Q-and-A: Engineer Talks Future of 3D-Printed Rocket Engines

Just last month, Aerojet Rocketdyne tested out a rocket engine fuel injector that was produced by 3D printing. AmericaSpace spoke with Jeff Haynes, a program manager working to develop this technology for space flight purposes. He talked about the first test, what the limitations are for 3D-printed parts, as well as the ramifications for the future of space flight. Photo Credit: Aerojet Rocketdyne / NASA

On July 11, 2013, the aerospace firm Aeroject Rocketdyne and NASA tested a rocket engine fuel injector that was constructed using additive manufacturing, or “3D printing” as it is more commonly called. By all accounts the test was a complete success. Wanting to bring more information on the subject to you, we reached out to the folks at Aerojet Rocketdyne to get some background on what the company has planned for this innovative new technology.

If you are looking for an injector for your engine, one of these diesel fuel injection parts suppliers can help you to find the parts that you need.

AmericaSpace sat down with Aerojet Rocketdyne’s Jeff Haynes, who is a program manager working to develop additive manufacturing technology for the Sacramento, Calif.-based company. Haynes spoke about where the technology is currently, as well as where it could be heading.

AmericaSpace: First of all, Jeff, thanks for taking the time to sit down and talk to us.

Haynes: It’s my pleasure!”

AmericaSpace: Tell us a bit about the fuel injector which was tested recently by Aerojet Rocketdyne and NASA.

Haynes: “That particular fuel injector uses both fuel as well as an oxidizer in the system.”

AmericaSpace: Is it true that the 3D-printed injector worked better than a traditionally manufactured injector?

Haynes: “We used an existing engine system to compare this injector to; the goal was to achieve an equivalent capability in terms of flow distribution of the fluids and things like that. When we tested it for the first time after we printed it, it achieved equivalent metrics to the traditionally manufactured component.”

On the left side of this image is the 3-D printed fuel injector as it looked immediately after coming out of the laser melting printer. The right image shows the Injector after it had been inspected and polished. Photo Credit: NASA / MSFC
A different 3D printed fuel injector, this one tapped to fly on SLS was tested at the space agency’s Marshall Space Flight Center. On the left side of this image is the 3-D printed fuel injector as it looked immediately after coming out of the laser melting printer. The right image shows the Injector after it had been inspected and polished. Photo Credit: NASA / MSFC

AmericaSpace: So it matched a traditionally produced injector?

Haynes: “Correct. That’s in terms of the propellant distribution and variability—not so much the performance.”

AmericaSpace: Given that, are there limitations to the technology? We know that this was just a specific component within the engine itself. Perhaps you can elaborate as to what sort of restrictions still impact the use of this technology?

Haynes: “Sure. Well, the key to the technology is that it has the ability to replicate very detailed features, and in most cases where we combine a lot of details into one assembly, and now you have the opportunity to consolidate that into one large component without having to join all of these details together. So the key that we had to produce was those fine-featured details.”

“Now some of the limitations are that, as you’re building layer by layer with 3D printing or additive manufacturing you have to provide support structures in the areas where the angle changes to more of a flat overhang. So in those areas, where you want to have an assembly and you can’t get the support structures removed, you are limited in terms of what you can produce for a design solution. So if you take a printer production engine and try to replicate it with 3D printing, those are your challenges—if you’re going to have those kinds of geometrical features that you’re going to have to account for. A new design solution is really preferred so that you can go in and optimize the geometry to account for that limitation.”

AmericaSpace: I take it efforts are underway to counter these issues so that you can begin producing more challenging components?

Haynes: “We have a design optimization group for additive manufacturing at Aerojet Rocketdyne that is looking to understand those limitations, and as we get into new design solutions, these are key considerations in our tool box that we’re developing. So our designers, you know, have to open their framework from what they are used to, to understand this new manufacturing process—to be able to accept and adopt it.”

Computer generated imaged of fuel injector NASA image posted on AmericaSpace
Before the component was deemed safe to be tested, engineers scanned it to make sure that it had been produced as advertised. In this case, a computer tomography scan was employed. This image comes from NASA from the agency’s testing of the injector at Marshall Space Flight Center. Image Credit: NASA / MSFC

AmericaSpace: A lot of the 3D printers that most folks are familiar with typically print using a material that is similar to resin. That obviously is not what you are using here. Can you give us a little bit of background regarding the actual materials that are used in this process?

