CALC-075

What electric propulsion does to the restated institutional spread

Feeds CLM-0683. Answers QST-004. Basis: GBP2024/MWh, 20-year asset life, 3 per cent real - matching CALC-064's basis exactly.

Why it was needed. CALC-067 identified launch cost and transfer architecture as the largest real driver of the 19-37x spread. Electric propulsion is the one architecture change that moves that driver by a large, source-stated amount. Substituting it in tests whether the spread is an artefact of NASA's chemical-transfer assumption.

Inputs.

Locator note, and it is a repair that propagated correctly. CALC-056 recorded that CLM-0408 and CLM-0410 mistook SRC-012's PDF indices for printed page numbers, and that the 62 per cent quotation is on printed p.74. CLM-0683 cites "SRC-012 PDF p.91" - explicitly labelled as a PDF index. Those are the same page under two labelling systems, and this claim disambiguates where the earlier ones did not. Worth naming because it shows a repair reaching a later claim rather than stopping at the one that was corrected.

Working, transcribed.

NASA RD1  564.0 x (1 - 0.62)  =  214.3 GBP2024/MWh   [NASA's own 62 per cent]
NASA RD1  564.0 x (1 - 0.668) =  187.3 GBP2024/MWh   [the 66.8 per cent at CALC-056]
NASA RD2  921.7 x (1 - 0.69)  =  285.7 GBP2024/MWh   [NASA's own 69 per cent]

Re-running the max/min against Frazer-Nash BEIS FOAK at 28.1:
  214.3 / 28.1  =   7.6 x     [RD1, NASA's own reduction]
  187.3 / 28.1  =   6.7 x     [RD1, the CALC-056 reduction]
  285.7 / 28.1  =  10.2 x     [RD2]

Against the previously established spread:
  RD1   19.0 - 20.1 x   ->    6.7 - 7.6 x
  RD2   32.8 - 36.5 x   ->         10.2 x

Result: the spread falls from 19-37x to roughly 7-10x. A real, large reduction - roughly threefold to fourfold - and not a closing of the gap.

State what the 7.6x and 10.2x are ratios OF. Each is NASA's restated design, with NASA's own electric-propulsion saving applied, divided by Frazer-Nash BEIS's restated FOAK figure with no equivalent sensitivity applied to it. It is therefore an asymmetric ratio by construction, and its residual is an upper bound on the true one rather than an estimate of it.

Conclusion. A defensible electric-propulsion architecture removes most, but not all, of the institutional spread. The remaining 7-10x sits inside the "genuine, unquantified architecture and scope difference" category that CALC-062's companion claim CLM-0386 had already identified as the irreducible remainder, and this record does not resolve it.

Where this fails. - One-sided by construction, as set out above. If Frazer-Nash's own figures would also fall under an equivalent sensitivity, the true residual could be smaller still, and that is not established here. The claim says exactly this. - It does not net off the radiation-degradation power loss quantified at CALC-078. Applying that would move the result in the opposite direction, by a smaller amount. - The 66.8 per cent figure is the CALC-056 model output and is presented alongside NASA's 62 per cent. They are not two independent confirmations - the CALC-056 model was built from NASA's own inputs - and the same trap is what produced the manufactured corroboration recorded at CALC-078 below. - VER-441 CONFIRMED 2026-09-04.

Cite this