Aerojet Rocketdyne building new RS-25 engines for recertification testing

by Philip Sloss

First all-new engine to certify subsequent set of six flight engines

Following Retrofit 2 using E0528, development engine 0525 will take the testing duties for the Retrofit 3 series. While retaining their Shuttle-era designations, both development engines are largely becoming full rebuilds; for Retrofit 3, with the exception of its powerhead E0525 will have all of its major components replaced with newly-produced hardware from the next set of engine components to be delivered from their assembly lines.

“For most components we’re getting three [test] samples, some we’re getting two or one, it just depends what it is,” Bradley said.

“Powerhead, for instance, you only need one. There are different rules that we go by that you have to test certain parts to get them certified and so that just builds up. We’ll test Retrofit 2, we’ll take it apart. In the meantime we’re building up Retrofit 3, we’ll put it in the stand.”

“Throw the Core Stage [Green Run] in there in the midst of that, because we have to avoid that testing,” he added. “Then based on the results we get from that we will already have been building a cert engine. The cert engine is basically new, the nozzle is new, the powerhead is new, all the hardware, pumps and everything, that’s basically a brand new engine.”

“And then after that’s done we have our design certification review with NASA, to make sure this is what we said we were going to do, this is how we thought the engine ought to operate, yes it did, and then we’ll be certified after that.”

Credit: Aerojet Rocketdyne.

(Photo Caption: The powerhead of Engine 2063 is lowered over the main combustion chamber in one of the final assembly decks at AR’s Stennis facility. A similar build-up will take place at Stennis for the certification engine, the first all-new engine to be built under the production restart contract. Both the powerhead and the MCC for the certification engine are being completed at AR’s Canoga Park facility prior to shipment to Stennis for final assembly.)

“We try to bring things in as early as we possibly can, because if you find problems you want to find them early and so that’s the idea why we’re bringing in all these things early,” Bradley noted. “But all the parts are different, they’ve got different requirements.”

“Some of them are heat related, so maybe you need [a number of] tests on them. Turbopumps you need time, you need seconds on them, so they’re all different. We’ll try to put a generic test program together that will fulfill all those requirements and we’ve got some extra stuff where we can get some risk mitigation and get some early warning signs on some of the hardware.”

“We’re always overachievers; when it comes to the cert program we’ve got more time than we originally committed to, and that’s just a good thing,” he added.

The A-1 Test Stand was itself retrofitted in 2018 to support gimbaling RS-25 engines mounted in the stand and at the time some of those tests where the engine is moved in the stand were scheduled for the Retrofit 2 series; now, those tests are looking like they will be performed on Retrofit 3 and the certification engine.

“[In the] Shuttle days it was tougher,” Bradley noted. “We’ve got real high-fidelity computer models now that you can gimbal. They’re movable, they’re dynamic models so you can see where all your clearances are.”

“We don’t trust those completely so probably we’ll end up doing one dry gimbal test for everything and then we’ll do another one on a different engine just so you didn’t get lucky and so we’ll probably just do a couple,” Bradley noted that the plan isn’t finalized yet, but they aren’t going to do gimbal checks in every test series.

“When you do a gimbal test, you move it one degree check all the clearances, move it one degree check all the clearances, it could take you a week. So we don’t think we need to do that.”

“I think what we decided is that we gimbal Retrofit 3, a complete dry gimbal, then do the cert unit as a double-check,” Bradley added. “That’ll be a quicker double-check. We’ll do that in the test stand and we’re set up to gimbal there anyway, so that’ll be an easy one.”

The testing verifies internal clearances within a single engine. “[It’s] not engine-engine [clearances], [it’s] component to component — make sure that duct isn’t bumping into an actuator or something like that,” Bradley explained.

“We’re just going to do it a lot more efficiently. We can make all the measurements before we gimbal, then we’ll dry gimbal it, make sure that the computer model matches, we’re good to go. That’s just another example of how the technology has come along and we take advantage of that, and just save a lot of time building engines.”

