Episode 38: Alan Hiken, Kane Robotics
Alan Hiken, COO of Kane Robotics, talks about historic, current and future composites use in aircraft fuselage structures and the role of automation in next-gen manufacturing environments.
#777x #787 #trends
Episode 38: Alan Hiken, Kane Robotics
Alan Hiken, COO, Kane Robotics. Photo Credit: Alan Hiken
This is episode 38 of CW Talks. In this episode, CW’s Scott Stephenson, director for strategic initiatives and events, talks to Alan Hiken, COO of Kane Robotics. Alan has long experience with composites use in commercial aerospace applications — specifically the fuselage. Scott will talk to Alan about his work on the Large Aircraft Composites Fuselage Program, NASA’s Advanced Composites Technology Fuselage, development of IML tools for the 787 fuselage, the pros and cons of IML and OML fuselage design, and the role of automation in a fast-changing manufacturing world.
Transcript of Alan Hiken interview with CW Talks
Scott Stephenson (SS): From your bio, I think you have an interesting perspective on the evolution of the composites fuselage, having worked at an OEM Northrop for many years on many projects, specifying and contracting tooling materials and then switching to a tooling supplier Rubber Craft, their tooling materials that were used in a lot of programs for the composite fuselage. In any chapter of the book, you describe the evolution of this being methodical, tenacious process between many partners military and defense, academia and industrials. Can you discuss briefly some lessons learned from the Larger Aircraft Composites Fuselage program in the ‘80s, and NASA's advanced composites technology fuselage?
Alan Hiken (AH): Yeah, I like to say that I've had the opportunity to sit on both sides of the desk as you mentioned as an OEM, and then as a supplier to the OEM and all tiers of the industry. I feel fortunate to have had the opportunity to stay engaged with the composites community even as my career has evolved into those different environments. The LACF program is certainly a trip down memory lane for me it, was the first real program I worked on straight out of college, and I had no real idea or appreciation for all the things that we were we were doing. LACF was a right labs Air Force program directed towards the definition and demonstration of manufacturing methods for co carrying stringer stiffened fuselage panels, typical of a larger aircraft composites primary structure, all in support of what was called linear manufacturing in those days. Today's terms we would call it lean, straight line throughput linear flow, automation and forming processes, no diversions to the autoclave. Among the ground rules for that program was the use of existing qualified materials, automated skin fabrication, IML controlled tooling, and the use of non-autoclave technology. It was really kind of throw the kitchen sink at a level of development, and to this day I'm amazed at all the technology we were experimenting with on one program. Automation was applied via fiber placement and continuous roll forming. We used integrally heated tooling, we developed a cross ply lamination cell and, did drape form technology, all pretty amazing in retrospect.
SS: So Alan, you did mention the US Air Force Large Aircraft Composites Fuselage program in the ‘80s LACF. What about the lessons learned from NASA's Advanced Composites Technology or ACT fuselage program? How are they different from the LACF and was it just a continuation of technology that you picked up on the LACF that went further in the ACT program?
AH: Actually, the ACT program was much larger, had many more companies and universities involved, larger budgets as well that go with that and actually a much more laser focus on what they were trying to achieve. It really dug into the details of the preferred structure or arrangement for large commercial aircraft composite fuselage, skins and frames, and the associated manufacturing processes to fabricate those preferred structural arrangements. Another significant component of the ACT program was the development of a cost model for evaluating the cost effectiveness and reproducibility of the various designs, tooling approaches, and materials. I think that that was again all things that fed Boeing as they develop the structural concepts for what will lead into the 787. They studied stiffened skin structures with combinations of co curing, co bonding, bonding, and mechanical attachment of the stringers and frames to either monolithic skins or sandwich panels. They looked at AFP and contoured tape length for skin fabrication. For frames, they were looking already at resin transfer molding, compression molding, even pultrusion and stretch forming of thermoplastics, and various configurations like Cs, or Zs, or Js. One significant lesson that they had was the blind nature of trying to fully integrate the skin frame stringer structure using flexible call plates and custom fit rubber bags was just too risky to develop that fully integrated structure. So they settled on the cocuring hearing of stringers in the IML tooling, and the mechanical attachment of frames as the preferred concept, and that was what ended up going into the 787.
