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

FMC DELAY 1NS 8CHA

OHWR0006 · revision R1

FMC DELAY 1NS 8CHA (OHWR0006 R1) is an open-hardware circuit board from the CERN Open Hardware Repository, designed in Altium Designer: a 6-layer board of 81.6 × 69 mm with 354 components and 279 nets, published under CC-BY-SA-4.0, CC0-1.0, CERN-OHL-W-2.0, GPL-2.0-or-later, LGPL-2.1, LGPL-2.1-or-later.

review 68% walked · 13 findings: 1 blocker · 8 major · 4 minor (47 items)

Open in the interactive viewerRead the full review

FMC DELAY 1NS 8CHA (OHWR0006 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 13 findings: 1 blocker · 8 major · 4 minor (47 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. NetCopperSplitblockerOpen

    GND's copper is in 9 pieces: R65C.2 and 7 other piece(s) are not connected to the rest of the net

    GND's copper is in 9 pieces: R65C.2 and 7 other piece(s) are not connected to the rest of the net. Piece 1: B1.1, B2.1, B3.1, B4.1, C1.2, C10.2, C13.2, C14.2 and 345 more pin(s), on TOP_LAYER, L4_DIFF and BOTTOM_LAYER, 376 drill(s), 1141.19 mm2 of pour, over (68.08, 67.79) to (149.35, 136.8). Piece 2: R65C.2, on BOTTOM_LAYER, over (68.7, 112.66) to (70.78, 116.37). Piece 3: R65D.2, on BOTTOM_LAYER, over (68.83, 122) to (70.78, 125.9). Piece 4: R107.2, on BOTTOM_LAYER, over (69.18, 85.96) to (72.17, 87.8). Piece 5: D4A.1, on BOTTOM_LAYER, over (73.09, 95.23) to (75.86, 97.32). Piece 6: D4B.1, on BOTTOM_LAYER, over (73.09, 105.25) to (75.86, 106.85). Piece 7: D4D.1, on BOTTOM_LAYER, over (73.19, 123.7) to (75.86, 125.9). Piece 8: R74.2, on BOTTOM_LAYER, over (74.01, 90.64) to (75.86, 93.77). Piece 9: C49A.1, on BOTTOM_LAYER, over (74.26, 100.17) to (79.19, 102.44). The export files every one of these pins under the net, so every connectivity check passes; the copper the fab builds leaves them apart. Usually a pour island that touches only one via of its own net, or a track ending short of its copper: join the piece to the net and export again.

    At split:GND

  2. PartLifecyclemajorOpen

    BAR66 on 5 placements (D4A, D4B, D4C, D4D, D6): the distributor catalogue reports "Obsolete" for Infineon Technologies's BAR66

    BAR66 on 5 placements (D4A, D4B, D4C, D4D, D6): the distributor catalogue reports "Obsolete" for Infineon Technologies's BAR66 — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. That is the maker this board's BOM names (INFINEON). 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/BAR66/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.

    At D4A

  3. DecapViaFanoutmajorOpen

    C77 on 'P3V3': its rail pad reaches its via over 2.70 mm of track, adding about 0.86 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads

    C77 on 'P3V3': its rail pad reaches its via over 2.70 mm of track, adding about 0.86 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads. C77.1 (rail) runs 2.70 mm along its copper to its nearest via, 1.70 mm past the 1 mm beside the pad: 0.86 nH; C77.2 (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 1.60 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.86 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 1.38 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 1.28 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    At C77

  4. RegulatorCapNotAtPinmajorOpen

    IC33 input pin 1 on 'NetC30_1': the nearest C_IN (C30, 3.6 mm) is on the other side of the board, and none on the regulator's side is within the datasheet's 10 mm

    IC33 input pin 1 on 'NetC30_1': the nearest C_IN (C30, 3.6 mm) is on the other side of the board, and none on the regulator's side is within the datasheet's 10 mm. IC33.1: C30 is 3.6 mm from the pin, within the distance, but on the top with the regulator on the bottom: its loop runs through the board's thickness and a via pair, and none sits on the regulator's side; its datasheet asks for C_IN within 10 mm (LP5951MF-3.0/NOPB datasheet p13, layout_max_cin_distance). Place C_IN on the regulator's side within the stated distance, its vias at its pads; the distance is the pads' centres in the board plane, and the routed copper between them can only be longer.

    At IC33.1

  5. DecapOrdermajorOpen

    IC4 supply pins 38, 57 on 'P3V3_PLL': the smallest capacitor C10 sits 4.6 mm away, behind C56 at 0.8 mm

    IC4 supply pins 38, 57 on 'P3V3_PLL': the smallest capacitor C10 sits 4.6 mm away, behind C56 at 0.8 mm. IC4.38: the smallest value, C10 (10 nF), is 4.6 mm from the pin and C56 (100 nF) 0.8 mm: the high-frequency part sits behind more inductance than the bulk part, which needs it least. IC4.57: the smallest value, C50 (10 nF), is 3.6 mm from the pin and C53 (100 nF) 0.5 mm: the high-frequency part sits behind more inductance than the bulk part, which needs it least. Swap them, or bring the smallest value to the pin: it serves the fastest current and is the one the loop inductance defeats. Distances are pad centre to pin centre in the board plane. The 5 mm radius, the 10x value step and the 1 mm / 1.5x tie tolerance are Probe's starting values.

    At IC4

  6. PartLifecyclemajorOpen

    PartLifecycle: 5 part numbers×5

    4 Obsolete, 1 Discontinued at DigiKey · on 29 placements

    At C13, IC11, IC21, IC33, SW1A

    Each finding, with its own headline
    • C1316TQC10M on 20 placements (C13, C17, C20, C21, C22, C23, C27, C34, …): the distributor catalogue reports "Obsolete" for Panasonic Industry's 16TQC10M

      16TQC10M on 20 placements (C13, C17, C20, C21, C22, C23, C27, C34, …): the distributor catalogue reports "Obsolete" for Panasonic Industry's 16TQC10M — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names SANYO, and that is one line with Panasonic Industry: Panasonic made SANYO Electric a wholly owned subsidiary on 2011-04-01; SANYO's capacitor lines (OS-CON, POSCAP) are sold under the Panasonic name. 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/16TQC10M/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 (2023-12-31, 1,050 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.

    • IC11LT3503EDCB#PBF on IC11: the distributor catalogue reports "Discontinued at DigiKey" for Analog Devices Inc.'s LT3503EDCB#PBF

      LT3503EDCB#PBF on IC11: the distributor catalogue reports "Discontinued at DigiKey" for Analog Devices Inc.'s LT3503EDCB#PBF — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names LINEAR TECHNOLOGY, and that is one line with Analog Devices Inc.: Analog Devices completed its acquisitions of Hittite Microwave on 2014-07-22, Linear Technology on 2017-03-10 and Maxim Integrated on 2021-08-26. 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/LT3503EDCB#PBF/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.

    • IC21AP1117YG-13 on 3 placements (IC21, IC6, IC9): the distributor catalogue reports "Obsolete" for Diodes Incorporated's AP1117YG-13

      AP1117YG-13 on 3 placements (IC21, IC6, IC9): the distributor catalogue reports "Obsolete" for Diodes Incorporated's AP1117YG-13 — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. That is the maker this board's BOM names (DIODES INCORPORATED). 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/AP1117YG-13/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 (2015-07-27, 4,129 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.

    • IC33LP5951MF-3.0/NOPB on IC33: the distributor catalogue reports "Obsolete" for National Semiconductor's LP5951MF-3.0/NOPB

      LP5951MF-3.0/NOPB on IC33: the distributor catalogue reports "Obsolete" for National Semiconductor's LP5951MF-3.0/NOPB — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names TEXAS INSTRUMENTS, and that is one line with National Semiconductor: Texas Instruments completed its acquisition of National Semiconductor on 2011-09-23. 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/LP5951MF-3.0/NOPB/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.

    • SW1AAQY221N3M on 4 placements (SW1A, SW1B, SW1C, SW1D): the distributor catalogue reports "Obsolete" for Panasonic Industry's AQY221N3M

      AQY221N3M on 4 placements (SW1A, SW1B, SW1C, SW1D): the distributor catalogue reports "Obsolete" for Panasonic Industry's AQY221N3M — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. That is the maker this board's BOM names (PANASONIC ELECTRIC WORKS). 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/AQY221N3M/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.

