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.

  • TI Thinks Resolved

BQ27510-G3: Strange Capacity readings

Genius 4445 points

Replies: 39

Views: 202

Part Number: BQ27510-G3

Hello,

I've been plugging and unplugging the battery as we do modifications to our design (unrelated to the fuel gauge) with some charging and discharging in between and noticed the battery fuel gauge is now way off.  It was reading OK when we first programmed it and sealed it and I did a few cycles.  Now it gives very inaccurate readings.  It says the battery is empty already at 4.0V.  Below are the readings when I jumpered it to an EV2400.  The full charge capacity is completely off and full available capacity is completely off as well.  Thoughts?

  • I got the device to unseal and here are all the registers:

    export.gg.csv

  • In reply to Andrew5821:

    I will take a look tomorrow and get back to you.

  • In reply to Andy Liu (BMS):

    It looks like the gauge chooses the wrong battery profile. Check 5.7.1 Battery Profile Storage and Selection in the TRM for details.

    Your second profile doesn't seem to have correct data. QMax Cell1 is 1000mAh. Your taper current is 282mA. That would be a huge C/3.5 so FCC will diminish accordingly.

    Please make sure that the second profile reflects your battery appropriately and please make sure that the profile selection is correct.

  • In reply to Dominik Hartl11:

    Dominik,

    What is the best way to modify my current golden image to correct this?  It looks like I can't just open a golden image file and edit it in bqStudio

    Thanks,

    Andrew

  • In reply to Dominik Hartl11:

    I'm still trying to figure out what happened. Are my Ra Tables reasonable?

    So my plan is to copy Qmax Cell 0 and Qmax Cell 1 from "Gas Gauging -> State" to "OCV Tables -> OCVa0 Table" and "OCV Tables -> OCVa1 Table". 

    Set Update status to "2" in "Ra Tables".

    I noticed that SOC1 Set Threshold is 150% and SOC1 Clear Threshold is 175%.  I'm not sure why these are those values?  What is more reasonable?

    Just to confirm as well.  I have GPOUT going to the BQ25895 charger and I have PFC_CFG[1:0] set to a "01" which should disable the charger by letting the pin go high of the temp is outside the Charge Inhibit Temp Low and High

  • In reply to Andrew5821:

    You can export the gg file, copy-paste the values for Ra and Qmax so that both cells match and re-import this modified gg file.

    About the other values: SOC1 set/clear are thresholds to set the SOC1 flag (they don't affect gauging). If you want to use this flag in your system, I would set the thresholds to reasonable values (150 and 170 don't make sense. Default is 10 and 12).

    The PFC_CFG config is correct for you use case (temperature charger inhibit) 

  • In reply to Dominik Hartl11:

    I noticed only one Ra had values and a status of 00.  The rest were defaults with status being FF.  So I just copied the Ra table values from the one with a status of 00 to the others that were FF.

    I think what happened was I changed all the values, but never wrote them out to the fuel gauge before creating the golden image.  This is why I had registers that were still defaults in my previous golden image.

    Attached is my final .gg file.

    2-20-20_Rev2.gg.csv

  • In reply to Andrew5821:

    Hello Andrew,

    Can you run an accuracy cycle with this file?

    The update status shows a Qmax update, but not resistance updates.

    Please select your chemID via GPCCHEM and then run the learning cycle.

  • In reply to Kang Kang:

    I'm starting over with a new golden image.  I've completed the learning cycle.  Do I set update status to 0x02 in the OCV Tables?

  • In reply to Andrew5821:

    Hello Andrew,

    I do see you have gotten resistance updates. At this point, can you run a charge to full, relax 2 hours, discharge to empty using C/5 and relax for 2 hours log file?

    Can you attach it. We can do the accuracy analysis for you for this file.

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.