Flow in Pipes: Systems & Networks - Theory & Concepts

Flow in Pipes: Systems & Networks

$Learning Objectives$

Apply continuity and energy conservation to series, parallel, branching, and looped pipe systems.
Express pipe and local losses in signed nonlinear resistance form and derive equivalent-system relations.
Solve junction and three-reservoir problems without losing flow-direction information.
Apply Hardy Cross corrections with consistent loop signs and convergence checks.
Explain nodal-head methods, pressure-dependent demand, pumps, valves, tanks, and extended-period simulation.
Distinguish steady-state network adequacy from transient pressure safety.
Apply wave-celerity, Joukowsky, and valve-closure relations only under their stated hydraulic-transient assumptions.
Select and evaluate practical surge-mitigation measures for both high- and low-pressure transients.

$Pipe network$

A connected set of hydraulic nodes and links whose flows and heads must simultaneously satisfy continuity, energy relations, and component operating constraints.

$Network Elements and Unknowns$

Nodes: junctions, reservoirs, tanks, outlets, hydrants, and demand points.
Links: pipes, pumps, turbines, valves, meters, and other connecting components.
Boundary conditions: fixed reservoir heads, tank states, prescribed inflows or demands, pump schedules, and valve controls.
Typical unknowns: link discharge, junction head, pressure, tank level, and active control state.

$Signed head-loss relation$

A component equation that preserves both head-loss magnitude and the direction associated with an assumed positive discharge.

General Signed Resistance Form

Represents a monotonic link loss with a power-law exponent while preserving flow direction.

h_L=rQ|Q|^{n-1}

Variables

Symbol	Description	Unit
$h_L$	Signed head change attributed to loss in the chosen positive direction	m
$r$	Resistance coefficient consistent with the chosen loss law	-
$Q$	Signed discharge	$m^3/s$
$n$	Flow exponent	-

Darcy-Weisbach Resistance at Fixed f

Discharge form of Darcy-Weisbach when the Darcy factor is treated as known for the iteration.

h_f=r_fQ|Q|

r_f=\frac{8fL}{g\pi^2D^5}

Variables

Symbol	Description	Unit
$r_f$	Darcy major-loss resistance coefficient	$s^2/m^5$
$f$	Darcy friction factor	-
$L$	Pipe length	m
$D$	Inside diameter	m

$Darcy Resistance Changes When f Changes$

The quadratic discharge form is exact only while $f$ is treated as fixed. In a network iteration, changing discharge alters Reynolds number and can alter $f$ . A robust Darcy-Weisbach solver updates the factor or includes its derivative in the nonlinear solution.

Local-Loss Resistance

Expresses a lumped local coefficient in discharge form for a circular pipe.

h_m=r_KQ|Q|

r_K=\frac{8K}{g\pi^2D^4}

Variables

Symbol	Description	Unit
$r_K$	Local-loss resistance coefficient	$s^2/m^5$
$K$	Sum of local coefficients referenced to the pipe velocity	-
$D$	Pipe diameter associated with the reference velocity	m

$Pipes in series$

Pipes connected sequentially without intermediate external inflow or withdrawal, so the same discharge passes through every series element.

$Series-System Rules$

For pipes in series,

Q_1=Q_2=\cdots=Q

and the total irreversible head loss is

h_{L,\text{total}}=\sum_i h_{L,i}

If all elements obey the same exponent $n$ , their resistance coefficients add: $r_{\text{eq}}=\sum_i r_i$ .

Series and parallel pipe systems

Series paths share flow; parallel branches share endpoint head difference.

$Pipes in parallel$

Branches that connect the same pair of hydraulic nodes, so each branch experiences the same node-to-node head difference while branch discharges sum algebraically.

Parallel Equivalent Resistance

Equivalent resistance for branches sharing the same positive flow direction and common power-law exponent.

r_{\text{eq}} = \left( \sum_i r_i^{-1/n} \right)^{-n}

Variables

Symbol	Description	Unit
$r_{\text{eq}}$	Equivalent parallel resistance	-
$r_i$	Branch resistance	-
$n$	Common positive flow exponent	-

$Equivalent-Pipe Results Are Model-Specific$

An equivalent diameter or resistance depends on the selected loss law, lengths, roughness, local losses, exponent, and flow range. An equivalent derived with fixed- $f$ Darcy-Weisbach is not automatically equivalent under Hazen-Williams or after Reynolds number changes.

$Interactive Series and Parallel Network Tool$

The network simulation visualizes continuity and equal node-to-node head change. In design work, update flow-dependent friction factors rather than treating all resistances as permanently fixed.

