Internal Flow and Pipe-System Analysis

Introduction and Context

Internal flow is the motion of a fluid through a confined passage such as a pipe, duct, valve, nozzle, or machine passage. Unlike external flow over a body, internal flow is strongly influenced by wall shear, confinement, entrance development, fittings, elevation changes, and the interaction of several branches. Pipe-system analysis is therefore central to water supply, hydraulic machinery, process plants, cooling systems, lubrication circuits, compressed-air networks, and fuel systems.

The engineer’s central task is to determine one or more of the following: discharge, pressure distribution, head loss, pipe diameter, pump power, or operating condition. The analysis combines conservation of mass, the energy equation, and empirical or semi-empirical descriptions of wall friction.

Learning Objectives

By the end of this chapter, the reader should be able to:

• distinguish laminar, transitional, and turbulent pipe flow;
• calculate entrance length and recognize fully developed flow;
• determine major and minor head losses;
• use the Darcy friction factor and Moody-chart procedures;
• analyze equivalent pipes and series or parallel networks;
• size pipes using energy, velocity, and economic considerations;
• construct and interpret energy-grade lines and hydraulic-grade lines;
• evaluate a practical pipe system and identify modeling limitations.

1. Governing Conservation Principles

For steady flow through a pipe, conservation of mass gives

where Q is volumetric flow rate (m³/s), A is cross-sectional area (m²), and V is mean velocity (m/s). For a circular pipe, A = πD²/4, where D is internal diameter.

For an incompressible fluid at a junction, the algebraic sum of flows is zero:

The steady-flow energy equation between sections 1 and 2 is

Here, z is elevation head, p/(ρg) is pressure head, V²/(2g) is velocity head, h_p is pump head added, h_t is turbine head removed, and h_L is total head loss. The density is ρ, gravitational acceleration is g, and α is the kinetic-energy correction coefficient. For turbulent flow in a nearly uniform pipe, α is commonly approximated as 1; for laminar flow in a circular pipe, α = 2.

The equation assumes a single-phase fluid and time-averaged quantities where turbulence exists. It is written per unit weight, so every term has units of metres of fluid.

Figure: A pipe system with a pump and the energy-grade and hydraulic-grade lines. The vertical separation between EGL and HGL is the velocity head.

2. Flow Regimes and Entrance Effects

The Reynolds number compares inertial and viscous effects:

where μ is dynamic viscosity (Pa·s) and ν = μ/ρ is kinematic viscosity (m²/s). For flow in a circular pipe, the usual classification is:

These boundaries are guidelines, not universal physical discontinuities. In carefully controlled conditions, laminar flow may persist above 2300, while disturbances can trigger turbulence at lower values.

At a pipe entrance, the velocity profile is initially nearly uniform. Wall friction slows fluid adjacent to the wall, and boundary layers grow inward until they meet at the centerline. The entrance length is the distance required for the profile to become fully developed. Useful engineering estimates for a circular pipe are

and

The turbulent range depends on Reynolds number, inlet geometry, and roughness. Entrance losses should not be confused with fully developed friction loss. In short pipes, the developing region can occupy a substantial fraction of the total length.

3. Fully Developed Laminar Flow

For steady, incompressible, fully developed laminar flow in a circular pipe, the velocity profile is parabolic:

where u is local velocity, r is radial position, R = D/2 is pipe radius, ∆p is pressure drop, and L is pipe length. Integration over the cross-section gives the Hagen–Poiseuille relation:

The corresponding pressure-drop form is

The Darcy–Weisbach friction factor in laminar flow is

The Darcy–Weisbach head loss is

The symbol f denotes the Darcy friction factor, not the Fanning friction factor. The relation f = 64/Re applies only to fully developed laminar flow in a circular pipe with Newtonian-fluid behavior. The fourth-power dependence on diameter makes diameter especially influential in low-Reynolds-number systems.

4. Turbulent Flow and the Moody Chart

In turbulent flow, wall roughness and Reynolds number affect the friction factor. Relative roughness is

where ε is the characteristic absolute roughness height. The Colebrook–White equation is widely used for commercial turbulent pipe flow:

It is implicit because f occurs on both sides. It applies principally to fully developed turbulent flow in pipes and requires consistent use of absolute roughness and internal diameter. An explicit approximation useful for hand calculation is the Swamee–Jain equation:

The Moody chart graphically represents friction factor as a function of Reynolds number and relative roughness. The procedure is: calculate Re, calculate ε/D, locate the operating point, and read f. In the hydraulically smooth region, roughness has little influence. In the fully rough region, friction factor is nearly independent of Reynolds number and depends mainly on ε/D.

