HWNEWS
Katalyst's Link satellite tumbled days into its chase of NASA's aging gamma ray observatory, and engineers have only months to recover both spacecraft before the rescue window closes.
A satellite NASA hired to save its $500M Swift gamma-ray observatory is now the one that needs saving, and the failure mode is exactly the shape the compressed commercial timeline left room for.
Link, built and operated by the Phoenix-based startup Katalyst Space Technologies, spun out of control about a week after its July 3, 2026 launch while climbing toward Swift, an aging NASA orbital asset that detects gamma-ray bursts. Engineers at Katalyst's control center near Denver now have months, not years, to stop the spin, finish the rendezvous, and tug Swift to a higher orbit before Swift slips too low to grab.
The contract is the first time NASA has outsourced satellite servicing to a commercial company, and the scale is asymmetric. Swift is worth roughly $500 million; Katalyst's contract is $30 million. NASA gave the company less than a year to design, build, and fly the mission. The compression was the point: Swift was already descending, and waiting for a slower procurement cycle meant losing the asset to reentry.
About a week after launch, Link lost two of its three reaction wheels, the spinning masses satellites use to point themselves precisely. Some of its cold-gas thrusters, the small jets normally used for finer attitude adjustments, also malfunctioned, leaving the spacecraft without its normal toolkit for stabilizing its orientation and regaining reliable communications. Rendezvous hardware, robotics, and power systems remain healthy, but pointing at a target tens of meters across from a few kilometers away is the part of a servicing mission that cannot tolerate slop.
The recovery plan turns the spacecraft's main engines into its steering system. Link carries three xenon-fueled electric thrusters, the same plasma engines meant to push it to a higher orbit, mounted on a two-axis gimbal. Engineers are using them to gradually despin the spacecraft, trading efficiency for control. The xenon thrusters are not the right tool for fine pointing, but they are the only working tool that can change the spacecraft's angular momentum fast enough to matter.
"We are not going to recover Link's original attitude control. We are going to do attitude control on the orbit-raise engines," Ghonhee Lee, Katalyst's CEO, said in an interview with Ars Technica.
NASA needed urgency because Swift was about to fall, and the commercial-servicing pitch is explicitly cheaper and faster by design. A one-year clock for a first-of-kind robotic servicing mission compresses exactly the testing margin that catches a multi-axis tumble before launch. Link's specific failure mode, losing two of three reaction wheels and then losing the cold-gas backup, is the kind of single-fault-then-compound-fault pattern that longer test campaigns are designed to flush out.
Katalyst bought NASA a mission that would not otherwise have flown in time, and the company is now working the anomaly with the same spacecraft, same propellant, and same ground team that was already paid for. If recovery holds, the contract still ends with Swift in a stable orbit and a precedent for the next commercial rescue. If it does not, the $500 million asset reenters, and the lesson is that "faster" and "ready" are not the same axis on a servicing timeline.
The watch item is a clock. Swift will be too low to rescue in a few months. The next Katalyst update on Link's spin and comms status is the test of whether the $30 million, less-than-a-year commercial servicing model is the future of orbital asset recovery, or its cautionary tale.