What to Know About Blade Quality, w/ Kronborg Consult

Allen Hall: Jacob, welcome to the program.
Jacob Kronborg: Thank you. Thank you for giving me the opportunity to talk about my new company.
Allen Hall: Well, it’s going to be a great discussion, because rarely do you have the chance on the podcast to talk to someone who’s worked as long as you have in blades and composites for a large operator. There’s just a lot of knowledge in your head, so hopefully today we can get some of that out to the world, because some of the things you have seen over the last 15-plus years are applicable right now to a lot of operators and how they’re dealing with OEMs when it comes to blades, blade inspections, and how to manage their blade fleet. There are so many issues, and I think globally everybody has their own little niche as to what their issue is. But we’ll just start there. What are you seeing as some of the real blade issues that owners and operators are fighting with today?
Jacob Kronborg: I think there are some actual problems. Offshore, there are blades in the water. Onshore, there are blades on the ground. And everybody’s wondering why that is happening, because we’ve been designing blades for decades now. Blades still break.
Basically, I would say all the major players, the OEMs, know how to design a blade in the engineering office. They should also know how to manufacture it. Sometimes there’s a little gap between what they thought and whether it can actually be manufactured, but usually that’s not a big issue. I think the issue could be the actual produced quality. Manufacturing of blades is very manual, labor-intensive work, and that comes down to training of personnel.
If you call a plumber or a carpenter, you would expect them to have a certain level of skill and know how to fix your plumbing or whatever you need. There’s no education for somebody to make a blade. That’s always in-house. So you have to get the training in place. You have to work out what the really important things are, where you need somebody with long experience, and what the less important things are, where you can take newcomers and train them. And that’s very often where things go wrong.
Allen Hall: I think that’s very true, because of the way the industry has grown so rapidly over the last 10 to 15 years. 20-meter blades were the thing for quite a while, and then all of a sudden we’re up to 100-meter blades. The scale of that is massive, and the location of factories has changed so much that it becomes very hard to have highly qualified people who have spent years in the factory doing this work. There tends to be a lot of in and out, and blade design and blade manufacturing are not easy. You’ve seen it up close. These 80-meter, 100-meter blades are super complex in terms of how many people are working simultaneously. How complex is it to put a 100-meter blade together? What are we talking about in terms of people, floor size, and scope?
Jacob Kronborg: It’s hundreds of people working in multiple shifts around the clock. A modern blade factory never stops. The big investment is the mold and the factory, so it has to run basically all the time to make a profit. It literally runs 24/7. That means you also have handovers from one shift to the next, and everybody’s always complaining about the night shift. It’s always the night shift that makes the mistake.
There might be 20 or 30 people working at the same time on the mold, taking breaks, coming back, with different skills in play. It is quite complex. People always say, “Oh, we just need to do it like the automotive business.” Well, in the automotive business you have series production. You have robots. You have one station with one operation, and then you move to the next station, where you have an expert in that operation, and then you move to the next. Here you have one big mold, and multiple different things you have to do in the span of maybe 24 hours. That’s roughly what it takes to build one blade, and that’s always been the case: one blade per day, per 24 hours, sometimes a little above, sometimes a little below. That’s usually a good aim to have in a blade factory.
Speaking of automotive, the good factories have actually adopted some of the lean manufacturing principles. How do we train people? How do we make sure the information flow is there? If something goes wrong, how do we handle that? How do we continue to improve? So there are learnings from automotive, but you cannot copy it.
Allen Hall: I want to ask you about the 24-hour cycle and how that plays into blade manufacturing, because it does seem like the blade industry has forever been able to produce one blade a day, and every morning the same person would arrive and do the same level of work. But we’ve crossed over that threshold where some of the blades are now taking longer than 24 hours. So the same person who on day one is laying fabric into the mold and getting things started comes back for day two of that same blade and is working on something else. Now they’re having to pick up a second set of skills, so they’re doing double work, and they can’t lock in on one thing every day. That automotive mentality really left the building as soon as we crossed 80 or 90 meters. Has the length added a level of complexity to how we build blades?
