• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 720 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: ISD issue

Ruixin Feng
Intellectual 2445 points
Part Number: BQ27742-G1


Tool/software:

Hi team,

Customer feedback battery pack will randomly trigger ISD fault after completing charge. We record log for your reference. And my question is that is selfdischargecurrent accurate? how often selfdischargecurrent will refresh?  How to set ISD threshold is correct?  

Below is battery configuration and log data which ISD is triggered.

Battery 2- ISD triggered.xlsx

Below is battery configuration and log data which ISD isn't triggered.

Battery 1 - ISD untriggered.xlsx

Thanks!
Rayna

  • 0
    TI__Mastermind 29245 points

    Hello Rayna,

    This question has been assigned and will be answered promptly.

    Thank you,
    Alan

  • 0 in reply to Alan Frias
    TI__Intellectual 2445 points

    Hi Alan,

    could you pls help to push this answer? Customer has been waiting for a week. Thanks!

    Rayna 

  • 0
    TI__Mastermind 29245 points

    Hello Rayna,

         The before giving a solution I wanted to walk through my thought process for it. The ISD Current Threshold only gets trigged when "Design Capacity" / " ISD Current threshold" is greater than "SelfDischargeCurrent" ; typically we don't ever recommend to change the Design capacity since it's based on the cells so we narrowed our focus to changing ISD Current Threshold. Going back to your excel I looked into the "SelfDischargeCurrent" Column between the triggered and not triggered excel and oddly noticed that the excel labeled "ISD untriggered"(-14mA) had a discharge current compared to "ISD triggered"(-3mA).
         Although the shift range seem to be around 14mA , could you say what you have your "SelfDischargeCurrent" value at because what you would need to do is take the SelfDischargeCurrent > "Design Capacity" / " ISD Current threshold" and find out what value you can change you "ISD current threshold"  and add 25% margin on top of it. Currently your  ISD Current threshold is set at 30,000Hr rate and te default is 10. please refer to the technical reference manual at 5.3.5.8 to understand the setup and background for this question 

    technical reference manual:
    https://www.ti.com/lit/ug/sluuax0c/sluuax0c.pdf?ts=1740673176869&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FBQ27742-G1

    Thank you,
    Alan

  • 0 in reply to Alan Frias
    TI__Intellectual 2445 points

    Hi Alan,

    Customer configured that Design capacity is 1800mAh, ISD current is 3000Hr rate, So ISD threshold is configured as -1800/3000=-0.6mA. Is this wrong?

    Customer's main concern is that selfdischargecurrent is not accurate and then cause the threshold is hard to set. 

    Could you pls help share how gauge work out selfdischargecurrent and make sure it's accurate enough? Thanks!

    Rayna

  • 0 in reply to Ruixin Feng
    TI__Mastermind 29245 points

    Hello Ruixin,

         The way the SelfDischargeCurrent() is calculate is through "computed based on two opencircuit voltage reads separated by a 1-hour interval. from there, the charge passed and time elapsed between the two OCV points is used to compute the amount of SelfDischargeCurrent()". Essentially the capacity should hold within a certain range while factoring IC combustion.
         The accuracies of the gauge is more precise than 1mA and is more than enough for the application however we noticed that based on the current picture above of the data memory you have the ISD threshold set at 30,000Hr rate which means you ISD threshold is -1,800/30,000=-0.06mA which is extremely low and would definitely explain all the false positives. 
         Most clients leave the ISD threshold at the default(10Hr rate) but if you would like to set it to conservative rate for it to only work on new and quality cells that try to set your SelfDischargeCurrent() to trip at 1mA or -1800/X=1mA -> X=180Hr rate

    *Reference previously give material 


    Thank you,
    Alan

  • 0 in reply to Alan Frias
    TI__Intellectual 2445 points

    Hi Alan,

    Thanks for support here!

     May I know how to calculate OCV from selfdischargecurrent? As the log data showed, selfdischargecurrent of battery 1 is -7. How to know the OCV points where -7 comes from? Thanks!

    Rayna

  • 0 in reply to Ruixin Feng
    TI__Mastermind 29245 points

    Hello Ruixin,

    There might be some confusion since the selfdischargecurrent is measured in mA meaning that your system is discharging at 7mA. OCV is measured from the BAT and VSS pin during periods of no load on the cell. OCV and selfdischargecurrent don't necessarily effect each other directly since OCV measurements wouldn't be taken during discharge periods.

    Thank you,
    Alan 

  • 0 in reply to Alan Frias
    TI__Intellectual 2445 points

    Hi Alan,

    Alan Frias said:
    The way the SelfDischargeCurrent() is calculate is through "computed based on two opencircuit voltage reads separated by a 1-hour interval. from there, the charge passed and time elapsed between the two OCV points is used to compute the amount of SelfDischargeCurrent()".

    I'm a little confused. You mentioned OCV is used to compute selfdischargecurrent. Customer was wondering how OCV is used to compute selfdischargecurrent.

    The reason why asked this question is that if you checked the log data, the voltage of battery didn't changed a lot. But the selfdischargecurrent indicates battery voltage changed a lot. And for a normal battery, the selfdischarge current should be in uA level. So 7mA is too large to believe. Customer doubt if the selfdischargecurrent is correct.

    Thanks!
    Rayna

  • 0 in reply to Ruixin Feng
    TI__Mastermind 29245 points

    Hello Ruixin,

    I misspoke on SelfDischargeCurrent(), but in terms of the way it is calculated you can find it in section 2.2.1.5 of the technical reference manual. However to properly answer the question from yesterday, you would know the OCV points are coming from an 1 hour interval and the charge elapsed would compute the selfDischargeCurrent(). If the customer does see a 7mA than there is some kind of load within there system and also I agree that this is unusual but I also believe that it wouldn't be coming from the battery gauge end due to the lower power use used on the IC.

    Thank you,
    Alan