Referencefluids & pipingLast reviewed: 2026-07-29

Use pipe roughness when a Darcy-Weisbach calculation needs a turbulent-flow friction factor and the actual internal surface is not modeled directly.

What This Means

Absolute roughness, epsilon, represents a characteristic height of internal surface irregularities. Relative roughness, epsilon / D, compares that height with the pipe's inside diameter.

The distinction matters because the same surface texture is more influential in a small passage than in a large pipe. In laminar flow, the Darcy friction factor is set by Reynolds number and is not calculated from roughness. In turbulent flow, both Reynolds number and relative roughness can influence the friction factor.

A handbook roughness value is an engineering model input, not a guaranteed measurement for every pipe carrying the same material label. Manufacturing method, coating, joints, corrosion, scale, fouling, and service history can make the effective hydraulic roughness different from a clean-pipe reference value.

Key Relationships

relative roughness = epsilon / D

Darcy-Weisbach:
h_L = f (L / D) V^2 / (2 g)

Colebrook form for turbulent flow:
1 / sqrt(f) = -2 log10[(epsilon / D) / 3.7 + 2.51 / (Re sqrt(f))]
  • epsilon is absolute roughness.
  • D is the actual inside diameter used by the flow model.
  • f is the Darcy friction factor.
  • Re is Reynolds number.
  • L is pipe length and V is average velocity.

Use consistent length units for epsilon and D so relative roughness is dimensionless.

Use This When

  • Estimating straight-pipe pressure loss with Darcy-Weisbach.
  • Reviewing why a pressure-loss result changed after selecting a different pipe material or condition.
  • Comparing clean, aged, lined, scaled, or corroded pipe assumptions.
  • Documenting a conservative roughness assumption when the installed condition is uncertain.
  • Checking whether a friction-factor correlation is using Darcy or Fanning convention.

Assumptions

  • The pipe can be represented by a uniform effective roughness over the analyzed length.
  • The characteristic diameter is the actual hydraulic inside diameter, not nominal pipe size.
  • The selected friction-factor relationship is appropriate for the Reynolds-number regime.
  • Local losses from fittings, valves, entrances, exits, and abrupt geometry changes are handled separately.

Limitations

  • Published roughness values are representative and may not describe a specific installed pipe.
  • Effective roughness can change with corrosion, deposition, biological growth, lining degradation, or cleaning.
  • Corrugated, ribbed, noncircular, flexible, or intentionally rough passages may need a specialized correlation.
  • Near the laminar-to-turbulent transition, small input changes can produce unstable regime assumptions.
  • A calibrated system model may absorb unmodeled valve and fitting losses into an apparent roughness value; that value should not automatically be reused elsewhere.

Common Mistakes

  • Using nominal pipe size as D instead of the schedule-dependent inside diameter.
  • Entering roughness in millimeters while diameter is in meters.
  • Applying a clean new-pipe value to old scaled or corroded service without documenting the assumption.
  • Mixing Darcy and Fanning friction factors; the Darcy value is four times the Fanning value.
  • Adding fitting losses through both equivalent length and separate K values.
  • Assuming roughness changes laminar friction factor in the same way it changes turbulent friction factor.

Sources

This reference uses White's Fluid Mechanics for relative roughness, flow-regime, Colebrook, and Darcy-Weisbach relationships. Crane TP-410 provides corroborating practical piping context for roughness selection and friction-loss calculations.

  1. Frank M. White. Fluid Mechanics, 7th ed., McGraw-Hill, 2011. ISBN 978-0-07-352934-9.
  2. Crane Co.. Flow of Fluids Through Valves, Fittings, and Pipe (Crane TP-410), Technical Paper No. 410, Crane Co., 2009. TP-410.