  7. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 19 boards×19

    Checked by the review, stands · Claude walker W3 (Cern session) — Checked: the reference break is read from the copper as stated. Whether it matters depends on the signal's edge rate and the receiver's margin, which no datasheet or app note on file fixes for this net; the impedance step and reflection figures are Probe's model, and the thresholds are its starting values. Layout practice, not a datasheet requirement. Left for the owner.

    At refchange:BYPASS|GND, refchange:GND|M2V, refchange:GND|P3V3, refchange:GND|P3V3_PLL, refchange:GND|P8V, refchange:M2V|P3V3 and 13 more

    Each finding, with its own headline
    • refchange:BYPASS|GND'PLL_STATUS' changes reference between BYPASS and GND at a via, none with a reference that continues across (fastest: PLL_STATUS at 100 ps 10-90 %, ASSUMED)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of BYPASS 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: 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 BYPASS 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. None is partial: at every one, no reference continues across the via.

      • 'PLL_STATUS' at (112.65, 104.83), TOP_LAYER -> BOTTOM_LAYER: BYPASS on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 23.6 mm of track past the via on BOTTOM_LAYER: 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. BYPASS 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/9/reference-planes?group=refchange%3ABYPASS%7CGND serves each with its own layer_image region; all of them lie in x 110.65..114.65, y 102.83..106.83 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|M2V9 nets change reference between GND and M2V at 9 vias, none with a reference that continues across (fastest: NetIC1A_6 at 100 ps 10-90 %, ASSUMED)

      At 9 via(s), on 9 net(s), the signal leaves a layer referenced by one of GND and M2V 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: 9 with none. 9 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 M2V 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.

      • 'NetIC1B_7' at (90.32, 100.15), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 19.6 mm of track past the via on L4_DIFF: 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 M2V 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. Measured for reference: C54A (10 uF 1206) 1.7 mm away: 1.12 nH in the 1.080 mm cavity + 0.80 nH ESL + 0.97 nH mounting (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 1.28 mm depth, ASSUMED): 90.9 ohm at the knee, 182 % of 50 ohm (assumed track impedance). None.
      • 'NetIC1A_7' at (90.22, 115.98), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 13.6 mm of track past the via on L4_DIFF: 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 M2V 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. Measured for reference: C54C (10 uF 1206) 1.9 mm away: 1.16 nH in the 1.080 mm cavity + 0.80 nH ESL + 0.97 nH mounting (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 1.28 mm depth, ASSUMED): 92.1 ohm at the knee, 184 % of 50 ohm (assumed track impedance). None.
      • 'NetIC1B_6' at (90.37, 108), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 10.8 mm of track past the via on L4_DIFF: 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 M2V 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. Measured for reference: C54B (10 uF 1206) 1.8 mm away: 1.15 nH in the 1.080 mm cavity + 0.80 nH ESL + 0.97 nH mounting (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 1.28 mm depth, ASSUMED): 91.7 ohm at the knee, 183 % of 50 ohm (assumed track impedance). None.
      • 'NetIC1A_6' at (90.26, 124.02), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 8.3 mm of track past the via on L4_DIFF: 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 M2V 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. Measured for reference: C54D (10 uF 1206) 1.8 mm away: 1.15 nH in the 1.080 mm cavity + 0.80 nH ESL + 0.97 nH mounting (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 1.28 mm depth, ASSUMED): 91.8 ohm at the knee, 184 % of 50 ohm (assumed track impedance). None.
      • 'TDC_CAL_STOP4' at (91.41, 96.65), TOP_LAYER -> BOTTOM_LAYER: M2V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 3.1 mm of track past the via on BOTTOM_LAYER: 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 M2V 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.
      • 'NetIC25D_6' at (82.55, 127.05), TOP_LAYER -> BOTTOM_LAYER: M2V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 1.1 mm of track past the via on BOTTOM_LAYER: 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 M2V 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.
      • 'NetIC25C_6' at (82.55, 119.08), TOP_LAYER -> BOTTOM_LAYER: M2V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 1.0 mm of track past the via on BOTTOM_LAYER: 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 M2V 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.
      • 'NetIC25A_6' at (82.58, 103.2), TOP_LAYER -> BOTTOM_LAYER: M2V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 1.0 mm of track past the via on BOTTOM_LAYER: 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 M2V 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.
      • 'NetIC25B_6' at (82.58, 111.2), TOP_LAYER -> BOTTOM_LAYER: M2V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 1.0 mm of track past the via on BOTTOM_LAYER: 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 M2V 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/9/reference-planes?group=refchange%3AGND%7CM2V serves each with its own layer_image region; all of them lie in x 80.55..93.41, y 94.65..129.05 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|P3V391 nets change reference between GND and P3V3 at 147 vias; 90 of them partial (fastest: CAL_PULSE_BUF at 100 ps 10-90 %, ASSUMED)

      At 147 via(s), on 91 net(s), the signal leaves a layer referenced by one of GND and P3V3 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: 147 with none. 147 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 P3V3 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. 90 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.

      • 'NetIC8_5' at (135.15, 108), TOP_LAYER -> BOTTOM_LAYER: P3V3 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 88.4 mm of track past the via on BOTTOM_LAYER: 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 P3V3 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.
      • 'NetIC8_4' at (135.89, 106.07), TOP_LAYER -> BOTTOM_LAYER: P3V3 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 88.0 mm of track past the via on BOTTOM_LAYER: 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 P3V3 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.
      • 'TDC_RDN' at (136.68, 102.29), TOP_LAYER -> BOTTOM_LAYER: P3V3 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 81.2 mm of track past the via on BOTTOM_LAYER: 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 P3V3 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.
      • 'TDC_WRN' at (139.45, 101.04), TOP_LAYER -> BOTTOM_LAYER: P3V3 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 75.9 mm of track past the via on BOTTOM_LAYER: 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 P3V3 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.
      • 'DMTD_FB_IN' at (80.72, 90.65), TOP_LAYER -> L4_DIFF: GND on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, GND continues through a via of GND 1.43 mm away; stitching none, 66.3 mm of track past the via on L4_DIFF: 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 P3V3 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.
      • 'DMTD_FB_IN' at (134.14, 85.78), TOP_LAYER -> L4_DIFF: GND on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, GND continues through a via of GND 1.8 mm away; stitching none, 66.3 mm of track past the via on L4_DIFF: 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 P3V3 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. Measured for reference: C40 (100 nF 0603) 1.4 mm away: 0.35 nH in the 0.360 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 54.3 ohm at the knee, 109 % of 50 ohm (assumed track impedance). None.
      • 'NetJ6_h19' at (130.96, 83.36), TOP_LAYER -> L4_DIFF: GND on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, GND continues through a via of GND 1.9 mm away; stitching none, 62.9 mm of track past the via on L4_DIFF: 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 P3V3 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.
      • 'TRIG_TO_FPGA' at (83.19, 92.08), TOP_LAYER -> L4_DIFF: GND on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, GND continues through a via of GND 1.99 mm away; stitching none, 62.8 mm of track past the via on L4_DIFF: 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 P3V3 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.
      • 'TRIG_TO_FPGA' at (134.19, 84.53), TOP_LAYER -> L4_DIFF: GND on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, GND continues through a via of GND 1.85 mm away; stitching none, 62.8 mm of track past the via on L4_DIFF: 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 P3V3 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. Measured for reference: C40 (100 nF 0603) 1.9 mm away: 0.39 nH in the 0.360 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 55.5 ohm at the knee, 111 % of 50 ohm (assumed track impedance). None.
      • 'VADJ' at (135.38, 126.44), TOP_LAYER -> BOTTOM_LAYER: P3V3 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 60.4 mm of track past the via on BOTTOM_LAYER: 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 P3V3 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. Measured for reference: C85 (100 nF 0603) 2.4 mm away: 1.28 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 83.4 ohm at the knee, 167 % of 50 ohm (assumed track impedance). None.
      • 'OUT_EN1' at (96.09, 105.87), TOP_LAYER -> L4_DIFF: P3V3 on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, P3V3 continues through a via of P3V3 5.88 mm away; stitching none, 58.4 mm of track past the via on L4_DIFF: 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 P3V3 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.
      • 'NetIC13_1' at (130.96, 102.92), TOP_LAYER -> L4_DIFF: P3V3 on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; partial, P3V3 continues through a via of P3V3 3.46 mm away; stitching none, 55.3 mm of track past the via on L4_DIFF: 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 P3V3 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 135 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/9/reference-planes?group=refchange%3AGND%7CP3V3 serves each with its own layer_image region; all of them lie in x 78.72..145.31, y 76.99..136.37 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|P3V3_PLL21 nets change reference between GND and P3V3_PLL at 21 vias, none with a reference that continues across (fastest: FPGA_TDC_START_N at 100 ps 10-90 %, ASSUMED)

      At 21 via(s), on 21 net(s), the signal leaves a layer referenced by one of GND and P3V3_PLL 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: 21 with none. 21 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 P3V3_PLL 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.