SS: You mentioned a IML. What are the differences between IML internal line control servicing and OML?
AH: Traditionally, aircraft had been built from the outside in not the inside out, you set the skins and LML controlled fixtures and you start adding stringers and frames building towards the center of the aircraft. 777 was really the first airplane to be built from the inside out. In traditional LML controlled tooling, the variability from the fabrication and assembly float to the inside is absorbed by the assembly process you have variable gaps between structures, and it's addressed with shims at different sizes and thicknesses that are custom fit for each gap. With IML controlled tooling, the process variability flows to the outside of the aircraft, which is the aerodynamic surface accumulated tolerances have minimal effect on the aerodynamic performance and can be addressed by ensuring the surfaces smooth and sanded. Well melt tooling is a lot less complex, it's less expensive, and can be initiated early in the process as soon as the LML is defined. It's very forgiving of changes to the structure and it's often not even impact. On the other hand, IML tooling is a lot more complex and expensive. Tooling can't start until the structure has been fully defined, including all the pipe buildups and laminate thicknesses. It's not very forgiving of changes to the thickness to the structure, and it often requires modifications if there are changes. It is better at maintaining stringer locations and it's easier to vacuum bag, but the real benefit comes in on the assembly process where you're going to save a lot of time and money in that process because of paying the penalty upfront and having to wait and define your structure before you start your tooling.
SS: Can you discuss a little bit how tooling design and manufacturing advances sort of coincided with each other, and were co-developed during the process of this evolution of the composite skews alive?
AH: Tooling was clearly an enabler to building a large composite fuselage and the ability to produce very large inaccurate tools was a key factor in that the ACT program certainly proved how important the tooling was to produce ability of large integrated composite structure, large multi axis machining capability, laser measuring and tracking systems, the wide use of an availability of Invar. Those were all key enablers to being able to build the 787 fuselage and the A350, and when Invar was too heavy Spirit incorporated a BMI cured tooling. Boeing also used composite tooling on sections in 46 and 47 after they had some early generation room temperature-cure tooling challenges, they evolved into the BMI care tools as well. Interesting part of BMI tools as I know it Northrup in the ‘90s, we used BMI cured tooling for autoclave cured BMI tooling for the F 18 program, and it was a far cry from what had been used on the C and D variants, which were wet laminated fiberglass tools that had to be frequently replaced. We used to like to say that the tools were better than the parts coming off of them on that program.
SS: Interesting. You mentioned both Boeing and Airbus. Can you talk a little bit about the different tooling approaches employed by Boeing versus Airbus?
