[FR] Assign footprint and see pin names


I have read that the whole “Assign footprint” is near to be completely re-designed to implement it in a single window with the footprint viewer.
Also, wouldn’t it be good to show in the footprint window the pin names?
Let’s make a little, dirty, ugly example, I have this part:

To be sure that the pin name-number association is correct I need to check 3 windows: schematics, pdf and footprint viewer

This way I need to check only two: pdf and footprint viewer

I need to check 0 as I directly connect symbol with footprint. So once checked never more needs any attention.

There’s one problem: when you create a footprint, it’s not part of a pcb design. Because in KiCad symbols and footprints aren’t tightly coupled together but one footprint can be assigned to several symbols, and the footprint is agnostic about it, the footprint can’t have pin names when it’s being designed.

In v5.99 you can see the pin name (I think it’s “function”) when the footprint has been attached to a symbol in a design.


Yes i am aware of this and the philosophy of kicad, I don’t think it is a problem, I am not talking about linking forever that footprint with that symbol. Instead I was thinking about showing just in that view the pin name and pin number. This way the end user doesn’t have to check multiple Windows

Actually you check 2 (or even 3) :sweat_smile:

I can’t see how the footprint can figure out the pin name before the symbol-footprint link.
Same footprint, different arrangements

I have never seen the process of “Assign footprints” but from first OP post I understood that he suggest that during footprint selection the footprints to be selected from should be shown (I don’t know if they are shown or not) showing pin numbers and names (coming form actual symbol for which the footprint to assign are you searching). So not footprint will figure out the name but the assign footprint process.

Sorry @Piotr, I’m not sure what you mean.

I think you mean that all the footprint candidates (in fact the whole library) could show the pin name?

In the case I showed let’s take BAT54A and BAT54C to make it simple.
When we select BAT54C pins are 1-A, 2-A and 3-K, so any footprint candidate will show 1-A, 2-A and 3-K.
And if we select BAT54A the pads of any footprint candidate will show 1-K, 2-K and 3-A?

A footprint candidate is any footprint in the selected library no matter the number of pads of the footprint. And each time I select a different component, all the relationships between the symbol and all the footprints of the library must be updated to show the new pin arrangements. I keep on not seeing it.
It’s a kind of “I will phone you to ask you your phone number”

I agree with @aster that we need to check 3 windows each time, mainly for sot-23 transistors, the most non-standard pin assigment footprint in my opinion.


I really never used "assign footprints’ so I don’t know how footprints candidate are selected but suppose that it is not so that for diode you can be suggested to select DIL14 footprint. I suppose that symbol knows that for him the 2 pin footprints should only be proposed (or 3 pins for some diodes).

Depends what you understand by updated. What I (and probably the OP) suggest I would not call it ‘update’ - only the selection window showing you the shape of suggested footprints would add to each pin number also the pin name coming from symbol.
I can’t imagine what a problem you see in it?

Some years ago we decided that in our new access control system there will be no ‘from top’ given functions of each input/output. That we will allow for flexibility, but not by allowing to select for specified output what is its function (it is simple) but in opposite direction - for each function to select input/output (from existing pool) to be used for it. So you have a controller + several extension devices (different kinds) each adding some (different numbers) inputs/outputs to be used by controller. Then during configuration of 8 doors served by controller you have to specify for each door which in/out in which extension device (or controller) will be used for its functions (in total about 12 in/outs per door, not all have to be used). There are some additional restrictions - some in/out sets have to be located in the same physical device. What I am doing in that configuration program to show the current set of in/outs you can select from for function you are specifying now you can name ‘updating’ (after each selection done the lists for all other selection boxes are recalculated) but it is only updated in RAM. No information have to be saved before the full configuration process is finished.
To show pin names during footprint selection only something (much simpler) like that have to be done. I don’t understand why you call it updating library and what problem you see in it.

By update I meant re-assining symbol pins and numbers to pad footprints. It can be made in RAM.

OK, no need to assing pin names to every footprint of the library, just to the clicked footprint. Then this footprint can be assigned to the symbol or not. No need of suggested footprints either. And it doesn’t matter if the footprint has more pins than the symbol.

Now I think it could be possible to implement this function.

While assigning a footprint to a symbol, why not, as long as the footprint is compatible with the symbol, i.e. has the same pin numbers.

As I understand it:

Schematic symbols have both pin names and pin numbers.
During assignment of footprint links to symbols, the pin names from the schematic symbol could be placed next to the pins in the Footprint view.

So the orange text next to this Footprint screenshot from the first post is not taken from the Footprint library, but it is generated “live” from the schematic symbol this Footprint is going to be assigned to.

I like the idea. Especially for SOT-23 where pin numbering is ambiguous, or for TO-92 where I always tend to forget if numbering is from left to right or right to left, what is the front or back, and some datasheets have top view, while others have bottom view.

1 Like

Yes, it can be genetated on the fly. One footprint at a time, no need to map all the footprints of the library as I thought before.
Good for 3 pin symbols, it is not useful for high number of pins like BGAs

It would also help with diodes.
For diodes both the pin numbers and pin names in Eeschema are hidden by default. Pin numbers “1” and “2” do not mean much with diodes anyway. The generic diode does have pin names “K” and “A” and showing these names during footprint assignment does seem useful.

I usually double and triple check myself. For example during drawing traces on the PCB I verify diode symbols o Silkscreen with their net connections.

1 Like

Yes @paulvdh this is what I wanted to purpose. If other people think this is useful should we open a FR? where does usually the kicad dev track them?


1 Like

ok, I will wait a few more days for some other opinions before opening the FR

opened https://gitlab.com/kicad/code/kicad/-/issues/4576

There were misunderstanding. I explained, I hope.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.