Jacob Kronborg: It has added some complexity, but of course there are also some modern tools. Take the layup of the glass fibers. In the old days you would just look at a list: okay, I have to put this one here. Now they have laser projection, so you know this layer number goes in this position, and it actually marks quite exactly where it has to be. So as long as you know the system, how the laser projector works, and that this number means I take this mat and not that mat, you don’t necessarily have to do exactly the same thing day after day or on the next shift. That helps a lot.
Allen Hall: Well, the thickness of the composite structures has changed dramatically. I’ve seen some of these longer blades now, and it’s incredible how thick the composite is. And the wrinkles become a problem.
Jacob Kronborg: Yeah. Normally when you say “wrinkle,” you think of a wrinkle in your skin. In reality, a wrinkle is just a fold, a bump in one of the layers among multiple layers. It might just be that the direction of the fiber is off by five or ten degrees and then bumps back again. That is the nature of a wrinkle. If it’s in the wrong place, with high loads, it doesn’t have to be a huge fold. It could actually be quite a minor thing if you look at it from the outside.
Allen Hall: It’s sort of like the old princess in the story, the princess who could detect a pea way down under all those mattresses. A blade is just like that. Every wrinkle, it’s going to notice at some point in its lifetime.
Jacob Kronborg: Princess blades. I think that’s a good one.
Allen Hall: Yes. Well, because you’ve seen so much inside these blade factories, I know a lot of operators would really like to go see a blade factory, be inside, and get a sense of it.
Jacob Kronborg: I think that’s one of the keys as an operator. Do they actually have good work ethics? Do they do good quality? How do they handle training and making sure quality is okay? I guess that goes for any workshop. I’ve also been in production of composites and smelled my share of styrene, and you get a feeling for what’s going on in a workshop. Are they able to produce a quality product in this workshop?
Allen Hall: Let me ask this question, because I think it’s a very good insight: when you get to a factory, what should you look for? Is the first pass just cleanliness and organization? If there are people wandering around like they don’t have any instructions, that’s a red flag. But also cleanliness. It does seem like some blade facilities are sometimes so busy that they haven’t done a good job of cleaning.
Jacob Kronborg: Yeah, and that again comes from automotive: 5S, cleaning up your workplace, keeping things in order. It’s a mentality. When you have so many people working in the same space, you should do that continuously to get good quality. You have to have some kind of order. I would not point my camera around my own office. It’s a mess. But that’s only me working here, and I know my way around my own mess. If my wife comes in and asks where she can find this or that, I know where it is, but I’m the only one working here, so it’s okay. It’s my mess. But if you’re in a workshop with maybe 20, 30, 50 people working on each shift, 24/7, and you don’t know where to find stuff and you just throw things around, then it’s impossible to keep good quality.
Allen Hall: Well, a defect found today on an offshore blade, the cost to repair it is just so much higher than on a 30- or 40-meter blade. Even the complexity of it, and the amount of downtime from an operator standpoint. Say I have a wrinkle in the blade and it shows up once it’s installed and in service, offshore. Those are not easy problems to solve. It would be a lot easier onshore, of course, but offshore adds so much more complexity, and the scale of the blade adds to the cost of the repair and the downtime. Money-wise, it gets to be a lot. Shouldn’t the emphasis right now be on the factory? We can catch it after the factory, but honestly, if we caught it in the blade assembly process, we would save ourselves millions of dollars easily.
Jacob Kronborg: Because that’s where you can make a difference. In modern blade manufacturing, you make mistakes. It’s difficult not to. That’s even part of manufacturing. But you produce them in the factory and you fix them in the factory. You have controlled areas. You can scan it afterwards. And when you do a repair, you do it in proper quality. If you do a composite repair, that’s one of the beauties of composites: you can actually restore the structural integrity back to original if you do it properly.
