Sparks fly as NASA pushes the limits of 3-D printing technology

Sep 01, 2014 by Joshua Buck
Engineers just completed hot-fire testing with two 3-D printed rocket injectors. Certain features of the rocket components were designed to increase rocket engine performance. The injector mixed liquid oxygen and gaseous hydrogen together, which combusted at temperatures over 6,000 degrees Fahrenheit, producing more than 20,000 pounds of thrust. Credit: NASA photo/David Olive

(Phys.org) —NASA has successfully tested the most complex rocket engine parts ever designed by the agency and printed with additive manufacturing, or 3-D printing, on a test stand at NASA's Marshall Space Flight Center in Huntsville, Alabama.

NASA engineers pushed the limits of technology by designing a engine injector —a highly complex part that sends propellant into the engine—with design features that took advantage of 3-D printing. To make the parts, the design was entered into the 3-D printer's computer. The printer then built each part by layering metal powder and fusing it together with a laser, a process known as selective laser melting.

The process allowed rocket designers to create an injector with 40 individual spray elements, all printed as a single component rather than manufactured individually. The part was similar in size to injectors that power small and similar in design to injectors for large engines, such as the RS-25 engine that will power NASA's Space Launch System (SLS) rocket, the heavy-lift, exploration class rocket under development to take humans beyond Earth orbit and to Mars.

"We wanted to go a step beyond just testing an injector and demonstrate how 3-D printing could revolutionize rocket designs for increased system performance," said Chris Singer, director of Marshall's Engineering Directorate. "The parts performed exceptionally well during the tests."

Using traditional manufacturing methods, 163 individual parts would be made and then assembled. But with 3-D printing technology, only two parts were required, saving time and money and allowing engineers to build parts that enhance rocket engine performance and are less prone to failure.

This video is not supported by your browser at this time.
3-D Printed Rocket Injector Roars to Life: The most complex 3-D printed rocket injector ever built by NASA roars to life on the test stand at NASA’s Marshall Space Flight Center in Huntsville, Alabama.

Two rocket injectors were tested for five seconds each, producing 20,000 pounds of thrust. Designers created complex geometric flow patterns that allowed oxygen and hydrogen to swirl together before combusting at 1,400 pounds per square inch and temperatures up to 6,000 degrees Fahrenheit. NASA engineers used this opportunity to work with two separate companies—Solid Concepts in Valencia, California, and Directed Manufacturing in Austin, Texas. Each company printed one injector.

"One of our goals is to collaborate with a variety of companies and establish standards for this new manufacturing process," explained Marshall propulsion engineer Jason Turpin. "We are working with industry to learn how to take advantage of additive manufacturing in every stage of space hardware construction from design to operations in space. We are applying everything we learn about making rocket engine components to the Space Launch System and other space hardware."

Additive manufacturing not only helped engineers build and test a rocket injector with a unique design, but it also enabled them to test faster and smarter. Using Marshall's in-house capability to design and produce small 3-D printed parts quickly, the propulsion and materials laboratories can work together to apply quick modifications to the test stand or the rocket component.

"Having an in-house additive manufacturing capability allows us to look at test data, modify parts or the based on the data, implement changes quickly and get back to testing," said Nicholas Case, a propulsion engineer leading the testing. "This speeds up the whole design, development and testing process and allows us to try innovative designs with less risk and cost to projects."

Marshall engineers have tested increasingly complex injectors, rocket nozzles and other components with the goal of reducing the manufacturing complexity and the time and cost of building and assembling future engines. Additive manufacturing is a key technology for enhancing rocket designs and enabling missions into deep space.

Explore further: NASA begins engine test project for space launch system rocket

add to favorites email to friend print save as pdf

Related Stories

Recommended for you

Europe sat-nav launch glitch linked to frozen pipe

14 hours ago

A frozen fuel pipe in the upper stage of a Soyuz launcher likely caused the failure last month to place two European navigation satellites in orbit, a source close to the inquiry said Wednesday.

Cyanide ice in Titan's atmosphere

16 hours ago

Gigantic polar clouds of hydrogen cyanide roughly four times the area of the UK are part of the impressive atmospheric diversity of Titan, the largest moon of Saturn, a new study led by Leiden Observatory, ...

Video: Alleged meteor caught on Russian dash cam (again)

20 hours ago

Thanks to the ubiquity of dashboard-mounted video cameras in Russia yet another bright object has been spotted lighting up the sky over Siberia, this time a "meteor-like object" seen on the evening of Saturday, Sept. 27.

User comments : 9

Adjust slider to filter visible comments by rank

Display comments: newest first

antialias_physorg
5 / 5 (3) Sep 01, 2014
test faster and smarter.

