Why we limit velocity at all

You could push fluid through a smaller, cheaper pipe by running it faster. The reason we don't is that velocity costs you in four ways, and at least one of them always bites before the others. Sizing a line is really just finding the velocity where none of these has crossed its threshold.

  • Pressure drop. Friction loss scales with velocity squared. Double the velocity and you roughly quadruple the pressure drop — and the pump power that pays for it.
  • Noise. Fast gas and steam generate flow noise. Above roughly 30 m/s in gas service, lines get loud and start vibrating.
  • Erosion. Liquids carrying solids, and any wet/two-phase flow, scour the pipe wall and especially the elbows. This thins the wall over time.
  • Water hammer. The faster a liquid moves, the bigger the pressure surge when a valve slams shut. Surge pressure is proportional to velocity.

Why it matters in practice

Velocity is the lever you pull when choosing a line size. Pick a diameter, and the velocity is fixed by the flow rate. If it lands too high you get erosion and noise complaints; too low and you've oversized the pipe and risk dropping solids out of suspension. The whole point of a hydraulic line-sizing exercise is to land inside the comfortable band for that service.

Core engineering concept — erosional velocity

For two-phase and gas-bearing lines, the usual reference is the erosional velocity from API RP 14E:

Ve = C / √ρm

Ve is the erosional velocity, ρm is the mixture density, and C is an empirical constant — traditionally 100 for continuous service (US units, ft/s and lb/ft³), higher for clean, non-corrosive, solids-free flow. The takeaway from the formula is the trend: lighter (lower-density) fluids erode at lower velocities, which is why the gas and steam limits are so much lower than the liquid ones.

Reality check. API 14E's C-factor is a rule of thumb, not a law of physics. For solids-laden or sour service, model the erosion properly rather than leaning on a single constant.

Design guidance — typical target velocities

These are the ranges most process specs design to. Treat them as a starting band, then check pressure drop and erosion for your specific case.

ServiceTypical velocityGoverned by
Pump suction (liquid)0.6 – 1.5 m/sNPSH, cavitation
Pump discharge (liquid)1.5 – 3.0 m/sPressure drop, erosion
General liquid lines1.0 – 3.0 m/sPressure drop
Gas / vapour lines10 – 20 m/sNoise, erosion
Saturated steam15 – 30 m/sNoise, erosion
Superheated steam30 – 60 m/sNoise

How to use the band

  • Aim for the middle of the range, not the edge. It leaves room for flow rate growth and uncertainty.
  • For pump suction, stay low — high suction velocity eats into NPSH and invites cavitation.
  • For long lines, pressure drop usually sets the size well before erosion does. Check it.
  • For two-phase or solids service, compute the erosional velocity and keep a margin below it.

Common mistakes

  • Treating one velocity limit as universal. The right limit depends on the fluid and the failure mode you're guarding against.
  • Ignoring suction-side velocity. A fast pump suction line is a classic cause of cavitation and NPSH problems.
  • Sizing on velocity alone for long lines. A velocity that looks fine can still produce an unacceptable total pressure drop over distance.
  • Going too slow in liquid lines with solids. Below the settling velocity, solids drop out and you get sludge and blockages.
  • Forgetting surge. High liquid velocity plus a fast-closing valve equals water hammer. Account for it in long pipelines.
Key takeaway

There's no fixed velocity limit — there's the first problem that appears. For liquids it's usually pressure drop and surge; for gas and steam it's noise and erosion. Size to the middle of the typical band, then verify pressure drop and erosional velocity for the actual fluid.