Remaining tool review and reference sheets

Reviewed 2026-11-09 against the actual fragments and shared implementations. This record complements the basic, RF and SMPS review records; it does not qualify excluded code. Thermal tools and harness builders are deliberately outside the published catalog at the user’s request.

Legacy fragment	Decision	Evidence / reason
ohms-law-wheel	Retain	All six independent positive input pairs, repeated edits, invalid values and remounting checked. Fixed stale calculated inputs.
reactance-calc	Retain	Ideal Xc, Xl and LC resonance; SI conversion and invalid-value clearing checked.
rc-time-calc	Retain	Ideal RC step response; 50%/90% charge/discharge and invalid input checks.
ev-powertrain-calc	Restore narrowly	Steady flat-road drag + rolling resistance, constant efficiency and usable energy. No acceleration, grade, weather, auxiliary load or battery model. At 72 km/h, 1000 kg, Crr .01, Cd .3, area 2 m²: drag 147 N, rolling 98.1 N, electrical power 5.447 kW at 90% efficiency.
magnetic-core-calc	Replace	Shared implementation mixes a valid constant-permeability reluctance model with generic material presets and an arbitrary 80% saturation “safe” solver. New magnetic-reluctance page retains only supplied geometry/permeability, linear L/B/energy and explicit no-fringing/no-loss limits.
awg-wire-ampacity	Exclude	Unsourced chassis/bundle ampacity table is used to declare “Safe & Compliant”; fusing formula does not establish insulation or installation compliance. DC resistance could be separated later.
ipc-2221-trace-width	Exclude	Empirical legacy temperature-rise fit is presented with automatic constraint satisfaction. It is a thermal screening tool without board/stackup/heat-flow validation.
pcb-clearance-calc	Exclude	Hardcoded table contains uncertain comments and no verified standard revision/applicability. A spacing compliance result needs the exact applicable standard and environment.
fusing-current	Exclude	Onderdonk-style adiabatic thermal calculation is not qualified for the offered PCB geometry and fault-clearing conditions. Thermal/fusing claims remain out.
planar-inductor	Exclude	Geometry/field display uses simplified effective-permeability weighting. Applicability, construction and current/loss claims lack the required reference comparison.
emi-filter-calc	Exclude	Resonant-frequency calculations assume specific common/differential-mode L and C definitions not established by the input labels; an LC corner alone is not insertion loss. The rebuilt SMPS filter offers an explicit loaded circuit instead.
power-electronics-calc	Exclude	Rectifier code explicitly mixes approximate resistive-load and continuous inductive-load equations; waveforms include a rough controlled-three-phase envelope. The combined suite is not restored on the strength of its sound buck subset.
battery-suite	Rebuilt/restored	Replaced generic chemistry/aging curves with a source-conditioned reference catalog and explicit OCV/R/temperature/load model. See battery-analysis-review.md.
bldc-winding-helper	Exclude	Code labels phase-belt winding assignment as simplified; no winding-factor/harmonic/connection validation supports practical winding recommendations.
ac-motor-dynamics	Exclude	Approximate Thevenin parameters are combined with a separate exact input-current branch and an operating-region “stable” verdict. Needs consistent model and reference cases.
motor-controller	Exclude	dq capability sweep uses coarse Id search and shared rendering/extraction assumptions; voltage/current conventions, field-weakening accuracy and model comparisons remain unverified.
voltage-margining-dac	Exclude	Valid feedback-node algebra coexists with default-masking inputs and ambiguous bipolar current-DAC step/count conventions. Requires separately checked solve paths and sign conventions before restoration.
web-serial-interface	Exclude	Device I/O utility, not a formula calculator. Shared chart/lifecycle/protocol behavior has no reviewed hardware acceptance record. No serial device was accessed in this task.
plot-extractor	Exclude	Linear-only digitization returns a plausible constant on coincident calibration points; lacks degenerate/calibration-reset checks and repeat-mount isolation.
fishbone	Exclude	Labels independent OR probability as a Bayesian fault tree; defaults and conversions do not establish a defensible causal probability model.
filter-designer	Exclude	Shared overrides draw a Butterworth-order magnitude based only on LC/LCL/CLC topology rather than the actual terminated network. Rebuilt SMPS filter uses its actual complex circuit.
loop-compensator-p-z	Exclude	Mixed shared Type-II/III topology/initializer paths and a fixed auto-solve component set lack a circuit-bound reference check. Measured-loop lead/lag exploration is now in the rebuilt SMPS tuner.
smps-loop-tuner	Rebuilt/restored	The old shaped Bode/transient recommendations are replaced; see smps-tool-review.md for actual RLC and measured-loop scope.
magnetics-suite	Exclude	Transformer area-product branch explicitly uses a “pseudo Ap” placeholder; generic window utilization/current density and turns/gap sizing cannot qualify a winding design.
transformer-designer	Exclude	Fixed approximate mean turn length, generic core/thermal inputs and safe-power recommendations do not represent supplied core/winding construction. Thermal model remains out.
pv-yield-simulator	Rebuilt/restored	Replaced synthetic clear-sky yield with supplied POA interval integration and monthly irradiation estimates, explicit temperatures/losses/clipping and provenance. See pv-generation-review.md.
thermistor-r-t-curve-ntc	Exclude	Shared Steinhart–Hart mode uses a placeholder beta-model plot. A valid beta equation alone does not defend the offered combined thermal/sensor tool.
heatsink-simulator	Exclude	Unverified heat-transfer/fin/thermal model; explicitly excluded.
thermal-network-calc	Exclude	Unverified thermal model; explicitly excluded.
cable-thermal	Exclude	Unverified dynamic cable thermal model; explicitly excluded.
busbar-capacity	Exclude	Capacity/temperature assumptions do not establish installation current rating; thermal/current-capacity result not restored.
multipoint-harness	Exclude	Harness builder explicitly excluded.
magnetics-harness	Exclude	Combined builder fragment outside the requested analytical catalog.
can-bus	Exclude	Harness builder; blanket length/stub PASS/FAIL thresholds lack transceiver/sample-point/delay budgets.

