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

UTCA RTM 8 SFP

OHWR0029 · revision R1

UTCA RTM 8 SFP (OHWR0029 R1) is an open-hardware circuit board from the CERN Open Hardware Repository, designed in Altium Designer: a 4-layer board of 182.5 × 148.5 mm with 363 components and 312 nets, published under CERN-OHL-1.1.

review 91% walked · 13 findings: 5 major · 8 minor (67 items)

Open in the interactive viewerRead the full review

UTCA RTM 8 SFP (OHWR0029 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: 5 major · 8 minor (67 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. observationOBS-302majorOpen

    100 R pull-downs R2/R3/R4/R5/R13 on the serial-register control lines RTM_IO2..RTM_IO6 need about 23 mA per line from the host to reach a 74HC04 logic high

    The 100 R pull-downs R2, R3, R4, R5 and R13 on the serial-register control lines RTM_IO2..RTM_IO6 need about 23 mA per line from the off-board host to reach a 74HC04 (IC1) logic high, about 115 mA for the five lines.

    The lines

    Each host-driven line goes straight to a 74HC04 input with a 100 R (R0603_100R_1%) resistor to GND, and nothing else on the net.

    NetSchematic name (sheet 6)Connector pinIC1 inputPull-down
    RTM_IO2REG_OEnJ6.C7pin 11R5
    RTM_IO3REG_CLKnJ6.C8pin 3R13
    RTM_IO4REG_STRnJ6.C9pin 9R2
    RTM_IO5REG_DINnJ6.C10pin 5R3
    RTM_IO6REG_PLJ2.A1pin 1R4

    Source: netlist; schematic sheet 6.

    The numbers

    QuantityValueSource
    74HC04 VIH min1.5 V at VCC = 2.0 V; 3.15 V at VCC = 4.5 V; 4.2 V at VCC = 6.0 V; about 0.7 x VCC74HC04 datasheet, p.5
    P3V3 rail0.758 x (1 + 4.3/1.3) = 3.27 V (IC16 VFB2 on R53 4k3 / R18 1k3, TEST2 tied to V5FILT)TPS53126 datasheet, p.5; netlist; derived
    VIH on that railroughly 2.3 Vderived
    Current per line held high2.3 V / 100 R = 23 mAderived
    Five linesabout 115 mA, continuously while highderived

    Why it matters

    100 R is a termination value, not a bias value. As a pull-down it is a DC load most FPGA/CPLD LVCMOS33 outputs are not rated to drive to a valid high (typical maximum drive settings are 12-24 mA, with VOH specified at far lower current). The clock line (RTM_IO3/REG_CLKn) must also slew that load at the shift rate.

    If the host cannot meet it, REG_OEn/REG_STRn/REG_CLKn/REG_DINn/REG_PL never read high, the register chain cannot be loaded, and OE/STROBE stay in their power-up state. The host is off-board (the AMC), so whether its drivers can source 23 mA per pin is a requirement question.

    Suggested fix: a 10 k (or 4.7 k) pull-down would give the same default with negligible load. Worth the designer confirming the value is intended.

  2. observationOBS-312majorOpen

    P3V3 converter current limit (R246 5k1 on TRIP2) leaves no margin over the designer's own >5 A load estimate for eight 2 W SFP+ modules

    The guaranteed current limit on P3V3 (~5.0 A at 25 C) is no higher than the designer's own load estimate (>5 A), so a board fully loaded with 2 W SFP+ modules can run into current limit.

    What sets the limit

    IC16 TPS53126 channel 2 makes P3V3 (VO2 pin 6) with low-side FET IRF9910 Q2 in T6 (sheet 3 note 'B - Q2 - Low'), L1 1.5 uH, 700 kHz (TEST2 tied to V5FILT), Vin 12 V. Limit per TPS53126 p.16 eq.13: IOCL = VTRIP / RDS(on) + Iripple / 2, with VTRIP = R246 x ITRIP.

    QuantityMin / worstTypSource
    ITRIP8.5 uA10 uATPS53126 p.6
    VTRIP = 5.1 k x ITRIP43.4 mV51 mVR246 5k1
    Q2 RDS(on) at VGS 4.5 V, 25 C11.3 mOhm (max)9.1 mOhmIRF9910 p.2
    Iripple p-p = (12 - 3.27) x 3.27 / (1.5 uH x 700 kHz x 12 V)2.27 A2.27 Acalc
    IOCL4.97 A6.7 Aeq.13
    Designer's load estimate (sheet 3)>5 A ('SFP max 2W ... *8 = 4.8A, clocks 200mA, total I >5A')schematic

    Why it matters

    The estimated load sits at or above the guaranteed minimum limit before the OCP compensation offset (-15..+15 mV at 25 C, p.6), inductor tolerance and frequency variation are counted.

  3. PartLifecyclemajorOpen

    PartLifecycle: 3 part numbers×3

    2 Obsolete, 1 Discontinued at DigiKey · on 5 placements

    At IC14, J3, T5

    Each finding, with its own headline
    • IC14PCA9554DB on IC14: the distributor catalogue reports "Obsolete" for Texas Instruments's PCA9554DB

      PCA9554DB on IC14: the distributor catalogue reports "Obsolete" for Texas Instruments's PCA9554DB — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names no manufacturer for it, so whether that is the fitted maker's product was not checked. A distributor delisting is not the same claim as the part being unavailable; check the vendor's own lifecycle/product-change notice for the status and any named successor. Re-read the source: GET /api/odb/parts/PCA9554DB/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 (2024-03-28, 962 day(s) ago), so this is a substitution rather than a buy. The distributor reports 25,904 left. No approved alternate is recorded for this position.

    • J385_SMA-50-0-101/111_NE on 2 placements (J3, J4): the distributor catalogue reports "Obsolete" for Huber+Suhner, Inc.'s 85_SMA-50-0-101/111_NE

      85_SMA-50-0-101/111_NE on 2 placements (J3, J4): the distributor catalogue reports "Obsolete" for Huber+Suhner, Inc.'s 85_SMA-50-0-101/111_NE — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. That is the maker this board's BOM names (HUBER+SUHNER). 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/85_SMA-50-0-101/111_NE/providers (which provider said it, the maker and part number it matched, its own words, and when we asked). THE LAST BUY DATE HAS PASSED (2025-12-31, 319 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.

    • T5IRF9910Pbf on 2 placements (T5, T6): the distributor catalogue reports "Discontinued at DigiKey" for Infineon Technologies's IRF9910PBF

      IRF9910Pbf on 2 placements (T5, T6): the distributor catalogue reports "Discontinued at DigiKey" for Infineon Technologies's IRF9910PBF — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names INTERNATIONAL RECTIFIER, and that is one line with Infineon Technologies: Infineon completed its acquisitions of International Rectifier on 2015-01-13 and Cypress Semiconductor on 2020-04-16. 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/IRF9910Pbf/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.

  4. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 41 boards×41

    41 checked by the review by Claude walker W4 (Cern session) — 12 different reasons

    At gap:MIDPLANE_1:hole -8.99,77.41|-8.74,84.62, gap:MIDPLANE_1:hole -8.99,84.37|-5.73,87.63, gap:MIDPLANE_2:hole -110.78,101.34|-110.52,105.91, gap:MIDPLANE_2:hole -110.78,110.24|-110.52,114.81, gap:MIDPLANE_2:hole -110.78,115.59|-110.52,120.16, gap:MIDPLANE_2:hole -110.78,124.49|-110.52,129.06 and 35 more

    Each finding, with its own headline
    • gap:MIDPLANE_1:hole -8.99,77.41|-8.74,84.62'NetIC12_1' crosses one gap in MIDPLANE_1's negative-plane field at 1 place, the longest 0.25 mm (fastest: NetIC12_1 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC12_1' on TOP_LAYER, MIDPLANE_1 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-8.87, 83.7). 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 -8.99..-8.74, y 77.41..84.62. 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.

