πŸŽ“ Lesson 5 D3

Building Your First EPANET Model

EPANET is a free computer program that lets engineers simulate how water flows and pressure changes in pipes, tanks, and pumps of a drinking water system.

🎯 Learning Objectives

  • βœ“ Construct a basic EPANET model from field data including pipe lengths, diameters, and node elevations
  • βœ“ Analyze hydraulic performance by interpreting pressure head and flow results at critical nodes
  • βœ“ Apply demand allocation methods to assign residential water consumption to junctions
  • βœ“ Validate model accuracy by comparing simulated pressures against measured field data

πŸ“– Why This Matters

Every day, millions rely on safe, reliable drinking water β€” but leaks, low pressure, or contamination can go unnoticed until it’s too late. EPANET allows engineers to virtually 'test' design changes, emergency responses, or infrastructure upgrades before spending a dollar on construction. In mining, it’s essential for camp water systems, dewatering networks, and tailings facility supply lines β€” where failure risks safety, compliance, and production.

πŸ“˜ Core Principles

An EPANET model rests on three foundational layers: (1) Topology β€” the physical layout of pipes, nodes, and components; (2) Hydraulics β€” governed by the continuity equation (mass balance) and energy equation (head loss via Hazen-Williams or Darcy-Weisbach); and (3) Time-based simulation β€” enabling analysis of diurnal demand patterns or pump scheduling. Nodes represent points of demand, supply, or elevation change; links represent flow paths with resistance properties. Model calibration ensures simulated outputs match real-world measurements β€” the cornerstone of credible analysis.

πŸ“ Hazen-Williams Head Loss

The Hazen-Williams equation estimates frictional head loss in water distribution pipes under turbulent flow β€” the default method in EPANET for municipal and mining camp systems due to its empirical fit for common pipe materials and flow velocities.

Hazen-Williams Head Loss

h_f = 10.67 Γ— L Γ— Q^1.852 / (C^1.852 Γ— D^4.871)

Calculates frictional head loss (m) in full-flow circular pipes under turbulent conditions.

Variables:
SymbolNameUnitDescription
h_f Friction head loss m Energy loss due to pipe wall friction
L Pipe length m Centerline distance between nodes
Q Volumetric flow rate mΒ³/s Flow through the pipe
C Hazen-Williams roughness coefficient unitless Empirical measure of pipe interior smoothness
D Internal pipe diameter m Hydraulic diameter for circular pipes
Typical Ranges:
New HDPE pipe: 140 – 150
Aged cast iron: 80 – 100
Mining camp PVC distribution: 130 – 150

πŸ’‘ Worked Example

Problem: A 300-mm HDPE pipe (C = 150) carries 45 L/s over 650 m. Ground elevation drops 12 m from start to end. Calculate head loss and residual pressure at downstream node if upstream pressure is 52 m head.
1. Step 1: Convert flow Q = 45 L/s = 0.045 mΒ³/s; diameter D = 0.3 m; L = 650 m; C = 150
2. Step 2: Apply Hazen-Williams: h_f = 10.67 Γ— L Γ— Q^1.852 / (C^1.852 Γ— D^4.871) = 10.67 Γ— 650 Γ— (0.045)^1.852 / (150^1.852 Γ— 0.3^4.871)
3. Step 3: Compute: h_f β‰ˆ 8.3 m. Elevation drop = 12 m β†’ downstream elevation = upstream βˆ’ 12. Residual head = 52 βˆ’ 8.3 = 43.7 m. Residual pressure = 43.7 βˆ’ (downstream elevation) β€” verify against minimum required (e.g., 20 m for mining camps per WHO/MSA guidelines).
Answer: The head loss is 8.3 m, yielding a residual pressure head of ~43.7 m β€” well above the 20–30 m minimum recommended for remote mining operations.

πŸ—οΈ Real-World Application

At the Telfer Gold Mine (Western Australia), engineers used EPANET to redesign the camp water network after population doubled. They modeled 12 km of PVC piping, 3 elevated tanks, and 5 booster pumps. By simulating peak morning demand (2.8 L/s per person), they identified two low-pressure zones (<15 m head) near dormitory blocks. Adding a small booster station and adjusting tank setpoints increased minimum pressure to 28 m head β€” verified via 17 field pressure loggers β€” avoiding costly pipe replacement.

✏️ Model-Building Exercise

Using the provided sketch (4 junctions, 5 pipes, 1 reservoir, 1 tank), assign: pipe lengths (120–450 m), diameters (150–300 mm), Hazen-Williams C values (140 for aged steel, 150 for new HDPE), and nodal demands (35–120 L/s). Build the model in EPANET v2.2, run a 24-hr extended period simulation with diurnal pattern (pattern factor peaks at 7–9 AM and 6–8 PM), and report: (a) minimum pressure head at Junction J3, (b) maximum velocity in Pipe P2, and (c) tank level fluctuation range. Submit .inp file and summary table.

πŸ“‹ Case Connection

πŸ“‹ Calibration of Lagos Metropolitan Water Network

Persistent model–field mismatch (>25% pressure error) due to undocumented pipe replacements and unaccounted demand growt...

πŸ“‹ Water Quality Model Validation for Singapore’s Deep Tunnel Sewerage System (DTSS) Supply Branch

Disinfectant residual dropping below 0.2 mg/L at farthest nodes despite design dosing; suspected wall reaction dominance

πŸ“š References