Probe E# — E-Sharp AB
Probe is made by E# (E-Sharp AB) in Sweden. Want your own boards reviewed like this? Book a meeting

MACBETH

OHWR0082 · revision R1

MACBETH (OHWR0082 R1) is an open-hardware circuit board from the CERN Open Hardware Repository, designed in KiCad EDA: a 12-layer board of 400 × 200 mm with 1559 components and 1212 nets, published under CC-BY-SA-4.0, CERN-OHL-S-2.0, GPL-3.0.

review 69% walked · 23 findings: 2 blocker · 12 major · 9 minor (228 items)

Open in the interactive viewerRead the full review

MACBETH (OHWR0082 R1) circuit board, top view: copper, silkscreen and components
Top view, rendered by Probe from the design's own fabrication data.

Design review

Probe recorded 23 findings: 2 blocker · 12 major · 9 minor (228 items) on this design. Each is shown here with its reasoning; the full review links every one to the pins, nets and datasheet pages it rests on.

  1. LedNoCurrentLimitblockerOpen

    LedNoCurrentLimit: 6 components×6

    Corrected · evaluation@esharp.se — D4, D5, D7, D8, D10, D11 are WS2812B addressable LEDs (BOM value/MPN WS2812B), not bare LED dies. WS2812B datasheet (hash 670517e4) p.2 pin table: 1 VDD power supply, 2 DOUT, 3 VSS ground, 4 DIN; supply range +3.5..+5.3 V. Pins 1/3 on 5V0_IN/GND is the specified supply connection (p.5 typical application circuit shows VDD straight to 5V); LED current is set by the integrated driver, so no series resistor is required.

    At D10, D11, D4, D5, D7, D8

    Each finding, with its own headline
    • D10D10 sits straight across '5V0_IN' and 'GND' with nothing in series

      D10 is an LED with pin 1 on the supply net '5V0_IN' (5.044 V) and pin 3 on the ground net 'GND' — nothing sits in series with it, so its forward current is set by the rail's compliance and the die's knee rather than by the board. Structural, read from the netlist and the resolved net domains: a resistor, a driver or a GPIO on either end would put this LED on a different net and this rule would say nothing. If the supply net is in fact a current-limited driver output classified as a rail, refute this on that fact. Pin(s) 1 of D10 read as a supply input from the pin NAME only: if this is an LED with an integrated driver, file the datasheet's type cell or a quoted role (submit_pin_behavior) and it is exempt on that fact.

    • D11D11 sits straight across '5V0_IN' and 'GND' with nothing in series

      D11 is an LED with pin 1 on the supply net '5V0_IN' (5.044 V) and pin 3 on the ground net 'GND' — nothing sits in series with it, so its forward current is set by the rail's compliance and the die's knee rather than by the board. Structural, read from the netlist and the resolved net domains: a resistor, a driver or a GPIO on either end would put this LED on a different net and this rule would say nothing. If the supply net is in fact a current-limited driver output classified as a rail, refute this on that fact. Pin(s) 1 of D11 read as a supply input from the pin NAME only: if this is an LED with an integrated driver, file the datasheet's type cell or a quoted role (submit_pin_behavior) and it is exempt on that fact.

    • D4D4 sits straight across '5V0_IN' and 'GND' with nothing in series

      D4 is an LED with pin 1 on the supply net '5V0_IN' (5.044 V) and pin 3 on the ground net 'GND' — nothing sits in series with it, so its forward current is set by the rail's compliance and the die's knee rather than by the board. Structural, read from the netlist and the resolved net domains: a resistor, a driver or a GPIO on either end would put this LED on a different net and this rule would say nothing. If the supply net is in fact a current-limited driver output classified as a rail, refute this on that fact. Pin(s) 1 of D4 read as a supply input from the pin NAME only: if this is an LED with an integrated driver, file the datasheet's type cell or a quoted role (submit_pin_behavior) and it is exempt on that fact.

    • D5D5 sits straight across '5V0_IN' and 'GND' with nothing in series

      D5 is an LED with pin 1 on the supply net '5V0_IN' (5.044 V) and pin 3 on the ground net 'GND' — nothing sits in series with it, so its forward current is set by the rail's compliance and the die's knee rather than by the board. Structural, read from the netlist and the resolved net domains: a resistor, a driver or a GPIO on either end would put this LED on a different net and this rule would say nothing. If the supply net is in fact a current-limited driver output classified as a rail, refute this on that fact. Pin(s) 1 of D5 read as a supply input from the pin NAME only: if this is an LED with an integrated driver, file the datasheet's type cell or a quoted role (submit_pin_behavior) and it is exempt on that fact.

    • D7D7 sits straight across '5V0_IN' and 'GND' with nothing in series

      D7 is an LED with pin 1 on the supply net '5V0_IN' (5.044 V) and pin 3 on the ground net 'GND' — nothing sits in series with it, so its forward current is set by the rail's compliance and the die's knee rather than by the board. Structural, read from the netlist and the resolved net domains: a resistor, a driver or a GPIO on either end would put this LED on a different net and this rule would say nothing. If the supply net is in fact a current-limited driver output classified as a rail, refute this on that fact. Pin(s) 1 of D7 read as a supply input from the pin NAME only: if this is an LED with an integrated driver, file the datasheet's type cell or a quoted role (submit_pin_behavior) and it is exempt on that fact.

    • D8D8 sits straight across '5V0_IN' and 'GND' with nothing in series

      D8 is an LED with pin 1 on the supply net '5V0_IN' (5.044 V) and pin 3 on the ground net 'GND' — nothing sits in series with it, so its forward current is set by the rail's compliance and the die's knee rather than by the board. Structural, read from the netlist and the resolved net domains: a resistor, a driver or a GPIO on either end would put this LED on a different net and this rule would say nothing. If the supply net is in fact a current-limited driver output classified as a rail, refute this on that fact. Pin(s) 1 of D8 read as a supply input from the pin NAME only: if this is an LED with an integrated driver, file the datasheet's type cell or a quoted role (submit_pin_behavior) and it is exempt on that fact.

  2. CopperShortBetweenNetsblockerOpen

    unconnected-(U3-NC-Pad3) and unconnected-(U3-NC-Pad4) copper touch on L1_TOP.CU, at (401.67, -197.1): the two nets are shorted in the copper as exported

    unconnected-(U3-NC-Pad3) and unconnected-(U3-NC-Pad4) copper touch on L1_TOP.CU, at (401.67, -197.1): the two nets are shorted in the copper as exported. Where: L1_TOP.CU at (401.67, -197.1) (one net's copper edge lies in the other's copper). The netlist still puts the two nets apart, so every connectivity check passes; the copper the fab builds joins them. The usual cause is a pour filled before the other net's via or pad was placed (or moved), so it runs through it without a clearance: refill the zones, check the CAD tool's DRC is clean, and export again. copper-near at a site shows both nets' copper at distance 0.

    At short:unconnected-(U3-NC-Pad3)|unconnected-(U3-NC-Pad4)

  3. observationOBS-77majorOpen

    TimIQ linear regulators U53, U54, U55, U48 (+5 V) and U59 (-5 V) dissipate 4.5 to 5.9 W each: junction temperature exceeds 125 °C even at 25 °C ambient

    The four TimIQ +5 V rails are each made by a TPS7A4700 LDO from +12 V, and the -5 V rail by a TPS7A3301 from -24 V. The typical load on each rail puts 4.5 to 5.9 W into a 3.5 x 3.5 mm VQFN. With the datasheet thermal resistance, the junction is about 150 to 190 °C above ambient, well past the 125 °C operating limit. Each regulator will sit in thermal shutdown or near it.

    What each regulator carries

    RegulatorRailVin → VoutMain loads (typical, datasheet)Iout typP = (Vin−Vout)·IθJATj at 25 °C ambientLimit
    U53 TPS7A4700T_5V_112 → 5 VU81 ADL5375 194 mA; U83, U84 HMC705 2×190 mA; U79, U80 LT6600 2×28 mA; U77, U78 LTC6655 2×5 mA; U74 LTC6241 5 mA≈ 650 mA≈ 4.5 W32.5 °C/W≈ 173 °C125 °C
    U54 TPS7A4700T_5V_212 → 5 Vsame set: U99, U101, U102, U97, U98, U95, U96, U92≈ 650 mA≈ 4.5 W32.5 °C/W≈ 173 °C125 °C
    U55 TPS7A4700T_5V_312 → 5 Vsame set: U117, U119, U120, U115, U116, U113, U114, U110≈ 650 mA≈ 4.5 W32.5 °C/W≈ 173 °C125 °C
    U48 TPS7A4700T_5V_412 → 5 VU132 ADL5375; U134, U135, U136 HMC705 3×190 mA; U130, U131; U128, U129; U125; U133 SN74ABT541≈ 840 mA≈ 5.9 W32.5 °C/W≈ 216 °C125 °C
    U59 TPS7A3301T_-5V−24 → −5 VV− of eight LT6600 8×28 mA; four LTC6241 4×5 mA; eight LTC2054≈ 245 mA≈ 4.7 W33.7 °C/W≈ 183 °C125 °C

    Only typical supply currents are used. With the maximums (LT6600 36 mA, ADL5375 203 mA) the numbers get worse. For the junction to stay at or below 125 °C at 25 °C ambient, U53 to U55 would need θJA of 22 °C/W or better, and U48 would need 17 °C/W. The regulators rely on their exposed pad alone, and no heatsink is drawn on schematic sheet 25 (TimIQ PSU).

    ``mermaid flowchart LR J28["J28 +12 V in"] --> U53["U53..U55 TPS7A4700<br/>drop 7 V each"] --> R1["T_5V_1..3: ADL5375 + 2×HMC705 + 2×LT6600 …<br/>≈0.65 A each"] J28 --> U48["U48 TPS7A4700<br/>drops 7 V"] --> R4["T_5V_4: ADL5375 + 3×HMC705 …<br/>≈0.84 A"] J29["J29 −24 V in"] --> U59["U59 TPS7A3301<br/>drops 19 V"] --> R5["T_-5V: 8×LT6600 V− …<br/>≈0.25 A"] ``

    Why it matters

    The TPS7A4700 has thermal shutdown with hysteresis (datasheet p.6). At these dissipations the RF chain (ADL5375 modulators and HMC705 dividers) will lose its supply or cycle on and off once the part warms up. The −5 V regulator drops 19 V to feed about a quarter of an amp.

    What would fix it

    Lower each LDO's input voltage (for example, a pre-regulator to about 6 V for the +5 V rails and about −6.5 V for −5 V). The other options are to split the load or to move the voltage drop into a switcher. The designer should confirm the intended input voltages and the thermal design.

  4. observationOBS-78majorOpen

    Clock-output LDOs U31, U35, U39 (LT3040): only low-ESR output cap is a 4.7 µF 6.3 V ceramic at 5.4 V bias, below the 4.7 µF / <50 mΩ stability minimum

    The three clock-output supplies VCC_CLK_IO (U31, U35 and U39, all LT3040) have only one low-ESR ceramic on each output: a 4.7 µF 6.3 V 0603 part biased at about 5.4 V. The datasheet asks for at least 4.7 µF with ESR below 50 mΩ. The nominal value only equals the minimum. After the -10 % tolerance, and after the DC-bias loss of running a 6.3 V class-2 part at 86 % of its rating, the effective low-ESR capacitance is well below the minimum. The parallel 4.7 µF tantalum does not qualify, because a tantalum A-case has ESR far above 50 mΩ.

    What is fitted on each output

    OutputRegulatorLow-ESR ceramicOther capsVout (from VIN divider)Datasheet requirement
    Kria_CLK_OUT0/VCC_CLK_IOU31C156 4.7uF-6.3V-10p-Ceramic-0603C157 4u7F-20V tantalum 1206, C158 10 nF12 V × 1k/(1k21+1k) ≈ 5.43 V≥ 4.7 µF, ESR < 50 mΩ
    Kria_CLK_OUT1/VCC_CLK_IOU35C168 4.7uF-6.3V-10p-Ceramic-0603C169 tantalum 4.7 µF, C170 10 nF≈ 5.43 V (R199/R200)≥ 4.7 µF, ESR < 50 mΩ
    Kria_CLK_OUT2/VCC_CLK_IOU39C180 4.7uF-6.3V-10p-Ceramic-0603C181 tantalum 4.7 µF, C182 10 nF≈ 5.43 V (R222/R223)≥ 4.7 µF, ESR < 50 mΩ
    QuantityLimitActualMargin
    Low-ESR output capacitance≥ 4.7 µF4.7 µF nominal, 4.23 µF at -10 %, and much less at 5.4 V on a 6.3 V part (typically well under half)none, and negative once DC bias is counted
    Capacitor working voltagenone stated5.43 V on 6.3 V rated (86 %)small

    ``mermaid flowchart LR V12["T_+12V"] --> U31["U31 LT3040<br/>VIN = 5.43 V set by R176/R177"] --> OUT["VCC_CLK_IO"] OUT --> C156["C156 4.7 µF 6.3 V ceramic<br/>(only low-ESR cap)"] OUT --> C157["C157 4.7 µF tantalum<br/>(ESR ≫ 50 mΩ)"] OUT --> U33["U33 SN74ABT541 VCC<br/>8 outputs paralleled into J11 coax"] ``

    Why it matters

    The LT3040 has about 1 MHz of bandwidth. The datasheet ties its stability to the low-ESR output capacitor (p.17), and the load here is a fast-edged clock buffer driving 50 Ω. An under-compensated output can ring or oscillate, and the result is jitter on the clock outputs J11 and its siblings.

    What would fix it

    Use a 10 µF or larger ceramic rated 10 V or 16 V (X7R/X5R, 0805 or bigger) on each VCC_CLK_IO, or add a second one, so that the effective low-ESR capacitance at 5.4 V stays above 4.7 µF.

  5. observationOBS-91majorOpen

    WS2812B LED chain (D4…D11) data line driven by open-drain U9 through only a 100 kΩ pull-up (R27): rise time ≈1.8 µs, far too slow for the 0.4/0.8 µs bit timing

    The WS2812B LED chain (D4, D5, D7, D8, D10, D11) is driven from U9, an NC7WZ07 open-drain buffer. The only pull-up on the data line is R27, 100 kΩ to 5V0_IN. That RC is far too slow for the WS2812B's 800 kHz one-wire protocol. The DIN line cannot rise to the input high threshold within a '1' or '0' high time, so the first LED cannot decode the data, and the whole chain stays dark or random.

    Numbers

    QuantityLimit / requirementActualMargin
    High time to be recognisedT1H = 0.8 µs ±150 ns (max 0.95 µs); T0H = 0.4 µs ±150 ns——
    Input high thresholdVIH = 0.7 × VDD = 3.5 V at 5 V——
    Rise time to VIH through R27must be ≪ 0.4 µs100 kΩ × 15 pF (DIN input capacitance alone) = τ 1.5 µs; t(0 → 70 %) = 1.2 τ ≈ 1.8 µs, more with trace and buffer capacitancefails by more than 2x even at best case

    ``mermaid flowchart LR FPGA["Kria SERIAL_LED (3.3 V)"] --> U9["U9 NC7WZ07<br/>open-drain, pulls low only"] U9 --> DIN["Net-(D4-DIN)"] R27["R27 100 kΩ to 5V0_IN<br/>(only source of the rising edge)"] --> DIN DIN --> D4["D4 WS2812B DIN<br/>(C_in up to 15 pF)"] --> CH["D5 → D7 → D8 → D10 → D11"] ``

    Why it matters

    With an open-drain driver, the rising edge is set only by the pull-up. A 100 kΩ pull-up is suited to static logic, not to a 1.25 µs bit period. All six status LEDs ('System LEDs' and 'CLK outputs LEDs', sheet 7) will not work.

    What would fix it

    Reduce R27 to about 1 kΩ or less (τ ≈ 15–30 ns). Or replace the open-drain buffer with a push-pull 5 V-capable buffer (for example a 74AHCT1G125 on 5 V) so that DIN gets a proper 5 V CMOS edge.

  6. observationOBS-537majorOpen

    MACBETH: 20 V tantalums C150, C162, C174 on T_+12V run at 60% of rating

    C150, C162 and C174 (4.7 µF 20 V tantalum, 1206, electrolyte not stated) sit on T_+12V, the input of the LT3040 clock LDOs, at 60% of their rating. If they are MnO2 parts, that is above the 50% manufacturers recommend, and an under-derated MnO2 tantalum on a stiff 12 V rail can fail short.

    PartRailVoltageApplied / 20 VSource
    C150, C162, C174T_+12V12 V60%netlist; schematic sheets 14-16 ('4.7uF 20V', + on T_+12V)

    MnO2 tantalum limit (capacitor guide, citing KEMET T491 p.14, Vishay 40246/EIA-809, AVX TDS-PTNO-0024 1.2.2): about 50% of rated up to 85 C, about 33% above. A polymer tantalum rated over 10 V is allowed 80% (KEMET T2076).

    The BOM value '4u7F-20V-10p-Tantalum-1206' names no MPN, so the electrolyte is unknown. Fix: a 25/35 V part, or a named polymer type. Orientation is correct. The same part on VCC_CLK_IO (about 5.4 V, 27%) and 5V0_IN (25%) is fine.

  7. observationOBS-673majorOpen

    C251 and C252 (10 uF 16 V ceramic) sit on the -24 V input rail T_-24V of the TPS7A3301 regulators

    C251 and C252 are 10 uF capacitors rated 16 V, and they sit between T_-24V and GND. That is the -24 V input rail of the negative LDOs U59 and U60 (TPS7A3301), so about 24 V is applied across a 16 V part.

