KBWaldron
Solar Enthusiast
I’ve been thinking, and as my wife would be the first to tell you, that can be a dangerous thing. Mostly, she says, because I’m not equipped for it.
This is with regard to the various parallel and serial cell combinations we throw together in order to achieve our needs. And in particular with regard to the use of battery balancers and getting the most power out of our battery banks.
So, here are my thoughts:
No two cells are exactly alike. Some reach a certain voltage level, whether charging or discharging, faster than others. This is due to them having an innately lower capacity despite having the same physical characteristics with the other cells in the bank.
Connecting purely in series (S) results in the bank not being able to utilize the complete capacity without risking over-discharging the lowest capacity cell. Conversely the bank cannot be charged to it’s highest capacity without risking over-charging that same cell.
This is where a battery balancer comes in to play. Effectively, by balancing the lower capacity cell with higher capacity cells it diverts energy to or away from the higher capacity cells from the lower capacity cells, until their voltages match. Thus when charging or discharging the complete capacity of the banks, albeit an unbalanced capacity, can be utilized.
Now, what additional energy percentage, on average, does one obtain for the average series set when using a balancer? Interesting question.
Now then, if we put series cells in parallel (SP) with other series cells, we need an additional balancer for each new set of series cells. Therefor more cost. Due to the parallel connection the multiple series sets maintain a balanced voltage between the sets, as when either discharging or charging any differences between each sets SOC will take energy more from the high set to low set. If one of the sets has an innately lower capacity then it will not limit the overall capacity as, during charging it will reach a higher voltage first and cause the other sets to take more current in order to maintain the voltage balance. Similarly during discharge, when the lower capacity voltage drops faster it will cause the other sets to provide more current to maintain the voltage equivalence.
So far, so good, I think.
If one connects cells in parallel (P), then when charged, they will act as one cell, and any imbalance in capacity will self-balance due to the effect discussed above. Thus when charged they will all achieve full capacity, and similarly discharged to a common SOC, albeit a different absolute energy content per cell. The difference being that some cells are taking on more of the workload.
Connecting a number of these parallel set in series (PS) to achieve the required voltage and adding a battery balancer should then allow the full capacity of the complete battery to be used. The difference is that only one battery balancer is required.
The one difference, besides cost, is that the second arrangement will, when under heavy load, generate a need for higher balancing currents, thus larger wires, and possibly a more robust balancer. More cost?
For those who are building their battery packs from single cells it seems prudent to me to match the cells in parallel so that each parallel set has the same overall capacity. Note I did not say each cell in a parallel set has the same capacity, only that their total capacity match all the other parallel sets' total capacity. Thus they should then not require a lot of balancing current.
Does this all make sense? From the above logic, and I don’t necessarily say it is correct; remember my Wife; it would seem to imply that a PS arrangement is better than an SP arrangement if only for the reduction in battery balancer investment.
This is with regard to the various parallel and serial cell combinations we throw together in order to achieve our needs. And in particular with regard to the use of battery balancers and getting the most power out of our battery banks.
So, here are my thoughts:
No two cells are exactly alike. Some reach a certain voltage level, whether charging or discharging, faster than others. This is due to them having an innately lower capacity despite having the same physical characteristics with the other cells in the bank.
Connecting purely in series (S) results in the bank not being able to utilize the complete capacity without risking over-discharging the lowest capacity cell. Conversely the bank cannot be charged to it’s highest capacity without risking over-charging that same cell.
This is where a battery balancer comes in to play. Effectively, by balancing the lower capacity cell with higher capacity cells it diverts energy to or away from the higher capacity cells from the lower capacity cells, until their voltages match. Thus when charging or discharging the complete capacity of the banks, albeit an unbalanced capacity, can be utilized.
Now, what additional energy percentage, on average, does one obtain for the average series set when using a balancer? Interesting question.
Now then, if we put series cells in parallel (SP) with other series cells, we need an additional balancer for each new set of series cells. Therefor more cost. Due to the parallel connection the multiple series sets maintain a balanced voltage between the sets, as when either discharging or charging any differences between each sets SOC will take energy more from the high set to low set. If one of the sets has an innately lower capacity then it will not limit the overall capacity as, during charging it will reach a higher voltage first and cause the other sets to take more current in order to maintain the voltage balance. Similarly during discharge, when the lower capacity voltage drops faster it will cause the other sets to provide more current to maintain the voltage equivalence.
So far, so good, I think.
If one connects cells in parallel (P), then when charged, they will act as one cell, and any imbalance in capacity will self-balance due to the effect discussed above. Thus when charged they will all achieve full capacity, and similarly discharged to a common SOC, albeit a different absolute energy content per cell. The difference being that some cells are taking on more of the workload.
Connecting a number of these parallel set in series (PS) to achieve the required voltage and adding a battery balancer should then allow the full capacity of the complete battery to be used. The difference is that only one battery balancer is required.
The one difference, besides cost, is that the second arrangement will, when under heavy load, generate a need for higher balancing currents, thus larger wires, and possibly a more robust balancer. More cost?
For those who are building their battery packs from single cells it seems prudent to me to match the cells in parallel so that each parallel set has the same overall capacity. Note I did not say each cell in a parallel set has the same capacity, only that their total capacity match all the other parallel sets' total capacity. Thus they should then not require a lot of balancing current.
Does this all make sense? From the above logic, and I don’t necessarily say it is correct; remember my Wife; it would seem to imply that a PS arrangement is better than an SP arrangement if only for the reduction in battery balancer investment.
