Flow Through Pipes: A Complete Guide to Reynolds Number, Friction Factor & Head Loss
Every pipe size on a drawing, every pump head on a schedule, every duct dimension in a ceiling void is the output of the same handful of equations. Learn them once and you stop looking things up in tables you do not understand, you start knowing why the table says what it says, and when it stops applying.
- TL;DR
- Key Concepts: An Overview
- Flow Velocity & Pipe Diameter
- Laminar vs Turbulent Flow
- Reynolds Number
- The Bernoulli Equation
- The Darcy–Weisbach Equation
- Friction Factor & Pipe Roughness
- Interactive Moody Chart & Pipe Flow Solver
- Major & Minor Head Losses
- Hydraulic Grade Line, Series & Parallel
- Siphons, Water Hammer & Nozzles
- Power Transmission & Heat Power
- Where This Takes You
- FAQs
- Sources & Further Reading
TL;DR
Key takeaways
- Continuity is the starting point: Q = V × A. Flow is fixed by the load, you choose a velocity, and the diameter follows.
- Reynolds number tells you which regime you are in: laminar below ~2300, turbulent above ~4000, an unpredictable transition zone between. It decides which friction equation applies.
- Real HVAC piping is almost always turbulent: a DN100 chilled water pipe would need to run below 0.023 m/s to go laminar, which is roughly 0.65 m³/h.
- Bernoulli conserves energy along the flow as pressure head plus velocity head plus elevation head, minus losses. Every hydraulic calculation is a bookkeeping exercise on those four terms.
- Darcy–Weisbach gives head loss: hf = f (L/D)(V²/2g). Loss scales with velocity squared, which is why oversizing pipe is cheap insurance and undersizing is a permanent tax.
- The friction factor f comes from the Moody diagram or the Colebrook equation, as a function of Reynolds number and relative roughness, not from a single lookup value.
- Minor losses at fittings are usually not minor. In a compact plant room they routinely exceed the straight-pipe friction.
- Water hammer is the violent case: stopping 2 m/s of water instantly in a steel pipe generates roughly 24 bar of surge on top of working pressure.
- The same equations govern air in ducts as water in pipes, only the fluid properties change.
Understanding Flow Through Pipes: Key Concepts
Flow through pipes is the study of how fluid moves through a closed conduit under pressure, and how much energy it loses on the way. The pipe runs full, so flow is driven by a pressure difference rather than by gravity acting on a free surface, which is what distinguishes it from open-channel flow in a drain or river.
For an MEP engineer this is not academic. It is the machinery behind four everyday deliverables:
- Pipe sizing for chilled water, condenser water, hot water and domestic services — see our dedicated pipe sizing calculation guide for the full step-by-step method.
- Pump head calculation: the sum of every loss around the index circuit, which becomes the pump's duty point.
- Duct sizing, because air in a duct obeys exactly the same equations with different fluid properties.
- System behaviour: why a branch starves, why a valve bangs, why a pump runs off its curve.
This guide works through the concepts in the order you would actually apply them: get the velocity, find the regime, find the friction factor, calculate the loss, add the fittings, then deal with the special cases that bite.
Flow Velocity & Pipe Diameter from a Known Flow Rate
Everything starts with continuity: the statement that for incompressible steady flow, whatever goes in must come out.
- Q volumetric flow rate, m³/s
- V mean velocity, m/s
- A internal cross-sectional area, m²
- D internal diameter, m
The one trap: use internal diameter, not nominal size. A "DN100" pipe is not 100 mm inside. Schedule 40 steel at DN100 has an internal diameter closer to 102.3 mm; heavier schedules are smaller. Since head loss varies with roughly the fifth power of diameter, a few millimetres of error is not trivial.
Master Pipe Flow & HVAC Design
Learn Darcy-Weisbach, pipe and duct sizing, pump head build-up, and full HVAC system documentation — structured for India and GCC MEP careers.
Laminar vs Turbulent Flow Regimes
Fluid moves through a pipe in one of two fundamentally different ways, and which one decides everything downstream in the calculation.
