How CNC Onsite Cuts Repair Delays for US Operators

Welcome to Uptime Spotlight, shining light on wind energy’s brightest innovators. This is the progress powering tomorrow.

Allen Hall: Søren, welcome back to the program.

Søren Kellenberger: Thank you very much, Allen. It’s a pleasure.

Allen Hall: Well, you have some really exciting news in the United States, because CNC Onsite has been growing tremendously. You’re now worldwide, and you’ve established a presence in the United States.

Søren Kellenberger: That’s correct. We have been working in the US for quite some time, but we see that the demand is definitely there. The North American market has more than 100,000 turbines above one megawatt. So of course it’s an interesting market, and we think that now is a good time for us to make that move and establish ourselves in the US. So we are really looking forward to that.

Allen Hall: So you’ve seen a lot of growth in the company over the last couple of years as operators worldwide realize they have problems that they can’t solve, and that will only be solved by precision machining, which is what CNC Onsite does. Where have you been doing work lately? I’ve seen things on LinkedIn from Asia, but I know you guys are pretty busy right now.

Søren Kellenberger: As you started saying, we are basically working globally. So we have had technicians in Australia and Japan, and in other Asian countries as well. Our main market is still Europe, where we also started, and we are widely spread across Europe. But the US, including Canada, is becoming a very interesting and important market for us.

Maybe the best way to explain it is that even though we are based in Denmark and working out of Denmark, 90% of what we do is outside of Denmark. And once you travel — whether you travel to Finland or if you travel to somewhere in the US — it doesn’t make a huge difference. Our technicians are away from home and family anyway, so it’s just the language that is a bit different.

Allen Hall: What brings people to your business? What are they calling about today? I know the yaw ring was the number one item. That’s what I remember as being, when you were getting started, the yaw ring with the teeth being broken. Is that still your number one request, or has that changed?

Søren Kellenberger: It is definitely often a door opener, because everyone can relate to that. So it brings a lot of interest to the company. And then when we start talking to customers about other wear issues in the turbine, it often turns out that we start to talk about worn sliding surfaces on the yaw ring, or it can be worn out rotor lock holes, or wear and tear to generator shafts or main shafts.

There’s a whole lot of mechanical things that obviously get worn when you have a turbine running for so many years and hours. And then we just try to see if we can come up with smart ways to fix it uptower instead of deploying cranes and doing large disassemblies.

Allen Hall: That’s the real problem. I think a lot of operators realize that they have worn assemblies, but they don’t really know of any way to address it, so they just live with it, or they try to accommodate as best they can.

With the yaw ring teeth, the way I had seen that until I talked to you was they would get a guy with a welder up there and they’d try to weld these teeth back on. But that’s really a temporary fix and it’s not a long-term answer. You’re doing something that is actually better than what the OEM installed to begin with, right?

Søren Kellenberger: Yeah. Instead of welding — and I won’t say that welding doesn’t have its place, because for smaller repairs welding can be a good fix as well. We do also work with welding companies for certain smaller repairs. But imagine you are up in a very confined space, and when you have to start rebuilding a tooth by welding, the more of the tooth that is missing, the more you need.

One thing is the welding, but grinding it back to shape again is also a manual process. It’s obvious that the more of a tooth that is missing, the more difficult it is to get that actual shape of the tooth afterwards by grinding. So what we do is that we machine away the damaged teeth, and then we bring prefabricated segments that have the actual shape of the original tooth, so that we get very good contact between the yaw gear and the yaw ring after the repair.

And I think that is probably the main reason why welding is sometimes not lasting as long as when you have to rebuild a full tooth and then manually grind it. It’s very difficult to get that well-defined contact surface. So you get some peak points where you have higher loads, and that will just fatigue those repairs over time, where we have better chances of getting that very good fix. We get very accurate machining because we use CNC-controlled machining, so our repair is within a few hundredths of a millimeter.

Allen Hall: Yeah, absolutely. Well, the other approach besides welding and your approach — and this is what I have seen more recently — is the whole nacelle comes off. So the rotor pops off, they lift the nacelle off with a crane, so they have to have a crane on site, and that’s really expensive. And they set the nacelle down, and then they have to go in and physically remove the yaw ring, or these sections of it, and replace it.

