I forgot to thank everyone for their helpful responses to my question. The ensuing message thread also brought up additional interesting information.
I was able to complete my board and get the files to the board house for production. I did end up doing a lot of manual modification to the auto-routed traces; but had I been more experienced I think there would have been less of this. For example, it would have made sense to manually connect the isolation capacitors to their respective IC's, and THEN run the auto-route, because the auto-router knew nothing of these location considerations. I'll know better next time. :=) Thanks again! Vaughn On Mon, 2006-10-02 at 10:27, Vaughn Treude wrote: > Hello everyone: > I've been checking my board layout with the Design Rule Checker command > in PCB. I was encountering some mysterious behavior and wondering if > anybody could explain this a little. > > 1. The default minimum spacing is 10 mils, yet DRC flags everything that > is exactly 10 mils. This is a problem because the auto-router put traces > 10 mils apart in many places. I tried changing the minimum spacing to > 9.9 or 9, but apparently that's not valid because it does not retain > that value. (In fact, the preferences dialog experiences an error if I > do this, and refuses to close the second time I bring it up.) So I set > the min-spacing to 5 mils and DRC found a few problems in the lines I'd > added manually, which I was able to correct. Does this mean the actual > minimum spacing should be 15 mils or something like that, so that > auto-routed traces will pass the DRC 10 mil guideline? > > 2. I've been seeing "potential for broken trace" error a lot. In some > cases I can see that a line has some unnecessary jags in it which I've > then straightened out. In other cases the only problem seems to be that > two connecting lines that go in the same direction have for some reason > not been merged into a single line. It's my understanding that if the > trace _looks_ continuous (at a pretty high zoom) it _is_ continuous. Is > DRC flagging nonexistent errors here or is my understanding incorrect? > > 3. Sometimes when DRC reports "copper areas too close", I go to the > coordinates specified in the message and I cannot for the life of me > find anything closer than 10 mils. (And I still have it checking for 5 > mil spacings.) Does DRC sometimes get fooled into thinking that > connected lines should be separated? Or am I misinterpreting the > errors? > > Thanks! > Vaughn T > > _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user