• State TI Thinks Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 3 answers
  • Subscribers 62 subscribers
  • Views 378 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ20Z655-R1: Battery PCM circuit is confused.

Iz Olmez
Prodigy 60 points
Part Number: BQ20Z655-R1
Other Parts Discussed in Thread: BQ294522

I am using a battery in one of our products, which employs BQ20Z655-R1, along with BQ294522 in its battery charge mgmt and protection circuit.

When our product is idle, we use a 200uS pulse every 7 seconds to load the battery with 3A (equivalent operating current), read the VBAT voltage, latch it to a LED indicator circuit and keep it until the next cycle. The LED indicator is a very low current 2-color LED ( to show 3 regions: "full", "half", and "near empty" ) which draws ~ 500uA. This is a 24/7 operation as long as the system is idle, i.e. shelf time.

We have discovered that after a couple of months of shelf life some of the units display low battery voltage; although the battery reports full charge through its SM bus. The battery does not take charge current in this condition, but if I take it out of the product for ~1 hr and place it back in, the voltage recovers to "full" region and it also takes charge current.

Seems like something -either the voltage monitor pulses, or the constant LED current for a long time, puts the battery in this mode, where it thinks it is fully charged and also shows high ISR -hence the voltage drop. Has anybody encountered a similar problem?

Thank you.

  • 0
    TI__Mastermind 42535 points
    This part is not designed for such use. Under shutdown your current draw is very high. If the gauge is on and in sleep mode it will take OCV readings every 5 hours and adjust capacity accordingly. The interval for taking OCV may also be shorter depending on the dV/dt conditions. I suspect that once your battery is fully charged the chg fet turns off. It remains in this state until another successful OCV reading happens after the rest period. Once this reading is taken and it is seen that the FCC can be adjusted per the OCV, the chg fet is closed again to allow charging.
  • 0 in reply to Batt
    Prodigy 60 points
    Hi Batt,

    Thank you for the answer. I would like some clarification as to which current draw you are referring to. Is it the 3A draw for 200 microseconds every 7 seconds or the constant LED indicator current of ~500 microamps?

    Thanks.
  • 0 in reply to Iz Olmez
    TI__Mastermind 42535 points
    Actually both will cause the gauge to not take OCV. OCV is the open circuit voltage measurement that we use as a reference to adjust capacity. In your application, the gauge literally never really goes to sleep or satisfies the dV/dt condition required to take an OCV.
  • 0 in reply to Batt
    Prodigy 60 points
    Hi Batt,

    I found an app note "Application Report SLUA364B–November 2005" on the subject that explains the operation, and based on your answers and my understanding of the app note I conducted a couple of tests today which I would like to discuss the results with you. Where are you located? Is there a phone # I can call you to discuss this in detail?

    Thanks,
    _Iz
  • 0 in reply to Iz Olmez
    TI__Mastermind 42535 points
    Sorry, I'm not available over the phone. We can discuss this over here if you like.
  • 0 in reply to Batt
    Prodigy 60 points

    Hi Batt,

    I'd like to refer you to the TI app note  "SLUA364B–November 2005–Revised December 2006".

    According to the app note, page 3: the Quit Current should not exceed C/20. My battery's capacity is 3A, and 3A/20 = 150mA.

    Also on page 3, figure 2: There seems to be a 30 min period of relax time needed before the 1st DOD reading. After that time the IC starts the DOD reading cycle to update the Qmax and decide whether to take new charge or not. 

    I think in our product, after the initial charge, if the unit is shelved for long time, our pulse per 7S keeps discharging the battery very slowly, but the IC never sees it relaxed so that it never updates the Qmax to open the gates for recharging.

    My questions are these:

    1) Am I correct in my assumption above?

    2) When in this mode (not taking charge) can we wake the IC another way to take charge current? i.e. applying the full load -operating the unit- for 10S? 20S, 30S?

    Below is what I did to make the battery take charge current again, based on what I deducted from figure 2 of the app note SLUA364B:

    My circuit idle + LED current is <1mA, so I left the LED on, but disabled my 200uS per 7S pulse that draws 3A; waited 0.5 hr + a few minutes and enabled my pulse. On the 2nd pulse the battery voltage recovered and the battery started taking charge current.

    I would appreciate your answers, especially to question 2, and your suggestions on how to wake the system up.

    Thank you,Evaluation 3_ Theory_Implementation Impedance Track Battery Fuel-Gauging Algo (R ev. B.pdf

    _Iz

  • 0 in reply to Batt
    Prodigy 60 points
    I have not heard a suggestion or answer to my questions I posed on my last post dated Jan 15, 2019 6:24PM
  • 0 in reply to Iz Olmez
    TI__Mastermind 42535 points
    1. Yes
    2. Yes, if you wake the IC from shutdown it will take readings and you may get a better accuracy.