Darcy vs Fanning Friction Factors: Catch the Fourfold Pipe-Loss Error
Two pipe-loss worksheets can use the same flow, diameter and length yet disagree by a factor of four. Before changing the roughness or adding a design allowance, inspect the friction-factor definition. A Darcy factor and a Fanning factor describe the same frictional behavior through different coefficient conventions. Each must be paired with its matching equation.
This review focuses on steady, fully developed, single-phase, incompressible flow in a straight circular pipe. The worked example is an equation-input audit with explicit illustrative assumptions. It does not select a pipe, establish a project pressure-loss limit or predict the operating point of an installed pump.
1. Write the convention beside the factor
Using the Darcy convention, straight-pipe friction head is hf = fD × (L/D) × v²/(2g). Using the Fanning convention, it is hf = 4fF × (L/D) × v²/(2g). Here hf is head loss in metres of the flowing liquid, L is pipe length in metres, D is internal diameter in metres, v is mean velocity in m/s and g is gravitational acceleration in m/s².
The two dimensionless factors are related by fD = 4fF. AFT’s published verification example explicitly identifies this conversion when importing a Fanning factor into a Darcy-based calculation. The conversion concerns the coefficient definition; it is not an additional physical loss or a multiplier to apply to the entire pump-head total.
The Ansys lesson on major pipe losses gives the Darcy pressure-loss equation and its relationship to the wall-friction coefficient. For the present constant-diameter, horizontal example, the friction pressure drop is Δpf = ρg hf, equivalently fD × (L/D) × ρv²/2. Density ρ must be expressed consistently with the chosen units.
2. Prove that both conventions give the same result
Assume a straight length of 100.0 m, internal diameter of 0.0500 m and mean velocity of 1.000 m/s. Take density as 998 kg/m³, dynamic viscosity as 0.00100 Pa·s and g as 9.81 m/s². The corresponding volume flow is πD²v/4 = 0.0019635 m³/s, or 1.9635 L/s. Reynolds number is ρvD/μ = 49,900.
For this audit only, assume a Darcy factor of 0.0240. Its equivalent Fanning factor is 0.00600. The coefficient is an imposed example input, not a value established from Reynolds number alone. A project calculation would need an appropriate roughness basis and friction correlation, or other justified data, for the actual pipe and fluid conditions.
With the Darcy equation, hf = 0.0240 × (100.0/0.0500) × 1.000²/(2 × 9.81) = 2.4465 m. With the Fanning equation, hf = 4 × 0.00600 × (100.0/0.0500) × 1.000²/(2 × 9.81) = 2.4465 m. Both produce the same friction pressure drop: 998 × 9.81 × 2.4465 ≈ 23,952 Pa, or 23.952 kPa.
Carry unrounded intermediate values when reproducing the result; report precision appropriate to the inputs. The extra decimal places retained during this audit help compare equations. They do not establish that the physical installation’s head loss is known to that precision.
3. Locate the fourfold error in the affected term
Original constant-flow example: 100 m of straight pipe, 0.0500 m internal diameter and 1.000 m/s mean velocity. Assumed fD = 0.0240 and fF = 0.00600. Fourfold changes apply to the pipe-friction component, not automatically to total circuit head.
If 0.00600 is entered directly into the Darcy equation without conversion, the calculated pipe head becomes 0.6116 m, one-quarter of the correct value. Its friction pressure drop is 5.988 kPa. If 0.0240 is entered into the Fanning equation with the leading factor of four still present, pipe head becomes 9.7859 m and friction pressure drop becomes 95.808 kPa.
These are two opposite convention errors. Neither is fixed by changing the fluid density, adding extra decimal places or averaging the answers. Trace the coefficient back to the source equation. Then either convert the coefficient once or change the receiving equation to its matching convention.
The error in total circuit head depends on how much of that total is affected. Suppose other correctly calculated components contribute 7.500 m at the same flow, and only this straight-pipe term has the convention problem. Correct total head is 7.500 + 2.4465 = 9.9465 m. The low erroneous total is 7.500 + 0.6116 = 8.1116 m, about 18.45% below correct.
The high erroneous total is 7.500 + 9.7859 = 17.2859 m, about 73.79% above correct. The pipe term changes by fourfold; the whole total does not. This comparison holds the flow and all other terms fixed. An actual pump’s operating flow can move when the system resistance changes, so these percentages do not predict installed flow or energy savings.
4. Check the correlation before trusting the number
AFT’s pipe-pressure-drop documentation distinguishes roughness-based calculation from explicit friction-factor entry. An imported coefficient can bypass the calculation that would otherwise reconcile flow regime and roughness. Record whether the value came from a chart, correlation, manufacturer source or direct entry, and preserve the source’s convention.
A useful identification check for fully developed laminar flow of a Newtonian fluid in a circular pipe is fD = 64/Re versus fF = 16/Re. If a chart or spreadsheet documents that laminar branch, it can help confirm its convention. Do not apply either laminar expression to the turbulent worked example or assume every noncircular or specialized correlation has the same constant.
Keep absolute roughness and relative roughness distinct. The ratio ε/D requires matching length units; a millimetre value entered as metres creates a different error from the factor-of-four issue.
Keep dynamic viscosity μ in Pa·s separate from kinematic viscosity ν in m²/s. Reynolds number can be written ρvD/μ or vD/ν when ν = μ/ρ.
Do not convert a fitting-loss coefficient K by four merely because it appears beside a friction factor. Check its defining equation and reference velocity separately.
Keep manufacturer equipment pressure drops at their stated flow and fluid conditions. A Darcy/Fanning correction belongs only where that convention actually enters the calculation.
5. Build a short regression check into the worksheet
Name input columns explicitly: Darcy friction factor or Fanning friction factor. A generic column headed f hides the information needed to review it.
Use the worked pair 0.0240 and 0.00600 as a controlled test. Matching equations must return identical pipe losses for identical geometry and velocity.
Inspect formula cells for both missing and duplicated factors of four. Check imported data, hidden conversion columns and any intermediate coefficient normalization.
Reconcile the result by component: straight pipe, fittings and equipment. If only one section differs fourfold, investigate that section before modifying the rest of the circuit.
Keep the coefficient source, convention, units and operating condition visible in the calculation record. Recheck the real design inputs after the arithmetic test passes.
For assembling the wider head-loss calculation, the paid Pump Head Calculator Design Suite v1.0 is a relevant resource covering pipe, fitting and equipment losses. This article’s hand-check remains useful alongside any software: match the factor to its equation, convert once and verify the affected component before using the total in pump selection.
Comments