Now nearing the culmination of a nine year project, Siemens Energy has decided on a site in western Denmark where it intends to assess direct drive technology for its suitability in wind turbine applications. The first of the two 3.6 MW test machines was installed in July, with the second to follow later in 2008. The turbines and their drive units will be subjected to comprehensive testing for a minimum of two years. The purpose of this project is to assess whether direct drive technology is competitive with geared machines for large turbines.

The main advantage of direct drive is the smart, straightforward design that it lends to the wind turbine, eliminating altogether the need for a gearbox. With fewer moving parts DD technology has the potential to reduce maintenance costs, which could result in higher turbine availability. But typical direct drive machines are known to be heavier and more expensive in manufacturing than geared wind turbines. One of the main objectives of this Siemens project is to establish if, and for which turbine sizes direct drive machines are competitive with geared turbines. A special focus will be on offshore applications, where machine robustness and reliability are of particular importance.

Pros and cons

Wind turbines are conventionally made with a gearbox that forms an integral part of the transmission system. The main shaft carries on one side the wind turbine rotor and on the other a large, multi-stage gearbox. The gearbox typically has a ratio in the range of 1:100 and converts the high-torque, low-speed rotation of the main shaft into a low-torque, high-speed rotation of its output shaft that is connected to a normal high-speed generator.

The geared transmission system has many advantages. It permits the use of a standard generator operating at 1000-1500 rpm. All components of the system can be purchased from a number of experienced suppliers, which makes it easy for a wind turbine designer to use a geared approach. And last, but not least, a geared transmission system is normally considerably cheaper than a directly driven, high-torque generator.

A geared transmission system does have some drawbacks, however. The main drawback is the complexity of the gearbox – a typical gearbox for a large turbine has three stages, two planetary stages and one helical stage, with a total of 13 gears and pinions and 22 bearings. To ensure trouble-free performance for 20 years for such a complex type of component, several supporting systems are required, including a sophisticated oil conditioning system that can maintain clean, cool and water-free lubrication at all times.

The alternative to a geared wind turbine is a concept where the gearbox and high speed generator are replaced with a low-speed multi-pole generator, which has a fixed connection to the wind turbine rotor and therefore rotates at the same speed.

The main advantage of the direct drive concept is that it eliminates the need for a gearbox and thereby greatly reduces the complexity of the turbine assembly as a whole. It is also slightly more efficient than a gearbox-generator combination.

Unfortunately, these advantages do not come free of charge. In the absence of the gearbox the generator has to be able to convert the full rotor torque. As a result the dimensions of a direct drive generator are very large compared to a high-speed generator with similar power rating, and a DD generator is typically much heavier than the combination of gearbox and high-speed generator for the same power rating. Only a few manufacturers of electrical machines are able to handle the weights and dimensions of direct drive generators for megawatt-scale wind turbines, particularly if offshore-grade insulation with vacuum pressure impregnation (VPI) is required. All of these factors combine to make direct drive generators rather expensive compared to the geared alternative.

Assessing direct drive

Siemens has been producing geared wind turbines since 1980 and despite occasional problems with gearboxes over the years it has always been, and remains, happy with the technology. However, the company has also realised that for large wind turbines direct drive generators might in serial production become competitive in due course with geared solutions. If this turned out to be the case then the simplicity and robustness of direct drive technology would be a decisive advantage for the offshore wind turbine installations that employed it.

Siemens has decided therefore to establish hands-on experience with direct drive technology in order to establish whether or not it can be made competitive with geared technology, and if so from what power level.

The project has four phases:

1. Selection of the preferred generator technology

2. Design and manufacturing of prototype generators

3. Testing under laboratory conditions

4. Field testing in wind turbines.

Phase 1 was carried out during 1999-2005. All known combinations of relevant technologies were investigated, including elements such as flux direction (radial flux or axial flux), excitation (electrical or permanent magnet), winding system (concentrated or distributed windings, slot/pole/phase combinations, round or rectangular wires, etc.), protection and cooling (open or closed, air or liquid cooling) and others.

In the end a permanent magnet excited synchronous generator design was selected. Permanent magnet generators have a number of advantages. They can be made extremely simple and robust, requiring no excitation power, slip rings or excitation control systems. Owing to the high field strength of modern permanent magnets they can be made very compact. And since no power is used for excitation, efficiency is very high, even at low load.

Permanent magnet machines have the disadvantages that the output voltage is a function of rotor speed and power, and that the power factor cannot be regulated directly. These drawbacks prevent the use of permanent magnet generators for normal, grid-connected power generation. But for a wind turbine generator these difficulties are irrelevant. A directly grid-connected synchronous generator cannot be used in a wind turbine anyway, since, by being locked to the grid frequency, it will have much too stiff characteristics for wind turbine use. The needs of the turbine act in the opposite direction – it should preferably be allowed to operate at variable speed, independent of the grid frequency. Therefore, in modern wind machines fitted with a synchronous generator a power converter is installed between the generator and the grid. The power converter first rectifies the variable voltage, variable frequency output current of the generator into direct current, and then inverts it into fixed voltage alternating current at 50 or 60 Hz depending on the local grid frequency.

