[ccp4bb] correct pdb encoding for split-personality solvent HETATM?

[Alastair Fyfe]

On the subject of  not (willfully) contaminating the pdb database, any 
suggestions on the following are appreciated. I am refining the solvent 
structure of some RNAs. Inspection of difference and anomalous 
difference maps suggests that some closely-spaced positions are 
alternately taken  by distinct entities with partial occupancy.  One 
pattern suggests partial occupancy by a Ca++ ion with a nearby site 
partially occupied by water (confirmed by an anomalous peak only for the 
former). I  am refining this with shelxl and expressing this in  the ins 
file is straightforward (and I'd like to express much gratitude for the 
multi-core version of shelxl!). However it's not clear to me how to say 
this in PDBese : choosing the HETATM residue name as either HOH or CA 
and assigning alternate positions with HOH and CA in the atom name field 
seems to make trouble for pdb reader software, eg mmdb.
thanks,
Alastair Fyfe