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

SVEC 3V3

OHWR0037 · revision R1

SVEC 3V3 (OHWR0037 R1) is an open-hardware circuit board from the CERN Open Hardware Repository, designed in Altium Designer: a 2-layer board of 55.5 × 23.2 mm with 20 components and 13 nets, published under CERN-OHL-1.2.

review 95% walked · 11 findings: 7 major · 4 minor (26 items)

Open in the interactive viewerRead the full review

SVEC 3V3 (OHWR0037 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 11 findings: 7 major · 4 minor (26 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-315majorOpen

    Current limit set by R4 (5k1) can fall below the 4 A the sheet promises

    The TPS53126 IC1 valley current limit set by R4 (5k1) can fall to about 3.2-3.4 A worst case, below the "3.3V, 4A supply" written on the sheet, and at the limit the controller can latch P3V3 off.

    How the limit is set

    R4 (5k1, 1 %) sits between IC1 pin 18 (TRIP1) and GND and programs the valley current limit against the low-side FET RDS(on) (TPS53126 datasheet, p.16):

    I_OCL = Vtrip / RDS(on) + (VIN - VO) x VO / (2 x L x fsw x VIN)

    Inputs

    QuantityValueSource
    Vtrip, typical10 µA x 5.1 k = 51 mVderived
    Vtrip, worst case at 25 C8.5 µA x 5.1 k = 43.4 mVderived
    fsw350 kHz (TEST2, pin 14, on GND)netlist: R4
    L11.5 µH ±30 %, Bourns SRR1260-1R5Ynetlist: L1
    VIN / VO5 V / 3.3 Vstatement
    Ripple2.14 A p-p, 1.07 A halfderived
    Low-side FETT1 half on pins 1/2/7/8: gate on DRVL1, source on GNDnetlist: T1
    IRF9910 RDS(on) at VGS = 4.5 V (what VREG5 supplies from 5 V)14.6 mΩ typ, 18.3 mΩ maxIRF9910 datasheet, p.2

    Results

    CaseArithmeticLimit
    Typical51 mV / 14.6 mΩ + 1.07 A4.56 A
    Worst case at 25 C43.4 mV / 18.3 mΩ + 1.07 A3.44 A
    Worst case, L1 at +30 %3.2 A

    The ±15 mV OCP compensation offset at 25 C (TPS53126 datasheet, p.6) lowers it further. Even the lower-RDS(on) half (11.3 mΩ max) gives only about 3.6 A with that offset.

    Why it matters

    At the limit the output droops, and below 70 % of the reference the controller latches off (UVP; 'turns off both top and bottom MOSFET drivers and latches off', TPS53126 datasheet, p.13). A 4 A load can therefore shut P3V3 down until EN or VIN is cycled.

    Suggested fix: a larger R4 (for example 8k2 to 10k, within the 30 to 300 mV range) would put the worst-case limit above 4 A.

  2. PartLifecyclemajorOpen

    ECASD40G157M016K00 on C4: the distributor catalogue reports "Obsolete" for Murata Electronics's ECASD40G157M016K00

    ECASD40G157M016K00 on C4: the distributor catalogue reports "Obsolete" for Murata Electronics's ECASD40G157M016K00 — 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/ECASD40G157M016K00/providers (which provider said it, the maker and part number it matched, its own words, and when we asked). No approved alternate is recorded for this position.

    At C4

  3. SinglePointNetmajorOpen

    IC1.13 (TRIP2) is an IC input left unconnected

    Pin IC1.13 (TRIP2) 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.

    At IC1.13

  4. PartLifecyclemajorOpen

    IRF9910Pbf on T1: the distributor catalogue reports "Discontinued at DigiKey" for Infineon Technologies's IRF9910PBF

    IRF9910Pbf on T1: 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 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/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.

