If I get a cell that looks like a self discharger (eg 3.85v after 4 weeks) I test it again. Many times they have been fine the second time. This is probably due to me removing them from the charger too early.
On my LittoKala testers on the full test cycle, they charge, discharge, display a value and charge again. It shows END when the discharge is done and then charges it back up again. There is no indication of the end charge cycle completing.With the constant current and then constant voltage charge curve, if I look at it and the voltage says 4.2, I might be pulling it off at 80% charge if it has only just entered the constant voltage part of the charge. (so it might come off saying 4.2v, but resting voltage might have dropped back to something like 3.9v very quickly)
The other thing worth considering is that two 2400mah cells can have quite different discharge curves. I was discarding any cells that went under 4v after 4 weeks. When I looked at them more closely though, some cell models might have 10% of their capacity above 4v, others might dip below 4v after 2% of their capacity is discharged. I've found some cells at 3.98v after 4 weeks were still at 3.98v after a further 3 weeks (I'd use these).
If they dont work first time they are out. If they would give more the 2nd time i qould question "why" 