish either flow area.
Q = 0.012 m³/s; A1 = π × 0.080² / 4 = 0.00502655 m²; A2 = π × 0.125² / 4 = 0.01227185 m².
V1 = Q / A1 = 2.3873 m/s; V2 = Q / A2 = 0.97785 m/s.
The volumetric flow is unchanged, but the downstream velocity is about 41% of the upstream velocity. Retain a separate area and velocity column for each section instead of overwriting one shared velocity cell when the pipe size changes.

Original assumed 12 L/s example; sudden-enlargement model only.
Calculate the irreversible loss directly
The classical Borda–Carnot sudden-enlargement model expresses the head loss in terms of the velocity difference. IIT Guwahati's NPTEL fluid-flow teaching material develops this relationship. It is a simplified separated-flow model, not a general substitute for manufacturer data or a detailed flow analysis where the geometry and velocity profiles depart from its assumptions.
hL = (V1 − V2)² / (2g) = (2.3873 − 0.97785)² / 19.62 = 0.10125 m of water head.
This is the irreversible mechanical-energy loss assigned to the enlargement. It is not simply the difference between two static-pressure gauge readings. The downstream section also has a lower velocity head, and some of that reduction appears as static-pressure recovery. Label the result as head loss so that a reviewer does not confuse these quantities.
Express the same answer on either velocity basis
With upstream velocity: K1 = (1 − A1/A2)² = (1 − 0.4096)² = 0.348572; hL = K1 V1² / (2g) = 0.10125 m.
With downstream velocity: K2 = (A2/A1 − 1)² = (2.441406 − 1)² = 2.077652; hL = K2 V2² / (2g) = 0.10125 m.
The two coefficients look very different because their reference velocities differ. They describe the same modeled loss. A comparison table that ranks fittings only by K, without reporting the reference diameter and velocity, can therefore be misleading.
If the spreadsheet mistakenly combines K1 with V2, it returns approximately 0.0170 m. That is not a more efficient fitting; it is an inconsistent calculation. An independent check using the velocity-difference equation exposes the mistake without requiring a second coefficient table.
Build a reviewable fitting row
Record the fitting type, flow direction, upstream bore, downstream bore, coefficient source, reference velocity and resulting head loss in one row. Keep the area ratio visible rather than burying it inside an unexplained constant. When the route contains several sizes, reference each fitting row to the correct local pipe section.
Add a units check: the coefficient is dimensionless, velocity squared divided by gravitational acceleration has units of length, and pressure loss would require multiplication by the stated fluid density and gravity. Do not add metres of head to kilopascals in the same total. For a reverse-flow operating case, reassess the geometry as a contraction rather than applying the enlargement result backwards.
Know when this example stops applying
A gradual transition depends on geometry and flow conditions that this sudden-expansion model does not represent. Nearby elbows, swirl, disturbed profiles and installation details can also alter the actual loss. Use applicable measured or manufacturer data with its stated conditions when selecting a real component; confirm which velocity basis that data uses.
The useful deliverable is a transparent calculation row, not a universal fitting allowance. Keep straight-pipe friction and other local losses separate, then assemble the complete system-head calculation for the actual operating case. The related pipe-design handbook can support the wider review workflow without replacing project-specific data or competent engineering judgment.
Continue the calculation review
Sources and calculation basis
Primary references checked on 26 October 2026. The worked examples and diagrams are original illustrative calculations, not manufacturer selections or project approvals. Apply the stated limitations and the requirements governing the actual project.

Comments