When the power goes out, the water runs
A residential estate, three generators, and a better question
The estate where I live in Cyprus had three diesel generators. They cost about two thousand pounds a year each to maintain, they were ageing, and replacing them would run to nine thousand apiece. They existed because Mediterranean power grids are not perfectly reliable, and when the electricity goes, things stop working.
But which things? The generators ran pool pumps, path lighting, and the domestic water system. I started there — not with the generators, but with the question of what actually matters when the power cuts out.
The pools can sit for a day without circulation. No visible problem. You'd need several days before water quality starts to deteriorate. The path lights are a convenience. But the water supply is different. An extended outage means no water pressure, and that leads quite quickly to unsanitary conditions. Seventy Nine homes with no running water is not a nuisance. It's a health concern.
So the question wasn't "how do we replace the generators?" It was "how do we keep water pressure up when the grid goes down?" That's a much smaller problem, and smaller problems have cheaper solutions.
The existing pumping system had its own flaw. When the grid voltage floated towards the upper end of its nominal range — which happens regularly here — the electronic controller would lock out entirely. A system designed to provide water was failing not because of power cuts, but because of power that was technically present but slightly out of spec. I removed it.
Then I looked at the topology of the estate. Water tanks sit at various heights across the site. I chose the highest point for the new pump installation — not arbitrarily, but because gravity does free work. Every metre of elevation you start with is energy you don't have to put in from a battery. The less the pump has to push, the longer the battery lasts, and the smaller the battery needs to be in the first place.
This is where the engineering becomes a chain of reasoning rather than a parts list. How much water does the estate use per hour in realistic conditions? What flow rate does that demand? What pressure is the minimum for usable domestic supply? Now work backwards: impeller efficiency at that operating point, motor losses, inverter conversion losses, and you arrive at a number — the actual energy budget, in watt-hours, for a meaningful period of backup.
I proposed a budget. It was approved. The system was built and commissioned thirty percent under that budget.
The three generators have been removed. The estate now redirects six thousand pounds a year towards structural improvements — actual, visible work on the place where people live. When the power goes out, the water runs. Most residents have no idea how, and they don't need to. But the thinking behind it is the same thinking I bring to everything, and that's what I'm trying to show here.
Several people — including some who were firmly supportive of me in other matters — later confided that they hadn't believed it would work. I understand that. A battery-backed pump doesn't sound like it should replace three diesel generators. But it does, because the generators were answering a question nobody had examined carefully enough. Once you ask the right question, the answer is often smaller, simpler, and cheaper than anyone expected.