[ehcrea]

Requested by Eren · project thread

Follows the formula validation against the JPE Precision Point articles. Where a calculation now differs from JPE, the calculator says so in a new Differences from the JPE article card, with numbered literature references. The card also appears in the PDF export.

Changes

• Leaf spring: C_y now includes shear deformation (Timoshenko, κ = 5/6). The JPE bending-only value is shown next to it, and a warning appears when b/L > 0.2. For the default 20 × 10 mm spring, C_y drops to 56 % of the JPE value.
• Folded leaf spring: the C_z shear term uses κ = 5/6 instead of the full area. C_z is labelled as valid for a guided end only.
• Hertz contact: f₂ and f₃ now come from the exact elliptic-integral solution (Johnson 1985) instead of the Brewe–Hamrock fit, which was off by up to 12 % near ε_rr = 800.
• Cross spring pivot: adds the end moment factor n = max(|4 − 6a|, |6a − 2|), plus the stress and maximum angle for the chosen pole position. Adds a note on the factor-2 inconsistency in JPE's F_b,ref.
• Materials: corrects PTFE α (13 → 135 µm/m·K) and phosphor bronze λ (208 → 69 W/m·K), each with a reference. Flags implausible JPE values: spring steel and tungsten carbide yield above ultimate strength, and phosphor bronze fatigue strength above yield. A notes card lists all of these.
• Updated DE/NL/TR translations.

Testing

• npm test: 822 passed. New checks: tabulated elliptic integrals, consistency of the Hertz solution with Johnson's δ = p₀·b·K(e)/E*, pressure integrating to F, leaf spring C_y with shear, and the cross pivot factor n.
• Checked the leaf spring, cross pivot, Hertz and materials pages, plus a PDF export, in Chromium. No console errors.

---

Generated by Claude Code