First real run on a Mac died before the dialog could open:
config.py line 15: CELL_UM_OVERRIDE: float | None = None
TypeError: unsupported operand type(s) for |: 'type' and 'NoneType'
KiCad macOS bundles Python 3.9, where PEP 604 unions in annotations
are evaluated at import time unless deferred - config.py was the one
annotated module without the __future__ import (errors.py has no
annotations, __init__.py is empty).
A new tripwire test walks every shipped module with ast and fails if
a file uses annotations without deferring them, so the next module
added (progress.py was born only last week) cannot regress this
silently while the Windows suite stays green.
Verified for real this time, not audited: the full suite passes under
CPython 3.9.25 with the exact stack a Mac venv resolves (numpy 2.0.2,
scipy 1.13.1, matplotlib 3.9.4, PySide6 6.10.0, pyamg 5.2.1) - 137
passed.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
After a solve, the per-layer current-density maps can be pushed into
the open board as reference images on User.9..User.12 (stackup order,
top first) - visible right in the editor, toggled like any layer,
never plotted to gerbers. Dialog checkbox, default OFF; every push
replaces all reference images on those layers.
Rendering (fill_resistance/overlay.py, KiCad-free and tested headless):
one pixel per grid cell, opaque over copper with the log scale lifted
off the colormap's near-black bottom (dark canvas), transparent
elsewhere, one pixel of half-alpha edge bleed so the overlay reaches
the drawn outline instead of stopping half a cell short. Pushing lives
in board_io (ReferenceImage via the IPC API, KiCad >= 10.0.1; scale =
width / (pixels * 1 inch / 300 PPI), position = image center); kipy
0.7.1 swallows creation errors, so the per-item status is read from
the raw CreateItemsResponse. pipeline.run takes an optional overlay
callback; failures are reported, never fatal.
tools/kicad_overlay_test.py pushes a fiducial alignment pattern (KiCad
bbox readback verified placement to half a pixel); tools/
kicad_heatmap_overlay.py runs the whole thing headless against the
open board, filtering marker rectangles of other nets' analyses via
the exact copper test.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Pads are the junctions where traces and thermal-relief spokes actually
meet; without their copper a multi-track junction necks down to the
accidental overlap of the rounded track ends, or is severed outright.
gather_smd_pad_copper fetches the exact shape of every undrilled pad
on the net (one API call per pad, layer from the padstack) and
build_problem stamps them onto their own layer (INCLUDE_SMD_PADS).
Selected SMD-pad contacts get their real copper as a side effect
(previously electrode-shape intersected whatever lay underneath).
Dead-end pads become floating islands that the existing connectivity
restriction drops.
Closes the documented SMD-pad-copper limitation (prompted by the
padne comparison).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- THT pad copper is now part of the conductor: the exact pad outline
(incl. oblong/custom shapes) is fetched from KiCad once per pad and
stamped onto every included layer (the outer shape stands in for
inner rings). Annular rings bridge antipads, and joints land on real
copper instead of only pour coverage.
- The internal lead conductor is modeled in every solder-filled hole:
a cylinder of drill - THT_LEAD_CLEARANCE_MM (0.25 fab rule) with
THT_LEAD_RHO_OHM_M (copper default; config for brass/steel leads),
in parallel with the solder annulus and the plating.
- Drill mouths of THT pads: populated pads keep conducting mouth
copper (stands in for the solder plug - conservative, the plug is
worth ~200 um of copper equivalent); DNP pad holes are cut open on
every layer like uncapped via mouths.
- Oblong pads: the coat uses the exact pad shape; the lead cone tapers
within the inscribed circle (new pad_min_nm on ViaLink/Electrode) so
the long axis is not overstated sideways.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The clipped lead of a soldered THT contact protrudes
THT_LEAD_PROTRUSION_MM (1.5 mm default, 0 disables) out of the hole on
the side opposite the component, and a solder cone wraps it: full
protrusion height at the drill wall, tapering linearly to zero at the
pad edge. Painted as per-cell extra conduction-equivalent copper via
stack.thick_scale - the tall solder column at the wall pulls the joint
vicinity to lead potential (equivalent to extending the barrel wall
vertically), the taper carries the radial spreading. DC-exact additive
conductance; at f > 0 the factor multiplies the skin-corrected sheet
conductance like the via mouths (documented approximation).
