• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 63 subscribers
  • Views 3099 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ76940: Schematic for multiple pack configurations, and removing discharge transistor

Dillon Stewart
Prodigy 80 points
Part Number: BQ76940
Other Parts Discussed in Thread: BQ76920

Hello,

I have two questions:

  1. I am using the BQ76940 in a design where I would like to give the end user the ability to choose the number of cells they want to use to the best of the device’s ability. (9s to 15s) I want to make sure this schematic will work. It is for a 14s battery pack. I have shorted the VC13 pin to VC14 after the resistors and capacitors. Will this work correctly? I could not find any information other than which pins to short in the datasheet.
  2. I have not finished my schematic, but can I leave out the pack discharge transistor, and just include the charge transistor and coulomb counter? I do not need discharge protection.

Thanks very much for the help!

Regards,

Dillon

  • 0
    TI__Expert 6085 points

    Hi Dillon,

    Since you are not using an external balancing system, I would prefer the VC14/VC13 to be shorted at the device pin. Hence you can save a entire leg of passive components. If you going with the  already drawn,  populate Rc17 with 1K as per schematic. Remove Rc16; move the connection to short out the Cc17.

    (2) if not using the  SPR/SRN connection with sense resistor for current measurement and your application does not mandate OCD, SCD, please  connect

    SRN/SRP to VSS.

    thanks

    vish

  • 0 in reply to Vish Nadarajah
    TI__Expert 6085 points
    correction with Rc17 with 1K as in schematics *NOT* with zero ohms
  • 0 in reply to Vish Nadarajah
    Prodigy 80 points

    Thank you so much for your help so far!

    I have made the revisions you have suggested (I removed the leg of passive components), and this is the final schematic if there are no errors. Could you please check over to make sure there are no errors?

    Thanks again for all your help Vish!

    Regards,

    Dillon

  • 0 in reply to Dillon Stewart
    TI__Guru 54715 points

    Hi Dillon,

    A few comments:

    • You must keep Cc16 to connect between BMS1 pins VC12 and VC13 for proper filtering.
    • The net names "VC10X" and "VC5X" are used for nets different from the pin names which is OK, but the comon node for Cc7 and Cc8 should connect to TS2GND, the common node for Cc13 and Cc14 should connect to TS3GND
    • Q1 is often used with a drain resistor to reduce power in the transistor and for isolation from transients, you might see the application note bq769x0 Family Top 10 Design Considerations section 9.  A blocking diode maintains voltage during short circuit or heavy load.
    • Rf1, Rf2, Rf3 are rather small for a '940 design for the reasons in section 3. 1k is more typical, but you will know your system best.
    • Cc1 to Cc18 are 1uF which gives good filtering, but if you plan to use balancing you may want smaller values, see section 4.

    Hope these are helpful, others readers may have valuable comments too.

  • 0 in reply to WM5295
    Prodigy 80 points

    Hi WM5295,

    Thank you for that comprehensive list!

    You say that I know my application best, but to be quite honest I do not. I used the evaluation board schematic and the typical application as reference for my design. I read through the design considerations and followed as best as possible. I really appreciate all the help.

    I fixed all the issues you described I believe. Hopefully this is the final schematic check.

    Some things that I am not sure are correct as I could not find any information in the datasheet or the design considerations:

    1. Rgate resistor value

    2. Rd resistor value

    3. Rdrain resistor value

    4. Cc1 - Cc18 values - I made them smaller but I was not sure how much smaller would be best. I cut the value in half. I am balancing a 9 ah 18650 battery pack. 

    Thanks again for all your help!

    Regards,

    Dillon

  • 0 in reply to Dillon Stewart
    TI__Guru 54715 points

    Hi Dillon,

    1. The Rgate resistor is not strictly needed but provides some isolation for the transistor from the circuit in case of transients such as ESD.  100 to 100k or possibly wider are OK.

    2. Rd will drain any peak-detected transient off of the Q1 drain back to PACK+.  The smaller the value the faster a transient will drop, it should not be so small that it eliminates the purpose of D5 which is to keep REGSRC from pulling to near 0 during a heavy load which pulls PACK+ near 0.  Since you don't need a discharge FET in your design this may be less of a concern than in some other systems.

    3. Rdrain resistor provides some filtering of the transients on PACK+ acting with the net capacitance of the Q1 drain, since the net capacitance and Rdrain are often small there may not be much filtering.  In some cases an added capacitor is desired here to provide better filtering. Rdrain also works in DC to drop some power to reduce the dissipation in Q1.  For example if you have 35V across Q1 and Rdrain at 10 mA, you will have 10V across the 1k Rdrain and 0.1W in the resistor.  If D5 drops 1V you have 24 V x .01A = 0.24W in Q1.  A larger resistor drops more power but also more voltage, you don't want to starve Q1 at low battery voltage and high current.

    4. The Cc cap size is a tradeoff of of filter effectiveness vs. voltage artifacts from balancing. Cc does not affect the balance current, but the settling time after the internal balance switch shuts off to allow voltage measurement which can result in a voltage measurement error during balancing.  The voltage artifact will depend on the voltage, which cells, and how many cells are balanced.  While the filter network has many inputs, a 1k and 0.1uF Rc and Cc gives the same simple RC as the 100 ohm 1uF used on the bq76920.