Allen Hall: I think that’s an interesting point, because a lot of buyers of blades think the blade is perfect, that they should be getting perfect blades, and that’s really hard to do. But catching it in the factory is an order of magnitude less expensive than outside the factory, because they can control it and rebuild the blade. They can actually bring it back to the original strength.
Jacob Kronborg: That’s the whole key. There might be a little exception with carbon fiber. It still seems to be really difficult to repair in good quality. That might be a specialty we’ll see more of in the future, if we have issues with carbon fiber.
Allen Hall: That gets me to the next question. We understand fiberglass really well, we’ve repaired it for a couple of generations now, and we know how to do that. We know wrinkles cause problems. There are other things that can be an issue, like bond lines, but we generally understand what those are. Then there’s the carbon fiber complexity, particularly as we get to these carbon fiber planks that are installed into the blade. Those are a slightly different type of performance. You use carbon because it’s much stronger than fiberglass. It cuts weight out of the blade and brings a lot of savings and improvements, but dealing with damaged carbon is much more complex. Are you looking for a higher level of skilled technician to do that work? What are you looking for exactly when you’re talking to an OEM about a carbon blade?
Jacob Kronborg: In my experience, we’re actually not yet at the level where we could even look for a technician.
Allen Hall: Oh, wow. Okay.
Jacob Kronborg: It’s simply so complex. Some of the carbon fibers they use are produced as pultrusions, with very straight fibers, and they use them to the limit. Carbon fiber is more sensitive to wrinkles or abnormalities in these planks than glass fiber. That means you really want perfectly laid-up fibers. So if you have a damaged carbon fiber plank, you’d have to cut it out and put something in that has the same quality, seamlessly. It is also still relatively new to have carbon fiber in the blades. I think the oldest blades with pultrusions are not even five years old yet. There have been other versions in the past, different attempts at using carbon fiber, but the new ones with pultrusions don’t have that many hours on the meter yet.
Allen Hall: That’s a very good point. The industry moved towards carbon a while ago, or at least dabbled with it in different forms, to cut some of the cost out. Pultrusions were the answer because of the strength and the consistency. You weren’t hand-building it in a factory. You had a formed piece of carbon, so to speak, that would just be installed into the blade. But there are a number of issues with carbon pultrusions that I think operators need to be aware of. The straightness of the fiber is the easy one. If there are problems there, that’s going to cause a reduction in strength. But bonding the pultrusion into the greater blade has occasionally been an issue too. How are operators supposed to detect and monitor that? If you have a fiberglass blade mentality, you’re not thinking something inside your blade is going to come loose or have a bonding issue. It’s a different world now.
Jacob Kronborg: At least in the ones I’ve seen, they actually co-cure it. You make a half beam in carbon fiber that you infuse with glass fiber around it to make a smooth, nice geometry that will fit the big blade. Then you take that precast beam, put it into the mold, put the dry glass on top of it, and infuse the whole blade. So you don’t have traditional bond lines. You don’t bond it in, so you won’t have the traditional issues with bond lines, like too little or too much adhesive, because it will be completely embedded. At least so far, I haven’t seen that being an issue.
Allen Hall: We’ve seen some carbon pultrusions that have a few issues, and sometimes it has to do with peel ply not being removed when it should have been, which seems like a training issue. But again, speed matters in a factory.
Jacob Kronborg: That’s back to the quality of the work. But there’s a lot of focus on the carbon fiber, which is also a very expensive part of the blade. You don’t want to scrap it. So that’s maybe where the most skilled workers are positioned and where most of the quality control is focused.
Allen Hall: I think that’s an interesting point, because nothing in blades is really consistent from generation to generation. There’s so much new technology and so many new approaches being used. We’re putting new materials, carbon and all kinds of advanced technologies, into blades to make them lighter and longer. That adds to the things you should know as an operator, which is really difficult. When you were on the operator side, was every discussion with the OEM about the next generation of blades almost like starting over? “Here, Jacob, here’s our new blade.” “All right, let me go figure out what’s inside it and see if I’m okay with it.” Or is it easier than that?
