At 09:55 AM 2/3/2004, Leo Potjewijd wrote:
I changed the schematic (just a single sheet) to include all connectors and their interconnections (almost all of them got a new refdes); then I changed the PCB manually to add the new ones, delete the unwanted ones and renumbered them all according to the new schematic.

This was the mistake. You changed the schematic and you changed the PCB and then you tried to synchronize..... Change one and take the changes to the other, that's the way it is supposed to work.....


When I run 'update PCB' the synchroniser does not recognise two (different) connectors: it claims that J7 is not on the PCB (which it is)

There is now a J7 on the PCB, yes, but it is not the original J7 that you changed to something else. As I understand this, the synchronizer is trying to implement the schematic change that you made, J7 to JXX. It looks for J7 as the original J7, *not* your new J7. Presumably that footprint is still there, but it now has a different name, so the sychronizer will not make the change J7 -> JXX. Or something like that.


The synchronizer depends upon hidden unique identifiers to allow it to do renumbering. You did extra work, actually. Probably all you needed to do was change the schematic and let the synchronizer do the rest. Or change the PCB and take that back into the schematic.

and that J13 is not on the schematic (which it is).

Yes, but your original J13 has been renamed to something else, I'd bet.


I checked all fields, names etc of the components in question but cannot find anything wrong.

Surely I am missing something extremely obvious but haven't a clue.....

You are trying to run the synchronizer as if it were a mere netlist updater, as if the PCB and Schematic had never before been synchronized. This is why you expect J7 on the schematic to become linked to J7 on the PCB. But, because the system is set up to do renumbering for you, components are matched by the hidden links. They are only matched by refdes on an original synchronization, or if there are no conflicts. (I.e., you added a new symbol to the schematic and placed a footprint for it on PCB and gave them the same reference designators, manually. Because the symbol and footprint have never before been synchronized, the synchronizer will correctly match them with ease.)


There are ways to reset the hidden links. As I recall, one method is to write the PCB as ASCII, which does not record those links. Then you synchronize the schematic to the new unlinked PCB, in the process creating new links that match the reference designators you now have.

There ought to be a button for "Unlink PCB" or something like that. But it is easy enough to handle once you realize what is going on.

I remember a time when I did a fairly complex PCB design for a client. This was in Tango PCB (DOS). The client then wanted to do some extensive changes. "And, by the way, I've renumbered the schematic."

[...Sinking feeling...]

There were two choices here: dump the original PCB and pretty much start from scratch, or find a way to renumber the PCB to match the new schematic. Since the customer had not kept any was-is list -- and it could not be generated from the schematic -- I wrote the new schematic and the prior, unrevised schematic to ASCII, and created a database for each schematic giving the locations of each symbol and its reference designator. Matching the two gave me a list of common symbols to both the old and new schematic, i.e., a was-is list. It was not complete, of course, because of the new parts and also because a few of the symbols had been moved, but it was 95% of the job....

But it would have been *much* simpler if the client had not renumbered the schematic until after all the changes had been complete. (At that point, even if a was-is list had not been created, I could have come up with a 100% was-is list for any common symbols, as long as I had the unrenumbered and renumbered schematics with no other differences.) Renumbering quite commonly creates more trouble than it is worth; if it is going to be done, *how* it is done is crucial, else a real mess can be created.




* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * 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] * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

This message is sent using PostCast Server Professional Trial:
http://www.postcastserver.com/

Reply via email to