Decisions — Chosen vs Rejected

Every architecture slot recorded as a decision — what was picked, what was rejected, and the reason, traced back to the power and research pages.

One row per slot. "Rejected" here means considered and not chosen, not bad — most of the rejected options are perfectly good parts that lost on one specific axis.

Board topology

Board P must be separate — it is a reusable switched-15 V sink module and its bring-up sequence requires testing it completely alone before anything is attached.

The MCPCB daughterboard was rejected on the thermal budget's own numbers: that page puts the plain-FR-4 ceiling at roughly 1 W per LED site and reserves metal core for the concentrated-COB case (3.6 W in one 13.4 mm package). This design runs 0.19 W per site. A third board would buy nothing thermally and would cost a second interconnect carrying the driver's ~440 kHz switching node down a wire harness.

The split boundary is still designed in — the array is a self-contained block on four nets (LED_P, LED_N, NTC_SENSE, GND) — so the enclosure phase can lift it onto a daughterboard later without touching anything else.

LED

CCT decided it. This is a fire/candle lamp; 3000 K warm white is the aesthetic requirement, and it is the only place a warm bin coincides with real JLCPCB stock (399,813 units). Every 5730 candidate is 4000 K or cooler, which reads as daylight, not flame.

The COB was rejected twice over: all three CXA1304 SKUs show zero stock, and at 3.6 W in one package it would force the MCPCB the board split just rejected. The stock-available fallback (C22398559) has a scanned datasheet with no machine-readable Vf or flux data — not something to design a thermal budget around.

3-series, not 4-series. Headroom analysis called 4s "tight". Against this part's datasheet Vf ceiling of 3.2 V, a 4s string is 12.8 V, putting the buck at ~88% duty by the datasheet's own equations — far outside the AL8860's recommended 25–75% DSW band. 3s plus ballast lands at 77%: 2 points above the band rather than inside it, and far below the 98% DSW(MAX). See Board L for the worked figures — the older 85%/74% pair used Vout/Vin without the diode and resistive terms.

Current sharing between parallel strings

led-candidates lists exactly these three. One driver per string means eight AL8860s and eight complete passive sets on a board whose entire output is 5.5 W — absurd. Vf-binned procurement is a supply guarantee the JLCPCB catalog cannot make.

Ballast costs 1.03 W (19% of delivered power) and that is the honest price of the choice. What it buys: strong sharing within a matched lot — not unconditional safety. At the datasheet's full 2.6–3.2 V Vf window the hottest branch would still reach 142.5% of the LED's 90 mA absolute maximum, which is why incoming lots are gated on the ≤0.30 V string-spread acceptance bound (issue #35); within that bound the datasheet worst-case branch peak stays at 98% of the absolute maximum (see board-l's current-sharing analysis against the primary-confirmed 306 mW limit).

33 Ω specifically is a two-sided optimisation: larger ballast improves sharing but pushes the duty cycle further past the driver's recommended band; smaller ballast does the reverse. 33 Ω puts nominal duty at 77% — accepted 2 points above the 25–75% DSW band for the sharing it buys. It is not "the largest value that keeps duty under 75%": no ballast value on this rail does that, because 3s alone is already at 69–81% across the Vf bin. If Gate 4 shows a duty-band problem, 22 Ω pulls nominal to ~72% at the cost of ~40% worse current sharing.

LED driver IC

Driver input rating vs a conditioned clamp screening point

The driver research screened each candidate against the SMAJ20A's 32.4 V table point. That point is specified only at 12.3 A, a 10/1000 µs pulse, and 25 °C; it is a conservative selection reference, not a prediction or universal ceiling for the installed rail. The ratings matrix keeps the actual waveform open, and the same screening point is published canonically as fact-high-diode-smaj20a-clamp with its exact conditions attached.

PT4115 is rejected by this conservative screening rule, not because 32.4 V has been proven as the installed rail waveform. AL8860 provides more static headroom at the same cost, while the actual transient still requires the source-impedance, parasitic, simulation, and bench evidence tracked in issue #34.

TPS92511 offers more static voltage headroom, but was rejected on two counts: its 500 mA maximum sits exactly at this design's operating current with zero headroom, and the research pass could not confirm its IADJ formula or its full passive count from a primary datasheet, so it is not schematic-ready. The AL8860 is characterised end to end.

The AL8860 also brought the decisive control-side property nobody was looking for: a CTRL pin with a guaranteed 5–100% analog dimming range (fact-al8860-ctrl), which is what made flicker-free dimming possible (see below).

The cross-component behaviour of the whole driver stage — the sense resistor, inductor, catch diode, ballast and LED envelope together — is published as rule-al8860-led-stage, including the exact statement of what it refuses to promote to a PASS.

