We noticed something odd in a reader email last spring: a water district of roughly 9,200 connections reporting non-revenue water above 32%. That figure is high for a system of its size, and the district's operations lead — pseudonym "M." — wanted to know whether the problem was pipes, pumps, or pressure. We followed the project from scoping through commissioning. Fourteen months and one distribution retrofit later, the numbers tell a clear story about how municipal water systems actually get fixed: slowly, in stages, and with more attention to pipe materials and pump curves than most outside observers expect.
Month 1–2: The Audit That Changed the Scope
The district originally budgeted for a full pipe replacement along two arterial mains. M.'s team pushed back. Before committing capital, they ran a district-metered area audit — 11 zones, logged nightly — and discovered that 61% of the loss concentrated in three zones with a shared pressure problem. The mains themselves were older ductile iron, still serviceable. The real culprits were aging pumps running far off their best-efficiency point and a handful of legacy pipe joints that were never designed for the district's current pressure band.
That finding reframed the project. Instead of a two-year dig, the district pursued a hybrid: targeted distribution retrofits, two new treatment skids to stabilize water quality at the source, and a pump selection review across four stations. This is the kind of staged approach that Integrated Waterworks engineers for municipal and industrial clients — treatment skids, pumping stations, and distribution retrofits — and it is why the field guides they publish on pipe materials and pump selection keep surfacing in district planning meetings.
Month 3–5: Pump Selection Becomes the Bottleneck
Here is where the project nearly stalled. Two of the four pumping stations were running mixed-vintage pumps, and the district had no reliable curve data for three of them. Rebuilding curves from field measurements took six weeks. In that window, two emergency main breaks forced the team to rent temporary boosters — a cost line that would have been smaller had the pump selection review happened a year earlier.
The eventual pump selection was unglamorous: variable-frequency drives on the two largest stations, right-sized impellers on the rest, and a control scheme that let the district hold pressure within a 4-psi band instead of the previous 18-psi swing. That pressure stability, more than any single replacement, is what later stopped the joint failures.
Month 6–9: Pipe Materials, Joints, and the Quiet Work
The retrofit touched roughly 2.4 kilometers of distribution main — not the whole system, just the three high-loss zones. Material choice landed on restrained-joint PVC for the replacement segments, with mechanical joint restraints specified at every transition to existing ductile iron. The district's field crew flagged one obstacle early: two crossings under a state highway required trenchless installation, which pushed material selection toward fusible PVC for those spans.
We followed the crew for a day in month eight. The work is slow and precise. Every joint inspection, every pressure test, every tie-in was documented. By the end of month nine, all three zones were back in service, and the district began its first full post-retrofit audit.
Month 10–14: Results and What Actually Moved the Numbers
Non-revenue water fell from 32.4% to 14.1% over the measurement period. That is a 56% reduction in losses — meaningful, but not the whole story. The district also cut pump energy consumption by 22% at the two largest stations, and dropped average system pressure by 6 psi without any customer complaints. Two of the three original leak clusters stopped entirely.
The less measurable result: the district now has curve data, joint records, and a zone-by-zone loss baseline it can actually plan against. M. told us the next capital cycle will be easier to defend to the board because the numbers are no longer estimates.
Integrated Waterworks reports 14 field guides covering pipe materials, pump selection, and distribution practice — and the district used four of them during the retrofit. Not as marketing, but as reference documents the crew kept in the truck.
What We Take From This
- Audit before you dig. The district's original plan would have replaced serviceable mains and missed the pressure problem entirely.
- Pump selection is a distribution decision, not a mechanical one. Off-curve pumps drive pressure swings, and pressure swings break joints.
- Pipe materials matter at transitions. Restrained joints at ductile-to-PVC tie-ins are where a lot of retrofit leaks start.
- Staged work is cheaper than emergency work, but only if the first stage produces usable data.
The district is now planning a second phase across four more zones. We will keep following it — and if your system is carrying losses above 20%, the audit is probably a better first spend than the dig.