Rainwater harvesting

How does Rainwater harvesting work, step by step?

1. 1

   Step 1: Establish the demand and the yield

   The designer works out how much non-potable water the building will actually use - which fittings and outlets will be served, and at what pattern of demand - and how much rain the collection roof can be expected to deliver over the year. The two are then reconciled against a storage volume that carries the system through dry periods without becoming so large that the water sits stagnant. The result also tells the client what proportion of demand the system will meet, which is usually rather less than people expect. All of these figures are specific to the building, the roof area and the location, and they come out of the designer's calculation rather than off a chart.

2. 2

   Step 2: Design the collection and filtration arrangement

   Only suitable roof areas are collected from. Downpipes serving those areas are routed to the filter, and downpipes serving unsuitable surfaces are kept out of the system entirely. Leaf guards and a first filtration stage remove debris before the water reaches the tank; on many systems an arrangement is included to divert the first, dirtiest flush of a rainfall event to drain rather than into storage. Filters are positioned where they can actually be reached, because a filter that needs a confined space entry or a road closure to clean will not be cleaned. The overflow from the tank is designed as part of the site drainage and discharges to the point the drainage designer has specified.

3. 3

   Step 3: Install the storage tank

   Below-ground tanks are installed as an excavation and bedding operation, set on a prepared base, surrounded and backfilled in the arrangement the tank manufacturer specifies, and provided with an access chamber and cover suitable for whatever will drive over them. Buoyancy has to be considered where the water table is high, because an empty tank in wet ground will float. Above-ground tanks are simpler but need a supporting structure designed for the full weight of water and a location where a leak or an overflow is not a disaster. Either way the tank is dark, vented, insect-proof and accessible for inspection and cleaning, and the inlet is arranged so that incoming water does not stir up the settled material at the bottom.

4. 4

   Step 4: Install the pump set and controls

   A pump takes water from the tank and delivers it to the non-potable outlets, either directly on demand or via a small header arrangement. The control panel manages the pump, monitors the tank level, and switches the supply to mains top-up when the stored water runs out. Controls also cover the alarms - low level, pump fault, overflow - and how those alarms reach somebody who will act on them. Pumps and panels are located where they can be serviced and where noise is not transmitted into occupied space, and the electrical supply and isolation are coordinated with the electrical contractor.

5. 5

   Step 5: Provide the mains backup through an air gap

   When the tank runs dry the non-potable system still has to work, so a mains top-up is provided. The critical point is how that connection is made. The mains supply discharges into a break tank or a top-up chamber across a physical air gap, so that there is no continuous path from the harvested water back into the drinking water supply under any circumstances - not when the pump runs, not when the mains pressure drops, and not when something is left open. A direct connection between the two systems, even through a valve, is not acceptable. This arrangement is designed by the services engineer and inspected as a specific item.

6. 6

   Step 6: Install and mark the non-potable distribution separately

   The harvested water runs in its own pipework, entirely separate from the drinking water system, and that pipework is marked so that nobody who works on the building in future can mistake one for the other. Marking means the specified colour or identification along the whole run, at every accessible point, in every riser and in every plant space - not a label at the pump and nothing beyond it. Outlets served by harvested water carry warning signage indicating that the water is not for drinking, and any outlet that could be used to draw water for washing or drinking is either avoided or clearly identified. Cross connection between the two systems is the one failure mode with a genuine public health consequence, and every control on the installation exists to prevent it.

7. 7

   Step 7: Commission, label and hand over

   Commissioning proves the whole sequence: that the filters pass water and can be removed and cleaned, that the tank fills and overflows to the right place, that the pump delivers to every outlet, that the level controls switch to mains top-up and back again, and that the alarms operate and reach somebody. A cross connection check is carried out as a distinct and recorded item. The handover pack includes the as-installed drawings showing every metre of non-potable pipework, the operating and maintenance instructions, the filter and tank cleaning intervals, the pump service requirements and the labelling schedule. The client is told plainly what the maintenance obligation is.

8. 8

   Step 8: Maintain it, or lose the benefit

   A harvested water system is a live installation, not a buried tank. Filters need cleaning at the intervals the supplier states and more often in autumn. Tanks need periodic inspection and desludging. Pumps and controls need servicing. Water quality needs attention appropriate to the uses served, particularly where water may be sprayed or where the system serves anything beyond flushing and irrigation. The identification and signage need checking after any alteration, because the greatest long-term risk to a building with two water systems is a future contractor who does not know the second one is there.