A recent question on what to do with harvested rainwater got me thinking about pumped energy storage systems. Since both systems require storage of water for later use, is there a cost-effective way in which the two could be combined? Specifically in small-scale applications (farms, rural communities, campuses) where rooftops are already available for rainwater harvesting.

As background, pumped energy storage involves pumping water to an elevated storage reservoir when energy is cheap and plentiful (excess wind, low demand, unused solar, etc) then letting it run through a turbine to generate power at times of low supply.

Some ideas about how this might work:

  • Rain water caught on the roofs of tall and/or elevated buildings could flow through a turbine (generating power) to shorter/lower reservoirs where it could be used for irrigation or other purposes
  • Overhead water tanks could be fitted with solar panels to drive pumps bringing up rain water from lower rooftops, where it would later be flowed through a turbine to provide power
  • A reservoir or tank could be built near a community or farm requiring water. Wind or solar power could pump water into the tank. When water is needed it could be flowed through a turbine

Have any such systems been implemented? Could it be cost-effective, or would it simply over-complicate things?

  • So basically you are looking for ways to divert more rainwater to the upper and lower reservoirs of a pumped storage hydropower facility? – THelper Aug 5 '16 at 12:26
  • 1
    Actually, that sounds like a large scale system - I was thinking smaller scale, such as on a farm or in a rural community, where you've already got rooftops doing most of the work catching rain. – LShaver Aug 5 '16 at 12:31
up vote 10 down vote accepted

Water loss is not the issue, because you need a lot of water to make it work at all. Even if your dams lose 30% of their water every year, that's about 0.1% per day. System losses in the best large-scale systems are about 20%. In other words, 70-80% efficiency of large scale systems (wikipedia). The big issue, though, is likely to be making it work at all. Conveniently that wikipedia article has a small-scale example:

1000 kilograms of water (1 cubic meter) at the top of a 100 meter tower has a potential energy of about 0.272 kW·h

To make it useful you'd want to store as much energy as a modern small battery system, say 10kWh. Assuming your "house" is 100m tall (25-30 storeys), you get 100m of head and using the 0.27kWh/m³, 10 kWh is about 40m³ (40,000 litres) of storage. That's a small farm dam or big farm tank, it's definitely not a rainwater tank as most urban households think of it. Here's a pair of 38,000 litre tanks sitting in front of a house:

9000 gallon water tank

The other problem is getting an efficient pump that can push water from your bottom 40m³ dam to the top 40m³ dam. Amazon will sell you a 2kW pump capable of 3m/hr to 100m for $1500. Efficiency is not very high - that's 2kWh input for 0.8kWh stored (3x0.272kWh from above). To get from 40% efficient to 90% efficient is going to cost a lot more, and if you're not 90% efficient on the storing side, you're going to struggle to get 80% round-trip efficiency.

Note that you can trade head for volume linearly, but the efficiency of your pipes, pumps and turbines is crucial. The reason moderate head systems are preferred is that it's easier to get them up to the efficiency required. If you have 1000 metres, or 10 metres, of head, it becomes very difficult to make the system efficient. For example, most bore pumps with over 100m of head are simply multistage versions of lower-head pumps, so the efficiency drops even further. But flow turbines for low-head systems usually peak at 70%-80% efficient rather than the 90% you can get out of a pelton wheel:

water turbine efficiency graph

(via wikidot - relative efficiency of various turbines as flow rate varies as a proportion of rated)

So this is not something you can do at all as an urban household, and it will be difficult for most rural landowners.

When I had a microhydro system on a property I owned, I had about 20 litres/second available but the head was over 100m because the block of land was steep (hence, cheap :)). The main limitation was my willingness to drag heavy pipe up a steep hill. Vehicle access to the top of the block was strictly "all wheel drive with winch" and took a whole day - that sort of terrain is brilliant for hydro, but steep land with high rainfall is a disaster for ground vehicles.

Because of the high rainfall, the "reservoir" at the top of that pipe was just an inlet filter anchored to bedrock using a cubic metre of concrete, there wasn't a dam per se, let alone storage. Getting the top anchor in place was over a months work for one man using machinery. Getting a tank like the one above to the top would have needed a helicopter. A big, expensive helicopter that would have had to come from a long way away. A dam would be easier, but still ridiculously difficult.

  • The good news, though, is that the 100m tall tower to hold the upper water tank will also work well to hold a 2-5MW wind turbine to power the pump! :) – Ⴖuі Aug 5 '16 at 23:43
  • Note that the 90% peak efficiency is of the turbine, not of the turbine+generator, so you would only get ~80% efficiency gravitation to electrical if the pipe, generator and invertor somehow managed a combined efficiency of 90%. Which is high. – Móż Aug 9 '16 at 0:06

Your Answer


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.