How can paint reduce wind turbine blade lightning damage by 73%?

by Neal Fine

June 27, 2023

A Brief History

In 2017, a colleague and I were studying the electrical properties of an active flow control device called a dielectric barrier discharge plasma actuator when we discovered that the presence of a particular thin, semi-conductive coating on the surface of the actuator had the rather remarkable effect of completely eliminating the build-up of electrostatic charge on the surface. While the coating had a very interesting impact on the performance of the actuators, we also immediately began to think about other potential applications for the discovery.

At the time, we were interested in developing plasma actuators as active lift control devices for installation near the tips of wind turbine blades. The idea was that the additional control over the local sectional lift of the blade would allow wind turbines to react nearly instantaneously to changes in the wind associated with gusts and turbulence. Controlling the local lift in response to gusts, for example, could reduce the unsteady fatigue loads experienced by the turbine’s blades, nacelle, tower, and foundation. The topic of active lift control for wind turbines has a rich history and will be the subject of a future blog post. But for now, understand that any device - be it a sensor or an actuator - that is installed near the tip of a wind turbine blade must be able to survive a lightning strike. The threat of lightning was therefore a potential obstruction to the active lift control concept and was on the top of our minds.

And so, it was natural for us to consider whether our discovery might have application in protecting wind turbine blades (including lift control devices) from the threat of lightning. Having just finished writing a proposal to the Department of Energy’s SBIR program on the subject of active lift control for wind turbines - with an unlikely two weeks to spare before the deadline - I made the rather rash decision to quickly write a second proposal on the subject of lightning protection. Since the build-up of electrostatic charge on the surface of a wind turbine blade weakens the dielectric strength of the fiberglass, perhaps the surface coating would help to reduce damage caused by lightning. With the support of our lightning guru and mentor, Andy Plumer of NTS Lightning Technologies, we were able to put together a compelling proposal and submit it just minutes before the deadline.

Unfortunately, the active lift control proposal was rejected. But the hastily-written lightning protection proposal was awarded. (Not to worry, we later won a grant under the 2018 ARPA-E OPEN program that supported our active lift control development - again, reference the future blog post!)

The awarded DOE SBIR project was titled “Development of a Wind Turbine Blade Surface Coating to Reduce Damage Due to Lightning.” We went on to win two follow-on awards: a Phase II award in 2019 and a Phase IIB award in 2021, supplemented by several small grants from the State of Rhode Island. These grants allowed Arctura to develop, test, refine, and validate a simple and robust coating that reduces damage to wind turbine blades caused by lightning. While it didn’t work precisely in the way that we originally envisioned, our investigation led to further understanding and additional discoveries that eventually produced a coating that works remarkably well.

How Wind Turbine Lightning Protection Systems Work

Most modern wind turbine blades are equipped with lightning protection systems (LPS). The most common LPS uses one or more surface-mounted lightning receptors connected to a grounded down conductor (see the illustration below). A typical lightning strike begins when the strong electric field induced by a charged storm cloud and downward stepped leader causes upward streamers and leaders to emanate from the grounded lightning receptors on the blade. A direct strike results when one of the upward leaders from the receptor connects with the downward stepped leader from the cloud passing a large amount of electrical charge.

Unfortunately, this system has some flaws. Streamers on the blade can frequently originate from the internal down conductor (or other metallic components inside the blade) instead of the receptors, leading to, at best, small punctures in the skin and, at worst, split trailing edges or major delaminations. Punctures, if left untreated, also allow moisture to infiltrate the blade, which can weaken the structure over time and may lead to catastrophic damage from subsequent strikes creating explosive increases in steam pressure.

Even when upward leaders originating from inside the blade do not connect to form a direct strike, they can penetrate the skin and weaken the structure. Over time, this can reduce the blade skin’s insulating properties, making it more susceptible to future lightning damage.

What’s needed is a way to ensure that the leaders that form over the exterior surface of the blade at the receptors “out-compete” those that form inside the blade and win the race to connect with the downward leader from the cloud.