      • 'NetIC12_1' on TOP_LAYER at (-8.87, 83.7), 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/31/reference-planes?group=gap%3AMIDPLANE_1%3Ahole%20-8.99%2C77.41%7C-8.74%2C84.62 serves each with its own layer_image region; all of them lie in x -9.99..-7.74, y 82.7..84.7 mm (board frame, Y up).

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

    • gap:MIDPLANE_1:hole -8.99,84.37|-5.73,87.63'NetIC12_2' crosses one gap in MIDPLANE_1's negative-plane field at 1 place, the longest 0.36 mm (fastest: NetIC12_2 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC12_2' on TOP_LAYER, MIDPLANE_1 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-8.77, 84.59). 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 -8.99..-5.73, y 84.37..87.63. 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.

      • 'NetIC12_2' on TOP_LAYER at (-8.77, 84.59), 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/31/reference-planes?group=gap%3AMIDPLANE_1%3Ahole%20-8.99%2C84.37%7C-5.73%2C87.63 serves each with its own layer_image region; all of them lie in x -9.95..-7.59, y 83.59..85.59 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.78,101.34|-110.52,105.913 nets cross one gap in MIDPLANE_2's negative-plane field at 3 places, the longest 0.25 mm (fastest: SFP_LED1_3 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_3' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 103.91). 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. 3 places on 3 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 101.34..105.91. 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.

      • 'SFP_LED1_3' on BOTTOM_LAYER at (-110.65, 103.91), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE3' on BOTTOM_LAYER at (-110.65, 103.2), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT3' on BOTTOM_LAYER at (-110.65, 103.56), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C101.34%7C-110.52%2C105.91 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 102.2..104.91 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.78,110.24|-110.52,114.814 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN2 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_DETECTN2' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 111.88). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 110.24..114.81. 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.

      • 'SFP_DETECTN2' on BOTTOM_LAYER at (-110.65, 111.88), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS2' on BOTTOM_LAYER at (-110.65, 112.24), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_2' on BOTTOM_LAYER at (-110.65, 112.95), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_2' on BOTTOM_LAYER at (-110.65, 112.6), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C110.24%7C-110.52%2C114.81 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 110.88..113.95 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.78,115.59|-110.52,120.162 nets cross one gap in MIDPLANE_2's negative-plane field at 2 places, the longest 0.25 mm (fastest: SFP_TXFAULT2 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXDISABLE2' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 117.25). 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 115.59..120.16. 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.

      • 'SFP_TXDISABLE2' on BOTTOM_LAYER at (-110.65, 117.25), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT2' on BOTTOM_LAYER at (-110.65, 117.61), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C115.59%7C-110.52%2C120.16 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 116.25..118.61 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.78,124.49|-110.52,129.064 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN1 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS1_1' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 126.58). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 124.49..129.06. 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.

      • 'SFP_DETECTN1' on BOTTOM_LAYER at (-110.65, 127.29), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS1' on BOTTOM_LAYER at (-110.65, 126.22), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_1' on BOTTOM_LAYER at (-110.65, 126.94), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_1' on BOTTOM_LAYER at (-110.65, 126.58), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C124.49%7C-110.52%2C129.06 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 125.22..128.29 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:MIDPLANE_2:hole -110.78,129.84|-110.52,134.413 nets cross one gap in MIDPLANE_2's negative-plane field at 3 places, the longest 0.25 mm (fastest: SFP_LED1_1 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXFAULT1' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 132.57). 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. 3 places on 3 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 129.84..134.41. 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.

      • 'SFP_LED1_1' on BOTTOM_LAYER at (-110.65, 132.22), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE1' on BOTTOM_LAYER at (-110.65, 131.86), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT1' on BOTTOM_LAYER at (-110.65, 132.57), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C129.84%7C-110.52%2C134.41 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 130.86..133.57 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:MIDPLANE_2:hole -110.78,81.74|-110.52,86.314 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN4 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS1_4' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 84.1). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 81.74..86.31. 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.

      • 'SFP_DETECTN4' on BOTTOM_LAYER at (-110.65, 83.38), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS4' on BOTTOM_LAYER at (-110.65, 83.74), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_4' on BOTTOM_LAYER at (-110.65, 84.45), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_4' on BOTTOM_LAYER at (-110.65, 84.1), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C81.74%7C-110.52%2C86.31 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 82.38..85.45 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.78,87.09|-110.52,91.662 nets cross one gap in MIDPLANE_2's negative-plane field at 2 places, the longest 0.25 mm (fastest: SFP_TXFAULT4 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXFAULT4' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 89.74). 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 87.09..91.66. 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.

      • 'SFP_TXDISABLE4' on BOTTOM_LAYER at (-110.65, 89.26), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT4' on BOTTOM_LAYER at (-110.65, 89.74), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C87.09%7C-110.52%2C91.66 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 88.26..90.74 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:MIDPLANE_2:hole -110.78,95.99|-110.52,100.564 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN3 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LOS3' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.65, 97.99). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.78..-110.52, y 95.99..100.56. 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.

      • 'SFP_DETECTN3' on BOTTOM_LAYER at (-110.65, 97.63), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS3' on BOTTOM_LAYER at (-110.65, 97.99), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_3' on BOTTOM_LAYER at (-110.65, 98.7), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_3' on BOTTOM_LAYER at (-110.65, 98.35), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.78%2C95.99%7C-110.52%2C100.56 serves each with its own layer_image region; all of them lie in x -111.78..-109.52, y 96.63..99.7 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.79,10.71|-110.54,15.284 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_TXFAULT8 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_I2C_1.SCL' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 13.26). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 10.71..15.28. 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.

      • 'SFP_I2C_1.SCL' on BOTTOM_LAYER at (-110.66, 13.26), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_1.SDA' on BOTTOM_LAYER at (-110.66, 13.61), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE8' on BOTTOM_LAYER at (-110.66, 12.54), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT8' on BOTTOM_LAYER at (-110.66, 12.9), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C10.71%7C-110.54%2C15.28 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 11.54..14.61 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.79,19.61|-110.54,24.184 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN7 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_DETECTN7' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 21.25). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 19.61..24.18. 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.

      • 'SFP_DETECTN7' on BOTTOM_LAYER at (-110.66, 21.25), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS7' on BOTTOM_LAYER at (-110.66, 21.61), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_7' on BOTTOM_LAYER at (-110.66, 22.32), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_7' on BOTTOM_LAYER at (-110.66, 21.97), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C19.61%7C-110.54%2C24.18 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 20.25..23.32 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.

    • gap:MIDPLANE_2:hole -110.79,24.96|-110.54,29.534 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_TXFAULT7 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_I2C_2.SDA' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 27.86). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 24.96..29.53. 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.

