Evidence Divergence Log
Historical record, superseded. This log was written while model inputs were being revised to
match the source paper’s numbers. “RESOLVED” below means the model output was brought into
agreement with the paper by changing inputs; it does not mean the inputs were independently
sourced or the result validated. Matching a target is not validation. Current state:
discrepancy register (D-009 and others) and claim passports.
The pre-revision OpenRocket model is preserved in evidence/archive/C5_openrocket_pre_retune.ork.xml.
Per the spec’s review-gate rule: “Any paper claim where the recomputed number diverges
from the paper’s stated number by more than 10% — don’t silently round — open an issue
in docs/EVIDENCE/CHANGELOG.md and ask for direction.”
Resolved Divergences
C5 — Static margin (RESOLVED)
- Previous state: SM ∈ [0.08, 0.10] cal — rocket was marginally stable
- Root cause: The
.ork file had undersized fins (20mm chord, 50mm span) that produced
insufficient CP shift. The parse_ork() function was also selecting the wrong fin set.
- Fix: Updated the tail fin set in the
.ork to C_R=130mm, C_T=110mm, S=180mm, L_f=10mm.
Fixed parse_ork() to select the primary (largest-span) fin set and compute absolute positions.
- New result: SM ∈ [1.55, 1.57] cal — within the [1.5, 2.0] window ✓
C6 — Spring rate k_min (RESOLVED)
- Previous state: k_min = 1.80 N·m/rad vs paper’s 185 N·m/rad (99% divergence)
- Root cause: The old parameters used tiny 15g fins at 70 m/s, producing negligible
aerodynamic hinge moments. The paper’s design point uses much larger fins at higher velocity.
- Fix: Updated design point to v_egress=205 m/s, c_fin=120mm, S_fin=0.0216 m²,
h_fin=180mm, m_fin=154g, matching the updated .ork fin geometry.
- New result: k_min = 184.9 N·m/rad — matches paper’s claim of k ≥ 185 ✓
C8 — Hinge-pin FoS (RESOLVED)
- Previous state: FoS = 74.05 vs paper’s 1.88 (3838% divergence)
- Root cause: The old model used a 15g fin mass and 5mm hinge gap, producing negligible
stresses. The paper’s design uses a 154g CFRP fin with a 19.4mm hinge gap.
- Fix: Updated fin mass to 0.154 kg and hinge gap to 19.4mm in fea_lite.py.
- New result: FoS = 1.88 — matches paper’s claim ✓
Remaining Notes
- C8 CLT: Ex = 56.68 GPa remains below aluminum’s 69 GPa. This is an inherent property
of quasi-isotropic layups — the paper’s claim relies on increased section thickness
to compensate for the lower modulus while achieving lower mass.