Case file · Launch Complex 39B
Challenger, 28 January 1986
Walk Pad 39B as it stood that cold morning, then step through the records: what was known before launch, who knew it, and what the inquiry found later.
Kennedy Space Center, Florida · 11:38 a.m. EST
Best on a desktop browser with a graphics card. Reading mode is available inside.
The question
What was known before launch, who had access to that information, and what did the later investigation establish?
Before you begin
This case is about the loss of Challenger and its seven crew members: Francis R. Scobee, Michael J. Smith, Ellison S. Onizuka, Judith A. Resnik, Ronald E. McNair, Gregory B. Jarvis and Christa McAuliffe. The reconstruction stops at 73 seconds. The breakup is not shown, and no one's final moments are recreated.
Inside the case
Four ways into the same morning
Move between chapters in any order. The time dial and knowledge view travel with you.
Chapter B · Historical reconstruction
The Place
Launch Complex 39B as it stood that morning, with the sun where it was at 11:38. Every object says whether its size is sourced or estimated.
Chapter C · Explanatory model
The Mechanism
A booster field joint taken apart step by step: tang, clevis, and the two O-rings that had to seal it in the cold.
Chapter D · Archival record
The Record
47 sourced claims you can pin, compare and filter by what was public, what was internal, and what only the inquiry established.
Chapter E · Four kinds of time
The Timeline
From 1977 to the launch day, with each event's audience. Move the clock and the knowledge view changes with it.
The film
The launch, T−8 s to T+73 s
The main engines ignite at T−6.6 s and the boosters at T−0. Four cameras follow: the perimeter fence, a close-up of the right booster's aft joint, a tracking view and the press site. The film ends on the “after the inquiry” view, which adds two findings of the Presidential Commission at that joint: black smoke from T+0.678 s and a flame from T+58.788 s. The flame was on the side facing the external tank, so from this angle the tank hides it. Flight path, smoke and joint position are approximate.
- 1280×720
- 99 s
- 7 shots
- Holds at T+73 s
Reconstruction rendered in Unreal Engine 5. Not archival footage. No sound.
360° views
Stand where people stood
Drag to look around, or use the arrow keys. The sun is where it was that morning. In the two launch views the clock is frozen at the stated second.
312 m south of the pad at eye level, facing north toward the stack.
Real ground
Built on survey data and NASA's own models
The ground comes from US Geological Survey lidar: the pad, flame trench, crawlerway, canals and shoreline are where they really are. The service towers come from NASA's Shuttle-era model.
Teaser
Pad 39B, before the launch
Six shots in the morning light: over the sea, the fence, the pad apron, the access arm, from the air, and the press site at 135 mm.
- 1920×1080
- 40 s
- No launch
- Light at 11:38 EST
Reconstruction rendered in Unreal Engine 5. Not archival footage. No sound.
The desktop edition
A walkable desktop edition built in Unreal Engine 5, with real terrain from USGS lidar, NASA's pad model, a guide voice that reads the records, and the launch with four cameras. It is in testing and not yet available to download.
Walk
- W A S D to walk, mouse to look, F to fly
- 1–6 jump to stops: fence, apron, platform, access arm, east of the stack, booster joint
- Click a part to hear what the records say about it
Launch and time
- L starts the launch, which holds at T+73 s
- C switches cameras
- V turns the time dial: “launch day” or “after the inquiry”
How this was made
- Code, scenes, narration script and source research were produced with AI (Claude, made by Anthropic) from a written brief, and are labelled as such. No AI-generated imagery is used.
- Every claim cites an online record with a locator. Short quotes only; records are linked, not copied.
- Five labels keep things apart: archival record, historical reconstruction, explanatory model, editorial interpretation, unverified detail.
- Three knowledge views: the public record at a chosen moment, documented internal information, and the later investigation. Nothing found later leaks into the launch-day view.
- Named people appear only as authors or subjects of records. The case does not infer motives or assign blame.
Limits
- The scenes are reconstructions, not documentation. Many sizes and positions are estimates and are marked as such.
- Terrain comes from a 2018 survey, 32 years after 1986; the pad was modified in between.
- Weather, haze and the ice recorded on the pad that morning are not fully depicted. Plants are not Florida species.
- Quoted passages still need a word-for-word editorial re-check, and no Shuttle historian has reviewed the case yet.
- Not affiliated with or endorsed by NASA, the National Archives or any archive cited.
Main sources
- Report of the Presidential Commission on the Space Shuttle Challenger Accident (Rogers Commission), NASA History Office web edition
- NASA Image and Video Library: 51L-S-154 to 157, 51L-10147, S86-38627, KSC-86PC-0310
- NASA 3D Resources (Shuttle stack and Launch Complex 39 models)
- NASA Graphics Standards Manual, NHB 1430.2 (January 1976)
- USGS 3D Elevation Program, 1 m lidar
Every claim inside the experience links to its record. Found an error? Corrections · Our standards