15 July 2015
This blog first appeared in Renewable Energy Focus (http://www.renewableenergyfocus.com/view/42606/tidal-power-a-different-perspective/) 

Kepler Energy is taking a different approach to tidal power, applying a design that can be used in benign sea conditions in a location closer to end-demand and which could generate utility-scale quantities of power, as Peter Dixon explains.The physical environment of tidal turbines is clearly very different to that of wind turbines, not least because water is a lot denser than air! However, despite the different tidal projects that have been developed but not commercialised in recent years, the full implications of this are not fully understood. The physics involved is not what has been assumed.

Logic and economics imply the use of transverse horizontal axis turbines that span across the tidal flow. Using these transverse axis machines, tidal energy "fences" can potentially extract utility-scale quantities of power.

Kepler Energy’s approach has been, with the help of Oxford University’s Department of Engineering Science, to understand the physical environment and the consequences for energy extraction, then apply a turbine design to achieve this effectively.

What’s Different?

In the early days of the tidal industry, it seemed obvious that tidal turbines should be like wind turbines – basically, horizontal axis axial machines – but stronger. It also seemed clear that these machines would be subject to the Betz Limit, which describes the maximum amount of energy that can be extracted from a flowing fluid, such as air. But there is a major difference between the underwater environment and air – there is a free surface, and this causes the fundamental physics to be subtly different. The theory of this is now well understood.

The key insight is that, in circumstances where there is significant blockage (where the area of the turbine is large relative to the flow of water – for example where the diameter of a transverse turbine is 50% of the depth of water), the power output becomes more a function of head efficiency than of the velocity of the flow. In other words, the turbines extract potential energy as well as kinetic energy.

So when three Professors at the Department of Engineering Science at Oxford University decided that they should try to design a more effective tidal turbine, this was a key factor. But to Professors Guy Houlsby, Martin Oldfield and Malcom McCulloch, one further (in fact rather obvious) insight was that, in order to ‘capture’ the maximum amount of tidal flow, the rotor of the turbine itself should be transverse and horizontal – in other words, it presents to the tidal flow a rectangular shape rather than a circular shape. Wind turbines can be extended by increasing the diameter of the rotor, within limits, but conventional tidal turbines are limited in their diameter by the depth of water. Being able to ‘stretch’ the rotor horizontally is a great advantage.

Whilst the shape in itself is not novel, there is a problem with making a structure strong enough to withstand the very large forces involved in generating vary large quantities of power. The solution is to make the entire rotor structure a three dimensional stressed truss – and this patented solution permits the construction of rotors of considerable size, enough to generate over 2MW each at tidal velocities of 2 metres per second.

To maximise power output using these machines, blockage needs to be optimised, and this can be done by stretching a series of transverse horizontal axis machines across the tidal flow in the form of a tidal fence, and it is envisaged that in the right location, these fences could be 10 to 15 kilometres long, with peak power outputs of 600+MW.

To investigate the potential power output of tidal fences in the Bristol Channel, detailed modelling has been undertaken at Oxford by both the ETI Perawat project and Kepler Energy. The work shows the potential power output of fences in various locations, and demonstrates how fences might interact with each other.

In addition, the testing of the prototype has given power output results which validate the results of the computer modelling. It is partly this work which has convinced Kepler Energy to move ahead with a proposed 30MW demonstration fence. The other part of the equation has to do with cost.

A different approach to cost

The industry as a whole has found it difficult, to get turbines into the water and generating – it’s been time consuming and expensive. Any new technology will require time and development effort to succeed. But the locations chosen to develop the turbines have also been the most difficult in which to exploit new ideas. Because power output is proportional to the cube of the tidal velocity, the natural instinct was to head for the areas with the highest velocities, such as the north of Scotland and The Orkneys.

The difficulty with this is that the sea conditions – high winds, high seas and, obviously, rapid tides – and the depths required, mean that accessing the machines is challenging and so expensive, even with the most modern and powerful vessels. In addition, the remote locations mean that there is little local demand for any electricity generated, meaning that an expensive transmission system must be constructed. So, the costs all add up.

Why not make life a bit easier and go for more benign sea conditions in a location closer to end-demand? The issue here is that, almost by definition, the tidal velocities are lower. However, because tidal fences which exploit the blockage phenomenon can have very high power outputs despite the apparently quite low kinetic energy available, these more benign conditions can be exploited. For example, the Bristol Channel, excluding the shipping lane, has appropriate depths for a tidal fence, has sea conditions which permit access for a higher proportion of the time than in the Orkneys, using ‘off the shelf’ equipment, and has grid transmission system and power demand close by.

Will the proposition make money?

The benign conditions alone are not enough to ensure that costs are low enough to bring costs down to below the level of offshore wind, although there is clearly a big impact upon operating costs. Kepler Energy has worked closely with its supply chain to produce robust designs which will satisfy the arduous demands of operation offshore at a highly competitive cost. Programme management has been handled by Altran, famous for its Solar Impulse PV powered plane, and Gurit has been responsible for detailed design of the carbon composite stressed truss. Royal Haskoning DHV has advised on environmental requirements, whilst GE has worked with Kepler Energy on the electrical system.

The price for the generated electricity is clearly a critical input to the profitability equation. Commentators often ask how it is that renewable sources of energy have ‘such high costs when the price of electricity is only (say) £45/MWh’? But they don’t also ask the question, why is the ‘wholesale’ price what it is?

Published work from DECC indicates that even so called low cost technology, CCGT, would have a full cost of around £60 to £110/MWh depending upon the capex and hydrocarbon price assumptions, and even more with Carbon Capture and Storage. The conclusion must be that current price levels are set by the marginal economics of fully depreciated old assets (coal and nuclear) and that future prices must rise to encourage new build of capacity – whether these prices are made available in some free market system or through a system of price support such as the Feed in Tariff with Contracts for Difference, one or the other or both are necessary.

Kepler Energy, through its work referred to above, is confident that tidal fences in the UK can be profitable under the current CfD regime. But there are many other areas of the world where such technology can be used. For example, appropriate cost versions are ideally suited to the islands of Indonesia, where the competition comes from diesel generation. Areas of the coastline in Korea, Japan and China are well suited to the application of tidal fences since they have the characteristics – lower peak tidal velocities and lower depth – for which the Kepler Energy turbine is so well suited.


We have seen that the physical environment of tidal turbines is clearly very different to that of wind turbines and that the full implications of this are not well understood in the industry. Logic and economics imply the use of transverse horizontal axis turbines that span across the tidal flow. Using these transverse axis machines, tidal fences can potentially extract utility scale quantities of power.

Kepler’s approach has been, with the help of Oxford University’s Department of Engineering Science, to properly understand this physical environment and the consequences for energy extraction, then apply a turbine design which will generate electricity profitably.

Subject to planning and financing, the initial Bristol Channel 1km £143m tidal fence, which is likely to be located in the Aberthaw to Minehead stretch of water, could be operational by 2020/21.

ABOUT THE AUTHOR: Peter Dixon is Chairman of Kepler Energy

FURTHER INFORMATION: Kepler Energy: http://www.keplerenergy.co.uk