• State
  • Locked Locked
  • Replies 7 replies
  • Subscribers 63 subscribers
  • Views 1086 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ25792: BQ25792 and a battery with PCM

Sergey K.
Prodigy 10 points
Part Number: BQ25792

Dear support,

I've been testing BQ25792 with a 3S Li-Ion battery pack that has protection circuitry (PCM) inside.  In case the battery goes flat (<2.7V per cell), the PCM disconnects a discharge switch between the cells and the battery pack terminal (negative rail only).  At this time only some phantom voltage is present at the battery terminals: around 1.4V~2V.  To reset, the PCM requires that the load is removed and an external charger to be present to the battery terminals.  The way PCM detects a charger is by the voltage applied to the battery terminals: it should be higher than the voltage of the cells (10V~12V usually works, 12.6V is recommended by the manufacturer).

However, low battery terminal voltage means that the BQ25792 does not see any battery at all.  In this case, the charger chip only applies 2.2V (Trickle Charging mode) and gives up after 1 hour.  The problem is that 2.2V does not reset the PCM so it will not show the real voltage of the cells to the charger.  In its own turn, the charger does not start providing higher voltage to the battery terminals until its voltage is high enough.  As a result of this lock-up, the battery cannot be recovered or charged.

Do you have any suggestions / solutions / best practices to resolve such situations when the charger does not apply full charging voltage to the battery pack (as it does not see the real cell voltage due to PCM protection) and the PCM does not reset its protection (as it does not see the full 12.6V voltage applied externally to the battery terminals)?  Batteries with PCM protection are predominantly used these days so it must be a very common scenario, I presume.

TIA!

  • 0
    TI__Guru**** 166866 points

    Hi Sergey,

    In trickle charge, the BAT pin sources a small current (100mA typ) which is supposed to slowly charge a battery pack cells, with open pack protector discharge FET.  The threshold is typically 2/3 of the battery regulation voltage.   Are you saying that the current is not high enough to charge the cells to turn on the FET? 

    If the PCM discharge switch was completely disconnecting the cells, the trickle charge current would charge the capacitors on the BAT pin to the battery regulation voltage (12.6V). Where is the 100mA trickle charge current going?      

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 10 points

    Hi Jeff,

    I agree that if the charger behaved as a constant current source as you described that could reset the PCM.  However, it looks like the BQ25792 IC limits its output voltage to 2.2V in the Trickle Charge mode regardless of the configured number of cells in the battery pack.  This voltage is obviously not high enough to reset PCM protection.

    While in Trickle Charge mode, I measured the charging current (using current sensing resistor R13 on BQ25792EVM board) and it is 0mA (which was sort of expected as PCM had opened protection switch, and even if not, 2.2V was still too low to start charging 3x cells in series).  The charger stays at that level (2.2V / 0mA) for exactly 1 hour (Trickle Charge Timer) and then gives up indicating Charge Error (LED D13 starts blicking and the voltage across the battery terminal drops back to the initial 1.4...2V level).

    It seems to me that the behaviour I have observed matches the BQ25792 datasheet but limiting the voltage to 2.2V in trickle charge mode does not make much sense because it will never recover a multicell battery.

    Best regards,

    Sergey

  • 0 in reply to Sergey K.
    TI__Guru**** 166866 points

    HI Sergey,

    The charger is not regulating 2.2V at BAT.  If you remove your pack, you will see the BAT voltage rise to BATREG value. Other customers have used it successfully with other multicell battery packs.  Unfortunately, your multi-cell pack seems to be incompatible.

    One more thought.  Did you use the shorting jumper to short the shipFET on the EVM?  

    Regards,
    Jeff

  • 0 in reply to Jeff F
    Prodigy 10 points

    Hi Jeff,

    Unfortunately, the charger does not raise the voltage to BATREG and stays at 2.2V - I just double-checked that again.

    Here are screenshots from BMS - it clearly shows the same things as I measure with a DMM:

    1. There is no battery detected (as it is isolated by PCM)

    2. Charger is in trickle charge mode

    3. Charger only supplies 2V to the battery.

    4. Current to the battery is 0mA

    5. System voltage is stable at 9V.

    No, I do not use the shorting jumper on shipping FET as I intend to use that FET in future.

    This is the PCM IC: http://en.cellwise-semi.com/prod_view.aspx?TypeId=67&Id=218&FId=t3:67:3

    The actual PCM circuit is as per the IC reference schematics:

    Best regards, Sergey

  • 0 in reply to Sergey K.
    TI__Guru**** 166866 points

    HI Sergey,

    The ADC does not measure the trickle charge current as it comes from a current source with output at the BAT pin and does not go through the internal BATFET for measurement. I do not see anything unreasonable about the CW1244, except the ground before RSENSE,  unless it is pulling all of the trickle current in order to power up and not leaving any to charge the cells. 

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Prodigy 10 points

    Hi Jeff,

    Thanks for your response.  I understood about the ADC not measuring the trickle charge current (although a separate current source for trickle charging is not shown in the Functional Block Diagram of BQ25792. Page25 of the datasheet).  I have measured the current flowing into the battery by other means: via voltage drop across the sensing resistor (R13 on BQ25792) and also directly with an ammeter - there is no current flowing into the protected battery.

    However, the input current (measured by a separate DMM) does change if the charging is enabled.  The reading changes from 128mA (there is a small system load connected to J3) to 185mA.  The voltage at the battery terminals oscillates between 1.95V and 2.45V as measured with an oscilloscope (the DMM showed that as 2.2V average previously).  The situation does not change even if I unplug the battery: the input current values and the battery voltage remain the same (which is expected as the battery is isolated by PCM).

    So the conclusion is that there is nothing to pull the trickle charge current and the voltage at the BAT pin is still too low to reset the PCM.

    Anything else I could try or that is expected behaviour for this charger IC?

    Best regards,

    Sergey

  • 0 in reply to Sergey K.
    TI__Guru**** 166866 points

    Hi Sergey, 

    The IC appears to be working correctly.  It must not be compatible with your pack.  Can you provide the pack model and vendor?  I will try to order one and test.  

    Regards,

    Jeff