[PCBnew] How to keep filled zone out of NPTH area?

Ok, I got this footprint here (a non-plated TH with a soldermask free area around it):

This happens when I put it into a layout and have a filled zone going over it:

The copper fill doesn’t stay free like in the footprint preview, but instead approaches the central through hole pad. I don’t want that.

This is in BZR6608. Gonna test some current nightly later on this. Anyone got an idea how I can keep the circular area from being filled?

Footprint for you to play:
z_NPTH.kicad_mod (1.0 KB)

1 Like

In such a case we can clearly see that the Keep Out layer is a missing feature.

I also wonder why Pcbnew zone fill procedure ignores the settings in the Local clearance tab.

1 Like

It is a usability issue.

From my experience, using a drill larger than the pad usually results in unplated holes. Pad clearance should work in this case.

Can you elaborate please?

How is the mounting hole footprint built for that?
I can’t define a NPTH/TH pad with a drill diameter larger than the pad diameter, KiCAD doesn’t allow it.

Here’s a mounting hole footprint that I’ve used a few times. I don’t know if it’s exactly what you’re looking for or not, but it DOES keep the copper fill pulled back from the mounting holes. I think the secret is in how I defined values in the “Local Clearance & Settings” tab of the pad’s “Properties” dialog.

MountingHole_125mil_Washer.kicad_mod (1 KB)



Thanks, that solved my problem.

The question remains. Why Thermal Gap setting from Copper Zone section does not play a significant role here?

I think my memory failed me. I might have read something along those lines on ‘oshpark’ (drill > pad). Apparently (drill = pad) works as well, I’ve done it before (even with normal TH pads) with various Chinese manufacturers.

Hmmmmm . . . . possibly because the NPTH has no electrical definition - that is, it has no electrical connection to any net - the concept of a thermal gap between the NPTH and any electrically-defined fill zone has no meaning. Or possibly because the NPTH has no annulus from which to calculate the gap?