      • 'SFP_I2C_2.SCL' on BOTTOM_LAYER at (-110.66, 27.5), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_2.SDA' on BOTTOM_LAYER at (-110.66, 27.86), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE7' on BOTTOM_LAYER at (-110.66, 26.79), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT7' on BOTTOM_LAYER at (-110.66, 27.15), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C24.96%7C-110.54%2C29.53 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 25.79..28.86 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.

    • gap:MIDPLANE_2:hole -110.79,33.86|-110.54,38.434 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN6 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_DETECTN6' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 35.5). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 33.86..38.43. 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.

      • 'SFP_DETECTN6' on BOTTOM_LAYER at (-110.66, 35.5), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS6' on BOTTOM_LAYER at (-110.66, 35.86), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_6' on BOTTOM_LAYER at (-110.66, 36.57), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_6' on BOTTOM_LAYER at (-110.66, 36.22), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C33.86%7C-110.54%2C38.43 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 34.5..37.57 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:MIDPLANE_2:hole -110.79,39.21|-110.54,43.784 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_TXFAULT6 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_I2C_3.SCL' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 41.75). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 39.21..43.78. 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.

      • 'SFP_I2C_3.SCL' on BOTTOM_LAYER at (-110.66, 41.75), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_3.SDA' on BOTTOM_LAYER at (-110.66, 42.11), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE6' on BOTTOM_LAYER at (-110.66, 41.04), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT6' on BOTTOM_LAYER at (-110.66, 41.4), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C39.21%7C-110.54%2C43.78 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 40.04..43.11 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -110.79,48.11|-110.54,52.684 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN5 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_DETECTN5' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 49.75). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 48.11..52.68. 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.

      • 'SFP_DETECTN5' on BOTTOM_LAYER at (-110.66, 49.75), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS5' on BOTTOM_LAYER at (-110.66, 50.11), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_5' on BOTTOM_LAYER at (-110.66, 50.82), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_5' on BOTTOM_LAYER at (-110.66, 50.47), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C48.11%7C-110.54%2C52.68 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 48.75..51.82 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:MIDPLANE_2:hole -110.79,5.36|-110.54,9.934 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN8 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LOS8' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 7.36). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 5.36..9.93. 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.

      • 'SFP_DETECTN8' on BOTTOM_LAYER at (-110.66, 7), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS8' on BOTTOM_LAYER at (-110.66, 7.36), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_8' on BOTTOM_LAYER at (-110.66, 8.07), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_8' on BOTTOM_LAYER at (-110.66, 7.72), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C5.36%7C-110.54%2C9.93 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 6..9.07 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.

    • gap:MIDPLANE_2:hole -110.79,53.46|-110.54,58.034 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_TXFAULT5 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXDISABLE5' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-110.66, 55.29). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -110.79..-110.54, y 53.46..58.03. 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.

      • 'SFP_I2C_4.SCL' on BOTTOM_LAYER at (-110.66, 56), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_4.SDA' on BOTTOM_LAYER at (-110.66, 56.36), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE5' on BOTTOM_LAYER at (-110.66, 55.29), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT5' on BOTTOM_LAYER at (-110.66, 55.65), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-110.79%2C53.46%7C-110.54%2C58.03 serves each with its own layer_image region; all of them lie in x -111.79..-109.54, y 54.29..57.36 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:MIDPLANE_2:hole -119.08,120.41|-111.03,120.67'SFP_LED1_2' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.36 mm (fastest: SFP_LED1_2 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_2' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-115.69, 120.54). 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 -119.08..-111.03, y 120.41..120.67. 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.

      • 'SFP_LED1_2' on BOTTOM_LAYER at (-115.69, 120.54), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.08%2C120.41%7C-111.03%2C120.67 serves each with its own layer_image region; all of them lie in x -116.82..-114.56, y 119.41..121.67 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:MIDPLANE_2:hole -119.08,91.91|-111.03,92.17'SFP_LED1_4' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.36 mm (fastest: SFP_LED1_4 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_4' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-115.51, 92.04). 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 -119.08..-111.03, y 91.91..92.17. 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.

      • 'SFP_LED1_4' on BOTTOM_LAYER at (-115.51, 92.04), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.08%2C91.91%7C-111.03%2C92.17 serves each with its own layer_image region; all of them lie in x -116.64..-114.38, y 90.91..93.17 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.09,15.53|-111.05,15.79'SFP_LED1_8' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.36 mm (fastest: SFP_LED1_8 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_8' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-115.62, 15.66). 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 -119.09..-111.05, y 15.53..15.79. 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.

      • 'SFP_LED1_8' on BOTTOM_LAYER at (-115.62, 15.66), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.09%2C15.53%7C-111.05%2C15.79 serves each with its own layer_image region; all of them lie in x -116.75..-114.49, y 14.53..16.79 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.

    • gap:MIDPLANE_2:hole -119.09,29.78|-111.05,30.04'SFP_LED1_7' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.36 mm (fastest: SFP_LED1_7 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_7' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-115.99, 29.91). 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 -119.09..-111.05, y 29.78..30.04. 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.

      • 'SFP_LED1_7' on BOTTOM_LAYER at (-115.99, 29.91), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.09%2C29.78%7C-111.05%2C30.04 serves each with its own layer_image region; all of them lie in x -117.12..-114.86, y 28.78..31.04 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.09,44.03|-111.05,44.29'SFP_LED1_6' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.36 mm (fastest: SFP_LED1_6 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_6' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-116.55, 44.16). 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 -119.09..-111.05, y 44.03..44.29. 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.

      • 'SFP_LED1_6' on BOTTOM_LAYER at (-116.55, 44.16), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.09%2C44.03%7C-111.05%2C44.29 serves each with its own layer_image region; all of them lie in x -117.68..-115.42, y 43.03..45.29 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:MIDPLANE_2:hole -119.09,58.28|-111.05,58.54'SFP_LED1_5' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.36 mm (fastest: SFP_LED1_5 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_5' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-115.6, 58.41). 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 -119.09..-111.05, y 58.28..58.54. 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.

      • 'SFP_LED1_5' on BOTTOM_LAYER at (-115.6, 58.41), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.09%2C58.28%7C-111.05%2C58.54 serves each with its own layer_image region; all of them lie in x -116.73..-114.47, y 57.28..59.54 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:MIDPLANE_2:hole -119.58,101.32|-119.32,105.923 nets cross one gap in MIDPLANE_2's negative-plane field at 3 places, the longest 0.25 mm (fastest: SFP_LED1_3 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXDISABLE3' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 103.2). 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. 3 places on 3 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 101.32..105.92. 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.

      • 'SFP_LED1_3' on BOTTOM_LAYER at (-119.45, 103.91), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE3' on BOTTOM_LAYER at (-119.45, 103.2), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT3' on BOTTOM_LAYER at (-119.45, 103.56), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C101.32%7C-119.32%2C105.92 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 102.2..104.91 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.58,110.23|-119.32,114.834 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN2 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS1_2' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 113.24). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 110.23..114.83. 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.

      • 'SFP_DETECTN2' on BOTTOM_LAYER at (-119.45, 112.53), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS2' on BOTTOM_LAYER at (-119.45, 112.89), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_2' on BOTTOM_LAYER at (-119.45, 113.6), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_2' on BOTTOM_LAYER at (-119.45, 113.24), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C110.23%7C-119.32%2C114.83 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 111.53..114.6 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.58,115.57|-119.32,120.173 nets cross one gap in MIDPLANE_2's negative-plane field at 3 places, the longest 0.25 mm (fastest: SFP_LED1_2 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXDISABLE2' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 117.25). 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. 3 places on 3 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 115.57..120.17. 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.