    PartValue as drawn and in the BOMNetSource
    C25110 uF 16 V ceramic 0805T_-24V (pin 1) and GNDschematic sheet 25, BOM, netlist
    C25210 uF 16 V ceramic 0805T_-24V (pin 1) and GNDschematic sheet 25, BOM, netlist
    RailT_-24V feeds U59/U60 IN pins 13/15/16netlist

    Why it matters

    A ceramic capacitor run above its rated voltage loses most of its capacitance and risks dielectric breakdown, which would short the -24 V input. The neighbouring bulk part C283 on the same rail is a 25 V aluminium capacitor, which is also close to the rail voltage. If T_-24V really carries -24 V, these need parts rated 35-50 V. If the rail is lower than its name says, a note on the schematic would settle it.

  8. CapVoltageDeratingmajorOpen

    CapVoltageDerating: 28 components×28

    28 checked by the review by Claude walker W2 (Cern session) — 28 different reasons

    At C251, C252, C264, C270, C273, C276 and 22 more

    Each finding, with its own headline
    • C251C251 sits on a 24 V rail and its voltage rating is not in the BOM

      C251: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-24V' sits at -24 V; a voltage rating is a magnitude, so it is judged at 24 V.

    • C252C252 sits on a 24 V rail and its voltage rating is not in the BOM

      C252: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-24V' sits at -24 V; a voltage rating is a magnitude, so it is judged at 24 V.

    • C264C264 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C264: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C270C270 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C270: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C273C273 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C273: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C276C276 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C276: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C277C277 sits on a 24 V rail and its voltage rating is not in the BOM

      C277: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'Net-(J29-Pin_1)' sits at -24 V; a voltage rating is a magnitude, so it is judged at 24 V.

    • C279C279 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C279: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C281C281 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C281: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C282C282 sits on a 24 V rail and its voltage rating is not in the BOM

      C282: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'Net-(J29-Pin_1)' sits at -24 V; a voltage rating is a magnitude, so it is judged at 24 V.

    • C283C283 sits on a 24 V rail and its voltage rating is not in the BOM

      C283: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-24V' sits at -24 V; a voltage rating is a magnitude, so it is judged at 24 V.

    • C284C284 sits on a 24 V rail and its voltage rating is not in the BOM

      C284: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-24V' sits at -24 V; a voltage rating is a magnitude, so it is judged at 24 V.

    • C316C316 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C316: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C317C317 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C317: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C332C332 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C332: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C333C333 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C333: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C384C384 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C384: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C385C385 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C385: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C400C400 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C400: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C401C401 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C401: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C452C452 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C452: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C453C453 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C453: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C468C468 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C468: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C469C469 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C469: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C519C519 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C519: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C520C520 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C520: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C535C535 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C535: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

    • C536C536 sits on a 15.05 V rail and its voltage rating is not in the BOM

      C536: on a 15.05 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source. 'T_-15V' sits at -15.05 V; a voltage rating is a magnitude, so it is judged at 15.05 V.

  9. CapVoltageDeratingmajorOpen

    CapVoltageDerating: 53 components×53

    53 checked by the review by Claude walker W4 (Cern session) — 53 different reasons

    At C1, C150, C151, C153, C162, C163 and 47 more

    Each finding, with its own headline
    • C1C1 sits on a 12 V rail and its voltage rating is not in the BOM

      C1: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C150C150 sits on a 12 V rail and its voltage rating is not in the BOM

      C150: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C151C151 sits on a 12 V rail and its voltage rating is not in the BOM

      C151: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C153C153 sits on a 12 V rail and its voltage rating is not in the BOM

      C153: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C162C162 sits on a 12 V rail and its voltage rating is not in the BOM

      C162: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C163C163 sits on a 12 V rail and its voltage rating is not in the BOM

      C163: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C165C165 sits on a 12 V rail and its voltage rating is not in the BOM

      C165: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C174C174 sits on a 12 V rail and its voltage rating is not in the BOM

      C174: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C175C175 sits on a 12 V rail and its voltage rating is not in the BOM

      C175: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C177C177 sits on a 12 V rail and its voltage rating is not in the BOM

      C177: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C193C193 sits on a 12 V rail and its voltage rating is not in the BOM

      C193: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C215C215 sits on a 12 V rail and its voltage rating is not in the BOM

      C215: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C216C216 sits on a 12 V rail and its voltage rating is not in the BOM

      C216: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C217C217 sits on a 12 V rail and its voltage rating is not in the BOM

      C217: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C218C218 sits on a 12 V rail and its voltage rating is not in the BOM

      C218: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C219C219 sits on a 12 V rail and its voltage rating is not in the BOM

      C219: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C250C250 sits on a 24 V rail and its voltage rating is not in the BOM

      C250: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C256C256 sits on a 24 V rail and its voltage rating is not in the BOM

      C256: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C257C257 sits on a 12 V rail and its voltage rating is not in the BOM

      C257: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C258C258 sits on a 24 V rail and its voltage rating is not in the BOM

      C258: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C259C259 sits on a 12 V rail and its voltage rating is not in the BOM

      C259: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C260C260 sits on a 24 V rail and its voltage rating is not in the BOM

      C260: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C261C261 sits on a 12 V rail and its voltage rating is not in the BOM

      C261: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C262C262 sits on a 15 V rail and its voltage rating is not in the BOM

      C262: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C265C265 sits on a 15 V rail and its voltage rating is not in the BOM

      C265: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C266C266 sits on a 24 V rail and its voltage rating is not in the BOM

      C266: on a 24 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C267C267 sits on a 12 V rail and its voltage rating is not in the BOM

      C267: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C268C268 sits on a 15 V rail and its voltage rating is not in the BOM

      C268: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C271C271 sits on a 15 V rail and its voltage rating is not in the BOM

      C271: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C274C274 sits on a 15 V rail and its voltage rating is not in the BOM

      C274: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C30C30 sits on a 12 V rail and its voltage rating is not in the BOM

      C30: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C31C31 sits on a 12 V rail and its voltage rating is not in the BOM

      C31: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C318C318 sits on a 15 V rail and its voltage rating is not in the BOM

      C318: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C319C319 sits on a 15 V rail and its voltage rating is not in the BOM

      C319: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C32C32 sits on a 12 V rail and its voltage rating is not in the BOM

      C32: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C33C33 sits on a 12 V rail and its voltage rating is not in the BOM

      C33: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C336C336 sits on a 15 V rail and its voltage rating is not in the BOM

      C336: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C337C337 sits on a 15 V rail and its voltage rating is not in the BOM

      C337: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C35C35 sits on a 12 V rail and its voltage rating is not in the BOM

      C35: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C386C386 sits on a 15 V rail and its voltage rating is not in the BOM

      C386: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C387C387 sits on a 15 V rail and its voltage rating is not in the BOM

      C387: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C404C404 sits on a 15 V rail and its voltage rating is not in the BOM

      C404: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C405C405 sits on a 15 V rail and its voltage rating is not in the BOM

      C405: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C454C454 sits on a 15 V rail and its voltage rating is not in the BOM

      C454: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C455C455 sits on a 15 V rail and its voltage rating is not in the BOM

      C455: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C472C472 sits on a 15 V rail and its voltage rating is not in the BOM

      C472: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C473C473 sits on a 15 V rail and its voltage rating is not in the BOM

      C473: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C5C5 sits on a 12 V rail and its voltage rating is not in the BOM

      C5: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C521C521 sits on a 15 V rail and its voltage rating is not in the BOM

      C521: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C522C522 sits on a 15 V rail and its voltage rating is not in the BOM

      C522: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C539C539 sits on a 15 V rail and its voltage rating is not in the BOM

      C539: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C540C540 sits on a 15 V rail and its voltage rating is not in the BOM

      C540: on a 15 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

    • C6C6 sits on a 12 V rail and its voltage rating is not in the BOM

      C6: on a 12 V rail but its voltage rating is not stated in the BOM — add it so the rating can be compared to the rail at all. No datasheet is on file for this part; the BOM short-spec is the source.

  10. DecapViaFanoutmajorOpen

    DecapViaFanout: 11 components×11

    11 checked by the review by Claude walker W4 (Cern session) — 11 different reasons

    At C147, C149, C159, C161, C171, C173 and 5 more

    Each finding, with its own headline
    • C147C147 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C147 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C147.1 (rail) runs 2.27 mm along its copper to its nearest via, 1.27 mm past the 1 mm beside the pad: 0.64 nH; C147.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.64 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C149C149 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C149 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C149.1 (rail) runs 2.27 mm along its copper to its nearest via, 1.27 mm past the 1 mm beside the pad: 0.64 nH; C149.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.64 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C159C159 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C159 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C159.1 (rail) runs 2.27 mm along its copper to its nearest via, 1.27 mm past the 1 mm beside the pad: 0.64 nH; C159.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.64 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C161C161 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C161 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C161.1 (rail) runs 2.27 mm along its copper to its nearest via, 1.27 mm past the 1 mm beside the pad: 0.64 nH; C161.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.64 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C171C171 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C171 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C171.1 (rail) runs 2.27 mm along its copper to its nearest via, 1.27 mm past the 1 mm beside the pad: 0.64 nH; C171.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.64 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C173C173 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C173 on 'T_3V3_1': its rail pad reaches its via over 2.27 mm of track, adding about 0.64 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C173.1 (rail) runs 2.27 mm along its copper to its nearest via, 1.27 mm past the 1 mm beside the pad: 0.64 nH; C173.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.64 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C212C212 on 'PL_1V8': its rail pad reaches its via over 2.62 mm of track, adding about 0.81 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C212 on 'PL_1V8': its rail pad reaches its via over 2.62 mm of track, adding about 0.81 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C212.1 (rail) runs 2.62 mm along its copper to its nearest via, 1.62 mm past the 1 mm beside the pad: 0.81 nH; C212.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.81 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C260C260 on 'T_+24V': its rail pad reaches its via over 2.45 mm of track, adding about 1.03 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C260 on 'T_+24V': its rail pad reaches its via over 2.45 mm of track, adding about 1.03 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C260.1 (rail) runs 2.45 mm along its copper to its nearest via, 1.45 mm past the 1 mm beside the pad: 1.03 nH; C260.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 4.40 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 1.03 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.43 nH (pad 0.20 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 0.39 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C283C283 on 'T_-24V': its rail pad reaches its via over 2.45 mm of track, adding about 1.03 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C283 on 'T_-24V': its rail pad reaches its via over 2.45 mm of track, adding about 1.03 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C283.2 (rail) runs 2.45 mm along its copper to its nearest via, 1.45 mm past the 1 mm beside the pad: 1.03 nH; C283.1 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 4.40 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 1.03 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.43 nH (pad 0.20 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 0.39 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C577C577 on 'T_5V_4': its rail pad reaches its via over 2.57 mm of track, adding about 0.79 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C577 on 'T_5V_4': its rail pad reaches its via over 2.57 mm of track, adding about 0.79 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C577.1 (rail) runs 2.57 mm along its copper to its nearest via, 1.57 mm past the 1 mm beside the pad: 0.79 nH; C577.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.56 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.79 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C69C69 on 'ETH_2V5': its rail pad reaches its via over 3.14 mm of track, adding about 1.07 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

      C69 on 'ETH_2V5': its rail pad reaches its via over 3.14 mm of track, adding about 1.07 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C69.1 (rail) runs 3.14 mm along its copper to its nearest via, 2.14 mm past the 1 mm beside the pad: 1.07 nH; C69.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.55 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 1.07 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

  11. EsdProtectorPlacementmajorOpen

    EsdProtectorPlacement: 2 components×2

    Checked by the review, stands · Claude walker W4 (Cern session) — Checked (W4, fleet re-walk): the placement is measured from the copper as stated (U21.1 on '/Top/Kria/Kria_FTDI/USB_JTAG_DM_P': 2 via(s) between J6.3 and U21.1; U21.2 on '/Top/Kria/Kria_FTDI/USB_JTAG_DM_N': 2 via(s) between J6.2 and U21.2). Protector placement guidance (on the line, near the connector, no vias/stubs) is layout practice from protector app notes; its effect depends on the ESD event and the protected part's rating, not on file here. Left open for the owner.

    At U21

    Each finding, with its own headline
    • U21U21.1 on '/Top/Kria/Kria_FTDI/USB_JTAG_DM_P': 2 via(s) between J6.3 and U21.1

      U21 (TPD4S012DRYR) protects '/Top/Kria/Kria_FTDI/USB_JTAG_DM_P' at J6.3. Along the copper: 7.75 mm of track from the contact, 2 via(s) before it, on a 0.79 mm branch off the line; the line runs 12.46 mm on to U24.8; U21 sits at J6, within 1.21 mm of the nearest place outside it to the contact: the track to it is mostly the escape out of J6's own pins, which no placement shortens.

      2 via(s) between J6.3 and U21.1: the strike reaches an inner layer before it is clamped.

      Measured on the copper the ODB states: tracks, the pads they land on, and the netlist's vias. Not a straight line between centroids.

    • U21U21.2 on '/Top/Kria/Kria_FTDI/USB_JTAG_DM_N': 2 via(s) between J6.2 and U21.2

      U21 (TPD4S012DRYR) protects '/Top/Kria/Kria_FTDI/USB_JTAG_DM_N' at J6.2. Along the copper: 7.72 mm of track from the contact, 2 via(s) before it, on a 0.77 mm branch off the line; the line runs 12.44 mm on to U24.7; U21 sits at J6, within 1.22 mm of the nearest place outside it to the contact: the track to it is mostly the escape out of J6's own pins, which no placement shortens.

      2 via(s) between J6.2 and U21.2: the strike reaches an inner layer before it is clamped.

      Measured on the copper the ODB states: tracks, the pads they land on, and the netlist's vias. Not a straight line between centroids.

  12. PartLifecyclemajorOpen

    PartLifecycle: 4 part numbers×4

    2 Discontinued at DigiKey, 2 Obsolete · on 25 placements

    At U106, U111, U115, U6

    Each finding, with its own headline
    • U106LT1012ACS8 on 8 placements (U106, U107, U121, U122, U70, U71, U88, U89): the distributor catalogue reports "Discontinued at DigiKey" for Analog Devices Inc.'s LT1012ACS8

      LT1012ACS8 on 8 placements (U106, U107, U121, U122, U70, U71, U88, U89): the distributor catalogue reports "Discontinued at DigiKey" for Analog Devices Inc.'s LT1012ACS8 — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names no manufacturer for it, so whether that is the fitted maker's product was not checked. A distributor delisting is not the same claim as the part being unavailable; check the vendor's own lifecycle/product-change notice for the status and any named successor. Re-read the source: GET /api/odb/parts/LT1012ACS8/providers (which provider said it, the maker and part number it matched, its own words, and when we asked). No approved alternate is recorded for this position.

    • U111LT1360CS8 on 8 placements (U111, U112, U126, U127, U75, U76, U93, U94): the distributor catalogue reports "Obsolete" for Analog Devices Inc.'s LT1360CS8

      LT1360CS8 on 8 placements (U111, U112, U126, U127, U75, U76, U93, U94): the distributor catalogue reports "Obsolete" for Analog Devices Inc.'s LT1360CS8 — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names no manufacturer for it, so whether that is the fitted maker's product was not checked. A distributor delisting is not the same claim as the part being unavailable; check the vendor's own lifecycle/product-change notice for the status and any named successor. Re-read the source: GET /api/odb/parts/LT1360CS8/providers (which provider said it, the maker and part number it matched, its own words, and when we asked). No approved alternate is recorded for this position.

    • U115LT6600CS8-2.5 on 8 placements (U115, U116, U130, U131, U79, U80, U97, U98): the distributor catalogue reports "Discontinued at DigiKey" for Analog Devices Inc.'s LT6600CS8-2.5

      LT6600CS8-2.5 on 8 placements (U115, U116, U130, U131, U79, U80, U97, U98): the distributor catalogue reports "Discontinued at DigiKey" for Analog Devices Inc.'s LT6600CS8-2.5 — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names no manufacturer for it, so whether that is the fitted maker's product was not checked. A distributor delisting is not the same claim as the part being unavailable; check the vendor's own lifecycle/product-change notice for the status and any named successor. Re-read the source: GET /api/odb/parts/LT6600CS8-2.5/providers (which provider said it, the maker and part number it matched, its own words, and when we asked). No approved alternate is recorded for this position.

    • U6INA260AIPW on U6: the distributor catalogue reports "Obsolete" for Texas Instruments's INA260AIPW

      INA260AIPW on U6: the distributor catalogue reports "Obsolete" for Texas Instruments's INA260AIPW — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names no manufacturer for it, so whether that is the fitted maker's product was not checked. A distributor delisting is not the same claim as the part being unavailable; check the vendor's own lifecycle/product-change notice for the status and any named successor. Re-read the source: GET /api/odb/parts/INA260AIPW/providers (which provider said it, the maker and part number it matched, its own words, and when we asked). THE LAST BUY DATE HAS PASSED (2025-12-30, 320 day(s) ago), so this is a substitution rather than a buy. The distributor reports NONE left. No approved alternate is recorded for this position.

