• State Resolved
  • Locked Locked
  • Replies 14 replies
  • Answers 4 answers
  • Subscribers 62 subscribers
  • Views 551 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.

BQ27742-G1: Recommended shutdown/startup procedure

Christopher Buckley
Intellectual 330 points
Part Number: BQ27742-G1


Hello,

I made a prototype board which uses the BQ27742-G1 and was unhappy to find that I must attach a charger or short Pack+ and Cell- according to Section 2.1 of the eval board's user guide. https://www.ti.com/lit/pdf/sluuax8 else I can't power my circuit. Is this mentioned anywhere in the actual datasheet? I would have thought if I was within the required voltage ranges, I could power the circuit when attaching a battery.

Anyways, I'm curious what is the design intent for this for startup/shutdown. I'm guessing the intent was to have a connected LiPo at all times and never disconnect, but have low-voltage cutoffs like the one included in the BQ27742-G1. Aren't there any worries about the LiPo continuing to drain through leakage currents? I see this all the time with old smartphones, that I can't even power on, even with a charger attached, after they've sat in the closet for a few months.

I'm working with a LiPo powered device that can see anywhere from a few minutes to several months of down time. So we're trying to implement a rocker switch (or similar) to completely cut off the cell from the rest of the circuit to prevent as much drain as possible, while keeping it safe from undervoltage dangers. Since the BQ27742-G1 cuts everything off, then even with a full battery, I switch on the rocker switch, my host controller will never be powered, so I can't do anything from the firmware side of things. So I guess I can use one of those 3-position rocker switches that push-release to short these two terminals and the other two positions would be the normal connect/disconnect. Is that a viable option for this or would this cause trouble somehow I'm not foreseeing? If this is seen as an issue, could you please provide a viable alternative?

Best regards,
Christopher

  • 0
    TI__Guru 57697 points

    Hello Christopher,

    The BQ27742-G1 is a pack side gauge which means it must always be connected to a battery. 

    Use the shutdown mode feature if you will put the pack in storage. See RTM section 2.2.2.2.1.5 SHUTDOWN WAIT Mode

  • 0 in reply to Shirish
    Intellectual 330 points

    Thank you for your reply Shirish. Do you know where I can find further information about the power consumption of the Fuel Gauge and LDO that are stated to be left on in this "Shutdown Wait" mode? If this draw is too much, do you have any recommended alternative from TI that are designed for battery disconnects as I mentioned?

  • 0 in reply to Shirish
    Intellectual 330 points

    Also, I'm curious what is the standard practice for preventing over discharge from the draws due to the Fuel Gauge and LDO? Surely their power consumption is very small, but it isn't zero, and I'm not a fan of just letting the battery continue to drain past a safe lower threshold.

  • 0 in reply to Christopher Buckley
    TI__Guru 57697 points

    Hello Christopher,

    The shutdown current consumption will be in the datasheet. You can use a protector to prevent discharging past undervoltage

  • 0 in reply to Shirish
    Intellectual 330 points
    Shirish said:
    You can use a protector to prevent discharging past undervoltage

    Could you please be more specific? Are there any TI parts you can link to that exemplify what you are recommending here?

  • 0 in reply to Shirish
    Intellectual 330 points

    Thank you, but I'm still not quite clear on how this would prevent discharging past undervoltage. The Shutdown Iq = 100 nA on the BQ297xx mentioned in your link, and the BQ27742-G1 claims a nominal 0.1 uA I_Shutdown as well. How would adding BQ297xx or any other protector prevent discharging past undervoltage any more than the BQ27742-G1 is already doing? I'm aware LiPos can self-discharge, so is this 100nA somewhat equivalent to this such that there is no point in trying to get less than that?

  • +1 in reply to Christopher Buckley
    TI__Guru 57697 points

    Yes, that is the minimum current and is comparable to self discharge. I do not think you need a rocker switch if it is only to communicate with the gauge. When a load is connected, then the gauge will communicate.

  • 0 in reply to Christopher Buckley
    Intellectual 330 points

    Thank you Shirish. I am trying to implement this now and tested it with a charger, and seems to work as expected. A follow up question to this though is what is the correct order of operations for shutting down when there is no charger attached? Is it it the same? In the guide it says:

    4.1.1.13 SET_SHUTDOWN: 0x0013
    Sets the CONTROL_STATUS [SHUTDN_EN] bit to 1, thereby enabling the fuel gauge to shutdown if
    conditions are met.
    When the [SHUTDWN] bit is set, the gas gauge opens both charge and discharge FETs and waits for
    charger removal. As soon as charger is removed, the gas gauge and protector will shutdown leaving both
    charger and discharge FETs open. The only way to recover from this SHUTDOWN mode is to connect a
    charger.

    But what if there is no charger attached in the first place? Can I just SET_SHUTDOWN and it enters the SHUTDOWN state right away (i.e. skips the SHUTDOWN WAIT state and goes straight to ANALOG SHUTDOWN state)? Is there any way to verify this is working as intended? Even the ProtectorState() 0x78 Extended Command doesn't seem to have a way to return that it is in the ANALOG SHUTDOWN state (which I guess makes sense), but I'd just like to verify I am doing this right.

  • 0 in reply to Christopher Buckley
    TI__Guru 57697 points

    Hello Chris,

    The documented conditions require charger removal. I am not sure if not having a charger will still send it into shutdown when the other conditions are met.

    Measuring power consumption is the correct way to check shutdown.

    As far as i know, Shutdown will disable communication, so that may be an alternate way to check if it enters this state

  • 0 in reply to Shirish
    Intellectual 330 points

    Thanks Shirish. I tested this a bit, and the current drops to something less than my multimeter can measure (good enough I guess), and its not responding to I2c, so I think that verifies shutdown as well as I can.

    Next step is how to startup again without attaching a charger. This is a battery powered device, so I don't want to rely on plugging it in to get it to start. I'm looking to implement something similar to a power button on a smartphone (long press to shutdown, long press to startup). Is this achievable with the ANALOG SHUTDOWN state, or would I have to rely on something a little more power hungry in order to allow the bq27742-g1 to respond to i2c comms?

    I tested one alternative. Here is a summary of that test:

    1. Attach battery as per the datasheet. One or both fets associated with the bq27742-g1 are off (i.e. no power to the circuit from the battery)
    2. Attach charger, and start firmware.
    3. Detach charger
    4. Firmware loops a few times and shuts down. This shutdown includes setting the shutdown bit on the bq27742-g1. This triggers fets associated with the bq27742-g1 to turn off and circuit is depowered.
    5. Short BATT+ with PACK+ like mentioned in the guide. I didn't check the state of the bq27742-g1, but fets turn on, and firmware starts as if a charger were attached. Steps 4-5 repeat.

    So going back to my original question, I guess it still makes sense for me to add some sort of button that can temporarily short BATT+ with PACK+ as a way to  startup the circuit after putting the bq27742-g1 into ANALOG SHUTDOWN. Is this advised against for any reason?

  • +1 in reply to Christopher Buckley
    TI__Guru 57697 points

    Hello Christopher,

    bq27742-G1 is a pack side gauge. Waking from shutdown is equivalent to resetting the gauge.

    Shutdown mode is only used when the battery is put into storage. Since self discharge could have discharged the battery to a very low level, it is good design practice to require a charger to be attached for wake up. You can short Bat+ to Pack+ to simulate that kind of condition but it that is not recommended.

    For normal system operation, sleep modes will automatically put the gauge power consumption below battery self discharge while still maintaining accurate gauging.

  • 0 in reply to Shirish
    Intellectual 330 points

    Got it, thanks Shirish. I think this is starting to make more sense to me now.