Two openings. One roof model.

Research record

The detail behind
the study.

Source-linked work for this revision. Reported checks describe the stated model or test; physical qualification and applicability to the current retrofit remain separate.

ENGINEERING REVIEW · R3 · ELASTIC DEMAND STUDY

Elastic demand comparison. No strength, stability, containment or manufacturing qualification is established.

Current recessed-shell R2 geometry · 12 T joints · 18 bearing stations · 3 continuous support rails.

All values are calculated under stated elastic and support assumptions. Movement is measured from an assumed fixed post-bottom reference. These are demands, not rated performance.

What changed

The current geometry has two distinct access frames, twelve T joints, twelve roof stock members, three continuous rails and eighteen bearing/post stations. The rails connect both frame groups.

The complete 3102 × 2101 mm comparison footprint carries 101.44 kPa, including loaded closures: 661,115.11488 N downward. Both global first moments are conserved.

Explicit load recovery includes eight 45 mm header-end regions, four 45 mm cut-stub regions and twelve 5 mm fitted-end strips. The four stub regions contribute 9,440.0064 N that would be lost by loading physical stock alone.

Rigid-region routes produce the same displacements but different connection forces. Finite connection flexibility also changes response. Each exported joint row retains its concurrent signed force/moment combination.

Same-load comparisons

One assumption changes at a time except where the case name identifies both changes. These comparisons are not qualified upper or lower bounds.

Verification

74 focused tests, 66 analysis checks and 234 independent topology/load/action checks passed. The independent audit covers all sixteen saved cases, including the influence fixture. Twenty-one geometry source hashes and eight prior CAD evidence hashes matched current local files.

Mesh refinement compares sampled movement, all signed T-joint actions and discrete energy. Analytic beam/offset fixtures, pressure scaling and patch/remainder superposition provide independent numerical checks.

Open engineering scope

Exact CAD member axes and 45 mm joint-face offsets are used. Section stiffness is homogenized from assumed laminates; local orthotropy, bending-twisting coupling, wall buckling and strength are not solved.

Deck, ribs, skins, curbs and closures route load through explicit alternatives but their stiffness and strength are omitted.

All 18 bearings use nominal projected-area centroids and bilateral vertical compatibility. Positive compression is necessary but cannot prove pressure distribution, no lift-off or contact strength.

Baseline roof-to-rail bearing rotations are free. Finite rotational springs are unqualified sensitivities, not evidence of a positive attachment.

Each post is an axial spring to a fixed bottom reference. Post bending/buckling, rail/post joint flexibility, lower frame, floor, bedding and soil deformation remain excluded.

The 190 mm roof/rail vertical separation is metadata only in the out-of-plane model; in-plane axial/rotation coupling and geometric stiffness are excluded.

Finite T-joint springs use dimensionless member-stiffness normalizations. They are not measured joint properties or qualified bounds.

All outputs are concurrent signed elastic demand cases. No fabrication layup, joint strength, allowable, tank rating or manufacturing release is provided.