Between roughly 2300 and 4000 lies the transition region: unstable, sensitive to upstream disturbance, and genuinely unpredictable. Flow may switch between regimes intermittently. Do not design a system to operate here. If a calculation lands in the transition zone, change the pipe size rather than picking a friction factor and hoping.
Why this matters less in practice than in the exam
Real HVAC piping is essentially always turbulent, and it is worth internalising just how decisively. Take a DN100 chilled water pipe: to drop below Re = 2300, the water would have to move slower than 0.023 m/s: a flow of about 0.65 m³/h, roughly a tenth of a litre per second. No commissioned system runs like that. Laminar flow matters in capillary tubing, in lubrication, in heavy oils and in microfluidics; it does not appear in a chiller plant. The reason to learn it is that it defines the boundary of where the turbulent equations apply.
Reynolds Number: The Regime Indicator
The Reynolds number is a dimensionless ratio of inertial forces to viscous forces. When inertia dominates the flow tears itself into turbulence; when viscosity dominates it stays orderly.
- ρ density, kg/m³
- V mean velocity, m/s
- D internal diameter, m
- μ dynamic viscosity, Pa·s
- ν kinematic viscosity, ν = μ/ρ, m²/s
The second form is the one to use in practice, because kinematic viscosity is tabulated directly. Use the value at the actual operating temperature. Water at 7°C is roughly 40% more viscous than at 20°C, and a 30% glycol solution at 7°C is around four times more viscous than water, a difference that changes the answer, not just the decimal place.
| Fluid | Temperature | ν (m²/s) | Where you meet it |
|---|---|---|---|
| Water | 7°C | 1.43 × 10−6 | Chilled water supply |
| Water | 20°C | 1.00 × 10−6 | Standard reference condition |
| Water | 32°C | 0.77 × 10−6 | Condenser water |
| Water | 60°C | 0.47 × 10−6 | Domestic hot water |
| 30% ethylene glycol | 7°C | ~4.0 × 10−6 | Low-temperature and freeze-protected loops |
| Air | 20°C | 1.51 × 10−5 | Ductwork |
Glycol values vary with concentration and supplier, always use the manufacturer's published property data for the actual mixture rather than a generic figure.
| Range | Regime | Friction factor from |
|---|---|---|
| Re < 2300 | Laminar | f = 64 / Re, roughness irrelevant |
| 2300 – 4000 | Transition, avoid | Indeterminate; do not design here |
| Re > 4000 | Turbulent | Colebrook equation or Moody diagram |
The Bernoulli Equation: Energy Conservation in Flow
Bernoulli is conservation of energy, written for a flowing fluid and expressed in metres of head so that every term can be compared directly.
- p/ρg pressure head: energy stored as pressure
- V²/2g velocity head: kinetic energy of the moving fluid
- z elevation head: potential energy from height
- hL head loss: energy irreversibly converted to heat by friction
Textbook Bernoulli omits hL and assumes a frictionless fluid. That version is useless for design. The whole discipline of pipe hydraulics is the business of quantifying that last term.
Read practically, Bernoulli says the three forms of energy can trade against each other but the total only ever decreases. Fluid speeding into a smaller pipe converts pressure head into velocity head, pressure genuinely drops. Fluid rising up a riser converts pressure head into elevation head. Neither is a loss; both are conversions. Only hL is gone for good.
The closed-loop insight that saves engineers embarrassment
A common early error is adding the static height of a building to the pump head in a closed chilled water circuit. In a closed loop the water that rises up the riser also comes back down, so the elevation terms cancel exactly, z at the start equals z at the end. The pump only has to overcome friction, not gravity. Static height matters for filling and for setting the expansion vessel pressure, not for pump duty. In an open circuit, a cooling tower, an open tank, the static lift is real and must be added.
The Darcy–Weisbach Equation: Head Loss
The central equation of pipe flow, and the one worth committing to memory. For a full worked example applying this to building services piping systems, see our guide on pressure drop piping calculations.
- hf head loss due to friction, metres of fluid
- f Darcy friction factor, dimensionless
- L pipe length, m
- D internal diameter, m
- V mean velocity, m/s
- g 9.81 m/s²
Three consequences are worth stating explicitly, because they drive every practical sizing decision:
- Loss scales with velocity squared. Double the velocity and friction loss quadruples. This is why "one size up" is such a cheap fix and why undersizing is a permanent tax on the pump.
