Source Review: NASA FSP 2024 — Fission Surface Power Project

Summary

Claim 1: "40 kWe electrical power output is the demonstrator-class target"

Verdict: Consistent Why: Establishes 40 kWe as the baseline demonstrator scale. Our calc requires 500 kWe (a 12.5x larger system), which is acknowledged as requiring either clustering or a separate larger-scale design. The 40 kWe figure is the program-of-record's near-term target.

Claim 2: "Mass constraint: under 6 metric tons" (40 kWe FSP target)

Verdict: Novel supporting Why: Gives a concrete 150 kg/kWe target ratio. Our calc used 250 kg/kWe (more conservative). Codex confirmed in pass-02-audit that "later concept work also discusses roughly 10 t for 40 kWe" — i.e. actual achievement may be 250 kg/kWe, validating our estimate. The 6-t-target-vs-10-t-actual divergence is itself informative.

Claim 3: "Three Phase 1 contracts at $5M each awarded 2022" + "Target launch pad delivery: early 2030s"

Verdict: Consistent Why: Establishes the program-of-record FSP development timeline. Our BAU 25-yr buildup has M4 (FSP installation) at year 7 of program start — slightly ahead of NASA's "early 2030s" framing if we assume program start in 2026, which is consistent.

Claim 4: "10-year operating lifetime without human intervention"

Verdict: Consistent Why: Direct support for our calc's 10-yr FSP lifetime assumption (one of the few components where my lifetime number is directly anchored to a published target).

Claim 5: "Nuclear reactors can operate in permanently shadowed areas and generate continuous power during lunar nights (14.5 Earth days)"

Verdict: Merits investigation Why: The framing of "FSP enables lunar-night operations" is a quasi-political framing that elides the alternative (PV + storage) cost comparison. For our calc, FSP is the chosen architecture, but the cost relative to a large PV-plus-storage alternative is not separately computed. Merits a follow-up on PV-plus-storage cost for a 500-kW industrial-scale lunar operation.

Cross-reference

• Anchors the energy-supply milestone of any sustained lunar manufacturing base.
• The 250 kg/kWe ratio used in our calc is conservative vs the 150 kg/kWe NASA target; consistent with realistic-achievable per Codex.
• Pairs with duchek-2024-fsps-falcon-heavy (350-kWth microreactor) as the larger-scale design point.
• Codex anti-hallucination check: all quoted text appears verbatim in the extract.md.

Abstract

NASA's Fission Surface Power (FSP) project extends the agency's Kilopower program (ended 2018) into a 40 kWe demonstrator class suitable for sustained lunar operations through the lunar night. Phase 1 awarded three $5M contracts in 2022 to commercial partners for initial reactor concept designs constrained to <6 metric tons total mass and 40 kWe output, with a 10-year operating lifetime requirement without human intervention. A Phase 2 solicitation was issued in 2025 with a target launch pad delivery in the early 2030s. NASA frames this as foundational lunar infrastructure: a small 5-10 kWe reactor can power one habitat or a rover charging station for full lunar day-and-night operation, while a 25-50 kWe reactor would serve multiple habitats plus commercial and industrial activity. Lockheed Martin, Westinghouse, and a third partner (later Rolls Royce / Brayton / GE for power-converter sub-contracts) are the Phase 1 awardees. The published mass constraint of 6 metric tons has been exceeded in at least one published conceptual design (see duchek-2024-fsps-falcon-heavy).

Key claims

• 40-kwe-target: "40 kWe electrical power output is the demonstrator-class target"
• 6-ton-mass-constraint: "Mass constraint: under 6 metric tons" (deployment requirement)
• 5-5m-phase1-contracts: "Three Phase 1 contracts at $5M each awarded 2022"
• 10-year-lifetime: "10-year operating lifetime without human intervention"
• early-2030s-delivery: "Target launch pad delivery: early 2030s"
• power-class-context: "40 kW can power approximately 33 U.S. households on average"
• enabling-shadow-and-night: "Nuclear reactors can operate in permanently shadowed areas and generate continuous power during lunar nights (14.5 Earth days)"

Reviewer notes

Anchors the energy-supply milestone of any sustained lunar manufacturing base. For a manufacturing base consuming O(100 kW) - O(1 MW) of continuous power, the FSP-class reactor at 40 kWe is one to two orders of magnitude below industrial-scale, implying 3-25 reactors per manufacturing site (or a much larger reactor not yet in NASA's program of record). The $5M Phase 1 figure is not the lifetime program cost — it's seed funding for concept design. Notably absent: total program development cost; cost per kWe; delta-cost vs solar-PV-plus-storage; how a TAI-grade fabrication push compresses development. Load-bearing for q5: provides one of the few authoritative dollar-and-mass anchors for a single piece of lunar surface infrastructure with a published technology development plan. Cross-references duchek-2024-fsps-falcon-heavy (the 350-kW(thermal) microreactor variant).