Charge Algorithm Comparison
This test will demonstrate how different charge algorithms determine the capacity and cycle count of NiMH batteries. The charge algorithms in question are
- negative delta V of 3mV (-dV) to simulate overcharge
- zero delta V (0dV)
- voltage termination at 1.48V (1.48V)
All tests are carried out on VARTA 2100 cells for the same reasons I used them in my High Current vs Low Current test.
The test conditions are the same for all cells - they are cycled with 1.5A current and discharged to 0.9V cut-off.
Results
| Algorithm | Capacity | Cycles to 80% | Cycles to 64% |
|---|---|---|---|
| -dV | 2023 (100%) | 139 (90%) | 176 |
| 0dV | 2024 (100%) | 154 (100%) | 199 |
| 1.48V | 1806 (89%) | 182 (118%) | 240+ |
Conclusion
-dV and 0dV yeald the same capacity, because both operate in overcharge region where any additional charge is completely lost as heat. 0dV exposes the cell to shorter damage which results in slower rise of internal impedance and more cycles. Therefore 0dV is better than -dV.
Some chargers terminate on voltage. An arbitrary 1.48V was choosen to compare with -dV and 0dV. Under the given conditions it resulted in less capacity and more cycles, and left the cell in still usable state when the test completed. On the other side the -dV and 0dV cells were totally worn down.