Finish KiCad Driver PCB
Budget: $8 – $15 USD
As Coordinator. First, the honest boundary: this is an **estimate**, not measured from the live `.kicad_pcb`/`.kicad_sch` files (I can't reach D:\ from here). It's based on the effort budget in `walkthrough.md` plus the known current state (PowerCell1 routed, open blockers). For exact numbers, see the MCP option at the end.
## Status by phase (effort-weighted)
| Phase | Session | Budget (h) | Est. done | Remaining |
|---|---|---|---|---|
| Hierarchical schematic | 1 | 2-3 | ✅ ~100% | — |
| Power Input | 2 | 1-2 | ✅ ~100% | — |
| PowerCell sheet | 3 | 2-3 | ✅ ~95% | (topology verification) |
| FPGA Core | 4 | 3-4 | ⚠️ ~30% | U402 real pinout/symbol |
| MCU Core | 5 | 1-2 | ⚠️ ~50% | U501 full-pin symbol |
| ADC + others | 6 | 2 | ⚠️ ~50% | U601/U602 symbols |
| ERC + Annotation | 7 | 1 | ⚠️ ~40% | PWR_GOOD error, J702/J703 short |
| Footprint assignment | 8 | 1-2 | ✅ ~90% | — |
| Netlist + open PCB | 9 | 0.5 | ✅ 100% | — |
| **PCB Layout** | 10 | **16-24** | ⚠️ **~20%** | 13× PowerCell repeat, center IC routing, outline, T_SENSE path, STATOR netclass |
| DRC + Gerber export | 11 | 1 | ❌ 0% | all |
## Overall ratio
Total budget ~30-45 h (mid ≈ 37 h). Effort-weighted:
- Schematic side (~10 h): **~80% done**
- PCB side (~25 h): **~20% done**
→ **Roughly 35% done / 65% remaining (≈ one third complete).**
This matches intuition: the schematic skeleton largely stands, but the heaviest item — **PCB layout (16-24 h)** — is barely started. Only 1 of 14 PowerCells is routed, with the stackup and power planes set up.
## The two things that move this number most
1. **Topology fork (critical):** If the PowerCell in KiCad is still the OLD single-switch design, the already-routed "template" PowerCell1 has to be reworked into the 2-FET synchronous-freewheel cell → the most-complete piece becomes partly invalid and the ratio drops **below 35%**. If KiCad already holds the NEW topology, the docs just need a cheap update and **35% holds**. That's why this verification comes first.
2. **FPGA pinout bottleneck:** U402's pinout is undefined, so the **center IC routing (the bulk of Session 10) is blocked**. PCB won't progress past ~20% until the pinout is locked.
---
For exact figures I can measure via MCP (Desktop session with pcb_* tools loaded): remaining ratsnest/airwire count, fraction of nets routed, live DRC/ERC error counts. Alternative: send me a current **DRC + ERC summary screenshot** and I'll revise the ratio with real numbers. Which do you prefer?
details in attachment files.
## Status by phase (effort-weighted)
| Phase | Session | Budget (h) | Est. done | Remaining |
|---|---|---|---|---|
| Hierarchical schematic | 1 | 2-3 | ✅ ~100% | — |
| Power Input | 2 | 1-2 | ✅ ~100% | — |
| PowerCell sheet | 3 | 2-3 | ✅ ~95% | (topology verification) |
| FPGA Core | 4 | 3-4 | ⚠️ ~30% | U402 real pinout/symbol |
| MCU Core | 5 | 1-2 | ⚠️ ~50% | U501 full-pin symbol |
| ADC + others | 6 | 2 | ⚠️ ~50% | U601/U602 symbols |
| ERC + Annotation | 7 | 1 | ⚠️ ~40% | PWR_GOOD error, J702/J703 short |
| Footprint assignment | 8 | 1-2 | ✅ ~90% | — |
| Netlist + open PCB | 9 | 0.5 | ✅ 100% | — |
| **PCB Layout** | 10 | **16-24** | ⚠️ **~20%** | 13× PowerCell repeat, center IC routing, outline, T_SENSE path, STATOR netclass |
| DRC + Gerber export | 11 | 1 | ❌ 0% | all |
## Overall ratio
Total budget ~30-45 h (mid ≈ 37 h). Effort-weighted:
- Schematic side (~10 h): **~80% done**
- PCB side (~25 h): **~20% done**
→ **Roughly 35% done / 65% remaining (≈ one third complete).**
This matches intuition: the schematic skeleton largely stands, but the heaviest item — **PCB layout (16-24 h)** — is barely started. Only 1 of 14 PowerCells is routed, with the stackup and power planes set up.
## The two things that move this number most
1. **Topology fork (critical):** If the PowerCell in KiCad is still the OLD single-switch design, the already-routed "template" PowerCell1 has to be reworked into the 2-FET synchronous-freewheel cell → the most-complete piece becomes partly invalid and the ratio drops **below 35%**. If KiCad already holds the NEW topology, the docs just need a cheap update and **35% holds**. That's why this verification comes first.
2. **FPGA pinout bottleneck:** U402's pinout is undefined, so the **center IC routing (the bulk of Session 10) is blocked**. PCB won't progress past ~20% until the pinout is locked.
---
For exact figures I can measure via MCP (Desktop session with pcb_* tools loaded): remaining ratsnest/airwire count, fraction of nets routed, live DRC/ERC error counts. Alternative: send me a current **DRC + ERC summary screenshot** and I'll revise the ratio with real numbers. Which do you prefer?
details in attachment files.