Wind-turbine blade manufacturing has come a long way over the last couple decades. Just ask Derek Berry, a Senior Engineer at the National Renewable Energy Laboratory (NREL) in Golden, Colorado, and the Director of the Wind Turbine Technology Area within the Institute for Advanced Composites Manufacturing Innovation (IACMI).

Derek Berry is a Senior Engineer at the NREL in Colorado, and the Director of the Wind Turbine Technology Area within the IACMI.

“When I first joined the wind industry in the early 90s, we were still using a wet layup process to manufacture laminates for blades and just making the transition to infusion,” shared Berry at the recent O&M and Safety Conference put on by the American Wind Energy Association in San Diego, CA. Wet layup is a molding process that combines layers of reinforced fiber with liquid resin to create a laminate, whereas infusion is a technique that uses vacuum pressure to pull resin from a laminate.

Berry got his start in materials some 25 years ago, designing and building aircraft components in the Air Force. Soon after he joined TPI Composites and learned that wind-turbine blade manufacturing was a whole different ball game. “I recall walking the manufacturing halls at TPI during my first interview, looking for autoclaves and the types of processing we used in the Air Force,” he said. “I quickly learned as a young wind engineer that the process was different, and instead of spending hundreds of dollars for a pound of composite material, it was five dollars a pound.”

Today, blade manufacturing is still evolving, and material costs, blade designs, and testing are keys to a successful industry. However, one of the first lessons Berry learned had less to do with materials or manufacturing, and more to do with communication.

“When I first started in wind, there was little interaction between different designers. For example, those working on the aerodynamics of blades were busy designing what they considered thin and high-performance airfoils, and then passing that onto the structural engineers,” he explained. Airfoils should minimize drag and blade-soiling losses. “So, the structural designers would simply use what they got, no questions asked, and integrate the airfoils into the blades.”

This process would work out fine if there were no room for improvement. “But there is always room for improvement,” said Berry. “The gap in communication between the aerodynamic and structural designers, and the manufacturers, meant airfoils were never fully optimized for their exact purpose. The result was heavier blades than necessary because of this basic lack of interaction.”

At the time, there was less communication than there is now with those the field, running the turbines. “The end users are often the key to innovation because they are the ones experiencing firsthand what works and what fails. It is important to ask the questions and to listen to one another in the industry,” said Berry. Fortunately, times are much different today. Over the years, a lot of effort went into changing the culture of communication in the industry between engineers, manufacturers, and operators. The result?

“Evolution,” said Berry. “Back in the early 2000s, when aerodynamic and structural designers began to talk, it prompted the first designs for flat-back airfoils.”

Blunt trailing edge or flat-back airfoils provide several structural and aerodynamic performance advantages over previous designs. Structurally, the flat-back increases the sectional area of a blade. But aerodynamically, it increases the sectional maximum lift curve.

NREL’s new CoMET facility in Boulder, Colorado innovates wind-turbine blade manufacturing by letting researchers design, prototype, and test composite blades and other components in one place.

“The inboard section of blade may not be as high-performance with flat-back airfoils, but it’s really not needed there,” explained Berry. “They also give structural designers more flexibility and the ability to use thinner laminate schedules and thinner blades to get better results for material strength and stiffness.”

Typically, the design of a turbine blade begins on a computer, but an optimal structure will incorporate feedback from the field. “Today we have longer and lighter blades because of that communication,” said Berry.

Manufacturing tolerances
Communication is an important element of good design. Open dialogue is also the first step in defining industry guidelines and agreed upon standards, which have evolved since Berry first began in the wind industry.

In fact, a new wind-turbine blade design and manufacturing document from the IEC (international standards organization, the International Electro-technical Commission) is currently under development. The aim is to provide an opportunity for credit to blade manufactures that properly quantify and control blade variations during production.

“This has been another important discussion, and one that treads a fine line,” said Berry. “Think about it for a moment: what is or should be considered a defect, and what is just variation in a blade?”

This is an important question to answer because blade designers must understand how much variation is acceptable during manufacturing. “It may sound basic or even trivial but if we held manufacturers to zero variation, blades would cost much more than they do today.” Berry said that by allowing a range of variance in manufacturing, it keeps blades relatively affordable. “The key here is to understand what is an acceptable level of variance, and to qualify it in terms of understanding how it affects the reliability and lifetime of a blade.”

Case in point: the material fiber direction in a spar cap in the center of a turbine blade. “It is worth determining how straight, wavy, or off-axis fibers are in a blade — anywhere in the blade but particularly in the spar cap — because variance can affect the reliability and longevity of that product over time,” he said. “A small wrinkle may be OK, but proper standards mean it should be controlled and under a certain level.”