Traditionally, permanent magnet machines have also had the disadvantage that the magnets would lose some of their flux density, and therefore their ability to generate power, over time. This problem is solved nowadays with the use of rare earth magnets, containing the elements neodymium and dysprosium. Such magnets are fairly expensive, but the advantages clearly compensate for the price.

Generator manufacture

Phase 2, the design and manufacturing of the generator, was carried out during 2005-07.

Siemens selected two leading manufacturers of large electrical machines as suppliers of the prototypes. The Large Drives business unit of the Siemens Industry Sector and Converteam of the UK each designed and manufactured one prototype. By selecting two manufacturers Siemens has been able to compare and assess the benefits of different technical solutions with respect to generator design.

Since it was expected that direct drive technology would be relevant only in the high power range, the rating of the test turbines was selected as 3.6 MW, corresponding to the largest Siemens turbine in serial production. Having a 107 metre rotor and 3.6 MW rated power the SWT-3.6-107 standard turbine is currently one of the company’s top selling volume products, together with the SWT-2.3-93 turbine type. While the 2.3 MW model is mainly used onshore, the 3.6 MW model is almost exclusively installed offshore. The first 25 x 3.6 MW offshore turbines became operational in the UK during 2007 in the Burbo Banks wind farm. Currently 54 x 3.6 MW turbines for the 180 MW Lynn & Inner Dowsing offshore wind farm (UK) are being installed, and Siemens recently announced a record order of 180 x 3.6 MW turbines for the Greater Gabbard offshore wind farm, also in the UK. Because of the success of the 3.6 MW unit, Siemens has become the market leader in the offshore wind power business.

Selecting a proven technological basis for the test machines has made it possible to use key main components from the 3.6 MW series-produced machine. The 107 m rotor, the tower, the controller and the power conversion system are all unchanged, and modifications are limited to the so-called nacelle, mounted on top of the rotor and supporting the generator. For the nacelle it was not possible to integrate DD technology with the standard components, and a completely new drive train has therefore been developed.

Integrating the direct drive generator into a proven wind turbine system substantially simplifies the testing arrangement, enabling the focus on generator performance, system reliability and cost-effectiveness during the test period.

The two generators are about 5.5 metre in diameter, 2.5 m in length and weigh 70-75 tonnes. At a rated torque of about 2500 kNm they are among the largest permanent magnet machines ever built. A totally enclosed design was selected. This approach has been used for all Siemens offshore turbines and has worked extremely well. Cooling is performed with heat exchangers, and the internal space of the generator is always kept dry and free of salt and dust.

Lab testing

Phase 3, laboratory testing, was started during 2008 and is still continuing. For the testing process a complete test bed was built at Siemens’ wind turbine factory in Brande, Denmark. The test bed has a continuous load capacity of 4 MW at 13 rpm, corresponding to a rated torque of about 3000 kNm.

Each of the two prototype generators has been mounted on the test bed, and a comprehensive testing programme has been carried out. Factors tested have included temperature, vibration and efficiency measurements through the entire power range and at continuous overload. In addition, the settings of the power converter system has been optimised for permanent magnet generator operation.

The tests on the first prototype generator have confirmed all theoretical assumptions made in the conceptual studies. At the time of writing this generator has completed all tests successfully and has been removed for mounting in the first turbine. The second prototype generator is currently being tested, and progressing according to schedule.

Field testing

Phase 4, field testing, commenced in 2008 and will continue until 2010. The first direct drive test turbine was installed in July, and the second will be installed later this year. Testing will commence in August.

On the test turbines the direct drive generator is supported by the main shaft rear end and is positioned behind the mainframe and tower, in the same position as the ‘normal’ gearbox position of a Siemens turbine. This allows for easy access to all relevant parts of the generator during testing, and it also facilitates possible corrective actions.

The generator cooling systems and the AC-DC power converter are all located at the rear of the nacelle, which is enclosed in a 6 metre diameter, 13 m long, conical and cylindrical steel canopy.

Laying the foundation

This direct drive project has attracted huge attention, not least because a potential shift in technology by a high-end manufacturer would be seen as a potential game-changer. However, even though direct drive wind turbines are an exciting technological option, it cannot at this stage be concluded whether they will in the end prove to be a competitive alternative to geared turbines. But with the installation of the two test turbines Siemens is laying the foundation for a fact-based assessment.

The company is continuing to develop its turbines using geared technology. For example, a new 2.3-MW machine with a 101 metre rotor diameter for sites with low or medium wind speeds is currently at the prototype stage.