S2 Episode 4 - Development of the Kern Fortis HD

Welcome to the Kern Competence Podcast. A podcast by Kern about all things, Kern and precision. Today I'm joined by Christian Meyer. Christian is working in our development. It's a rare thing to get one of our developers to join me for a podcast. So Christian, a very warm welcome to you. Hey Marvin. Hi you. I'm really good. It's pretty late in the day. And you look a bit tired because, well, we all know there is a big bird in the room. Okay, not in this room, but I'm also quite naked already. Exhibition preparation is, is towering at us. Yeah, yeah, Marvin, you're right. I am a little bit tired, but I try my best. Okay, very good. I've known you for a couple of years now, but our dear listeners don't know you yet. So I would ask you to kindly introduce a little bit about yourself. Where do you come from? Yeah, I introduced myself. Yeah, my name is Christian Meyer. Yeah, I'm born in Munich. Yeah, I went to school and I'm later. We changed the town. We went next to First and Feldbrook. And what's in the area? Also in the area. And later we changed the, yeah, the south of the Lake Stambeck, Stambeck, which is already approaching up up the area and quite close to Can. Yeah. And yeah, right now I live in Crossweil. It's a town 10 kilometers from here. So. Tell us a little bit about your professional or your academic formation. Where did you learn? Maybe it's a little bit funny. After school, I started as a carpenter. So I built tables and chairs and kitchens and stuff like that. You an apprenticeship as a carpenter, hospital, yes. Okay. Yeah, I did. I finished it. And after that I was working, someone as a carpenter. But yeah, for me, so it can't be all in my life. I have to do something different. Then I started studying wood. It's wood technique. It's kind of technology of wood. Technology in wood. How you build the machines for, okay. So some wood. And that we had one, Ausflug, like a tour. A tour with the university to company which develops and builds machines. Or milling machines of wood. No. Okay. For manufacturing, for milling. Which company was it? The year-on-year? It was altimeter. Altimeter. Altimeter. Okay. Yeah. Then I changed my mind. Okay. What are I doing here? Now you're studying wood technologies. And what can I do with that? And what do I have to learn to develop and build and use such machines? Okay. And then I checked how, yeah, I would be ready for this. And so I decided to change from wood technology to mechanically design. Okay. Yeah. You do know if this was like a Marvel movie, this would be your villain origin story, you know? The moment everything transformed. Okay. Yeah. So did you study in Munich? Yeah, I studied in Munich, not at the T.U. At the Hochschule in Munich. And I think I quit studying 2014 and the great idea was Diplom. Still Diplom now. Okay. Special master. Now they would call it Master's nowadays in Germany, but then was Diplom engine. Yeah. Very well renowned in German Diplom engine now. Okay. So I'm working almost 11 years at Kern, micro-technic. So you started studying? Yeah. The Kern was my first company and I started not in the R&D. Okay. I started in the application. Yeah. At the beginning there was a different structure in the applications. So we had a small department, it's called P.A.V. Position application. And there started as a system engineer. And one year later I had changed, changed to the research department with the task to develop the micro-HD. Yeah. So I know that from I started a couple of years after you when HD development was already pretty much finished, but you kind of were the project lead on micro-HD already. Yeah. It's quite the steep rise at Kern from like engineer applications to becoming project lead on the most important machine we've probably ever built. Yeah. Very cool. I think this is a really nice lead over and we probably have to do a second episode at a later date. Talk about the development of the HD. Yeah. You know, the reward for good work, for example, a good podcast is always more work. So you should have expected this. Let us talk a little bit about Forch's HD. Simon and I have already spoken one full episode and we teased it in an episode before that that big new machine is coming for this HD. Internally, we haven't called it Forch's at Kern. Your team and you gave it a project name, a working name. Please tell us the name and explain where it came from. Maybe before giving an explanation for the name, maybe he started a little bit before. Okay. It's okay. Sure. Tell something about how it comes to the project. Maybe. We can also start with it, one yes? Sure. Okay. I mean, Simon and I spoke a lot about how at Kern we pick up customer requirements. So we are talking a lot with our customers. We are actually listening to them. At least I think I do. I'm not sure whether Simon is always listening, but we are listening to the customer and they can give us input and we are also looking at the market. So we came up with this idea of it being a big machine. Now if this was a large mega corporation, you would probably have been given a design control grid and told to get to work. How did it work here? What did the project start? To be honest, it was different. At the beginning we had some meetings. Yeah, but that where we decided, okay, we want to build a bigger machine. And what I got from the meetings is, okay, the question, we want a machine which has the same accuracy like the micro HD for bigger workpieces, maybe like for parts who fit on the UPC clamping system, that's it. Nice. No design control grid, but it wasn't