Weather Guard Lightning Tech

Drone VS. Rover Inspections, AI Crack Monitoring

Allen Hall, Rosemary Barnes, Philip Totaro, and Joel Saxum discuss the evolution of wind turbine blade inspections, from external drones to internal rovers. They debate the potential of AI in predicting damage progression and managing repair priorities, with Rosemary emphasizing the complexity of crack propagation in composites. Joel highlights Top 7’s innovative drone technology for detecting lightning protection system faults in blades, as featured in PES Wind magazine.

Sign up now for Uptime Tech News, our weekly email update on all things wind technology. This episode is sponsored by Weather Guard Lightning Tech. Learn more about Weather Guard’s StrikeTape Wind Turbine LPS retrofit. Follow the show on Facebook, YouTube, Twitter, Linkedin and visit Weather Guard on the web. And subscribe to Rosemary Barnes’ YouTube channel here. Have a question we can answer on the show? Email us!

Pardalote Consulting – https://www.pardaloteconsulting.com
Weather Guard Lightning Tech – www.weatherguardwind.com
Intelstor – https://www.intelstor.com

Allen Hall: On the mean streets of Lowville, New York, Phil that’s your neck of the woods. The local Kraft Heinz plant has reclaimed the Guinness World Record for the largest cheesecake. Tipping the scales at a whopping 15, 008 pounds. Yes, that’s right. That’s seven and a half tons of creamy goodness.

Joel Saxum: I absolutely love cheesecake. My brother loves cheesecake so much that’s what he had at his wedding. He had a smorgasbord of different kinds of cheesecake that you could pick from.

Allen Hall: They broke the record, almost double the record that was held from a team from Russia. So here we go. Now we’re back into the 1980s.

Olympic hockey

Philip Totaro: exit no that’s great that’s good that’s a good thing we should be world domination in cheesecake size

Allen Hall: and

Philip Totaro: wait.

Allen Hall: Yeah why did we get an invite joel i don’t understand we should’ve been top of the list to come to lowville.

Philip Totaro: That’s yeah that’s what i’m saying like did they pass it out to everybody in town like how do you eat a cheesecake seven tons of cheesecake.

Allen Hall: They donated to local food bank is what they did after everybody had a slice or two or three. But 15, 000 pounds of cheesecake. What’s that in metric tons, Phil? Come on. I need a sense of this for the Europeans in our audience. 6. 8 metric tons. That’s a lot of metric tons, but this, these are the things you got to keep your eyes open for, right?

So if they’re going for a world record. And anything food related, they need to be calling the Uptime Podcast and at least give us a heads up so we can plan our travel accordingly, because this cheesecake thing seems like we missed out.

I’m Alan Hall and here are this week’s top news stories. In our first story, Vestas has secured its largest onshore wind project to date in Japan. The company has received a 134 megawatt order from Invenergy. for the Inaniwa Wind Energy Center. The order includes 32 V117 4. 2 megawatt wind turbines and a 20 year service agreement.

Deliveries are expected to begin in the first half of 2027 with commissioning planned for Q1 2028. Moving to Spain, Windar has started preparatory work to construct a new monopile factory. The facility will have the capacity to manufacture monopiles up to 12. 5 meters in diameter, 3, 500 tons, and 130 meters in length.

With an annual capacity of 100 to 120 monopiles, the factory aims to supply wind farms in the North Sea, the Baltic Sea, as well as the East Coast of the United States. In a significant development for the Mediterranean region, nine Southern European Union member countries have pledged to turn the area into a renewable energy hub.

Officials from Cyprus, Slovenia, Malta, Croatia, Greece, Italy, France, Portugal, and Spain are focusing on harnessing offshore wind and solar energy. They aim to set up a joint renewable energies project across borders and have called on the European Commission to conduct a study on the region’s renewable energy potential.

Shifting to Sweden, Northvolt has announced a revised scope of operations in response to challenging market conditions. The company will focus on ramping up its first 16 gigawatt hour production capacity at Northvolt ET. resulting in the redundancy of approximately 1, 600 employees. Northvolt will suspend the expansion project at Northvolt at this time.

and slowdown programs at Northvolt Labs. In India, Xcel Composites and its joint venture Koneko Xcel India have won a bidding process to supply poultry to Carbon Planks for wind turbine spar caps to Vestas. The products will be manufactured in KECI’s new factory in Goa, India, with deliveries estimated to begin in the last quarter of 2025.

