Rainwater Harvesting – Summary after 4 Months of Operation

The rainwater harvesting system went live at the start of December 2025, replacing the use of mains water for the toilets, the outside taps and the washing machine – and literally halving the mains water consumption in the process. The rainwater does need to be pumped out of the underground tank but a dedicated electricity meter shows the pump and its controller consume just 1 kWh of electricity per 1,000 litres of water delivered – which costs £0.05 compared with £2.61 for importing 1,000 litres of mains water.

The water saving is no surprise, since the ‘always needs to be mains water’ and ‘potential future rainwater’ consumption have been metered separately since the House was built. The low electricity consumption was largely expected by making careful choices when selecting the water pump and its controller (and by including a pressure vessel which gets ‘charged up’ whenever the pump runs, avoiding the need for the pump to start more than a few times a day) – but it is good to have this confirmed.

Back when I was specifying the systems for the House, the consensus in the low energy building community was heavily influenced by a 2010 report by the UK Environment Agency which concluded that domestic rainwater harvesting systems were 40% more carbon intensive than using mains water, with a split of 52% ‘operational’ carbon and 48% embedded. That ‘operational’ carbon was due to the electricity consumption of the rainwater pump – especially for ‘direct’ systems which need the pump for every rainwater draw-off (rather than periodically filling a cistern which then supplies water by gravity). Despite this report, I went ahead and installed parallel pipework systems in the house, for potentially splitting out rainwater usage from mains water usage. To a large extent, this was based on the expectation that the carbon intensity of the electricity supply would reduce over time.

More recent research has shown that things have changed and that rainwater harvesting systems are now typically beneficial from both a carbon- and a cost-saving standpoint. However, it still makes sense to reduce the embedded carbon – for example by avoiding tanks which need to be bedded in concrete – and to choose energy-efficient control systems.

My experience is that the water looks as clear as mains water with no hint of any smell. There is no sign of any algal or bacterial growth where the water gets stored at room temperature in toilet cisterns. The mains water here isn’t especially hard but the rainwater is completely soft so there’s no limescale build-up and washing detergents are more effective. I had wondered if it might be necessary to retrofit an inline filter for the washing machine in particular but based on current experience that is absolutely not necessary.

The filter on the inlet to the underground tank has been getting quite dirty with fine dust coming off the new clay roof tiles (a distinctive ‘Staffordshire Blue’ colour) and needed cleaning ever month or so, but that’s now reducing.

While the saving on the mains water bill is the main appeal, there are wider benefits too:

  • Provided the tank isn’t already full, hundreds of litres of rainfall will get stored in the tank, reducing the run-off from the site and helping mitigate the risk of flooding in the wider area
  • Less demand is placed on the water supply and delivery network
  • The tank holds an on-site reserve of water for firefighting purposes, which is useful since the nearest fire hydrant is literally miles away

There’s some more detail in the Technical Article here.

Validation of Passivhaus-Level Heating Performance

The main criterion for Passivhaus-level performance is achieving a space heating requirement better than 15 kWh per m2 of floor area per year (see e.g. this page from the Passivhaus Institut) when heating to an internal temperature of 20C.

One benefit of having a heating system listed on the Heatpump Monitor website is the Heatpump + Fabric dashboard, which orders systems by kWh/m2 and thus provides a direct view of whether buildings are performing at this level.

Screenshot of the “Heatpump + Fabric” dashboard on Heatpumpmonitor.org

A few points to note:

  • By default, this list is sorted by Electricity Input (Elec kWh/m2) rather than Heat Output (Heat kWh/m2) but that’s easy to change.
  • It’s not obvious but the Elec and Heat figures relate to the Combined performance of the heat pump, when delivering Space Heating and also when heating Domestic Hot Water.
  • There is no indication of the internal temperature achieved, so some buildings might be hotter (or cooler) than the nominal 20C used for the Passivhaus assessment.

The second point tends to increase the consumption figure – potentially quite significantly, because high-performance buildings can use just as much DHW as low-performance buildings, and the higher temperatures required for stored hot water (compared with space heating) results in lower efficiency from a heat pump.

Some systems listed on heatpumpmonitor.org provide additional data on whether they are delivering Space Heating or Hot Water at a particular point in time. If this is accurate, it allows the Hot Water heating periods to be filtered out from the results, providing a more representative view of the heat required for Space Heating. However, it’s clear that many systems don’t accurately report Space Heating mode – so the overall list is unrepresentative. It can be useful to filter out Hot Water production for a single system where the data is known to be accurate.

In my case, looking only at Space Heating drops the kWh / m2 from 13.8 to 13.1, demonstrating that this is even more compliant with the 15.0 kWh / m2 Passivhaus limit (even though the target indoor temperature is 21C, rather than 20C).