A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Friction factor is found differently depending on flow regime: laminar flows use f = 64/Re analytically; transitional flows (Re 2,300-4,000) are unpredictable; fully turbulent flows require the Colebrook equation or Moody chart lookup using Re and relative roughness. Understanding friction factor selection across regimes is critical for accurate pressure drop prediction.
You already know the Darcy-Weisbach equation: the head loss in a pipe is h_f = f·(L/D)·(V²/2g). This equation is exact — it makes no assumption about whether flow is laminar or turbulent. But it contains one quantity you cannot look up in a table or derive from geometry alone: the Darcy friction factor f. The friction factor encodes everything about how the fluid interacts with the pipe wall and the flow's internal structure. Determining f correctly is therefore the central skill in pipe flow engineering, and it depends entirely on the Reynolds number you calculated as a prerequisite.
In the laminar regime (Re < 2,300), the physics simplify dramatically. Flow is organized into neat concentric streamlines with no cross-stream mixing, and the parabolic velocity profile can be derived analytically from the Navier-Stokes equations. This derivation yields the exact result f = 64/Re — no empiricism, no roughness, no uncertainty. At Re = 1,000, f = 0.064; at Re = 2,000, f = 0.032. Notice that f decreases as velocity increases in laminar flow: more speed means higher Re which gives lower f. This counterintuitive behavior reflects the fact that, in laminar flow, the proportional friction loss actually decreases with velocity.
The turbulent regime (Re > 4,000) is fundamentally different. Random three-dimensional eddies transport momentum across the flow, creating much higher shear stresses at the wall than laminar flow would produce at the same velocity. Now f depends on two quantities: Re and relative roughness ε/D, the ratio of pipe wall roughness height to pipe diameter. The Colebrook equation captures this: 1/√f = −2 log(ε/3.7D + 2.51/Re√f). This equation is implicit in f — you cannot solve it directly. In practice, engineers use the Moody diagram, a log-log chart plotting f as a function of Re for families of ε/D curves, to read off f graphically. At very high Re, curves flatten to the "fully rough" limit where f depends only on ε/D, not Re — meaning doubling the velocity no longer changes f. The explicit Swamee-Jain approximation f ≈ 0.25/[log(ε/3.7D + 5.74/Re⁰·⁹)]² gives quick answers without iteration.
The transitional zone between Re = 2,300 and 4,000 is genuinely unpredictable. Flow can be laminar or turbulent depending on upstream disturbances, vibrations, and entrance conditions. Engineers treat this region conservatively, either assuming turbulent behavior (which gives higher f and thus a safer design for pressure drop calculations) or simply redesigning the system to operate clearly in one regime. The Moody diagram shows this region as a band of uncertainty rather than a clean curve. In real systems, pipe fittings, bends, and entrance effects destabilize laminar flow, so naturally occurring laminar pipe flow at Re close to 2,300 is rare in practice.