Worked Example 1: Friction Loss in a Water Main

Water at approximately 20 °C flows through a commercial steel pipe of diameter 0.10 m and length 40 m at Q = 0.010 m³/s. Take ν = 1.00×10^−6 m²/s, ε = 0.046 mm, and g = 9.81 m/s².

The area and velocity are

Thus,

The relative roughness is ε/D = 0.000046/0.10 = 0.00046. Using Swamee–Jain gives approximately f = 0.019. Therefore,

The pressure equivalent is ρgh_f, approximately 6.2 kPa for water. The result is valid as a fully developed turbulent estimate; entrance and fitting losses must be added for a complete system calculation.

5. Minor or Local Losses

Losses caused by entrances, exits, bends, valves, contractions, expansions, and tees are called minor losses, although they can dominate short or highly fitted systems. They are represented by

where K is a dimensionless loss coefficient and V is the appropriate reference velocity. The reference section must be identified because a fitting may connect pipes of different diameters.

For an abrupt expansion, an idealized result based on separation and momentum considerations is the Borda–Carnot expression:

For a gradual contraction or expansion, K depends on angle, area ratio, and geometry. Manufacturer data are preferable for valves and complex fittings. A sharp-edged entrance often has a coefficient near 0.5, but actual values vary with geometry and approach conditions. A pipe exit into a large reservoir has approximately K = 1 when the pipe velocity is used as reference, because the velocity head is dissipated in the reservoir.

The total loss is

or, when all local losses use the same velocity,

This compact form must not be used blindly when pipe diameters change.

6. Equivalent Length and Equivalent Pipes

A fitting can be represented by an equivalent length L_e of straight pipe satisfying

Therefore,

Equivalent length depends on the friction factor, so it changes with operating condition and should be used cautiously. Direct K values are generally more transparent.

For pipes in series, the same discharge passes through every pipe, and losses add:

If a single equivalent pipe has the same discharge and loss, its resistance can be defined by

when friction factors and properties are treated as fixed over the operating range. This is an engineering linearization, not a universal exact law.

For pipes in parallel between common upstream and downstream junctions, the head loss is the same in each branch:

and continuity requires

Because friction factors can depend on branch flow through Reynolds number, parallel systems generally require iteration.

7. Pipe Networks and Solution Methods

A network is governed by two conditions:

1. continuity at every junction;
2. energy consistency around every closed loop.

For a loop, algebraic head changes and losses must sum to zero. In a simple network, one may assume a flow split, calculate branch losses, and adjust the split until branch head losses match. More complex networks can be solved using Hardy Cross, Newton–Raphson, or specialized software.

The Hardy Cross method begins with assumed flows satisfying continuity. For each loop, calculate the residual head imbalance. A correction is then applied to loop flows. With a loss model of the form h = rQ|Q|, an often-used correction is

The sign convention must be consistent, and the correction is applied with opposite signs in branches shared by adjacent loops. Modern computational methods are usually faster and more robust, but Hardy Cross remains valuable for understanding network structure.

8. Pipe Sizing and Pump Requirements

Pipe diameter is selected by balancing capital cost, energy cost, allowable velocity, pressure requirements, noise, erosion, and maintenance. Increasing diameter generally reduces velocity and friction loss but increases material and installation cost. A rational design process is:

1. establish required flow and operating range;
2. identify fluid properties and allowable pressure or velocity limits;
3. select candidate diameters;
4. calculate Reynolds number, roughness, major and minor losses;
5. determine pump head and power;
6. check transient, structural, regulatory, and economic constraints;
7. select a commercially available size and verify off-design operation.

Pump hydraulic power is

If pump efficiency is η_p, shaft power is

For a pump connecting two reservoirs, both reservoir surface velocities are negligible. If both surfaces are exposed to the same atmospheric pressure, pressure-head terms cancel, and the pump head is approximately elevation rise plus all system losses:

Worked Example 2: Pump Head and Power

A pump supplies Q = 0.020 m³/s from a lower reservoir to an upper reservoir 18 m higher. The total pipe and fitting loss at this flow is 7.0 m. Water density is 998 kg/m³ and pump efficiency is 0.78.

A motor would normally be selected above this calculated value to accommodate uncertainty, control range, and starting conditions.

9. Energy-Grade and Hydraulic-Grade Lines

The energy-grade line is the locus of total head:

The hydraulic-grade line is the piezometric head:

Their separation is approximately V²/(2g) when α ≈ 1. In the direction of flow, the EGL falls because of losses, jumps upward across a pump, and drops across a turbine. The HGL follows the EGL minus velocity head. A sudden expansion produces a loss and changes velocity head; a valve creates a localized downward step.