AH: Yeah, Boeing definitely built on the results and lessons learned from the ACT program., and you can see many elements from that program in the final design. Boeing also took an approach where all the partners and all of these sections followed a similar manufacturing process in fabricating full barrel sections with cocured ???????. They all use AFP technology over IML controlled lab manuals that also serve as the cure tools. With the exception of the tail all use multi piece breakdown tools that are removed from inside the fuselage after cure. As we talked about earlier, one things most people don't realize is that because it uses IML tooling, each airplane has a slightly different LML. The IMO control tuning process produces a dimensionally uncontrolled outer surface and complaints are used to control the smoothness and the waviness of the OML but they float, and the actual surface location is going to vary from cure to cure based on the original resin content of the prepreg, how much resin might be lost during the care cycle, how much material might be removed during the sanding and the preparation process for painting. It's not the traditional way of building aircraft. Airbus on the other hand, did not use as uniform an approach to their manufacturing of the fuselage for the A350. They did use commonly Inbar tooling and longitudinally incorporated omega stiffeners which are just hatched by another name and those are common, but the A350 for example incorporates one complete barrel section like the 787 but it's the tail. It doesn't require breakdown tooling because the tail has enough draft in it that you're able to slide the skin off of it after cure so they didn't have to deal with the complexity of a breakdown tool. The rest of the A350 follows a more conventional panel assembly approach. Now Spirit is a common supplier on both programs, and the fabrication approaches used by Spirit do share several characteristics like IML controlled tooling. On the A350 Spirit uses sector tools and look a lot like a quadrant of the section 41, the barrel section from the 787. It uses inflatable mandrels to co cure the omegas to the skin on call plates to smooth the outer surface. Like the 787 process, bladders are removed after cure reused and they're up to 65 feet in length. The rest of the A350 fuselage uses co bonding to incorporate the Omega stiffeners with the fuselage skins. Earlier we talked about the co cured has different issues on the 787. A350 uses the same amount of pre cured omega stringers on the A350. I remember on a tour of the Clearfield facility at ATK, which is now Northrop hearing him say we manufacture more than two miles of precured Omega for every A350 in the cobinding process. The precured stiffeners are located on the green fiber place skin with a layer of film adhesive between the two of them, and an inflatable tube made from thin film is putting aside the void and open to the autoclave pressure to prevent the Omega from being crushed during the co bonding cycle. Definitely a more hodgepodge of various technologies incorporated on the A350 that what you see on a 787.
SS: So, they didn't go through the same amount of lessons learned or the same programs that were here in the US for LACF and ACT. Now they have these other programs that are similar to the large aircraft composites fuselage and NASA's advanced composites technology or ACT, and it seems like that clean sky and the wing of the future programs have had a recent big influence on composite aircraft technology. Is some of the tooling that they're using similar to what was developed for the 787 and A350 do you know?
AH: I'm not as familiar with the specifics on the tooling approaches that they're taking. Although I will say that Ginger has done an incredible job in Composites World of reporting on the progress of those programs and the technology that's being incorporated there is certainly very impressive. The Europeans have been very active and very public about their clean sky and wing of the future programs. My biggest question there is, how much of that technology being developed is actually going to be applied to the next aircraft? The skeptic in me says that much of the development is kind of for development sake and we'll never see the production, but I hope I'm wrong and the process can be made production ready. I know ????? has been doing a great deal of work and feel very confident there. I like to say again during my career I encountered many brilliant people that can build anything once, but they do it 1000 times is a much smaller group. The optimist in me sees the impact that the LACF and the ACT program had on subsequent aircraft especially the 787, and you'd have to believe that the clean sky and wind of the future programs will have a similar influence on future aircraft produced in Europe. At the end of the day though I'm from Missouri, and you know that the show me state will be interested to see how much of this technology is actually incorporated into a production program.
SS: I guess we'll have to see what how that shakes out. Recently we had this 737 max disasters and now COVID. Assuming further composite fuselage development eventually continues into single aircraft. What do you think are the major challenges from your experience besides rate? What might be some new enabling technologies to increase rate and overcome some of these technical challenges?
AH: Well, I think you're certainly going to start seeing more automation being applied. I hope that the industry is using this COVID pause to better understand their existing operations, to lean them out, to look at process improvements, and to be preparing themselves for when the industry does start to shake out again. I don't think that it's going to be overnight, especially for the commercial industry. I think the defense and the space industries have fared much better during this time, and they seem to have continued on as best they could despite the COVID crisis. In a lot of cases they have actually been helping to sustain the supply base that was that's lucky enough to have dual military and commercial programs.
SS: You discussed, Alan about the difference between the two philosophies chosen by Boeing and then by Airbus. The OML floating surface can be sanded and finished to dimension, whereas hundreds of thermoplastic clips are used for the Airbus panels. Do you see an advantage for one or the other? Where do you see maybe thermoplastics technology coming more into use and single construction?