  13. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 21 boards×21

    21 checked by the review by Claude walker W4 (Cern session) — 20 different reasons

    At along@L4_POWER.CU:ETH_2V5|GND, along@L4_POWER.CU:GND|PL_1V8, along@L4_POWER.CU:GND|PS_3V3, refchange:GND|PL_3V3, refchange:GND|PS_1V8, refchange:GND|PS_3V3 and 15 more

    Each finding, with its own headline
    • along@L4_POWER.CU:ETH_2V5|GND'/Top/Kria/Kria_ethernet/GEM_LED0' runs ALONG the split between the ETH_2V5 and GND planes on L4_POWER.CU at 1 place, the longest 0.9 mm; the impedance rises from 64 ohm to 80 ohm there at worst, 11.1 % of each edge reflected (fastest: /Top/Kria/Kria_ethernet/GEM_LED0 at 1 ns 10-90 %, suggested by its name)

      On L4_POWER.CU the ETH_2V5 and GND planes part, and '/Top/Kria/Kria_ethernet/GEM_LED0' runs down the slot between them at 1 place(s), rather than across it. This is not a crossing: the return still follows the track, in the plane on its other side and in the slot's edges, so it does not detour. What changes is the track's own impedance. With the plane gone from under it on this side for that length, its field reaches further, to the slot's edges and to the next plane beyond, so its capacitance falls and its impedance rises; every edge travelling down it is partly reflected where it enters the stretch and again where it leaves, by (Z2 - Z1)/(Z2 + Z1). The longer the stretch against the edge's rise, the more of the edge sees the full step. A crossing is generally the worse fault; a mismatch is not free. Over solid plane each track is 64 ohm; along the slot 80 ohm, reflecting up to 11.1 % of each edge (a method-of-moments field solve of the cross-section, one dielectric; were the slot's edges to carry none of the return, up to 86 ohm). No place stays within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is a warning. One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • '/Top/Kria/Kria_ethernet/GEM_LED0' on L3_SIG.CU at (407.8, -283.28), 5.91 mm, running ALONG the split for 0.9 mm, not across it, 64 ohm over solid plane, 80 ohm along it: 11.1 % of each edge reflected; other side: no plane on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=along%40L4_POWER.CU%3AETH_2V5%7CGND serves each with its own layer_image region; all of them lie in x 405.75..409.49, y -285.94..-278.75 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • along@L4_POWER.CU:GND|PL_1V82 nets run ALONG the split between the GND and PL_1V8 planes on L4_POWER.CU at 2 places, the longest 3.3 mm; the impedance rises from 61 ohm to 71 ohm there at worst, 7.4 % of each edge reflected (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_P at 1 ns 10-90 %, suggested by its name)

      On L4_POWER.CU the GND and PL_1V8 planes part, and 2 nets run down the slot between them at 2 place(s), rather than across it. This is not a crossing: the return still follows the track, in the plane on its other side and in the slot's edges, so it does not detour. What changes is the track's own impedance. With the plane gone from under it on this side for that length, its field reaches further, to the slot's edges and to the next plane beyond, so its capacitance falls and its impedance rises; every edge travelling down it is partly reflected where it enters the stretch and again where it leaves, by (Z2 - Z1)/(Z2 + Z1). The longer the stretch against the edge's rise, the more of the edge sees the full step. A crossing is generally the worse fault; a mismatch is not free. Over solid plane each track is 61 ohm; along the slot 71 ohm, reflecting up to 7.4 % of each edge (a method-of-moments field solve of the cross-section, one dielectric; were the slot's edges to carry none of the return, up to 71 ohm). No place stays within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is a warning. One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' on L3_SIG.CU at (285.53, -179.28), 11.25 mm, running ALONG the split for 1.95 mm, not across it, 61 ohm over solid plane, 71 ohm along it: 7.4 % of each edge reflected; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_ethernet/ETH_RX_D2' on L3_SIG.CU at (353.96, -175.49), 5.4 mm, running ALONG the split for 3.3 mm, not across it, 61 ohm over solid plane, 71 ohm along it: 7.4 % of each edge reflected; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=along%40L4_POWER.CU%3AGND%7CPL_1V8 serves each with its own layer_image region; all of them lie in x 280.72..357.32, y -184.09..-174 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • along@L4_POWER.CU:GND|PS_3V33 nets run ALONG the split between the GND and PS_3V3 planes on L4_POWER.CU at 3 places, the longest 41.62 mm; the impedance rises from 61 ohm to 71 ohm there at worst, 7 % of each edge reflected (fastest: /Top/Kria/Kria_SFP+/SFP1_RD_N at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and PS_3V3 planes part, and 3 nets run down the slot between them at 3 place(s), rather than across it. This is not a crossing: the return still follows the track, in the plane on its other side and in the slot's edges, so it does not detour. What changes is the track's own impedance. With the plane gone from under it on this side for that length, its field reaches further, to the slot's edges and to the next plane beyond, so its capacitance falls and its impedance rises; every edge travelling down it is partly reflected where it enters the stretch and again where it leaves, by (Z2 - Z1)/(Z2 + Z1). The longer the stretch against the edge's rise, the more of the edge sees the full step. A crossing is generally the worse fault; a mismatch is not free. Over solid plane each track is 61 ohm; along the slot 69 ohm to 71 ohm, reflecting up to 7 % of each edge (a method-of-moments field solve of the cross-section, one dielectric; were the slot's edges to carry none of the return, up to 71 ohm). At 1 place(s) the impedance cannot be said: the track's centreline is not over the slot. No place stays within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is a warning. One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • '/Top/Kria/Kria_ethernet/GEM_LED2' on L3_SIG.CU at (404.75, -278.88), 11.33 mm, running ALONG the split for 9.15 mm, not across it, impedance not said: the track's centreline is not over the slot; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP1_RD_N' on L3_SIG.CU at (388.12, -143.76), 43.13 mm, running ALONG the split for 41.62 mm, not across it, 61 ohm over solid plane, 71 ohm along it: 7 % of each edge reflected; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • 'Net-(U19-SEL)' on L3_SIG.CU at (367.38, -258.32), 12.62 mm, running ALONG the split for 11.19 mm, not across it, 61 ohm over solid plane, 69 ohm along it: 5.6 % of each edge reflected; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=along%40L4_POWER.CU%3AGND%7CPS_3V3 serves each with its own layer_image region; all of them lie in x 365.49..410.69, y -285.93..-142.76 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|PL_3V327 nets change reference between GND and PL_3V3 at 50 vias; 25 of them partial (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      At 50 via(s), on 27 net(s), the signal leaves a layer referenced by one of GND and PL_3V3 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 50 with none. 50 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and PL_3V3 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. 25 of them are partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go. The pure ones are listed first, then the partial ones with the farthest stitching via first.

      • '/Top/Kria/Kria_SFP+/SFP0_TD_P' at (283.09, -149.32), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 267.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_TD_P' at (436.82, -258.3), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU; partial, GND continues through a via of GND 3.71 mm away; stitching none, 267.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_TD_N' at (283.09, -149.92), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 267.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_TD_N' at (436.22, -258.3), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU; partial, GND continues through a via of GND 3.76 mm away; stitching none, 267.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' at (347.27, -156.93), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 211.9 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' at (455.57, -267.02), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU; partial, GND continues through a via of GND 2.56 mm away; stitching none, 211.9 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_RX_LOS' at (347.26, -156.29), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 200.4 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_RX_LOS' at (441.32, -267.05), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU; partial, GND continues through a via of GND 2.59 mm away; stitching none, 200.4 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_IOs/CLK_SR' at (349.6, -159), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 198.5 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_IOs/CLK_SR' at (201.45, -280.55), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU; partial, GND continues through a via of GND 1.7 mm away; stitching none, 198.5 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (388.07, -145.12), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 198.3 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (485.6, -269.8), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to USB_PWRD_VCCIO on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 198.3 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • and 38 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CPL_3V3 serves each with its own layer_image region; all of them lie in x 120..487.6, y -282.55..-143.12 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|PS_1V83 nets change reference between GND and PS_1V8 at 4 vias, none with a reference that continues across (fastest: /Top/Kria/Kria_SOM240_1_connector/WATCHDOG_WDI at 100 ps 10-90 %, ASSUMED)

      At 4 via(s), on 3 net(s), the signal leaves a layer referenced by one of GND and PS_1V8 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 4 with none. 4 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and PS_1V8 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/Kria_FTDI/~{EEPROM_WC}' at (468, -271.25), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to PS_1V8 on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 182.6 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SOM240_1_connector/WATCHDOG_WDI' at (393.7, -177.3), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to PS_1V8 on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 54.0 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SOM240_1_connector/SD_RST_L' at (395.6, -156.7), L8_SIG.CU -> L12_BOTTOM.CU: PS_1V8 on L7_POWER.CU, 5V0_IN on L9_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 5.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (20 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_FTDI/~{EEPROM_WC}' at (379.7, -169.85), L8_SIG.CU -> L12_BOTTOM.CU: PS_1V8 on L7_POWER.CU, 5V0_IN on L9_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 1.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CPS_1V8 serves each with its own layer_image region; all of them lie in x 377.7..470, y -273.25..-154.7 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|PS_3V39 nets change reference between GND and PS_3V3 at 10 vias; 2 of them partial (fastest: /Top/Kria/Kria_SD_card/SD_DETECT at 100 ps 10-90 %, ASSUMED)

      At 10 via(s), on 9 net(s), the signal leaves a layer referenced by one of GND and PS_3V3 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 10 with none. 10 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and PS_3V3 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. 2 of them are partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go. The pure ones are listed first, then the partial ones with the farthest stitching via first.

      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (388.07, -145.12), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 198.3 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SD_card/SD_DETECT' at (389.95, -278.43), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 144.8 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U19-DAT0A)' at (386.99, -273.24), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 135.1 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U19-DAT2A)' at (388.2, -279.6), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 134.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U19-DAT3A)' at (388.35, -278.6), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 134.0 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U19-CMDA)' at (387.9, -278.15), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 132.3 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U19-DAT1A)' at (386.3, -271.75), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 131.1 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (388.07, -145.12), L5_MIX.CU -> L10_SIG.CU: PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; pure, no reference continues across; stitching none, 56.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_ethernet/GEM_LED2' at (403.45, -273), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, PS_3V3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.46 mm away; stitching none, 2.8 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_ethernet/GEM_LED1' at (403.45, -273.87), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, PS_3V3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.13 mm away; stitching none, 2.2 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PS_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CPS_3V3 serves each with its own layer_image region; all of them lie in x 384.3..405.45, y -281.6..-143.12 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|USB_PWRD_VCCIO'/Top/Kria/Kria_FTDI/PS_POR_L' changes reference between GND and USB_PWRD_VCCIO at a via, none with a reference that continues across (fastest: /Top/Kria/Kria_FTDI/PS_POR_L at 2 ns 10-90 %, suggested by its name)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of GND and USB_PWRD_VCCIO for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 1 with none. 1 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and USB_PWRD_VCCIO are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (485.6, -269.8), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to USB_PWRD_VCCIO on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 198.3 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and USB_PWRD_VCCIO are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CUSB_PWRD_VCCIO serves each with its own layer_image region; all of them lie in x 483.6..487.6, y -271.8..-267.8 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|UTIL_3V32 nets change reference between GND and UTIL_3V3 at 4 vias, none with a reference that continues across (fastest: /Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L at 100 ps 10-90 %, ASSUMED)

      At 4 via(s), on 2 net(s), the signal leaves a layer referenced by one of GND and UTIL_3V3 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 4 with none. 4 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and UTIL_3V3 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L' at (392.05, -130.25), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to UTIL_3V3 on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 92.9 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and UTIL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L' at (381.85, -109.95), L8_SIG.CU -> L12_BOTTOM.CU: UTIL_3V3 on L7_POWER.CU, 5V0_IN on L9_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 91.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and UTIL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L' at (392.05, -130.25), L8_SIG.CU -> L12_BOTTOM.CU: UTIL_3V3 on L7_POWER.CU, 5V0_IN on L9_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 91.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and UTIL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RST_L' at (393.4, -153.2), L8_SIG.CU -> L12_BOTTOM.CU: UTIL_3V3 on L7_POWER.CU, 5V0_IN on L9_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 1.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and UTIL_3V3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CUTIL_3V3 serves each with its own layer_image region; all of them lie in x 379.85..395.4, y -155.2..-107.95 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:283.26,-176.84'/Top/TimIQ/TimIQ_PLL/PLL_LE' crosses where the outline of the T_+12V plane on L7_POWER.CU and the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.24 mm; 1 cleared on both sides of the track (fastest: /Top/TimIQ/TimIQ_PLL/PLL_LE at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_PLL/PLL_LE' on L8_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.24 mm under the track with no copper, at (283.26, -176.84), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 and of the pour at x 101..360, y -299..-101 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/TimIQ/TimIQ_PLL/PLL_LE' on L8_SIG.CU at (283.25, -176.84), 0.06 mm over no copper; other side: no plane on L9_POWER.CU / '/Top/TimIQ/TimIQ_PLL/PLL_LE' on L8_SIG.CU at (283.26, -176.84), 0.24 mm over no copper; other side: no plane on L7_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A283.26%2C-176.84 serves each with its own layer_image region; all of them lie in x 282.14..284.38, y -177.84..-175.84 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:413.81,-271.06'/Top/Kria/Kria_SOM240_1_connector/ETH_MDIO' crosses where the outline of the GND plane on L11_GND.CU and the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 1.06 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/Kria_SOM240_1_connector/ETH_MDIO at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SOM240_1_connector/ETH_MDIO' on L10_SIG.CU, the GND plane on L11_GND.CU has 1.06 mm under the track with no copper, at (413.73, -270.98), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 0.91 mm). No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 and of the pour at x 101..499, y -299..-101 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. 1 of the 2 crossing(s) have a plane on the track's other side at most twice as far away as the broken one (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The rest leave most or all of the return to the break, so the finding is a warning.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/Kria_SOM240_1_connector/ETH_MDIO' on L10_SIG.CU at (413.73, -270.98), 1.06 mm, running ALONG the bare copper for 0.91 mm, not across it over no copper; other side: 5V0_IN on L9_POWER.CU, 0.12 mm from the track against 0.15 mm to L11_GND.CU: 44% of the return on L11_GND.CU / '/Top/Kria/Kria_SOM240_1_connector/ETH_MDIO' on L10_SIG.CU at (413.81, -271.06), 0.2 mm over no copper; other side: no plane on L11_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A413.81%2C-271.06 serves each with its own layer_image region; all of them lie in x 412.46..415, y -272.54..-269.71 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:ETH_2V5@L4_POWER.CU:412.08,-277.82'/Top/Kria/Kria_ethernet/ETH_TD2_P' crosses where the outline of the GND plane on L2_GND.CU and the ETH_2V5 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.09 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/Kria_ethernet/ETH_TD2_P at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_ethernet/ETH_TD2_P' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.09 mm under the track with no copper, at (412.08, -277.82), running to the end of the track. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 and of the pour at x 405.5..418.75, y -279.75..-250 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/Kria_ethernet/ETH_TD2_P' on L3_SIG.CU at (412.08, -277.82), 0.09 mm over no copper, to the end of the track; other side: no plane on L4_POWER.CU / '/Top/Kria/Kria_ethernet/ETH_TD2_P' on L3_SIG.CU at (412.08, -277.82), 0.09 mm over no copper, to the end of the track; other side: no plane on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AETH_2V5%40L4_POWER.CU%3A412.08%2C-277.82 serves each with its own layer_image region; all of them lie in x 411.07..413.08, y -278.84..-276.75 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:ETH_2V5@L4_POWER.CU:412.47,-277.77'/Top/Kria/Kria_ethernet/ETH_TD2_N' crosses where the outline of the GND plane on L2_GND.CU and the ETH_2V5 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.12 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/Kria_ethernet/ETH_TD2_N at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_ethernet/ETH_TD2_N' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.12 mm under the track with no copper, at (412.47, -277.77), running to the end of the track. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 and of the pour at x 405.5..418.75, y -279.75..-250 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/Kria_ethernet/ETH_TD2_N' on L3_SIG.CU at (412.47, -277.77), 0.12 mm over no copper, to the end of the track; other side: no plane on L4_POWER.CU / '/Top/Kria/Kria_ethernet/ETH_TD2_N' on L3_SIG.CU at (412.47, -277.77), 0.12 mm over no copper, to the end of the track; other side: no plane on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AETH_2V5%40L4_POWER.CU%3A412.47%2C-277.77 serves each with its own layer_image region; all of them lie in x 411.47..413.52, y -278.85..-276.75 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L11_GND.CU:176.7,-253.9'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' crosses where the outline of the GND plane on L11_GND.CU is drawn back, at 1 place, the longest 0.33 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' on L12_BOTTOM.CU, the GND plane on L11_GND.CU has 0.33 mm under the track with no copper, at (176.7, -253.9), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 0.33 mm). No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' on L12_BOTTOM.CU at (176.7, -253.9), 0.33 mm, running ALONG the bare copper for 0.33 mm, not across it over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L11_GND.CU%3A176.7%2C-253.9 serves each with its own layer_image region; all of them lie in x 175.6..177.8, y -255.04..-252.8 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L11_GND.CU:176.8,-255'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' crosses where the outline of the GND plane on L11_GND.CU is drawn back, at 1 place, the longest 0.39 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' on L12_BOTTOM.CU, the GND plane on L11_GND.CU has 0.39 mm under the track with no copper, at (176.8, -255), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' on L12_BOTTOM.CU at (176.8, -255), 0.39 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L11_GND.CU%3A176.8%2C-255 serves each with its own layer_image region; all of them lie in x 175.8..177.8, y -256.19..-253.81 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L11_GND.CU:176.8,-256.15'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' crosses where the outline of the GND plane on L11_GND.CU is drawn back, at 1 place, the longest 0.39 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' on L12_BOTTOM.CU, the GND plane on L11_GND.CU has 0.39 mm under the track with no copper, at (176.8, -256.15), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_RF_out/FD1_DIV1' on L12_BOTTOM.CU at (176.8, -256.15), 0.39 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L11_GND.CU%3A176.8%2C-256.15 serves each with its own layer_image region; all of them lie in x 175.8..177.8, y -257.34..-254.96 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L11_GND.CU:440.01,-257.13'/Top/Kria/Kria_SFP+/SFP0_MOD_ABS' crosses where the outline of the GND plane on L11_GND.CU is drawn back, at 1 place, the longest 1.06 mm (fastest: /Top/Kria/Kria_SFP+/SFP0_MOD_ABS at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP0_MOD_ABS' on L10_SIG.CU, the GND plane on L11_GND.CU has 1.06 mm under the track with no copper, at (440.01, -257.13), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/Kria/Kria_SFP+/SFP0_MOD_ABS' on L10_SIG.CU at (440.01, -257.13), 1.06 mm over no copper; other side: no plane on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L11_GND.CU%3A440.01%2C-257.13 serves each with its own layer_image region; all of them lie in x 439.01..441.01, y -258.66..-255.59 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:407.07,-274.75'/Top/Kria/Kria_ethernet/GEM_LED1' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.1 mm (fastest: /Top/Kria/Kria_ethernet/GEM_LED1 at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_ethernet/GEM_LED1' on L1_TOP.CU, the GND plane on L2_GND.CU has 0.1 mm under the track with no copper, at (407.07, -274.75), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/Kria/Kria_ethernet/GEM_LED1' on L1_TOP.CU at (407.07, -274.75), 0.1 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A407.07%2C-274.75 serves each with its own layer_image region; all of them lie in x 406.04..408.11, y -275.79..-273.72 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:449.09,-256.09'/Top/Kria/Kria_SFP+/SFP1_TD_N' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.29 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TD_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_TD_N' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.29 mm under the track with no copper, at (449.09, -256.09), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/Kria/Kria_SFP+/SFP1_TD_N' on L3_SIG.CU at (449.09, -256.09), 0.29 mm over no copper; other side: no plane on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A449.09%2C-256.09 serves each with its own layer_image region; all of them lie in x 448.01..450.17, y -257.17..-254.96 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:449.65,-256.64'/Top/Kria/Kria_SFP+/SFP1_TD_N' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.34 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TD_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_TD_N' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.34 mm under the track with no copper, at (449.65, -256.64), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/Kria/Kria_SFP+/SFP1_TD_N' on L3_SIG.CU at (449.65, -256.64), 0.34 mm over no copper; other side: no plane on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A449.65%2C-256.64 serves each with its own layer_image region; all of them lie in x 448.53..450.77, y -257.76..-255.52 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:PL_1V8@L4_POWER.CU:347.36,-159.46'/Top/Kria/Kria_IOs/CLK_SR' crosses where the outline of the PL_1V8 plane on L4_POWER.CU and the PL_3V3 plane on L6_POWER.CU is drawn back, at 1 place, the longest 0.1 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/Kria_IOs/CLK_SR at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_IOs/CLK_SR' on L5_MIX.CU, the PL_1V8 plane on L4_POWER.CU has 0.1 mm under the track with no copper, at (347.36, -159.46), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 and of the pour at x 275..355.5, y -175..-127.5 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/Kria_IOs/CLK_SR' on L5_MIX.CU at (347.36, -159.46), 0.1 mm over no copper; other side: no plane on L6_POWER.CU / '/Top/Kria/Kria_IOs/CLK_SR' on L5_MIX.CU at (347.36, -159.46), 0.1 mm over no copper; other side: no plane on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3APL_1V8%40L4_POWER.CU%3A347.36%2C-159.46 serves each with its own layer_image region; all of them lie in x 346.31..348.41, y -160.46..-158.46 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:PL_3V3@L6_POWER.CU:357.36,-166.49'/Top/Kria/Kria_SD_card/SD_DETECT' crosses where the outline of the PL_3V3 plane on L6_POWER.CU is drawn back, at 1 place, the longest 0.06 mm (fastest: /Top/Kria/Kria_SD_card/SD_DETECT at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SD_card/SD_DETECT' on L5_MIX.CU, the PL_3V3 plane on L6_POWER.CU has 0.06 mm under the track with no copper, at (357.36, -166.49), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/Kria/Kria_SD_card/SD_DETECT' on L5_MIX.CU at (357.36, -166.49), 0.06 mm over no copper; other side: no plane on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3APL_3V3%40L6_POWER.CU%3A357.36%2C-166.49 serves each with its own layer_image region; all of them lie in x 356.33..358.39, y -167.49..-165.49 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

