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

WRS V2 AMC SFP

OHWR0015 · revision R1

WRS V2 AMC SFP (OHWR0015 R1) is an open-hardware circuit board from the CERN Open Hardware Repository, designed in Altium Designer: a 4-layer board of 180.6 × 73.5 mm with 49 components and 41 nets, published under CERN-OHL-1.2.

review 90% walked · 9 findings: 1 blocker · 6 major · 2 minor (12 items)

Open in the interactive viewerRead the full review

WRS V2 AMC SFP (OHWR0015 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 9 findings: 1 blocker · 6 major · 2 minor (12 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. CapVoltageDeratingblockerOpen

    C14 is rated 10 V on a 12 V rail

    C14: rated 10 V on a 12 V rail — UNDER-rated.

    At C14

  2. observationOBS-259majorOpen

    AMC SFP buck IC2 LM2574 uses a standard-recovery rectifier (D4 MRA4006T3G) as its catch diode

    The catch diode D4 of the AMC SFP buck IC2 (LM2574) is an MRA4006T3G, a standard-recovery rectifier, where the LM2574 datasheet calls for a Schottky or fast-recovery type.

    Where it is

    D4 is the freewheeling diode of the 52 kHz LM2574 buck that makes +3V3 from the AMC 12 V payload power: cathode (pin 2) on the switch node NetD4_2 (IC2 pin 12 OUTPUT, L3 330 uH), anode (pin 1) on GND.

    What the datasheets say

    PartFactSource
    MRA4006T3G1 A standard-recovery power rectifier, intended for line-frequency rectificationMRA4006T3G datasheet, p.1
    LM2574 catch diodeSchottky, or fast-recovery high-efficiency types; standard 60 Hz rectifiers are not suitableLM2574 datasheet, p.17

    Why it matters

    A standard-recovery diode conducts in reverse for microseconds at every switch turn-on, adding switching loss in D4 and the LM2574 switch and raising noise and temperature.

    Note: the page-12 evidence row attached at creation (60-Hz ripple on ON/OFF) is not relevant to this finding; the page-17 row is the citation.

  3. observationOBS-260majorOpen

    The published AMC SFP BOM mounts both halves of the R9/R10 '0*' option pair, which shorts +3V3 to GND through SFP pin 7

    The published AMC SFP BOM mounts both halves of the R9/R10 '0*' option pair, which shorts +3V3 to GND through SFP pin 7 (NetCN2_7).

    Where it is

    PartPin 1Pin 2
    R9, 0 OhmGNDNetCN2_7 (CN2 pin 7, the SFP cage's rate-select contact)
    R10, 0 Ohm+3V3NetCN2_7

    The schematic marks both '0*', the same assembly-option convention as on the MCH boards, and only one should be fitted.

    The BOM row

    BOM CSV row 'R9, R10': Name '0*', Value 0, LibRef R0603_0R_JUMPER, Mounted 'Yes'. An assembler following it shorts +3V3 to GND.

    For this review both are treated as not fitted.

  4. PartLifecyclemajorOpen

    ATmega8L-8AC on IC1: the distributor catalogue reports "Obsolete" for Microchip Technology's ATMEGA8L-8AC

    ATmega8L-8AC on IC1: the distributor catalogue reports "Obsolete" for Microchip Technology's ATMEGA8L-8AC — 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/ATmega8L-8AC/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 (2007-12-31, 6,894 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.

    At IC1

  5. FiducialMissingmajorOpen

    FiducialMissing: 2 boards×2

    Checked by the review, stands · Claude walker (cern session) — No fiducial component on this board: the BOM has no FID*/FTG*/fiducial row and the rule found no fiducial-attribute pad on either side. Not refuted; whether fiducials are carried on the production panel instead is for the owner.

    At bottom, top

    Each finding, with its own headline
    • bottomThe bottom side carries 22 SMT component(s) and no fiducial was found on it

      The bottom side carries 22 SMT component(s) but no fiducial was found on it — a placement machine has no optical reference to align to. Probe found no pad labelled .pad_usage=g_fiducial/l_fiducial (basis pad_usage); no single-pad, netless, undrilled component whose 0.5-3 mm round pad has a concentric solder-mask opening at least 1.5x its diameter with no other copper inside it (basis geometry); and no component whose refdes starts FID or whose part or package name contains FIDUCIAL (basis footprint-name). If your library names fiducials something Probe does not recognise, mark this benign — disposition_finding(rule='FiducialMissing', entity='bottom', status='benign', reason=…) — the name match only ever silences this check, it never asserts anything. If the side really has no fiducial, say so with status='confirmed': that answers the finding and keeps it gating release.

