• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 727 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.

BQ76930 used in 20S LiFePO4 cell , balance issue

jerry Yang
Intellectual 470 points
Part Number: BQ76930

Hi TI BMS Expert,
      My customer is using 2pcs BQ7693003 for 20S LiFePO4 Cells application,  when customer do cell balance test, they met as below issue, please help to check the reason,  Tks.

    1. Customer used P-MOSFET external balancing way to balance cell. the schematic as attached, when turn on Q40 Fet to balance cell 5 , after balance to measure  the 6th cell and 7th cell voltage, the voltage will change 15 to 16 mv .

    2.when turn on cell 1,cell4, cell 7,cell 10,cell13, cell16, cell 19 Mos-fet  to balance multi-cell, the 6th cell and the 7th cell  voltage will change 34mv to 50mv.

    3. so customer want to know if it is normal , after balance , the 6th cell and 7th cell voltage change like above. why turn on balance , only 6cell and 7cell voltage will be influenced

    4. please help to review the schematic ,if is ok,tks.BQ76930 for 20S schematic.pdf

   

   

  • 0
    Prodigy 10 points

    Hi all,

        is there any feedback? custoemer is urgent, tks.

  • 0 in reply to xiaoguo yun
    TI__Guru 54715 points

    Hi Jerry & Xiaoguo,

    1: During balancing at the cell group boundaries the common path resistance can show voltage effects on the adjacent cell:  Balancing cell 6 can show effect in cell 5 or balancing 5 can show effects in cell 6 and possibly cell 7 due to connection of the capacitors.  See http://www.ti.com/lit/an/slua749a/slua749a.pdf figure 8.  Minimize the common path resistance to avoid this, route the current paths to the board connection and use a heavy wire connection to the cell.

    2. 3: During balancing 12.5 ms is allowed during the timeline for the input filter to settle, then measurement begins at the bottom cell of the group and proceeds up.  See http://www.ti.com/lit/an/slua775/slua775.pdf figure 8.  If the input filter is still settling when balancing begins voltage the voltage on the bottom cell of each group is most likely to be affected.  The 1k and 1uF Rc and Cc used on the EVM show this effect, the customer used the same values.  Using smaller Cc such as 0.22 or 0.1 uF avoids this effect.

    4. The 5.6V zener diodes between the VCn inputs of the IC may contribute to the observed voltage differences.  During balancing the balanced cell's VCn lines will move approximately 1/2 of the cell voltage, the adjacent cell will see about 1.5x normal voltage.  If a cell is balanced at 4V the adjacent cell may see 6V during balance.  Balancing cell 5 with the attached schematic will pull down VC5 and D74 may pull down VC6 affecting cell 6 and cell 7 reported voltages. On cells within the same group the pull of the zener may not turn on the external balance circuit, and cell balancing is synchronized within the group so measurement effects are not expected.  The customer should consider if the values of the zeners are suitable or if adjustment is needed on some such as D74 or if a different topology should be used.  Some customers have found value in zener placement like http://www.ti.com/lit/an/slua749a/slua749a.pdf figure 21.

    In the schematic the power filter resistors should typically be 1k: R63, R182, R247