The protrusion side is looked up from the owning footprint (pads store
absolute positions; component on F.Cu -> lead tents on B.Cu), with a
logged B.Cu fallback. Dump schema gains protrusion_side and
tht_protrusion_nm (defaults keep older v6 dumps loading).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Selected vias and through-hole pads inject at the drill-wall ring on
every spanned layer (both contact models), not the whole pad face,
so the pad/pour spreading resistance is part of the result. Vias are
now selectable as contacts.
- Soldered THT joints: the hole is modeled solder-filled (core in
parallel with the plating, also for stitching THT barrels) and the
pad face carries an average-thickness solder coat over the modeled
copper (SOLDER_THICKNESS_UM).
- Vias with drills above a configurable threshold (dialog field,
default CAP_MAX_DRILL_MM = 0.5) keep open mouths even with capping
selected - the fab caps only small vias.
- Geometry dump schema v6: electrode barrel fields, cap_max_drill_nm.
- Verified against R = rho/(pi t)*acosh(d/2a) for two circular contacts
on a sheet (+2.6% at h = 0.15 mm, a = 1 mm; uniform model above the
equipotential one as required by the contact bracket).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Two display fixes for the adaptive grid, from field feedback:
- Equipotential contours showed leaf-sized staircase corners on plane
interiors: the potential is now expanded piecewise-LINEARLY from each
leaf's reconstructed gradient instead of constant-per-leaf, and the
default ADAPTIVE_MAX_CELL_UM drops 2 mm -> 1 mm (interior leaves
beyond that buy almost nothing).
- The raster map now overlays the adaptive mesh: boundaries of coarse
leaves draw in darker copper (fine regions stay plain = fully
resolved), with a legend entry. Uniform-grid runs are unchanged.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
ADAPTIVE_CELLS defaults to True (dialog checkbox stays; untick or
--no-adaptive for the uniform reference grid). With the adaptive grid
the auto cell sizer targets TARGET_CELLS_ADAPTIVE (8M fine cells,
~2x finer h) since unknowns no longer scale with the fine cell count -
memory of the masks/field arrays is the new bound.
The test suite pins ADAPTIVE_CELLS off via an autouse conftest fixture:
the exact-value tests define the uniform reference grid; adaptive tests
opt in per test.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Two-point fluxes across coarse-fine faces miss the tangential potential
gradient (offset leaf centers), biasing R ~0.5-2% low. After the first
solve, per-leaf gradients are reconstructed by least squares over face
neighbors and the known tangential term g*delta*Gt moves to the
right-hand side of a re-solve (ADAPTIVE_CORRECTION_PASSES, default 1).
The matrix is unchanged, so the new PreparedSolver reuses the LU
factorization / AMG hierarchy across passes; the corrected currents
satisfy KCL exactly, so the power-balance identity, via currents and
part fluxes all use them consistently (edge power = dV * I_corr).
Measured: strip worst case -1.74% -> -0.028% (1 pass); feature-dense
plate end-to-end -1.1% -> -0.011% at 7.2 s vs 25.9 s uniform (5.58M ->
823k unknowns). Tests tightened accordingly plus a passes-knob test.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
config.ADAPTIVE_CELLS (dialog checkbox "adaptive cells", off by
default; standalone --adaptive) routes run_solve through
fill_resistance/adaptive.py: per-layer balanced leaf grids where every
non-uniform fine cell (electrodes, 1D chain cells, buildup, via-mouth
thickness map) is pinned at the fine size, leaf faces via the
series-half-cell rule, chain links and barrels re-attached by node id,
connectivity restriction and both contact models on the leaf graph via
solver cores extracted for reuse (_equipotential_core, _uniform_core,
_conductance_params, _barrel_links). All fields (V, |J|, power density)
are computed per leaf and expanded to the fine grid, so plots, summary
and dumps are unchanged.
Element sizes: minimum = the grid cell size itself (auto / dialog /
CELL_UM_OVERRIDE); maximum = ADAPTIVE_MAX_CELL_UM (2 mm default);
ADAPTIVE_GUARD sets the clearance a block needs to grow.
