Dear George and Phil, thanks a lot for the fast answers. Things are unfortunately a bit more complicated and the usually very convenient way using SAINT-SADABS-XPREP has too much limitations for this datasets because
1. It starts with that one datasets has more than 2.000.000 reflections (space group P1, high redundancy and high resolution), so I already have to split the initial datasets in two. (SADABS can just process 2.000.000 reflections, probably something archaic from old days, when computers were not so fast?) 2. I can then of course combine them with xprep but the XPREP (version 2014/2) writes out just merged .sca file. Other formats hkl4, hkl3 are unmerged but can be just used with the command line version of xtriage and phenix merging statistics (which is a part of the phenix suite and needs additional inputs (spacegroup). However as aimless writes out all the statistics too, it is not necessary to run these phenix programs anyhow. Thank you both of you again, for the great programs and support. @Phil I will send you the SAINT manual and a small raw file offlist. @George can I write out unmerged .sca files from xprep and does it also print out CC*? (Of course I always use the latest ccp4, but I hoped the old one might work in my case).