Driver passive set

Logic rail

This is the one place where this pass overrides a research recommendation. The logic-rail page says an LDO is fine below ~30 mA, and the actual budget is ~25 mA, so on dissipation grounds the LDO wins.

The rail envelope overrules it. AMS1117's 15 V absolute maximum leaves no operating margin at the nominal rail voltage—an absolute maximum is not an operating target. HT7333's 18 V ceiling has little transient headroom. As a conservative screen, it also sits 14.4 V under SMAJ20A's conditioned 32.4 V table point; that subtraction does not predict the installed waveform, but it exposes evidence that the LDO choice cannot close.

AP63203's absolute maximum is 35 V DC and 40 V for 400 ms (DS41326 Rev 3). Its static limits cover the conditioned 32.4 V screening point, so it is the strongest candidate and costs about $0.60 plus three passives. This does not make the real clamp event zero-risk: source impedance, waveform, effective capacitance, layout, and bench behavior remain open.

XL7015E1 has still more static voltage headroom (80 V), but needs a feedback divider and a 150 kHz-class inductor whose researched candidate showed only 238 units in stock. AP63203 is fixed-output, needs no divider, and its inductor is a 250,000-unit part.

MCU

The MCU research frames this as cost versus toolchain friction and calls it "a preference call, not a technical blocker". At prototype quantity the cost axis is noise: the whole spread is about $1, against ¥8,460 of Extended-part setup fees. Toolchain friction, by contrast, is paid in hours during bring-up.

So the decision is toolchain-first, and STM32 wins that axis by the research's own assessment ("best-supported path on macOS by a clear margin" — official vendor tools, Homebrew, OpenOCD, and PlatformIO all first-class, no community forks).

CH32V003 is rejected specifically on its proprietary single-wire SDI debug interface, which only WCH-LinkE programmers speak. A one-vendor-probe dependency is a real cost on a board that will be debugged interactively.

G031 over C011 for 64 KB flash versus 16 KB: it keeps the recorded-flame LUT modulation approach reachable without a board respin, for about $0.40. Hardware headroom that removes a future respin is cheap at this price.

Knob

Current choice (2026-09-06): ALPS RK10J11E0034 / C470643, single-unit 10 kΩ ±30%, 270°±10° travel with physical stops. It replaces the tall EC11 encoder and provides absolute brightness position. The earlier encoder decision used the stock information then available; it is superseded by this stocked low-profile wheel.

Use PA0 ADC_IN0, R27 100 Ω, R22 100 kΩ pull-down and C22 100 nF. Calibrate and average ADC input, verify installed direction, and ramp from dark to the position after valid PD/thermal checks. Do not restore the old fixed 40% startup target.

Control scheme

What the knob controls

Three reasons, in order of weight:

1. A lamp knob that does not change brightness is a surprise. Every other mapping requires the user to learn the device.

2. Brightness is the parameter with a safety meaning. Turning it down is also turning down heat, and turning it fully down is off. Speed and depth have no such property, and neither does a mode-cycled knob whose current mode is invisible.

3. The selected potentiometer has no mode button. The stopped absolute wheel controls brightness directly; no gesture is needed.

Speed and depth become firmware constants. They want tuning against the real diffuser anyway — which does not exist yet, because the enclosure is out of scope — so freezing them in firmware costs nothing now and is a one-line change later.

Dimming method

The AL8860 datasheet recommends PWM below 500 Hz for good dimming resolution. The modulation research puts the peripheral-vision flicker threshold near 300 Hz. A 500 Hz carrier clears that by 1.7x — thin for a lamp that is deliberately meant to be seen out of the corner of the eye, and raising the carrier is exactly what the datasheet says degrades accuracy.

Analog dimming dissolves the conflict: LED current never chops, so there is no flicker frequency at all, and no PWM modulation of the inductor to become an audible whine. The datasheet's guaranteed 5–100% analog range (20:1) is ample for a flame effect.

The 20 kHz carrier is chosen to sit above the audio band entirely and to give the 208 Hz RC filter 40 dB of attenuation. It is an MCU-side frequency only — it never reaches the driver as a switching signal.

Modulation algorithm

modulation-algorithms notes that layering 2–3 filtered-noise generators "adds realism at the cost of a bit more CPU/RAM" — on a 64 MHz Cortex-M0+ running a 50 Hz animation loop, that cost is unmeasurable, so take the realism. The bounded-random-walk approach's character is folded into the slow layer rather than kept as a separate mode.

The LUT approach is rejected because the dataset does not exist. It needs a real flame's luminance curve extracted from video, which is a separate project. The G031's 64 KB of flash keeps the door open for it.