      • 'SFP_LED1_2' on BOTTOM_LAYER at (-119.45, 118.42), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE2' on BOTTOM_LAYER at (-119.45, 117.25), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT2' on BOTTOM_LAYER at (-119.45, 117.61), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C115.57%7C-119.32%2C120.17 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 116.25..119.42 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.58,124.48|-119.32,129.084 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN1 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS1_1' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 127.41). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 124.48..129.08. 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.

      • 'SFP_DETECTN1' on BOTTOM_LAYER at (-119.45, 128.21), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS1' on BOTTOM_LAYER at (-119.45, 127.05), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_1' on BOTTOM_LAYER at (-119.45, 127.85), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_1' on BOTTOM_LAYER at (-119.45, 127.41), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C124.48%7C-119.32%2C129.08 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 126.05..129.21 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.58,129.82|-119.32,134.423 nets cross one gap in MIDPLANE_2's negative-plane field at 3 places, the longest 0.25 mm (fastest: SFP_LED1_1 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXDISABLE1' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 131.86). 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. 3 places on 3 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 129.82..134.42. 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.

      • 'SFP_LED1_1' on BOTTOM_LAYER at (-119.45, 132.22), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE1' on BOTTOM_LAYER at (-119.45, 131.86), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT1' on BOTTOM_LAYER at (-119.45, 132.57), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C129.82%7C-119.32%2C134.42 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 130.86..133.57 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:MIDPLANE_2:hole -119.58,81.73|-119.32,86.334 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN4 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS0_4' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 85.32). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 81.73..86.33. 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.

      • 'SFP_DETECTN4' on BOTTOM_LAYER at (-119.45, 84.25), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS4' on BOTTOM_LAYER at (-119.45, 84.61), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_4' on BOTTOM_LAYER at (-119.45, 85.32), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_4' on BOTTOM_LAYER at (-119.45, 84.96), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C81.73%7C-119.32%2C86.33 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 83.25..86.32 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.

    • gap:MIDPLANE_2:hole -119.58,87.07|-119.32,91.673 nets cross one gap in MIDPLANE_2's negative-plane field at 3 places, the longest 0.25 mm (fastest: SFP_LED1_4 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_4' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 89.92). 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. 3 places on 3 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 87.07..91.67. 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.

      • 'SFP_LED1_4' on BOTTOM_LAYER at (-119.45, 89.92), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE4' on BOTTOM_LAYER at (-119.45, 88.31), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT4' on BOTTOM_LAYER at (-119.45, 89.11), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C87.07%7C-119.32%2C91.67 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 87.31..90.92 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.58,95.98|-119.32,100.584 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN3 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_DETECTN3' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.45, 98.49). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.58..-119.32, y 95.98..100.58. 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.

      • 'SFP_DETECTN3' on BOTTOM_LAYER at (-119.45, 98.49), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS3' on BOTTOM_LAYER at (-119.45, 98.85), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_3' on BOTTOM_LAYER at (-119.45, 99.56), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_3' on BOTTOM_LAYER at (-119.45, 99.2), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.58%2C95.98%7C-119.32%2C100.58 serves each with its own layer_image region; all of them lie in x -120.58..-118.32, y 97.49..100.56 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:MIDPLANE_2:hole -119.59,10.69|-119.34,15.294 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_LED1_8 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_TXFAULT8' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 12.28). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 10.69..15.29. 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.

      • 'SFP_I2C_1.SDA' on BOTTOM_LAYER at (-119.46, 13.5), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LED1_8' on BOTTOM_LAYER at (-119.46, 13.86), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE8' on BOTTOM_LAYER at (-119.46, 11.93), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT8' on BOTTOM_LAYER at (-119.46, 12.28), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C10.69%7C-119.34%2C15.29 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 10.93..14.86 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.

    • gap:MIDPLANE_2:hole -119.59,19.6|-119.34,24.24 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN7 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS0_7' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 22.32). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 19.6..24.2. 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.

      • 'SFP_DETECTN7' on BOTTOM_LAYER at (-119.46, 21.25), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS7' on BOTTOM_LAYER at (-119.46, 21.61), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_7' on BOTTOM_LAYER at (-119.46, 22.32), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_7' on BOTTOM_LAYER at (-119.46, 21.97), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C19.6%7C-119.34%2C24.2 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 20.25..23.32 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.

    • gap:MIDPLANE_2:hole -119.59,24.94|-119.34,29.544 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_LED1_7 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_7' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 28.11). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 24.94..29.54. 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.

      • 'SFP_I2C_2.SDA' on BOTTOM_LAYER at (-119.46, 27.75), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LED1_7' on BOTTOM_LAYER at (-119.46, 28.11), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE7' on BOTTOM_LAYER at (-119.46, 26.18), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT7' on BOTTOM_LAYER at (-119.46, 26.53), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C24.94%7C-119.34%2C29.54 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 25.18..29.11 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.59,33.85|-119.34,38.454 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN6 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS1_6' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 36.1). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 33.85..38.45. 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.

      • 'SFP_DETECTN6' on BOTTOM_LAYER at (-119.46, 35.39), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS6' on BOTTOM_LAYER at (-119.46, 35.74), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_6' on BOTTOM_LAYER at (-119.46, 36.45), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_6' on BOTTOM_LAYER at (-119.46, 36.1), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C33.85%7C-119.34%2C38.45 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 34.39..37.45 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.59,39.19|-119.34,43.794 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_LED1_6 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_6' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 42.36). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 39.19..43.79. 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.

      • 'SFP_I2C_3.SDA' on BOTTOM_LAYER at (-119.46, 42), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LED1_6' on BOTTOM_LAYER at (-119.46, 42.36), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE6' on BOTTOM_LAYER at (-119.46, 40.43), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT6' on BOTTOM_LAYER at (-119.46, 40.78), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C39.19%7C-119.34%2C43.79 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 39.43..43.36 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:MIDPLANE_2:hole -119.59,48.1|-119.34,52.74 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN5 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_RS0_5' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 50.82). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 48.1..52.7. 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.

      • 'SFP_DETECTN5' on BOTTOM_LAYER at (-119.46, 49.75), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS5' on BOTTOM_LAYER at (-119.46, 50.11), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_5' on BOTTOM_LAYER at (-119.46, 50.82), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_5' on BOTTOM_LAYER at (-119.46, 50.47), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C48.1%7C-119.34%2C52.7 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 48.75..51.82 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:MIDPLANE_2:hole -119.59,5.35|-119.34,9.954 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_DETECTN8 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_DETECTN8' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 7). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 5.35..9.95. 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.

      • 'SFP_DETECTN8' on BOTTOM_LAYER at (-119.46, 7), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LOS8' on BOTTOM_LAYER at (-119.46, 7.36), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS0_8' on BOTTOM_LAYER at (-119.46, 8.07), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_RS1_8' on BOTTOM_LAYER at (-119.46, 7.72), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C5.35%7C-119.34%2C9.95 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 6..9.07 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.

    • gap:MIDPLANE_2:hole -119.59,53.44|-119.34,58.044 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.25 mm (fastest: SFP_LED1_5 at 100 ps 10-90 %, ASSUMED)

      Along 'SFP_LED1_5' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.46, 56.61). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.59..-119.34, y 53.44..58.04. 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.