How the ArcGuide® Coating Works

The ArcGuide® coating is the result of the four-year DOE-funded SBIR project. It is a polyurethane-based topcoat that is applied to the exterior surface of wind turbine blades in the vicinity of the lightning receptors. The coating enhances the performance of the LPS by making it more likely that lightning will attach to the receptors and less likely that it will penetrate the skin of the blade in the process of attaching directly to the down conductor or other metal component inside the blade. The ArcGuide® coating contains a distribution of conductive particles of a certain size, shape, material, and concentration within the polyurethane matrix. The coating itself is not conductive. The low concentration of conductive particles disrupts the electric field over the surface of the blade in a way that leads to early formation of electric arcs (“streamers”) at the receptors. The receptor-fed streamers then grow faster and further over the surface of the blade, out-competing the streamers that form inside the blade in the race to connect with the downward leader from the cloud.

This activity was assessed visually in the high voltage lab at NTS Lightning Technologies using both square fiberglass panels and wind turbine blade tips. The panels and blade tips were subjected to a high DC voltage simulating the electric field moments before a lightning strike, and a photo of the streamer and leader activity was captured.

One such test featured a panel coated with the ArcGuide® coating on the left half and a baseline topcoat on the right half. The panel was arranged such that both halves were subject to the same initial electric field conditions. A mock lightning receptor was installed at the center of the panel. The figure below shows the resulting photo. The outline of the panel, center receptor, and dividing line between the coatings have been highlighted. Increased streamer and leader activity is clearly seen on the left side (coated with ArcGuide® coating) relative to the right side (coated with baseline topcoat). This was observed in all trials. As previously discussed, this increased activity is key to guiding strikes to the receptor and thus reducing the probability of punctures and damage elsewhere on the blade.

Measuring the Effectiveness of the Coating in Reducing Damage

The effectiveness of the ArcGuide® coating at reducing damage has been assessed using high voltage initial leader attachment tests per Annex D of IEC 61400-24. The testing was performed at the NTS Lightning Technologies test facility in Pittsfield, MA.

Several wind turbine blade tips from GE 1.5sle wind turbines were obtained after being retired from field use. This turbine model was chosen because it is the most common machine operating in the United States today and is known to suffer from lightning damage. Different blade variations exist depending on the time of production and manufacturer. The results presented here are all for a “37C” variety which has a single large lightning receptor only on the pressure side of the blade near the tip. Typical blade tips tested were 5 meters in length and had some existing damage from years of operation.

The IEC standard summarizes lightning risk, testing, and general protection requirements for the wind industry. The standard calls for the blade tips to be suspended over a ground plane with the receptor at high voltage. The blades were tested at angles of 60°, 30°, and 10° with respect to ground. Shallower angles are harder to protect as the electric field is increased inboard and away from the tip receptor.

Example of an initial leader attachment test setup from IEC 61400-24.

Possible angles and orientations for the initial leader attachment tests.

At each angle, the blade tips were rotated to different orientations to allow each side of the blade to experience the highest electric field.

Three trials (“strikes”) were conducted for each orientation. The blades were at negative polarity for all tests and the waveform and peak voltage were kept constant. Photos of each strike were captured, and the strike behavior (i.e., puncture or flashover to the receptor) was categorized based on the path seen in the photo. Suspected punctures were further confirmed by inspecting the blade for damage. For a given angle (60°, 30°, and 10°), if any strike at any orientation resulted in a puncture, that angle was denotated as “FAIL”. If all three strikes for all orientations for an angle successfully flashed over to the receptor, that angle was denoted as “PASS.” After each puncture, the resulting hole was patched with 5-minute epoxy. This helped increase the insulation resistance at that damage point, but the location remained a weak spot in subsequent tests.

