Strange connectivity behavior on project reported on the forum

Bug #1821475 reported by Rene Poeschl
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
KiCad
Fix Released
Medium
Jon Evans

Bug Description

We have gotten a cry for help regarding a strangely behaving project over at the forum: https://forum.kicad.info/t/before-giving-up-on-kicad-one-last-attempt-erc-misery/15933

We played around with it but are out of ideas.

It seems something is quite wrong with the schematic as it is. I will attach the original version and one that i cleaned up and assigned footprints such that one can also check what gets exported.

One strange thing is that ERC reports R13 as not connected (exporting to pcb_new will also leave R13 unconnected)
Even stranger are the crystal near R13, and its condensators. These are not reported as not connected but do not get a connection on the board if exported.
If i delete the wires between r13 and the crystal i can no longer highlight the remaining wire connecting the crystal to its capacitors.

A similar behavior exists with the net created by the global labels Rx, Tx, and Reset. (Here replacing the ardoino with some other symbol can fix the problem. But the one with the crystal persists.)

This behavior is experienced with multiple versions of kicad (At least 5.0.2 and 5.1.0 expirience this problem)
I seem to remember encountering something similar on the forum a long time ago but can no longer find the post. (Meaning something very similar could have already been reported here.)

Revision history for this message
Rene Poeschl (poeschlr) wrote :
Revision history for this message
Rene Poeschl (poeschlr) wrote :
Revision history for this message
Rene Poeschl (poeschlr) wrote :

Additional info by the original author:

---

The schematic was drawn from scratch in Kicad 5.02 (Linux Mint 19.1) BUT the arduino part with components was imported in an empty schematic, saved in Kicad format and then added to the mouseoleum schematic.

After that I replaced ALL parts with newly made parts because the parts I use for Veecad need a specific FOOTPRINT label.

Also most parts were rewired.

Then I read of problems in Kicad with ERC when IC’s have other labels than ‘PASSIVE’ so I changed all pins to passive.

A lot of tinkering within the schematic so maybe that’s the source of the problem although it should be possible to to that.

Revision history for this message
Rene Poeschl (poeschlr) wrote :

And version info of the original author

Application: kicad
Version: 5.0.2-bee76a0~70~ubuntu18.04.1, release build
Libraries:
wxWidgets 3.0.4
libcurl/7.58.0 OpenSSL/1.1.0g zlib/1.2.11 libidn2/2.0.4 libpsl/0.19.1 (+libidn2/2.0.4) nghttp2/1.30.0 librtmp/2.3
Platform: Linux 4.15.0-46-generic x86_64, 64 bit, Little endian, wxGTK
Build Info:
wxWidgets: 3.0.4 (wchar_t,wx containers,compatible with 2.8) GTK+ 2.24
Boost: 1.65.1
OpenCASCADE Community Edition: 6.9.1
Curl: 7.58.0
Compiler: GCC 7.3.0 with C++ ABI 1011
Build settings:
USE_WX_GRAPHICS_CONTEXT=OFF
USE_WX_OVERLAY=OFF
KICAD_SCRIPTING=ON
KICAD_SCRIPTING_MODULES=ON
KICAD_SCRIPTING_WXPYTHON=OFF
KICAD_SCRIPTING_ACTION_MENU=ON
BUILD_GITHUB_PLUGIN=ON
KICAD_USE_OCE=ON
KICAD_USE_OCC=OFF
KICAD_SPICE=ON

Revision history for this message
Rene Poeschl (poeschlr) wrote :

The user created a new schematic where they do not experience the same trouble. The main noticeable difference is the timestamps in that new schematic are 8 hex digits (=32 bit) long instead of 16 digits (=64 bits).

Revision history for this message
Rene Poeschl (poeschlr) wrote :

Ok somebody else went through the trouble of truncating the timestamps down to 8 digits and nothing changed with the schematics behavior. (Other than that the 32bit version can now open it.) So the timestamps might be a red hering.

Revision history for this message
Jon Evans (craftyjon) wrote :

Confirmed on:

Application: kicad
Version: 5.1.0-unknown-4896daa~82~ubuntu18.04.1, release build
Libraries:
    wxWidgets 3.0.4
    libcurl/7.58.0 OpenSSL/1.1.0g zlib/1.2.11 libidn2/2.0.4 libpsl/0.19.1 (+libidn2/2.0.4) nghttp2/1.30.0 librtmp/2.3
Platform: Linux 4.15.0-46-generic x86_64, 64 bit, Little endian, wxGTK
Build Info:
    wxWidgets: 3.0.4 (wchar_t,wx containers,compatible with 2.8) GTK+ 3.22
    Boost: 1.65.1
    OpenCASCADE Community Edition: 6.9.1
    Curl: 7.58.0
    Compiler: GCC 7.3.0 with C++ ABI 1011

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

FWIW the "cleaned up" version operates perfectly in the new connectivity algorithm, but that's not going to be very helpful to this user. I have not dug in to what's going on in the previous (V5/V4) connectivity algorithm and this schematic.

Changed in kicad:
status: New → Confirmed
importance: Undecided → Medium
Jon Evans (craftyjon)
Changed in kicad:
assignee: nobody → Jon Evans (craftyjon)
Revision history for this message
Jon Evans (craftyjon) wrote :

This ended up being similar to https://bugs.launchpad.net/kicad/+bug/1677282

Changed in kicad:
status: Confirmed → In Progress
Jon Evans (craftyjon)
Changed in kicad:
status: In Progress → Fix Committed
milestone: none → 5.1.1
Revision history for this message
Rene Poeschl (poeschlr) wrote :

The fixed project files you posted on the forum are missing the pictures and also some footprints seem to have changed. Did you use the original or cleaned up version as the source for the fixed files?

Changed in kicad:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.