Pipe Flow Rate Calculator

Introduction to pipe flow rate and hydraulic sizing

Every pressurised piping job comes down to two coupled questions: how much fluid actually moves through the bore, and how much pressure it costs to push it there. This calculator answers both. It starts from the continuity relation between cross-sectional area and mean velocity, adds the Reynolds number so you know which flow regime you are working in, solves the implicit Colebrook-White equation for the Darcy friction factor, and then applies the Darcy-Weisbach equation to convert that friction factor into head loss and pressure drop over a run of straight pipe. You can work in either direction: enter a velocity and read off the volumetric flow rate, or enter the flow rate your process demands and read off the velocity the pipe will actually see.

The single most common error in a quick pipe calculation is treating the nominal pipe size as if it were the bore. Nominal Pipe Size (NPS) is a label, not a measurement. A schedule 40 NPS 2 in steel pipe has an outside diameter of 2.375 in and a wall thickness of 0.154 in, which leaves an inside diameter of 2.067 in. That is 3.35 % larger than the "2 inch" name suggests, and because flow area scales with the square of the bore it is a 6.8 % difference in capacity. At NPS 1/2 in the gap is far worse: the bore is 0.622 in against a 0.500 in label, so a nominal-diameter calculation understates the flow area by 35 %. This calculator carries the actual ASME B36.10M schedule 40 bores from NPS 1/2 in to NPS 12 in, so the diameter that goes into the area is the diameter the fluid sees.

The second common error is ignoring the flow regime altogether. A great many pipe calculators quietly assume turbulent flow and apply a turbulent friction correlation to every case handed to them. That is fine for water in a 100 mm main, but it is badly wrong for lubricating oil in a 10 mm line, for glycol at low temperature, or for any viscous fluid creeping along at a fraction of a metre per second. Below a Reynolds number of roughly 2300 the friction factor is not a function of wall roughness at all — it is exactly 64/Re, a result that falls straight out of the Hagen-Poiseuille solution. This tool detects the laminar branch, uses the correct expression for it, and warns you explicitly when your operating point lands in the transitional band between about 2300 and 4000, where no correlation is reliable.

Typical design velocities are worth keeping in the back of your mind while you use the tool. Pumped water distribution normally sits between roughly 1 and 3 m/s: below about 0.6 m/s solids and air pockets tend to settle out of suspension, and above roughly 3 m/s noise, vibration, erosion of protective oxide films and the severity of any water-hammer transient all climb quickly. Suction lines to pumps are usually kept slower than discharge lines to protect the net positive suction head available. None of these are hard limits set by physics; they are engineering conventions that trade capital cost in pipe diameter against operating cost in pumping energy, and your own project specification always wins.

How to use this pipe flow rate and head loss calculator

1. Pick how you will specify the bore. Choose Schedule 40 nominal size to select a standard steel pipe and have the actual ASME B36.10M inside diameter filled in for you, or choose Custom inside diameter and type the true bore in millimetres, centimetres or inches. Always enter the inside diameter, never the outside diameter and never the nominal label.
2. Choose what you know. If you have a velocity from a pump curve, a design specification or an in-line meter, select I know the mean velocity. If you have a duty in litres per second, cubic metres per hour or US gallons per minute, select I know the volumetric flow rate and the calculator inverts the continuity relation to give you the velocity instead.
3. Enter the straight length of pipe. Head loss is proportional to length, so this drives the pressure drop result. Enter the developed length of straight pipe; fittings, bends and valves are handled separately and are discussed under the limitations below.
4. Select the pipe material. The material sets the absolute roughness used in the Colebrook-White equation. Choose Custom roughness if you have a measured or specified value, for example from a pipe manufacturer's data sheet or from a condition assessment of an ageing main.
5. Describe the fluid. Choose Water at a given temperature and enter the temperature in degrees Celsius: density and dynamic viscosity are interpolated from NIST Chemistry WebBook values between 0 °C and 100 °C at atmospheric pressure. For anything that is not water, choose Custom fluid and enter the kinematic viscosity in centistokes together with the density in kilograms per cubic metre.
6. Read the results panel and the size comparison table. The panel reports flow rate in five units, velocity, Reynolds number, flow regime, relative roughness, friction factor, head loss and pressure drop. Below it, a live table re-runs the same duty through every schedule 40 size so you can see at a glance which one lands in a sensible velocity band. Use the Copy result and Download CSV buttons to take the numbers away with you.

