Design · widgets

Micro Tilt Display — the fold, with the hardware

A tiny USB-powered status display: a 0.96" OLED on a friction-hinged pod that stands up to read and folds flat onto a serviceable base. This is the first-proof hinge + wire-path coupon (v0.2) — the riskiest interface reduced to two printable parts. Geometry only; unproven until the bench.

The display pod folding open and closed — FIG · live the assembled coupon **with hardware** — the 0.96" OLED rides in the pod, the ESP32-S3-Zero in the base. **Drag** to orbit; **slide** or hit ▶ to fold it flat → reclined back ~110° to view. Collision-free across the whole sweep (the GIF is the fallback if 3D won't load).

Ghosted shells showing the OLED in the pod and the ESP32-S3-Zero in the base — FIG · the hardware shells ghosted: the **0.96" SSD1306 OLED** rides in the pod, the **ESP32-S3-Zero** in the base (raised here for clarity), the interleaved M3 knuckle hinge between. Soldered leads cross at the hinge axis.

display

0.96" OLED 27×28 mm, active 21.74×10.86, +1.92 offset

brain

ESP32-S3-Zero 21×18 mm, USB-C to the rear

hinge

M3 friction holds anywhere, reclines ~110°, folds flat

wiring

soldered leads slim pod; slack loop at the hinge

geometry only · unproven v0.2 · 7 gates green (incl. swept-collision + ear-wall) · the print is the proof: friction hold, lead survival, and horizontal-barrel printability are bench items

The whole object reduces to one hard question: can a printed friction hinge hold a screen up, fold flat, and route the OLED leads without crushing them? So the first proof isn't the enclosure — it's a cropped pod stub + base cradle + the real M3 hinge, modeled on the calipered ESP32-S3-Zero and 0.96" OLED references.

It took two adversarial review cycles to make it real. The first caught the pod slab sweeping straight through both base knuckles for the entire fold — green gates and a clean render had both missed it. The fix (relieve the slab, supports to the rear) introduced a 1.05 mm-thin hinge ear, which the second cycle caught. Each blind spot earned a computed gate — a swept-boolean for the collision, an ear-wall check for the thin fin — so neither failure class can hide again. A render is never a fit; a green gate is never a print.

The base · where the hardware lives

…and the USB-C cutout

The hinge coupon had no port — so here is the real base: the ESP32-S3-Zero in a serviceable shell with a rear USB-C notch aligned to its port, a bottom plate that carries and releases the board, and the hinge cradle at the back. Drag to orbit; it opens facing the rear so the port reads (the shell is ghosted so you can see the board).

The base enclosure with the USB-C notch and the ESP32 inside — FIG · the base ESP32-S3-Zero (green) in the cavity, its USB-C seated in the rear-wall notch at port height; the bottom plate carries the board (service from below); hinge cradle + nut boss at the rear-top.

geometry only · unproven USB-C notch aligned for a direct/soldered board. Two print-readiness review cycles fixed a 148 mm³ post/plate overlap, a 0.32 mm nut wall, and sub-min-wall pockets — now guarded by a real mesh gate. Open fork: the closed top deck bridges 37 mm → needs print support or a service-direction redesign; the matching pod is modeled below.

The payoff · two proofs, one product

The whole thing — folded and open

The hinge coupon and the base were separate proofs at different hinge heights. The matching pod re-bases the proven pod onto the real enclosure hinge so they finally go together as one object. Drag the slider — it reclines back to ~110° so the screen tilts up toward you, then folds flat to stow. Pod, base, ESP32 and OLED are all here.

The assembled Micro Tilt Display, open — FIG · unified product base + matching pod on the M3 friction hinge; the OLED (screen biased +1.92 toward the hinge) faces the user when open, folds flat onto the apron. prod_swept_collision = 0 across the whole fold.

geometry only · unproven The tapered pod neck threads between the base knuckles — the first full-width cut rammed the pillars + nut boss, caught by a real swept-boolean gate at 64 mm³ and fixed to 0. A solid-slab integration proof; the open-back pod shell + OLED screw capture is the next refinement. Needs a print.