By day, the engineers work on NASA's new Ares moon rockets. By night, some go undercover to work on a competing design.
These dissenting scientists and their backers insist they have created an alternative rocket that would be safer, cheaper and easier to build than the two Ares spacecraft that will replace the space shuttle.
They call their project Jupiter, and like Ares, it's a brainchild of workers at the Marshall Space Flight Center and other NASA facilities. The engineers involved are doing the work on their own time and mostly anonymously.
A key Ares project manager dismisses their design as little more than a sketch on a napkin that won't work.
A spokesman for the competing effort, Ross Tierney, said concerned engineers at NASA and some contractors want a review of the Ares plans but can't speak out for fear of being demoted, transferred or fired.
The Jupiter design is being reviewed by 57 volunteer engineers, Tierney said. Those numbers are dwarfed by NASA's Ares workforce, which has thousands of workers.
The head of the Ares office at Marshall said he can't rule out the possibility that some of his people are involved with the underground program.
“I don't know what people do on their own time,” Steve Cook.
But Cook said he is familiar with the Jupiter project, and he's not impressed. NASA informally reviewed plans for the rocket last fall and determined the idea to be flawed.
Meanwhile, he said, work on the Ares I rocket is so far along that the first test flight is less than a year away.
The debate reflects disagreement over the direction of U.S. spaceflight as NASA prepares to retire the shuttle in 2010. By 2015, the agency plans to begin orbital flights with Ares I and a companion heavy-lift cargo rocket, Ares V.
Besides being a simpler, more powerful system, backers say, the Jupiter rockets would save NASA $19 billion in development costs and another $16 billion in operating costs over two decades.
Steve Metschan, an engineer and former NASA contractor who supports the Jupiter team accused NASA of suppressing information that shows Jupiter would perform better than Ares.
“Our concern is that by the time everyone figures this out, we will have destroyed our heavy-lift system,” said Metschan. “At the end of the day, all we're asking for is an independent review of all this.”