You are here:Home\Energy\Pumped storage

Pumped storage

When you look in Wikipedia under pumped storage, this is what you get:

 Diagram of the TVA pumped storage facility at Raccoon Mountain Pumped-Storage Plant.Pumped-storage hydroelectricity (PSH) is a type of hydroelectric power generation used by some power plants for load balancing. The method stores energy in the form of water,
 pumped from a lower elevation reservoir to a higher elevation. Low-cost off-peak electric power is used to run the pumps. During periods of high electrical demand, the stored water is released through turbines to produce electric power. Although the losses of the pumping process makes the plant a net consumer of energy overall, the system increases revenue by selling more electricity during periods of peak demand, when electricity prices are highest.

Pumped storage is the largest-capacity form of grid energy storage available, and, as of March 2012, the Electric Power Research Institute (EPRI) reports that PSH accounts for more than 99% of bulk storage capacity worldwide, representing around 127,000 MW.[1] PSH reported energy efficiency varies in practice between 70% and 80%[1][2][3][4], with some claiming up to 87%[5].

Today’s  most urgent problem is the CO2 emission due to fossil energy carriers like coal, oil, lignite, gas. If the CO2 emission cannot be stopped, we are heading for a global catastrophe the scale of which is even difficult to imagine. The only possible way to avoid such a catastrophe is to switch to renewable energy sources, such as wind or solar power. These energies, however, have one big disadvantage: they don’t follow in their generation pattern the consumption pattern of the energy consumer. Solar energy flows only when the sun shines, and wind energy can be produced only when the wind blows. Without changing the consumption pattern (which today nobody is really attempting), there are limits for an electricity grid as to what extent it can absorb renewable energies. Estimates say that with about 20% of renewable energy in a grid we are reaching the maximum possible. Above that, the grid becomes unstable. However,  with 20% of renewable energy and 80% of conventional energy, we won’t be able to solve the CO2 problem.

100 years ago, the whole of mankind comprised about 1.5 billion people. Today we have crossed 7 billion.  And every day we are getting more. In addition, our living standard is constantly rising, and so is our energy consumption. The only way to increase the ratio of renewable energies within a grid is to store the energy. The question is how?

As we can see from the report of Electric Power Research Institute from March 2012, 99% of grid electricity worldwide is stored in the form of hydro-energy in pumped storages. What we can also learn from this report is that the efficiency ratio is around 70%. That means, when we pump up water to a higher elevation, and then let it flow back through a turbine, we get 70% of the energy back that we invested in pumping the water up.

This scenario of a grid becoming unstable and the renewable energy not being able to be further developed is not a problem of the far future.  Very concretely we have reached in Tamil Nadu a situation where wind energy cannot be further developed, because it makes the grid unstable, because in peak season it cannot be transported from the wind-sites to the consumers, and in off-season it is missing in the energy balance of the grid.  So if we want to develop wind energy further in Tamil Nadu we either have to complement it with some other form of energy which compensates for the irregularities of the wind energy, or we have to create storage facilities for the peak-hour production energy. According to today’s technology this can be achieved only with pumped storages.

Tamil Nadu has one big pumped storage set-up near Coimbatore. But this is the only one which exists in the whole of Tamil Nadu, and of course one single pumped storage  is not enough  to have a decisive effect on the energy-situation. 

We urgently need more pumped storages in Tamil Nadu if we want to increase the ratio of renewable energies in our TNEB electricity grid.  Even more urgent, we will start requiring pumped storages when solar power becomes a bigger element in the energy mix in Tamil Nadu. The peak consumption time in Tamil Nadu is early evening, around 19.00h.  At that time solar panels give no current any more. Tamil Nadu will very soon reach also a saturation of solar energy if it doesn’t start installing pumped storages.

The only country in the world which is taking the threat of climate change seriously, and which has decided to reduce its CO2 emissions drastically, is Germany.  In Germany it has been decided by parliament, and been implemented by law, that till 2023 all nuclear power stations will be switched off and that Germany  will reduce its CO2 emissions by 2020 by 40% compared to 1990, and by 80% by 2050.

To achieve this, in Germany the capacities of pumped storages are planned to be increased by an additional 80% over the next 7 years.

For interested readers we recommend to go through the study Prospects for Pumped Hydro Storage in Germany.


If Auroville wants to be a front-runner in renewable energy technology  in Tamil Nadu, then a pumped storage with the latest technology would be an ideal complement to its existing wind park and its planned involvement in solar energy projects.

The type of pumped storage of the Racoon Mountain, given as an example by Wikipedia, is not the type we favor for Auroville. For Auroville we suggest a similar set-up to the Ludington Pumped Storage.

Ludington Pumped Storage

Picture of the Ludington Pumped StoragePicture of the Ludington Pumped Storage

Three similar pumped storages exist in the hills around Deggendorf, the home-town of Michael Bonke, though much smaller than the Ludington Pumped Storage set-up. The storages in Deggendorf belong to the Ruselkraftwerke, and serve exclusively to outbalance the differences of power demand and power generation of the German Electricity Grid.

