search

KiCad / kicad-symbols

Official KiCad schematic symbol libraries for Kicad 5
https://kicad.github.io/symbols
Other
704 stars 745 forks source link

Strange behaviour of the Cvpcb #648

Closed Misca1234 closed 6 years ago

Misca1234 commented 6 years ago

I am not sure if you have already seen this or already reported and I guess it is not a library issue really but a program one.

It seems that Cvpcb have difficulties with foot prints where the size of the pad is not 3 numbers

I have made a symbol ESP8089 which I associate with the "wrong" footprint (just to show that there is not an error with the symbol or so, the strange behaviour is described further down.)

bild

bild

bild

If I run the Cvpcb I get the following result (as expected)

bild

But if I change to the more correct foot print, QFN-32-1EP_5x5mm_P0.5mm_EP3.6x3.6mm I get the "wrong" behaviour

bild

bild

bild

It does not longer "catch" anything with the foot print filter

Misca1234 commented 6 years ago

Something is very strange here, here is a screen shoot of the footprint, the pin number is way wrong

bild

The only one using that footprint, is nRF52810-QCxx in MCU_Nordic.lib Shall I fix this footprint in same push as the symbol I want to push using this foot print ?

evanshultz commented 6 years ago

@Misca1234 Good find. So this doesn't have anything to do with CvPcb.

Yes, please fix the footprint. It won't be in the same PR as the symbol, but a complementary PR would be nice.

@poeschlr @Ratfink This footprint was added at https://github.com/KiCad/kicad-footprints/pull/518. I suspect the pin array feature was used four times but the starting pin number wasn't incremented on each row of pins and instead left at 1. Maybe it's best to show copper pads and pad numbers (not just for us but all contributors) so something like this can be caught at a glance? https://github.com/KiCad/kicad-footprints/pull/612 is another one where the copper wasn't enabled during the screenshot and without that info this kind of thing may be more likely.

Ratfink commented 6 years ago

Aw gosh darnit, I merged that one! I was probably expediting the review because of all the work to fix #520, got lazy, and didn't load it in KiCad before merging. Since I'm responsible for that footprint being merged, I'll fix it.

Misca1234 commented 6 years ago

So this doesn't have anything to do with CvPcb.

No, it seems that the pad numbering messed stuff up for the Cvpcb.

I'll fix it.

Ok, then I wont do anything with it then

poeschlr commented 6 years ago

And i created it so... Wasn't my best hour i guess.