• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 63 subscribers
  • Views 235 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.

some doubts on system side gauge(BQ27425)

liting Lei
Intellectual 325 points
Other Parts Discussed in Thread: BQ27510, BQ27500

I have some doubts regarding system side gauge.

1. power on device immediately when a well-relaxed battery is attached.(no adaptor is attached)

in above case, if the gauge can provide the accurate SOC？ My concern is that gauge will take some to take the OCV measurement when battery is inserted, maybe OCV measurement is not completed before power on device action. if so, the SOC prediction accuracy will be impacted.

2. when a not well-relaxed battery is attached for the first time, whether OCV will be taken or not?(no adaptor is attached)

if yes, gauge provides a inaccurate SOC prediction, correct?

3. the device with a deep depleted battery, then plug in an adaptor. how does the gauge work?

 

best wishes,

Jacky

  • 0
    TI__Genius 14985 points

    Hi Jacky,

    You point out some challenges for system-side gauges which are very valid.  This is why a pack-side gauge is always the ideal choice, but many customers must work with "dumb" batteries that do not have a gauge.  Also, consider the cases where the user might switch to third party batteries, or if you have multiple suppliers of cells/batteries.  How can a gauge be accurate if the type of battery being used might change at any moment?

    Our system-side gauges (like bq27510, bq27520, bq27425, bq27421) have features to mitigate the error sources you pointed out.  All gauges need to initialize their SOC "guess" when they are first powered up.  When a battery is first inserted, the best results will come if they can take an accurate voltage measurement of a very well-relaxed and stable voltage.  However, there is no guarantee that the battery was not recently being charged or discharged so the voltage might not be stable.  Also, the system might immediately boot and drain current, or if a charger is plugged in then it might begin charging immediately. 

    Our gauges have features to try to estimate the OCV (open-circuit votlage) even though current is flowing.  Of course, the smaller the current, the more accurate the estimation.  Therefore it is ideal to try to postpone high current activity from the battery for a few seconds when the system is first powered up so that the gauge can take a voltage reading with as little current as possible.

    Some of our gauges have a BAT_GD pin which can be used to disable the charger until the gauge has completed taking its voltage measurement.  There are also bits in the gauge registers which can be used to tell when the initialization is complete.  These bits can be polled, or some gauges have an interrupt to notify the host.

    Another consideration is that the user might be swapping between two batteries.  One might be old and one might be newer.  The best gauge accuracy will be obtained if the gauge has had an opportunity to learn the battery capacity and impedance.  If the gauge has learned the old battery characteristics but the user swaps in the new battery, there will of course be some error until the gauge learns the newly inserted battery characteristics.  This can take some time and the user might even swap back to the other battery before the gauge can learn.

    Therefore, some of our system-side gauges have a "battery arbitration" feature.  This allows it to learn two different battery Qmax and Ra values as well as remember the default values.  During a discharge, it can quickly try to match the battery being used to one of those three profiles and quickly regain accuracy by switching if necessary.  Gauges with this feature include bq27500, bq27510, and bq27520.  See section 1.5 of this app note for more details: SLUA439

    Despite these features, it may be impossible to achieve maximum accuracy instanty when an unknown battery is inserted to the system and introduced to the gauge.  However, we include other features such as Fast Resistance Scaling to help the gauge reach empty 0% at the correct time, as well as other features to allow the gauge to quickly learn the newly inserted battery.  The end goal is to ensure the user experiences no surprises and can still use the maximum available capacity from the battery.

    Besides "instant" SOC and SOH accuracy, putting the gauge inside the battery pack gives you many other features, including lifetime data (like a black box to store pack history), additional safety features and traceability, authentication (to ensure users are using genuine safe packs/cells), and more.

    You might notice that I've been talking about single-cell system-side gauges.  Typically multi-cell packs have a gauge inside the pack (just like your notebook computer) because they not only need instant accuracy, but the gauge also performs other critical safety functions like cell balancing and more.  Nonetheless, we do offer gauges like bq34z1xx which can be used with a voltage divider to allow you to gauge a multi-cell pack from the system-side.  The same concept can be applied to our dedicated single-cell gauges as in the example shown in SLUA496.

    Customers have a variety of requirements so we've tried to ensure they can get the benefits of Impedance Track accuracy no mattery where they must place the fuel gauge and no matter what the use case of the batteries is.