SpaceX plans to pack 22 Starlink mass simulators inside the payload bay of its first full-scale Starship V3 vehicle for Flight 12, the company’s next integrated test launch from Starbase in Boca Chica, Texas. Twenty of those simulators will mimic the weight and deployment mechanics of future Starlink V3 satellites. The final two carry a different job: they are fitted with cameras pointed back at the ship’s thermal protection tiles, tasked with capturing photographs of the heat shield while it endures temperatures exceeding 1,400°C during atmospheric reentry.
If the cameras survive long enough and the data links hold, SpaceX engineers will have something they have never had before: direct visual documentation of how individual tiles behave under peak heating on a vehicle 50 percent larger than any Starship that has flown to date.
Why V3 changes the heat-shield problem
Starship V3 is physically bigger than the V2 ships that flew on Flights 5 through 11. SpaceX has described a stretched upper stage with a nine-meter diameter and roughly 100 metric tons of additional payload capacity compared to V2, which itself carried about 50 tons to low Earth orbit in its test configuration. A larger vehicle means a larger surface area slamming into the atmosphere at roughly 27,000 km/h, and that translates directly into a bigger, more complex heat shield.
On earlier flights, SpaceX struggled with tile losses. Company updates after Flights 5 and 6 in 2024 acknowledged that tiles detached or were damaged during ascent vibration and reentry heating, exposing underlying steel to temperatures that warped structural panels. By Flights 9 through 11, SpaceX had improved tile attachment methods and adjusted the reentry flight profile, and the ships survived descent in progressively better condition. But those were V2 airframes. V3 introduces new tile geometry across a wider surface, and the margins validated on smaller ships do not automatically transfer.
That gap between simulation and reality is exactly what the two camera-equipped simulators are meant to close.
How the imaging experiment is expected to work
According to details shared by SpaceX and reflected in the company’s pattern of FCC experimental license filings, the two camera simulators will ride inside the payload bay alongside the other 20 units during ascent. After the upper stage reaches its target orbit and opens the payload door, all 22 simulators will be deployed in sequence, verifying the dispenser mechanism that will eventually release operational Starlink V3 satellites on future commercial missions.
The camera simulators, however, are not simply dead weight. Each is expected to carry imaging hardware oriented to capture the Starship upper stage as it performs its deorbit burn and descends belly-first into the atmosphere. The goal is to photograph the windward heat-shield surface during the period of maximum aerodynamic heating, when plasma temperatures around the vehicle peak and the hexagonal tiles absorb the brunt of the thermal load.
One major technical challenge is the plasma blackout. As the ship decelerates through the upper atmosphere, a sheath of ionized gas envelops the vehicle and can block conventional radio transmissions for minutes at a time. On recent Starship flights, SpaceX has partially mitigated this by relaying telemetry through its own Starlink constellation, routing signals around the plasma barrier. Whether the camera simulators will use a similar Starlink relay, transmit on dedicated experimental frequencies, or simply store imagery onboard for later recovery has not been confirmed in any publicly available FCC filing as of early June 2025.
The distinction matters. Real-time downlink would let engineers monitor tile behavior as it happens. Stored-and-recovered data would still be valuable but would require either a controlled landing of the simulators or a retrieval operation, neither of which SpaceX has described publicly for this mission.
The regulatory path still ahead
Every Starship flight requires parallel approvals from two federal agencies. The FCC must authorize the radio spectrum SpaceX uses for telemetry, video, and command links through Special Temporary Authority grants. The FAA Office of Commercial Space Transportation must separately license the launch and reentry, a process that includes safety analysis, environmental review, and coordination with airspace and maritime authorities.
For Flight 12, the FAA’s Starship-specific review process requires SpaceX to submit a reentry trajectory, planned splashdown or landing coordinates, and contingency procedures covering off-nominal scenarios. Because V3 is a new vehicle configuration, the agency may require updated debris and casualty-risk analyses that account for the larger airframe and heavier payload mass.
As of early June 2025, neither agency has publicly posted a completed authorization for Flight 12. SpaceX has targeted a launch window in the coming months, but the timeline depends on how quickly the regulatory reviews close. Both the FCC experimental licensing database and the FAA’s commercial space license index are publicly searchable, giving independent observers a way to track progress in near-real time.
What engineers hope to learn
SpaceX currently relies on computational fluid dynamics models to predict where tiles will experience the highest heat flux during reentry. Those simulations are sophisticated, but they involve assumptions about turbulent boundary-layer transition, gas-surface interaction chemistry, and tile-gap leakage that are difficult to validate without flight data. Temperature sensors embedded in the Starship structure have provided some measurements on past flights, but sensors give point readings at fixed locations. Cameras offer something different: a spatial map of the entire visible tile field, showing discoloration gradients, ablation patterns, and any gaps that open under thermal stress.
If the two simulators capture continuous or near-continuous footage through the heating peak, engineers could overlay that imagery onto their CFD predictions tile by tile. Discrepancies would point to areas where the models underpredict heating, potentially driving changes in tile thickness, material composition, or attachment hardware on subsequent V3 builds.
Even partial results would carry value. Images taken just before and just after the plasma blackout window could bracket the most extreme phase, showing the heat shield’s condition entering peak heating and its state immediately after. Combined with onboard thermocouple data, those bookend frames could constrain the models enough to improve confidence in the design without requiring a perfect, uninterrupted video feed.
What a failure would reveal
If the cameras fail early, if the downlink never locks during descent, or if the simulators tumble and lose their view of the heat shield, the experiment will still produce a result. It will demonstrate how difficult it is to instrument a vehicle’s own thermal protection system from an external vantage point during reentry. That difficulty is not unique to SpaceX. NASA’s Space Shuttle program spent decades refining post-flight tile inspection because real-time imaging of the underside during reentry was never operationally reliable.
SpaceX’s approach of using deployable camera platforms rather than fixed vehicle-mounted cameras is a workaround for the fundamental problem: any camera bolted to the heat shield’s surface would need to survive the same environment it is trying to observe. Separating the camera from the ship and photographing from a trailing or nearby orbit sidesteps that constraint, but introduces new ones around relative positioning, attitude control, and communication links.
The outcome of Flight 12’s imaging attempt will shape whether SpaceX continues refining this deployable-camera strategy, shifts to alternative instrumentation like embedded fiber-optic temperature sensors, or pursues some combination of both on future flights. No public FAA or FCC document ties a specific certification milestone to the success of these two simulators, but the engineering feedback loop is clear: better thermal data means faster iteration on the heat shield, and faster iteration is central to SpaceX’s goal of making Starship flights routine.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