  14. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 4 boards×4

    4 checked by the review by Claude walker W4 (Cern session) — 4 different reasons

    At refchange:5V0_IN|GND, refchange:GND|PL_1V8, refchange:GND|T_+12V, refchange:GND|T_5V_1

    Each finding, with its own headline
    • refchange:5V0_IN|GND66 nets change reference between 5V0_IN and GND at 133 vias; 73 of them partial (fastest: /Top/Kria/DAC2_CS_L at 100 ps 10-90 %, ASSUMED)

      At 133 via(s), on 66 net(s), the signal leaves a layer referenced by one of 5V0_IN and GND for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 133 with none. 133 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether 5V0_IN and GND are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. 73 of them are partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go. The pure ones are listed first, then the partial ones with the farthest stitching via first.

      • '/Top/Kria/Kria_SFP+/SFP1_MOD_ABS' at (453.97, -267.07), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 3.28 mm away; stitching none, 283.2 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_MOD_ABS' at (280.61, -176.61), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 0.64 mm away; stitching none, 283.2 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_MOD_ABS' at (439.72, -264.55), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 2.03 mm away; stitching none, 246.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_MOD_ABS' at (280.55, -175.34), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 0.84 mm away; stitching none, 246.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' at (280.58, -175.98), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 1.27 mm away; stitching none, 242.3 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' at (437.32, -262.93), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 1.1 mm away; stitching none, 242.3 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_SDA' at (452.38, -266.77), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 2.43 mm away; stitching none, 239.2 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (10 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_SDA' at (282.9, -183.3), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 2.06 mm away; stitching none, 239.2 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (10 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_SCL' at (453.17, -266.27), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 2.43 mm away; stitching none, 238.7 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (10 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_SCL' at (282.9, -182.65), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 1.7 mm away; stitching none, 238.7 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (10 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_FTDI/JTAG_TMS_C2M' at (487.24, -252.89), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 4.24 mm away; stitching none, 236.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_FTDI/JTAG_TMS_C2M' at (347.55, -159.85), L1_TOP.CU -> L10_SIG.CU: GND on L2_GND.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 0.89 mm away; stitching none, 236.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • and 121 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3A5V0_IN%7CGND serves each with its own layer_image region; all of them lie in x 121.5..489.24, y -282.5..-107.95 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|PL_1V827 nets change reference between GND and PL_1V8 at 30 vias; 18 of them partial (fastest: /Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_2/VCCOEN_PL_M2C at 100 ps 10-90 %, ASSUMED)

      At 30 via(s), on 27 net(s), the signal leaves a layer referenced by one of GND and PL_1V8 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 30 with none. 30 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and PL_1V8 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. 18 of them are partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go. The pure ones are listed first, then the partial ones with the farthest stitching via first.

      • '/Top/Kria/Kria_SFP+/SFP0_TD_P' at (283.09, -149.32), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 267.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_TD_N' at (283.09, -149.92), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 267.6 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' at (347.27, -156.93), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 211.9 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_RX_LOS' at (347.26, -156.29), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 200.4 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_IOs/CLK_SR' at (349.6, -159), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 198.5 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SD_card/SD_DETECT' at (352.25, -166.75), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 144.8 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_LED1' at (347.35, -151.85), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 97.7 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_LED2' at (347.3, -152.5), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 97.2 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_LED1' at (347.35, -153.75), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 84.1 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP0_LED2' at (347.31, -154.39), L1_TOP.CU -> L5_MIX.CU: GND on L2_GND.CU to PL_1V8 on L4_POWER.CU, PL_3V3 on L6_POWER.CU; pure, no reference continues across; stitching none, 83.2 mm of track past the via on L5_MIX.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' at (346.18, -170.57), L3_SIG.CU -> L10_SIG.CU: GND on L2_GND.CU, PL_1V8 on L4_POWER.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; partial, GND continues through a via of GND 1.11 mm away; stitching none, 68.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (3 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_2/CC_PL_PGOOD' at (312.1, -156.8), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, PL_1V8 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 3.55 mm away; stitching none, 54.4 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (20 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and PL_1V8 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • and 18 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CPL_1V8 serves each with its own layer_image region; all of them lie in x 273.4..359.5, y -175.8..-147.32 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|T_+12V26 nets change reference between GND and T_+12V at 51 vias, none with a reference that continues across (fastest: /Top/Kria/DAC3_CS_L at 100 ps 10-90 %, ASSUMED)

      At 51 via(s), on 26 net(s), the signal leaves a layer referenced by one of GND and T_+12V for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 51 with none. 51 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and T_+12V are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/DAC3_MISO' at (352.34, -157.56), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 227.8 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/DAC3_MOSI' at (352.38, -156.93), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 227.0 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/DAC3_SCK' at (350.05, -157.7), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 226.7 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/DAC3_CS_L' at (350.05, -157.1), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 226.1 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_CE' at (123.5, -269), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 201.5 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_CE' at (285.75, -174.7), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 201.5 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_LOCKED' at (128.25, -276.8), L3_SIG.CU -> L8_SIG.CU: GND on L2_GND.CU, T_3V3_2 on L4_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 200.5 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_LOCKED' at (287.65, -178.4), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 200.5 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_LE' at (133.5, -272.05), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 195.8 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_LE' at (286.85, -176.4), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 195.8 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_CLK' at (135.25, -272.3), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 195.4 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_PLL/PLL_CLK' at (285.73, -174.07), L1_TOP.CU -> L8_SIG.CU: GND on L2_GND.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching none, 195.4 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_+12V are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • and 39 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CT_%2B12V serves each with its own layer_image region; all of them lie in x 121.5..356.6, y -282.5..-154.93 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|T_5V_115 nets change reference between GND and T_5V_1 at 16 vias; 12 of them partial (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      At 16 via(s), on 15 net(s), the signal leaves a layer referenced by one of GND and T_5V_1 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 16 with none. 16 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and T_5V_1 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. 12 of them are partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go. The pure ones are listed first, then the partial ones with the farthest stitching via first.

      • 'Net-(U76-+)' at (134.12, -193.12), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.11 mm away; stitching none, 19.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U76-+)' at (134.5, -187), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.75 mm away; stitching none, 19.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_N0' at (124.5, -236), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 3.99 mm away; stitching none, 2.9 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/DAC0_CS_L' at (124.48, -182.25), L5_MIX.CU -> L12_BOTTOM.CU: T_5V_1 on L4_POWER.CU, PL_3V3 on L6_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 2.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_S0' at (123.75, -234.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.95 mm away; stitching none, 2.2 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/DAC0_MOSI' at (122, -173.7), L5_MIX.CU -> L12_BOTTOM.CU: T_5V_1 on L4_POWER.CU, PL_3V3 on L6_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 1.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_N2' at (123, -231.25), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.57 mm away; stitching none, 1.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/DAC0_SCK' at (122.53, -179.58), L5_MIX.CU -> L12_BOTTOM.CU: T_5V_1 on L4_POWER.CU, PL_3V3 on L6_POWER.CU to GND on L11_GND.CU; pure, no reference continues across; stitching none, 1.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_N1' at (122.75, -233), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.46 mm away; stitching none, 1.2 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' at (122.5, -229.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.82 mm away; stitching none, 1.1 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD1_N1' at (136, -219.25), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.1 mm away; stitching none, 1.1 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD1_S0' at (136.75, -215.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_1 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 0.92 mm away; stitching none, 1.1 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_1 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • and 4 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CT_5V_1 serves each with its own layer_image region; all of them lie in x 119.5..138.92, y -238..-170.1 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

  15. observationOBS-81minorOpen

    ±15 V TimIQ rails (U58, U60): output capacitors are 16 V parts at 15 V; effective capacitance after DC bias is marginal against the 10 µF stability minimum

    The ±15 V TimIQ rails use the same output capacitors as the 3.3 V and 5 V rails: five 10 µF 16 V 0805 ceramics each. At 15 V they run at 94 % of their rating. A class-2 0805 at that bias keeps only a small part of its nominal value, so the effective output capacitance can fall to about the 10 µF minimum that both regulators need for stability, or below it.

    Rails affected

    RailRegulatorOutput caps (BOM)NominalBias / ratingDatasheet minimum
    T_15VU58 TPS7A4700 (24 → 15 V)C262, C265, C268, C271, C274: 10uF-16V-10p-Ceramic-0805 (+16 × 10 nF)50 µF15 V / 16 V = 94 %≥ 10 µF
    T_-15VU60 TPS7A3301 (−24 → −15 V)C270, C273, C276, C279, C281: 10uF-16V-10p-Ceramic-0805 (+17 × 10 nF)50 µF15 V / 16 V = 94 %≥ 10 µF (recommended 10 to 47 µF)
    QuantityLimitActualMargin
    Effective output capacitance at 15 V≥ 10 µF50 µF nominal. A 16 V 0805 X5R/X7R at 15 V typically keeps roughly 15 to 30 %, so about 7.5 to 15 µF, before the −10 % tolerance and temperaturefrom about −3 to +5 µF: marginal
    Capacitor working voltagenone stated94 % of ratingnone

    ``mermaid flowchart LR P24["T_+24V"] --> U58["U58 TPS7A4700"] --> P15["T_15V: 5 × 10 µF 16 V 0805 at 15 V"] N24["T_-24V"] --> U60["U60 TPS7A3301"] --> N15["T_-15V: 5 × 10 µF 16 V 0805 at 15 V"] ``

    Why it matters

    Both LDOs specify a minimum output capacitance for loop stability, and the ±15 V rails feed the eight LT1360 DAC output amplifiers and the eight LT1012 reference inverters. The BOM gives no MPN or dielectric for these capacitors, so the real DC-bias curve cannot be checked. A 16 V part on a 15 V rail leaves no voltage derating either.

    What would fix it

    Use 25 V or 35 V rated capacitors (1206 or larger) on T_15V and T_-15V. Alternatively, confirm from the chosen MPN's DC-bias curve that five parts still give more than 10 µF at 15 V over temperature.

  16. observationOBS-88minorOpen

    Carrier EEPROM U16 (M24C64): the schematic note swaps the memory and identification-page I2C addresses (memory is 0x51, ID page 0x59)

    The note beside the carrier EEPROM U16 (M24C64) on schematic sheet 8 has its two I2C addresses the wrong way round. It reads "Identification page : 0x51, Memory page : 0x59". With U16's chip-enable straps, the datasheet gives the memory array at 0x51 and the identification page at 0x59.

    How the address is formed

    ItemSourceValue
    E0 (pin 1)connectivity: PS_3V31
    E1 (pin 2), E2 (pin 3)connectivity: GND0, 0
    Memory array device typeM24C64 datasheet p.12: "1010b (to select the memory)"1010 E2 E1 E0 = 1010 001 = 0x51
    Identification page device typesame page: "1011b (to select the Identification page)"1011 001 = 0x59
    Schematic notesheet 8, 'Kria EEPROM' blockID page 0x51, memory 0x59 (swapped)

    ``mermaid flowchart LR SOM["Kria SOM I2C (1.8 V)"] --> U17["U17 PCA9306"] --> BUS["I2C_SCK_LS / I2C_SDA_LS (3.3 V)"] BUS --> MEM["U16 memory array: 0x51"] BUS --> ID["U16 identification page: 0x59"] BUS --> INA["U6 INA260: 0x40"] ``

    Why it matters

    The addresses themselves are fine and do not collide (U6 INA260 is at 0x40 with A0 = A1 = GND). But firmware or a programming script written from the drawing would write the board data into the lockable identification page and read the main array as if it were the ID page. The fix is to correct the note to "Memory: 0x51, Identification page: 0x59".

  17. observationOBS-89minorOpen

    Top silkscreen: reference designators printed end to end without a gap in the TimIQ DAC blocks (e.g. 'C529C517', 'C530C518')

    In the dense TimIQ DAC areas, several reference designators on the top silkscreen are printed end to end with no gap, so they read as one string. Others are stacked in a list away from their parts. At about 1.1 mm the text itself is legible; the problem is working out which designator belongs to which part during assembly, inspection or rework.

    What was measured

    Top silk, copper and paste rendered at 35 px/mm over x 160-200 mm, y -240 to -205 mm (the DAC3 block around U51):

    DesignatorsWhat the silk showsEffect
    C529 + C517printed as "C529C517" in one vertical line beside U121reads as a single designator
    C530 + C518printed as "C530C518" beside U122reads as a single designator
    R584 + C53x"R584C53…" run together in a list at the block edgeambiguous
    R580, R582, C524, R584stacked in a column at the shield-frame edge, several mm from the partsnot attributable to a part without the assembly drawing

    Character height ≈ 40 px at 35 px/mm ≈ 1.1 mm (legible). The same layout is repeated in the four TimIQ DAC channels.

    ``mermaid flowchart LR A["C529 silk text"] --- B["C517 silk text"] B -. "no gap" .- C["read as 'C529C517'"] ``

    Why it matters

    This is cosmetic for the fab, but it affects manual rework and inspection of the analog DAC chains, which use 0.1 % parts. Adding a space or offset between abutting designators, or relying on an assembly drawing, resolves it.