    At T1

  5. ReferencePlaneDiscontinuitymajorOpen

    ReferencePlaneDiscontinuity: 15 boards×15

    e.g. “2 nets cross one void in the GND plane on TOP_LAYER at 2 places, the longest 0.25 mm (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)” · 12 different statements

    At along@TOP_LAYER:GND|P5V_VME, along@TOP_LAYER:NetC3_2|P5V_VME, refchange:GND|NetC3_2, refchange:GND|P5V_VME, split@TOP_LAYER:GND|P5V_VME, void:GND@BOTTOM_LAYER:hole 67.59,80.34|85.54,88.06 and 9 more

    Each finding, with its own headline
    • along@TOP_LAYER:GND|P5V_VME'NetIC1_22' runs ALONG the split between the GND and P5V_VME planes on TOP_LAYER at 1 place, the longest 0.51 mm; the impedance there is not said (fastest: NetIC1_22 at 100 ps 10-90 %, ASSUMED)

      On TOP_LAYER the GND and P5V_VME planes part, and 'NetIC1_22' runs down the slot between them at 1 place(s), rather than across it. This is not a crossing: the return still follows the track, in the plane on its other side and in the slot's edges, so it does not detour. What changes is the track's own impedance. With the plane gone from under it on this side for that length, its field reaches further, to the slot's edges and to the next plane beyond, so its capacitance falls and its impedance rises; every edge travelling down it is partly reflected where it enters the stretch and again where it leaves, by (Z2 - Z1)/(Z2 + Z1). The longer the stretch against the edge's rise, the more of the edge sees the full step. A crossing is generally the worse fault; a mismatch is not free. At 1 place(s) the impedance cannot be said: no plane lies on the track's other side or past the split layer, so the field along the slot has nothing to end on. No place stays within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is a warning. One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • 'NetIC1_22' on BOTTOM_LAYER at (76.49, 87.5), 0.51 mm, running ALONG the split for 0.51 mm, not across it, impedance not said: no plane lies on the track's other side or past the split layer, so the field along the slot has nothing to end on; 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/38/reference-planes?group=along%40TOP_LAYER%3AGND%7CP5V_VME serves each with its own layer_image region; all of them lie in x 75.24..77.75, y 86.5..88.5 mm (board frame, Y up).

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

    • along@TOP_LAYER:NetC3_2|P5V_VME'NetIC1_20' runs ALONG the split between the NetC3_2 and P5V_VME planes on TOP_LAYER at 1 place, the longest 2.12 mm; the impedance there is not said (fastest: NetIC1_20 at 100 ps 10-90 %, ASSUMED)

      On TOP_LAYER the NetC3_2 and P5V_VME planes part, and 'NetIC1_20' runs down the slot between them at 1 place(s), rather than across it. This is not a crossing: the return still follows the track, in the plane on its other side and in the slot's edges, so it does not detour. What changes is the track's own impedance. With the plane gone from under it on this side for that length, its field reaches further, to the slot's edges and to the next plane beyond, so its capacitance falls and its impedance rises; every edge travelling down it is partly reflected where it enters the stretch and again where it leaves, by (Z2 - Z1)/(Z2 + Z1). The longer the stretch against the edge's rise, the more of the edge sees the full step. A crossing is generally the worse fault; a mismatch is not free. At 1 place(s) the impedance cannot be said: no plane lies on the track's other side or past the split layer, so the field along the slot has nothing to end on. No place stays within 10 %, the tolerance controlled impedance is commonly built to (the stackup states none), so the finding is a warning. One decision covers the split line's slot: whether these tracks may run down it, or should be moved over solid plane.

      • 'NetIC1_20' on BOTTOM_LAYER at (83.06, 87.1), 2.48 mm, running ALONG the split for 2.12 mm, not across it, impedance not said: no plane lies on the track's other side or past the split layer, so the field along the slot has nothing to end on; 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/38/reference-planes?group=along%40TOP_LAYER%3ANetC3_2%7CP5V_VME serves each with its own layer_image region; all of them lie in x 81.25..85.58, y 85.74..88.1 mm (board frame, Y up).

