15.4 Main Maintenance, Unidirectional Flushing & Leak Detection

15.4 Main Maintenance, Unidirectional Flushing & Leak Detection

[!NOTE] Operational Excellence and Asset Management: Distribution infrastructure maintenance directly dictates regulatory compliance under Safe Drinking Water Act rules, fire suppression readiness under Insurance Services Office (ISO) ratings, and water conservation mandates enforced by the Arizona Department of Water Resources (ADWR) under the Groundwater Management Act. Certified operators must execute systematic preventative maintenance, manage real and apparent water losses, and apply rigorous disinfection protocols conforming to AWWA C651.

Water distribution systems are dynamic physical, chemical, and biological reactors. Without structured maintenance, valves seize in place, mineral scale and bacterial biofilms choke hydraulic carrying capacity, hidden subsurface leaks waste millions of gallons of scarce desert water, and pipe wall corrosion degrades water quality. Maintaining this subsurface infrastructure requires rigorous valve exercising, engineered unidirectional flushing, active acoustic leak detection, and advanced corrosion control.

---

Preventative Infrastructure Maintenance: Valves & Hydrants

Preventative maintenance ensures that distribution appurtenances function instantaneously during catastrophic main breaks, fire emergencies, or contamination containment events.

+-----------------------------------------------------------------------------------------+
|                        Distribution Preventative Maintenance                            |
+-----------------------------------------------------------------------------------------+
| Valve Exercising     | Cycles gate valves to clear mineral encrustation; records        |
|                      | operating torque, direction, and exact turn counts               |
| Fire Hydrant Flow    | AWWA M17 pitot testing; measures static, residual, and velocity  |
| Testing              | pressure; calculates available fire flow at 20 psi residual      |
| Unidirectional       | Systematic sequence from clean source outward; achieves scouring |
| Flushing (UDF)       | velocity ≥ 5.0 ft/sec; cleans pipe walls using 40% less water   |
| Pipeline Pigging     | Propels flexible polyurethane foam swabs or wire-brush pigs      |
|                      | under line pressure to scour heavy tuberculation and scale       |
+-----------------------------------------------------------------------------------------+

Valve Exercising Programs

Distribution valves that remain stationary for years become inoperable. In Arizona, groundwater supplies often exhibit elevated calcium carbonate hardness (> 250 mg/L as CaCO3) and dissolved silica. Minerals precipitate onto valve stems, bonnet guide tracks, and seating wedges. If an emergency main break occurs, operators attempting to close an unexercised valve frequently snap the operating stem, turning a routine isolation into an uncontrollable multi-block outage.

• Exercising Methodology: Operators use automated, truck-mounted hydraulic valve exercisers equipped with electronic torque sensors. The operator never forces a stuck valve closed. Instead, the machine cycles the valve back and forth: turning the stem several revolutions in the closing direction, reversing a few revolutions in the opening direction to allow high-velocity water to wash away sheared mineral scale, and repeating the cycle until full closure is achieved without exceeding maximum torque thresholds.
• Turn Counts: Standard resilient wedge and double-disc gate valves require a specific number of turns to complete full travel. A standard industry rule of thumb is:

$\text{Number of Turns} \approx (3 \times D) + (2 \text{ to } 3)$

(where D is the nominal pipe diameter in inches)

For example, an 8-inch valve requires approximately (3 * 8) + 2 = 26 to 27 turns. A 12-inch valve requires approximately (3 * 12) + 3 = 39 turns. If a valve stops rotating after only 15 turns, the operator knows the valve is obstructed by gravel or a bent stem. Once fully opened, the valve should be backed off a quarter-to-half turn to relieve packing stress and prevent thermal jamming.

Fire Hydrant Flow Testing (AWWA M17)

Hydrant flow testing determines available fire flows and evaluates the internal roughness (C-factor) of distribution mains. The test requires a minimum of two adjacent hydrants:

1. Test Hydrant (Residual Hydrant): Fitted with a calibrated pressure gauge on one nozzle cap. Operators record the initial Static Pressure (Ps) when no water is flowing. When the second hydrant flows, the operator records the Residual Pressure (Pr).
2. Flow Hydrant: Discharged fully through a 2.5-inch hose nozzle or 4.5-inch pumper nozzle. While flowing, the operator inserts a handheld pitot tube directly into the center of the discharge stream, holding the pitot orifice one-half the nozzle diameter away from the orifice face to record the Velocity Pressure (Pv).

