Value chain · CLM-0904
In plain terms
Nothing the satellite distributes to itself is free. Each distributed signal reintroduces a small part of the error the steering trick removed, and the size of that part is just how far its frequency sits from the ground signal's. Two separate effects that looked unrelated turn out to be one rule.
The claim as recorded
In a retrodirective array, any reference distributed at a frequency differing from the pilot's leaves a residual phase error proportional to the FRACTIONAL frequency difference times the structural displacement, once the static geometric part has been pre-compensated. The power beam offset gives a residual (df/f_t).k_t.s, degrading the flex correction by f_t/df = 124.8 for SRC-001's 20 MHz offset. The timing tick's propagation delay gives a residual (f_IF/f_p).k_p.s, degrading it by f_p/f_IF = 247.5 for the same hardware's 10 MHz beat. These are the same relation twice. Only a reference at exactly the pilot frequency, propagating along exactly the line of sight to the target, leaves no residual at all.
Evidence behind this: T0-derivation. A relation unifying two results already computed separately.
Scope and limits
A relation, not a budget. It says how each channel scales; the actual residual needs the displacement, and no displacement figure for a real structure is held. It does not cover the reference PHASE channel, which per CLM-0902 carries full weight and is the exception that makes the pattern worth stating - the rule is about frequency differences, and phase delivery error is not a frequency difference. The two figures are computed from different statements of one frequency plan. SRC-001 gives its frequencies twice, one megahertz apart: f_t/df uses the design beam at 2.496 GHz and f_p/f_IF the as-measured pilot at 2.475 GHz. Both statements give the same 20 MHz offset and the same 10 MHz intermediate frequency, and the alternative beam gives 124.75 against 124.80, so no figure here depends on the choice - but a record that draws on both plans should say so rather than let a reader discover it.
What it assumes
That static geometric terms are pre-compensated in each case. Without that, both residuals are far larger and dominated by design geometry rather than displacement.
Where it comes from
This is the project's own working, shown in full:
CALC-103 covers the power-beam offset and CALC-101 the tick. Both reduce to (delta_f/f).k.s once the static geometry is removed. Basis: dimensionless (fractional frequency difference)
Related records
- Answers question
- QST-013 not published in this release
Cite this
energyaperture.com/c/clm-0904
CLM-0904, Energy Aperture, as of 2026-09-13.