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

    • refchange:GND|NetC3_2'NetIC1_20' changes reference between GND and NetC3_2 at a via, none with a reference that continues across (fastest: NetIC1_20 at 100 ps 10-90 %, ASSUMED)

      At 1 via(s), on 1 net(s), the signal leaves a layer referenced by one of GND and NetC3_2 for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 1 with none. 1 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and NetC3_2 are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • 'NetIC1_20' at (84.89, 84.86), TOP_LAYER -> BOTTOM_LAYER: GND on BOTTOM_LAYER to NetC3_2 on TOP_LAYER; pure, no reference continues across; stitching none, 19.4 mm of track past the via on BOTTOM_LAYER: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and NetC3_2 are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=refchange%3AGND%7CNetC3_2 serves each with its own layer_image region; all of them lie in x 82.89..86.89, y 82.86..86.86 mm (board frame, Y up).

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

    • refchange:GND|P5V_VME2 nets change reference between GND and P5V_VME at 2 vias, none with a reference that continues across (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      At 2 via(s), on 2 net(s), the signal leaves a layer referenced by one of GND and P5V_VME for a layer referenced by the other, and no plane references both layers. Different nets: no via can carry the return across; only decoupling between the two planes near the via can. The capacitor carrying the return across was measured at each via, at the knee of the crossing net's edge against 50 ohm assumed: 2 with none. 2 of them at an ASSUMED edge (no edge on file for the net): state one (net_timing_state) and it is re-read. One decision covers the pair: whether GND and P5V_VME are decoupled to each other near these vias. A via whose layers have more references than these two is a site of each pair it joins. None is partial: at every one, no reference continues across the via.

      • 'NetIC1_22' at (84.63, 87.4), TOP_LAYER -> BOTTOM_LAYER: GND on BOTTOM_LAYER to P5V_VME on TOP_LAYER; pure, no reference continues across; stitching none, 17.6 mm of track past the via on BOTTOM_LAYER: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and P5V_VME are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.
      • 'NetC3_2' at (83.87, 86.51), TOP_LAYER -> BOTTOM_LAYER: GND on BOTTOM_LAYER to P5V_VME on TOP_LAYER; pure, no reference continues across; stitching none, 15.0 mm of track past the via on BOTTOM_LAYER: At the 5 GHz knee of a 100 ps edge (100 ps 10-90 %, assumed: no edge is on file for the net) no capacitor carries the return; only the planes do. GND and P5V_VME are not a thin pair (at most 0.2 mm), so their own capacitance does not carry the return. To serve it, change layers across the same plane, or move the signal to a layer referenced to ground. None.

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=refchange%3AGND%7CP5V_VME serves each with its own layer_image region; all of them lie in x 81.87..86.63, y 84.51..89.4 mm (board frame, Y up).

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

    • split@TOP_LAYER:GND|P5V_VME2 nets cross the split between the GND and P5V_VME planes on TOP_LAYER at 2 places, with no plane on the track's other side (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      On TOP_LAYER the GND and P5V_VME planes part, and 2 nets cross from one to the other at 2 place(s). The return current cannot follow a track across; it detours to wherever the two planes are joined (a stitching capacitor, or the source), and the loop it opens radiates and couples. One decision covers the boundary: whether it is stitched near these crossings, or the nets crossing it are slow enough not to care. Tracks that run ALONG this split, down the slot rather than across it, are a finding of their own: theirs is an impedance step, not a detour. At 0 of the 2 crossing(s) the track's other neighbour carries a plane, which the return can use there; 2 are on an outer layer, with no other neighbour. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetC3_2' on BOTTOM_LAYER at (76.5, 86.69), 0.5 mm; other side: none: the signal layer is outermost
      • 'NetIC1_20' on BOTTOM_LAYER at (76.5, 87.1), 0.5 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/38/reference-planes?group=split%40TOP_LAYER%3AGND%7CP5V_VME serves each with its own layer_image region; all of them lie in x 75.25..77.75, y 85.69..88.1 mm (board frame, Y up).