      • 'SFP_I2C_4.SDA' on BOTTOM_LAYER at (-119.46, 56.25), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_LED1_5' on BOTTOM_LAYER at (-119.46, 56.61), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXDISABLE5' on BOTTOM_LAYER at (-119.46, 54.68), 0.25 mm; other side: none: the signal layer is outermost
      • 'SFP_TXFAULT5' on BOTTOM_LAYER at (-119.46, 55.03), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.59%2C53.44%7C-119.34%2C58.04 serves each with its own layer_image region; all of them lie in x -120.59..-118.34, y 53.68..57.61 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -129.36,65.87|-110.86,66.13'NetC5_1' crosses one gap in MIDPLANE_2's negative-plane field at 1 place, the longest 0.25 mm (fastest: NetC5_1 at 100 ps 10-90 %, ASSUMED)

      Along 'NetC5_1' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.25 mm under the track at (-119.8, 66). 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 -129.36..-110.86, y 65.87..66.13. 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.

      • 'NetC5_1' on BOTTOM_LAYER at (-119.8, 66), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-129.36%2C65.87%7C-110.86%2C66.13 serves each with its own layer_image region; all of them lie in x -120.8..-118.8, y 64.87..67.13 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.

  5. SinglePointNetmajorOpen

    SinglePointNet: 3 findings (2 nets, 1 pin)×3

    3 checked by the review by Claude walker (cern session) — 2 different reasons

    At IC12.5, NetC5_1, NetIC11_1

    Each finding, with its own headline
    • IC12.5IC12.5 (A2) is an IC input left unconnected

      Pin IC12.5 (A2) 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.

    • NetC5_1'NetC5_1' continues through C5 to J4.1, a board connector contact

      Net 'NetC5_1' is not single-point: it continues through the fitted series part(s) C5 to J4.1, which is a board CONNECTOR contact. The node (NetC5_1, NetC5_2) enters the board there and terminates at no device on this board. ABSTAINED, not benign: whether a dead-ended host contact is acceptable depends on what the MATING side drives on it, which is not visible from this board. Ingest the mating board, or record the interface contract for that contact, to settle it.

    • NetIC11_1'NetIC11_1' continues through R391 to J2.C1, a board connector contact

      Net 'NetIC11_1' is not single-point: it continues through the fitted series part(s) R391 to J2.C1, which is a board CONNECTOR contact. The node (NetIC11_1, RTM_IO8) enters the board there and terminates at no device on this board. ABSTAINED, not benign: whether a dead-ended host contact is acceptable depends on what the MATING side drives on it, which is not visible from this board. Ingest the mating board, or record the interface contract for that contact, to settle it.

  6. observationOBS-301minorOpen

    SFP control shift registers (IC3/IC4/IC6/IC8/IC10) drive TX_DISABLE, RS0/RS1 and all 16 SFP LEDs from undefined data at power-up: OE and STROBE default active

    The five 74HC4094 SFP control shift registers (IC3, IC4, IC6, IC8, IC10) drive TX_DISABLE, RS0/RS1 and all 16 SFP LEDs from undefined data at power-up, because OE and STROBE default active when the host is not yet driving.

    The control path

    The OE (pin 15) and STROBE (pin 1) of all five registers come from inverters on IC1, whose inputs are pulled to GND by 100 R.

    ``mermaid flowchart LR A["RTM_IO2 REG_OEn<br/>R5 100 R to GND"] --> B["IC1 5A pin 11<br/>5Y pin 10"] B -- "NetIC1_10, high" --> R["IC3 IC4 IC6 IC8 IC10<br/>OE pin 15, STROBE pin 1"] C["RTM_IO4 REG_STRn<br/>R2 100 R to GND"] --> D["IC1 4A pin 9<br/>4Y pin 8"] D -- "NetIC1_8, high" --> R ``

    With the host not yet driving (AMC FPGA unconfigured, or no AMC), REG_OEn and REG_STRn read low, so OE and STROBE are both high.

    What the registers drive

    • SFP_TXDISABLE1-8 (to J5x pin 3), e.g. IC3.4 = SFP_TXDISABLE1 to J5H.3
    • SFP_RS0/RS1_1-8
    • BSS138 gates of LD1A-H and LD2A-H, e.g. IC8.11 = SFP_LED1_8 to T2A (LD1A); IC10.4-7/11-14 = SFP_LED2_1..8 to the T1x gates

    Why it matters

    Per the 74HC4094 datasheet (p.10), the storage register is loaded from the shift register while STROBE is high and appears on the outputs while OE is high. The shift register has no reset input (p.3), so the outputs present power-up content until the host shifts and strobes a word: TX_DISABLE, rate select and the LED gates are in an undefined state at power-up.

    The active-low naming and the pull-downs make "outputs enabled, latch transparent" the default. That is the opposite of the usual safe default, where outputs stay high-Z until firmware loads a known word, and the modules' own TX_DISABLE pull-ups and the 10 k gate pull-downs R10x/R12x would then give TX disabled and LEDs off.

    Class-1 SFP lasers make this a determinism issue more than a safety one, but a random rate select or TX enable before firmware is worth the designer confirming as intended.

    Suggested fix: a pull-up on REG_OEn (or removing the inverter on OE) would give a defined state.

  7. observationOBS-303minorOpen

    Temperature sensor IC12 (LM75A) address pin A2 (pin 5) is left floating, so its I2C address is undefined (0x48 or 0x4C)

    The address pin A2 (pin 5) of the LM75A temperature sensor IC12 is left floating, so its I2C address is undefined: 0x48 or 0x4C depending on how the CMOS input settles.

    Where it is

    PartA2 (pin 5)A1 (pin 6)A0 (pin 7)AddressSource
    IC12 LM75AIMunconnected ($NONE$, stub on sheet 15)GNDGND0x48 or 0x4Cnetlist; schematic sheet 15
    IC13 LM75AGNDGNDP3V3MP0x49netlist
    IC14 PCA9554GNDGNDGND0x20netlist
    IC15 24LC64GNDGNDGND0x50netlist

    All four share the management I2C bus (SDA NetIC12_1 / SCL NetIC12_2 to J6 D1/D2).

    The LM75A datasheet defines A2-A0 as user-set address inputs, to be tied to ground (0) or +VS (1) (p.3); the slave address is 1001 A2 A1 A0 (p.11). The sheet note "LM75 A6-A3 = 1001" shows the designer intended fixed addresses.

    Why it matters

    • There is no collision either way, but firmware cannot know which address IC12 has, and it may change with noise or temperature.
    • A floating input can also draw extra supply current.

    Fix: one connection. A2 to GND gives 0x48, matching the pattern of IC13, which differs only in A0.

  8. observationOBS-311minorOpen

    P2V5 converter current limit is programmed outside the TPS53126's specified range: R54 1k3 on TRIP1 gives about 13 mV against a 30-300 mV setting range

    The P2V5 converter's current limit is programmed outside the TPS53126's specified range: R54 = 1k3 on TRIP1 of IC16 gives about 13 mV against a 30-300 mV setting range, so the cycle-by-cycle limit on P2V5 is not guaranteed.

    The numbers

    IC16 TPS53126 channel 1 produces P2V5 (VO1 pin 1, VFB1 pin 2 on R195 6k8 / R248 3k0 = 2.48 V).