All validation is done in the page's own script rather than by the browser, so a blank, zero, negative or non-numeric entry produces a specific written message telling you which field is wrong and why, instead of silently blocking the form or leaving a stale answer on screen. The calculator will never show you a result built on a division by zero.

The formula set: continuity, Reynolds number and the Darcy-Weisbach equation

Cross-sectional area and the continuity relation

For a full, round pipe of inside diameter D, the flow area is

and the volumetric flow rate follows from the continuity equation for an incompressible fluid, where v is the area-averaged (mean) velocity:

Note the word mean. The velocity profile across a pipe is never flat: in fully developed laminar flow the centreline velocity is exactly twice the mean, and even in strongly turbulent flow the centreline runs roughly 15 to 20 % above the mean. If you measured your velocity with a pitot tube on the centreline, correct it to a mean before you enter it, or you will overstate the flow rate.

Reynolds number and the flow regime

The Reynolds number is the ratio of inertial to viscous forces and decides everything downstream of it:

Here ρ is density, μ is dynamic viscosity and ν = μ/ρ is kinematic viscosity. One centistokes equals 10⁻⁶ m²/s, and water at 20 °C has a kinematic viscosity of almost exactly 1 cSt, which is why the unit is so convenient in pipe work. Conventionally, flow below Re ≈ 2300 is laminar, flow above Re ≈ 4000 is fully turbulent, and the band in between is transitional and genuinely unpredictable — the same pipe can flip between regimes depending on inlet disturbances, vibration and upstream fittings.

In the laminar branch the Darcy friction factor is exact and independent of roughness:

Colebrook-White, relative roughness and the explicit approximations

In turbulent flow the friction factor depends on the Reynolds number and on the relative roughness, the ratio of the absolute wall roughness ε to the bore:

The governing relation is the Colebrook-White equation, which is the algebraic curve fit underlying the Moody diagram. It is implicit in f, so it has to be solved iteratively:

This calculator solves that equation numerically by fixed-point iteration on 1/√f, seeded with the Swamee-Jain estimate and driven to a relative tolerance of 10⁻¹², so the friction factor you see is the Colebrook-White root rather than an approximation of it. The well-known explicit alternative is the Swamee-Jain equation:

Swamee and Jain published this as an approximation, not as an identity, and it should always be described that way. It agrees with Colebrook-White to within about 1 % for relative roughness between 10⁻⁶ and 10⁻² and Reynolds numbers between 5 × 10³ and 10⁸; outside that window the error grows. In the worked example below it comes out 0.56 % high. The results panel reports both values so you can see the size of the approximation for your own case. Haaland's explicit form is another common choice and is typically within about 2 % over a similar range.

Head loss and pressure drop

With f in hand, the Darcy-Weisbach equation gives the friction head loss over a straight length L. This is the physically general form: it holds for any Newtonian fluid, any temperature and any regime, provided the friction factor is evaluated correctly.

The head loss is expressed in metres of the flowing fluid. To convert it into a pressure drop, multiply by the fluid's specific weight, using the standard acceleration of gravity g = 9.806 65 m/s²:

The group v²/2g is the velocity head, and it is worth watching separately. It sets the scale of every minor loss in the system, because fitting losses are conventionally written as K times the velocity head. If the velocity head is large, the fittings will dominate and a straight-pipe-only calculation will badly understate the true pressure drop.

Where Hazen-Williams fits, and where it does not

You will often see pipe flow presented through the Hazen-Williams equation instead, which in SI form is written with the hydraulic radius R = D/4 and the hydraulic gradient S = h_f/L:

It is popular in municipal water distribution because the roughness coefficient C does not depend on the Reynolds number, which makes network models converge easily. But it is an empirical fit, not a physical law, and it is valid only for water in turbulent flow at ordinary temperatures. Its coefficients were calibrated around a kinematic viscosity near 1.1 × 10⁻⁶ m²/s, which is water at roughly 15 °C, and it carries no viscosity term at all, so it cannot represent hot water, cold glycol, oil, slurry or any laminar case. This calculator therefore does not use Hazen-Williams for its answers; it is shown here because presenting it as a universally applicable pipe-flow formula is a real and common error.