This agreement extends an existing multi year frame contract and deepens the collaboration between the companies. And Denmark is making strides in wind turbine testing capabilities. R& D Test Systems has completed the foundation of a test bench for main bearings at the Offshore Renewable Center in Odense Port.

The facility will be able to test both geared and direct drive main bearing arrangements for 25 megawatt turbines. The project, supported by a 10 million euro grant from the Danish Greenlab program, is expected to be delivered in 2025. Lastly, Nikon Corporation is set to provide Eurus Energy Holdings Corporation with a riblet film for wind turbine blades, aimed at improving wind turbine generation efficiency.

In Japan’s first such verification test, the film will be attached to turbine blades at the Yurisoya Misaki wind farm. Nikon anticipates that the Riblet film could improve wind power generation efficiency by approximately 3%. That’s this week’s top news stories. After the break, I’ll be joined by my co host, renewable energy expert and founder of Pardalote Consulting, Rosemary Barnes, the founder and CEO of Intel Store, Phil Totaro, And the Chief Commercial Officer of WeatherGuard Lightning Tech, Joel Saxon.

Lightning is an act of God, but lightning damage is not. Actually, it’s very predictable and very preventable. Strike Tape is a lightning protection system upgrade for wind turbines made by WeatherGuard. It dramatically improves the effectiveness of the factory LPS, so you can stop worrying about lightning damage.

Visit weatherguardwind. com to learn more, read a case study, and schedule a call today.

Allen Hall: It’s almost fall in the northern hemisphere, which means that blade inspections are occurring. A lot of drone inspections are happening right about now at the end of season after all the repairs have been done.

So you’re typically seeing two scans in some wind farms. One in the spring, one in the fall, just to keep track of how things are progressing. But there’s been a recent big shift, I think, in from moving away from external drone inspections, which had been the norm, to a lot more of internal rover inspections.

With a little automated car with the cameras on it and the lights and the beepers and the whole thing. And I think the industry has really learned from that. I’ve been talking to a number of blade engineers and we were at Sandia last week that the significant structural defects are turning up sooner on the inside.

And that’s a big shift, Rosemary, I think, because for so long, we were just doing drone inspections and when we got to standardized drone inspection, that was the big deal. But. Now we’re doing inside and outside.

Rosemary Barnes: Yeah, I actually wondered this because of a job I was on recently They were going to go do internal inspections that I mean to be honest I hadn’t even been doing external inspections properly But yeah They were going to go do some internal inspections and I just wondered how they actually get into the blade like I know that you can get in between the two webs and trailing edge is mostly pretty clear, but in the leading edge between the front web and the leading edge, there are bulkheads all the way down these bits of foam or whatever.

They’re structural, mostly not, mostly to assist with manufacturing, but they’re blocking off the cavity. You can’t just crawl into the leading edge cavity and crawl all the way down, no matter how small you are, because you’re going to encounter these solid obstacles. What do they do about that?

Are they punching their way through them, which it being, like I said they’re not primarily structural, but they’re not structural. How do they manage that?

Allen Hall: I don’t think they’re doing anything at the bulkheads. I think a lot of it is right between the spars, the shear webs. And I think from what I can tell so far, are they not looking for problems with core splice areas, transverse cracks?

Which are the Cat 4, Cat 5s that you would need to know. I know they have leading edge and trailing edge dis bonds, but the big ones seem to be in the mid span region.

Rosemary Barnes: That’s true, that they would catch most of the catastrophic, like most of the catastrophic stuff would happen on the main laminate, which is, yeah, right down the middle where it should be possible to get a really long way down, so I guess they can probably capture most of it.

But yeah the system that, or the, one of the things that, I know that they’re looking for with drone inspections is, when lightning flashover happens, then that can cause damage on the inside. And you might not see it on the outside until it’s like a category five and the blades about to snap in half.

That can obviously happen on the leading edge. And yeah, but I guess maybe they’re not catching all of that, but they’re catching other damage that. Might cause big structural problems. Makes sense.