• Discharge Equation: Flow rate (Q) is calculated using the hydraulic discharge formula:

$Q = 29.84 \times C_d \times D^2 \times \sqrt{P_v}$

Where:

• Q = Discharge flow rate (gallons per minute, gpm)

• Cd = Hydrant nozzle coefficient of discharge (0.90 for smooth, well-rounded outlet; 0.80 for square, sharp-edged outlet; 0.70 for rough nozzle projecting into barrel)

• D = Inside diameter of the flowing nozzle (inches)

• Pv = Velocity pressure measured by the pitot gauge (psi)

• Available Fire Flow at 20 psi Residual: To project the available flow rate (QR) at the mandatory 20 psi residual threshold, operators use the Hazen-Williams flow formula:

$Q_R = Q_F \times \left(\frac{P_s - 20}{P_s - P_r}\right)^{0.54}$

(where QF is the total measured flow during the test in gpm)

---

Water Main Flushing Methodologies & Pigging

Periodic main cleaning is vital to remove accumulated silt, iron and manganese precipitates, and bacterial biofilms that consume disinfectant residuals.

                 Flushing Methodologies Comparison

   CONVENTIONAL (RANDOM) FLUSHING       UNIDIRECTIONAL FLUSHING (UDF)
   ┌─────────────────────────────┐      ┌─────────────────────────────┐
   │ Hydrants opened arbitrarily │      │ Sequenced from clean source │
   │ Water pulls from all loops  │      │ Boundary valves closed      │
   │ Low velocity (< 2.5 ft/sec) │      │ High velocity (≥ 5.0 ft/sec)│
   │ Pulls dirty water everywhere│      │ Scours biofilm & scale      │
   │ High water waste / dirty tap│      │ Uses 40% less water         │
   └─────────────────────────────┘      └─────────────────────────────┘

Conventional Flushing vs. Unidirectional Flushing (UDF)

• Conventional (Random) Flushing: Operators open hydrants in a neighborhood without closing intermediate valves. Water flows toward the open hydrant from all directions through interconnected network loops. Flow velocities rarely exceed 2.0 to 3.0 ft/sec. This velocity is sufficient to stir up loose sediments but completely insufficient to scour pipe walls. Worse, conventional flushing draws contaminated, discolored water from older unlined cast iron mains across pristine PVC zones, generating customer red-water complaints and wasting enormous water volumes.
• Unidirectional Flushing (UDF): UDF is an engineered, systematic approach. Flushing always originates from a clean, high-pressure water source (such as a water treatment plant clearwell or primary transmission main) and proceeds systematically outward toward the system periphery. Prior to flowing a target hydrant, operators systematically close specific boundary gate valves to isolate a single pipe segment. Water is forced down a single pipeline from a single direction at high velocity.
  • The 5.0 ft/sec Scouring Velocity Threshold: UDF programs are engineered to achieve a minimum scouring velocity of 5.0 ft/sec (1.5 m/s). Fluid mechanics dictates that at velocities ≥ 5.0 ft/sec, turbulent boundary layer shear stress on the pipe wall increases dramatically. This hydraulic shear physically scrubs adherent bacterial biofilms, loose tuberculation scale, and settled mineral particulates off the pipe barrel, carrying them out through the open hydrant. UDF uses approximately 40% less water than conventional flushing, restores line carrying capacity (C-factors), and exercises distribution valves as an integral part of the process.

Pipeline Pigging

Where pipelines suffer severe tuberculation, heavy mineral scaling, or accumulated sand that cannot be dislodged by 5.0 ft/sec water velocity, utilities deploy pipeline pigging. Flexible, bullet-shaped polyurethane foam cylinders ('pigs')—ranging from soft open-cell foam swabs to high-density foam pigs wrapped with criss-cross polyurethane bands or hardened wire brushes—are inserted into the pipe through dedicated pig launchers. Distribution main pressure (typically 30 to 50 psi) propels the pig through the pipeline at 2 to 5 ft/sec, physically scraping encrustations from the walls and pushing the debris out through a downstream retrieval structure or modified fire hydrant.

