After three years of work, military researchers are near the end of a project to find a faster, cheaper way to make tools for large aerospace parts like skins for wings and fuselages.
Results are looking good for large-scale additive manufacturing (AM) of carbon fiber-reinforced thermoplastic materials that can be autoclaved. Compared with metals, the researchers typically saw tool delivery that was up to five times faster and tools that were two or three times cheaper.
But questions remain before the process can move into full-scale production: Would the composite tools stand up to the heat of repeated autoclaving? Over time, will the vacuum integrity hold for an AM tool made from two parts and joined with a sealant?
“I think the big question is, what is the durability of these tools?” said Craig Neslen, low-cost attritable manufacturing lead at the Air Force Research Laboratory (AFRL). “Am I going to lose dimensional accuracy as I expose the tool to repeated thermal cycles? Is it still going to maintain its proper shape and, if so, how many quality parts will I get out of the tool?”
The lab’s results showed a tool made of Ultem 1010 with 10 percent carbon fiber fill withstood 57 cycles in a vacuum-sealed bag in an autoclave at 350°F (177°C) and 90 psi (620 kPa) while maintaining ±0.020in (0.508mm) dimensional accuracy compared with an aluminum tool that theoretically could be used for a “couple hundred cycles” at the same temperature, Neslen said.
Limited lifespan may not be a deal breaker. Storing and maintaining a tool that’s used infrequently or may ultimately no longer be needed costs money and time to track and maintain.
During Phase 1 of their two-phase project, the AFRL scientists found that an aluminum version of their tool would cost $14,000, more than twice that of one of the same dimensions but made of carbon-reinforced Ultem 1010, which cost $6,276 (see chart).
The difference between the aluminum and fiber-reinforced Ultem tool was much more significant for the time required to manufacture the tools: 10 days for aluminum and 11 hours to fabricate the additively manufactured polymer tool.
“There’s no getting around the lead time,” said Mark Benedict, AM lead at the AFRL. “And most of the savvy manufacturers realize that’s one of the biggest value propositions for additive, that you can get the thing in a week or two weeks instead of potentially six months to a year. “The other caveat there is rarely is the tool perfect the first time.”
For Phase 2 of their research, AFRL teamed up with Boeing and initially used Cincinnati Inc.’s Big Area Additive Manufacturing (BAAM) printer and chopped carbon fiber loaded polyphenylene sulfide material.
After fabricating several mid-scale tools using BAAM, AFRL and Boeing decided to also evaluate tool quality using Thermwood’s Large Scale Additive Manufacturing vertical layer printing technology.
The partners originally wanted a fuselage skin measuring 10x5x2.5 feet (305x152x76cm) made with chopped carbon fiber loaded polyethersulfone. However, they ultimately opted for two 5-foot length tools instead in order to use them to validate tool dimensional stability and vacuum integrity by autoclaving one at 350°F (177°C) and the other at 250F (121C).
“In addition to dimensional accuracy, another concern for us is tool vacuum integrity at 350°F for three, four, or five-hour cures,” Neslen said. “As it turns out, for the particular tools that we’ve printed, Boeing believes they can use these tools with the sealant they’ve identified for more than 75 percent of the (composite) parts they fabricate.”
Mike Matlack, a Boeing materials and process engineer who was his company’s project manager for the AFRL partnership, confirmed Neslen’s estimate, citing the composite tools’ significantly lower cost. While he and AFRL work out the tool durability issue, the company is using the technology for prototypes and low-production parts, “spares and repairs to get fleet readiness back up,” Matlack said.
“Traditionally on those prototype airplanes we need to get quicker to market and that actually cuts our tool fabrication cycle time down,” he said. “If the technology proves out to where we’re getting more cycles than we’re anticipating it could go into a medium-range type of production application.”
On another venture Matlack project managed, Boeing and Cincinnati teamed up with Oak Ridge National Laboratory to make a wing trim tool that achieved a Guinness World Records title of largest solid 3D-printed item.
If the aerospace industry were to adopt large-scale AM to make tools for big parts, are there enough manufacturers making these type of 3D printers? Currently there aren’t but a handful of companies making them.
“I wouldn’t say there’s, at the moment, adequate industrial base capacity for production of tooling but … the market is starting to emerge,” Benedict said.
A U.S. government materials scientist/polymer chemist “who’s really into 3D printing” has devised a practical additive manufacturing process to make aerospace-grade, short carbon fiber-reinforced thermoset composites and tune the materials for desired physical properties.
“In its current form, I would say the technology is suitable for high-volume, small and medium-size components—complicated parts that need to be light weighted—that would be difficult to manufacture conventionally,” said James Lewicki, director of the quantitative polymer aging and degradation laboratory at Lawrence Livermore National Laboratory. “However, if you want to look ahead there are ways to make the technology large-format.”
Lewicki’s technology uses direct ink writing, aka robocasting or robotic material extrusion, in which a paste-like material is extruded from a micro-nozzle onto a print platform and retains its shape without relying on solidification or drying. The printer has control over the high aspect ratio fiber’s orientation in three dimensions and, using computational design optimization, can orient the fibers in each of the printed part’s layers to achieve desired strength and stiffness, as well as thermal and electrical conductivity properties, singly or in combination, that are superior to randomly ordered carbon fibers.
To date, the largest part printed on Lewicki’s printer is about 1500cm3 (49 feet3), although the machine is capable of build volumes on the order of 20,000cm3 (656 feet3).
Current methods of manufacturing small-to-medium parts made of similar materials include fiber winding (for cylinder-shaped components) and hand layup. Large parts are fabricated with hand layup, automated fiber placement and automated tow placement.
A paper published in 2017 in Scientific Reports by Lewicki and his team describes their early work on the printer. Since then, the scientists have been focused on making the technology viable for scale in the manufacturing industry.
“Now we’re in a position where we can print, on a large scale, composites which are somewhat descended from the ones in the paper which are on the order of as strong as temper-6 6061 aluminum and about 70 percent as stiff,” Lewicki said. “The volume fractions of the fiber are much higher, the defects in the composites are minimized and the complexity and the scale of the parts have gone way beyond what we demonstrated (in the paper).”
Lewicki has more stretch goals for his team and his printer.
He is curious whether they can produce a part that approaches the properties of ferrous alloys. He thinks integrating longer fibers may help achieve a material that outperforms the mechanical properties of stainless steel, he said.
“I don’t see many remaining chemical challenges; I think we’ve got that down,” Lewicki said. “What I see are mechatronic and technical integration challenges—those and continuing to scale.
“I would love to print an aircraft wing but that’s years away and would require a partner.”
Connect With Us
