Hello Ltiu,
Perhaps we need to clarify the concept of draining NiMh cells...
When someone talks about totally draining a cell, I think that they are going to use it until the voltage reads 0 volts.
This is bad for the cell.
The design of the cell allows for use to 0.9 volts under moderate load or 1.0 volts under light load. In my mind, if you are totally draining the cell you are going beyond that, and damaging the cell.
The only way to determine the capacity of the cell is to first charge it, let it rest for awhile, then discharge it to the cut off voltage under load. The time of the discharge multiplied by the discharge current will give you the capacity of the cell.
Battery manufacturers use a series of tests to determine the capacity and life expectancy of the cells they sell. These tests are standardized. The capacity test involves charging the cell at 0.1C for 16 hours, let it rest for 1-4 hours, then discharge the cell at 0.2C down to 1.0 volts.
Now that the capacity of the cell is known, then next step is to determine the life expectancy of the cell. The standard calls for using the standard charge and discharge for a minimum of 500 cycles, maintaining 60% of the initial capacity.
The testing continues to check the self discharge rate, and cold weather performance. Additional testing can also be done at higher discharge rates, and using different charging algorithms.
The battery supplier reviews the test data, then chooses whatever capacity number they want, then they label and market the cells. By the time you purchase the cells for use, they may have degraded during storage, or they may have been optimistically labeled. This is why we spend time testing batteries. We are interested in seeing how close to the labeled capacity the batteries are performing to. With graphs we can also review the voltage under load during the discharge to help with circuit utilization or design.
Tom