Now that the first of NASA’s Space Launch System (SLS) Core Stages has shipped from the factory, prime contractor Boeing is refining production plans for the second flight article and continuing negotiations with the civilian space agency for future production. The production team at the Michoud Assembly Facility (MAF) in New Orleans is applying lessons learned from the process of completing the first Core to building the second one while NASA and Boeing are planning to build more.
NASA and Boeing have agreed to begin ordering long lead elements for the third Core that the agency prizes for the final flight in a series of launches for the high-priority Artemis 3 mission. Once the long lead items like machined structures and early parts arrive at MAF the current estimate is that production in the factory will take three years.
With the long lead orders placed, there should be enough margin to produce the rocket stage in time to fly Artemis 3 by the December, 2024, deadline; however, before that work can accelerate a new contract between the agency and the contractor still must be agreed to. NASA decided late last Summer to sole-source a production and evolution contract for future Core Stages to Boeing and negotiations on the terms are expected to continue through much of 2020.
With the experience of a full build under its belt and seeing improvements the production quality so far of the second Core Stage, Boeing is hoping to get the go ahead to establish a supply chain and a production line around its existing tooling and ramp-up to a production rate of one unit every eight months, or three Core Stages every two years. For now, Core Stage-2 is the last one under contract with delivery currently scheduled for Spring of 2022.
Applying lessons learned to second build and production plans
Even before the first build left the factory earlier this month, Boeing was incorporating lessons learned to production of the second flight article which will fly as a part of the SLS vehicle for the Artemis 2 lunar test flight. The prime contractor is also using the knowledge gained from the build for planning production of future Cores.
“The agency and the program were frustrated due to the lack of progress and our inability to hold schedule,” NASA SLS Program Manager John Honeycutt said recently. “That said though I’m fully aware that we were building the system to build a lot of these Core Stages and we’re putting the system together in parallel with building the first one.”
“Any time you do first-time operations, a) you go slow to make sure you do it right and b) stuff can go wrong,” John Shannon, Boeing’s Vice President and Program Manager for SLS, explained when the first Core rolled out of the factory on January 8. “Right up until the night before [rollout] we were doing things that we had never done before, these first-time operations.”
(Photo Caption: Core Stage-1 rolls out of the factory at Michoud on January 8 with much of the production team that assembled and tested it nearby. In this aerial view from a drone, the stage is also flanked by the buildings where it was put together. Building 110 is the high-bay where vertical manufacturing, assembly, and other processing occurs. To the left behind the workers, Building 103 is where the stage elements were first outfitted with electronics, wiring, and moving parts, and then where horizontal final assembly occurred. In the distance on the eastern end of Building 103, the newer high-bay Building 115 is also used by SLS for initial structural welding.)
Before Boeing could fully start production at MAF, new tooling had to be built to assemble the structures for the large, crew-rated rocket; during production, work instructions for putting the equipment together had to be established. Stuff went wrong in both areas.
“The issues that caused us to be two years late were associated with the new tooling that we used to weld the vehicle together and some issues that we had with that process,” Shannon said. “We had never done friction stir welding for material that was that thick and we learned some lessons from that and I’m happy to say the second Core Stage went together flawlessly with no issues so we’ve gotten past that problem.”
“The other issue that I think caused us some difficulty was we really underestimated the complexity of building the engine section, which is the bottom of the rocket that holds all of the propulsion elements and all of the TVC and hydraulic elements,” he added.
In addition to assembly and integration of the first flight article, structural qualification articles were assembled at MAF for structural testing performed at the SLS Program’s home at the Marshall Space Flight Center in Huntsville, Alabama. During the ups and downs assembling, integrating, and testing all the hardware needed to qualify the stage for flight and then to fly it, Boeing and NASA were capturing feedback on all the little things and the big things that they could do better.
“After action reviews, we have value stream maps, we do events after,” Amanda Gertjejansen, Core Stage-1 Operations Manager for Boeing, said. “I have a VST team, value stream team.”
“They’re a little group that sits on the floor and they just capture lessons learned and corrective actions for the future. There’s a quality and a ME and an IE on the team, a quality engineer, a manufacturing engineer, and an industrial engineer.”
“The manufacturing engineer is going to figure out the problem and then the industrial engineer is going to figure out the flow and save time from doing it or if it’s worth it, [whether] the fix is worth the savings that it’s going to give us,” she explained. “[Our] final assembly team we’ll all sit in a room before we go to Stennis for a day. We’ve been doing it throughout [the build] to make sure we capture ‘if I had this kind of tool, this would have made it a little better.'”
(Photo Caption: The three remaining sections of Core Stage-1 in April, 2019, in the SLS final assembly area at MAF; from left to right, the forward join, the LH2 tank, and the engine section/boattail. The orientation and sequence of final assembly of these elements was a late change that saved several months of schedule. Boeing plans the same horizontal final assembly for Core Stage-2.)
With the build of the second Core transitioning from assembly to integration, NASA has already seen obvious improvement in both the efficiency and the quality of the work. “I’m seeing a twenty-five percent savings in Artemis 2 as far as labor rates go,” NASA Deputy Administrator Jim Morhard said in December.
“Just the engine section, the labor rate for that is going down by forty percent and that’s the most complex part of it. The rework hours, it’s improved by ninety-seven percent. The baseline work hours have gone down by forty percent.”
