hatched zones can create unconnected pads/vias

[Thomas Pointhuber]

It is possible to create unconnected vias/pads using hatched zones.

I think there should furthermore always be some filling around vias (otherwise it could be possible to have the via connected only half to the zone and it is marked connected, but this could cause manufacturing issues)

# Version Information

Application: pcbnew
Version: (5.1.0-rc1-304-g3c2761778), debug build
Libraries:
    wxWidgets 3.0.4
    libcurl/7.64.1 OpenSSL/1.1.1b zlib/1.2.11 libidn2/2.1.1 libpsl/0.20.2 (+libidn2/2.1.1) libssh2/1.8.1 nghttp2/1.36.0
Platform: Linux 5.0.4-arch1-1-ARCH x86_64, 64 bit, Little endian, wxGTK
Build Info:
    wxWidgets: 3.0.4 (wchar_t,wx containers,compatible with 2.8) GTK+ 3.24
    Boost: 1.69.0
    OpenCASCADE Community Edition: 6.9.1
    Curl: 7.64.0
    Compiler: Clang 7.0.1 with C++ ABI 1002

Build settings:
    USE_WX_GRAPHICS_CONTEXT=OFF
    USE_WX_OVERLAY=ON
    KICAD_SCRIPTING=ON
    KICAD_SCRIPTING_MODULES=ON
    KICAD_SCRIPTING_PYTHON3=ON
    KICAD_SCRIPTING_WXPYTHON=ON
    KICAD_SCRIPTING_WXPYTHON_PHOENIX=ON
    KICAD_SCRIPTING_ACTION_MENU=ON
    BUILD_GITHUB_PLUGIN=ON
    KICAD_USE_OCE=ON
    KICAD_USE_OCC=OFF
    KICAD_SPICE=ON

[Wayne Stambaugh (stambaughw)]

@Thomas, would you please attach a sample board showing this issue to this bug report.

@JP, please take a look at this when you get a chance.

[Thomas Pointhuber (pointhi)]

here is a pcb showing the errors

[Jon Evans (craftyjon)]

FWIW, unless I haven't figured something out, even Altium can't do this automatically -- if you set your hatch size to be larger than ~2x the via diameter, it won't connect them. I suspect this is a low-priority bug for them because generally there isn't a good reason to have such a sparse hatching.

I think it definitely IS a kicad bug that we sometimes generate thermals / islands around vias that then don't get connected to the rest of the hatch -- those should at a minimum be removed in my opinion.

We could detect them and make the thermal spokes longer and longer until they eventually connect to the hatch, but should we? From an electrical POV this seems like a very poor design to me.

[Seth Hillbrand (sethh)]

What about an option that fills the hatch when it overlaps with the via cut-out ring? That would add some additional copper but allow for valid fills when this happens.

[jean-pierre charras (jp-charras)]

For me, there are 2 different cases:

- If a via is inside a hole, it is not inside the zone. The user has to place vias to the right place.
Removing the hole from the hatched shape is also do not really make a hatched shape.
It can really modify the expected hatched shape if there are a lot of stitching vias.
The Kicad code cannot guess the user wishes.

- Now the pads with thermal reliefs.
This is a different problem.
I do not have a strong opinion about that.
The main issue is a cosmetic issue.

In every case, a hatched zone is not a solid zone (obviously), and should be used only in specific cases, and the user has to know what he is doing.

[Wayne Stambaugh (stambaughw)]

Do we want to do anything about this? Given that Altium doesn't seem to handle this any better than we do, I'm not sure we should go to any great lengths to resolve the issue.

For vias, I agree with JP's assessment. There is no way for the hatched zone filling to determine the users intentions.

For pads with thermal reliefs, wouldn't making the pad a solid connection be a solution?

If there is nothing to do here, let's close this as "Invalid" or "Wont Fix". Otherwise we should determine a path forward. I'm fine with the former.

[Hildo Guillardi Júnior (hildogjr)]

I attached a proposed of how other EDA deal with (the resistors is not showing the pad because the software doesn't have full compatibly with Linux-Wine).

See that:
1) It creates a diagonal lines to internal close the hatched;
2) Derivate the connection horizontal-vertically from their.

[Seth Hillbrand (sethh)]

@Hildo- I don't think your image is directly comparable yet. You would need to make the hatch size larger than the hole size to see how your software handles this issue.

[Jeff Young (jeyjey)]

I'm going to close this as "Won't Fix".

If someone wants to re-open it, please suggest a strategy for how we might handle it.