Credit: NASA.

(Photo Caption: A new four-piece RS-25 nozzle jacket is machined. Aerojet Rocketdyne currently has an assembly line of seven nozzles in various stages of production at Canoga Park. The first new nozzle will be installed on E0525 for the Retrofit 3b test series.)

New RS-25 engine hardware is beginning to line up at Canoga Park to complete the retrofits, certification engine, and flight engine set that are part of the current contracts. Additional sets of the Retrofit 1 pogo accumulator assemblies and MCCs have already been completed. The second set has already been shipped to Stennis for installation on E0528 for Retrofit 2.

Another set of pogo and MCC units have already been constructed for the certification engine, which will be the first full, new engine build. Units farther down the assembly line will be installed on the six flight engines that are a part of the existing production restart contract.

The largest component of the engine is the nozzle, at a recent visit to the Canoga Park facility, an assembly line of six of the seven units that AR is contracted to build under the first SLS-based engine contract were in various stages of production. Nozzles are built starting with the bell-shaped structural jacket.

The nozzle jacket is now being assembled from four pieces, as opposed to the thirty-seven it took for SSMEs. After the four pieces are welded together, nine large rings called hatbands are welded to the outside of the jacket for hoop strength. Over one-thousand cooling tubes are then added to the inside of the jacket.

The tubes are brazed to the jacket in an oven; nozzle 6001, the first production restart unit, recently completed this process. The nozzle is also “stacked” with its forward and aft manifolds that helps to distribute a branch of the hydrogen flowing through the engine while it is running to the nozzle cooling tubes before it is circulated back for combustion.

The second new nozzle, 6002, has all of its cooling tubes fitted to the jacket; after quality control inspections it will go through the brazing and stacking process. Nozzle 6003 has all of its hatbands welded to the jacket and is moving towards coolant tube installation. Three other already-assembled nozzle jackets were seen going through earlier phases of production, all the way up to nozzle 6007.

Although the first nozzle is progressing through production, the Retrofit 3 test series was divided into two parts (Retrofit 3a and Retrofit 3b) so that other hardware can be tested without waiting for the nozzle. “We couldn’t quite make it for the start of three, so we actually split that program into two halves,” Bradley said.

“You can do that with the nozzle because you don’t need as many tests on it, so that’s OK. Some hardware you can’t, you’re just going to have to wait because you need all twelve of the tests or ten tests.”

“[For the] nozzle, you don’t need them all, just as long as you get some tests in there,” he added. “You don’t need a lot of starts on them but you do need to test them so you see what it’s doing to the engine.”

Credit: Aerojet Rocketdyne.

(Photo Caption: An AR brochure showing some of the parts that will use additive manufacturing on the production restart engines.)

The certification engine, which will include a complete set of newly built engine parts, will follow Retrofit 3 into the test stand. It will serve as a new ground test engine and will perform the final hot-fire test series for recertifying RS-25 production. The last major component to enter the certification testing series is the powerhead.

“The powerhead is a welded assembly which includes the hot gas manifold, all three injectors, the heat exchanger, and miscellaneous ducts,” Bradley said in a follow-up email. “All of those parts are welded together at the Canoga Park facility prior to shipment to Stennis for engine assembly.”

The first two new hot-gas manifolds are well into construction at Canoga Park. The already-completed first one which will be a part of the heart of the certification engine is already being assembled into the full powerhead.

The three injectors consist of a main injector and two preburners (one oxidizer, one fuel). The completed main injector for the certification engine was seen under protective covering during the recent Canoga Park visit; that will be welded to the hot-gas manifold along with the two preburner injectors on-site. (The heat exchanger had already been welded to the hot-gas manifold at the time of the visit.)

“The preburner injectors are complete and ready to install, the heat exchanger has already been welded to the hot-gas manifold. The other minor components such as fuel, lox and coolant ducts are being fabricated and will be installed after the injectors are welded to the hot-gas manifold.”

