We have the following points to mention:
In the case of multivolume, multi-file data sets, the assignment for a new scratch volume is identical to the previous volume, which in turn can be traced back to the primary data set assignment on the first volume.
For multivolume files, when a new volume is opened during output, the pool assignment is the same as for the preceding volume. If the assignment rules have been changed since the creation of the original volume, the assignment of the new volume is based on the new assignment rules.
CA TLMS does not assign subpools for foreign tapes, which includes (EXPDT=98000 or SPACE=(1,(1,1))).
CA TLMS assigns tape pools based only on the first file (primary data set) of the volume. Secondary data sets receive the same pool assignment as the primary data set for the same volume set.
Data sets which are not assigned to a tape pool by the assignment rules are automatically assigned to a generic tape pool which contains all tapes not defined to a tape pool.