Worked example: a 4-inch schedule 40 water main at 2 m/s

Take NPS 4 in schedule 40 carbon steel: outside diameter 4.500 in, wall thickness 0.237 in, so the bore is 4.500 − 2 × 0.237 = 4.026 in = 102.2604 mm. The line runs 100 m straight, carries water at 20 °C, and the design mean velocity is 2.0 m/s. New commercial steel has an absolute roughness of 0.045 mm.

1. Area. A = π × (0.102 260 4 m)² / 4 = 8.2133 × 10⁻³ m², or 82.13 cm².
2. Flow rate. Q = 8.2133 × 10⁻³ × 2.0 = 1.6427 × 10⁻² m³/s = 16.43 L/s = 985.6 L/min = 59.14 m³/h = 260.4 US GPM.
3. Fluid properties. At 20 °C and atmospheric pressure, NIST gives ρ = 998.21 kg/m³ and μ = 1.0014 × 10⁻³ Pa·s, so ν = 1.0032 × 10⁻⁶ m²/s (1.0032 cSt).
4. Reynolds number. Re = 2.0 × 0.102 260 4 / (1.0032 × 10⁻⁶) = 2.039 × 10⁵. That is comfortably turbulent, so the Colebrook-White branch applies.
5. Relative roughness. ε/D = 0.045 / 102.2604 = 4.401 × 10⁻⁴.
6. Friction factor. Iterating Colebrook-White gives f = 0.018 47. The Swamee-Jain approximation returns 0.018 58, which is 0.56 % high — a good illustration of why the explicit form should be labelled as an approximation.
7. Head loss. h_f = 0.018 47 × (100 / 0.102 260 4) × (2.0² / (2 × 9.806 65)) = 3.68 m of water over the 100 m run.
8. Pressure drop. Δp = 998.21 × 9.806 65 × 3.68 = 36.1 kPa = 0.361 bar = 5.23 psi, which is 1.59 psi per 100 ft — in line with published schedule 40 friction tables for that duty.

Now repeat the calculation with the nominal 4.000 in taken as the bore. The area falls to 81.07 cm², the flow rate at 2 m/s drops to 16.21 L/s, and you would have understated the pipe's capacity by 1.3 %. That sounds harmless until you try the same shortcut at NPS 1/2 in, where the same mistake costs you 35 % of the flow area. The table below shows how the size of the error moves with the nominal size, and why it is never safe to assume.

Comparison tables: nominal versus actual bore, roughness and regime

Two things stand out. The error does not shrink monotonically — NPS 1 1/4 in is worse than NPS 1 in — and above NPS 8 in the nominal label actually overstates the bore, so the shortcut flips from conservative to optimistic. There is no safe rule of thumb here; use the real dimension.

Interpreting the flow rate, velocity and head loss you get

Read the velocity first. It is the number that tells you whether the pipe is sensibly sized. A water line at 0.3 m/s is oversized and will let sediment drop out; the same line at 5 m/s is undersized, noisy, and burning pumping energy that grows roughly with the cube of the flow rate for a fixed pipe. The calculator flags velocities outside the customary 0.6 to 3.0 m/s window for pumped water so the point is hard to miss, but treat those thresholds as prompts to think rather than as pass or fail criteria — fire mains, cooling water, gravity drains and slurry lines all have their own conventions.

Read the head loss second, and specifically read the value per 100 m. That figure is what pump selection actually turns on, because the pump has to supply the static lift plus the friction head at the design flow. Doubling the flow through a fixed pipe roughly quadruples the friction head, since head loss scales with v² while f changes only slowly; going up one pipe size at fixed flow typically cuts the friction head by a factor of three to five. The live schedule 40 comparison table below the calculator makes that trade-off visible in one glance, which is exactly the calculation a designer wants when deciding whether extra pipe cost pays for itself in pump energy.

Read the Reynolds number third. If it comes back below 4000 for a fluid you assumed was turbulent, something in your assumptions is wrong — usually the viscosity or the velocity. If it comes back in the transitional band, do not design to the number: either change the pipe size to move the operating point clearly into one regime, or bracket the design using both the laminar and turbulent friction factors and take the worse case.