AH: I think that one of the major differences between the A350 and the 787 was the structural arrangements between the two programs. The fact that the IML tooling incorporated on the 787 and the close molding processes that were used to make the frames RTM processes, essentially meant that you had a tooled surface to tooled surface assembly, and where those two surfaces are going to mate. Boeing was able to bring the frames directly down to interface with the skins because the hats had a known height and they were tooled. They weren't going to wander in their locations, because they were they were tooled that way. Whereas Airbus wasn't as confident in the ability to make the frames to the skin because of the variabilities, and the uncontrolled skin on the IML, and the difficulties in trying to liquid or hard Shim each of those gaps. They chose instead to attach to the tops of the hats, and to address the intersection between the frames and the tops in the hat, Airbus develop the standard thermoplastic clip that they make by the 1000s, and that they use to adjust slightly on those different tolerances that are absorbed by that clip in their assembly. To me, that's the biggest difference in the tooling approaches between the two programs. Obviously Boeing is not sharing their cost data and is not sharing their lessons learned from the assembly process on the 787. I think that's going to manifest itself in what they do on on their next aircraft, and to see how they approach that aspect of it. I think that the success that Airbus has seen with the thermoplastic clips, I know Boeing has using some thermal plastic parts on the 787, I think you'll continue to see thermoplastics used in a right sized and a right application manner. I don't think you're going to see us ready to do in situ consolidation on the fly of thermoplastic fuselage panels for the next aircraft, I don't think that technology is ready for that. I don't know if you're going to see a lot of thermoplastic welding applications, or if that's going to be ready in time to be proven for the next aircraft if they are it's probably going to be on a less than primary structure. In a lot of cases the typical implementation path of new materials and aircraft structure which is start with bearings and non load carrying members and gradually earn your way into more and more sophisticated and highly loaded applications, and I think thermoplastics will follow that path as well.
SS: Fokker GKN has been doing a lot of work and I think it's been leading towards primary structures, but I'm not quite sure it's ready for primetime. Although our friend aren't offering and I disagree with that. Finally, in a recent editorial Jeff Sloan discussed the necessity for Boeing to develop a new mid-market aircraft to remain competitive with Airbus, and Merrill Lynch Bank of America analysts Ron Epstein has sort of said the same thing. Boeing has announced that they're going to do something so seemingly following their advice. Continuing on the theme that you just started in the last question, what type of design construction materials and processing do you think would be the key to their success with this aircraft based on your experience?
AH: I think there's been a lot of press lately, and based on what I've read, heard and experienced, I think Boeing really goes into the enemy, you know, 5x, smaller variant, with a low tolerance for risk and the need to get to market as quickly as possible to stop the bleeding and deterioration of the market share to Airbus. I think they're going to end up with a pretty conventional structural arrangement with existing qualified materials and proven engine technology. I think this includes the materials of construction, proven systems, proven prepregs, probably a lot of autoclave curing systems already being used on 787 are the KC 46 tanker or 777X. They've got a lot of autoclave capacity out there. I do think Boeing sticks with IML controlled tooling, and full barrel five replacement fabrication, but I think that they move to non breakdown composite tooling. I think that they're going to laminate that full barrel and cut it into upper and lower shelves to remove from the tool without breaking it down. I think that this really simplifies the process for making the next barrel, we'd reduce their rate to requirements, and it saves them a ton of floor space as they try and manage their way through the rate tooling. I think that one of the lessons learned and one of the concerns Boeing had going into the 787 with the full barrel construction was the cost of adding longitudinal splices in the assembly process, but I think the trade off between dealing with the longitudinal splices versus dealing with the breakdown tooling will lead them to an upper and lower shell type of construction that is similar to what the HondaJet uses. As a matter of fact, I think they're going to reel in the supply chain and stick with domestic partners, probably Spirit for most of the primary structure. I think they'll keep the foreign sales enticed with secondary structure, the wing technology that's on 787, and the seven triple 7x program will certainly continue with the enemy family. After all of their investments in wing technology, and the autoclave capacity, that wing center up in Everett, the decision to move 787 fabrication solely to South Carolina leaves them with some floor space, whether they can produce that next aircraft. So I think it's going to be pretty conventional and pretty risk adverse in most in most cases.