    QuantityValueSource
    TRIP1 (pin 18) to GNDR54 = 1k3netlist
    TRIP source current10 uA (8.5-11.5 uA)TPS53126 datasheet, p.6
    VTRIP on channel 11.3 k x 10 uA = 13 mV (11-15 mV)derived
    Specified setting range VTRIP(TRIPx-GND)30 mV to 300 mVTPS53126 datasheet, p.6
    OCP compensation offset (at VTRIP = 60 mV)-15..+15 mV at 25 C; -20..+20 mV over temperatureTPS53126 datasheet, p.6
    Channel 2 (P3V3), TRIP2 (pin 13)R246 5k1, so 51 mV, inside the rangenetlist; derived

    At 13 mV the offset is as large as the setting. Channel 2 sits inside the range, which suggests R54 is not a deliberate choice.

    Why it matters

    On this build the only load on P2V5 (through L20 to P2V5_CLEAN) is IC11 CDCLVD1212, which is marked not mounted (BOM "nm (CDCLVD1212RHAT)"). So the practical consequence today is small: an unloaded rail with an ill-defined limit.

    If IC11 is fitted, or P2V5 is used elsewhere, the limit may sit anywhere from near zero (nuisance limiting, failure to start with 120 uF to charge) to several amps.

    Fix: a TRIP1 resistor of 3 k or more (>= 30 mV) would put the setting back inside the specified range.

  9. observationOBS-313minorOpen

    Schematic title blocks carry no revision, designer, checker or document number on any of the 15 sheets

    The schematic title blocks carry no revision, designer, checker or document number on any of the 15 sheets, so the drawing cannot be tied by revision to the board, whose copper reads "MK RTM SFP v1.0".

    What the title blocks hold

    Sheets: pages 1-15, Clock Distribution, I2C Switch, PWR, RTM Connectors, RTM SFP TOP, Serial Registers, eight SFP channel sheets, USER IO IPMI.

    • Empty on every sheet: Rev, Designer, Drawn by, Check.by, Document.
    • Filled: project name "uTCA RTM 8 SFP+", sheet title, file name ("<name>.SchDoc"), sheet number ("x of 8"), a print date of 2026-09-22, "Creotech Instruments S.A.", Size A3.

    The 2026-09-22 print date is the PDF generation date, not a design-change date.

    Why it matters

    For an open-hardware release that other people build and modify, a schematic with no revision or author cannot be matched to a PCB/BOM revision or traced back to its source.

    Fix: fill Rev (matching "v1.0" on the board) and Designer/Document on the SchDoc templates.

  10. observationOBS-314minorOpen

    Both sides DO carry fiducials: FTG1-FTG8 are fiducial footprints the FiducialMissing rule does not recognise (refdes FTG, BOM text 'Use only in PCB')

    Both sides do carry fiducials: FTG1-FTG8 are fiducial footprints that the FiducialMissing rule does not recognise, because it matches only refdes FID* or part name FIDUCIAL. The two FiducialMissing rows can be closed on this evidence.

    The fiducials

    RefdesSidePosition (mm)PackageSource
    FTG2top(-149.0, 141.0)FIDUCIAL_TOP_C100-200placement
    FTG3top(-146.5, 0.5)FIDUCIAL_TOP_S100-200placement
    FTG5top(-0.5, 140.1)FIDUCIAL_TOP_S100-200placement
    FTG6top(0, 0)FIDUCIAL_TOP_C100-200placement
    FTG1bottom(-146.5, 0.5)fiducialplacement
    FTG4bottom(0, 0)fiducialplacement
    FTG7bottom(-0.5, 140.1)fiducialplacement
    FTG8bottom(-149.0, 141.0)fiducialplacement
    • They sit at three corners plus a fourth point on each side, and are visible as copper marks on the top-layer render.
    • They are correctly DNP in the BOM ("Use only in PCB", dnp=true): nothing is assembled on them.
    • The rule misses the FTG refdes; the package name does say FIDUCIAL.

    Local fiducials

    No local fiducials are present at the 0.5 mm-pitch QFNs (IC16, IC39), which most assembly houses do not require at that pitch. Whether they are wanted is the assembler's call.

  11. observationOBS-320minorOpen

    Only ICs carry reference designators in silk: the ~290 passives, LEDs, FETs, connectors and inductors (e.g. R1A, LD1A, T2A, J5A-H, L1, F1) have none

    Only the ICs carry reference designators in silk. The roughly 290 passives, LEDs, FETs, connectors and inductors (e.g. R1A, LD1A, T2A, J5A-J5H, L1, F1) have none, and no assembly drawing is in the released package on file.

    What is labelled

    The bottom overlay has 828 features for 305 bottom-side parts.

    • Bottom, labelled: IC1-IC10, IC12, IC13, IC16, IC39. Nothing else.
    • Bottom, no text: the LD1x/LD2x LEDs, their R1x/R11x 330 R, the T1x/T2x BSS138s, R10x/R12x, the 0402/0805 decoupling, T5/T6, L1/L2 and F1.
    • Top, labelled: IC11, IC14, IC15, IC48.
    • Top, no text: J5A-J5H, Cage1/2, J3/J4, J12/J13, PB1 and LED17.

    Source: layer render, bottom paste+silk at 11-80 px/mm (whole board, regions (-157..-125, 0..35) and LD1A), and top paste+silk region (-130..-105, 0..25) at 40 px/mm.

    Pin-1/polarity marks are present, so assembly orientation is not affected.

    Why it matters

    Omitting passive designators is a legitimate density choice when an assembly drawing is supplied, but none is in the released package on file. On an open-hardware board that others assemble, rework and debug, with eight identical SFP channels and letter-suffixed designators, a missing designator makes it hard to find, for example, R12C or the LED of channel 5.

    Worth the designer confirming an assembly drawing exists, or adding designators where space allows.

  12. ConnectorDomainMixedminorOpen

    J6 takes 2 different supplies off the board on one connector: NetF1_1 (12.0 V), P3V3MP (3.3 V)

    J6 carries 2 different supplies out of the board on one connector: NetF1_1 (12.0 V) on A1/A2/B1/B2, P3V3MP (3.3 V) on C2. Declare the intent. Two rails on one connector is ordinary in a docking or power-distribution interface and a hazard in a signal one, and the difference is not visible from the netlist — but note that adjacent contacts at different voltages are exactly what a mirrored, reversed or off-by-one cable shorts together, so the cost of getting the mating wrong is a rail collision rather than a dead signal. If this is a standard connector, declare its family (connector_family_declare) and file the standard's contact map: supplies on the contacts it assigns to power are then not reported.

    At J6

  13. ReferencePlaneDiscontinuityminorOpen

    ReferencePlaneDiscontinuity: 11 boards×11

    11 checked by the review by Claude walker W4 (Cern session) — 9 different reasons

    At gap:MIDPLANE_2:hole -119.08,109.73|-111.03,109.99, gap:MIDPLANE_2:hole -119.08,123.98|-111.03,124.24, gap:MIDPLANE_2:hole -119.08,81.23|-111.03,81.49, gap:MIDPLANE_2:hole -119.08,95.48|-111.03,95.74, gap:MIDPLANE_2:hole -119.18,73.34|-114.39,78.13, gap:MIDPLANE_2:hole -141.1,77.87|-118.92,78.13 and 5 more

    Each finding, with its own headline
    • gap:MIDPLANE_2:hole -119.08,109.73|-111.03,109.992 nets cross one gap in MIDPLANE_2's negative-plane field at 2 places, the longest 0.36 mm (fastest: SFP_I2C_7.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_7.SCL' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-115.65, 109.86). 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.08..-111.03, y 109.73..109.99. 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.

