Charging and Routing Attempt Report
Scope and immutable constraints honored
- No component placement, rotation, board outline, or ESP32 antenna/e-paper/USB location was changed.
- The board remains a 50 x 100 mm rounded rectangle with the existing four-copper-layer stackup: Top signal, Mid-Layer 1 Ground Plane, Mid-Layer 2 Power Plane, Bottom signal. Mid-Layer 1 is the solid internal ground reference.
- USB programming VBUS remains on
USB_PROG_VBUS (J2:VBUS_A, J2:VBUS_B, U8:VBUS) and is not connected to CHG_5V_PROT or U3:VIN.
- The remote charging topology remains
J6:1 -> F1:P1 -> F1:P2 / CHG_5V_PROT -> D4:1, C34:P1, U3:VIN; D4:2 and C34:P2 remain on GND.
Charge-current correction — completed
R3 was changed in place from 270 kOhm to 170 kOhm on U3:ICHG; connections and PCB placement were preserved.- Source check: the Injoinic IP2312 V1.0 current-setting section gives
ICC = 135000 / RICHG; the documented minimum table entry is 170 kOhm for 0.8 A. The source also states that an open/NC setting selects the 2.1 A default. The source URL is attached to U3 in the schematic.
- Electrical-rule review after the change returned no ERROR, WARNING, OPEN, or UNCERTAIN ERC checks (reviewed 2026-08-27T03:52:57Z).
USB and fine-pitch routing preparation — completed
- Made
USB_DP and USB_DM explicit High Speed nets with a 90 ohm target, 7.5 percent tolerance, and common USB2_DATA differential-pair identifier. No hard-coded width/gap rule was invented.
- Created a protected, via-free deterministic J2 USB-C D+/D- breakout using the live connector UID and net names. It created 15 trace segments, no vias, and no failed nets. The stackup solver used 0.300 mm tracks and 0.097 mm gap for target 90 ohm, but reported an achieved value of 100.36 ohm; this must be resolved with the fabrication house's actual impedance coupon/stackup before release.
- Generated 27 locked deterministic neckdowns for the 24-pin 0.5 mm e-paper FPC and other dense pads. These copper features were preserved.
Power-routing work and autorouter outcome — blocked/partial
- Manually routed the J6-to-F1
CHG_5V segment on Top with a 0.500 mm trace (J6:1 to F1:P1).
- Manually routed F1:P2 to D4:1 on
CHG_5V_PROT with a 0.500 mm trace. An earlier F1:P2-to-U3:VIN attempt caused two copper-overlap findings and was deleted; the overlaps cleared. U3:VIN and C34:P1 therefore remain to be routed on CHG_5V_PROT.
- Full-board auto-route attempt 1 finished partial: 22 routing nodes added, 146 remaining airwires, no unrouteable nets. It reported the same two J2 footprint preflight warnings below.
- Full-board attempt 2 was stopped by the router watchdog after 1186 seconds of inactivity. Inspection afterward found 41 route segments and still 146 remaining airwires, so it did not improve the result. No further bulk retry was made.
Final validation snapshot
- DRC review at 2026-08-27T03:52:54Z: one failing DRC category, Airwires. The review reports 148 failing check instances; the live routing-state report gives 146 remaining airwires plus two dangling USB breakout stub groups. No additional DRC category returned ERROR/WARNING/OPEN/UNCERTAIN in the requested full DRC set.
- The two dangling groups are the protected J2 USB deterministic-breakout runway stubs and must not be deleted as stale routing. They remain for a successful full-board route to continue from.
- The live airwire report includes unresolved USB data connections (the generated
USB_DP and USB_DM breakout stubs are not yet connected through the ESD path/downstream nodes), USB_CC2, USB_PROG_VBUS, charger CHG_5V_PROT, VBAT, 3V3, GND, e-paper, and other signal nets. Routing is not complete.
Preflight blocker requiring placement/footprint authorization
The board check was run before and after preparation. It persists with two warnings:
ROUTING_PIN_INSIDE_KEEPOUT — J2 CC2.ROUTING_PIN_INSIDE_KEEPOUT — J2 VBUS_B.
The router says it ignores this keepout. Inspection confirmed J2 has an explicit fixed position and edge-placement rule, not position:auto. The offending keepout was not exposed as an editable standalone layout node; changing the J2 footprint keepout boundary or moving/replacing J2 would exceed the immutable-placement constraint. This unresolved footprint/keepout anomaly, plus the autorouter watchdog/no-progress behavior, blocks a safe routing-complete claim.
Remaining fabrication holds (not guessed)
- Exact mated insertion tolerance for the e-paper FPC.
- Touch-electrode overlay stackup/thickness and mechanical behavior.
- Speaker enclosure/acoustic volume and venting geometry.
- CH340 crystal specification.
- Manufacturer part numbers remain unassigned for 62 parts; none were invented.
- USB differential impedance must be finalized from the selected fabricator's controlled stackup/coupon, because the breakout solver reported 100.36 ohm versus the 90 ohm target.