• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 366 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.

BQ34Z100-G1: State of Health/Capacity as batteries age.

Kirk Baumgartner
Prodigy 20 points
Part Number: BQ34Z100-G1
Other Parts Discussed in Thread: BQ34Z100

We are using a BQ34Z100-G1 on a 6S Li-Ion battery.  The fuel gauge works pretty well with new batteries, but we are running into a problem as the batteries age, particularly with batteries that have a cell that is weaker than the others.  I am discharging a battery with a weak cell using a representative load on an Arbin battery tester. The Arbin is set to stop at 20V.  The terminate voltage in the fuel gauge is set to 20.1V.  At the end of the discharge cycle the fuel gauge is still indicating that the SOC is about 15%. The full charge capacity value is reading higher than what I am able to pull out of the battery with the Arbin.  I realize that there is still charge in the battery, but that charge is inaccessible due to an imbalance in the cells.  I would like to adjust the fuel gauge to be able to accurately report battery capacity in this situation.

1- Does the fuel gauge need the battery to dwell at the terminate voltage for a certain period of time?

2- What is the difference between FCC and True FCC?

3- Does the battery need to see the terminate voltage to adjust the FCC value?

4- Is the SOH based on the FCC or True FCC?

5- I have noticed on some batteries that if the fuel gauge is overestimating the capacity, when it hits the terminate voltage the SOC will jump immediately to 0% (I know that is the definition of the terminate voltage).  Does the fuel gauge update it's FCC estimate when this happens?  I would like to avoid having the SOC drop from 10% to 0% instantaneously.  This was under a constant load.

Thanks,

Kirk

  • 0
    Prodigy 20 points

    Regarding question #5, I verified that the FCC is updated at that point.  However when given time to rest the SOC increased from 0 to 9% and the FCC returned to a value that was higher than before.  Is there a way to prevent this from happening?

  • 0 in reply to Kirk Baumgartner
    TI__Genius 12215 points

    For Q1, The fuel gauge does need the battery to dwell once in a while to update the Qmax, the capacity of 2 adjacent dwelling should be no less than 34% of design capacity so that the Qmax Update can happen. The voltage of the 2 dwelling point shall not be in the range between Q Invalid MaxV and Q Invalid Min V, which may need to check with TI from case to case as these 2 parameters are different for different chemical ID

    For Q2, FCC is the post processed True FCC, for bq34z100, the post process refers to the smoothing to the True FCC and True RM  over different temperature and load rate.

    For Q3, Not really, but the Gap of the Q of the 2 dwelling point, the more accurate the Qmax will be

    For Q4, SOH is calculated with the SOH FCC, which is calculated under give load(SOH Load I) and temperature 25C

    For Q5, The FCC will adjusted after discharge to the terminate voltage. You can set the bit Pack Configuration C[LOCK_0] to suppress such RSOC drifting.

  • 0 in reply to Steven Yao
    Prodigy 20 points

    Here is some data from a discharge I have done with this battery.  We use a 9X scaling factor on the current.

    Pre-Discharge:

    Cell 1: 4.231V
    Cell 2: 4.241V
    Cell 3: 4.241V
    Cell 4: 4.241V
    Cell 5: 4.241V
    Cell 6: 4.244V

    SOC: 95%
    FCC: 1164 mAh  (10.476 Ah) 9x scaling
    True FCC: 1164 mAh (10.476 Ah)
    Available Energy: 476mWhr
    SOH: 91%
    QMaxCell0: 1252 mAh
    Term Volt: 20.4V

    The battery was discharged at 610W to 20 V pack voltage

    Post Discharge:

    Arbin Discharge Capacity 7.9096 Ah

    FCC: 930 mAh
    True FCC: 1178 AH (10.602 Ah)
    Qmax Passed Q: 876mAh (7.884 Ah)

    I set the LOCK_0 bit in the pack configuration so the SOC stays at 0, but after about 1.5 hours in rest the FCC jumped from 930 mAh to 1154 mAh

    As you can see this pack has a weak cell.  What I am trying to prevent from happening is the following:

    The SOC drops from about 22% to 0% instantly.  Are there settings or other parameters that would allow the fuel gauge to learn it has a reduced capacity and prevent a sudden drop in the SOC?

    Why is there a discrepancy between the loaded versus unloaded SOC and FCC.  

  • 0 in reply to Kirk Baumgartner
    TI__Genius 12215 points

        FCC change is caused by the Qmax update and following simulation during relax, the discrepancy between the FCC during loaded and unloaded FCC is caused by the deviation of actual cell character and the cell chemistry model identified by Chem ID, as the bq34z100 only monitors the total voltage of the battery stack, it requires the cells in same battery pack grouped properly so that the single cell voltage does not vary too much, for the cells with voltage varies like the figures you uploaded above, the actual averaged cell character is very different to chem ID, so the gauge can not estimate the SOC and FCC accurately. As such deviation always exhibits itself during discharge, so the SOC and FCC during discharge can be very different to the value during relax.

        The cells to be grouped in same battery pack shall be selected appropriately. 

  • 0 in reply to Steven Yao
    Prodigy 20 points

    We receive these cells in a battery pack from our supplier, they should be pretty well matched.  The problem is that although they are matched, as they age, they all age differently and they may no longer be so well matched.  How does the SOH estimation deal with cells that fail early?  Is there a way to disable the FCC re-update after the fuel gauge is detecting a rest condition?

  • 0 in reply to Kirk Baumgartner
    TI__Genius 12215 points

        In present code, Qmax update should always followed by a FCC update. I do not see any walk around available