DARPA Backs SpaceNukes Kilopower Test With Active Control Rods

DARPA has given Space Nuclear Power Corporation, which trades as SpaceNukes, an Other Transaction Agreement to build and ground-test a Kilopower reactor run by active control rods, the company announced Oct. 21. Neither the dollar value nor the reactor's power output was disclosed.

The award sits under DARPA's Fast Capacitance program, called Fat Cat, which the Defense Sciences Office runs under program manager Tabitha Dodson, according to the SpaceNukes announcement published by SpaceNews. The goal is to develop and ground-test control technology for autonomous reactor operation in national security missions. Los Alamos National Laboratory, which licenses the Kilopower design to SpaceNukes, will help design, build and run the test.

What KRUSTY proved in 2018, and what it did not

The only precedent is KRUSTY, the Kilopower ground test that ran at the National Nuclear Security Administration's Nevada National Security Site from November 2017 through March 2018. NASA's May 2, 2018 release says the core was a solid, cast uranium-235 cylinder about the size of a paper towel roll, with passive sodium heat pipes carrying heat to Stirling engines that convert it to electricity.

The test ran in four phases. The first two had no power, the third raised power in steps, and the last was a 28-hour full-power run covering startup, ramp-up, steady operation and shutdown. The team also simulated power reduction, failed engines and failed heat pipes, and the system kept running. NASA described the Kilopower concept as capable of up to 10 kilowatts of electrical power for at least 10 years, with four units enough to establish an outpost.

SpaceNukes says the 2018 test used passive control. The follow-on changes that: the new reactor will use active control rod operation. The difference is the point of the award. A passive design settles itself through physics; a reactor meant to run autonomously for a defense customer has to decide, with no one at a console, when to move a rod.

	KRUSTY (2017-2018)	Fat Cat follow-on
Control	Passive	Active control rods
Sponsor	NASA and the National Nuclear Security Administration	DARPA Defense Sciences Office
Test site	Nevada National Security Site	Not disclosed
Power output	Not stated in the release	Not stated

Who is doing the work

Patrick McClure, SpaceNukes' chief operating officer, said the award "recognizes the maturity and continuity of the team that designed, built, and tested KRUSTY in 2018." Chief Executive Andy Phelps said: "There is no substitute for demonstrated technology when American national security objectives are at stake."

The continuity is literal. NASA's 2018 release names Marc Gibson as lead Kilopower engineer at the Glenn Research Center. SpaceNukes' 2026 announcement lists Gibson as its chief engineer and says integration will happen in its Medina, Ohio workshop, which has thermal and vacuum capabilities. The company's headquarters and a workshop with additive 3D printing are in Los Alamos, New Mexico.

What the announcement leaves out

The release gives no dollar value, no reactor power level and no date for the nuclear ground demonstration. It describes phased gates for design, procurement and integration before that demonstration, but not their dates. It also does not say whether the reactor is meant to fly. Fat Cat is framed around ground testing and national security missions, and the announcement does not claim a launch.

A web search on the day of writing turned up no independent coverage of the award beyond the company's own text, so every figure and quote above rests on the company's announcement or on NASA's 2018 account. DARPA's separate DRACO nuclear thermal propulsion effort, led by Lockheed Martin, is a different program.

The next visible milestone is the first design gate, and the company has not said when it falls. If SpaceNukes publishes a power level and a test date, the award becomes a schedule; until then it is a contract with a name on it. Our coverage of GM's sodium-ion battery deal shows the contrast: that story came with a dollar figure and a capacity in gigawatt-hours, and this one has neither. Autonomy is the other thread, covered in our Rivian piece, where the question is likewise what a system does without a person in the loop.