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

    • void:GND@BOTTOM_LAYER:hole 67.59,80.34|85.54,88.06'NetIC1_18' crosses one void in the GND plane on BOTTOM_LAYER at 1 place, the longest 1.35 mm; 1 over tracks of other nets routed inside it (fastest: NetIC1_18 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC1_18' on TOP_LAYER, the GND plane on BOTTOM_LAYER has 1.35 mm under the track with copper of P5V_VME, not a plane, at (68.19, 81.86), 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. 1 of the crossings pass over tracks of other nets routed inside the hole on the reference layer: the plane is absent there, and those tracks stand in it. What cleared it: the hole at x 67.59..85.54, y 80.34..88.06. 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.

      • 'NetIC1_18' on TOP_LAYER at (68.19, 81.86), 1.35 mm over copper of P5V_VME, not a plane, 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/38/reference-planes?group=void%3AGND%40BOTTOM_LAYER%3Ahole%2067.59%2C80.34%7C85.54%2C88.06 serves each with its own layer_image region; all of them lie in x 67..70.04, y 80.3..83.05 mm (board frame, Y up).

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

    • void:GND@TOP_LAYER:hole 74.42,86.1|75.12,87.122 nets cross one void in the GND plane on TOP_LAYER at 2 places, the longest 0.25 mm (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC1_20' on BOTTOM_LAYER, the GND plane on TOP_LAYER has 0.25 mm under the track with no copper, at (74.5, 87.1), 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x 74.42..75.12, y 86.1..87.12. 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.

      • 'NetC3_2' on BOTTOM_LAYER at (74.55, 86.69), 0.25 mm over no copper; other side: none: the signal layer is outermost
      • 'NetIC1_20' on BOTTOM_LAYER at (74.5, 87.1), 0.25 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=void%3AGND%40TOP_LAYER%3Ahole%2074.42%2C86.1%7C75.12%2C87.12 serves each with its own layer_image region; all of them lie in x 73.42..75.68, y 85.69..88.1 mm (board frame, Y up).

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

    • void:GND@TOP_LAYER:hole 75.38,86.1|76.08,87.122 nets cross one void in the GND plane on TOP_LAYER at 2 places, the longest 0.25 mm (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC1_20' on BOTTOM_LAYER, the GND plane on TOP_LAYER has 0.25 mm under the track with no copper, at (75.95, 87.1), 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x 75.38..76.08, y 86.1..87.12. 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.

      • 'NetC3_2' on BOTTOM_LAYER at (75.95, 86.69), 0.25 mm over no copper; other side: none: the signal layer is outermost
      • 'NetIC1_20' on BOTTOM_LAYER at (75.95, 87.1), 0.25 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=void%3AGND%40TOP_LAYER%3Ahole%2075.38%2C86.1%7C76.08%2C87.12 serves each with its own layer_image region; all of them lie in x 74.83..77.08, y 85.69..88.1 mm (board frame, Y up).

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

    • void:GND@TOP_LAYER:relief'NetIC1_22' crosses the thermal reliefs of the GND plane on TOP_LAYER at 1 place, the longest 0.61 mm (fastest: NetIC1_22 at 100 ps 10-90 %, ASSUMED)

      The GND plane on TOP_LAYER is drawn back round copper of its own net - pads spoked into it, in thermal reliefs - and tracks cross those reliefs: over the bare ring either side of the pad or spoke the return has no plane under it, and finds its way round through the spokes. 1 place(s) on 1 net(s), each the relief's bare stretches either side of that copper counted together, the longest 0.61 mm. One decision covers the pour's reliefs: whether tracks may cross them, or the reliefs or the routing should keep them apart. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetIC1_22' on BOTTOM_LAYER at (76, 87.5), 0.61 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=void%3AGND%40TOP_LAYER%3Arelief serves each with its own layer_image region; all of them lie in x 73.42..77.18, y 86.5..88.5 mm (board frame, Y up).

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

    • void:P5V_VME@TOP_LAYER:hole 77.42,86.1|78.12,87.122 nets cross one void in the P5V_VME plane on TOP_LAYER at 2 places, the longest 0.25 mm (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      Along 'NetC3_2' on BOTTOM_LAYER, the P5V_VME plane on TOP_LAYER has 0.25 mm under the track with no copper, at (77.55, 86.69), 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x 77.42..78.12, y 86.1..87.12. 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.

