Rev 3.0-rc1 — Fastening Redesign (Release Candidate)

Overview

Rev 3.0 redesigns the force plate fastening system: stainless 304 throughout (M10×60 DIN 7991 + Parlock all-metal locknut + DIN 125 plain washer), mirror shim configuration (qty 8 — 4 top + 4 bottom), perimeter chamfers added to both plates, and foot piece adjustments (bar Ø55 + base wall serrilhado).

Changes from v2.0.0

Mechanical:

• Foot piece: bar Ø60 → Ø55 mm (machine shop preference); base vertical wall serrilhado (knurled) for sensor threading grip
• Top plate: + perimeter chamfer 45° × 2.12 mm both faces (1/3 of thickness)
• Bottom plate: + perimeter chamfer 45° × 1.5 mm bottom face only (1/2 of thickness)
• Stack height: 43.35 mm → 44.35 mm (added bottom shim)

Fastening (full inox 304 redesign):

• Bolt: M10×50 cl 8.8 → M10×60 DIN 7991 inox 304 A2-70
• Nut: standard hex → Parlock all-metal locknut inox 304 (anti-vibration, reusable >50 cycles)
• Washer: unspec → DIN 125-A plain inox 304
• Shims: 4 (top only, 2 mm carbon/stainless) → 8 (mirror config, 1.5 mm nominal, inox 304 mandatory)
• Torque target: 50–60 N·m → 20–25 N·m (lower preload + anti-seize on cone)

Why:

• Galvanic compatibility with Al 5052-F plates (vs cl 8.8 carbon steel which corrodes)
• Mirror shim brings preload bearing FS on bottom plate from 1.05 → 11
• Parlock anti-vibration without nylon degradation, suitable for cyclic athlete impact loading

Costs

• AL Usinagem (foot pieces + plate finishing): R$ 1,480
• Inox 304 fastening (MercadoLivre): R$ 136,19
• Total Rev 3.0 unit cost: ~R$ 3,750 (vs Rev 2.0 estimate R$ 2,500). Still 8–27× cheaper than commercial (VALD FDLite ~R$ 60,000).

Validation status

⚠️ This is a release candidate. Design is fully documented (specs, structural analysis §8.4, drawings, assembly procedure) but has not been validated with physical assembly + calibration.

For verified production design, wait for v3.0.0 (no -rc suffix). Promotion criteria:

1. Steel tubes 35×35×2 mm received + dimensions confirmed
2. Plates+tubes bonded with structural epoxy (24 h cure)
3. Bonded box section height measured at 4 corners
4. Cell H actual measured (DYX-301)
5. Shim final thickness derived empirically (may differ from 1.5 mm nominal)
6. 8 stainless 304 shims fabricated at measured thickness
7. Test-fit single corner with target torque 20–25 N·m
8. Full 4-corner assembly without visible deformation
9. First calibration: zero-load consistent + known-load with <0.5% error

See the annotated tag message (git show v3.0.0-rc1) for the complete validation gate.

Reference

• Spec and implementation plan kept in private workspace
• Snapshot of previous design preserved at tag v2.0.0
• Branch feat/rev-3.0-fastening-redesign preserves the granular 9-commit changelog