LibrePCB
Conventions: What orientation markings to use in footprints
There are different ways to implement orientation markings (for pin 1), e.g.
- mark at the short side of a DIP package
- dot in the corner
- chamferred edge
Should only one of these markers be included in the footprint or all possibilities? The same part may have different markings depending on manufacturer (it may even be different for the same part from the same vendor), so I think it would be good to have all of them in one footprint.
Another would be a square pad for THT components, e.g. pin headers :)
It would be great to choose one approach as the recommended way to keep the look of footprints consistent. But I doubt we can come up with something that works everywhere...
Maybe we should suggest approaches depending on package? E.g.
- Through-hole pin headers: Pad 1 square, other pads round
- TO-220: Trapezoid shape, or line where the cooler is
- DIP: Semicircle marking on the short side
- SOT-23-6: Dot in the corner
- SOT-23-3 / SOT-23-5: Obvious from the outline + pads ...etc
The problem is that not all marking methods are interchangeable or can be easily deduced. If you have a package with a chamferred side and a corner circle on the silk screen - which way do you put it? I would have to look it up.
Yup, that's true... But putting a point next to a SOT-23-3 is pointless :)
Package-specific guidelines could account for that, right?
Yes, this is definitely specific per package type. One question yet is still: If multiple markings are used in a package type - should all be included into one footprint?
Marking is actually one thing which is described in IPC-7351, but I really don't like it xD See for example www.ocipcdc.org/archive/What_is_New_in_IPC-7351C_03_11_2015.pdf or https://www.ipc.org/committee/drafts/1-13_d_7351CGoals.pdf.
I don't think it's a good idea to have multiple pin-1 markings in one footprint. IMHO this would be quite confusing in the assembly drawing.
I'm not sure whether I like that approach or not. One advantage is that it's still visible with the part soldered on though. And it could be implemented in a mostly consistent way.
But it does look a bit strange...
Redundancy can be good, especially if the silk screen doesn't get printed very well. It's also useful if the pin 1 marking is still visible with the part installed. That way, the orientation can be easily checked after assembly.