- Loss scales inversely with diameter: and steeply, because velocity itself depends on D². At fixed flow, head loss varies with roughly D−5. Going up one pipe size can halve the loss.
- Loss scales linearly with length. The only term that behaves intuitively.
Darcy friction factor vs Fanning friction factor
Two friction factors exist in the literature and they differ by a factor of four: fDarcy = 4 fFanning. Civil and mechanical engineering (and the Moody diagram, and everything in this article) use Darcy. Some chemical engineering texts use Fanning. If a calculated head loss comes out four times too big or too small, this is almost always why.
Friction Factor & Pipe Roughness
The friction factor is the one term in Darcy–Weisbach that is not simply measured, it depends on the Reynolds number and on how rough the pipe wall is relative to its diameter.
No roughness term at all. In laminar flow the fluid layer at the wall is stationary and the flow never "feels" the surface texture, a rusty pipe and a polished one behave identically.
f appears on both sides, so it must be solved iteratively, or read graphically from the Moody diagram, which is exactly what the diagram was created to avoid doing by hand.
The Explicit Shortcut
If you need a friction factor in a spreadsheet without iteration, the Swamee–Jain approximation solves explicitly and lands within about 1% of Colebrook across the normal engineering range:
Absolute Roughness
| Material | ε (mm) | ε/D at DN100 | Notes |
|---|---|---|---|
| Drawn copper / brass | 0.0015 | 0.000015 | Effectively hydraulically smooth |
| PVC, PE, plastics | 0.0015 – 0.007 | ~0.00005 | Very smooth; does not degrade with age |
| Commercial / welded steel | 0.045 | 0.00045 | The HVAC default for CHW and CW pipework |
| Galvanised steel | 0.15 | 0.0015 | Roughly 3× the roughness of black steel |
| Cast iron | 0.26 | 0.0026 | Common in older and buried services |
| Concrete | 0.3 – 3.0 | up to 0.03 | Wide range with finish quality |
Roughness changes over the life of the system
The values above are for new, clean pipe. Steel pipework in a poorly treated system develops corrosion products and scale, and its effective roughness can rise by an order of magnitude over decades, which is why an old system's measured pressure drop can far exceed its original design figure with no change in flow. This is the hydraulic argument for water treatment, alongside the heat-transfer one. When designing a retrofit or investigating an underperforming system, do not assume new-pipe roughness.
Interactive Moody Chart & Pipe Flow Solver
The Moody diagram plots friction factor against Reynolds number for a family of relative roughness curves. Enter your pipe below and the solver computes velocity, Reynolds number, friction factor and head loss, then plots your operating point on the chart.
Figure 1: Moody diagram with your calculated operating point plotted. Note how the roughness curves flatten to the right, at high Reynolds numbers the friction factor becomes independent of Re and depends only on relative roughness. This is the fully rough or complete turbulence zone where most large HVAC pipework operates.
Worked Example
-
Velocity from flow
50 m³/h of chilled water through DN100 commercial steel, internal bore 102.3 mm, over a 100 m run.
V=50 / 3600π(0.1023)2 / 4=1.69m/s -
Reynolds numberRe=1.69×0.10231.004×10−6=172,000fully turbulent
-
Relative roughness and friction factorε/D=0.045102.3=0.00044Colebrook→f=0.0188
-
Darcy–Weisbach head losshf=0.0188×1000.1023×1.6922×9.81hf = 2.7 m of water ≈ 26 kPa ≈ 260 Pa/m
Major & Minor Head Losses
Total head loss around any circuit splits into two categories, and the naming is genuinely misleading.
