search

KiCad / kicad-footprints

Official KiCad Footprint Libraries for Kicad version 5
https://kicad.github.io/footprints
Other
613 stars 715 forks source link

How t oname variant on existing SC-59 #1195

Closed Misca1234 closed 5 years ago

Misca1234 commented 5 years ago

There exist a foot print SC-59

bild

However the component AH1806 https://www.diodes.com/assets/Datasheets/AH1806.pdf

Says it support SC-59

bild

But it is numbered in a different way.

Should I create a new foot print, if so, how to name it so others foot print filter SC59 wont catch it, that is, like "DIODES_SC-59" ?

poeschlr commented 5 years ago

I would suggest we add some form of suffix to make it obvious that the footprint is for some alternative pad numbering scheme.

evanshultz commented 5 years ago

https://github.com/KiCad/kicad-footprints/tree/master/Package_TO_SOT_SMD.pretty shows a couple possible options:

  1. Add a _Reverse suffix since the pin numbering is reversed.
  2. Add a _TabPin2 suffix (it's perhaps not really a tab, but it's similar to SOT-223 in this regard).

I think KiCad pinout does not have to match the vendors. For example, you can find similar variations in SOT-23 transistors from different vendors. That being said, with a transistor symbol you don't exactly need the pin numbers but for AH1806 the pin numbers are more important.

I'm not finding this clause in KLC v3 on the website, but KLC v2 section 7.6 says Footprints should not be duplicated to match a different pin ordering. This is to be handled in the symbol libraries.. See https://github.com/KiCad/kicad-library/wiki/Kicad-Library-Convention. Maybe that's invalid now? It that clause is still relevant, it would mean change pin numbers on the symbol for this part so it matches the existing SC-59 footprint. That is bound to be weird when compared with the datasheet, though, so perhaps this clause was intentionally removed in KLC v3.

Misca1234 commented 5 years ago

But wont the symbols which have a FP = "SC*59*" hit these ?

evanshultz commented 5 years ago

The symbol for AH1806 in SC-59 package would call out the new footprint, right? And because the new footprint would have a unique name it would not affect the default footprint for any other SC-59 symbol already in the lib.

But you're talking about the footprint filter. Even if it matches it's going to be obvious there's something special about a footprint that has _Reverse at the end. The AH1806 footprint filter could even be SC*59*Reverse* so ensure it only allows the proper footprint.

Misca1234 commented 5 years ago

No, I meant already existing (cant find one actually) or further symbols that use the "original" SC-59 foot print, if those set their foot print filter to "SC*59*" wont they hit SC_59_Reverese ?

poeschlr commented 5 years ago

The footprint filters can never be made such that they do not filter false positives. There is a reason why we require the footprint field to be filled out in most cases. If users ignore the pre filled footprint field then it is up to them to know if the footprint fits their need.

Edit: I might argue that this is even the case if there is no pre filled footprint field. Using cvpcb requires the user to know what they are doing.

Misca1234 commented 5 years ago

then it will be SC-59_Reverse