      • 'RESETN' at (115.34, 96.29), TOP_LAYER -> BOTTOM_LAYER: P3V3_PLL on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 50.3 mm of track past the via on BOTTOM_LAYER: 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 P3V3_PLL 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.
      • 'TDC_START_P' at (118.17, 97.99), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 48.3 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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.
      • 'TDC_START_N' at (118.97, 97.99), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 47.6 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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.
      • 'SPI_SCK_3V3' at (112.55, 100.1), TOP_LAYER -> BOTTOM_LAYER: P3V3_PLL on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 46.4 mm of track past the via on BOTTOM_LAYER: 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 P3V3_PLL 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. Measured for reference: C39 (10 nF 0603) 2.6 mm away: 1.30 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 84.0 ohm at the knee, 168 % of 50 ohm (assumed track impedance). None.
      • 'SPI_MOSI_3V3' at (112.98, 96.39), TOP_LAYER -> BOTTOM_LAYER: P3V3_PLL on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 34.8 mm of track past the via on BOTTOM_LAYER: 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 P3V3_PLL 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.
      • 'CLK125_P' at (123.52, 101.83), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 30.2 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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. Measured for reference: C56 (100 nF 0603) 2.5 mm away: 1.29 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 84.0 ohm at the knee, 168 % of 50 ohm (assumed track impedance). None.
      • 'CLK125_N' at (123.52, 100.97), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 29.0 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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.
      • 'SPI_PLL_CSN' at (114.99, 98.45), TOP_LAYER -> BOTTOM_LAYER: P3V3_PLL on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 28.6 mm of track past the via on BOTTOM_LAYER: 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 P3V3_PLL 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. Measured for reference: C39 (10 nF 0603) 2.2 mm away: 1.23 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 82.1 ohm at the knee, 164 % of 50 ohm (assumed track impedance). None.
      • 'CLK2A_N' at (118.15, 107.52), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 26.0 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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. Measured for reference: C53 (100 nF 0603) 2.0 mm away: 1.18 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 80.4 ohm at the knee, 161 % of 50 ohm (assumed track impedance). None.
      • 'CLK2A_P' at (118.96, 107.52), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 25.3 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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. Measured for reference: C53 (100 nF 0603) 2.1 mm away: 1.21 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 81.3 ohm at the knee, 163 % of 50 ohm (assumed track impedance). None.
      • 'NetIC4_46' at (123.65, 106.22), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.6 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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. Measured for reference: C56 (100 nF 0603) 2.3 mm away: 1.25 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 82.6 ohm at the knee, 165 % of 50 ohm (assumed track impedance). None.
      • 'CLK1B_N' at (124.26, 102.87), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.3 mm of track past the via on L4_DIFF: 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 P3V3_PLL 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. Measured for reference: C56 (100 nF 0603) 2.1 mm away: 1.22 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 81.6 ohm at the knee, 163 % of 50 ohm (assumed track impedance). None.
      • and 9 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/9/reference-planes?group=refchange%3AGND%7CP3V3_PLL serves each with its own layer_image region; all of them lie in x 110.55..126.28, y 92.16..113.71 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|P8V14 nets change reference between GND and P8V at 14 vias, none with a reference that continues across (fastest: NetC42_1 at 100 ps 10-90 %, ASSUMED)

      At 14 via(s), on 14 net(s), the signal leaves a layer referenced by one of GND and P8V 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: 14 with none. 14 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 P8V 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.

      • 'NetJ6_h19' at (82.6, 72.42), TOP_LAYER -> L4_DIFF: P8V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 62.9 mm of track past the via on L4_DIFF: 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 P8V 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.
      • 'TERM_ENABLE' at (84.33, 72.39), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 48.5 mm of track past the via on BOTTOM_LAYER: 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 P8V 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.
      • 'NetR12_1' at (82.02, 75.34), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 9.1 mm of track past the via on BOTTOM_LAYER: 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 P8V 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.
      • 'NetC42_1' at (118.82, 86.18), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 5.3 mm of track past the via on BOTTOM_LAYER: 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 P8V 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. Measured for reference: C21 (10 uF 1206) 2.6 mm away: 1.30 nH in the 1.080 mm cavity + 0.80 nH ESL + 0.97 nH mounting (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 1.28 mm depth, ASSUMED): 96.4 ohm at the knee, 193 % of 50 ohm (assumed track impedance). None.
      • 'NetIC6_1' at (91.87, 72.75), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 4.0 mm of track past the via on BOTTOM_LAYER: 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 P8V 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.
      • 'NetR130_1' at (82.07, 76.99), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 3.8 mm of track past the via on BOTTOM_LAYER: 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 P8V 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.
      • 'NetC49C_2' at (79.67, 115.31), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 3.6 mm of track past the via on BOTTOM_LAYER: 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 P8V 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. Measured for reference: C47C (0.68 nF 0603) 2.2 mm away: 1.23 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (pad 0.10 nH + no via of the second net within 1 mm of its pad: the study's 0.6 nH per mm of 1.28 mm depth, ASSUMED): 82.0 ohm at the knee, 164 % of 50 ohm (assumed track impedance). None.
      • 'NetC49A_2' at (79.67, 99.43), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 3.6 mm of track past the via on BOTTOM_LAYER: 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 P8V 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. Measured for reference: C47A (0.68 nF 0603) 2.2 mm away: 1.23 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 81.9 ohm at the knee, 164 % of 50 ohm (assumed track impedance). None.
      • 'NetC49B_2' at (79.67, 107.41), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 3.6 mm of track past the via on BOTTOM_LAYER: 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 P8V 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. Measured for reference: C47B (0.68 nF 0603) 2.2 mm away: 1.23 nH in the 1.080 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 81.9 ohm at the knee, 164 % of 50 ohm (assumed track impedance). None.
      • 'NetC49D_2' at (79.58, 124.99), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 2.9 mm of track past the via on BOTTOM_LAYER: 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 P8V 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.
      • 'NetD4A_3' at (74.37, 98.81), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 2.2 mm of track past the via on BOTTOM_LAYER: 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 P8V 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.
      • 'NetD4C_3' at (73.79, 118.52), TOP_LAYER -> BOTTOM_LAYER: P8V on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 1.2 mm of track past the via on BOTTOM_LAYER: 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 P8V 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 2 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/9/reference-planes?group=refchange%3AGND%7CP8V serves each with its own layer_image region; all of them lie in x 71.66..120.82, y 70.39..128.52 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:M2V|P3V34 nets change reference between M2V and P3V3 at 4 vias, none with a reference that continues across (fastest: NetIC1A_6 at 100 ps 10-90 %, ASSUMED)

      At 4 via(s), on 4 net(s), the signal leaves a layer referenced by one of M2V and P3V3 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 M2V and P3V3 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.

