Back to the future4 August 2002
Water wheels are generally perceived as being inefficient energy converters which belong to the past, with no role for the future. But as Gerald Müller, Klemens Kauppert and Rüdiger Mach explain, they can actually be efficient and cost-effective in low head micro hydro applications
IN EUROPE, a large number of low head micro hydro power sites (head = <5m, power = <100kW) exist. Water power was a prime power source during the industrial revolution and thousands of water mills were built at low head sites. Today however, the large majority of such hydro power sources is not exploited due to the lack of a cost-effective hydraulic power converter. Recently, a number of developments appear to have opened up the possibility to generate electricity economically at low head sites. These developments include the two oldest hydraulic machines, namely water wheels and the Archimedian screw - the latter working in reverse as a power source rather than as a pump.
Water wheels are today often considered to be relics from the beginning of the industrial revolution; romantic but inefficient hydraulic machines made of wood and belonging to the past. It is generally believed that turbines are much more efficient than water wheels and subsequently took over their role as hydraulic power converters. The statistics however show a different picture. In Bavaria - a German province with an area of 70,500km2 - there were 7554 operational water wheels counted as late as 1927, with power outputs ranging from 0.75 to 75kW. In the middle of the 19th century, a high stage of development was reached when Zuppinger designed the most modern and efficient water wheel. Engineers, manufacturers and mill owners must have regarded water wheels as commercially interesting power sources. During the 1940s however, virtually all water wheels seem to have disappeared.
Today, some companies in Germany (Bega, Hydrowatt) and the US (Water Wheel Factory) are again manufacturing water wheels for electricity generation. The performance characteristics of such wheels still appear to be largely unknown. Assessment of the available power potential, comparisons with other turbine types such as the Kaplan or the Ossberger (crossflow) turbine, and even the determination of optimum operating conditions for water wheels, relies on estimates.
'Modern' water wheels, ie water wheels built using scientific principles, are made of steel and employ only the potential energy of the water since in low head flows the potential energy exceeds the kinetic energy of the flow by far. These water wheels can be divided into three fundamental types:
• Overshot wheels: the water enters the wheel from above; 2.5m < H < 10m, Q <= 0.2m3/sec per m width.
• Breast wheels: the water enters the wheel approximately at the level of the axis; 1.5m < H < 3m, 0.3 <= Q <= 0.65m3/sec per m width.
• Undershot wheels: the water enters the wheel below the axis; 0.3m < H < 2.0m, 0.45 <= Q <= 1m3/sec per m width.
Water wheels were, in the large majority of cases, used to drive machinery and reached efficiencies of 75-89%. This development seems to have subsequently been forgotten.
In Karlsruhe, Germany, a small non-profit research company (IFMW - Institut für Forschung und Medien im Wasserbau) has been set up which specialises in hydraulic engineering research, and in particular in the development and promotion of low head hydro power. Within the company, a very detailed literature and market review on water wheels was conducted in order to assess the suitability of water wheels for electric power production. Since only over and undershot wheels are currently built, the discussion will be limited to those two types.
The overshot wheel receives its feeding water at the top of the wheel, catches the water in buckets or 'cells' and releases the water at the lowermost possible elevation. In order to make maximum use of the energy contained in the water, the cells are shaped so as to receive the water at its natural angle of fall and then to retain it as long as possible.
Some measurements of the performance characteristics of overshot wheels were conducted in 1928, as shown in the figure below. It was found that the efficiency of a water wheel reaches 85%.
These measurements were taken with a gear ratio of 1:25, so that the shaft efficiency will probably lie close to 90%. In addition, very high efficiencies are maintained over a wide range of flow rates.
Modern turbines require sophisticated (and expensive) design and control in order to approach this performance band width.
Undershot water wheels
The undershot wheel was developed for the utilisation of very low heads from 0.5-2m. Whereas in ancient times the kinetic energy of the flow was utilised with a paddle-type wheel, 'modern' undershot wheels built after Zuppinger's design employ the potential energy only. The figure below shows a Zuppinger wheel with the typical 'backwards' inclined curved blades.
Wheel diameters range from 4-7m, with head differences from 0.5 to 1.5m. The blades are arranged in a way so as to avoid losses at the water entry, then gradually reduce the head of water in each cell and finally to discharge the water, again with a minimum of losses. The wheel blades are curved to allow for a gentle decrease of the water level from upstream to downstream, and to minimise losses at the downstream end. In an engineering textbook from 1939 it was stated that efficiencies of 76% can be guaranteed for properly designed undershot wheels.