Two-Pipe System Solver

Learning objective: Observe how branch resistance and network arrangement redistribute discharge while satisfying continuity and energy compatibility.

Uses Darcy–Weisbach major loss plus lumped minor losses: $h_L=\left(fL/D+\sum K\right)V^2/(2g)$ .

Total system discharge (

Q

)100 L/s

Pipe 1

Inside diameter150 mm

Length100 m

Darcy friction factor0.020

Total minor-loss

K

1.0

Pipe 2

Inside diameter150 mm

Length100 m

Darcy friction factor0.020

Total minor-loss

K

1.0

Results

Pipe 1 flow

50.00 L/s

V₁ = 2.83 m/s

Pipe 2 flow

50.00 L/s

V₂ = 2.83 m/s

Pipe 1 head loss

5.848 m

Pipe 2 head loss

5.848 m

Common branch head loss

5.848 m

Continuity residual: 0.0e+0 m³/s
Parallel energy residual: 0.0e+0 m

The friction factors are user-supplied Darcy factors and are treated as constant. In a full design iteration they must be recomputed from Reynolds number and roughness as flow changes. Pumps, valves with control laws, and elevation differences require the full energy equation rather than this two-branch resistance model.

$Junction continuity$

The requirement that mass does not accumulate at a zero-storage node, so algebraic inflow equals external withdrawal plus algebraic outflow.

Junction Continuity

Steady incompressible node balance using a consistent sign convention.

\sum Q_{\text{in}}-\sum Q_{\text{out}}-Q_d=0

Variables

Symbol	Description	Unit
$Q_d$	External demand or withdrawal at the node	$m^3/s$

$Three-reservoir problem$

A junction-head problem in which three reservoirs of known head connect through hydraulic resistances to one common node of unknown head.

Signed Reservoir-Branch Relation

Relates each reservoir head to junction head and signed branch flow.

H_i-H_J=r_iQ_i|Q_i|^{n-1}

Variables

Symbol	Description	Unit
$H_i$	Known reservoir head	m
$H_J$	Unknown junction head	m
$Q_i$	Signed branch discharge under the chosen convention	$m^3/s$

$Three-Reservoir Solution with Direction Checks$

STEP-BY-STEP

Choose a sign convention, such as flow toward the junction positive.
Bound the trial junction head between physically relevant reservoir heads.
For a trial $H_J$ , determine the sign of each $H_i-H_J$ before calculating flow magnitude.
Evaluate the algebraic continuity residual.
Use bisection, safeguarded Newton iteration, or another reliable root method.
Confirm final branch directions from the solved head gradients.

Three-reservoir junction

Reservoir heads and an unknown junction head establish signed branch flows.

$Square Roots Give Magnitudes, Not Directions$

For a quadratic loss law, $\sqrt{|H_i-H_J|/r_i}$ gives only the magnitude. The sign must come from the head difference. Discarding sign information can produce a numerically tidy but physically impossible junction solution.

$Looped pipe network$

A network containing one or more closed hydraulic paths, so flows are coupled by both node continuity and path energy conservation.

$Network Conservation Conditions$

A physically valid steady network solution satisfies:

Continuity at every node: algebraic inflow and outflow balance the prescribed external demand.
Energy around every closed loop: the algebraic sum of head changes is zero.
Path independence: the head difference between two nodes is the same regardless of the path used to compute it.

$Hardy Cross method$

An iterative loop-balancing procedure that starts from flows satisfying continuity and applies loop corrections to reduce energy residuals.

Hardy Cross Flow Correction

Loop correction for links obeying h=rQ|Q|^(n-1), with each signed head loss referenced to the chosen loop direction.

\Delta Q = -\frac{\sum h_i} {\sum n_i r_i|Q_i|^{n_i-1}}

Variables

Symbol	Description	Unit
$\Delta Q$	Loop flow correction in the chosen traversal direction	$m^3/s$
$h_i$	Signed head loss of link i relative to the loop direction	m
$n_i$	Loss exponent for link i	-

$Hardy Cross Loop Procedure$

STEP-BY-STEP

Assign initial link directions and magnitudes that satisfy node continuity.
Select a loop traversal direction and sign every link head loss relative to it.
Calculate the loop energy residual and derivative denominator.
Apply $\Delta Q$ algebraically to links in that loop.
Apply corrections from all loops to shared links.
Update Darcy factors or other flow-dependent coefficients when required.
Continue until both node and loop residuals meet the stated tolerance.

Hardy Cross loop correction

Loop corrections reduce energy residual while preserving nodal continuity.