  18. observationOBS-121minorOpen

    TimIQ 3.3 V LDOs U56 (T_3V3_1) and U57 (T_3V3_2): 12 V → 3.3 V drop dissipates 1.5–2.3 W; junction exceeds 125 °C above ~50–65 °C ambient

    The two 3.3 V TimIQ LDOs, U56 (T_3V3_1, clock-output logic) and U57 (T_3V3_2, the LMX2541 PLL), are TPS7A4700s fed from +12 V. Each drops 8.7 V and dissipates 1.5 to 2.3 W. With the datasheet θJA of 32.5 °C/W, the junction runs 48 to 75 °C above ambient. At 25 °C that is acceptable, but the margin to the 125 °C limit is gone above roughly 50 to 65 °C ambient. The board uses commercial-grade parts (LT1012AC, LT6600C, LT1360C: 0 to 70 °C), so its likely rated ambient is up to 70 °C. No rated ambient is recorded anywhere.

    Numbers

    RegulatorRailLoad (datasheet)Iout typ / maxP = 8.7 V · ITj at 25 °CTj at 70 °CLimit
    U56 TPS7A4700T_3V3_13× NB4L52 (16 typ / 25 max mA), 3× MC100EPT21 (≈19 typ / 30 max mA), 6× 130R/82R input terminations (≈31 mA per channel), dividers and LED ≈6 mA≈0.20 A / 0.26 A1.8–2.3 W83–100 °C128–145 °C125 °C
    U57 TPS7A4700T_3V3_2U61 LMX2541 (170 typ / 204 max mA)0.17 A / 0.20 A1.5–1.8 W73–83 °C118–128 °C125 °C

    ``mermaid flowchart LR P12["T_+12V"] --> U56["U56 TPS7A4700<br/>drops 8.7 V"] --> R1["T_3V3_1: NB4L52 ×3, MC100EPT21 ×3, terminations"] P12 --> U57["U57 TPS7A4700<br/>drops 8.7 V"] --> R2["T_3V3_2: LMX2541"] ``

    Why it matters

    Both parts shut down thermally at their limit. The clock outputs and the PLL would drop out at the warm end of a commercial ambient range. This has the same root cause as OBS-77 (a 12 V input to low-voltage LDOs), but with less margin lost, so it is filed as minor. Lowering the LDO input (pre-regulating), or confirming a lower rated ambient, resolves it.

  19. observationOBS-214minorOpen

    LED footprint pad numbering is the reverse of the Kingbright drawing for D14-D17 (KP-3216SGC): pad 1 is the cathode on the board, terminal 1 is the anode in the datasheet

    The LED footprint for D14-D17 (KP-3216SGC) numbers its pads the reverse of the Kingbright drawing: pad 1 is the cathode on the board, terminal 1 is the anode in the datasheet. A pick-and-place programme that orients by terminal 1 fits the LEDs reversed, and they never light.

    Board versus datasheet

    The LEDs sit on footprint LED_1206_3216Metric_R.

    TerminalBoard (netlist)Kingbright KP-3216SGC drawing (p.1)
    1cathode, Net-(Dxx-K), to the Q18-Q21 drainsanode
    2anode, Net-(Dxx-A), via 180R to 5V0_INcathode, at the polarity-mark end

    Per LED: D14.1 to Q18.3 (drain), D14.2 to R138.2; likewise D15/Q19/R139, D16/Q20/R158, D17/Q21/R159.

    The drawing's terminal labels are not in the PDF text layer; they were read off the rendered page.

    Why it matters

    The electrical design is correct. The risk is at assembly: a reversed LED simply never lights. The same part sits on TIMIQ CPU (OHWR0060).

    Fix: assembly must orient by the cathode polarity mark, and the assembly documentation should say so. Renumbering the footprint pads to match the vendor drawing removes the ambiguity.

  20. DecapViaFanoutminorOpen

    C309 on 'T_5V_1': its rail pad reaches its via over 2.00 mm of track, adding about 0.50 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (within a quarter of the starting value)

    C309 on 'T_5V_1': its rail pad reaches its via over 2.00 mm of track, adding about 0.50 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (within a quarter of the starting value). C309.1 (rail) runs 2.00 mm along its copper to its nearest via, 1.00 mm past the 1 mm beside the pad: 0.50 nH; C309.2 (ground) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure. Each run past the pad as one conductor of a pair 1.56 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.50 nH in all. Put the vias at the pads (or in them), one per pad at least, the rail's and the ground's close together; the capacitor's own inductance as mounted is 1.33 nH (pad 0.10 nH + no via of the first net within 1 mm of its pad: the study's 0.6 nH per mm of 1.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    At C309

  21. ConnectorDomainMixedminorOpen

    ConnectorDomainMixed: 2 components×2

    2 checked by the review by Claude walker W4 (Cern session) — 2 different reasons

    At J24, J25

    Each finding, with its own headline
    • J24J24 takes 3 different supplies off the board on one connector: SOM_5V0 (5.044 V), PL_3V3 (3.315 V), PL_1V8 (1.8 V)

      J24 carries 3 different supplies out of the board on one connector: SOM_5V0 (5.044 V) on A59/A60/B60/C59/C60/D60, PL_3V3 (3.315 V) on B13/B14, PL_1V8 (1.8 V) on D01/D02. Declare the intent. Two rails on one connector is ordinary in a docking or power-distribution interface and a hazard in a signal one, and the difference is not visible from the netlist — but note that adjacent contacts at different voltages are exactly what a mirrored, reversed or off-by-one cable shorts together, so the cost of getting the mating wrong is a rail collision rather than a dead signal. If this is a standard connector, declare its family (connector_family_declare) and file the standard's contact map: supplies on the contacts it assigns to power are then not reported.

    • J25J25 takes 2 different supplies off the board on one connector: PL_3V3 (3.315 V), PL_1V8 (1.8 V)

      J25 carries 2 different supplies out of the board on one connector: PL_3V3 (3.315 V) on B59/B60/D59/D60, PL_1V8 (1.8 V) on A44/B42/C44/D42. Declare the intent. Two rails on one connector is ordinary in a docking or power-distribution interface and a hazard in a signal one, and the difference is not visible from the netlist — but note that adjacent contacts at different voltages are exactly what a mirrored, reversed or off-by-one cable shorts together, so the cost of getting the mating wrong is a rail collision rather than a dead signal. If this is a standard connector, declare its family (connector_family_declare) and file the standard's contact map: supplies on the contacts it assigns to power are then not reported.

  22. ReferencePlaneDiscontinuityminorOpen

    ReferencePlaneDiscontinuity: 2 boards×2

    2 checked by the review by Claude walker W4 (Cern session) — 2 different reasons

    At refchange:GND|T_5V_2, refchange:GND|T_5V_3

    Each finding, with its own headline
    • refchange:GND|T_5V_211 nets change reference between GND and T_5V_2 at 12 vias, each with another reference that continues across (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      At 12 via(s), on 11 net(s), the signal leaves a layer referenced by one of GND and T_5V_2 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 12 with none. 12 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and T_5V_2 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. Every one is partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go, and the finding is the lower tier (info). They are listed with the farthest stitching via first.

      • 'Net-(U94-+)' at (185.4, -152.61), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.93 mm away; stitching none, 19.7 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U94-+)' at (185.5, -146.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.75 mm away; stitching none, 19.7 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_N0' at (175, -195.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 3.58 mm away; stitching none, 2.4 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_S0' at (174.25, -194), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.5 mm away; stitching none, 1.7 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_N2' at (173.75, -190.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.95 mm away; stitching none, 1.4 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_DIV1' at (172.5, -186.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.26 mm away; stitching none, 1.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' at (173.5, -188.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.82 mm away; stitching none, 1.2 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD1_DIV1' at (187.67, -181.42), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 0.8 mm away; stitching none, 1.2 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD1_S0' at (187.5, -174.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.36 mm away; stitching none, 1.1 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_N1' at (173.5, -192.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.3 mm away; stitching none, 0.9 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD1_N0' at (186.75, -177.25), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.52 mm away; stitching none, 0.9 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD1_N1' at (186.75, -178.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.46 mm away; stitching none, 0.9 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CT_5V_2 serves each with its own layer_image region; all of them lie in x 170.5..189.67, y -197.5..-144.5 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|T_5V_311 nets change reference between GND and T_5V_3 at 12 vias, each with another reference that continues across (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      At 12 via(s), on 11 net(s), the signal leaves a layer referenced by one of GND and T_5V_3 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 12 with none. 12 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and T_5V_3 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. Every one is partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go, and the finding is the lower tier (info). They are listed with the farthest stitching via first.

      • 'Net-(U112-+)' at (229.25, -160.25), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.63 mm away; stitching none, 19.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'Net-(U112-+)' at (228.86, -166.36), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.05 mm away; stitching none, 19.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_N0' at (219, -209.25), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 3.26 mm away; stitching none, 2.7 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_S0' at (218.25, -207.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.82 mm away; stitching none, 1.9 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD1_DIV1' at (230.5, -194.92), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.57 mm away; stitching none, 1.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_N2' at (217.5, -204.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.52 mm away; stitching none, 1.3 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_N1' at (217.5, -206.25), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.27 mm away; stitching none, 1.2 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' at (217.25, -202.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.77 mm away; stitching none, 1.1 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_DIV1' at (216.25, -200.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 2.15 mm away; stitching none, 1.0 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD1_S0' at (230.5, -189.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.95 mm away; stitching none, 0.9 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD1_N0' at (232.25, -191), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.77 mm away; stitching none, 0.8 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD1_N1' at (232.25, -192.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_5V_3 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.75 mm away; stitching none, 0.8 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_5V_3 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CT_5V_3 serves each with its own layer_image region; all of them lie in x 214.25..234.25, y -211.25..-158.25 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

  23. ReferencePlaneDiscontinuityminorOpen

    ReferencePlaneDiscontinuity: 83 boards×83

    83 checked by the review by Claude walker W4 (Cern session) — 75 different reasons

    At refchange:5V0_IN|PL_3V3, refchange:5V0_IN|PS_3V3, refchange:5V0_IN|T_3V3_2, refchange:ETH_2V5|GND, refchange:GND|T_3V3_2, refchange:PL_3V3|T_+12V and 77 more

    Each finding, with its own headline
    • refchange:5V0_IN|PL_3V32 nets change reference between 5V0_IN and PL_3V3 at 2 vias, none with a reference that continues across (fastest: /Top/Kria/FD1_BYP at 100 ps 10-90 %, ASSUMED)

      At 2 via(s), on 2 net(s), the signal leaves a layer referenced by one of 5V0_IN and PL_3V3 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 2 served. Every site is served, so the finding is the lower tier (info). 2 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether 5V0_IN and PL_3V3 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (388.07, -145.12), L5_MIX.CU -> L10_SIG.CU: PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; pure, no reference continues across; stitching served, 56.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN on L9_POWER.CU is 0.120 mm from GND on L10_SIG.CU and PL_3V3 on L6_POWER.CU 0.120 mm from GND on L5_MIX.CU: their own capacitance carries the return. Served.
      • '/Top/Kria/FD1_BYP' at (255.8, -198.4), L5_MIX.CU -> L8_SIG.CU: GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching served, 38.3 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN on L9_POWER.CU is 0.120 mm from GND on L10_SIG.CU and PL_3V3 on L6_POWER.CU 0.120 mm from GND on L5_MIX.CU: their own capacitance carries the return. Served.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3A5V0_IN%7CPL_3V3 serves each with its own layer_image region; all of them lie in x 253.8..390.07, y -200.4..-143.12 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:5V0_IN|PS_3V3'/Top/Kria/Kria_FTDI/PS_POR_L' changes reference between 5V0_IN and PS_3V3 at a via, none with a reference that continues across (fastest: /Top/Kria/Kria_FTDI/PS_POR_L at 2 ns 10-90 %, suggested by its name)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of 5V0_IN and PS_3V3 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 1 served. Every site is served, so the finding is the lower tier (info). 1 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether 5V0_IN and PS_3V3 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/Kria_FTDI/PS_POR_L' at (388.07, -145.12), L5_MIX.CU -> L10_SIG.CU: PS_3V3 on L4_POWER.CU, PL_3V3 on L6_POWER.CU to 5V0_IN on L9_POWER.CU, GND on L11_GND.CU; pure, no reference continues across; stitching served, 56.6 mm of track past the via on L10_SIG.CU: At the 5 GHz knee of a 100 ps edge (2 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN on L9_POWER.CU is 0.120 mm from GND on L10_SIG.CU and PS_3V3 on L4_POWER.CU 0.120 mm from GND on L3_SIG.CU: their own capacitance carries the return. Served.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3A5V0_IN%7CPS_3V3 serves each with its own layer_image region; all of them lie in x 386.07..390.07, y -147.12..-143.12 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:5V0_IN|T_3V3_22 nets change reference between 5V0_IN and T_3V3_2 at 2 vias, none with a reference that continues across (fastest: /Top/TimIQ/TimIQ_PLL/PLL_LOCKED at 100 ps 10-90 %, ASSUMED)

      At 2 via(s), on 2 net(s), the signal leaves a layer referenced by one of 5V0_IN and T_3V3_2 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 2 served. Every site is served, so the finding is the lower tier (info). 2 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether 5V0_IN and T_3V3_2 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/TimIQ/TimIQ_PLL/PLL_LOCKED' at (128.25, -276.8), L3_SIG.CU -> L8_SIG.CU: GND on L2_GND.CU, T_3V3_2 on L4_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching served, 200.5 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN on L9_POWER.CU is 0.120 mm from GND on L10_SIG.CU and T_3V3_2 on L4_POWER.CU 0.120 mm from GND on L3_SIG.CU: their own capacitance carries the return. Served.
      • '/Top/TimIQ/TimIQ_PLL/PLL_ENout' at (124.5, -267.3), L3_SIG.CU -> L8_SIG.CU: GND on L2_GND.CU, T_3V3_2 on L4_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching served, 191.7 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (20 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. 5V0_IN on L9_POWER.CU is 0.120 mm from GND on L10_SIG.CU and T_3V3_2 on L4_POWER.CU 0.120 mm from GND on L3_SIG.CU: their own capacitance carries the return. Served.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3A5V0_IN%7CT_3V3_2 serves each with its own layer_image region; all of them lie in x 122.5..130.25, y -278.8..-265.3 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:ETH_2V5|GND4 nets change reference between ETH_2V5 and GND at 4 vias, each with another reference that continues across (fastest: /Top/Kria/Kria_ethernet/ETH_RX_CTRL at 1 ns 10-90 %, suggested by its name)

      At 4 via(s), on 4 net(s), the signal leaves a layer referenced by one of ETH_2V5 and GND for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 4 with none. 4 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether ETH_2V5 and GND are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. Every one is partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go, and the finding is the lower tier (info). They are listed with the farthest stitching via first.

      • '/Top/Kria/Kria_ethernet/ETH_RX_CTRL' at (407.5, -271), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, ETH_2V5 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 4.59 mm away; stitching none, 3.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. ETH_2V5 and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_ethernet/GEM_LED0' at (406.75, -276), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, ETH_2V5 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.8 mm away; stitching none, 2.7 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. ETH_2V5 and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_ethernet/ETH_RX_D0' at (410.5, -268.75), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, ETH_2V5 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 3.86 mm away; stitching none, 2.5 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. ETH_2V5 and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • '/Top/Kria/Kria_ethernet/ETH_RX_D2' at (409, -267.5), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, ETH_2V5 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.91 mm away; stitching none, 1.8 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (1 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. ETH_2V5 and GND are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AETH_2V5%7CGND serves each with its own layer_image region; all of them lie in x 404.75..412.5, y -278..-265.5 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:GND|T_3V3_2'/Top/TimIQ/TimIQ_PLL/PLL_LOCKED' changes reference between GND and T_3V3_2 at a via, each with another reference that continues across (fastest: /Top/TimIQ/TimIQ_PLL/PLL_LOCKED at 100 ps 10-90 %, ASSUMED)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of GND and T_3V3_2 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 1 with none. 1 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and T_3V3_2 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. Every one is partial: another reference continues across through a stitching via of its net, so only part of the return has nowhere to go, and the finding is the lower tier (info). They are listed with the farthest stitching via first.

      • '/Top/TimIQ/TimIQ_PLL/PLL_LOCKED' at (122.75, -267.3), L3_SIG.CU -> L12_BOTTOM.CU: GND on L2_GND.CU, T_3V3_2 on L4_POWER.CU to GND on L11_GND.CU; partial, GND continues through a via of GND 1.75 mm away; stitching none, 0.8 mm of track past the via on L12_BOTTOM.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and T_3V3_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AGND%7CT_3V3_2 serves each with its own layer_image region; all of them lie in x 120.75..124.75, y -269.3..-265.3 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:PL_3V3|T_+12V'/Top/Kria/FD1_BYP' changes reference between PL_3V3 and T_+12V at a via, none with a reference that continues across (fastest: /Top/Kria/FD1_BYP at 100 ps 10-90 %, ASSUMED)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of PL_3V3 and T_+12V for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 1 served. Every site is served, so the finding is the lower tier (info). 1 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether PL_3V3 and T_+12V are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/Kria/FD1_BYP' at (255.8, -198.4), L5_MIX.CU -> L8_SIG.CU: GND on L4_POWER.CU, PL_3V3 on L6_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching served, 38.3 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. PL_3V3 on L6_POWER.CU and T_+12V on L7_POWER.CU are adjacent, 0.150 mm apart: their own capacitance carries the return. Served.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3APL_3V3%7CT_%2B12V serves each with its own layer_image region; all of them lie in x 253.8..257.8, y -200.4..-196.4 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • refchange:T_+12V|T_3V3_22 nets change reference between T_+12V and T_3V3_2 at 2 vias, none with a reference that continues across (fastest: /Top/TimIQ/TimIQ_PLL/PLL_LOCKED at 100 ps 10-90 %, ASSUMED)

      At 2 via(s), on 2 net(s), the signal leaves a layer referenced by one of T_+12V and T_3V3_2 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 2 served. Every site is served, so the finding is the lower tier (info). 2 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether T_+12V and T_3V3_2 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • '/Top/TimIQ/TimIQ_PLL/PLL_LOCKED' at (128.25, -276.8), L3_SIG.CU -> L8_SIG.CU: GND on L2_GND.CU, T_3V3_2 on L4_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching served, 200.5 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. T_+12V on L7_POWER.CU is 0.120 mm from GND on L8_SIG.CU and T_3V3_2 on L4_POWER.CU 0.120 mm from GND on L3_SIG.CU: their own capacitance carries the return. Served.
      • '/Top/TimIQ/TimIQ_PLL/PLL_ENout' at (124.5, -267.3), L3_SIG.CU -> L8_SIG.CU: GND on L2_GND.CU, T_3V3_2 on L4_POWER.CU to T_+12V on L7_POWER.CU, 5V0_IN on L9_POWER.CU; pure, no reference continues across; stitching served, 191.7 mm of track past the via on L8_SIG.CU: At the 5 GHz knee of a 100 ps edge (20 ns 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. T_+12V on L7_POWER.CU is 0.120 mm from GND on L8_SIG.CU and T_3V3_2 on L4_POWER.CU 0.120 mm from GND on L3_SIG.CU: their own capacitance carries the return. Served.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=refchange%3AT_%2B12V%7CT_3V3_2 serves each with its own layer_image region; all of them lie in x 122.5..130.25, y -278.8..-265.3 mm (board frame, Y up).