Philip Totaro: Wouldn’t it make sense to have a channel running the length of the blade adjacent to the one of the, like the front or the rear spar so that you could put, rather than a Rover, couldn’t you use one of those, like a larger size of a surgical camera and just run it down this channel and have it do the inspection on the trailing edge bond line?

Rosemary Barnes: They do boroscope inspections. Although often that’s also from the outside. I think that they, find lightening damage and then poke around inside with a a broscript to see if there’s a lot of internal damage so that they would need to, cut it open and fully repair it or not.

But I was just thinking, I thought you were going to say shouldn’t the manufacturers make that channel so it makes maintenance and operations easier? I think they should and that they never will because no one cares about the full lifetime, life, yeah, full life cycle of a wind turbine.

It’s all just about that, upfront cost.

Philip Totaro: Not when they’re in the business of selling aftermarket blades. Yeah. Yeah. And they’re service. It’s Hey, we got a brand new blade you could buy instead

Rosemary Barnes: of repairing the old one. It’s basically like the inkjet blade. Printer model now and wind turbines like, Oh, give me the wind turbine for free.

But yeah, extort you on the service and and spare parts. But I was at this conference this week. It was a systems engineering conference. They brought me in to do yeah, one of the keynote speeches there. And so of course I stayed around for most of the rest of the conference and learned some interesting things.

And one of the guys I got talking to he, he does a system analysis. He’s worked with offshore wind in the Netherlands. For installations actually, but he was saying that, systems engineering approach, what the industry desperately needs is to do a systems engineering approach right from the start, where you take into account like the full life cycle of the turbine so that you are when you’re designing a wind turbine, you are actually thinking about, repairs and all that sort of thing that is going to need over the next 10, 20, 30 years of its lifetime.

And that should be a yeah, a design requirement early on. And I was thinking that, yeah, we do need that, but I just can’t see the mechanism that we could ever get that. Cause it would just cost money for the manufacturer and they wouldn’t be able to sell it for anymore. So yeah, they’re obviously not going to do it.

Joel Saxum: We were at the Sandia conference last week talking with a lot of operators and you’re starting to see a difference in approaches with internal inspections. So you see some that are, yeah, we do internal inspections just like we do external inspections. We inspect everything. There’s some that are even large operators looking at, we just do targeted ones on issues that we know exist, or blades that have a history of certain things, and some doing the, hey, we do 10 percent of a wind farm because they’re expensive, right?

They’re, in the United States, they’re between 1, 000 and 1, 800 per turbine. So they’re, five times the cost of an external inspection. So you have a lot of people doing 10 or 20 percent of their fleet or 10 or 20 percent of a wind farm. And if they find an issue, then they’ll go and start expanding the scope.

But one of the things that was common, right? Is we know that there’s quite a few people starting to look at doing crawlers and inspections internal in blades. With not just with people, but with actual, with the crawlers to get a good baseline from commissioning. So they’re either doing it right when they come out of the factory or right when they get delivered on site to make sure that they’ve got documented what that blade looked like before it was hung up.

And to me that is a fantastic idea.

Allen Hall: Are the crawlers still being operated by the ISPs or are the operators starting to rent them basically to do the internal inspections? You’re getting a little bit of

Joel Saxum: both and that’s driven by cost model, right? So same thing with external inspections where we started to see that Hey, you can rent a drone from us or you can buy a drone from us and you can fly whenever you have a good day or winds are negligible or whatever that may be in the external side.

You’re starting to see that in the internal side. I know our good friends Armando at Earthwind and Rodolfo and the team down there from Brazil they’re start, they have some of them up in the United States now. So they’re putting them out there and that their idea is they made it easy enough to pilot them.

Easy enough to collect the data where they can basically put one in a box, send it out to a site, they can do their inspections, and then the data goes back to the mothership, down in Sao Paulo, and they do all of the analysis and stuff down there. You have, there’s other places like like the Aarons team, they’re very efficient at doing these in a cost effective way because When they’re on site, a lot of times they’re doing LEP, external inspections, internal inspections, cleaning towers, all kinds of stuff, so they have the same crew, less mobilization costs.