really a problem because. I mean, it sounds nice. It's like a blank canvas you can start. Yeah. Did they require you for it to be high-priced? Yes. It was one of the recon's. Yeah. Okay. This was hot. Requirement, yes. Yeah. And so with the project, Albatross was the internal name. Maybe I explain later why Albatross. It was also the first project at Cannes in which we wanted to introduce a new process for introduce new products in Cannes. It's called NPI process. can you product introduction and we started with this project. So at the beginning I also had to make a plan how could this process be and how is it? So they require a couple of boundary conditions from you, how much work will be done on this? Yeah, how much work? How much money? What are the different faces? What? Especially what milestones? What is really important for the feasibility face and so on? Where are the risks? Did you have to sign one of those fancy NPI contracts? No, I don't know. I think the only engineer they ever had jumped through those hoops where you actually sign a contract with like the management that you're developing something. I wrote the project card at the beginning. The project is called ABC and yeah it's just one picture and you assign it and you sign one of those. With the algorithms project it starts. And hopefully I can tell something about the NPI. We already did an episode with Sebastian about that one. So if all listeners go back to season one they will know a little bit more about that one. I think what would be nice to know is you got this really blank canvas you got told okay it needs to be hydrostatic, it needs to be bigger, you got told like a rough size, it needs to be five x's I think this was never without question. And how do you go on from that moment? Like what is the next step? Like even before setting up that NPI process you probably set down and brainstorm with a couple of people? At the beginning yes. Then we check the market. How are the machines are which are ready for UPC clamping systems, how much? Maybe to define it for listeners UPC that's like the ROA, the large ROA palette. I think it's 320 by 320 but it doesn't mean your part needs to be limited to this one just like our current ITS 184 millimeter pellets, your part can't be much larger but this is the pellet interface for the robot and for the zero point clamping on the machine. So with this information we check the whole market especially our competitors like Reidas, Makino, also group I think because bigger machines had need to have more productivity so group is also. Did you ever consider building a horizontal machine? Yes. We were thinking about this. Okay. So we analyzed the market and after that we were able to get some values how big, just like some performance indicators what the machine needs to be able to do. Which kind of spindle is useful? How many tools? The weight and yeah on and on we were able to write our own design control grid. This is actually really nice. Yeah. Okay. And after we had the design control grid, yeah we talked about with our management. If it's okay if we go in this kind of direction and said yeah maybe a little bit more there. And so in the end we had a complete design control grid which was the base of our whole development. How does one start development there? Do you, for example, start with like a small mock-up of the machine like in solid works and then start adding details or do you get to work on like individual groups of components? So is it like a bottom up or a top down approach? It's a mix. It's a little bit mixed. We knew from our other machines but we have to include to the machine. Like okay we need a spindle, we need a tool change, we need a four and fifth axis and some aggregates for doing something. And so it was the structure from the different machine parts. A component was clear from the beginning. So I made an overview on a white paper pencil and wrote okay we need spindle. Like I said we need the hydrostatic linear axis, we need the linear drives, the weight balancer and so on. And with this I was able to see how long it takes and how much money it costs, what's all included in the project. And the next step was to take the different parts and give it to some people in the company, some employers who were defined before, was in the meetings before the project officially started. How many people do we need? And they all got their task. And at the beginning there was no CID, just write it on papers because we had a lot of papers and sketches. So to decide in which direction do we want to go? For example one step, let's go for the same concept like the micro HD kind of kinematics or should we do something complete different which maybe has more risks. The end we decided. I mean I know that we can't really define like what is your team because most of the developers that worked in your team on this project, they also worked on other projects. But overall how many people do you consider, if you can put a number on it, consider to be your core team that worked on you with you on this. I think at the beginning it's not the core team, we defined I think about 16 team members but over the whole company. So we would also like people from marketing or sales or service. So 16 and the core team think the core makes one electric part, high-grade, one designer, constructor, Andreas Lähnerer. We was very important in the development. Another designer Johannes Whitman. A guy you really know but at this moment he's not in the company anymore. He's now building hydraulic systems for the rains or for excavators. You cannot escape the oil. Yeah, yeah, he