      • 'NetIC1B_7' at (90.32, 100.15), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 19.6 mm of track past the via on L4_DIFF: 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. M2V and P3V3 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. Measured for reference: C68A (100 nF 0603) 2.2 mm away: 0.00 nH in the 0.000 mm cavity + 0.51 nH ESL + 0.36 nH mounting (pad 0.10 nH (vias beside the pads) + 1 via pair(s) at 3.38 mm pitch, 0.25 mm drill, 0.20 mm to the farther plane), then C81A (100 nF 0603) 2.2 mm away: 0.41 nH in the 0.360 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 83.6 ohm at the knee, 167 % of 50 ohm (assumed track impedance). None.
      • 'NetIC1A_7' at (90.22, 115.98), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 13.6 mm of track past the via on L4_DIFF: 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. M2V and P3V3 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. Measured for reference: C68C (100 nF 0603) 2.3 mm away: 0.00 nH in the 0.000 mm cavity + 0.51 nH ESL + 0.36 nH mounting (pad 0.10 nH (vias beside the pads) + 1 via pair(s) at 3.38 mm pitch, 0.25 mm drill, 0.20 mm to the farther plane), then C81C (100 nF 0603) 2.1 mm away: 0.40 nH in the 0.360 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 83.3 ohm at the knee, 167 % of 50 ohm (assumed track impedance). None.
      • 'NetIC1B_6' at (90.37, 108), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 10.8 mm of track past the via on L4_DIFF: 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. M2V and P3V3 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. Measured for reference: C68B (100 nF 0603) 2.2 mm away: 0.00 nH in the 0.000 mm cavity + 0.51 nH ESL + 0.36 nH mounting (pad 0.10 nH (vias beside the pads) + 1 via pair(s) at 3.38 mm pitch, 0.25 mm drill, 0.20 mm to the farther plane), then C81B (100 nF 0603) 2.2 mm away: 0.41 nH in the 0.360 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 83.7 ohm at the knee, 167 % of 50 ohm (assumed track impedance). None.
      • 'NetIC1A_6' at (90.26, 124.02), TOP_LAYER -> L4_DIFF: M2V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 8.3 mm of track past the via on L4_DIFF: 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. M2V and P3V3 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. Measured for reference: C68D (100 nF 0603) 2.2 mm away: 0.00 nH in the 0.000 mm cavity + 0.51 nH ESL + 0.36 nH mounting (pad 0.10 nH (vias beside the pads) + 1 via pair(s) at 3.38 mm pitch, 0.25 mm drill, 0.20 mm to the farther plane), then C81D (100 nF 0603) 2.1 mm away: 0.40 nH in the 0.360 mm cavity + 0.51 nH ESL + 0.87 nH mounting (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 1.28 mm depth, ASSUMED): 83.4 ohm at the knee, 167 % of 50 ohm (assumed track impedance). None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/9/reference-planes?group=refchange%3AM2V%7CP3V3 serves each with its own layer_image region; all of them lie in x 88.22..92.37, y 98.15..126.02 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:P3V3|P3V3_PLL15 nets change reference between P3V3 and P3V3_PLL at 15 vias, none with a reference that continues across (fastest: FPGA_TDC_START_N at 100 ps 10-90 %, ASSUMED)

      At 15 via(s), on 15 net(s), the signal leaves a layer referenced by one of P3V3 and P3V3_PLL 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: 15 with none. 15 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 P3V3 and P3V3_PLL 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.

      • 'TDC_START_P' at (118.17, 97.99), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 48.3 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'TDC_START_N' at (118.97, 97.99), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 47.6 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK125_P' at (123.52, 101.83), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 30.2 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK125_N' at (123.52, 100.97), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 29.0 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK2A_N' at (118.15, 107.52), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 26.0 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK2A_P' at (118.96, 107.52), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 25.3 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'NetIC4_46' at (123.65, 106.22), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.6 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK1B_N' at (124.26, 102.87), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.3 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK1A_N' at (119.63, 110.29), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.3 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK2B_N' at (123.62, 105.31), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 19.8 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK2B_P' at (123.62, 104.44), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 19.8 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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.
      • 'CLK1A_P' at (120.45, 110.24), TOP_LAYER -> L4_DIFF: P3V3_PLL on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 19.6 mm of track past the via on L4_DIFF: 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. P3V3 and P3V3_PLL 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 3 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/9/reference-planes?group=refchange%3AP3V3%7CP3V3_PLL serves each with its own layer_image region; all of them lie in x 116.15..126.28, y 92.16..112.29 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:P3V3|P8V'NetJ6_h19' changes reference between P3V3 and P8V at a via, none with a reference that continues across (fastest: NetJ6_h19 at 100 ps 10-90 %, ASSUMED)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of P3V3 and P8V 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 P3V3 and P8V 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.

      • 'NetJ6_h19' at (82.6, 72.42), TOP_LAYER -> L4_DIFF: P8V on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 62.9 mm of track past the via on L4_DIFF: 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. P3V3 and P8V 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/9/reference-planes?group=refchange%3AP3V3%7CP8V serves each with its own layer_image region; all of them lie in x 80.6..84.6, y 70.42..74.42 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@PLANE_L2_PWR:GND|P8V'NetJ4_1' crosses the split between the GND and P8V planes on PLANE_L2_PWR at 1 place, with no plane on the track's other side (fastest: NetJ4_1 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the GND and P8V planes part, and 'NetJ4_1' 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 0 of the 1 crossing(s) the track's other neighbour carries a plane, which the return can use there; 1 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetJ4_1' on TOP_LAYER at (76.11, 93.76), 0.25 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AGND%7CP8V serves each with its own layer_image region; all of them lie in x 75.02..77.2, y 92.67..94.85 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.

    • split@PLANE_L2_PWR:M2V|VDDC_TDC5 nets cross the split between the M2V and VDDC_TDC planes on PLANE_L2_PWR at 8 places, with no plane on the track's other side (fastest: TDC_ADR0 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the M2V and VDDC_TDC planes part, and 5 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 0 of the 8 crossing(s) the track's other neighbour carries a plane, which the return can use there; 8 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'TDC_ADR0' on TOP_LAYER at (94.18, 100.25), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR1' on TOP_LAYER at (92.89, 97.05), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR1' on TOP_LAYER at (94.18, 99.82), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR2' on TOP_LAYER at (93.19, 97.36), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR2' on TOP_LAYER at (94.18, 99.39), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR3' on TOP_LAYER at (93.5, 97.66), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR3' on TOP_LAYER at (94.18, 98.96), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_CAL_STOP4' on TOP_LAYER at (91.11, 95.27), 0.36 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AM2V%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 90.11..95.31, y 94.09..101.38 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.

    • split@PLANE_L2_PWR:P12V0|P3V32 nets cross the split between the P12V0 and P3V3 planes on PLANE_L2_PWR at 2 places, with no plane on the track's other side (fastest: GA0 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the P12V0 and P3V3 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 0 of the 2 crossing(s) the track's other neighbour carries a plane, which the return can use there; 2 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'GA0' on TOP_LAYER at (140.06, 125.83), 0.2 mm; other side: none: the signal layer is outermost
      • 'SDA' on TOP_LAYER at (140.51, 125.83), 0.2 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AP12V0%7CP3V3 serves each with its own layer_image region; all of them lie in x 139.06..141.51, y 124.73..126.93 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@PLANE_L2_PWR:P12V0|P8V'NetC3_2' crosses the split between the P12V0 and P8V planes on PLANE_L2_PWR at 1 place, with no plane on the track's other side (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the P12V0 and P8V planes part, and 'NetC3_2' 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 0 of the 1 crossing(s) the track's other neighbour carries a plane, which the return can use there; 1 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetC3_2' on TOP_LAYER at (111.68, 74.32), 0.25 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AP12V0%7CP8V serves each with its own layer_image region; all of them lie in x 110.56..112.81, y 73.32..75.32 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@PLANE_L2_PWR:P3V3_PLL|VDDC_TDC2 nets cross the split between the P3V3_PLL and VDDC_TDC planes on PLANE_L2_PWR at 3 places, with no plane on the track's other side (fastest: NetC22_1 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the P3V3_PLL and VDDC_TDC planes part, and 2 nets cross from one to the other at 3 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 0 of the 3 crossing(s) the track's other neighbour carries a plane, which the return can use there; 3 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetC22_1' on TOP_LAYER at (110.13, 91.03), 0.25 mm; other side: none: the signal layer is outermost
      • 'NetC22_1' on TOP_LAYER at (110.68, 91.45), 0.36 mm; other side: none: the signal layer is outermost
      • 'SPI_MOSI_3V3' on TOP_LAYER at (110.67, 98.16), 0.31 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AP3V3_PLL%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 109.13..111.86, y 89.91..99.23 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@PLANE_L2_PWR:P3V3|P3V3_PLL2 nets cross the split between the P3V3 and P3V3_PLL planes on PLANE_L2_PWR at 3 places, with no plane on the track's other side (fastest: NetC80_2 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the P3V3 and P3V3_PLL planes part, and 2 nets cross from one to the other at 3 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 0 of the 3 crossing(s) the track's other neighbour carries a plane, which the return can use there; 3 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetC80_2' on TOP_LAYER at (110.26, 109.86), 0.25 mm; other side: none: the signal layer is outermost
      • 'SPI_DAC_CSN' on TOP_LAYER at (127.56, 105.36), 0.36 mm; other side: none: the signal layer is outermost
      • 'SPI_DAC_CSN' on TOP_LAYER at (127.84, 105.94), 0.36 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AP3V3%7CP3V3_PLL serves each with its own layer_image region; all of them lie in x 109.13..128.97, y 104.18..110.86 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.

