• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 63 subscribers
  • Views 1505 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.

BQ24195L can't be disabled by MCU

S L61
Intellectual 780 points
Other Parts Discussed in Thread: BQ24195, BQ24195L

I am using MSP430 to control BQ24195 but no matter how I did, I couldn't disable it from charging.

I have tried set NCE pin to HIGH and set REG01[5:4] = 0x00. But it was still charging. Even I disconnected MCU from BQ24195 and then connected NCE pin to HIGH, the result was the same.

I think seting NCE to HIGH or REG01[5:4] = 0x00 can disable it charging. Is there any misunderstanding about disabling charger?

  • 0
    TI__Expert 8690 points

    Hi S L61,

    I noticed the problem from before was fixed, I'm glad.

    Yes you are correct, either setting NCE pin to HIGH or REG01[5:4] to 0x00 will disable charging. Can you confirm whether the bq24195L is being written correctly? Do you reset watchdog timer to prevent all registers from resetting? Also, can you measure voltage at NCE pin? Read I2C and post results on REG01, REG08 and REG09.

    Thanks,

    StevenB

  • 0 in reply to Steven Bangdiwala
    Intellectual 780 points
    Thanks for your reply, StevenB.
    I can see the written results from I2C and I disabled watchdog to prevent it resetting registers. The voltage at NCE pin was 4V (when I sent DISABLE command). REG01 = 0x26, REG08 = 0xB4 and REG09 = 0x00.
  • 0 in reply to S L61
    TI__Expert 8690 points

    Hi S L61,

    To have charge disable you need to maintain NCE pin at HIGH 4V, in that case change MCU NCE digital control line to stay on HIGH.

    In addition, I have noticed that REG01[5:4] reads 10 (OTG mode) instead of 00 (charge disable), which is not what you aim for.
    1. Make sure you are writing to slave address 0x6B
    2. That you are writing bitwise 00 to specific bits REG01[5:4]
    3. If not in bitwise, I recommend reading the whole register REG01, applying AND mask (11001111), and writing back to REG01
    4. To reset watchdog:
    - Bitwise 1 to REG01[6]
    - If not bitwise, read REG01, apply OR mask (01000000) and write to REG01

    ++ REG01 should read 0x0B if done correctly.

    Try the following steps if the above doesn't work, just to troubleshoot the bq24195L.
    1. Do not use MCU
    2. Remove all power to bq24195L (disconnect VBUS and VBAT)
    3. Reconnect VBAT and VBUS
    4. Voltage at the VBAT pin between 4.1V and 3.3V
    5. Apply a HIGH on NCE pin (power supply 4V on NCE pin)
    6. See if battery still charges by measuring the current to the battery, or measure the STAT pin (TP10) (HIGH means charging done or charge disabled, while LOW means charging)

    If problems persist, do a read of all the registers before and after you write REG01[5:4] to disable charger, something else might be the culprit.

    Thanks,
    StevenB

  • 0 in reply to Steven Bangdiwala
    Intellectual 780 points
    Thanks StevenB. I have tried your sugesstion step by step but things become more confusing.

    I found the charger was disabled (REG08[5:4] = 0x00, STAT pin was HIGH) after writing REG01[5:4] = 0x00 and I could enable it again (REG08[5:4] = 0x10, STAT pin was LOW) by setting REG01[5:4] = 0x01. However, there was still current from BAT to the battery. I don't think this is a disable state.

    Thanks.
  • 0 in reply to Steven Bangdiwala
    Intellectual 780 points
    I followed the steps to troubleshoot the bq24195L. I found there was current from BAT when NCE pin was HIGH and LOW. Changing NCE didn't give me a different result. Maybe the chip has been corrupt. Is there any way to verify the chip? (I measured the VBUS, BAT. They seems normal.)

    Thanks,
    S L61
  • 0 in reply to S L61
    TI__Expert 8690 points

    Hi S L61,

    Please read all the registers before and after writing 00 to REG01[5:4] and post them here.

    Thanks,
    StevenB

  • 0 in reply to Steven Bangdiwala
    TI__Expert 8690 points

    Hi S L61,

    Also, let me know what is the equipment you are using to test and troubleshoot, how you are connecting VBAT and VBUS, where in the board are you measuring voltage and how are you measuring voltages and currents? Are you using our EVM or a different board?

    If on our EVM, verify that there is no jumper on NCE to GND at JP5.


    Thanks,
    StevenB

  • 0 in reply to Steven Bangdiwala
    Intellectual 780 points
    Hi StevenB,

    I am using EVM and there is no jumper at JP5.

    Once connected with power supply, there will be current from BAT no matter which condition the bq24195 is in (like disable charging or enable charging). If meeting charge termination condition like fully charged or safety timer expired, there will be no current from BAT. Am I right?
  • 0 in reply to S L61
    TI__Expert 8690 points
    Hi S L61,

    There could still be current from the battery if it is in supplement mode (use battery to supply current to system) or if it is in OTG boost mode (use battery to supply input), both of which have output current. If safety timer expired or if battery is fully charged, there is no current going into the battery.

    This is why it is important that you give me the reading of all the registers before and after writing from MCU.

    Are you using the GUI that comes with the EVM? Go to the "tools and software" page from bq24195L and download the bq2419x Graphical User Interface (Rev. A). From there you can read and write to the I2C without hassle. This is a sure way to verify that the EVM is working correctly.

    Thanks,
    StevenB
  • 0 in reply to Steven Bangdiwala
    Intellectual 780 points
    Thanks for your reply. I think I have got the answer. The charger state can be changed by MCU while there is still current from BAT. I misunderstood that there should be no current from BAT after DISABLE charging.