K is the dimensionless loss coefficient for each fitting. Note the same V²/2g velocity-head term appears in both major and minor loss, which is why velocity is the master variable in all of hydraulics.
| Fitting | K (typical) | Note |
|---|---|---|
| Long-radius 90° bend | 0.2 – 0.4 | The preferred elbow |
| Standard 90° elbow | 0.5 – 0.9 | Roughly double the long-radius loss |
| Tee, flow through run | 0.2 – 0.4 | Straight-through path |
| Tee, flow through branch | 1.0 – 1.8 | Branch path costs far more |
| Gate valve, fully open | 0.15 – 0.2 | Low loss, isolation duty |
| Globe valve, fully open | 6 – 10 | Very high, regulating duty |
| Swing check valve | 2 – 2.5 | Often overlooked in pump head build-up |
| Butterfly valve, fully open | 0.3 – 0.9 | Varies strongly with disc size |
| Sudden exit to tank | 1.0 | All velocity head is lost |
Indicative ranges for orientation. Use manufacturer data for valves, a specific globe valve's K can differ substantially from the generic figure, and Crane Technical Paper No. 410 or the ASHRAE Handbook for design calculations.
"Minor" losses are frequently the majority
The terminology is historical and misleading. In a long buried transmission main, fitting losses genuinely are minor. In a plant room: where a few metres of pipe carry two isolation valves, a check valve, a strainer, a balancing valve, four bends and a pump, the fittings can easily exceed the straight-pipe friction several times over. A single fully-open globe valve at K = 8 is worth about 20 long-radius bends. Never skip fitting losses because the pipe run looks short; the short runs are exactly where they dominate.
Hydraulic Grade Line, Series & Parallel Pipes
The Hydraulic Grade Line
The hydraulic grade line (HGL) is a plot of pressure head plus elevation head along a pipe, physically, the height to which water would rise in a vertical tube tapped into the pipe at each point. The energy grade line (EGL) sits above it by the velocity head, and always slopes downward in the direction of flow.
The HGL is the fastest diagnostic tool in hydraulics because it makes three things visible at a glance:
- Steeper slope means higher friction loss: so undersized sections show up immediately as steep segments.
- A pump appears as a vertical step up, equal to the pump head added.
- Where the HGL falls below the pipe centreline, the pressure is sub-atmospheric: a warning of cavitation, air release or pipe collapse. This is the single most valuable thing the sketch tells you.
Pipes in Series and Parallel
Why parallel branches self-distribute, and why that's the balancing problem
Because both branches connect the same two points, they must experience the same head loss. Flow therefore redistributes itself automatically until that is true, more through the easy path, less through the hard one. This is not a fault; it is physics, and it is precisely why a chilled water network delivers too much to the near coil and too little to the far one. Balancing valves work by deliberately adding resistance to the easy branches until every path has equal loss. Understanding this equation is understanding why balancing exists at all.
Siphons, Water Hammer & Flow Through Nozzles
Siphons
A siphon carries liquid over an elevation and down to a lower level with no pump. The driving force is the net elevation difference between the two free surfaces; atmospheric pressure on the upstream surface is what supports the column over the crest.
The design constraint is at the summit, where pressure is lowest. If it falls to the liquid's vapour pressure, the liquid boils, the column breaks, and flow stops. For water at ambient conditions the theoretical ceiling is about 10.3 m; in practice, keep the crest below roughly 7 m above the upstream surface to leave margin for friction, velocity head, temperature and altitude.
Water Hammer
Water hammer is the pressure surge produced when flowing liquid is stopped or started rapidly. The moving column has momentum; a valve slamming shut converts that momentum into a pressure wave that travels the pipe at the speed of sound in the fluid, reflecting back and forth until damped.
- a pressure wave speed, ~1,200–1,400 m/s for water in steel, ~300–500 m/s in plastic
- ΔV change in velocity, m/s
The number that should change how you specify valves
Stopping water moving at just 2 m/s instantaneously in a steel pipe generates roughly 250 metres of head, about 25 bar: on top of the working pressure. A typical chilled water system runs at 4–10 bar. This is why fast-acting solenoid valves and motorised valves with rapid actuators burst pipes and destroy joints, and why the mitigations exist: slow-closing actuators (the cheapest and most effective), surge vessels or air chambers, lower design velocity, and pump non-return valves selected for slam characteristics. Note the material effect too, plastic pipe's lower wave speed gives it a natural surge advantage over steel.
