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.

BQ77PL900EVM Charging and Balancing

Other Parts Discussed in Thread: BQ77PL900

Hi,

I am testing the BQ77PL900EVM in standalone mode with 7s Li ion and I was able to get it charged to 4.056V; however, it is no longer charging. I was also testing balancing, and made one cell lower or higher than the rest; however, the system just responds with status of DSG and OV. I've tried looking through the forums and also tried manually setting non adjacent cells to balance in host mode; however, I don't seem to be getting any balancing, or it is just taking more than a few hours to have any change. What can be causing this problem? Is it a setting problem? My OVH is set at 200mV and my OV is 4.2 while my delay is 1000ms. I've enabled CBEN, ZVC, and SOR. ALso do the FETS get warmer when balancing? My FETS don't seem to get warmer although they do seem to be working when measured.Thank you

  • The part will operate balancing itself only in standalone mode.  If you go to host mode you must implement the balancing control and algorithm.

    Staying in standalone mode, when you charge from normal to the OV threshold for the delay (4.2V from your setting), it will go to OV mode and begin balancing until the cell voltages fall below OV-OVH, or 4.0V.  Remember there are tolerances.  Generally I would expect it to still be balancing at the 4.056V if it is still in OV mode (OV status).

    Cell balancing is low current with the bq77PL900 and its EVM.  At 4.2V current will be nominally < 4.2/(2 x 510) = 4.1 mA.  There will be some resistance internal to the IC and the IC operates with a duty cycle to measure during balance.  It can take a long time to balance out a forced cell mismatch. The power FETs are not involved in balancing and will not heat, Balancing dissipation will be in the input resistors for the cells balanced and some in the IC.  Since the EVM was designed with the "solution box" approach with the circuit in a small footprint with no test points for VCx pins, the best way to observe balancing will be to turn the EVM over and measure the voltage across the input resistors R4 through R14, see the user's guide pages 20 and 22. A cell that is balancing will have a positive voltage across its top resistor and a negative voltage across its bottom resistor.  This may be simpler to see if one cell is high since that is the cell which would be balancing.  Since it is voltage based balancing remember that the IC has a tolerance on its voltage measurement.  To be certain which cell will be balanced, a larger voltage offset is desirable which may mean a large capacity difference which will take a long time for the part to correct.

    If you go to host mode and balance, the balancing operation can be observed in the same manner.  In host mode you will know what cell you are balancing regardless of voltage.  You can observe balancing without adjusting the cell voltages.

  • Hi,

    Thanks for the response, so you are saying that the only way to find out if it is actually balancing is by checking the resistors for each cell? Therefore, it would be a good idea to add some test points when making my own board. I got the it to charge to 4.2 by setting OV to 4.3V and OVH to 0.1V, so the charge voltage has to be set to anything less <= (OV-OVH)*number_cells and > current pack voltage. So the cells will only ever fully charge to (OV-OVH) since anything higher is over voltage? So it is safe to set the OV to 4.3V even though the batteries are rated for 4.2V? Thank you for all the help

  • Right, if you are in host mode you will know what register bits you have commanded to balance, but the part does not signal that it is balancing or what it is balancing.  Measuring the resistor voltage is the certain way.  Measuring the VCX - VCX-1 pin voltages is also good.  Test points are very helpful.  For either case, check how your meter responds to the duty cycle, it may give a better reading with averaging or some special mode selection.

    Check with your cell manufacturer for what is safe for the cell, and consider the IC measurement tolerance. Typically you may want the OV set just above the normal cell voltage and within a safe operating point of the cells. (Standalone) balancing will never occur unless the pack becomes inbalanced.  If you want balancing to run consistently on charge, you would want to set your OV lower so that the normal charge pushes the IC into OV so that balancing occurs.  OVH must be large enough to allow balancing to run some time, but small enough so that a reasonable load pulls the part out of OV and turns on the charge FET.

    Balancing in host mode gives you many more options for balancing algorithms.

  • Thank you for the help :)