0
\$\begingroup\$

Some battery types, like lithium-ion are sensitive to over-voltage or over-charging. How do chargers deal the differences between cells?

For example, a lithium-ion cell is 4V, so a 12V battery I guess will have sets of 3 cells arranged in series. The problem arises what if two cells are charged but the third cell is lagging behind? If the charger is applying 12V and two cells are full, but the third is low, then it could lead to an overcharge situation developing. How does the charger detect and deal with this?

\$\endgroup\$

2 Answers 2

4
\$\begingroup\$

Some battery chemistries self-balance well enough to allow this aspect to be ignored, or to be accommodated by compromise solutions, but many modern batteries require specific attention to this aspect.

With eg LiIon batteries the normal approach is to use "balancing' circuitry specifically aimed at ensuring cell balance. Various approaches are possible. It is possible to cut off charge to cells which reach a certain voltage first, so that they "wait for the others", or to reduce charge rates on higher voltage cells so that the others 'catch up'

With LiFePO4 chemistry - some manufacturers make 12V 4 cell batteries for automotive use. Some have exposed intercell connections and others provide no access. The ones with no access claim that it is not needed - but experienced sellers who offer a range of such batteries may suggest that such versions be used only for 'float' operation and that deep discharge applications employ balancing.

4 cell lead-acid car batteries have been used for decades with no attempt to address balancing issues. The application of topping or boost cycle occasionally serves to bring up any lower voltage cells. However, serious users of large capacity lead acid batteries check balance on a per cell basis, measure and record acid specific gravity per cell and concern themselves with issues such as acid stratification.

Ye olde NiCd cell packs were noticeably prone to unbalance issues when more than about 3 cells were used in series. When V_cell_min x N > V_cell_max x (N-1) for N cells then it is possible to have N-1 fully charged cell while one cell can have less or even no charge while the battery voltage is in spec. Eg NiCd battery. 6 cells. Vmin/cell = 1V say. Vmax/cell = 1.2V say.
5 x 1.2 = 6V, 6 x 1.0 = 6V. So 5 charged cells and 6 discharged cells can have the same voltage. As V_cell_charged is actually > 1.2 V "we have problems". NimH is less prone to imbalance issues than NiCd BUT modern NimH cells MUST NOT be trickle charged* - so, while it was often possible to balance a NiCd pack by long enough trickle charging, this must not be done with NimH cells.

*When a NimH cell is overcharged Hydrogen & Oxygen gas are evolved. Earlier NimH cells used a recombination system that converted the H2 + O2 back to water. As energy densities rose manufacturers dispensed with the recombination system so as to pack in more active material. This means that in modern NimH cells, if ANY gas is produced the water is lost permanently - cell pressure will rise, if vents are provided, gas will escape. If vents are NOT provided gas will (still escape) - Hydrogen can escape from Houdini proof confinement (THROUGH steel if necessary - embrittling it on the way). Oxygen has a harder job but will either escape via leakage or worst case cause an explosion.

The limit for inclusion of recombination system was was somewhere in the 1500-1800 mAh range AFAIR. By about 2000 mAh trickle charging was not allowed and certainly at 2500 mAh + capacity NimH cells will be damaged by even extremely small rates of trickle charging.

I long long ago in the early days of NiCd batteries saw a solid metal encased NiCd C cell that had exploded after gross overcharging, but that was rare.

_________________

TI - Battery Cell Balancing: What to Balance and How - 8 pages

TI - Cell balancing buys extra run time and battery life

Battery University - BU-803a: Cell Matching and Balancing Discover that quality cells are the best foundation for a lasting battery pack

Electropaedia - Cell Balancing

LT - Battery Cell Balancers

LTC3300:

enter image description here

\$\endgroup\$
7
  • \$\begingroup\$ I've often wondered whether devices which use multiple cells for purposes of increasing energy capacity rather than power-handling ability might benefit from "rotating" cells, so that usage which would repeatedly drain all cells by 40% and then charge them fully would instead drain cells by 80%, half as often, between charges. Do you know if that's done at all, and how effective it is or would be? \$\endgroup\$
    – supercat
    Jul 22, 2016 at 16:11
  • 1
    \$\begingroup\$ @supercat Alas, deeper discharge has substantial negative effect on cycle life. For LiIon and probably most or all chemistries, discharging a cell half as much twice as often results in more whole-of-life Ah or W,h. \$\endgroup\$
    – Russell McMahon
    Jul 22, 2016 at 16:19
  • \$\begingroup\$ There's an optimal discharge depth, which is why I said 80% rather than 95%. Further, by my understanding most battery packs become useless while most of the cells still have the vast majority of their capacity. If one cell in a six-cell pack has degraded to 20% of normal capacity but the other cells are at 80%, using all cells as a unit would yield 20% of normal battery life and accelerate wear on the bad cell, while rotating through cells could give closer to 70% of normal capacity. \$\endgroup\$
    – supercat
    Jul 22, 2016 at 16:38
  • \$\begingroup\$ @SamGibson Thanks - yes - that was a typo - should have read "... modern NimH ...". I have corrected that and added detail re why this is the case. \$\endgroup\$
    – Russell McMahon
    Jul 23, 2016 at 12:48
  • \$\begingroup\$ @supercat A complex subject (most involving batteries are :-) ). The 80% discharge depth is substantially harder on LiIon cells than 40%. To get REALLY long cycle life you also limit the top end voltage - at the cost of large or very large reductions in capacity per cycle. The LiIon cells in the Mars rovers get 8000 cycles but charge to 3.6V max [!]. I don't think balancing is a major issue if active steps are taken to balance cells in a pack ... \$\endgroup\$
    – Russell McMahon
    Jul 23, 2016 at 12:53
1
\$\begingroup\$

You have two choices: Ignore it and put up with the lifetime and capacity issues or use cell-balancing circuitry that monitors the cell voltage and equalizes the cells (mostly by bleeding off some charge from the cell with the highest voltage, though there are active circuits that can recover the energy.)

One battery protection and cell-balancing part can be seen in the datasheet below, though there are lots of options:

Cell balancing and protection device datasheet

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.