Haynes: “That’s a good point. 3D printing, as far as the mainstream population is concerned, is mostly the plastics and polymer-based systems that you know and can see every day. We’re really focused on the metallic manufacturing, for which we use a powdered metal process. It’s a powdered ‘bed,’ essentially—a 10-by-10 cube that you place powdered metal into and use a laser to build very thin layers, sliced from a computer model. You fuse with the laser; you melt and fuse the layers together to build your part. The alloys that we’re interested in—the ones used in rocket engines—typically are nickel-alloy-based super alloys, for example. We’re also exploring other alloys, such as those using titanium, aluminum, and copper.”

AmericaSpace: Engines of all types are constructed from a wide range of parts and a lot of different materials. Given this, is there the possibility that in the near future we could see a rocket engine that’s constructed almost entirely by additive manufacturing? Could you give us a guess as to when that point might be, when we might see a rocket engine with components produced through additive manufacturing?

Haynes: “Well, the first thing that we’d need to be sure to address is that when you use the term ‘rocket engine,’ there is a wide range of rocket engines—some use turbo machinery to drive them, while others are pressure-driven. Those that use pressure-driven designs are typically simpler in design. For some of these simpler engines, with these 3D printers you could produce one of these engines today almost in its entirety. The problem is that you have to qualify this new material system. It’s not like casting; it’s not like forging; it’s something that is uniquely different that requires its own set of design data associated with it. That really is the key to what we’re doing; it’s really the barrier to implementation. It’s a new product form, so we have to develop the design data that supports that, and that takes time. So, you know, usually it’s 2-5 years to get that kind of a database built up on these alloy systems.”

AmericaSpace: We’d imagine that, like anything used in aerospace, it has to undergo rigorous testing before being approved on actual missions.

Haynes: “That’s accurate. It’s a risk-based approach that we take to everything that we do. So even in terms of the recent injector test that we ran in collaboration with NASA-Glenn, we had to do over a year and a half of material testing of that material that was produced with 3D printing to be able to get to that level of test capability. Once we’re able to demonstrate the material had a robust capability, we went into that component-level test. Then, once you generate that level of data, you’re ready to start putting these things into engines and begin engine-level testing, which will take another 2-4 years to get through that level of validation.”

Aerojet Rocketdyne's Jeff Haynes. Photo Credit: Aerojet Rocketdyne
Aerojet Rocketdyne’s Jeff Haynes. Photo Credit: Aerojet Rocketdyne

AmericaSpace: For those of us that follow developments in this field, we’ve all watched the television show “Star Trek,” and we’ve seen the characters, whenever they needed a new component, walk over to the replicator and order up a fuel injector or whatever other part they needed. Naturally, the process between what is science–fiction and what you do as science-fact is totally different. However, in 2015, a 3D printer, constructed by a company called Made-In-Space, is going to be sent to the International Space Station. Is that something that Aerojet Rocketdyne is working on? Could we one day see astronauts on their way to Mars, who lose an engine component and can print a new one, then go out and do an EVA and replace it? Is that what you guys are trying to accomplish?

Haynes: “That is certainly an interest that both our NASA customer and the U.S. Air Force have—to, once you’re up there, have that kind of capability to be able to replace parts. This would be very beneficial in terms of missions that encounter issues, such as what took place on Apollo 13. Our core focus, however, has been on not only getting them there, but getting them there more affordably.”

“We really see this technology enabling these propulsion systems to lower the cost of access to space. Now once you demonstrate that and you’re able to validate that, I think the in-space activities become viable as well. There are different 3D printing systems, though. The one that we use, as mentioned, uses a powder-based system, which would make use in the micro-gravity environment problematic.”

“What the team over at Made-In-Space is using is a plastics-based design, and then there is also wire-fed additive systems, where you can feed a wire into these energized beam sources to build your parts. Those obviously can be used in zero gravity. I know that there is a lot of work going on toward that, but that’s currently not what we’re focused on. What we’re trying to do is to get the people working on those things, to those destinations, in a more affordable manner.”

AmericaSpace: Is it faster to produce an additive-manufactured component as opposed to a traditionally produced component?

Haynes: “That’s a really complicated answer, as it depends on the complexity of the part that you’re trying to produce. Now if I had a simple part that I wanted to produce today, say a single element, which can be produced if it’s a forged or machined, you have to wait for that forge upwards of about four months. Once you get that forging, it takes about two weeks to make that part so-to-speak. Now if you have a complex assembly like these injectors that have hundreds of details within them—but you can print them in one or two details—then you’ve eliminated that full-value stream of waiting and support time. So, in the case of that one part, once you’ve got the forging, it’s kind of a one-to-one trade off. But if you want to construct a full assembly with hundreds of details in it … like on this injector, we built that with about a 70 percent savings in terms of both time and cost to make that assembly part. While it really depends on the complexity of the part, we’re really looking at the value added to this process where you are able to take these very complex assemblies and you’re now able to print them in one or two pieces.”

