• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 309 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.

BQ25890: Several questions on charing with BQ25890 and BQ27441

Feng Xu
Intellectual 390 points
Part Number: BQ25890

I use BQ25890 as the charger chip, and use the BQ27441 as the voltameter.

I read the register of the two chips every 10 seconds, and plot the data into 5 graphs.

Since the graphs are large, they are attatched as a RAR below.

I have 9 questions about the charging progress, the 9 questions are labeled in the graphs.

And the data are labeled with different colors, if the data name is begin with "27441", it means it's the data of BQ27441. And other data are all BQ25890 data.

Q1 point: Why does the charging current begin to decrease at this point, where the SOC is only 70% (get from 27441)?

       Yes, maybe the 27441 has not learned, the SOC is not accurate. I just want to know how does BQ25890 knows that the battery is nearly full?  It decreases the current when the VBAT rise to 4.1V?

Q2 point: When the charging current decreases to 400mA, the charging current begins to switch between 800mA and below 400mA(200mA - 400mA). Why?

      I have checked that, the REG0D_VINDPM is always 950mA through the charging period in the graph.

Q3 point: At this point, the charging current decreases from 800mA to 780mA and then jumps to 0. The chip get into charging termination state.

     What causes the termination at this point? The VBAT has only got 4.284V, not reaching REG06_VREG(I set 4.336V).

Q4 point: The SOC of BQ27441 is 91%. If I discharge the battery to 0% and recharge it again, can it get to 100%?

Q5 point: The chip is always in IDPM Status. I still don't understand the IINDPM. What is IINDPM, I didn't get clear from the datasheet.

      I understand VINDPM. For example, if the VINDPM_OS is 0.6V, when USB is plugged in, the 25890chip will test the adapter. If the VBUS is 5.2V at the pin of 25890 when no load is put, the 25890 will control the charing current so that hte VBUS pin is always above 4.6V.

      But for IINDPM, I don't know. How does it work? And in my test, why IDPM Status is always 1?

Q6 point: NTC warm is got. That is because I put the board and battery too close, but hasn't got NTC hot. And I see REG0E_THERM_STAT is always 0, hasn't got into thermal regulation.

       Does NTC warm affect the charging progress?

Q7 point: At this point, why the chip quit IDPM Status?

Q8 point: There are two points in VINDPM status. One point is at 771 in x axis, where the VBUS is 4.6V. Another point is at 726 in x axis, where the VBUS is 4.7V.

                 Why it goes into VINDPM when VBUS is 4.7V?

Q9 point: At this point, why the chip terminate charing? The VBAT is only 4.205V, hasn't got 4.336V.

DataGraph.rar

  • 0
    TI__Guru** 105590 points

    Hello Feng,

    It looks like the .rar file shared only had 5 images in it, and I don't see the where the questions are pointed out.

    Q1: Usually the charge current begins to fall because you have reached the Constant Voltage portion of the charge.

    Q2: I'm not sure why your current is jumping, can you share your register settings so I can take a look for the bq25890?

    Q3: If you are reading the VBAT with the charger ADC is may be due to the ADC accuracy, combined with the voltage regulation error it could cause a difference that you see.

    Q4: You may need to adjust your charger termination to coincide better with your battery.

    Q5: IINDPM means your charger will not pull more than that limit from VBUS.

    Q6: Yes the charger is can run a JEITA charging profile, this is dependent on the temperature.

    Q7: Not sure where this point is in your pictures.

    Q8: It depends on the ADC accuracy you are using and the +/- percentages of the threshold.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 390 points

    Hi Keller,

           I put "Qx" label on the images and put question discription in the post. And I have updated the graphs with short line to point the data point where the question is about.

           Q1: REG06_VREG is set as 4.336V, is the corresponding Constant Voltage 4.1V?

                  Corresponding to the Q1 point in image5, the BATV is only 4.1V in image1, and the BATV increases to 4.3V in image1 in the left time.

           Q5 & Q7:

                  Q7 is shown in image3, at the point of 574 in the x axis, at that time the 27441avgCurrent is 483mA.