But yeah, if it was me, the cost of these things are to rent them, it’s not and a robot internal crawler is not like a drone that requires a lot of maintenance. Drone after so many hours, you gotta replace motors, you gotta do this, it’s very difficult to fly, you gotta have special license, or not very difficult to fly, but can be, you gotta have special licenses, all this stuff a crawler is can your six year old kid drive a remote control truck?

Okay? They can basically do an inspection, so the technicians can do that themselves. Fairly easily with a little bit of training.

Allen Hall: And the latest generation of crawlers, the images are outstanding. It’s scary. The one,

Joel Saxum: the one I really like is what Aaron’s has done for the the root inspections, because of course, the root in some of these blades is huge, right? Especially offshore. Some of these things are. three meters across at the root. They’re massive. So what they did was they have, of course, their regular camera setups at the nose that can see in, 180 degrees as they go through the whole thing. But they actually took a full frame CMOS sensor, like what would be, like a Sony a seven R type thing.

And they put it on the back of the drone and it’s perpendicular, it’s pointed perpendicular to the path. Then it rotates in a 270 degree angle up. So you’re getting a very high resolution image of the root area as you go into the root or through the root area.

Allen Hall: Is that really changing the dynamic though?

I think that the key to all this data is in the, because the images are so heavy and detailed now, are we just into the phase of, we need some AI to go through the images to find out what’s going on.

Joel Saxum: Yeah, everybody that’s a big player in the internal inspection world is developing some kind of AI to handle the analysis, right?

Because to be honest with you, the most important thing in internal inspection robotics, whether it’s with the Helios drones or whether it’s with the crawler or whatever, is the lighting, the quality of the lighting. Because if your lighting isn’t good in there, or if your lighting is too vibrant or too high color, you end up with sheen and shear off the fiberglass and you can’t see, or sometimes you’re kicking up dust, or you got some other things going on, so if the lighting is right, and the lenses are correct, then you can get these amazing images, and you get a ton of them, right?

You’re gonna get a whole pile of data, so I think Marone’s is doing some AI stuff. I think Earth Wind’s doing a little bit of AI stuff. Of course, every trade show you go to, there’s another one kind of popping out.

Allen Hall: When does UT happen? I, we at Sandia, we ran into a number of people that were doing ultrasonic tests and looking at specific parts of the blade that was particularly the root area was a big emphasis.

We need to see what’s going on down there. Are we doing a lot of ultrasonic inspections in the fall? quantify what’s happening on our blaze before it gets

Joel Saxum: cold? Not really. The thing with UT is it’s very specialized. It is an amazing tool. In Rosemary, you know this better than probably all of us, but NDT is an amazing tool to get what you want.

However, it’s expensive. It’s difficult to get really highly skilled, level two, level three, NDT technicians out there. The processing Is in, is intensive because composite NDT is very difficult compared to steel. Steel NDT is pretty cut and dry.

Rosemary Barnes: Yeah. And also, really to get really good visuals, it needs to be quite flat and wind turbine blades mostly not flat.

Even in the factory where they’ve got like the full big kit and the, full time employee who’s a expert, they still can’t just scan everything and see everything inside a blade. It’s like there’s certain parts that you can get a pretty good image of. And yeah stuff does get missed, but I guess they’re developing the technology all the time to get more portable machines and also to just improve the kinds of image you can get and the amount of, Complexity that it can deal with in terms of curvature and that sort of thing.

Joel Saxum: And then the other thing about NDT that people maybe don’t realize as much, if you’ve never played in that space is NDT is not one technology, right? Like CIC NDT and the team over there, those guys have I think a half a dozen or even more, maybe even 10 different kinds of technologies where, you know, laser shearography and just the UT that you would think.

And then you have, you’re looking at microwave for composites and all kinds of stuff. It’s right now. I think that when there’s specific and targeted issues and wind turbines, yes, they’re getting some of them are getting regular quarterly inspections, right? You’re seeing the root sections of some of these turbine models being inspected quarterly out in the field, but that’s because they know there’s an issue and there’s a specific problem there, but to have a like an autonomous external drones style, we’re just going to look at everything.