---

Water Loss Auditing & AWWA M36 Methodology

In Arizona, where water supplies are governed by strict conservation mandates enforced by the Arizona Department of Water Resources (ADWR) within Active Management Areas (AMAs), utilities are held to rigorous water loss accounting standards under the AWWA M36 Water Audits and Loss Control Programs framework.

                      AWWA M36 Water Audit Architecture

   ┌────────────────────────────────────────────────────────────────────────┐
   │                       SYSTEM INPUT VOLUME                              │
   │        (Total finished water produced by plants and wellheads)         │
   ├───────────────────────────────────┬────────────────────────────────────┤
   │    AUTHORIZED CONSUMPTION         │            WATER LOSSES            │
   ├─────────────────┬─────────────────┼──────────────────┬─────────────────┤
   │ Billed          │ Unbilled        │ Apparent Losses  │ Real Losses     │
   │ Authorized      │ Authorized      │ (Paper Losses)   │ (Physical)      │
   │ - Billed Metered│ - Fire fighting │ - Customer meter │ - Main ruptures │
   │ - Billed Unmet. │ - Main flushing │   under-register.│ - Service leaks │
   │                 │ - Street clean. │ - Data errors    │ - Tank overflow │
   │                 │                 │ - Water theft    │                 │
   └─────────────────┴─────────────────┴──────────────────┴─────────────────┘
   │◄─────────────── REVENUE WATER ───►│◄──────── NON-REVENUE WATER ───────►│

Non-Revenue Water (NRW) Breakdown

Non-Revenue Water (NRW) represents water that is extracted, treated, and pumped into the distribution system but generates zero tariff revenue for the utility. NRW is divided into three distinct operational tiers:

1. Unbilled Authorized Consumption: Water authorized by utility policy for non-metered municipal operations, including fire department firefighting and training, unidirectional main flushing, street sweeping, and park maintenance.
2. Apparent Losses ('Paper Losses'): Water that reaches customer taps and is consumed, but is not properly recorded or billed due to measurement or administrative failures:
   • Customer Meter Under-Registration: Mechanical nutating disc and oscillating piston meters wear over time. As mechanical tolerances degrade, low flows (< 0.5 gpm) slip past the measuring chamber unrecorded. Aging residential meters systematically under-register consumption, costing the utility revenue while understating customer demand.
   • Billing and Data Handling Errors: Systematic software bugs, estimated meter reads, and data transfer discrepancies between Automated Meter Reading (AMR/AMI) systems and utility billing software.
   • Unauthorized Consumption: Direct water theft via illegal hydrant taps, unauthorized meter jumpers, and unmetered construction services.
3. Real Losses ('Physical Losses'): The physical escape of potable water from the pressurized network before reaching customer meters:
   • Catastrophic distribution main breaks and joint blowouts
   • Chronic subsurface pinhole leaks on mains and utility-owned service laterals
   • Finished water storage reservoir floor seepage and structural overflow discharges

• Infrastructure Leakage Index (ILI): The premier operational metric of the AWWA M36 methodology is the Infrastructure Leakage Index (ILI), which represents the ratio of Current Annual Real Losses (CARL) to the theoretical Unavoidable Annual Real Losses (UARL):

$\text{ILI} = \frac{\text{CARL}}{\text{UARL}}$

An ILI value near 1.0 indicates world-class physical loss management, whereas an ILI greater than 5.0 indicates severe, chronic deferred maintenance and structural network leakage.

---

Active Leak Detection Technologies

Because desert soils (caliche, porous sands, fractured basalt) are often highly permeable, pressurized water escaping from broken pipes rarely surfaces. Water tracks downward into dry alluvial aquifers for months without surfacing, dissipating millions of gallons. Utilities utilize sophisticated non-invasive active leak detection technologies:

