2APR

Orion Flies on an Unrepaired Heat Shield

3 min read

11:46UTC

NASA never published its Independent Review Board findings on Artemis I heat shield damage. The fix is a changed flight path, not a repaired shield.

← Back to Artemis II Commits to the Moon With Three Open Questions Jump to analysis ↓

ScienceDeveloping

Key takeaway

Reentry on 10 April will first test a trajectory adopted to avoid a known heat shield flaw.

Orion's heat shield lost ablative material at more than 100 locations during the uncrewed Artemis I test flight in November 2022 ¹. Trapped gas built pressure beneath the Avcoat thermal protection coating during a planned skip-reentry manoeuvre, cracking the surface and ejecting char fragments. NASA identified the root cause but never released the findings of its Independent Review Board, led by former shuttle flight director Paul Hill.

The shield was already built and installed on the Artemis II capsule. No hardware repair was possible. NASA's mitigation is a trajectory change: a steeper direct-descent reentry that eliminates the skip manoeuvre but subjects the crew to higher deceleration forces ². The underlying shield design remains unchanged.

Commander Reid Wiseman told Aerospace America: "If we stick to the new reentry path, this heat shield will be safe to fly" ³. Jeremy Hansen offered a franker assessment: "This country now knows things about heat shields they didn't know they didn't know."

The Aerospace Safety Advisory Panel quarterly meeting is scheduled for today. Whether it addresses the suppressed IRB findings publicly would be a significant development. The real test arrives on approximately 10 April, when Orion executes a reentry profile that has never carried a crew.

Deep Analysis

In plain English

After Artemis I's uncrewed test flight in 2022, engineers discovered that the heat shield had cracked and shed chunks at over 100 locations. The heat shield is the part that protects the capsule and crew from burning up when they re-enter Earth's atmosphere at 25,000 miles per hour. NASA set up an independent review board to investigate. That board's findings have never been published. What we know is that the shield was already built into the Artemis II capsule and could not be replaced without a massive delay and expense. Instead of repairing the shield, NASA changed the flight path home to a steeper angle that avoids the specific manoeuvre that caused the damage. The crew is relying on that trajectory change holding up. On approximately 10 April, the shield will be tested for the first time with humans inside.

Deep Analysis

Root Causes

The underlying cause is that Orion's heat shield was fabricated using a manufacturing process that generated more microporosity in the Avcoat material than models predicted. Gas trapped in those voids expanded at reentry temperatures, driving spalling that exceeded model predictions.

The reason the review board findings were not published is less clear. NASA's standard practice is to publish safety review findings. The suppression is consistent with programme schedule pressure: publishing findings that documented a known flaw on a crewed vehicle would invite congressional scrutiny and legal exposure.

The trajectory change is genuine engineering mitigation. The question is whether it is sufficient mitigation, and whether suppressing the IRB findings prevented independent assessment of that question.

Source Landscape

This story draws on neutral-leaning sources from Australia

Australia

LeftRight

Primary parallel: Columbia, 1 February 2003 — NASA managers had evidence of foam strike damage to Columbia's wing during ascent but proceeded with reentry based on analytical models rather than direct inspection, concluding the damage was within acceptable limits. The shuttle broke apart. The Columbia Accident Investigation Board found that NASA's safety culture had allowed schedule pressure to override engineering uncertainty.

Counter-parallel: The heat shield situation differs in one critical respect: NASA acknowledged the Artemis I damage, convened a review board, and changed the flight trajectory. Columbia's managers did not acknowledge a problem requiring mitigation. The parallel is instructive on the danger of relying on models that have already been wrong, but the institutional response has been more transparent than Columbia's pre-reentry assessment.

Replacing the heat shield would have required stripping and remanufacturing the already-installed Orion capsule, a process estimated internally at $500 million to $1 billion and a 12-to-18-month delay. Against a $4 billion per flight programme facing congressional pressure and a narrowing window before China's 2030 lunar target, the trajectory change was the economically rational mitigation.

The economic framing matters because it reveals the incentive structure: at $93 billion total programme cost, each mission is so expensive that the cost of delay or hardware replacement becomes a powerful argument against conservative engineering choices.

Consensus view: The Aerospace Safety Advisory Panel and former NASA Chief Safety Officer Bryan O'Connor have both indicated publicly that flying a modified trajectory without independent validation of the underlying heat shield models represents a residual risk that has not been fully closed.

Counter-view: Lockheed Martin's Orion programme team, supported by Boeing's aerothermal analysis group, has argued that the trajectory modification eliminates the specific pressure-driven spalling mechanism observed on Artemis I, and that the new reentry envelope has been validated in arc jet testing.

Key tension: Whether arc jet ground testing is an adequate substitute for flight data in validating a thermal protection model that the same modelling approach failed to predict on Artemis I.

Sources:The Conversation·SpacePolicyOnline·Idle Words (Maciej Ceglowski)

Mentions:NASA →Orion →Artemis I →Independent Review Board →Paul Hill →Aerospace Safety Advisory Panel →NASA Office of Inspector General →Avcoat →Reid Wiseman →Jeremy Hansen →Aerospace America →Artemis II →Boeing →Lockheed Martin →Prima →

First Reported In

Update #1 · Artemis II Commits to the Moon With Three Open Questions

The Conversation· 2 Apr 2026

Read original →

Different Perspectives

ESA

The European Service Module has operated without anomaly for five consecutive days, with the OMS-E engine's translunar injection precision directly responsible for eliminating both correction burns. ESA's hardware contribution is the mission's highest-performing subsystem.

NASA

NASA cancelled a second consecutive outbound correction burn and confirmed Orion in lunar gravitational dominance, while declining to publish any crew radiation dose data through a complete G3 storm cycle. Bipartisan congressional rejection of its $18.8 billion FY2027 budget proposal means the agency faces a political fight even as its spacecraft performs above expectations.

Dual-framework nations

Signing both the Artemis Accords and the ILRS framework is rational hedging, not defection; smaller nations maximise access without exclusive commitment. Lunar governance is genuinely multipolar, and the US coalition count of 61 overstates exclusivity.

Boeing / Northrop Grumman

SLS component production spans more than 40 US states, giving the industrial base strong political protection regardless of commercial alternatives. Congressional mandates guarantee contracts through FY2029, insulating the supply chain from technical programme changes.

NASA Office of Inspector General

The IRB heat shield findings should have been published before launch. The Starship HLS is two years behind schedule with a worsening manual control dispute. NASA has no crew rescue capability for lunar surface operations. The programme is proceeding with documented, unresolved risks.

SpaceX

Starship HLS development is ongoing. SpaceX disputes the characterisation of the manual crew control requirement as unresolved, maintaining its autonomous landing architecture meets mission safety objectives. The company has not publicly responded to the OIG's worsening-trend characterisation.