Antipad area should not exclude spokes

[Stephen Eaton]

Version: nightly build kicad-20100629-r2407-windows.tar.gz
Platform: Windows7

Problem:
(files attached)

Zone on top Copper Layer fails to attach/connect two adjoining pins (25 & 26) that are connected to net GND on component U2. Other pins on the component on the same net attach to the zone ok.

[jean-pierre charras (jp-charras)]

There is no really a bug.
Just there is no room to create thermal reliefs for pads 25 and 26.
With your antipad values and the pad sizes, pad 25 creates a 200 mils antipad area that prevents Pcbnew to create a good thermal relief for pad 26, and pad 26 creates a 200 mils antipad area that prevents Pcbnew to create a ggod thermal relief for pad 25.

[Stephen Eaton (seaton)]

Thanks Jean-Pierre for explaining, it makes sense.

However shouldn't the anti-pads get over-ridden or combined on adjacent pads on the same net? So that the zone nets can complete on the pads.

Stephen

-----Original Message-----
From: <email address hidden> [mailto:<email address hidden>] On Behalf Of jean-pierre charras
Sent: Tuesday, 6 July 2010 6:29 PM
To: <email address hidden>
Subject: [Bug 602176] Re: Zone fill fails on component with adjoining pins attached to zone

There is no really a bug.
Just there is no room to create thermal reliefs for pads 25 and 26.
With your antipad values and the pad sizes, pad 25 creates a 200 mils antipad area that prevents Pcbnew to create a good thermal relief for pad 26, and pad 26 creates a 200 mils antipad area that prevents Pcbnew to create a ggod thermal relief for pad 25.

--
Zone fill fails on component with adjoining pins attached to zone
https://bugs.launchpad.net/bugs/602176
You received this bug notification because you are a direct subscriber
of the bug.

Status in KiCad: Triaged

Bug description:
Version: nightly build kicad-20100629-r2407-windows.tar.gz
Platform: Windows7

Problem:
(files attached)

Zone on top Copper Layer fails to attach/connect two adjoining pins (25 & 26) that are connected to net GND on component U2. Other pins on the component on the same net attach to the zone ok.

To unsubscribe from this bug, go to:
https://bugs.launchpad.net/kicad/+bug/602176/+subscribe

__________ Information from ESET Smart Security, version of virus signature database 5254 (20100706) __________

The message was checked by ESET Smart Security.

http://www.eset.com

__________ Information from ESET Smart Security, version of virus signature database 5254 (20100706) __________

The message was checked by ESET Smart Security.

http://www.eset.com

[Jeff Young (jeyjey)]

After studying this case a bit (I've attached a picture of the area of interest), I don't think the anitpad distance should be overridden.

The antipad clearance is about limiting heat-transfer away from the pad; it has nothing to do with electrical connection. Two pads from the same net will transfer heat away just as fast as two pads from different nets.

If that's not an issue, then the antipad distance can be decreased.

[Jeff Young (jeyjey)]

That didn't sit quite right with me, so I noodled on it some more.

The second pad doesn't actually contribute to the heat transfer profile because it's on the other side of the zone. The only factors that control the heat transfer to/from the pad are the number of spokes, the spoke width and the spoke length. Essentially the antipad distance controls the length of the neck-down (ie: the spoke).

So the anti-pad area shouldn't exclude spokes at all (whether on the same net or not -- net-to-net isolation is specified by the zone clearance distance not the antipad distance).

[Jeff Young (jeyjey)]

The thermal relief algorithm currently knocks out the 4 polygons around the spokes/pad. To fix this we'd need to change that to knocking out a single polygon and then adding back in the spokes. It's too late in the 5.0 dev cycle for that kind of change, so I'm pushing this to 6.0.

(FWIW, an algorithm as described would more easily handle different numbers of spokes: https://bugs.launchpad.net/kicad/+bug/1661963)

[Jeff Young (jeyjey)]

Raising to medium as there isn't really a good workaround to this.

[Rene Poeschl (poeschlr)]

Is https://bugs.launchpad.net/kicad/+bug/1740633 a duplicate of this?

[Jeff Young (jeyjey)]

See also the bug Rene linked too (which has been marked as a duplicate).

[KiCad Janitor (kicad-janitor)]

Fixed in revision 5c89e4490ede81a45e77e258f81d54dca65a6536
https://git.launchpad.net/kicad/patch/?id=5c89e4490ede81a45e77e258f81d54dca65a6536