Kicad eeschema generating random ERC pin not driven errors

Bug #1827773 reported by Aaron Michael Williams
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
KiCad
Fix Released
Unknown

Bug Description

I have a project which has 96 WS2813-mini LEDs on it in 9 rows. Since it is so repetitive, I created the first row then used either block copy/paste or block duplicate and trimmed off the unneeded LEDs in the rows as needed. The circuit is quite simple. My problem is that DRC reports pins as being unconnected even though they are connected. Redoing the wires doesn't fix it unless I add a new wire in a different location which usually fixes the problem. However, when one problem is fixed, a new random one pops up.

Additionally, when I go into the PCB editor, no parts are displayed for placing them onto the PCB even though all footprints have been assigned.

My project is at https://github.com/aaronw2/ears-ws2813-mini

Application: kicad
Version: 5.1.2-5.1.2, release build
Libraries:
    wxWidgets 3.0.3
    libcurl/7.60.0 OpenSSL/1.1.0i zlib/1.2.11 libidn2/2.0.4 libpsl/0.20.1 (+libidn2/2.0.4) libssh/0.7.5/openssl/zlib nghttp2/1.31.1
Platform: Linux 4.12.14-lp150.12.58-default x86_64, 64 bit, Little endian, wxGTK
Build Info:
    wxWidgets: 3.0.3 (wchar_t,STL containers,compatible with 2.8) GTK+ 2.24
    Boost: 1.66.0
    OpenCASCADE Community Edition: 6.9.1
    Curl: 7.59.0
    Compiler: GCC 7.3.1 with C++ ABI 1011

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

Tags: eeschema erc
Revision history for this message
Aaron Michael Williams (aaronw2) wrote :
tags: added: eeschema erc
summary: - Kicad eeschema generating random DRC unconnected errors
+ Kicad eeschema generating random ERC unconnected errors
Revision history for this message
jean-pierre charras (jp-charras) wrote : Re: Kicad eeschema generating random ERC unconnected errors

I am not seeing a ERC error saying a pin is unconnected.
It is the opposite: it says the pin is connected.

Changed in kicad:
status: New → Incomplete
Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

My mistake in reporting, I was going from memory. The errors say "Pin connected to other pins, but not driven by any pin" which is incorrect.

Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

I get the following:
ERC report (Sun 05 May 2019 04:36:56 AM PDT, Encoding UTF8 )

***** Sheet /
ErrType(3): Pin connected to other pins, but not driven by any pin
    @(142.24 mm, 135.89 mm): Pin 1 (Power input) of component D63 is not driven (Net 18).

***** Sheet /Ears-top and bypass/
ErrType(3): Pin connected to other pins, but not driven by any pin
    @(45.72 mm, 35.56 mm): Pin 5 (Power input) of component D69 is not driven (Net 11).

 ** ERC messages: 2 Errors 0 Warnings 2

Both errors are invalid. If I re-route the traces I can make the errors go away, but then a new random error pops up (same type of error).

Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

I might also add that the pcb layout is working, unlike my first post, though I'm early in the process.

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

I can confirm that ERC errors are generated that I don't think should be. The two instances of "pin connected but not driven" should not be there.

Changed in kicad:
status: Incomplete → Confirmed
importance: Undecided → Low
summary: - Kicad eeschema generating random ERC unconnected errors
+ Kicad eeschema generating random ERC pin not driven errors
Revision history for this message
Jon Evans (craftyjon) wrote :

I can ensure this is fixed for 6.0 as I will be moving the pin-to-pin ERC checks to use the new connectivity algorithm. If this is to be fixed for 5.x it will need to be investigated in the old algorithm

Jon Evans (craftyjon)
Changed in kicad:
milestone: none → 6.0.0-rc1
Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

Are there any suggestions on how I can "fix" my design, or should I just ignore this?

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

I looked at it a bit more, these errors are coming from the fact that your +5V and GND nets are coming from an off-board connector (so no pins on the nets are set as outputs) You can add PWR_FLAG components to the affected nets to get rid of them. I'm going to think about if there is any way to make this more intuitive for V6.

Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

How do I go about setting this flag? I am unable to edit the connector pin properties.

Revision history for this message
Rene Poeschl (poeschlr) wrote :
Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

Ideally it would be nice if each pin on a connector could be configured like in the symbol editor to specify if it's power input, power output, input, output, bidirectional, etc.

Revision history for this message
Michael Kavanagh (michaelkavanagh) wrote :

@Aaron, a wish list item already exists for that: https://bugs.launchpad.net/kicad/+bug/1526213
Click the "Does this bug affect you?" to increase the 'heat'.

Revision history for this message
Paul van der hoeven (paulvdh) wrote :

OP has also mentioned this on the user forum:
https://forum.kicad.info/t/getting-random-erc-errors/16811

I do not see a bug, as as mentioned before, placing the PWR_FLAG on the +5V and GND nets makes the presumed errors go away.

Before that I also dragged the ws2812 with the error, and the wire stays attached to the pin.

OP is also a beginner with KiCad and unfamiliar with the PWR_FLAG.

@JonEvans:
Just renaming the error message can already help a lot.
"Pin connected to other pins, but not driven by any pin" is not a very intuitive message.
The fact that it is connected to other pins is not even relevant here.

How about:
"Power input pin not connected to a power output pin or PWR_FLAG. (shown only once)"

or:
"This net has power input pins, but is not connected to a power output".
(this is even better, as it moves the focus from a single pin to the whole net!)

Revision history for this message
Paul van der hoeven (paulvdh) wrote :

Or, state the problem and add the solution:
"This net has power input pins, and must also be connected to an power output pin or PWR_FLAG".

Revision history for this message
jean-pierre charras (jp-charras) wrote :

Just renaming does not work:
The ERC message is not specific to power pins, this ERC warning happens every time an input pin is not driven, i.e. not connected to an output or something similar.

Revision history for this message
Paul van der hoeven (paulvdh) wrote :

So the new message only acounts for the most confusing instances, and is easy to distinguish from the others by the inclusion of (at least one) power pin in the net...

Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

I have actually created at least a dozen boards with KiCad. This is the first time I have run into this, however.

Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

Since this can apply to any input pin as well, I will restate that one must be able to configure connector pins as input, output, power, etc.

Revision history for this message
Aaron Michael Williams (aaronw2) wrote :

An alternative would be to suppress the error if the wire is connected to a connector and assume that power or a signal is being received from the connector pin.

Revision history for this message
Seth Hillbrand (sethh) wrote :

@Jon- This report appears to be a case of users not fully understanding the error and its causes/solutions. Perhaps this is best addressed by documentation? I'd hate to change this out from under our users who have already learned how to interact with this error.

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

@jp-charras input pins do not generate this ERC message so the suggested rewording might actually be a good idea. Tested with 5.1.4 so this rewording could potentially even be done in v5.1.x

See screenshot.

I also noticed the third reported error points to the wrong pin. the source of the error is the power input pin of U1E but it wrongly points to the pin of the resistor. (will open a separate report for this.)

Revision history for this message
KiCad Janitor (kicad-janitor) wrote :

KiCad bug tracker has moved to Gitlab. This report is now available here: https://gitlab.com/kicad/code/kicad/-/issues/1895

Changed in kicad:
status: Confirmed → Expired
Changed in kicad:
importance: Low → Unknown
status: Expired → 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.