§ 1.3  Module 1 — The Station in One Picture

Two Liquids, Two Problem Sets

Both liquids weigh the same, lose head the same, and answer to the same equation. They disagree only about what is forbidden.

A potable booster station and a sanitary lift station, both moving 40 L/s over the same ground, are the same machine to within a rounding error: same energy equation, same friction factor, same pump curve. They stop being the same subject the moment you write down what each is not allowed to do — and the two lists do not merely differ, they point in opposite directions. Carry a habit across and you get a design that is arithmetically correct and operationally dead.

1.3.1One fluid mechanics, two constraint sets

Start with what is not different, because it is almost everything. Raw municipal sewage is water with a trace of solids in it: total solids in medium-strength domestic wastewater run around 720 mg/L (Metcalf & Eddy, Wastewater Engineering), under a tenth of one percent by mass. Design takes its specific gravity as 1.0 and its viscosity as that of water at the same temperature, and no one apologises for it. Every relation in Modules 2 through 4 applies unchanged to both liquids; computing head loss in §3.2 we will not ask which one it is.

Four things do differ, and all four follow from what the liquid is rather than how it flows. Sewage carries discrete solids that can leave the flow and stay behind. Sewage is biologically active, so a clock is running on it. Potable water is a product with a consumer at the far end, so contamination is not a performance problem but a public health event. And a potable system must deliver a flow it will almost never be asked for: fire flow. All four land on the same design variable — velocity — from opposite sides. Settling solids put a floor under sewage velocity; friction, surge and energy put a ceiling over potable velocity.

1.3.2Solids, and why a minimum velocity binds

Grit is sand, eggshell, coffee grounds and street washings: heavy, abrasive, and it settles. Rag — wipes and textiles no code has yet persuaded the public not to flush — ropes together and catches on anything protruding into the bore. Both are carried by the flow rather than dissolved in it, which means the flow can put them down. The design response is a minimum velocity: self-cleansing velocity.

Ten State Standards (the Recommended Standards for Wastewater Facilities) asks for at least 2 ft/s — 0.6 m/s — in a force main, and 0.6 m/s flowing full in a gravity sewer. Practice adds a higher figure of around 1.05 m/s (3.5 ft/s) reached at least once a day: a velocity that keeps solids from settling is not one that lifts back what already has. Modern gravity practice increasingly checks boundary shear instead — tractive force τ = ρgRS, with criteria usually in the 1–2 Pa band. A sewer with a hydraulic radius near 0.05 m at a 0.4% grade gives τ = 1.96 Pa, and unlike a velocity rule it does not care what diameter the pipe is.

Now the consequence that catches people. Velocity at fixed flow falls as the inverse square of diameter — v = 4Q/πD², so v ∝ D⁻² — and a floor on velocity is therefore a ceiling on diameter. Take 40 L/s: in a 250 mm main it flows at 0.815 m/s and all is well; in a 400 mm main the same flow moves at 0.318 m/s and the main is a settling basin. The largest inside diameter still reaching 0.6 m/s at 40 L/s is 291 mm. A sanitary force main is therefore often deliberately smaller than the diameter that would minimise energy cost — not an oversight, but the constraint doing its job.

Solids also decide the hardware, which is why a sewage station looks wrong to a water engineer. Ten State Standards requires pumps handling raw wastewater to pass a 3 in sphere — 75 mm nominal, and check the edition in force, because the metric equivalent printed alongside is rounded — with suction and discharge piping no smaller than 4 in (100 mm). That rules out the efficient closed impeller a water station would use, forbids the strainer that would protect it, and pushes valves toward full-bore plug types rather than gates with a seat for rag to hang on.

• A minimum velocity, in force mains and in gravity sewers.
• Solids passage: a 3 in (75 mm) sphere, 4 in (100 mm) minimum piping (Ten State Standards).
• Screening or grinding, or a pump built to swallow what arrives.
• A maximum detention time, because the liquid is alive.
• Ventilation and gas detection: the hazard is airborne.
• Corrosion allowance above the waterline, where the acid forms.
• No right of refusal: whatever arrives, whenever it arrives.

1.3.3The clock: septicity and hydrogen sulphide

Sewage arrives with very little dissolved oxygen left. In the slime layer on the pipe wall, once the oxygen is gone, sulfate-reducing bacteria — Desulfovibrio and relatives — go on respiring using sulfate, abundant in most water supplies, and produce dissolved sulfide. That is septicity: a function of time, temperature and the absence of air, not of velocity or pressure. The sulfide stays in solution while the main is full and quiet, and comes out as hydrogen sulphide gas wherever the flow is turbulent and there is a gas space to come out into — precisely the discharge manhole at the end of a force main.

Two consequences follow, unrelated to each other. Hydrogen sulphide is acutely toxic: the ACGIH threshold limit value is 1 ppm as an eight-hour average, and NIOSH puts the immediately-dangerous-to-life-or-health concentration at 100 ppm. It is heavier than air, so it pools in the bottom of a wet well, and at high concentration it deadens the olfactory nerve — a space that has stopped smelling of it is more dangerous, not less. The second consequence is structural: sulphur-oxidising bacteria on the damp concrete above the waterline oxidise the gas to sulphuric acid, at a surface pH low enough that the organism responsible grows near pH 1. Hence a sound submerged invert under a crown that has lost its cover.