    • split@PLANE_L2_PWR:P3V3|VDDC_TDC5 nets cross the split between the P3V3 and VDDC_TDC planes on PLANE_L2_PWR at 5 places, with no plane on the track's other side (fastest: TDC_ADR0 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the P3V3 and VDDC_TDC planes part, and 5 nets cross from one to the other at 5 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 0 of the 5 crossing(s) the track's other neighbour carries a plane, which the return can use there; 5 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'SPI_MOSI_3V3' on TOP_LAYER at (104.79, 99.79), 0.25 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR0' on TOP_LAYER at (94.62, 100.69), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR1' on TOP_LAYER at (95.05, 100.69), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR2' on TOP_LAYER at (95.48, 100.69), 0.36 mm; other side: none: the signal layer is outermost
      • 'TDC_ADR3' on TOP_LAYER at (95.91, 100.69), 0.36 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AP3V3%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 93.49..105.88, y 98.7..101.81 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.

    • split@PLANE_L2_PWR:P8V|VDDC_TDC'NetC22_2' crosses the split between the P8V and VDDC_TDC planes on PLANE_L2_PWR at 1 place, with no plane on the track's other side (fastest: NetC22_2 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the P8V and VDDC_TDC planes part, and 'NetC22_2' 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 0 of the 1 crossing(s) the track's other neighbour carries a plane, which the return can use there; 1 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetC22_2' on TOP_LAYER at (117.09, 92.02), 0.36 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AP8V%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 115.97..118.22, y 90.9..93.15 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:BYPASS@PLANE_L2_PWR:hole 113.59,102.11|113.84,104.17'LF' crosses one void in the BYPASS plane on PLANE_L2_PWR at 1 place, the longest 0.02 mm (fastest: LF at 100 ps 10-90 %, ASSUMED)

      Along 'LF' on TOP_LAYER, the BYPASS plane on PLANE_L2_PWR has 0.02 mm under the track with no copper, at (113.6, 103.64), 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. What cleared it: the hole at x 113.59..113.84, y 102.11..104.17. 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.

      • 'LF' on TOP_LAYER at (113.6, 103.64), 0.02 mm over no copper, to the end of the track; 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/9/reference-planes?group=void%3ABYPASS%40PLANE_L2_PWR%3Ahole%20113.59%2C102.11%7C113.84%2C104.17 serves each with its own layer_image region; all of them lie in x 112.59..114.61, y 102.64..104.64 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:BYPASS@PLANE_L2_PWR:hole 113.59,103.68|113.84,105.972 nets cross one void in the BYPASS plane on PLANE_L2_PWR at 2 places, the longest 0.02 mm (fastest: CP at 100 ps 10-90 %, ASSUMED)

      Along 'CP' on TOP_LAYER, the BYPASS plane on PLANE_L2_PWR has 0.02 mm under the track with no copper, at (113.6, 105.64), 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x 113.59..113.84, y 103.68..105.97. 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.

      • 'CP' on TOP_LAYER at (113.6, 105.64), 0.02 mm over no copper, to the end of the track; other side: none: the signal layer is outermost
      • 'PLL_STATUS' on TOP_LAYER at (113.6, 105.14), 0.02 mm over no copper, to the end of the track; 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/9/reference-planes?group=void%3ABYPASS%40PLANE_L2_PWR%3Ahole%20113.59%2C103.68%7C113.84%2C105.97 serves each with its own layer_image region; all of them lie in x 112.59..114.61, y 104.14..106.64 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@PLANE_L5_GND:hole 141.5,112.45|143.54,120.57'VADJ' crosses one void in the GND plane on PLANE_L5_GND at 1 place, the longest 5.42 mm (fastest: VADJ at 100 ps 10-90 %, ASSUMED)

      Along 'VADJ' on BOTTOM_LAYER, the GND plane on PLANE_L5_GND has 5.42 mm under the track with no copper, at (143.28, 113.84), 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. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 3.02 mm). What cleared it: the hole at x 141.5..143.54, y 112.45..120.57. 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.

      • 'VADJ' on BOTTOM_LAYER at (143.28, 113.84), 5.42 mm, running ALONG the bare copper for 3.02 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/9/reference-planes?group=void%3AGND%40PLANE_L5_GND%3Ahole%20141.5%2C112.45%7C143.54%2C120.57 serves each with its own layer_image region; all of them lie in x 140.4..144.28, y 111.34..117.79 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.

  8. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 3 boards×3

    Checked by the review, stands · Claude walker W3 (Cern session) — Checked: the reference break is read from the copper as stated. Whether it matters depends on the signal's edge rate and the receiver's margin, which no datasheet or app note on file fixes for this net; the impedance step and reflection figures are Probe's model, and the thresholds are its starting values. Layout practice, not a datasheet requirement. Left for the owner.

    At refchange:GND|P12V0, refchange:GND|VDDC_TDC, refchange:P3V3|VDDC_TDC

    Each finding, with its own headline
    • refchange:GND|P12V02 nets change reference between GND and P12V0 at 2 vias, none with a reference that continues across (fastest: NetC41_2 at 100 ps 10-90 %, ASSUMED)

      At 2 via(s), on 2 net(s), the signal leaves a layer referenced by one of GND and P12V0 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 with none. 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 GND and P12V0 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.

      • 'NetC41_2' at (119.96, 71.45), TOP_LAYER -> BOTTOM_LAYER: P12V0 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 5.3 mm of track past the via on BOTTOM_LAYER: 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 P12V0 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.
      • 'NetC57_1' at (140.06, 68.73), TOP_LAYER -> BOTTOM_LAYER: P12V0 on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 1.4 mm of track past the via on BOTTOM_LAYER: 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 P12V0 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/9/reference-planes?group=refchange%3AGND%7CP12V0 serves each with its own layer_image region; all of them lie in x 117.96..142.06, y 66.73..73.45 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|VDDC_TDC46 nets change reference between GND and VDDC_TDC at 46 vias, none with a reference that continues across (fastest: CAL_PULSE_BUF at 100 ps 10-90 %, ASSUMED)

      At 46 via(s), on 46 net(s), the signal leaves a layer referenced by one of GND and VDDC_TDC 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: 46 with none. 46 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 VDDC_TDC 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.

      • 'NetIC8_5' at (95, 78.36), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 88.4 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'NetIC8_4' at (92.99, 78.31), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 88.0 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'TDC_RDN' at (97.89, 81.99), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 81.2 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'TDC_WRN' at (98.15, 78.36), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 75.9 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'RESETN' at (97.03, 100.08), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 50.3 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'TDC_START_P' at (91.03, 84.43), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 48.3 mm of track past the via on L4_DIFF: 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 VDDC_TDC 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.
      • 'TDC_START_N' at (91.01, 85.73), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 47.6 mm of track past the via on L4_DIFF: 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 VDDC_TDC 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.
      • 'TDC_OEN' at (97.49, 96.19), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 45.4 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'CAL_PULSE_BUF' at (92.99, 92.05), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 42.8 mm of track past the via on L4_DIFF: 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 VDDC_TDC 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.
      • 'NetIC4_46' at (110.64, 96.8), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.6 mm of track past the via on L4_DIFF: 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 VDDC_TDC 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.
      • 'TDC_CAL_STOP1' at (93.09, 90.2), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 14.7 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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.
      • 'TDC_CAL_STOP2' at (93.98, 90.25), TOP_LAYER -> BOTTOM_LAYER: VDDC_TDC on PLANE_L2_PWR to GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 13.2 mm of track past the via on BOTTOM_LAYER: 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 VDDC_TDC 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 34 more

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/9/reference-planes?group=refchange%3AGND%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 88.99..112.64, y 75.65..102.08 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:P3V3|VDDC_TDC6 nets change reference between P3V3 and VDDC_TDC at 6 vias, none with a reference that continues across (fastest: CAL_PULSE_BUF at 100 ps 10-90 %, ASSUMED)

      At 6 via(s), on 6 net(s), the signal leaves a layer referenced by one of P3V3 and VDDC_TDC 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: 6 with none. 6 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 P3V3 and VDDC_TDC 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.

