For getting a spid's total tempdb space usage try using the pssinfo() function, eg:
select pssinfo(<spid>,'tempdb_pages')
I've got a couple monitoring scripts that use this function and I've found it's quite good at giving me a breakdown of how much tempdb space (#temp tables, worktables) is being used by each spid.
------------
As for the NULL monProcessObject.partitionsize values ...
1 - this table provides a point-in-time snapshot of objects being referenced/managed by the SPID; if your MDA query was run near the start of the murex query it may be the worktables had been created but not yet populated
2 - there could be a limitation with the population of the partitionsize column for worktables, eg, MDA may be presenting the size of the object as measured at the beginning of the query *and* it may not check for a change in size during execution; you could run some tests to verify/refute this hypothesis, eg, for a SELECT/INTO <{perm|temp}_table> does monProcessObject show the table's size growing as the query progresses, or does it show NULL/0 for the entire run?
