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.

bq2050h - Maximum available capacity?

Other Parts Discussed in Thread: BQ2050H

We are using battery with capacity 40Ah (25-29V) and bq2050h as Battery Menegment System controller.

But  there is  a problem with discharge (charge) rate which is twice more then expected by controller description.

Can bq2050h  handle that capacity at all?

  • Hello Mikhail,

    If the sense resistor is sized to 15mohms, then the computed mVh for a 40A battery would be 600mVh (.015 * 40000).  I would select the next lower value of 563mVh and let the gauge learn a higher value when the bqttery is fresh.  Also, if the discharge rate is high, the availble capacity will be less than the 40Ah that is normally rated for a C/5 load.  The maximum sense resistor drop during discharge is 300mV, so you could have up to a 20A (C/2) discharge rate with no problem.  If your discharge rate is higher, then you would need to choose a smaller sense resistor value and program the capacity to a lower mVh value as appropriate to your sense resistor choice.  If you go too small with the sense resistor, you may run into a problem with accurately measuring low value currents.  Since the maximum VFC measurement offset is 150uV, if you want your measurement accuracy for you lowest load current to be 5%, for example, then you would need to have at least 150uV/5% or 3mV at your lightest system load.

    Bill

  • Hello Bill,

    Thank you for reply!  I need to add some details to my question.

    We are using sens resistore 2mOhm,   so  PFC=40Ah*2mOhm=80mVh.

    We select slightly less, 76.8mVh, as it is recomended in the specification (using pullup and pull down resistors).

    Registers PPU and PPD  have correct value after reset.

    So Scale is equal 1/640mVh/count  and  sens resistor voltage drop 11.25mV is equivalented 2count/sec  (it is from the Table in the specification).

    Then we discharge battery with constant current about 10A and measured sens resistor voltage drop is 22mV.

    So expected discharge rate should be  (22mV/11.25mV)*2count/sec = ~4count/sec.

    But we see that Nominal Avalable Capacity register (NAC)  is decreased with rate about 8count/sec and goes to zerro far before the battery is full discharged!

    And it is the MAIN problem because in that situation we can not get correct information from NAC register about available capacity.

    Also I was confused with  information from http://focus.ti.com/docs/prod/folders/print/bq2050h.html   where the Max Battery capacity for bq2050h equals 6Ah!!

    I would be very appreciated for some explanations.

     

     

  • Hello Mikhail,

    You should verify that the program resistors are being correctly read by verifying the content of the PPU and PPD registers against the desired values for PROG3 and PROG4 especially.

    I expect your problem is due to an improper "Kelvin" connection to the sense resistor, especially from Vss of the gauge.  The gauge SR measurement is from the SR pin to the Vss pin.  If the Vss connection is not made directly to the sense resistor, but  the load current flows through some pcb etch resistance between the sense resistor and Vss connections, then there will be some additional effective sense resistance due to the resistance of your pcb etch.  This should be easy to verify by probing with a DVM from the IC Vss pin to the IC SR pin (or resistor lead connecting to it) while load current is flowing and verify whether you are getting the expected drop or approximately twice that drop.  The very low resistance of your sense resistor will make this layout extremely critical.  You should have all the Vss connections of the gauge connected together and only tied to your system ground at the sense resistor connection that is made at exactly the point on the sense resistor where the 2mohm resistance is specified.

    Bill

  • Hello Bill,

    You are right about accurate voltage measurement.

    I have measured voltage drop (22mV) directly between VSS and SR pins on  the controller.

    Also I know the current (10A) and know voltage drop jn the sense resistore 2mOhm*10A=20mV, it is the same as measured voltage between VSS and SR pins.

    So the problem is incorrect Available Capacity register decrease rate (if battery is discharged) when we have correct voltage drop on sens resistor.

    My  assumption are

    - we are working on the  limit of accuracy because we are using  very small sens value,

    - this controller can't handle such big capacity (as it  pointed in    http://focus.ti.com/docs/prod/folders/print/bq2050h.html  )

    Which is right?

     

    Mikhail.

     

     

  • Hello Mikhail,

    Please disregard any reference to the maximum capacity that this gauge can handle.  It was info created by marketing to cover general selection criteria and was not based on actual device capability.

    Please veryify that PPU value is 0xX7 or 0xXF, depending on whether PROG4 is Z or H, and that PPD value is 0xX0.  LMD should be 0xC0.  There is internal register (undocumented in data sheet) at address 0x7C that should be initialized to 0x08 based on your programming.   The VFC that measures the SR input will output a count every time the passed charge accumulates to 3.125uVh.  With 11.25mV from SR to Vss, the VFC should accumulate 11.25/0.003125 = 3600 counts every hour or 1 per second.  The gauge will subtract the value in 0x7C divided by 4, or 2 counts for every VFC output (when discharging), or 2 counts per second at this rate.  Since the fully-charged condition of NACH*256 = PPFC = 49152, it should count NAC down to zero in 49152/(3600*2) = 6.83 hours.  This is a battery capacity of 6.83*11.25mV = 76.8mVh or about 38Ah using a 2mohm sense resistor.

    If the programming of PPU, PPD, LMD, and 0x7C are correct and the count rate is still too high, please verify that when the load is reduced to zero, that there is no counting.  If the gauge is still counting with no current, there may be some offset caused by your pcb layout.

    Bill