      • 'NetC3_2' on BOTTOM_LAYER at (77.55, 86.69), 0.25 mm over no copper; other side: none: the signal layer is outermost
      • 'NetIC1_20' on BOTTOM_LAYER at (77.55, 87.1), 0.25 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=void%3AP5V_VME%40TOP_LAYER%3Ahole%2077.42%2C86.1%7C78.12%2C87.12 serves each with its own layer_image region; all of them lie in x 76.42..78.68, y 85.69..88.1 mm (board frame, Y up).

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

    • void:P5V_VME@TOP_LAYER:hole 78.38,86.1|79.08,87.122 nets cross one void in the P5V_VME plane on TOP_LAYER at 2 places, the longest 0.25 mm (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      Along 'NetC3_2' on BOTTOM_LAYER, the P5V_VME plane on TOP_LAYER has 0.25 mm under the track with no copper, at (78.95, 86.69), 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. 2 places on 2 net(s) cross it, listed below, the longest first. What cleared it: the hole at x 78.38..79.08, y 86.1..87.12. 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.

      • 'NetC3_2' on BOTTOM_LAYER at (78.95, 86.69), 0.25 mm over no copper; other side: none: the signal layer is outermost
      • 'NetIC1_20' on BOTTOM_LAYER at (78.95, 87.1), 0.25 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=void%3AP5V_VME%40TOP_LAYER%3Ahole%2078.38%2C86.1%7C79.08%2C87.12 serves each with its own layer_image region; all of them lie in x 77.83..80.08, y 85.69..88.1 mm (board frame, Y up).

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

    • void:P5V_VME@TOP_LAYER:hole 79.52,86.85|80.42,87.28'NetIC1_20' crosses one void in the P5V_VME plane on TOP_LAYER at 1 place, the longest 0.9 mm (fastest: NetIC1_20 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC1_20' on BOTTOM_LAYER, the P5V_VME plane on TOP_LAYER has 0.9 mm under the track with no copper, at (79.97, 87.1), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 0.9 mm). What cleared it: the hole at x 79.52..80.42, y 86.85..87.28. 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.

      • 'NetIC1_20' on BOTTOM_LAYER at (79.97, 87.1), 0.9 mm, running ALONG the bare copper for 0.9 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/38/reference-planes?group=void%3AP5V_VME%40TOP_LAYER%3Ahole%2079.52%2C86.85%7C80.42%2C87.28 serves each with its own layer_image region; all of them lie in x 78.52..81.42, y 86.1..88.1 mm (board frame, Y up).

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

    • void:P5V_VME@TOP_LAYER:hole 80.68,86.85|81.58,87.28'NetIC1_20' crosses one void in the P5V_VME plane on TOP_LAYER at 1 place, the longest 0.9 mm (fastest: NetIC1_20 at 100 ps 10-90 %, ASSUMED)

      Along 'NetIC1_20' on BOTTOM_LAYER, the P5V_VME plane on TOP_LAYER has 0.9 mm under the track with no copper, at (81.13, 87.1), the same plane on both sides. No single drill's anti-pad and no landing explains it: a clearance is excused only when it holds a drill and is drawn round it, or the track ends in it on the net's own pad or via, or runs through its own via's clearance and no farther. The return current detours round the void. 1 of them run ALONG the bare copper, parallel to its edges, rather than across it (the longest 0.9 mm). What cleared it: the hole at x 80.68..81.58, y 86.85..87.28. 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.

      • 'NetIC1_20' on BOTTOM_LAYER at (81.13, 87.1), 0.9 mm, running ALONG the bare copper for 0.9 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/38/reference-planes?group=void%3AP5V_VME%40TOP_LAYER%3Ahole%2080.68%2C86.85%7C81.58%2C87.28 serves each with its own layer_image region; all of them lie in x 79.68..82.58, y 86.1..88.1 mm (board frame, Y up).

