Engineers at the Sandia National Laboratories in Albuquerque provided a sort of state-of-the-research report at its recent Sandia Wind Turbine Blade Workshop with detailed presentations on a turbine array to study wake induced turbulence, the design for a 100-m blade, and more. The Lab is working on a range of wind-industry projects, several of which are advanced enough to warrant a public review.

A lab in Lubbock
The first project is a turbine array planned for the dusty plains west of Lubbock Texas. The area does not yet look like a test facility but soon will as Sandia crews begin erecting the first three turbines, one contributed by Vestas. The company is assisting with the conversion on the turbines from fix speed (governed by pitch) to variable-speed generators. Their rpm will range from 0 to 55, faster than original.

It has been well known that turbines produce wakes that drift far behind a working turbine. But how they affect turbines downstream is less well known. Because CFD and wind-tunnel studies can only tell so much, real-world experience is needed. Hence, the “Lubbock Lab” and the project called Scaled Wind Farm Technology Facility (SWIFT) will reveal those secrets.

The dessert near Lubbock was chosen because it has no trees, foliage, or buildings, it has Class IV and V wind consistently from the south, and it’s flat for miles around, much like an offshore site would be. Also, partner Texas Tech University has a 200-m met tower there with weather instruments at several levels. The tower is tall enough to detect low level jet streams. The University also maintains the Mesonet, a network of about 60 anemometers across the state. Other sensors, such as lidar and sodar, will dot the landscape to better characterize the air space.

In Phase II, two other turbines will be added for a column of four machines. These turbines will include reliable, 1994 vintage and rebuilt 225-kW units with 27-m rotor diameters. These are a minimum size that would work in the program but still give good results. Their turbines characteristics, for instance, provide a Reynolds number of over 1.7 million and will be over two million after upgrades. This allows results from the smaller machines to scale to the characteristics of a 1.5 MW machine. What’s more, the technology on these 225-kW turbines is widely used so there are no problems with proprietary equipment. These small turbines were also chosen because they convert easily from fix speed to variable-speed operation and they are inexpensive to buy and erect.

Project lead Jon White says crane rental cost is another reason for the smaller turbines. For instance, a crane needed to erect a conventional utility-sized turbine on an 80-m tower can cost $450,000/day ($250,000 is a typical charge for mobilization and several days on site) and must be scheduled months ahead for time, while smaller and more readily available cranes are scheduled a week ahead at about $5,000/day.

Phase III will erect two more turbines for a triangular shape to the array and Phase IV will provide more turbines for an array of ten.

White also acknowledges partnerships that make such a large undertaking possible, organizations such as DOE, Texas Tech, Group Nire, and Vestas. Phase I of the project is expected to become operational late summer or early fall.

Quest for the 100-m blade

No company has built a 100-m blade–yet. The closest anyone has gotten is Siemens in a recently announced 75-m blades for offshore turbines. Larger machines will need the 100-m blade, or as one engineer called it, the 0.1 km blade.

A Sandia report on its progress designing a 100-m blade came from team member and structural engineer Todd Griffith. Study goals were to identify challenges and barriers to a cost effective design. No prototype will be manufactured but the design will be publically available for companies that might want to investigate the process and improve on the results. A turbine accounts for 28% of an installation cost so the study was one way to reduce costs.

Griffith says the study was purely a conceptual exercise but as realistic as possible with layups and a complete suite of analyses. The design was all fiberglass without carbon materials and no other optimizations. There is room for improvement because no aerodynamic or power improvements were sought.
The geometry started with a 60-m blade from Dutch firm Dowich. Design was guided by accepted certifications. For example, IEC guided the structure and GL guided safety. Principle analyses was for strength, fatigue, deflection, and load-plane calculations.

“The challenge in finding such a long blade begins by recognizing that simply doubling the length of a 50-m blade boosts the weight by nearly fourfold, so gravity becomes a big design driver,” says Griffith. The design settled on would weigh 114,000 kg but Griffith acknowledges was not optimized for weight or power enhancements. Nonetheless, it would be appropriate for up to a 13.2-MW turbine. The root section would measure about 7.2-m dia.

“For blades of this size, buckling will be a big issue,” says Griffith. To handle that problem, the design team initially added a foam thickness but that added too much weight. “Instead, we settled on a third shear web to the chord for the last 14.6 m of the blade. Fatigue life, however, was calculated at a healthy 1,290 years. And flutter would be a problem only at a maximum operating speed.”

The baseline design, called Sandia 100-000 is based on conventional technology. A next step, says Griffith, is to apply carbon-fiber materials, bend-twist coupling (passive control characteristics), and flat back airfoils.

In another presentation, Wetzel Engineering President Kyle Wetzel mentioned that his company would be producing a 100m blade for a 10 MW offshore turbine what will be produced within two years. However, he offered few other details.

Sandia’s White says the goal of the Lab is not the basic research that a university would provide or the commercialization that a company would fund. “Ours is in the middle, to get basic ideas into simulation and scale testing for products that industry can carry forward.” WPE

How close is too close? Spacing and other questions relating to wind farm efficiency will be addressed with the Scaled Wind Farm Technology Facility or SWIFT project. One row of four turbines will let Sandia wind researchers measure the effects of rotor wakes on downstream machines.