      • 'SFP_I2C_7.SCL' on BOTTOM_LAYER at (-115.65, 109.86), 0.36 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_7.SDA' on BOTTOM_LAYER at (-115.12, 109.86), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.08%2C109.73%7C-111.03%2C109.99 serves each with its own layer_image region; all of them lie in x -116.78..-113.99, y 108.73..110.99 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.08,123.98|-111.03,124.242 nets cross one gap in MIDPLANE_2's negative-plane field at 2 places, the longest 0.36 mm (fastest: SFP_I2C_8.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_8.SCL' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-117.95, 124.11). 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.08..-111.03, y 123.98..124.24. 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.

      • 'SFP_I2C_8.SCL' on BOTTOM_LAYER at (-117.95, 124.11), 0.36 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_8.SDA' on BOTTOM_LAYER at (-116.97, 124.11), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.08%2C123.98%7C-111.03%2C124.24 serves each with its own layer_image region; all of them lie in x -119.08..-115.84, y 122.98..125.24 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:MIDPLANE_2:hole -119.08,81.23|-111.03,81.492 nets cross one gap in MIDPLANE_2's negative-plane field at 2 places, the longest 5.3 mm (fastest: SFP_I2C_5.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_5.SCL' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 5.3 mm under the track at (-113.59, 81.24). 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. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 4.95 mm). 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.08..-111.03, y 81.23..81.49. 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.

      • 'SFP_I2C_5.SCL' on BOTTOM_LAYER at (-113.59, 81.24), 5.3 mm, running ALONG the bare copper for 4.95 mm, not across it; other side: none: the signal layer is outermost
      • 'SFP_I2C_5.SDA' on BOTTOM_LAYER at (-116.97, 81.36), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.08%2C81.23%7C-111.03%2C81.49 serves each with its own layer_image region; all of them lie in x -118.1..-110.12, y 80.23..82.49 mm (board frame, Y up).

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

    • gap:MIDPLANE_2:hole -119.08,95.48|-111.03,95.742 nets cross one gap in MIDPLANE_2's negative-plane field at 2 places, the longest 0.36 mm (fastest: SFP_I2C_6.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_6.SDA' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-113.39, 95.61). 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.08..-111.03, y 95.48..95.74. 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.

      • 'SFP_I2C_6.SCL' on BOTTOM_LAYER at (-112.88, 95.61), 0.36 mm; other side: none: the signal layer is outermost
      • 'SFP_I2C_6.SDA' on BOTTOM_LAYER at (-113.39, 95.61), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.08%2C95.48%7C-111.03%2C95.74 serves each with its own layer_image region; all of them lie in x -114.51..-111.76, y 94.48..96.74 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:MIDPLANE_2:hole -119.18,73.34|-114.39,78.134 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.36 mm (fastest: RTM_IO10 at 2 ns 10-90 %, suggested by its name)

      Along 'RTM_IO10' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-118.16, 77.11). 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. 4 places on 4 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -119.18..-114.39, y 73.34..78.13. 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.

      • 'RTM_IO10' on BOTTOM_LAYER at (-118.16, 77.11), 0.36 mm; other side: none: the signal layer is outermost
      • 'RTM_IO11' on BOTTOM_LAYER at (-118.87, 77.82), 0.36 mm; other side: none: the signal layer is outermost
      • 'RTM_IO9' on BOTTOM_LAYER at (-118.52, 77.47), 0.36 mm; other side: none: the signal layer is outermost
      • 'RTM_IO8' on BOTTOM_LAYER at (-117.27, 76.22), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-119.18%2C73.34%7C-114.39%2C78.13 serves each with its own layer_image region; all of them lie in x -120.05..-116.18, y 75.13..78.82 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:MIDPLANE_2:hole -141.1,77.87|-118.92,78.132 nets cross one gap in MIDPLANE_2's negative-plane field at 4 places, the longest 0.36 mm (fastest: RTM_IO11 at 2 ns 10-90 %, suggested by its name)

      Along 'RTM_IO11' on BOTTOM_LAYER, MIDPLANE_2 is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (-133.61, 78). 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. 4 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x -141.1..-118.92, y 77.87..78.13. 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.

      • 'RTM_IO11' on BOTTOM_LAYER at (-133.61, 78), 0.36 mm; other side: none: the signal layer is outermost
      • 'RTM_IO11' on BOTTOM_LAYER at (-140.49, 78), 0.36 mm; other side: none: the signal layer is outermost
      • 'RTM_IO9' on BOTTOM_LAYER at (-134.11, 78), 0.36 mm; other side: none: the signal layer is outermost
      • 'RTM_IO9' on BOTTOM_LAYER at (-139.59, 78), 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/31/reference-planes?group=gap%3AMIDPLANE_2%3Ahole%20-141.1%2C77.87%7C-118.92%2C78.13 serves each with its own layer_image region; all of them lie in x -141.62..-132.48, y 76.87..79.13 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:NetC219E_1@MIDPLANE_2:hole -120.44,82.8|-119.39,83.85'SFP_I2C_5.SCL' crosses one void in the NetC219E_1 plane on MIDPLANE_2 at 1 place, the longest 0.82 mm (fastest: SFP_I2C_5.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_5.SCL' on BOTTOM_LAYER, the NetC219E_1 plane on MIDPLANE_2 has 0.82 mm under the track with no copper, at (-119.72, 83.51), 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 -120.44..-119.39, y 82.8..83.85. 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.

      • 'SFP_I2C_5.SCL' on BOTTOM_LAYER at (-119.72, 83.51), 0.82 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/31/reference-planes?group=void%3ANetC219E_1%40MIDPLANE_2%3Ahole%20-120.44%2C82.8%7C-119.39%2C83.85 serves each with its own layer_image region; all of them lie in x -120.91..-118.32, y 82.32..84.88 mm (board frame, Y up).

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

    • void:NetC219F_1@MIDPLANE_2:hole -120.44,97.05|-119.39,98.1'SFP_I2C_6.SCL' crosses one void in the NetC219F_1 plane on MIDPLANE_2 at 1 place, the longest 0.82 mm (fastest: SFP_I2C_6.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_6.SCL' on BOTTOM_LAYER, the NetC219F_1 plane on MIDPLANE_2 has 0.82 mm under the track with no copper, at (-119.72, 97.76), 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 -120.44..-119.39, y 97.05..98.1. 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.

      • 'SFP_I2C_6.SCL' on BOTTOM_LAYER at (-119.72, 97.76), 0.82 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/31/reference-planes?group=void%3ANetC219F_1%40MIDPLANE_2%3Ahole%20-120.44%2C97.05%7C-119.39%2C98.1 serves each with its own layer_image region; all of them lie in x -120.91..-118.32, y 96.57..99.14 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:NetC219G_1@MIDPLANE_2:hole -120.44,111.3|-119.39,112.35'SFP_I2C_7.SCL' crosses one void in the NetC219G_1 plane on MIDPLANE_2 at 1 place, the longest 0.82 mm (fastest: SFP_I2C_7.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_7.SCL' on BOTTOM_LAYER, the NetC219G_1 plane on MIDPLANE_2 has 0.82 mm under the track with no copper, at (-119.72, 111.63), 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 -120.44..-119.39, y 111.3..112.35. 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.

