Settle the numpad layout: 25 keys, no encoder

Nav block across the top with an isolated inverted-T below it, operators in a
right-hand column beside 9/6/3 where a numpad puts them, and a 2u Enter plus
+ on a stepped thumb cluster spanning the three nav columns. Six by five,
~114 x 105 mm.

The volume encoder is dropped, which makes this a completely plain QMK
keyboard - no encoder, no pointing device, no consumer control. Firmware notes
and BOM updated; the numpad board is now ~$4.11 in parts.

Co-Authored-By: Claude Opus 5 <[email protected]>

README.md

@@ -41,16 +41,27 @@ either.
 |---|---|---|---|
 | `Schematics/Left`   | 1 | **34** | MCU, matrix, USB-C, SBU split link |
 | `Schematics/Right`  | 1 | **41** | MCU, matrix, USB-C, SBU split link |
-| `Schematics/Numpad` | 1 | 25 (16 numpad + 9 nav) | MCU, matrix, 1 encoder, **CH334R hub**, 3 downstream ports |
+| `Schematics/Numpad` | 1 | **25** (14 numpad + 9 nav + 2 thumb) | MCU, matrix, **CH334R hub**, 3 downstream ports |
 
 **Numpad layout is settled** — see `Documents/numpad-layout.svg`, and
-`Documents/three-board-overview.svg` for all three at the same scale. Six columns by five
-rows, ~114 × 95 mm. Arrows take the **outer** left column on the principle that
-the most-used keys go furthest out, PgUp/PgDn sit beside them, and Del/Home/End
-fill the rest of the two nav columns. **No Num key** — the firmware resolves
-keypad vs number-row at press time, so the pad never needed one. The 2u that
-frees up at the top holds the single volume encoder, which means a full-size
-knob fits rather than two cramped ones colliding at 19.05 mm centres.
+`Documents/three-board-overview.svg` for all three at the same scale. Six
+columns by five rows, ~114 × 105 mm, **25 keys and no encoder**.
+
+- **Nav block across the top**: Home/End/PgUp over Del/PgDn, with the arrows in
+  a standard **inverted-T** isolated below by three empty cells — the same
+  breathing room a 104 gives them. Del sits two rows above `←` with a gap
+  between, so the destructive key is never adjacent to the one you hit blind.
+- **Numpad on the right**: digits in a 3×3 with `0` as a 2u, and `/ * −` in a
+  right-hand column beside 9/6/3, where a full-size numpad puts its operators.
+- **`Enter` (2u) and `+` on a stepped thumb cluster**, spanning exactly the
+  three nav columns. That's what lets the digits drop from four columns to
+  three and keeps the board at six wide.
+- **No Num key** — the firmware resolves keypad vs number-row at press time, so
+  the pad never needed one.
+- **No encoder.** Dropped along with the scroll wheel; volume lives on a layer
+  or on the halves' function row. This makes the numpad a completely plain QMK
+  keyboard — no encoder, no pointing device, just a matrix. It is the simplest
+  of the three boards by a distance, and ~$4.11 in parts.
 
 Cabling:
 
@@ -302,19 +313,19 @@ Plan is belt-and-braces:
 Only eleven keys need this (1–9, 0, `.`). `KP_SLASH`/`ASTERISK`/`MINUS`/`PLUS`/
 `ENTER` mean the same thing either way.
 
-**One encoder, volume only.** The scroll wheel was dropped — no good physical
-implementation. That removes more than it looks like:
+**No encoder anywhere.** The scroll wheel went first, then the volume knob. That
+strips a surprising amount out of the numpad's firmware:
 
-- No `POINTING_DEVICE_ENABLE`, no `MOUSEKEY_ENABLE`, no hi-res scroll resolution
-  multiplier. **The numpad stops enumerating as a mouse entirely** — it's a
-  keyboard plus consumer control and nothing else.
+- No `ENCODER_ENABLE`, no `POINTING_DEVICE_ENABLE`, no `MOUSEKEY_ENABLE`, no
+  hi-res scroll multiplier.
+- The numpad **enumerates as a plain keyboard and nothing else** — no mouse
+  interface, no consumer-control interface unless media keys go on a layer.
 - One fewer USB endpoint, which matters because the L103's USBFS has eight and
-  QMK's composite device was going to be tight against that.
-- The detentless-EC11 sourcing risk is gone. A stock EC11 is detented, which is
-  exactly what volume wants, so `C2991196` is simply correct with nothing to
-  verify.
+  QMK's composite device was tight against that.
+- The detentless-EC11 sourcing risk is gone, because there is no EC11.
 
-Rules file is now just `ENCODER_ENABLE` + `EXTRAKEY_ENABLE`.
+Volume has to live on a layer, or on the halves' function row. Decide that
+rather than discover it.
 
 ---