I am trying to recreate an old PCB that is double sided, but pads were removed from the top layer when no connections were made on the top layer. I know that would be difficult, but some of the connectors don’t have any copper traces going to them on the top layer so I thought I could make a part with only pads on the bottom layer. It seemed easy enough and the editor let me edit each pad and make them NPTH instead of Through Hole. Then I picked only the bottom layer and only the bottom solder mask. That seemed to work. However, when I did that the pad numbering went away, so I tried to re-enter them. It let me do it, but when saved all numbers are gone. Why can’t NPTH pads be numbered?
Here are some photos… The 3D view looks like what I want, but if the pads aren’t numbered how will the schematic know what is connected to what? I started putting parts on the board without the schematic being designed just to see the layout temporarily.
Sorry, I didn’t see the Footprints subsection when I posted…
You say there is no copper attached to it? Look at me 3D photos. It shows copper on the bottom pad, just like I want it, so why can’t it be numbered and the program only allow traces to connect on the bottom where the copper is and maybe have a keepout area on the top to avoid having copper go between pads too close to the hole? Seems logical to me, but I came up old school starting out designing with stick-on pads and taped traces!
Also, if you can’t use the pad for anything, why let me modify it in that way in the program?
… and these holes are not supposed to be used for electrical connections. I do not really know why. Sometimes non plated holes are used on purpose to make rework easier, but this is rare.
After re-reading your post, I do not see a mention of whether your old PCB has plated or non-plated holes.
If the holes are plated, then you can simply use THT and remove the top pads as I wrote earlier. If the holes can’t be plated, then (for now) you have to combine a NPTH with an SMT pad. This is a nuisance, you can (probably will) have troubles with the tracks not reaching the center of the pad because of the hole. Such subtle things may also change with KiCad versions, and I am not sure of the exact behavior in V8.
Thanks for the reply. My old board does have plated through holes. It is double-sided and does have pads and vias on both sides, but we just removed the top pads where they weren’t connected. I have gotten used to simply using pads on top and bottom, but just though I would experiment with my options. I was just thinking that since I didn’t need plating through on those pads it wouldn’t matter if they were NPTH, but it might make it more complicated for the PCB manufacturer. Plating through all of them doesn’t matter to me. Our old boards have plating on all of them whether pads are on top or not and I guess it does give a little more mechanical stability since the solder can wick into the holes.
In looking at my options when not choosing NPTH, I can’t simply remove the top pads. It gives me these choices:
Plated holes are easier for PCB manufacturers, as that is what they do standard. Sometimes they even charge extra if there is even a single NPTH, because that needs an extra step in the production process. But usually it’s combine with the routing at the end and it does not matter much for the PCB manufacturer, it’s only an extra automatic tool change for the machine.
