In World War II in Europe, a key to victory over the Nazis was the strategic bombing of factories and industrial sites. In the early part of the war, that process wasn’t going well. Losses of bombers were excessively high; on October 14, 1943, for example, a raid on ball-bearing factories in Schweinfurt resulted in the loss of 77 airplanes and 650 aircrew killed or captured.
The loss rate was an unsustainable 26 percent, and the reason was readily apparent. Cannon-armed German Air Force fighter planes attacked the lumbering bombers at will, while fuel-thirsty American and British fighter planes could only protect them for a fraction of the route.
The solution to this problem was the long range escort fighter, which could protect the B-17 and B-24 bombers all the way to their targets and back. The ultimate expression of this race for a long-range, high-performance fighter plane was the North American P-51D Mustang. The P-51D was the combination of a US-built advanced airframe with a powerful Rolls-Royce designed and Packard-built Merlin engine. It’s even said that Nazi Air Force chief Hermann Goring said, “I knew the war was lost when I saw Mustangs over Berlin.”
This winning combination of a great engine and an advanced airframe made the P-51 a legendary airplane—and seven decades later, the Mustang remains a favorite of aviation collectors and enthusiasts worldwide.
Rob Connelly, former vice president of additive manufacturing for Proto Labs, now retired, owns and flies an outstanding example of the breed. His is a pristine 1944 P-51D named “Hurry Home Honey.” It carries the unusual name for a reason: it was once flown by Major Richard Peterson, a highly-decorated World War II triple ace with 15 and-a-half victories and 150 combat missions over Europe.
Peterson’s girlfriend Elaine ended each letter with the expression, “Hurry home, honey,” and so the Major added it as “nose art” to his aircraft. The combination of the aircraft’s historical significance, high performance, and rarity make it a historical artifact, but this is no museum piece: Connelly flies the aircraft regularly.
Fighter aircraft have always been maintenance heavy, but operating a mechanically complex machine decades longer than its designers ever dreamed, while also complying with modern standards for flight safety, is a challenge. Spare parts range from scarce to nonexistent, and it goes without saying that they’re expensive.
“The engine’s putting out 1,500 horsepower, so there is tremendous wear and tear,” noted John Murray, president of Concept Laser Inc., which General Electric has recently acquired to form a division called GE Additive. He is also a friend of Rob’s, as well as a fellow pilot.
“As Rob and I talked, we thought, ‘How could we reach back 70 years and improve the design or make something more reliable?’ The exhaust stack came to mind because it’s visible, it goes through heat cycles, and it’s under tremendous stress in the airflow,” said Murray. “These exhaust stacks produce a level of thrust, adding about three to five knots of airspeed. There are some interesting aerodynamic tricks on this aircraft, and the exhaust stacks are one of those. So, we thought, the original part was built from stainless steel. Can we can print a new one today in stainless steel, and have a brand new part for an aircraft that’s 70 years old?”
To answer this question, Murray decided to use the replacement project as a test case in end-to-end advanced manufacturing, from 3D scanning and rendering to SLS part production, part finishing, and validation. The Concept Laser team began by examining the original stainless steel part. The stack represented typical 1940s technology: a machine flanged combined with two formed sheet stainless clamshells and a clamp ring in a hand-welded assembly. The original offered dimensional data, but little else.
Modern digital renderings would clearly be needed for the metal AM build, but Murray’s team realized that modern simulation and CFD software would let the group do something the original Rolls Royce engineers could never imagine.
Murray describe the prebuild process: “We took the part to Phoenix Analysis and Design Technologies (PADT), in Phoenix, Arizona. Rey Chu is one of the principals there, and they’re experts on reverse engineering, 3D scanning, and are ANSYS experts on thermal and fluid flow simulation. So we thought, ‘let’s ask PADT to build that CAD model, and then do some really advanced flow characteristic testing on it.’”
The build progressed beyond a simple aviation MRO solution into true research: how well did the original exhaust stack flow? Could the laser sintered part perform better? What about the effect of surface finish on flow? Would the elimination of 34 inches of weld seam affect performance?
Rey Chu applied state-of-the-art hardware and software to the data problem. “We took the existing four-pieced welded assembly and used a blue light scanner, which has a very high resolution, giving us about 5 million data points per square inch of surface area. We scanned the whole assembly, took the point cloud data and turned it into a meshed surface model that smooths all the weld seams and all the machining surfaces. Then we created three-dimensional CAD data and converted it into a 3D printer data format called STL. I sent that data back to John, and they were able to put it in their machine and print this part.”
Chu chose blue light scanning over conventional laser scanning for performance reasons, explaining, “They all have pros and cons. Blue light is a much faster technique. We did this in less than a minute, with much higher resolution versus the laser scanner.”