That is extremely expensive and time-consuming compared to what you’re describing. What is the delta in cost, roughly?

Søren Kellenberger: That is very difficult to say, because it depends how fast you can actually get a crane out. And I guess there is also, in the US, a huge spread on what the deployment cost is for a crane, how close you are to a local crane center, and what the deployment cost is.

Then it also depends on your access roads. If the turbine has been out there for 5 or 10 years, maybe your access roads are unfit for the crane anymore. Your hard stand or the crane pad is not in sufficiently good condition to support your crane. So there can be a lot of extra costs that are adding up — not just the rental of the crane, but the overall cost of preparing it and the whole project management of getting that set up. Then you need your support fixtures to place on the nacelle. You need to get your plates down.

So it’s a lot of logistics, and just preparing the turbine for disassembly is probably about a week. And then you have all the craning and replacing and putting it back together. So if you say you could have everything ready, your turbine is down for, I would guess, two to three weeks. But in reality, the downtime before you can get a crane out there is much, much longer.

And we can basically — we have repairs, spare parts, we have them in stock, so we can basically move out the day after we get the call and get your turbine up and running. Installing a segment, which repairs up to six teeth, takes about a day for us to do. Then we need a day to get up in the turbine and get ready, and a day to get back down. So what takes from several weeks to potentially months before you’re up and running again, we can do in just about a week.

Allen Hall: In the United States, one of the things I hear a lot of times, particularly dealing with cranes, is that procurement, when they want to buy a fix of some sort, doesn’t necessarily consider the cost of the crane, the downtime, all those other factors. Obviously, the engineers on the backside are thinking about that, but when it comes to decisions for your technology, that doesn’t always get into the mix. Is that a worldwide problem or an American problem? Because that feels like an American problem.

Søren Kellenberger: No, it’s not. I think it’s a worldwide problem.

I think what makes the difference is probably that the turbines are typically a bit older in Europe. So in that way, we are a few years ahead in that life cycle. Procurement and also technical organizations have maybe become a bit more used to including that thinking, where this is still a relatively new issue in the US because your fleet is newer. So I think it will change in the coming years, as it will develop naturally as it has done in Europe.

But it’s also sometimes the structure of the company — that you have a certain organization that takes care of the main component replacement, and then you have a part of the organization that takes care of day-to-day service. And now you don’t need to replace your main component, but the work we do is on the main component, so it sort of falls in between those two areas of responsibility in the company. And that can also sometimes be challenging.

Allen Hall: So it’s not only that you have to get a crane to do a lot of these repairs that you can now do with CNC Onsite, right, internal to the nacelle — but there’s also weather, right?

I think the one thing we don’t consider a lot of times, or have kind of preconceived notions about, is that the weather’s, like, 30% downtime. Yes, if you’re working with a crane and you’re trying to lift heavy objects, there will be 30% downtime because of weather. But when you’re doing the repair in the nacelle, what limitations do you have on weather? Is it a much wider window than a crane would have?

Søren Kellenberger: Absolutely. We can work up to much higher wind speeds — basically the same wind speeds that you can do ordinary maintenance work within — because our machine is always mounted on the component that we are machining. So even if the nacelle is moving a little bit in the wind, we move with the component, and it doesn’t really affect our tolerances. So yes, we can work in much higher wind speeds.

And another factor that people often do not think about is also the risk of damage to the component when you lift it. You have, of course, a safety risk every time you lift a main component. So that we completely eliminate.

When we speak of the yaw ring, where you need to take down other components — every time you lift a blade, blades are extremely fragile. So every time you lift a blade, you have a huge risk of imposing a damage to that blade. Those follow-on costs, or the risk of having these damages which are actually not related to the original issue, are often also not considered. But they can be quite substantial risks of damaging some of the other components that were fine to begin with. Because you needed to take them down to get another repair done, you end up damaging some of those components, or even hurting people during lifting, which is, of course, the worst thing you can imagine.

Allen Hall: Right. When you lift in the nacelle, there are a lot of components that are not supposed to be touched that you’re touching, for sure. Electrical, hydraulic, everything, and all the mechanical, all the bolts and all that. When you lift in the nacelle, it is a major ordeal, and I don’t recommend that you do that anymore, especially since CNC Onsite can come up and do the work internally and get it done.