Mr. Eder, who is managing the Ruselkraftwerke and has built himself one of the three Deggendorf Pumped Storages, will retire next year, and has agreed to help us construct a pumped storage in Auroville.

Here are a few pictures of one of the Pumped Storages in Deggendorf. Click to enlarge and view comments.

Upper lake of the Parstweiher pumped storageLower lake of the Parstweiher pumped storage. At this moment the upper lake is full and the lower lake is empty. Therefore one can see nicely the 1.5m wide pressure-pipe, which carries the water from the upper lake to the turbine houseSouthern slope of dam containing the lake on the hill. The Parsteweiher (the upper lake) is a man-made lake on top of a hill. The hill itself is a man-made hill which is built on the slope of a mountain

Southern slope of the dam containing the lake on the hillThe turbine house of the pumped storageInside the turbine house. The picture shows the Franklin Turbine used for this pumped storage

A big shovel-turbine wheel of a Franklin turbine which is at this moment under serviceThe inlet of the water coming from the upper lake shooting into the turbineWater inlet into the turbine

Control room of the three pumped storagesMr. Eder, manager of the Ruselkraftwerke, in the control-roomTransformers of the Ruselkraftwerke for the Parstweiher pumped storage


How could a pumped storage be built in Auroville?

There are in principle three options for building a pumped storage in Auroville:

a) If we create a hill with the earth from the excavation of the future Matrimandir lake and build a smaller lake on top of this hill. In such a system the upper and lower lake would have a height difference of approximately 30 meters. The upper lake would have a volume of approximately 160,000 cubic meters.

b) Suppose we do not want to involve the Matrimandir lake in a system of pumped storage, we could use thedifference between the beach and the highest Auroville area. The highest area is the Matrimandir area, so the upper lake should be not too far from Matrimandir.

c) There could be a combination of the two systems. If we take the proposed hill, which we intend to build with the earth from the excavation of the Matrimandir lake, then we dig next to this hill the bed for a second lake. The earth from this second excavation would then increase the hill in height. The second lake next to the hill could then be combined with a lake which we dig near the beach. The earth from this excavation could again be used to increase the hill in size.

If we stick to option a), a pumped storage on a hill constructed with earth from the excavation of the Matrimandir lake, then we would arrive at a storage capacity of electricity which could bridge a power-cut throughout the whole of Auroville lasting up to 48 hours. Scaled up to a population of 40,000 people, the system could still bridge a power-cut of 2 hours.

The following Excel-sheet, made by Toine, can give you the different options. You may vary certain parameters, like population etc, and thereby get the corresponding duration of possible bridging of power-cuts.

Pumped Storage Project
Energy Storage Capacity Matrix
g = 9,81
Storage Eff. = 80%
Population = 10 000
Generator Eff. = 80%
Cos Ф = 0,85
E (kWh) = g x h x v x 2.77 x 10-7
Energy Storage in kWh
Vol -> 100 125 150 160 175 200 225 250
x 106 l x 106 l x 106 l x 106 l x 106 l x 106 l x 106 l x 106 l
Avg. Height of Upper Reservoir
20 m 4 348 5 435 6 522 6 956 7 609 8 696 9 783 10 869
25 m 5 435 6 793 8 152 8 696 9 511 10 869 12 228 13 587
30 m 6 522 8 152 9 783 10 435 11 413 13 043 14 674 16 304
35 m 7 609 9 511 11 413 12 174 13 315 15 217 17 119 19 022
40 m 8 696 10 869 13 043 13 913 15 217 17 391 19 565 21 739
45 m 9 783 12 228 14 674 15 652 17 119 19 565 22 011 24 456
50 m 10 869 13 587 16 304 17 391 19 022 21 739 24 456 27 174
Energy Storage in Hours
Pop -> 10 000 10 000 10 000 10 000 10 000 10 000 10 000 10 000
kWh / Annum / Capita
500 kWh 7,62 9,52 11,43 12,19 13,33 15,23 17,14 19,04
1 000 kWh 4,76 5,95 7,14 7,62 8,33 9,52 10,71 11,90
1 500 kWh 3,81 4,76 5,71 6,09 6,67 7,62 8,57 9,52
2 000 kWh 3,33 4,17 5,00 5,33 5,83 6,67 7,50 8,33
2 500 kWh 3,05 3,81 4,57 4,88 5,33 6,09 6,86 7,62
3 000 kWh 2,86 3,57 4,28 4,57 5,00 5,71 6,43 7,14
3 500 kWh 2,72 3,40 4,08 4,35 4,76 5,44 6,12 6,80
Generator Capacity in MVA
Pop -> 10 000 10 000 10 000 10 000 10 000 10 000 10 000 10 000
kWh / Annum / Capita
500 kWh 0,54 0,54 0,54 0,54 0,54 0,54 0,54 0,54
1 000 kWh 1,07 1,07 1,07 1,07 1,07 1,07 1,07 1,07
1 500 kWh 1,61 1,61 1,61 1,61 1,61 1,61 1,61 1,61
2 000 kWh 2,15 2,15 2,15 2,15 2,15 2,15 2,15 2,15
2 500 kWh 2,69 2,69 2,69 2,69 2,69 2,69 2,69 2,69
3 000 kWh 3,22 3,22 3,22 3,22 3,22 3,22 3,22 3,22
3 500 kWh 3,76 3,76 3,76 3,76 3,76 3,76 3,76 3,76

