• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 210 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.

BQ76942: High current sensing with uC ?

AhtoM
Prodigy 30 points
Part Number: BQ76942

Knowingly, BQ76942 has it's own current sensing capability with SRP and SRN, allowing to measure on a current sense resistor in the range of +-200mV (higher with fault current).

In my application low side current sense resistor is not feasible.
I need to be able to measure 250A peaks (fault current) and about 100A-175A continuous (@26Vbat max).
With a current sense resistor of 1mR, the resistor would dissipate roughly 30.6W and 62.5W peak - not feasible due to space constraints of such sense resistor and just can't allow that power loss continuously.

I was thinking if I could keep the power loss through current sense path low but still manage to use the BQ functionality.

1. Would it be possible NOT to use BQ76942's current sense pins, but use f.i a Hall Effect current sensor (<100uR series resistance), interface this to a microcontroller and via I2C communicate the current value to BQ76942 ?

This way to I'd hope to keep BQ informed about the current and still be able to use built-in features ?

2. Instead use still a Hall Effect current sensor, but convert it's output range (+-2.5V) to +-200mV to the BQ76942's SRP and SRN ?
I imagine this to be more 'autonomous', if smth were to happen with the uC, the current can still be measured.

Or
in Your humble experience, is there any other solution to allow the least power loss when sensing high current but still have BQ's protection and measurement features?

  • 0
    TI__Genius 16010 points

    Hi Ahto,

    The solution we generally recommend for enabling higher current is to decrease the value of the sense resistor. This will decrease the power dissipated across the resistor. In addition to that, using several larger resistors in parallel would allow the power dissipation to be divided across the resistors while keeping the voltage the same.

    Regards,

    Max Verboncoeur

  • 0 in reply to Maxwell Verboncoeur
    Prodigy 30 points

    Thanks for the reply.

    Sure, paralleling reduces the dissipation for a single component in the circuit,
    but in total, I still have high power dissipation, which I want to reduce.

    To put in numbers again. In order to measure, lets say, 100A with BQ SNS pins, I could use
    0.2V / 100A = 2mR current sense resistor. This would dissipate 20W !
    Paralleling 10x, would bring down the power dissipation for one 20mR sense resistor to 2W BUT I'm still consuming 20W to get 0.2V full scale measurement for a 100A peak.

    So I was thinking, how about using something like a 100uR resistor (1W @100A), which I have seen available, let an external sensor amplify the voltage drop to BQ's SNS inputs to it's recommended range (+-0.2V)...

    OR

    do current sense externally altogether with uC  but communicate current to BQ...

    - Does any of the two scenarios affect the BQ76942's ability to engage any of the overcurrent/discharge protections 'autonomously' ?

    Fundamentally I'm trying to understand if I DON'T use BQ's SNS pins, do I have to implement external methods to switch protection fets with a host uC ?
    Could firmware just communicate current to BQ and BQ is still able to engage it's protections ?

    Sorry, I haven't yet played around with the BQ software that much.

    (corrected calculation error)

  • +1 in reply to AhtoM
    TI__Genius 16010 points

    Hi Ahto,

    Reducing the resistance to 0.5mOhms could work for you. At 100A that would equate to 50mV across the sense resistor, and you shouldn't need to amplify the voltage at all.

    The +/-200mV range of the ADC should not be the target voltage across the resistor, it's a limit. The desired voltage should be much lower so that any high current events (short circuits or transients) that exceed the expected current remain within the limit and are properly detected by the device.

    It's worth noting that the ADC step size is around 7.6uV, so there shouldn't be much of a loss in accuracy by decreasing the ADC gain to match the lower resistance.

    Regards,

    Max Verboncoeur

  • 0 in reply to Maxwell Verboncoeur
    Prodigy 30 points

    Ah, good to know, that I don't necessarily need to use the full scale - of course. BQ allows setting different current gains and depending on one's definition of smallest allowed current to be measured the sense value could be set accordingly.  
    Based on this I've currently settled with two parallel 300uR which would result in a more manageable power dissipation.
    Thank you!