$Convergence Is a Numerical Criterion, Not Proof of Good Design$

Small corrections indicate that the chosen nonlinear equations are nearly satisfied. They do not prove adequate pressure, acceptable velocity, valid component states, water quality, or transient safety.

$Nodal-head method$

A simultaneous network solution that treats junction heads as primary unknowns and enforces continuity using head-dependent link-flow relations.

Nodal Continuity Residual

Generic nonlinear residual equation solved for a junction head.

R_j(H) = \sum_i Q_{ij}(H_j,H_i)-Q_{d,j}=0

Variables

Symbol	Description	Unit
$R_j$	Continuity residual at node j	$m^3/s$
$H_j$	Unknown head at node j	m
$Q_{d,j}$	External demand at node j	$m^3/s$

$Demand Models and Feasibility$

Demand-driven: prescribed withdrawal is enforced regardless of computed pressure; useful when pressure remains adequate.
Pressure-dependent: delivered demand decreases when available pressure is insufficient.
Emitter or leak model: a relation such as $Q=Cp^m$ approximates pressure-sensitive outflow.

A demand-driven solution with strongly negative pressure can be mathematically converged but physically infeasible.

$Pump operating point$

The discharge and head at which the pump characteristic and the complete hydraulic system requirement are simultaneously satisfied.

$Pumps, Valves, and Controls$

A pump adds head, so it acts as a negative-loss element in a signed path equation. Throttle valves add loss, while pressure-reducing, pressure-sustaining, and flow-control valves change their state to meet a target when physically possible. Check valves change network topology when they close. Control assumptions must be verified after each nonlinear solution.

$Extended-period simulation$

A sequence of hydraulic solutions linked through time-varying demands, controls, pump schedules, and tank storage.

Tank Storage Balance

Relates tank volume change to algebraic inflow and outflow.

\frac{dV}{dt}=\sum Q_{\text{in}}-\sum Q_{\text{out}}

Variables

Symbol	Description	Unit
$V$	Stored tank volume	$m^3$
$t$	Time	s

$Steady Network Design Checks$

After solving, evaluate service pressure, peak demand, fire flow, velocity, friction loss, pump range, NPSH, tank turnover, redundancy, isolation, leakage, energy use, and future-demand uncertainty. A single converged peak-hour case is not a complete network design.

$Water hammer$

A hydraulic transient in which a change in velocity generates pressure waves that propagate through the compressible liquid and deformable pipe system.

$Wave celerity$

The propagation speed of a small pressure disturbance through the coupled fluid-pipe system.

Simplified Elastic Wave Celerity

Thin-wall approximation for a liquid-filled elastic pipe under a simplified restraint condition.

a= \sqrt{ \frac{K/\rho} {1+KD/(Ee)} }

Variables

Symbol	Description	Unit
$a$	Pressure-wave celerity	m/s
$K$	Liquid bulk modulus	Pa
$\rho$	Liquid density	$kg/m^3$
$D$	Inside pipe diameter	m
$E$	Pipe Young's modulus	Pa
$e$	Pipe wall thickness	m

$Celerity Depends on Pipe Restraint and Wall Model$

The displayed equation is a simplified thin-wall form. Actual wave speed can depend on Poisson effects, axial restraint, pipe-wall geometry, lining, viscoelastic behavior, entrained or trapped gas, and fluid properties. Use a transient model consistent with the real pipe system.

$Joukowsky relation$

The one-dimensional elastic-wave relation between a rapid change in liquid velocity and the associated pressure change at a wavefront.

Joukowsky Pressure and Head Change

Ideal elastic-wave relation for a sufficiently rapid velocity change before system reflections alter the local event.

\Delta p=\rho a\,\Delta V

\Delta H=\frac{a\,\Delta V}{g}

Variables

Symbol	Description	Unit
$\Delta p$	Signed pressure change associated with the wave	Pa
$\Delta H$	Equivalent signed head change	m
$\Delta V$	Signed velocity change	m/s

$Joukowsky Is Not a Universal Maximum-Surge Formula$

The relation describes the initial ideal wave response under one-dimensional elastic assumptions. Reflections, friction, branches, pump inertia, air pockets, valve law, column separation, and vapor cavities can produce later pressures that differ materially from the simple estimate.

$Characteristic round-trip wave time$

For a simple reservoir-pipe-valve reach of length $L$ , the time $2L/a$ for a pressure wave to travel to the upstream boundary and return.

Round-Trip Wave Time

Characteristic closure-time scale for a simple reservoir-pipe-valve configuration.