“So you’re seeing already with just Artemis 2 you’re seeing a huge improvement,” Morhard added. “Will that always be that way? No, at some point that’ll flatten out, too. We’re in the ramp-up stage now and the hope is we’re going to start seeing the bell curve, we’re going to be on the other end of it.”
Shannon explained that the team now has the experience of a full build and those lessons learned were applied to the work already completed for the second vehicle. “The biggest thing is that our tools are dialed in now and we’re not having to make changes to the tooling,” he said.
“As we built the structural test articles (STA), as we built the first Core Stage, not only has the team learned how to do different things but the tooling has been perfected for doing those tasks.” In addition to making changes to the tooling for welding and bolting the elements of the rocket’s structure in between the first and second builds, some of the infrastructure at MAF was upgraded to improve efficiency; for example, better environmental controls were added to extend work windows for tasks with temperature and humidity requirements and time limits.
“And then it’s a lot of just mundane things,” Shannon added. “We spent a lot of time drilling holes the first time that we built Core Stage-1 and you think that shouldn’t take long but we put in all these brackets for all these wiring harnesses in and we drill a hole for every single one of those. We learned wherein the sequence to do that drilling for maximum efficiency.”
“Also we learned how to build a lot of subsystems off of the vehicle and then literally fly them in with a crane into the vehicle,” he said. “We don’t have the constraints of ‘I can only get ten people inside the engine section.'”
“I can have fifty people working on subsystems outside the engine section. When their particular part is finished we pick it up with a big crane and we lower it into the engine section and we install it and that is a much faster way to do it than the way we did it the first time which was really a bespoke way of each wire harness coming in, each bracket, and making sure that we knew how to build it.”
“If you’ve ever put together something for your kids for Christmas and you had to put two of them together, the first one you’re reading the instructions very carefully, you’re laying out all the parts,” he added. “For the second one it’s just ‘I know how this fits together’ and the production system is of use to you then and you can get it done much faster.”
“So training of the team, tooling being dialed in, and the fact that we’ve just learned a lot of simple lessons on the way to put it together that has really helped us to our benefit on Core Stage-2.”
(Photo Caption: The liquid hydrogen (LH2) tank for Core Stage-2 is removed from the Vertical Assembly Center (VAC) welding tool at MAF last Fall. After welding the structures for the first Core Stage, NASA and Boeing made changes in 2018 to the VAC and added air-conditioning hookups to allow welding operations to run more smoothly. The LH2 tank was the last Core Stage-2 structure to be welded.)
Additional production improvements have been in development for implementation on the second build. Application of the thermal protection system (TPS) on the propellant tanks was partly automated, partly manual for the set produced for the first launch, which included test and flight articles.
Spray-on foam insulation (SOFI) was applied robotically to the barrels of the tanks and then the hemispheric domes were sprayed manually. Originally the dome sprays were planned to be done robotically, but were set aside to get a certified process in place for the first tanks; now for the second build, engineers have completed development of the automated dome spray.
“We have the enhanced spray system we’re doing for the acreage on the domes, we’ve swapped that from a manual application to an automated application, which was huge,” Michael Alldredge, NASA SLS TPS Subsystem Manager, said. “It took us a few weeks to do the hand application for domes, now we’ve got it down to twenty minutes.”
In other cases, Alldredge said TPS applications were switched from one process to another based on a better understanding of the overall work context. “We learned a lot about what works and a lot about what we sat back and said ‘don’t ever do that again,'” he said.
“There were a number of applications that were planned as pours that we swapped over to sprays because it gave us flexibility and there were also some that we did a hand spray on that and we said ‘hey, this would make a lot more sense to pour on’ based on orientation, based on parallel work.”
“As with anything the first time you do it you learn a lot,” he added. “For TPS, this rocket is very hard to insulate in certain areas. There’s a lot of war stories, a lot of scars that we’ve definitely already moved it to our engineering and already said ‘aha, we’ve got to fix this in the paperwork going forward to take advantage of those lessons learned.”
“Some of them we’re just having to fight because there’s no other way to do it, so we knew it was tough going into it and it’s still tough, but others we’ve been able to try to capitalize on some [improvements].”
Third Core production dependent on contract negotiations
The knowledge gained from the first build experience is being put into plans for future production, but executing those plans has to wait for a contract. Although NASA needs three Core Stages to play their part in the launch of the three Artemis missions to achieve the Administration’s goal of landing U.S. astronauts on the Moon by the end of 2024, only two of them have been bought.
Boeing’s development contract established early in the SLS Program only covers two flight articles. NASA has authorized the purchase of long-lead materials and parts for a third vehicle, but Boeing is not contracted to put those pieces together. “Not yet,” Shannon said.
“We got approval and funding from NASA for spares for Core Stage-2, just in case in the build we have some issues we’ll have some spares available to us, and long lead items for Core Stage-3, which was a fairly specific set of hardware. Mostly big structural elements and avionics parts and that’s where we are right now.”
The latest step in the regulated U.S. federal procurement process for the new contract was for NASA to decide to sole-source it to Boeing; a formal justification notice for the sole-sourcing was publicly released in mid-October following the endorsement of the documentation in late August/early September.
Negotiations on the terms of a “production and evolution” contract are expected to take most of this year, if not longer. “The Marshall team is working on the contract to allow us to do the full build for [Core Stage] Three and go all the way out through Twelve, so that contracting work is still to be done.”