Okay. Thank you. Doesn’t happen yet in the NDT world. And it’s cool that we’re developing robotics for, cause that could un, unlock a little bit more power in the NDT world. But right now you have like experts like Jeremy Hanks over at CIC NDT that they have the teams that know how to do it. It’s just not in super high demand.

Allen Hall: And does thermal imaging start taking some of that space in terms of inspection because of the, sort of the localization that has to happen with. Ultrasonic inspections, where thermal inspections, it gives you a quick sense of how the blade’s doing. If you can get images, the resolution you want, is that the next generation of inspection before you hit wintertime?

I

Joel Saxum: think thermal inspections are a good, or can be, when it gets developed a little bit better than it is now, can be a good screening tool, but I don’t think they’re an inspection method that is something you bank on.

Allen Hall: Going back to Rosemary’s comment, you can’t see the leading edge, you can’t see the trailing edge, because the access is poor.

Thermal imagery should get you some image of the

Joel Saxum: leading and trailing edge. I don’t think the leading and trailing edge besides the trailing, a trailing edge crack that’s doing this, but you can see that most likely with a visual drone, you don’t need a thermal drone for that. The ones you’re looking for are the ones that are under the gel coat that are causing some friction or something of that sort, or a little hairline crack.

That’s why I think it’s a screening tool. You may look and see, okay, we’ve got, Hey, there’s something weird going on here. Let’s go up and look at it.

Allen Hall: Going back to your 10 percent rule of look at 10 percent of the turbines, would you quick scan a number of turbines just to get a gross sense of, Hey we have these, we may have this potential issue, then come back with UT to, to suss it out.

Joel Saxum: And if it’s fast enough, like I know the Rom R MotionCam guys have been working on this for a while. The Rom R MotionCam concept. Is a very quick and speedy inspection process because you don’t need lotto, you don’t shut the turbines down, you don’t do any of that. Especially if you’re in like, cornfields or something, or cut cave fields, you just put that remote cam unit in the back of a truck and drive and boom, as the turbine spins, you take a peek at it, take some inspection images, you move on to the next one.

Like I just, in my opinion, that’s a good screening tool, but that’s not an end all be all inspection. You may find an or isolate an issue like, Hey, we need to look at blade B on this turbine because there’s something going on there, but you’re not going to be able to diagnose what the issue is from that.

At the end of the day, right now, thermal imaging technology, unless you’re getting some kind of ITAR approved thermal imaging system from a, like an Apache helicopter, like you’re not going to see any kind of detail that you need to be able to diagnose a problem.

Allen Hall: The next generation after thermal is terahertz.

We saw some of that at Sandia where the, you in theory can see individual layers of fabric in the layup because the wavelengths are so small you can measure that and detect dis bonds and it’s giving you, from what I saw, a really accurate look at an evenness in the composite structure, much more than ultrasound can do, but I’m sure it’s not nearly as fast as ultrasound is, Which is the problem, right?

Is so as technology develops, are we going to just know too much?

Philip Totaro: And look, the reality with AI is that the image quality we have today results in AI that can only be a certain level of accurate. If we address the issues that Joel’s mentioned, and you get higher quality and higher resolution, not only with visual, but some of these other technologies, and you start layering things together, it can be data overload if you don’t take AI to scrub it, and figure out how to actually take that results of the AI analysis and do something useful with it.

Which is where, that’s really where a lot of the AI developed predictive maintenance solutions have fallen down. Up until now, is, we’ve got a ton of data, people aren’t really using it right. With or without AI, it’s still too much to analyze even today, and you don’t even get the quality, so there’s a lot of false positives in the data sets that we have.

Assuming that you can improve the quality of the imagery and reduce the false positive rate down to a point where you could actually do visual or any other type of inspection, get actionable information out of it, the question is then, how are you leveraging that actionable information to be able to address things like We see something that’s maybe a cat two or three damage right now.

I need to be able to determine how fast that crack’s going to propagate to become a cat four or five so that we can avoid a crane call out. Do we prioritize fixing that now? Or do we only have the budget to be able to fix the cat four and five issues that we’ve got? That’s still a huge challenge for operators at the moment.

So that’s where AI can take us.