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

    • void:P5V_VME@TOP_LAYER:relief'NetIC1_22' crosses the thermal reliefs of the P5V_VME plane on TOP_LAYER at 1 place, the longest 0.51 mm (fastest: NetIC1_22 at 100 ps 10-90 %, ASSUMED)

      The P5V_VME plane on TOP_LAYER is drawn back round copper of its own net - pads spoked into it, in thermal reliefs - and tracks cross those reliefs: over the bare ring either side of the pad or spoke the return has no plane under it, and finds its way round through the spokes. 1 place(s) on 1 net(s), each the relief's bare stretches either side of that copper counted together, the longest 0.51 mm. One decision covers the pour's reliefs: whether tracks may cross them, or the reliefs or the routing should keep them apart. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetIC1_22' on BOTTOM_LAYER at (78.95, 87.5), 0.51 mm over no copper; other side: none: the signal layer is outermost

      Every site is a row of this finding's evidence, keyed by net, layers and position. GET /api/odb/boards/38/reference-planes?group=void%3AP5V_VME%40TOP_LAYER%3Arelief serves each with its own layer_image region; all of them lie in x 76.42..80.08, y 86.5..88.5 mm (board frame, Y up).

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

    • void:routed@BOTTOM_LAYER|TOP_LAYER3 nets on BOTTOM_LAYER cross the tracks routed through the GND pour on TOP_LAYER at 3 places, the longest 6.16 mm; 3 run ALONG the cut round another net's track (fastest: NetC3_2 at 100 ps 10-90 %, ASSUMED)

      The GND pour on TOP_LAYER is cleared round tracks routed through it, and tracks on BOTTOM_LAYER, which take that pour as their reference, cross them: for the width of each cut the return has no plane under it and detours round the cut. 3 place(s) on 3 net(s). 3 of the crossings run ALONG the cut round another net's track rather than across it, parallel to its edges (the longest 3.07 mm): over or right beside that track, broadside to it as well as off the plane. One decision covers the layer pair: whether these layers may take a pour their own tracks cut as the reference, or the routing or the stackup should give the tracks a plane that is not cut where they cross. No crossing has a plane on the track's other side, so the whole return meets the break.

      • 'NetIC1_22' on BOTTOM_LAYER at (70.71, 84.25), 6.16 mm, running ALONG the bare copper for 3.07 mm, not across it over copper of P5V_VME, not a plane, to the end of the track; other side: none: the signal layer is outermost
      • 'NetIC1_20' on BOTTOM_LAYER at (72.69, 85.52), 5.9 mm, running ALONG the bare copper for 2.45 mm, not across it over copper of NetC3_2, P5V_VME, not a plane, to the end of the track; other side: none: the signal layer is outermost
      • 'NetC3_2' on BOTTOM_LAYER at (73.09, 85.17), 4.28 mm, running ALONG the bare copper for 2.16 mm, not across it over copper of P5V_VME, not a plane, 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/38/reference-planes?group=void%3Arouted%40BOTTOM_LAYER%7CTOP_LAYER serves each with its own layer_image region; all of them lie in x 68.5..74.5, y 82..88.5 mm (board frame, Y up).

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

  6. SinglePointNetmajorOpen

    SinglePointNet: 2 pins×2

    Checked by the review, stands · Claude walker (cern session) — TPS53126 IC1 channel 2 is unused: VFB2 (pin 5) and VO2 (pin 6) are open, along with TRIP2 and the channel-2 gate/switch pins (netlist). The datasheet pin table (p.3) says VO2 connects to the output and VFB2 to a divider, and gives no instruction for an unused channel, so whether open VFB2/VO2 is allowed could not be cited. Left checked.

    At IC1.5, IC1.6

    Each finding, with its own headline
    • IC1.5IC1.5 (VFB2) is an IC input left unconnected

      Pin IC1.5 (VFB2) 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.

    • IC1.6IC1.6 (VO2) is an IC input left unconnected

      Pin IC1.6 (VO2) 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.

  7. FiducialMissingmajorOpen

    The top side carries 19 SMT component(s) and no fiducial was found on it

    The top side carries 19 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.

