After reading discussions here on Master Parts, I wrote myself a note saying that master parts could be inserted in both the top assembly and reusable sub assemblies. Then I thought that it was nonsensical to build a reusable sub assembly on a master part created for a unique top assembly.
Now I am thinking that master parts may be good for laying out complex assemblies in resource lean segment assemblies. When the segment assemblies are inserted into the top assembly there will be multiple instances of the master part, and its design table, in the top assembly.
If one instance of the design table is changed, will the others be independent, or in lock step with the changed one?
Was this the problem with Christopher's windows?
Is the workaround to insert uniquely named, but identical, configurations of the design table in the top assembly?
If instead of master parts I employ Layout planes in sub assemblies, then when the sub assemblies are inserted into a top assembly, are just the solids inserted, or are the layout planes inserted too?
If the former, then what happens to mates to the layout plane?
Which SW training course addresses these issues?
John,
Can you clarify a few things here. What do you mean by resource lean segment assemblies and reference the previous thread to which you refer. Are you talking about this in the context of top down design? Wht do you mean by master parts?
I say this because if you are talking about top down design in assemblies this is an issue that I'm giving some thought to at the moment. Cannot believe that I've been using the software for this long and still cannot come up with decent method for doing this.
The main approach taken by SW, and confirmed in discussions today with my VAR is to use in context relations. I've done the advances assembly course and we did not discuss top down design much. The problem I have with incontext relations is that they can easily become corrupted.
If your reference to a master part is what I think then I suspect the main reason for this is to allow editing by others working on the model.
Another approach is to use equation driven parts that get mated into the main assembles. In his thread with the record length thread title on the subject, Mauricio Martinez-Saez discusses this in detail. That is a method I'm trialling at the moment. It seems to be more robust than in context relations but it is time consuming and fiddly if you are using the simple approach of highlighting dimensions to set up equations.
I am interested to hear others thoughts on the topic.