T_c=\frac{2L}{a}

Variables

Symbol	Description	Unit
$T_c$	Round-trip wave time	s
$L$	Relevant pipe length to the reflecting boundary	m

$Rapid and Slow Valve Closure$

In a simple reservoir-pipe-valve model, a closure completed before the first reflected relief wave returns is commonly called rapid relative to $T_c$ . A much slower closure can reduce the initial surge because boundary reflections interact with the continuing valve motion.

For a simplified frictionless system with approximately linear deceleration, one commonly taught estimate is

\Delta p\approx\frac{2\rho L V_0}{T}

This is a teaching approximation, not a universal valve-closure law.

Water-hammer wave travel

A valve disturbance propagates and reflects through the pipeline.

$Interactive Water-Hammer Model$

Use the simulator to explore celerity, round-trip time, and simplified rapid/slow closure estimates. Treat its results as one-dimensional screening calculations rather than a substitute for a method-of-characteristics or equivalent transient analysis.

Water Hammer Simulator

Learning objective: See how wave speed, pipe properties, initial velocity, and valve-closure time govern the idealized transient pressure response.

Elastic-pipe wave speed and rapid/slow valve-closure pressure rise.

Initial velocity (

V_0

)2.0 m/s

Valve closure time (

T

)0.00 s

Zero represents an ideal instantaneous closure.

Pipe length (

L

)100 m

Diameter (

D

)0.20 m

Wall thickness (

t

)10 mm

Fluid bulk modulus (

K

)2.20 GPa

Pipe elastic modulus (

E

)200 GPa

Fluid density (

\rho

)1000 kg/m³

Wave celerity (

c

)1343 m/s

Critical closure time (

T_c=2L/c

)0.149 s

Closure classificationRapid

Maximum pressure rise (

\Delta p

)2.69 MPa

Equivalent head rise (

\Delta H

)273.8 m

Normal flowCompression waveReflected low-pressure phase

Rapid closure uses the Joukowsky relation $\Delta p=\rho c\Delta V$ . For closure slower than $T_c=2L/c$ , this educational model uses the classical gradual-closure approximation $\Delta p\approx2\rho L\Delta V/T$ . Real systems may require method-of-characteristics analysis including friction, valve law, pipe restraint, entrained air, and vapor-column separation.

$Transient Protection$

Possible measures include controlled valve motion, pump ramping or flywheels, surge tanks, hydropneumatic or air vessels, relief valves, bypasses, non-slam check valves, air/vacuum valves, and changes in pipe diameter or pressure class. A measure that limits maximum pressure can worsen minimum pressure, so both extremes must be checked.

$Steady-State Results Do Not Bound Transient Pressure$

A system can have acceptable steady pressure yet experience transient overpressure, sub-atmospheric pressure, vapor formation, or column separation after a rapid event. Surge analysis must use absolute pressure limits and realistic boundary conditions.

$Pipe-System Analysis Workflow$

STEP-BY-STEP

Define node heads, elevations, demands, link data, component controls, and a consistent sign convention.
Select loss relations and identify which coefficients change with flow.
Solve continuity and energy equations simultaneously.
Verify residuals, directions, pressures, velocities, and active pump or valve states.
Evaluate peak, fire, low-demand, outage, and failure scenarios.
Perform extended-period analysis when tanks and controls change with time.
Perform transient analysis for rapid operations, long/high-velocity mains, pump trips, or other surge-sensitive conditions.
Check both maximum and minimum transient pressure against structural and vapor-pressure limits.

Key Takeaways

Series pipes share discharge and add losses; parallel branches share a node-to-node head difference and divide flow nonlinearly.
Signed head-flow relations preserve physical flow direction and prevent square-root sign mistakes.
Equivalent-pipe relations depend on the selected loss model and its assumptions.
Hardy Cross requires continuity-consistent initial flows, loop sign discipline, and residual-based convergence checks.
Modern network analysis often solves nodal heads and explicitly models pumps, valves, tanks, leaks, and pressure-dependent demand.
Wave celerity depends on both fluid compressibility and pipe-system elasticity and restraint.
Joukowsky and $2L/a$ are powerful transient concepts but are not universal bounds for complex networks.
Simplified slow-closure formulas must state their valve-law and system assumptions.
Surge protection must be evaluated for both high-pressure and low-pressure transients.

Previous TopicFlow in Pipes: Fundamentals & Losses - Examples & Applications

Quiz Me

Next TopicFlow in Pipes: Systems & Networks - Examples & Applications

Prev Next

Quiz Me