Over on the saveitforparts YouTube channel, Gabe has recently posted two videos where he attempts to receive the Artemis 2 signal. His setup consists of a surplus satellite dish inside a geodesic radar dome at his "Sandland" radio observatory, a 3D-printed feed, a HackRF One SDR, and various LNAs, including a dedicated S-band unit from LMA Scientific. He used GPredict for tracking and SDR++ for spectrum analysis, targeting the expected downlink frequency around 2216.5 MHz.

The main challenges were the capsule's low elevation angle from his location in Minnesota, rapidly changing orbital elements that made TLE-based tracking unreliable during the trans-lunar injection burn, and the fact that all telemetry is encrypted. During his first overnight session, he was only able to detect what appeared to be an extremely faint carrier at approximately 2216.49 MHz, which is consistent with the expected Doppler-shifted frequency, which disappeared when the dish was moved off-target. In a second session timed to catch a handover between NASA's Goldstone and Canberra Deep Space Network stations, he received a noticeably stronger carrier signal and even observed sideband activity, though still not strong enough to resolve any modulation detail.

He notes that NASA's original citizen science RFP called for ~9 meter dishes, far larger than his ~2.5 meter setup, and that the capsule also uses a laser communications system for high-bandwidth data. The Canadian Space Dashboard and DSN Now websites proved useful for predicting optimal observation windows during ground station handovers.