had the project for all the tasks to develop spinel and swivel and rotary axis and also an important guy comes later in the project because in the beginning you don't need the software but you have to talk with them that you want to be like that so that they know what they have to which program they have to do. So five, six, maybe six people in the beginning with me. Yeah, really the core team. A larger range of people you could get access to to help you. So for example, I know Christoph Stydmeier from applications development was hugely involved. Things like that. Okay. What are some of the challenges you faced? So I mean, this is a huge project and I don't mean it as a joke because the machine is huge. It's more like this is the, I would probably say this is the biggest machine we've ever built. Yes, larger than the filming nano. It's larger in the bandchens and also in the complexity. Yeah. So maybe talk us through a couple of challenges you faced there. For example, maybe let's talk about the motion system. It's hydrostatics. Yeah, but it's not the hydrostatics from the HD. Obviously you couldn't take those. Yeah, you can't take the same hydrostatic from like the micro HD has. Why? Good question. There would be why. Why can't you just slap the hydrostatics from the HD on a larger machine? Yeah, you have different kind of move masses. You have bigger forces or other forces. Oh, I will listen to this. Can't see Christy on face, but he's currently looking at me like I'm a little bit stupid for asking that question. So the question is what the people not as deep in. Problem is scaling. So you cannot take technology. This would be just like taking a very tiny bicycle tire and putting it on a motorbike. Yeah, it doesn't work. It can't scale. So developing new hydrostatic technology. Yeah, it was at the beginning. We also made a list of all the risk and challenges and to scale the hydrostatics with the linear motion system was a big challenge because we wanted to scale the micro gap hydrostatics, not the standard hydrostatics. Yes. Helicetics with higher gap sizes. I wanted to scale the micro gap sizes. So you have different challenges to get the parts with that high accuracy you need for the small. Because typically for example the flatness of your guideway needs to be within a certain range of your film height. And let's for example say it needs to be 20%. Obviously if your guideway is much smaller this is easier to achieve a good flatness than having the same flatness as a large part. So it was a hard fight with our suppliers to get this. We needed to train them. Okay. Did you actually don't recall? Did you put up a test bench for the hydrostatics? Yeah, you did. Especially for the bearings on the side. We have kind of a amount of bearings in each linear axis. And especially at the side bearings we made a big change compared to the micro HD. There we built a test bench to check the visibility. And yeah in the end we were really happy that the test bench worked. And with some improvements it was able to bring the technology from the test bench to the machine. I would imagine that's quite the exciting moment isn't it because I mean for the test bench you manufacture kind of some mock up guideways and you manufacture the pocket technology and the restrictors and you set up all of the hydraulics and everything but it's all put together and you don't know whether it's working. And then I imagine I wasn't there but I imagine it's kind of like flipping a switch and seeing whether it lifts off or not. Yeah. And is this a real risk? Yes, and did it lift up? It lifted up. Imagine years or what years you did all the calculations in the different parts. How they move, how they bend. And I mean this is not only like simple FEM simulation. This is not like you click it together in Anziz. Are we using Anziz? We're using Anziz. You not just click it together in Anziz and you can calculate the whole model and tells you where this works right. It's more complicated than this. Yeah. Okay. So because you have addition that's one part you have to use Anziz to get the risk a little bit down. But you have the manufacturing torrents is too. For example, there are more things need to be done. And for me the moment where we switched on the aggregate to lift up. I think the first axis was the X axis to lift up the X axis. It was very special and I was excited because I did all the calculations and will it work or not? And after it worked because it worked I was very happy. Okay. I can imagine. Please tell me, I guess you didn't, but please tell me you did like a space flight control base with a countdown and then you said there was lift off. No, you didn't. No. Okay. This is a mischance. Yeah. Okay. I could imagine that this, I mean this was probably like the technological, most advanced single event happening where you were like, oh, this is really a win. But there were some larger component groups you also put together. And I imagine this is also quite difficult to simulate. If you have like just like let's say you have a tool changer and the creeper for the tool changer is easy to simulate. But if you want to simulate the whole tool changer that's probably horrible. Were there any other like larger component groups where like, okay, this is really a challenge and something we did differently to compare to the HD. Maybe let's let's go out of the group. Just go one step before the axis lifted. It was lifted. Just to get the part as part as a part, I mean, for example, a y axis or a set axis in this kind of dimensions. There was a huge risk to manufacture this part. We