Engineering Concept · Rev E · Sibling project to Groove
A 48 mm round, adhesive-mount sensor that lives inside the fridge, watches for the interior light switching on after a set hour, and fires a piezo siren to deter late-night raids. No app, no NFC, no audio playback — a sense-and-alarm circuit.
The last two are per-part renders of the actual STL files — a print/assembly reference: print both shells separately, seat the PCB on the rear shell's standoff posts, then press the front shell on until its pegs seat in the rear shell's sockets. No glue or fasteners in the base design; friction-fit pegs hold the two halves together.
* limited by the CR2032's own shelf life, not by circuit consumption — see the power budget below.
No phone app, no wireless pairing. Two side buttons set an "armed window" (e.g. 11pm–6am) directly into the MCU's own battery-backed real-time clock. From then on the circuit sleeps at microamp-level current until the phototransistor sees the fridge's interior light switch on. If that happens inside the armed window, the MCU drives the piezo; outside the window, nothing happens — a 2pm fridge raid doesn't trigger it.
The math says over a decade, which is longer than a CR2032 lasts on the shelf before self-discharge takes over (typically 5–10 years). In practice, sleep current is so far below the siren's rare bursts that this circuit will never be the reason the battery dies — swap the cell on a calendar, not based on usage.
| Component | Real part | Price | Status |
|---|---|---|---|
| MCU + internal RTC | ST STM8L051F3P6TR, TSSOP-20 | $0.58 | Verified |
| Ambient light sensor | Senba SMD0805-20 | $0.057 | Verified |
| Piezo siren | TDK PS1240P02CT3, 12.2mm | $0.164 | Verified† |
| Battery cell | QIJEY generic CR2032 | $0.024 | Verified‡ |
| Battery holder | QIJEY CR2032-BS-6-1 | $0.109 | Verified |
| Hour + confirm buttons ×2 | ALPS SKSCLBE010 | $0.148 | Verified |
| Status LED | Nationstar NCD0603R1 | $0.008 | Verified |
| PCB | JLCPCB 2-layer, ⌀ <100mm baseline | ~$0.15 | Est. |
| Enclosure (2-piece shell) | — | ~$0.45 | Est. |
| Adhesive mount pad | — | ~$0.04 | Est. |
| Estimated BOM total | $1.73 | ||
† Sound-pressure rating (dB) wasn't confirmed in this pull — verify against TDK's datasheet before treating "siren" as settled. ‡ Generic/unbranded cell price; a name-brand CR2032 (Panasonic, Murata) will cost more.
What the sensor itself needs to charge, independent of customer acquisition. See Groove's spec for the full formula derivation.
| Channel | Break-even (0%) | Price @ 15% | Price @ 20% | Contribution @ 20% |
|---|---|---|---|---|
| Pre-order | $5.23 | $6.15 | $6.54 | $1.31 |
| Retail / wholesale | $5.88 | $6.92 | $7.35 | $1.47 |
The product itself is cheap to justify — under $7 clears a 20% margin easily. Whether that leaves room for customer acquisition is a separate question, answered below.
Same M values and sourcing as Groove's spec: $6/unit for pre-order/crowdfunding, $70/unit for standalone retail DTC.
| Channel | Contribution @ 20% price | CAC (M) | Gap | Fully-loaded price to still hit 20% after M |
|---|---|---|---|---|
| Pre-order | $1.31 | $6.00 | −$4.69 | $14.04 |
| Retail DTC | $1.47 | $70.00 | −$68.53 | $94.85 |
Same pattern as Groove: at the $6.54 product-only price, neither channel's acquisition cost is covered by the contribution margin available. The $14.04 pre-order price is the one to actually plan around — this product's economics only work through a channel that supplies its own discovery (crowdfunding, marketplace search, word of mouth), not through funded advertising, where $94.85 would be required and isn't remotely competitive for a simple sensor.
Both shells demold with straight pulls except the two button bosses at opposite points on the rim, which are perpendicular to the main pull but simple single-direction side-actions. The light and LED windows are open apertures in this model — production would cover them with a clear IP-rated film to keep the sealed interior dust- and splash-resistant without blocking the sensor or LED. Four standoff posts spaced around the rim (not a full perimeter ring) support the PCB, keeping clearance from the outer wall and avoiding the thin-wall pinch that a tight-fitting ring risks at this scale. The round shell also drops the injection-mold complexity of the rounded-rectangle's four distinct corner radii to a single constant-radius core.