Magnet price hikes hit direct drive's future

However, rare earth element price increases are putting financial pressure on the concept. Eize de Vries reports.

In recent years, permanent-magnet type direct-drive generators have become increasingly popular in the wind industry. However, recent price hikes of several rare earth elements (REEs) including neodymium and dysprosium, commonly used for the powerful magnets, could disrupt this momentum.

Main wind-turbine drive systems can be functionally subdivided into those with and others without a gearbox. The primary function of both systems is to convert power in the wind into mechanical power and, finally, mechanical power into electric power. In direct-drive turbines a gearbox has been eliminated, making the rotor speed and generator speed identical.

The benefits of direct drive include the elimination of fast-speed components - not least failure-prone gearboxes - and a substantial reduction in component numbers. Conversely, a large diameter multi-pole ring generator is required, which is much bulkier and heavier than fast-speed equivalents of similar power rating.

Direct-driven ring generators have been used in hydropower applications for the last century. These come with quoted diameters of up to ten to 12 metres.

But in contrast to hydropower, wind-turbine operation is characterised by continuously varying wind speeds and, as a consequence, substantially fluctuating rotor speeds and generator loads. That in turn puts high demands on the generator outer-shape retention while exposing it to continuously changing combinations of mechanical, thermal and electrical loads.

In 1992, German wind-turbine manufacturer Enercon faced up to these challenges and became the world's first to introduce a commercial wind turbine with direct drive, the 500kW E-40. According to wind-industry observers, the turbine was a good compromise between manufacturing costs and overall systems efficiency.

Since then, the E-40 platform has been the basis for a comprehensive range of direct-drive models with power ratings up to 7.5MW. For many years Enercon nearly monopolised the direct-drive segment. From the start the company has used classic synchronous ring generators with external field excitation of the electromagnets and a choice in diameters and, more recently, liquid-cooling systems for specific-power ratings.

As of early September, Enercon's cumulative track record (30kW - 7.5MW) is over 18,000 units. Over the years, several hundred more direct-drive turbines with external field excitation have been built by competitors including Lagerwey, Emergy Wind Technologies (EWT), Northern Power Systems and MTorres.

Nearly all of these have chosen substantially lighter and more compact permanent magnet generators (PMGs) for their latest designs. GoldWind and XEMC, both of China, have together already manufactured thousands of such 1.5-2MW class turbines.

Goldwind's 1.5MW concept originates from Vensys in Germany, while XEMC manufactures 1.5-2MW turbines based upon a former Dutch Zephyros concept. The list of established suppliers and newcomers entering the direct-drive PMG segment has also grown substantially in the past years. It includes among others Alstom, Avantis, EWT, GE, Lagerwey Wind, Nordex, NPS, Siemens and STX.

Direct drive PMG benefits and challenges

In the second half of the 1990s, the head mass (nacelle + rotor) of direct-drive turbines was quoted on average 15-30% higher than fast-speed geared turbines of similar power rating.

However, that initial direct-drive disadvantage has changed in the past years, especially for new models applying PMGs. In fact for some of the latest direct-drive PMG turbines the head mass is even less than that of geared machines of similar power rating and rotor size. Another main PMG advantage is superior partial-load efficiency, a common operating condition.

On the other hand, being able to externally regulate field strength in classic generators offers an extra control variable, such as easier starting up. Enercon has also taken substantial strides towards reducing head mass, especially for the 800-900kW and the 2-2.3MW series.

Until recently, it seemed only a matter of time before the industry would switch to direct-drive PMGs. But the rise in price for REEs used for magnets in recent months has caused a major increase in generator manufacturing costs. This has fuelled a rethink on remedying options including alternative drive-system configurations.

Diverse options

If this trend continues, staying with direct drive could involve a choice between radical and conservative solutions. The former option could involve reducing both magnet quantity and copper requirement - and thus costs - by substantially increasing generator diameter. Disadvantages to this include more complex road transport logistics, and a necessity to increase structural-steel mass to ensure sufficient generator stiffness and strength.

An alternative option is to shift to less expensive lower-grade REEs. But such a measure puts extra demands on magnet-temperature management and overall generator cooling and is likely to drive up generator mass. Another possibility is a switch to non-REE magnet materials. A number of alternatives are reported to be in development. Again another option is a shift away from PMGs and a return to classic direct-drive generators with external field excitation, which eliminates the use of REEs.

Lastly, if a company is determined to use a PMG direct-drive mechanism plus high-grade REEs, there is the example of Siemens to follow, which is taking a share in an REE mining company, a measure among others aimed at increasing supply-chain control.

High magnet prices impact direct-drive PMGs significantly as they reportedly contain up to a factor of ten more magnets than fast-speed geared wind systems. A number of recent new drive-system developments build on the 5MW Multibrid concept, a hybrid between direct drive and fast-speed geared that comprises a single-stage planetary gearbox and a low-speed PMG.

One of those concepts is German manufacturer Winergy's HybridDrive introduced earlier this year. It comprises a two-stage planetary gear system and a permanent-magnet type synchronous generator integrated into a single product, focused on both offshore and onshore wind turbines up to 6-7MW.

A main product-development focus was to enable a compact design whereby a flange connection between gearbox and generator eliminates an otherwise required intermediate shaft. Winergy claims that its HybridDrive requires only about 20% of the REE quantity compared to direct-drive PMGs of similar power rating.

Several other new products are reported to focus on combining a low-speed or medium-speed gearbox with a PMG, induction, or classic synchronous generator.

NOVEL SOLUTIONS - A look at the rationale behind Nordex and vestas drive-drive designs

Since the beginning of the year several suppliers, including Alstom, Nordex, NPS, Siemens and XEMC have presented new offshore-dedicated direct-drive PMG wind-turbine concepts. Gamesa and Vestas opted for medium-speed geared PMGs.

With its new 6MW N150/6000 offshore wind turbine Nordex has switched to direct drive. It hopes for higher reliability and availability due to significantly fewer components and better yield thanks to reduced drive-train losses.

A distinct design feature of the 6MW N150/6000 is a liquid-cooled PMG located behind the tower. Chief technical officer Eberhard Voss pointed to its lowest cost of energy (CoE), optimised serviceability and easy major components exchange as the main design drivers.

For its new 4MW onshore turbine Nordex by contrast has gone for a medium-speed concept with two gear stages. This time Voss considers a maximum main component or sub-assembly mass limitation of 75 tonnes essential, whereas key design drivers are maximised reliability, design flexibility and optimised CoE.

With regard to the V164-7.0MW offshore turbine design, Finn Strom Madsen, president of Vestas Technology R&D, says that the decision-making process was driven by two objectives beyond lowering the machine's CoE.

These were drive-system choice and serviceability. Initially, the company followed two parallel drive-system development tracks, direct drive and geared. The design team asked itself what were the main factors behind turbine downtime.

In-house studies supplemented by third-party reports from 2010 showed, according to Madsen, that faulty electrical components cause more failures than gearboxes. A second main finding he refers to is that other mechanical components such as the fail-safe brake and blade bearings also cause more downtime compared with gearboxes.

"Direct-drive generators contain a factor four more electrical components compared to medium-speed geared," says Madsen. For Vestas, the main conclusion was that wind-turbine reliability discussions should not focus on gearboxes. "We had also decided that no technology solution should create high and unwanted dependence on REEs used for the PMGs," adds Madsen.

The comparative drive-system study resulted in a three-stage medium-speed drive system, which steps up ten or 12 rotor revolutions to a 400rpm-rated generator speed.