Safe default-OFF

control-safety is unambiguous that this must be a hardware guarantee. The wrinkle the research could not know — because the driver was not picked yet — is that the AL8860's CTRL pin floats to ON, so a pull-down is mandatory rather than merely prudent.

The 2026-08-01 review argued R21 is bounded: the AL8860's soft-start rate (1.5 ms/nF to charge CTRL) implies an internal bias current of ≤1.67 µA, which into 33 kΩ is ≤55 mV — 3.6x below the 0.2 V off threshold. That argument currently has no retained fact. Neither the 1.5 ms/nF rate nor the ≤1.67 µA bound appears anywhere under .claude/skills/component-al8860mp-13/; the record's only CTRL fact is fact-al8860-ctrl, which gives the thresholds and the dimming range but no bias current. Under this project's own rule that UNSOURCED evidence cannot close a domain, mechanism 1 is not closed and Gate 3's CTRL measurement is a real gate, not a confirmation.

The soft-start capacitor is the mechanism that stands on its own: the driver physically cannot reach full current for 150 ms, and the MCU takes control within about 1–2 ms of power-on reset. Two mechanisms, one of them currently evidence-backed. An MCU internal pull-down was rejected because it depends on firmware configuring it — which is the very window being protected.

Interconnect

The interconnect research framed the small-header class's 3 A contact rating against VH's 10 A headroom. That framing assumed the load was the 3 A contract cap. It is 410 mA. Each doubled contact carries 205 mA — 6.8% of nameplate, a 14x derated margin. The headroom problem VH exists to solve does not occur in this design.

The selected direct header pair also fits the six required conductors exactly (2 power + 2 ground + ATT + PDOK) and eliminates the previously unresolved cable housing, crimps, wire, and completed-harness BOM. Its tradeoff is mechanical: it is unkeyed and requires reviewed PCB alignment, standoff, and retention.

Molex KK was never a real alternative — the research page includes it explicitly to show the 3 A-per-contact ceiling is a whole-class property, not an XH quirk.

Protection

The local TVS costs one BOM line already on the order (D5 is the same part) and clamps transients the ~30 cm harness can develop on its own inductance — the whole point of the TVS margin rule is to clamp close to what you are protecting.

The PPTC exists because there is a 7x gap between what the source will deliver and what the board needs: the PD contract allows 3 A, and Board L draws 410 mA. Without a fuse, a Board L fault — a shorted LED string, a failed driver — can pull 45 W through a 2.5 mm harness before anything intervenes. A 750 mA hold bounds that at under 2x the normal draw.

Conditioned open point: the PPTC's 30 V rating (fact-bhfuse-voltage-max) is 2.4 V below the SMAJ20A's 32.4 V table value at 12.3 A, 10/1000 µs, and 25 °C. That subtraction is a useful screening lead, not an installed-waveform result or an accepted over-rating. Actual source impedance, clamp current, duration, parasitics, and fuse stress remain part of the issue #34 simulation and bench gate.

Summary of open conditioned risks

Each of those conditioned figures is published as a calculation on the integration page, with the expression, every input fact, and the conditions under which the number means anything — so the arithmetic can be recomputed rather than trusted. These first three rows are conditioned arithmetic, not permission to energize or call either protected input chain PASS. The current no-PD gate is tracked in issue #34.

Historical toggle decision — superseded by the short slide switch

The previous revision chose Dailywell 1MS1T1B1M1QES-5 / C496154, the earlier toggle fallback, with live JLCPCB stock and wave-solder assembly support verified during planning. The WR11AS choice on 2026-09-05 used external distributor stock and missed the JLCPCB sourcing requirement; that decision is superseded.

SW2 mounts on the rear PCB, clear of Board P and the pillars, and cuts the fused positive feed to both converters. Common 2 feeds pin 1 in ON; pin 3 is unwired but live in OFF. Board P remains powered. The EC11 retained brightness control in that revision. Use the exact solder-lug SKU in the reviewed plated-slot footprint and support its threaded bushing in the enclosure. The old J5 pads and external harness are removed. See selection, mounting and qualification.

Low-profile controls replace the tall controls (2026-09-06)

The current power control is G-Switch SS-12D01-G020 / C5446803, maintained SPDT rated 1 A at 24 V DC. Common 2 receives V15_FUSED; throw 1 feeds V15; throw 3 stays isolated on the PCB and becomes live in OFF. The corrected family model retains the 3.5 mm case, 2 mm actuator and 0.5 mm shoulder, about 6 mm mounted height. RV1 replaces SW1 with the stopped ALPS wheel. Board P, pillars and J2 alignment remain fixed; enclosure and soldering fit still need prototypes.