    • topThe top side carries 22 SMT component(s) and no fiducial was found on it

      The top side carries 22 SMT component(s) but no fiducial was found on it — a placement machine has no optical reference to align to. Probe found no pad labelled .pad_usage=g_fiducial/l_fiducial (basis pad_usage); no single-pad, netless, undrilled component whose 0.5-3 mm round pad has a concentric solder-mask opening at least 1.5x its diameter with no other copper inside it (basis geometry); and no component whose refdes starts FID or whose part or package name contains FIDUCIAL (basis footprint-name). If your library names fiducials something Probe does not recognise, mark this benign — disposition_finding(rule='FiducialMissing', entity='top', status='benign', reason=…) — the name match only ever silences this check, it never asserts anything. If the side really has no fiducial, say so with status='confirmed': that answers the finding and keeps it gating release.

  6. PartLifecyclemajorOpen

    LM2574M-ADJ on IC2: the distributor catalogue reports "Obsolete" for Texas Instruments's LM2574M-ADJ

    LM2574M-ADJ on IC2: the distributor catalogue reports "Obsolete" for Texas Instruments's LM2574M-ADJ — a sourcing/longevity risk to confirm against the MANUFACTURER before acting. This board's BOM names National Semiconductor, and that is one line with Texas Instruments: Texas Instruments completed its acquisition of National Semiconductor on 2011-09-23. A distributor delisting is not the same claim as the part being unavailable; check the vendor's own lifecycle/product-change notice for the status and any named successor. Re-read the source: GET /api/odb/parts/LM2574M-ADJ/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 (2024-09-11, 795 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.

    At IC2

  7. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 2 boards×2

    Checked by the review, stands · Claude walker W2 (Cern session) — Checked: the crossings and reference changes are read from the copper and stand as stated. No edge-rate or impedance requirement on file makes them a defect: most nets involved are slow control/IO, and the stackup's stated planes give a nearby return for most crossings as reported. Layout practice, left for the owner.

    At gap:SUPPLY:hole 10.49,-23.85|49.63,-23.6, gap:SUPPLY:hole 44.76,-15.09|53.39,-6.45

    Each finding, with its own headline
    • gap:SUPPLY:hole 10.49,-23.85|49.63,-23.62 nets cross one gap in SUPPLY's negative-plane field at 2 places, the longest 0.36 mm (fastest: NetCN1_3 at 100 ps 10-90 %, ASSUMED)

      Along 'UTCA_GA0' on BOTTOM_LAYER, SUPPLY is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (10.54, -23.72). 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 10.49..49.63, y -23.85..-23.6. 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.

      • 'UTCA_GA0' on BOTTOM_LAYER at (10.54, -23.72), 0.36 mm; other side: none: the signal layer is outermost
      • 'NetCN1_3' on BOTTOM_LAYER at (11.46, -23.72), 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/18/reference-planes?group=gap%3ASUPPLY%3Ahole%2010.49%2C-23.85%7C49.63%2C-23.6 serves each with its own layer_image region; all of them lie in x 9.54..12.46, y -24.98..-22.6 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:SUPPLY:hole 44.76,-15.09|53.39,-6.453 nets cross one gap in SUPPLY's negative-plane field at 3 places, the longest 0.36 mm (fastest: UTCA_GA0 at 100 ps 10-90 %, ASSUMED)

      Along 'UTCA_GA1' on BOTTOM_LAYER, SUPPLY is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (46.25, -7.95). 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 44.76..53.39, y -15.09..-6.45. 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.

      • 'UTCA_GA0' on BOTTOM_LAYER at (46.86, -8.56), 0.36 mm; other side: none: the signal layer is outermost
      • 'UTCA_GA1' on BOTTOM_LAYER at (46.25, -7.95), 0.36 mm; other side: none: the signal layer is outermost
      • 'UTCA_GA2' on BOTTOM_LAYER at (45.42, -7.11), 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/18/reference-planes?group=gap%3ASUPPLY%3Ahole%2044.76%2C-15.09%7C53.39%2C-6.45 serves each with its own layer_image region; all of them lie in x 44.24..48.04, y -9.56..-6.11 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.

  8. ConnectorDomainMixedminorOpen

    CN1 takes 2 different supplies off the board on one connector: +12V (12.0 V), +3V3_MP (3.3 V)

    CN1 carries 2 different supplies out of the board on one connector: +12V (12.0 V) on 18/2/27/42/57/72/84/9, +3V3_MP (3.3 V) on 4. 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 CN1

  9. ReferencePlaneDiscontinuityminorOpen

    ReferencePlaneDiscontinuity: 2 boards×2

    Checked by the review, stands · Claude walker W2 (Cern session) — Checked: the crossings and reference changes are read from the copper and stand as stated. No edge-rate or impedance requirement on file makes them a defect: most nets involved are slow control/IO, and the stackup's stated planes give a nearby return for most crossings as reported. Layout practice, left for the owner.