How could Auroville benefit concretely from such a pumped storage?

a) A collective power back-up with a pumped storage would be much cheaper than the individual UPS systems which are in use now.  Pumped storage as power back-up comes to only half the price in its installation costs (including creating the hill) compared to the total price of all the UPS systems which could store a similar quantity of energy.

b) Pumped storage as energy back-up is much more efficient than battery-driven UPS systems. The efficiency of UPS systems is below 50%, while the efficiency of pumped storage is usually above 70%.

c) With a pumped storage we would have 24/7 uninterrupted current in Auroville.  We could either use the pumped storage right away as generator, similar to a huge diesel-generator, or we could use a more intelligent system:  a grid-based pumped storage,  which supplies energy to the grid, and which would be treated as a power-producing device which would wheel (to wheel energy is the technical term for feeding energy into the grid at its source, and draw it from the grid at its consuminginto the AV transformer. Even with wheeling the energy from the wind generators from the Auroville wind-park, we could already have uninterrupted power. If we purchase for Auroville a separate transformer into which we wheel the energy from the wind generators, then this transformer could run under a special provision that it would never be shut down.  However, this system has one big disadvantage:  The electricity which is coming in by wheeling and the electricity going out due to consumption will have to match time-whise. . The generation and the consumption will be counted according to time zones (peak consumption time – low consumption time, etc), and a different tariff applies for each time zone. At peak consumption time around 7 p.m.  the rate is at present between 15 and 18 Rs per kWh,  while in the low-consumption time the rate may only be 2.8 Rs. If we wheel energy into such a special 24/7 transformer, then the special electricity meter will compare the electricity produced by our wind-park near Coimbattore  timewhise with the electricity consumed in Auroville.   When we over-produce electricity in a particular time zone, we get only the lowest rate (2.8 Rs). But when we produce less electricity in a particular time zone than we consume, we have to pay the full tariff (for example 18 Rs / kWh during peak consumption hours).

That is why wheeling in combination with a NO-POWER-CUT transformer comes very expensive. But if we would have a pumped storage, then we could treat the pumped storage as an energy producer and feed electricity into the grid (thereby wheel it) at exactly the time when our wind generators produce less and when the rate is high. This way we could buffer the irregular wind energy or solar energy with our pumped storage.

Again in brief: we wheel our energy from the wind-park into the AV transformer. At times when the electricity rate is low, or when our wind generators are producing more electricity than Auroville consumes, we pump the water from the MM lake to the lake on top of the hill.  At night, when the electricity rate is very high, we let the water flow back, create electricity with our turbine, and feed it into the grid. Thereby we feed electricity at the time when it is most costly, and can thus buffer up our wheeling pattern.

d) Auroville could maybe also arrive at a special tariff with TNEB and offer to sell its stored electricity at peak hours for a high rate, like the German pumped storages do.

How would the environment benefit from a pumped storage in Auroville?

a) A pumped storage would reduce electricity consumption due to a higher efficiency in storage than the presently used UPS systems.

b) A pumped storage could help to outbalance the irregular power producing profiles of alternative energy devices such as wind generators or solar powerThereby more renewable energy can be produced without destabilizing the grid.

c) Pumped storage is presently regarded in Tamil Nadu as something very special and complicated. It is not common technology here in South India. If Auroville can install its own pumped storage, then this would demonstrate that even small private entities are able to install such a device. This could help encourage TNEB or private investors to install other such devices, which would help to increase the ratio of renewable energies in the grid.

d) If Auroville could negotiate with TNEB a special tariff for peak-hour power input from pumped storages, and if such a rate would prove commercially viable, then this could be a breakthrough for private investors, who could copy the Auroville storage system.

A few details of pumped storages

A pumped storage does not need two pipelines or a separate pump and separate turbine. The turbine which produces electricity is at the same time the pump which pumps up the water. When it runs in one direction it pumps and consumes energy, when it runs in the other direction it produces energy.

The horizontal distance between the upper and the lower lake do not affect the efficiency of the system. The horizontal distance at the Parstweiher pumped storage is several kilometres. Still Mr. Eder estimates the pressure loss due to resistance of the pipes is less than 1%.