Hans Kroeger
Solar Enthusiast
- Joined
- Dec 30, 2020
- Messages
- 165
I am charging with CC_CV as follows
bulk: CC <= 0.25 C (A)
absorb: 3.5 V, cutoff @ 0.02 C (A)
float: 3.35 V
I use a JK 2A active balancer starting at 3.48 V with balance trig. voltage set to 6 mV.
As a consequence balancing takes place during the absorption phase at a time when the charging current is tapering off from high to low. The week cell in my 4S battery shows a voltage bump being caused by a slightly increased internal resistance of the week cell. Therefore the balancer transfers charge prematurely from the week cell to the strong cells causing an imbalance, which is contraproductive. Once the charging current becomes very small, as does the voltage drop across the cell resistance, the true cell voltages are finally seen by the balancer. Now the missing charge from the week cell must be transfered back from the strong cells. This fact has been described here by various members.
Running multiple tests I realized that most of the charge being transferred during balancing was caused by wrong voltage readings at the cell terminals. The voltages at the terminals include the voltage drop across the cell internal resistance, including bad connections of bus bars etc.....
As a consequence a good balancing procedure requires the charge current being low when starting balancing.
When I modified my balancing procedure such that it started when the absorption current went below 0.01 C (A), the balancing time suddenly became very short.
During various tests I did not only verify my "new" balancing procedure, but also learned that only 2 facts are the main reason for cells to become unbalanced:
1. premature balancing
2. spread of self discharge
When I tested various scenarios I also found out, that the JK 2A active balancer provides only 1.35 A balancing current. This balancer also has a passive balancing function!! It dissipates more than 2 Watt during balancing.
Attached are 2 drawings trying summarise some of my findings.
bulk: CC <= 0.25 C (A)
absorb: 3.5 V, cutoff @ 0.02 C (A)
float: 3.35 V
I use a JK 2A active balancer starting at 3.48 V with balance trig. voltage set to 6 mV.
As a consequence balancing takes place during the absorption phase at a time when the charging current is tapering off from high to low. The week cell in my 4S battery shows a voltage bump being caused by a slightly increased internal resistance of the week cell. Therefore the balancer transfers charge prematurely from the week cell to the strong cells causing an imbalance, which is contraproductive. Once the charging current becomes very small, as does the voltage drop across the cell resistance, the true cell voltages are finally seen by the balancer. Now the missing charge from the week cell must be transfered back from the strong cells. This fact has been described here by various members.
Running multiple tests I realized that most of the charge being transferred during balancing was caused by wrong voltage readings at the cell terminals. The voltages at the terminals include the voltage drop across the cell internal resistance, including bad connections of bus bars etc.....
As a consequence a good balancing procedure requires the charge current being low when starting balancing.
When I modified my balancing procedure such that it started when the absorption current went below 0.01 C (A), the balancing time suddenly became very short.
During various tests I did not only verify my "new" balancing procedure, but also learned that only 2 facts are the main reason for cells to become unbalanced:
1. premature balancing
2. spread of self discharge
When I tested various scenarios I also found out, that the JK 2A active balancer provides only 1.35 A balancing current. This balancer also has a passive balancing function!! It dissipates more than 2 Watt during balancing.
Attached are 2 drawings trying summarise some of my findings.