    At gap:SUPPLY:hole 44.76,6.86|49.45,11.56, gap:SUPPLY:hole 58.34,11.3|60.17,13.13

    Each finding, with its own headline
    • gap:SUPPLY:hole 44.76,6.86|49.45,11.562 nets cross one gap in SUPPLY's negative-plane field at 2 places, the longest 0.36 mm (fastest: UTCA_SCL_L at 10 ns 10-90 %, suggested by its name)

      Along 'UTCA_ENABLE' on BOTTOM_LAYER, SUPPLY is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (46.12, 8.23). 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 44.76..49.45, y 6.86..11.56. 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.

      • 'UTCA_ENABLE' on BOTTOM_LAYER at (46.12, 8.23), 0.36 mm; other side: none: the signal layer is outermost
      • 'UTCA_SCL_L' on BOTTOM_LAYER at (47.73, 9.83), 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/18/reference-planes?group=gap%3ASUPPLY%3Ahole%2044.76%2C6.86%7C49.45%2C11.56 serves each with its own layer_image region; all of them lie in x 44.94..48.91, y 7.23..10.83 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:SUPPLY:hole 58.34,11.3|60.17,13.13'UTCA_SDA_L' crosses one gap in SUPPLY's negative-plane field at 1 place, the longest 0.36 mm (fastest: UTCA_SDA_L at 10 ns 10-90 %, suggested by its name)

      Along 'UTCA_SDA_L' on BOTTOM_LAYER, SUPPLY is a negative layer (copper everywhere inside the profile), and its field is cleared for 0.36 mm under the track at (59.49, 12.45). 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 58.34..60.17, y 11.3..13.13. 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.

      • 'UTCA_SDA_L' on BOTTOM_LAYER at (59.49, 12.45), 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/18/reference-planes?group=gap%3ASUPPLY%3Ahole%2058.34%2C11.3%7C60.17%2C13.13 serves each with its own layer_image region; all of them lie in x 58.31..60.67, y 11.45..13.45 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.

Main parts

Part numberMakerDescriptionWhere
1n4148Schottky DiodeD3
ATmega8L-8AC8-Bit AVR Microcontroller with 8K Bytes of In-System Programmable Flash MemoryIC1
HLMP-1700AVAGO TECHNOLOGIESLED Red 1C 2AD1
LEDChip LED (any color)D2
LM2574M-ADJNational SemiconductorSIMPLE SWITCHER 0.5A Step-Down Voltage RegulatorIC2
MRA4006T3GSchottky DiodeD4
UTCA_AMC_TONGUE1AMC1.0 connectorCN1
uTCA_HSPush buttonS1
R0805_10K_1%_0.125W_100PPMGENERICResistor - 1%R13
R0805_16K9_0.1%_0.1W_10PPMGENERICResistor - 0.1%R12
CC0805_22PF_50V_5%_NP0GENERICNon-polarised capacitorC7, C8
R0805_330R_1%_0.125W_100PPMGENERICResistor - 1%R1, R2
10SP100MSANYOPolarised capacitorC13, C14, C15
CC1206_10uF_10V_10%_X7RGENERICNon-polarised capacitorC1, C3, C5
CC0805_100NF_50V_10%_X7RGENERICNon-polarised capacitorC10, C11, C12, C9
CC0603_100NF_50V_10%_X7RGENERICNon-polarised capacitorC16, C17, C2, C4, C6
R1206_100K_1%_0.25W_100PPMGENERICResistor - 1%R14, R15, R16, R17, R18, R19
R0603_10K_1%_0.1W_100PPMGENERICResistor - 1%R11, R3, R4, R5, R6, R7, R8

About this design

Design
WRS V2 AMC SFP — OHWR0015 R1
Designed in
Altium Designer
Board
4 copper layers, 180.6 × 73.5 mm, 49 components, 41 nets
Licence
CERN-OHL-1.2
Source
https://gitlab.com/ohwr/project/wr-switch-hw at commit 54f8a604978b
Attribution
CERN Open Hardware Repository, project ohwr/project/wr-switch-hw (commit 54f8a604978bdc65bbde5b1f4c7ec9bee3505a44, 2025-04-09). Licence: CERN-OHL-1.2. Source: https://gitlab.com/ohwr/project/wr-switch-hw. 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)