John Branthoover wrote:
This is an artifact of the schematic rules. In fact, that's why the auto-junction featureHello All, I have a schematic of a board that I just had fabricated. When I drew the schematic, I did so with the Auto-Junction feature turned off. I always draw schematics like this because of problems that I have had in the past with the Auto-Junction feature placing junctions where I don t want them. During testing of the new board, my boss brought me the schematics asking why I failed to attach one end of a pull-up resistor to the +5 volt supply. I had simply forgot to place a junction at the point where the wires intersected. My mistake.
In looking at the board, the connection was made. I ran a netlist. It showed that the connection was made. How can this be?
is there, I think. Any time there is a wire segment with vertices on some point, and
another wire segment that either passes through the same point, or has a vertex at that
point, these are considered connected,
whether there is a junction dot shown or not. The rules are this way so that two
wire segments that are in a straight line are connected, as you'd have no way to
tell there was a break in the wire. The auto-junction only places a junction dot
where a wire segment terminates ON another wire, so it is really just an INDICATOR
of the connection, not the thing that CAUSES the connection. The real usefulness
of the auto-junctions is that it SHOWS were an unintended connection may have been
made. And, clicking and deleting the junction dot will NOT remove the connection,
except in one special case. If you have two wires that cross without there being any
vertex at the crossing point, and you place or remove a junction dot there, the
dot DOES control the connectivity.
I added theYes, you need to have a break in the wire segments at the crossing point. That break,
junction, created another netlist. The netlist showed that the connection
was still made, as expected. I then deleted the dot, created another
netlist. The connection was still there. The only way that I could remove
the connection was to delete the wires and redraw them. After that the
schematic started to behave as expected. No junction, no connection. I
also tried the procedure on other junctions on the same schematic page. I
could not reproduce the error.
without auto-junction, is totally INVISIBLE - and, of course, that IS the danger there!
I have no idea how this error (bug) originally happened. It has made meNo, don''t lose faith in the package. Protel (and Accel's Tango) has ALWAYS behaved
loose all faith in Protel s schematic capture side. Has anyone else seen
this behavior? How can I deal with this problem without checking the entire
netlist before I bring it over to the PCB? Argh!!!!!
this way, it has worked like this for 15 years or more. The auto-junction feature, when
left on, almost removes the danger of this.
But, you have to understand the behavior of every program you use, and know the
pitfalls. EVERY program has a few pitfalls. This particular one is something I've
known about for many years, and since the auto-junction feature, it is not really
a problem.
Jon
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * To post a message: mailto:[EMAIL PROTECTED] * * To leave this list visit: * http://www.techservinc.com/protelusers/leave.html * * Contact the list manager: * mailto:[EMAIL PROTECTED] * * Forum Guidelines Rules: * http://www.techservinc.com/protelusers/forumrules.html * * Browse or Search previous postings: * http://www.mail-archive.com/[EMAIL PROTECTED] * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
