Charging a lithium pack inner workings...

Thread Starter

Externet

Joined Nov 29, 2005
2,625
Hi.
Been charging packs that I have no dedicated charger for by applying a very fixed voltage never exceeding 4.2V times the number of cells to the output terminals of the pack. Always has worked fine from just a dimmer + a selectable winding transformer + rectifier + filtering + voltmeter. (Works up to ~3 Amperes / 85 Volts)

But I wonder how the internal charge manager circuitry works. If a higher voltage is applied to the pack terminals -say 5V times the number of cells-; would the charger manager circuitry keep the cells at the proper voltage under the maximum ?
Is that the way chargers work ? Feeding more voltage than needed and leaving the management to the internal circuit board; or, is the voltage applied hyper regulated at the charger output ?

These days am charging half a dozen 40V Kobalt packs. Applying voltage until the pack shows 42.0V and then disconnect.

1756305555068.png
 

KeithWalker

Joined Jul 10, 2017
3,603
I can't answer your question about how the chargers/manager circuitry works for your battery packs but I do have a recommendation for you. I charge batteries from a DC power supply, set to 4.2V per cell, the same as you. I limit the current to a safe level. It works well but you have to be aware that if one cell fails, the charge will ruin the whole pack. I avoid that by measuring the voltage of each cell before and during charging. That way, I can replace any cell that is failing. That will be difficult in your case as you are charging assembled commercial packs.
Check the battery specs to see what chargers they recommend. That will give you a better idea of whether they have built-in charge monitoring circuitry and what it does. Some packs (R/C) have no built-in circuitry and rely on the charger to do the monitoring, but they have an extra multi-pin connector for the charger to access each cell.
 
Last edited:

Irving

Joined Jan 30, 2016
4,996
A proper charger monitors the voltage on each cell. Initially the charger applies a compliance voltage greater or equal to n * 4.2 (or whatever the specific chemistry wants, eg 3.65 for LFP) but at a fixed constant current, typically a %age of the cells rated C capacity eg 200mA for a 2000mA 18650 cell or more for fast chargers (with other precautions). When the individual cell gets to its rated voltage the charger starts bypassing that cell to reduce its charge current while maintaining the same charge rate for cells that are still catching up. On a well designed and pre-balanced pack (ie cells selected as matched sets) there shouldn't be much difference between cells, but that often isn't the case. Once all cells are close to their terminal voltage and are needing bypassing the charger switches to constant voltage of n * terminal voltage and continues balancing while gradually dropping the charging current until the charge current reaches, typically, 1/100C (ie 0.1A for a 10Ah pack). This constant current/constant voltage (hence often referred to as CC/CV charging) is typical for most Lithium-based chemistries. Done well its the fastest and safest way to charge a large pack - a good charger will monitor each cell voltage and should raise an alarm/stop charging if an individual cell either increases voltage too quickly or is falling behind, as this indicates a faulty cell that could be dangerous. I personally would be extremely wary of fast charging any pack that I couldn't monitor the individual cell voltages.

These graphs illustrate a good charger at work. Left shows the pack, at around 10% capacity, but cells all near identical voltage. Fast charging at 30A (60Ah pack) you can see individual cells charge differently, but at the end of the CC phase they are all balanced to within 5mV as they all creep towards the final 3.650 terminal voltage. The balancing CV phase takes almost as long as the CC phase and 'bad' chargers stop early but that's a false economy as a properly balanced pack will achieve a better balance at high discharge rates and depths and will last longer. This particular pack is nearly 7y old and after >700 charge/discharge cycles is still >97% original capacity.

1756310031959.png
 

Thread Starter

Externet

Joined Nov 29, 2005
2,625
...be aware that if one cell fails, the charge will ruin the whole pack...
Correct. In relation to this; wouldn't the BMS abort/alarm/terminate charging process if a cell is kaput ?


...a good charger will monitor each cell voltage...
You say 'charger'. Did you mean the BMS circuitry inside the packinstead of the 'charger' ? An external charger, to monitor each cell would need a dozen+ pin connectors ! Which I have never seen, less at 80V packs or 120V ! at any factory dedicated charger. Here the 80V (20cells) has 4 pins and the 120V (30cells) has 5 pins pack-to-charger.

1756320453522.png
 
Last edited:

Irving

Joined Jan 30, 2016
4,996
You say 'charger'. Did you mean the BMS circuitry inside the packinstead of the external charger ? An external charger, to monitor each cell would need a dozen+ pin connectors ! Which I have never seen, less at 80V packs or 120V ! at any factory dedicated charger. Here the 80V (20cells) has 4 pins and the 120V (30cells) has 5 pins pack-to-charger.
Yes, the LFP packs I build have an internal BMS with active balancing and I'd expect those to have the same. However don't let the lack of pins fool you. The newer breed of chips for EV battery management are distributed systems. The charge control and CC/CV management is done in the external charger and the balancing/cell voltage monitoring is in the pack and the comms between them is a fast differential serial link, so 4 or 5 wires makes sense, especially if its a dedicated charger for the pack or a range of packs.
 

Thread Starter

Externet

Joined Nov 29, 2005
2,625
4 or 5 pins only make sense because one is resistive signature and other is thermal limits. There is NO multiple pins for the charger to monitor each cell !
The BMS does that, not the charger !


Rewording the question: A lithium pack with all its cells good; receives the end-of-charge voltage from the charger or receives more than that leaving the BMS to manage charging ?
 

BobTPH

Joined Jun 5, 2013
11,463
Rewording the question: A lithium pack with all its cells good; receives the end-of-charge voltage from the charger or receives more than that leaving the BMS to manage charging ?
I found a compatible charger on Amazon. It clams to output 40V.
 

Thread Starter

Externet

Joined Nov 29, 2005
2,625
Hi. If outputs 40 V will not reach the 42.0 V needed as end-of-charge voltage to fully charge that pack, right ?

Am not after buying a proper charger, but learning how is it supposed to work. My home-built contraption works fine as built for any battery I want to charge as can be dialed to any V. But what would happen if this home-made supply is set to feed 50V instead to the 40V pack; would its internal BMS control the excess and end properly charged ? Is not one of the functions of a BMS to prevent overcharge ?
 
Top