The HGL is especially useful for detecting negative gauge pressure, possible cavitation, and inadequate clearances. If the HGL lies below the pipe centerline, the gauge pressure may be negative. This is not automatically unsafe, but it requires checking vapor pressure, pipe strength, air admission, and transient effects.

10. Illustrative Case: Cooling-Water Distribution in a Machine Plant

Background

An illustrative manufacturing plant must deliver cooling water from a storage tank to three machine tools. The proposed system includes a pump, a 60 m main, two branches, isolation valves, elbows, and return piping.

Problem

The first design used a small main pipe to reduce installation cost. During simultaneous operation, the most distant machine received insufficient flow, and measured pump discharge pressure was higher than expected.

Analysis

The engineer first verified flow continuity and measured the actual internal diameters. Reynolds numbers showed turbulent flow. The initial model included straight-pipe friction but omitted several valve and tee losses. The fittings were then assigned manufacturer-based K values, and the network was solved iteratively. The HGL showed a large decline across the undersized main and a further decline in the distant branch. The pump operating point was also checked against its characteristic curve rather than assuming constant head.

Approach and Solution

The main was increased by one commercial diameter, and balancing valves were installed in the shorter branches. The pump control was set to maintain the required differential pressure, while a minimum-flow bypass protected the pump during low demand. The revised design included a transient review for rapid valve closure and a water-quality plan to limit corrosion and biological growth.

Outcome

The redesign reduced friction loss, equalized branch delivery, and moved the pump closer to its efficient operating region. Although pipe capital cost increased, energy use and production interruptions decreased. The case demonstrates that pipe sizing must consider the whole network, not merely the cheapest initial pipe.

Learning Points

• omitted minor losses can be important in compact systems;
• parallel branches must be analyzed with equal branch head loss;
• pump curves and system curves must be considered together;
• measured dimensions and operating data improve model reliability;
• controls, water quality, and transients are part of engineering design.

11. Limitations, Safety, and Contemporary Practice

The Darcy–Weisbach method is powerful but depends on reliable roughness, viscosity, geometry, and loss coefficients. It does not by itself model water hammer, multiphase flow, non-Newtonian behavior, severe cavitation, flexible-wall interaction, or rapidly varying flow. For those cases, transient or specialized computational models are required.

Professional design should follow applicable plumbing, pressure-vessel, fire-protection, environmental, and occupational-safety regulations. Engineers must prevent contamination, leaks, flooding, excessive noise, and unsafe pressure conditions. Material selection should address corrosion, temperature, chemical compatibility, and life-cycle impacts.

Digital twins, sensor networks, computational fluid dynamics, and AI-assisted anomaly detection increasingly support pipe-system operation. These tools do not eliminate conservation laws or engineering judgment. AI predictions should be validated against measured flow, pressure, calibration records, and physically plausible bounds. Transparent assumptions and version-controlled models are essential for safety-critical systems.

Best-Practice Checklist

• use actual internal diameter, not nominal pipe size;
• maintain consistent units and fluid properties at operating temperature;
• distinguish Darcy and Fanning friction factors;
• include entrances, exits, fittings, valves, and changes in diameter;
• verify whether flow is developing or fully developed;
• use manufacturer data for important fittings;
• check both normal and off-design conditions;
• plot EGL and HGL for complex systems;
• compare predicted and measured pressures and flow rates;
• document assumptions, uncertainty, and safety margins.

Chapter Summary

Internal pipe-flow analysis combines continuity and the steady-flow energy equation with models for friction and local losses. Reynolds number classifies the flow regime, while entrance length indicates whether a velocity profile is fully developed. Fully developed laminar flow follows Hagen–Poiseuille behavior and has Darcy friction factor f = 64/Re. Turbulent friction depends on Reynolds number and relative roughness and may be obtained from the Moody chart or Colebrook-type equations.

Major losses are calculated with the Darcy–Weisbach equation, while fittings and geometric disturbances are represented by loss coefficients. Pipes in series carry the same flow and have additive losses; parallel branches have equal head loss and divide the total flow according to their hydraulic resistances. Pipe sizing requires technical, economic, safety, and operational evaluation. EGL and HGL provide a powerful visual means of interpreting pressure, velocity head, pumps, turbines, and losses. Real systems require attention to transients, measurement, regulation, sustainability, and model limitations.

Key Terms

Reynolds number: Ratio of inertial to viscous effects in a flow.

Laminar flow: Flow with predominantly orderly layers and limited transverse mixing.

Turbulent flow: Flow containing velocity fluctuations and strong mixing.

Entrance length: Distance required for the velocity profile to become fully developed.

Fully developed flow: Flow whose velocity profile no longer changes in the streamwise direction.