Limitations and assumptions of this hydraulic model

These are the assumptions this calculator makes. Every one of them can be violated by a real system.

• Steady, fully developed, single-phase flow in a completely full round pipe. Partly full gravity pipes obey open-channel hydraulics instead, which needs Manning's equation and the actual depth of flow. Two-phase, flashing, cavitating or air-entrained flows are outside the model entirely.
• Newtonian fluid, incompressible over the run. That covers water, most oils, glycols and dilute solutions. It does not cover polymer solutions, drilling muds, fresh concrete, high-solids slurries or any other shear-thinning or Bingham-plastic fluid. It also does not cover compressible gas flow where the pressure drop is a significant fraction of the absolute pressure — there, density changes along the pipe and you need an isothermal or adiabatic compressible formulation.
• Straight pipe only. No elbows, tees, reducers, valves, strainers, entrances or exits are included. Real systems add these as minor losses, either as K × velocity head or as an equivalent length of straight pipe. A compact skid can easily have more fitting loss than pipe loss, so a straight-pipe answer is a floor, not a total.
• Roughness is an assumption, not a measurement. The tabulated ε values are for clean, new pipe. Corrosion, tuberculation, scaling and biofilm can raise the effective roughness of an old ferrous main by an order of magnitude, and the Colebrook data on which the correlation rests carry a scatter of roughly ±5 to 10 % even for new pipe. Treat the friction factor as good to about two significant figures unless you have measured your line.
• Water properties are interpolated between 0 °C and 100 °C at atmospheric pressure. Outside that range, or for pressurised hot water above its atmospheric boiling point, switch to the custom fluid mode and supply your own density and viscosity.
• The transitional band is genuinely uncertain. Between Re 2300 and 4000 no correlation is dependable. The calculator will give you a number and will tell you not to trust it.
• Hazen-Williams is not used here and should not be substituted casually. It applies only to water in turbulent flow at ordinary temperatures, and it has no viscosity term, so it silently gives wrong answers for hot water, cold or viscous fluids, and every laminar case.
• Nominal sizes are schedule 40 carbon steel per ASME B36.10M. Copper tube, PVC, PEX, HDPE, ductile iron and other schedules all have different bores for the same nominal label. If you are not using schedule 40 steel, measure or look up the real inside diameter and use the custom option.
• This is a preliminary design and checking tool. It is not a substitute for a network hydraulic model, a code-compliant fire protection calculation, or a stamped engineering review.

Common questions about pipe flow rate calculations

Sources and standards behind these numbers

Friction factor and head loss: C. F. Colebrook, "Turbulent Flow in Pipes, with Particular Reference to the Transition Region between the Smooth and Rough Pipe Laws," Journal of the Institution of Civil Engineers 11(4), 1939, pp. 133–156; and L. F. Moody, "Friction Factors for Pipe Flow," Transactions of the ASME 66, 1944, pp. 671–684 (the Moody diagram and its absolute roughness table). Explicit approximation and its stated accuracy: P. K. Swamee and A. K. Jain, "Explicit Equations for Pipe-Flow Problems," Journal of the Hydraulics Division, ASCE 102(HY5), 1976, pp. 657–664. Roughness values as tabulated for industrial practice: Crane Co., Flow of Fluids Through Valves, Fittings, and Pipe, Technical Paper No. 410; the ASHRAE Handbook—Fundamentals, Chapter 22 (Pipe Sizing) tabulates the same figures. Darcy-Weisbach in an institutional manual: U.S. Bureau of Reclamation, Water Measurement Manual, 3rd edition (2001), Chapter 2, Section 16, Normal Flow Equations and Friction Head Loss. Pipe dimensions: ASME B36.10M, Welded and Seamless Wrought Steel Pipe — schedule 40 outside diameters and wall thicknesses, with inside diameters computed as OD − 2 × wall. Water density and viscosity from 0 °C to 100 °C at 0.101 325 MPa: NIST Chemistry WebBook, SRD 69, Thermophysical Properties of Fluid Systems, which implements IAPWS-95 for density and the IAPWS 2008 Formulation for the Viscosity of Ordinary Water Substance. Standard gravity g = 9.806 65 m/s² per the SI convention adopted by the CGPM in 1901.