Rosemary Barnes: Can it though? It’s I never. Never heard anybody that thought that you could tell you to tell how fast a crack was going to propagate over composites It’s just it’s unpredictable. You know, like a cat one crack could become cat five tomorrow or might never progress and Yeah, and

Joel Saxum: There’s people that are working on that project, right?

Rosemary Barnes: Yeah, but it’s like the holy grail of composites for at least the last 20 years. So I don’t know if we’re close and It might be the sort of thing that AI can crack. Ha, funny intended. It might be the sort of problem that AI can crack if if they had the data, but how are they going to get the data?

I guess if you like, maybe a cool research project would be to put cameras inside a hundred thousand blades and just have them, like watching cracks, and then they might, AI might be able to pick up something that does give you a, a tip off for when something’s about to happen.

You would need at least tens of thousands of data points. It’s not something that you get from, the amount of data that’s being collected today and and if, yeah, someone was to go through it, it’s all just, like one off snapshots and I don’t see us learning anything unless we set out to learn it.

Joel Saxum: The trouble with that, Rosemary, is exactly like you’re saying, you need all those data points, but then you also have to. You also have to filter all that data by blade model, sub model, manufacturer, resin type, core ramp. There’s so many things that you can’t actually, if all the blades were the same, then yeah, we could probably do that.

But there’s so many variations and out in the field that. It’s not possible. I don’t at this stage.

Philip Totaro: No and that’s, but that’s the point, isn’t it? As, as technology evolves and becomes cheaper, because it’s more reliable, whatever, it’ll get used more, and then we’ll start building that catalog of information over time.

But now is when we really, up until now, what we’ve been doing is just Stumbling through the methodology of like, how do you leverage, technology to do internal and external inspections. We’re now at a point where the technology is sophisticated enough to actually start giving us meaningful data and feedback that could eventually be used to build models that’ll do some of that prediction.

I think you’ll start getting more useful information out of it from what we have the opportunity to collect. But you’re right, I we are just at the beginning of the meaningful data collection process to be able to do anything with all that.

Rosemary Barnes: Yeah. There is heaps of, in useful information that is coming out from ai looked at images and especially compared to a few years ago, it was doing, it was worse than.

Worse than not looking at the images because you just got so many damn false positive positives that no, no one like was just totally overwhelming. Cat five grease

Joel Saxum: smudges.

Rosemary Barnes: Oh yeah. Like on two thirds of the blades and one of the ones I looked at there, had cat five damage and everyone else spent, oh yeah, that’s cool.

And continued on, like no one ever looked into it. So it was just like, it meant that they didn’t. They did these inspections instead of other ones and then couldn’t use the information. So it was just as if I never did any inspections really. So they went backwards, only a few years since then, like we do say a lot of really useful information now from AI stuff.

And of course, like for cat five, you’re going to look at the image yourself like with the human eyeballs as well and get up to, get a better look around. We haven’t totally divorced the need for humans involved at this point and I’m sure it will get better. But there’s definitely some issues are much, much easier for AI than others.

And yeah, I think that’s the thing, like AI, you just need just so much data. And especially with crack propagation, that is such a such a complex topic to start with. I don’t think that is going to be one of the next ones that we say, but. I would love to be surprised by that.

Joel Saxum: If we rewind to what Phil said, like Rosemary, you’re a consultant in the wind energy and renewable energy world.

So you deal with people on blade issues and other things. So you hear this from your clients. I’m sure Phil, you and I have talked about this in the off air. Many times, Alan, you and I talked to everybody else about this. Everybody just wants to be told what to do. They would rather not even have to look at these inspections.

Just tell me on turbine 7 you need to do this and on turbine 8 you need to do this and you’ll be fine. Great. Thank you. That’s where I want to go. Because we know there’s this, we always talk about the shortage of technicians but there’s a Even worse shortage of good engineers.

Rosemary Barnes: I’m happy to tell people what to do.

If someone needs to be told what to do, they can, they

Philip Totaro: can call me and I’ll tell them. The, Joel, the other thing there’s a shortage of is budget. Because I hear all the time that, most of the owner operators, not all of them, and there was one I spoke to back in Minneapolis at the wind power event who, I asked him like, Hey, do you guys like ever run out of budget before you have the opportunity to fix your, Cat 2, Cat 3 damage on blades or whatever else.