Darcy friction factor: Dimensionless factor used in the Darcy–Weisbach equation.

Relative roughness: Ratio of absolute roughness to pipe diameter.

Moody chart: Graph relating friction factor to Reynolds number and relative roughness.

Major loss: Head loss caused by wall friction along a pipe length.

Minor loss: Local head loss caused by fittings, entrances, exits, or geometry changes.

Loss coefficient: Dimensionless coefficient K relating local loss to velocity head.

Equivalent length: Length of straight pipe producing the same loss as a fitting.

Pipe in series: Arrangement in which the same flow passes through successive pipes.

Parallel pipes: Branches connecting common upstream and downstream junctions.

Hydraulic grade line: Elevation plus pressure head.

Energy grade line: Hydraulic grade line plus velocity head.

Pump head: Energy per unit weight added to the fluid by a pump.

System curve: Relationship between required system head and flow rate.

Relative roughness: Absolute roughness divided by internal diameter.

Cavitation: Vapor formation and collapse caused by pressure falling below vapor pressure.

Review Questions

Short Answer Questions

1. Define Reynolds number and state its significance in pipe flow.
2. What is the physical meaning of entrance length?
3. Distinguish major losses from minor losses.
4. State the difference between the EGL and HGL.
5. Why must branch head losses be equal in a parallel pipe system?

Long Answer Questions

1. Explain laminar and turbulent pipe flow, including the role of viscosity, roughness, and Reynolds number.
2. Derive the Hagen–Poiseuille relation from the parabolic velocity profile and state its assumptions.
3. Explain how the Moody chart is used to determine turbulent friction factor.
4. Describe a systematic procedure for sizing a pipe and selecting a pump.
5. Explain the analysis of pipes in series and parallel, including the governing continuity and energy conditions.

Analytical / Application Questions

1. Water flows through a 75 mm pipe at a specified discharge. Develop a calculation sequence to determine Reynolds number, friction factor, major loss, and total loss including two fittings. State all data required.
2. A parallel network delivers water to two consumers, but one receives excessive flow. Explain how measurements, branch resistance, balancing valves, and HGL analysis could be used to correct the distribution.
3. A pump system experiences low pressure at the suction side during peak demand. Use the energy equation and HGL concepts to identify possible causes and recommend technically justified checks and remedies.

Multiple Choice Questions with Answers

Question 1

For fully developed laminar flow in a circular pipe, the Darcy friction factor is:

A. 64/Re

B. Re/64

C. 0.25/Re

D. 64Re

Correct Answer: A. 64/Re

Question 2

Which quantity directly represents the ratio of inertial to viscous effects in pipe flow?

A. Froude number

B. Reynolds number

C. Mach number

D. Strouhal number

Correct Answer: B. Reynolds number

Question 3

A local fitting loss is commonly calculated using which expression?

A. h_m = K V²/(2g)

B. h_m = 2gV²/K

C. h_m = fD/(LV²)

D. h_m = Q/A

Correct Answer: A. h_m = K V²/(2g)

Question 4

For two branches connected in parallel between the same junctions, which condition applies?

A. The velocity is always equal in both branches

B. The pipe diameters must be equal

C. The head loss is equal in each branch

D. The Reynolds number is equal in each branch

Correct Answer: C. The head loss is equal in each branch

Question 5

A pipe has diameter 0.10 m and discharge 0.010 m³/s. Its mean velocity is approximately:

A. 0.127 m/s

B. 1.27 m/s

C. 12.7 m/s

D. 127 m/s

Correct Answer: B. 1.27 m/s

Question 6

The vertical separation between the EGL and HGL in a uniform pipe is approximately:

A. Pressure head

B. Elevation head

C. Velocity head

D. Friction factor

Correct Answer: C. Velocity head

Question 7

In the fully rough turbulent region, the friction factor depends primarily on:

A. Relative roughness

B. Pipe length only

C. Pump efficiency only

D. Elevation only

Correct Answer: A. Relative roughness

Question 8

For a pump delivering 0.020 m³/s through a head of 25 m to water of density 998 kg/m³, the hydraulic power is closest to:

A. 0.49 kW

B. 2.45 kW

C. 4.90 kW

D. 49.0 kW

Correct Answer: C. 4.90 kW

Question 9

For pipes in series, which quantity is the same through each pipe under steady incompressible conditions?

A. Head loss

B. Discharge

C. Velocity, regardless of diameter

D. Reynolds number

Correct Answer: B. Discharge

Question 10

The Colebrook–White equation is used primarily to estimate:

A. Pump efficiency

B. Entrance length in laminar flow

C. Turbulent Darcy friction factor

D. Fluid density

Correct Answer: C. Turbulent Darcy friction factor