      The reference on each layer is the first plane the track meets leaving the via, anti-pads and the net's own copper skipped. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:ETH_2V5|PS_3V312 nets cross the split between the ETH_2V5 and PS_3V3 planes on L4_POWER.CU at 12 places, all with a plane on the track's other side (fastest: /Top/Kria/Kria_SOM240_1_connector/ETH_GTX_CLK at 1 ns 10-90 %, suggested by its name)

      On L4_POWER.CU the ETH_2V5 and PS_3V3 planes part, and 12 nets cross from one to the other at 12 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 12 of the 12 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SOM240_1_connector/ETH_GTX_CLK' on L3_SIG.CU at (409.13, -263.63), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RX_CLK' on L3_SIG.CU at (406.33, -266.33), 1.2 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RX_D1' on L3_SIG.CU at (405.13, -266.88), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RX_D3' on L3_SIG.CU at (405.13, -267.63), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D0' on L3_SIG.CU at (409.13, -264.68), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D1' on L3_SIG.CU at (409.13, -265.18), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D2' on L3_SIG.CU at (409.13, -265.78), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D3' on L3_SIG.CU at (409.08, -266.33), 1.2 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_ethernet/ETH_RX_CTRL' on L3_SIG.CU at (405.13, -268.38), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_ethernet/ETH_RX_D0' on L3_SIG.CU at (405.58, -266.33), 1.2 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_ethernet/ETH_RX_D2' on L3_SIG.CU at (407.08, -266.33), 1.2 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_ethernet/GEM_LED2' on L3_SIG.CU at (405.13, -273.73), 1.06 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AETH_2V5%7CPS_3V3 serves each with its own layer_image region; all of them lie in x 403.75..410.5, y -275.1..-262.25 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|PL_1V852 nets cross the split between the GND and PL_1V8 planes on L4_POWER.CU at 52 places, all with a plane on the track's other side (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and PL_1V8 planes part, and 52 nets cross from one to the other at 52 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 52 of the 52 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/BOARD_RST_L' on L3_SIG.CU at (356.01, -142.3), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (274.5, -172.39), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_P' on L3_SIG.CU at (274.5, -171.81), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_N' on L3_SIG.CU at (274.5, -157.79), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_P' on L3_SIG.CU at (274.5, -157.21), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (279.82, -179.8), 0.75 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_EN_L' on L3_SIG.CU at (334.25, -175.51), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU at (334.8, -175.51), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' on L3_SIG.CU at (288.61, -175.39), 0.67 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_EN_L' on L3_SIG.CU at (335.75, -175.51), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_FAULT_L' on L3_SIG.CU at (336.14, -175.51), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_RST_N' on L3_SIG.CU at (290.48, -175.51), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • and 40 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CPL_1V8 serves each with its own layer_image region; all of them lie in x 272.99..359.64, y -180.95..-141.3 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|PS_3V338 nets cross the split between the GND and PS_3V3 planes on L4_POWER.CU at 55 places, all with a plane on the track's other side (fastest: /Top/Kria/Kria_SD_card/SD_DETECT at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and PS_3V3 planes part, and 38 nets cross from one to the other at 55 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 55 of the 55 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_FTDI/PS_POR_L' on L5_MIX.CU at (409.26, -146.4), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_FTDI/PS_SRST_C2M_L' on L3_SIG.CU at (367, -169.06), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_FTDI/PS_SRST_C2M_L' on L3_SIG.CU at (409.26, -169.06), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SD_card/SD_DETECT' on L5_MIX.CU at (367, -166.49), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_LED1' on L5_MIX.CU at (367, -152.81), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_LED1' on L5_MIX.CU at (409.26, -152.81), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_LED2' on L5_MIX.CU at (367, -153.91), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_LED2' on L5_MIX.CU at (409.26, -153.91), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_RD_N' on L3_SIG.CU at (367, -149.41), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_RD_N' on L3_SIG.CU at (409.26, -149.41), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_RD_P' on L3_SIG.CU at (367, -149.69), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_SFP+/SFP0_RD_P' on L3_SIG.CU at (409.26, -149.69), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • and 43 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CPS_3V3 serves each with its own layer_image region; all of them lie in x 365.49..410.76, y -286.01..-143.04 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|T_3V3_114 nets cross the split between the GND and T_3V3_1 planes on L4_POWER.CU at 19 places, all with a plane on the track's other side (fastest: /Top/Kria/FD1_DIV1 at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and T_3V3_1 planes part, and 14 nets cross from one to the other at 19 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 19 of the 19 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/FD1_DIV1' on L5_MIX.CU at (232.01, -244.16), 1.38 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/FD1_DIV1' on L5_MIX.CU at (236.68, -239.5), 1.43 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_EN_L' on L3_SIG.CU at (251.75, -272.51), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_EN_L' on L3_SIG.CU at (284.76, -239.5), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU at (258.9, -272.5), 1.41 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU at (291.91, -239.5), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' on L3_SIG.CU at (247.75, -239.5), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' on L3_SIG.CU at (248.55, -239.51), 1.38 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_EN_L' on L3_SIG.CU at (295.51, -240.8), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_FAULT_L' on L3_SIG.CU at (276.8, -272.51), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_FAULT_L' on L3_SIG.CU at (295.51, -243.3), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT1_RST_N' on L3_SIG.CU at (282.34, -239.5), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • and 7 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CT_3V3_1 serves each with its own layer_image region; all of them lie in x 230.49..297.01, y -274.01..-237.99 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|T_5V_18 nets cross the split between the GND and T_5V_1 planes on L4_POWER.CU at 8 places, all with a plane on the track's other side (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and T_5V_1 planes part, and 8 nets cross from one to the other at 8 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 8 of the 8 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/CLK_OUT0_D_N' on L3_SIG.CU at (146.01, -231.54), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_OUT0_D_P' on L3_SIG.CU at (146.01, -230.96), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (140.26, -221.29), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_P' on L3_SIG.CU at (140.26, -220.71), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (140.36, -181.6), 1.05 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MISO' on L5_MIX.CU at (140.26, -174.05), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MOSI' on L5_MIX.CU at (140.26, -174.64), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_SCK' on L5_MIX.CU at (140.26, -180.14), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CT_5V_1 serves each with its own layer_image region; all of them lie in x 138.75..147.51, y -232.54..-173.05 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|T_5V_26 nets cross the split between the GND and T_5V_2 planes on L4_POWER.CU at 10 places, all with a plane on the track's other side (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and T_5V_2 planes part, and 6 nets cross from one to the other at 10 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 10 of the 10 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/CLK_RF1_N' on L3_SIG.CU at (191.01, -176.81), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_P' on L3_SIG.CU at (190.99, -176), 1.4 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (156.77, -187.37), 1.24 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (191.01, -187.65), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MISO' on L5_MIX.CU at (157, -174.05), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MISO' on L5_MIX.CU at (191.01, -173.11), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MOSI' on L5_MIX.CU at (157, -174.64), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MOSI' on L5_MIX.CU at (191.01, -173.89), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_SCK' on L5_MIX.CU at (157, -180.14), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_SCK' on L5_MIX.CU at (190.99, -180.06), 1.47 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CT_5V_2 serves each with its own layer_image region; all of them lie in x 155.49..192.51, y -188.65..-172.11 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|T_5V_38 nets cross the split between the GND and T_5V_3 planes on L4_POWER.CU at 16 places, all with a plane on the track's other side (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      On L4_POWER.CU the GND and T_5V_3 planes part, and 8 nets cross from one to the other at 16 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 16 of the 16 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (231.06, -210.76), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (235.01, -206.81), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_P' on L3_SIG.CU at (230.24, -210.76), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_P' on L3_SIG.CU at (235.01, -205.99), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_N' on L3_SIG.CU at (200.75, -167.07), 1.43 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_N' on L3_SIG.CU at (235.01, -157.79), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_P' on L3_SIG.CU at (200.75, -166.24), 1.42 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF1_P' on L3_SIG.CU at (235.01, -157.21), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (200.75, -187.65), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (235.01, -187.65), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MISO' on L5_MIX.CU at (200.75, -173.11), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • '/Top/Kria/DAC0_MISO' on L5_MIX.CU at (235.01, -174.66), 1.43 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU
      • and 4 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CT_5V_3 serves each with its own layer_image region; all of them lie in x 199.24..236.51, y -212.26..-156.21 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|T_5V_44 nets cross the split between the GND and T_5V_4 planes on L4_POWER.CU at 8 places, all with a plane on the track's other side (fastest: /Top/Kria/CLK_OUT0_D_N at 1 ns 10-90 %, suggested by its name)

      On L4_POWER.CU the GND and T_5V_4 planes part, and 4 nets cross from one to the other at 8 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 8 of the 8 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/CLK_OUT0_D_N' on L3_SIG.CU at (157, -231.54), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_OUT0_D_N' on L3_SIG.CU at (196.41, -231.54), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_OUT0_D_P' on L3_SIG.CU at (157, -230.96), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_OUT0_D_P' on L3_SIG.CU at (196.41, -230.96), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (157, -221.29), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (196.41, -221.29), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_P' on L3_SIG.CU at (157, -220.71), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU
      • '/Top/Kria/CLK_RF0_P' on L3_SIG.CU at (196.41, -220.71), 1.01 mm; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CT_5V_4 serves each with its own layer_image region; all of them lie in x 155.49..197.91, y -232.54..-219.71 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L4_POWER.CU:GND|USB_PWRD_VCCIO'/Top/Kria/Kria_FTDI/PS_POR_L' crosses the split between the GND and USB_PWRD_VCCIO planes on L4_POWER.CU at 1 place, all with a plane on the track's other side (fastest: /Top/Kria/Kria_FTDI/PS_POR_L at 2 ns 10-90 %, suggested by its name)

      On L4_POWER.CU the GND and USB_PWRD_VCCIO planes part, and '/Top/Kria/Kria_FTDI/PS_POR_L' crosses from one to the other at 1 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 1 of the 1 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_FTDI/PS_POR_L' on L5_MIX.CU at (485.16, -254.5), 1.01 mm; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L4_POWER.CU%3AGND%7CUSB_PWRD_VCCIO serves each with its own layer_image region; all of them lie in x 484.16..486.16, y -256..-252.99 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L7_POWER.CU:PS_1V8|T_+12V4 nets cross the split between the PS_1V8 and T_+12V planes on L7_POWER.CU at 4 places, all with a plane on the track's other side (fastest: /Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L at 100 ps 10-90 %, ASSUMED)

      On L7_POWER.CU the PS_1V8 and T_+12V planes part, and 4 nets cross from one to the other at 4 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 4 of the 4 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L' on L8_SIG.CU at (362.78, -161.44), 3.6 mm; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RST_L' on L8_SIG.CU at (362.78, -165.7), 3.6 mm; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/SD_RST_L' on L8_SIG.CU at (362.78, -169.3), 3.6 mm; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/WATCHDOG_WDI' on L8_SIG.CU at (362.78, -170.61), 3.6 mm; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L7_POWER.CU%3APS_1V8%7CT_%2B12V serves each with its own layer_image region; all of them lie in x 359..364.6, y -171.61..-160.44 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • split@L7_POWER.CU:PS_1V8|UTIL_3V32 nets cross the split between the PS_1V8 and UTIL_3V3 planes on L7_POWER.CU at 2 places, all with a plane on the track's other side (fastest: /Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L at 100 ps 10-90 %, ASSUMED)

