Mail Archives: geda-user/2012/12/13/23:14:10
On Dec 13, 2012, at 7:59 PM, Markus Hitter wrote:
>
> Am 14.12.2012 um 02:22 schrieb DJ Delorie:
>
>> Even something as simple as adding a single trace could "short"
>> multiple existing subnets, and if some of those subnets have been
>> assigned to nets but some subnets are as yet unassigned (because they
>> have yet to connect to something known to be in the netlist), you get
>> lots of arbitrary choices to be made about how everything needs to be
>> resolved.
>
> Don't wanna defend Mr. Doty here, but at least his appearance gave me some inspiration. How would gschem handle this "shorted tracks" problem?
>
> It doesn't, because gschem tracks don't short by just lying atop of each other.
I don't wanna be impolite to my defender, but the gschem case is different. Gschem is a schematic editor: it has no need to understand connectivity. Indeed, for maximum flexibility it should not. But a conductor on a printed circuit board is usually a part of a specific net, and if it isn't, any contact with any other conductor needs to be considered a "short" unless specifically authorized.
> Except in certain conditions, where the short is actually wanted.
>
> Conveniently, this pretty much applies to rat lines as well; enter pcb.
>
> To get a more gschem-like behaviour in pcb, pcb would have to give up drawing arbitrary, disconnected tracks. It would have to stop connecting tracks by just drawing them at the same place. Sounds ridiculous, doesn't it?
It seems to me that a freshly drawn piece of conductor should have defined affinities. Of course, for a "floating" object (HV guard ring, antenna element, etched-in text) it might just be "floating".
>
> Possible solution: instead of drawing tracks, board design starts with rat lines. Like we currently have them. Then, these rat lines are - sort of - pinned down to become, or being morphed into tracks. Perhaps with a tool similar to how paths are edited in drawing applications. Add vertices, drag these vertices, join them to forks, and so on, until the board is done. But never disconnect a track in this process.
That would make a lot of sense to me, anyway.
>
> This way, tracks are never disconnected from a net. Finding a short becomes trivial. Probably a number of other tasks, like track length measurement, too.
>
> OK, I have no code. But I couldn't resist to forward this inspiration, either.
>
>
> Thanks for listening,
> Markus
>
> - - - - - - - - - - - - - - - - - - -
> Dipl. Ing. (FH) Markus Hitter
> http://www.jump-ing.de/
>
>
>
>
>
>
John Doty Noqsi Aerospace, Ltd.
http://www.noqsi.com/
jpd AT noqsi DOT com
- Raw text -