I'm trying to figure out how systems work in Revit MEP 2012, as well as the "system type" which is now user-defined in 2012. Say I have 2 systems: SA-1 and SA-2. Diffusers fed from AH-1 are put on the SA-1 system and diffusers fed from AH-2 are put on the SA-2 system. In Revit 2011, if you drew a duct that connected a diffuser on the SA-1 system to a diffuser on the SA-2 system, the diffusers would remain on their assigned systems, and the ductwork connecting them would be on one system or the other, though I never figured out how it decided which system it would be on.

Now, in 2012, if you do the same thing and accidentally draw a connection between two systems, it combines the two systems, so AH-1 and AH-2 would be on the same system! Why would you want it to do that? A simple mistake, like connecting to a hidden duct, and you lose all the information you put in about what diffuser is fed from what air handler.

We also now can make our own "system types" which we couldn't do before. How are people using these? I'm thinking about setting up the model so that every air handler would be its own system type. Following the example I described above, SA-1 and SA-2 would each be both systems and system types. The only system of the type "SA-1" would be the system "SA-1". If they're different system types, they won't override the system assignments of diffusers when ducts are crossed, which seems to work around this stupid new feature of overriding defined systems for diffusers based on connectivity.