SS: So you think that will allow them to be quicker to the market.
AH: I think that will be certainly critical for them at this stage is to get to market as quickly as they can. With the 787 they tried so many different things, just going to the full barrel was probably enough without the lithium batteries, and without some of the avionics changes and some of the other things that they tried with that aircraft. I don't think that they can afford to do the same thing with this next aircraft considering the market condition that they're in.
SS: So Alan, to achieve some of these rates that we're talking about with single aisles and some of the new enabling technologies that you've discussed to increase rate and overcome technical challenges. What are some of the technical and production areas that you think, smaller, menial task robots such as your new company is developing might be employed to help solve these challenges?
AH: I know something the industry has been talking about for a long time is the aging workforce, and the ongoing loss of skills and experience for both engineers and mechanics on and above the shop floor. The fact that new college graduates haven't been flocking into aerospace careers for the past decade or so hasn't really improved that situation although I do think that has been improving, and will continue to improve. What it means at the end of the day is getting more out of the labor that you have, making sure you're skilled laborers working on their most valuated tasks, and things that require skill. First off, I'm not a robot person or automation expert, but if those solutions were available to me in my previous life, I definitely would have been interested. So much of what I bring to the table in this area is my experience in aerospace and defense, composite structures, assembly, and the perspective of somebody who would be purchasing these systems for implementation into an environment that I was responsible for. In general I know what I want the robots to do, and we have bright young engineers that know how to how to make the robots do it. The opportunity to employ lightweight, low cost automation based on these collaborative robot platforms to perform dull, dirty and repetitive tasks that are currently being performed by highly skilled labor. Typically, things your mechanics don't want to do anyway, like grinding, sanding and finishing. Anytime you have to put on a bunny suit and gloves, then take your gloves to the bunny suit and add a mask or respirator, goggles, and a hood is probably not a job you really enjoy doing and not a job you enjoy doing for very long. The little cost of these systems allow you to think of the robot more as the next generation of power tool for the mechanic for them to use to help them perform their job. We like to call it the 80/20 rule that the robot do 80% of the work, the acreage the stuff that mechanic is forced to do along the way, because it needs to be done anyway. Let the skilled mechanic perform the 20% of the task that requires their skill and their creativity and their judgments while they're doing that task. The low cost and fast ROI of the systems means it's okay for the robot to sit for a few minutes while the mechanic performs a skilled task. Unlike industrial automation, which has higher cost or disruption to the factory and takes a longer time to implement. There is plenty of growth and opportunity out there for industrial automation, and we'll continue to see AFP applications in automated drilling, but unlike monument robots, as we like to call them, these lightweight, low cost robots can be applied with a different mindset. We think of industrial robots like an out of autoclave enthusiast thinks about an autoclave. It's a monument in your factory where you divert all your flow to it, so you want to avoid doing that when you can put in a mobile system. These new types of robots plug into 110 volts, and they don't require 480 volt three phase power. I mean that's huge when you're considering the factory planning and layout that goes into an automated installation. I think the marketplace for these lightweight, low cost solutions are just going to continue to explode, and it's going to allow small and medium sized enterprises to implement automation where historically they never would have been able to consider that type of an expense.
CompositesWorld's editors report on the technologies and products that caught our eye at JEC World 2016, in early March.
Although North America currently dominates this market, Asia and the Middle East, in fact, are expected to be the primary drivers of global aerospace industry demand in the coming years.
Demand for advanced fiber, virgin and reclaimed, is on an upswing after a dip during the recent recession. Fiber suppliers are gearing up to meet an expected increase in demand as manufacturers in a number of composites end-markets scale up or start up part development and production programs.