                  How does exactly the IINDPM works?

                  For example, in my situation the REG0D_VINDPM is 950mA. Why the IINDPM status is always 1 before Q7 point, and it becomes 0 at Q7 point?

           Q2: The registers are shown below:

    [0x00] 0x7F
    [0x01] 0x06
    [0x02] 0x15
    [0x03] 0x1A
    [0x04] 0x19
    [0x05] 0x11
    [0x06] 0x7F
    [0x07] 0xC9
    [0x08] 0x01
    [0x09] 0x44
    [0x0A] 0x73
    [0x0B] 0x6F
    [0x0C] 0x00
    [0x0D] 0x12
    [0x0E] 0x0D
    [0x0F] 0x0B
    [0x10] 0x3A
    [0x11] 0x97
    [0x12] 0x00
    [0x13] 0x42
    [0x14] 0x5D

            Thank you.

    DataGraph.zip

  • 0 in reply to Feng Xu
    TI__Guru** 105590 points

    Hello Feng,

    Voltage differences this close can be from IR losses, depending on where you measure the voltages. You can use IR compensation if you would like to mitigate the voltage difference and your power path has a known resistance.

    It looks like you're going in and out of NTC fault and hitting VDPM, this may be causing your oscillations near end of discharge.

    IINDPM limits the input current so the adaptor does not become overloaded. 

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 390 points

    Hi Keller,

         Q1. Do you mean that, normally, it should rise to 4.336V first, and then the chip get into the Constant Voltage portion, in which the current drop slowly.

                And in my case, maybe the power path resistance between battery and bq25890 is too large, so that it begins Constant Voltage portion when bq25890 thinks it's 4.336V while it's actually 4.1V at the battery point?

                Do I understand right?

         Q2. I used to think that the NTC_FAULT = TS_WARM won't affect the charging progress, only TS_HOT matters. Now I see, TS_WARM also matters. Thanks.

         Q5 & Q7. "IINDPM limits the input current so the adaptor does not become overloaded. "

          Is that what the VINDPM do? What is the difference between VINDPM and IINDPM?

          How does bq25890 knows the adaptor is overloaded? It monitors the voltage drop in the VBUS PIN. I think that is VINDPM. What is IINDPM?

          Thank you.

    Frank

  • 0 in reply to Feng Xu
    TI__Guru** 105590 points

    Hello Feng,

    Yes that's correct, that is why we have implemented the IR compensation in our chargers.

    Both VINDPM and IINDPM are designed to prevent overloading of the adaptor, VINDPM limits based on the adaptor voltage, the other limits based on adaptor current.

    Sincerely,

    Wyatt Keller

  • 0 in reply to Wyatt Keller
    Intellectual 390 points

    Hi Keller,

         "the other limits based on adaptor current"

         Can you explain more, how it judges how much mAs the current should be limited? In my case, the adapter is a 5V2A adapter for HUAWEI phone, but the IDPM_LIM is set 950mA by the BQ25890. I see the REG11_VBUSV is stil 4.8V when in IDPM_LIM, since the VINDPM_OS is 0.6V, and the VBUS is 5.2V when no load is attatched, I think the chip can draw more current since there is still 0.2V torrelance.

         Q2: And if the IDPM_LIM is 950mA, the BQ25890 should limit the input current below 950mA. Will it limit the current to for example 750mA or 650 mA when IDPM_LIM==950mA?

         Q3: Will the IDPM_LIM changes during working if the VBUS is not removed? Or it is just determined when the VBUS is plugged in and will not change unless VBUS is removed?

    Frank

  • 0 in reply to Feng Xu
    TI__Guru** 105590 points

    Hello Feng,

    Your adaptor may have a very lossy cable and cause the VINDPM to be hit earlier than IINDPM, you would need to take scope captures at the different nodes with the corresponding currents to identify what is the cause.

    The charger will limit it slightly below the threshold, IBUS should not be 300mA below IINDPM.

    Please check section 9.2.4 Input Current Optimizer (ICO) of the datasheet for when the algorithm runs.

    Sincerely,

    Wyatt Keller