>>>>> "Dan" == Dan McMahill <[EMAIL PROTECTED]> writes:
> David Kuehling wrote: >>>>>>> "DJ" == DJ Delorie <[EMAIL PROTECTED]> writes: >> >>>> http://user.cs.tu-berlin.de/~dvdkhlng/clearance-problem.png >> >>> Could you post (or send me privately) the .pcb file? >> Here is a simplified file that only contains the problematic >> footprint. Quite possibly this is just a problem with the footprint? >> After all this was hand-coded in M4... > I'd be more likely to suspect the footprint if "after all it was hand > drawn instead of generated programattically"... > Ok, here's the deal. It is a bug in pcb. Square (or rectangular) > pads are checked by growing one of them in X and Y on all 4 sides by > the minimum space. This of course means that the corners grew by > sqrt(2) more and thats why you got a failures. I'll try to cook up a > patch tonight. Yes, I now remember having seen some illustration in the PCB user guide (section 7), illustrating how lines (pads) with rectangular ends are drawn by a rectangular aperture (on page 57). The clearance is drawn as a rectangle, so one could take the current behaviour as being right. Else somebody would have to fix the documentation :) I already reworked my footprint, moving the pads a little outwards. Hopefully this will work in production (but since there isn't an oficially recommended pcb layout for the chip in question the footprint is a rough guess anyway). Thanks also for your patch. But since my layout already changed, there might not be much testing that I can do. I'm going to upgrade PCB anyway before exporting the final gerber files (I heard of some problems which aren't fixed yet in pcb-2006082) and will then test the patch. regards, David -- GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg Fingerprint: B17A DC95 D293 657B 4205 D016 7DEF 5323 C174 7D40 _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