      • 'SFP_I2C_7.SCL' on BOTTOM_LAYER at (-119.72, 111.63), 0.82 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/31/reference-planes?group=void%3ANetC219G_1%40MIDPLANE_2%3Ahole%20-120.44%2C111.3%7C-119.39%2C112.35 serves each with its own layer_image region; all of them lie in x -120.91..-118.32, y 110.27..112.82 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:NetC219H_1@MIDPLANE_2:hole -120.44,125.55|-119.39,126.6'SFP_I2C_8.SCL' crosses one void in the NetC219H_1 plane on MIDPLANE_2 at 1 place, the longest 0.83 mm (fastest: SFP_I2C_8.SCL at 10 ns 10-90 %, suggested by its name)

      Along 'SFP_I2C_8.SCL' on BOTTOM_LAYER, the NetC219H_1 plane on MIDPLANE_2 has 0.83 mm under the track with no copper, at (-119.72, 125.88), running to the end of the track. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. What cleared it: the hole at x -120.44..-119.39, y 125.55..126.6. 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.

      • 'SFP_I2C_8.SCL' on BOTTOM_LAYER at (-119.72, 125.88), 0.83 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/31/reference-planes?group=void%3ANetC219H_1%40MIDPLANE_2%3Ahole%20-120.44%2C125.55%7C-119.39%2C126.6 serves each with its own layer_image region; all of them lie in x -120.91..-118.32, y 124.48..127.07 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.

    • void:P3V3_CLEAN@MIDPLANE_2:hole -142.53,66.82|-142.27,76.7'RTM_IO11' crosses one void in the P3V3_CLEAN plane on MIDPLANE_2 at 1 place, the longest 7.17 mm (fastest: RTM_IO11 at 2 ns 10-90 %, suggested by its name)

      Along 'RTM_IO11' on BOTTOM_LAYER, the P3V3_CLEAN plane on MIDPLANE_2 has 7.17 mm under the track with no copper, at (-142.4, 72.69), 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 6.81 mm). What cleared it: the hole at x -142.53..-142.27, y 66.82..76.7. 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.

      • 'RTM_IO11' on BOTTOM_LAYER at (-142.4, 72.69), 7.17 mm, running ALONG the bare copper for 6.81 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/31/reference-planes?group=void%3AP3V3_CLEAN%40MIDPLANE_2%3Ahole%20-142.53%2C66.82%7C-142.27%2C76.7 serves each with its own layer_image region; all of them lie in x -143.4..-141.27, y 68.15..77.22 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
24LC64-I/MSMICROCHIP TECHNOLOGY64K I2C Serial EEPROMIC15
570BBC000121DGSILICON LABSAny-Rate I2C Programmable XO Oscillator Si570 SerieOSC1
74HC04DNXP SEMICONDUCTORSHex InverterIC1
FIN1101K8XLVDS TO LVDS repeater3.3VIC48
HDT0001C&KPush-Button Switch SPDTPB1
LED0603-P-RW120 or LYN971-ZLED 0603 redLED17
MAX7357ETG+8-channel I2C-bus multiplexer with reset, INT and timeoutIC39
PCA9554DB8-bit I2C-bus SMBus I/O port with interruptIC14
SMD150F-2TYCO ELECTRONICSFuseF1
TPS53126RGETTEXAS INSTRUMENTSDual Synchronous Step-Down Controller For Low Voltage Power RailsIC16
2007132-1TYCO ELECTRONICSSFP+ 1x4 Cage (46 Pins)Cage1, Cage2
85_SMA-50-0-101/111_NEHUBER+SUHNERJ3, J4
IRF9910PbfINTERNATIONAL RECTIFIERDual SO-8 MOSFET For POL CconverterT5, T6
LM75AIMI2C temperature sensorIC12, IC13
U.FL-R-SMT-1(10)HIROSE (HRS)J12, J13
74HC165DNXP SEMICONDUCTORS8-Bit Parallel-In/Serial Out Shift RegisterIC2, IC5, IC7, IC9
74HC4094Dshift registerIC10, IC3, IC4, IC6, IC8
1888247-1TYCO ELECTRONICS20 Position SFP+ Connector (Small Form-factor Pluggable)J5A, J5B, J5C, J5D, J5E, J5F, J5G, J5H
HSMC-C120AVAGO TECHNOLOGIESLED Red Avago C120LD2A, LD2B, LD2C, LD2D, LD2E, LD2F, LD2G, LD2H
HSME-C120AVAGO TECHNOLOGIESLED Green Avago C120LD1A, LD1B, LD1C, LD1D, LD1E, LD1F, LD1G, LD1H
BSS138LT1GON SEMICONDUCTORN-CHANNEL MOSFET WITH DIODET1A, T1B, T1C, T1D, T1E, T1F and 10 more
R0603_3K_1%_0.1W_100PPMGENERICResistor - 1%R248
CC0603_100NF_50V_10%_X7RGENERICNon-polarised capacitorC11, C280
CC0603_1uF_16V_10%_X7RGENERICNon-polarised capacitorC386, C84
IHLP2525CZER1R5M01VISHAY DALEInductor With Magnetic coreL1, L2
MPZ1608R391ATDKInductorL19, L20
R0603_1K3_1%_0.1W_100PPMGENERICResistor - 1%R18, R54
R0603_4K3_1%_0.1W_100PPMGENERICResistor - 1%R391, R53
R0603_6K8_1%_0.1W_100PPMGENERICResistor - 1%R17, R195
CC1206_10uF_16V_10%_X7RGENERICNon-polarised capacitorC10, C12, C14, C9
R0402_49R9_1%_0.0625W_100PPMGENERICResistor - 1%R312, R313, R314, R315
R1206_1M_1%_0.25W_100PPMGENERICResistor - 1%R19, R20, R21, R22
R0603_100R_1%_0.1W_100PPMGENERICResistor - 1%R13, R2, R3, R4, R5, R6
R0603_0R_JUMPERGENERICResistor - 1%R107A, R107B, R107C, R107D, R107E, R107F, R107G, R107H
R0603_5K1_1%_0.1W_100PPMGENERICResistor - 1%R23, R24, R246, R25, R26, R27 and 4 more
CC1206_22UF_10V_10%_X7RGENERICNon-polarised capacitorC15, C176, C179, C180, C181, C182 and 5 more
CC0201_100NF_6.3V_10%_X5RGENERICNon-polarised capacitorC391, C392, C393, C394, C395, C396 and 6 more
BLM21PG221SN1DMURATAInductorL11A, L11B, L11C, L11D, L11E, L11F and 10 more
R0603_330R_1%_0.1W_100PPMGENERICResistor - 1%R11A, R11B, R11C, R11D, R11E, R11F and 13 more
CC0805_10UF_10V_10%_X7RGENERICNon-polarised capacitorC13A, C13B, C13C, C13D, C13E, C13F and 20 more

About this design

Design
UTCA RTM 8 SFP — OHWR0029 R1
Designed in
Altium Designer
Board
4 copper layers, 182.5 × 148.5 mm, 363 components, 312 nets
Licence
CERN-OHL-1.1
Source
https://gitlab.com/ohwr/project/utca-rtm-8-sfp at commit 39c351266e93
Attribution
CERN Open Hardware Repository, project ohwr/project/utca-rtm-8-sfp (commit 39c351266e93bbd1d63f426c85dfa18e61e17c00, 2025-04-09). Licence: CERN-OHL-1.1. Source: https://gitlab.com/ohwr/project/utca-rtm-8-sfp. 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)