    At top

  8. observationOBS-316minorOpen

    Unused channel 2: PGND2 and TRIP2 are left floating

    On the TPS53126 IC1, only channel 1 is used, and the unused channel 2 has PGND2 and TRIP2 left floating, while the datasheet says PGND1, PGND2 and GND must be tied strongly together near the IC.

    Channel 2 pins

    PinNameConnection
    7EN2GND
    14TEST2GND
    5VFB2no net
    6VO2no net
    12PGND2no net; marked no-connect, ends in an open stub on the schematic
    13TRIP2no net

    For comparison, PGND1 (pin 19), GND (pin 3) and the exposed pad (pin 25) are on GND.

    Why it matters

    • PGND2 is the ground of the channel 2 low-side driver and current comparator, and the datasheet gives no exception for a disabled channel ('Connect PGND1, PGND2 and GND strongly together near the IC.', TPS53126 datasheet, p.3).
    • TRIP2 is an input with a 10 µA source current and a 0.3 V absolute maximum; floating, it has no defined level. The drawing appears to run TRIP2 into the grounded EN2/TEST2 line, but the exported netlist (and the engine ERC) leaves pin 13 unconnected, so the wire probably ends without a junction.

    The part probably works with channel 2 disabled, but that is outside the documented configuration.

    Suggested fix: tie PGND2 to GND (it sits next to the exposed pad), and TRIP2, VFB2 and VO2 to GND, to match the datasheet.

  9. observationOBS-317minorOpen

    Title block carries placeholder document number, revision and sheet fields

    The only schematic sheet's title block carries placeholder document number, revision, sheet and date fields.

    Title block of the SVEC 3.3V adapter sheet:

    • Document number: EDA-XXXXX-VX-X
    • Rev: *
    • Sheet: * of *
    • Drawn by GK, date XX/XX/XXXX (Check.by MC)

    A published design should carry a real document number and revision so a reader can tie the schematic to the fabricated board. The silkscreen has no revision either.

  10. observationOBS-318minorOpen

    No fiducials on the populated top side

    The populated top side has no fiducial marks, although it carries all 19 SMT parts including the 0.5 mm pitch QFN IC1.

    No fiducial is present on the top copper, paste or overlay; IC1's footprint is QFN50P400X400X100-25N (24 pins, 0.5 mm pitch).

    Automated placement needs at least two global fiducials per populated side. Without them the board has to be aligned on pads or hand-assembled.

  11. observationOBS-319minorOpen

    Designators C6, C8 and C13 are overprinted by other silkscreen

    The designators C6, C8 and C13 are overprinted by other silkscreen on the top overlay.

    • C6 and C8: crossed by the capacitor outline bars of C10 and C9 (the labels sit between the C10/C9 and C6/C8 pads).
    • C13: runs into the descender of the "ohwr.org" text (the g).

    Overprinted designators are hard to read at assembly and rework; moving the three labels clear of the other silk would fix it.

Main parts

Part numberMakerDescriptionWhere
IRF9910PbfDual SO-8 MOSFET For POL Cconverter (library IRF9910Pbf)T1
TPS53126RGETDual Synchronous Step-Down Controller For Low Voltage Power Rails (library TPS53126RGET)IC1
ECASD40G157M016K00Polarised capacitor (library CAEM7343-20_150UF_4V_20%_LESR0016, tolerance ±20%)C4
SRR1260-1R5YInductor, Bourns SRR1260 (house code IND_1.5UH_30%_BOURNS_SRR1260, tolerance ±30%, rated current 8.8A)L1

About this design

Design
SVEC 3V3 — OHWR0037 R1
Designed in
Altium Designer
Board
2 copper layers, 55.5 × 23.2 mm, 20 components, 13 nets
Licence
CERN-OHL-1.2
Source
https://gitlab.com/ohwr/project/svec-3v3 at commit cac8ee8f1242
Attribution
CERN Open Hardware Repository, project ohwr/project/svec-3v3 (commit cac8ee8f12422e99fb7388403ff9c5e501963c10, 2025-04-09). Licence: CERN-OHL-1.2. Source: https://gitlab.com/ohwr/project/svec-3v3. 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)