Square pads or Round?


#1

I notice in some of the supplied footprints that, while most of the pads are round, one pad will be square. I guess there is some significance to this, some convention. Would someone please explain?
Thanks


#2

Without knowing what footprints you mean, it is difficult to tell.

But, with that said, an old standard (that I personally still follow for all components and therefore must maintain my own libraries for) is that pin1 is a square pad. This makes it easy to identify pin numbers if the silkscreen is either hidden by the component body, or on the other side of the board. This makes troubleshooting and/or applying modifications much easier.


#3

These footprints follow the KLC :


#4

Thanks folks. That’s what I wanted to know.


#5

Once V5 is released, there would be an advantage to using rounded rectangular rather than square for pin 1. When I route around that square pad, I always seem to get clearance problems due to the sharp corner.


#6

You didn’t quote the exception:

Regarding my own practices, when I said all I meant all including non-polarized parts. I know it seems that it shouldn’t matter on non-polarized parts to know pin 1, and I agree with that for assembly. But, I find it easier for troubleshooting and documentation to simply be able to say “R2-1” (R2 pin 1) instead of “the end of R2 closest to U3” or some other feature otherwise unrelated to R2. I also have pin numbers turned on even for non-polarized components in my schematic symbols so I can positively identify the pins of all parts when troubleshooting.

I gave up my fight to implement this in the official libs in the early days of KLC, so I just maintain my own libraries. I guess my corner was to acute to fit anyone but myself in it… :wink:


#7

The forum script quoted the entire KLC section I linked (exception included). Don’t blame me wrong.


#8

Yeah, sorry about that. I think I realized that after writing my response. I didn’t intend to impugn you.

I probably should have said that the forum software didn’t include the exception in the link quote because it is further down the linked page.


#9

I use the “square pad is pin 1” convention for non-polar components, and for the same reasons. I don’t show pin numbers (of non-polar parts) in my general-purpose schematics, but DO show pin numbers when I extract portions of a schematic to illustrate test procedures, troubleshooting aids, or failure analyses.

Dale


#10

Yay! I don’t feel so alone. :smiley:


#11

What about polar components?

I can’t find suitable footprints for a Tantalum THT capacitor, ones that have one square and one round pad at a specified pitch (Would someone please name one for me. Thanks). If I have one, is the square pad +ve or -ve?


#12

We are aware of the fact that these footprints are missing but we have not yet had the time to create them. More detials see: https://github.com/KiCad/kicad-footprints/issues/474


#13

Pin 1 is always square, and is +VCC for nearly every passive component; including the capacitor you mention in your post.

Diodes however (except zerners), are bassakwards from the normal; pin 1 is still square, but it ties to the cathode of the diode.


#14

What makes diodes backwards, and why are zeners an exception?


#15

Because zerner diodes only operate correctly when they are installed on the schematic upside-down.


#16

That still doesn’t explain why you think diodes are backwards. I assume you are referring to the typical zener shunt regulator, what makes the zener “upside-down”?


#17

I don’t know why you want me to give you a detailed answer as I know you already know.

I’m having a real difficult time at the moment finding a quality link, but I know the following to be true (because I got it wrong the first time).

With resistors,capacitors, inductors, and transformers the prefered convention is to connect Pin 1 to “positive” and Pin 2 to “negative”. This is important with passive elements that have an actual polarity requirement, like an electrolytic capacitor.

The diode, in it’s most common forward biased operation, the actual polarity is opposite of the convention; having Pin1 as “negative” and Pin 2 as “positive”.

This is important as KiCad matches the footprint Pad numbers to the symbol Pin numbers.

Is this answer satisfactory?


#18

Conventions for pin 1 in PCB footprints are set by IPC-7351, which says pin 1 is Cathode


#19

Perhaps the greatest use of giving a detailed answer is that newbies like myself who happen to be following the thread do not know the things that 1.21Gigawatts could take for granted. By giving the detailed answer you open a door to some of the arcane, Druidic secrets which abound in the cosmology of electronics. :grin:


#20

That seems to me to be correct.

I’ve been digging around on documents saved to my computer that I thought I had saved about this very issue over a year ago.; no joy.

I KNOW I have a saved DataSheet from a manufacture of a diode that (for a forward biased diode) lists both the pin number and polarity; but I can not find it.

Good news is that with a 2 pin part, one can always flip the part opposite of the silk and still have a working board.