Flow Through Nozzles
A nozzle converts pressure head into velocity head by reducing area. Continuity forces velocity up; Bernoulli says the pressure must fall to pay for it. For an ideal nozzle discharging to atmosphere from a reservoir of head H, the exit velocity is given by Torricelli's relation:
Cv is the velocity coefficient, typically 0.95–0.99 for a well-formed nozzle, accounting for friction. The same principle runs in reverse in a venturi or orifice plate flow meter: create a known constriction, measure the pressure drop, and infer flow, which is how much of the flow measurement in a commissioned building actually works.
Power Transmission & Heat Power in Fluid Flow
Hydraulic Power
A classical result worth knowing: maximum power transmission through a pipe occurs when friction loss equals one third of the total head, at which point transmission efficiency is only 66.7%. This matters in hydraulic power transmission, but it is a trap in building services. In HVAC you are optimising for efficiency, not for maximum power throughput, so design head loss far below that point.
Heat Power, Where Fluid Mechanics Meets HVAC
The reason any of this matters in a building: the fluid is not the product, the heat it carries is.
This equation closes the loop between the two halves of a hydronic design. Fluid mechanics tells you the cost of moving the water; thermodynamics tells you why you are moving it. A larger ΔT means less flow for the same cooling duty, smaller pipes, smaller pumps, less friction, less energy. That is the whole argument for high-ΔT design in a chilled water system, and it is why the two equations should always be considered together rather than in sequence.
Where This Takes You
These equations are the substrate under every piece of hydraulic and air-side design work an MEP engineer does. Learn them properly and three things change.
You stop treating sizing tables as magic. A ductulator, a pipe sizing chart and a manufacturer's selection program are all just Darcy–Weisbach with the friction factor pre-solved, and knowing that tells you exactly when their assumptions stop applying to your project: different fluid, different temperature, glycol, aged pipework, altitude.
You can diagnose systems rather than guess at them. A branch that starves, a valve that bangs, a pump running off its curve, a coil that will not deliver, each has a hydraulic explanation, and the explanation is in this article.
And you can defend your numbers. The difference between a drafter and a design engineer is usually the ability to explain why the pipe is that size when a contractor proposes a cheaper alternative.
For how these fundamentals become project deliverables, see our guides to pump efficiency and BEP, HVAC duct design, and air distribution systems.
Frequently Asked Questions
Sources & Further Reading
- Moody, L.F. (1944), "Friction Factors for Pipe Flow", Transactions of the ASME. The original paper presenting the Moody diagram used throughout this article.
- Colebrook, C.F. (1939): the implicit turbulent friction factor correlation on which the Moody diagram's turbulent curves are constructed.
- Swamee, P.K. and Jain, A.K. (1976): explicit approximation to Colebrook, accurate to roughly 1% across the normal engineering range.
- ASHRAE Handbook, Fundamentals, ASHRAE. Chapters on Fluid Flow, Pipe Sizing and Duct Design, including friction charts, fitting loss data and fluid property tables for water and glycol solutions.
- Crane Technical Paper No. 410, Flow of Fluids Through Valves, Fittings and Pipe: the standard industry reference for loss coefficients, valve data and worked hydraulic examples.
- ASHRAE Duct Fitting Database: fitting loss coefficients for the air-side equivalent of this analysis.
Calculation basis for this article's tools
The pipe flow solver computes velocity from continuity, Reynolds number as VD/ν, and the friction factor from 64/Re below Re = 2300 or from an iterative solution of the Colebrook–White equation above it. Head loss uses Darcy–Weisbach; pressure drop uses Δp = ρghf. Fluid properties are representative values at the stated temperatures, use actual property data for the specific fluid and glycol concentration on a real project. The Moody chart curves are generated from the same Colebrook solution and validate against published values (for example f = 0.018 at Re = 105 for smooth pipe). The water hammer calculator applies the Joukowsky equation for instantaneous closure, which is an upper bound; real surge analysis for critical systems requires transient modelling that accounts for pipe elasticity, valve closure characteristics and reflections. These are teaching and first-pass tools, issued designs should be produced in validated hydraulic software.
This article was last reviewed on 1 August 2026.
Ready to start your engineering career journey?
Investing in structured training is one of the best ways to set yourself apart in MEP engineering. Our Courses cover HVAC, electrical and plumbing design in depth, plus tools like HAP software — the exact skillset that sets you apart in a competitive field.