Composite molding compound replaces Invar for lightweight small satellite structures

Complexity of molding V7 design

For the next step, the team decided to mold a housing for their V7 optic (7-inch/18.5-centimeter aperture). “We provided Patz with the CAD model of this Invar part, and they attempted to mold it using the composite compound, which presented a number of challenges,” says Ravizza.

“The V7 housing wasn’t just larger, it was also more complex, with two pieces held together by more than 40 fasteners using four different types of molded-in inserts,” says Bernhard. “The key issue was just molding the design. The compression molding process itself can be pretty challenging, especially to produce very intricate parts and for this type of closed cylinder.” The material properties needed were calculated and became the key performance parameters (KPP) for the molding compound in this application (Fig. 3).

 

V7 molded composite housing. Photo Credit: Patz Materials & Technologies

Patz succeeded in molding the V7 housing. “And it demonstrated that they had the capability to make a complex part,” says Ravizza. “However, what we decided going forward for larger parts, was to make them into separate pieces and bolt them together in a ring-type structure. The idea was to design parts which were more appropriate for molded composites versus machined metal.”

“The V7 part was a middle step in that process and the peak of our complexity curve,” he continues. “From that point forward, and based on that experience, we designed parts that were optimized for the composite molding compound and ultimately, were a better design. We ended up with a more robust way of holding the optic, and it was really a mutually beneficial exercise to transition from a complex design which had some limitations to a  design that was lower in complexity and more functional.”

“I’ll sing the praises of that V7 design, in that it was really a one-for-one replacement,” says Smilo. “There were no geometry changes from the Invar to the V7 composite counterpart and Patz Materials was able to figure out how to do that in a very complex molded part that survived our testing.” It also reduced weight to 1/5^th of  the equivalent Invar part. The two-part Invar housing totaled 6.0 pounds while the composite version totaled 1.3 pounds. “So, it was a success,” adds Smilo, “but process-wise, we knew we could make the manufacturing easier and also better optimize the design’s use of carbon fiber.”

Optimizing for composites

Bernhard summarizes the overall development: “The housing design went from a single piece to a two-piece design and then to the final larger housings, which used a four-piece design incorporating new technology for how to hold the optics, which has multiple benefits.” This four-piece design is being used for LLNL’s V10 optic (10-inch/25-centimeter aperture). A key driver for this housing was simplifying the molded part, which began with tooling (Fig. 3).

  

Figure 3. Successful V10 composite housing. Key performance parameters (KPP) were defined based on the V7 design (top). Tooling was then redesigned (center) collaboratively for the larger V10 housing that LLNL had designed as a four-piece ring (two pieces are shown here, bottom) to achieve a more robust optics support but also simpler tooling which produced better parts. Photo Credit: Patz Materials & Technologies, LLNL

“It’s always challenging when the initial goal is to just replace metal,” says Patz. “You’re designing around subtractive manufacturing which has a certain toolset, set of rules and way of design. But when you’re designing around composites, especially compression molded composites, it’s a different set of rules and toolset. You’ve really got to work with the customer and help guide them in terms of how the materials are used, where their advantages are and then how to design the structure to exploit those to achieve optimum performance.”

Patz explains that the transition from the V7 to the V10 part involved a review of the tooling. “LLNL had given us a CAD drawing of the part and we showed them the resulting mold was expensive because of the complexity that had to be machined into the metal compression molding tool. So, they could see the impact of design options, such as certain types of angles. And then we started to discuss that we could make a simpler mold which would not only be less expensive, but also make better parts. It was a collaborative process.”

“That really does highlight the difference between the V7 and V10 housing,” says Smilo. “The V7 was a blind replacement for Invar while the V10 was an iterative collaboration to really get the benefits from the composite material and molding process. Our goal was something easy to manufacture.”

A parallel development took place within the LLNL monolithic optics team. “The new V10 design also emerged from our research last year into an entirely new way of mounting larger optics,” explains Ravizza. “In parallel to actually producing the composite V7 design, we also used metal components to test this new approach that reduced complexity and made the design more robust. So, the V10 design was really the coming together of improved tooling, molding and this new approach for how we support the optics.”