Statistically significant quantitative results pose a challenge for this type of testing. The stochastic nature of lightning means that many tests are required to gain confidence. Every test, however, also electrically weakens the blade, and using a “new” blade for every test is not feasible. Furthermore, the blade tips tested in the lab were far from new due to lightning strikes and other wear and tear over years of use in the field. To minimize these biases as best as possible, tests were conducted on three different blades of similar age and condition. Tests for each blade configuration were ordered from least to most likely to puncture to minimize degradation and repair work. Each blade was also tested in both baseline and ArcGuide® configurations. For blades #1 and #3, the baseline configuration was tested first and then the ArcGuide® coating was applied. For blade #2, the ArcGuide™ coating was applied first, then it was sanded off and a baseline topcoat was applied before testing again.

The figure below shows two composite images of all strikes to blade #3 hung at an angle of 60°. The baseline configuration (left) shows the fail condition (punctures) while theArcGuide® configuration (right) shows the pass condition (all flashovers to the receptor).

Composite image of blade # 3 (hung at a 60 deg angle with respect to the floor) with the baseline coating, showing failures on two of six strikes.

Composite image of blade #3 with the ArcGuide™ coating, which passed all trials at that angle.

The blades coated with the ArcGuide® coating performed significantly better than the blades without the coating. Blades that were coated with ArcGuide® incurred no damage when tested at angles of 30 deg or higher. At 10 degrees, the blades with the ArcGuide® coating suffered half as many punctures as the baseline cases.

In total, 154 strikes of four GE 1.5sle blade tips were conducted in the lab. For each angle and each configuration (baseline or ArcGuide®), a failure rate was calculated as the number of punctures divided by the total number of strikes (either punctures or flashovers to the receptor). This is plotted in the figure below. Limited testing below 10° and at 90° indicate a 100% and 0% failure rate can be assumed at these angles respectively for both configurations.

The failure rate is defined as the number of punctures divided by the number of strikes at each angle (for all orientations).

Data on the attachment angles of lightning strikes on operational wind turbines was obtained from a publication that analyzed video footage of winter lightning strikes on 12 turbines over a 5-year period in Japan. The attachment angles of 172 strikes in total were quantified. The histogram in the figure below shows the reported strike frequency as a function of blade angle. In this case, 90° refers to the blade extending straight up, and 0° refers to the blade extending horizontally either ascending or descending.

Histogram of attachment angles of 172 lightning strikes to 12 turbines over a five year period. From S. Vogel, “Realistic Lightning Exposure System for Optimized Wind Turbine Reliability”, Technical University of Denmark, 2018.

Of note is that 98% of the observed strikes attached at a blade angle greater than 30°, and no punctures in the lab were observed with the ArcGuide® coating at a blade angle of 30° or higher. Applying the baseline failure rate observed in the lab (linearly interpolating between angles tested) to this distribution of strikes, 6.4% of strikes would be expected to puncture through the blade skin and cause damage. Applying the ArcGuide® failure rate to this distribution, only 1.7% of strikes (the three at 0°) would be expected to cause punctures. Thus, a 73% reduction in the frequency of punctures is predicted in the field with application of the ArcGuide® coating for typical GE 1.5sle blades. The usual caveats about the stochasticity of lightning are relevant here and these results should be interpreted with caution. Even with the hundreds of strikes observed in the lab and in the field for this analysis, the datasets may not be completely converged.

Concluding Remarks

Lightning is a top cause of wind turbine blade damage, and lightning damage repairs represent a significant maintenance cost for the industry. Over four years of empirical testing and optimization (in collaboration with NTS Lightning Technologies), With important contributions by all members of my talented team of engineers at Arctura, we have developed the ArcGuide® coating to reduce wind turbine blade lightning damage. In addition to the evidence presented here, a significant amount of data has supported the efficacy and performance of the coating. Much of that data can be found elsewhere on our website and in publications that can be easily found on the web (see this academic conference paper, for example).

The ArcGuide® coating is now available to purchase as the ALEXIT® BladeRep® topcoat ALP 20 with ArcGuide®. For more information, visit www.bladerep.com.