Run time eval with log for the one day where you say the problem is.
Expand the day fully.
Find all ADDDB9000.
You will know where the time type is being added to.
It is possible that on one day the addition is occurring more than once in different PCR's; or the same PCR called more than once.
You can also see this when you expand the time eval schema fully with RPDASC00 program with selection inputs:
- schema = main schema name
- schema type = T for time eval (note it is C for PY)
- all other sel.inputs = *
Execute to show the expanded schema. Then find all ADDDB9000.
This expanded schema can be saved as a .TXT file for documentation of major changes. This can also be used to restore a past version of a PCR or part of it.