And they were like,

Rosemary Barnes: no, we’re good. Cat 2 and 3. People are running out of budget before they get to all the Cat 4 damage in my experience. Maybe it’s just an Australian thing, but definitely Cat 2 is never being repaired unless maybe if they’re up there anyway, that it might get done.

Philip Totaro: One owner operator in the U.

S. is doing that, or so they say. But that’s exactly my point, is most of the time, companies run out of budget fixing the Cat 4 and 5 damage that they obviously have to fix, because you gotta get the turbines back up and running, or prevent further catastrophic failure. The issue with that is, can we figure out a way, because, something that’s a Cat 4 today wasn’t always a Cat 4, and that’s why I keep going back to, can AI help us figure out how you can avoid a crane callout, Because that’s where it just gets preposterously expensive.

That’s why everybody runs out of budget. Because they’re getting to these too late, they’re being reactionary in their repairs, and not just on blades, but even on things like gearboxes or anything else, generators. They’re being, they’re still being reactionary.

Rosemary Barnes: With a good spreadsheet of every blade and every inspection and damage and tracking over time.

what it was, knowing, you can know from that what the average cost is for different categories of damage. I think you could definitely come up with a better strategy for repairs. Yeah, and because it would be statistical then, you don’t need to know. This particular blade is going to go from cut three to cut five on, August 27th, it’s going to say that statistically, if you leave this, it’s going to cost you more than what it would cost you to get up and fix a few that were never going to progress.

It’ll tell you where that point is that at category two, or is that at category three? for that you should be doing them that day. Or I think that definitely like a larger operator with a good spreadsheet and people usually don’t have good spreadsheets in my experience.

The, like it’s always like a lot of the work that I do is so tedious. You’re going back through. manually opening PDFs for this blade, okay this, was there a, chasing up the, I can’t see the report, in this year, was there one, and you can’t expect AI to make anything of that either, but if you put it all in a in an organized database, then I think that you could definitely pull out a Better repair strategy.

I wonder if people are already doing that.

Joel Saxum: When you open any one of those spreadsheets at any one of these operators, and you go filter by lightning damage, that spreadsheet is going to tell you to call WeatherGuard Lightning Tech and get strike tape installed because it’s going to lessen your costs over time.

Philip Totaro: Well, Joel if that database doesn’t exist, they should be calling Intel store and we’ll help them build that database because that’s where, you The real value is in, in understanding how those things, how this can escalate. I’ve had literally three different asset owners say that if that kind of tool existed, they’d buy it.

Joel Saxum: As busy wind energy professionals, staying informed is crucial and let’s face it, difficult. That’s why the Uptime Podcast recommends PES Wind Magazine. PES Wind offers a diverse range of in depth articles and expert insights that dive into the most pressing issues facing our energy future. Whether you’re an industry veteran or new to wind, PES Wind has the high quality content you need.

Don’t miss out. Visit peswind. com today.

Allen Hall: The new PES win magazine is out and it is thick, Joel. I was looking at it in the office this morning. I thought I was getting a workout from holding this magazine. My gosh, there’s a lot in this episode or this issue of the magazine. And obviously the thing about PES Wind is that the images and the photographs are amazing.

So it’s very easy just to look at the pictures, which is something I like to do when I first get it, like Picture book kind of guy? It is! It’s The quality of the images and the printing is just way beyond what you usually see in a print Magazine today, but the article I stopped on was one from top seven because it said lightning and anytime I flip through and I see something about lightning, I have to pause and look at it.

But top seven, which is a essentially a drone company or drone based company has this technique where they can determine if your lightning protection system is broken. Without crawling inside or doing a bunch of resistance measurements. And Joel, you want to describe how this device works, how the system works?

Yeah,

Joel Saxum: absolutely. So tech, classically. LPS measurements are usually done by resistance. So you put physically put a meter on the end of a receptor and either a meter on, the ground or within the blade. If you’re just checking in the blade, you have to have a technician in there and a technician on ropes or something of that sort.

So it’s usually done by a resistance measurement. So if you have very low resistance, your LPS looks good. The challenge with that can be is low resistance is how do I explain this? The cable, the LPS cable within a turbine is so large that you can have damage to it, you can have cracks in it, you can have strands, you can have partial disconnects, where the next lightning strike that comes will break that thing.