      • 'TDC_START_P' at (91.03, 84.43), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 48.3 mm of track past the via on L4_DIFF: 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. P3V3 and VDDC_TDC 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.
      • 'TDC_START_N' at (91.01, 85.73), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 47.6 mm of track past the via on L4_DIFF: 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. P3V3 and VDDC_TDC 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.
      • 'CAL_PULSE_BUF' at (92.99, 92.05), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 42.8 mm of track past the via on L4_DIFF: 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. P3V3 and VDDC_TDC 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.
      • 'NetIC4_46' at (110.64, 96.8), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 20.6 mm of track past the via on L4_DIFF: 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. P3V3 and VDDC_TDC 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.
      • 'TDC_STOP_P' at (91.08, 82.4), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 11.1 mm of track past the via on L4_DIFF: 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. P3V3 and VDDC_TDC 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.
      • 'TDC_STOP_N' at (90.99, 83.33), TOP_LAYER -> L4_DIFF: VDDC_TDC on PLANE_L2_PWR to P3V3 on PLANE_L3_P3V3, GND on PLANE_L5_GND; pure, no reference continues across; stitching none, 11.0 mm of track past the via on L4_DIFF: 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. P3V3 and VDDC_TDC 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/9/reference-planes?group=refchange%3AP3V3%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 88.99..112.64, y 80.4..98.8 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.

  9. SinglePointNetmajorOpen

    SinglePointNet: 2 pins×2

    Checked by the review, stands · Claude walker (cern session) — IC8 is an acam TDC-GPX; DStop2N/DStop2 (pins 83/84) are open on the netlist (only DStop1, pins 96/97, carries TDC_STOP). The TDC-GPX pin table (datasheet p14, 'if not used' column) gives '(10kOhm to GND)' for both DStop2N and DStop2. The board leaves them floating instead, so the finding stands for the owner (StopDis2 behaviour and register setup may make it harmless, but nothing on file shows that).

    At IC8.83, IC8.84

    Each finding, with its own headline
    • IC8.83IC8.83 (DStop2N) is an IC input left unconnected

      Pin IC8.83 (DStop2N) is an IC INPUT left UNCONNECTED (on the no-connect bucket). Marked no-connect by the designer, but an input normally needs a defined level - confirm it is truly meant to float. If the datasheet defines this input left open (an internal pull, "read as low when left floating", or a tri-level input whose open state is one of its settings), file it on the pin as if_unused (pull_up, pull_down or may_float) with submit_pin_behavior, citing the page, and this finding clears.

    • IC8.84IC8.84 (DStop2) is an IC input left unconnected

      Pin IC8.84 (DStop2) is an IC INPUT left UNCONNECTED (on the no-connect bucket). Marked no-connect by the designer, but an input normally needs a defined level - confirm it is truly meant to float. If the datasheet defines this input left open (an internal pull, "read as low when left floating", or a tri-level input whose open state is one of its settings), file it on the pin as if_unused (pull_up, pull_down or may_float) with submit_pin_behavior, citing the page, and this finding clears.

  10. observationOBS-119minorOpen

    SPI chip selects of IC19 (MCP23S17) and IC14 (AD5662) have no pull-up, so both float until the carrier gateware drives them

    Two of the three SPI slave selects on this mezzanine have nothing holding them inactive before the carrier FPGA is configured. IC19 (MCP23S17 GPIO expander) CS is driven by buffer IC12 (SN74LVC1G125, OE tied to GND), and IC12's input goes straight to FMC J6-G21. IC14 (AD5662 VCXO-tuning DAC) SYNC goes straight to FMC J6-D24. Neither net has a pull-up. The third select, for IC4 (AD9516), is safe because that part has an internal 30 kohm pull-up on CS.

    What sits between what

    ``mermaid flowchart LR G21[J6-G21 SPI_IO_CSN_2V5] --> IC12[IC12 LVC1G125, OE=GND] --> CS19[IC19 pin 7 CS, active low] D24[J6-D24 SPI_DAC_CSN] --> SY14[IC14 pin 5 SYNC, active low] D27[J6-D27 SPI_PLL_CSN] --> CS4[IC4 pin 17 CS, internal 30k pull-up] ``

    Evidence

    SelectDriver before carrier configStrap on the netDevice's own holdResult
    IC19 CS (SPI_IO_CSN_3V3)IC12, whose input J6-G21 is undrivennonenone statedundefined
    IC14 SYNC (SPI_DAC_CSN)J6-D24, undrivennonenone: AD5662 logic input current is only +/-2 uAundefined
    IC4 CS (SPI_PLL_CSN)J6-D27, undrivennone30 kohm internal pull-upheld high (inactive)

    SCK and MOSI reach the slaves through IC28 and IC29, whose inputs are also unpulled. So during the window between power-up and gateware load, both the select and the clock of IC19 and IC14 are undefined.

    Why it matters

    • IC14: a floating-low SYNC with noise on SCLK can load an arbitrary code into the DAC that tunes OSC5. The AD5662-1 otherwise powers up to zero scale and holds it until a valid write.
    • IC19: a spurious write could change IODIR/OLAT, for example driving OUT_EN1-4 or TERM_ENABLE. Those control the output PhotoMOS relays and the input termination.

    Whether this happens depends on the carrier: many FPGAs enable weak pull-ups during configuration, but that is outside this board. A pull-up on SPI_IO_CSN_3V3 (or on the IC12 input) and on SPI_DAC_CSN would make the idle state the board's own. The AD9516 already does this internally.

  11. observationOBS-120minorOpen

    SPI_MISO drives 3.3 V CMOS from IC19 (MCP23S17) and IC4 (AD9516) back to the FMC VADJ (2.5 V) domain through 51 ohm only, while every other SPI line is level-translated

    The SPI bus crosses from the FMC's VADJ domain (2.5 V on this design) into the board's 3.3 V domain. In the host-to-slave direction it is translated. SCK, MOSI and the IO chip select pass through SN74LVC1G125 buffers IC28, IC29, IC12 powered from P3V3, whose inputs accept up to 5.5 V. The return direction is not translated. IC19 (MCP23S17) SO and IC4 (AD9516) SDO are 3.3 V CMOS outputs, and they reach FMC contact J6-C27 through only R68 / R114 (51 ohm each). The carrier's VADJ-bank input therefore sees a 3.3 V high level.

    What sits between what

    ``mermaid flowchart LR subgraph VADJ_2V5[FMC VADJ domain, 2.5 V] C26[J6-C26 MOSI] ; D26[J6-D26 SCK] ; G21[J6-G21 IO_CSN] ; C27[J6-C27 MISO] end C26 --> IC29[IC29 LVC1G125 @P3V3] --> SLV[IC19 / IC4 / IC14 @3.3 V] D26 --> IC28[IC28 LVC1G125 @P3V3] --> SLV G21 --> IC12[IC12 LVC1G125 @P3V3] --> SLV SLV -- IC19 SO via R68 51R --> C27 SLV -- IC4 SDO via R114 51R --> C27 ``

    Levels

    QuantityValueSource
    IC19 SO high levelVDD - 0.7 V min at -3 mA, i.e. up to ~3.3 V unloadedMCP23S17 D090
    IC4 SDO high level2.7 V min at VS = 3.3 V ("3.3 V CMOS SDO")AD9516 p.12 / p.18
    Carrier bank supplyVADJ, 2.5 V on this design (SPI_*_2V5 nets; TPS3307 SENSE3 divider R39 5k1 / R40 8k2 trips at ~2.03 V)netlist
    Excess over VADJ~0.8 V, limited only by 51 ohm into the receiver's clampcomputed

    Why it matters

    Whether this is harmless depends on the carrier FPGA. Many FPGA I/O banks specify input voltage only up to VCCO plus a few hundred millivolts (some as an absolute maximum). A 3.3 V high on a 2.5 V bank then forward-biases the receiver's clamp, and the 51 ohm resistors limit that current to a few mA without removing it. The fix is cheap and in keeping with the rest of the design: a fifth LVC1G125 (or equivalent) powered from VADJ on the MISO return, or a resistive divider sized for the SPI rate.