      On L7_POWER.CU the PS_1V8 and UTIL_3V3 planes part, and 2 nets cross from one to the other at 2 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 2 of the 2 crossing(s) the track's other neighbour carries a plane, which the return can use there. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_1/PWROFF_C2M_L' on L8_SIG.CU at (376.7, -155.8), 2.26 mm; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU
      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RST_L' on L8_SIG.CU at (389.7, -155.5), 2.01 mm; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=split%40L7_POWER.CU%3APS_1V8%7CUTIL_3V3 serves each with its own layer_image region; all of them lie in x 374.9..391.2, y -157.6..-154 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:280.31,-173.54'/Top/Kria/Kria_SFP+/SFP1_MOD_ABS' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.1 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_MOD_ABS at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_MOD_ABS' on L10_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.1 mm under the track with no copper, at (280.31, -173.54), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_MOD_ABS' on L10_SIG.CU at (280.31, -173.54), 0.1 mm over no copper; other side: GND on L11_GND.CU, 0.15 mm from the track against 0.12 mm to L9_POWER.CU: 56% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A280.31%2C-173.54 serves each with its own layer_image region; all of them lie in x 279.28..281.35, y -174.57..-172.5 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:280.33,-173.51'/Top/Kria/FD1_BYP' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.1 mm (fastest: /Top/Kria/FD1_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/FD1_BYP' on L8_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.1 mm under the track with no copper, at (280.33, -173.51), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/FD1_BYP' on L8_SIG.CU at (280.33, -173.51), 0.1 mm over no copper; other side: T_+12V on L7_POWER.CU, 0.12 mm from the track against 0.15 mm to L9_POWER.CU: 44% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A280.33%2C-173.51 serves each with its own layer_image region; all of them lie in x 279.27..281.36, y -174.54..-172.51 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:280.53,-174.95'/Top/Kria/FD1_N2' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.2 mm (fastest: /Top/Kria/FD1_N2 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/FD1_N2' on L8_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.2 mm under the track with no copper, at (280.53, -174.95), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/FD1_N2' on L8_SIG.CU at (280.53, -174.95), 0.2 mm over no copper; other side: T_+12V on L7_POWER.CU, 0.12 mm from the track against 0.15 mm to L9_POWER.CU: 44% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A280.53%2C-174.95 serves each with its own layer_image region; all of them lie in x 279.43..281.63, y -175.95..-173.95 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:283.25,-176.85'/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' crosses where the outline of the GND plane on L11_GND.CU and the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.24 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L10_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.24 mm under the track with no copper, at (283.25, -176.85), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 and of the pour at x 101..499, y -299..-101 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L10_SIG.CU at (283.25, -176.85), 0.06 mm over no copper; other side: no plane on L9_POWER.CU / '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L10_SIG.CU at (283.25, -176.85), 0.24 mm over no copper; other side: no plane on L11_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A283.25%2C-176.85 serves each with its own layer_image region; all of them lie in x 282.13..284.37, y -177.85..-175.85 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:347.34,-159.41'/Top/Kria/FD0_BYP' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.1 mm (fastest: /Top/Kria/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/FD0_BYP' on L10_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.1 mm under the track with no copper, at (347.34, -159.41), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/FD0_BYP' on L10_SIG.CU at (347.34, -159.41), 0.1 mm over no copper; other side: GND on L11_GND.CU, 0.15 mm from the track against 0.12 mm to L9_POWER.CU: 56% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A347.34%2C-159.41 serves each with its own layer_image region; all of them lie in x 346.3..348.37, y -160.45..-158.38 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:347.36,-159.46'/Top/Kria/Kria_IOs/CLK_LF' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.1 mm (fastest: /Top/Kria/Kria_IOs/CLK_LF at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_IOs/CLK_LF' on L8_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.1 mm under the track with no copper, at (347.36, -159.46), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_IOs/CLK_LF' on L8_SIG.CU at (347.36, -159.46), 0.1 mm over no copper; other side: T_+12V on L7_POWER.CU, 0.12 mm from the track against 0.15 mm to L9_POWER.CU: 44% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A347.36%2C-159.46 serves each with its own layer_image region; all of them lie in x 346.31..348.41, y -160.46..-158.46 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:349.68,-159.41'/Top/Kria/Kria_FTDI/JTAG_TMS_C2M' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.3 mm (fastest: /Top/Kria/Kria_FTDI/JTAG_TMS_C2M at 2 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_FTDI/JTAG_TMS_C2M' on L10_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.3 mm under the track with no copper, at (349.68, -159.41), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_FTDI/JTAG_TMS_C2M' on L10_SIG.CU at (349.68, -159.41), 0.3 mm over no copper; other side: GND on L11_GND.CU, 0.15 mm from the track against 0.12 mm to L9_POWER.CU: 56% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A349.68%2C-159.41 serves each with its own layer_image region; all of them lie in x 348.52..350.83, y -160.41..-158.41 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:349.68,-159.46'/Top/Kria/Kria_IOs/CLK_LF' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.34 mm (fastest: /Top/Kria/Kria_IOs/CLK_LF at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_IOs/CLK_LF' on L8_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.34 mm under the track with no copper, at (349.68, -159.46), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_IOs/CLK_LF' on L8_SIG.CU at (349.68, -159.46), 0.34 mm over no copper; other side: T_+12V on L7_POWER.CU, 0.12 mm from the track against 0.15 mm to L9_POWER.CU: 44% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A349.68%2C-159.46 serves each with its own layer_image region; all of them lie in x 348.53..350.86, y -160.46..-158.44 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:350.07,-156.71'/Top/Kria/DAC3_MISO' crosses where the outline of the T_+12V plane on L7_POWER.CU and the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.26 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/DAC3_MISO at 2 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/DAC3_MISO' on L8_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.26 mm under the track with no copper, at (350.07, -156.71), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 and of the pour at x 101..360, y -299..-101 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/DAC3_MISO' on L8_SIG.CU at (350.08, -156.71), 0.13 mm over no copper; other side: no plane on L9_POWER.CU / '/Top/Kria/DAC3_MISO' on L8_SIG.CU at (350.07, -156.71), 0.26 mm over no copper; other side: no plane on L7_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A350.07%2C-156.71 serves each with its own layer_image region; all of them lie in x 348.94..351.2, y -157.71..-155.71 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:352.5,-166.45'/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.2 mm (fastest: /Top/Kria/Kria_SFP+/SFP0_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' on L10_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.2 mm under the track with no copper, at (352.5, -166.45), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' on L10_SIG.CU at (352.5, -166.45), 0.2 mm over no copper; other side: GND on L11_GND.CU, 0.15 mm from the track against 0.12 mm to L9_POWER.CU: 56% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A352.5%2C-166.45 serves each with its own layer_image region; all of them lie in x 351.43..353.58, y -167.53..-165.38 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:5V0_IN@L9_POWER.CU:357.39,-166.61'/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' crosses where the outline of the 5V0_IN plane on L9_POWER.CU is drawn back, at 1 place, the longest 0.06 mm (fastest: /Top/Kria/Kria_SFP+/SFP0_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' on L10_SIG.CU, the 5V0_IN plane on L9_POWER.CU has 0.06 mm under the track with no copper, at (357.39, -166.61), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 100.9..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP0_TX_DISABLE' on L10_SIG.CU at (357.39, -166.61), 0.06 mm over no copper; other side: GND on L11_GND.CU, 0.15 mm from the track against 0.12 mm to L9_POWER.CU: 56% of the return on L9_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3A5V0_IN%40L9_POWER.CU%3A357.39%2C-166.61 serves each with its own layer_image region; all of them lie in x 356.36..358.42, y -167.61..-165.61 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:ETH_2V5@L4_POWER.CU:412.08,-279.88'/Top/Kria/Kria_ethernet/ETH_TD2_P' crosses where the outline of the ETH_2V5 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.25 mm (fastest: /Top/Kria/Kria_ethernet/ETH_TD2_P at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_ethernet/ETH_TD2_P' on L3_SIG.CU, the ETH_2V5 plane on L4_POWER.CU has 0.25 mm under the track with no copper, at (412.08, -279.88), running to the end of the track. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 405.5..418.75, y -279.75..-250 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_ethernet/ETH_TD2_P' on L3_SIG.CU at (412.08, -279.88), 0.25 mm over no copper, to the end of the track; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AETH_2V5%40L4_POWER.CU%3A412.08%2C-279.88 serves each with its own layer_image region; all of them lie in x 411.08..413.08, y -281..-278.75 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L11_GND.CU:433.37,-257.13'/Top/Kria/Kria_SFP+/SFP0_SDA' crosses where the outline of the GND plane on L11_GND.CU is drawn back, at 1 place, the longest 1.21 mm (fastest: /Top/Kria/Kria_SFP+/SFP0_SDA at 10 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP0_SDA' on L10_SIG.CU, the GND plane on L11_GND.CU has 1.21 mm under the track with no copper, at (433.37, -257.13), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP0_SDA' on L10_SIG.CU at (433.37, -257.13), 1.21 mm over no copper; other side: 5V0_IN on L9_POWER.CU, 0.12 mm from the track against 0.15 mm to L11_GND.CU: 44% of the return on L11_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L11_GND.CU%3A433.37%2C-257.13 serves each with its own layer_image region; all of them lie in x 432.37..434.37, y -258.73..-255.52 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L11_GND.CU:446.25,-266.05'/Top/Kria/Kria_SFP+/SFP1_SDA' crosses where the outline of the GND plane on L11_GND.CU is drawn back, at 1 place, the longest 1.36 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_SDA at 10 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_SDA' on L10_SIG.CU, the GND plane on L11_GND.CU has 1.36 mm under the track with no copper, at (446.25, -266.05), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_SDA' on L10_SIG.CU at (446.25, -266.05), 1.36 mm over no copper; other side: 5V0_IN on L9_POWER.CU, 0.12 mm from the track against 0.15 mm to L11_GND.CU: 44% of the return on L11_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L11_GND.CU%3A446.25%2C-266.05 serves each with its own layer_image region; all of them lie in x 444.57..447.93, y -267.05..-265.05 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:127.75,-221.25'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.48 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.48 mm under the track with no copper, at (127.75, -221.25), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' on L3_SIG.CU at (127.75, -221.25), 0.48 mm over no copper; other side: T_5V_1 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A127.75%2C-221.25 serves each with its own layer_image region; all of them lie in x 126.75..128.75, y -222.49..-220.01 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:127.75,-222.4'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.48 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.48 mm under the track with no copper, at (127.75, -222.4), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' on L3_SIG.CU at (127.75, -222.4), 0.48 mm over no copper; other side: T_5V_1 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A127.75%2C-222.4 serves each with its own layer_image region; all of them lie in x 126.75..128.75, y -223.64..-221.16 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:127.83,-220.18'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.3 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.3 mm under the track with no copper, at (127.83, -220.18), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator/FD0_BYP' on L3_SIG.CU at (127.83, -220.18), 0.3 mm over no copper; other side: T_5V_1 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A127.83%2C-220.18 serves each with its own layer_image region; all of them lie in x 126.75..128.9, y -221.34..-219.1 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:162.28,-254.86'/Top/TimIQ/TimIQ_phase_shifters/I3_N' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.43 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/I3_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/I3_N' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.43 mm under the track with no copper, at (162.28, -254.86), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/I3_N' on L3_SIG.CU at (162.28, -254.86), 0.43 mm over no copper; other side: T_5V_4 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A162.28%2C-254.86 serves each with its own layer_image region; all of them lie in x 161.28..163.28, y -256.08..-253.65 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:162.28,-256.01'/Top/TimIQ/TimIQ_phase_shifters/I3_N' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.43 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/I3_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/I3_N' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.43 mm under the track with no copper, at (162.28, -256.01), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/I3_N' on L3_SIG.CU at (162.28, -256.01), 0.43 mm over no copper; other side: T_5V_4 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A162.28%2C-256.01 serves each with its own layer_image region; all of them lie in x 161.28..163.28, y -257.23..-254.8 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:162.28,-257.16'/Top/TimIQ/TimIQ_phase_shifters/I3_N' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.43 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/I3_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/I3_N' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.43 mm under the track with no copper, at (162.28, -257.16), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/I3_N' on L3_SIG.CU at (162.28, -257.16), 0.43 mm over no copper; other side: T_5V_4 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A162.28%2C-257.16 serves each with its own layer_image region; all of them lie in x 161.28..163.28, y -258.38..-255.95 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:178.75,-180.75'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.48 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.48 mm under the track with no copper, at (178.75, -180.75), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' on L3_SIG.CU at (178.75, -180.75), 0.48 mm over no copper; other side: T_5V_2 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A178.75%2C-180.75 serves each with its own layer_image region; all of them lie in x 177.75..179.75, y -181.99..-179.51 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:178.75,-181.9'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.48 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.48 mm under the track with no copper, at (178.75, -181.9), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' on L3_SIG.CU at (178.75, -181.9), 0.48 mm over no copper; other side: T_5V_2 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A178.75%2C-181.9 serves each with its own layer_image region; all of them lie in x 177.75..179.75, y -183.14..-180.66 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:178.83,-179.68'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.3 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.3 mm under the track with no copper, at (178.83, -179.68), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator1/FD0_BYP' on L3_SIG.CU at (178.83, -179.68), 0.3 mm over no copper; other side: T_5V_2 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A178.83%2C-179.68 serves each with its own layer_image region; all of them lie in x 177.75..179.9, y -180.84..-178.6 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:222.5,-194.5'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.48 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.48 mm under the track with no copper, at (222.5, -194.5), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' on L3_SIG.CU at (222.5, -194.5), 0.48 mm over no copper; other side: T_5V_3 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A222.5%2C-194.5 serves each with its own layer_image region; all of them lie in x 221.5..223.5, y -195.74..-193.26 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:222.5,-195.65'/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.48 mm (fastest: /Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP at 100 ps 10-90 %, ASSUMED)

      Along '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.48 mm under the track with no copper, at (222.5, -195.65), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/TimIQ/TimIQ_phase_shifters/TimIQ_IQ_modulator2/FD0_BYP' on L3_SIG.CU at (222.5, -195.65), 0.48 mm over no copper; other side: T_5V_3 on L4_POWER.CU, 0.12 mm from the track against 0.15 mm to L2_GND.CU: 44% of the return on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A222.5%2C-195.65 serves each with its own layer_image region; all of them lie in x 221.5..223.5, y -196.89..-194.41 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:283.25,-176.86'/Top/TimIQ/TimIQ_PLL/PLL_ENout' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 0.06 mm (fastest: /Top/TimIQ/TimIQ_PLL/PLL_ENout at 20 ns 10-90 %, suggested by its name)

      Along '/Top/TimIQ/TimIQ_PLL/PLL_ENout' on L3_SIG.CU, the GND plane on L2_GND.CU has 0.06 mm under the track with no copper, at (283.25, -176.86), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/TimIQ/TimIQ_PLL/PLL_ENout' on L3_SIG.CU at (283.25, -176.86), 0.06 mm over no copper; other side: no plane on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A283.25%2C-176.86 serves each with its own layer_image region; all of them lie in x 282.22..284.28, y -177.86..-175.86 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:433.37,-257.13'/Top/Kria/Kria_SFP+/SFP0_TX_FAULT' crosses where the outline of the GND plane on L2_GND.CU and the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.21 mm; 1 cleared on both sides of the track (fastest: /Top/Kria/Kria_SFP+/SFP0_TX_FAULT at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP0_TX_FAULT' on L3_SIG.CU, the GND plane on L2_GND.CU has 1.21 mm under the track with no copper, at (433.37, -257.13), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 and of the pour at x 101..499, y -299.41..-101 is drawn back there. At 1 place(s) the track's plane is cleared on both sides of it, the same stretch on each reference layer: it has no plane on either side there, counted once. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • one stretch, cleared on 2 reference layers: '/Top/Kria/Kria_SFP+/SFP0_TX_FAULT' on L3_SIG.CU at (433.37, -257.13), 1.21 mm over no copper; other side: no plane on L4_POWER.CU / '/Top/Kria/Kria_SFP+/SFP0_TX_FAULT' on L3_SIG.CU at (433.37, -257.13), 1.21 mm over no copper; other side: no plane on L2_GND.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A433.37%2C-257.13 serves each with its own layer_image region; all of them lie in x 432.37..434.37, y -258.73..-255.52 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L2_GND.CU:481.05,-230'/Top/Kria/Kria_FTDI/JTAG_TDO_M2C' crosses where the outline of the GND plane on L2_GND.CU is drawn back, at 1 place, the longest 1.97 mm (fastest: /Top/Kria/Kria_FTDI/JTAG_TDO_M2C at 2 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_FTDI/JTAG_TDO_M2C' on L1_TOP.CU, the GND plane on L2_GND.CU has 1.97 mm under the track with no copper, at (481.05, -230), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. No crossing has a plane on the track's other side, so the whole return meets the break.

