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.

BQ27510-G3: Learning cycle doesn't seem to start

Part Number: BQ27510-G3

Hello,

I've been attempting to complete a learning cycle with a new battery pack with a 1S3P configuration. I'm using a program to automate the charge discharge cycle set to charge at C/2 and discharge at C/5. Here are the steps I took:

.Calibrate temp, voltage, and current measurements

. Answer the Parameter Q&A and Save to Gauge

. Discharge battery pack

. Send commands IT_ENABLE and RESET

. Rest 5 hours

. Charge battery pack

. Rest 2 hours

. Discharge battery pack

. Rest 5 hours

When I sent the IT_ENABLE and RESET commands, RUP_DIS, VOK, and QEN remained cleared but the IT Enable went from 00 to 01. The learning cycle ran overnight but Update Status 0 never changed. I've attached the register log file and the auto exports of the data memory. What could've gone wrong?

I did notice that when I calibrated the internal temperature, the temp would decrease by 10C - 15C while the battery was discharging.

18-1013_LearningCycleData.zip

Thanks,

Rodrigo

  • Hi Rodrigo,
    QEN ought to get set when you sent the IT enable command. Did you hit the refresh or scan button to be sure that QEN flag did not get set? Can you use the external thermistor of the gauge instead.

    also, Can you verify that the corresponding parameters you changed via the Q and A actually got updated in data memory. If they didn't you would want to update those manually by going into data memory and effecting the changes. see documentation below.

    www.ti.com/.../slua902

    thanks
    Onyx
  • Hi Onyx,

    I did hit refresh a couple of times and was also logging so the registers should've updated either way. I still did not see any changes to QEN. The pack has been resting now for well over 5 hours now so I can start the cycle again. Sending the commands IT_ENABLE and RESET does not set QEN , VOK, nor RUP_DIS. Earlier today, the RUP_DIS bit was set but it eventually cleared while resting. This can be seen in the screenshot.

    Unfortunately we require to use the internal temp sensor for our application. Can I use an external thermistor for the sake of the learning cycle then switch to the internal temp sensor for the application?

    Checked the parameters within the Data Memory and they did change to those specified in the Q&A section.

    Would the issue have anything to do with the battery pack configuration? It's 3 batteries in parallel which are each nominally 2050mAh so a total pack capacity of 6000mAh. Each individual cell has a Chem ID of 1253 but I'm not sure if having them in parallel will affect the learning cycle. I conducted a chemistry learning cycle but the GPC tool returned with an error of >3% so it seems like the ID generated is unusable.

    Thanks,

    Rodrigo

  • A chem id greater than 3% is for sure unusable. When you did the chem id identification, did you remove any protectors that may have been attached to the cell? This tend to contribute to not being able to find a match.

    It is weird that QEN doesn't get to set. Can you try a different IC or even try programming the default firmware on the gauge.

    thanks
    Onyx
  • Hi Onyx,

    It turns out all I needed was to place an external thermistor on the eval board. I removed it since I wanted to simulate the application but placing it back on the board set the QEN and VOK bits. I left the auto cycler running over night and the Update Status 0 register incremented to 01 and Qmax Cell 0 updated to >6000mAh. I am now waiting on for the pack to rest to confirm a successful learning cycle.

    Thanks,

    Rodrigo

  • Hi Onyx,

    The learning cycle has completed and Update Status 0 was incremented to 02. I believe this constitutes a successful learning cycle.

    Thanks,
    Rodrigo
  • yes it does. Good job
    thanks
    Onyx