    Not in scope here: the two chip selects driven directly from VADJ (SPI_PLL_CSN, SPI_DAC_CSN) meet their receivers' VIH of 2.0 V, so the 2.5 V to 3.3 V direction is within spec.

  12. DecapViaFanoutminorOpen

    DecapViaFanout: 6 components×6

    Checked by the review, stands · Claude walker W3 (Cern session) — Checked: the fan-out length is read from the copper as stated. The 0.5 nH limit is Probe's starting value, not a datasheet or vendor requirement on file; whether it matters depends on this capacitor's role in the rail's PDN target. Left for the owner.

    At C1, C36A, C44A, C44C, C44D, C63

    Each finding, with its own headline
    • C1C1 on 'P12V0': its rail pad reaches its via over 2.60 mm of track and its ground pad over 0.00 mm, adding about 0.85 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 1 uF bulk capacitor: its inductance matters only far above the band it serves)

      C1 on 'P12V0': its rail pad reaches its via over 2.60 mm of track and its ground pad over 0.00 mm, adding about 0.85 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 1 uF bulk capacitor: its inductance matters only far above the band it serves). C1.1 (rail) runs 2.60 mm along its copper to its nearest via, 1.60 mm past the 1 mm beside the pad: 0.85 nH; C1.2 (ground) runs 0.00 mm along its copper to its nearest via, beside the pad. Each run past the pad as one conductor of a pair 1.80 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.85 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 0.73 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 0.20 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C36AC36A on 'P3V3': its rail pad reaches its via over 2.12 mm of track, adding about 0.57 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)

      C36A on 'P3V3': its rail pad reaches its via over 2.12 mm of track, adding about 0.57 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). C36A.1 (rail) runs 2.12 mm along its copper to its nearest via, 1.12 mm past the 1 mm beside the pad: 0.57 nH; C36A.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 1.60 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.57 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 1.16 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 0.92 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C44AC44A on 'P8V': its ground pad over 1.82 mm, adding about 0.51 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 10 uF bulk capacitor: its inductance matters only far above the band it serves)

      C44A on 'P8V': its ground pad over 1.82 mm, adding about 0.51 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 10 uF bulk capacitor: its inductance matters only far above the band it serves). C44A.1 (rail) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure; C44A.2 (ground) runs 1.82 mm along its copper to its nearest via, 0.82 mm past the 1 mm beside the pad: 0.51 nH. Each run past the pad as one conductor of a pair 2.90 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.51 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 1.89 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 1.48 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C44CC44C on 'P8V': its ground pad over 1.84 mm, adding about 0.53 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 10 uF bulk capacitor: its inductance matters only far above the band it serves)

      C44C on 'P8V': its ground pad over 1.84 mm, adding about 0.53 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 10 uF bulk capacitor: its inductance matters only far above the band it serves). C44C.1 (rail) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure; C44C.2 (ground) runs 1.84 mm along its copper to its nearest via, 0.84 mm past the 1 mm beside the pad: 0.53 nH. Each run past the pad as one conductor of a pair 2.90 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.53 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 1.89 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 1.48 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C44DC44D on 'P8V': its ground pad over 2.75 mm, adding about 1.09 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 10 uF bulk capacitor: its inductance matters only far above the band it serves)

      C44D on 'P8V': its ground pad over 2.75 mm, adding about 1.09 nH to its loop, past Probe's 0.5 nH starting value; put the vias at the pads (a 10 uF bulk capacitor: its inductance matters only far above the band it serves). C44D.1 (rail) not measured: the pad sits in a pour of its net on its own layer, which has no track length to measure; C44D.2 (ground) runs 2.75 mm along its copper to its nearest via, 1.75 mm past the 1 mm beside the pad: 1.09 nH. Each run past the pad as one conductor of a pair 2.90 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 1.09 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 1.89 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 1.48 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

    • C63C63 on 'P3V3': its rail pad reaches its via over 2.18 mm of track, adding about 0.60 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)

      C63 on 'P3V3': its rail pad reaches its via over 2.18 mm of track, adding about 0.60 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). C63.2 (rail) runs 2.18 mm along its copper to its nearest via, 1.18 mm past the 1 mm beside the pad: 0.60 nH; C63.1 (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 1.60 mm apart (the pad pitch), 0.2 l ln(D/R) nH with R 0.127 mm (the track's width is not read): 0.60 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 1.38 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 1.28 mm depth, ASSUMED). 0.5 nH is Probe's starting value, about the whole mounted inductance of a well-mounted 0402.

  13. ReferencePlaneDiscontinuityminorOpen

    ReferencePlaneDiscontinuity: 5 boards×5

    Checked by the review, stands · Claude walker W3 (Cern session) — Checked: the reference break is read from the copper as stated. Whether it matters depends on the signal's edge rate and the receiver's margin, which no datasheet or app note on file fixes for this net; the impedance step and reflection figures are Probe's model, and the thresholds are its starting values. Layout practice, not a datasheet requirement. Left for the owner.

    At along@PLANE_L2_PWR:GND|P8V, along@PLANE_L2_PWR:M2V|VDDC_TDC, gap:PLANE_L2_PWR:hole 139.19,125.73|149.83,125.93, split@PLANE_L2_PWR:GND|P12V0, split@PLANE_L2_PWR:GND|P3V3

    Each finding, with its own headline
    • along@PLANE_L2_PWR:GND|P8V'CAL_PULSE_BUF' runs ALONG the split between the GND and P8V planes on PLANE_L2_PWR at 1 place, the longest 3.1 mm; the impedance rises from 51 ohm to 54 ohm there at worst, 2.5 % of each edge reflected (fastest: CAL_PULSE_BUF at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the GND and P8V planes part, and 'CAL_PULSE_BUF' 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 51 ohm; along the slot 54 ohm, reflecting up to 2.5 % 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 78 ohm). At every place the rise is within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is the lower tier (info). One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • 'CAL_PULSE_BUF' on TOP_LAYER at (81.23, 79.63), 3.17 mm, running ALONG the split for 3.1 mm, not across it, 51 ohm over solid plane, 54 ohm along it: 2.5 % of each edge reflected; 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/9/reference-planes?group=along%40PLANE_L2_PWR%3AGND%7CP8V serves each with its own layer_image region; all of them lie in x 78.66..83.8, y 78.63..80.65 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.

    • along@PLANE_L2_PWR:M2V|VDDC_TDC'TDC_ADR0' runs ALONG the split between the M2V and VDDC_TDC planes on PLANE_L2_PWR at 1 place, the longest 0.85 mm; the impedance rises from 65 ohm to 69 ohm there at worst, 3 % of each edge reflected (fastest: TDC_ADR0 at 100 ps 10-90 %, ASSUMED)

      On PLANE_L2_PWR the M2V and VDDC_TDC planes part, and 'TDC_ADR0' 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 65 ohm; along the slot 69 ohm, reflecting up to 3 % 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 92 ohm). At every place the rise is within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is the lower tier (info). One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • 'TDC_ADR0' on TOP_LAYER at (92.28, 96.46), 1.21 mm, running ALONG the split for 0.85 mm, not across it, 65 ohm over solid plane, 69 ohm along it: 3 % of each edge reflected; 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/9/reference-planes?group=along%40PLANE_L2_PWR%3AM2V%7CVDDC_TDC serves each with its own layer_image region; all of them lie in x 90.99..93.58, y 94.97..97.93 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.

