CALC-066

First-of-a-kind against nth-of-a-kind, across two independent studies

Calculation · Cited by CLM-0381 , CLM-0389 , CLM-0390

What this is

This is the project's own arithmetic, published in full so it can be checked rather than taken on trust. It is not a source. Every input it uses is cited to a record, and the conclusion it reaches only carries as far as those inputs do.

Feeds CLM-0381. Answers QST-004. Basis: dimensionless ratios of native-currency figures.

Script. restate-lcoe.py SECTION 7, sub-heading (c), emitted at line 717. CLM-0381's `src` says Section 4(c); its `support` says Section 7(c), which is right.

Why it was needed. Learning-by-doing is the standard explanation offered for why a first-unit cost estimate should not be compared with a mature one. If it were large enough, it would explain part of the spread. Measuring it against two independent studies of the same design settles whether it can.

Inputs.

Quantity	Value	Source	Stated or assumed
Frazer-Nash BEIS FOAK p50 / NOAK 5th	GBP66.30 / GBP49.70 per MWh, 2018 prices	SRC-021	STATED
Frazer-Nash ESA FOAK p10 / NOAK 10th p50	EUR106 / EUR88.5 per MWh, 2022 prices	SRC-078	STATED
Frazer-Nash ESA FOAK p90 / NOAK 10th p90	EUR206 / EUR155.5 per MWh	SRC-078	STATED
That the two studies model the same CASSIOPeiA design	-	both are Frazer-Nash CASSIOPeiA studies	STATED by the sources

Working, transcribed. Ratios of native figures, so no deflator or FX enters this record.

Frazer-Nash BEIS   NOAK(5th)  / FOAK p50  =  0.750    25.0 per cent reduction
Frazer-Nash ESA    NOAK(10th) p50 / FOAK p10  =  0.835  16.5 per cent reduction
Frazer-Nash ESA    NOAK(10th) p90 / FOAK p90  =  0.755  24.5 per cent reduction

Result: a 16.5 to 25.0 per cent reduction from first unit to fifth or tenth, consistent across two studies published a year apart.

State what these are ratios OF. Each is a NOAK figure divided by a FOAK figure from the same study, in the same currency, at the same basis year and the same hurdle rate. Nothing is restated. That is what makes them clean - the deflators and FX of CALC-060 cancel entirely, so this is the one record in this section that survives unchanged if every recalled rate in CALC-060 turns out wrong.

Conclusion. A real and consistently measured effect that is an order of magnitude too small to explain a 19-37x spread, and which does not touch the core comparison at all, because that comparison already holds unit count fixed at FOAK on both sides. Ranked fifth of six drivers.

Where this fails. Two studies by the same contractor on the same design; not an independent replication in the usual sense. And the ESA pairing is p50-against-p10 in one row and p90-against-p90 in the other, which are not the same comparison - the first mixes percentiles. That is the claim's own construction and is transcribed as found, not smoothed.

Cite this

energyaperture.com/calc/calc-066

CALC-066. The working above, not a source document.