The scanning was quick, but more than four hours were dedicated to data processing to eliminate noise and create a print-ready STL file. While the team’s primary goal was an exact replacement for the stainless steel stack, Concept Laser has a wide variety of possible working metals and few limitations on the part form, which suggested possibilities for a part better than the original.
According to Chu, “We reverse-engineered the existing product, so that’s the exact same geometry that it had for the past 70 years. The next step is to do analysis on the existing component’s geometry, stresses, thermal gradients, and also the gas flow dynamics and work from there. Is there a way for improvement or optimization for the current manufacturing techniques? We can definitely cut a lot of weight out of this, and we can increase performance and reduce thermal stress by changing some of the geometry. That’s the next project we can work on.”
The design of the part is key, but in additive manufacturing, the as-printed product includes support structures which are critical to dimensional accuracy and efficiency in the build.
Murray described how Concept Laser approached the build. “At first, we look at the part orientation; you’re trying to reduce the number of supports you’re going to have and where they’re needed to optimize the build. In this instance, we built two parts at one time, with supports on the down-facing surface and no supports going up vertically through the exhaust stack.
“That worked out beautifully, because we were going to finish machine the base anyway since that’s the surface that mates to the aircraft. These are the kinds of things you’re thinking about when you’re orienting a part: where these supports are going to be, what’s easiest to remove them, what features are going to be post-processed hit with EDM or machining,” Murray said.
Concept Laser built two exhaust stacks in two days on their M2 cusing platform, which has approximately a 10” x 10” x 10” (x,y,z) build volume. Murray considers the stack a simple part, made from a straightforward stainless steel material. Concept Laser does, however, build from a wide variety of metals, including many of the exotic aerospace superalloys, such as the Inconel grades often used in jet engine hot-section components.
The firm also works with customers to develop exotic materials for high-performance applications, Murray stated: “We do engineering work for companies that bring a particular alloy and say, ‘can you run this?’ We’ll say, ‘we’ll work on that and the build process.’ There are more alloys out there than people even realize. Once a company develops an alloy, they’re not necessarily going to make that public. We do a lot of work in consulting with our customers to help them really push the limits.”
In pushing those limits, additive technology is quickly racing ahead of regulatory bodies such as the FAA in many applications. Murray predicts that the recently certified GE jet engine fuel nozzle and Airbus’s experience with airframe components will open a flood of flight-ready 3D parts, each of which will require certification for production aircraft. “I think it’s only going to accelerate. Since they’ve given certification on the GE jet engine parts, that’s a huge step forward,” Murray said.
Like Connelly, Murray is experimenting with additive manufacturing for his personal aircraft, a Beechcraft single that he flies extensively. “The Bonanza I fly is a general aviation aircraft, so I can build owner-produced parts for that aircraft, and I’m doing some ducts, some different things just to work with the technology and improve the designs. You’re allowed to do that as an owner, but you can’t obviously sell those components. The FAA, I think, is definitely on board on the commercial side, and they’re going to continue to come up to speed on 3D printing. It’s moving so quickly, no doubt it’s going to be a challenge for them to keep up with the pace of this technology.”
The stainless steel Mustang exhaust part is a good example of why metal additive manufacturing is unbeatable in many critical applications. Although built with a conservative material, the 3D part is stronger, more durable, reduces the part count from four parts to just one, and flows exhaust gases better than the original. And with no need for custom tooling or fixturing, it’s considerably lower in cost than short run or one-off production for this rare application.
These planes have not been produced since the 1940s—the aircrafts themselves and the parts for them are rather scarce.
Connelly, a 3D printing pioneer himself, has no qualms about flying his Mustang with FAA-approved additive parts. “These planes have not been produced since the 1940s,” he said. “And they didn’t make that many of them in the first place. Very few of them survived their military service. So, as you can imagine, the aircraft themselves and the parts for them are rather scarce. If an owner or a pilot has a part that’s broken or worn beyond serviceable levels, they have to reach out to the warbird community in hopes that someone has a military surplus part lying around, or something from their own operations that they can spare.”
“Often, they’re waiting a long time and paying a premium. Fast forward to the future with 3D printing, and we can scan all these components in, refer to the original drawings that were used to create them, and create three-dimensional databases. We’ll be able to pull from those data sets to make components in the future. It’s definitely within the realm of possibility,” Connelly said.
Like Connelly, Murray carries a dual passion for aviation and additive manufacturing technology, and he agrees that there are few limits to widespread application of 3D printing in the highly conservative, safety-driven aerospace industry.
“Parts will be safer, stronger, and perform better with a lower overall part count,” Murray said. “There are just so many benefits to this technology. In the Mustang stack part, for example, the elimination of 34 inches of weld seam is a major advantage from a production and performance standpoint, so it’s definitely going to step up the game in every industry, whether it’s aviation, medical, automotive, and many more. This is touching every aspect of our lives. It’s an exciting future.”