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.

BQ27532-G1: Problem with entering hibernate mode

Part Number: BQ27532-G1
Other Parts Discussed in Thread: BQSTUDIO

Hi there,

I want to put the gauge bq27532 into hibernate mode in our project. So I followed the device state diagram on technical reference page 15, after the board entered the sleep mode, i send the "set_hibernate" command, but still did't see its going into hibernate mode. My device usually works at around 37 to 100 uA and i set "sleep current" = 1, "hibernate current" = 1, "quit current" = 1, "Dsg Relax Time" = 2, "Chg Relax Time" = 2.

Can i get any suggestion about how to test for the next step?

Thanks,

Shiping

  • Hi Shiping,
    For the device to enter hibernate mode, an OCV measurement has to be taken. It could take up to 15 mins for a valid OCV measurement to be taken depending on if your battery is well relaxed. Also, you shouldn't be communicating with the gauge as that will take it out of hibernate mode. after you send the hibernate command, you need to disconnect the ev2300 from the gauge to ensure there is no communication with the gauge. You have to monitor the current drawn by the gauge to know if you are in hibernate mode or not.
    thanks
    Onyx
  • Hi Onyx,

    Thanks for your reply, the picture below is my newest setting for gauge module, and my device usually work at 33uA ~ 120 uA. As you said, i wait until the gauge goes to "sleep mode" and send "set hibernate" command, then removing EV2300 to ensure no communication, but after waiting for quite a while i still didn't see any power consumption change for entering hibernate mode.

    Is there anything you think i could check to help finding the reason?

    Thanks,

    Shiping

  • Shiping,

    i suggest keeping the defaults for sleep current and hibernate current as the default valies and follow the instructions for getting into hibernate as seen below.. You need to send the hibernate command, when you don't have any current going into or out of the battery, and then disconnect the ev2300. The OCV measurement could take up to 30 mins.

  • Hi Onyx,

    Thanks for your patient reply.

    Your suggestion is really helpful and I tried as you suggest but still didn't see changing in power consumption. Also feel it really takes a lot of time to do one test, so wonder is there any ways to make the test more efficient? Or is there a way to know if the gauge is in relaxation mode, feel the document didn't really talk much about the relaxation mode.

     

    Thanks a lot,

    Shiping

  • Hi Shiping
    If you communicate with the gauge after sending the hibernate command the gauge will not enter hibernate mode. You need to send the hibernate command, verify that the flag is set. disconnect the unit from the ev2300 and let the pack sit. You can measure the TS pin to see that it isn't pulsing once every 20s or measure the current consumed by the gauge to determine if it is in hibernate mode.
    thanks
    Onyx
  • Hi Onyx,

    I checked the document about TS pin and from the document "screenshot 1" below it looks like TS is a input pin, how did it actually pulse that pin every 20 secs if it's a input pin?

    Also, in our design, our battery didn't have 3 pin but only 2 pins, so here is how we design it in our application at "screenshot 2". And we set [TEMPS] bit to be 0 to only measure the internal temperature. I have a concern that will this affect the observation of pulse at TS pin? Or is there any other pin i could check to observe the "wait hibernate" state?

    screenshot 1:

     screenshot 2:

    Thanks a lot!

    Shiping

  • Hi Shipping
    So are you saying you aren't using the external thermistor for temperature measurement? If that is the case, then i am not sure if the pin would pulse, every 20 s. It is something worth checking otherwise simply use the current measurement drawn by the gauge to determine hibernate mode.

    thanks
    Onyx
  • Hi Onyx,

    Yes, i am not using the external thermistor.

    I still have something unsure about.
    1. If i could use the external thermistor, which state will the TS pin pulse every 20 s, when it is in "wait hibernate" state or in "hibernate" state?
    2. My current problem is to find out why it doesn't go to hibernate mode even after i send the "set hibernate" command when it's in sleep mode. Since the OCV measurement will take up to 30 minutes(really long to wait) before entering hibernate mode, and in previous experiments i didn't see any change in power consumption after 30 mins, my idea is to find a way to make sure it does go to "wait hibernate" mode as the previous stage for hibernate mode. Hope by doing that it could help me figure out the reason it doesn't go to hibernate mode. Do you have any idea about this?


    Thanks!
    Shiping
  • Hi Shiping,
    In hibernate almost all the internal circuity is turn off. The ts pin is will not pulse in hibernate mode. It only pulses in sleep.

    Here is what i need you to do so we see if you can get into hibernate.

    -Set your hibernate current to 5mA
    -Set your sleep current to 10mA
    -Power your board with a power supply instead of with a battery. Chances are that your battery wasn't getting relaxed for the gauge to take an OCV reading that allows the OCVtaken flag to get set . The gauge will only force an OCV reading after 1 hour if a valid OCV measurement was unable to be taken due to the battery not being relaxed.
    -Make sure you have an ammeter in series with the power supply to measure the current being drawn by the gauge to know when you are in hibernate mode.
    -Make sure you do not have any load on the system that draws any current. Make sure IT is enabled
    -Send the hibernate command in bqstudio
    -disconnect the i2c cable so there is no communication.
    -Given you are using a power supply which has a stable voltage and you do not have any current flow in the system, the gauge will be able to take an OCV measurement and you will be able to enter hibernate mode in less than 12 minutes.

    i hope this helps.
    thanks
    Onyx
  • Hi Onyx,

    Appreciate your help with the testing.

    I followed the steps and have a question about instruction "-Make sure you do not have any load on the system that draws any current. Make sure IT is enabled", why do i need to make sure "no load on system that draws any power"? To make sure the current is below the threshold?

    The reason i ask is as shown below, there is pulse in current consumption on our device. Is this still fine to let gauge go to hibernate mode?

    I am also trying to build a clean environment to test the gauge going into hibernate mode, any idea about that?

    Thanks,

    Shiping

  • Hi Shiping,
    If this current spikes are above your hibernate current value in the data flash, the gauge will not enter hibernate mode. You need to adjust your hibernate and sleep currents to be higher than these current spikes.

    With regards to the environment, you just need to make sure you don't touch the comm lines of the gauge after you send the hibernate command and disconnect the i2c lines. Reason being touching the lines could cause pulses that might make the gauge think there is comm about to start and then it wouldn't enter hibernate mode. This might not really be necessary to strictly follow but i am just trying to make sure there is nothing preventing you from getting into hibernate.

    thanks
    Onyx