Value chain · CLM-0860
In plain terms
An open-source antenna modelling package puts the first side peak of a beam in the right place, to better than three parts in ten thousand, for the one antenna layout tested. A separate finding about how strong it reports one feature to be has been withdrawn, because it reversed when the aperture was reconstructed in an equally reasonable way.
The claim as recorded
LyceanEM 0.1.131 places the first sidelobe of a uniformly illuminated circular aperture pattern within 0.03 per cent of an independent direct summation over the same element positions, 11.5600 against 11.5570 degrees, on the one off-centre 209-element lattice run; no first-null comparison and no comparison on any other lattice is recorded. THE SIDELOBE LEVEL RESULT PREVIOUSLY STATED HERE IS WITHDRAWN: see `limits`. No claim is made about the reproduction of sidelobe LEVELS by this record.
Evidence behind this: Own computation against an independently implemented reference. The artefact is the source and the observation of its behaviour is made here. The tool's output is the measured object; the reference is a direct array sum computed here.
Scope and limits
ONE configuration only - an 8.17-wavelength aperture, uniform illumination, one frequency, one range. The agreement must not be assumed to hold at other aperture sizes, tapers, frequencies or ranges. The element set is a specific off-centre square grid, now stated in `support`. An independent rebuild from the earlier wording, which gave no lattice or inclusion rule, reproduced the continuous-aperture reference to four decimals but could not reproduce the 209-element set, because a centred grid of the same pitch holds 213. The sidelobe LEVEL offset previously stated here is withdrawn and must not be quoted from anywhere: it reverses sign under a lattice any competent reader would build. The POSITION agreement this claim states is a separate measurement and is not withdrawn, but it was compared on the off-centre grid only.
What it assumes
That the comparison reference is correct. Its independence is the point: the reference was implemented here, the tool was not.
Where it comes from
This is the project's own working, shown in full:
Run locally on an NVIDIA RTX 4070, LyceanEM 0.1.131 from PyPI. 1 m circular aperture at 2.45 GHz, 209 elements at half-wavelength pitch d = 0.061182134 m on a square grid built per axis as np.arange(-a, a+1e-12, d) with a = 0.5 m, giving 17 points from -0.500000 to +0.478914 m, a grid off-centre by -0.010543 m, keeping the elements inside the 0.5 m radius (a centred grid of the same pitch holds 213), uniform illumination, observation points at 1000 m. LyceanEM: first sidelobe 11.5600 deg at -17.167 dB. Independent direct array sum over identical element positions and weights: 11.5570 deg at -17.526 dB. Continuous-aperture Airy reference: 11.5388 deg at -17.5704 dB. Discretisation accounts for +0.044 dB of the +0.403 dB total; residual +0.359 dB unexplained ON THE OFF-CENTRE 209-ELEMENT LATTICE ACTUALLY RUN. THAT RESIDUAL IS WITHDRAWN - it reverses to -0.255 dB on a centred 213-element lattice of the same pitch. See `limits`. Sampling plateau table in CALC-037.
Basis: degrees (angular position); dB (level, relative to on-axis peak)
Related records
- Answers question
- QST-016 not published in this release
Cite this
energyaperture.com/c/clm-0860
CLM-0860, Energy Aperture, as of 2026-09-05.
Connected records
Other published records linked to this one. A list, not a judgement.