Boeing has cracked the code on digital design for the T–7A Red Hawk military trainer.
Episode Summary:
The Boeing/Saab T-7A Red Hawk military jet trainer will be built using model-based engineering principles. Model-based engineering is proving to be a perfect fit for the aerospace industry: more accurate, precise parts and assemblies that fit better and faster on the assembly line. Model-based engineering also offers a hidden bonus that few in the industry talk about.
Access all episodes of End of the Line on engineering.com TV along with all of our other series.
Transcript of this week’s show:
For such a high-tech enterprise, it can be surprising how slowly things move in aviation. For the U.S. Air Force, training pilots to fly advanced jet aircraft has been built largely around the Northrop T–38 and that excellent aircraft has been flying for, incredibly, half a century. Jets are fast, but military procurement is slow, and good as it is, the T 38 probably should have been replaced two decades ago. With modern 4 ½ and fifth-generation military aircraft on the flight line however the need for an advanced trainer with modern systems simply forced the hands of Pentagon planners, and the winning team, Boeing and Saab, have produced a very useful looking airplane in the T–7A Red Hawk. The aircraft is capable, affordable, at least by military standards, and maintainable and the USAF plans to buy some 350 of them. By automotive standards, the industry I come from, numbers like that represent a Tuesday afternoon, but in the aerospace world, that’s a multiyear production program, and the unit costs reflect it. But cost-reduction pressures are just as important in the aerospace industry as any other form of production engineering, and Boeing has turned to model-based engineering systems to get the Red Hawk off the drawing board and into production. The challenge was simple, using the static test airframe. The rear fuselage assembly is built by Saab in Sweden, with the front built by Boeing St. Louis. Two different companies, on two different continents, building safety critical major assemblies that must be mated under assembly-line conditions. Normally, we would expect to see lots of shimming, and maybe some match drilling, but the two assemblies fit like Lego blocks, going together in 30 minutes, some 95% faster than industry standard. That’s remarkable and is a quantum leap over even fully digital designs like Boeing’s own 787 Dreamliner. What model-based engineering does is take the same processes that were done before, namely design, static test article fabrication and fitting, then redesign, and take it into the virtual realm. The key factor is that the interaction of components, subassemblies and assemblies are revealed that the software level, especially important where the high complexity of modern systems make it almost impossible for humans to anticipate how they interact dynamically. Mistakes are still made, miscommunication between teams still exists, and vendor supplied components still deviate from nominal specifications, but now the hammering and cutting and swearing happens virtually, shortening the test/redesign cycle to minutes or hours instead of days and weeks. Living with a substandard part because the design is frozen is much less likely when an engineer can iterate 100 tweaks in the time it would take to prototype and fit check one. And there’s a hidden benefit few talk about with advanced design systems like this: politics. Design changes are expensive, but they are really expensive when they involve complications like altitude and low temperature testing, fatigue failure testing, complex fabricated test articles and of course the spectre of the FAA. Engineers are incentivized to think conservatively, and colour within the lines. In the virtual world, you can do anything. Wonder if you can take half the weight out of that bracket? Try it. Want to clamp that condensate line in four places instead of five? Excite the structure, and see the bending modes, in living colour. We talk a lot about additive manufacturing and generative design as breakthrough technologies in production engineering, but I think the real breakthrough is going to be not in how design is automated, but how it allows designers to get creative and break things, at essentially no cost. And in every engineering organization in which I’ve ever worked, “no cost” is always the right answer.