have many different steps in this aluminum parts. So the axis are still cast aluminum. Yes. And Simon, I think he teased it in the last episode that it's two tons of weight x, y, z about a bit less, but yeah. So maybe for our listeners to get some sense for the dimension because you and I have seen the machine, we've seen the dimension of the y axis. But the y axis body, can you tell us like just roughly what length it is, what width and like how heavy it is? Just round about it's one point for me this long 800 high, 600 wide. And it's about 700 kilos. Okay. That is a massive part for casting aluminum. Yeah. And with the information that it was the second bigger aluminum part, cast, cast, aluminum part can ever design. The first ones were the parts from the micro HD. We did it really well because we could use the first part for the first machine. Nice. Yeah. Like a So this was also a great moment to see the parts. The first time I followed from the moment where we lifted them up. I mean, for me this was always like we talked about your villain origin moment. And mine kinda was the first time scene seeing something. It's something you only envision in your mind became reality. And I can imagine that this feels on a much larger scale. It feels also very gratified when the parts arrive in the factory. And you can see really see them. It must have been really cool. Did you actually see them outside the factory first? Did you go to the casting supplier? Yeah. Oh, it was. It was. We've been at all our suppliers. Okay. Who make the important parts. Okay. Did they ever tell you know where they like you can't cast this or where they always like you? We can do it. It was different. Suppliers who deliver us the parts now. Some of them said, "Are you can't do it like this?" And then we made changes. So that they can do. And there were some who said there was no problem. But with the other side having the opinion from the suppliers said you can't do it, we then believe the others. So we decided to go to the guys who think this is a very valuable moment for suppliers of ours listening to us to here. If you want to sell something to Kurt and you have to tell them it's not possible and suggest a different way, right? I said that wasn't the only reason. I can imagine. Okay. So you told me that for you, the one big moment in the project was to see the first milling of the machine. It's possible. It's not possible. Some, yeah. This time some people told me, "Kürtjean, are you excited?" Do you think it will be able to mill something? To me personally wasn't that special moment because I was sure it would work. Because what happened before? Before we measured the whole parameters from the machine and we checked them. So for me it was clear that it will be able to mill. This sounds very confident. But something I now, our listeners, don't know is you actually were on vacation when we did the first chips. You're like so confident. It's happening. You were on vacation. And I know very well. People asked me, "Hey, Marvin, can you hack a post-processor together so we can run a pro-hoydle milling program on that machine?" And I was there with Christoph Stadman. I was a Christoph Stadman and I were the first two people milling on it. And I kind of looked around and was like, "Where's Christian? Why isn't he here?" And everybody was like, "He's on vacation." And I was like, "Shouldn't we wait for him?" And everybody was like, "No, no, he said it's good. We can do milling." We actually have video footage. I sent it to you in your vacation back then with video footage of the first chips. It's quite the beefy cut with an insert cutter already. And I also kept the chips. We have three chips of what is. Which is a bit good. Yeah. Between the box. But I'm hiding them very well away. So you kind of need to pay me for you. No, and I mean for us, this was a really cool moment because I'm from all of the stupid questions. I'm asking I wasn't very involved in this problem at all. I was mostly sometimes doing a bad joke about it. And when we took the first chips, I was like, "Hey, people, this is an historic moment." For me, it had more weight than for you. And I told them, "Hey, we're going to film it and please don't talk over the recording. We want some sound of the machine of the first chips hitting the enclosure." And it's kind of funny because you had like the 607 people from your core team standing around. Christoph and me operating the machine. And the camera was rolling and everybody was silent for like 10 seconds. And then everybody started high-fiving and shouting because they were so happy it's working. That was a really cool moment. Just like see it make chips. And you weren't there. I'm a bit sorry. Okay. Very cool. I think something, another major change in the design of the machine compared to the HD is not only the hydrostatics but also the cooling management. And here I don't mean the coolant management but like cooling the actual machine parts. Yeah. Maybe not a huge change. If you know the micro HD, you know the area where the cooling aggregate is, it's an area next to the machine. It's a separate aggregate. You need separate space. So the machine had a kind of orders. We decided that we want to integrate it under the machine at the back of the machine. And this was a very tough decision with many consequences because it was really hard to make the design that it fits under the machine. Everybody said, "Are you crazy? You never get it under the machine. It will never work." But I had no, we decided this. And so we just did it. Not works. It's pretty cool. You have