The same goes with the bond gaps or the adhesive glue joints between components, which Berry said should be tested for thickness and potential voids. This is where industry communication and discussion is important to ensure that reasonable yet high-quality standards develop and evolve over time. “We’ve actually come a long way in getting material fibers straighter and in creating more reliable blades,” said Berry, who credits those improvements to better designs, new materials, and better checks and balances.

Non-destructive evaluation
NDE is an assessment method typically used at end of the manufacturing process as a final check to ensure a new blade measures up and is acceptable for use in the field.

“NDE basically tells us if we’ve made a good blade or a bad one,” said Berry. “However, we now want to use NDE much earlier on and as an up-stream process, so we’ll know during production — rather than afterwards and once it’s too late — if a blade is going to be a ‘bad’ one.”

In addition to prototyping of new blade materials and production methods, NREL’s CoMET facility will also offer workforce development and training for the composites industry.

Just imagine a laser system that can detect just how straight blade fibers are when they’re laid down, Berry said. If the laser detects a defect, it is possible to correct or re-process it at this point rather than after the blade is fully formed.

“The point is to make up-stream NDE invisible within the manufacturing process so it never becomes a hassle or hindrance to technicians or engineers, but rather helps them do their job better and with fewer errors,” he said.

NREL is also evaluating the potential of testing components and smaller blade sections to see how well they predict performance of the entire blade. “Testing is critical but as blades get larger, the process is more challenging and costly,” Berry explained. He said a full-scale structural test of an entire turbine blade can take six months or longer, and the cost ranges between $300,000 and $750,000.

“So the trend is toward component-level testing for use as certification instead of a full-scale blade test each and every time,” he said. “We’re not entirely sure this will work but we are evaluating different options.” He said NREL is also considering testing smaller-scaled blades and looking at how results correlate to a full-sized structure. With turbine blades reaching over 80-m, it makes sense to want to scale down the structures for more efficient testing.

“But we have to also consider what the industry will accept as far as scaling test results of smaller-scale structures to larger blades, and what testing facilities and equipment are available to do so in a manner that ensures correlation with how that blade will ultimately hold up in the field,” said Berry.

Better materials
Blade materials are also evolving. One such material, thermoplastic resin, is currently undergoing testing for use in turbine blades. “Almost every single megawatt-sized turbine blade produced today has been made with a thermal set material, such as epoxy, vinyl ester, or polyester,” said Berry.

Although thermoplastic resin systems are new to wind, they have been around a long time in other industries. Much like metals, thermoplastics soften with heating, and can eventually melt and then re-harden with cooling.

Berry said there are a number of potential cost-effective benefits to using thermoplastics for turbine blades. “The material could cut manufacturing times, getting more blades out the door more quickly. It could also decrease capital costs.” Unlike an epoxy, thermoplastics do not have to post-cure after molding, so manufacturers can save costs on ovens and utilities. “Thermoplastics could actually decrease the amount of energy it takes to make a blade,” he said.

However, the most impressive benefits according to Berry are the material’s ability to thermo-weld, repair, and recycle.

“If you heat up thermoplastic, it melts and you can reform it. That cannot happen with epoxy thermoset,” he said. This means a manufacturer can take two parts of a turbine blade and, instead of using an adhesive, simply heat up and meld the two parts together. “In the future, this could potentially mean moving away from problematic adhesive joints and, therefore, creating better and more reliable blades.”

Thermo-welding may also allow for easier blade repairs. “Rather than grinding up thermoset materials and then putting parts back together by hand layup or by infusion, it may be possible to melt down and reform or repair blades in the field.” The time and cost savings here could be significant, though Berry said more research is required.

This is something that NREL, in conjunction with IACMI, is working on at the new Composites Manufacturing Education and Technology (CoMET) facility. The CoMET is located at NREL’s National Wind Technology Center in Boulder, CO, and will let the organization lead composite research projects and rapid prototyping of new blade materials and production methods for the wind industry.

“We’re just learning about thermoplastics, how they can potentially be used with infusion and pre-preg processes,” said Berry. NREL has already produced a thermoplastic nine-meter blade using molds provided by TPI Composites, and plans to manufacture full-scale blade components using tooling donated by GE Energy.

“One final plus related to thermoplastics is their ability to provide for more recycling options at the end of a turbine blade’s life,” added Berry. “Instead of cutting up and throwing blade materials into a landfill, we can melt and re-use them, making wind power truly a renewable energy source.”