Narrow replacements

Magnetic reluctance: ℜ = le/(μ0 μr Ae) + lg/(μ0 Ae); L = N²/ℜ, B = LI/(N Ae), E = LI²/2. Uniform geometry, constant permeability, no fringing, leakage, saturation curve, core/winding losses or thermal result. Primary background: TDK ferrite definitions. Tests check 125.664 µH ungapped and 5.984 µH at a 0.5 mm total gap for the stated 50 mm / 25 mm² / μr 2000 / 10-turn example, plus zero current and bad geometry.
Skin depth replaces the old AC-resistance interpolation: δ = 1/√(πfμσ) for a homogeneous good conductor and Rdc = length/(σA). No AC-resistance, ampacity or temperature claim. MIT circuit notes give the skin-depth derivation. Tests check 66.085 µm at 1 MHz and 21.9524 mΩ/m DC resistance for a 1 mm wire at 58 MS/m, plus DC/invalid-input behavior.

New reference sheets

DDR, Ethernet, SerDes and LVDS sheets provide independently written summaries and link the primary vendor documentation beside the relevant sections. No universal trace-length/skew, dielectric, impedance, capacitor or reach prescription is inferred from an unrelated device. DDR includes a worked clock/data-rate example; Ethernet separates MAC–PHY and MDI paths and identifies 8P8C pair groups; SerDes covers channel/clock/coupling terminology; LVDS covers point-to-point termination, common mode and topology limits. Tables scroll locally on mobile.

These are references, not standards-compliance calculators. They contain no simulated eye diagrams or unsupported hardware pass/fail badges. Local checks cover unique anchors, internal navigation and source-link presence; browser QA covers representative layout, theme, and interactive calculators separately.