The MCC is the other core element of the engine, which will be joined with the powerhead at Stennis. “Fabrication of the MCC for the certification engine has been completed. The MCC will be shipped to SSC for engine assembly after special test instrumentation and some minor bolt-on parts have been installed.”

Welding of the second hot-gas manifold was partially completed at the time; that unit will help form the core of the first production restart flight engine.

Also supporting Core Stage Green Run

Aerojet Rocketdyne is also supporting the upcoming critical tests that will culminate the SLS Core Stage Green Run campaign underway in the B-2 Test Stand at Stennis. Although the propellant loading and stage firing tests are focused on testing the stage and the four former SSMEs installed in the Core Stage are all flight-proven, AR plays a big role in the test campaign.

The Core Stage schedule that runs through Stennis for the Green Run campaign is the critical path to get to the Artemis 1 launch, and the schedules for the other SLS elements and the Orion spacecraft that will be sent to the Moon are being coordinated to the availability of the Core Stage at the Kennedy Space Center launch site. After the hot-fire test is satisfactorily completed, the stage will be refurbished and put on the Pegasus barge for delivery to KSC.

The largest block of refurbishment work after the hot-fire test is on the core’s RS-25 engines. “We had a certain amount of refurbishment work between each Shuttle mission. It took about thirty days if you were in a hurry and we were working two shifts and all that,” Bradley explained.

“So we looked at all those things that we used to do when it came to the Core Stage because now we’re going to be critical path again. We go from having sixteen complete engines, we’ll be critical path so we’ve got to take time out of the critical path.”

“We looked at all those changes and saw OK based on our experience, a million seconds [of run time], what do we not need to do,” Bradley said. “There were certainly things that were overkill.”

Credit: NASA/Joel Kowsky.

(Photo Caption: NASA Administrator Jim Bridenstine tours the engine access platform level of the B-2 Test Stand at Stennis on February 10. The first working article of the SLS Core Stage is being used for the Green Run test campaign there before traveling to KSC to be the first flight article. The nozzles of the four RS-25 engines are covered by yellow bags.)

Reducing the turnaround time for the engines from their last shutdown to the next ignition was part of a 2018 technology demonstration for the U.S. Defense Advanced Research Projects Agency’s (DARPA) Experimental Spaceplane Program (XSP). AR assembled two AR-22 engines for XSP using heritage SSME hardware and the new Honeywell engine controller unit, and one of them was test-fired ten times in the span of ten days in mid-2018.

XSP was effectively terminated in January when prime contractor Boeing withdrew from its role; however, AR is leveraging the “ten in ten” technology demonstration in an effort to reduce the refurbishment time after the Green Run hot-fire. The AR-22 experience is applicable to the initial post-shutdown period of time.

Once physical access to the engine is possible after the stage and its surroundings have been inerted, the engine is purged with dry air to get the moisture out of it. “There are different places where you’ll physically take a fitting off a plug and you’ll put a drying hose on there or tubing and then they’ll do that,” Bradley said.

“We’ve got measurements of dew point so we know when we’re dry and we know when we’re not, so that the challenge was to find that process that’s going to get you those dew point checks drier and drier, faster and faster,” Bradley said. “So that’s directly transferable.”

“I think we use they call it missile-grade air, it’s just really dry air because it’s a steam engine, right? We’re burning hydrogen and oxygen, it turns to steam.”

“After you shut down the engine you’ve got all that steam in there but then you’ve got a cold soak,” he explained. “Most of the engine is cold, and so it freezes everything and you’ve got that water stuck in there and now you can’t start it because you’ve got a bunch of orifices that are closed off and you don’t know what that’s going to do.”

“You know you’re going to have some hot spots in the engine so it has to be dry, we’ve known that from the beginning [for] any hydrogen engine. We’ve worked out through the years what’s the most efficient way to dry it and you can imagine there are different ports you can go in.”

