• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 1423 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.

Original question:

BQ76940EVM: bq78350-R1 SOC always 0%

BQ76940EVM: bq78350 SOC drop at some point

user5769503
Prodigy 180 points
Part Number: BQ76940EVM
Other Parts Discussed in Thread: BQSTUDIO, GPCCEDV

Hello,

I'm currently working with the BQ76940EVM, EV2300 and the latest BqStudio version (1.3.86.3).

When I discharge by battery pack, at some point the SoC and the remaining capicity are dropping down,

for example from 80 % to 7 %, the voltage is still displayed correct.

Sometimes after a reset everything is correct for 2 seconds and then it fails again but it also happens that the SoC is permanently wrong after the drop happens and I have to programm it with the default firmware to get correct values again.

The same happens during charging (goes up to 100%).

I currently try to do test for the GPC packger but I cannot finish a discharge test because the by76940 stops it at some point.

My battery pack has 10 serie cells, max voltage of the howl pack: 42 V and min 25 V

The test I'm running: 25 degree, 2.9 A constant discharge. When the pack reach for example 33 V the discharge is stopped.

Do you have any idea what the problem is and how I can fully discharge my pack?

I attach a zip file with log and gg files.

ti-files.rar

  • 0
    Intellectual 350 points

    Hi,

    I have pretty much the same issue:

  • 0
    TI__Guru* 81034 points
    Thanks for attaching your gg and log files. I will review these and reply back tomorrow.

    Best regards,
    Matt
  • 0 in reply to Matt Sunna
    TI__Guru* 81034 points
    I looked through your log and .gg files. The jump down to 7% indicates that one of the cells crossed below the EDV2 voltage threshold. It looks like this happens at 3500mV for your configuration.

    Have you run voltage calibration and loaded the correct chemID for your device in bqStudio? This is very important to initialize the correct voltage curve for gauging.

    Best regards,
    Matt
  • 0 in reply to Matt Sunna
    TI__Guru* 81034 points
    I also just noticed in your gg file that the EDV_CMP should be set. Otherwise it will use the fixed EDV2 default voltage of 3500mV. If this is not correct for your battery, you will see the SOC jump down to 7% too early.

    You may also want to follow this other thread where another user is debugging a similar issue:
    e2e.ti.com/.../723974

    Best regards,
    Matt
  • 0 in reply to Matt Sunna
    Prodigy 180 points

    Hi Matt,

    discharging worked after changing EDV2 but now it does not recognize when the battery is empty... it stops at 33545 mV (20 % SOC) while my battery tester (digatron) already reached the minimum of the pack (29V). It seems like it is the wrong value again.

    I did the calibration and selected the chemID in bqstudio. I'm using SAMSUNG SDI INR18650-29E 2G81 and selected the chemID 1737. I set the EDV_CMP bit to 1 and corrected the EDV2 to 3299.Do I have to change EDV0 and EDV1 as well? I couldnt figure out what they mean.

    Is it possible to get the OCV curve that bqstudio is using for the chemID 1737? It looks like the curve is sligthly different to the one that I'm using but if  I would get the curve from bqstudio, I could use this one for my calculations instead.

    Could I already use the test I atteched for the GPC packer, even if its wrong?

    I attach the log and gg files again.

    Best regards,

    Anna

    Test_1.rar

  • 0 in reply to user5769503
    TI__Guru* 81034 points
    Hi Anna,

    Thanks for sending the new log and gg files. I will look through these and reply by tomorrow.

    Best regards,
    Matt
  • 0 in reply to user5769503
    TI__Guru* 81034 points

    Hi Anna,

    I think the issue is that the battery was not fully charged and that a fault occurred during the charging that caused it to stop. Right before charging stopped, one of the cell voltages showed a measurement go up by 2V from its previous value to 6V. Is it possible that the connections to the cells are compromised since these appear to be false readings?

    Regards,

    Matt

  • 0 in reply to Matt Sunna
    Prodigy 180 points

    Hi Matt,

    I already noticed that issue and solved it. The problem with the voltage jump at a singel cell was caused through the cell balancing. After I disabled CB, this issue did not happen anymore. But the problem I explained before is still the same. Even if the cells are fully charged, the SOC is wrong towards the end of a discharge and completely wrong at the total end. After a reset the SOC is perfect.

    regards

    Anna

  • 0 in reply to user5769503
    TI__Guru* 81034 points
    Hi Anna,

    Once you are able to do a full charge and discharge cycle with no issue, the next step is to collect logs and use the GPCCEDV tool to obtain the optimized CEDV gauging parameters. However, I want to make sure there are no other issues still present since that step can be time consuming. Can you send me your latest log with the fault conditions resolved?

    Thanks,
    Matt
  • 0 in reply to Matt Sunna
    Prodigy 180 points

    Hi Matt,

    attached the current log file, thanks that you have a look. The full charge and discharge is working now. The GPC Packer only requires the discharge log or? Just if you wonder why the voltage values are a bit over the maximum voltage (42 V), the reason is that the voltage does not get 100% correct detected from the device.

    Could you have a look at my EDV values as well and tell me why they are not getting used? I disabled EDV_CMP again. I know I will get the calculated values from Ti after I send the GPC log files but it would be nice to know the reason.

    regards

    Anna

    Test_ti.rar

  • 0 in reply to user5769503
    TI__Guru* 81034 points
    Hi Anna,

    Yes, the GPC Packer only requires the discharge log. I think you are ready to collect the log files for the GPC tool.

    For the log files, it is good to discharge to ~100mV below your system terminate voltage to collect a little more data for the curves. When you create your config.txt file for the GPCCEDV tool, you will set the actual system terminate voltage. For the 'high' discharge rate, you will want to pick the highest expected current for your system (so if you expect up to 4A spikes, you will want to discharge with 4A). You can see the full instructions here: http://www.ti.com/tool/GPCCEDV

    Best regards,
    Matt