NASA image of SLS Marshall Space Flight Center Fuel Injector 3d printed additively manufactured photo credit NASA MSFC posted on AmericaSpace
NASA and its various commercial partners have begun testing and validating 3D printed components on the ground while other 3D printers are being prepared to make the journey to orbit. Photo Credit: NASA / MSFC

AmericaSpace: To that, given that you’ve now scanned and even printed one of these injectors, it would seem that it’d be far easier to just print one of these injectors as opposed to getting them the old-fashioned way. Is this correct?

Haynes: “Yeah, the two-piece injector we printed, we printed each section in about six days-each. Then that has to get assembled and put together in a system. Traditionally? That would have taken well over a year. With all the details that you have to get from your sources-stock, forge, cast, and machine, each one of those, then you have to plate them, engrave (braze) them, join them, and then put them together as an assembly-that takes over a year.”

“However, once you’ve established this process where you can print the two pieces in six days each and then assemble them and then you validate that process—now you’ve got a capability to do that on a repeatable basis.”

AmericaSpace: If you could impart anything to the public that is learning about 3D printing in some cases for the first time, what would you like to tell them is the most important aspect of what this new technology could provide us with?

Haynes: “That’s a good question. A lot of people can buy these machines—well not necessarily a lot of people, as these types of systems run upwards of a million dollars each, but you can buy one and learn how to print parts—but really to understand what you’re producing when you print those parts is the key to what we’re trying to do. We’re trying to gather that fundamental data base, which is currently a new process, a new technology. It’s not a casting; it’s not a forging. So we have to establish a whole new product form with all the data that supports it.”

AmericaSpace: Jeff, thanks for taking the time to chat with us today about what you’re working on and how it could benefit space flight.

Haynes: “No problem, it was great chatting with you!”

As highlighted during the interview, during the Commercial Resupply Services 5 mission, conducted by Space Exploration Technologies (SpaceX) Made In Space will have one of its 3D printers sent to the International Space Station. If all goes according to plan, this mission will take place in 2015. The use of 3D printing technology is on the rise. A number of commercial companies in partnership with NASA are testing 3D printed rocket components. The NASA centers involved include Glenn Research Center in Ohio, Stennis Space Center in Mississippi and Marshall Space Flight Center in Alabama.

Want to keep up-to-date with all things space? Be sure to “Like” AmericaSpace on Facebook and follow us on Twitter: @AmericaSpace


  1. So Jason what does this article mean….You asked the questions but he didn’t completely answer anything….It will take 2 to 5 years to create the engineered drawings templates before testing can begin on space usage?…He said that it was taking 6 days to “print” a part but…didn’t say how long it would take to assemble it…He suggested a cost and time savings of 70% …is that right?

    • Mr. Lovett,
      Article’s Meaning: Haynes provided information about 3D printing. While there were certain aspects he didn’t expound on, most would agree he was very informative.
      Haynes plainly states the amount of time/revenue this process saves, so I’m uncertain why you’re asking.
      Jason Rhian – Editor, AmericaSpace

  2. what I didn’t understand is how does this technology get implemented. You suggested that a part goes bad and like on Star Trek they go and replicate a new part. Well he suggested that yes his clients would like that ability and then he talked about 6 days to “print” and then assemble. What I was hoping to understand was how does this part get replaced in real time….Are the parts monitored by predicative error software suggests a failure in say 2 months and the process is started to “print” a new part or is it likely that one spare part of the engine components would be carried aboard as a spare and then after the brakedown the spare part is then replaced with the new “Printed part?

    I am certainly glad to see that companies are working on these issues..

  3. Tracy,
    You’re taking that humorous reference far too literally. As to the “breakdown” he says that these parts are printed on the ground. So, naturally, printed parts would have to be carried on board as the type of process he’s involved with isn’t able to work on orbit.
    Jason Rhian – Editor, AmericaSpace

  4. Great interview and article Jason! This is definitely a “make a hard copy to show colleagues” AmericaSpace article. Talk of “Star Trek”? In the year 1913 engineers would have thought that THIS was science fiction. Imagine, to be an engineer in 2113.

One Ping

  1. Pingback:

NASA Selects Innovative Technology Proposals for Suborbital Flights

Original Delta Clipper Experimental Team Inducted Into International Space Hall of Fame