I know one of the questions in the United States, and probably Australia, because I talk to Australians quite a bit, is the delay, right? So they may see they have a rotor lock that’s oblong or isn’t where it should be, or they have a yaw gear tooth that’s broken, and they’re trying to weigh that “how long do I wait” thing, which is usually tied to a crane, right? So I can’t get a crane tomorrow. I can get a crane in six months, or if it’s late in the season, it could be longer. It could be eight months before I can get a crane on site. There are costs associated with that delay. What are we talking about in terms of sort of raw numbers and costs on waiting for some of these repairs that can now be done with CNC Onsite?

Søren Kellenberger: That goes up, obviously, with the size of the turbine, because of your lost production. If you have your turbine down, then it’s lost production, and that very quickly becomes a substantial number. Again, with increased sizes of turbines — and we put up larger and larger turbines — they produce more and more megawatts. But then, of course, the loss is also bigger when they are standing still. So that number gets really high. It depends on your PPAs and stuff like that, how much that number really is. So that can also vary quite a bit.

And you often also see that with the smaller turbines in the kilowatt size, or maybe even the small one megawatt turbines, you don’t need as large cranes as you do with a three, four, five megawatt turbine. So the business case is different depending on your turbine model and where it’s located and your PPAs and stuff like that. But I would say once we are above one megawatt, I’ve never seen that the business case is not better for repairing on site than deploying a crane.

Allen Hall: Oh, that makes total sense. And I think as more operators become aware of CNC Onsite — you know, the thing about America is it takes a little bit of time for everybody to realize that there’s another opportunity. And now with the office in Houston, that’s going to really expand your market. I guarantee you it will do that.

I want to talk about the other solution you have, which is for the root inserts into the blades, and that’s a more complicated repair, because in this case you are taking the blades down. But you’ve been doing that work globally, and you’ve done a lot of them at this point. What is the outcome of that? How is that system working, where you essentially drill into the root, remove that old insert, and put in a more advanced insert from We4Ce?

Søren Kellenberger: As you describe it, those are basically the simple process steps, right? You need to bring the blade down, because we do actually drill out the old bushing and, maybe also quite importantly, the affected material around the bushing. So the problem is basically not the bushing itself, but it’s the bonding between the blade root, the glass fiber, and the bushing.

And in that layer where you have a resin — typically polyester or epoxy — we’ve seen the most issues on polyester-based blades. But that one is based on epoxy, and it gives some better bonding conditions to the blade.

I’m not a blade specialist, but anyway, that’s overall the process of that repair. And it’s correct, we’ve done a major project in Asia, installing more than 1,000 bushings, and those turbines are up and running full speed again. And it’s also opened up a lot of new markets in Asia, and we do also see quite a lot of interest for the solution in the US.

So I think it’s a very good way of repairing blades instead of completely scrapping and installing new blades. The cost of a new blade is high, and you have lead time. Blades are not so easy to recycle, so if it’s otherwise in good condition, I think it’s a very good idea to repair rather than replace. Which is basically what we do in general with the components that we can repair instead of replacing them with new ones.

Allen Hall: And at what point should somebody call you for the root insert repair? I know there’s a lot of discussion in the industry that if I see a gap of one millimeter, that usually is an indicator that this situation is not going to get better, and then you’re really monitoring it. The number I hear more recently is that if it’s a three millimeter gap between the root of the blade and the pitch bearing, you better be doing something about it pretty quickly. Are you hearing the same things?

Søren Kellenberger: Yes. Those numbers are quite good as a rule of thumb. I think when you get to three millimeters or above, you’re about at that limit where you should stop your turbine or you risk the blade coming off.

Again, I’m not a blade expert, and I’m not a root cause specialist in these things. We are the machining expert. But this is as with any problem almost, right? The sooner you detect it, the better. And there are different solutions out there to monitor these gaps. It doesn’t mean that you need to repair them tomorrow, but if you start monitoring, then you can also start planning.

And if you want to bring down your repair cost, planning it is the best way to do it. It is, of course, more expensive — you will have more downtime if you call us the day your turbine is down, rather than if you start monitoring and you can say, “Okay, let’s plan for this repair in three months,” then we can get everything set up. You minimize your downtime and everything. It’s much cheaper compared to if you have a really an emergency stop and you need to get it fixed now. Then the cost will go up.