Jacob Kronborg: I think in a lot of things, it is the same old thing. It’s the same type of glass fiber. Yes, now carbon is being used, but it’s still embedded inside the blade, and you would expect that you don’t have to repair the carbon.
Allen Hall: Right.
Jacob Kronborg: So a lot of it is the same. But then of course there are changes of materials and changes of details in the blade, where you have to ask how you actually maintain that. Classic glass fiber repairs can be done in situ on the blade, hanging in a rope or something like that. It’s known technology. Carbon fiber is a little more difficult. And then it’s more about new materials coming in that you haven’t used before.
Allen Hall: What about recyclable resins? That’s a new area a lot of manufacturers are moving into, or have been in for a year or two or three. What should we be watching out for with recyclable resins?
Jacob Kronborg: If you look at the structural properties, the fatigue properties, everything, it’s the same. Some really clever chemists figured out how to open up some of the bonding between the molecules.
Allen Hall: Really?
Jacob Kronborg: Yes, some clever chemists nailed it and designed a molecule that you can dissolve with a hot acid. And I’m not too worried that that would happen during operation.
Allen Hall: Really? Okay. Because I’ve heard operators be reluctant to switch. You get used to seeing certain things in certain blades, and any change is scary.
Jacob Kronborg: No, I think the reluctance is the cost. Up till now, the recyclable epoxy has been made in small batches, and that just means the price is higher. In the chemical industry, it makes a difference if you make one ton at a time or 50 tons. But a recyclable blade also has value for the operator. They can say, “We are green, and we can recycle our blades.” I think most operators have said they will recycle, and they haven’t figured out how to do it. This is a fast track to say, “Now we have a resin that can be dissolved, and now we are extra green.”
Allen Hall: Right. At the end of the day, the project has to work financially. And speaking of money, one of the big areas where a lot of money is spent is warranty claims and the definition of serial defect. I think a lot of operators struggle with what a serial defect is and how to get there. What are some of the precursors here? If I’m buying a new blade that doesn’t have a lot of service history, I feel like I’m taking a lot of risk. Should I be a little more proactive on inspections and maybe file some warranty claims early, just so I can understand it and set that framework up with the OEM, like, “Hey, we need to negotiate through this problem financially”? How does that work with the new blades coming out today? Is there an established process for operators, or is everybody just trying to figure this out?
Jacob Kronborg: I think every operator has their own idea of it. But ultimately you could end up in court, fighting over whether this is a warranty defect or normal business you could anticipate as normal maintenance. Which is it? And there you need evidence. Evidence, evidence. And that means data. When did this happen? How fast did it develop? How many turbines does it affect?
Allen Hall: Right.
Jacob Kronborg: You will never get all your costs redeemed as an operator. So it’s also a matter of being cautious and saying, okay, this is a new product, there are some new technologies, how do we find out if somebody overlooked something? It could be a new factory. It could be a whole new design. Maybe you should look a little extra at some of these blades and make sure that in good time before the end of warranty, which is typically year five, you have enough data.
Allen Hall: I want to ask a very specific question, because you’re raising a really good area I think we should dig into for a moment. What about leading edge erosion? At what point does an operator say this leading edge erosion is too much and it’s a serial defect, versus normal wear and tear based on where the turbine is installed? If I’m in those first two to five years, where is the line being drawn today that says this design is not going to work out and we need to get to a financial conclusion?
Jacob Kronborg: If you buy a turbine today, I would expect it not to have erosion within five years. Maybe a very, very small number of the blades have some kind of wear, but not a serial defect where things fall off or start to erode super fast. There are systems out there that can protect against leading edge erosion, and the OEMs know them best. So they should be able to offer a system they can warranty, at least for five years, so you will not have erosion in that time period.