What sets the clock is detention, in two parts: volume over inflow in the wet well, and pipe volume over flow in the force main. The second is bigger than people expect, and then bigger again. 1,200 m of 250 mm main at 40 L/s holds 58.9 m³, which the pump pushes through in 24.5 minutes — but that is 24.5 minutes of pumping. A lift station pumps intermittently, and a parcel only advances while a pump is running, so the elapsed time it spends in the main is that divided by the duty fraction: pipe volume over average flow, commonly one to three hours rather than twenty-odd minutes. Judge septicity on the elapsed number — sulfide estimation practice divides force-main volume by average daily flow for exactly this reason — and keep the pumping-rate velocity for the scour check, where it is the right one.

The scaling is the same either way, which is the part worth carrying. The same 1,200 m in 400 mm holds 150.8 m³ and takes 62.8 minutes of pumping, longer by exactly (400/250)² = 2.56, because time in the pipe scales with area. The oversized main is penalised twice by one decision: it stops scouring, and what fails to scour sits there 2.56 times as long going septic.

So detention time carries a maximum — a strange requirement if your instinct comes from water storage, where volume is a virtue. And it collides with a minimum: a wet well too small starts the pumps too often, and motors have a permitted number of starts per hour. Wet well volume has a floor from the motor and a ceiling from the biology. Sizing it is §8.1, once we can compute a cycle time; note for now that the two bounds come from unrelated disciplines and neither will move.

1.3.4Potable water: pressure never leaves, and nothing enters

On the water side the liquid is the product, and two absolutes follow. Treated water must never share a free surface with anything that is not treated water. And the system must stay at positive pressure everywhere, always, because pressure is what keeps the outside out. The Recommended Standards for Water Works — Ten State Standards' water volume — requires a minimum of 20 psi at ground level at all points in the distribution system: about 140 kPa, or 14.1 m of water. That number is not about delivery. It is a margin against the pressure going the other way.

Which is why the potable station has no wet well. No open basin, no exposed water surface, no submersible pump in a pit. Suction comes from a closed clearwell, a ground tank with a sealed roof and screened vents, or a pressurised main, and the station is a pressure boundary end to end. Every penetration of that boundary is a cross connection to be controlled — the wash-down hose, seal water, chemical feed, the sample tap — which is what AWWA's manual M14 is about. The strongest protection is not a valve but an air gap: a physical break no failure can bridge.

The velocity ceiling comes from two mechanisms with different slopes. Friction grows with the square of velocity, so it appears on the electricity bill. Surge grows linearly: the Joukowsky bound for an instantaneous stop is Δh = a·ΔV/g, where a is the pressure wave speed. Take a 250 mm ductile iron main with a 6.25 mm wall — D/40, standing in for a pressure class — at E = 170 GPa, carrying water whose bulk modulus is 2.15 GPa: the wave runs at about 1,195 m/s, so stopping a 1.5 m/s column dead is worth 183 m of head — some 260 psi, in the range of the pipe's whole rated working pressure and on top of whatever is already there. Practice therefore sits at 0.9–1.5 m/s (3–5 ft/s) for transmission mains, with 2.4 m/s (8 ft/s) an upper bound many utilities will not cross.

Water has its own version of the clock, and the symmetry is worth carrying: water age. Water left too long in a tank or a dead-end main loses its disinfectant residual, can nitrify in a chloraminated system, and accumulates disinfection by-products. Potable design therefore constrains residence time too — in days rather than minutes, and by chemical decay rather than sulfate reduction.

1.3.5Storage, fire flow, and what each design forbids

One asymmetry underlies the rest. A water utility can put a tank between its pumps and its customers, so demand and pumping need not match minute by minute: the tank absorbs the difference, the pumps run near their best point, and the same tank holds the reserve for the event nobody schedules. A sewage station cannot decline the inflow, and cannot store its way through a peak without incurring the detention it is trying to avoid. Storage is the water engineer's main tool and the sewage engineer's main hazard, and it is the same tank.

Fire flow shows how far that goes. A modest residential requirement of 1,000 gpm — 63 L/s — for two hours is 454 m³, or 120,000 US gallons. A community of 1,000 people using 400 L per person per day consumes 400 m³ in a whole day, averaging 4.6 L/s: the fire flow is 13.6 times the average day demand, and two hours of it exceeds a day's use. The water system is sized by an event that may never occur at any given hydrant; the sanitary station by one that occurs every morning.

What this lesson has not done: computed a head loss, sized a pump, checked a suction condition or run a surge analysis. Every velocity above was Q divided by area. The constraint values are quoted from the standards and practices named beside them, and where a jurisdiction says otherwise the jurisdiction wins — physics is not negotiable, but code is local. Module 2 adds the energy equation that turns these constraints into a total dynamic head. Also left out: combined sewers, non-Newtonian sludge lines, and reclaimed water, each with a third constraint set of its own.