Recently, the undershot wheel has also experienced a small renaissance. Hydrowatt has built and installed 15 Zuppinger wheels over the last nine years, with diameters ranging from 4-7.5m, and widths of 0.5-3m. Hydraulic heads utilised ranged from 1-2.2m, with typical flow rates of 1.5-3.1 m3/sec, giving power outputs from 4-45kW of electrical power. The overall efficiency (from hydraulic power available to electric power out) was estimated as ranging from 60 to 65%.
In the UK, many smaller streams were made navigable by building weirs, many of which still exist. Generally, the head differences were in the range between 1.2-1.8m. The undershot wheel may offer a possibility to produce electric power from such weir sites. The picture below shows a typical weir situation (Eel weir on the Lagan river in Northern Ireland) with a virtual water wheel inserted. The water wheel actually fits into a 'natural' environment very well, indicating that a modern machine can become a visually attractive feature too.
The Archimedian screw
The Archimedian screw has been known since antiquity as a simple machine for the lifting of water. Today, Archimedian screws are still in widespread use as pumps for sewage, grain and so on. It has the advantage of being a very simple machine, with only one moving part and two bearings. It was however only recently noticed that the screw could also - in its reverse role - be employed as energy converter, termed hydraulic screw. Large scale experiments with a hydraulic screw of 8.6m length and 2.35m drop were conducted at Prague Technical University in the Czech Republic in order to assess the performance of the hydraulic screw in its power generation mode.
The screw shown in the picture above was designed for a maximum flow rate of Qmax = 0.35 m3/sec. In experiments, it reached an efficiency of 70% for Q/Qmax = 0.4, and 80% for 0.6 < Q/Qmax < 1.0. The screw rotates at 53rpm, so that fewer gear ratios than for a comparable water wheel are required to achieve the speed necessary for electricity generation. To the author's knowledge, six hydraulic screws have been installed already. Some design guidance for hydraulic screws, based on the design experience with screw pumps and generator experiments, is also available.
It should be noted that Ritz-Atro of Nürnberg, Germany holds the patent for the Archimedian screw as a hydraulic power converter.
micro hydro converters
The economics of micro hydro converters are a function of variable boundary conditions such as electricity prices and so on. In Germany, overshot water wheels are currently built (including installation and grid connection) for 4360-4850 US$/kW. Undershot wheels cost 7760-9700 US$/kW, Archimedian screws approximately 8250-8730 US$/kW installed capacity. For comparison, low head Kaplan turbines cost 14500-15500 US$/kW. Although water wheels and the Archimedian screw have significant cost advantages over turbines, micro hydro installations are economical only if the owner uses the generated electricity at least partially, such as for a small business. Assuming 50% of self use, 6000 hours of operation per year at nominal capacity, a small business electricity price of 11 c/kWh and a price of 7.3 c/kWh for electricity fed into the grid, the following pay back periods apply:
• Overshot wheel: 7.7 - 8.5 years
• Undershot wheel: 13.7 - 17.1 years
• Archimedian screw: 14.5 - 15.4 years
• Kaplan turbine: 25.6 - 27.4 years
Both water wheels and Archimedian screws therefore appear to be competitive alternatives to turbines.
The general perception amongst the public as well as many engineers is that water wheels are inefficient energy converters and belong to the past. Water wheels, and in particular the overshot variety, are however very efficient and cost-effective energy converters for low head micro hydro power applications. Today, the wider application of water wheels seems to suffer from a lack of information on water wheels, the lack of any design guidance and - possibly the most important aspect - the lack of actual data about the performance of such wheels.
Apparently, no performance data at all exists for undershot wheels, whereas some information is available on overshot wheels in old reports. IFMW Karlsruhe is currently conducting a detailed review and analysis of the experimental data available on overshot wheels with a view to application of such wheels for electricity generation.
The determination of performance characteristics for overshot wheels, and the publication of such data, will be the next task of the company, followed by a detailed evaluation of undershot wheels.
The use of the Archimedian screw for power generation is a recent idea. A lot of data and design expertise does however exist from the application of the Archimedian screw as a pump, and large scale experiments in the generator role were conducted so that the design methodology of Archimedian screws is quite developed.
Ecological aspects are today a major issue in the design of hydro power installations. Both the water wheel and the Archimedian screw are considered to be very fish friendly because of the large compartments for the water and the slow speed, even for long fish like eel, thus giving them a considerable advantage over the fast rotating Kaplan turbines.
It appears therefore that currently the two oldest hydraulic machines are beginning to experience a renaissance in rejuvenated form, and for a new purpose: the generation of electricity. Both types of energy converter can economically fill a niche where standard turbines appear not to be able to operate cost-effectively, and therefore deserve the attention of the hydraulic engineer.