But it will still give you a good ohm reading. What this drone does is this is not a from a physics and electrical engineering standpoint, this is not a new concept. But it is the first time we’ve seen it commercially readily available on a drone and able to do this. What they’re, what the concept is you put a, basically an electromagnetic pulse through the down conductor.

And then with the drone, you have a sensor on the drone that has an electromagnetic field reader. And then you chase that pulse along the blade. So you don’t actually touch the blade, you can do it remotely, so you’re not having to fly right into it or up against it. You can still do it back a little ways.

But what you’re gonna do, or what this solution does is it pinpoints where the damage is or where the break is within the blade. Something that technicians have wanted forever, right? So technicians will be, hey, we’ve got a broken LPS, we gotta fix it, where is it in the blade? We don’t know, we know it’s broken.

So you can put someone in the blade and you can get so far along it until probably, it gets pretty tight once you get about to the 40 percent mark. Where a guy the size of me is definitely not gonna make it. I’d probably make it to the 50 percent mark.

Allen Hall: I was gonna say the root, but

Joel Saxum: Yeah, scoot along with a skateboard on my belly or something.

Allen Hall: Yeah, I’d

Joel Saxum: peek in there.

Allen Hall: I can’t get through the access panel. I’m way too tall for that craziness, yeah.

Joel Saxum: So what this tool does is it’s able to let the technicians pinpoint exactly where the damage is. So if you need to open up a part of the blade to fix the LPS, you can do it in that specific spot. It’s a great time saver.

It is. It has the capability. Now, I’ve never seen this piece of kit run in the field, so I don’t know exactly what the operation looks like, but in theory, I get it and I like it. So again, Top 7, they’ve got high resolution cameras on drones, they’ve got some other things, they’re a drone company, but this is their kind of flagship device that we’ve heard about for a while, able to find where the break is

Allen Hall: in the LPS cable.

So if you haven’t read this Top 7 article, you better get that. To PESWIN. com, download it and read it. You get this great information in there.

Joel Saxum: Okay, I was just hanging out with some of my cousins from San Diego that live near La Jolla Beach, but this wind farm is called La Joya. Wind farm in New Mexico, and I don’t want to get that wrong.

It’s an Avangrid wind farm. It was built by Wanzik in February of 2020. They finished it in May of 2021 with balance of plant and everything else. The project has 111 turbines, and this is the interesting part of this wind farm. 74 of those 111 turbines are on state trust land. The other ones are on private land.

So there is Some SGRE G114s and 76 GE 2. 82127 machines in this wind farm. And to get it built on the state trust land, why they did it and how they did it. They had a bunch of support from U. S. Senator Tom Udall, U. S. Senator Martin Heinrich, a Congresswoman, Deb Highland from New Mexico’s 1st District.

And the governor of New Mexico, Michelle Luan Grisham. And one of the reasons that they pushed so hard to do this was on the state trust land, is it’s going to give a ton of money back to the state. New Mexico classically is a state that doesn’t have a huge tax base. So nice for them to get some income from a resource that they have in spades, which is wind.

So we’re seeing, the Sun Zia project. That’s in New Mexico. So New Mexico is becoming a haven for wind farm developers. There’s hundreds of thousands of acres of state trust land. That’s prime for renewable energy generation. Avon grid took advantage of this partnered up with PNM, which is the power corporation in New Mexico.

Helped PNM hit some other state emissions free goals. They’re well on their way to the 100 percent carbon free by 2040 benchmark that they’ve set with this wind farm in place. And some of the other ones that are moving there. So the Avon grid. La Joya Wind Farm in New Mexico is Torrance County.

Allen Hall: You are the wind farm of the week.

That’s going to do it for this week’s Uptime Wind Energy podcast. Thanks for listening and please give us a five star rating on your podcast platform and subscribe in the shows below to the Uptime Tech News newsletter and check out Rosie’s YouTube channel Engineering with Rosie. We’ll see you here next week on the Uptime Wind Energy podcast.

https://weatherguardwind.com/drone-rover-inspections-ai-crack-monitoring/