Allen Hall: Yeah. It makes sense for operators listening to this podcast today. If you have that one millimeter gap between the root and the pitch bearing, you better be calling Søren and getting something scheduled, because the end is near. You need to be working on that.

Now, one of the things I want to get more information on is how you’re getting equipment up into the nacelle. All your machines are high accuracy machines where you’re doing really, really fine work. When I see fine machining done, it’s usually in a factory with concrete pads and big machines, and a lot of lasering in on making sure that there’s submicron tolerances, and you’re like, “Wow, this is really complicated.” But you’ve taken all that complexity and you’re putting it uptower. How does that work? What’s all involved in moving a highly complex machine uptower and getting it to do fine machining?

Søren Kellenberger: All our machines are designed to be able to be lifted with the internal crane of that specific nacelle.

You can say that a lot of our machines need to be adapted a bit to each specific turbine platform. Sometimes — especially for the yaw ring repair solution — it’s more or less a dedicated machine for the specific platform, because the available space is so limited that we need to move gears and motors around to fit it. For other machines, it’s a bit more simple. It’s maybe just a different mounting plate.

But they can all be broken down to smaller sub-assemblies, weighing, depending on the crane capacity, from maybe 100 to 200, 300, 400 kilos, depending on what crane capacity you have. And then you get these sub-assemblies up in a nacelle and you move them to that dedicated area where you need to do your machining, and then we assemble the machine uptower.

I don’t know if I make it sound easy or complicated, but anyway, it’s what we do on a daily basis. So we are really specialized in on-site machining. It’s what we’ve been doing since the company was founded. Our engineers are used to that mindset, where you need to be able to take a machine apart and then reassemble it and make sure you still have those micron, very fine tolerances, and our technicians are used to operating them.

And I think that is a little bit where the magic is happening — that we have technicians running and operating those machines in the field, and they feed back to our engineers, so they get that hands-on experience into the design as well. And then you do also need some skilled technicians. So the craftsmanship is important. You can’t just buy our machine and then ask anybody to run it. It’s not like pushing a button on a workshop CNC machine, where somebody can do the preprogramming, and then you just install your component and press the go button, and it will make you a nice component. Yeah, we are not quite there yet, so we still need skilled people.

Allen Hall: Well, I want to talk about the rotor lock issue, because you raised this with me at one of the conferences and I didn’t know much about it. On gearboxes, obviously there has to be a way to lock the gearbox so that you can work on it. However, some of the rotor locks that go in, they kind of get worn over time, because there have been so many repairs done to them, honestly. So you have this rotor lock system, which is a little wobbly on the low speed side. But if you’re working on the high speed side with a rotor lock on the low speed side, there still can be a lot of movement in the mechanism, which would prevent you from actually working safely. Can you describe that problem a little bit, and then what that solution looks like?

Søren Kellenberger: You can say a rotor lock has two main purposes, right? It needs to make sure that the turbine isn’t spinning, that the safety is there if you need to either go into the hub, or if you need to work on some of the rotating equipment.

But it also has a practical purpose, as you say, Allen. The rotor lock is basically a hole and a pin, and the pin goes into the hole and locks the rotor, prevents it from spinning. And when you use that pin-hole system over time, then at some point you risk that the hole gets worn, and it oblongs, and then you can start to have a bit of movement.

It might still be okay from a safety perspective in terms of maybe going into the hub, because it’s a very little movement. If you’re not used to it, it’s probably not that nice a feeling, that movement, but it won’t kill you. But if you do work on the high speed side of things, even small movements in front become relatively large movements on the high speed side due to that gear ratio.

So it can make it dangerous in terms of you getting your fingers squeezed or whatever when you are working on it. But it can also make certain maintenance tasks impossible, because you need to maybe assemble a spline shaft or something like that, and if that is just moving from side to side all the time, you cannot do your repair.

So it has different purposes, and it’s quite an easy fix actually, once you get up there and you have the right equipment for it. So we typically just machine out the hole and then install a bushing to recreate a good fit between the hole and the pin. There is a bit more to it, because you need to align the original center with the center of the new bushing, and that’s where the accuracy and the experience come into play. But we have that, and know how to do it. So it’s actually, for us, quite an easy fix.