      • '/Top/Kria/Kria_FTDI/JTAG_TDO_M2C' on L1_TOP.CU at (481.05, -230), 1.97 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L2_GND.CU%3A481.05%2C-230 serves each with its own layer_image region; all of them lie in x 480.05..482.05, y -231.98..-228.02 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:250.88,-180.39'/Top/Kria/DAC0_CS_L' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.92 mm (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/DAC0_CS_L' on L5_MIX.CU, the GND plane on L4_POWER.CU has 2.92 mm under the track with no copper, at (250.88, -180.39), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (250.88, -180.39), 2.92 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A250.88%2C-180.39 serves each with its own layer_image region; all of them lie in x 248.42..253.34, y -181.39..-179.39 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:256.58,-207.41'/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.79 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_N at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.79 mm under the track with no copper, at (256.58, -207.41), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' on L3_SIG.CU at (256.58, -207.41), 2.79 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A256.58%2C-207.41 serves each with its own layer_image region; all of them lie in x 254.6..258.57, y -209.39..-205.42 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:257.4,-207.42'/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 3.01 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_P at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' on L3_SIG.CU, the GND plane on L4_POWER.CU has 3.01 mm under the track with no copper, at (257.4, -207.42), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' on L3_SIG.CU at (257.4, -207.42), 3.01 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A257.4%2C-207.42 serves each with its own layer_image region; all of them lie in x 255.33..259.46, y -209.48..-205.35 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:258.41,-183.4'/Top/Kria/CLK_RF0_N' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.28 mm (fastest: /Top/Kria/CLK_RF0_N at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/CLK_RF0_N' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.28 mm under the track with no copper, at (258.41, -183.4), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (258.41, -183.4), 2.28 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A258.41%2C-183.4 serves each with its own layer_image region; all of them lie in x 256.61..260.22, y -185.2..-181.59 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:259.06,-205.76'/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.15 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_P at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' on L3_SIG.CU, the GND plane on L4_POWER.CU has 1.15 mm under the track with no copper, at (259.06, -205.76), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_P' on L3_SIG.CU at (259.06, -205.76), 1.15 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A259.06%2C-205.76 serves each with its own layer_image region; all of them lie in x 257.65..260.46, y -207.16..-204.35 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:259.2,-184.3'/Top/Kria/DAC1_MOSI' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 3.09 mm (fastest: /Top/Kria/DAC1_MOSI at 2 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/DAC1_MOSI' on L5_MIX.CU, the GND plane on L4_POWER.CU has 3.09 mm under the track with no copper, at (259.2, -184.3), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/DAC1_MOSI' on L5_MIX.CU at (259.2, -184.3), 3.09 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A259.2%2C-184.3 serves each with its own layer_image region; all of them lie in x 256.66..261.74, y -185.3..-183.3 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:259.2,-184.75'/Top/Kria/DAC1_MISO' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.17 mm (fastest: /Top/Kria/DAC1_MISO at 2 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/DAC1_MISO' on L5_MIX.CU, the GND plane on L4_POWER.CU has 2.17 mm under the track with no copper, at (259.2, -184.75), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/DAC1_MISO' on L5_MIX.CU at (259.2, -184.75), 2.17 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A259.2%2C-184.75 serves each with its own layer_image region; all of them lie in x 257.12..261.29, y -185.75..-183.75 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:261.3,-180.51'/Top/Kria/CLK_RF0_N' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.27 mm (fastest: /Top/Kria/CLK_RF0_N at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/CLK_RF0_N' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.27 mm under the track with no copper, at (261.3, -180.51), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/CLK_RF0_N' on L3_SIG.CU at (261.3, -180.51), 2.27 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A261.3%2C-180.51 serves each with its own layer_image region; all of them lie in x 259.5..263.11, y -182.31..-178.7 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:263.1,-180.39'/Top/Kria/DAC0_CS_L' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 4.11 mm (fastest: /Top/Kria/DAC0_CS_L at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/DAC0_CS_L' on L5_MIX.CU, the GND plane on L4_POWER.CU has 4.11 mm under the track with no copper, at (263.1, -180.39), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/DAC0_CS_L' on L5_MIX.CU at (263.1, -180.39), 4.11 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A263.1%2C-180.39 serves each with its own layer_image region; all of them lie in x 260.05..266.16, y -181.39..-179.39 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:279.43,-179.5'/Top/Kria/DAC0_SCK' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.19 mm (fastest: /Top/Kria/DAC0_SCK at 2 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/DAC0_SCK' on L5_MIX.CU, the GND plane on L4_POWER.CU has 2.19 mm under the track with no copper, at (279.43, -179.5), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/DAC0_SCK' on L5_MIX.CU at (279.43, -179.5), 2.19 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A279.43%2C-179.5 serves each with its own layer_image region; all of them lie in x 277.56..281.62, y -180.5..-178.18 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:279.5,-179.15'/Top/Kria/TEMP_CLK' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.27 mm (fastest: /Top/Kria/TEMP_CLK at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/TEMP_CLK' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.27 mm under the track with no copper, at (279.5, -179.15), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/TEMP_CLK' on L3_SIG.CU at (279.5, -179.15), 2.27 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A279.5%2C-179.15 serves each with its own layer_image region; all of them lie in x 277.37..281.64, y -180.15..-178.15 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:279.5,-179.75'/Top/Kria/TEMP_DQ' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.34 mm (fastest: /Top/Kria/TEMP_DQ at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/TEMP_DQ' on L3_SIG.CU, the GND plane on L4_POWER.CU has 1.34 mm under the track with no copper, at (279.5, -179.75), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/TEMP_DQ' on L3_SIG.CU at (279.5, -179.75), 1.34 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A279.5%2C-179.75 serves each with its own layer_image region; all of them lie in x 277.83..281.17, y -180.75..-178.75 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:282.93,-195.17'/Top/Kria/FD1_DIV1' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.04 mm (fastest: /Top/Kria/FD1_DIV1 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/FD1_DIV1' on L5_MIX.CU, the GND plane on L4_POWER.CU has 1.04 mm under the track with no copper, at (282.93, -195.17), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/FD1_DIV1' on L5_MIX.CU at (282.93, -195.17), 1.04 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A282.93%2C-195.17 serves each with its own layer_image region; all of them lie in x 281.57..284.3, y -196.53..-193.8 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:283.69,-180.3'/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 7.35 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_N at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' on L3_SIG.CU, the GND plane on L4_POWER.CU has 7.35 mm under the track with no copper, at (283.69, -180.3), running to the end of the track. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_N' on L3_SIG.CU at (283.69, -180.3), 7.35 mm over no copper, to the end of the track; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A283.69%2C-180.3 serves each with its own layer_image region; all of them lie in x 280.47..287.76, y -183.52..-176.65 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:284.02,-194.08'/Top/Kria/FD1_DIV1' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.04 mm (fastest: /Top/Kria/FD1_DIV1 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/FD1_DIV1' on L5_MIX.CU, the GND plane on L4_POWER.CU has 1.04 mm under the track with no copper, at (284.02, -194.08), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/FD1_DIV1' on L5_MIX.CU at (284.02, -194.08), 1.04 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A284.02%2C-194.08 serves each with its own layer_image region; all of them lie in x 282.65..285.38, y -195.45..-192.72 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:289.59,-179.31'/Top/Kria/Kria_SOM240_2_connector/AUX_CONN5' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.29 mm (fastest: /Top/Kria/Kria_SOM240_2_connector/AUX_CONN5 at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SOM240_2_connector/AUX_CONN5' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.29 mm under the track with no copper, at (289.59, -179.31), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SOM240_2_connector/AUX_CONN5' on L3_SIG.CU at (289.59, -179.31), 2.29 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A289.59%2C-179.31 serves each with its own layer_image region; all of them lie in x 288.59..290.59, y -181.45..-177.17 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:342.15,-228.45'/Top/Kria/Kria_CLK_outputs/CLK_OUT2_EN_L' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.18 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT2_EN_L at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT2_EN_L' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.18 mm under the track with no copper, at (342.15, -228.45), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT2_EN_L' on L3_SIG.CU at (342.15, -228.45), 2.18 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A342.15%2C-228.45 serves each with its own layer_image region; all of them lie in x 340.38..343.92, y -230.22..-226.68 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:352.05,-176.04'/Top/Kria/Kria_SOM240_1_connector/ETH_RX_CLK' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.49 mm (fastest: /Top/Kria/Kria_SOM240_1_connector/ETH_RX_CLK at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SOM240_1_connector/ETH_RX_CLK' on L3_SIG.CU, the GND plane on L4_POWER.CU has 1.49 mm under the track with no copper, at (352.05, -176.04), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SOM240_1_connector/ETH_RX_CLK' on L3_SIG.CU at (352.05, -176.04), 1.49 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A352.05%2C-176.04 serves each with its own layer_image region; all of them lie in x 350.3..353.8, y -177.04..-175.04 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:357.56,-172.17'/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D0' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.57 mm (fastest: /Top/Kria/Kria_SOM240_1_connector/ETH_TX_D0 at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D0' on L3_SIG.CU, the GND plane on L4_POWER.CU has 0.57 mm under the track with no copper, at (357.56, -172.17), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SOM240_1_connector/ETH_TX_D0' on L3_SIG.CU at (357.56, -172.17), 0.57 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A357.56%2C-172.17 serves each with its own layer_image region; all of them lie in x 356.27..358.84, y -173.17..-171.17 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:435.71,-258.29'/Top/Kria/Kria_SFP+/SFP1_TX_FAULT' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.18 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_FAULT at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_FAULT' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.18 mm under the track with no copper, at (435.71, -258.29), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TX_FAULT' on L3_SIG.CU at (435.71, -258.29), 2.18 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A435.71%2C-258.29 serves each with its own layer_image region; all of them lie in x 434.71..436.71, y -260.38..-256.2 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:437.75,-269.35'/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.68 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU, the GND plane on L4_POWER.CU has 0.68 mm under the track with no copper, at (437.75, -269.35), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU at (437.75, -269.35), 0.68 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A437.75%2C-269.35 serves each with its own layer_image region; all of them lie in x 436.41..439.09, y -270.35..-268.35 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:438.92,-262.5'/Top/Kria/Kria_SFP+/SFP1_TX_FAULT' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 5.17 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_FAULT at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_FAULT' on L3_SIG.CU, the GND plane on L4_POWER.CU has 5.17 mm under the track with no copper, at (438.92, -262.5), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 0.7 mm). No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TX_FAULT' on L3_SIG.CU at (438.92, -262.5), 5.17 mm, running ALONG the bare copper for 0.7 mm, not across it over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A438.92%2C-262.5 serves each with its own layer_image region; all of them lie in x 435.28..442.38, y -263.5..-261.33 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:439.25,-269.35'/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.96 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU, the GND plane on L4_POWER.CU has 0.96 mm under the track with no copper, at (439.25, -269.35), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU at (439.25, -269.35), 0.96 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A439.25%2C-269.35 serves each with its own layer_image region; all of them lie in x 437.77..440.73, y -270.35..-268.35 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:440.75,-269.35'/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.96 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU, the GND plane on L4_POWER.CU has 0.96 mm under the track with no copper, at (440.75, -269.35), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU at (440.75, -269.35), 0.96 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A440.75%2C-269.35 serves each with its own layer_image region; all of them lie in x 439.27..442.23, y -270.35..-268.35 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:442.25,-269.35'/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.68 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TX_DISABLE at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU, the GND plane on L4_POWER.CU has 0.68 mm under the track with no copper, at (442.25, -269.35), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TX_DISABLE' on L3_SIG.CU at (442.25, -269.35), 0.68 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A442.25%2C-269.35 serves each with its own layer_image region; all of them lie in x 440.91..443.59, y -270.35..-268.35 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:442.8,-257.62'/Top/Kria/Kria_SFP+/SFP0_RX_LOS' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 4.31 mm (fastest: /Top/Kria/Kria_SFP+/SFP0_RX_LOS at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP0_RX_LOS' on L5_MIX.CU, the GND plane on L4_POWER.CU has 4.31 mm under the track with no copper, at (442.8, -257.62), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 1.21 mm). No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP0_RX_LOS' on L5_MIX.CU at (442.8, -257.62), 4.31 mm, running ALONG the bare copper for 1.21 mm, not across it over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A442.8%2C-257.62 serves each with its own layer_image region; all of them lie in x 441.8..444.32, y -259.94..-253.85 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:445.4,-163.9'/Top/Kria/Kria_SFP+/SFP1_RX_LOS' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.21 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_RX_LOS at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' on L5_MIX.CU, the GND plane on L4_POWER.CU has 2.21 mm under the track with no copper, at (445.4, -163.9), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' on L5_MIX.CU at (445.4, -163.9), 2.21 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A445.4%2C-163.9 serves each with its own layer_image region; all of them lie in x 443.62..447.18, y -165.68..-162.12 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:447.35,-257.12'/Top/Kria/Kria_SFP+/SFP1_RX_LOS' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.5 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_RX_LOS at 3 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' on L5_MIX.CU, the GND plane on L4_POWER.CU has 0.5 mm under the track with no copper, at (447.35, -257.12), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 44% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_RX_LOS' on L5_MIX.CU at (447.35, -257.12), 0.5 mm over no copper; other side: PL_3V3 on L6_POWER.CU, 0.12 mm from the track against 0.15 mm to L4_POWER.CU: 44% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A447.35%2C-257.12 serves each with its own layer_image region; all of them lie in x 446.35..448.35, y -258.37..-255.88 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:449.01,-233.75'/Top/Kria/Kria_SFP+/SFP1_TD_N' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.67 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TD_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_TD_N' on L3_SIG.CU, the GND plane on L4_POWER.CU has 1.67 mm under the track with no copper, at (449.01, -233.75), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TD_N' on L3_SIG.CU at (449.01, -233.75), 1.67 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A449.01%2C-233.75 serves each with its own layer_image region; all of them lie in x 448.01..450.01, y -235.59..-231.92 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:449.29,-233.75'/Top/Kria/Kria_SFP+/SFP1_TD_P' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.67 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_TD_P at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_TD_P' on L3_SIG.CU, the GND plane on L4_POWER.CU has 1.67 mm under the track with no copper, at (449.29, -233.75), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_TD_P' on L3_SIG.CU at (449.29, -233.75), 1.67 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A449.29%2C-233.75 serves each with its own layer_image region; all of them lie in x 448.29..450.29, y -235.58..-231.91 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:455.38,-247'/Top/Kria/Kria_SFP+/SFP1_RD_P' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 1.46 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_RD_P at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_RD_P' on L3_SIG.CU, the GND plane on L4_POWER.CU has 1.46 mm under the track with no copper, at (455.38, -247), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_RD_P' on L3_SIG.CU at (455.38, -247), 1.46 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A455.38%2C-247 serves each with its own layer_image region; all of them lie in x 454.38..456.38, y -248.73..-245.27 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:GND@L4_POWER.CU:455.77,-246.94'/Top/Kria/Kria_SFP+/SFP1_RD_N' crosses where the outline of the GND plane on L4_POWER.CU is drawn back, at 1 place, the longest 2.63 mm (fastest: /Top/Kria/Kria_SFP+/SFP1_RD_N at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_SFP+/SFP1_RD_N' on L3_SIG.CU, the GND plane on L4_POWER.CU has 2.63 mm under the track with no copper, at (455.77, -246.94), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 101..499, y -299.41..-101 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SFP+/SFP1_RD_N' on L3_SIG.CU at (455.77, -246.94), 2.63 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3AGND%40L4_POWER.CU%3A455.77%2C-246.94 serves each with its own layer_image region; all of them lie in x 454.77..456.77, y -249.26..-244.63 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:PL_1V8@L4_POWER.CU:310.19,-161.5'/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_2/VCCOEN_PL_M2C' crosses where the outline of the PL_1V8 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.1 mm (fastest: /Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_2/VCCOEN_PL_M2C at 100 ps 10-90 %, ASSUMED)

      Along '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_2/VCCOEN_PL_M2C' on L3_SIG.CU, the PL_1V8 plane on L4_POWER.CU has 0.1 mm under the track with no copper, at (310.19, -161.5), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 275..355.5, y -175..-127.5 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_Power_Supply_Units/Kria_PSU_-_2/VCCOEN_PL_M2C' on L3_SIG.CU at (310.19, -161.5), 0.1 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3APL_1V8%40L4_POWER.CU%3A310.19%2C-161.5 serves each with its own layer_image region; all of them lie in x 309.19..311.19, y -162.55..-160.45 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:PL_1V8@L4_POWER.CU:347.34,-159.42'/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' crosses where the outline of the PL_1V8 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.11 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU, the PL_1V8 plane on L4_POWER.CU has 0.11 mm under the track with no copper, at (347.34, -159.42), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 275..355.5, y -175..-127.5 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU at (347.34, -159.42), 0.11 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3APL_1V8%40L4_POWER.CU%3A347.34%2C-159.42 serves each with its own layer_image region; all of them lie in x 346.29..348.38, y -160.44..-158.39 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:PL_1V8@L4_POWER.CU:349.55,-173.59'/Top/Kria/Kria_ethernet/ETH_RX_D2' crosses where the outline of the PL_1V8 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.13 mm (fastest: /Top/Kria/Kria_ethernet/ETH_RX_D2 at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_ethernet/ETH_RX_D2' on L3_SIG.CU, the PL_1V8 plane on L4_POWER.CU has 0.13 mm under the track with no copper, at (349.55, -173.59), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 275..355.5, y -175..-127.5 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_ethernet/ETH_RX_D2' on L3_SIG.CU at (349.55, -173.59), 0.13 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3APL_1V8%40L4_POWER.CU%3A349.55%2C-173.59 serves each with its own layer_image region; all of them lie in x 348.49..350.62, y -174.59..-172.59 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:PL_1V8@L4_POWER.CU:349.67,-159.39'/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' crosses where the outline of the PL_1V8 plane on L4_POWER.CU is drawn back, at 1 place, the longest 0.3 mm (fastest: /Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L at 1 ns 10-90 %, suggested by its name)

      Along '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU, the PL_1V8 plane on L4_POWER.CU has 0.3 mm under the track with no copper, at (349.67, -159.39), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. No hole cleared it: the outline of the pour at x 275..355.5, y -175..-127.5 is drawn back there. One decision covers the void: whether these tracks may cross it, or the plane should be filled under them. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_CLK_outputs/CLK_OUT0_FAULT_L' on L3_SIG.CU at (349.67, -159.39), 0.3 mm over no copper; other side: GND on L2_GND.CU, 0.15 mm from the track against 0.12 mm to L4_POWER.CU: 56% of the return on L4_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3APL_1V8%40L4_POWER.CU%3A349.67%2C-159.39 serves each with its own layer_image region; all of them lie in x 348.52..350.82, y -160.39..-158.39 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

    • void:routed@L7_POWER.CU|L8_SIG.CU'/Top/Kria/Kria_SOM240_1_connector/I2C_SCK' on L8_SIG.CU crosses the tracks routed through the T_+12V pour on L7_POWER.CU at 1 place, the longest 3.8 mm (fastest: /Top/Kria/Kria_SOM240_1_connector/I2C_SCK at 10 ns 10-90 %, suggested by its name)

      The T_+12V pour on L7_POWER.CU is cleared round tracks routed through it, and tracks on L8_SIG.CU, which take that pour as their reference, cross them: for the width of each cut the return has no plane under it and detours round the cut. 1 place(s) on 1 net(s). One decision covers the layer pair: whether these layers may take a pour their own tracks cut as the reference, or the routing or the stackup should give the tracks a plane that is not cut where they cross. Every crossing has a plane on the track's other side at most twice as far away as the broken one, so at most 56% of the return meets the break and the rest follows the track on that plane (to first order each plane beside a stripline carries a share of the return inversely proportional to its distance from the track). The finding is the lower tier (info).

      • '/Top/Kria/Kria_SOM240_1_connector/I2C_SCK' on L8_SIG.CU at (362.88, -163.8), 3.8 mm over copper of UTIL_3V3, not a plane, to the end of the track; other side: 5V0_IN on L9_POWER.CU, 0.15 mm from the track against 0.12 mm to L7_POWER.CU: 56% of the return on L7_POWER.CU

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/82/reference-planes?group=void%3Arouted%40L7_POWER.CU%7CL8_SIG.CU serves each with its own layer_image region; all of them lie in x 359..365, y -164.8..-162.8 mm (board frame, Y up).

      Read from the copper as drawn on the reference layer, along each track's centreline at exact breakpoints. The reference is the copper beside the track, never the layer's type or name. Whether it matters depends on each signal's edge rate, which nothing on file states for these nets: a slow static line crosses a split harmlessly, a fast edge does not.

Main parts

Part numberMakerDescriptionWhere
531BC125M000DGX3
531BC156M250DGRX2
93LC56B-I/SNU20
ABM8G-12.000MHZ-8-B2X-TY1
DP83867CRRGZRU18
DS1620S+T&RU52
ECS-2016MV-250-CN-TR3X1
FDN336PQ14
FT4232HLU24
GAT-5+U64
INA260AIPWU6
KMR231NGLFSSW1
LMX2541SQ2690EU61
M24C64-DRDW8U16
MAX9152EUEU29
MMSD4148T1GD1
MP8772GQ-ZU4
MPM3810AGQBE-AEC1-ZU3
NDS331NQ3
PCA9306DCTRU17
PMA2-43LN+U62
RJ45_BEL_L829-1J1T-43J4
SD-Card_5025700893J5
SN74AVC4T245RSVRU23
SN74HCS09PWRU15
TLV840MADL30DBVRU12
TPD4S012DRYRU21
TPS3430WDRCRU14
TSM-103-01-S-SVJ3
TXS02612RTWU19
USB_B_Micro_Molex_473460001J6
TPS7A3301U59, U60
LT3040EDDU31, U35, U39
MC100EPT21DR2GU32, U36, U40
MPM3840GQV-ZU10, U11, U7
NB4L52MNGU30, U34, U38
NC7WZ07P6XU28, U8, U9
74LVC1G17DBVRU42, U43, U44, U46
ADL5375-05ACPZ-R7U117, U132, U81, U99
KP-3216SGCD14, D15, D16, D17

About this design

Design
MACBETH — OHWR0082 R1
Designed in
KiCad EDA
Board
12 copper layers, 400 × 200 mm, 1559 components, 1212 nets
Licence
CC-BY-SA-4.0, CERN-OHL-S-2.0, GPL-3.0
Source
https://gitlab.com/ohwr/project/macbeth at commit 5e339d8fde71
Attribution
CERN Open Hardware Repository, project ohwr/project/macbeth (commit 5e339d8fde71c07dd9b5ebac33dd2321c47bc0c0, 2025-07-25). Licence: CC-BY-SA-4.0, CERN-OHL-S-2.0, GPL-3.0. Source: https://gitlab.com/ohwr/project/macbeth. This is a snapshot of the copy published on the OHWR GitLab pages (R1), not CERN's released record. CERN's source of truth is EDMS (https://edms.cern.ch), and the current release may be newer.
Reviewed with
Probe by E# (E-Sharp AB)