For three years the National Aeronautics and Space Administration has been testing a variety of 3D-printed rocket engine parts. But in a recent test, NASA has come closer than ever to building an entire rocket engine—one that could potentially power a Mars lander—solely from additive manufacturing.
Turbopumps, injectors, and valves are engine parts that work together to get a rocket airborne, and they represent about 75% of the parts needed to build a 3D-printed rocket engine. NASA has 3D-printed and tested all three individually with great results, as Fortune has previously reported.
In a new test conducted this month, a NASA team at the federal agency’s Marshall Space Flight Center in Huntsville, Ala., connected the 3D-printed parts together so they work the same as they do in a real rocket engine. What they found was that the 3D-printed engine produced more than 20,000 pounds of thrust—what would be needed for a Mars lander—and withstood temperatures of 6,000 degrees Fahrenheit.
“By testing the turbopumps, injectors, and valves together, we’ve shown that it would be possible to build a 3D-printed engine for multiple purposes such as landers, in-space propulsion or rocket engine upper stages,” said Elizabeth Robertson, project manager for the 3D-printed engine just tested in Huntsville, in a release. NASA also posted a video of the test on YouTube.
MORE: Google and NASA Show Off Blazingly Fast Computer
The parts themselves were manufactured out of powdered metal through a selective laser melting 3D-printing process, a standard for the 3D-printed industry. But for an agency looking to fly to Mars, making the most of its roughly $17 billion budget is a priority, and that’s where this new 3D-printed engine test is another giant leap for NASA.
By 3D printing the turbopumps, the injector, and the valves, NASA was able to design each part with demonstrably fewer components than typically needed. For instance, NASA’s 3D-printed injector—the basketball-size part that sends hydrogen gas and liquid oxygen into a combustion chamber to create the thrust needed to launch a rocket—is just two parts, compared with the more than 200 parts that make up a traditionally manufactured injector. Valves, which normally take more than a year to build, took just few months. That saves both time and money, an important benefit as NASA begins crafting its next class of rocket engines to power its deep-space missions starting in 2017.
“The advances in the technology are finally getting to the point where we can see parts additively manufactured for demanding NASA applications,” Dale Thomas, associate technical director at the Marshall Space Flight Center, told Fortune last year. “We’re never going to get away from the traditional manufacturing process. But additive is going to have some real game-changing benefits.”
Sign up for Data Sheet, Fortune’s daily newsletter about the business of technology.
For more Fortune coverage of NASA, watch this video: