Te relevance of the weldment comes from the fact that the hinge has only one part number cause we dont buy it in multiple parts. And the angle between the doors cause you cant use peer variables( since the hinge is an lower level assembly ) in alternate members.
Again, weldment or regular assembly yield the same result. They are both assemblies.
Also, you can indeed use variables from lower level assemblies in alternate position members. That's what I've been doing.
Edit: if you click on the hinge in the lower level adjustable assembly then right click on the angular constraint and select "override relationship" it promotes the constraint to the highest level of the adjustable assembly. That's how it's available in the variable table and to alternate position members.
Edit2: You can also do what others have said: don't constrain the angle in the hinge assy. Instead constrain the door angle at the next level up assy while the adjustable hinge assy angle follows along.
I am a bit late to the party on this, but here is how I typically handle door hinges:
Most of our hinges are weld-on, lift-off type pin hinges. I model them accurately as two individual pieces and then bring the two pieces together into a hinge assembly model contrained except for rotation about the hinge pin.
I build my assembly using the hinge assemblies, door, and frame structure and make the hinges adjustable assemblies. Then, I use an angular relationship to control the door opening angle. This works because the rotational axis of the hinges are free when the hinges are set as adjustable assemblies.
With the angular relationship in my assembly I can open open and close the doors as required by adjusting the angular relationship, but I don't have a need to show it both ways on my drawings, so I don't create an adjustable assembly. It would be an easy next step though, if I had the need.
These hinges can be a bit tricky to position because of the weld prep. profile of the hinge so I typically place a sketch (circle) in the frame part where the hinge attaches and axially align the hinge to the frame skecth, and a corresponding circle sketch in the door.
I often use peer variables to drive assemblies buried down below the top level during the design process, but I seldom use family of assemblies.
I am a firm believer in modeling assemblies like hinges accurately as someone else noted, especially if I need them to behave as they do in real life. We obviously buy the hinge as a single part number so my convention is to tack a -01, -02, etc. onto the individual parts that make up the purchased hinge assembly. My parts lists never call in atomic lists, so the hinge parts don't ever appear in my parts lists, only the hinge assembly itself as we purchase it.
As with most things in SE there are a multitude of ways to accomplish the same task and it often boils down to your personal preference or your company procedures.
That is my $.02 for a brisk Saturday morning.
After reading through my previous post, I thought I should clarify my procedure where I mentioned that I used sketches for positioning the hinge and the door. Ultimately, the door swinging action is controlled by an axial alignment between the circular sketch in the door and the circular sketch in the door frame which is colinear with the hinge axis. The door would swing without the hinge even being there as it is just along for the ride as an adjustable assembly. The adjustable hinge assembly allows the hinge to look correct regardless of the door opening.