Measured end-to-end (feature-dense 120x120 plate, h=50um): 25.9 s ->
5.4 s, 5.58M -> 823k unknowns, R -1.1%. Accuracy documented honestly:
coarse-fine interfaces carry a first-order tangential flux error
biasing R low by ~0.5-2% depending on geometry (worst on narrow
strips); the earlier assumption that linear fields solve exactly on the
leaf graph was wrong - offset centers across size transitions leave an
unpaired residue. Gradient-corrected interface fluxes remain as phase 4
if tighter accuracy per leaf is needed.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Each via now contributes its ring/pad copper (full-thickness disc of
the pad diameter on every spanned layer) and its drill mouth: with
"vias filled + capped" (dialog checkbox, default on, VIAS_CAPPED) the
mouth carries a CAP_PLATING_UM (15 um) thin copper cap on the outer
layers and is an open hole on inner layers; unchecked, mouths are open
everywhere. Mouth coverage is area-weighted per cell (4x4
supersampling) through a per-cell thickness map feeding the existing
harmonic-mean face machinery, so sub-cell mouths perturb the sheet by
their true covered fraction instead of whole cells. Fully swallowed
cells leave the mask; the barrel then attaches through the ring via the
existing pad-footprint search. THT-pad copper and drills stay outside
the model. Ring discs paint before 1D trace chains (chains see them as
regular copper), mouths after wide tracks (drills go through trace
copper). standalone gains --uncapped.
Tests: cap==foil identity against the feature-off reference, strict
R(solid) < R(cap) < R(hole) ordering, ring bridging a fill gap that a
ringless barrel cannot cross, gentle sub-cell perturbation at coarse
grids, and JSON roundtrip of the new fields.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Layer-to-layer the cap (>=15 um plating per fab spec) is parallel to
the annular-ring contact, so it cannot change the DC path. In-plane the
model treats every via mouth as solid layer-thickness copper; the error
is bounded by the dilute-hole correction (~ +2x mouth-area-fraction
locally), partially offset by the unmodeled annular-ring copper, and
typically well under a few percent of total R.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Tracks are now first-class Problem objects (TrackSeg: centerline +
width, dump schema v5), so the wide/narrow decision replays at raster
time: traces at least TRACK_1D_FACTOR (3) cells wide rasterize from
their outline as before; narrower ones mark the cells their centerline
crosses as copper and connect them with explicit conductance links
carrying the trace's TRUE arc length per link - no staircase inflation
for diagonals or arcs, and no discretization error in the trace R, at
any grid size. Links across cells already joined by pour faces are
skipped (union, not sum); chain-only cells get no sheet faces (their
copper is narrower than a cell). Electrodes, via barrels, connectivity
restriction and the skin-effect scaling all work on chain cells
unchanged.
This removes the need to shrink the cell size for thin traces: a 0.2 mm
bridge at 500 um cells now matches its finely-rasterized ground truth
within a few percent (tested), including diagonal and arc traces.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Straight tracks become capsule outline polygons (rectangle +
semicircular caps), arc tracks annular bands with end caps, both
tessellated to the same sagitta tolerance as zone-fill arcs; they merge
into the per-layer copper next to the fills, so rasterization, via
stitching, the solver and the plots handle them unchanged. Trace-only
layers and trace-only nets now qualify as candidates. Dialog checkbox
(on by default, INCLUDE_TRACKS) toggles them per run.
Hole-less polygons (every track outline) now paint the layer mask
directly instead of allocating a full-frame temporary each.
Tests: exact N-cell chain on a rasterized capsule, analytic annular-
sector convergence for an arc trace, capsule/arc-band outline geometry
invariants, collinear-arc degradation, and fill+trace union solve.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Hybrid rasterizer: PIL scanline fill for the bulk, with cells in a
~2 px band around each ring edge re-tested exactly against the
polygon - cell-for-cell identical to the old center-in-polygon pass
(equivalence test added) but O(vertices + cells) instead of
O(vertices x cells). Measured 4.5 s -> 0.11 s at 1.45M cells with
8.8k polygon vertices.
- AMG-preconditioned CG (pyamg, new requirement) above 500k unknowns:
measured 7.0 s vs 15.3 s spsolve at 1.4M unknowns at a fraction of
the memory, R identical to 1e-6; the old Jacobi-CG (kept as fallback
when pyamg is missing) needed tens of minutes there. spsolve stays
the default below 500k where it is exact and fastest.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
DC/AC resistance, power dissipation, and via/injection-area currents of
copper zone fills. KiCad 10 IPC-API plugin (kicad-python/kipy):
multi-layer via-coupled FDM solver, multi-part terminals via User.1/User.2
marker layers, pads as contacts, uniform-injection and equipotential
contact models, per-foil skin effect, optional solder/copper buildup on
mask openings. 54-case test suite incl. exact analytic references.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
janik