" There's an extra face on the inside of the tube. Synchronous wants clean topology."
Why can't SE give the user a clue as to where the problem lies? As it is you get nothing at all.
Amen @bshand. I'm banging my head against the wall trying to figure what's causing these errors.
I re-created the part in ordered and easily created a family of parts with the same shape, but different dimensions. Easy. I think I'm done with sync. I'll just finish the the book so I know how, but plan to use ordered for real work.
I imagine that level of error checking would take a lot of time to write, and it would never be done.
The way I found the error was to find the first way that the dimension failed to drive the model. First, there were red and blue dimensions on the same face.
Second, you have to approach the problem with the idea that there is always a rational solution, and this geometry is pretty simple. So there must be something you can't see in there complicating the situation. So I turned the model wireframe. This is an old troubleshooting trick that I learned a long time ago that helps you find issues where the problem seems supernatural.
Once I saw the extra interior edge, I just deleted that face, and the changes worked as expected.
The problem here isn't with Synchronous, it's with building a good model. Even in ordered, that would be a sloppy model. On a section view on a drawing, you'd get an extra line and wonder where that came from. Not sure how that edge came to be there, but my guess would be that the smaller tube was extruded through, and then cut off on the inside, instead of just extruding up to the big tube.
You might be able to argue that Solid Edge should have merged that saddle with the main cylindrical face, regardless of the technique you used to create it. They are the same radius, with the same axis.
You did several things the hard way. The thin tube with closed ends can be made in a single step with the Thin Wall feature, by just not selecting a face to open. (although to get this to work, you have to change the wall thickness value, it won't work if you don't change the value regardless of what the value is).
Also, you could have just modeled both tubes solid and thin walled it in a single step.
Anyway, I don't think avoiding Synchronous is going to solve your problems.
The real problem is that people dont even try to work properly. This is not a part, this is a tube welded assembly. If this would be modeled correctly it would corect the problem . This is why yuou should REALLY try to do it with frames (tube extrusion). This would help represent the reality a lot better and help for manufacturing drawings ( cut lengths, weld properties,....).
If users are going to be using frames a lot then, yeah, take the time to learn it and set it up. Otherwise there's nothing wrong IMO with just using standard part modeling to get you through what may be an uncommon design in your company. This is besides whether or not the modeling task was well thought out.
I don't agree however with the idea of abandoning sync early on rather than getting better with it.
There are too many strengths with it. But it can be very frustrating especially in more feature rich models when it "inexplicably" fails. We can't always take the time to wait for a Matt to clarify things.
The same geometry could be a blow molded plastic part or a casting, so it's not a given that it's a tube or frame structure. But in general, yes, I do agree that working haphazardly gives you haphazard results.