nothing next to the machine. You have all the aggregates for the cooling, the axis and the spindle and the supply for the hydrostatics under the machine. If you want or if you need, you can pull it out and make some service or something. Take a look at it. It's beautiful. Without stopping the process and the machine. So take a look at it now. I think everybody says it has to be like this. Nobody would make it different. But we had this picture at the beginning. It's quite the challenge. Typically you stack everything in the hydraulic aggregate and for some components, for example, the pumps. It just makes sense to stack the pumps on top of the cooling tanks. And I will imagine if you put it under the machine, that's difficult. But it's still wholly isolated from the machine in terms of vibration, not entering the machine. And also the performance of our temperature management is the same like the market. Very cool. A little bit more cool. Yeah. A little bit larger as well. My machine has been crying after coolant for the past. I would say four or five machines. The four does I'm working on. And the little bit more I always tell Josef from construction. I'm like, Josef, it's asking for coolant. Where do I put cool in? And he's like, I still have time. So yeah. Maybe let's quickly chat about the design of the machine. Because I mean, I've seen it. And we're definitely not going to post a picture before Aimo about it. So you just need to explain a little bit. But it's pretty special, isn't it? It is pretty special. And I think we can be proud of the design. I mean, this is the moment where we can tell the internet that it's spherical in shape. You know, we have the pyramid nano, we have triangular, the HD, which is boxy. So now forges is like a sphere, right? Yeah. Is that you know? Really? Not. It's pretty cool. If you stand in front of the machine, it's really pretty. It has some features you kind of know from HD, right? Yeah. With the decision, we make the whole machine design similar to the microHD. We also decided to put some design parts from the microHD into the FAUTIS HD. Maybe if you see them together. In my opinion, now you just see a microHD in bigger. I mean, I have to say I spend a little bit more pretty, but bigger. I mean, I spent the past five weeks now milling on FAUTIS. And I looked at the spindle hours today. I've done over 450 spindle hours in that time. So I'd say I've quite a heavy user on that machine already. And I have to say what I find exceptional about that machine is it feels completely like an HD. It's just like, and this is something Simon and I talked about extensively in the last podcast. It's just like that magic of the HD, it's there as well. And I think this is the biggest kudos I can give you building a machine which is more than twice the size. I mean, we're always talking about, it's more than twice the size, but this means for the volume of the machine and all of the errors that can happen that it's more like six times as large or even more. And if you think about that, it's amazing that you manage to scale that up by not scaling that up. And it feels exactly as magical. Not scaling up the errors. Exactly. So from my side, this is a huge win for Kern and also for your team. That's just like it feels like an HD. But it's nice to hear. So I think we made the right decision to build it like a micro HD. And it is like a micro HD. Yeah. Before we wrap up, maybe looking back on that project, what would you change if you had to do it again? I would say in technical detail, I can't say what I would do different, but maybe in the team structure, I would change a little bit. Maybe at the beginning, smaller team with one or two people who have a wider knowledge. For example, that somebody can do something with electronics and mechanics. Calculating and designing. Or yeah, this is to move a bit quicker and to move a bit quicker to keep the interfaces between the people small. Yeah, to decrease the amount of meetings you need because you have to spend a lot of time. If you want to do it good, you have to prepare for the meeting. Then you have it and later. Okay. So I think we have to be faster with the first prototype to make more and faster improvements when we hit the market. Till we hit the market. Okay. Yeah. I think the last real content question before we have to discuss a couple of other things is, would you do it again? Would you consider doing another leading and another development project as current? At this moment, I can't give you answer. That is totally fine. Before we wrap it up, are you going to be at Emo? I will be there. Will you be at the open house day in Chicago as well? No. So Emo, every day for Monday till Wednesday. Monday till Wednesday. So dear listeners, I will be posting a picture of Chris Chen with this podcast. He won't do it. Yeah. He wondered that. No, he didn't want that. But he's going to be at the current booth. You can talk to him. He's super cool. He knows so much. He can explain to you about the machine. Emo is in the middle of September, I think it's starting at the 15th of September. No, 22. 22nd. 22nd. 22nd till 26th September. Nice. You know better than me. Very good. Christian will be there Monday till Wednesday. I will also be there Monday till Wednesday. I'll be back on Thursday. So nice. Okay. We will all be there. Come up, chat with technical people. Talk to them. I think everybody has worked on the project. It's really enthusiastic about it and what they made there. Christian, thanks a lot for your time. Thanks for chatting with us. Until the next episode. We'll see if it happens.