Allen Hall: Let me ask you this. Should anybody be buying a blade today without specialized leading edge erosion coating?
Jacob Kronborg: Unless you are in a very, very dry, non-dusty area, no. But you should of course ask how much you actually need, depending on where you put it up. There are areas with 2,000 millimeters of rain per year, and there you would say, okay, I’ll probably have some rain erosion. There are areas with a lot of dust and sand, where I’ll probably need a really good system. But there are also areas where it’s very dry and you don’t really have a lot of rain, so maybe I need less.
Allen Hall: I agree with that.
Jacob Kronborg: And then the debate should be more about whether you need 10, 20, or 30 meters of leading edge protection, and what type of protection you need. Maybe you need something super strong, and probably also expensive, on the first five to ten meters, and then something else for the rest. Or you need to know up front: we only put ten meters on, and we expect that in year ten we will have to go out and do a maintenance campaign. And you could do that with these materials.
Allen Hall: And that relationship is fine. You can have that discussion about the cost of leading edge protection, the varieties of leading edge protection, and how much leading edge protection. But I think the discussions about no leading edge protection, or just a little bit, are not really helpful, because at the end of the day you’re going to have leading edge erosion. Most places I’ve seen on this planet have had some leading edge erosion. You’re going to need something, so why not work with the OEM to find out what that is? And Jacob, you’re right. I think the OEMs know pretty well which materials work and which don’t, and what they like and don’t like, and you’re going to have to pay a little more for them. But isn’t it worth paying a little bit up front to extend the lifetime of the blades?
Jacob Kronborg: It’s so much more expensive if you have to do it hanging in a rope or take the blades down. I’ve also been involved in projects where it was so expensive that we figured out it was actually better to take the blades down, put them onshore, repair them in a factory where you could do it in a controlled way with good quality, and then put the blades up again. Of course, you’d have a set to swap, so the downtime is only what it takes to swap three blades, which is very little. Then you save the blades, repair them, and maintain them. But it’s still quite expensive, and you don’t want to do that.
Allen Hall: Yeah, that’s it. Even five years ago, leading edge erosion materials were still being evaluated. I think the industry has done a tremendous job in the last couple of years of understanding the problem better, with test facilities available to every operator and OEM, and the quality of the products being delivered today is so much better than five years ago. There’s no reason to have blades with leading edge erosion coming out of the factory. That shouldn’t be a thing. But it is a financial decision.
Jacob Kronborg: I’d say today all the series OEMs have it built in carefully, so the warranted power curve you buy includes LEP, and you are outside that discussion. And there are some tweaks to the profile, tweaks to how you finish the edges of the LEP, and things like that, that will at least minimize the AEP loss, plus putting extra VGs on. That’s another set of tricks you can play to at least reduce the effect of leading edge erosion.
Allen Hall: Well, this is why you want to call Jacob and talk to him about these issues, from what’s happening in the factory, to what’s happening with the OEM on the design side, to how the blades are performing out in the field once they’ve left the factory. You need experts on your side. These are hundreds of millions of dollars, hundreds of millions of euros, billions of euros being thrown around on these projects. You need expertise, and you need to get it early in the process. Jacob has set up his own consulting firm, and I know a lot of people are reaching out to you because of your experience with offshore wind turbines, which is hard to get. It’s hard to find people like you walking around. How do people get hold of you and start working with you, Jacob?
Jacob Kronborg: They can find me on LinkedIn, and they can also find Kronborg Consult on LinkedIn, or go to my website, kronborgconsult.com.
Allen Hall: We’ll put the website in the show notes. Jacob, it’s been so great to talk to you. We have to have you back, because there’s so much knowledge in your head that I need to extract. The industry and operators around the world need your expertise. They need to understand what’s happening in the blade factory, what’s happening in the field, and what makes sense financially. It’s so good to talk to experts like you. Jacob, thank you so much for being on the podcast. I’d love to have you back.