The turbo 35 can monitor each cell, right? It could be that one poor cell is pulling the pack down. Or it could be incorrect charge/discharge settings. Or the whole pack has had a problem, perhaps from misuse.
I've had a couple of problems with my GP3300's - slightly mistreating them (charging before they are fully cool, at a 6A rate. 0.03v cutoff per pack) has caused them to get VERY hot during the charge. I've noticed a drop off in runtime of around 10% on about 4-5 cells out of 3 packs. Very annoying - and essentially my fault. My Sanyo 3000's were more forgiving of slight mistreatment.
So my advice - LET THE CELLS COOL TOTALLY BEFORE CHARGING - I know this is always said, but make sure - one oversight could cost you.
As for part charging - I don't really see the point - you buy the cells discharged, and they shed up to 10% of their charge per day, so a part-charged NiMH will be vitually flat within a week anyway. So long as the cells aren't stored trayed to 0v, I don't believe there will be a problem.