Li-Ion BMS

Home resources white_papers wp_charger_control

White Paper - Charger Control

Effective ways to let the BMS control the charger

On-off control anchor

The BMS must be able to control the charger to protect the battery pack and to enable balancing.

  • With a top-balanced pack, the BMS turns off the charger only once, at the end of charge
  • With an unbalanced pack, and without a charger that the BMS can control down to the balance current, the BMS turns the charger off and on many times during balancing
  • With an unbalanced pack, and with a charger that the BMS can control down to the balance current, the BMS turns off the charger only once, at the end of charge

(The above is true of all chargers, even those that have a "Li-ion profile".)

At the very least, the BMS must be able to turn the charger on and off (the ability to also control the current is optional).

There are 3 ways for the BMS to turn the charger on and off:

  1. Low power (signal) control, for chargers that have such a provision
  2. Relay (or contactor) at the charger's AC input
  3. Relay (or contactor) at the charger's DC output
3 ways to control a charger

1: Signal control

There are two ways of doing so:

  • Wire between the BMS and the charger's control input, possibly through a relay or opto-isolator if isolation is required;
    Ideally, this must be a "fail safe" connection: open = charger off; closed = charger on
  • Communication link (such as a CAN bus) between the BMS and the charger; a bit in a message turns the charger on or off

This is by far the best way, as it is simple, low cost, and in general is the most reliable.

2: Relay at the charger's AC input

A relay or contactor with contacts rated for AC is placed between the AC line and the charger's AC input. Its coil is driven by the BMS.
Ideally, this must be a "fail safe" system by using the relay's NO (Normally Open) contacts: open = charger off; closed = charger on

This is the best solution for chargers that do not have provisions for low power control.

However, this is disadvantageous if the charger includes a low voltage output (such as 12 V), which will become unavailable when the BMS turns off the charger.

3: Relay at the charger's DC output

A relay (or contactor) is placed between the charger's DC output and the battery. Its coil is driven by the BMS.

This is the worse solution, as it has many disadvantages.

  • If the BMS opens the relay and disconnects the charger from the battery while charging at full power, bad things may happen:
    • Some chargers simply blow up (Manzanita Micro)
    • Other chargers recover from it, but emit an extremely high bursts of EMI, which disrupts the operation of nearby electronics
    • Other chargers go into an over-voltage fault, which must be manually reset
  • If the BMS closes the relay when the charger output is 0 V, the inrush of current from the battery to the output capacity of the charger may weld the relay's contacts
  • DC rated relays and contactors are less available and more expensive than AC rated ones

Primary reasons for placing a relay between the charger and the battery are:

  • The charger is off-board (e.g.: this is a plug-in vehicle using a DC charging port, and the charger is stationary)
  • Having to meet regulation written by people who feel that the charger should only be connected to the battery if actually charging
  • Stationary applications using a non-isolated charger: when the load is connected, it will be also connected to the AC line voltage through the charger, which can be a safety issue

Secondary reasons to isolate the charger from the battery include:

  • The battery pack has high impedance, and the charger's output capacitors are not able to handle the ripple generated by the motor controller
  • The charger has a resistor permanently across its output (it will discharge the battery)

These should be weighted against the risk associated with placing a relay between the charger and the battery, and may be better solved by using a rectifier instead of a relay.

Redundant methods

If you don't trust the charger to shut off, and you want to add redundancy, then use the signal control method to turn the charger on and off during normal operation; in addition, use a relay between at the charger's AC input that is driven by a Fault output of the BMS, which is only triggered in case of extreme overvoltage of any cell. That way, the relay is never switched under load, except in case of emergency.

As an additional safety measure, make absolutely sure that the CV (Constant Voltage) setting of a CCCV charger is set for the maximum cell voltage times the number of cells in series. For example, for 100 LiFePO4 cells in series, set the charger for 3.6 V x 100 = 360 V maximum.
This will protect a top-balanced battery even if the BMS is unable to shut down the charger: the current will decay to 0 when the battery is full, naturally. However, this will not protect an unbalanced battery pack.

Current reduction anchor

While not required, it may be desirable to be able to reduce the charger's output current:

  • Because of extreme temperature (hot or cold)
  • Because one of the cells is full and being balanced at a low current

In applications using large battery packs, typically the current from the charger is far less than the battery can handle. If so, usually there is no reason to reduce the charger current.

For example, a 200 AH, 500 V pack may be charged with a 3 kW charger; that charger can only put out 6 A, which is a current of 0.03 C, far less than the 0.1 C termination current specified by the cell manufacturer, and low enough that it can be used to safely charge a Li-ion battery even below 0 °C.

It's nice to be able to reduce the charger current down to the balancing current (say, 100 mA); that way, the charger is only turned off once, at the end of charging; also, the most charged cells are not subject to many cycles of being fully charged by the charger and then slightly discharged by balancing.

However, those are minor issues. Therefore, again, usually there is no reason to reduce the charger current.

Having said that, if it's easy for the BMS to reduce the charging current, one might as well implement it.

CCL control of a charger

There are two ways of doing so, using the CCL (Charge Current Limit):

  • Wire between an analog output of the BMS and an analog input on the charger to reduce the current
    Often, this is a temperature sensor output, and what gets reduced is the voltage, not the current.
  • Communication link (such as a CAN bus) between the BMS and the charger; a value in a message sets the charger's maximum current

The CCL can be used by itself, without on/off control, if the CCL is able to reduce the charger current down to zero.

Davide Andrea, Elithion, 5/10/12

© 2008~2017 Davide Andrea. All rights reserved, except where noted by CC mark. Page published on May 10 2012. Graphic design by morninglori