• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 3 answers
  • Subscribers 65 subscribers
  • Views 727 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.

BQ35100: Using a coulomb counter with a lithium primary cell in a smart meter

user1082833
Intellectual 330 points
Part Number: BQ35100
Other Parts Discussed in Thread: CC1101, MSP430FR5994

I want to monitor the current consumption of a smart meter with IOT communication. It should work with a small 3,6V Li-SOCL2 battery for many years.

When all the communication is off, the resulting current is smaller than 10uA. When the IOT wakes up, we could have very short pulses up to 800mA for several mS each (an additional capacitor protects the battery from the pulses). The problem is that we don't know when they occurs. The integration of the sleep current and the IOT current gives us an average current of nearly 50uA.


Now we search a device to predict the state of the battery (coulomb counter). If I read the datasheets, it seems to me that all the coulomb counter needs a current of approximately 100uA, when they are on. Is that right? Because in that case, the coulomb counter will need two third of the battery capacity.

Without CC: 2600mAH at 50uA average => 6 years for one Battery
With CC: 2600mAH at 150uA average => 2 years for one Battery


Please could you help me to better understand the work of a coulomb counter with an IOT device, where we don't know, when the pulses occurs. Witch one could you advise us to use for this application?

  • 0
    TI__Guru 52115 points

    Your question has been assigned to our engineer and we will get back to you.

    Andy

  • 0
    TI__Genius 10370 points

    The bq35100 is designed to be asleep most of the time and it heavily revolves around Host controlling its actions. If you have no idea when the load will occur then this part will most likely not be the right fit for you. 

    The only option you would have is to leave it active the entire time which is most likely going to be too power expensive. 

    I would encourage you to find a way to detect a pulse will happen prior to it actually happen so you can wake the gauge.

    thanks,

    Eric Vos

  • 0 in reply to Vos
    Intellectual 330 points

    Thank you for your reply.

    In that case, the selected coulomb counter wouldn't be the right device for our application, because we don't know, when the pulses occurs, and so the power consumption of the coulomb counter would be to important.

    We have an NB-IOT device (with UART) and a CC1101 (with SPI) on our PCB. Both devices works in an autonomous hearing mode (PSM and hearing strobes alternate). And they only gives an interrupt to the micro-controller, when they received wireless data. So the micro-controller (MSP430) can't know when the pulses occurs. Furthermore the output power of the NB-IOT device is automatically adjusted to the needed power by the device itself (depending on the reception quality).

    Do you know another approach to the problem of measuring the real power consumption in IOT devices?

    Eventually the problem could be resolved by a measure of the internal resistance of the batteries (the voltage isn't an indicator for Li-SOCl2 batteries). Do someone know something about it, or tested this approach?

  • 0 in reply to user1082833
    TI__Mastermind 24870 points

    Hello User,

    For IOT, this will not be a good fit. It does not meet many of the power requirements.

    I have seen some simple implementations with a FET, but that requires your host processor.

    You could try putting this device into shutdown and justmeasure the raw resistance.

    I encourage you to try it on an EVM first prior to implementing in system.

  • 0 in reply to Kang Kang
    Intellectual 330 points

    Thank you very much for your reply. This sounds very interesting.

    Unfortunately, I couldn't find the relevant informations about the measuring of the raw resistance (the measuring of the current by a FET).

    Please could you show me a link, where I could find these informations. How does it work? We actually use the MSP430FR5994, so we have a free ADC12 , that we could use for the measuring, if necessary.

  • 0 in reply to user1082833
    TI__Genius 10370 points

    The bq35100 also does gauging based on impedance. This is the EOS mode of the device. 

    For your situation to use this mode you would need to install a "Learning Load" on the board that you can use to control when the load is applied. 

    So you would not use your radio as the source to measure the impedance, but instead just some resistor on your board. This will use power, but the amount it uses will be minimal based on the amount of time it needs to be on.

    Thanks,

    Eric Vos