NASA and SpaceX just finished late-2025 wind tunnel testing on the Super Heavy Version 3 booster that will support Starship Human Landing System development for Artemis III. The tests took place at NASA’s Ames Research Center in California’s Silicon Valley, using both a transonic and a supersonic wind tunnel to measure the aerodynamic forces and moments the rocket can face during atmospheric re-entry.

Why it matters right now: the whole Artemis III plan is banking on a rocket that can be guided reliably through re-entry to return for refurbishment and reuse. NASA and SpaceX wanted more information about both steady aerodynamic loads and unsteady, potentially vibration-inducing “buffeting” as the vehicle comes back to the launch site. Their wind tunnel data is designed to flow into how the flight software will guide the rocket during re-entry.

This is not a one-off science project. NASA’s Ames work in 2024 already set a baseline, and this new test series builds on previous wind tunnel testing at the same facility. NASA is also using the agency’s broader unsteady aerodynamics experience from earlier programs. In particular, similar testing at the Ames Unitary Plan Wind Tunnel supported adding strakes to SLS for Artemis II, so the organization is treating what it learns as a compounding capability across multiple programs.

The partnership structure is also a tell. Even though Starship Human Landing System is designed and built by SpaceX, NASA is collaborating with commercial industry partners for access to specialized testing facilities like the Ames wind tunnels, plus the technical expertise needed to set up testing and analyze results. On the NASA side, Manish Mehta, discipline lead engineer for the HLS Plume and Aero Environments team at NASA’s Marshall Space Flight Center in Huntsville, Alabama, ties the effort to past wind tunnel workflows: NASA has experience setting up and analyzing tests across the space shuttle, SLS, and Orion spacecraft, which helps move Super Heavy Version 3 testing efficiently from concept to usable data.

What changed on Super Heavy Version 3 is the engineering reason wind tunnels had to matter more. SpaceX’s Starship consists of a 33-engine first-stage Super Heavy booster and the second-stage Starship. Version 3 brings upgrades including new propulsion systems and new Raptor 3 rocket engines, plus major thermal and structural changes: no engine section skirt, individual engine shrouds, and an integrated large-scale base heat shield. Aerodynamics also shift materially, with three gridfins on the booster instead of four, and each fin now 50% larger. Another upgrade is an integrated hot stage replacing a previous single-use protective interstage.

Given these changes, NASA and SpaceX needed visibility into both steady and unsteady aerodynamic environments during re-entry. Jayanta Panda, an unsteady aerodynamics subject matter expert at NASA Ames and part of the Human Landing System Plume and Aero Environments team, frames the difference in plain terms: steady aerodynamic forces are the smoother effects of air flowing over the surface, like during ascent. Unsteady aerodynamic forces are harder to predict, where air hits areas at less predictable times and can create buffeting and vibrations. In board-level language, that is the difference between “modelled loads” and “loads that surprise you,” and surprise loads are exactly what complicate guidance, control, and structural margin decisions.

So the testing setup is built to bracket the relevant flight regime. NASA and SpaceX used a 1.2% scale model of Super Heavy Version 3 in the transonic wind tunnel and the supersonic wind tunnel at Ames. The transonic tunnel runs at Mach 0.2 to Mach 1.4, where Mach 1 is about 761 miles per hour. The supersonic tunnel covers Mach 1.55 to Mach 2.5. In late 2025, they used both tunnels to measure steady and unsteady air flows on Super Heavy surfaces, collecting resulting wind tunnel data on steady forces and moments. That data helps predict how the rocket reacts to atmospheric forces during re-entry so flight software can guide the rocket effectively. Separately, the unsteady pressure data is used to understand the environment as the vehicle re-enters Earth’s atmosphere, feeding into a software analysis of loads on the rocket.

Zoom out and you can see the strategic stake. NASA expects Starship Version 3 and Super Heavy to form the basis for the Starship Human Landing System for Artemis III in 2027, and for a later crewed lunar landing in 2028. Wind tunnels are not glamorous. But they are a direct input to safety-critical guidance and structural analysis. And in a reusable-rocket world, re-entry behavior is where a lot can go wrong quickly: the rocket is coming back hot, stressed, and depending on accurate modelling to avoid chasing surprises. For executives watching aerospace execution, this is a reminder that “testing” is not a box-check. It is a risk-reduction pipeline that turns new vehicle configurations, like Version 3’s engine, thermal, and fin changes, into software-ready knowledge before the mission timeline forces real decisions.