1. Acoustic Listening Sticks & Ground Microphones: Turbulent water jetting through a pipe crack or failed gasket vibrates the pipe wall and surrounding soil, generating acoustic frequencies between 500 Hz and 2,000 Hz. Operators place mechanical listening rods directly onto valves, hydrants, and customer curb stops to detect acoustic hiss. High-sensitivity ground microphones placed along the pavement surface isolate the point of peak acoustic amplitude.
2. Digital Leak Noise Correlators: Operators deploy dual electronic acoustic sensors on two valves or hydrants flanking a suspected leak location. The sensors transmit audio waveforms via radio to a central processor. The processor executes a mathematical cross-correlation to measure the precise time differential (Δt) required for the leak sound wave to reach Sensor 1 versus Sensor 2. Knowing the speed of sound through the specific pipe material (v) and the distance between sensors (D), the correlator calculates the exact distance (L1) to the subsurface leak within inches:

$L_1 = \frac{D - (v \times \Delta t)}{2}$

1. Permanent Acoustic Noise Loggers: Small magnetic acoustic loggers deployed permanently on valve operating nuts throughout the grid. Loggers are programmed to record acoustic amplitude between 2:00 AM and 4:00 AM, when ambient street traffic noise and domestic consumption are minimal. Persistent high-amplitude anomalies are flagged and transmitted via cellular network to utility dispatch for pinpointing.
2. Satellite Synthetic Aperture Radar (SAR): Satellite-mounted L-band microwave radar penetrates asphalt, concrete, and dry desert soils to depths of several feet. The radar system measures the dielectric signature of the subsurface soil, identifying the unique electromagnetic resonance of treated, chlorinated drinking water leaks beneath street pavement while filtering out natural groundwater signatures.

---

Internal and External Corrosion Mitigation

Corrosion deteriorates pipe structural integrity, leaches toxic metals (lead and copper) into drinking water, and drives main rupture rates.

Internal Pipe Corrosion Control

Internal corrosion involves electrochemical oxidation of the pipe metal by dissolved constituents in potable water.

• Langelier Saturation Index (LSI): A chemical equilibrium model evaluating the tendency of water to either precipitate or dissolve a protective layer of calcium carbonate (CaCO3) on internal pipe walls:

$\text{LSI} = \text{pH} - \text{pH}_s$

(where pH is measured water pH, and pHs is the saturation pH calculated from calcium hardness, total alkalinity, dissolved solids, and temperature)

• Corrosion Inhibitors: In addition to pH and alkalinity adjustment (using sodium hydroxide, lime, or soda ash), utilities dose orthophosphate or zinc orthophosphate (0.5 to 2.0 mg/L as PO4). Orthophosphate reacts chemically with divalent metal ions on the pipe wall to form an insoluble microscopic passivation barrier (such as insoluble lead carbonate, lead phosphate, or zinc hydroxide complexes), preventing metal leaching under the EPA Lead and Copper Rule.

External Soil Corrosion Mitigation

Arizona desert soils are notoriously aggressive toward metallic infrastructure due to high salt concentrations, alkaline chemistry, and fluctuating desert moisture.

• Soil Resistivity Testing (Wenner Four-Pin Method): Prior to pipeline construction, geotechnical crews measure soil electrical resistivity. Soils with resistivity < 1,500 ohm-cm are classified as severely corrosive, requiring comprehensive cathodic protection.
• Sacrificial Galvanic Anodes: High-purity magnesium (Mg) or zinc (Zn) ingots buried in conductive bentonite clay backfill adjacent to the pipeline are electrically bonded to the ductile iron or steel main via insulated copper wire. Because magnesium is more electrochemically active than iron in the galvanic series, the magnesium anode sacrificially corrodes, discharging protective electrons into the pipe (cathode) to suppress rust formation.
• Impressed Current Cathodic Protection (ICCP): For large regional transmission pipelines, galvanic anodes lack sufficient electrical potential. ICCP systems utilize an external AC-to-DC transformer-rectifier to drive protective direct current from an inert deep-well mixed-metal oxide (MMO) anode groundbed through the soil into the pipeline.
• Dielectric Isolation: Installing insulating flange kits, nylon bolt sleeves, and dielectric unions at transitions between dissimilar metals (e.g., copper service lines connected to ductile iron mains) to interrupt electrical conductivity and eliminate galvanic corrosion cells.

---

AWWA C651 Water Main Disinfection & Bacteriological Clearance

Whenever a new water main is installed or an existing main is depressurized for emergency repair, it is exposed to bacterial pathogens, dirt, and trench contamination. Returning the main to service without rigorous disinfection and microbiological testing threatens public health.