Divorce from Invar, success via partnership

V10 optics housing molded using PMT-F16+CNT/HM63 molding compound. Photo Credit: Patz Materials and Technologies

Tooling for the V10 composite housing was machined from aluminum, scaled to accommodate its expansion at the 375°F press molding temperature. V10 parts were then molded using a charge weight of 390 grams. They were bolted together and tested successfully, including vibration testing. Weight for the V10 housing was again reduced to 1/5^th that of Invar: 17.5 pounds for Invar to 3.5 pounds using the PMT-F16+CNT/HM63 molding compound.

“Going forward, we can divorce ourselves as much as possible from using Invar,” says Ravizza. “That’s a tremendous benefit to the program not only for cost, but for lead time, because like mass, delivery time is a killer in the small satellite industry.”

Smilo notes that Invar parts are hard to get any faster than 8 weeks, “and it’s typically closer to 12 weeks. But with the Patz molding compound, once we have a design done, we’re able to get these parts turned around with a new mold and whatever post-processing is required by us, in a matter of 3-4 weeks. And when we want to make a second part, we’re simply molding new parts and we’re able to shave lead time down even further.”

He notes that some in the space industry believe there must be a tradeoff with this type of weight and lead time reduction. “They think you have to give something up, but we’re not seeing that with our parts. This composite housing is stronger, lighter, cheaper and faster — and not just by a week or two, but months faster. We’re in the business of trying to make satellites very quickly, in a responsive fashion. Having something that can be provided in a weeks-to-month timeframe versus several months can make the difference for a program win.”

What was the biggest challenge? “It wasn’t necessarily a materials challenges,” says Patz, “but being able to work together to utilize the material in the most optimum way to provide the best benefits.” LLNL agrees. “I think the story here is the partnership between Patz Materials and our team at LLNL,” says Smilo, “and our collaboration to overcome the issues for insertion of these materials onto space platforms.” Ravizza adds, “although the use of composites has a compelling case, the reason why we had success is that we’ve been able to develop a good design, but that required an iterative process and also the Patz team’s knowledge and willingness to keep working on this.”

Future developments, composite radiators

Ravizza believes that his team at LLNL were the perfect partners for this type of development because their role is to do R&D and push the capabilities of new technologies, “and the small satellites that we’re developing are a good proving ground. But we see opportunity to expand this material and approach beyond that.” He concedes there may be more process control and rigor required for composite materials, because of how they’re manufactured versus a metallic part, “but those issues can be overcome with proper quality assurance.”

“We are looking to transition all the metallic components over to composite materials where they make sense,” Ravizza continues. “Not just optics, but the supporting elements that are key to making a functional payload. And we’d like to transfer this technology and get it into the hands of industry.”

Smilo agrees. “As a mechanical engineer who is also very close to the build schedule of the project, I’m interested in how we can replace all involved parts with something that is faster to produce, while still maintaining the kind of reliability needed.” Here, he gives an example, saying, “We’re currently working on a carbon fiber radiator, which takes heat from inside the payload and dissipates that to the outside environment.” To achieve this, the molding compound was modified to use pitch-based fibers instead of the standard PAN-based fibers to increase the thermal conductivity to numbers much higher than aluminum. Patz adds that this can be done at a lower density while retaining all the other required properties for space applications.

Smilo notes that in this case, the composite isn’t being used for its strength properties, but purely for its ability to transfer and reject heat from the payload. “That’s not the classic way to use carbon fiber, but we’re able to benefit from the high thermal conductivity of the fibers and with some intentional design and orientation of the fibers, we can use this material as a radiator panel.”

“Thermal management is a big concern in space systems where temperature can have adverse effects on optics and other equipment,” explains Ravizza. “In space, you are in a vacuum, so the only way to reject heat is through gray body radiation which requires surface area. And copper, though one of the best metallic thermal conductors, is a high-density material. So, for this area of thermal systems, copper would be the analog to Invar. This is a really exciting area that we’ve just started to investigate. We have produced a first example of a radiator using this higher conductivity composite that Patz developed and we’re now testing it.”

“We’ll see more such projects going forward,” says Patz. “We see this molding compound being adopted into more applications as commercial space grows and more and more parts are needed.”