Simply reading a 1 tb disk (at around 150 mb a sec) should take about two hours, if my maths is correct. Recuva does rather more, it runs a normal scan first (which should be a few minutes only) and during the deep scan has to correlate the used (live) clusters with the unused, compare the first section of each cluster with a table of known file headers, build up a list in memory of what valid files it has found, and do all the other things I can't think of at the moment or just don't know about. In any event three days is silly. I doubt if anything good will come if Recuva ever finishes, but what it's doing heaven only knows. I guess it's an HDD, USB?