Allen Hall: It sounds so easy the way you describe it, but I know there’s a lot of complexity in engineering on the backside of that. And that ties into customers that have used CNC Onsite. I talked to some of them — obviously I don’t know all of them, but the ones I talk to say it’s amazing. What you guys do has saved them so much time and money, and actually has made them money. What do you hear from your customers after you complete some of these more complex jobs?

Søren Kellenberger: To be honest, I’m quite proud about the feedback we get, and that is probably what makes me really happy to go to work and see what our technicians deliver out in the field on an everyday basis.

Some of the things we are handling, it’s not always that the customers want the world to know about it, so it can sometimes be a little bit difficult to share what we are actually doing. But at least we can share internally when we get good feedback from customers. So whenever I have the chance and I get some feedback from a customer — we always appreciate our customers taking their time and giving feedback to our technicians, because they are the ones making it happen out there every day.

And the growth that we’ve been able to have wouldn’t have been possible if we didn’t deliver high quality. I’m actually quite proud that we’ve been able to go from 15 technicians to 65 in four years, and we have basically no warranty cases. We have no complaints. We have a lot of happy customers.

And working in Denmark, we don’t like to brag, you know? We like to underplay, so it’s difficult actually for me to say this. If you have spoken to Danes, you may have heard about Janteloven. And we live very much by that. But I am actually really, really proud about what we’ve been able to accomplish — even though we’ve been growing so much, that we’ve been able to onboard that many people and still deliver that high quality every day. That is something I’m really proud about.

Allen Hall: Well, and now with your move to the United States, which is a major business move, what do you expect to happen in the United States over the next 12 months? I’m sure as soon as they hear this podcast, there’s going to be a lot of emails and phone calls coming in. So if you’re listening in the United States, you better get in line when you hear this podcast. But what do you see coming up over the next couple of months, Søren?

Søren Kellenberger: I know we will have quite a lot of work coming up in the US, and that is also why we have chosen to establish a company over there. We have been working over the last years in the US quite regularly, but we need to be more present.

It’s a huge market. People often tend to look at it as a country, and you are one country, but when you look at the sheer size, it’s like looking at Europe and America. And we always look at single countries in Europe — at least we do. I don’t know how it is from your side. And then we look at the US as one country, but it’s so huge, and of course we need to be present in the US. It’s a huge wind market.

The number of turbines that are reaching a certain level of years in operation is increasing. So I think that further underlines or justifies that we should really establish ourselves in the US. We have also been over for the last couple of years at different conferences and fairs to see if there is an interest, and that has also backed our decision to establish ourselves in the US. So I’m very much looking forward to it.

We will start by having Danish technicians traveling to the US the first couple of months. It’s our plan to have them then relocate to the US, and as soon as we have a base of, I don’t know, three to five technicians in the US, we will start bringing on local technicians and get them trained in our technology, because we want to base our American office on American workers. But we need to bring over the technology and get them trained, and so I’m quite looking forward to that.

Allen Hall: You’re going to have a great time in the United States for sure. You will be busy. How do operators get ahold of you, Søren, and get ahold of CNC Onsite worldwide? A lot of people in Australia listen to this, and in India listen to this podcast, and all over Europe. So how do they connect with you?

Søren Kellenberger: It’s very easy to go to our webpage. We’ve actually just changed to an international webpage, to also underline that we are a global company, so it’s now cnc-onsite.com. And there we have all our contact information. So we have my direct email, phone number, and there’s also the CNC Onsite contact information.

So I think our webpage is a good place. And we are also on LinkedIn, so you can follow our company there. We try to get all the newer stuff out on LinkedIn, to keep people updated if they are interested.

Allen Hall: Yeah. So you should check out CNC Onsite’s LinkedIn page, where you can see some of these repairs being done and get a better sense of what is actually happening and how complex these machines are to do these wonderful repairs.

Søren, it’s so great to have you back on the podcast. I love having you. You’ve got to come back soon, and next time you’re in the United States, we’ve got to connect up, because I want to see this new facility.

Søren Kellenberger: Absolutely. I’ll give you a coffee.

Allen Hall: Thanks, Søren.