• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 920 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.

bq76920 - sensing high currents

Peter Nink
Prodigy 40 points
Other Parts Discussed in Thread: BQ76920

Dear All,

I have seen several post looking to increase the current capability of the bq76920 by paralleling MOSFETs and using low shunt resistor values.

We are looking to design a battery management system for a 4 cell LiFeP stack with 250A max continuous discharge current. This would include state of charge (gauge), cell balancing, overcurrent protection and voltage supervisor.

The resulting shunt resistor values would likely be between 0.1 and 0.2 mR. That is likely to be beyond the intended range of the AFE current amplifier. 

Would adding a zero drift amplifier between the shunt and the AFE inputs be sensible? Has anyone tried that?

Or is there an altogether better solution to designing a battery management system for a 12V 200Ah LiFeP battery?

Love to hear your thoughts on this - Peter

  • 0
    TI__Guru 54715 points

    Hi Peter,

    Think your questions were largely addressed in another post.

    The part has a fixed dynamic range on the current sense.  The sense resistor needs to be picked to provide measurement over the desired range and allow selection of the protection limits.  Once you have a resistance value parallel resistors can be selected if needed to provide the desired value while spreading power as needed.  The part is not characterized for operation with an external current sensor, it is probably possible and people have likely done it, but the datasheet does not provide guidance.  Since the inputs were designed to operate at GND, you should expect the input to run into a power supply limit if you attempt to add a common mode.  Operating about zero would allow the part to operate in its intended range.  Drift and calibration may take special consideration if using a sensor circuit.

  • 0 in reply to WM5295
    Prodigy 40 points

    Hi WM5295,

    Thanks for your answers. You are correct this post is largely an overlap with my other post. I did not realise that the posting was delayed by moderation and thought it was lost. Maybe the moderator can join them together?

    I understand that the AFE is not designed to be used in conjunction with current amplifiers. Questions is how can I adapt the circuit to handle high currents without having significant losses on the shunt in order to provide the +/- 200mV input range for the AFE? I would be looking at +25mV and -5mV in my scenario. Would that result in a workable system in your view?

    Out of curiosity: I have never seen current shunts with kelvin pick-up being used in parallel. If that was the case I would guess you would need to keep the 2 kelvin pick-ups separate or lose the accuracy of the system?

    (edit: corrected typos)

  • 0 in reply to Peter Nink
    TI__Guru 54715 points
    I think it depends on the dynamic range of your system. If 250A is represented by 25 mV, then 1A would be 100 uV. The CC would only have about 11.8 counts for the first amp. Low currents will be rather noisy but you may not care. The maximum OCD threshold is 100 mV so you could shift the operating point up if your SCD allows it. You have to choose.
    The Kelvin connection really only works to one resistor, I typically see it put on a central resistor. If you connect to multiple resistors it forms a parallel current path to the high current path and the Kelvin connection is then really at the edge. There will be inaccuracy with the parallel resistors, calibration is important.