AWWA C651 Disinfection Methods

Under AWWA C651, three distinct chlorination methodologies are approved for main disinfection:

+-----------------------------------------------------------------------------------------+
|                        AWWA C651 Main Disinfection Methods                              |
+-----------------------------------------------------------------------------------------+
| Continuous Feed Method | Minimum initial dose: 25 mg/L free chlorine                    |
|                        | Hold time: 24 hours                                            |
|                        | Minimum 24-hr residual: ≥ 10 mg/L throughout entire main       |
| Slug Method            | Minimum initial dose: 100 mg/L free chlorine                   |
|                        | Contact time: 3 hours with every interior surface              |
| Tablet / Granular      | Calcium hypochlorite tablets/granules glued to pipe invert     |
| Method                 | Hold time: 24 hours; initial 25 mg/L, residual ≥ 10 mg/L       |
|                        | STRICTLY prohibited if trench mud or water enters pipe         |
+-----------------------------------------------------------------------------------------+

1. Continuous Feed Method: The standard, most reliable method. Potable water is introduced into the new main at a controlled rate while concentrated sodium hypochlorite (or chlorine gas solution) is continuously injected through an upstream corporation stop. The chemical dose is adjusted to achieve a minimum of 25 mg/L free chlorine throughout the entire length of the pipe. The chlorinated water is held in the isolated line for 24 hours. At the end of the 24-hour retention period, the free chlorine residual at all sampling points must be at least 10 mg/L. If the residual drops below 10 mg/L, the main must be re-chlorinated and held for an additional 24 hours.
2. Slug Method: Utilized on large transmission pipelines to conserve water and chemical volume. A concentrated 'slug' of water containing a minimum free chlorine concentration of 100 mg/L is slowly moved through the pipeline. Pumping rates are throttled to guarantee that every square inch of the interior pipe wall contacts the 100 mg/L slug for a minimum contact time of 3 hours.
3. Tablet / Granular Method: During construction of dry, clean pipelines up to 24 inches in diameter, operators glue 5-gram calcium hypochlorite tablets (65% available chlorine) to the top inside surface of each pipe section using food-grade adhesive. The main is filled slowly with potable water at a velocity ≤ 1.0 ft/sec to dissolve the tablets evenly, achieving a 25 mg/L dose held for 24 hours (with ≥ 10 mg/L residual). This method is strictly prohibited if trench water, mud, or rain enters the pipe during construction, as tablets cannot penetrate mud cakes.

Neutralization & Chemical Dechlorination

Highly chlorinated water (10 to 100 mg/L) is lethal to aquatic life. Under Arizona environmental regulations and Clean Water Act discharge permits, superchlorinated main flushing water must never be discharged directly into washes, storm drains, or retention basins without chemical dechlorination. Operators inject chemical neutralizing agents—such as sodium bisulfite (NaHSO3), sodium thiosulfate (Na2S2O3), or ascorbic acid (Vitamin C)—at the discharge diffuser to reduce free and total chlorine residuals to non-detectable levels (< 0.05 mg/L) before water reaches ground surfaces.

Mandatory Bacteriological Clearance Sampling

Following successful disinfection, the superchlorinated water is flushed out using potable water until the disinfectant residual in the new main matches the normal background residual of the distribution system. The pipeline must remain isolated while microbiological clearance testing is performed.

• ADEQ Sampling Protocol: Pursuant to A.A.C. R18-4-213 and AWWA C651, the utility must collect two consecutive sets of acceptable bacteriological samples collected at least 24 hours apart from dedicated copper sampling whips installed along the main (spaced no more than 1,200 feet apart, with samples taken at all dead-ends and branches). An alternate special protocol permits two sample sets taken 15 minutes apart after a 16-hour stagnation period if specifically authorized in the utility's approved engineering specifications.
• Pass Criteria: Both sets of samples must be analyzed by an ADEQ-certified drinking water laboratory using an approved method (such as Colilert® or membrane filtration) and test negative (absent) for Total Coliform bacteria and E. coli. If any sample tests positive for coliforms, the pipeline fails clearance: the main must undergo immediate re-flushing, re-chlorination, and a complete re-sampling cycle before it can be legally connected to the active public water distribution network.