    • gap:PLANE_L2_PWR:hole 139.19,125.73|149.83,125.93'SCL' crosses one gap in PLANE_L2_PWR's negative-plane field at 1 place, the longest 0.2 mm (fastest: SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SCL' on TOP_LAYER, PLANE_L2_PWR is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.2 mm under the track at (141.05, 125.83). The layer does not state the nets either side, so whether this is a split between two planes or a void in one is not said; either way the return current has to go round it. What cleared it: the hole at x 139.19..149.83, y 125.73..125.93. 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.

      • 'SCL' on TOP_LAYER at (141.05, 125.83), 0.2 mm; 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/9/reference-planes?group=gap%3APLANE_L2_PWR%3Ahole%20139.19%2C125.73%7C149.83%2C125.93 serves each with its own layer_image region; all of them lie in x 140.05..142.05, y 124.73..126.93 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@PLANE_L2_PWR:GND|P12V0'EXT_RESET_N' crosses the split between the GND and P12V0 planes on PLANE_L2_PWR at 1 place, with no plane on the track's other side (fastest: EXT_RESET_N at 20 ns 10-90 %, suggested by its name)

      On PLANE_L2_PWR the GND and P12V0 planes part, and 'EXT_RESET_N' 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 0 of the 1 crossing(s) the track's other neighbour carries a plane, which the return can use there; 1 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'EXT_RESET_N' on TOP_LAYER at (136.96, 76.07), 0.36 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AGND%7CP12V0 serves each with its own layer_image region; all of them lie in x 135.83..138.08, y 74.95..77.2 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.

    • split@PLANE_L2_PWR:GND|P3V33 nets cross the split between the GND and P3V3 planes on PLANE_L2_PWR at 3 places, with no plane on the track's other side (fastest: SPI_MOSI_3V3 at 2 ns 10-90 %, suggested by its name)

      On PLANE_L2_PWR the GND and P3V3 planes part, and 3 nets cross from one to the other at 3 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 0 of the 3 crossing(s) the track's other neighbour carries a plane, which the return can use there; 3 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'EXT_RESET_N' on TOP_LAYER at (135.52, 100.97), 0.29 mm; other side: none: the signal layer is outermost
      • 'SPI_MOSI_3V3' on TOP_LAYER at (126.03, 100.97), 0.2 mm; other side: none: the signal layer is outermost
      • 'SPI_SCK_3V3' on TOP_LAYER at (125.38, 100.97), 0.2 mm; 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/9/reference-planes?group=split%40PLANE_L2_PWR%3AGND%7CP3V3 serves each with its own layer_image region; all of them lie in x 124.38..136.62, y 99.86..102.07 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
24AA64T-I/MCMICROCHIP TECHNOLOGY1.7-5.5V 64K I2C Serial EEPROMIC22
571-0122-100FDIALIGHTLED x2 Green 1C2A-BOT_3C4A-TOPLD1
AD5662BRMZ-1ANALOG DEVICES2.7V to 5.5V, 250uA, Rail-to-Rail Output 16-Bit nanoDACIC14
AD9516-4BCPZANALOG DEVICES14-Output Clock Generator with Integrated 1.6 GHz VCOIC4
ASP-134604-01SAMTECConnector VITA 57, 160 Male contacts in 4 row cdghJ6
DS18B20U+DALLAS MAXIMProgrammable Resolution 1-Wire Digital ThermometerIC13
LM336M-2.5/NOPBTEXAS INSTRUMENTSReference DiodeIC10
LP5951MF-3.0/NOPBTEXAS INSTRUMENTS150mA Micropower Low-Dropout CMOS Voltage RegulatorIC33
LT1931AES5#PBFLINEAR TECHNOLOGY1.2MHz/2.2MHz Inverting DC/DC Converters in ThinSOTIC30
LT3503EDCB#PBFLINEAR TECHNOLOGY1A, 2.2MHz Step-Down Switching RegulatorIC11
MC100EPT20DTGON SEMICONDUCTOR3.3V LVTTL/LVCMOS to Differential LVPECL TranslatorIC5
MCP23S17-E/MLMICROCHIP TECHNOLOGY16-bit I/O Port Expander With Serial InterfaceIC19
SMBJ10AFAIRCHILD SEMICONDUCTORZener DiodeD7
TDC-GPXACAM MESS-ELECTRONICMultifunctional High-End Time-to-Digital Converter (4 different operation modes, 81 ps up to 10 ps resolution)IC8
TPS3307-33DGNTEXAS INSTRUMENTSTriple Processor SupervisorsIC16
TS3USB221DRCRTEXAS INSTRUMENTSHigh-Speed USB 2.0 (480-Mbps) 1:2 Multiplexer/Demultiplexer Switch With Single EnableIC17
VM53S3-25.000-2.5/-30+75MERCURY ELECTRONICSOscillator 4 pins 1=VC 2=GND 3=OUT 4=+VsOSC5
1N4148WSFAIRCHILD SEMICONDUCTORDiodeD1, D5
MBRM140T1GON SEMICONDUCTORSchottky DiodeD2, D3
MC100EP29DTGON SEMICONDUCTOR3.3V/5V ECL Dual Differential Data and Clock D Flip-Flop With Set and ResetIC3A, IC3B
SY100EPT23LZGMICREL3.3V Dual Differential LVPECL/LVDS/CML to LVTTL/LVCMOS TranslatorIC1A, IC1B
AP1117YG-13DIODES INCORPORATED3-Terminal Regulator 1=Adj 2=OUT 3=INIC21, IC6, IC9
BSH103NXP SEMICONDUCTORSN-CHANNEL MOSFET WITH DIODET1, T2, T3
SN74AUP1G08DCKTTEXAS INSTRUMENTSSingle 2-Input AND GateIC15, IC18, IC20
AD8009ARZANALOG DEVICESOperational AmplifierIC25A, IC25B, IC25C, IC25D
AQY221N3MPANASONIC ELECTRIC WORKSPhotoMOS RelaySW1A, SW1B, SW1C, SW1D
SY89295UMGMICREL2.5V/3.3V 1.5GHz Precision LVPECL Programmable DelayIC2A, IC2B, IC7A, IC7B
1206L012WRLITTELFUSEFuseF1A, F1B, F1C, F1D, F5
BAR66INFINEONSerial DiodeD4A, D4B, D4C, D4D, D6
EPK.00.250.NTNLEMOJ1, J2, J3, J4, J5
SN74AUP1G80DCKTTEXAS INSTRUMENTSSingle Positive-Edge-Triggered D-Type Flip-FlopIC24, IC27A, IC27B, IC27C, IC27D
SN74LVC1G125DCKTTEXAS INSTRUMENTSSingle Bus Buffer Gate With 3-State OutputIC12, IC23, IC26A, IC26B, IC26C, IC26D and 4 more
R0603_22R_1%_0.1W_100PPMGENERICResistor - 1%R57
BLM21PG221SN1DMURATAInductorL17, L3
CC0603_330NF_16V_10%_X5RGENERICNon-polarised capacitorC18, C51
R0603_10R_1%_0.1W_200PPMGENERICResistor - 1%R16, R22
R0603_4K12_1%_0.1W_100PPMGENERICResistor - 1%R237, R31, R44
R1206_150R_1%_0.25W_100PPMGENERICResistor - 1%R110, R116, R117
VLCF5020T-4R7N1R7-1TDKInductorL1, L2, L4
R0603_330R_1%_0.1W_100PPMGENERICResistor - 1%R71A, R71B, R71C, R71D

About this design

Design
FMC DELAY 1NS 8CHA — OHWR0006 R1
Designed in
Altium Designer
Board
6 copper layers, 81.6 × 69 mm, 354 components, 279 nets
Licence
CC-BY-SA-4.0, CC0-1.0, CERN-OHL-W-2.0, GPL-2.0-or-later, LGPL-2.1, LGPL-2.1-or-later
Source
https://gitlab.com/ohwr/project/fmc-delay-1ns-8cha at commit a0712d66aa1d
Attribution
CERN Open Hardware Repository, project ohwr/project/fmc-delay-1ns-8cha (commit a0712d66aa1d3a015aa1b5d3ac3c39d73312c907, 2025-04-12). Licence: CC-BY-SA-4.0, CC0-1.0, CERN-OHL-W-2.0, GPL-2.0-or-later, LGPL-2.1, LGPL-2.1-or-later. Source: https://gitlab.com/ohwr/project/fmc-delay-1ns-8cha. 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)