• TI Thinks Resolved

BQ78350-R1: cell voltage fluctuation on cell 6 and cell 10

Prodigy 95 points

Replies: 10

Views: 128

Part Number: BQ78350-R1

We are seeing fluctuation, or noise, on cell 6 and cell 10 while periodically poling the cell voltages from the BQ78350-R1. This happens systematically across all units we have measured. Is this a known issue? The fluctuations are 100mV-200mV. Is this normal?

  • Hi Steve,

    This is not normal. Can you share your schematic and we can help review for any issues. Additionally, you can view this video that walks through the schematic creation process to help avoid common issues: https://www.youtube.com/watch?v=B5dsIc2C_xA 

    Best regards,

    Matt

     

  • In reply to Matt Sunna:

    Where should I upload / send my schematics?

  • In reply to Steve Coffin:

    Hi Steve,

    I sent you an E2E friend request. You should be able to send me your schematic through a private message.

    Thanks,

    Matt

     

  • In reply to Matt Sunna:

    Hi Steve,

    Thank you for sending your schematic. I think I see the problem. There are two filter resistors indicated in the picture below that are missing on the VC5X and VC10X supplies. These should match the value of your resistor from Cell15 to BAT. Otherwise, I think everything else looks good. 

    Best regards,

    Matt

     

  • In reply to Matt Sunna:

    Thank you Matt. I'll add those resistors to my design. Any particular reason the resistors on VCX5 and VCX10 should match the one on BAT to cell 15? It looks like the reference design uses 100 ohms on the BAT, and 1K on the other 2.

  • In reply to Steve Coffin:

    Hi Steve,

    The Rf and Cf make up the filter for each cell group. Generally, you want to use the same values for each cell group to ensure consistent measurement results for all the cells.

    Matt

     

  • In reply to Matt Sunna:

    Matt, Another question. Can we power REGSRC from a 5V supply which is powered by either the battery, or the charger if the charger is plugged in? The problem is that if the batteries deep discharge due to leakage current and long term shelf storage, the shunt regulator in the reference design can't function. If the pack voltage drops below 14V the system gets into an unrecoverable state. The BMS has no power so it  can  not enable the charge FET's. By powering REGSRC with a 5V power supply that can be powered by either the charger or the batteries, we should be able to recover from accidental deep discharge because the BMS will get power when the charger is plugged in. Do you see any problem with this approach?

    Thanks,

    Steve  

  • In reply to Steve Coffin:

    Hi Steve,

    I think this should be okay. Although, 5V may be low during normal operation (when powered by the battery). There are two figures (Figure 12 and Figure 15) in the Top Design Considerations app note that show the FET voltage vs REGSRC voltage: http://www.ti.com/lit/an/slua749a/slua749a.pdf

    Matt

     

  • In reply to Matt Sunna:

    Matt, I'm using the BQ76200 high side driver to control the FET's. It only requires 1.2V for a logic high so 5V into REGSRC should fine. Do you agree?

    Thanks,

    Steve

  • In reply to Steve Coffin:

    Hi Steve,

    Can you supply 6V to REGSRC? The datasheet minimum is 6V. I checked the characterization data and this is the lowest voltage that the device is characterized. There is no guarantee that the LDO output performance will be acceptable since we do not have data at REGSRC voltages below 6V.

    Matt