“Some of them we’re putting air in, some of them we’re draining air out, and then we’ll just measuring the dew point, we’ll measure how much moisture is in it and when we get down to a certain amount you know you’re good and dry,” he noted.

“Even in the Shuttle days that was a pain, sometimes it would take ninety-six hours of drying or something long like that so we were optimizing things in the Shuttle days. For the AR-22 I think they went even a little bit further and found some ways to dry the engine faster, that was the big challenge.”

“The checks that we need to do are pretty quick — spin a pump, put a borescope in it at periodic times — but the drying just physically takes time to draw the water out of there.” AR was able to meet the ten firings in ten days objective, getting the time between test firings down as low as seventeen hours by reducing the post-shutdown drying period down as low as six hours.

Prior to getting into drying out the components, an immediate post-shutdown engine purge is run to evacuate residual propellant. “You want to evacuate all that hydrogen and oxygen out,” Bradley explained.

“The components are still hot and if you don’t evacuate it out you can have localized spots where you might have some burning or a pop of hydrogen or something like that which wouldn’t be very good. So from the very beginning of the program, we’ve got a helium purge because we’ve got helium onboard. Right after you shut down, those purges open up and evacuate all that stuff out of the engine.”

“It doesn’t take long. So that’s mainly the purges to get the propellants out. Not as much for moisture, but it does help since you’re getting the propellants out.”

Credit: NASA/Joel Kowsky.

(Photo Caption: NASA Administrator Jim Bridenstine is given a tour of engine access platform on the B-2 Test Stand on February 10, surrounded by the SLS Core Stage RS-25 engines and the base heat shield of the stage’s boattail. The administrator is standing under the manhole of the boattail, one of multiple access doors leading into and providing work access to the engine section and boattail of the stage.)

Once the engine is dry, then a series of checks need to be made and any issues identified would need to be addressed prior to the next firing. Since the plan would be for the next ignition to be a few seconds before Artemis 1 lifts off, some of the overall engine refurbishment work will be deferred or “traveled” to KSC where it can be done in parallel with other launch preparations.

“Several of those changes we were going to make in the Shuttle program already, we just ran out of time, the Shuttle program ended,” Bradley noted. “So those were easy, we were able to rationalize those; and then there were others that we looked closer at.”

“I think at first we said it was going to take us sixty-three days and then we got it down to forty-two days and then we got it down to twenty-one if we do the traveled work or something like that so we took big chunks out of that part of it as well. Not sacrificing any reliability, there’s still checks in there that some people would say we don’t need.”

“I tend to be conservative, I’m usually the one that wants to leave stuff in, it’s cheap insurance,” he added. “But even I was able to look at many of the things that we were doing and say we don’t need to do that anymore. They were checks that we were doing since the Eighties and they persist.”

As their name implies, the flight engines are typically only used for flights which minimizes wear and tear. In this case of a development program’s initial stage test in the test stand on the ground, there aren’t enough ground test engines to fill out the needs of a Core Stage, so the four flight engines are being used.

Stage Green Run testing was originally planned to be more expansive and AR established its plans conservatively in terms of maintenance and support for the RS-25 engines. The engines were last used in a Space Shuttle Program in an operational mode; now they will be used to support the SLS Program that is still in development.

“We had a ‘six-six, three-three’ requirement,” Bradley said. “That means the first two flights they had to be good for six tests or [firings] without any real changes in how we had to inspect them. The next two [flights] would be three.”

“We only picked that out because we didn’t know the requirements, that was really early in the [SLS] program when we were putting them together and matching all the hardware. We didn’t know so we just took a best guess.”

“At one time there were going to be two Core Stage tests on the engine and so we put an abort in for both of those, that gets you up to four,” he explained. “Then we said put in an on the pad abort for the flight and then the flight, so we came up with six. Then we said after you do the first two you’re not going to be doing two [ground tests] anymore so we said three.”

Related Articles