What does 'test smarter' even mean? Have test been done in a dumb way in the past?
RealScience
5 / 5 (2) Sep 01, 2014
@AA - In my experience "testing smarter" has been a 'marketing' name coined for putting a bit of thought into testing procedures so that one can extract a comparable amount of data from fewer tests or especially fewer test samples.

This can be as simple as instead of running 100 parallel tests with some variable tweaked to 1, 2, 3, ...100 respectively, running sequential tests with that variable at 50, 25, 37, 31, 34, 33, 32 to find the optimum value requires far fewer tests.

Smart testing can also refer to running tests that do not damage a part first, and then running tests that do not damage critical features, and then running a massively destructive test last, allowing the same part to be reused for multiple tests.

Not everyone does this - parts for testing are usually late, and I've seen people so eager to get data that they just grab parts and start testing, and when new tests are thought up later, they realize that they don't have enough parts to run the new tests.
Iceberg86300
5 / 5 (4) Sep 01, 2014
First of all, to be precise, it's selective laser sintering.

As far as testing smarter, @Realscience, everything you said I'd correct, but I would add the SLS rapid prototypes to the list. These are way cheaper, and produced much more quickly than a conventionally manufactured part. Instead of destroying ten $10k parts, they can destroy ten $500 parts, all the while either identical parts or variants are being rapid prototyped.
RealScience
not rated yet Sep 01, 2014
I agree, Iceberg, that if the final part had to be machined, then testing rapid prototype parts first for tests where they were sufficient would be smart testing.

However in this case the goal is a final part from selective laser sintering, so I would count the use of SLS itself as smart manufacturing rather than smart testing. (Of course the rapid turn-around facilitates smart testing by allowing iterative sequential testing without long delays between tests, so it contributes to smart testing).

What is really cool is that this allows integrating large number of spray elements right into the main parts without having to add them later, which allows building a better engine as well as building an engine faster.
Iceberg86300
5 / 5 (2) Sep 01, 2014
Thanks, missed that part. Honestly I thought it was going to be about testing a Saturn V F-1 engine turbo pump with a vastly better design/modern manufacturing. As the F-1 had been solid modeled from laser scans of a RTF F-1 put in storage and the one in the Smithsonian, and a refurb turbopump tested, I've been hoping they would optimize the F-1, and fix shortcomings (most of us know about the pogo problem, but just because they reduced the pogo vibrations to an acceptable level doesn't mean they solved the problem in my eyes, at least on the Saturn V). I still consider this to be the best liquid fueled engine ever produced. With modern FEA methods the engine could be optimized in itself, as well as optimized for whatever application is needed, including the vehicle interface to practically end the pogo effect. Also, it could be easily scaled to F-1/2 or 3 for use on smaller vehicles. Jeez, the turbo pump alone for an F-1 could probably launch some cube SATs into low earth orbit.
RealScience
not rated yet Sep 01, 2014
Yes, the Saturn V was an amazing engine!

If an improved version could be 'printed' at reasonable cost through laser sintering, it would be amazing again. Low volume production of metal parts with fine internal features is a great use of SLS.

Ad Astra - the meek will inherit the earth, the bold will go to the stars! (Or at least colonize the asteroid belt...)
antialias_physorg
not rated yet Sep 02, 2014
has been a 'marketing' name coined for putting a bit of thought into testing procedures so that one can extract a comparable amount of data from fewer tests or especially fewer test samples.

Yeah - but in my experience that has always been the case. No one has done frivolous tests in the past. Tests cost money. If anything companies have skimped on tests.
RealScience
not rated yet Sep 02, 2014

Yeah - but in my experience that has always been the case. No one has done frivolous tests in the past. Tests cost money. If anything companies have skimped on tests.


I've never seen frivolous tests, but I have seen not-well-planned tests.

I agree on skimping on testing being common. Skimping on testing in the original plans often causes extra testing later (typically something that was expected to be essentially invariant in a part turns out to be surprisingly variable in the test samples, and this is not suspected until many tests for suspected causes have produced mysterious results). By the time the meed for more testing becomes clear, the project is behind schedule and everyone is in a rush so the test plans are often not re-thought with the same care that went into the original planning, leading to sub-optimal testing.
TheGhostofOtto1923
1 / 5 (1) Sep 03, 2014
I just read 'Rescue Mode' by Ben bova about an ill-fated mars mission. The ship gets hit in the carbon nanotube spine by a meteor, and so they printed a replacement. I wondered - how do you print nanotubes?

Anyway it was an interesting example of how replacement parts could be printed in space.